JiNSTITUTl0N^N0lini!iSNI^NVIN0SHil/JS^S3 I dVy a ll^LI B RAR I ES^SMITHSONIAN^INSTITl ssaiavaan LIBRARIES SMITHSONIAN INSTITUTION NOliflillSNI NVIN0SHillNS^S3 I d V CO Z . CO ^ Z T CO 2 _1NSTITUTI0N NOIiniliSNI_NVmOSHillNS S3iaVHan_LIBRARI ES'^SMITHSONIAN^INSTITL _ \h;asi^ q X^osHj ^S3 I ava a n “"lI B RAR l ES^SMITHSONIAN“'lNSTITUTION^NOIiniliSNl"'NVINOSHillMS ^S3 I a V ^INSTITUTION^ N0'J-nillSNI^NVIN0SHilWS^S3iavaan^LIBRARIEs|^SMITHS0NIAN"lNSTITU ~ ^ •- ? .'i i I |^S3 I avaa LI B RAR l ES^SMITHSONIAN^INSTITUTION^^NOlinillSNI^NVINOSHilWs'^Sa I aVi i ^ CO ■■■- Z - _INST!TUTI0N^N0liniliSNI_NVIN0SHillNs|^S3 I a Vd a II^LI B RAR I Es'^SMITHSONlAN^INSTITUl LniliSNI NVIN0SHiIWS^S3 lavyan^LIBRARI ES^SMITHSONIAN^INSTITUTION^NOlinillSNI^t RARIES SMITHSONIAN INSTITUTION NOIinillSNI NVINOSHillAiS S3 I aVd B H LI B R AR I ES W 2 ^ t. ' 5: ^mgsva^ > : iniliSNl^NVIN0SHilWs‘^S3 I a Vd 0 H^LI B RAR I Es'^SMlTHSONIAN INSTITUTION NOlifUliSNI jVA»Vl^ O — RARIES SMITHSONIAN INSTITUTION"’NOIiniliSNl"'NVINOSHilWS S3ldVaan LIBRARIES Z r- 2 r- . 2 r- . — . 2 _ CO ^ — c/> X ^ IfUliSNI NVINOSHimS S31dVaan libraries SMITHSONIAN institution NOlinillSNI ;n 5 IRAR I ES“ SMITHSONIAN INSTITUTION NOlifUliSNI NVINOSHims“'S3 I d Vd a h" LI B RAR I ES' X2iu52i' o ^ H • 77 w'iisv ” ;□ w O \Om pc^ iniliSNCNVIN0SHimS^S3 I dVd a n~LI B RAR I ES^SMITHSONIAN^INSTITUTION^NOUniUSNr ! RARIES SMITHSONIAN INSTITUTION NOliOiliSNI NVINOSHilWS S3ldVdan LIBRARIES ■">1^' > 2 >■' i XivosH^/ >■ ^ I iniIiSNI_NVINOSHilWs‘^S3 I d Vd a II^LI B RAR I Es'^SMITHSONIAN INSTITUTION _NOIiniliSNI_ s- JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. XXV No. 3 (112) January 1966 CONTENTS Page Check List of Elopoid and Clupeoid Fishes in East African Waters, by G. F. Losse 166 A List of Birds seen in Southern Tanzania, by C. J. Tweedy 179 Keys to the Genera of Insectivora, Chiroptera and Rodentia of East Africa, by J. B. Foster and A. Dulf-Mackay 189 The Habitat of the Rock Hyrax, by J. B. Sale 205 Daily Food Consumption and Mode of Ingestion in the Hyrax, by J. B. Sale 215 Notes on East African Cowries, by E. Robson 225 Sight Additions to the Avifaunal List of Ethiopia, by Edward W. Beals 227 A Reef Heron at Lake Nakuru, Kenya, by Myles E. W. North 231 A further Note on Reef Herons in East Africa, by Alec Forbes-Watson 233 Nature Note: Lake Abiata, Ethiopia, Weavers and Gabar Goshawks, by P. M. Allen 230 Review 234 Appreciation 235 (Published 15.4.66) Price Shs. 15/- EAST AFRICA NATURAL HISTORY SOCIETY PRESIDENT: Dr. M.J. Coe. VICE-PRESIDENT: Dr. A.D.Q. Agnew. EXECUTIVE COMMITTEE. J.A. Wood, Esq. L.H. Brown, Esq. M.E.W. North, Esq. R.H. Carcasson, Esq. Miss. E.J. Blencowe. Miss. J.R. Ossent. Mrs. D. Fieminq. B. Parsons , Esq. J. Smart, Esq. HON. EDITOR. ^Dr. P.J. Greenway. HON. TREASURER. A.G.T. Garter, Esq. SECRETARY Mrs. F. Nq'weno. All correspondence in connection with this Journal should be addressed to: The Secretary, East Africa Natural History Society, P,.0. Box 4486, Nairobi, Kenya. JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. XXV No. 3 (III) January 1966 CONTENTS Page Check List of Elopoid and Clupeoid Fishes in East African Waters, by G. F. Losse 166 A List of Birds seen in Southern Tanzania, by C. J. Tweedy 179 Keys to the Genera of Insectivora, Chiroptera and Rodentia of East Africa, by J. B. Foster and A. Duff-Mackay 189 The Habitat of the Rock Hyrax, by J. B. Sale 205 Daily Food Consumption and Mode of Ingestion in the Hyrax, by J. B. Sale 215 Notes on East African Cowries, by E. Robson 225 Sight Additions to the Avifaunal List of Ethiopia, by Edward W. Beals 227 A Reef Heron at Lake Nakuru, Kenya, by Myles E. W. North 231 A further Note on Reef Herons in East Africa, by Alec Forbes-Watson 233 Nature Note: Lake Abiata, Ethiopia, Weavers and Gabar Goshawks, by P. M. Allen 230 Review 234 Appreciation 235 (Published 15.4.66) i;!'V: -V Prirrl'v”^ ■■ ■-•■" c ■' ''^> v! ■ i-r/i'i''''^" ■ ■ .77' :'' ,■:;. ■■v-*‘^i?,iyilW l,«*»l|i«lM ..;,.ffi-.!-,l.J...lLMll,f :7'':4M !■ . ' «ilt>j '!•'" '1,1 'Af'i > .,'i, ; 'Vt II; HL ''■■\^„ '.. ■ ■ .’ . ••<>! jii ■„. ,. •f'i'. l;•:•■' . ■/- ' "'■ , ■■ ■ y-\:7 . ’ ' *■•' ,\ 'i, ' v/v ■ • ,' ■'■ ■(" ' ■!' ^'.ir ,. . , , ■;■ ',| _ ^ ',^, ;' * , ; ,■ I' id; ( '/•;.< s^jT ja .., h V *" - ^ -»•• •■ 1 T- , r^*-: •■';;,< -if ' ' pjSv.iiA,' ' 7>,\,0%!li. pa mm. iiii^i J.E.Afr.Nat .Hist .Soc. Vol.XXV No. 3 (112) January 1966 CHECK LIST OF ELOPOID AND CLUPEOID FISHES IN EAST AFRICAN COASTAL WATERS By G.F. LOSSE (East African Marine Fisheries Research Organization, Zanzibar.) Introduction During preliminary biological studies of the economically Important fishes of the suborders Elopoidei and Clupeoidei in East African coastal waters, it was found that due to considerable confusion in the existing literature and the taxonomy of many genera, accurate identifi- cations were often difficult. A large collection of elopoid and clupeoid fishes has been made. Specimens have been obtained from purse-seine catches, by trawling in estuaries and shallow bays, by seining, handnetting under lamps at night, and from the catches of indigenous fishermen. A representative part of this collection has now been deposited in the British Museum (Natural History), London, where I was able to examine further material from the Western Indian Ocean during the summer of 1964. Based on these collections this check list has been prepared; a review on the taxonomy, fishery and existing biological knowledge of elopoid and clupeoid species in the East African area is in preparation. Twenty-one species, representing seven families are listed here; four not previously published distributional records are indicated by asterisks . Classification to familial level is based on Whitehead, P.J.P. (1963 a). Keys refer only to species listed and adult fishes. In the synonymy reference is made only to the original description and other subsequent records from the area. Only those localities are listed from which I have examined specimens. East Africa refers to the coastal waters of Kenya and Tanzania and the offshore islands of Pemba, Zanzibar and Mafia; Eastern Africa refers to the eastern side of the African continent, i.e. from the Red Sea to Natal. 166 Check List of Elopoid and Clupeoid Fishes Key to the Suborders and Families 1. Lateral line present on body; abdominal scutes and pelvic scute always absent Suborder ELOPOIDEI (i) Mouth terminal; upper jaw bordered by maxillae and premaxillae: (a) Last dorsal ray not filamentous; pseudobranch exposed Family Elopidae (b) Last dorsal ray filamentous; pseudobranch not exposed Family Meqalopidae (ii) Mouth inferior; upper jaw bordered by premaxillae only Family Albulidae 2. Lateral line absent on body; abdominal scutes often present, pelvic scute always present Suborder CLUPEOIDEI (i) Abdominal scutes absent: (a) Body highly compressed; jaw teeth large and fang-like; dorsal origin much nearer caudal base than snout Family Chirocentridae (b) Body rounded; jaw teeth small, not fang-like; dorsal origin about mid-way between snout tip and caudal base Family EXjssumieriidae (ii) Abdominal scutes present: (a) Mouth terminal or sub-terminal, snout not pig-like; maxilla does not extend beyond posterior border of eye Family Clupeidae (b) Mouth inferior, snout pig-like; maxilla extends beyond posterior border of eye Family Enqraulidae Suborder ELOPOIDEI Family ELOPIDAE Tenpounders ELOPS Linnaeus , 1766 Elops Linnaeus, 1766, Svst.Nat . ed. 12, ^ : 518 (type: Elops saurus). A single species in East African waters. Elops machnata (Forskal) Arqentina machnata Forskal, 1775, Descript .Animal : xii, 68 (type locality: Djedda, Red Sea) . Elops machnata: Gunther, 1866, Fishes of Zanzibar : 121, fig. (caudal) ( Zanzibar) : Whitehead, 1962, Ann. Maq.nat. Hist. (13) ^ : 321 (Revision all spp. ; Indo-Pacific specimens): Losse, 1964, E.A.M.F.R.O. Ann. Rep. 1963 : 12 (Zanzibar Channel). Elops saurus: (part) Gunther, 1868, Cat .Fish. Brit .Mus . 7 : 470 ( Zanzibar, East Africa); Boulenger, 1909, Cat. Fresh Water Fish Africa . 1 : 25, fig. 17 (Zanzibar); (part) Copley, 1952, 167 J.E.Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 Game Fishes of Africa : 53, fig. 34 (Malindi, Kilifi, Mombasa); Morrow, 1954, Ann „ Mag. nat. Hist. (12) 7 ; 803 (Shimoni, Kenya). VERNACULAR NAMES: Makuronqa (Malindi) , Munyimbi (Shimoni) , Mkizi (Zanzibar) . LOCALITIES: Dar-es -Salaam , Zanzibar, Shimoni, Mombasa, Malindi, Fundishu, Lamu. RANGE: Eastern coast of Africa from the Red Sea to Algoa Bay; Seychelles, Aldabra , Comores, Madagascar, Mauritius. Widespread in Indo-Pacific to the East Indies, China, Japan and Hawaii. Family MEGALOPIDAE Tarpons MEGALOPS Lacepede, 1803 Meqalops Lacep&de, 1803, Hist .Nat. Poiss . , b : 289 (types: Meqalops f ilamentosus Lacepfede = Clupea cypri"oides Broussonet) . A single Indo-Pacific species. Meqalops cyprinoides (Broussonet) Clupea cyprinoides Broussonet, 1782, Ichthyol. (no pagination) pi. 9. Elops cyprinoides: Gunther, 1866, Fishes of Zanzibar : 122 (East Africa) ; Martens, 1869, in Deckens Reise Ost Africa. 3(1) : 143 (Pangani Riyer, ZanzibarJ"! Megalops cyprinoides: Gunther, 1868, Cat.Fish.Brit.Mus . 2 • 471 (Zanzibar); Copley. 1952, Game Fishes of Africa : 54, fig. 35, pi. 3 (upper fig.) (Mombasa, Tanga, Zanzibar, Mafia); Losse, 1964, E.A.M.F.R.O. Ann. Rep. 1963 : 12 (Zanzibar Channel). VERNACULAR NAMES: Pawale (general), Kumpanqu (Malindi). LOCALITIES: Dar-es-Salaam , Zanzibar, Mombasa, Malindi. RANGE: Eastern coast of Africa from Somalia to Algoa Bay; Seychelles, Aldabra, Comores, Madagascar, Mauritius. Widespread in Indo- Pacific, to the East Indies, Australia, Philippines and China. Family ALBULIDAE Lady Fishes ALBULA Scopoli, 1777 Albula Scopoli, 1777, Introd. Hist. Nat. : 450 (on Gronow) (type: Esox yulpes Linnaeus ) . A single species in all tropical seas. Albula yulpes (Linnaeus) Esox yulpes Linnaeus, 1758, Syst.Nat.ed.lO. 1 : 313 (On Bone Fish Catesby, 1737, Hist. Nat. Carolina . pi. 27 fig. 1; Bahamas ). 168 Check List of Elopoid and Clupeoid Fishes Butirinus qlossodontus ; Gunther, 1866, Fishes of Zanzibar : 120 (Zanzibar) . Albula conorhynchus : Gunther, 1868, Cat. Fish Brit.Mus., 7 : 468 (Zanzibar) . Albula vulpes ; Copley, 1952, Game Fishes of Africa, : 56, pi. 3 (lower fig.) (East African coast; Lamu , Malindi, Kilifi, Mombasa). VERNACULAR NAMES; Mborode (Malindi), Mnymbi (Zanzibar, Shimoni). LOCALITIES: Zanzibar, Mombasa. RANGE: Eastern coast of Africa from the Red Sea to Algoa Bay; Seychelles, Aldabra, Comores, Madagascar, Mauritius, Reunion. Cosmopolitan in tropical seas. Suborder CLUPEOIDEI Family CHIROCENTRIDAE Wolf Herrings CHIROCENTRUS Cuvier, 1817 Chirocentrus Cuvier, 1817, Reqne Animal. 2 : 178 (type: Clupea dorab Forskal). “ A single species in East African waters. Chirocentrus dorab (Forskll) Clupea dorab Forskal, 1775, Descript. Animal : xii, 72 (type locality: Dj edda) . Chirocentrus dorab; Valenciennes, 1846, Hist. Nat. Poiss. . 19 : 150, pi. 565 (Zanzibar); Gunther, 1866, Fishes of Zanzi^r : 120 (Zanzibar); Idem . 1868, Cat. Fish. Brit. Mus . . 7 : 475 (Zanzibar); Bonde, 1934, Ann. Natal MusT^ (7), ^ ; 437 (Zanzibar); Copley, 1952, Game Fishes of Africa, : 56, pi. 5 (upper fig.) (East Africa: Lamu, Mombasa, Zanzibar, Dar-es-Salaam , Mafia); Morrow, 1954, Ann. Mag. nat. Hist. (12) 2 • ^04 (Pemba); Losse, 1964, E.A.M.F.R.O. Ann. Rep. 1963 ; 12 (Zanzibar Channel). VERNACULAR NAMES: Bahanafu (Malindi), Panqa (Mombasa), Mkonqe (Zanzibar, Shimoni ) . LOCALITIES: Dar-es-Salaam, Zanzibar, Mombasa, Malindi, Lamu. RANGE: Eastern coast of Africa from the Red Sea to Natal; Seychelles, Aldabra, Comores, Madagascar, Mauritius. Widespread in Indo- Pacific, eastwards to the Philippines, China, Australia and Melanesia. 169 J. E.Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 Family DUSSUMIERIIDAE Round Herrings Key to the Genera Branchiostegal rays 14-16; adults large ( >130 mm. standard length) Branchiostegal rays 6-7; adults small ( < 80 mm. standard length) EXjssumieria Spratelloides DUSSUMIERIA Valenciennes. 1857 Dussumieria Valenciennes, 1847, Hist. Nat. Poiss. . ^ ; 467 (type; Dussumieria acuta Valenciennes ) . A single species widespread in the Indo-Pacific region. Dussumieria acuta Valenciennes Dussumieria acuta Valenciennes, 1847. Hist .Nat . Poiss . , ; 467, pi. 606 (type locality; Bombay); Whitehead, 1963, Bull. Brit. Mus.nat.Hist. (ZooU, IQ (6) ; 312, figs. 1-5 (Revision, synonymy Indo-Pacific specimensTT Losse, 1964, E.A.M.F.R.O. Ann. Rep. 1963 ; 12 (Zanzibar Channel). VERNACULAR NAMES; Daoaa la upapa (Zanzibar). LOCALITIES; Dar-es-Salaam, Zanzibar, Kenya, (a single specimen from the stomach of a sailfish). RANGE; Eastern coast of Africa from the Red Sea to Madagascar. Indo- Pacific east to Japan. SPRATELLOIDES Bleeker, 1852 Spratelloides Bleeker, 1852, Natuurk.Tiidschr.Ned.Ind. . 2 ; 214 (type; Clupea argyrotaeniata Bleeker = Clupea gracilis Schelegel) Key to the Species 1. Anal rays 11-14; a prominent silver lateral band ... S. gracilis 2. Anal rays 9-11; no silver band, whole of sides silvery S. delicatulus Spratelloides gracilis (Schlegel) Clupea gracilis Schlegel, 1846, Faun. Japan Poiss . . pts. 10-14 ; 238, pi. 108, fig. 2 (type locality; Japan) . Spratelloides iaponicus; Morrow, 1954, Ann. Mag. nat. Hist. (12) 7 ; 804 (Mkoanl Harbour, Pemba) . “ 170 Check List of Elopoid and Clupeoid Fishes Spratelloides gracilis: Whitehead, 1963, Bull « Brit ,Mus . nat ,Hist . ( Zool. ) , 10 (6) : 388, figs. 15-18 (Revision, synonymy: Red Sea and Pacific specimens). VERNACULAR NAMES: None known. LOCALITIES: Mafia, Zanzibar. RANGE: East African coast. Elsewhere: Red Sea, Indo-Pacific east to Japan and Samoa, Spratelloides delicatulus (Bennett) Clupea delicatula Bennett, 1831, Proc . zool .Soc . Lond. . 1 : 168 (type locality: Mauritius). Spratelloides delicatulus: Jatzow & Lenz, 1899, Abhandl. Senckenberq. Naturf .Ges . . xxi, _3 : 526 (no locality. East African collection); Morrow, 1954, AnnTMaq. nat. Hist. (12) ^ : 804 (Mkoani Harbour, Pemba); Whitehead, 1963, Bull.Brit.Mus’ .nat .Hist. (Zool.), 10 (6) : 345, figs, 16-17 (Revision, synonymy; Indo-Pacific specimens ) . VERNACULAR NAMES: Dagaa (general). LOCALITIES: Mafia, Zanzibar, Kilifi. RANGE: Eastern coast of Africa from the Red Sea to Zululand; Seychelles, Aldabra , Comores, Madagascar, Mauritius. Widespread in Indo-Pacific, eastwards to Australia. Family CLUPEIDAE Herrings Key to the Genera 1. Anal short, of ).ess than 30 rays; hypomaxillary bone absent: (i) Upper jaw without deep median notch: (a) Few (3-5) fronto-parietal striae, last two anal rays not enlarged, vertical striae of scales generally complete Herklotsichthvs (b) Many (more than 7) fronto-parietal striae; last two anal rays enlarged; vertical striae of scales incomplete Sardinella (ii) Upper jaw with prominent deep median notch Hilsa 2. Anal long, of more than 30 rays; a small toothed hypomaxillary bone present Pellona HERKLOTSICHTHYS Whitley, 1951. Herklotsichthys Whitley, 1951, Proc .Roy. zool. Soc .N.S ,W. 1949-50 : 67. Harenqula Valenciennes (part i.e. Indo-Pacific species only), 1847, Hist . nat . Poiss . , ^ : 301 (type: Harenqula latulus Val. = Clupea 171 J.E.Af r. Nat .Hist .Soc. Vol.XXV No. 3 (112) January 1966 macropthalma Ranzanzi, designated by Gill, 1861, Proc.Acad.nat. Sci . Philad . : 36) . Indo-Pacific species of Harenqula have been separated from New World species on the basis of a toothed hypomaxilla being present in the latter (Berry, 1963); Herklotsichthys Whitley replaces Harenqula Valenciennes for the Indo-Pacific species (Whitehead, 1963 bT"i Two species occur in the Western Indian Ocean, H. punctatus (RUpp.) and H. vittatus (Val.); although the latter has been recorded from Natal (Barnard, 1925; Smith, 1961), Mauritius (Baissac, 1951), Madagascar (Fourmanoir, 1957) and the Gulf of Aden (Whitehead, 1964), I can find no record of occurrence in East African waters. Herklotsichthys punctatus (Ruppell) Clupea punctata Ruppell, 1837, Neue Wirbelth.Fische : 78, pi. 21, fig. 2 (type locality; Red Sea) . Alosa punctata: Gunther, 1866, Fishes of Zanzibar : 23 (Zanzibar). Clupea venenosa; Gunther, 1868, Cat. Fish. Brit. Mus. . 7 : 449 (Zanzibar) Harenqula punctata; Regan, 1917, Ann. Maq.nat. Hist. (8) W : 390 (East Africa) ; Losse, 1964, E.A.M.F.R.O, Ann.Rep. 1963 : 11 (Zanzibar Channel) . VERNACULAR NAMES; Daqaa la upapa (Zanzibar), Simu (Kenya), Daqaa (small specimens - general). LOCALITIES: Dar-es-Salaam, Zanzibar, Tanga, Mombasa, Milindi, Formosa Bay. RANGE: Eastern coast of Africa from the Red Sea to Durban; Seychelles Aldabra, Madagascar, Mauritius. Widespread in Indo-Pacific, east to Japan, Australia and Polynesia. SARDINELLA Valenciennes, 1847 Sardinella Valenciennes. 1847, Hist .Nat . Poiss . . 20 : 28 (type; S.ard_inella aurita Val., designated by Gill,'°T861, Proc.Acad.nat. SH. Philad.; 35). Key to the Species 1. Abdominal scutes strongly keeled, sharp and exposed; abdomen highly compressed; 46 or more gill rakers on lower part of 1st gill arch: (i) Pelvic rays 9; more than 150 gill rakers on lower part of 1st gill arch S. lonqiceps (ii) Pelvic rays 8; less than 60 gill rakers on lower part of 1st gill arch; (a) Body depth 31.0% (generally 32% -34%) of standard length; 46-54 gill rakers on lower part of 1st gill arch S. bulan (b) Body depth 25.0% - 31.6% (generally 26% - 28%) of standard length; 47-57 gill rakers on lower part of 1st gill arch S. iussieu Check List of Elopoid and Clupeoid Fishes 2. Abdominal scutes feebly keeled, hardly exposed, abdomen smooth and rounded: 35-43 gill rakers on lower part of 1st gill arch S. sirm Sardinella lonqiceps Valenciennes Sardinella lonqiceps Valenciennes, 1847, Hist .Nat .Poiss . . 20 : (198) 273 (type locality; Pondicherry); Regan, 1917, Ann.Maq.nat.Hist. (8) ^ : 379 (Mombasa). LOCALITIES: Mombasa. RANGE: Eastern coast of Africa from the Gulf of Aden to Mombasa; Seychelles. Indo-Pacific east to Indonesia and the Philippines. Sardinella bulan (Bleeker) Clupalosa bulan Bleeker, 1849, Verb. Bat. Gen. . 22 ; 12 (type locality: Java) . Alosa kowal: Gunther, 1866, Fishes of Zanzibar ; 123 (Zanzibar). Clupea kowal; Gunther, 1868, Cat.Fish.Brit.Mus . , 7 : 450 (Zanzibar). Sardinella perforata: Regan, 1917, Ann.Maq.nat.Hist. (8) : 382 ( Indian Ocean) ; Losse, 1964, E.A.M.F.R.O. Ann. Rep. 1963 : 11 (Zanzibar Channel). Sardinella bulan; Whitehead, 1964, Bull. Brit .Mus .nat. Hist . (Zool.) 12 (T) : 250 (Zanzibar) . ^ VERNACULAR NAMES: Daqaa la upapa (Zanzibar), Daqaa (small specimens general) . LOCALITIES; Zanzibar, Pangani estuary, Tanga, Mombasa. RANGE; Eastern coast of Africa from the Gulf of Aden to Lourenco Marques; Madagascar. Indo-Pacific to the East Indies, Philippines, Siam, Amoy and Polynesia. Sardinella jussieu (Lac4pfede) Clupanodon iussieu Lacepfede, 1803, Hist .Nat . Poiss . . ^ : 469, 474, pi. 11, fig. 2 (type locality: Mauritius ) . ?Clupea tembanq; Jatzow & Lenz , 1899, Abhandl .Senckenberq .Naturf .Ges . . xxi, 3 : 526 (Zanzibar). Sardinella qibbosa; Regan, 1917, Ann.Maq.nat.Hist. (8) ^ : 383 (Mombasa, Indian Ocean) . Sardinella iussieu; Bonde, 1934, Ann. Natal Mus. (7) 3 : 437 ( Zanzibar) ; Losse, 1964, E.A.M.F.R.O. Ann. Rep. 1963 : 11 (Zanzibar Channel). VERNACULAR NAMES: Daqaa la upapa (Zanzibar), Simu (Kenya), Daqaa (small specimens - general). LOCALITIES: Dar-es-Salaam , Zanzibar, Tanga, Mombasa, Malindi. RANGE: Eastern coast of Africa from the Red Sea to Natal; Seychelles, 173 J. E. Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 Mauritius, Madagascar. Widespread in Indo-Pacif ic ; East Indies, China, Australia, Micronesia and Polynesia. Sardinella sirm (Walbaum) Cluoea sirm Walbaum, Artedi Pise., 3 ; 38 (on Forskal, 1775, Descript. Animal ; 17); Gunther, 1868, Cat. Fish. Brit. Mus. , 7 ; 425 (Zanzibar) . Alosa sirm: Gunther, 1866, Fishes of Zanzibar : 123 (Zanzibar). Sardinella sirm; Regan, 1917, Ann. Mag . nat . Hist . (8) ^ : 385 (Zanzibar, Indian Ocean); Losse, 1964, E.A.M.FTR.O. Ann. Rep. 1963 : 11 (Zanzibar Channel). VERNACULAR NAMES: Daqaa la upapa (Zanzibar), Simu (Kenya). LOCALITIES: Dar-es-Salaam , Zanzibar, Mombasa. RANGE: Eastern coast of Africa from the Red Sea to Madagascar; Seychelles, Aldabra. Widespread in Indo-Pacif ic ; East Indies, Philippines, China, Micronesia and Polynesia. HILSA Regan, 1917. Hilsa Regan, 1917 Ann. Mag. nat. Hist. (8) ^ : 303 (type: Clupea durbanensis Regan) . A single species in East African coastal waters. Hilsa kelee (Cuvier) Clupea kelee Cuvier, 1829, Reqne Animal . , ed. 2, 2 : 320 (name in footnote, based on Kelee Russell, 1803, Fishe's of Coromandel. 2 : 75, pi. 195; type locality: Vizagapatam) . Alosa chapra: Gunther, 1866, Fishes of Zanzibar : 123 (Zanzibar). Clupea ilisha: Gunther, 1868, Cat. Fish. Brit. Mus. . _7 ; 445 (Zanzibar). Hilsa kelee: Whitehead, 1964, Bull . Brit .Mus .nat. Hist. (Zool.) 12 (4) : 129, fig. 8 (Revision, synonymy; Sabaki estuary, KenyaTT VERNACULAR NAMES: Makrenqe (Mombasa), Pawali (general). LOCALITIES: Ruvu estuary, Pangani estuary, Mombasa, Malindi, Sabaki estuary. RANGE: Eastern coast of Africa from the Gulf of Aden to Natal; Madagascar. Western Indo-Pacifdc to Burma and Siam. 174 Check List of Elopoid and Clupeoid Fishes PELLONA Valenciennes, 1847. Pellona Valenciennes, 1847, Hist .Nat . Poiss . . 20 ; (218) 300 Ttype : Pellona orbygnyana Val., designateH by Gill, 1861, Proc .Acad.nat.Sci.Philad. , : 38) . A single Indo-Pacific species*. Pellona ditchela Valenciennes Pellona ditchela Valenciennes, 1847, Hist .Nat . Poiss . . 20 : (228) 314 (on Ditchelee Russell, 1803, Fishes of CoromandeTT 2 : 72 pi. 188; type locality: Vizagapatam) . Pellona ditchoa; Gunther, 1866, Fishes of Zanzibar : 122 (East Africa); Idem . 1868, Cat . Fish . Brit .Mus . , 7 : 455 (Zanzibar, East Africa) . Neosteus ditchela: Norman, 1923, Ann.Maq.nat.Hist. (9) : 17 (East Africa) . VERNACULAR NAMES: Chaa (general), Simu koko (Malindi) , Simu (general). LOCALITIES: Ruvu estuary, Pangani estuary, Zanzibar, Mombasa, Malindi, Formosa Bay. RANGE: Eastern coast of Africa from Kenya to Delagoa Bay; Madagascar. Coasts of India. * The genus Pellona , primarily composed of New World species, differs from the Indo-Pacific genus Ilisha by the possession of a toothed hypomaxilia, lying between the end of the pre-maxillary and centre of maxillary jaw bones. In Ilisha this bone is replaced by a ligament; this genus has not been recorded from our area. Family ENGRAULIDAE Anchovies Key to the Genera 1. Abdominal scutes confined to pre-pelvic region 2. Post-pelvic scutes present; (i) No pre-pectoral scutes; pseudobranch exposed (ii) Pre-pectoral scutes present; pseudobranch not exposed STOLE PHORUS Lac4p^de, 1803. Stolephorus Lac4pbde, 1803, Hist .Nat .Poiss . , 5 : 381 (type: Stolephorus commersonii LacepedeJ. Key to the Species 1. Anal origin under or behind last dorsal ray; muscular portion of isthmus short, not reaching posterior border of branchiostegal membranes: (i) Posterior tip of maxilla pointed, projecting well behind 2nd supramaxilla S. heterolobus (ii) Posterior tip of maxilla truncated, hardly projecting beyond 2nd supramaxilla . .S . buccaneeri . . Stolephorus . . . . Thrissina Thryssa 175 J. E.Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 2. Anal origin under posterior third of dorsal base; muscular portion of isthmus long, projecting forward beyond hind border of branchiostegal membranes : (i) Posterior tip of maxilla does not project beyond posterior border of pre-operculum; abdominal scutes 4-5 S. indicus (ii) Posterior tip of maxilla reaches gill opening; abdominal scutes 4-7 S. commersoni *Stolephorus heterolobus (Riippell) Enqraulis heteroloba Riippell, 1837, Neue Wirbelth.Fische : 79, pi. 21, fig. 4 (type locality: Massaua). VERNACULAR NAMES; Daoaa uronda (Zanzibar), Kumbu (Kenya). LOCALITIES; Dar-es-Salaam , Zanzibar, Mombasa, Malindi. RANGE; East African coast, Madagascar, Red Sea and the Suez Canal (Whitehead, 1964); Madras, East Indies and Australia. ♦Stolephorus buccaneer! Strasburg Stolephorus buccaneer! Strasburg, 1960, Pacific Science. ^ (4) ; 396 (type locality; Hawaii). “ VERNACULAR NAMES; None known. LOCALITIES; Mombasa (Port Tudor). RANGE; East African coast (only known from Mombasa); elsewhere; Durban, Red Sea, Persian Gulf, "Arabia" (Whitehead, 1964) and Hawaii . Stolephorus indicus (Van Hasselt) Enqraulis indicus Van Hasselt, 1823, Alq.Konst-en Letter-Bode, j,, (23) 329 (type locality; Java). ~ Enqraulis russellii; Jatzow & Lenz, 1899, Abhandl .Senckenberq Naturf. Ges . . xxi, 3 ; 525 (Zanzibar). Anchoviella indica; Morrow, 1954, Ann. Maq.nat. Hist. (12) 7 ; 804 (Mkoani harbour, Pemba). ~ Stolephorus indicus; Losse, 1964, E.A.M.F.R.O. Ann. Rep. 1963 ; 12 (Zanzibar Channel) . VERNACULAR NAMES; Daqaa uronde (Zanzibar), Kumbu (Kenya). LOCALITIES; Dar-es-Salaam, Zanzibar, Mombasa, Malindi. RANGE; Eastern coast of Africa from the Red Sea to Natal; Madagascar, Aldabra. Widespread in Indo-Pacific to the East Indies, Philippines, China, Formosa, Melanesia, Micronesia and Polynesia. ♦ Not previously recorded from East Africa. 176 Check List of Elopoid and Clupeoid Fishes Stolephorus commersoni Lacep^de Stolephorus commersoni Lac^pfede, 1803, Hist. Nat. Poiss. . 5 : 381, pi. 12, fig. 1 (no locality, on CommersonTi Enqraulis brownii: Gunther, 1866, Fishes of Zanzibar ; 123 (Zanzibar). Enqraulis commersonianus : Gunther, 1868, Cat .Fish. Brit ,Mus . . 7 : 388 ( Zanzibar) . VERNACULAR NAMES; Kumbu (Kenya). LOCALITIES: Zanzibar, Ruvu estuary, Pangani estuary, Mombasa, Malindi. RANGE: East African coast south to Durban; Madagascar, Mauritius. Widespread in Indo-Pacif ic ; Arabia, East Indies, Philippines, China, Formosa, Korea and Polynesia. THRISSINA Jordan & Seal, 1925 Thrissina Jordan & Seal, 1925, Copeia No. 141 : 30 ( type: Clupea baelama ForskSl ) . A monotypic , Indo-Pacific genus. Thrissina baelama (Forskal) Clupea baelama Forskal, Descript. Animal, : 72 ( type locality: Dj edda) . Enqraulis boelama (mis-spelt); Gunther, 1866, Fishes of Zanzibar : 123 (Zanzibar) ; Idem. 1868, Cat .Fish. Brit .Mus . . 2 • 393 (Zanzibar); Idem. 1871, Proc. Zoo l.Soc. London : 671 (Zanzibar). VERNACULAR NAMES: Daqaa (Zanzibar), Simu (Kenya. Localities; Dar-es-Salaam , Zanzibar, Mombasa, Malindi. RANGE: Eastern coast of Africa from the Red Sea to Madagascar; Seychelles, Mauritius, Reunion. Widespread in Indo-Pacific, to the Philippines, Melanesia, Micronesia and Polynesia. THRYSSA Cuvier, 1829 Thrvssa Cuvier, 1829, R^qne Animal, ed. 2, 2 ; 323. Ttype: Clupea setirostris BroussoneT) . Key to the Species 1. Maxilla does not extend beyond tip of pectoral T. vitrirostris 2. Maxilla extends beyond tip of pelvic fins T. setirostris *Thryssa vitrirostris (Gilchrist & Thompson) Enqraulis vitrirostris Gilchrist & Thompson, 1908-11, Ann ,S .Afr.Mus . ^ : 201 (localities: Natal; inner harbour, Durban) . * Not previously recorded from East Africa. 177 J.Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 VERNACULAR NAMES: None known. „ n- ,■ LOCALITIES: Zanzibar, Ruvu estuary, Pangani estuary, Mombasa, Malindi, Formosa Bay. „ , RANGE: East African coast south to East London; Madagascar, Aldabra. Persian Gulf of Oman (Whitehead, 1964), coast of India. ♦Thrvssa setirostris (Broussonet) Clupea setirostris Broussonet, 1782, Ichth. , pi. 2 (type locality: Society Islands) . VERNACULAR NAMES: None known. LOCALITIES: Ruvu estuary, Pangani estuary, Malindi, Formosa Bay. RANGE: Eastern coast of Africa from the Red Sea to Natal; Madagascar, Aldabra. Widespread in Indo-Pacif ic , to China, Queensland and Polynesia. * Not previously recorded from East Africa Acknowledgements The author wishes to express his thanks to the Director and Staff of the East African Marine Fisheries Research Organization, Zanzibar, for critically reading the manuscript. References BAISSAC, J. DE B. , 1951, Contributions a I'etude des poissons de I'lle Maurice, iv, Proc .Roy. Soc. Arts & Sci. Mauritius . 1 (2) : 123-152. BARNARD, K.H. , 1925. A monograph of the marine fishes of S, Africa. Ann.S .Af r.Mus . . (2) ; 1-814. BERRY, F.H., 1963. A hypomaxillarv bone in the herring genus Harenqula (Pisces : Clupeidae). Pacific Science (in press). FOURMANOIR, P. , 1957. Poissons Teleostdens des aux Malagaches du Canal de Mozambique. Mems. Inst .Sci. Madagascar. Ser. F. , 1 : 1-317. MORROW, J.E., 1954. Fishes from East Africa, with new records and description of two new species. Ann.Maq.nat.Hist. ( 12) 7 : 797-820. SMITH, J.L.B., 1961. The Sea Fishes of Southern Africa (4th edition) Central News Agency Ltd. , South Africa. WHITEHEAD, P.J.P. 1963 a. A contribution to the classification of Clupeoid fishes. Ann.Maq.nat.Hist. (13) b : 737-750. 1963 b. Herklotsichthvs Whitley 1951 to replace Harenqula Valenciennes 1847 for Indo-Pacific Species (Pisces : Clupeidae). Ann.Maq.nat.Hist. (13) 6 : 273-284. 1964. A review of the Elopoid and Clupeoid fishes of the Red Sea and Adjacent regions. Bull. Brit .Mus .nat. Hist . ( Zool. ) . 12 (7) : 227-281 (in press). “ (Received for publication 1st July 1965) 178 • .'■.:• V' ,:• ■ '.'.vC'ffftO ■ _ ■ .^5 (••+ rfW:’ n* lllU .. .',, Ur- ; ■ ■■■"■'' r. . ( r'g)(fj» - « iBpawigto jtoX *^4Ux~: . „.«kiuj«w4,» 4 -.ii.ii'nis^- -tx 'i f^r i j4. . , sJi-' .;■ ^4'*'' 1^4 I'ilBiiM. 9 i WjB' iZij^ i' i'J>/ -\ ■^^^. -"' v ,» . r. r*wr:iei3 .i^t ''ifcji-»l[_ HA, ?4&2 ’^lifl •, Js-^UtttfM S' 352 X. , ■■ " *■' -■ h:^a/'^:^ .,v'i-»=t^-l ^^mim t:j4' *rt»? -Tilf. ; ■/ ■ --r'--v«U'i ‘ Xe7l' ' x,:skW^Mi^^^ i '■*¥?• '■ «*3«V '■■■ * ■■) fi&J: 3^111^ S-. aiiT ...i.d^?l .-■'.ft. ■fVBrt'l -i. A .■/« .CA^. tyiy ^ c ffl ;Aigia'.ftiA. : :,4w^'?t»' f ^ li? • 3 l.?V!. Y^irim . ),. ^ jU ;. J4ii. .-: A) ,i:^v J. E. Af r. Nat .Hist. Soc Vol.XXV No. 3 (112) January 1966 A LIST OF BIRDS SEEN IN SOUTHERN TANZANIA By C.J. TWEEDY Introduction Scope of the List The attached list of birds seen in southern Tanzania between July 30th 1963 and July 18th 1964 gives records of 105 species exactly identified, together with a further general note on 7 families of which representatives were seen but not specified with absolute certainty. Of these 105 species, 71 were seen within the school grounds of St. Joseph's College, Chidya, near Masasi, in the Mtwara region, an area of perhaps three quarters of a square mile in extent; and a further 13 species were seen within five miles of the school boundary, half of them within a mile of it. Thus only 21 species represent what I assume to be the large number of southern Tanzanian - birds - especially waders and waterfowl - which seldom or never come near Chidya. At the same time, even with regard to the birds on the list, it must be borne in mind that they are the result of observations made largely in spare time over less than one calendar year, during which I was absent from southern Tanzania from December 15th to January 10th; and that even within the limited time available many birds were seen which I could not certainly identify. It will thus be seen that the negative evidence of this list is only to a small extent valuable, and that it is basically little more than a preliminary check-list of the birds of St. Joseph's College, Chidya. Chidya Ecologically Chidya is a settlement consisting of the buildings directly connected with a junior secondary boarding school for about 300 boys, and with the small local medical dispensary. Situated some twenty miles east-north-east of Masasi and three due west of the main Makonde Plateau escarpment, it is on a low ridge about five miles long, at a height of about 2,000 ft above sea-level. It is some 500 ft at its highest above the large plain stretching to the Ruvuma River fifty miles to the south, and a hundred or two miles towards the southern highlands in the far west, in the direction of Lake Nyasa. Geologically it is an outrider to the Makonde Plateau. The buildings at Chidya are scattered about an area to various degrees cleared of the surrounding Brachysteqia woodland. There are many mango trees, two open playing-fields and cultivated lands. During the year concerned, the remaining old-style African school- buildings were being steadily replaced by or transformed into modern buildings with corrugated iron roofs, but a number of staff houses were being retained unaltered with thatched roofs. There is permanent water in two wells about a couple of hundred feet below the school, some of which is piped up as a constant supply. (Measurements of heights and distances in this Introduction are all approximate) . Exceptional Weather It will be noticed that a number of passerine species usually 179 Birds seen in Southern Tanzania typical of Chidya were reduced in numbers or absent for several weeks during the rainy season. It is possible that this was connected with an unusually long dry spell between January 22nd and February 22nd, with not even a trace of rain from January 28th to February 6th. Details of Presentation Species in the list are numbered as in Mackworth Praed and Grant, 'Birds of Eastern and North Eastern Africa', Second Edition, 1957. Subspecies are not normally mentioned; no unusual ones were identified. The phrase 'generally common' means that one was never surprised to see the bird throughout the year in any typical habitat. 'Chidya' in the list means 'within the school boundary'. 'Near Chidya' means within a mile of this, sometimes much less. Msati is a very small settlement immediately outside the school boundary to the south-east. 'Mwiti bridge' refers to the small stretch of river-valley visible from the concrete bridge over the river just below Mwiti village, about 5 miles south-east of Chidya on the Chiwata-Ma j embe road. Nangoo and Nyangao are on the Masasi-Lindi road. Records ascribed to 'Rondo Plateau' were all seen from the road between Nyangao and St. Cyprian's Theological College, Ngala. Just one more Bird So many people connected with the school remarked that they had in previous years, apparently usually in December, seen the presumably unmistakable Pennant -wing Nightjar, Semelophorus vexillarius Gould, that I do not doubt that this also is a genuine Chidya bird. The Birds. ANHINGIDAE - Darters 28. Darter, Anhinqa rufa (Lac^pede & Daudin) One, Masasi dam. Sept. 14th. ARDEIDAE - Herons and Egrets 33. Grey Heron. Ardea cinerea Linnaeus One, Ruvuma River near Luatala, Aug. 11th. 38. Yellow-billed Egret, Mesophovx intermedius (Wagler) Small parties seen from Lindi-Masasi road, July 30th; Masasi dam. May 7th, SCOPIDAE - Hammerkops 53. Hammerkop, Scopus umbretta (Gmelin) One occasionally, Chidya, October and November. FALCONIDAE - Birds of Prey 129. Pygmy Falcon, Poliohierax semitorquatus (A. Smith) One near Chidya, Oct. 11th. 132. Kite, Milvus miqrans (Boddaert) Common generally, but not noticed at Chidya between end of October and beginning of April. Single bird feeding in flight at c. 25 ft., off small bird held in talons, Oct. 26th. 154. Black-chested Harrier-Eagle, Circaetus pectoralis Smith One in flight, Chidya, Apr. 21st. 159. Bateleur, Terathopius ecaudatus (Daudin) Fairly common in ones and twos, but no records between Sept. 2nd and March 7th. 180 Birds seen in Southern Tanzania 176. Shikra, Accipiter badius (Gmelin) Common in February and March at Chidya; Masasi, May 7th. Bird in the hand, Feb. 26th., was clearly A.b. polvzonoides Smith. 177. African Goshawk, Accipiter tachiro (Daudin) Common from early August to late October at Chidya. 185. Harrier-Hawk, Polyboroides typus Smith One, Masasi, Oct. 12th; one, Chidya, April 21st. PHASIANIDAE - Game Birds. 189-207 and 211-213: No species of Francolin or Quail were certainly identified. One or two small birds, thought to be Coqui Francolin, Francolinus coqui (Smith) or a Quail, not all of the same species, were frequently flushed from long grass on the Chidya hillside on dates between September and Noyember 1963. A large species, thought to be the Scaly Francolin, Francolinus squamatus (Cassin) was once seen and frequently heard on Mtandi hill, Masasi, in May 1964. 208. Red-necked Spurfowl, Pternistis cranchii (Leach) Shot 9 in the hand, Chidya, Sept. 30th; three on road near Mwiti, Dec. 11th. CHARADRIIDAE - Stilts. 296. Black-winged Stilt, Himantopus himantopus (Linnaeus) One, Machole Salt Farm, near Lindi, July 30th. SCOLOPACIDAE - Waders. 313. Green Sandpiper, Trinqa ocrophus Linnaeus One, Mwiti bridge, Dec. 11th. COLUMBIDAE - Pigeons and Doyes. 386. Red-eyed Doye, Streptopelia semitorquata (Riippell) Generally common. 388. Ring-necked Doye, Streptopelia capicola (Sundeyall) Generally common. 392. Laughing Doye, Stiqmatopelia seneqalensis (Linnaeus) Common along main roads. 397. Emerald-spotted Wood-Doye, Turtur chalcospilos (Wagler) Generally common. 401. Green Pigeon, Treron australis (Linnaeus) One near Chidya, May 16th. CUCULIDAE - Cuckoos and Coucals. 407. Black Cuckoo, Cuculus cafer Lichtenstein A yery yocal rainy season yisitor, early December till at least the middle of March. 423. White-browed Coucal, Centropus superciliosus . Hemprich and Ehrenberg. Common; ubiquitous. MUSOPHAGIDAE - Turacos. 425-437; Turacos were certainly seen on three occasions, though neyer specified with certainty. One at Chikundi near Ndanda on Noy. 1st and one at Chidya on Jan. 16th were thought to be Liyingstone ' s or possibly Reichenow’s Turaco , Tauraco liyinqstonii Gray, or Tauraco reichenowi (Fischer). CORACIIDAE - Rollers. 460. Lilac-breasted Roller, Coracias caudata Linnaeus. Lindi, May 23rd; Mtwara, July 18th. 463. Broad-billed Roller, Eurystomus qlaucurus (Muller). Several at and near Chidya, Nov. 3rd to Jan. 23rd; a rainy season visitor. 181 Birds seen in Southern Tanzania ALCEDINIDAE - Kingfishers 471. Pygmy Kingfisher, Ispidina picta (Boddaert) Single birds at Chidya, Oct. 23rd, Nov. 10th, Nov. 30th, Dec. 10th. 476. Brown-hooded Kingfisher, Halcyon albiventris (Scopoli) One, forest near Nyangao , Nov. 1st. 479. Striped Kingfisher, Halcyon chelicuti (Stanley) One, Makonde Plateau, July 4th. MEROPIDAE - Bee-eaters 488. Little Bee-eater, Melittophaqus pusillus (Muller) Three, once at Chidya, Aug. 29th; Masasi, May 8th; Lindi, May 23rd. 493. White-fronted Bee-eater, Melittophaqus bullockoides (Smith) One, Ruvuma River near Luatala, Aug. 11th; one, Nangoo , May 23rd. 496. Swallow-tailed Bee-eater, Dicrocercus hirundineus (Lichtenstein) One, Rondo Plateau, Nov. 1st. BUCEROTIDAE - Hornbills 497. Trumpeter Hornbill, Bycanistes bucinator (Temminck) One, Rondo Plateau, Nov. 1st. 509. Crowned Hornbill, Tockus alboterminatus ( Buttikorf er) . Generally fairly common, but not seen at Chidya nearer than Mwiti bridge, Dec. 11th. 515. Ground Hornbill, Bucorvus leadbeateri (Vigors). One, Chidya, Dec. 1st; two, Chidya, Feb. 2nd. PHOENICULIDAE - Wood-Hoopoes Green Wood-hoopoe, Phoeniculus purpureus (Miller) Generally common; always in small parties of up to eight. Party throwing down chunks of loose bark 1 ft. or more long, Chidya, Sept. 3rd. Party taking flies in Flycatcher style, Chidya, March 25th. STRIGIDAE - Owls. 528 Barn Owl, Tvto alba (Scopoli). Chidya, Nov. 16th. 533. African Wood-Owl, Ciccaba woodfordii (Smith). Chidya, Dec. 7th and May 31st. 543. Spotted Eagle-Owl, Bubo africanus (Temminck). Chidya, April 8th and 13th. 544. Verreaux's Eagle-Owl, Bubo lacteus (Temminck), Chidya, June 11th; one openly perched in Flamboyant, Delonix reqia discovered 3p.m. and undisturbed by extensive observation, though awake. CAPRIMULGIDAE - Nightjars. 548 or 551. A nightjar thought to be either the Dusky, Caprimulqus pectoralis Cuvier or the Fiery-necked, Caprimulqus fervidus Sharpe was seen and observed closely at Chidya on September 24th and a mile or so away on June 26th. The bird, certainly larger than the Gaboon Nightjar, Caprimulqus fossil Hartlaub and generally of a warm brown, was not certainly distinguishable on plumage-details taken, but might be on behaviour (June 26th); Expert at dodging in and out among trees, a few feet above the ground. Continually does this (for a few yards only, when gently pursued) with only a slight whirr on take-off, and there- after silent as it cunningly manoeuvres, changing direction before landing. Ground on this occasion either open with dead leaves, or with 182 J.E.Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 long grass, or dead leaves, twigs etc., under fairly light trees. Almost invisible among dead leaves when facing openly at c. 15 ft. 560. Gaboon Nightjar, Caprimulqus fossii Hartlaub. Single birds seen close and frequently at and near Chidya, especially on open grass path near buildings, and on sandy tracks, from Feb. 25th onwards. Two hawking flies, Chidya, June 8th. COLIIDAE - Mousebirds. 566. Speckled Mousebird, Colius striatus Gmelin. Two, Chidya, Oct. 9th only. CAPITONIDAE - Barbets. 597. Golden-rumped Tinker-bird, Poqoniulus bilineatus (Sundevall) One, Chidya, Sept. 23rd and Oct. 14th; often heard at this time of year. PICIDAE - Woodpeckers. 623. Cardinal Woodpecker, Dendropicos fuscescens (Vieillot) . One near Chidya, Feb. 25th; Lindi , May 23rd. 629. Bearded Woodpecker, Thripias namaquus (Lichtenstein). Two near Chidya, March 15th; pair mating, Chiwata, July 4th. APODIDAE - Swifts. 644 White-rumped Swift, Apus caffer (Lichtenstein). Common at Chidya all the year round, though less in evidence in January and February. Frequently in and out of African buildings from August onwards, often with nest on verandahs. Easterly movement of 100+ swifts overhead at Chidya on April 9th appeared to be of this species. ALAUDIDAE - Larks. 654 -- 689: No Larks were specified exactly, but the Family Alaudidae were represented by a tree-perching Lark with conspicuous white stripe over and behind eye, at Chidya, seen at least twice (Oct. 26th and Dec. 8th); and by two pairs in grass at Lindi airport and one pair at Mtwara airport (both July 18th) of a larger species - a plump, long- legged upstanding creamy-eyebrowed Lark of c. 6 ins., perhaps the Rufous-naped Lark, Mirafra africana A. Smith. MOTACILLIDAE - Wagtails. 691. African Pied Wagtail, Motacilla aquimp Dumont Often seen and heard at Chidya in August; full song, Aug. 31st. MOTACILLIDAE - Pipits. 702 - 715, No Pipits were specified with certainty, but they were represented by an upstanding Pipit of longish and stoutish beak standing conspicuously on a rooftop in Masasi and thought to be the Long-billed Pipit, Anthus similis Jerdon. 716. Yellow-throated Long-claw, Macronyx croceus (Vieillot) Lindi airport grass, Dec. 28th. PYCNONOTIDAE - Bulbuls. 741. Black-capped Bulbul, Pycnonotus xanthopyqos (Hemprich and Ehrenberg) , Ubiquitous and abundant, most of the year, though less in evidence during most of February and perhaps before this. 183 Birds seen in Southern Tanzania MUSCIDAPIDAE - Flycatchers. 778, Spotted Flycatcher, Muscicapa striata (Pallas) One, Chidya, Dec. 9th, and March 25th; occasional. 799, South African Black Flycatcher, Melaenornis pammelaina (Stanley) One, Chidya, Nov. 10th only. 809. Livingstone's Flycatcher, Erythrocercus livinqstonei Gray Chidya, one. Sept. 26th (seen clearly to be E. 1. thomsoni Shelley.); two, March 14th. 815. Puff-back Flycatcher, Batis capensis Linnaeus Masasi, May 5th. 817. Chin-spot Puff-back Flycatcher, Batis molitor (Hahn & Kiister) Chidya, June 9th. Batis Flycatchers fairly common round Chidya and Masasi; exact identifications of 815 and 817 refer only to two individuals. 823. Black-throated Wattle-eye, Platysteira peltata Sundevall One male, Chidya, Sept. 10th only. 834. Grey-headed Paradise Flycatcher, Tchitrea plumbeiceps Reichenow One, Chidya, Dec. 10th, frequently calling. TURDIDAE - Thrushes, Chats, etc. 871. Red-tailed Chat, Cercomela familiaris (Stephens) Masasi, one May 5th; at least four. May 8th. 876. Cliff-chat, Thamnolaea cinnamomeiventris (Lafresnaye) Masasi; pair, Dec. 14th; several, both sexes, May 2nd, and 5th. 910. Red-backed Scrub-Robin, Erythropyqia zambesiana Sharpe Pair at Chidya, Dec. 7th, male in full song (and sometimes ventriloquial) , seen displaying tail by waving slowly up and down, fully fanned. SYLVIIDAE - Warblers. 979. Black-breasted Apalis , Apalis flavida (Strickland) One, Chidya, Oct. 24th;, two, Nov. 4th. 1009. Green-backed Camaroptera, Camaroptera brachyura (Vieillot) One, Chidya, Dec. 7th and 10th. SYLVIIDAE - Cisticolas or Grass Warblers. 1016 - 1044; Three types of Cisticola were distinguished at Chidya, though absolute certainty of species seemed impossible in the field. Call in addition to plumage features favoured Zitting Cisticola, Cisticola juncidis (Rafinesque) on Dec. 5th and Croaking Cisticola, Cisticola natelensis (Smith) on March 8th, May 11th and June 24th, as probable. SYLVIIDAE - Prinias or Long-tails. 1045. Tawny-f lanked Prinia, Prinia subflava (Cmelin) Fairly common throughout the year. HIRUNDINIDAE - Swallows. 1061. Wire-tailed Swallow, Hirundo smithii Leach Common about houses at Chidya throughout the year, not decreasing during some of rains as White-rumped Swift and Striped Swallow. A pair built a nest at 12-15 ft. against concrete wall just below a corrugated iron roof above my door. This fell down twice, and was twice rebuilt, once in March. No young were at any time seen or heard; but by April 4th the two had been seen nightly on the nest for several weeks, and were recorded as still present on May 12th and June 14th (never recorded as missing between those dates.) On April 4th and June 14th I was 'attacked' by the presumed male sweeping very 184 J.E.Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 fast and close towards my face, and only veering away just in time. The usual penetrating 'tweet' shrilly accompanied these attacks. Small broken egg-shell below nest, June 14th. 1063. Mosque Swallow, Hirundo seneqalensis Linnaeus Mtwara airport, Dec. 14th; small parties overhead at Chidya on various dates in April and May. 1065. Striped Swallow, Hirundo abyssinica Guerin Abundant and vocal at Chidya often round houses in Aug., Sept., Oct.; few only Jan. and Feb.; none in March; occasional in April; common and vocal again by May 17th; abundant by June 24th. Breeding: family of four inside African house, Aug. 1st; beaks of two crammed with mud, Oct. 6th. 1077. Eastern Rough-wing Swallow, Psalidoprocne orientalis Reichenow Quite common but rather irregular at and near Chidya. Written records only in August, May and June. CAMPEPHAGIDAE - Cuckoo-Shrikes. 1081. Black Cuckoo-Shrike, Campephaga sulphurate (Lichtenstein) Occasional at Chidya: one male, Aug. 31st, Nov. 17th, Nov. 24th; 2 females. Sept. 16th; pair Nov. 23rd. On these occasions they were not at all 'shy and inconspicuous', as described by Praed and Grant, II, p. 557. DICRURIDAE - Drongos. 1088. Drongo , Dicrurus adsimilis (Bechstein) Single birds at and near Chidya fairly common, probably throughout the year. PRIONOPIDAE - Helmet-Shrikes. 1096. Chestnut-fronted Shrike, Siqmodus scopifrons Peters Party of six, Chidya, August 6th only. LANIIDAE - Shrikes. 1112. Red-backed Shrike, Lanius collurio Linnaeus Single male near Chidya, April 8th. 1125. Tropical Boubou, Laniarius aethiopicus (Gmelin) Generally common, but the bell-call of this and 1144, some of the most typical bird-sounds of Chidya, were hardly heard at all in January and February. Striking duet, Aug. 31st. 1128. Black-headed Puff-back, Dryoscopus cubla (Shaw) Very common at Chidya: continually heard between August and April. Rump 'puffed', Nov. 8th only. 1133. Black-headed Bush-Shrike, Tchaqra seneqala (Linnaeus) Generally common; either this bird or li34 often seen from main roads. 1134. Brown-headed Bush-Shrike, Tchaqra australis (Smith) Generally common; see 1133. Full song first heard in late November. 1138. Sulphur-breasted Bush-Shrike, Chiorophoneus sulf ureopectus ( Lesson) . Much less in evidence at Chidya than all other shrikes except 1112, but I think present in small numbers throughout the year. 1144. Grey-headed Bush-Shrike, Malaconotus blanchoti Stephens Very common at Chidya, though hardly seen or heard at all in Jan,, Feb. and March. Its bell-call (seen being made) very characteristic of Chidya, though Praed and Grant seem only to refer this to 1125. Striking duet witnessed, Oct. 20th. Multiple chase with great variety of calls, Oct, 26th. 1148. Nicator, Nicator chloris (Valenciennes) Extremely elusive but easily audible and I think common 185 Birds seen in Southern Tanzania throughout the year, Masasi once, Feb. 29th; intensive observation at Chidya from March to May. Always single bird, except for two calling to each other. May 17th. Great variety of calls. Has habit of keeping ’one tree ahead' of walker, very audible close at hand, but usually invisable. ORIOLIDAE - Orioles. 1164. Golden Oriole, Oriolus oriolus (Linnaeus) One male, Chidya, Dec. 7th; pair, Chidya March 8th and 14th. 1165. African Golden Oriole, Oriolus auratus Vieillot Occasional in dry season; two, Chidya, Sept. 3rd (full song); one male May 16th. 1167. Black-headed Oriole, Oriolus larvatus Lichtenstein Common and more visible at Chidya than 1164 or 1165; one Lindi, May 23rd. CORVIDAE - Ravens and Crows. 1172. Pied Cow, Corvus albus Muller Present throughout the year at Chidya in very varied numbers; almost absent in October and much of November, when they were replaced by Ravens as below. Numerous and noisy with bill-chattering and calling at Lulindi and Newala, Apr. 25th- 27th. Two drinking blood at Masasi slaughter house. May 2nd. 1175. White-necked Raven, Corvultur albicollis (Latham) Occasional throughout the year at Chidya, but much more numerous in October and November, occasionally in flocks of from ten to twenty. STURNIDAE - Starlings. 1184. Violet-backed Starling, Cinnyricinclus leucogaster (Boddaert) Quite common at Chidya in September and October; not seen at any other time. Not strongly gregarious; usually in ones and twos; seven, Oct. 29th. 1203. Red-winged Starling, Onychoqnathus morio (Linnaeus) Common round Masasi; only one Chidya record, Aug. 4th. ZOSTEROPIDAE - White-eyes. 1219. Yellow White-eye, Zosterops seneqalensis Bonaparte Quite common in small flocks for most of the year. Possibly more than one species present. NECTARINIIDAE - Sunbirds. 1241. Little Purple-banded Sunbird, Cinnyris bifasciatus (Shaw) Common at Chidya from end of September to end of November; apparently absent at other times. One Lindi, May 23rd. 1249. White-bellied Sunbird, Cinnyris talatala (Smith) Masasi, Sept. 14th; May 1st and 7th. 1261. Amethyst Sunbird, Chalcomitra amethystina (Shaw) Two in the hand at Chidya, March 7th. (juvenile male) and 17th (male). 1263. Scarlet-chested Sunbird, Chalcomitra seneqalensis (Linnaeus) Common, Masasi; not seen at Chidya save for juvenile in hand, March 7th. 1271. Collared Sunbird, Anthreptes collaris (Vieillot) Masasi, May 5th and 7th. PASSERIDAE and PLOCEIDAE - Sparrows, Weavers, Waxbills, etc. 1300. Grey-headed Sparrow, Passer qriseus (Vieillot) Generally common. 186 J.E.Afr.Nat.Hist.Soc. Vol.XXV No. 3 (112) January 1966 1312. Black-headed Weaver, Ploceus cucullatus (Muller) Breeding colony, Masasi, Feb. 22nd. No weavers ever seen at Chidya. 1365. Black-winged Red Bishop, Euplectes hordeacea (Linnaeus) One, Nangoo, May 23rd. 1367. Yellow Bishop, Euplectes capensis (Linnaeus) One seen from Masasi-Newala road, March 19th. 1379. Bronze Mannikin, Spermestes cucullatus Swainson Abundant in dry season, at Chidya; absent for at least the end of January and most of February. 1382. Magpie Mannikin, Amauresthes f rinqilloides (Lafreshaye) One, Masasi, May 5th. 1406. Peters' Twin-spot, Hvparqos niveoquttatus (Peters) Pair at Chidya, Oct. 27th only. 1410. Green-winged Pytilia, Pytilia melba (Linnaeus) Single birds at and near Chidya, quite often throughout the year. 1411. African Fire-finch, Laqonosticta rubricate (Lichtenstein) Generally common throughout the year. 1418. Waxbill, Estrilda astrild (Linnaeus) Occasional, Chidya, Feb. -April. 1430. Cordon-bleu, Uraeqinthus anqolensis (Linnaeus) Generally common, but not recorded at Chidya between Nov. 24th and Apr. 9th. 1434 - 1440: Hvopchera spp. Indigo-birds, possibly of more than one species, were often seen on ground and on shrubs among houses at Lulindi on April 25th and 26th. Birds estimated at c. 4'^" , wholly dull black in plumage except for dull dark brown, only once seen clearly on wings and never on tail; no wash of either blue, purple or green distinguished on the general black; tail short; relatively large finch-type beak usually white, once pinker; legs apparently brown. 1441. Pin-tailed Whydah, Vidua macroura (Pallas) Near Nangoo, Nov. 1st. Flock, both sexes at Kilwa airport, Dec. 14th; quite common in March at Chidya, but not other- wise recorded there; Lulindi, Apr. 25th. 1444. Paradise Whydah, Steqanura paradisaea (Linnaeus) Only once near Chidya, June 28th, immediately after seeing 1445 a few yards away. Several individuals seen from Masasi-Newala road, March 19th and Apr. 24th. 1445. Broad-tailed Paradise Whydah, Steqanura orientalis (Heuglin) Single males often seen near Chidya, usually at Msati, bet- ween May 17th and July 11th. Call recorded as a rattle, similar to but gentler than that of Mistle-Thrush ( Turdus viscivorus ) ; remarkable display with tail on three levels , July 11th. See J. E. Afr .Nat. Hist .Soc. Vol. XXV, No. 2, p.l08, June 1965. FRINGILLIDAE - Finches. 1148. Yellow-fronted Canary, Serinus mozambicus (Muller) Small flocks regularly in Casuarina trees at Chidya, Sept., Oct. and Nov. 1456. Streaky-headed Seed-eater, Serinus qularis (Smith) Not identified certainly till July 4th near Chiwata, but suspected to be generally common. 187 Birds seen in Southern Tanzania EMBERIZIDAE - Buntings. 1469. Golden-breasted Bunting, Emberiza flaviventris Stephen Pair near Chidya, March 4th; one June 1st; probably not uncommon. 1476. Cinnamon-breasted Rock-Bunting, Frinqillaria tahapisi A, Smith. One at Chidya, June 23rd, only. (Received for publication 8th August 1965) 188 J. E.Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 KEYS TO THE GENERA OF INSECTIVORA, CHIROPTERA AND RODENTIA OF EAST AFRICA By J.B. FOSTER, Department of Zoology, University College, Nairobi and A, DUFF-MACKAY, National Museum, Nairobi. It is the purpose of these keys to help newcomers to the field to become more rapidly acquainted with the small mammals of East Africa; a task which is usually found difficult due to the lack of comprehen- sive literature dealing with identification. The keys extend only to generic level since a taxonomic revision is needed in most groups. Having determined the genus more specialised works, of which the fol- lowing will be found most useful, can be referred to for the species; for all orders; Allen (l) Ansell (2) Ellerman (9) Hollister (11) Moreau (12) Swynnerton & Hayman (14 and 15), for Chiroptera ; Harrison (10), and for Petrodomus (elephant shrews), Corbet ( 6) . References to original descriptions and old revisions may be found in the above. Walker (16) has been used as the authority for some of the more problematical genera, while Romer (13) has been used at the higher taxonomic level (e.g. for the affinities of Anomalurus and Pedetes ) . To use the keys it is generally necessary to have both the skin and skull for Chiroptera . and although the Insectivora and Rodentia can be keyed on the skull alone, skin characters are included. None of the keys can be used for the skin alone. This work has been carried out at the National Museum using mainly the National Museum collection of small mammals. Key to the Insectivores . 13 genera Zygomatic arch present, (golden moles, hedgehogs, and elephant shrews) 1 Zygomatic arch absent, (shrews and otter shrews) 6 1. Cheek teeth zalambdodont (cusps V-shaped); body mole-like, with a smooth leathery pad on the nose, and very large claws on the front feet; rudimentary eyes covered with skin; no externally visible ears or tail, (family Chrysochloridae) Amblysomus ( = Chlorotalpa ) Cheek teeth dilambdodont (cusps W-shaped; body not mole-like 2 189 Insectivores 2. Palate extends well beyond the tooth row; body heavily armed with spines; nose not long and mobile, (family Erinaceidae) Erinaceus Palate does not extend beyond the tooth row; body not armed with spines; nose long and mobile, projecting far beyond the nasal bones, (family Macroscelididae) 3 3. *Seven or eight teeth in the upper jaw; no large palatal foramina; skull more than 57 mm long. Pelage coarse, black and red or checkered Rhynchocvon Ten teeth in the upper jaw; large palatal foramina present; skull less than 57 mm long. Pelage softj light grey-brown 4 4. A pair of large palatal foramina lie between the molariform teeth. Ears shorter than 30 mm; five toes on the hind foot 5 No large palatal foramina between the molariform teeth. Ears longer than 30 mm; four toes on the hind foot Petrodomus 5. Ten teeth in the lower jaw. Chest gland present on the skin Elephantulus Eleven teeth in the lower jaw. Chest gland present (N. fuscipes) or absent (N. brachyrhynchus ) Nasilio dilambdodont) . Nose not long and mobile; body adapted for swimming, with a long and, at least to some extent, laterally compressed tail, ( family Potamoqalidae ) 7 Not more than nine upper and seven lower teeth (dilambdodont). Nose long and mobile, projecting far beyond the nasal bones; body not adapted for swimming, (family Soricidae).. 8 7. Teeth zalambdodont . Tail much compressed laterally; nose with a horny or leathery rhinarium; hind feet not webbed Potamoqale Teeth dilambdodont. Tail slightly compressed laterally; nose with a fleshy rhinarium; hind feet webbed Micro potamoqale 8. Nine teeth in the upper jaw (four upper unicuspids) 9 Eight teeth in the upper jaw (three upper unicuspids) 11 9. Seven teeth in the lower jaw; braincase strongly angled in the squamosal region. No bristle hairs on the tail Myosorex Six teeth in the lower jaw; braincase not strongly angled laterally. With or without bristle hairs on the tail 10 + In very old individuals the crowns of two rooted teeth may be so worn as to be completely divided, thus appearing as two separate teeth. 190 J.E .Afr. Nat .Hist .Soc . Vol.XXV No. 3 (112) January 1966 10. Condylo - basal length of the skull well over 20 mm. Length of the head and body 120 to 150 mm; possesses a remarkably strong, thick, vertebral column; no bristle hairs on the tail .............................................. Scutisorex Condylo - basal length less than 20 mm. Head and body length well under 100 mm; vertebrae not specialized as above; tail with bristle bristle hairs on the tail................ Suncus 11. Six or seven teeth in the lower jaw; braincase strongly angled laterally in the squamosal region. Body rather mole-like; with long claws, particularly on the front feet; ear not visible above the fur; tail short, about twice the length of the hind foot, without bristle hairs Surdisorex Six teeth in the lower jaw; braincase not strongly angled laterally. Habit shrew-like; tail more than three times the length of the hind foot, with bristle hairs Crocidura Key to the Chiroptera 29 genera Special attention should be paid to counting teeth in bats; some teeth are minute and must be looked for with great care. External features readily characterize the families of bats, and these are illustrated in Fig. 1. Crowns of the molars smooth, with a longitudinal groove. The first and second fingers with a family Pteropodidae) 1 Crowns of the molars not smooth. Only the first finger with a claw; eyes small or minute, (suborder Microchiroptera) 6 1. *Cheek teeth 3/5. White tufts of fur by the ears 2 Cheek teeth 5/6. No white tufts of fur by the ears 4 2. Rostrum large, laterally compressed and rectangular in profile. Lips with a large flap Hvpsiqnathus Not as above 3 3. Orbit to the tip of the nasals more than the lacrymal breadth Epomophorus Orbit to the tip of the nasals less than the lacrymal breadth Micropteropus 4. First upper cheek tooth minute. Head and shoulders paler than the body Pteropus Not as above 5 ♦ (i.e. 3 upper and 5 lower cheek teeth) 191 CHIROPTERA MEGADERM I OAE EMBALLONURIDAE YELLOW WINGED t FALSE VAMPIRE BATS SHEATH TAILED BATS PTEROPODIOAE FRUIT BATS • FLYING FOXES Fig. 1 192 Chiroptera 5. Bulla with an auditory meatus. Body and femur covered with yellowish fur Eidolon Bulla without an auditory meatus. No yellowish fur Rousettus 6. Two inflated bulbs on top of the rostrum. Free tail, about as long as the head and body, (family Rhinopomatidae) Rhinopoma No bulbs on top of the rostrum. Tail shorter than the head and body 7 7. Tail perforates the upper surface of the interfemoral membrane, (family Emballonuridae) 8 Tail does not perforate the interfemoral membrane 9 8. Three lower incisors; frontals concave. Forearm 45 to 55 mm long Coleura Two lower incisors; frontals not conspicuously concave. Forearm 50 to 83 mm long Taphozous 9. Dish face depression on the rostrum. Tail ends in a T-shaped tip, (family Nycteridae) Nycteris Tail does not end in a T-shaped tip 10 10. Dental formula 0/2, 1/1, 1/2, 3/3. Tailless; tragus divided, (family Meqadermidae) 11 Bats with tails; tragus, if present, not divided . 12 11. Palate extends anteriorly slightly beyond the nasals. Overall colour grey; nose leaf does not extend half way to the base of the ears Meqaderma ( = Cardioderma ) Palate extends anteriorly well beyond the nasals. Wings yellowish in life; nose leaf extends more than half way to the base of the ears Lavia 12. No tragus present 13 Tragus present, may be small 17 13. Dental formula 1/2, l/l, 2/3, 3/3. Leaf nosed; Two joints in the first toe, three in the others, (family Rhinolophidae) Rhinolophus Dental formula l/2, 1/1, 1-2/2, 3/3; zygoma large. Leaf nosed; two joints on all the toes, (family Hipposideridae) 14 14. Cheek teeth 5/5. Nasal structure may or may not have three pointed flaps above 15 Cheek teeth 4/5. Nasal structure horseshoe- shaped with three pointed flaps above 16 15. Nasal structure squarish Hipposideros Nasal structure horseshoe shaped with three pointed flaps above; ear notched Triaenops 16. Ears very short, hardly projecting above the fur Cloeotis Ears large, naked Asellia 193 J.E.Afr. Nat. Hist, Soc. Vol.XXV No. 3 (112) January 1966 17. Braincase thick, flat, broad. Tail projects beyond the edge of the interfemoral membrane; fur short, velvet-like, (family Molossidae) 18 Braincase generally delicate, round. Tail extends only to the edge of the interfemoral membrane; hair often long, silky, (family Vespertilionidae) 20 18. Skull height 1/3 of the width Platymops Skull height at least 1/2 of the width 19 19. Flange on the zygomatic arch large. A pale greyish area on the upper back Otomops Flange on the zygomatic arch indistinct. Generally uniform colour, sometimes spotted with white Tadarida (= Nyctinomus . and including subgenera Chaerophon and Mops ) 20. Two upper incisors. Ear more than 15 mm long Laephotis Ear less than 15 mm long 21 21. Cheek teeth 6/6 22 Cheek teeth less than 6/6 23 22. First two upper cheek teeth small. Fur long and woolly; ear rather large, pointed and funnel- shaped; margin of the interfemoral membrane fringed with hair Second upper cheek tooth minute. Fur not long and wooly; tragus erect and tapering 23. Cheek teeth 5/6, the first upper being minute. In the longest finger (3rd.) the second bone from the"wrist" is about l/3 the length of the third bone; tail long, about the length of the head and body Miniopterus the second bone from the "wrist" is well oyer 1/3 the length of the third bone 24 24. Cheek teeth 5/5. Generally very small species Pipistrellus Cheek teeth 4/5 25 25. Condylo-basal length of skull less than 16 mm 26 Condylo-basal length more than. 16 mm. Usually lemon-yellow or cream coloured fur below Scotophilus 26. One upper incisor Nycticeius Two upper incisors 27 27, Braincase very deep. A fleshy lobe at the base of the mouth connected by a ridge to the base of the ear Glauconycteris Not as above Eptesicus Kerivoula , . . Myotis 194 ERRATUiy[ to J.E. Afr. Nat. Hist. Soc., Vol.XXV, No3 (112) Jan 1966, p.l94. Key to Genera of Chiroptera: No. 20 should read: 20. Two upper incisors, cheek teeth less than 6/6. Ear more than 15ram long Ear less than 15mra long, or else one’ upper'** Incisor or cheek teeth 6/6 . . . . .T Laephot is 21. m li' Key to the Rodents 46 Genera The key is based on skull characters. Skin characters given for each genus are seldom key characters and are intended to be used mainly as a measure of confirmation of an identification already arrived at by using the skull. The skin characters are short and some- times, perhaps, a little vague; this is due to the necessity for brev- ity and to include the genus in all its forms through out East Africa. Unless it is specifically stated, the lower jaw is not used in the key, and the fur colours given are the general overall colours of the back and sides. For the sake of simplicity, M is loosely used for all molariform teeth. The animals are divided into size groups as follows;- "Very large" - any size greater than Rattus rattus ; "Large" - size about that of Rattus rattus ; "Medium" - size about that of Rattus (Mastomvs) coucha; "Small" - the size of Mus musculus or smaller. In the case of the Muridae a well haired tail means well haired for a Murid , and would be very poorly haired when compared with, for instance, a squirrel or a dormouse ( Graphiurus ) . A common example of a well haired tail is that of Arvicanthis . and a poorly haired tail that of Rattus rattus. A pencilled tail is one which becomes very narrow and well haired towards the tip, a common example is Grammomys . It has been found inconvenient to key the Murids according to their subfamilies, of which there are three:- Dendromyinae . Murinae and Otomyinae. There is, furthermore, some disagreement as to the class- ification of some genera. Delanymys . Dendromus . Deomys . and Steatomys . with the possible inclusion of Beamys and Saccostomus , are Dendromy- inae , Otomys belongs to the Otomyinae . and the rest are all Murinae. In the case of the genus Rattus a departure has been made from the general plan, so that the five subgenera, all of which have form- erly held full generic status, have been keyed out separately. Infraorbital foramen AB = Condylo-basal length Fig. 2 19b J. E.Afr. Nat. Hist. Soc Vol.XXV No. 3 (112) January 1966 Infraorbital foramen small, rarely as large as the surface of the largest molar, or else absent, (suborder Sciuromorpha ; some mole- rats and all squirrels 1 Infraorbital foramen present, larger in one dimension than the surface of the largest molar 8 1. Angular portion of the lower jaw turned out- wards; incisors white. Fossorial; tail short, (superfamily Bathyerqoidea . family Bathverqidae) 2 No outward turn to the lower jaw; incisors yellow; generally arboreal; tail long and bushy, (superfamily Sciuroidea . family Sciuridae) 4 2. Cheek teeth 3/3 or 2/2. Size medium; body naked. . .Heterocephalus Cheek teeth more than 3/3. Size large; body furred 3 3. Cheek teeth always 4/4; palate extends well beyond the tooth row, as far as, or beyond, the roots of the upper incisors. Fur grey-brown, and usually with a white patch on top of the head Cryptomys Cheek teeth 6/6, though usually no more than 5/4 present at one time, often 4/4; palate does not extend far beyond the tooth row, neyer reaching the roots of the upper incisors. Fur light grey-brown with no white patch on head Heliophobius 4. Palate extends well beyond the tooth row. Size large to yery large; fur bristly Xerus (including subgenus Euxerus) Palate does not extend conspicuously beyond the tooth row. Fur soft 5 5. Infraorbital foramen egg-shaped, about the size of the surface of the largest cheek tooth. Cheek teeth 4/4. Size yery large; ventral surface poorly furred, sharply divided from the well furred sides and back Protoxerus Infraorbital foramen smaller than the surface of the largest cheek tooth. Ventral surface with less fur than the sides, but there is no sharp division 6 6. Cheek teeth 4/4. Size very large, (including subgenus Aethosciurus ) Heliosciurus Cheek teeth 5/4. Size medium to large 7 7. Both upper and lower Cheek teeth flat crowned in the adult. Four mammae, small round ears Funisciurus Lower cheek teeth tend to remain cuspidate in the adult. Six mammae, ears not unusually small Paraxerus (including subgenus Tamiscus ) 8. Infraorbital foramen generally V-shaped, (suborder Myomorpha: most rats and mice) 13 196 Rodents Infraorbital foramen large, oval or round, usually larger than the foramen magnum; four cheek teeth. Size always very large, (suborder Hvstricomorpha) 9 9. Volant or saltatorial. Fur soft, (superfamily Anomaluroidea) 10 Cursorial. Fur bristly or spiny, (superfamily Hvstricoidea) 11 10. Infraorbital foramen slightly smaller than the foramen magnum. Volant adaptations, with a membrane between the limbs; sharp scales present at the base of the tail, (family Anomaluridae) Anomalurus Infraorbital foramen larger than the foramen magnum. Saltatorial, with long hind legs and tail; no scales at the base of the tail, (family Peditidae) Pedetes 11. Upper incisors with 3 grooves. Fur bristly, (family Thryonomidae) Thrvonomvs ( = Choeromvs ) Upper incisors not grooved. Pelage spiny, with long, hollow quills, (family Hvstricidae) 12 12. Top of the skull flat. Tail about as long as the head and body Atherurus Top of the skull very convex. Tail short Hvstrix 13. Temporal muscles cover the cranium, and are divided by a sagital crest in mature individuals; cheek teeth 3/3; incisors yellow. Size large; fossorial adaptations, lips joined behind the upper incisors; tail Temporal muscles originate only on the sides of the cranium, no sagital crest. Lips not joined behind the upper incisors 14 14. Cheek teeth 4/4. Size small-medium; fur short, soft and dense, light grey or grey-brown in colour; tail thick and bushy, (family Gliridae) Graphiurus Cheek teeth 3/3. Tail, except for Lophiomvs . relatively sparsely haired, never bushy 15 15. Cheek teeth either laminate, with laminae separated by wide folds, or cuspidate with cusps in 2 rows, (family Cricetidae) 16 Cheek teeth, if laminate, with laminae tightly pressed together; usually cuspidate with cusps in 3 rows, (family Muridae) 19 16. Temporal fossae roofed over by plates of bone arising from the frontals, parietals, and jugals; upper surface of the skull granulated; incisors ungrooved, white. Size very large; hair long with a black and white erectile crest on the back; tail short and bushy, (subfamily Lophiomvinae) Lophiomvs 197 J.E.Afr. Nat. Hist. Soc Vol.XXV No. 3 (112) January 1966 Temporal fossae open; upper surface of the skull not granulated; upper incisors grooved, yellow. Size small to large; fur short and generally golden brown in colour; tail long and well haired or tufted at the tip, (subfamily Cricetinae) 17 17. Zygomatic plate projects less than half way from the posterior edge of the infraorbital foramen to the incisors. Size small, hind foot generally less than 24 mm long, with soles naked, subgenus Dipodillus , or completely haired, subgenus Gerbillus Gerbillus Zygomatic plate projects about half way from the posterior edge of the infraorbital foramen to the incisors. Size medium to large; hind foot generally more than 24 mm long...... 18 18. Second pair of palatal foramina shorter than the length of Ml. Size large, generally weighing over 75 gms . ; hind foot usually more than 35 mm long with the sole entirely naked Tatera Second pair of palatal foramina longer than the length of Mi. Size medium, generally weighing less than 75 gms.; hind foot usually less than 35 mm long with a narrow band of fine hairs across the sole Taterillus 19. Incisors not grooved 20 Incisors grooved 44 20. Condylo-basal length of the skull more than 50 mm. ; palatal foramina shorter than, and not nearly reaching, the tooth row. Size very large; fur short, fine and grey-brown in colour; tail about equal in length to the head and body, fairly well haired, dark in colour for at least 3 or 4 inches from the base, changing abruptly to white distally; possesses cheek pouches Cricetomvs Condylo-basal length of the skull less than 50 mm.... 21 21. Palate ends far behind Mg, the end acute, V-shaped 22 Palate does not end far behind Mg, the end rounded or square 23 22. Incisors outstandingly pro-odont. Size medium; fur coarse, brown; tail very much shorter than the head and body and fairly well haired Uranomvs Incisors orthodont. Size small - medium; fur spiny, varying in colour from light red-brown to dark grey, belly white; tail slightly shorter than the head and body, bicoloured Acomvs 198 Rodents 23. Mj relatively large, larger than M2+ M3; the ■^wearing surfaces of the upper incisors notched. Size small; fur fine, though sometimes "crisp", grey-brown in colour, belly sometimes white; tail shorter than head and body Mus Mj^ not longer than M2+M3, the wearing surfaces of the upper incisors seldom notched 24 24. Condylo-basal length of the skull less than 20 mm. Size small; fur soft, brown, with a black patch between the eye and the nostril; tail much longer than the head and body, poorly haired Delanymys Condylo-basal length more than 20 mm 25 25. Palatal foramina short, not nearly reaching 26 Palatal foramina end about level with, or posterior to, Mj^ 27 26. Rostrum flat in profile; zygomatic arch with a small flange. Size large; fur soft, brown; tail slightly longer than the head and body, poorly haired Malacomvs Rostrum convex in profile; zygomatic arch with no flange. Size medium; fur short, soft, uniform light grey; tail slightly shorter than the head and body, very finely haired, grey at the base, white distally; possesses cheek pouches Beamys 27. Frontals perfectly flat in the young, and generally concave in the adults. Size large; fur soft, brown, with a dark patch extending from the nose backwards to surround the eyes, belly white; tail longer than the head and body, pencilled Thallomys Frontals at least slightly convex, or convex in parts 28 28. Molars massive, M3 as long as M2 29 M3 shorter than M2 30 29. Width across both palatal foramina less than the width of M]^ ; cusps normal. Size large; fur long and soft, brown; tail about equal in length to the head and body, poorly haired Dasymys Width across both palatal foramina greater than the width of M]^ ; cusps large and of uniform size, with the wearing surfaces strongly directed backwards in the upper jaw; the outer row of cusps of M3 very conspicuously reduced. Size large, fur long and soft, grey-brown with a deep rufous nose, ears, and rump; belly white; tail longer than the head and body, poorly haired Oenomys 30. Incisors narrow, markedly pro-odont; palate and palatal foramina wide; no supraorbital ridge. Size medium; fur soft, short, grey-brown; tail shorter than the head and body, finely haired Zelotomvs Incisors not pro-odont 31 199 J.E.Afr. Nat. Hist. Soc Vol.XXV No. 3 (112) January 1966 31. Distance from the anterior edge of the zygomatic plate to the tip of the nasals less than V6 times the length of the tooth row, measuring along the crowns 32 Distance from the anterior edge of the zygomatic plate to the tip of the nasals more than times the length of the tooth row, measuring along the crowns 34 32. Width of the rostrum immediately in front of the zygomatic plate equal to, or greater than, the length of the rostrum (measured from the zygomatic plate) , and about equal to the length of the tooth row; no black membrane covering the skull. Size medium-large; fur short, coarse, yellowish-brown to light grey, flecked with black; tail shorter than the head and body. well haired, bicoloured Arvicanthis Width of the rostrum less than its length (from the zygomatic plate); a thin black membrane covers the surface of the skull (this easily removed) 33 33. Zygomatic plate generally sharp pointed anteriorly. Size medium; fur short, coarse; general colour either dark brown with a black dorsal stripe and numerous longitudinal white stripes (L. barbarus) or broken stripes (L. striatus) on the back and sides, or else light orange-brown with a single black dorsal stripe (L. qriselda) ; tail about equal in length to the head and body, well haired, bicoloured Lemniscomvs Zygomatic plate rarely sharp pointed anteriorly. Size medium; fur short, coarse, grey-brown in general colour, with 4 dark longitudinal dorsal stripes; tail shorter than the head and body, well haired, bicoloured Rhabdomvs 34. Anterior ventral edge of the foramen enclosed by the zygomatic arch is level with, or posterior to, the anterior edge of Mt ; no supraorbital ridge. Size medium to large; fur short, coarse (or "harsh") with the colour varying from dark grey to deep red- brown; tail shorter than the head and body. poorly haired Lophuromvs Anterior ventral edge of the zygomatic foramen lies anterior to Mj^ 35 35. Anterior edge of the zygomatic plate turned outwards; zygomatic arch greatly flattened and deflected inwards ventrally; antero- internal cusp (T;^) of Mj^ and M2 absent, postero- internal cusp (Ty) present. Size medium - large; fur short, soft, brown; tail very short, well haired; possesses cheek pouches Saccostomus Zygomatic plate not turned outwards; zygomatic arch not greatly flattened; present 36 200 Rodents 36. Tj of Mj^ and 1^2 well developed, so that there ^are 3 well defined cusps on the inner side of and WI2. Size medium-large; fur soft, short, grey-brown; tail much longer than the head and body, pencilled Thamnomys Ty of Ml and M2 absent, or else so much reduced that there are only 2 well defined cusps on the inner side of M]_ and M2 37. Supraorbital ridge absent; in profile the top of the rostrum is in a straight line with the top of the cranium. Size medium - large; fur short, soft, and thick, with numerous guard hairs on the back and belly; deep brown in colour; tail longer than the head and body. poorly haired; hind foot very long, the metatarsals loosely knit together so that the foot can be greatly expanded in width Colomys Supraorbital ridge present; rostrum curved downwards. Hind foot normal 38 38. Outer borders of the palatal foramina strongly angled outwards. Size medium; fur short, soft, red-brown (redder on the rump) with a dark dorsal stripe; tail slightly shorter than the head and body, poorly haired Hybomys Outer borders of the palatal foramina straight or smoothly curved. Fur with no dorsal stripe 39 39. Distance from the anterior end of the palatal foramina to the incisors greater than the length of Mj^ ; supraorbital ridge prominent. Size large; fur very short, slightly coarse, and with long guard hairs; colour ranging from light brown to dark grey; tail about the same length as the head and body, poorly haired; of the hind foot extends well beyond the base^ of D^. Rattus ( Rattus ) Distance from the anterior end of the palatal foramina to the incisors less than the length of Ml ; or else about equal to the length of Ml, in which case the animal is medium sized, or smaller 40 40. Distance from the palatal foramina to the incisors less than the length of Mi. Size large or medium 41 Distance from the palatal foramina to the incisors about equal to the length of Mi. Size small - medium or medium 42 41. Zygomatic plate small, extends only very slightly anterior to the zygomatic arch; supraorbital ridge not very prominent. Size medium; fur short, soft, from light grey to red-brown in colour; tail longer than the head and body, pencilled; D5 of the hind foot reaches first joint of D4 Grammomys Zygomatic plate extends well in front of the zygomatic arch; supraorbital ridge very prominent. Size large; fur short; soft but not silky, with guard hairs; tail about the 201 J.E.Afr.Nat .Hist .Soc. Vol.XXV No. 3 (112) January 1966 same length as the head and body; of the hind foot does not reach the base of D4 R. (Aethomys) 42. Zygomatic plate small; extends only very slightly anterior to the zygomatic arch. Size small-medium; fur soft, brown to grey- brown with, usually, a darker patch surrounding the eye; tail longer than the head and body, finely haired; of the hind foot extends well beyond the first joint of D4 R. (Hvlomyscus) Zygomatic plate extends well in front of the zygomatic arch (at least 1 mm). of the hind foot does not reach the first joint of D4 43 43. Septum dividing the palatal foramina swollen only in the anterior half, the foramina extend posteriorly to at least the second root of Size medium; fur short, soft, grey-brown; tail about the same length as the head and body, poorly haired; D5 of the hind foot reaches, but does not extend beyond, the base of D4 Swelling in the septum dividing the palatal foramina extends beyond the anterior half, the foramina do not reach the second root of Size medium; fur short, soft, grey- brown; tail longer than the head and body, poorly haired; D5 of the hind foot extends a little beyond the base of D4 44. Both upper and lower incisors grooved; molars laminate; in the upper jaw M3 is the largest cheek tooth. Size large; fur long, soft, deep brown flecked with grey-black; tail much shorter than the head and body, well haired Otomys Only the upper incisors grooved; cheek teeth cuspidate, M3 being the smallest 45 45. Upper Incisors with 2 grooves; no zygomatic plate. Size large; fur short, slightly coarse, red-brown on the back and white on the belly, tail longer than the head and body, pencilled, bicoloured Deomys Upper incisors with only one groove; at least a small zygomatic plate present 46 46. Antero-internal cusp (Ti) of Mi and M2 absent. Size small 47 Ti of M2 and M3 present. Size medium-large or large 48 47. Tooth rows parallel to each other. Fur soft; light brown with a dark dorsal stripe; tail slightly longer than the head and body, finely haired; only 3 functional digits on the front feet Dendromus Tooth rows diverge anteriorly. Fur short, soft, grey to red-brown in colour with the belly white; tail shorter than the head and body, well, but finely, haired; 4 functional digits on the front foot Steatomys R. (Mastomys) R. (Praomys) 202 Rodents 48. Palatal foramina project beyond the anterior edge of Mi. Size large; fur long and coarse, orange-brown flecked with black; tail slightly longer than the head and body, well •haired with hairs up to 4 mm. long, bicoloured Mylomys Palatal foramina do not reach Mp. Size medium- large; fur rather long, coarse, light brown flecked with black, and often with a dark dorsal stripe; tail about the same length as the head and body, well haired, but the hairs less than 4 mm long, bicoloured Pelomys Acknowledgements Grateful acknowledgement is extended to Dr. G.B. Corbet of the British Museum (Nat. Hist.) for the loan of some material, and to Messrs. J.G. Williams, J.R. Wyman, M. Modha, I. Furtado and A. Menzes for help in checking the keys. Dr. J.B. Foster expresses his gratitude to the Canadian Goyernment who haye financed his appointment at the Uniyersity College. References 1. ALLEN, G.M. 1939. A Checklist of African Mammals. Bull .Mus . Comp. Zoo 1 .Haryard 83. 2. ANSELL, W.F.H. 1960. "Mammals of Northern Rhodesia - A Reyised Check List with Identification Keys, Maps and Notes on Distribution, Breeding and Ecology" . (Goyernment Printer) (Dept. Game and Fisheries) 3. BROWN, J.C. 1964. Obseryations on the Elephant Shrews (Macro- scelididae) of Equatorial Africa. Proc.zool.Soc.Lond. 143; pp. 103-119. ““ 4. BUTLER, P.M. 1956. The Skull of Ictops and the Classification of the Insectiyora. Proc.zool.Soc.Lond. 126: pp. 453-481. 5. COPLEY, H. 1950. "Small Mammals of Kenya". Highway Press, Nairobi. 6. CORBET, G.B. and NEAL, B.R. 1965. The Taxonomy of the Elephant Shrews of the Genus Petrodomus . with particular reference to the East African Coast. Rey. zool . Bot . Af r. . LXXI . 1-2, pp. 49-78. 7. DELANY, M.J. 1964. A Study of the Zoology and Breeding of Small Mammals in Uganda. Proc.zool.Soc.Lond. 142; pp. 347-370. 8. ELLERMAN, J.R. 1940-41. "The Families and Genera of Liying Rodents". British Museum (Nat. Hist.) Vols. 1 and 2. 9. ELLERMAN, J.R. , MORRISON- SCOTT , T.C.S., andHAYMAN, R.W. 1953. . "Southern African Mammals . 1758 -to 1951; A Reclassification". British Museum (Nat. Hist.) 10. HARRISON, D.L, 1960. A Checklist of the Bats (Chiroptera) of Kenya Colony. J. E ,Afr. Nat .Hist .Soc . 23; pp 286-295. 11. HOLLISTER, N. 1918-19. East African Mammals in the United States National Museum. Bull .U.S .Nat .Mus . 99; pt. 1,2. 12. MOREAU, R.E. and PAKENHAM, R.H.W. 1941. The Land Vertebrates of Pemba, Zanzibar and Mafia: A Zooqeoqraphic Study. Proc.zool. Soc.Lond. 110: pp 97-128. 13. ROMER, A.S. 1945. "Vertebrate Palaeotology'.' Uniy. Chicago Press. 203 J. E.Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 14. SWYNNERTON, G.H. and HAYMAN, R.W, 1951. A Checklist of the Mammals of Tanganyika Territory and Zanzibar Protectorate. J. E.Afr. Nat. Hist. Soc. 2Q: pp. 274-392. 15. SWYNNERTON, G.H, and HAYMAN, R.W. 1958. Additions to the Check- list of Tanganyika Mammals. J. E .Afr. Nat .Hist .Soc . 23; pp. 9-10. 16. WALKER, E.P. et al. 1964. "Mammals of the World". Vols. 1, 2 and (Bibliography) 3. Hopkins Press. (Received for publication 24th. November 1965) 204 J.E.Afr .Nat .Hist .Soc . Vol.XXV No, 3 (112) January 1966 THE HABITAT OF THE ROCK HYRAX By J.B, SALE (Zoology Department, University College, Nairobi) Introduction There are three genera of hyrax and two of these, Procavia and Heterohyrax , are -rock hyraces, living in existing cavities in rocky outcrops. Procavia is the larger of the two and is distributed throughout Africa and the Middle East while Heterohyrax is found down the eastern coast of Africa (Sale, I960). Both genera are gregarious, living in colonies among the rocks and in this respect differ from the solitary tree hyrax, Dendrohyrax , which, unlike the rock hyraces, is strictly nocturnal. Procavia and Heterohyrax in East Africa show no basic differences in behaviour and selection of habitat as far as "lowland" areas are concerned. Only the rather specialised P. John- ston! mackinderi Thomas inhabits the cold alpine zone of Mount Kenya, however. The tendency of some workers to refer to Heterohyrax as the "bush" hyrax (e.g. Roche, 1962) and accord it an intermediate position, as far as habitat selection is concerned, between Dendrohyrax and Procavia seems to me to be unjustified. Heterohyrax is just as much a rock dweller as Procavia and it is significant that while there are separate native names for tree and rock hyraces, no distinction is made between the two genera. In all social animals, the structure of the group depends to a large extent on the nature of the habitat and may have to be modified from time to time to fit in with changes in the habitat. If the animal is one which, like many rodents, modifies the habitat to suit its own social needs then a more rigid social structure can be main- tained and will be modified only to a minor extent by such factors as terrain, soil type, predominant vegetation and climate. King (1959) says of the prairie dog, Cynomys sp. "By building towns that occupy acres and even square miles, prairie dogs modify their environment and make it more suitable to their needs", and similar statements could be made about many burrowing social rodents. The rock hyrax, however, is not a borrower and, as Bruce (1790) points out, the fleshy part of the toes projects beyond the nails precluding their use in digging. Occasionally one finds evidence of the hyrax having scratch- ed out a small quantity of loose soil in order to enlarge a cavity or widen its entrance but even such minor modification of the environment as this is uncommon and certainly does not confirm Powell's statement (quoted in Shortridge, 1934) that they "do a considerable amount of digging" . The importance of suitable shelter to the distribution of the rock hyrax will be discussed more fully later but it can be stated herethat wherever there are rocky cliffs, outcrops or boulder screes providing cavities in which the colonies can shelter, one can reason- ably expect to find hyrax ( Drake-Brockman 1910; Shortridge, 1934). Within this basic requirement it is difficult to define a "typical" hyrax habitat in terms of environmental necessities, although where colonies exist there are certain visible indications of their presence 205 The Habitat of the Rock Hyrax such as urine deposits on the rocks (see Plate 1 a). Geology Throughout the highlands of Abyssinia and East Africa comparative- ly recent volcanic activity and faulting have resulted in the presence of frequent outcrops of igneous rock. Very many of these outcrops occur along fault scarps where the weathering of the scree has left a loose collection of boulders at the foot of the cliff (see Fig. lA) . These boulder heaps, with large interstices, constitute an ideal habi- tat for hyrax colonies. Outcrops formed in this way are common throughout the highlands and particularly in the region of the Rift Valley where, in addition to the great walls of the valley itself, there are numerous secondary rifts in the valley floor. In the region of volcanoes, outcrops often represent the front of a lava flow that has had the slag removed by erosion, leaving boulders of volcanic rock (see Fig. IB). In the large glaciated valleys radiating from the peaks of Mount Kenya similar collections of boulders have resulted from the weathering of lateral and terminal moraines (see Fig. 1C and Plate lb) . Hyrax colonies are not confined to boulder screes, however, and frequently occupy cracks and crevices that have formed in exposed rock faces due to cooling and erosion. In some places, such as Lukenya Hill in the Ukambani area east of Nairobi, large vertical cracks have resulted in the distal portion of rock falling away and coming to rest as a massive boulder on the slope below (Fig. 2). Where this has happened a shear vertical rock face is left behind in the form of a cliff. Further erosion by wind and water has made horizontal cavi- ties at the base of such cliffs by the removal of soil. These miniature caves form ideal hyrax shelters and are often connected to other similar cavities by horizontal ledges. If the boulder has not rolled far down the slope, its flat upper face forms an ideal basking surface near to the hole (Fig. 2). Any kind of rock may form a suit- able habitat as long as it provides shelter and outside the highlands particularly, hyrax are frequently found inhabiting metamorphic Basement System rocks and occasionally sedimentary rocks, such as sandstone. There are several accounts in the literature of hyrax living in the disused holes of other animals (Roberts, 1951). Thomas (1946) describes how one population of P. capensis Pallas in South Africa increased greatly following the destruction of wild predators (espec- ially jackal) by local sheep farmers. The result was over-population of the rocky habitat, forcing the hyrax onto the plains where they lived in the holes of the antbear, Orycteropus afer Grote, and Meercat, Suricata suricatta Schreber. They also took refuge in road culverts, holes in stone walls and any other available shelter. It is thus apparent that hyrax are adaptable in the matter of the type of substrate in which they will occupy holes. Although hard rocks are most commonly used, holes in softer rock and even soil are inhabited in some areas. The colony site It is also difficult to establish definite rules about the size or extent of holes required. Floor space would seem to be the criti- cal factor in hole size, as many holes are merely horizontal crevices no deeper than an adult hyrax in the crouching position (14 cm. for Procavia and 11 cm. for Heterohyrax) . In fact, holes with a high ceiling are generally not used as living quarters. Providing there 206 J.E.Afr.Nat .Hist .Soc. Vol.XXV No. 3 (112) January 1966 are several holes large enough to house a family group of about five adult hyrax (about 1 m2 of floor space), then there is the possibility of a small colony becoming established there. It is extremely rare to find a group living in a single isolated hole or crevice. In boulder screes, of course, there is usually an entire ramifying system of interstices comprising large cavities interlinked by smaller holes and cracks. In weathered rock faces, however, there are often large indi- vidual holes or crevices separated by a considerable distance from other cavities. Unless a cavity has similar shelter within about 10m. it will not be used as living quarters by hyrax, although they may take temporary refuge in it during flight from an enemy. Well worn paths among the rocks or along ledges serve to link together holes that have no internal connection. Such paths are mainly horizontally disposed and often a number of them, at approxi- mately the same level on the cliff, are linked together forming horizontal trunk paths which may extend along the entire length of a cliff or ridge (Fig. 4) . Occasionally a vertical minor path links together the horizontal ones. Vertically arranged paths also lead directly from the holes to the main feeding areas which are generally above or below the rocky cliff. These paths often pass near to an isolated boulder under which shelter can be taken if the animals are disturbed during feeding. In any case, the directness of these verti- cal feeding paths allows very rapid retreat to the living holes on such occasions. It is extremely difficult to determine what happens inside the the holes. It is impossible to dig out the system, except in a few cases where there is soil on one side of the hole, and the thickness of the dense rock precludes the use of radioactive tagging methods (Godfrey, 1954). Hence it is not possible to say with certainty whether a group of hyrax always live in the same part of a hole system or whether there is any functional subdivision of the system. Circum- stantial evidence, however, suggests that there is some constancy in use of various areas of the holes. For example, individual animals tend to use one entrance more than others. The habit of hyrax of having common urinating and defaecating places for each set of holes is widely known (Sale, 1960). It has also been observed that individ- uals in a group tend to occupy a regular place on the rocks when basking. These facts show that there is a positional constancy in hyrax, and in all probability this also applies when they are inside their holes. Its application must, of course, be adapted to the particular arrangement of holes in which the group is living and it would not be possible to produce the rather stereotype burrow plan that has been presented for many burrowing mammals. Climate and predators in relation to the habitat It can be seen from this brief account of colony sites that there is no fixed pattern for the hyrax dwelling. The animal adapts itself to any shelter that provides adequate protection from the elements and predators. In areas where there is a tendency to strong winds the animals avoid using rocks which are facing the prevailing wind, or if they do so they select a set of holes with protected entrances. For example some holes may have entrances on the leeward side of a large boulder. The Uaso Kedong gorge, where many of the present observations have been made, well illustrates the relationship of colony sites to prevailing wind (Fig. 3). The gorge runs in a north-south direction and at times there is a strong easterly wind in the area, especially during the night. This wind blows over the top of the east wall of 207 The habitat of the Rock Hyrax the gorge but catches the rocks on the west wall with considerable force and enters any east-facing cavities in these rocks. There are a number of Procavia colonies along the sheltered east wall but no perma- nent colonies on the west wall and the rocks are heavily covered with lichen. One or two, apparently lone animals have been observed from time to time on the west wall but no real colonies. Other colonies in the area are also on rocks that face west and are thus sheltered from the strong east wind. In areas where, because of the physiography, there is little wind, no such directional bias in the distribution of colony sites is observed. Climatic data for three areas where Procavia spp. are very common is given in Table 1 and it can be seen that the distribution of this genus shows wide tolerence in relation to altitude, temperature and rainfall. The main connection between rainfall and an herbivore popu- lation is the vegetation and this factor has been discussed in a paper on feeding (Sale, 1965). The wide tolerence of the rock hyraces in respect to food plants is largely responsible for the fact that rain- fall is not a primary factor in their distribution. Hyrax are adapted to a wide temperature range by their daily behaviour cycle which results in the animals avoiding the thermally extreme parts of the habitat . Location of Procavia spp. Altitude Mean Annual . Temperature Mean Annual Rainfall (approx.) Maximum Minimum Magadi 613 m. 35° C 17.8°C 56 cm. Naivasha 1,900 25 9.4 56 Mt. Kenya 4,200 10 -5.0 89 Table 1. The exposed rocks on which the lowland hyrax bask during the early part of the day often reach a very high temperature by the middle of the afternoon, with correspondingly low humidity. Air temperature recorded under a canopy above one of these exposed rocks at Uaso-Kedong was frequently up to 39° 0 with a relative humidity of around 10%. At Magadi even higher temperatures were recorded (Table 2). The night temperature at Uasa-Kedong was 16° - 18° C and humidity up to 85%. Thus the outside air around the rocks shows diurnal fluctuations of tempera- ture and humidity of 23° C and 75% respectively. A simultaneous recording of the temperature (but not humidity) 2m. inside a hyrax hole showed a maximum of 17° C and minimum of 14° C, i.e. an absolute fluctua- tion of 4°C. No doubt the humidity in the hole was of a similar rela- tive order. Table 2 compares these temperature measurements at Uaso- Kedong with similar recordings at Magadi, a much hotter area, and the Hausberg Valley on Mount Kenya where air temperatures are much lower (recordings here were not made above an exposed rock). 208 J.E .Afr. Nat .Hist .Soc . Vol.XXV No. 3 (112) January 1966 Table 2. LOCATION PERIOD OUTSIDE TEMP. °C HYRAX HOLE TEMP. °C Mean Max . Mean Min. Range Mean Max. Mean Min. Range Magadi 5 days 41.6 26.0 16.6 31.6 26.7 5.9 Uaso Kedong 7 38.6 17.5 22.1 16.6 14.4 3.2 Mt. Kenya* 4 9.0 -4.0 14.0 9.0 0.6 9.4 In all three areas the air temperature of the hole shows much less fluctuation (range) than that outside and thus provides a rela- tively constant microclimate into which the animals can withdraw in order to avoid extremes outside. At Magadi the mean minimum outside and in the hole is about the same but the high maximum temperatures on the rocks are never reached in the hole. The converse is true on Mount Kenya, where the mean maxima are the same but the low night temperatures outside are avoided in the hole where it seldom goes below freezing point. The shade of trees and other vegetation provides the animals with a moderate alternative to the extreme microclimate of the exposed rocks. Measurements of temperature and humidity in heavily shaded areas of the habitat confirm the anticipated intermediate range bet- ween figures for the rocks and the hole. There would appear to be a relationship between local predators and the maximum size of hole entrance regularly used by members of a colony. Except on high mountains, the leopard, Felis pardus pardus L. , is a common resident of the rocky areas where hyrax are found and is undoubtedly their main enemy. It is certainly the only major predator that would be likely to enter a hyrax hole. It is noticable that holes that are apparently otherwise ideal for hyrax habitation but have an entrance large enough to allow a large cat, such as a leopard, to enter are never used except for very temporary shelter. Holes, however, that are quite near to the surface but with a confined entrance are quite often inhabited and one can sometimes observe a group of hyrax huddled inside such a hole at very close quarters. They appear to "feel quite secure" even when so near to such a large enemy as man. A comparable manifestation of this "intelligent calculation" by hyrax was frequently seen in my captive colony of Mount Kenya hyrax. On an open roof the animals were extremely afraid of any strange intruder, such as a new human or a dog, and would immediately flee into the shelter of their home. When they were later placed in a wire mesh enclosure, however, a Boxer dog frequently tried to molest them through the wire, barking and rushing at them with open jaws. The * These figures were kindly made available by the Queen Elizabeth College Expedition to Mt. Kenya, 1964/65. 209 The Habitat of the Rock Hyrax The hyrax very quickly realised that they were quite safe with the wire separating them from the dog and although the younger animals showed threat, the adults took very little notice at all, sometimes even deliberately sitting with their noses against the mesh while the less intelligent dog charged them ferociously. On Mount Kenya there appears to be only a single pair of leopard in most of the large valleys. Considering the very large hyrax population in these val- leys, the rate of predation by leopard must be very low indeed, especially as hyrax is not its only food (Coe, 1963). It is interest- ing, therefore, to note that the Mount Kenya hyrax often lives in holes with quite large entrances, in some cases large enough to permit human entry. There would seem to be a connection between this habit and the comparative rarity of the leopard, especially as there is no shortage of holes with more confined entrances. Summary. Rock hyraces do not burrow but inhabit any type of rock providing suitable cavities as dwelling holes. Although the harder types are most commonly used, holes in sedimentary rocks and even soil, are inhabited in some areas. There is no fixed pattern for the hyrax dwelling but isolated holes are not generally used. The animals adapt themselves to any shelter that provides adequate protection from the elements and preda- tors. The temperature inside the hyrax hole never reaches the extremes of the outside air temperature. Holes facing the prevailing wind or with entrances big enough to allow the entry of a large predator are avoided, especially in areas where such predators are numerous. References BRUCE, J. (179C). Travels to discover the Source of the Nile, V: 139-145. CCE, M.J. (1963). Contributions to the ecology of the alpine zone of Mount Kenya. Ph.D. thesis. London. DRAKE -BRCCKMAN, R.E. (191C). The mammals of Somaliland. London: Hurst and Blackett. GCDFREY, G.K, (1954). Use of radioactive isotopes in small mammal ecology. KING, J.A. (1959). The social behaviour of prairie dogs. Scientific American 201: 128-140. Nature 174: 951-'^ ROBERTS, A (1951). The mammals of South Africa. South Africa: Central News Agency. ROCHE, J. (1962). Nouvelles donnees sur la reproduction des hyracoides. Mammalia 26: 517-529. SALE, J.B. (1960TT The Hyracoidea : a review of the systematic position and biology of the hyrax. J.E.Afr.nat.Hist .Soc. 23: 185-188. SALE, J.B. (1965). The feeding behaviour of Rock Hyraces (genera Procavia and Heterohvrax in Kenya. E.Af r.Wildl . J. J: 1-18. SHORTRIDGE, G.C. (1934). The mammals of South West Africa 1. London: Heinemann. THOMAS, A.D. (1964). The Cape Dassie. Afr.wild life. 1: 64. (Received for publication, 13th January 1966) 210 THE HABITAT OF THE ROCK HYRAX A. A FAULT SCARP Outer boulders enclosing cavities formed by the washing away of the finer materials. B. A LAVA FLOW Cavities formed by removal of slag. LATERAL MORAINES OF A GLACIATED VALLEY ( Mt. Kenya ). Fig. 1 211 THE HABITAT OF THE ROCK HYRAX SECTION OF ROCK OUTCROP SHOWING FORMATION OF HYRAX HABITAT BY VERTICAL CRACKING (e.g. Lukenya). Fig. 3 SECTION OF GORGE AT UASO-KEDONG SHOWING RELATION OF COLONIES TO PREVAILING WIND (shown by arrows) 212 THE HABITAT OF THE ROCK HYRAX 213 THE HABITAT OF THE ROCK HYRAX la. Part of the east wall of the Uaso Kedong gorge in the Rift Valley. Fault scarps of this kind are frequently occupied by rock hyrax, whose presence is indicated by white urine stains on the rocks ,e.g. the rocks on the extreme left. lb. The head of the Teleki Valley, Mount Kenya. The lateral moraine in the centre provides an extensive boulder scree habitat, housing a large colony of P.iohnstoni mackinderi. 214 J.E.Afr.Nat.Hist ,Soc. Vol.XXV No. 3 (112) January 1966 DAILY FOOD CONSUMPTION AND MODE OF INGESTION IN THE HYRAX By J.B. SALE (Zoology Department, University College, Nairobi.) Introduction Probably the earliest experiment on hyrax concerned its feeding and was carried out by the 18th Century explorer Bruce, in Ethiopia. He describes (Bruce,- 1790) how he shut up a hyrax, which had been starved for a day, in a cage with a chicken. The latter was not eaten by the hungry hyrax. A further experiment involved two small birds which, after several weeks, were also unharmed. From these experiments Bruce concluded that the hyrax was not a carnivore. Most observers agree that hyraces are herbivores and in a recent paper (Sale, 1966a) on the feeding behaviour of rock hyraces, Procavia and Heterohyrax . in Kenya it has been shown that these have catholic dietary habits. A study of the social behaviour of feeding shows rock hyraces to have some patterns similar to the feeding behaviour of ungulates. The earlier paper emphasised the speed and intensity of feeding in rock hyraces. The total time any colony has been observed group feeding is under one hour per day which, by comparison with many herbivores, is extremely brief. This suggests that either the food consumption of the hyrax is relatively low or that the mode of inges- tion permits an unusually rapid rate of food intake. The present paper represents an attempt to elucidate these two factors. In addition to extensive field observation of feeding colonies, detailed information on the amount eaten and mode of ingestion has been obtained from a captive colony of Mount Kenya hyrax, Procavia iohnstoni mackinderi Thomas. For the estimation of food and water intake and the collection of urine and faeces a metabolism cage, to house a single animal, was built (the details of which are being pub- lished elsewhere) and kept in a temperature-controlled room. Amount Eaten. It is always difficult to assess the amount of food eaten by animals under natural conditions. A certain amount of information can be obtained by weighing the stomach contents of dead animals but this method is unreliable for several reasons. Unless a very large number of animals have been previous-ly examined, it is impossible to state with accuracy the degree to which a stomach is filled. Hence, the contents of what appears to be a full stomach, may in fact represent only half a full meal. In any case, few animals eat the same amount at every meal. Only in rare cases where the behaviour of an animal, for a considerable period prior to death, has been observed, will the exact significance of the stomach contents be known. The method is destructive, uneconomic and denies the possibility of a series of measurements from the same individual, which is essential if any fluc- tuations are to be recorded. A very large number of "spot" observa- tions of this kind are needed for a given species before a true assessment of the quantity of food it eats can be obtained. 216 Food Consumption in the Hyrax Hyrax in the wild normally have an abundance of food except during rare drought conditions (Sale 1965a). It can thus be assumed that unless some disturbance curtails a feeding period, an animal will eat as much as it requires during each day. An animal that is well settled in captivity and being liberally fed ought, therefore, to give a fairly accurate assessment of the quantitative food requirements of a member of its species of the same age group and sex. A fully adult Mount Kenya male was kept in the metabolism cage and accurate records were maintained of the amount of food put into the cage and the amount uneaten at the end of 24 hours. During the period October 1962 to March 1963, when the room temperature was constant around 21* C (rela- tive humidity varied between 44 and 56%) , two contrasting types of food material were given. Freshly-collected sow thistle (Sonchus sp.) was fed for the first six weeks and then replaced by lucerne hay. The animal had eaten both these foods equally well during three months in captivity prior to the experimental period. Water-content, nitrogen and crude protein determinations of samples of the food were carried out at intervals and average values for each food over the period obtained. While being fed on Sonchus . the animal had no water but was given water ad lib while being fed on lucerne. A summary of the data obtained is given in Table I. It can be seen from Table I that there is a very great difference between the daily intake of Sonchus (896 gm) and lucerne (119 gm) . This can largely be accounted for by the fact that the former contains much more water (86.6%) than the latter (15.2%). The dry matter in- take of the two feeds shows very little difference and appears to be the main factor determining the total amount of food consumed. The much larger quantity of fresh Sonchus was apparently not taken in order to obtain water. The urine output decreased significantly when water was mainly obtained through drinking (while feeding lucerne) , suggesting that in the first period, water obtained from the Sonchus was in excess of requirements. There was also a great increase in the concentration of the urine in the second period, further demon- strating that water was consumed in excess during the first period. It is clear from these experimental results that the water content of the food greatly influences the amount consumed. Hence the weights of wet stomach contents are a completely unreliable guide to the amount of food being consumed by an animal. If expressed as dry matter, a more useful comparison can be made but will still be subject to the objections stated above. While the dry matter intake of the two foods was similar, the Sonchus provided over 100% more nitrogen and crude protein than the lucerne. Since there was always uneaten lucerne in the cage, it can be assumed that the nitrogen and crude protein provided by the daily intake of this food was adequate. This means that the nitrogen and crude protein provided by the Sonchus intake were in great excess of requirements and, like the water provided by the Sonchus . were not critical factors in determining the amount of the food consumed. The only factor which shows approximate equality in the two food intakes is the dry matter which must therefore basically determine the amount of a food consumed. The "voluntary" lowering of the water intake with a lessening of protein intake may reflect a built-in mechanism which cuts down water requirements when the quality of vegetation decreases. Such a protein decrease in the plants takes place during the dry season when water 216 J.E.Afr.Nat.Hist.Soc. Vol.XXV No. 3 (112) January 1966 is also short. A mechanism of this kind has recently been demonstrated in native cattle (Payne, 1963) and, if present in hyrax, would partial- ly account for its ability to live for long periods in very arid areas such as the Sahara (Monod, 1963; Grasse, 1956). The data obtained from this one animal suggest that a large adult male Mount Kenya hyrax will eat approximately 111 gm. (average figure) of dry matter per day. For this animal, weighing 3.3 kg., this is 33.6 gm. per kg. body weight per day. Table 2 compares this ratio with that of other species. The hyrax ratio is in the same range as that of sheep which vary according to body weight, e.g, 31 gm, per kg. body weight for an animal of 79 kg. (Spector, 1956); 34,5 for an animal of 60 kg. (Woodman 1948), This means that for its size the hyrax has a modest food intake, as the ratio normally increases as body weight decreases. The reason for a relatively low food intake may be that the hyrax is not a very active animal and spends the major part of the day hud- dled in a hole or lying outside on the rocks (Sale, 1965). It is a marked feature of Table 2 that very active animals such as the Wallaby and Howler Monkey have abnormally high food intakes. Figures could not be obtained for markedly inactive mammals but it seems likely that the converse of the above trend would operate in such cases. An addi- tional factor is that hyrax exhibit a higher degree of thermolability than many mammals (unpublished observation) and will thus use less energy for their size. The low food intake is undoubtedly a signifi- cant factor contributing to the ability of the rock hyraces to inhabit areas where vegetation is sparse or of a poor nutritional quality and may also be partially responsible for the fact that they spend relatively little time feeding. Mode of Ingestion Use of the feet; The feet are not extensively used in hyrax for manipulating food. Sometimes a tall herb or small shrub will bear shoots and leaves out of the reach of a feeding animal. If the plant is not stout enough to be climbed, the hyrax may raise itself up and press on the stem about 20 cm, from the ground with its forefeet, thus bending or breaking the stem and bringing the edible parts of the plant within reach. A similar technique is often used by goats and browsing antelopes such as the Gerenuk, Litocranius waller! Brooke. If a difficult piece of food is encountered on the ground one of the forefeet may be employed to steady it while a portion is being bitten off. This happens far more frequently in captivity than in the wild. Food such as carrots or mealies which tend to roll around are often held in this way while the side of the mouth is brought into position and a bite taken. Hyrax have never been observed carrying food into their holes in the wild, although occasionally one finds evidence that a small twig has been dragged into the hole and stripped of its leaves. Even shy newly-captured animals rarely take food into the dark part of the accommodation but generally eat it outside when undisturbed. Mollaret (1962), who has kept both Procavia and Dendrohyrax in captivity, states that only Procavia uses its feet to hold food which it often drags into its shelter to eat there. He does not offer any explana- tion for the lack of such behaviour in Dendrohyrax but it may be connected with the fact that this genus is more easily tamed than Procavia . which he admits is the more aggressive and difficult to handle. A tame animal would probably be less afraid of eating in an 217 Food Consumption in the Hyrax exposed place than one which was uneasy in captivity. Also Dendro- hvrax . being nocturnal, always eats at night. Unlike many small mammals, hyrax do not pick up objects with their forefeet. The pad-like structure of the feet and absence of really separate prehensile digits (Plate Ib) which make it difficult to do so. The gait of hyrax too, is not predisposed to such action. Mammals, such as many of the rodents, that grasp and lift objects with their hands have long hind limbs on which they are able to walk and sit erect easily, without overbalancing. Although hyrax can stand erect momentarily (tame animals do it when begging for food) , they are unable to walk for more than a few steps or sit still in this upright position. The body-shape of the hyrax is similar to that of a bear but the shorter trunk of the latter enables it to balance in the sitting position more easily. Use of the teeth. The upper incisors of hyrax are widely separated and developed into a pair of sharp tusks (Plate la), triangular in cross-section. The lower incisors, of which there are two pairs, are flattened and deeply incised so as to form comb-like structures used in cleaning the fur, like those of the lemur. Hence the incisors are unsuitable for biting off small shoots and they are little used in ingestion. The normal mode of browsing is to turn the head sideways ( at 90® to the body) and bite off the shoot or leaf with the molar teeth and take it in through the side of the mouth (Plate Ic). Hyrax look rather clumsy when feeding and remind one of a carnivore gnawing a large bone which remains projecting from the side of the mouth (^Plate Ib) . The use of the molars in cropping leads to the distinctive flat-topped appearance of the tussocks of Festuca sp. on Mount Kenya (Plate Ila) . Hyrax can be seen with their heads twisted, so as to bring the side of the mouth into a horizontal position during this cropping operation. The relatively long cutting edge provided by the molars (Plate lib) enables a large amount of food material to be taken at each bite and thus assists in rapid feeding. The greatest advantage of this will be realised during grazing or cropping when the whole length of the molar row is in use. A rough comparision of the length of the cutting edge and dry matter intake rate in relation to body weight in the hyrax and two grazing ruminants, where the lower incisors form the cutting edge, is given in Table 3 (for authorities see Table 2). It is thus clear that for a herbivore of its size the hyrax can take in food at a very great rate and this must largely account for the relatively brief feeding time. There is occasional use of the incisor teeth and tongue in inges- tion, as follows. If the material is a little out of reach, then, with the neck outstretched, the upper incisors may be used in conjunc- tion with the tongue, the latter pressing the leaf up onto the incisors. The neck is then contracted and as the leaf is pulled, it either breaks along the line of the incisor perforations or it breaks off at the base of its petiole. In the latter case the whole leaf is obtained and can be seen impaled on the upper incisors and projecting from the front part of the animal's mouth. It is removed with the aid of the tongue and the lower incisors. The lips do not appear to be used in the ingestion of food mater- ial but are employed in drinking which strongly resembles that of ungulates. The lips are lowered to just below water level and water is sucked into the mouth in gulps as the lips are slightly parted. The tongue plays very little part in the process. 218 J.E.Af r. Nat .Hist .Soc . Vol.XXV No. 3 (112) January 1966 Rumination and refection. Ingested food material is rapidly chewed in a side-to-side motion before swallowing. Throughout my observations of rock hyraces I have found no evidence of rumination. Hyrax will sometimes produce a chew- ing motion without having recently ingested and such action is partic- ularly common when they are confronted by something which is strange to them. It has been observed, for example, when captive animals are introduced to an unfamiliar animal species such as a caged bird. Newly-captured and nervous animals frequently show it when being observed by humans. This motion reminds one forcibly of a ruminant and is probably responsible for the statement by some observers that hyrax chews the cud (Bruce, 1790). Very low intensity pilo-erection is also manifest on such occasions indicating a conflict response to the strange situation (Sale, 1965b). Recently Hendrichs (1963) claims to have observed rumination in P. capensis in captivity in Europe. He informs me (Hendrichs 1965) that the animals chewed the cud for hr. (in 24 hr.) when fed on dried grass (? hay). Until more details of these observations are available it is unwise to comment but my own view is that although hyrax sometimes chew in the absence of ingestion, they do not regurgi- tate material from the stomach for further mastication. The simple structure of the stomach would appear to make such action extremely unlikely. Should rumination in the hyrax be established, the accuracy of the Bible (Lev. XI, 5), where the coney is stated to chew the cud, will be attested. Refection would seem to be a more likely phenomenon in the hyrax than rumination. The process is known to exist in the wild rabbit (Madsen, 1939; Southern 1940, a) and Southern (1940,b) has drawn attention to its possible usefulness during enforced long periods in the burrow due to bad weather or disturbance preventing feeding. Coe (1962) instances a colony of Mount Kenya hyrax that "remained below ground for three days" during a period of bad weather. Although I have no concrete evidence of such prolonged periods without feeding, it seems likely that hyrax can stay in their holes for more than 24 hr. Coprophagy would clearly be a potential mode of nourishment during periods of confinement. Captive hyrax have frequently been seen to sniff and lick fresh faeces but the occurence of actual ingestion has not yet been established. The occurence of refection in the hyrax would provide an interesting comparison with the elephant, where the eating of a quantity of fresh faeces has recently been reported (Dougall and Sheldrick, 1964). Discussion. The development of a pair of upper incisors as defensive tusks is one of a number of characteristics that hyrax have in common with the elephant. In both cases this has precluded the use of the incisors for biting and alternative modes of ingestion are used. The elephant has developed the trunk as a highly efficient organ of prehension and suction, unparalled among the mammals. The hyrax have a less unconven- tional mode of ingestion using the molar teeth, which, because of their long cutting edge allow a large amount of food to be taken in at a time, especially when cropping leafy vegetation. It is interesting that while the elephant is the largest ungulate-type mammal, the hyrax IS the smallest. The efficiency of the trunk in ingesting large amounts of food material has undoubtedly contributed to the great size of the modern elephant and enabled the group to radiate out from Africa where it had its origins. The hyraces, on the other hand, have 219 Food Consumption in the Hyrax become smaller in size and largely remained confined to Africa where they appear to have had a common origin with the probiscideans in the late Eocene (Romer, 1933). This decrease in size is, however, proba- bly due to the fact that the hyrax has remained plantigrade and has never been capable of the rapid locomotion that enables modern ungulates to escape from predators. Hyraces have therefore left the plains for the protection of rocky outcrops or hollow trees, their sharp incisors being ideal for the defence of the entrance to a hole. This change in niche has necessitated a decrease in size as few existing holes are large enough to house the Oligocene hyraces which were the size of a large hog. Changes in niche and body size have been accompanied by behavioural changes. In general these have involved decreasing activity, resulting in a lowering of relative food requirements . The survival advantage, to a small herbivore such as hyrax, of a short feeding period, involving maximum exposure to predation, is clear (Sale, 1965a). The factors contributing to the shortness of the feeding time in hyrax, viz a low food intake and rapid mode of ingestion, would seem to have been produced by the physiological and anatomical changes accompanying the change of habit from plains- dwelling grazer to a rock-dwelling browser. Summary The dry matter content of a food determines the amount of it that a hyrax will eat. Foods with a low protein content probably demand a lower water intake than those rich in protein. The dry matter intake of an adult Procavia was found to be 33.6 gm per kg. body weight per day, which is low for an animal of this size and may be connected with the relatively inactive life and poor temperature regulation of the rock hyrax. Hyrax rarely use their feet in grasping food material which is seldom carried into the holes. The development of the incisors for defence and toilet purposes makes them unsuitable for use in ingestion. The molar teeth are used to bite off plant material, an action often necessitating the turning of the head sideways. The cutting edge provided by the molars is relatively long and enables the animal to take in a large amount of food material, thus contributing to the rapid feeding of hyrax. 220 J.E.Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 Acknowledgements The analyses of food materials were kindly carried out for me by the Biochemical Unit in the Animal Husbandry Division of the East African Agricultural and Forestry Research Organisation, by courtesy of Dr. L.J.A. Payne to whom I am greatly indebted. I am also grateful to the Rockefeller Foundation, New York and to the former Royal College, Nairobi for financial assistance. References ALBRITTON, E.C. (1954). Standard values in nutrition and metabolism. London. BRUCE, J. (1790. Travels to discover the source of the Nile, OOE, M.J. (l962). Notes on the habits of the Mount Kenya hyrax (Procavia johnstoni mackinderi Thomas) Proc. zool. Soc. Lond. 138: 639-644. DOUGALL, H.W. and SHELDRICK, D.L.W. (1964). The chemical composition of a days diet of an elephant. E.Af r.Wildl. J. . 2: 51-59 GRASSE, P.P. (1956). Traite de zoologie. XVII. Mammiferes. Paris; Masson. HENDRICHS, H. (1963). Wieder Kauen bei klippschllef ern und Kanguruhs. Naturwissenschaften 50 (12); 454-55. HENDRICHS , H . ( 1965) . Personal communication MADSEN, H. (1939). Does the rabbit chew the cud? Nature . Lond., 143; 981. MOLLARET, H.H. (1962). Naissance de demans en captivite. Mammalia 26 530-532. PAYNE, L.J.A. (1963). Personal communications. ROMER, A.S. (1933). Man and the Vertebrates, London: Pelican. SALE, J.B. (1965a). The feeding behaviour of Rock Hyraces (genera Procavia and Heterohyrax) in Kenya. E.Afr.Wildl. J. 1-18. SALE, J. B. ( 1965b) . Some aspects of the behaviour and ecology of the rock hyraces (Genera Procavia and Heterohyrax) . Ph.D. thesis. University of London. SOUTHERN, H.N. (1940, a). Coprophagy in the wild rabbit. Nature, 145: 262. SOUTHERN, H.N. (1940,b). The ecology and population dynamics of the wild rabbit (Oryctolaous cuniculus). Ann. Appl . Biol . . 27: 509- 526. ““ SPECTOR, W.G. (1956). Handbook of biological data. Philadelphia and London: Sanders. WOODMAN, H.E. (1948). Rations for livestock. Min, of Aqric. and Fish Bull, no. 48. London: H.M. Stationery Office. (Received for publication 13th January 1966) Food Consumption in the Hyrax Period in days Water Content DAILY CONSUMPTION IN GM. Dry matter Nitrogen Fresh Sonchus Lucerne hay 86.6 15.2 1940 280 3.68 1.64 23.08 10.28 SPECIES BODY WT. Dry matter per kg. body wt. per day Procavia 3.3 kg. 33.6 gm. Elephant (African) 3409 29 Zebra (Grevy) 409 20 Wallabv (M. aailis) 5 280 Wombat 9 63 Giraffe 1134 28 Howler monkey 3 238 Beef cattle 800 15 Sheep 60 34.5 Rat 0.3 50 A comparison of the daily food (dry matter) intake of various mammals. Compiled from Albritton (1954), Spector( 1956) and Woodman (1948). TABLE 2. A. Body Wt. B. Cutting edge A B C. Dry matter per kg. body wt. per day D. Time spent feeding per day C D Procavia 3.3 kg. 3 cm. 1.1 33.6 gm. .66 hr. 50.8, Sheep 79 3 26 31 10.5 2.95 Cow 800 8 100 15 6.5 2.3 TABLE 3. 222 FOOD CONSUMPTION IN THE HYRAX Plate la The incisor teeth of a male hyrax. The lower incisors of this old animal are worn down to peg-like stumps and are no longer comb-like. Plate Ib An animal eating a leaf which remains projecting out of the side of the mouth. The peculiar form of the feet can also be seen. Plate Ic Browsing hyrax, showing how the head is turned sideways as vegetation is bitten off with the molars. 223 FOOD CONSUMPTION IN THE HYRAX Plate Ilb A skull from the side showing molar teeth. The outer cutting edge of the upper molars overlaps the narrower lower molars when the jaws close, thus producing a scissors action. 224 J.E.Afr. Nat. Hist. Soc Vol.XXV No. 3 (112) January 1966 NOTES ON EAST AFRICAN COVWIIES By E. ROBSON This up to date list has gradually been compiled by various East African collectors. My particular thanks are due to Ken Fuller and Misha Fainzilber for data concerning the more recent additions, numbers 7, 31 and 42, which have been found in Dar-es-Salaam and Zanzibar. I have just added No. 43 which has been found at Shelly Beach and at Shimoni. Quite a few E. marginalis with their distinct violet base and B. oweni with deep cut teeth have been found at Zanzibar and also the little P. microdon. The once rare Cribraria chinensis was quite common in 1965 along the Diani Beaches. The Cowry collection at the National Museum has been sorted out and a representative collection is on view but is not yet complete. Nos. 23, 29, 31, 41 and 43 being still required to complete the series. I am preparing a collection of colour slides of shells - and cowries in particular, and hope to show these to interested people during 1966. There is considerable interest in East African varieties of such cowries as A. histrio . A. arabica. P. ziczac. M. moneta and L. mappa as we seem to have more than the usual known varieties here. Information on lesser known reefs and pools between Jadini and Shimoni is required, particularly details of specimens found, habitat and localities. A list of shells from the Lamu area and of beaches north of Malindi would also be welcome. Mrs. L. Cameron is collect- ing information on the colour and appearance of the bodies of Cowries, which should, when available, assist in the identification of specimens caught alive. List of Cowry Shells (CYPRAEIDAE) Recorded from the E. African Coast Genus Species Subspecies or Author Common Name ' Subfamily ADUSTINAE 1. Adusta 2. Cribraria 3. Palmadusta i 4. " 5. " 6. ; 7. " onyx cribraria clandestina ziczac ziczac ziczac contaminata variety adusta passerina diliculum virginalis misella distans Lamarck, 1959 Perry, 1811 Melvill, 1858 Reeve, 1845 Schilder, 1939 Perry, 1811 Schilder, 1939 onyx spotted cowry false 3 banded ziczac ziczac pale ziczac ' Subfamily Cvoraeinae 8.. Lyncina 9. Cypraea 10. Leporicypraea ! 11. " ■ 12. Ponda 13. Mystaponda lynx tigris mappa mappa carneola vitellus lynx alga mappa sowerbyi dama Linnaeus, 1758 Linnaeus, 1758 Perry, 1811 Linnaeus , 1758 Anton, 1839 Perry, 1811 lynx tiger map map flesh fallow deer Subfamily ERRONEINAE 14. Bistolida 15. Blasicrura 16. 17. Ovatipsa I 18. '' : 19. Talostolida stolida kieneri oweni caurica chinensis teres diaudes kieneri caurica violacea subfasciata Melvill, 1888 Hidalgo. Sowerby . Linnaeus , 1758 Rous, 1905 Link, 1807 square spot false swallow Owen violet long 225 Notes East African Cowries Genus Subfamily EROSARIINAE Species Subspecies or variety 20. Erosaria erosa erosa 21. " lamarcki lamarcki 22. marginalis pseudocellata 23. nebrites mozambicana 24. " turdus turdus 25. Monetaria moneta moneta 26, moneta icterina 27. Ornametaria annulus camelorum 28. Ravitrona caputserpentis caputserpentis 29. gangranosa reentsii 30. helvola argella 31. " poraria Author Common Name Linnaeus , 1758 Gray, 1825 Schilder, 1939 Schilder, 1939 Lamarck, 1810 Linnaeus , 1758 Lamarck, 1810 Iredale, 1939 Linnaeus, 1758 Dunkir, 1852 Melvill, 1888 Linnaeus , 1758 margined thrush money money gold-ring serpents head star porous 36, Mauritia immanis dispersa scurra mauritiana Schilder, 1939 Schilder, 1939 Qnelin, 1791 Gmelin, 1791 Linnaeus, 1758 arabian harlequin jester hump back Subfamily Ni^IlNAE 37. Evenaria 38. 39. Milicerona 40. Paulonaria 41. " 42. " 43. Evenaria asellus punctata felina f imbriata gracilis microdon hirundo asellus punctata felina durbanensis notata chrysalis hirundo Linnaeus, 1758 Linnaeus , 1758 Gmelin, 1791 Schilder, 1939 Gill, 1859 Kiener, 1843 Linnaeus, 1758 three banded punctate cat small toothed swallow Subfamily PUSTULARIINAE 44. Pustularia cicercula lienardi 45. " globulus brevirostris Josseaume,1843 Schilder, 1939 Subfamily STAPHYLAEINAE 46. Nuclearia nucleus 47. Staphylaea limacina 48. " staphylaea madagascarensis interstictina laevigata Gmelin, 1791 Wood, 1828 Dautzenburg , 1932 Subfamily TALPARIINAE 49. Arestorides 50. Basilitrona 51. Chelycypraea 52. Talparia argus Isabella testudinaria talpa contrastriata Isabella ingens imperialis Perry, 1811 Linnaeus , 1758 Schilder, 1939 Schilder, 1939 pheasant Isabella tortoise mole (Received for Publication October, 1965) 226 J.E.Afr.Nat.Hist .Soc. Vol.XXV No. 3 (112) January 1966 SIGHT ADDITIONS TO THE AVIFAUNAL LIST OF ETHIOPIA By EDWARD W. BEALS During three years of residence in Ethiopia, from September 1962 to July 1965, I have had the opportunity of travelling widely in that country. From bird records kept, I have found nine species, identi- fied without any doubts, that are apparently not previously recorded for Ethiopia. I have had no opportunity to collect birds, but these records may be of interest as additions to the hypothetical list of Ethiopian birds. Locations are numbered on the map. I must point out that the distribution of these records on the map reflect more the time spent in various areas than the choicest birding sites. The concentration of records is in the Main Ethiopian Rift, where I have done the most field work. Comments on previously known range are taken from Praed and Grant (1957-1960). Anthropoides virqo (Linn.) Demoiselle Crane. Two flocks seen, one of 12 birds on October 16th 1962, in a grassy marsh just north of Bishoftu (Loc. 1); the other of 17 birds on January 21st 1963, in a grass field on the west side of Lake Zwai (Loc. 2). Previously stated to visit the Sudan "in enormous numbers." I might also add that Meqalornis qrus (Linn.), the Common Crane, which was reported by Smith (1957) in Eritrea, has been found in the Rift Valley in open Acacia savanna 20 km. west of Shashamanne (Loc. 3), a flock of 5 birds on February 8th 1963. Trochocercus albonotatus Sharpe. White-tailed Crested Flycatcher. Two birds seen, one on the edge of secondary forest (in Podocarpus zone), 15 km. south of Shashamanne (Loc. 4) on April 6th 1963; and a second in similar but wetter forest 10 km. southwest of Jlmma (Loc. 5) on June 14th 1964. White spots in tail were seen clearly. Uganda and western Kenya were previously considered its northern limit. Sylvia ruppelli Temminck. Riippell’s Warbler. One male bird seen in Acacia scrub just west of Massawa (Loc. 6) on February 17th 1964. White moustache and black throat were seen clearly. Two other birds in the vicinity may have been females of this species. Previously described as a "locally common palaearctic winter visitor" to the Sudan. Phvlloscopus bonelli (Vieillot). Bonelli’s Warbler. One bird seen in Terminal ia-Anoqeissus woodland, on the escarp- ment west of Massawa (Loc. 7) on January 20th 1965. Yellow rump and whitish underparts were unmistakable. Previously recorded as far south as the Sudan. Psalidoprocne albiceps Sclater. White-headed Rough-wing Swallow. Four black swallows, two with white heads, were seen on August 14th 1963, on the edge of a Balanites -Acacia forest along the south- west shore of Lake Margherita (Loc. 8). It has previously been recorded as far north as the southern Sudan. Corvus ruficollos Lesson. Brown-necked Raven. In a flock of six black corvids seen on the coast at Assab (Loc. 9) 227 Sight Additions to the Avifaunal List of Ethiopia on January 8th 1963, two were distinctly larger than the other four, has more distinctly brown heads, and had proportionately heavier bills. Their call was a high-pitched croak, in. contrast to the 'caw* of the four smaller birds. Blair (1961) and North (1962) suggest that C. edithae Phillips and C. ruficollis are distinct species. Bird records from Ethiopia have been referred to C. edithae (Smith assumes this for the Eritrean coast). The two larger birds were definitely C. ruficollis. Nectarinia kilimensis Shelley. Bronze Sunbird. One male seen on August 16th 1963, in open Combretum savanna 25 km. west of Dilla (Loc. 10). The green head, bronzy chest, and elongated tall feathers were clearly seen. Described previously as common and widespread in Kenya. Nectarinia erythrocerca Hartlaub. Red-chested Sunbird. One male seen in Acacia scrub along the east shore of Lake Margh- erita (Loc. 11) on March 17th 1964. The red band quite across the chest and the elongated tail feathers were distinctive. Previously reported from the southern Sudan southwards. Uraeqinthus cyanocephalus (Richmond), Blue-capped Cordon-Bleu. Two males and at least one female were seen in a large mixed flock, (including many U. benqalus (Linn.), the Red-cheeked Cordon Bleu), on April 23rd 1965, in the Fafan Valley, 20 km. south of the Harar-Jijiga road (Loc. 12), in Acacia woodland. The clear all-blue head (of the males) and red bill (of both sexes) were distinctive. Previously reported from Kenya and southern Somalia southwards. References BLAIR, C.M.G. 1961. Hybridization of Corvus albus and Corvus edithae in Ethiopia. Ibis 103; 499-502. NORTH, M.E.W. 1962. Vocal affinities of Corvus corax edithae. "Dwarf Raven or Somali Crow". Ibis 104: 431. PRAED, C.W.M., and GRANT, C.H.B. 1957-1960. Birds of eastern and north eastern Africa, 2nd ed. 2 vols. Longmans, London. SMITH, K.D. 1957. An annotated check list of the birds of Eritrea. Ibis. 99: 1-26, 307-337. (Received for publication 21st July 1965) 228 AVIFAUNAL LIST OF ETHIOPIA 229 :■* *: iT'* L... NATURE NOTE Lake Abiata, Ethiopia, Weavers and Gabar Goshawks. Lake Abiata is one of a group of lakes at about 5,000 ft. altitude, about 150 miles south of Addis Ababa. They are a continuation North- wards of the Rift Valley chain of lakes; but at this point the Rift has no escarpment and the road descends gradually into flat thorn country with birds mostly familiar to an observer from East Africa. A river runs into the lake, and along the banks are several colon- ies of the Black-headed Weaver, Ploceus cucullatus (Muller) one of which I saw being searched by three Gabar Goshawks, Micronisus gabar (Daudin) , two grey and a black, intent on robbing the nests. They did not tear the nests to pieces as recorded by Praed and Grant, but each seized hold of the entrance to a nest and thrust its head inside. I have seen a Harrier Hawk do this; but the Gabar has not got such a long neck and is anyway a smaller bird. To get the head well into the nest the shoulders had to go too, while the wings were half closed and hung down like those of a moth. At this point the male Weaver would spring into action, striking the hawk's body with an audible thwack. The head would then be withdrawn. I never saw them pull anything out. A young bird or sitting female would no doubt have been devoured; but the Weavers were not at that stage. The females were still bringing leaves to line the nests, and the hawks would potter about in the bush before beginning their search long enough to give everyone time to get out. Most of them withdrew to a safe distance, and only a few males, perhaps those whose nests were being searched, remained to put up a resistance. The hawks would go away, but returned again and again during the afternoon. The relationship bet- ween them puzzled me. One was slightly larger than the others, so I took them to be a female and two males, perhaps a family party. In that case the black one must have been a male, as it was one of the two smaller. (Jackson says the black ones are usually females). P.M. Allen. 23/7/65. 230 •i’i'.u t. ^J- ■ ';. i> •Kl'^ .>‘*(i. 'I i^A •?!« 'i5 I'', ■':.* '-'■ ;j,^ 0.:: ' ^u}*d^ ^ ._ if, 'litj «ikr;n&- & »lbdU/* tTrliu-f ■ ’ii^ r, ^i.v- ■ . ;U ” o v n' t ant . -<^15 , A' j.,^3 ,T:W»t>W 1 .i';«# *# ‘ vfj ■ wn'<3i fcy ' « <^ #^;.' . ChitsiVto#' jn . 5«'S»i,,.'^'?^’ - .,_. i.iil ' \-il5: f‘‘ !■■":» -rveiiiaf ■WfM .J ,T iiw ^’i'. I U - ■ M 1 :f L ,: iWiStt' .CP c*a w*' &f(^i >t.:,t1J ►•l^Tfl)*)- ‘^i^T ■ ') In H t>oP 6 m ,. fcv -1 t?^i. ^ .' t ',< ( -v'8»i.. i w-t i;c - .** . ..iaP'iiK*4 . • -tJi''' i -’fm • ' • 'ftlaa b.-tfi plfiiiBt. s-‘fts»p# r cnivw^-'-* “<»■" '''■ ,.,.■■■*1;* VP. • ■>»/ iji F.-.fWV f.' •:!' 1^..^,. *.^ i«Tj ^ 'tr'! • -?i/n iJ* .■ ni^it h'f i.'c ^'. ■ .i''tf<« ir. ., , Mi.ftnr ,.Ctf'* -.if^ - . ;-V» ‘Ml, «|*t. !■»»*«».•'>•;■' #> .1' *'■•>■> - _’«» .; 'iw,- vV-'J''’> 5' ’aiWt,.:\.. t i'';... r -'f fsrtjpr. -T .. *> '•■ I- i «.i*tfe^'i -t ^ . , . .'.;■ ,vp.^.-r« j-r!?* J ••' ■' ....I,. ;,„. '«•»'* L?'*- < •>si <.. iipflci f ». » ‘T.i'a^,^ •' I I . / ,■ ,;•. ■■ H7^' * ; ; <■■' A .1 w ,/be ^ ^ a Mi' ' 1'-^; 'C^ .. .-,Mi , ■'!?P''^‘‘' . . V n , ■-•■’ i.|« 1, ..1.6^ - 1 '.(!5•^^.■'^l«’.^^*• •*"' 2^5$ 55f» ,. -w' , .,■ - .7, , i>V6riVaM‘f -V -....ie^lil.fl^li. J. E.Afr. Nat .Hist .Soc . Vol.XXV No. 3 (112) January 1966, A REEF HERON AT LAKE NAKURU, KENYA. By MYLES E.W. NORTH On 3rd May 1965 at Lake Nakuru, a Reef Heron, Eqretta schistacea (Ehrenberg), was seen and positively Identified by Mrs. A.J. Hanna, her son Harold and myself. Praed and Grant (1952:46) gives the range of this species in Eastern Africa as the coast of the Red Sea, the Gulf of Aden and probably Socotra Island, so it would appear that the bird which we saw was nearly a thousand miles south-west of its normal haunts. Our bird was observed for a considerable period - nearly an hour - fishing in the shallows at the north-western corner of the lake. It was very tame, allowing us excellent views through binoculars at a range of 30 to 60 yards. Luckily, a Little Egret, Eqretta qarzetta (Linnaeus), in the normal white plumage, was also fishing nearby, and its shape and size at once proved that our bird could not be a Little Egret in the grey, melanistic plumage, which was, of course, the first alternative to consider. Our bird’s plumage was uniformly dark grey, except for the chin, which was white. It had an extremely lonq. oranqe-yellow bill, neutral greeny-coloured legs and briqht. lemon-yellow feet and toes. It looked exactly similar to the coloured illustration of the'blue’ phase of the Reef Heron in Meinertzhagen (1954:399), except that the bill was more orange, the pectoral plumes more prominent and the dorsal plumes less prominent. Compared with the Little Egret already mentioned, our bird was larger and broader, with longer legs, and its yellow bill was longer and thicker than the short, slender black bill of the Little Egret. A Great White Egret, Casmerodius albus (Linnaeus), which alighted beside our bird was considerably the taller. Our bird fished in a few inches of water near the lake edge by making little dashes on foot with wings slightly raised to catch fish 2-3 ins. long, and it caught many in this way while we were watching. Unfortunately we had no long-focus camera handy, and when Mr. D.A. Turner made a special journey to Nakuru a few days later to photograph the bird, it had disappeared. Therefore, this is purely a sight- record, though I have no doubt concerning its correctness. Mr. J.G. Williams has shown me a specimen of a Reef Heron from Kenya in the National (formely Coryndon) Museum at Nairobi. The label shows that this bird was collected at Lake Rudolf in the Northern Province on 22nd March 1947, but no further information is provided. In this specimen the exposed portion of the culmen measures 104 mm as compared with only 89 mm for a typical Little Egret. The long bill was a prominent feature of the bird which we saw at Nakuru. I spent much time on the southern shore of the Gulf of Aden when stationed in ex-Italian Somaliland during the years 1941-5. Here I became familiar with Reef Herons, which looked and behaved exactly like our Nakuru bird. Since this species appears to be normally coastal, it is odd that the two Kenya records mentione'd here are both from inland. However, 231 Reef Heron at Lake Nakuru Williams (1963:25) states that there are unconfirmed reports of its presence on the northern Kenya coast. Records of Reef Herons from anywhere in East Africa south of the Gulf of Aden would be of consid- erable interest. For identification, the first feature to observe is the dark legs and contrasting yellow toes which occur only in three East African herons - the Reef Heron, the Little Egret and the Black Heron, Melanophoyx ardesiaca (Wagler), The Reef Heron has a long yellow bill; plumage usually grey, sometimes white. The Little Egret has a short black bill; plumage usually white, sometimes grey. The Black Heron has a short black bill; plumage black. We would like to thank Messrs J.G, Williams and D.A. Turner for their advice and help. References . PRAED, C.W.M. and GRANT, C.H.B. (1952). Birds of Eastern and north eastern Africa. Longmans Green, London, MEINERTZHAGEN , R. (1954). Birds of Arabia. Oliver and Boyd, Edinburgh. WILLIAMS, J.G. (1963). Field Guide to Birds of East and Central Africa. Collins, London. (Received for publication 20th September 1965) 232 J. E.Afr. Nat .Hist .Soc . Vol.XXV No. 3 (112) January 1966 A FURTHER NOTE ON REEF HERONS IN EAST AFRICA By ALEC FORBES -WATSON I have been privileged to see Mr. Myles North's article before publication. He has asked me to append the following summary of my observations. Like his, they are solely sight-records unsupported by specimens. I have seen undoubted Reef Herons at Aden and one on Socotra Island in 1964, and they were breeding alongside Little Egrets (and other herons) at Tananarive, Madagascar in October 1965 (both species were in the white phase). Apart from a group of c.20 white birds seen on a reef near an islet off the west coast of Pemba Island in November 1963, all the Reef Herons I have seen in East Africa have been of the dark grey phase, except for one paler 'lavender' bird identical to similar birds seen at Aden, and four white birds (probably of this species) seen at Kilifi. Between July 1960 and October 1962 I was the Game Warden in charge of that portion of N.W. Kenya which includes Lake Rudolf. When- ever I visited Ferguson's Gulf, about half-way up the west coast of the lake, I looked for Reef Herons, and there was usually only one to be seen somewhere near the old rest-house. All these were of a rather dark grey (but not blackish) colour, except for the 'lavender' bird mentioned above, which was a pale washed-out looking pinkish- grey; all had a whitish chin. One evening in July 1958 I was in a boat some way up Kilifi Creek with my brother, the late Mr. N.M. Forbes-Watson , and Mr. Ian Parker, when three dark and four white egret-sized herons flew down the creek towards Kilifi township. Mr. Parker was then the Game Warden at Kilifi, and he told me he had seen similar birds near Kilifi on several occasions. I have never identified a dark-phase Little Egret, and strongly suspect that these coastal birds were also Reef Herons. I cannot improve upon the field-characters already given by Mr. North. For all the birds positively identified - that is, with the exception of those seen at Kilifi - the very long bill, part of which, at least, is yellow, is the best field-character for distinguishing Reef Herons from Little Egrets. (Received for publication 20th September 1965) 233 Book Review "THE BUTTERFLIES OF MALAWI" By D. Gifford 1965. Society of Malawi. Price 50/- Shs. This small book of one hundred and thirty-six pages deals with all the five hundred odd species of butterflies recorded from Malawi in an admirably clear and concise fashion. The book is constructed as a running key which will enable Lepidopterists to identify most of the species mentioned with little trouble. All original references are given, as well as type localities, synonymy (for Malawi only), distribution and habitat in Malawi and food plants when known. Descrip- tive material has been kept down to a minimum for obvious reasons of economy, but the characters on which the key is based are adequate for the identification of the majority of species. Classification and nomenclature are in accordance with the most recent published work and there is a full bibliography which occupies the last nineteen pages of the book. There are nine plates in full colour, illustrating one hundred and forty species. As Sir Malcolm Barrow points out in his foreword, although comprehensive works have been published on the butterflies of South Africa, Rhodesia, Kenya and Uganda, no such works exist on the very rich butterfly fauna of the intervening areas and the present work will help to fill a gap of long standing in the literature. The value of the book would have been enhanced at little extra cost by the addition of a few halftone plates and of more information on the general distribution of the species. "The Butterflies of Malawi", as well as being an important contribution to our knowledge of African butterflies will be of the greatest value to the intelligent amateur. R .H.C. 234 J.E.Afr. Nat. Hist. Soc. Vol.XXV No. 3 (112) January 1966 APPRECIATION Mr. N.P. Mitton. The news of the death of Mr. N.P. Mitton in a car accident has been a sad shock to his many friends and admirers. Norman Philip Mitton was born in West Bromwich in 1916. At an early age he developed a passionate love of nature, and an absorbing interest in the techniques connected with the preservation and dis- play of Natural History objects. He was trained as a taxidermist in Birmingham, and came out to Africa in 1938, where he knew he would find greater opportunities for perfecting his techniques and more scope for his talents. His war service in the army took him to Ethiopia and East Africa during the early part of the war, and it was then that he came to know and love the East African scene which he was to portray so effectively in some of his later work. After 1942 Norman Mitton served as a gunner in the Middle East and in Italy until the end of the war, and then returned to Africa in 1945. He was employed by the Transvaal Museum as Taxidermist for a few years, but seized the first opportunity to return to his beloved East Africa, when he was offered a post on the staff of the Coryndon Museum by Dr. L.S.B. Leakey, in 1950. At that time, thanks to the enthusiasm and dedication of Dr. Leakey, funds had been raised for extensions to the old Museum which more than doubled its original exhibition space. Mr. Allen Turner, a man of advancing years and failing health, had up to then been responsible for the preparation and display of all exhibition material. The task of preparing new exhibits for the large new empty galleries was clearly too much for one man, and Norman Mitton assisted him at first and took over complete charge of the work a little later, when Allen Turner died. Norman Mitton met the formidable challenge of re-organising the entire Museum with hard work, selfless dedication, unerring taste and with an uncanny grasp of technical problems. Step by step, despite a perennial shortage of finance, staff and equipment, the Coryndon Museum became one of the finest in Africa and a source of pride and prestige to Kenya. Much of what the visitor of today admires in what is now the National Museum is the work of Norman Mitton, and a permanent tribute to his great talents; his Natural Habitat Groups are outstanding, and the perfection of his life-like casts of fish and reptiles unsur- passed; indeed the entire Museum bears the stamp of his genius. Not one to rest on his laurels, he was never satisfied with his achievements, and forever experimented with improved techniques and more satisfying display methods; in fact he was a great artist and a great all round naturalist, as well as a brilliant technician. Norman Mitton was an exceptionally well read man with a wide range of interests covering literature, the theatre and visual arts as well as current scientific and technical developments. His out- standing talents and achievements, his unfailing generosity and attractive personality, earned him universal respect and popularity, and his untimely death will be a great personal loss to his many 235 Appreciation friends, whose deepest sympathy goes out to his widow and children. Norman Mitton's services to Kenya were outstanding and his loss to our National Museum is one that can never be made good. R.H.C. 236 EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM Notice to Contributors Contributions. The Committee is pleased to consider contributions on natural history for publication in the Journal on the understanding that these are not also being offered, wholly or partially, to any other Journal. They should be addressed to the Secretary, P.O. Box 4486, Nairobi. Typescript. Articles should be typed on one side of the paper, in double spacing and with wide margins. As this Journal is printed by the photo lithographic process, alterations and additions to manuscripts which have reached the proof stage can only be accepted at the discretion of the Editor. Illustrations. These should be in a form suitable for reproduc- tion. The Editor cannot be expected to re-draw. Line drawings should be in Indian ink on Bristol board or thick white paper. Reproduction will be better if they are drawn larger than it is intended that they should appear. An indication of the degree of reduction is advisable. Photographs should be printed on glossy paper and a better reproduc- tion is achieved from prints slightly darker in tone than normal. Nomenclature. Where a recent standard work for the area is avail- able (e.g. Praed and Grant for birds) the names given there (both English and scientific) should be used. Initial capitals should be used for specific English names, e.g. Pied Wagtail and small initial letters for group names, e.g. wagtails. Scientific names must always be underlined and followed by the naming authority when mentioned for the first time. Where an English name is used, it must, on first mention be accompanied by the scientific name. References. These are usually abbreviated in the text and listed more fully in alphabetical order of author? at the end of the article. For example, in the text a book reference might be (Pinhey 1956: p.20K At the bottom of the contribution: Jackson, F.J., 1938. Birds of Kenya and Uganda. Pinhey, E.C.G., 1956. The Emperor Moths of Eastern Africa. J. E .Afr .Nat .Hist .Soc . XXIII No. 1. (9S). With short articles it may not be worth making a list of references at the end, but the whole reference in the most abbreviated comprehensible form should then be inserted in the text. Reprints. Provided that they order at the time of submitting their articles, authors may have up to 25 reprints free (other than Nature Notes). Additional copies can be supplied on payment. A list of available back numbers of the Journal and reprints can be obtained from the Librarian, The National Museum, P.O. Box 658, Nairobi. EAST AFRICA BY AERAD LITHOGRAPHIC PRINTERS JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. XXVI No. 1 (113) JUNE 1966 CONTENTS Page Cyperaceae of East Africa — IV 1 By D. M. Napper Notes on the Birds of Lengetia Farm, Man Narok 18 By P. H. B. Sessions Rhinoptera javanica Muller & Henie from Kenya Waters (Piscess: Rhinopteridae) 49 By G. F. Losse Prophylls and Branching in Cyi»eraceae 51 By R. Wheeler Haines Book Reviews 71 (Published 7/6/1967) Price Shs. 20/- EAST AFRICA NATURAL HISTORY SOCIETY NOTICE TO CONTRIBUTORS Contributions. The Committee is pleased to consider contributions on natural history for publication in the Journal on the understanding that these are not also being offered, wholly or partially, to any other Journal. They should be addressed to the Secretary, P.O. Box 4486, Nairobi. Typescript. Articles should be typed on one side of the paper, in double spacing and with wide margins. Illustrations. These should be in a form suitable for reproduction. The Editor cannot be expected to re-draw. Line drawing should be in Indian ink on Bristol board or thick white paper. Reproduction will be better if they are drawn larger than it is intended than they should appear. An indication of the degree of reduction is advisable. Photographs should be printed on glossy paper and a better reproduction is achieved from prints slightly darker in tone than normal. Nomenclature. Where a recent standard work for the area is available (e.g. Praed and Grant for birds) the names given there (both English and scientific) should be used. Initial capitals should be used for specific English names, e.g. Pied Wagtail and small initial letters for group names, e.g. wagtails. Scientific names must always be underlined and followed by the naming authority when mentioned for the first time. Where an English name is used, it must, on first mention be accompanied by the scientific name. References. These are usually abbreviated in the text and listed more fully in alpha- betical order of authors at the end of the article. For example, in the text book reference might be (Pinhey 1956: p. 20). At the bottom of the contribution: Jackson, F. J., 1938. Birds of Kenya and Uganda. Pinhey, E. C. G., 1956. The Emperor Moths of Eastern Africa. J.E. Afr. Nat. Hist. Soc. XXIII No. 1, (98). With short articles it may not be worth making a list of references at the end, but the whole reference in the most abbreviated comprehen- sible form should then be inserted in the text. Reprints. Provided that they order at the time of submitting their articles, authors may have up to 25 reprints free (other than Nature Notes). Additional copies can be supplied on payment. A list of available back numbers of the Journal and reprints can be obtained from the Librarian, National Museum, P.O. Box 658, Nairobi. JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. XXVI No. 1 (113) JUNE 1966 CONTENTS Page Cyperaceae of East Africa — IV 1 By D. M. Napper Notes on the Birds of Lengetia Farm, Mau Narok 18 By P. H. B. Sessions Rhinopiera javanica Muller & Henle from Kenya Waters (Piscess: Rhinopteridae) 49 By G. F. Losse Prophylls and Branching in Cyperaceae 51 By R. Wheeler Haines Book Reviews 71 (Published 7/6/1967) Price Shs. 20/- EAST AFRICA NATURAL HISTORY SOCIETY President: DR. M. J. COE Vice-President: DR. A. D. Q. AGNEW Executive Committee: L. H. BROWN, ESQ. M. E. W. NORTH, ESQ. R. H. CARCASSON, ESQ. MISS E. J. BLENCOWE MRS. D. FLEMING B. PARSONS, ESQ. J. SMART, ESQ. P. M. OLINDO, ESQ. C. J. BEECHER, ESQ. DR. J. 3. FOSTER J. S. KARMALI, ESQ. Hon. Editor: DR. P. J. GREENWAY Hon. Treasurer: MISS J. OSSENT Secretary: MRS. F. NG’WENO All correspondence in connection with this Journal should be addressed to The Secretary, East Africa Natural History Society, P.O. Box 4486, Nairobi, Kenya. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 1 CYPERACEAE OF EAST AFRICA — IV By D. M. Nappep CYPERUS L. Cyperus is without doubt the largest genus ot sedges occurring in the tropics. Even in the narrower interpretations of the genus such as the one used here several hundred species of the tropical and subtropical regions are included, many of which are pantropical. In East Africa Cyperus species are found in damp places and in water, but some also flourish in the short rainy season of the near desert areas; few have been recorded above 8,000 ft. except in the forest undergrowth, though from the sea shore to about 7,000 ft. they are abundant in all suitable habitats. Though mostly leafy perennials of moderate size ranging from about 1 to 4 ft. high, a few species, including C. papyrus, are conspicuous for their great size, while others are conspicuous for the absence of foliage leaves, or their small very ephemeral or bulbous habit. The variety of form of rootstock and rhizome in the genus is striking and should be noted when material is being gathered, also, for rhizomatous or stoloniferous specimens, it should be noted whether the internodes are short or long and whether the stolons (where present) bear tubers or tunicate bulbs. Not all species with tunicate bulbs have the aerial stem and leaves arising directly from them; many have a short stolon growing out of the bulb which gives rise to the plant. Elongated pseudobulbs such as occur so commonly in Mariscus are not seen in Cyperus and help to separate these two genera at a point where inflorescence differences are minimal. The inflorescence may be subtended by few or many leafy bracts, rarely there is only one looking like a continuation of the stem with the inflorescence apparently borne laterally. Though usually spreading and consisting of a simple or compound umbel of spikes there are a few species where the inflorescence is contracted into a simple head and the number of spikelets is correspondingly reduced. An exception to this is C. pulchellus in which the inflorescence closely resembles a Kyllinga, but the distichous and flattened spikelets with opposite instead of spirally arranged glumes as in the terete Kyllinga spikelets serve to separate them easily. The spikes of which the inflorescences are built up may be elongate with a rhachis much longer than the spikelets or digitate with the spikelets clustered at the tips of the peduncle. The spikelets themselves are few to many-flowered and normally comprise a basal pair of empty glumes with a series of distichously arranged fertile glumes above them in the axil of Page 2 Cyperaceae of East Africa — IV each of which are the 3 stamens (sometimes only 1-2) with oblong or linear, rarely crested, anthers and an ovary which passes gradually into the simple style with a 3-fid stigma. In a few anomalous species the style is frequently 2-fid and the nutlet lenticular as in Pycreus, but such anomalous species may be distinguished by the orientation of the nutlet; in Cyperus it is dorso-ventrally compressed (the anterior angle of a normally trigonous nutlet is fiattened and the flat side faces the rhachis) and in Pycreus it is laterally compressed (the edge of the nutlet is nearest the rhachis). The mature nutlets show little diversity of form or ornamentation, being for the most part smooth and ellipsoid-cylindric, with a triangular section. In a very few species the surface is warted or undulate (transversely waved) and some have a lenticular, not triangular, section as mentioned above. The species designated as comprising Juncellus in the Flora of Tropical Africa vol. 8, and for which provision was made in the key to species in the first paper in the series, have here been included in Cyperus. Examination showed that most of them do not consistantly have 2-fid stigmas, the main point of deviation between the genera, and that there is less similarity between the species so grouped together than there is between each and the species of Cyperus with which they would be associated. Certain marked omissions will be obvious on a comparison between the species given here and the account in the Flora of Tropical Africa. The affi.nity of these species is with Mariscus rather than Cyperus and they will be included in the next, and final, paper dealing with Pycreus, Kyllinga and Mariscus (including Courtoisia). Key to Species 1. Inflorescence a solitary dense head (readily confused with Mariscus and Kyllinga) 77 Inflorescence spreading, branched, simple or compound (rarely a solitary spike) . 2 2. Inflorescence of one or more elongate spikes having a distinct, and often very long, rhachis 3 Inflorescence of short spikes of clustered or digitately arranged spikelets, rhachis scarcely discernable or absent 49 3. Plants leafless, or with few, very short leaves 4 Plants leafy 8 4. Stems terete, articulated 5 Stems not articulated 6 5. Bracts almost as long as the inflorescences, with a few indistinct articulations 11. C. corymbosus Bracts very short 10. C. articulatus 6. Spikes shortly cylindric with numerous small spikelets . . . 1. C. papyrus Spikes broadly ovate 7 /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 3 1. Stem-bases coated with fibres; glumes 2.5-3 mm long . . 12. C. schimperianus Stem-bases not surrounded by fibres; glumes 4-5 mm long . . . 9. C. holstii 8. Spikes long cylindric with numerous spikelets 9 Spikes shorter, broadly ovate with fewer spikelets or irregular in outline . . 17 9. Spikelets terete 10 Spikelets more or less compressed 11 10. Culms stout triangular not winged glumes obtuse 2. C. grandis Culms more slender, triangular with winged angles; glumes subacute 3. C. digitatus var. auricomus 11. Spikelets rather turgid, inflated 8. C. alopecuroides Spikelets flat 12 12. Spikelets distant 4. C. exaltatus Spikelets crowded together 13 13. Spikes all sessile or subsessile 14 Spikes peduncled and sessile together 16 14. Glumes very small, scarcely over 1 mm long, acuminate . . 7. C. imbricatus Glumes 1. 5-2.5 mm long, obtuse or acute 15 15. Glumes obtuse, golden, 1.5-2 mm long 36. C. iria Glumes acuminate, green or brown, 2-2.5 mm long . . .35. C. eleusinoides 16. Very robust plants with leaves over 20 mm wide; rays up to 10 ins. long 5. C. immensus Stout plants with leaves less than 20 mm wide; rays up to 4 ins. long . . 6. C. dives 17. Spikelets quadrangular in section or subquadrangular 18 Spikelets terete or compressed 22 18. Annual; glumes 3-4 mm long 32. C. zollingeri Perennials 19 19. Glumes dark grey-green 11. C. marangucnsis Glumes orange to dark red on the sides 20 20. All rays long, more or less equal 9. C. holstii Rays unequal in length, some very short 21 21. Stems scabrid above 31. C. schweinfurthianus Stems smooth throughout 30. C. tenuiculmis 22. Annual 28. C. sphacelatus Perennials 23 23. Stems or stolens bearing bulbs 24 Bulbs absent, rhizome and stolons often tuberous 30 24. Stems distant from the bulbs, bulbs underground 25 Stems arising directly out of the bulbs which are often visible at ground level . 27 25. Bulbs large, 8-20 mm diam. at maturity; spikelets well spaced . . 22. C. usitatus var., macrobulbus Bulbs rarely over 8 mm diam.; spikelets densely spicate 26 26. Bracts 2-3; bulbs brown 22. C. usitatus var. usitatus Bracts 4-8; bulbs black 23. C. stuhlmannii 27. Stems slender; bracts very short; spike solitary . . . . 26. C. blysmoides Stems stouter; bracts 3-6, at least as long as the inflorescence; spikes several . . 28 28. Spikelets dark red, 10-20 mm long; rays up to 5 mm long . . 25. C. bulbosus var., melanolepis Spikelets tawny, up to 10 mm long; rays short or absent 29 Page 4 Cyperaceae of East Africa — IV 29. Spikes contracted almost into a head; rays scarcely present ... 24. C. grandibulbosus var. grandibulbosus At least some of the spikes on well developed rays 24. C. grandibulbosus var. amplus 30. Stem-bases woody, tuberous, stolons usually stout; leaves never over 20 mm wide . 31 Stem-bases usually woody but not tuberous; leaves of some species over 20 mm wide 37 31. Nutlets strongly transversely ridged 14. C. undulatus Nutlets smooth 32 32. Glumes whitish, 1-1.5 mm long 16. C. maculatus Glumes white, red, dark chestnut or black, 1.5-3 mm long 33 33. Glumes whitish or red, up to 2.5 mm long 34 Glumes black, rarely reddish, up to 3 mm long 36 34. Glumes 1.5-2 mm long, white with dark patches on the sides . . 15. C. longus var. tenuiflorus Glumes 2-2.5 mm long; spikelets 1-1.5 mm wide 35 35. Spikelets reddish 15. C. longus var. longus Spikelets whitish 15. C. longus var. pallidus 36. Spikelets erect, crowded 17. C. rigidifolius Spikelets spreading, distant 18. C. kilimandscharicus 37. Stolons present, usually long and slender, bearing tubers (on young plants look carefully for early signs of stolon development) 38 Stolons absent, plants without tubers 42 38. Spikelets turgid tumid or subcompressed 40 Spikelets strongly compressed 39 39. Plants slender: spikelets reddish brown, 1-2 mm wide . . 19. C. rotundus Plants more robust; spikelets black, 2.5 mm wide . . . . 20. C. merkeri 40. Spikelets up to 2 mm wide 41 Spikelets 2-2.5 mm wide 19. C. rotundus ssp. tuberosus 41. Spikelets red 19. C. rotundus Spikelets orange 21. C. esculentus 42. Spikelets terete; glumes rounded on the back 43 Spikelets compressed; glumes mostly keeled on the back 44 43. Snikelets very numerous, crowded; leaves over 25 mm wide . . 13. C. latifolius Spikelets few. distant; leaves under 12 mm wide . . . . 29. C. gracilinux 44. Spikelets 3-5 mm wide 45 Spikelets up to 2 mm wide . 46 45. Glumes oale throuoghout; spikelets compressed . . . . 37. C. compressus Glumes dark red on the sides; spikelets tumid . . . . 38. C. pustulatus 46. Snikelets less than 1 mm wide 33. C. distans Spikelets over 1.5 mm wide 47 47. G'umes eolden or yellow, obtuse 36. C. iria Glumes black or brownish 48 48. Glumes dark red with broad green keels 34. C. aterrimus Glumes light reddish brown 35. C. eleusinoides 49. Perennials usually stout, with woodv rhizomes and numerous leaves ... 50 Annuals or slender-stemmed or leafless perennials 60 50. Snikelets large, 10-20 mm long 51 Spikelets up to 10 mm long 52 51. Snikelets subterete, 2.5-3 mm wide, pale or brownish ... 62. C. maritimus Spikelets compressed, 3-5 mm wide, red 63. C. frerei E. Afr. nat. Hist. Soc. VoL XXVI No. 1 (113) Page 5 52. Spikelets sharply keeled; glumes veined on the back only; bracts very long and very numerous 39. C. alternifolius ssp. ftabelliformis Spikelets often somewhat turgid; glumes never sharply keeled, veined on both back and sides 53 53. Glumes ovate when mature, up to 1.5 mm long 54 Glumes oblong 55 54. Culms up to 6 ft. high; spikelets 2-4 mm long 46. C. renschii Culms up to 2 ft. high; spikelets 6-7 mm long; nutlets conspicuously protruding 45. C. diffusus var. buchholzii 55. Glumes acute, conspicuously mucronate 56 Glumes obtuse or minutely mucronate 59 56. Glumes up to 1.5 mm long; leaves 1-4 mm wide .... 40. C. ajax Glumes 2 mm long; leaves up to 2 cm wide 57 57. Spikelets plae greeny buff 43. C. glaucophyllus Spikelets chestnut to dark red 58 58. Spikelets lanceolate . . .42. C. pseudoleptocladus var. pseudoleptocladus Spikelets linear 42. C. pseudoleptocladus var. polycarpus 59. Leaves up to 1 in. wide; glumes closely overlapping, 2-2.5 mm long, minutely mucronate or emucronate 41. C. dereilema Leaves up to i in. wide but usually less; glumes distant; 1.5-2 mm long, obtuse 44. C. fischerianus 60. Glumes obtuse, more or less furrowed, membranous; leaf-blades flat or absent . 61 Glumes truncate or subacute and mucronate but never obtuse, tougher; leaves narrow and channelled 74 61. Stems very stout, 2-6 mm wide; plants usually leafless 62 Stems narrower, up to 2 mm wide; leafy annuals or perennials or leafless perennials with stems not over 2 mm wide 65 62. Rays very numerous, equal 58. C. prolifer Rays usually fewer, unequal . 53 63. Spikelets linear, light brown or reddish, 8-10 mm long at maturity 57. C. denudatus Spikelets narrowly lanceolate, light to dark red, up to 8 mm long at maturity . . 64 64. Spikelets reddish 56. C. platycaulis var. platycaulis Spikelets shining dark blackish red . . . 56. C. platycaulis var. lucenti-nigricans 65. Glumes dark; perennials 66 Glumes greenish tinged with orange, red or yellow; annuals or slender perennials 67 66. Plants very leafy; spikelets 2.5-4 mm long; glumes black with pale green keels 48. C. dichroostachyus Leaves few or absent; spikelets 6-8 mm long; glumes dark red 55. C. phceorrhizus 67. Glumes with recurving mucros, often very short or with recurving tips ... 68 Glumes emucronate or with a straight mucro 70 68. Leaves 8-10 mm wide 53. C. foliaceus Leaves up to 4 mm wide .69 69. Spikelets 3-4 mm wide; glumes spreading at maturity ... 50. C. reduncus Spikelets 1-2 mm wide 52. C.' tenuispica 70. Spikelets 3-5 mm wide Spikelets 1-2 mm wide ' . \ 12 71. Glumes pale throughout; spikelets compressed . . 37. C. compressus Glumes dark red on the sides; spikelets tumid . . 38. C. pustulatus 72. Glumes shortly mucronate . . 54. C. haspan Glumes emucronate ’ / ' . 73 Cyperaceae of East Africa — IV 73. Inflorescence up to 5 mm diam 51. C. suhmicrolepis Inflorescence 6-12 mm diam. 49. C. difformis 74. Slender annuals .75 Woody-based perennials 76 75. Glumes excurrent into a long recurved mucro . . . . 60. C. cuspidatus Glumes with a short straight mucro 61. C. amabilis 76. Spikelets 6-12 mm long; glumes not widely spreadying . 59. C. tenax var. tenax Spikelets up to 20 mm long; glumes larger, more distant and widely spreading . . .59. C. tenax var. pseudocastaneus 77. Spikelets turgid; glumes rounded on the back 78 Spikelets compressed; glumes keeled 80 78. Plants leafless . 65. C. laevigatas Plants leafy 79 79. Rhizome stout; leaves 3-6 mm wide 62. C. maritimus Rhizomes slender; leaves narrow 64. C. chordorrhizus 80. Perennials; glumes obtuse or shortly mucronate 81 Annuals, sometimes quite tufted 90 81. Stem-bases swollen, tuberous or bulbous; plants leafy 83 Stem-bases not swollen; plants leafless 82 82. Stems acutely angled, 2-4 mm wide 70. C. colymbetes Stems slender, angles not acute 71. C. nudicaulis 83. Spikelets up to 8 mm long; plants slender 66. C. pulchellus Spikelets over 8 mm long; plants stouter . 84 84. Glumes pinkish brown; maritime shores 63. C. frerei Glumes white, yellow or red 85 85. Rhizome horizontal with long internodes; glumes acute . . 67. C. angolensis Rhizome descending or horizontal with short internodes, or absent; glumes obtuse 86 86. Spikelets 2.5-4 mm wide 88 Spikelets 4-10 mm wide .87 87. Spikelets 3-7, subcompressed; a rare species .... 68. C. margaritaceus Spikelets 5-20 or more, compressed; a common, very variable species 69. C. obtusifloms 88. Glumes whitish or green 47. C. mapanioides Glumes dark red 89 89. Bracts 2-3; bulbs brown 22. C. usitatus Bracts 4-8; bulbs black 23. C. stuhlmannii 90. Glumes emucronate 32. C. zollingeri Glumes conspicuously mucronate 91 91. Bracts 2.5-4 ins. long, abruptly widening at the base ... 75. C. pyginaeiis Bracts shorter and not abruptly widened below . 92 92. Basal sheaths entire, membranous 93 Basal sheaths becoming fibrous 72. C. bellus 93. Spikelets 4-5 mm wide 74. C. teneriffae Spikelets up to 3 mm wide .94 94. Very slender plants with spikelets up to 1.5 mm wide ... 60. C. cuspidatus Slender plants slightly swollen at the base; spikelets wider . . 73. C. kaessneri i. 1E. Afr. nat. Hist. Soc. Vot. XXVI No. 1 (ll3) Page 7 1. C. papyrus L. Leafless, stout-stemmed, rhizomatous perennial up to 18 ft. high. Inflorescence compound, spreading, with 1-2 cm long spikes of linear, narrow 6-10 mm long spikelets with green-keeled glumes. Rivers, dams, lakes, swamps, but always in water; 1,500-7,000 ft. KENYA— Widespread. TANGANYIKA— Widespread. UGANDA— Widespread. 2. C. grandis C.B.Cl. Stout perennial up to 6 ft. high with leaves i-li in. wide. Inflorescence large with numerous spikelets 8-18 mm long and over 1 mm wide in cylindric spikes 2-5 cm long; glumes ovate, obtuse. Swamps, sea level — 500 ft. KENYA— Coast. TANGANYIKA— Coast. ZANZIBAR — Zanzibar and Pemba Islands. 3. C. digitatus Roxb. ssp. auricomus (Spreng.) Kukenthal (C. auricomus Sieber ex Spreng.) Tufted perennial 1^-5 ft. high with numerous leaves i-i in. wide. Inflorescence large and spreading with numerous golden or reddish linear spikelets 10-20 mm long in loose cylindric spikes 2.5-6.5 cm long. River banks, seasonal and permanent swamps; 1,500-7,500 ft. KENYA — Western, Rift Valley, Central Regions and Nairobi. TANGANYIKA — Widespread in suitable places. UGANDA — ^Westem and Eastern Provinces. 4. C. exaltatus Retz. Stout tufted perennial 1-4 ft. high with leaves up to \ in. wide. Inflorescence large and compound with numerous distant compressed golden spikelets 4-10 mm long in cylindric spikes. Glumes ovate, obtuse, with a recurved mucro. In shallow water at the edge of dams, pools, lakes and rivers; sea level — 6,000 ft. KENYA — The coast and Nairobi. TANGANYIKA — Northern, Tanga Regions and the coast. ZANZIBAR — Zanzibar Island. 5. C. immensus C.B.Cl. Large tufted perennial up to 8 ft. high with leaves i-l^ ins. wide. Inflorescence very large with numerous linear-lanceolate compressed spikelets 6-15 mm long in sessile or subsessile dense, broadly cylindric spikes. Glumes with a rigid conspicuous mucro and green keel. Rivers, swamps, dams and seasonally flooded places; 200-8,000 ft. KENYA— Widespread. TANGANYIKA— Widespread. UGANDA — Buganda, and probably throughout the country. ZANZIBAR— Zanzibar Island. 6. C. dives Del. (C. exaltatus Retz. var. dives (Del.) C.B.Cl.) Large tufted perennial up to 5 ft. high with leaves up to i in. wide. Inflorescence large, with numerous compressed spikelets 4-8 mm long in dense cylindric spikes. Glumes with a short straight mucro. Differs from C. immensus chiefly in the smaller size, and from C. imbricatus by some or all of the spikes being pedunculate. Swamps; 2,000-6,000 ft. KENYA— Widespread. TANGANYIKA — Northern and Tanga Regions. UGANDA — Western Province and Buganda. 7. C. imbricatus Retz. (C. radiatus Vahl) Stout tufted perennial 1^-4 ft. high with narrow leaves. Inflorescence of numeorus oblong-lanceolate compressed spikelets 3-7 mm long in dense sessile spikes. Glumes small, obtuse, greenish, with a short recurving mucro. Wet sandy places and river banks, uncommon; sea level — 4,000 ft. TANGANYIKA — ^Western and Southern Highland Regions and along the Rufiji River. UGANDA — Western Province. Page 8 Cyperaceae of East Africa — iV 8. C. alopecuroides Rottb. (Juncellus alopecuroides (Rottb.) C.B.Cl.) Stout leafy perennial up to 5 ft. high. Inflorescence large with numerous lanceolate, acute, somewhat turgid spikelets 4-8 mm long in oblong cylindric spikes. Glumes 2-2.5 mm long, rounded on the back, tawny or greenish. Swamps, dams, rivers, usually in standing water; sea level — 5,000 ft. KENYA — ^Widespread but not common. TANGANYIKA — Widespread in the northern part of the country. 9. C. holstii Kukenthal (C. zollingeri var. robusta K. Schum.) Tufted plant with long stout stolons, stout stems, and the leaves reduced to the sheaths only. Inflorescence large with numerous quadrangular tawny or light chestnut spikelets 15-40 mm long in subdigitate spikes of 3-8 spikelets. Glumes obtuse, 4-5 mm long. Damp grassland and swamps; sea level — 800 ft. KENYA— Coast. TANGANYIKA— Coast. The spikelets have a marked resemblance to C. tenuiculmis but the long and more numerous rays give the inflorescence a very different appearance. 10. C. articulatus L. Leafless perennial with distant septate terete stems 2-6 ft. high. Inflorescence spreading with compressed spikelets 8-30 mm long. Glumes reddish, obtuse, green keeled. Dams, swamps, lakes; sea level — 5,500 ft. KENYA— Widespread. TANGANYIKA-Widespread, and Mafia Island. UGANDA— Widespread. ZANZIBAR — Zanzibar Island. 11. C. corymbosus Rottb. Very similar to C. articulatus but differs in the obscurely septate stems and the 2-4 scabrid-margined leaf-like bracts almost as long as the inflorescence. Rare in dams, swamps and lakes; 1,000-2,000 ft. TANGANYIKA— Tanga Region. 12. C. schimperianus Steud. Perennial 2-3 ft. high with trigonous or subterete stems surrounded by fibres at the base. Inflorescence with linear-oblong spikelets 8-18 mm long. Glumes distant, 2.5-3 mm long, rounded. Damp places, stream banks; 3,000-5,000 ft. KENYA— Nairobi. TANGANYIKA — Northern, Western and Tanga Regions. UGANDA — Northern Province. 13. C. latifolius Poir. Tufted perennial 1-6 ft. high. Leaves over 1 in. wide. Inflorescence large with broadly ovate dense spikes of 10-20 mm long spikelets. Glumes brown with conspicuous white margins. Swamps and wet places; 1,000-7,000 ft. KENYA — Western, Central and Eastern Regions and Nairobi. TANGANYIKA — Northern, Tanga, Western and Southern Highland Regions. UGANDA — Western Province and Buganda. 14. C. undulatus Kukenthal Stoloniferous perennial up to 2 ft. high with leaves i-y in. wide. Inflorescence spreading with broadly ovate dense spikes. Spikelets spreading, linear, 10-25 mm long with obtuse, tawny glumes with white margins and green keels. Nutlets conspicuously transversely wrinkled. Damp and seasonally inundated places in dry bush and grassland; 1,000-3,500 ft. KENYA — East of the Rift Valley, not common. TANGANYIKA — Northern, Tanga and Eastern Regions. 15. C. longus L. var. longus Tufted rhizomatous perennial up to 3 ft. high with scabrid-margined leaves. Inflorescence spreading with distant, suberect bright chestnut linear spikelets 10-25 mm long, up to 2 m.m wide. Glumes 2-2.5 mm long, obtuse, with broad white margins. Seasonally inundated grassland; 1,000-4,000 ft. KENYA — Northern Region. TANGANYIKA — Tanga and Western Regions j. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 9 var. pallidus Boeck. This variety differs from the above only in the pale, off-white spikelets. Seasonally inundated places, very uncommon; 2,000-4,000 ft. TANGANYIKA— Western Region. UGANDA — Eastern Province. var. tenuifiorus (Rottb.) Boeck. Plant more slender than the above and rarely over 1-1^ ft. high. Spikelets linear, 10-25 mm long but only 1 mm wide, whitish but with chestnut patches on the sides and wide white margins. Near water, lake shores and river banks; 500-4,000 ft. KENYA — Western Region. TANGANYIKA — Lake, Northern, Western Regions and the Coast. UGANDA — Western Province and Buganda. 16. C. maculatus Boeck. Tufted perennial 1-H ft. high very similar to C. longus var. tenuifiorus but with much narrower leaves. Spikelets similar but shorter, 1 mm wide and lacking the dark patches of the above. Damp places and lake shores; 2,000-4,000 ft. TANGANYIKA— Western Region. UGANDA — Western Province. 17. C. rigidifolius Steud. Stoloniferous perennial 4-12 ins. high with a stout woody rhizome. Inflorescence with numerous oblong-lanceolate spikelets 6-10 mm long, 2 mm wide, crowded into very dense ovate spikes, but often very reduced. Glumes dark red or black, green keeled. Damp places in grassland and near water; 4,000-11,000 ft. KENYA— Widespread. TANGANYIKA — Widespread, but less common towards the south. UGANDA— Widespread. There is considerable variation in the reduced forms normally included in this species and it is possible that a few of these should be ascribed to the rather similar C. fissus. 18. C. kilimandscharicus Kukenthal Rhizomatous perennial 2-3 ft. high with tuberous-swollen culm-bases coated with fibrous sheath remnants. Inflorescence rather dense with oblong-lanceolate black spikelets 5-7 mm long, 2-2.5 mm wide, with shortly mucronate green keeled glumes 2.5-3 mm long. Seasonally swampy places, usually on black cotton soils; 3,500-5,500 ft. KENYA — Central Region and Nairobi. TANGANYIKA — ^Lake, Northern and Central Regions. 19. C. rotundus L. Very variable perennial 6 ins.-2 ft. high with slender stolons bearing tubers. Inflorescence spreading with linear-lanceolate compressed or terete spikelets 10-20 mm long, 2-2.5 mm wide in short ovate spikes. Glumes ovate, acute or subobtuse, pale or dark red usually with a green keel and white margins. Grassland, swamps, damp places; sea level — 6,000 ft. KENYA — Widespread but not very abundant. TANGANYIKA — Lake, Northern, Tanga, Western and Eastern Regions. UGANDA— Widespread. ZANZIBAR— Zanzibar and Pemba Islands. Several subspecies and varieties have been described but the interpretation and synonymy of these is so complex that no attempt will be made here to distinguish them. 20. C. merkeri C.B.Cl. (C. rotundus L. var, spadiceus Boeck., C. rotundus ssp. merkeri (C.B.Cl.) Kukenthal). Tufted perennial li-2 ft. high. Stems often scarbrid. Inflorescence with strongly compressed spikelets 15-20 mm long, 2.5 mm wide, with dark red shining green-keeled glumes. Damp places in woodland and grassland; sea level — 6,500 ft. KENYA— Widespread. TANGANYIKA— Widespread. UGANDA — Western and Northern Provinces. ZANZIBAR — Zanzibar Island. 21. C. esculentus L. Perennial 4 ins.-ly ft. high with slender tuber-bearing stolons. Inflorescence spreading, with ovate spikes of golden or tawny liner-oblong obtuse spikelets 5-12 mm long. Glumes Page 10 Cyperaceae of East Africa — IV truncate, obtuse, rounded on the back. Seasonally swampy places, ditches and as a weed; sea level — 7,000 ft. KENYA — Widespread above about 4,500 ft. TANGANYIKA— Widespread. UGANDA — Western, Eastern Provinces and Buganda. ZANZIBAR — Zanzibar Island. 22. C. usitatus Burch, var. usitatus Small, single-stemmed plants up to 8 ins. high developed away from the bulb to which they are attached by a short slender stolon. Bulbs brown, shiny. Inflorescence simple, dense, 15-30 mm diam. with linear-lanceolate acute shining red spikelets 6-15 mm long, 2 mm wide. Glumes ovate, shortly mucronate. Grasslands, seasonally swampy places and as a weed; 3.500- 6,000 ft. KENYA— Rift Valley. TANGANYIKA — Lake, Northern, Western and Central Regions. UGANDA — Karamoja. var. macrobulbus Kukenthal Similar to the above except for its greater size, larger bulb etc. It is easily confused with the following species. TANGANYIKA— Central Region. 23. C. stuhlmannii C.B.CI. Small plants up to 1 ft. high developed apart from the black 6-8 mm diam. bulbs. Inflorescence more ample than that of the above species, 12-18 mm diam. with linear- lanceolate acute spikelets 6-8 mm long, 2-2.5 mm wide. Glumes dark red subobtuse or acute. Grasslands and seasonally swampy places; 2,500-5,500 ft. KENYA — Western Region and the Rift Valley. TANGANYIKA— Lake Region. 24. C. grandibulbosus C.B.CI. var grandibulbosus Perennial up to H ft. hgh with a solitary stem arising from a black 6-10 mm diam. bulb. Inflorescence contracted, with tawny-orange linear spikelets 8-10 mm long, 3 mm wide in a dense head. Sandy soils in seasonably damp places and on rocky outcrops; 1.500- 4,500 ft. KENYA — Northeastern, Eastern and Central Regions. TANGANYIKA — Northern and Tanga Regions. var. amplus Kukenthal Plants larger than the above and the inflorescence spreading. Sandy soils in seasonally damp places; 1,500-4,500 ft. KENYA — Northeastern and Eastern Regions. 25. C. bulbosus Vahl var. melanolepis Kukenthal Perennial up to li ft. high with a single stem arising out of a black or very dark brown bulb 6-10 mm wide. Inflorescence spreading with distant compressed spikelets 10-20 mm long. 1.5 mm wide, with dark red glumes. The inflorescence of this species is readily confused with C. merkeri, but the habit is very different. Damp places in dry grassland; 2,500-5,000 ft. KENYA — Southern Region. TANGANYIKA— Northern and Tanga Regions. 26. C. blysmoides C.B.CI. (C. bulbosus Vahl var. spicatus Boeck., C. blysmoides Hochst. nom. nud.) Stems 4-12 ins. high arising directly out of a black bulb. Inflorescence a single spike without bracts, of 3-6 linear spikelets 8-12 mm long. Glumes shortly mucronate, dark red or pale. Weed, usually in damp places; 1,500-6,000 ft. KENYA — Northern, Rift Valley, Central and Eastern Regions and Nairobi. TANGANYIKA — Northern and Tanga Regions. 27. C. maranguensis K. Schum. Tufted perennial up to 3 ft. high. Inflorescence of broadly ovate spikes of dark grey- green linear spikelets 15-20 mm long, 1.5 mm wide. Glumes elliptic obtuse. Upland and mountain forests; 2,000-7,000 ft. KENYA — Northern and Central Regions and Nairobi. TANGANYIKA— Northern, Tanga and Eastern Regions. UGANDA— Buganda. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 11 28. C. sphacelatus Rottb. Tufted annual up to 2 ft. high. Inflorescence rather scanty with distant linear-lanceolate acute greyish-yellow spikelets 6-20 mm long, 1.5 mm wide. Glumes obtuse, green-keeled with white margins. Woodland and damp places; 1,000-4,000 ft. TANGANYIKA — Lake, Western and Southern Regions. UGANDA — Buganda. 29. C. gmcilinux C.B.Cl. Stoloniferous perennial 9 ins.-l-^ ft. high. Inflorescence rather scanty with spreading linear-lanceolate acute spikelets 16-24 mm long, 2 mm wide. Glumes subobtuse, reddish with white hyaline margins and green keels. Seasonally swampy places, vlei; sea level — 2,000 ft. TANGANYIKA— Tanga Region. 30. C. tenuiculmis Boeck. fC. zollingeri non Steud.) Tufted perennial 8 ins.-3 ft. high with short stolons and swollen stem-bases. Inflorescence simple, with yellow or reddish linear acute spikelets 15-20 mm long and 3 mm wide. Glumes distant, obtuse, green-keeled. Grassland, woodland and damp places; 3,000-6,000 ft. KENYA — Western Region. TANGANYIKA — Western and Central Regions. UGANDA — Western Province and Buganda. 31. C. schweinfiirthianus Boeck. (C. zollingeri auctt. var. schweinfurthianus (Boeck.) Kukenthal) Perennial very similar to the above but the scabrous stems, and spikelets up to 30 mm long, 2.5 mm wide. Woodland, swamps and damp places; 3,000-5,000 ft. TANGANYIKA — Western and Southern Highland Regions. UGANDA — Western and Eastern Provinces. 32. C. zollingeri Steud. (C. rubroviridis Cherm., C. sphacelatus var. tenuior C.B.Cl.) Annual 6-12 ins. high. Inflorescence simple, spreading or contracted with short dense spikes. Spikelets linear-lanceolate, subquadrangular, 8-30 mm long, 1.5-2 mm wide, with distant glumes 3-4 mm long. Damp places, stream banks, and grasslands; sea level — 500 ft. KENYA— Coast. TANGANYIKA— Coast. ZANZIBAR — Zanzibar Island. 33. C. distans L.f. (Including C. ferrugineo-viridis var. distantiformis Kukenthal in part) Tufted perennial 1-3 ft. high with scabrid-margined leaves. Inflorescence large and spreading with dense spikes of dark spikelets obliquely spreading at first, becoming reflexed. Spikelets dark red or black, 10-20 mm long, less than 1 mm wide, with distant obtuse glumes with a rounded green back becoming spreading at maturity. Damp places, stream banks, lake shores and rivers; sea level — 8,000 ft. KENYA— Widespread. TANGANYIKA— Widespread. UGANDA— Widespread. ZANZIBAR — Zanzibar and Pemba Islands. 34. C. aterrimus Steud. Tufted perennial up to 4 ft. high with scabrid margined leaves. Inflorescence large, spreading, with black erect linear-lanceolate spikelets 10-20 mm long, 2 mm wide, with distant, keeled, obtuse glumes. In mist and mountain forests, damp places and seasonally flooded grassland; 3,500-10,000 ft. KENYA — Widespread but not common. TANGANYIKA — Widespread but not common. UGANDA — ^Western Province. 35. C. eleusinoides Kunth Perennial up to 4 ft. high with woody rhizome. Inflorescence large, with dense spikes of greeny-brown linear-oblong, acute spikelets 4-8 mm long. Glumes ovate-elliptic, mucronate. 2 mm long. Locally common on river banks and in swampy places; 3,000-4,500 ft. UGANDA — Karamoja and the Eastern Province. Page 12 Cyperaceae of East Africa — IV 36. C. iria L. Tufted perennial up to 2 ft. high. Inflorescence simple or compound with oblong compressed, obtuse, golden or yellow spikelets 5-10 mm long, in ovate or ovate-oblong short spikes. Glumes distant, broadly ovate, green-keeled, rounded, 1-1.5 mm long. Swamps, dams and ponds, rare; 3,000-4,500 ft. KENYA — Western Region. TANGANYIKA— Northern Region. 37. C. compressus L. Tufted annual 6 ins.-l| ft. high. Inflorescence simple, often contracted, with subdigitately arranged linear-oblong spikelets 10-20 mm long, 2.5-4 mm wide, with closely overlapped ovate, conspicuously mucronate glumes. Swamps and damp places, often in standing water; sea level^ — 3,000 ft. KENYA — Northeastern Region and the Coast. TANGANYIKA — Lake, Tanga and Southern Regions and the Coast. ZANZIBAR — Zanzibar Island. 38. C. pustulatus Vahl Tufted annual 1-H ft. high. Spikelets subterete, 6-12 mm long, 3-3.5 mm wide, in very short or subdigitate spikes. Glumes rounded, obtuse, grey-green streaked with red. Stream banks, swampy places; 3,500-4,500 ft. TANGANYIKA — Lake and Western Regions. 39. C. alternifolius L. ssp. flabelliformis (Rottb.) Kukenthal Tufted leafless perennial H-6 ft. high. Inflorescence spreading with numerous leaflike bracts partially concealing the small clusters of 3-7 pale or brownish linear-oblong spikelets 5-8 mm long 1.5-2 mm wide. Damp places, swamps and stream banks; sea level — 5,000 ft. KENYA— Widespread. TANGANYIKA— Widespread. UGANDA— Widespread. ZANZIBAR — Zanzibar and Pemba Islands. 40. C. ajax C.B.Cl. Stout perennial 3-5 ft. high with leaves up to H ins. wide. Inflorescence large with ovate or oblong-lanceolate reddish spikelets 2-5 mm long, in clusters of 3-5. Glumes conspicuously mucronate 1.5 mm long. Mountain and bamboo forests; 2,000-10,000 ft. KENYA — Mountains over 7,000 ft. TANGANYIKA — Northern, Tanga and Southern Highland Regions. UGANDA — Western Province. 41. C. dereilema Steud. (C. deckenii Boeck., C. dereilema ssp. deckenii (Boeck.) Kukenthal) Stout perennial 3-5 ft. high with leaves up to 1 in. wide. Inflorescence large with digitate clusters of ovate-lanceolate spikelets 4-10 mm long, 2-2.5 mm wide. Glumes tawny, brown, or dark red, obtuse. Many of the specimens examined show the tendency to leafy proliferation of the inflorescence common in most species of this group (species 40-46). Bamboo and mountain forests; 6,000-10,000 ft. KENYA— Widespread. TANGANYIKA— All upland areas. UGANDA— All upland areas. 42. C. pseudoleptocladus Kukenthal var. pseudoleptocladus (Including C. deckenii C.B.CL, not of Boeck.) Stout leafy perennial rather similar to the above but with leaves in. wide. Inflorescence dense with oblong-ovate spikelets 5-10 mm long, 1.5-2 mm wide. Glumes 2 mm long, brown or green conspicuously mucronate. Swamps and forests; 3,000-7,000 ft. KENYA — Eastern Region. TANGANYIKA — Northern, Tanga and Eastern Regions. UGANDA — ^Western and Eastern Provinces and Buganda. var. polycarpus Kukenthal Differs from the above in the more branched and less crowded inflorescence and the slightly larger, narrower and darker spikelets. Locally abundant in open places; 3,000-8,000 ft. KENYA — Western, Rift Valley, Northern and Eastern Regions. TANGANYIKA — Kilimanjaro and the Usambara Mts. UGANDA — ^Eastern Province. /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 13 43. C. glaucophyllus Boeck. Stout perennial up to 3 ft. high with leaves less than i in. wide. Inflorescence large with solitary or clustered oblong-linear pale greenish spikelets 4-15 mm long, 1.5-2 mm wide. Glumes obovate, mucronate. Very similar to C. dereilema except for the spikelets. Forest; 3.000- 7,000 ft. TANGANYIKA— Pare and Usambara Mts. UGANDA — Western Province. 44. C. fischerianus A. Rich. Tufted perennial 1-3 ft. high with scabrid-margined leaves less than i in. wide. Inflorescence large, compound, with oblong-lanceolate spikelets 4-6 mm long, 1.5-2 mm wide. Glumes ovate-oblong, reddish with a green keel, obtuse. Inflorescence usually proliferating and arching down to the ground and taking root. Mountain forest; 4,000- 8,000 ft. KENYA— All Highland areas. TANGANYIKA— Kilimanjaro. UGANDA — All mountain ranges. 45. C. diffusus Vahl ssp. buchholzU (Boeck.) Kukenthal Tufted perennial 1-2 ft. high, rather slender. Inflorescence large with brownish spikelets 6-7 mm long, 2-2.5 mm wide. Glumes with recurved mucros. Nutlets protruding conspicuously except from very immature spikelets. Damp places in dense bush and forest; 1.000- 7,000 ft. KENYA — ^Western and Rift Valley Regions. TANGANYIKA — Western, Central and Tanga Regions. UGANDA — Western, Eastern Provinces and Buganda. 46. C. renschii Boeck. Tufted perennial up to 6 ft. high with leaves up to 1 in. wide. Inflorescence large with ovate spikelets 2-4 mm long. Glumes green, 1 mm long, with a recurved mucro. Forest clearings; sea level — 7,000 ft. KENYA — Coast, Southern Region. TANGANYIKA— Widespread. UGANDA — Western Province. ZANZIBAR — Zanzibar Island. 47. C. mapanioides C.B.Cl. var. major (Boeck.) Kukenthal (C. dichromenaeformis Kunth var. major Boeck.) Tufted perennial 1-2 ft. high with stems somewhat swollen at the base. Inflorescence a simple contracted globose whitish head of compressed spikelets 6-14 mm long, 2.5-3.5 mm wide. Forests; 3,000-6,000 ft. KENYA— Elgon. TANGANYIKA— Western Region. UGANDA — Sese Islands, Buganda. 48. C. dichroostachyus Hochst. Tufted perennial 1-3 ft. high, with long stolons. Inflorescence rather dense with dark spikelets 2.5-4 mm long, 1-1.5 mm wide. Glumes black with broad, light green keels. Swampy places, stream banks dams; 4,000-8,000 ft. KENYA— Widespread. TANGANYIKA — Widespread, but more common in the north. UGANDA — Western Province. 49. C. difformis L. Erect perennial 9 ins.-H ft. high, often flowering as an annual. Inflorescence with numerous spikelets congested into dense greenish or red globose heads 6-12 mm diam. Spikelets 4-8 mm long, up to 1.5 mm wide. Dams, rivers, swampy places and seepage zones on rocky outcrops; sea level— 5,500 ft. KENYA — Western, Central and Southern Regions, Nairobi and the Coast. TANGANYIKA — Widespread in the northern part of the country. UGANDA — Western and Eastern Provinces and Buganda. 50. C. reduncus Boeck. Annual 4-12 ins. high. Inflorescence of clustered pale green spikelets 5-10 mm long, 3-4 mm wide, with oblong-lanceolate recurving glumes. Swamps and rivers; 3,000-4,000 ft. UGANDA — Buganda and Eastern Province. Page 14 Cyperaceae of East Africa — IV 51. C. submicrolepis Kukenthal (C. microlepis C.B.Cl. non Baker). Tufted annual 2-12 ins. high. Inflorescence of greenish clustered spikelets up to 4 mm long, 1.5 mm wide. Glumes obovate, closely overlapping. Poor shallow soils on rocky outcrops; 3,000-4,000 ft. U G AND A — Buganda. 52. C. tenuispica Steud. (C. flavidus C.B.Cl.) Slender densely tufted ephemeral with yellowish leaves. Inflorescence of rather yellowish spikelets 4-12 mm long with spreading truncate glumes less than 1 mm long. Swampy places; sea level — 3,500 ft. TANGANYIKA — Northern, Tanga, Western, Eastern, Southern Regions and Mafia Island. UGANDA — Buganda. ZANZIBAR — Zanzibar Island. 53. C. foliaceus C.B.Cl. Slender annual similar to the above, 1 ft. high. Spikelets 5-8 mm long, 1 mm wide, with truncate glumes 1.5 mm long. Stream banks; 500-4,000 ft. KENYA— Coast. TANGANYIKA — Tanga and Southern Regions. UGANDA — Buganda. 54. C. has pan L. (C. cancellatus Ridl.) Slender stoloniferous perennial up to H ft- high with short leaves or leafless. Inflorescence branched, with clustered spikelets 5-10 mm long, 1 mm wide. Glumes 1-1.5 mm long, shortly mucronate. Standing water, swamps, ditches, rivers; sea level — 5,500 ft. KENYA — Western. Eastern and Coast Regions. TANGANYIKA — Widespread, especially in the north. UGANDA — Western and Eastern Provinces and Buganda. 55. C. phceonhizus K. Schum, Slender perennial up to H ft. high. Inflorescence simple, of dark red shining spikelets 6-8 mm long with ovate-oblong truncate glumes. Dams, rivers, seasonally swampy places and seepage zones; 3,500-10,000 ft. KENYA— Widespread. TANGANYIKA — ^Widespread, especially in the north. UGANDA— Widespread. 56. C. platycaulis Baker var. platycaulis (C. denudatus var. platycaulis C.B.Cl.) Leafless perennial 1-3 ft. high with a stout woody rhizome and 3-winged stems. Inflorescence dense with numerous red spikelets 5-8 mm long, up to 1 mm wide, with small obtuse glumes. Very similar to C. denudatus. Streams, lake shores, dams; sea level — 8.000 ft. KENYA — Central Region and the Coast. TANGANYIKA— Lake, Northern, Tanga and Western Regions. There is much confusion over the precise limits of this species, and it is possible that most of the above localities refer to the following variety which is much more abundant in East Africa. var. lucenti-nigricans (K. Schum.) Kukenth. Leafless perennial with clums up to 8 mm wide. Inflorescence similar to the above but with dark red or black shining glumes. Swamps, dams and lake shores; 5,000-10,000 ft. KENYA — Widespread. Western, Rift Valley, Central and Southern Regions and Nairobi. TANGANYIKA— Widespread. UGANDA — ^Western Province. 57. C. denudatus L.f. Leafless perennial 1-3 ft. high with triangular stems and creeping rhizome. Inflorescence similar to the above species, but the spikelets 8-10 mm long, up to 2 mm wide, with ovate- lanceolate glumes. Lake shores, streams, swamps and seasonally flooded places; sea level — 7.000 ft. KENYA — Central, Eastern and Coast Regionsj /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 15 TANGANYIKA — Widespread in the north of the country. UGANDA — Western Province. ZANZIBAR — Zanzibar Island. There has been much confusion between C. denudatus, C. phoeorrhizus and C. platycaulis var. luceiiti-nigricans and they are possibly not all specifically distinct. 58. C. prolifer Lam. Rhizomatous leafless perennial 1-3 ft. high with terete or 3-sided stems. Inflorescence umbellate with equal rays, usually proliferating. Spikelets linear, 6-12 mm long, 1 mm wide with obtuse, light red glumes. Swamps, seasonally flooded places and standing water, looking like a small papyrus; sea level — 1,500 ft. KENYA— Southern Region and the Coast. TANGANYIKA — ^Eastern, Southern Highland and Southern Regions and the Coast. ZANZIBAR — Zanzibar Island. 59. C. tenax Boeck, var. tenax (C. grantii C.B.Cl.) Tufted leafy perennial 3-12 ins. high with glaucous, coriaceous leaves crowded at the base. Ipflorescence spreading with red or dark chestnut spikelets 6-12 mm long, 1-1.5 mm wide. Glumes obtuse, green-keeled. Sandy soils in bush and damp places; sea level — 3,000 ft. KENYA — Eastern Region and the Coast. TANGANYIKA — Southern Region, the Coast and Mafia Island. ZANZIBAR — Zanzibar and Pemba Island. var. pseudocastaneus (Kukenthal) Kukenthal (C. boehmii Boeck.) Differs only in the darker spikelets and the more distant, spreading glumes. Thicket and woodland on sandy soils, damp places; 3,500-6,000 ft. TANGANYIKA — Lake, Western, Central and Southern Highland Regions. UGANDA — Western Province. 60. C. cuspidatus H.B.K. Slender annual 2-6 ins. high with filiform leaves. Inflorescence of shortly pedunculate clusters often reduced to sessile, clusters appearing lateral to the bract. Spikelets chestnut 8-15 mm long, linear, with long-mucronate recurving glumes. Damp places, seepage zones on rocky outcrops; sea level — 4,500 ft. KENYA — Northern and Eastern Regions. UGANDA — Buganda and Eastern Province. ZANZIBAR— Pemba Island. This species can easily be confused with slender plants of Mariscus maderaspatanus. 61. C. amabilis Vahl Slender annual 2-10 ins. high with few leaves. Inflorescences of linear orange or brown spikelets 8-18 mm long, 1 mm wide. Seepage zones on rocky outcrops, lake shores and swampy places; sea level — 5,000 ft. KENYA — Northern, Rift Valley, Southern Regions and the Coast. TANGANYIKA — Northern, Tanga, Western, Central and Southern Regions and the Coast. UGANDA — Eastern Province. ZANZIBAR — Zanzibar Island. 62. C. maritimus Poir. Stout leafy perennial up to 1 ft. high. Inflorescence rather dense with oblong-lanceolate spikelets 12-30 mm long, 3-4 mm wide, with crowded obtuse glumes. Dry sandy places on river banks and the sea shore; sea level — 500 ft. KENYA— Coast. TANGANYIKA— Coast and Mafia Island. ZANZIBAR — Zanzibar Island. 63. C. frerei C.B.Cl. Rhizomatous perennial 1 ft. high with narrower leaves than C. maritimus. Inflorescence dense 1-2 ins. diam., or occasionally with rays up to 2 ins. long. Spikelets cinnamon or reddish-purple, 16-40 mm long, 3-5 mm wide. Sandy damp places; sea level. KENYA— Coast. Page 16 Cyperaceae of East Africa — IV 64. C. chordorrhizus Chiov. Low-growing perennial with a slender branched rhizome. Lateral shoots long, sterile, with semiterete glaucous leaves 1-2 ins. long crowded at the tips. Terminal shoots short, fertile, with leaves up to i in. long. Inflorescence capitate, 4-6 mm wide with bracts dilated at the base. Spikelets ovate-lanceolate, 3-7 mm long with whitish glumes. Coastal sand dunes; sea level. KENYA— Coast. 65. C. laevigatas L. {Juncellus laevigatas (L.) C.B.Cl.) Rhizomatous perennial up to 2 ft. high with stout leafless stems. Inflorescence a contracted, pseudolateral head of pale or dark spikelets 5-20 mm long, 2-2.5 mm wide. Glumes rounded on the back, 3 mm long. Swamps, lake shores and dams; sea level — 6,500 ft. KENYA— Widespread. TANGANYIKA — Lake, Northern and Western Regions. UGANDA — Western Province. 66. C. palchellas R. Br. (C. leacocephalas Nees non Retz.) Tufted perennial up to H ft. high with bulbous-based stems. Bracts reflexed. Inflorescence a dense globose head 5-8 mm diam. of whitish or cinnamon coloured com- pressed spikelets 3-6 mm long, 2-3 mm wide. Glumes at the base of the spikelets falling early. Swamps and damp places; sea level — 5,000 ft. KENYA — Eastern Region and the Coast. TANGANYIKA— Western Region. 67. C. angolensis Boeck. Sparingly leafy perennial 4 ins.-2 ft. high. Rhizome woody with long internodes and the stems bulbous based. Inflorescence a dense whitish hemispheric head up to 25 mm diam., the spikelets often purple tipped. Bush, open forest where burning is frequent; 2,000-9,000 ft. TANGANYIKA — Western, Southern Highland and Southern Regions. UGANDA — Imatong Mts. 68. C. margaritaceas Vahl Slender culmed perennial up to 2 ft. high. Stems with swollen bases on a woody rhizome. Spikelets few, usually 3-7, in a dense head, oblong ovate, up to 20 mm long, 8-10 mm wide. Glumes loosely overlapped usually rather shiny. Similar to C. obtusiflorus but most uncommon. Open bush and sandy places; sea level — 1,000 ft. KENYA— Coast. TANGANYIKA — Western Region and the Coast. ZANZIBAR — Zanzibar Island. Various varieties have been recorded in East Africa which, from the descriptions, seem scarcely distinguishable from the more robust form of C. obtasifioras. In the absence of type material of these two species and their numerous forms it has not been possible to distinguish fully between them. 69. C. obtasifioras Vahl (C. compactus Vahl) Leafy perennial up to 1 ft. high. Stems with swollen bases. Inflorescence a dense hemispheric head 15-20 mm wide with crowded, obtuse compressed spikelets 8-12 mm long, 4-8 mm wide. Glumes obtuse white. Woodland, rocky grassland, damp places, always on sandy soils; sea level — 6,500 ft. KENYA— Widespread. TANGANYIKA— Widespread. UGANDA— Widespread. ZANZIBAR — Zanzibar Island. 70. C. colymbetes Kotschy & Peyr. Leafless perennial up to 2 ft. high with stout sharply triangular stems 2-4 mm wide. Bract solitary as long as the dense 10-30 mm diam. head. Spikelets pale, lanceolate, 8-14 mm long, 6-8 mm wide, compressed. Styles occasionally bifid. In standing water, dams, swamps etc.; 2,000-4,000 ft. KENYA — Western Region. TANGANYIKA— Tanga Region. UGANDA — Western Province. /. E. Afr. nat. Hist. Soc. Vol. XXVI No. I (113) 71. C. nudicaulis Poir. Leafless perennial 1-2 ft. high with slender obtuse-angled stems. Bracts 2, scarcely exceeding the dense head. Spikelets 8-20 mm long, 4-5 mm wide with acute, tawny, green- keeled glumes. Styles 3 — 2-fid. Swamps and rivers; 3,000-6,500 ft. KENYA — Western and Central Regions. TANGANYIKA — Western and Tanga Regions. UGANDA — Western Province. 72. C. bellus Kunth var. tanganyicanus Kukenthal Slender tufted plants up to 5 ins. high becoming rather fibrous at the base. Inflorescence a solitary head of few compressed spikelets 8-12 mm long, 1-2 mm wide. Glumes brown, 2 mm long, with a conspicuous mucro. Rocky upland grassland; 2,500-7,000 ft. TANGANYIKA— Western, Southern Highland and Southern Regions. 73. C. kaessneri C.B.Cl. Annual 3-9 ins. high. Head dense, hemispheric, 10-25 mm diam. with linear-oblong obtuse spikelets 6-20 mm long, 2 mm wide. Glumes ovate, dark red but paling above. Open grassland, rock crevices, scrub; sea level — 4,000 ft. KENYA — Eastern and Coastal Regions. TANGANYIKA — Tanga Region and the Coast. UGANDA — Karamoja. 74. C. teneriffae Poir. Tufted annual up to 9 ins. high somewhat bulbous at the base with membranous sheaths. Head dense, 10-20 mm diam. with 5-10 mm long spikelets broader than in the above species. Glumes purplish or reddish 2.5-5. 5 mm long, with a recurved mucro. Damp grassland and seasonal pools; sea level — 6,500 ft. KENYA— Widespread. TANGANYIKA — Lake, Northern and Tanga Regions. UGANDA — Western Province and Karamoja. ZANZIBAR— Zanzibar Island. 75. C. pygmaeus Rottb. (C. michelicmus IL.) Link subsp. pygmaeus (Rottb.) Aschers, & Graeb., Juncellus pygmaeus (Rottb.) C.B.Cl.) Tufted annual up to 6 ins. high. Bracts 3-8, vep' long and spreading, dilated at the base. Head dense, 6-12 mm diam. with numerous spikelets 3-4 mm long. Glumes almost spirally arranged, lanceolate-ovate, 1-2 mm long. Damp sandy places; 1,000-4,000 ft. KENYA — Eastern Region. TANGANYIKA — Lake, Western and Southern Highland Regions. (Received for publication 11th March, 1966) /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 18 NOTES ON THE BIRDS OF LENGETIA FARM, MAU NAROK By P. H. B. Sessions Introduction The following paper is the summary of five years observation and notes on the birds of this high-altitude mixed farm. The notes were made in order to resolve their status, distribution, breeding and migration dates, and the effect on them of the change-over from natural to agricul- tural conditions. From 1953, when the farm was first occupied, to 1960 only a very few notes were miade, and this summary is mainly based on notes made during the following five years. The 250 species listed have been identified from the farm either by sight or sound, except where instanced in the text. I have included a locally-extinct species, the Ostrich, as being of interest, although unlikely to re-occur. The title heading for each species, giving the English and scientific for the bird under reference, plus the names of any other bird mentioned in the text, follow Praed and Grant. Only the English names are normally used where these birds are also mentioned in the text. The nomenclature of those species not found in Praed and Grant is taken from Witherby. Topography and Climate Beyond the western wall of the central Rift Valley, 35 miles west of Gilgil, lies a high plateau, between 9,000 ft. and 10,000 ft. above sea-level. This plateau runs more or less north-west and south-east for about 50 miles, and consists of open grassland, wedged in by thick forest on either side; it is occupied mainly by Masai tribesmen and is virtually in its primaeval state. Traversing this open plain from east to west is the farming district of Mau Narok, about 12 miles long by 3 miles wide. This farmland has been intensively developed from its virgin state over the last 13 years mainly with cereal crops and exotic grass leys for sheep. Lengetia Farm lies at the western end of this district and consists of 1,900 acres of both fiat and steeply ridged land. The south and west sides are bordered by cedar and olive forest, and the remainder is quite open; before 1953 the latter was covered with sour grasses and light bush, but now more than half is under crops and leys. Across the farm a number of windbreaks of pine and cypress have been planted. Three semi- permanent streams run north and south through the farm, and in the Page 19 Notes on the Birds of Mau Narok middle of a flat waterlogged fleld lies a small dam of about an acre of water. The annual rainfall averages 41", the rainy periods occurring in April/May, heaviest in July/ August, and again in November/December. The average shade temperature at noon is 70°F, and at night there is often a frost. The prevailing wind is from the east, and is fairly strong in the dry season. The climate is generally equable, and does not suffer the extremes that occur at lower altitudes. The Habitats The area contains four habitats (I) Forest, (II) Open Grassland, (III) Rocky Stream-beds and Lightly Wooded Valleys, and (IV) Aquatic. (I) The Forest covers 200 acres along a two mile boundary being the edge of the main forest which lies inside the Masai Reserve. In fact many birds seem to prefer this edge of forest, which has been subjected to successive fires and where much secondary growth has arisen. Further in, where there is less ground cover, and less light penetrates the towering cedars, bird-life seems comparatively scarce. Over 50 of the resident species live in the forest edge habitat. The garden should be included here, as although it has been made a quarter of a mile from the forest, it has gradually attracted many species from there. (II) The Open Grassland covers about 1,500 acres and is practically treeless. 20 species are resident or breed in this area, but it is a great attraction to migrants and birds of prey. (III) The Rocky Stream-beds and lightly Wooded Valleys are the haunt of the Mackinder’s Owl nightjars, Black Duck and Wryneck. (IV) The Aquatic habitat consists of a dam set in an open windy marshland, this never-the-less has been the primary attraction for 45 species of birds. The effect of cultivation does not seem to have an adverse action on any species, except the Capped Wheatear, and most birds which existed on the natural grasses seem to thrive even better on planted crops. The Ostrich, which has been locally exterminated, is more likely the victim of “civilisation” than of cultivation. Species unobserved before cultivation, which are beginning to colonise the farm include the Black-shouldered Kite, Ring-necked Dove and Pied Wagtail; other birds, especially certain species of sparrows and weavers not yet listed for the farm, are moving nearer each year. Vegetation The vegetation is divided sharply between the forest and open grass- land; there is hardly any park-like land. The forest consists mainly of J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 20 cedar, Juniperus procera Hochst. ex Endl., and brown olive, Olea africana Mill., with a few Kenya olive, O. hochstetteri Baker; most of this timber is in a dry and over-mature state. There is hardly any bamboo, podo or mukeo, which all occur on the farms a few miles to the east, and are indicative of a slightly higher rainfall. The secondary growth along the edge of the forest consists of a large number of species of shrubs and small trees, as well as numerous young cedars. The grassland is composed mainly of short grasses flowering not more than 18" high, together with some heather and light bush, such as Berheris holstii Engl., on the ridges where the soil is deeper. Where cattle were stockaded in the past, nettles, thistles and tussock grass cover the ground. Migration The large number of migratory birds, comprising some 50 species, seen regularly on passage, seem to indicate that this area lies across a definite migration route, possibly two routes. Flanked by thick forest, the Mau Narok belt of open land may act as a funnel for birds moving north and south, or east and west, especially as it includes a number of small dams which provide convenient resting places. Breeding Most of the small birds have their main breeding season in April/May, and the forest birds may breed again in November /December. The breeding dates are derived from any associated evidence, i.e. from display to the presence of fledglings, and refer roughly to the egg-laying period. The word “pair” is user rather loosely in the text, generally as opposed to “single” or “flock”, rather than to denote “male and female”. Systematic List OSTRICH, Stmthio camelus massaicus Neum. Formerly a resident, breeding locally between 9,000' and 10,000', according to a reliable Masai tribesman. Lunge Ole Kisaga, who has lived here for 50 years. He says they were seen in numbers up to a 100, away from the forest on the open grasslands. When I arrived in 1953, I found only a cock and two hens, and these were soon wantonly destroyed. GREAT CRESTED GREBE, Podiceps cristatus Linnaeus. Rare visitor to the dam: two records only, a single bird on 7.vi.63, and two birds on 4.xi.63. LITTLE GREBE, Poliocephalus mficollis (Pallas). Resident, only absent for short spells. Found on the dam and sometimes on smaller pools; breeds, juveniles being seen in August. In April 1964, 15 birds stayed on the dam for 2 days, and were in every stage of plumage, from near-white immatures to almost black adults. Page 21 Notes on the Birds of Mau Narok WHITE-NECKED CORMORANT, Phalacrocorax carbo lugubris Riippell. Passage migrant. Three records for October and December 1963, one bird being immature. LONG-TAILED CORMORANT, Phalacrocorax africanus (Gmelin). Uncommon passage migrant, but occurring most months. [WHITE PELICAN, Pelecanus onocrotalus Linnaeus.] A party of about a dozen birds seen over Lutyens’ farm on 13.vii.65. The birds were really too high for us to distinguish the species, but this seemed the most likely. GREY HERON, Ardea cinerea Linnaeus. Occasional visitor to dam in winter months; immatures seen November 1960 and September 1965. Rather shyer than following species, and I consider they are mostly palaearctic migrants. BLACK-HEADED HERON, Ardea melanocephala Vigors & Children. Visitor, more frequent than the Grey Heron, although only singly as that species. Much more common, and seen in small flocks on farms a few miles to the east, where there is wetter land. PURPLE HERON, Pyrrherodia purpurea (Linnaeus). Uncommon visitor, although one immature bird stayed 6 months. Has been seen locally in small flocks. GREAT WHITE EGRET, Casmerodius albus (Linnaeus). Regular visitor, about alternate months, always singly, with yellow bill. Usually stays for one or two days only. YELLOW-BILLED EGRET, Mesophoyx intermedius (Wagler). Occasional visitor to dam, and is probably a passage migrant from the south, going north to breed, as all records are for the month of May. BUFF-BACKED HERON, Bubulcus ibis (Linnaeus). Regular annual passage migrant in small numbers in March and April, when a few birds are seen resting on their way north. SQUACCO HERON, Ardeola ralloides (Scopoli). One record for 17.xi.65, the bird staying for two days on the farm. The bird was heavily streaked with brown on its flanks, and may have been an immature, or even been A. idae (Hartlaub). It perched freely on tall cedar trees when disturbed. GREEN-BACKED HERON, Butorides striatus (Linnaeus). One record for October 1964. This was a pale bird, looking like a miniature Grey Heron; it had bright yellow legs, but the colour of the legs seems to vary, for the colour of legs of birds seen in Narok and Baringo were both different; no two textbooks give the same description, which may be due to marked seasonal variations. HAMMERKOP, Scopus umbretta (Gmelin). Visitor to dam and small pools; recorded for most months. Always singly. Usually very tame, and stays several days. WHITE STORK, Ciconia ciconia (Linnaeus). “Winter” visitor and passage migrant, with some birds staying on through June and July. In 1980, many of the birds that stayed behind appeared to be sick and dying, and observers from Molo reported the same thing; they may have eaten locusts or other insects killed by spraying. White Storks become most numerous in the New Year when ploughing starts, and they are generally to be seen following the tractors, sometimes lining the furrow for a hundred yards or more. They appear to exhaust the food supply of the fresh-turned soil quickly, and then stand patiently until the tractor returns. They eat mainly small flying and larval insects, although their prize catch is the mole-rat {Tachyoryctes), which, if they are allowed to keep, occupies them a quarter of an hour in trying to swallow. They make /. E. Afr. nat. Hist. Hoc. Vol. XXVI No. 1 (113) no attempt to rob each other, but often lose their prey to Marabou Storks (Leptoptilos) and Steppe and Tawny Eagles (Aquila rapax) which are often at hand and take advantage of the Stork’s tameness which allows it to approach within a few feet of the tractor and so find the mole-rats before they can dive beneath the furrow out of which the plough has just thrown them. At dusk the Storks roost either on an open field of short grass, or on the tops of the tall cedars on the edge of the forest; as they settle down for the night they indulge in some bill-clattering. There seems to be some danger in the ground roosting habit, as many birds have been found with broken legs (about a dozen on the farm last year) and they may take off in the night when alarmed by a small predator such as a jackal, and fly into the wire fences. The frequency of these accidents must constitute a considerable factor in their death rate. In the morning they fly back to the working tractors about an hour after sunrise; if the tractor has moved into another field they go there directly, and I have seen them move low down from a ground roost to some new ploughing over a hill, which indicates they find their way by ear. From January to February there are about 200 birds on the farm, and probably 2,000 in the district. On a brilliant cloudless morning in March I have watched small parties from many farms joining up into a group of a 1,000 birds spiralling up on a thermal, the flock nearly disappearing out of sight overhead, before gliding off north- wards at great speed. BLACK STORK, Ciconia nigra (Linnaeus). Rare “winter” visitor, recent dates being 13.ii.61, and l.iii.61. Both were in pairs, and unaccompanied by other species. WOOLLY-NECKED STORK, Dissoura episcopus (Boddaert). One record of 50 to 60 birds flying east, 9.00 a.m., 27.xi.60, low over the house, into a strong wind, on a clear sunny day, when their white necks showed up clearly. ABDIM’S STORK, Sphenorynchus abdimii (Lichtenstein). A not uncommon visitor, only in the “winter” months, there being no records for May to October inclusive. Often in company with White Storks. By far the most common occurrences are in November and December, when it occurs in flocks up to 200. OPEN-BILL, Anastomus lamelligerus Temminck. On 29.xi.64. 150 to 200 birds flew eastwards over the farm about midday. They made their way into a strong wind by circling up until they were almost out of sight, then gliding down until they nearly reached the ground. Finally they must have picked up a more favourable wind, as they made a bee-line for the Rift Valley at a great height. Again on 18.xi.65, in company with Mr. Leslie Brown, I saw a large flock of nearly 400 birds 4 miles west of the farm; these also were flying due east, and the following day I saw another flock of 200. A few days previously several of these storks were actually seen resting on the farm. SADDLE-BILL, Ephippiorhynchus senegalensis (Shaw). Passage migrant: one record, 29.xi.60, bird flying low over farm eastward, 10.00 a.m. Also observed by others on dams to the east, up to 9,500 ft., at least three times. MARABOU, Leptoptilos cmmeniferus (Lesson). Variable visitor, sometimes a few, sometimes many, at times staying for weeks, sometimes not seen for months. On 31.iii.53, several hundred roosted in cedar trees on farm. Often appear with White Storks, (q.v.), on which they prey, robbing them of the mole-rats they catch when following the plough. Appear at lambing time to feed on the afterbirth with vultures, but generally not often seen to eat carrion. WOOD-IBIS, Ibis ibis (Linnaeus). One immature bird visited dams on this and other farms from July to October 1963. Very shy. SACRED IBIS, Threskiornis aethiopicus (Latham). Occasional visitor to dam. Recorded January, March and April over 4 years. One to three birds at a time, staying only a day or so. HADADA, Hagedashia hagedash (Latham). Regular visitor, but resident and breeds locally. Nested on a dead tree overhanging a dam on a neighbouring farm. May 1962, the eggs being destroyed. Praed and Grant are Page 23 Notes on the Birds of Mau Narok rather misleading in stating that birds seen over 7,000 ft. would be the Green Ibis, Lampribis olivacea (Dubus), as I have been unable to confirm the presence of this bird, while the Hadada is fairly common between here and Molo at 9,000 ft. However our Hadada do seem much darker than birds of this species seen at lower altitudes. AFRICAN SPOONBILL, Platalea alba Scopoli. One bird seen beside the dam on 20.X.65. Lesser flamingo, Phoenwonalas minor (Geoffroy). Occasional weak birds fall from migrating flocks; this occurred especially in 1961 when many lakes were dry, and birds drifted about all over the Rift Valley, some being picked up dead in tiny pools on mountain streams. I have occasionally heard large flocks passing overhead at night, and from the sound I judged them to be travelling north or south. From the map it would seem that these birds do not necessarily follow the line of the Rift Valley Lakes when moving to another feeding ground, but may take a more direct overland route. AFRICAN POCHARD, Ay thy a erythrophthalma (Wied). Occasional visitor, especially in years 1954-1956, lately only seldom, perhaps one or two a year. SHOVELER, Spatula clypeata (Linnaeus). Occasional winter visitor between November and February; singly or in pairs. YELLOW-BILLED DUCK, Anas undulata Dubois. Commonest duck found on the dam, most months; generally a pair, but up to 26 seen. Not found to breed, but may do so when surrounding cover grows. BLACK DUCK, Anas sparsa Eyton. Probably resident, but difficult to find at times, as during the dry weather it retires into the forest. Newly hatched ducklings seen locally September 1963, and a family party of 8 including f-grown birds on 18.vi.61. Otherwise only seen singly or in pairs. WIGEON, Anas penelope Linnaeus. Two birds in a large mixed pack of duck on dam, October 1962. GARGANEY, Anas querquedula Linnaeus. Three records: a pair in November 1962, a pair in October 1963, and a single bird that stayed for two weeks in December 1964. CAPE WIGEON, Anas capensis Gmelin. One bird, 12.iii.65. HOTTENTOT TEAL, Anas punctata Burchell. Three single records, September and December 1960, and April 1961. RED-BILL, Anas erythrorhyncha Gmelin. Common visitor 1954-1956, now only rarely, about twice a year, in pairs or small parties. Usually seen with Yellow-bills. PINTAIL, Anas acuta Linnaeus. Two records, 18.xi.63 and l.i.65, the second being for a male and female. FULVOUS TREE-DUCK, Dendocygna bicolor (Vieillot). Three records, all single birds; May, November and December. KNOB-BILLED GOOSE, Sarkidiornis melanotos (Pennant). Uncommon visitor. EGYPTIAN GOOSE, Alopochen aegyptiacus (Linnaeus). Irregular visitor, which bred once. About 6 months after the dam had filled for the first time, a pair hatched 5 goslings, mid-October 1954, all of them surviving and leaving December 13th. Seen occasionally in singles, pairs or parties up to 15, usually March/April and October/December. J. E. Afr. nat. Hist. Soc. Vol XXVI No. 1 (113) Page i4 SPUR-WINGED GOOSE, Plectropterus gambensis (Linnaeus). Rare visitor, with one record for 8.ix.61, and two or three birds seeti on local dams. SECRETARY BIRD, Sagittarius serpentarius (Miller). Resident in the district. Has bred on the farm once, when two young were reared from a nest on a low flat topped cedar tree in a river valley. Birds sitting on eggs have been seen in March and September. What appeared to be a display was observed when two birds circled about 150 ft., above the house, one of them making a loud, deep, creaking noise, quite unlike any other bird sound; this may be compared with Priest, p. 187. On another occasion I witnessed a Secretary Bird fly down a Snipe. The latter flew about a 100 yards at a time, dropping down into the tussock vlei grass, only to dart up as the Secretary Bird approached. Finally the Snipe left it too late, and the pursuer made a successful grab with its talons. When I arrived at the spot there was hardly a feather to be seen. RUPPELL’S GRIFFON, Gyps ruppellii (Brehm). Possibly the commonest vulture in the area, sometimes in mixed flocks, at other times by themselves in flocks of up to 50 birds. Roost in cedar trees, leaving in the morning about 10.00 a.m., to pick up a thermal, usually over a ploughed field. On a cold wet day, they often start off much earlier, fiapping off in a long, leisurely line into the Masai reserve. The breeding status of the vultures has not been mentioned, as although they are seen all the year round, it is unlikly that any of them breed near here. WHITE-BACKED VULTURE, Pseudogyps africanus (Salvador!). Quite a common vulture which is seen most months, with up to 20 birds in a mixed flock of vultures. The white back is not easily seen, and the best identification is from the under-wing pattern. LAPPET-FACED VULTURE, Torgos tracheliotus (Forster). Generally only in pairs or singly in mixed flocks of vultures, but seen more regularly than other species. WHITE-HEADED VULTURE, Trigonoceps occipitalis (Burchell). Only two positive records, October and November 1962, in flight. EGYPTIAN VULTURE, Neophron percnopterus (Linnaeus). Three records; October/November 1962, and April 1963. HOODED VULTURE, Necrosyrtes monachus (Temminck). Fairly common, about 5 to 15 birds in a mixed vulture flock of 30 birds. The surrounding Masai country still provides a lot of carrion, due to the frequent deaths of the native cattle, the afterbirth from domestic stock, and the remains from hyena kills. About the only animal they seldom touch is a dead hyena itself; they can finish off an entire leopard carcase in an hour, although on occasions they will not touch that either. [PEREGRINE, Falco peregrinus Tunstall.] Typical peregrine type seen occasionally flying west, usually very fast, and obviating accurate identification. Most records have been for October, and November. HOBBY, Falco subbuteo Linnaeus. Winter passage migrant. Hobbles are seen fairly often but the speed at which they fly normally precludes naming of the species. Singly or in pairs. Mostly seen November. December and April, once September. AFRICAN HOBBY, Falco cuvieri Smith. Status uncertain, but rarely seen. I have watched it hunting the Black-winged Plover, by quartering the ground at great speed, causing the plovers to take off, but have yet to see it make a kill. EASTERN RED-FOOTED FALCON, Falco amurensis Radde. Uncommon passage migrant. Records in November, December and April. Page 25 Notes on the Birds of Mau Narok KESTREL, Falco tinnunculus Linnaeus. Winter passage migrant, but generally not distinguished from the Lesser Kestrel. From the certain identifications made, it would appear that the latter is much the commoner. One bird was observed on the 8.vi.63, hovering overhead, and might well have been the ABYSSINIAN KESTREL, F.t. carlo (Hart & Neum). LESSER KESTREL, Falco naumanni Fleischer. Winter passage migrant, usually in small flocks of 6 to 20 birds. As with many of the European migrants, they are generally seen moving in an easterly or south-easterly direction on both the spring and autumn passages. Records for both species of Kestrels occur from October to April only, and they are seen mostly in these two months, with a slight increase of numbers .again in December. KITE, Milvus migrans (Boddaert). Generally present on farm in small numbers, except for July and August when there are no records. In January and February they are found in flocks of from 20 to 30 birds, which drift around and retire to roost in cedars on the edge of the forest in the evening. I have not yet found their nests, but they may well breed here. A number of these birds have been definitely identified as the migratory European race, M.m. migrans (Bodd.), with black bills. BLACK-SHOULDERED KITE, Elanus caeruleus (Desfontaines). Occasional visitor. This is one of the birds that appears to be extending its range west- wards with the increase of cultivation. It used to be observed regularly about 15 miles to the east, and in March 1962 it was seen three miles away. Six months later it was seen on the next farm, and a year later it was on this farm. Since then it has been seen with increasing frequency, though normally staying only for a week or so. STEPPE/TAWNY EAGLE, Aquila rapax (Temminck). Praed and Grant, in the amendments contained in their 1956 edition, have combined the Steppe and Tawny Eagles into one species, the Steppe being a migrant and winter visitor, and the Tawny a resident. Whether this is a valid arrangment or not, the two races appear to be distinct in the field. The Steppe Eagle, renamed Aquila rapax orientalis Cab, arrives around the second week of October, and may be recognised by the pale upper-tail coverts, which in the young bird look almost white; there is also a distinctive under- wing pattern which is formed by pale tips to the under-wing-coverts, and appears as a V across the wings. Sometimes these migrant Eagles are seen in numbers up to a dozen, and it seems hard to believe all these would be juveniles, and yet all of them will show the under-wing and upper- tail patterns, and distinguish the bird from the resident Tawny Eagle. Other characteristics include, of course, its habit of flocking, and also a more pronounced tendency to settle on the ground; I have not yet been able to separate their call notes. They spend much time in company with White Storks.waiting to rob them of mole-rats which the latter sometimes catch. The Steppe Eagle departs early in April. The Tawny Eagle is a fairly conrunon resident, which may undergo some local movement. There are probably two pairs on the farm, and they must nest at the turn of the year, as I usually see very young birds in April. There are at least three other pairs in Mau Narok, which all breed about the same time. AFRICAN HAWK-EAGLE, Hieraaetus spilogaster (Bonaparte). Uncommon visitor, which seldom gives enough time for sure identification. One bird dispersed an enormous flock of Pink-breasted Doves, which were feeding on some newly planted wheat seed, and which I had failed to move by gunshot. The next morning I found the remains of many dead doves scattered over the field, which I presumed had been killed by this Eagle. AYRES’ HAWK-EAGLE, Hieraaetus dubius (Smith). One definite record, 9.viii.65, and possibly other occasions when confirmation was difficult. The white over the eye was a prominent field character. MARTIAL EAGLE, Polemaetus bellicosus (Daudin). Visitor, which probably breeds locally. Two juveniles were shot locally in July 1961 and May 1962, while stealing poultry. Another young bird was seen in August 1963. From the appearance of the young, sometimes seen with their parents, I would guess that they breed at the beginning of the year. I have only twice heard this bird call, each time from i. £. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 26 the same place, and nearly two years apart. Once the bird was flying, the other time it was in a tree, flying off silently when disturbed. The call was high and flutey “hoy-yok hoy-yok hoy-yok” and distinct enough for me to remember after the two year gap. They do not appear to take carrion, as I have seen a bird follow vultures to a corpse, and return without feeding. Another one I saw in March 1953, alighted on the back of a White Stork, which was feeding in company with about a hundred others, and killed and ate it, with three Tawny Eagles watching from a few feet away. The Stork seemed quite healthy. CROWNED HAWK-EAGLE, Stephanoaetus coronatus (Linnaeus). Resident, breeds. One pair have a nest in an 80 ft. tall dead cedar about 800 yards below the house, about 100 yards inside the forest and 50 yards from the stream bed. The nest is clearly visible from the farm road, and I have had it under close observation for nearly 4 years. The nest is about 6 feet across and the same in depth, and is placed in a fork about two-thirds up the tree, against the trunk. During a very heavy storm in 1963 half of the nest collapsed, but fortunately the young bird was just able to fly. On 17.xi.63, when the juvenile was over a year old, all three birds reassembled at the nest tree, and each in turn climbed all over the collapsed nest. The young bird was especially excited, and spent a long time on it, calling continuously, while the parents perched quietly nearby. At the end of June 1964, both adults arrived to repair the nest, and commenced their 2-year breeding cycle once more. I have deduced this cycle as follows: farm workers tell me that the birds bred in 1956, and 1958, while I have observed them breeding in 1960, 1962 and 1964. The timetable is worth recording, if only to confirm work by L. H. Brown. The adults begin displaying and paying attention to the nest about the end of June, the female starts to sit in mid-August, and there is no sign of eggs hatching until mid-October, when the Eagle stands up, or makes a stiff flight to a nearby tree. Even then the chick is not visible until a month later at least, but it grows fast thereafter, with the female feeding it daily and tearing its food up until February. Early in this month the feeding becomes less frequent, and the young bird starts its hunger call, a rapid plaintive whistle that it will keep up all day if necessary. It also starts to climb around the nest tree about now, and by the end of the month it may have made its first flight, although it continues to be fed at the nest; sometimes the nestling takes another month before it decides to fly. The young bird continues to use its hunger call irregularly for another 8 to 12 months within half a mile of the nest, and finally disappears when it is about 18 months old. Some can be quick at finding their own food, for my headman, Mr. Kamonde, who is a reliable observer, saw an immature bird in May, still in its white plumage, take a half-grown Colobus Monkey off the limb of a dead tree with scarcely a pause in its flight. The female appears to do all the building herself, as well as adding green branches during the time she is on the nest, but the male can generally be seen sitting quietly by. They make a devoted pair, and once during a torrential downpour, (and the rain is cold at this altitude), I watched the female brooding with outspread wings over the chick, while the male perched on a branch about ten feet away; during this time the two birds maintained a continuous soft piping to each other. The display of these Eagles takes place mainly during the 18 months between the start of nesting to the departure of the juvenile. As the birds breed in alternate years, there is a gap of about six months when they are seldom heard. The display is well described by L. H. Brown (Eagles p. 186) and usually takes place over the forest, not far from the nesting site; on occasions the bird will rise to a very great height, and carry on calling far out over the grasslands. Either one or two birds may be seen in the display flight, but I suspect only the male does the calling, a far carrying “ke — wik ke— wik ke — wik”; the female’s call is different, described by Chapin as “pee-ou” rapidly repeated. I have generally heard this call from the nest, and it appears to be a call for its mate; standing on the nest beside its chick the bird flattens itself right out and opens its bill wide to get maximum sound, varying the speed and pitch of the notes; the call lasts about two minutes, and then the bird listens and looks around before starting up again, and so on for half an hour. TTiere are about six or seven pairs in the district, all with territories bordering on the forest, and 3-4 miles betwen each pair; altogether they must be fairly numerous over the Mau. LONG-CRESTED HAWK-EAGLE, Lophoaetus occipitalis (Daudin). Uncommon visitor. This eagle is usually seen over the forest if at all, whereas at Njoro it is a bird of the open farmland. Six records for 1962, none in 1963. BLACK-CHESTED HARRIER-EAGLE, Circaetus pectoralis Smith. Regular visitor, which possibly breeds locally. Immature seen on farm in June and July 1961.. Generally singly, once or twice in pairs. They spend a lot of time hovering, but I Page 27 Notes on the Birds of Mau Narok have yet to see one make a kill. Always seems to be in the air, rarely perches on a fencepost and never seen in a tree. BATELEUR, Terathopius ecaudatus (Daudin). Status as for preceding species; it appears about alternate months, and the farm must be on the edge of a territory. On 27.ii.62 an immature bird was chased by an adult in a wonderful aerial exhibition. Their relationship to Tawny Eagles is curious, as I have seen a pair chase the latter away with much gusto, but another time I watched a Tawny Eagle devour a large Mole-snake, while a Bateleur sat on the ground a few feet away, for over an hour, not daring to approach. Perhaps their mastery is confined to the air. FISH EAGLE, Cuncuma vocifer (Daudin). One record, 20.ii.55, a bird perched in a small cedar tree above a tiny pool, in the late evening. STEPPE BUZZARD, Buteo vulpimis (Gloger). Winter visitor. This is not an easy bird to distinguish from the Mountain Buzzard at a distance, and it is not possible to say if these birds reside for the winter, or are on passage only. They even appear to consort with the Mountain Buzzards, which confuses the situation. However, I have recorded it for all months from September to April. MOUNTAIN BUZZARD. Buteo oreophilus Hartert. Resident. One or two pairs, but I have not found their nests. They are seldom seen any distance away from the forest and are generally found circling over the trees or perched quiet on a bough; when disturbed, they slip silently off and move to another tree a little further on. Their “mew” is very like that of a European Buzzard’s B. buteo (Linnaeus). They have a fast “switchback” disply flight, during which they mew loudly. AUGUR BUZZARD, Buteo rufofuscus (Forster). Common resident. About six pairs breed on the farm, mainly April /May and October/ November. Highly beneficial, as one of their staple foods is the mole-rat, which causes immense damage to pasture and grain crops. About one in three is black-phased, and these seem to predominate in one particular family. They seem more active here than at lower altitudes, and are flying and hovering most of the day. They are very audacious in defence of their territory, and I have seen them chase off Tawny, Steppe, Crowned and Bateleur Eagles in a most aggressive manner. If a single bird starts the chase, it is only a matter of moments before the mate has arrived. Only once have I failed to see both birds, and that was when a female Crowned Hawk Eagle had been calling for half an hour, fed her young, and then joined in a simultaneous display with her mate. In the course of this flight, they came too close to the cedar tree where the Augur Buzzards were nesting. A small adult Buzzard came shooting out, which I took to be the male, and engaged with the female Eagle. The normally lethargic Buzzard put up a series of terrific power dives, and the Eagle had to turn over on her back each time, and I could see her claws lash out at her pursuer, ducking her head as he came shooting past. The Eagles then retired to a tree in the forest; I think the second Buzzard must have been sitting too tight to be able to leave her nest. In spite of being so common. Augur Buzzards are, for a large bird, quite hard to recognise in the field, due to their innumerable colour phases; a young bird with no barring on wings or tail (which is much longer than an adult’s,) dark brown above and below, took a long time to identify. They can often be seen flying back to their roosting tree after sunset, when they look very owl-like, and in the early morning they may be seen on their favourite look-out perch long before the sun is up. LITTLE SPARROW-HAWK, Accipiter minullus (Daudin) Occasionally seen. RUFOUS SPARROW-HAWK, Accipiter rufiventris Smith. This Sparrow-Hawk has been clearly identified on occasions, but this group is particularly hard to verify in the field, as usually all one sees is a flash of grey or brown, and the bird has disappeared into the thickest trees; the relative status of each species is hard to assess. ]. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 28 GREAT SPARROW-HAWK, Accipiter melamoleucus Smith. Resident nearby. Tlris bird is sometimes seen on the farm, but its real habitat is in the thick forest outside the farm boundary. An immature female was caught in the poultry-run on 10.iii.65. I have also seen a black-phased bird flying through the garden. SHIKRA, Accipiter badius (Gmelin). Occasional visitor, one bird staying near the house for two days in July 1962. For most of the time it remained in some young pine trees, making a purring “coooorr coooorr”, perhaps akin to the sound made by the Ovampo Sparrow-Hawk, as described in Praed and Grant. AFRICAN GOSHAWK. Accipiter tachiro (Daudin). Occasionally seen, but may well be a resident. Generally one only catches a fleeting glimpse of it, but on 20.xi.62, I saw a single bird dive down to attack a Common Sandpiper which was running on the rocky edge of a stream. The wader evaded the attack by jumping into the little pool, whereupon the hawk squatted down on the rock alongside. I was able to observe the bird at close range with binoculars for some minutes, and had to disturb it eventually to see its appearance in flight. DARK CHANTING-GOSHAWK, Melierax metahates Heuglin. One record, January 1965, in the garden. MONTAGU’S HARRIER, Circus pygargus (Linnaeus). Regular winter passage migrant, but less common than the Pale Harrier. They are seen throughout the winter months, but not apparently staying on the farm. They arrive in early October, when they move fast and purposefully. Later they are generally seen leisurely quartering, and moving eastwards. PALE HARRIER, Circus macrourus (Gmelin). Common winter passage migrant, generally moving east at both seasons. Both these harriers are most often seen in pairs, the male bird 200 to 300 yards in front. If the pre- dominant east wind is at all fierce, the birds will drop to the ground and rest for half an hour before drifting into the wind again. One female or immature that had apparently lost its mate, uttered a high “kitter kitter kitter” every minute or so; the only time I have heard these harriers make any sort of noise. The date of autumn arrival is the first week of October, once late September. MARSH HARRIER, Circus aeruginosus (Linnaeus). Regular passage migrant, and some years a winter visitor, when one or two birds will frequent the dam for weeks. The early birds are seen mid-October and then only a few until the return passage in March/ April. The African race, C.a. ranivorus (Daud), has only been confirmed on one occasion, June 1963, and can only be an unusual visitor. The Marsh Harrier is seen in much smaller numbers than the preceding harriers, and seems a more sedentary bird. It will spend hours on a fence post ,or sitting on the ground or in reeds beside the dam. The earliest arrival was an adult male on 25.ix.65. HARRIER-HAWK, Polyboroides typus Smith. Uncommon visitor, mainly March and December. It is seen more regularly a few miles to the east; where I suspect it to breed, as I have seen an immature on this farm in March. Generally observed near forest, where it is often seen clinging with outstretched wings to the trunk and foliage of cedar trees looking for prey. MONTANE REDWING FRANCOLIN, Francolinus psilolaemus Grey. This bird is called F. shelleyi theresae Meinertzhagen in Praed and Grant, and was identified for me by Mr. John Williams (following the revision by Hall 1963) from a specimen I sent him. In voice, habits and appearance it resembles closely the Redwing, F. levaillanti, (Valenciennes), and Shelley’s Francolin, F. shelleyi O. Grant. Normally found in pairs or with young, which stay with parents until next breeding season. The young chicks are seen from June to August, and there is some indication that this species breeds again in September. About five young are reared. SCALY FRANCOLIN, Francolinus squamatus (Cassin). Two or three pairs resident on farm. Breeding season May and June. Page 29 Notes on the Birds of Mau Narok JACKSON’S FRANCOLIN, Francolinus jacksoni O. Grant. Resident, with about six pairs on farm. A pair seems to keep to the same small area for years, and although breeding successfully their numbers do not increase. They breed at any time of the year, and stay in family coveys for about 8 months, after which time they are seen in pairs. They frequent the forest edge, moving out into the grass leys and ‘shambas’ in the morning and evening. They are very partial to beds of nettles, which is perhaps why serval cats are, as well ! I have seen up to seven chicks in a brood. QUAIL, Coturnix coturnix (Linnaeus). The European race has not yet been confirmed in hand, but probably occurs. The African race, C.c. africana Temminck & Schlegel, is a resident, and probably commoner than in the past due to increased cultivation. It is usually seen in pairs, and may breed regularly, but I have rarely seen young birds, usually September and October. Up to 50 pairs on the farm. HARLEQUIN QUAIL, Coturnix delegorguei Delegorgue. Irregular visitor, chiefly May, June and July. I have recorded this bird with chicks in June and July 1954. On two occasions in June there has been evidence of night migration, with birds hitting lighted windows in some numbers. [CRESTED GUINEA-FOWL, Guttera edouardi (Hartlaub).] I have not seen this bird on the farm, but include it as it has been taken nearby on Mr. Grainger’s farm, and I have also seen it only a few miles away inside the forest. It is a most beautiful bird, with brilliant metallic blue plumage. BLACK CRAKE, Limnocorax flavirostra (Swainson). Only two brief glimpse so far on the dam, but it is now resident on larger and better- covered dams in the district. No records prior to 1964. AFRICAN MOORHEN, Gallinula chloropus meridionalis (Brehm). Two records of single birds on this farm, and it is an uncommon resident on local dams, breeding in November. RED-KNOBBED COOT, Fulica cristata Gmelin. Resident, breeds. One pair appear to have bred three times in less than 12 months, and reared a total of 10 chicks, nesting in September, February and July 1961/1962. As the young grew, they seemed to take on the function of tending the latest hatched, as immature birds were seen attending tiny chicks and feeding them, each bird to a separate chick; this is a habit normally employed by the adults. The numbers on the dam vary, and there are generally one or two pairs, which breed January and February, June and October. Bad- tempered, they do much to frighten av/ay other water-fowl, and I have seen them chase Egyptian Geese. SOUTH AFRICAN CROWNED CRANE, Balearica regulorum (Bennett) Resident, breeds. Two pairs always reside on the farm, one by the dam, the other a mile away by a small weir. They nest regularly, but have difficulty in rearing young. The young stay with parents for 8-9 months, and the adults may then start to nest again immediately. Eggs have been laid in most months of the year. The immature birds appear to join a small flock a few miles away, on leaving their parents. JACKSON’S BUSTARD, Neotis denhami jacksoni Bann. Uncommon visitor to farm, but may be resident nearby in the Masai, where it is more open, and where I have seen it from time to time. JACANA, Actophilomis africanus (Gmelin). Two records only, April 1963 and June 1964, single birds that only stayed a day. RINGED PLOVER, Charadrius hiaticula Linnaeus. Uncommon winter passage migrant. One bird seen on 17.iv.66 was with a party of 6 Common Sandpipers. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 30 THREE-BANDED PLOVER, Charadrius tricollaris Vieillot. Passage migrant seen most years, generally November to January, and June to August, staying for a few days. Call is high and staccato, “peet peet”, when alarmed, and a soft “pit-pit-pit” on alighting. At other times the single alarm call is run into a variation of a number of notes. Nearly always within a short distance of the dam, often in pairs. CASPIAN PLOVER, Charadrius asiaticus Pallas. Uncommon winter migrant, arriving August and September, and not seen much until return passage in April. Groups number 3 to 20, and are often in larger flocks of the Blackwinged Plover. Their alarm call note is a soft “tsik tsik”. My dates of arrival are a lot earlier than given in Praed and Grant, and I have a good sight record for a pair on the 28th and 29th August 1962. CROWNED LAPWING, Stephanibyx coronatus (Boddaert). Rare visitor: two birds stayed a few days in April 1965. BLACK-WINGED PLOVER, Stephanibyx melanopterus (Cretzschmar). Partial resident, which undergoes considerable local migrations. Thus, the first single birds appear in late January, and numbers increase slowly to March/April, when breeding commences if conditions are right, otherwise not until May/June. By June, the early juveniles flock with the non-breeders, and by July there are flocks of 100 or more over the district. In August/September these have grown into vast flocks of over 1,000, perhaps even to 10,000, and then they all suddenly disappear, although they can be heard passing over at night or in thick mist during this period. They have been seen down on the Narok plains, about 40 miles south of the farm and 3,000 ft. lower, and this may be the ground to which they move. The bulk of the breeding takes place in May /June, 2-4 eggs being laid in what can only be described as a scrape, although bits of straw and sheep droppings are added some- times. The eggs are often laid on newly turned earth, the action of a tractor cultivating the soil causing a remarkable stimulation of egg-laying. As the tractor passes, the hen birds squat down in broody postures all around, and by the time the tractor returns the first egg has been laid, and the bird sits in a threatening attitude until the tractor is within inches of its bill. I have observed this sequence of actions on several occasions, but the occurrence must depend on many synchronising factors. It is hard to say what happened before the arrival of tractors, but perhaps the ploughed land has attracted a concentration of breeding birds. The liking for bare earth could perhaps be due to the increase of soil temperature from cultivations, or to the improved camouflage against the darker earth. As the birds fly in they make a vivid pattern of black and white, but the moment they close their wings they become nearly invisible, so perfectly do their colours blend with the soil; the eggs too are very hard to see on the bare ground even when the nest is marked. AVOCET, Recurvirostra avosetta Linnaeus. One record of a small party on dam in 1954, Mr. & Mrs. Grainger saw a single bird in a puddle on their farm road, about three miles from here, 27.iv.65. BLACK-WINGED STILT, Himantopus himantopus (Linnaeus). Occasional visitor, staying several days at a time, in two’s and three’s. Records from late September to March, suggesting birds are northern migrants. GREAT SNIPE, Capella media (Latham). Four records, for December, February and March; single birds only. AFRICAN SNIPE, Capella nigripennis (Bonaparte). Resident, although numbers fluctuate according to weather. In the dry season only one or two birds may be found in isolated damp spots, but after prolonged rain, there may be fifty pairs on the farm. I have found newly hatched chicks, but the only date for them is October 1964, and I have been unable to find a nest. If breeding coincides with display, it would appear that the Snipe breeds at any time of the year when conditions are suitable, although this would tend to occur mainly in July /August. Display consists of two types; one in which the bird stands on a low mound or a fence post, and utters a rapid, piping “pic-pic-pic-pic” for about a 5-second burst, and continues for a quarter of an hour or more. The other is the drumming which resembles that of the European species C. gallinago Page 31 Notes on the Birds of Mau Narok (Linnaeus), but is not identical. The African species makes a circular flight, climbing rapidly to about 50 ft., then diving almost to the ground with a few quick wing beats, and making the drumming sound with its tail. The drumming bird has been seen to do this in circles round a standing piping bird, and I thought they might be male and female, but I have seen one bird make both sounds, so it might have been coincidental. The “scape” alarm call is also used as an antagonistic note during chase, repeated several times. CURLEW SANDPIPER, Calidris testacea (Pallas). Two records on dam; single birds on 7.xi.60 and 14.iii.61. LITTLE STINT, Calidris minuta (Leisler). Regular winter visitor and passage migrant, some birds staying on dam for several months, others passing straight on. Usually 3 to 10 birds. Early date 15th August, late date 19th April, when birds were in breeding dress. TEMMINCK’S STINT, Calidris temminckii (Leisler). One record, a single bird with 7 Caspian Plovers in a waterlogged wheatfield. 7.X.62. RUFF, Philomachus pugnax (Linnaeus). Irregular winter passage migrant. Most of the ones I have seen here have had orange legs. Latest date, 19th April. COMMON SANDPIPER, Tringa hypoleucos Linnaeus. Winter visitor and passage migrant. Those that stay appear to prefer the rocky streams, where they may be found for 2 or 3 months, while those passing through are generally seen on the dam. An early arrival was noted on 11th July, and another on 15th August. Although these might have been resident Kenya birds, they seemed from their exhausted state to be migrants. These birds are probably on their way much further south, as their stay is very brief. The winter visitors that come to stay arrive about October. Late date, 21st April, bird in fat condition and breeding plumage. GREEN SANDPIPER, Tringa ocrophus Linnaeus. Winter visitor, in varying numbers, but usually 3 or 4 single or paired birds on farm. Prefers wet grassy streams, but also seen on dam. On 17.iii. 63, a pair was seen resting on the dam spillway, in bright plumage, and the male called a continual “tweeeeeee twi twi twi twi”, until alarmed, when they flew off with their usual call. From Witherby’s remarks this would appear to be their song. Earliest arrival 29th August; late date, 20th April. WOOD SANDPIPER, Tringa glareola Linnaeus. The commonest of the winter visitor Sandpipers, and seen in one’s and two’s, although sometimes passage migrants appear in small parties. Eight birds were seen in late December, passing through, and on 27.iv.62 at 2.30 p.m., I saw 20 resting birds on the dam weed, which had gone again in two hours. Latest date, 5.V.61. MARSH SANDPIPER, Tringa stagnatilis (Bechstein). Uncommon winter passage migrant, with three records, a pair and two singles, for October, November and December. GREENSHANK, Tringa nebularia (Gunnerus). Uncommon winter passage migrant, usually in pairs, once a party of six. Generally October to December; late date 16.iv.64. TEMMINCK’S COURSER, Cursorius temminckii Swainson. Occasional passage migrant, singly or in small numbers. GREY-HEADED GULL, Larus cirrocephalus Vieillot. Occasional visitor, staying for a few days on the dam, from one to four birds at a time. WHITE-WINGED BLACK TERN, Chlidonias leucoptera (Temminck). One bird on dam, 4.i.64. Another bird was seen by Mrs. Lutyens on her dam in Novem- ber 1965. 7. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 32 OLIVE PIGEON, Columba arquatrix Temminck & Knip. Local migrant, undergoing large movements of vast flocks; some years the pigeon is resident all 12 months, other years it occurs only sporadically. Breeding may take place either rarely, or on and off through the year, chiefly November/December. The display flight consists of a few slow wingbeats taking the bird about 20 ft. above the tree tops, and then a stiff-winged glide to another tree about 50 yards away, during which it utters a loud mewing sound; perhaps described as a nasal “twaa twaa”. This is best heard on a damp misty evening, when the birds sound as loud as a flock of bleating goats. The song is a deep musical “boobooboohoo”. PINK-BREASTED DOVE, Streptopelia lugens (Ruppell). Very common resident, which may breed throughout the year. It congregates at certain times into large flocks of 100 to 200, sometimes many more, causing darnage to newly planted wheat, when they pick up the grain. They roost in the forest, but flight off to feeding grounds in the early morning. They are always moving about, usually in small parties, probably due to their need for water. RED-EYED DOVE, Streptopelia semitorquata (Ruppell). Only three records. RING-NECKED DOVE, Streptopelia capicola (Sundevall). Unrecorded for 7 years, but now appears to have spread west from Rift Valley escarp- ment, and is resident in small numbers, mainly near tree plantations. Nesting April, 1962, juvenile, June 1961. LAUGHING DOVE, Stigmatopelia senegalensis (Linnaeus). One record, a very wild bird on edge of Masai grassland, 14.viii.62. NAMAQUA DOVE, Oena capensis (Linnaeus). One bird stayed for about a week in March, 1965, usually being seen on a dusty road; another was seen a few miles away by Mrs. T. Hamilton-Fletcher, also on a farm road. No other records. CUCKOO, Cuculus canorus Linnaeus. Both subspecies, the European (C.c. canorus) and African (C.c. gularis Stephens), are rare passage migrants through the farm. One bird seen in forest on 27.X.62, was the hepatic variation of the European race, having a bright chestnut plumage, barred blackish, with no white markings, and corresponding exactly with the description in Witherby. A bird of the African race was seen to eat a large and very hairy caterpillar, ll.xi.63. GREAT SPOTTED CUCKOO, Clamator glandarius (Linnaeus). One record, 21.xii.65. BLACK-AND-WHITE CUCKOO, Clamator jacobinus (Boddaert). Uncommon passage migrant, being seen once by me, in November 1964, and twice by Mrs. Lutyens, in July and September 1965. HARTLAUB’S TURACO, Tauraco hartlaubi (Fischer & Relchenow). Common resident in the forest, although some years it becomes scarce from July to October. I have not yet found a nest or other signs of breeding. I would expect them to breed at the turn of the year, as from May onwards they are seen in small flocks of up to 20 birds. Unlike many other forest birds, which follow the riverbeds away from the true forest, I have not yet seen this bird in the open at all, as suggested in Praed and Grant. RED-HEADED PARROT, Poicephalus gulielmi (Jardine). A resident of fluctuating numbers, which like the last species is sometimes absent during the middle months of the year. Again, I have found no signs of their breeding, and there is no description of their nest in Praed and Grant. Often they are to be seen in the early morning setting off for their feeding grounds, accompanied by loud whistles and screeches, in bands of a dozen or two, often with large numbers of Olive Pigeons. Page 33 Notes on the Birds of Man Narok EUROPEAN ROLLER, Coracias garmlus Linnaeus. Rare winter passage migrant, as I have only three definite records for April 1964 and 1965 and November 1965. However they are more often seen singly, on farms to the east, and their route seems to lie nearer the floor of the Rift Valley. LILAC-BREASTED ROLLER, Coracias caudata Linnaeus. Two records only, for September 1961 and October 1962. RUFOUS-CROWNED ROLLER, Coracias naevia Daudin. One record, December 1962. [MALACHITE KINGFISHER, Corythornis cristata (Pallas).] A small Kingfisher with red underparts has been seen three times on this farm on the small streams by different people, and I myself have seen this bird on a small dam on another farm nearby. As I have seen this species at Narok, this is likely to be the one that occurs here. BEE-EATER, Merops apiaster Linnaeus. Regular passage migrant, mainly on the spring movement. Generally it passes south in late September to mid-October, and returns mid-March to end of April, late dates 8th May 1964, 9th May 1965. Parties of about 20 birds are most usual. WHITE-THROATED BEE-EATER, Aerops albicollis (Vieillot). One record for July 1964, a party of six birds seen resting and occasionally feeding by the dam. A large group of mixed African swifts, swallows and martins were around and the bee-eaters may have been in loose company with them. Their flight cal! is similar to that of M. apiaster, and they may be over-looked when passing overhead. CINNAMON-CHESTED BEE-EATER, Melittophagus oreobates Sharpe. Local migrant, which visits the farm for about six months during which it breeds. It arrives about October, breeds in December/January, and leaves in April. The call is a sharp, high-pitched, metallic “deck”, and is often the first hint of the birds’ presence. The song is a little tinkle of about a dozen notes of a variation of this call-note, and is not often heard. CROWNED HORNBILL, Tockus alboterminatus (Biittikorfer). Local resident, not always present on farm, but display and song observed, so must breed nearby. They are especially fond of the berberis fruit, which takes them out over the grassland some distance from the forest. When they come to the garden their main quarry seem to be chameleons, which they are very adept at finding. Normally seen in parties of 4-6 birds along the forest edge or up the wooded valleys. GROUND HORNBILL, Bucorvus leadbeateri (Vigors). Resident, sometimes absent for 'short periods. A very young bird, seen with adults in October, had the bare skin around the head a dull straw colour, and the plumage itself was duller black than that of the adults. The feathers were very rough and quite lax, and the bird appeared to have left the nest recently, as it showed no fear of me at all. The female v/hich had no casque, had the bare skin red, not blue as in Praed and Grant, and v/as very courageous in staying by the young bird, while the male ran on ahead, blowing out its wattles. Another female I observed had the bare skin round the face entirely blue; this bird was perched on a post near the borehole pump which was making a dull clanking noise, and in time to this mechanical noise, the Hombill was making a low booming sound in reply. It kept this up even when I approached to within a few feet and its wings were out-stretched ready to fly. When it flew off it alighted on the edge of the forest, when it started booming again. As it boomed it blew its blue wattles right up. sucking them in again after each series of notes. This description compares with Praed and Grant’s note that the female may make an answering call on a lower tone; possibly the bird mistook the deep clank of the well-head for the sound of a male hornbill ? GREEN WOOD-HOOPOE, Phoeniculus purpureas (Miller). Passage migrant, single birds on 6.X.62 and 10.xi.63. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 34 WHITE-HEADED WOOD-HOOPOE, Phoeniculus bollei (Hartlaub). The common resident Wood-hoopoe of the cedar forest. In parties of about a dozen birds, their presence proclaimed by their loud chattering cries; these are on two notes, one harsh, the other a more whistling one; they also make a soft piping “chi chi, peepeepeepee”. Young, with streaky crown, seen in February. BARN OWL, Tyto alba (Scopoli). One record, 22.i.64. AFRICAN MARSH-OWL, Asio capensis (Smith). One record for l.iv.66. MACKINDER’S OWL, Bubo capensis mackinderi Sharpe. Resident, at least 4 pairs nesting withm two miles of the house. They are partial to the lightly timbered valleys with a small stream running down, and they make their nests on the bare earth under some light bush or overhanging rock, above the water. Normally they roost in the cedar trees by day, not necessarily against the trunk; but where the trees are absent they become ground roosters, talcing cover in heather or bunch grass. Mr. Stephenson, on the next farm, found a pair of these birds nesting on a grassy slope some 30 ft. above a 3 acre dam. Three eggs, measuring 58 X 45, 58 X 46 and 59 X 46 mm, were laid before 26.ii.64, and hatched about 18.iii.64. This pair had another clutch, found on 29.xii.64, and these eggs measured 59.5 X 45, 58 X 45 and 56 X 45 mm. The chicks were hatched with white down, and by a month old had acquired a greyish-brown, rather sooty plumage, with dark brown barring from nape to tail. The eye was a chrome yellow with bright blue pupil; feet and bill dark horn. At a later stage they show a conspicuously pale face marking, surrounding the gape. It was difficult to tell the sexes of the adults apart and to say if the male took part in incubation, but he was found quite near the nest in daytime. The three chicks were wantonly destroyed at two months old, as were some others Mr. Stephenson found in another nest, probably by Africans for superstitious reasons. Large nestlings have been found, in November (three times) and in January, small ones in December and March, young birds seen flying in January, April and May. Working back from these dates, egg-laying would appear to take place from August to February. The same nest is used many times, and then may be moved only a few yards. From casts and from litter in abandoned nests, it appears that food consists mainly of mole-rats; twice I have found crab shells. They have a deep hoot, usually with a long and short note: “hooooo...hu”; in addition they make a barking “wak wak”, and this may be heard with the hooting, but it is also an alarm call, as I have seen a bird make it as it watched my dogs running below its tree-roost. An immature bird made a harsh, staccato squawk when mobbed by shrikes. The bird is well distributed, as nearly every valley has its pair, and local Masai tribesmen, who know it well, confirm this. VERREAUX’S EAGLE-OWL, Bubo lacteus (Temminck). Rare visitor, April 1964 and March/April 1965, when it grunted through the night outside the house. NIGHTJAR, Caprimulgus europaeus Linnaeus. Two records identified from male specimens picked up from the road, 29.iii.61 and 2.iv.65. Birds not identified in the hand but appearing to be this species are often seen in the winter months, and it may be that it is a common winter visitor. ABYSSINIAN NIGHTJAR, Caprimulgus poliocephalus Riippell. Common resident, and partial migrant. SPECKLED MOUSEBIRD, Colius striatus Gmelin. Common resident along the forest edge; sometimes forms flocks of up to 100. NARINA’S TROGON, Apaloderma narina (Stephens). Observed twice in March 1961; not since. MOUSTACHED GREEN TINKER-BIRD, Viridibucco leucomystax (Sharpe). Generally common and vocal from October to April; then it becomes silent. I have seen it once or twice in July and August, but in spite of it being a difficult bird to see, I suspect that numbers move away from May to September. The song is very loud for such Page 35 Notes on the Birds of Man Narok a small bird, a very fast “pip-pip-pIp-pip”, accented on the last syllable. This is repeated with scarcely a pause over and over again. Tire note changes at times to become either higher or lower, but the phrase is always constant. I have only seen these birds feed on small berries. GOLDEN-RUMPED TINKER-BIRD, Pogoniulus bilineatus (Sundevall). Less common than the Green Tinker-Bird, with a shorter song-period, from October to December, once in February. The song is the well known “quok quok quok quok quok”, at about a fifth of the speed of the last species, and nothing like so shrill. I have seen them in May and August, silent, so it may be that some birds remain the year round. They are also capable of making a loud hissing sound, just like a snake’s, although I do not know for what reason, BLACK-THROATED HONEY-GUIDE, Indicator indicator (Sparrman). This bird is only known from its distinctive song, which is heard from January to March, and from one brief sight record. The song has only been heard from one small area on the edge of the farm, and it would appear that at the most there is only one pair, which may be migratory. LESSER HONEY-GUIDE, Indicator minor Stephens. Resident, possibly only one or two pairs. Inconspicuous outside song period, which extends from July to November. Tlie song of one bird is delivered from the same branch high up in an olive tree year after year, and consists of “wheeoo, pleep pleep pleep”, the “pleep” repeated 14 or 15 times at one second intervals, with a three minute pause; the cold dri’zly weather which occurs at that time of the year does not seem to deter the bird from singing. I have seen this bird deliberately perch amid a swarm of wild bees, peering around intently as if to see where they were coming from. This Honey-Guide is usually found in the forest, but sometimes up the river valleys. FINE-BANDED WOODPECKER, Campethera taeniolaema Reichenow & Neumann. Resident, possibly the commonest of our three resident woodpeckers, although these species are often seen in company with one another. Young seen in August and December. Their call is a loud “che che che che”, but they are rather silent birds. Twice I have watched two females feeding close together, which kept up a low buzzing note that sounded like “chwoor-ic wor-ic”. This may be compared with J. H. Owen’s description of a parent Great Spotted Woodpecker, Dendrocopus major Hartert, which he gives as a conversational “Too-ut”. (Quoted from Witherby). CARDINAL WOODPECKER, Dendropicos fuscescens (Vieillot). Resident, and fairly common. Young seen in August and January. They have a number of calls, one of which is a deliberate “keekeekeekee” and another a strident rattle. BEARDED WOODPECKER, Thripias namaquus (Lichtenstein). Probably resident. It is rather less common than the other two Woodpeckers, and may have been overlooked. No breeding records. A fair amount of drumming is heard on the farm, but I have not yet managed to identify the bird performing. GREY WOODPECKER, Mesopicos goertae (Muller). Rare passage migrant, the only records being for November 1963 and March 1965, the birds crossing some open fields near the house. RED-BREASTED WRYNECK, Jynx ruficollis Wagler. Fairly common resident, and there may be twenty pairs on the farm, occurring mainly on the forest edge and up the lightly timbered valleys, but sometimes wandering across more open country where they can examine a line of fence posts for food. Their common song is either a harsh “chwoi chwoi chwoi” or a softer, piping “twee twee twee”, uttered from a bare bough, the bill nearly closed, and the head thrust forward with each note. Before breeding commences in May, three or more birds will be found in a tree indulging in a lot of display, with much posturing, jumping about and excited calls. The red under- tail coverts are brought into prominence by the displaying bird perching over the head of another. The birds ca'l a continuous “tuk tuk” and I have heard a wheezy chirrupping song at the same time. Witherby described the “tuk tuk” call as an alarm, in the article on the European Wryneck, Jynx torquilla, Linnaeus, but with the present species I have noted the alarm call J. E. Air. nat. Hist. Soc. Vol. XXVI No. 1 (113) as a much more metallic “chip chip chip”, the “tuk tuk” being used for display. The birds may be heard throughout the year and at any time of the day, though least from December to March. The dark Une down the centre of the back is a good field character, and in flight, which is slightly dipping, the bird looks chunky, and rather larger than most small brown birds. COMMON SWIFT, Apus apus (Linnaeus). As Praed and Grant remark, the identification of Swifts in the field is extremely difficult, especially if one has had little experience. They are most noticeable when heavy storm clouds are about, as these bring them lower. I include the present species because large numbers of swifts pass through in April together with House Martins, flying rapidly East and obviously migrating. [NYANZA SWIFT, Apus niansae (Reichenow.)] Ordinary brown swifts are present throughout the year in fairly large numbers, and I consider that they may be this species. I have found no sign of nesting. WHITE-RUMPED SWIFT, Apus caffier (Lichtenstein). Three or four birds amongst a party of African Sand Martins, on 20.ii.62, answered closely to this species’ description. I noted deep-forked tail, glossy blue-black underparts and small size. HORUS SWIFT, Apus horus (Heugl). Regularly seen in small numbers from the end of April to early July. RED-CAPPED LARK, Calandrella cinerea (Gmelin). Resident with some local movement. Breeds in April and May, flocking into some hundreds from July to October, after which more breeding may take place. A hen flying from a nest on April 24th had two eggs, 22mm X 15mm, which were heavily blotched earth brown on a coffee-coloured background. The male has a fine song which includes snatches of songs of other grassland birds, and it is uttered from the ground, from a fence post or in display flight up to 300 ft. above ground. They are aggressive birds, and are often seen chasing Black-winged Plover; their song includes good imitations of this bird, as well as several others such as Stonechat and Streaky Seed Eater. AFRICAN PIED WAGTAIL, Motacilla aguimp Dumont. Rare visitor until 1963, when a pair bred in June, and there are now two or three pairs breeding round the buildings. This is one of the birds that has been colonising slowly west- wards with the spread of cultivation and habitation. MOUNTAIN WAGTAIL, Motacilla clara Sharpe. Regular visitor, but not common; it may breed further down the valleys in the forest. They are generally in pairs and sing frequently, both on the dams and near the rivers. At Molo and Elburgon I have seen their nests on the spillways of dams at 8,000 ft. and 9,000 ft; the latter on the 20.vi.64 had two chicks just hatched and one egg, 21mm X 14 mm, had small pinky-brown splashes over a buff background. The nest was a deep cup made of bunch-grass leaves woven into the tussock about 12" above ground. WELLS’ WAGTAIL, Motacilla capensis wellsi O. Grant. Only two records, September 1962 and December 1963. Both were for tame silent birds on the dam; these seemed rather exhausted, and I assumed them to be migrants. GREY WAGTAIL, Motacilla cinerea Tunstall. Uncommon winter passage migrant occurring singly. Early date SeptemOer 25th. Occurs on fast flowing streams in or near the forest; one bird stayed for a week. I have seen this bird on the Aberdares at 10,000 ft., on a dam at Molo at 9,000 ft., and at Elmenteita at 6,000 ft., (in March) but nowhere does it appear common. YELLOW WAGTAIL GROUP, Budytes general notes. Three species, including a total of five races, occur as winter visitors; as they are not really identifiable on arrival, and their habits are very similar, it is proposed to treat them as a group before listing them individually. A few birds arrive early in September, and by the end of October small parties of up to 50 birds are scattered over the farm, mostly among Page 37 Notes on the Birds of Mau Narok the flocks of sheep. When ploughing starts in earnest in March, they congregate in numbers round the tractors and are difRcult to count, but run into many hundreds. They become very tame at this time, hardly moving out of the way of the implements; from the driver’s seat, with binoculars, one can see every detail of their plumage, so that adult males of the various species can be recognised. B. luteus would appear to be the most common, followed by B. fiavus, with B. thunbergi only occasional. On the 21.X.62, I saw an interesting display by two birds out of a small flock of Budytes. I considered them to be immature B. fiavus, both very pale ash-grey above, almost white below, except for the yellow under-tail coverts and brown pectoral band. The birds shuffled round each other in a dusty path among the sheep in a squatting fashion, with their mantle feathers raised on end, their tails right over their backs, and the yellow ventral feathers puffed out. Since the birds were immature, this observation, though of interest, may have no breeding significance; however, two further observations do indeed appear to indicate that occasional birds may breed locally, even though this has not yet, so far as I know, been recorded for Tropical Africa. The first observation was for a single B. luteus singing in November, and the second for a bird in juvenile, not immature, plumage in April. Witherby maintains that the moult from first winter plumage to adult takes place in January, so that this juvenile might well have been born in winter quarters. BLUE-HEADED YELLOW WAGTAIL, Budytes fiavus (Linnaeus). Regular winter visitor, all birds appearing to correspond with the nominate race. YELLOW WAGTAIL, Budytes luteus (Gmelin). Very common winter visitor; an early date, probably for this species, was 8.ix.61, but the majority stay here between October and April. A bird I saw on 5.V.64 looked sick, and another 1 saw on 20.V.63 was very wild, and appeared a straggler. Jackson doubts the preference these birds have for following livestock, thinking it might be the type of ground they prefer, but there is no question that Yellow Wagtails like feeding in a flock of sheep, which doubtless disturb the insects for the birds to feed on. The Wagtails even perch on a sheep’s back. Both races of B. luteus occur, as a few birds in March may be seen with the whole head a clear yellow, and these must be old males of B. t. luteus. DARK-HEADED YELLOW WAGTAIL, Budytes thunbergi (Billbcrg). Not a common winter visitor, and not usually identified until March or April. One bird I saw on 26.ii.63 was in brilliant spring plumage, and had the chin and throat pure white extending back to the neck, while another bird had a yellow chin. It would thus seem that both races, B. t. thunbergi and B. t. cinereocapillus occur. I have only seen this species on ploughland. RICHARDS PIPIT, Anthus novaeseelandiae Gmelin. Common resident, identified from specimens sent to Mr. J. G. Williams. Breeds January and June, probably other months as well. The bird is found on cultivated as well as indigenous grassland. Other species of pipits sometimes occur, but not yet identified in the field; a large pale bird I have seen occasionally may be the Sandy Plain-backed Pipit, Anthus vaalensis Shelley. TREE PIPIT, Anthus trivialis (Linnaeus). Winter visitor, but not in large numbers. Their arrival dates are regular, but later than most migrants, viz. 7.xi.60, 7.xii.61, 26.xi.62, 3.xii.63, and 29.X.64. They leave about the end of March. Very tame indeed, one pair spending two winters around the kitchen door, although generally they are to be found near the forest edge, flying up into low branches when alarmed. It sometimes takes short soaring flights and its “cheez cheez” call may then be heard. RED-THROATED PIPIT, Anthus cervinus (Pallas). Winter visitor, fairly common but rather wild; partial to damp and marshy places. Arrives October/November and leaves end of March. Found singly or in small flocks, with one or two birds usually showing reddish cheeks; but the surest guide to identification is their call note, a very high thin “teep” or “tseep”, reminiscent of a Redwing, Turdus musicus Linnaeus. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 38 SHARPE’S LONG-CLAW, Marconyx sharpei Jackson. Common resident, which breeds April to August. Fairly tame and seldom flies far. The song, usually uttered in flight as it circles about 30 ft. up, and heard throughout the year, is a thin, plaintive “tip tee teu” or “tip tip teee” this is expanded a little during the breeding season. The bird is found only in pairs or with yoUng, on the open grassland. A nest found under a small tussock on 18.i.66 contained two newly-hatched young, and one egg measuring 24 X 17 mm; the egg was splashed all over with light brown. ABYSSINIAN HILL-BABBLER, Pseudoalcippe abyssinicus (Riippell). Fairly common resident throughout forest; sings all the year, but most strongly in April and May and again in November and December. Normally in pairs, sometimes seen in little flocks. The call note is a low chuckle “kwa kukukukkluk”, or else a soft “quup quup” similar to the Greenbul’s. The song as remarked in Praed and Grant, is very fine though rather short; occasionally I have heard a longer burst. The pure flute-like quality of the tone is the outstanding feature; for this, the bird can have few rivals. DARK-CAPPED BULBUL, Pycnonotus tricolor (Hartlaub). Very common resident which congregates after breeding in June and July, often in com- pany with numbers of Olive Thrushes. OLIVE-BREASTED MOUNTAIN-GREENBUL, Arizelocichla tephrolaema (Gray). Well distributed but not very common throughout the forest. It is a silent bird, so may well be overlooked a certain amount. The only sounds I have heard are a low “quup quup”, a scolding “schurr” and a thrush-like cackling; and once I noted a low bulbul type of song which was unremarkable. These birds are often associated with Hill-Babblers and are found frequently feeding within feet of each other. I have no record of their breeding except for a family party including young birds in April 1964. SPOTTED FLYCATCHER, Muscicapa striata (Pallas). Only one record, for 15.iv.63, the bird unexpectedly using large clods of earth on the edge of a ploughed field as a perch from which to catch insects. It was a greyish bird, which might imply that it belongs to the Asiatic race, M.s. neumanni Poche. DUSKY FLYCATCHER, Alseonax adustus (Boie). Common resident throughout forest and along forest edge, as well as in the garden. It breeds from December to February and from May to August, and the fledglings are very conspicuous when they are being fed by their parents. They are usually in pairs or in family parties, but sometimes before breeding one may see a number of them chasing each other around among the treetops with a lot of excited twittering. The call note is a very high squeak “tit” or “tsit”, and the song is an expanded variation of this, rather stuttering. They are tame little birds, and when they turn in the air back to their perch after catching an insect, their wings give an audible snap. WHITE-EYED SLATY FLYCATCHER, Dioptrornis fischeri Reichenow. Common resident in garden and forest edge. Breeds from May to July, once in March. Song heard only rarely, and seemed as high as a bat’s squeak, unlikely to be heard except at very close quarters. MOUNTAIN YELLOW FLYCATCHER, Chloropeta similis Richmond. Common resident in forest. Fine songster through the year. The alarm note is a sharp “chak chak” similar to that of the Blackcap or Brown Woodland-Warbler. Generally in pairs. No breeding dates, but a display was seen 5.X.62, when a bird was seen singing, raising its crest, swinging its head from side to side, and moving the tail and body in opposite direc- tions. The mouth was opened wide and displayed to the second bird, the colour being a vivid red. BLACK-THROATED WATTLE-EYE, Platysteira peltata Sundevall. One record, a female, on 30.ix.63, on the forest edge. The white on the chin appeared as only a spot. WHITE-TAILED CRESTED FLYCATCHER, Trochocercus albonotatus Sharpe. Resident through the forest, generally where it is damper, and almost always in small parties of 5-6. A very active bird that flicks its wings and fans its tail without ceasing, and Page 39 Notes on the Birds of Man Narok continually utters its call note, a sharp “ti-ek”. The song is heard less often, mainly in June and July, and it is rather a disjointed and staccato effort, but with many fine notes; I have heard it mimic the Brown Woodland Warbler accurately. PARADISE FLYCATCHER, Tchitrea viridis (Muller). Uncommon visitor or migrant, from July to November. The black and white variety has been seen once, deep in the forest. OLIVE THRUSH, Turdus olivaceus (Linnaeus). Very common throughout farm except where there is no cover at all. In the garden it becomes very tame and rather a pest, due to its fondness for fruit. It breeds from March to June and again in November. TTie song is a wild strident call of three or four notes “chow chee cher, chow chee cher chee”, which does not seem to tally with Praed and Grant’s descrip- tion of a “low, sweet song”. The song period is short compared with that of other song birds. Flocks of up to 30 birds occur in July and August, sometimes with numbers of Bulbuls, when they fly high through the forest amongst the cedars, feeding silently in the topmost branches of the trees. ABYSSINIAN GROUND-THRUSH, Goekichla piaggiae kilimensis Neum. Uncommon bird of the densest and dampest parts of the forest, where it may be resident. I found an adult with a juvenile in June 1961, the young bird making a very high pitched “seeep” the whole time; it was tawny above, with black spots on a brown background under- neath. I have recorded the song in April and August, and seen birds apparently taking food to the nest in April and May, so the breeding season seems to be during the middle of the year. The song is loud and thrush-like, and well-described by van Someren in Praed and Grant for the nominate race, but it is a shy singer and not often heard. The race was identified for me by Mr. J. G. Williams from a specimen. WHEATEAR, Oenanthe oenanthe (Linnaeus). Winter passage migrant, the first birds passing through late September or October. It is not common, and nearly always seen singly. Only an occasional bird is seen after October until the return migration in February, when the birds are seen in breeding dress. None have been seen after the middle of March. PIED WHEATEAR, Oenanthe leucomela (Pallas). Rare winter migrant, two records for early March, one in December, the March ones in company with Oe. oenanthe. In flight the tail appears entirely white, and at rest almost black. CAPPED WHEATEAR, Oenanthe pileata (Gmelin). Resident, and perhaps a partial migrant, breeding in April /May, exceptionally in January. Originally this wheatear was common throughout the district, but for some reason the increase in cultivation has caused a diminution of numbers; probably the rat holes they use for nesting have been ploughed out. Outside the farm boundary on the Masai grasslands the bird is still common. Its short song period is restricted to the start of the breeding season, but it has a very fine, sustained song at this time. The song consists of a series of loud pipes mingled with grating sounds, and a number of mimicked phrases interspersed. The mimicry includes calls of Quail, Blackwinged Plover, Glossy Starlings and Red-Capped Lark, but the ability varies with individuals. ANTEATER CHAT, Myrmecocichla aethiops Cabanis. Common resident in the open country in the vicinity of earth banks. Breeds April /May. During display, up to 15 may be seen together, grouped on a row of fenceposts, the males(?) keeping up a continuous piping, with outstretched wings and up-pointed bills. They display indiscriminately, with four or five birds sometimes on a single post. STONECHAT, Saxicola torquata (Linnaeus). Common resident from forest edge to open grassland. Breeds March to June, usually 3 to 4 eggs in a nest placed in the shelter of a tuft of grass or heather. ROBIN-CHAT, Cossypha caffra (Linnaeus). Common resident where there is any cover, breeding from April to August. A tame bird in the garden, and a fine songster. Although they usually lay three eggs, they seem only to be able to raise one or two young at a time. L E. Afr. nat. Hist. Soc. Vol. XXVI No. I (113) Page 40 WHITE-STARRED BUSH-ROBIN, Pogonocichla stellata (Vieillot). Fairly common forest resident. The song is a series of soft piping notes followed by a bubbling warble, and sometimes a croaking rattle. Young have been seen being fed in March, July and September to November. A field character that is quoted in all reference books is the white spot at the base of the throat; I have observed more than 30 birds at very close range, some of them in display when their eyespots appear raised and enlarged, and 1 have been unable to discern any sign of the spot on the throat. This spot is also shown prominently in the illustrations in these works, including Chapin’s, but Chapin states he never saw this spot after watching more than a dozen birds. Priest (Vol. Ill p. 196) made the same comment, but attributed the characteristics to immaturity. It certainly appears to be common to all races of the species. [GARDEN WARBLER, Sylvia borin (Boddaert).] I have three records for a typical Slyvia, two in April and one in November, which might be for this species. The birds appeared very pale ash brown above, and whitish below. BLACKCAP, Sylvia atricapilla (Linnaeus). Common winter visitor, arriving mid-November onwards. The subsong is heard in December, and by February they are singing loudly through the forest. Some years they collect into large flocks before leaving at the end of March, and the volume of noise they make in chorus is remarkable. In 1961 I reckoned there were a 1,000 birds congregated along a mile of forest edge. In December 1963, on a cloudy, windy night, a number of birds flew into the lighted window-panes of a neighbour’s house. I have only seen them feeding on berries. SEDGE WARBLER, Acrocephalus schoenobaenus (Linnaeus). One record for 16.iv.66, when I saw a bird feeding off floating weed on a stream, and in the rushes growing nearby. CINNAMON BRACKEN-WARBLER, Sathrocercus cinnamomeus (Riippell). Common resident through forest and forest edge. A skulking bird that keeps to the cover of low bush. The song is as described in Praed and Grant, and may be recognised by the initial aspirate note. The call is a loud “seerk”. They also make a soft “prrrrip” when they have young, and this is usually the only indication of breeding. WILLOW WARBLER, Phylloscopus trochilus (Linnaeus). Winter migrant, mostly seen in April, the latest date being the 28th. One arrival record for 28.ix.60, and the remainder single birds through the winter. One bird flew into the window at 9.00 p.m., on 12.xii.63, which indicates that they move about at night in the non- breeding season like the Blackcap. CHIFF-CHAFF, Phylloscopus collybita (Vieillot). A single bird stayed in the garden close to the house from the beginning of February 1966 to the 19th of that month. The bird sang daily from low shrubs, from the tops of pine-trees and especially from a spindly Acacia baileyana. The song was always vigorous, and I sometimes heard the soft warble which may be given at the end. The call note “hweet” I also heard several times. The only bird I saw was very tame, and from a few feet away I could easily distinguish the blackish legs and buff eye-stripe. I heard a Qiiff-Chaff’s song at the end of January 1964 in the garden, but the bird did not stay long enough for confir- mation. I made a tape-recording of this bird’s song, and Mr. M. E. W. North was kind enough to confirm the identification from this tape. Mr. North suggested this was probably the first time the Chiff-Chaff has been recorded singing in Kenya. WOOD-WARBLER, Phylloscopus sibilatrix (Bechstein). One record for 3.X.62. I distinguished this bird from the Willow-Warbler by its pale green upper-parts, yellow breast and white belly; altogether a larger and brighter bird. BROWN WOODLAND-WARBLER, Seicercus umbrovirens (Riippell). Common resident in the forest. In habits and appearance very like a Leaf-Warbler; its song is also similar, but it has a wider and more varied range of notes that make it superior Page 41 Notes on the Birds of Mau Narok to that of any phylloscopine warbler that I have heard. The call is a distinctive 3-note “weer tee wew”, and the alarm is a sharp “tchuk”. GREY APALIS, Apalis cinerea (Sharpe). Common resident of the forest and forest edge. It is a conspicuous bird by reason of its loud song and variety of calls heard throughout the year, and is invariably found in mixed bird parties. The notes are adequately described in Praed and Grant. Birds in juvenile plumage have been noted in family parties in February, April and October. Display includes a deal of fanning of the tail to show off the white outer feathers. CHESTNUT-THROATED APALIS, Apalis porphyrolaema Reichenow & Neumann. Common resident of the forest. Distribution as for the previous species, with which it is often in company. It also has a distinctive song which betrays its presence throughout the year, which I describe as “pip-preeeeeeee pip-preeeee”. They are also remarkable for the variety of call notes they make, mostly being churrings and rattlings similar to those of the European tit family (Paridae) I have seen young in February and July; as with the Grey ApaUs, they appear to remain in family parties for a month or so after nesting, and then disperse to join mixed bird groups. Once the song is known, one realises how common they are, about equal in numbers to the Grey Apalis. The young have pale yellow chin and throat, (where the chestnut in the adult would be,) dull white underparts, and pale flesh legs which become almost red in the adult. WING-SNAPPING CISTICOLA, Cisticola ayresii mauensis van Someren. Fairly common resident on the open land. I can include the sub-specific name, since this was identified by Mr. J. G. Williams from a bird which Mr. North collected on 23.xi.64 from a nest. This nest, which contained three eggs, is worth describing, as Lynes was unable to do so, and Praed and Grant appear to have taken Lynes’ notes on C.a. ayresii. The nest was in a tuft of green grass, and made from thistledown, lined with a “down” of a small silvery weed. The eggs measured 17 X 12.5 mm and were blue with a circlet of red-brown spots and scratches at the large end, and a few scratches towards the narrow end. The song, which is uttered between bouts of wing-snapping, consists of a thin piping of four notes: “der der dee du”. Lynes mentions that there is no song during ascent, but I have definitely heard the bird sing as it rises into the air. This cisticola has adapted itself well to the advent of cultivation, and it may often be seen rising from a field of tall wheat or from short, dry stubble. HUNTER’S CISTICOLA, Cisticola hunteri Shelley. Very common resident of forest edge, garden and fight bush. Breeding recorded for March, May and October, and may well breed the year round. Two nests were found by Mr. Daniell in his garden in 1962, placed in some small annual flowers and quite close together. One nest had been deserted with two eggs in it, and the other had five eggs, which were duly hatched. It would seem that one female had laid seven eggs. This Cisticola seems to be subject to some albinism, as I have now found three birds affected. One bird near the house had only the flight feathers and tail white; another bird a mile away was white with a few brown markings on its back and wings, which disappeared with successive moults, until it became completely white, and the third about 8 miles away, was pure white, but I saw it only once. The value of such albinos it, of course, that they are individually recognisable and can thus give an idea of their length of fife and of their range of movement. Neither of the two birds I observed for a total of two years were ever seen more than 30 yards away from a central point, generally much nearer. One was seen for about 10 months, and the other for nearly two years, and in spite of being pure white, it seemed to be able to accom- plish its duets, when it sang with and actually touched a normal brown bird. STOUT CISTICOLA, Cisticola robusta aberdare Lynes. A common resident, chiefly restricted to the wet grassland and open valleys. There is some overlapping with C. hunteri. I have added the sub-specific name after a specimen was identified at the National Museum. The song is a rattle followed by a few emphatic notes, “trititititit trit trit trit”. The call is a nasal “chwer chwer chwer”. The young are distinctive by being bright ochreous yellow below, shading to white on the belly. Breeds in February and May. BANDED PRINIA, Prinia bairdii (Cassin). An uncommon forest bird that is probably resident. It is partial to nettles and similar low vegetation in shady forest, and generally keeps well hidden. I have only once heard J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 42 anything like a song, an undistinguished warble, but they have a number of calls some of which may be used as a song, as I have heard them duetting: one bird made a loud “pwee pwee pwee”, the the other simultaneously making “keow keow”, and when they ceased each time, both birds resumed their “chip chip” call, (“pink pink” in Chapin) which is their most usual call when feeding. This prinia is usually found in pairs, and family (?) parties of 5-6 have only been seen in May, November and December, which may indicate when they breed. SWALLOW, Hirundo rustica Linnaeus. Common winter passage migrant. They are always seen flying east and south, and at the height of migration, pass in groups of half a dozen every few minutes. I have rarely seen them perch or pause to feed, and they seem intent only on reaching the Rift Valley floor. They are seen throughout the winter months, but large numbers only occur in October, February and March. Late date l.v.64. ANGOLA SWALLOW, Hirundo angolensis Bacage. Tropical migrant, which breeds in May /June. I had thought them to be passage migrants only. However, Mr. J. R. Stephenson showed me on 20.xi.65 a small colony of these birds on the next farm, where they were nesting under a rock overhung by a stream. There were about a dozen nests, in various stages of breeding; some had new laid eggs, average clutch about 3 eggs, some with small nestlings, some with large; other nests were empty; also there were nests that had survived from previous years. Seven eggs averaged 20 X 13 mm, and were marked like those of H. rustica. From notes of other parts of the farm where I have seen them, they appear to like fairly open country, near water, and with a suitable cliff-face to build on. RED-RUMPED SWALLOW, Hirundo daurica Linnaeus. Resident and partial migrant, as sometimes it seems to disappear for a month or two. The main breeding takes place in April and May, but it is often upset if the rains are late, and it cannot get mud for building. They sometimes start building in November, but I have not found them rearing young at this time. MOSQUE SWALLOW, Hirundo senegaitnsis Linnaeus. Resident and partial migrant. A fair number on the farm in April and May, when they breed, but the majority disperse and leave from August until about December. AFRICAN SAND MARTIN, Riparia paludicola (Vieillot). A common resident, which undergoes considerable local movements, as the numbers vary from hundreds to only a few. Breeds April to July. BANDED MARTIN, Riparia cincta (Boddaert). Visitor, mainly from February to July, but never common. I have seen it carrying small bits of grass in its bill, but have not yet found it breeding on the farm; however, Mr. Stephenson found a small colony breeding on the 30th June on the next farm in a murram bank near the river. There were 2-3 eggs per nest, but breeding may have been disturbed as the eggs were cold; two measured were 22 X 15 mm 23 X 16 mm. Birds are first seen about March, and these seem mainly migratory, the later ones staying to breed. By August, they have nearly all gone, and are seldom seen again before the following year. AFRICAN ROCK MARTIN, Ptyonoprogne fuligula (Lichtenstein). An occasional visitor, which is usually seen round the eaves of the house and farm buildings for a day or two, before it disappears again. This bird is resident elsewhere in the district, and it may colonise this farm eventually. HOUSE MARTIN, Delichon urbica (Linnaeus). Winter passage migrant, rather more apparent in the autumn than in the spring months, and especially when migrating. Ofton in company with swifts, and are never seen to perch. Late date 13th March. BLACK ROUGH-WING SWALLOW, Psalidoprocne holomelaena (Sundevall). Resident, normally found within a short distance from the river, and either near or in the forest. The call is a thin, pleasant “sweeeeee”, something like a Sand-Martin’s. I found them nesting in tunnels in hard rock by a river-bed deep in the forest in February. On the farm, they nest in murram-banks, where I imagine they excavate the holes themselves. ?e 43 Motes on the Birds of Mau Narok BLACK CUCKOO-SHRIKE, Campephaga sulphurata (Lichtenstein). Visitor of uncertain status. I have only seen them on half a dozen occasions, either singly or in pairs, and they seem likely to be overlooked due to their inconspicuous habits. I have not seen a male with yellow wing shoulder. GREY CUCKOO-SHRIKE, Coracina caesia (Lichtenstein). Resident, fairly common throughout forest. In the field the bird appears entirely grey the lores, throat, wings and tail being uniform. The eye seems to have a very pale ring round it. These birds appear rather stolid and silent, but on occasions they indulge in a lot of excited chases; these take place in May and June, and again in November and December, and I believe they may breed twice a year; the only breeding record I have is of an adult feeding a mottled juvenile on 30.xi.63. The call is a very high-pitched, batlike squeak, “tsip tsip”, and the song is a number of notes run together as an elaboration of the call. LESSER GREY SHRIKE, Lanius minor Gmelin. Regular winter passage migrant, which occurs only in April, with one record for the 1st May. These are conspicuous birds with strong flight which I never fail to see yearly as they make their way north through tlie farm in ones and twos. FISCAL, Lanius collaris Linnaeus. Common resident over the whole farm, using fences and telegraph wires where no natural perches available. Breeds January to April. RED-BACKED SHRIKE, Lanius collurio Linnaeus. Regular winter passage migrant, generally seen in April, but twice in October, viz. 27.X.62, and 31.X.63. They tend to keep nearer to the forest edge, and are more likely to be missed than L. minor. The migratory dates of both these shrikes are remarkably consistent. TROPICAL BOUBOU, Laniarius aethiopicus (Gmelin). Fairly common along forest edge, in the garden and up river beds. Occasionally I have seen them gather into loose noisy flocks in order to feed in a particular forest tree. DOHERTY’S BUSH-SHRIKE, Telophorus dohertyi (Rothschild). Not uncommon resident in the forest where there is enough secondary growth. Rather skulking, and keeps to the inside of a bush, and its presence is usually given away by its call, a loud ringing “wip wip-wip”. This call is heard all the year, especially during the rains, and in the breeding season it adds a few notes to this whistle. It has a scolding rattle “crrrrr”, and a sharp alarm note “jeb”. I found two pairs with young in July 1962, and another pair with a juvenile in August. WHITE BREASTED TIT, Pams albiventris Shelley. Common throughout forest and forest edge. Often a member of mixed bird parties. Young seen in April. [GOLDEN ORIOLE, Oriolus oriolus (Linnaeus).] Three records, October 1960, and November 1964 and 1965, but not confirmed, as the birds afforded only a brief glimpse. From these dates, it would seem more likely to be this species than the African Golden Oriole. BLACK-HEADED ORIOLE, Oriolus larvatus Lichtenstein. Not uncommon, but irregular visitor generally from June to December. Frequents open woodland and garden, particularly favouring rows of pine trees, in which it finds numbers of hairy black and white caterpillars. The bird has a melodious subsong which includes a lot of mimicking, rather like a Jay’s, {Garrulus glandarius); in spite of this I have seen no sign of breeding. PIED CROW, Corvus albus Muller. One record for a pair of birds, one visibly larger than the other, which I saw around the district for a few days in August 1965. They were severely mobbed by Black-winged Plover when they ventured too close to a breeding colony, and later I saw them a few miles away trying their luck at an African shop. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) WHITE-NECKED RAVEN, Corvultur albicollis (Latham). Regular visitor to farm, generally throughout the year. Mostly seen in pairs, but in February 1962 I saw 4 birds, and in February 1963 I saw 5 together, and these may have been family parties. They have a curious affinity for Tawny Eagles, which I have observed here and on Mt. Kenya. The ravens do not seem to be aggressive, but will fly wing-tip to wing-tip with the Eagles, and then sit beside them for no apparent reason. WATTLED STARLING, Creatophora cinerea (Menschen). Irregular visitor, mostly between July and December, in small numbers from 1-6, and always in company with Oxpeckers or Glossy Starlings. They seldom stay more than a few days. VIOLET-BACKED STARLING, Cinnyricinclus leucogaster (Boddaert). Rare straggler, with two records only, from the garden. Records were for single females, one on '20.vii.61, and the other on ll.viii.64. SHARPE’S STARLING, Pholia sharpii (Jackson). A spasmodic visitor, which usually arrives about December and leaves April or May, although some years it does not come at all. There is some evidence that they may breed, as at times they are found in pairs, and there is a lot of song and display. The latter includes postures like a Wryneck’s, with the neck outstretched stiffly, and the bright yellow iris, (always conspicuous), appearing even more prominent. The song consists of several squeaky but not unattractive notes, which remind me of a rusty tricycle being pedalled down a cobbled street! The call note is a staccato “peek” repeated several times, and often precedes the song. One bird seen in a party of 20 in April 1961, was clearly a juvenile, with a dusky grey back, a dingy white front with a few indistinct streaks, and no golden iris. As adult birds had been seen paired in November, this juvenile might well have been born on the farm. The birds are always found in the forest or on the edge of it, and so far I have only seen them eating small berries, but they also cling to the bark of a tree like a Nuthatch Sitta europaea Linnaeus, and this may be to search for insects. The flight is swift and powerful, and slightly un- dulating. BLUE-EARED GLOSSY STARLING, Lampwcolius chalybaeus chalybaeus (Hemprich & Ehrenijerg). A common, at times abundant resident which breeds in the forest in April and May, and flocks from July onwards, with up to 200 in a flock, wandering over the whole farm. The wing measures up to 155 mm so this must be the nominate race, WALLER’S CHESTNUT-WING STARLING, Onychognathus walleri (Shelley). One record of several short-tailed redwing starlings in loose company with Sharpe’s Starlings flying amongst the tops of some tall forest trees. The date was April 1961. SLENDER-BILLED CHESTNUT-WING STARLING, Onychognathus tenuirostris (Riippell). Only a few isolated records of long-tailed redwing starlings have occurred, which is curious, as only a few miles to the east, at about the same altitude, they appear to be resident and fairly common. The records are all from December to May. RED-BILLED OXPECKER, Buphagus erythrorhynchus (Stanley). Common resident. All the conditions they require exist on the farm, with plenty of undipped cattle for feeding on, and tall timber for nesting in. They appear to breed in September and October. GREEN-WHITE-EYE, Zosterops virens Sundevall. Common resident of the forest and forest edge; nearly always found in small flocks in mixed bird parties. MALACHITE SUNBIRD, Nectarinia famosa (Linnaeus). Regular visitor from about September to April. The males are in breeding dress when they arrive, and there is a lot of song so I suspect them of breeding on the farm. During the months of absence I have seen them at Njoro, (2,000 ft. lower, and 30 miles away). The song is very similar to that of the Double-Collared Sunbird, and consists of a few sharp notes, followed by a high- pitched reel. The call is a thin “tsit tsit” by both male and female. This is the least common of the Sunbirds that occur on the farm. Page 45 Notes on the Birds of Man Narok TACAZZE SUNBIRD, Nectarinia tacazze (Stanley). A common resident, especially in the garden and the forest edge. Aggressive birds that always seem to be chasing each other or other species of Sunbirds. They make a wide variety of call notes including a sharp “tac tac” and some more squeaky. The song is either rather like a Chiffchaff’s viz. “chip chip chap chip chap chap” etc., or else a harsh reel. The males appear not to undergo any change of plumage in the non-breeding season; there are at least 6 of them in the garden under observation. They seem quite promiscuous and are not seen in pairs like the Malachite or Double-Collared Sunbirds. The female builds the nest which is suspended at the end of a thin branch and made of dry grass, leaves, bits of old string etc., and lined with feathers; the nests may be from 4 ft. to 30 ft. above ground. Spiders’ webs are also used in construction, and the female is often seen hovering under the eaves of the house, extracting bits of cobweb. Breeding takes place in April and May, and October to December, and once in August. This sunbird seems more insectivorous than other species, and may be seen hawking for flying insects. GOLDEN-WINGED SUNBIRD, Drepanorhynchus reichenowi Fischer. A breeding visitor, usually absent from June to August or September. Their arrival coincides with the flowering of a wild tree Crotalaria to which they are addicted, although they are less particular after a few weeks, when they invade the garden and remain to breed. Here they are especially fond of a heavy flowering shrub, Streptosolen jamesonii. They breed twice, in April and October. Their song is a loud reel, “chissississississ”, rather as for other sunbirds, and their usual call note, a deep “jer-wit jer-wit”. Like other montane Sunbirds, they are primarily dwellers of the forest edge where there are plenty of flowering shrubs and wild plants, but they will venture into the open when food is available. One evening in November, Mr. M. E. W. North and I witnessed a large number coming in from the grasslands to roost amid the big timber. EASTERN DOUBLE-COLLARED SUNBIRD, Cinnyris mediocris Shelley. A common resident of the forest, forest edge and garden, although only seen in pairs or with young. They frequent the green undergrowth that arises after forest fires, and seem to find their food from very small flowers; occasionally they ascend into tall trees, but I have not seen them in bird parties, as suggested in Praed and Grant. They seem to confine them- selves to a very limited area. I have only seen the female building the nest, usually in April and October, once in June, but the male is a more faithful mate than other male Sunbirds, singing from a nearby perch during the building operation or following close behind while she carries the material. The call note is a soft “jeb jeb”, and the female’s is the same but even softer and lower in pitch. The song is composed of two or three call notes followed by a silvery reel, uttered from a prominent perch; these birds are persistent singers, and may be heard most of the year in all sorts of weather, from dawn to dark. As with the Malachite Sunbird, the pectoral tufts are not always visible; on some birds they can be seen and on others they cannot: they do not seem able to display these tufts at will. I have seen birds in magnificent fresh plumage and full nuptial song, with not a sign of the tufts; other birds in worn-out plumage creeping about in some nettles, showed the whole tuft quite distinctly. GREY-HEADED SPARROW, Passer griseus (Vieillot). Rare straggler; two occurrences, the only recorded month being June. It is curious that it should be so uncommon, when I have seen both this bird and the Rufous Sparrow ten miles to the east, at 8,000 ft. BLACK-BILLED WEAVER, Heterhyphantes melanogaster (Shelley). A very shy and retiring forest bird of doubtful status. I have recorded it from June to December, singly or in pairs, once, (October), in a small party. Their presence is often only detected by the sound of dry leaves being scratched about under a bush; their song is most curious, and seldom heard, being a whistle followed by a buzzing reel and then a series of clicks, the whole being uttered extremely rapidly and possibly by two birds. The call is a sharp “preet preet”, seldom heard. Probably the bird is resident a little further inside the forest, where the undergrowth is thicker. The party I saw in October appeared to include young birds with dusky plumage, and from this and the dates recorded for song, I would conclude that they breed in August or September. REICHENOW’S WEAVER, Othyphantes reichenowi (Fischer). A common resident, especially around the garden and near cultivations. From the face markings of the male, it seems that the nominate race is the one that occurs here. I have seen three or four xanthochroic individuals on the farm, and with their bright golden backs ]. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 46 they look like miniature Orioles. This Weaver is found in parties of up to 30 birds, but the male and female of a pair usually keep close together. They breed during most of the year, especially in January, April and October. BROWN-CAPPED WEAVER, Phormoplectes insignis (Sharpe). An uncommon resident found along the forest edge, which appears to undergo some local movement. It is invariably found in pairs or with young. Their manner of feeding and the way they climb about the branches are reminiscent of the Nuthatch Sitta europaea Linnaeus. Their normal call is a soft conversational “tuk tuk tuk”, but with young they make a loud “chwee chwee”. The song is simple and mainly consists of a soft buzzing reel like that of the Black-Billed Weaver. Young birds have been seen in March, September and December, and as they do not appear to stay long in company with their parents, the breeding dates would be about a month earlier in each case. GROSBEAK WEAVER, Amblyospiza albifrons (Vigors). A rare straggler, one recorded only. Mr. Arthur Loveridge and I found this bird nesting on the Narok dam in May 1960, which is much lower, but only about 30 miles away. The single bird I saw on the farm was flying into a strong wind on 28.1.65, du^ east, and appeared to be migrating. It paused for a moment at the top of a pine tree and then flew on again. RED-BILLED QUELEA, Quelea quelea (Linnaeus). Fortunately the Queleas are rare stragglers to this wheat-growing district, where they might do immense damage. They are usually seen two or three at a time, and are in non- breeding plumage, which makes them hard to identify. This is the commoner of the two queleas so far confirmed. CARDINAL QUELEA, Quelea cardinalis (Hartlaub). Rare straggler. RED-NAPED WIDOW-BIRD, Coliuspasser laticauda (Lichtenstein). Visitor from October to December, generally only in small numbers. Although the males are in full breeding dress during this period, I have seen no signs of nesting. The song is like the noise of a miniature sewing-machine, and is uttered in flight. JACKSON’S WIDOW-BIRD, Drepanoplectes jacksoni Sharpe. A common resident in the grassy valleys, especially near dams. They undergo some local migrations in the dry season from January to March, when most of the males are in non-breeding dress. Nesting seems to take place from April to October, and birds become dispersed over a wide area, flocking into hundreds when breeding has finished. GREY-HEADED NEGRO-FINCH, Nigrita canicapilla (Strickland). I have only seen this bird three times so it must be rare. The birds were seen on the forest edge near the river. Another time I saw a small party a mile away, deep in the Masai forest, in August. RED-FACED CRIMSON-WING, Cryptospiza reichenovii (Hartlaub). Only three definite records, but there is some confusion between this and the next species, especially with females and juveniles. However, on 18.X.60, I saw with binoculars, a crimson- wing at close quarters, perched on a low tree, with a piece of dry grass in its bill. Jackson’s description of C. salvadorii, states that the eyelids are red, but how clearly these would show in the field I am not sure; this bird and the other two recorded had the red lores as well. Other birds seen with an impression of red around the eye I have discounted, as they may well have been C. salvadorii; some birds even seem to have the iris red. ABYSSINIAN CRIMSON-WING, Cryptospiza salvadorii Reichenow. Fairly common resident. Very active little birds generally seen darting about in the forest making their soft “chip chip” call-note. Usually in small numbers, or pairs, but after August may be found in flocks of up to 50 when they feed on tall grasses on the edge of cultivations. QUAIL-FINCH, Ortygospiza atricollis (Vieillot). Uncommon visitor. Occurred from December 1962 to March 1963, and again November/ December 1965, when it was seen on grassland and stubbles in pairs or parties up to six. They Notes on the Birds of Man Narok were very shy, seldom allowing a close approach, and it took a long time to identify them; the cail-note, tiny size and pale margins to the tail feathers were distinctive features. YELLOW-BELLIED WAXBILL, Coccopygia melanotis (Temminck). Resident and local migrant in small numbers. Most common in August, when small parties of up to 20 birds are seen, and when I have seen them carrying nesting material. Their call is a high, thin “seep seep seeip see-ip”. WAXBILL, Estrilda astrild (Linnaeus). Common resident on the grasslands near water. On one occasion I heard a single bird uttering a definite song, but I have no written description of it. In flocks of up to 50. YELLOW-CROWNED CANARY Serinus fiavivertex (Blanford). A very common resident, found all over the farm. It occurs in small and large flocks, sometimes in many hundreds. It is a very persistent singer, singing from the tops of the tallest cedar trees even in high winds. The song consists of a few short notes followed by a con- tinuous reel for many minutes, when the bird pauses briefly, and then starts again. Breeding takes place chiefly in October. STREAKY SEED-EATER, Serinus striolatus (Riippell). Very common and ubiquitous resident. They are a nuisance in the garden, where they destroy flowers and seedlings, and in the stores where they make holes in the grain sacks, spilling large amounts of grain. They breed most of the year round; I have not found more than two eggs or young in a nest. TTiey have a variety of call notes, the most usual being a quick “sirrup”, another being a harsher “shwee-ip”. The song is hardly “reminiscent of a Bulbul” as suggested by Praed and Grant, but it is more typically finchlike. THICK-BILLED SEED-EATER, Serinus burtoni (Gray). A fairly common resident, found in the forest and in small trees along the forest edge; occasionally it is met with out in the open. In silhouette its stocky build and stout bill remind one of a Hawfinch Coccothraustes coccothraustes (Linnaeus). Lack of white on forehead or throat identify the race as gurneti (Gyldenstolpe). TTie call-note is a high, sibilant “syip syip”, and as they usually feed with hardly a movement, their call is the first indication of their presence. The song is only heard in what is presumably the breeding season, i.e., July, and is a pleasant “sippi swee swee sippi swee”. The birds are found in small parties of 6 to 7, but once I found 30 of them moving across more open country, indicating some local migration. AFRICAN CITRIL, Carduelis citrinelloides (Ruppell). Fairly common resident, but never in flocks of more than a dozen, and often in pairs. Mostly found along the forest edge, where it feeds especially on tall thistles. The call note is a single very high squeak, almost inaudible. The song is very characteristic: a three-note whistle, “pee per pee”; sometimes an extra note is tacked on the end. The only time I observed a variation was when I saw a male displaying to a female; perched high up in a tree, he sang a long trill like a Canary, throwing his head right back and fluttering his wings. The female in this instance paid no attention, but on another occasion, a female crouched on a twig, and with much wing-fluttering, begged for food, which the male pretended to give her. Breeding records. May and December. CINNAMON-BREASTED ROCK-BUNTING, Fringillaria tahapisi A. Smith. The only buntings I have seen here were two of these birds on a murram bank on the 6.xi.65. Neither showed much black on head or chin, and they may have been immature. Summary of species identified on the farm (i) Breeding Regular breeding residents, some moving off the farm in non-breeding season : 84 Migrants which breed on the farm, then leave; . . . . 15 Total breeding species ...... 99 J. E. Air. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 48 (il) Non-breeding visitors or migrants From within Africa, regular . From within Africa, irregular . Palaeartic winter migrants 36 62 53 Total non-breeding species 151 Acknowledgments I am most grateful to Mr. M. E. W. North for encouraging me to write this paper in the first place, for reading over the manuscript and commenting on it, and in general for assisting me with his invaluable experience; he has also generously allowed me to use his habitat photographs of the farm. Mr. J. G. Williams has very kindly identified the specimens I have sent him, and Mr. Leslie Brown gave me many helpful suggestions concerning the birds of prey. Many neighbours have helped by reporting various species seen on their farms, especially Mr. and Mrs. Grainger, Mr. and Mrs. Daniell, Mr. and Mrs. Hamilton-Fletcher, Mr. and Mrs. Lutyens, and Mr. J. R. Shephenson, who found the nests of the Mackinder’s Owl, the Angola Swallow and the Banded Martin. Mrs. Elizabeth Nicoll kindly made a competent job of typing out a very untidy manuscript. Brown, L. H. 1955. Eagles. London: Michael Joseph. Chapin, J. P. 1932-1954. The birds of the Belgian Congo. New York: The American Museum of Natural History. Hall, B. P. 1963: The Francolins, a study in speciation. Bull. Brit. Mus. (Nat. Hist.) Vol. 10, No. 2. Jackson, Sir F. J. 1938. Birds of Kenya Colony and Uganda Protectorate. London: Gurney and Jackson. Lynes, H. 1930. Review of the genus Cisticola. London : B.O.U. Praed, C. W. M. and Grant, C. H. B. 1952-1955. Birds of eastern and north-eastern Africa. London: Longman, Green & Co. Priest, C. D. 1933. The birds of Southern Rhodesia. London: Clowes. WiTHERBY, H. F. et al. 1943. The handbook of British Birds. London: Witherby. References (Received for publication 25th April, 1966) J. E. Afr. nat. Hist. Soc. Vul. XXVI No. I (113) PLATE 1. Forest canopy, showing Crowned Hawk-Eagle’s nest in dead cedar tree. (Photo by M. E. W. North). The Birds of Mau Narok PLATE 2. View of cultivated land with two Masai hills in the background. (Photo by M. E. W. North). (Photo by M. E. W. The Birds of Man Narok PLATE 4. Young Mackinder’s Owl just able to fly. The white part round the gape is particularly evident in poor light. J. E. Afr. nat. Hist. Soc. Vol. XXVI. No. 1 (1 13) Page 49 RHINOPTERA- JAVANICA MULLER & HENLE FROM KENYA WATERS (PISCES : RHINOPTERIDAE) By G. F. Losse East African Marine Fisheries Research Organization, Zanzibar Rays of the family Rhinopteridae, commonly known as cow or bull rays, are rare in East African waters. Few are caught and none have been recorded previously by the East African Marine Fisheries Research Organization in Zanzibar. It was therefore of interest when two specimens of Rhinoptera javanica Muller & Henle were caught in a shark tangle-net set at the edge of a coral reef off Kikambala, Mombasa district, in March 1965. One of these speciments was purchased and is described below. I would like to thank Mr. Peter Nicholas of Mombasa for supplying the specimen and catch data and Dr. J. F. C. Schwartz of Maryland University (U.S.A.) for confirming the identification. RHINOPTERA JAVANICA Muller & Henle, 1841. Rhinoptera javanica Muller & Henle, 1841, Syst. Beschr. Plagiostomen: 182, pi. 58 (type locality: Java); Day, 1878, Fishes of India: 744, pi. 195, fig 4 (teeth) (India); Fowler, 1941, Bull. U.S. nat. Mus., No. 100: 476 (references; Pacific specimens); Smith, 1953-1965, Sea Fishes Southern Africa: 504, fig. 77a (South Africa); Idem, 1958, in Natural History of Inhaca Island, Mocambique: 131 (Inhaca, Mocambique); Fourmanoir, 1964, Cahiers — ORSTOM, ocednogr.. No. 6: 40, pi. 7 (Madagascar). Rhinoptera jayakari Boulenger, 1895, Ann. Mag. nat. Hist., (6) 15: 141 (Muskat, Arabia). DESCRIPTION: Based on the single specimen, 1194 mm. total length (head notch to tip of tail), a male approximately 26 lb. in weight from Kikambala, Mombasa district. Body, head and pectoral fins form a broad lozenge-shaped disc. Head distinct, divided into two lobes by a deep median anterior notch. A pair of rostral fins not joined with front of skull and not continuous at sides of head with pectoral fins. Upper lip of mouth with a fringed edge, lower lip with numerous small papillae. Eyes prominent, placed laterally. Spiracles large, close behind eyes and open laterally. Gill openings of moderate size, the last smallest. Pectorals falcate, the leading edges convex, the hind edges concave. Dorsal fin set above basal part of tail. Page 50 Rhenoptera javanaca Pelvic fins long and narrow. Anal fin absent. Tail long, slender and whiplike with a single basal serrated spine. Teeth wide, angular and fiat, set in pavement; nine rows in each jaw, the median row the widest, the outer narrowest. Skin smooth, without dermal denticles or tubercles; minute denticles on rostral fins. Disc width 872 mm. (across pectoral fins), length 560 mm. (head notch to hind border of pectoral); maximum body depth 105 mm.; pre-oral distance 91 mm. (from head notch); pre-nasal distance 63.5 mm. (from head notch); eye diameter 30 mm.; inter-orbital distance 106 mm.; inter- spiracular distance 125 mm.; pre-pelvic distance 91 mm. (from head notch); dorsal fin base 58 mm., height 69 mm.; pelvic fin length 127 mm.; clasper length 114 mm.; tail length 322 mm. (? broken); serrated spine, length 54 mm., maximum width 7 mm. COLOUR: Fresh, upper surfaces uniformly bluish-grey, ventral surface of disc greyish-white, darker towards tip of pectoral fins. Rostral fin denticles black. DISTRIBUTION: Kenya (Mombasa district). Elsewhere, Indo-Pacific region: recorded from South Africa, Madagascar, Mozambique, Arabia, India, Ceylon, East Indies and China. REMARKS: Two Indo-Pacific species are generally recognised, R. javanica and R. adspersa M. & H. These have been distinguished primarily on the presence or absence of dermal denticles or tubercles on the dorsal surface (i.e. back rough, tuberculate or smooth), and on the number of tooth rows in the jaws. Thus, Day (1878) and Fowler (1941) follow Muller and Henle (1841) and describe the dental formulae for R. javanica and R. adspersa as 7/7 and 9/7 respectively. That there must be some variation in the dental pattern is shown by the Kenya specimen with 9/9 tooth rows. I have now been informed by Dr. Schwartz (in litt.) that the dental formula of R. javanica may indeed vary from 7/7 to 9/9, or a com- bination of these. There is apparently but a single specimen of R. adspersa in existence (a stuffed mount in the Paris Museum); in this specimen the dorsal surface is warty or tuberculate. Boulenger’s (1895) description of R. jayakari from Muscat (nine tooth rows in each jaw, skin smooth) is very close to the specimen described here and may be regarded as a synonym of R. javanica. Additional specimens of Rhinoptera are required from East African waters for further studies. (Received for publication 3rd June, 1966) J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 51 PROPHYLLS AND BRANCHING IN CYPERACEAE R. Wheeler Haines Department of Anatomy, Medical School, Makerere University College Kampala (Uganda) Introduction The prophylls of Cyperaceae have been discussed in a full paper by Blaser (1944), and others, particularly Holttum (1948), Koyama (1961) and Kern (1962), have considered them in relation to spikelet structure and phylogeny. But they do not usually figure in taxonomic descriptions or drawings and are still relatively little known, particularly as regards their distribution in the whole plant. Here a few Nigerian species growing in the forest zone near Lagos, supplemented by a few plants from elsewhere, are used to illustrate their range of development and their bearing on the evolution of the family. Most of these species are also found in East Africa. The proper identifi- cation of the prophylls and the branch systems associated with them underlies many problems of morphology and classification. In dicotyledons the prophyll or prophylls are taken to be the first leaf or leaves of the shoot, which may be modified in various ways but usually grade into the foliage leaves. In monocotyledons a particular form of prophyll is often found at the base of the shoot, never more than one, always placed dorsally, that is between the shoot and its parent axis, and usually having two more or less equally developed main vascular bundles each with its own keel rather than the single main bundle and single keel of the foliage leaves. It is this unique organ, well developed in Cyperaceae, that is considered in this paper. I have to thank Miss S. Hooper, of the Kew herbarium, for identifying most of the material, and for her kindly guidance through the literature of the subject. Tubular Prophylls Fimhristylis ohtusifolia (Lam.) Kunth (fig. 1, B-E) is a common tufted perennial of open trodden ground near ponds and lagoons. The foliage Page 52 Prophylls and Branching in Cyperaceae leaves are all basal, but some ensheathe the lower part of the rounded culm (B and C). New branches spring as buds from the axis of the leaves, the three branches a, h and c of the fragment shown from the leaves, 3, 4 and 7 of the parent stem. At its base each branch is surrounded by a tubular prophyll, aP, bP and cp. The prophyll carries a pair of ciliate keels ending in short points, and these keels embrace the parent axis (D). As the shoot expands the prophyll is often split to the base on the abaxial side (E). The prophyll is followed by a variable number of foliage leaves arranged spirally, and these, at the culm apex, by the bracts that subtend the branches of the inflorescence. Fimhristylis obtusijolia has a condensed inflorescence not suitable for analysis, but in F. dichotoma (L.) Vahl, a common weed, usually annual, of gassy places, banks and flelds, the parts are well spread (fig 1, A and F-L). The lowest bract (F-H, b) is longer and set more vertically than the others, and subtends the longest and most complex branch (I), while the succeed- ing bracts, set, as in most Cyperaceae, at near limit divergence (Hirmer, 1931), decrease regularly in size and bear less complex branches. Each branch is surrounded at its base by a tubular, two-keeled prophyll (I.J) similar to, but much smaller than those at the bases of the culms. The lower branches (1 and 2) are themselves further branched, the upper not so. Each branch ends in a terminal spikelet with the secondary branches, if any, springing immediately below (K,L). The glumes of the terminal spikelet follow the spiral of the bracts without interruption (H,L). One spikelet is terminal to the culm itself and is sessile amongst the major branches (A and G, f), and so lies near the base of the whole inflorescence. In Fimbristylis dichotoma the branching at the base resembles that in F. obtusijolia, but the lower leaves are reduced to short-bladed sheaths. The leaves have a sub-distichous arrangement, so that the branching comes to be largely in one plane (fig. 2, F and G). In the fragment analysed the leaves and a^ of the oldest shoot subtend shoots b and /, while a^ has the prophyll of an undeveloped shoot eP hidden in its axil. The shoot c is subtended by the leaf b^ and d by in succession so that all the prophylls of this side of the fragment face towards a. But on the other side the shoot g is subtended by and its prophyll faces the other way, spoiling the symmetry. Within the spikelet glumes there are no prophylls: nothing corres- ponding to the paleas of Gramineae. Since every shoot of the plant, from the largest culm axis to the smallest branch, carries a prophyll at its base and a spikelet at its apex, the numbers of prophylls and spikelets over the plant as a whole are, in principle, equal. But while the spikelet of a minor branch lies near its prophyll and its glumes follow the prophyll immediately (fig. 1, K, L), the terminal spikelet of the infloressence is J. E. Air. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 53 separated from its prophyll by the whole length of the culm, and there are many foliage leaves and inflorescence bracts between. In Eleocharis mutata Roem. & Schultes (fig. 2, H, I) the culm ends without branching in a single spikelet, so that there are normally no prophylls except at the culm bases, where they are constant. In the normal spikelet there are no prophylls, nor are they found at any stage of develop- ment (Mora, 1960, Eleocharis palustris R. Br.). But in proliferating spikelets of Eleocharis, sect. Multicaules where the florets are replaced by vegetative shoots, Koyama (1961) found them, the only clearly recorded examples of intraspicular prophylls in Cyperaceae. The Fimhristylis species discussed seem, so far as their branch-systems and prophylls are concerned, fair samples of unspecialized Scirpeae. It is generally agreed that the presence of perianth bristles and the relatively simple insertion of the floret in Scirpus are more primitive characters than the absence of bristles and winged rachilla of Fimhristylis (Koyama, 1961). The branch structure in S. sylvaticus L., picked by Monoyer (1934) as the nearest approach in living material to a primitive Scirpus, is similar to that in Fimhristylis. In S. pedicellatus Fernald Blaser (1944), found a gra- dual transition from a large two-keeled many-veined prophyll of the lowest ray of the umbel to the smallest type which was hyaline and usually without keels or veins. Blaser also noted that in the species of Scirpus he examined the prophyll never subtended an axillary bud. A fragment of S. hrachyceras Hochst. (fig. 2, J), however, shows one prophyll (a) at the base of the branch, and others (b, c) at the bases of the latest spikelets. In the axil of a is an undeveloped spikelet, with its own prophyll (d), and similarly placed spikelets may become fully developed. Prophyll branching at the plant base Fimhristylis hispidula (Vahl) Kunth is a tufted, hairy, fast growing annual of dry disturbed places (fig. 2, A-E). The leaves are relatively re- duced, the culms serving as the main assimilating organs. The inflorescence is usually less complex than in F. dichotoma and is often without secondary branching, as in the head figured by Nelmes and Baldwin (1952), but is of the same structure, with tubular prophylls (E). At the base of the plant the prophylls are closely packed and con- spicuous, and most of the new shoots spring from their axils. A prophyll subtends one shoot, and the prophyll of this shoot the next, without waiting for the parent shoot to mature, a process which leads to a mass of shoots and prophylls facing in different ways (B, C). The young prophylls are conical with only a narrow opening (D) but become split as the shoot expands. Such basal prophyll branching is uncommon but in Eleocharis Page 54 Prophylls and Branching in Cyperaceae acicularis (L.) Roem. & Schultes, and E. palustris (L.) R. Br. tuft formation in the season following establishment of a new plant depends on this mechanism (Tutin, 1954; Mora, 1960). Prophyll branching in the inflorescence Fuirena umbellata Rottb. (fig. 3, A-0) is a soft-stemmed upright plant of wet places and streamsides, often, at least at flood time, partly submerg- ed. The rhizome is creeping, made up of the horizontal bases of successive shoots. The stem is eventually ascending with the leaves spaced along it, the lower reduced to sheaths, the upper foliaceous. The spikelets are set in clusters of about six, arranged in an elongate inflorescence. New shoots break through the bases of the old scale leaves (B), protected at first by short conical prophylls which are split as the shoots expand (C, D), and soon wither. Usually only one such shoot develops from each old stem, but several buds are formed, and these may develop to give a branching rhizome. The leaves below the inflorescence subtend no branches, nor can any trace of buds be found in their axils. But in the flower beai’ing region two branches, a major and a minor, appear at the ligule margin of each of the larger bracts (E and F, mj, mn). Traced to their origins at the bract attachment the two branches are found enclosed together in a delicate tubular prophyll (G), and the minor branch has, besides, its own prophyll (H), arranged as shown in the plan (I). Thus the minor branch springs from the axil of the prophyll of the major branch. Tubular prophylls similar to those of Fimhristylis also surround the smaller branches of the inflorescence (J). But in Fuirena the individual spikelets are sessile in the cluster, and their prophylls are shortened so as to be wider than long (K). Further each prophyll bears on its adaxial surface a pulvinus which swells as the inflorescence reaches maturity so as to force each branch or spikelet away from the axis that carries it. At maturity the prophyll persists at the base of the spikelet rachilla while the glumes and nutlets fall from it (L). Here again the numbers of prophylls and spikelets correspond. The scheme (M) represents the spikelets carried at a single node. The major and minor branches each have a prophyll at the base and end in a terminal spikelet (f) which occupies the centre of a cluster and has no prophyll at its base. The other spikelets are arranged in spiral order, each subtended by a bract and each bearing a prophyll at its base. The distinction between bracts, prophylls and glumes is clear. Occasionally, as in the axil of the bract b'^, a spikelet (in) may spring from the prophyll of another spikelet. Another example, this time aborted, is seen at the base of the fruiting J. E. Afr. nat. Hist. Soc. Vol XXVI No. 1 (113) Page 55 Spikelet (L, in)^ with its own prophyll. But whereas prophyll branching is an important feature of the main branch system of the inflorescencej it is relatively rare in the spikelet clusters. Owing to the elongation of the internodes of the main stem the terminal spikelet of the inflorescence lies near its apex (E, t). It does not usually lie quite at the apex, for the uppermost part of the main stem is deflected and a secondary branch continues its direction. The floret in this species has three perianth scales, of which the largest lies between the ovary and the rachilla of the spikelet (N, O), so in the position of a prophyll. But it is not two-keeled, and its texture and veining agree with those of the two smaller scales. Further all three scales end in a weak, flexuous bristle, suggesting that the wide basal part of the scale is an expansion of a bristle such as is found in many Scirpeae (fig. 2, I) and, in species of Fuirena with six perianth members, in the outer three. Nees Van Essenbach (1835) believed that the three scales of Fuirena were attached within the three stamens and Kern (1962) was also ‘pretty sure’ that the scales were inside the whorl of stamens. If this were indeed the case the scales could not belong to a perianth and would presumably be bracts. So Mattfeld (1938) and Kern (1962) took each separate stamen in Fuirena as a unisexual flower subtended by a bristle, and the ovary as a terminal female flower, the whole bisexual structure being a composite ‘synanthium’. They accepted the implication that the florets of other Scirpeae must, since their structure was rather similar, also be synanthia. But though the filaments of the stamens lie outside the scales both Blaser’s (1941, a) beautiful sections and my dissections of fresh material (fig. 3, N) show their attachments at most on a level with the perianth segments and not outside them. There is then no need to consider these florets as synanthia. It seems probable that the many peculiarities of Fuirena umhellata, its various types of prophyll with and without a pulvinus, prophyll branching, hollow 5-angled stems and leaf with ligule and adjustor tissue, are, together with the scaly perianth, all specializa- tions. It is more difficult to decide whether the long internode of the culm and shorter internodes of the inflorescence in Fimhristylis, or the more uniformly expanded internodes giving the leafy stem and elongated inflorescence of Fuirena are the more primitive, or whether both are derived from yet another arrangement. Dulichium has also been interpreted by Mattfeld (1938), Schultze- Motel (1959) and Kem (1962), as having synanthial florets. Of the 8 bristles the 5 abaxial are inserted below the stamens but the 3 abaxial above. Each group of bristles was interpreted by Mattfeld as the surviving veins of a dissected bract, but Blaser (1944) found the anatomy Page 56 Prophylts and Branching in Cyperaceae of the floret to be of normal scirpoid type, except that the inner bristles were a little displaced, presumably by pressure in the bud. Kern pointed out that neither of the two groups of bristles could be a reduced prophyll as in Cyperaceae the prophylls have no midveins, but still believed that the bristles were reduced bracts, not perianth segments. In fact Blaser’s (1944) figures of prophylls at the base of a major branch and of a spikelet, taken together with Schultze-Motel’s (1959) figures of the lowest floret of a spikelet which includes the prophyll (unlabelled) at the spikelet base, show Dulichium to be, in general struc- ture, of ordinary scirpoid type. Koyama (1961) has removed it from Cypereae to Scirpeae in spite of the distichous glumes, which may be found in other Scirpeae such as Scirpus roylei (Nees) A. A. Beetle and several species of Fimbristylis and Bulhostylis. Indeed this and other characters such as the cylindrical hollow stems, the evenly spaced equal inflorescence branches, and the shift of the inner bristle suggest a peculiarly specialised type, rather than a relatively primitive form, ‘com- bining characters of Scirpeae, Cypereae and Rhynchosporeae’, as suggested by Mattfeld. A new study of the plant as a whole is needed. Serial branching Though Fuirena umhellata usually grows in swamps, it is sometimes found in open places which dry out each year for several months. In this case the rhizomes are straight, with only occasional branches, and are studded with closely packed, hard, black ‘bulbs’ (fig. 3 P,Q), which sprout at a favourable opportunity. The ‘bulbs’ lie alternately to the left and right of the sympodial rhizome, and the leaves are set in alternate right and left-handed spirals (arrows in R). Successive branching from the second leaves, P, g^, h^, etc. gives the straight rhizome with two scales, the first and second leaves of the succeeding culm, between each pair of ‘bulbs’. These two scales are intact only in the last shoot j. In the others they are split as the succeeding shoot develops. The third leaf of each shoot makes the outer covering of its ‘bulb’. Such ‘serial’ growth is rare in Scirpeae, but is found in many Cypereae, always dependant on budding from the axil of the second leaf. Buds may be found in the axils of other leaves, as giving branching rhizomes. Prophylls in Cypereae Cyperus tenuis Swartz (fig. 4, A-H) is common tufted weed of grassy clearings and pathsides in dry, disturbed bush, usually dying off in the dry season. The base of the culm is swollen (B) and carries a variable number of scale and foliage leaves. The lower leaves are split (C) so that it is J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) difficult to place them, but when they are torn off their positions become clear from the arrangement of the buds developed in their axils. The strictly i phyllotaxy, found in all Cyperaceae with sharply triquetrous culms, is attested by three vertical series of buds, each protected by a conical prophyll (B). The tristichy is derived from a distichy in the seed- ling (Hirmer, 1931). The primary and secondary umbel branches have tubular, two-keeled bifid prophylls (D,E), resembling those of a Fimhristylis. But at the attachment of the prophyll, and probably a part of the prophyll, is a pulvinus which adjusts the position of both the branch and the umbel bract that subtends it. This action is fully discussed by Mora (1960) for Cyperus papyrus L., where each of the larger bracts subtends many branches set side by side like a hand of bananas, each branch with its own prophyll. In Fimbristylis some of the spikelets are solitary, others grouped into small clusters. This is a common arrangement in other general of Scirpeae, but in some, such as Holoschoenus, all the spikelets are grouped. In Cyperus the unit is a group, most of the groups carried in the secondary umbels, but some solitary on the shorter branches and one terminal to the culm. In the group one spikelet is terminal, the rest lateral (F,G). Each lateral spikelet has two empty scales at its base, one the subtending bract (b) , the other the prophyll (p), followed immediately by the fertile glumes. The prophyll is scale-like, the tip entire, the keels hardly dis- tinguishable and the pulvinus confined to the adaxial surface (H), so that it has little resemblance to the prophylls of the umbel rays. The terminal spikelet (f) is, like the lateral, sessile, and is often pushed to one side by the uppermost prophyll. It has one empty scale (R) at the base, presumably an empty glume, an exception to the general statement that there are no empty glumes in Cyperus. Similar arrange- ments, with sharp differentiation of the different kinds of prophyll, have been found in other species of Cyperus examined. Cyperus subumbellatus Kukenth., another common weed of open, well drained places, has rounded stems, and the scales and leaves follow an in- determinate phylotaxy (fig. 4, 1, J). The fragment illustrated, part of a large clump, illustrates the arrangement well. The parent shoot bears leaves labelled arbitrarily a\ a^, etc., and of these a^, a^, and subtend buds. That in the axil of is expanded and has burst through its subtending leaf. It bears a two-keeled prophyll p and the leaves 2, 3, 4, etc. Now it has been claimed by several authors whose work is sum- marized by Arber (1934), with important additions of her own, that where the prophyll has two keels one is the original midrib of the prophyll and Page 58 Prophyils and Branching in Cypemceae the other an enlarged lateral. Supposing that the angles between succes- sive leaves remain relatively constant, then following the series 4,3,2,p, down the shoot, the keel x not y appears as the midrib. Blaser (1944) has objected to Arber’s suggestion on the grounds that her main criterion for distinguishing the keels was that of size, and ‘many of her figures show the largest vein in the wrong keel’. But her claim that not more than one shoot is subtended by a prophyll, and that this shoot springs opposite the keel representing the midrib still stands, and is true of most Cyperaceae. The prophyll a part of the disseminule In all Scirpeae and in most Cypereae the glumes fall away from the persistant rachilla of the spikelet to set free the fruits. But in some species of Cyperus of the section Mariscus, including C. subumhellatus. the spikelet falls as a whole. The spikelets are arranged in dense clusters (fig. 4, K), the clusters mostly pedunculate but with a single sessile cluster terminal to the culm. Each individual lateral spikelet has a subtending bract and prophyll at its base (L). On ripening the spikelet falls away taking the prophyll with it, but usually leaving the subtending bract behind. The prophyll is not actually attached to the spikelet, and can be teased away, but clasps the base of the spikelet tightly. In Kyllinga, possibly derived from a Mariscus by further concentration of the inflorescence and reduction of the stigmas, the disseminule again falls with the clasping prophyll (Fig. 4, M,N.). In Lipocarpha the prophyll may be five-ribbed, but only two of the ribs are of vascular structure (Blaser, 1944) or it may be smooth surfaced without ribs or keel. Hemicarpha and Remirea are related (Kern, 1962). Ascolepis (fig. 4, 0,P, A. capensis Ridl.) is usually placed in Mapanieae, separated from other genera by the chypogynous scales’ or ‘bracteoles’ being ‘united’ to enclose the otherwise naked flower. But there is no reason to believe that the enclosing structure has resulted from a union, for it occupies the position and has the two keels of the prophyll normally found at the spikelet base in sedges. All these genera may belong to Cypereae, to the group defined by Nees von Essen- bach (1835, group I b) as having ‘spiculae uniflorae’. Possibly in Isolepis also the single median hypogynous scale figured by Clarke (1909) is a prophyll and the genus belongs in Cypereae rather than Scirpeae, a sug- gestion which follows Pax (1886), who associated it with Hemicarpha. This is a rare instance of doubt as to whether or not a particular scale is a prophyll, and so as to how much of the inflorescence constitutes a spikelet. 1. E. Air. not. Hist. Soc. Vot. XXVI Ho. 1 (113) Page 59 Geometrical branching in Kyllinga The excellent habit drawing of Kyllinga erecta Schum. Thonn. in the Flora of West Tropical Africa shows the straight knotted rhizome with two scales between the origins of successive culms, and my fig. 4, Q, R presents an analysis of a fragment. As in the bulbous variety of Fuirena umbellata the scales are a prophyll and a second leaf, and the direction of the phyllotaxy is reversed at each new shoot. Besides the main buds con- tinuing the rhizome from the axils of each second leaf, other buds are found in the axils of the third and fourth leaves of each shoot. In this Kyllinga the internodes of the horizontal parts of the shoots that build the rhizome are well developed, though the internodes at the bases of the culms are short so that the leaf attachments are crowded. In other Cyperaceae with horizontal rhizomes there may be only one inter- node between successive culms, so that the rhizome bears only prophylls, as in Eleocharis palustris (Walters, 1950), or four with a prophyll and three scales as in Scirpus lacustris L. (Mora, 1960). In Cyperaceae which spread by long horizontal rhizomes bearing tubers at their ends, each rhizome has a prophyll at its base and an indefinite number of elongated internodes in its horizontal part (Pax, 1886). Leafy stems in Cyperaceae Most Cypereae have all the stem leaves attached at the base of the culm, but in Cyperus mundtii Kunth. there are long horizontal leafy stems, usually half-floating in slowly moving water, clothed throughout with even- ly spaced leaves. Still leafy each stem eventually becomes vertical and ends in a short bare culm and inflorescence. Occasional new branches, each with its prophyll (fig. 4, S) spring from the horizontal parts. Remirea maritima Aubl. has a similar spreading habit but grows in loose coastal sand. The leaves are reduced to scales on the horizontal parts; on the short vertical parts the leaves are tightly packed, and there is no bare culm between the leaves and inflorescence. These peculiar species are un- doubtedly specialised and indicate at least the possibility of a culm becoming leafy. But in other tribes the direction of evolution is less certain, and the habit of the early Cyperaceae cannot be determined. Prophyll branching in the inflorescence of Rhynchospora Rhynchospora corymhosa Britten (Fig. 5, A-K) grows in the shallow waters of open swamps to about 2 m, usually in pure stand. The long coarse leaves are mostly basal, but others spring from the culm, and there is no gap between these and the inflorescence bract. Each major inflorescence Prophylls and Branching in Cyperaceae branch has a delicate tubular prophyll, minutely denticulate on the keels, at its base, hidden in the sheath of the subtending leaf (B). Whereas in most sedges all the spikelets of a given inforescence are at about the same stage of development, in this Rhynchospora young and fruiting spikelets are mixed (C and D, young spikelets 3, 5 and 7). The culm, once it is developed, continues to bear new spikelets, and may survive the dry season to fruit again later, all with little change in its out- ward appearance. This is done by prophyll branching. The smaller fragment (E, F) has three successive branches with prophylls aP, bP, and CP, and the larger fragment (C, D) includes the spikelet t terminal to the branch bearing the prophyll tP, at its base, and four bracts, each of which subtends a shoot, 1, 4, 6 and 8. From the prophylls of these spring 2, 5 and 6 and from the prophyll of 2 shoot 3. Blaser (1944) gives a section of Rhynchospora inexpansa Vahl with a similar arrangement. In the dwarf Rhynchospora alba (L.) Vahl (fig. 5, L-T, Irish material) the leaves are distichous and some subtend young shoots. In each shoot the prophyll faces the parent stem, but the new distichy is at right angles to the old. The culm leaves subtend no buds below the inflorescence, whose major branches bear long tubular prophylls (P), reduced to scales in the minor branches (Q, R). Blaser (1944) studied this species and found the smaller prophylls veinless and only slightly keeled, but had no doubt as to their identity. Tandem branching in Cladium Cladium mariscus R. Br. (Fig. 6, A-H, Irish material) is tall and rough approaching in habit our larger tropical sedges, though of temperate regions. In the inflorescence two branches, major and minor, noted by Mora (1960), spring from each node of the culm (mj, mn), as in Fuirena. The prophylls of these branches are not tubular as in most sedges, but are split to the base ventrally, with the edges overlapping. The minor branch is set in tandem ventral to the major, not at its side and not enclosed in its prophyll.. The spikelets are crowded (D, E) but their arrangement is normal, each of the lateral spikelets having a bract and prophyll at its base, the terminal spikelet neither. Loss of prophylls in Schoenus Schoenus nigricans L. (Fig. 6, I-M, Irish material) has long, open, membranous pointed prophylls at the base (I). In the fragment figured the main stem has formed its culm, and this carries leaves a^-a^. Of these a^ and subtend shoots. The prophyll of the younger shoot, bP, is still entire but that of the older shoot, cp, is split into two single keeled halves. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) The inforescence is carried on a long bare culm which becomes bracteate near its apex. Groups of spikelets spring from the axils of the larger bracts each spikelet on a distinct pedicel, but without a prophyll (Mora, 1960 and Fig. 6 J). Slight differences of size, position and state of development suggest successive prophyll branching (K), as at the base of Fimhristylis hispidula Kunth, b developing from a and giving rise to c and d, but it is difficult to be sure. Near the apex of the inflorescence the bracts are smaller and subtend only one spikelet. As there are no prophylls the terminal and lateral spikelets are similar (L, M). The spikelet in Rhynchosporeae A terminal spikelet of Rhynchospora corymbosa (fig. 5 G) has three empty glumes 1, 2 and 3 at the base, and removal of glume 4 (H) does not expose a flower. But the next glume 5 which hides the lowest flower from view does not subtend it, for the glume and flower lie on opposite sides of the rachilla. In fact this flower is subtended by 4, but is closely enwrap- ped by the overlapping membranous margins of 5. Similarly removal of 5 does not expose the next flower for it is enwrapped by 6 (I) and there may be yet another flower with no ovary, subtended by 6 and enwrapped by 7, 7 being sterile (J). There is usually only one bisexual flower, the lowest (K), but there may be two. R. alba has a similar spikelet structure, but with fewer parts (fig. 5, S, T). Clarke’s (1909) diagram of R. wallichiana and of Eriospora pilosa Benth., now transferred from Sclerieae to Rhynchospcreae on account of its perianth (Reynal, 1963), agree well. Cladium (fig. 6, F-H) has similar enwrapping glumes. Thus in both the terminal spikelet (F, G) and the lateral spikelet (H) glume 3 is wrapped round the lowest flower, which is subtended by 2. In Schoenus (L-N) the spikelet has up to five flowers set on a flexuous rachilla, each flower springing from the rachilla above the subtending bract rather than from the actual axil of the bract. Supraaxillary buds are common in Cyperaceae as in the rhizome of Eleocharis (Walters, 1950) and the culm base of Cyperus tenuis (fig. 4, B). Pax (1886) described the spikelets of Asterochaete and Elynanthus as having a structure similar to that of a Rhynchospora or Cladium. But he showed the bract between the two flowers, corresponding to 6 in fig. 5 J, as a two keeled prophyll. He explained its presence by supposing the lower flower to terminate the main axis of the spikelet, and the second flower to terminate a secondary axis springing from the uppermost bract of the main axis and carrying the supposed prophyll. In Schoenus again he and Celakovsky (1887) supposed the spikelet to be a sympodium, each flower above the first terminating a short axis and bearing a prophyll from Page 62 Prophylls and Branching in Cyperaceae which the flower above sprang. The sympodial structure separated Rhynchosporeae from other tribes. Blaser (1941,b) showed the lower flower of Rhynchospora macrostachya Torr. ex A. Gray as terminal, but the disputed bract as having only one vein and no keels. Hamlin’s (1955) description of the rachilla of Schoenus “carrying the glume above the subtended flower” was clearly based on Pax’s theory. Kern (1962) again figured the spikelet of Schoenus as a ‘rhipidium’, but, since he found ‘exactly the same’ structure in Cyperus, believed that here also the spikelet was a rhipidium. Holttum (1948) pointed out the weakness of Pax’s theory. The supposed prophylls of Asterochaete and Elynanthus had no keels and were in fact glumes, though they might be compressed by the flowers between which they lay and assume angular forms (fig. 5 J, glumes 6 and 7). Again Mora (1960) has described the inflorescence unit of Cladium and Schoenus as a cymose “Scheinahrchen”, identifying the more distal bracts as prophylls subtending the flowers. But his carefully drawn sections show the supposed prophylls to have the same structure as the undisputed bracts, with one keel and one vascular bundle, both median, not two of each. His drawings of a developing unit are also indecisive, for his “sterilen Spelze” appears to subtend the lowest flower and his “Vorblatt 1 and 2 the succeeding flowers of a racemose spikelet. Koyama (1961) appears justified in giving a reduced scirpoid spikelet structure for Rhynchosporeae, and the arrangement of the prophylls, not considered by him, supports this. The peculiarity of the tribe lies in the enwrapping glumes, not in a cymose structure. Branching in Sclerieae Scleria naumanniana Boeck. is a loosely tufted perennial, about l.m high, of dry open bush on laterite or sandy soil (Fig. 7, A-I). The buds at the base have short conical prophylls, soon splitting as the bud bursts through the sheath of the subtending leaf (B). The florets are unisexual, set in spikelets of three kinds, male, female and bisexual (C). Young buds are mixed with the old. Removal of the subtending bract (D) exposes a series of prophylls, inbricated one within the other (E). Branch 1 carries the prophyll Ip and several spikelets, removed in D, and Ip subtends branch 2, and so succes- sively to the youngest shoot 6. As in Rhynchospora corymhosa, which has similar prophyll branching, a fruiting culm can survive the dry season and fruit again. In a bisexual spikelet (F, G) glumes 1 and 2 are sterile, 3 subtends a female flower supported by a three lobed gynophore, but without perianth, J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) and the remaining glumes take up a subdistichous arrangement, sub- tending male flowers, of which most have only two stamens. The upper- most glume is small and sterile. The glumes of the male part of the spikelet are pushed aside by the female flower, the more so as the fruit expands. The female spikelets have only three well developed glumes and the flower appears terminal. But in some spikelets there is a minute, sterile, fourth glume representing the male part of the spikelet (H, I, m). Scleria verrucosa Willd., a coarse plant of permanent swamps, m, has a shortly creeping, branched rootstock, with new branches arising where the shoots turn vertically to form the culms (J). The inflorescence branches arise two at each node (K), one from the prophyll of the other, the prophylls, especially that of the minor shoot, being very delicate (L). The spikelets are unisexual, a sessile female usually springing from the prophyll of a pedicelled male spikelet (M, N). The male spikelets have numerous glumes, distichous below but spiralled above, some of the lower and the uppermost sterile, often with one or more glumes (e.g. 5) enwrap- ping the flowers below (O, P). The female spikelets have three to five glumes, with the flower apparently terminal (Q, R). The lobes of the gynophore enlarge and evert as the fruit ripens. The male part of the bisexual spikelets of Scleria has been interpreted, by Goebel (1888), Mora (1960) and Kern (1962) among others, as a secondary branch arising from a bract of the main axis which carries a terminal female flower. Kern showed an intraspicular prophyll in the position of glume 4 (fig. 7, F, G), marking off the male part. But the dissections of Clarke (1909), Koyama (1961) and Raynal (1963) show no prophyll in this position in the many species they have studied, and Blaser’s (1944, b), section of Scleria reticularis Michx. shows the scale in question to have a midvein but no keels. In Hoppia and in Diplacrum, a genus believed closely related to Scleria, the sessile male spikelets are subtended by glume-like scales below the female spikelet, but they are set off by keeled prophylls leaving no doubt as to their spicular nature. Indeed in all Sclerieae the prophyll arrangemnet is a safe guide to spicular identity. It might be difficult, in the absence of any remnant of the male part of the spikelet, to say whether the female flower was lateral or terminal. But since in bisexual spikelets it is always lateral, it is presumably so in all species of Scleria, and probably in all species of Cyperaceae, even when it appears terminal. The lobes of the gynophore in this and other Sclerieae have been interpreted as perianth segments. But Blaser’s (1941, b) sections show vascular bundles on their way to the ovary deviated into the lobes, not ending in them as they would if they were perianth segments. Page 64 Prophylls and Branching in Cyperaceae In most Cyperaceae the region of the culm between the bud-subtend- ing basal leaves and the bracts of the inflorescence bears no buds and is unbranched (Mora, 1960), and even if the culm is leafy as in Fuirena or Rhynchospora the culm leaves subtend no buds. But in Scleria naumanniana buds are found in the axils of the culm leaves, though they seldom develop unless the culm is cut. In Scleria harteri Boeck. the culms are profusely branched and scramble over bushes to 5m., clinging by the retrorse spicules of the stems and leaves. As in Cyperus mundtii this branching is clearly a secondary, not a primitive, feature of the group, as it appears to be in Cyperaceae as a whole. Mapanieae Mapanieae are mostly large tropical plants with wide leaves. Their basal parts resemble those of other tribes. Hypolytrum sp., nr. heterophyl- lum Boeck, (flg. 8, A), growing in rich damp places in semi-shade, has wide basal leaves with spreading scaly stolons (B) breaking through their attach- ments. The prophylls of these stolons are short and conical, resembling those of a Cyperus or Scleria. But in the inflorescence the branching is peculiar. Riiter (1918), quoted by Blaser (1944), showed the prophyll in Scirpodendron and Mapania more or less divided into two separate bracts, each having a single keel and each subtending a shoot (C). In Hypolytrum the main branches again come off three together, the group as a whole subtended by a large bract, but each branch has its own prophll (D). The minor lateral branches carrying heads of spikelets have small tubular prophylls at the base (E). Kern (1962) has noted that, as in other Cyperaceae, the terminal head in Mapanieae has no prophyll near its base. Each bract of the head subtends a reduced spikelet enwrapped by a pair of ciliate keeled scales, united below and subtending a pair of stamens (F) with a naked female flower between (G). This is the most reduced form of spikelet found in Mapanieae, for other genera have unkeeled scales, often joined to a tube, with or without single stamens in their axils, set between the keeled pair and the female flower (H), richly developed in Chorisandra (I) and Scirpodenron. The keeled scales appear to be derived from a single prophyll sub- tending two flowers, each reduced to a single stamen. , They are found throughout the tribe except in Chrysithrix, where, as in Eleocharis the spikelet is solitary and terminal to the culm (Clarke, 1909), so that there could be no prophyll at its base. In general the structure is imiform and, contrary to Kern’s (1962) proposal, it seems reasonable to retain the Scirpodenron group and the Scirpus group at subfamily level. Bentham (1877) believed that the peculiar spikelet was in fact a single floret with numerous perianth segments and stamens, but Goebel (1888), }. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 65 after detailed consideration, set this suggestion aside. Indeed Bentham’s proposal did not explain the frequent presence of scales within the outer stamens. But in the reduced form seen in some species of Mapania the spikelet may resemble a single floret of a Scirpus, particularly S. memhranaceus Thunb. which has a pair of keeled scales, and on such a comparison the modern synanthial theory of Holttum (1948), Koyama (1961) and Kern (1962) is based. The complex spikelet of unisexual florets of such general as Scirpodendron is supposed to have been reduced to a simpler type resembling that in Mapania, and so to the floret of a Scirpus. The florets became re-arranged in new spikelets of a higher order, and from Scripeae other tribes were derived. Evidence from the prophylls, so far as they are known, does not favour the synanthial theory. The scirpoid type of single prophyll subtending only one bud, if any, opposite one of the keels can be matched in many families of monocotyledons, so it is likely to be primitive for Cyperaceae, while the divided type of prophyll subtending two buds is restricted to Mapanieae, and appears to be a specialization. Possibly Scirpus memhranaceus should be transferred to the Mapanieae which it resembles in general habit and massive inflorescence, as well as in spikelet structure (Compare Clarke’s 1909 flgure with my flg. 9 F). Reductions of the florets by loss of perianth, stamens or ovary occur in other tribes of Cyperaceae where they are always regarded as specializations, and a solitary, apparently terminal ovary is found in some species of Scleria. A monograph giving the general structure of Mapanieae with plans of the branching is still needed (Kunth in 1837 wrote of them “structura mihi adhuk obscura”), but the little we know favours an evolution from Scirpus to Scirpodendron (J-L) rather than the reverse. Note on Cariceae The Cariceae are, on the whole, an extratropical group, but are well represented on the mountains of East Africa. Fully discussed by Schultze-Motel (1959), Koyama (1961), Kern (1962) and Raynal (1963), with references to earlier work, they are unique in the placing of the solitary female flower in the axil of a utricle, a modified prophyll at the base of a spikelet, and not of a glume. How the flower reached that position is unknown, but there is no need to follow Gilley (1952) in his proposal for a new family, Kobresiaceae, for in Mapanieae also the prophyll is fertile, though it subtends two male flowers, not one female. Hamlin (1955) says that authorities generally look to the Rh5mchos- poreae for the ‘progenitors’ of the tribe, but they may come from Sclerieae in which the flowers were already unisexual and without a perianth. Page 66 Prophylls and Branching in Cyperaceae Bentham (1877) actually placed Kohresia in Sclerieae and Schoenoxiphum rufum Nees as figured by Clarke (1909), resembles Scleria naumanniana. The point cannot be determined on spikelet structure alone, but anatomical or embryological studies might settle the matter. Nature of the prophyll Blaser (1944) believed the cyperaceous prophyll to be ‘merely a leaf, ‘occasionally distinguishable by position and modifiable in various ways’, and Koyama (1961) suggested an origin within the Cyperaceae from a ‘metamorphosed bract scale’ which ‘would have become empty by the abortion of its axillary flower secondarily’. The ‘occasionally’ seems weak, for there is seldom any difficulty in making the distinction. Koyama’s further statement that prophylls are homologous within genera also seems too limited, for they seem homologous throughout and even beyond, the family. Their very rare absence, as in the inflorescences of Schoenus (fig. 6, J) and Chrysithrix (fig. 7, D) is at once apparent. Nor does their presence ‘differentiate rachillae from branches’ for prophylls are found at the bases of both, and only at the bases of the florets are they absent. Origin and development of Cyperaceae Ziegenspeck (1963), Schultze-Motel (1959) and Takhtajan (1959) have summed up the evidence for the classical conception of an origin of Cyperaceae from Liliiflorae, probably through Juncaceae, evidence based on the structure of the plant, pollen and embryo, with adequate references. But Kern (1962) found a derivation ‘from Liliiflorae, especially Juncaceae, impossible’. If the scirpoid t5rpe of spikelet is the most primitive in Cyperaceae it is certainly not impossible, and fig. 9 suggests how it may have happened. Wind pollination, with reduction of the flowers and their grouping in clusters are characteristics of Juncaceae. The arrangement of the bracts, prophylls, branches of different orders and lengths, and sessile terminal and stalked lateral flower clusters in Juncus lamprocarpus, Ehr, and the loss of prophylls within the clusters in Luzula campestris DC. (Buchanau, 1866) precisely parallel the inflorescence structure of typical Scirpeae. Further reduction of the perianth, association of the florets in spikelets and reduction of the ovules to one would then lead to a form of the Scripus sylvaticus habit. From this other Cyperaceae could have evolved, including the paniculate Desmoschoenus, chosen by Koyama (1958) as closest to the prototype of Scripus. The only form difficult to derive in this way is Oreoholus, where the flowers are not arranged in spikelets. Possibly this comes, by reduction of the ovules to one, from a juncaceous plant similar to Distichea, but with bisexual flowers. It should, in this case, be removed from the Cyperaceae. J. E. Afr. nat. Hist. Soc. VoL XXVI No. 1 (113) Page 67 Primitive features would then be, as Tutin (1954) stated, leafty stems, spikelets of many bisexual florets spirally arranged with few or no sterile glumes, each flower with a perianth, three stamens and three stigmas, and the ovary ripening to a trigonous nut not enclosed by a perigon or closely enfolded glume. The leafy stem is somewhat doubtful, but there may be added the presence of a prophyll at the base of all branches down to spikelet level and the absence of branching from the axils of these prophylls. Classiflcation could follow orthodox lines without the changes in order implied by the synanthial theory, returning to Clarke’s system as modified by Marloth in the Flora of South Africa (1915), given by Koyama (1961), only substituting Rhynchosporeae for the less familiar Schoeneae: 1. Scirpeae (Primitive Cyperaceae) 2. Cypereae (From Scirpeae. Glumes distichous, perianth lost) 3. Rhynchosporae (From Scirpeae. Flowers usually few, each often enwrapped by the glume above). 4. Mapanieae (Possibly from Rhynchosporeae. Prophylls of spikelets more or less split and subtending two male flowers, usually with further monandrous male flowers below the solitary female floret, perianth lost). 5. Sclerieae, including Lagenocarpeae (From Rhynchosporeae. Flowers unisexual, the female solitary, the male above the female or in separate spikelets, perianth lost). 6. Cariceae (Possibly from Sclerieae, Prophylls of spikelets often form- ing utricles, subtending one female flower, with male florets above or in separate spikelets.) Tropical Cyperaceae Corner (1954), developing his ‘durian theory’, suggested that many extra-tropical plants were the dwarfed and secondarily simplified descendants of larger and more primitive types of tropical forests, where many of these forms still lived. Holttum (1948) and Kern (1962) believed that a study of tropical Cyperaceae might throw light on the origin of the family. But near Lagos and Kampala few Cyperaceae grow in dense forest, in contrast to the many specialized forest Gramineae, mostly Paniceae with broad leaves and green stilt roots. Only Cyperus maculatus Boeck. and C. soyauxii Boeck. are restricted to shade, and these are not specially remarkable species. Some larger species of Scleria, S. racemosa Poir. and verrucosa Willd. and the large Cyperus renschii Boeck, and Page 68 Prophylls and Branching in Cyperaceae fischerianus Schimp. ex Hochst. grow in forest, but along streamsides where there is some break in the canopy, and they also grow vigorously where it is damp but there are few trees. Fuirena is tolerant of light shade, but grows better in the open and only on ground which is damp for at least part of the year. Scleria naumanniana is a plant of dry bush, but only where the bush is open, usually from human inter- ference. Even the clim.bing Scleria harteri is confined to low secondary bush and is not found in tall forest. In fact most of our Cyperaceae are plants of open damp places, par- ticularly shallow swamps rather than of shady or dry situations. Their abundance depends on destruction of forest or unsuitability of the habitat for forest growth, and they have undoubtedly been spread by human habitation, cultivation and burning. Many are found in the precincts of towns. Some Mapanieae do grow in damp forest conditions, but their morphology appears peculiar and they seem to represent an end product of evolutionary specialization. Cyperaceae appear to have invaded the forest from more open situations rather than the reverse. Even if the essentials of the durian theory were accepted this would not imply the primitiveness of tropical forms. For, as Corner (1954) pointed out, Juncaceae are mainly ‘leptocaul and extra-tropical’, though possibly derived from a pachycaul such as Prionia. (There are no Juncaceae in the forests and savannahs of Nigeria proper or lowland Uganda, but they grow on the highlands of East Africa and the Came- roons.) If then Cyperaceae are derived from Juncaceae the transition probably occurred outside the rain-forest. References Arber, a. (1925). Monocotyledons. Cambridge: Univ. Press. Bentham, G. (1877). On the distribution of the monocotyledonous orders into primary groups, more especially in reference to the Australian flora, with notes on some points of terminology. J. Linn. Soc., Bot., 15: 490-520. Blaser, H. W. (1941,a). Studies in the morphology of the Cyperaceae I. Morphology of flowers. A. Scirpoid genera. Amer. J. Bot., 28 : 541-551. Blaser, H. W. (1941,b). Studies in the morphology of the Cyperaceae I. Morphology of of flowers. B. Rhychosporoid genera. Amer. J. Bot., 28: 832-838. Blaser, H. W. (1944). Studies in the morphology of the Cyperaceae II. The prophyll. Amer. J. Bot., 31: 53-64. Buchenau, F. (1866). Der Bliithenstand der Juncaceen. Jb. wiss. Bot., 4: 383-440. Celakovsky, L. (1887) Ueber die ahrchenartigen Partialinflorescenzen der Rhynchosporen Ber. dtsch. Bot. Ges., 5: 148-152. Clarke, C. B. (1909). Illustrations of Cyperaceae. London. /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Page 69 Corner, E. J. H. (1953-1954). The Durian Theory extended. Parts I, II & III. Phytomorphology 3: 465-76, 4: 152-65, 263-74. Gilley, C. L. (1952). Phylogenetic development of the inflorescence and generic relationships in the Kobresiaceae. Iowa State Coll. J. Sci., 26: 210-212. Goebel, K. (1888). Morphologische und biologische Studien III. Ueber den Bau der Aerchen und Bliiten einiger Javanischen Cyperaceen. Ann. lard. Bot. Buit, 7 : 120-138. Hamlin, B. G. (1955). Key to the general of Cyperaceae in New Zealand. Tautera, 6: 27-38. Hirmer, M. (1931). Zur Kenntnis der Schraubcnstellungen im Pflanzenreich. Planta 14: 132-206. Holttum, R. E. (1948). The spikelet in Cyperaceae. Bot. Rev., 14: 525-541. Hutchinson, 1. and Dalziel, J. M. 1963). Flora of West Tropical Africa. London: Crown Agents. Kern, J. H. (1962). New look at some Cyperaceae mainly from the tropical standpoint. Advancement of Science. 19: 141-148. Koyama, T. (1961). Classification of the family Cyperaceae (1). J. Fac. Sci. Univ. Tokio, III, Bot., 8: 37-148. Kunth, C. S. (1837). Cyperographia synoptica sive enumeratio cyperacearum omnium hucusque cognitarum ajectis characteribus, differentius et synomymis. Stutgardiae et Tubingae: J. C. Cottae. Mattfield, j. 1938). Das morphologische Wesen und die Phylogenetische Bedeutung der Blumenblatter Ber. Dtsch. Bot. Ges., 56: 86-116. Monoyer, A. (1934). Contribution a I’anatomie du genre Scirpus. Arch. Inst. Bot Univ. Fiege, 11: 1-185. Mora, L. E. (1960). Beitrage zur Entwickingsgeschichte und vergleichenden Morphologie der Cypereen. Beitr. Boil. Pfianz., 35: 253-341. Nees von Essenbeck, C. G. (1835). Ubersicht der Cyperaceengattungen. Finnaea, 9: 273-306. Nelmes, E. and Baldwin, J. T. (1952). Cyperaceae in Liberia. Amer. J. Bot., 39: 368-393. Pax, F. (1886). Beitrage zur Morphologie und Systematik der Cyperaceen Bot. lb., 7; 287-318. Raynal, j. (1936). Notes cyperologiques, I. Afrotrilepis, nouveau genre africain. Adansonia, 3: 250-265. Ruter, E. (1918). Uber Vorblattbildung bei Monokotylen. Flora, 110: 193-261. Schultze-Motel, W. (1959). Entwicklungsgeschichtliche und vergleichend-morphologische Untersuchungen im Bliitenbereich der Cyperaceae. Bot. Jb., 78: 129-170. Takhtajan, a. (1959). Die Evolution der Angiospermen. Jena: Fischer. Tutin, T. G. (1954) Cyperaceae. In Clapham, A. R., Tutin, T.G. and Warburg, E. F., Flora of the British Isles. Cambridge: Univ. Press. Walters, S. R. (1950). On the vegetative morphology of Eleocharis R. Br. New PhytoL, 49: 1-7. ZiEGENSPECK, H. (1938). Die Phylogenie der Glumiflorae. Bot. Arch., 39: 177-205. Page 70 Prophylls and Branching in Cyperaceai INDEX TO SPECIES Ascolepis capensis Ridley Cladium mariscus (L.) Pohl . Cypcrus fischerianus Schimp. ex. Hochst. C. maculatus C.B.U C. mundtii Nees C. subumbellatus Kuk. C. tenuis Swartz .... Eleocharis mutata Roem. & Schult. Fimbristylis dichotoma (L.) Vahl F. hispidula (Vahl) Kunth F. obtusifolia (Lam.) Kunth Fuirera umbellata Rottb. Hypolytrum sp. nr. heterophyllum Boeck. Kyllinga erecta Schumach. . K. pumila Mich. .... Remirea maritima Aubl. Rhynchospora alba. (L.) Vahl R. corymbosa (L.) Britt. Schoenus nigricans L. ... Scirpus brachyceras Hochst. Scleria barteri Boeck. S. naumanniana Boeck S. racemosa Poir. .... S. verrucosa Wild Page Fig. 58 4 60 6 68 67 59 4 57—58 4 56—61 4 53 2 52 1—2 53— 61 2 51 1 54— 56 3 64 8 59 4 4 59 60—61 5 59—61 5 60 6 53 2 64—68 62— 68 7 67 63— 67 7 (Received for publication 28th March, 1966) J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Prophylls and Branching in Cyperaceae J. E. Air. nat. Hist. Soc. VoL XXVI No. I (113) disseminule. Q — K. erecta, stolon. R — plan of Q. S — C. mundtii, fragment of stem. Prophylls and Branching in Cyperaceae /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 1 (113) Fig. 6. RHYNCHOSPOREAE: Cladium, Schoenus. of B. D — cluster of spikelets. E — terminal spikelet and bases of three others. F — Third, enwrapping glume. G — spikelet spread. H — plan of J. L — two spikelets. M — plan of L. N — rachilla and florets. Prophylls and Branching in Cyperaceae Fig. 7. Scleria: D — prophyll branching. E — plan of D. F — bisexual spikelet, spread. G — plan of F. H — female spikelet, fruit fallen. I — plan of H. J — from K. M — fragment of inflorescence. N — plan of spikelet group. O — male spikelet, spread to fifth glume. P — plan of O. Q — female flower. ]. E. Afr. Hist. Soc. Vdl. XXVI No. ! (113) Prophylls and Branching in Cyperaceae Fig. 9. Origin of Cyperaceae, orthodox theo A — Liliiflorous plant, with single flower scirpoid plant with flower /. E. Air. nat. Hist. Soc. Vol. XXVI. No. 1 (113) Page 71 BOOK REVIEWS FliORA ZAMBESIACA This flora, produced by a distinguished team of British, Portuguese and Rhodesian botanists deals with the seed plants of Malawi, Zambia, Botswana, Rhodesia and Portuguese East Africa. Although the first part appeared in 1960 the existence, excellence and usefulness in East Africa of the work seem much too little known to naturalists in this part of the world. The four parts which have so far appeared cover the Gymnosperms and all Dicotyledonous families before the Leguminosae in the sequence of Bentham and Hooker. Every taxon is fully described, almost every genus is illustrated in black and white and there is a fine coloured plate at the beginning of each part. The key to families in vol. 1, part 1, provides the best available method by which an unknown indigenous East African plant can be assigned to its family. While Flora Zambesiaca will clearly be more useful to naturalists in southern Tanzania than to those further north, as long as ‘The Flora of Tropical East Africa’ is incomplete it is a ‘must’ for every botanist in East Africa and, on account, especially, of the plates, it will remain useful even when F.T.E.A. is complete. Particulars of the parts which have so far appeared are as follows. Vol. 1, part 1 (1960) up to Polygalaceae, 336 p., 59 pi., 25/-. Vol. 1, part 2, (1961) Caryophyllaceae-Sterculiaceae, 245 p., 49 pL, 25/- Vol. 2, part 1 (1963) Tiliaceae — Icacinaceae, 351 p., 75 pL, 30/- Vol. 2, part 2 (1966) Aquifoliaceae — Connaraceae, 300 p., 62 pi., 30/- All parts are obtainable from the Govt. Bookshop, P.O. Box 569, London S.E.l. and Samcax Book Services Ltd., P.O. Box 2720, Nairobi. J.B.G. GRASSES OF TANGANYIKA by D. M. Napper, and other works on E. African grasses. The appearance in January, 1966 of Miss Napper’s splendid little book, whose title should have been ‘Grasses of Tanzania’ since Zanzibar and Pemba are included, means that we now have excellent manuals for the identification of grasses in each of the three East African countries. All the 700 grass species, included in 172 genera, which are known to occur in Tanzania, are keyed out, readily visible characters being used, in order to help the field worker, wherever possible. A brief descriptive Page 72 Book Reviews note and a summary of the known distribution are supplied for each species, and the fact that specimens are cited will be a great help to the herbarium taxonomist. The essential features of 219 species, including the typical forms of genera and difficult or unusual species, are illustrated in 20 full page composite plates. A map, short glossary and full' index complete the work. The Government Printer Dar es Salaam and his staff are to be congratulated on the high standard of printing. Only two errors have been detected. That on p. 127, where the caption for fig. 188 has been misapplied to fig. 187, whose true caption should be Digitaria diagonalis, spikelets x 8, is important. Particulars of the companion works, for Kenya and Uganda as well as those of ‘The grasses of Tanganyika’ are given below. It should be noted that the 123 full page plates, illustrating over 420 species which appear in the Uganda book form a most valuable supplement to the works for Kenya and Tanzania. A. V. Bogdan ‘A revised list of Kenya Grasses, with keys’ (1958) obtainable from Govt. Printers, P.O. Box 30128, Nairobi, 5/-, postage 30 cents. K. W. Harker & D. M. Napper ‘An Illustrated Guide to the Grasses of Uganda’ (1961). Obtainable from Govt. Printer, P.O. Box 33, Entebbe 22/50, postage 65 cents; or leading Nairobi bookshops. D. M. Napper ‘Grasses of Tanganyika’ (1965). Obtainable from Govt. Printer, P.O. Box 9124, Dar es Salaam, 21/- postage 1/25; or leading Nairobi bookshops. J.B.G. BOOKS ON THE FERNS OF TROPICAL AFRICA For many years naturalists in East Africa have felt the need for books, especially illustrated books, which would help them to identify ferns. Until recently almost the only work available was T. R. Sim ‘The Ferns of South Africa’ Cambridge U.P., 2nd Edn. 1915. This illustrates 288 species in 181 plates and is still of great value, but the representation of tropical species is inadequate, the nomenclature is much out of date and the book can be obtained, through second hand booksellers, only occasionally and with difficulty. For a really satisfactory account of East African ferns we must await the appearance of the relevant part of ‘The Flora of Tropical East Africa’ which is now being prepared by Dr. F. M. Jarrett at Kew. Those who /. E. Afr. nat. Hist. Soc. Vol. XXVI. No. 1 (113) Page 73 now collect ferns in East Africa, especially if they are able to visit the less well known areas such as the Eastern slopes of Mt. Kenya, the Nyambeni range, the Nguru and Uluguru mountains in Tanzania, south eastern Tanzania in general and the more inaccessible mountains of Uganda, have the opportunity to contribute to Dr. Jarrett’s work. For their guidance several recent works dealing with the ferns of West Tropical Africa which have received little or no attention in East Africa will be found of great value. In 1953 appeared ‘Les Pteridophytes de I’Afrique intertropicale Francaise’ (Memoire 28 of I.F.A.N.) by Mme M. L. Tardieu-Blot, in which 256 species in 66 genera are described; 205 of these are illustrated in 44 plates. This work is completed in the first section of Memoire 50 of I.F.A.N. (1957) in which 46 species in 7 genera of fern allies are described, 24 of them illustrated in 8 plates. ‘The Ferns and Fern Allies of West Tropical Africa’ (1959) by A. H. G. Alston keys out 305 species in 72 genera: 15 species are illustrated in 11 plates. Lastly in 1964 appeared Vol. 3 ‘Pteridophytes’ of the ‘Flore du Cameroon’ in which Mme. Tardieu-Blot describes 256 species in 69 genera, 181 species being illustrated in 55 plates. Naturally this repeats to a large extent the author’s earlier work, in fact 13 of the plates are identical, but the treatment of many of the species has been considerably improved. In order to assist those who may wish to obtain these works the addresses from which they may be obtained with current prices are given below. Postage is in each case extra. M. L. Tardieu-Blot ‘Les Pteridophytes de I’Afrioue intertropicale Francaise’ 1 Memoire de I.F.A.N. 28. 16 N.F. (c. 25/-) 2 Mem. de I.F.A.N. 50. 20 N.F. (c. 30/-) Librairie Clairafrique, 2 Rue Sandiniery, B.P. 2005, Dakar, Senegal. H. G. Alston ‘The Ferns and Fern Allies of West Tropical Africa’ The Govt. Bookshop, P.O. Box 569, London S.E. 1. 7/50. M. L. Tardieu-Blot ‘Flore du Cameroun, 3, Pteridophytes’ Mus. Nat. D’His. Naturelle, Lab. de Phanerogamic. 16 Rue Buffon, Paris 5c. 57 N.F. (c. 85/50). J.B.G. JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM CONTENTS Page A June- July Census of Small Mammals on the Athi Plains, Kenya 1 By D. S. Hartman Some Gasteromycetes from Eastern Africa 5 By D. M. Bring and R. W. Rayner Report on Birdringing for 1961 — 1966 47 By J. B. Smart Preliminary Observations on the Effects of Water Flow on Protozoan 53 Populations By Apollo H. Ogambo-Ongomo The Reef Heron, Egretta schistacea Ehrenb., in Interior East Africa 61 By Oscar T. Owre Identification of Aloes in East Africa 65 By J. B. Gillett Marine Botany of the Kenya Coast, I A First List of Kenya Marine Algae 75 By Wm. Edwyn Isaac Notes on a Collection of Amphibians from Ethiopia 85 By Emil K. Urban Nature Notes: Migration of Lesser Spotted Eagle, 87 R. V. Bowles Possible Occurrence of the Whale-headed Stork in Ethiopia, 87 E. K. Urban Book Reviews 88 (Published 20/11/67) EAST AFRICA NATURAL HISTORY SOCIETY NOTICE TO CONTRIBUTORS Contributions. The Committee is pleased to consider contributions on natural history for publica- tion in the Journal on the understanding that these are not also being offered, wholly or partially, to any other Journal. They should be addresssed to the Secretary, P.O. Box 4486, Nairobi. Typescript. Articles should be typed on one side of the paper, in double spacing and with wide margins. Illustrations. These should be in a form suitable for reproduction. The Editor cannot be expected to re-draw. Line drawing should be in Indian ink on Bristol board or thick white paper. Reproduction will be better if they are drawn larger than it is intended that they should appear. An indication of the degree of reduction is advisable. Photographs should be printed on glossy paper and a better reproduction is achieved from prints slightly darker in tone than normal . Nomenclature. Where a recent standard work for the area is available (e.g. Praed and Grant for birds) the names given there (both EngUsh and scientific) should be used. Initial capitals should be used for specific English names, e.g. Pied Wagtail and small intitial letters for group names, e.g. wagtails. Scientific names must always be underlined and followed by the naming authority when mentioned for the first time. Where an English name is used, it must, on first mention be accompanied by the scientific name. References. These are usually abbreviated in the text and listed more fully in alphabetical order of authors at the end of the article. For example, in the text book reference might be (Pinhey 1956: p. 20). At the bottom of the contribution: Jackson, F. J., 1938. Birds of Kenya and Uganda. Pinhey, E. C. G., 1956. The Emperor Moths of Eastern Africa. J.E. Afr. Nat. Hist. Soc. XXIII No. 1, (98). With short articles it may not be worth making a list of references at the end, but the whole reference in the most abbreviated comprehensible form should then be inserted in the text. Reprints. Provided that they order at the time of submitting their articles, authors may have up to 25 reprints free (other than Nature Notes). Additional copies can be supplied on payment. A list of available back numbers of the Journal and reprints can be obtained from the Librarian, National Museum, P.O. Box 658, Nairobi. JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM CONTENTS Page A June- July Census of Small Mammals on the Athi Plains, Kenya 1 By D. S. Hartman Some Gasteromycetes from Eastern Africa 5 By D. M. Bring and R. W. Rayner Report on Birdringing for 1961 — 1966 47 By J. B. Smart Preliminary Observations on the Effects of Water Flow on Protozoan 53 Populations By Apollo H. Ogambo-Ongomo The Reef Heron, Egretta schistacea Ehrenb., in Interior East Africa 61 By Oscar T. Owre Identification of Aloes in East Africa 65 By J. B. Gillett Marine Botany of the Kenya Coast, I A First List of Kenya Marine Algae 75 By Wm. Edwyn Isaac Notes on a Collection of Amphibians from Ethiopia 85 By Emil K. Urban Nature Notes: Migration of Lesser Spotted Eagle, 87 R. V. Bowles Possible Occurrence of the Whale-headed Stork in Ethiopia, 87 E. K. Urban Book Reviews 88 (Published 20/11/67) Price Shs. 25/- EAST AFRICA NATURAL HISTORY SOCIETY President: DR. M. J. COE Vice-President: DR. A. D. Q. AGNEW Executive Committee: L. H. BROWN, ESQ. M. E. W. NORTH, ESQ. R. H. CARCASSON, ESQ. MISS E. J. BLENCOWE MRS. D. FLEMING B. PARSONS, ESQ. J. SMART, ESQ. P. M. OLINDO, ESQ. C. J. BEECHER, ESQ. DR. J. B. FOSTER J. S. KARMALI, ESQ. Hon. Editor: DR. P. J. GREENWAY Hon. Treasurer: MISS J. OSSENT Secretary: MRS. F. ng’weno All correspondence in connection with this Journal should be addressed to: The Secretary, East Africa Natural History Society, P.O. Box 4486, Nairobi, Kenya. J. E. Afr. nat. Hist. Soc. Vol XXVI No. 2 (114) Page 1 A JUNE-JULY CENSUS OF SMALL MAMMALS ON THE ATHl PLAINS, KENYA By D. S. Hartman (Conservation Department, Cornell University, Ithaca, New York) INTRODUCTION Ecological studies of small mammals in East Africa are scarce. What little intensive work that has been undertaken centres on observations of habitat distribution and breeding cycles. Southern and Hook collected data on the distribution of Soricids and Murids in Uganda high forest (1963a) and later studied the reproductive condi- tion of some insectivores and rodents of Uganda and Kenya (1963b). Delany (1964) summarized all quantitative work on small mammals in Africa south of the Sahara and added his own results from trappings in Uganda. The following superficial observations, the first on the Athi Plains, are presented merely as a guidepost for further investigations. Briefly, the study area falls within the semi-arid “scattered tree” or Acacia- Themeda grassland biome (Edwards & Bogdan, 1951). The area is characterized by an abundance of red oat grass, Themeda triandra Forsk., whose dominance is ensured by periodic fires and impeded soil drainage, both of which factors retard the advance of tree and bush constituents (Ibid). Where conditions are favourable, stunted whistling thorn. Acacia drepanolobium Harms ex Sjdstedt, is found in association with Themeda. On the plains trees and shrubs are otherwise found only in riparian depressions. (For a fuller description of floral constituents see Heriz- Smith, 1962.) Mean annual rainfall on the plains varies between 20 and 30 inches. Years with less than 20 inches are not uncommon. Drought is a feature of the habitat. Rain, when present, is characteristically sporadic, localized, torrential, and short-lived. Run-off is rapid; evapo-transpiration loss, high. Possibly more imoprtant with regard to the ecologic distribution of small mammals is the prevalence of a heavy black clay — the infamous “black cotton” soil of East Africa. This clay becomes glutenous and waterlogged during the rains, concretionary and impenetrable during drought, subjecting the homogeneous surface horizon to its annual kneading and mixing and perhaps making conditions for burrowing and life in the subsoil intolerable for shrews and rodents. The purpose of the sampling programme was primarily to determine the species composition and relative numbers of the small mammals of the plains habitat. The two months devoted to the study coincided with the onset of the long dry season during which period the vegetation was either overgrazed, mowed, burned, or simply drying up — in all cases depleting food and cover sources. It was hoped a significant correlation between environmental pressures and population density would be evident. Secondarily, ancillary observations were to be made on reproductive condition, niche diversity, and habitat preference. METHODS AND MATERIALS The species composition and relative density of small mammals were obtained by line-trapping. A slight modification of the NACSM B-type line (Calhoun, 1951) was adopted. Page 2 A June-July Census of Small Mammals Two parallel lines, 60 snap traps to a line, were set over 400 yards apart but in the same vegetative faciation. Each line was divided at 50 ft. intervals into 20 stations. At every station two standard 7 x 3" rat traps and one 4 x 2" mouse trap were set within two yards of the centre of the station in spots most likely to capture animals. Various baits were separately experimented with, including cheese, raisins, and aniseed oil, none of which yielded any greater trapping success than the ultimately chosen mixture of peanut butter and rolled oats. Traps remained set for three conse- cutive days after which the transects were relocated in a new area. The distribution of trapping locations is presented in Figure 1 . Typically, two locations were sampled in a week. Vegetative faciations sampled included: riverine forest (6 days); over- grazed, mown or burned grassland and bush (9 days); and undisturbed Acacia- Themeda (21 days). Figure 1. Distribution of trapping locations on the Athi Plains. MAJOR ROAD . • LOCATION OF TRAP LINES /VAA. = MAJOR RIVER /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 3 Trapping was conducted from May 30th through July 28th, 1966. While operative, the trap lines were checked daily at 6:30 a.m. and 6:30 p.m. Collected animals were subjected to routine necropsy procedures in the laboratory with especial emphasis on reproductive condition. Study skins were sent to the Cornell University Museum. RESULTS In the course of 4,320 trap-nights a total of only 12 mammals of but two species was collected (Table 1). All captures were made at night. The categorization of the faciations in which the trap line pairs were set is not included as trapping success showed no bias toward a particular vegetative type. The results are therefore lumped under the plains habitat as a whole. Table 1 SPECIES NUMBERS, COMPOSITION, AGE AND SEX DISTRIBUTION, AND REPRODUCTIVE CONDITION No. Species Age Sex Reprod. Cond. Date 1 Insectivora : Crocidura fumosa Thomas A F breeding 28-6-66 2 A M non-breeding 7-7-66 3 „ „ A M non-breeding 8-7-66 4 „ „ A F breeding 12-7-66 5 A F breeding 28-7-66 6 A F breeding 28-7-66 1 Rodentia: Mastomys coucha A. Smith J M non-breeding 8-7-66 2 A M breeding 12-7-66 3 jj „ A M breeding 12-7-66 4 „ „ A M breeding 14-7-66 5 A F breeding 14-7-66 6 J M non-breeding 14-7-66 In addition, mole-rats, Tachyoryctes splendens (Rueppel), were observed tunneling on several occasions, usually in the loosened black soil around aardvark diggings. Though only a single animal was captured in the month of June as opposed to the remaining 1 1 in July, signs of presence (faeces, runways, middens, and cuttings) were much in evidence in early June but rare by late July. Correspondingly, insects, seeds, and green shoots were abundant at the beginning of the study and showed a progressive decline into the dry season. DISCUSSION Any inference from such scant results must be made with care. The fact that 80% of the adults captured (100% of the females) were reproductively viable stands in opposition to the assumption that a breeding peak should coincide with the end of the rains when environmental pressure would be minimal, when the replacement of vegetation would provide optimal conditions of cover and food. Approaching the end of the sampling period, reproducing individuals showed no trend toward declining reproductive stages. Otherwise, the numbers trapped are simply too small either to statistically analyze or to gauge cause and effect factors. Page 4 A June-July Census of Small Mammals Small mammal populations have been shown to be cyclic, to fluctuate regularly in numbers, where unstable environmental circumstances prevail, notably on the arctic tundra. The reason or reasons for their predictable fluctuations are still a matter of dispute. It is doubtful that the populations of the Athi Plains cycle as such but probable that they fluctuate irregularly dependent upon climatic and affiliated changes. From personal inquiries I gather that there have been eruptions of small mammals on the study area. No one can recall, however, the duration of these periods or the seasons in which they occurred. The two month sample provides little insight into the question other than a presumed ebb in numbers if indeed a fluctuation is present. Equally undiscovered are the adaptations the animals have evolved to tide them through the annual droughts on the grasslands. The results give no indication of mass death, aestivation or emigration to more favourable areas (water catchment basins, for example). The data gives only an index of the density and reproductive status of small mammals over a two month interval and professes to be no more than a framework for a future study which would demand at least four years. SUMMARY Coinciding with the advent of the “long dry season” in Kenya, a two month sampling of the small mammal population of the Athi Plains was carried out deter- mining (1) the species composition and numbers and (2) the reproductive status of the animals collected. The results of over 4,300 trap nights indicated a dearth of small mammals, yielding only 12 individuals of two species, 8 of which were in breeding condition. Environmental factors influencing the paucity of animals are discussed. The possibility of a fluctuation in numbers is inferred. ACKNOWLEDGEMENTS The author wishes to thank Dr. J. B. Foster of the Dept, of Zoology, University College, for his suggestions and criticisms, Alexander Duff-Mackay of the National Museum for his help in identification of the specimens, and Miss Jean Angwin also of the Dept, of Zoology for her assistance in the preparation of study skins. REFERENCES Calhoun, J. B., 1951. North American census of small mammals. Release No. 4 (1950 Annual Report). Roscoe B. Jackson Memorial Lab., Bar Harbor, Maine. 135 pp. Delany, M. j., 1964. A study of the ecology and breeding of small mammals in Uganda. Proc. Zool. Soc. Land., 142: 347-370. Edwards, D. C. & Bogdan, A. V., 1951. Important Grassland Plants of Kenya. Pitman, Nairobi. 124 pp. Heriz-Smith, S., 1962. Wild Flowers in the Nairobi National Park. D. A. Hawkins, Ltd., Nairobi. 56 pp. Southern, H. N. & Hook, O., 1963a. A note on small mammals in East African forests. J. Mamm., 44: 126-129. Southern, H. N. & Hook, O., 1963b. Notes on breeding of small mammals in Uganda and Kenya. Proc. Zool. Soc. bond., 140: 503-515. {Received on 3rd November, 1966) J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) SOME GASTEROMYCETES FROM EASTERN AFRICA By D. M. Dring and R. W. Rayner INTRODUCTION During recent months a number of collectors have sent gatherings of gasteromycete fungi from E. Africa to the Kew Herbarium. This paper has been written to record the results of their collecting as well as to give them, and other workers in the area, an admittedly incomplete but, it is hoped, useful guide to the puffballs and their allies in the East African region. Recently collected material has been supplemented by studies of the older collections in Kew (K) and elsewhere, particularly the E. African Herbarium (EA). Though this study is centred on E. Africa in a restricted sense we have also included some material from Malawi, Zambia, Rhodesia and Mozambique where considered appropriate. In a like manner Somalia has also been included. Perhaps, however, it is more important to note that we have also included the rather few gasteromycetes which we have examined from the Mascarene Islands. Though their nearest mainland is the east coast of Africa, these islands are known to have strong floristic affinities with Asia and Australasia rather than with Africa. So that, unless there is evidence to the contrary it should not be assumed that species recorded from these islands will occur on the African mainland. The colour names used, excepting those describing microscopic characters, are based on Dade, Colour Terminology in Biology ed. II, Mycol. Pap. 6, 1949. COLLECTING Gasteromycetes are easy to collect and preserve. All except Clathraceae and Phallaceae should simply be dried quickly and placed in boxes or packets with the usual data on place of collection, date, etc. They should never be pressed. If the collector has the time, inclination and very modest skill required, he should make pencil sketches, or better, water-colour paintings of the fresh material. In the case of phalloids, thou^ dried material is better than none, the best method of preservation is to put at least part of each collection into bottles of spirit, once again making adequate collector’s notes and if possible a painting or other indication of the colour of the fresh material. GLOSSARY OF TERMS USED ADAXIAL ALVEOLUS AMYGDALIFORM APICULUS APOPHYSIS towards the axis small depression or hollow in a surface almond shaped short projection swelling at the base of the spore-sac BALLISTOSPORE spore which is violently propelled from its mother-cell BASIDIUM the spore-mother-cell of badidiomycetes, bearing spores on short spines or sterigmata CADUCOUS CAESPITOSE CAMPANULATE CAPILLITIUM CLAMP-CONNEXION CLATHEU^TE CLATHROID (n) COLLAR COLUMELLA CORTEX CRENULATE falling away early growing from a single point bell-shaped mass of sterile, thread-like hyphae mixed with the spores (cf. paracapillitium) characteristic protuberance at the septum in certain hyphae of some basidiomycetes in the form of a lattice a member of the Clathraceae of a phalloid, the pad of tissue at the apex of the cap surrounding the apical perforation, if any; of Tulostoma, that part of the outer peridium which adheres to the head and surrounds the socket a sterile prolongation of the stipe into the gleba (cf. pseudo- columella) of the peridiole of Nidulariaceae, the dark-coloured layer of sterile bases, longitudinally wrinkled Page 6 Somef Gasteromycetes from Eastern Africa DIAPHRAGM of Lycoperdaceae, a membrane separating gleba from subgleba and confluent with the endoperidium ECHINATE EGG ENDOPERIDIUM EVANESCENT EXOPERIDIUM spiny of phalloids, the immature fruit-body and its enclosing universal veil the inner layer of the peridium in Lycoperdales fleeting the outer layer of the peridium in Lycoperdales FARINA FLOCCOSE, FLOCCULENT FUNICULUS FUSIFORM floury coating cottony of Nidulariaceae, an elastic cord joining the peridiole to the cup wall spindle-shaped GLEBA GLEBIFEROUS spore-mass bearing the gleba HYMENIUM fertile layer; in basidiomycetes it is composed of basidia IMBRICATE INDUSIUM overlapping like tiles on a roof of “Dictyophora”, the net-like organ hanging from near the stipe apex, under the cap MAMMOSE MULTISERIATE breast-like of the stipe wall of a phalloid, consisting of more than two layers of chambers NAKED of a stoma, without a differentiated peristome OSTIOLE stoma, mouth PARACAPILLITIUM PERCURRENT PERIDIAL SUTURE PERIDIOLE PERIDIUM PERISTOME PHALLOID (n) PILEUS PLACENTA PSEUDOCOLUMELLA PSEUDOSTEM PUNCTATE PYRIFORM hyphae resembling those of the true capillitium except that they are hyaline, collapsed, and with frequent septa of a columella, extending right through the gleba of clathroids, membrane joining the universal veil (outer peridium) to the receptacle discrete portion of the gleba, surrounded by its own wall wall or membrane enclosing the fertile part of the fruit-body area surrounding the stoma a member of the Phallaceae or of the Phallales, according to context cap of the agaricoid forms of Sclerodermatales, tissue which nourishes the spore after its discharge from the badidium a ± densely woven, central mass of capillitium stem-like structure with tissues not orientated along the long axis of the fruit-body, usually consisting of ± spongy tissue having minute warts or depressions pear-shaped RECEPTACLE RETICULATE REVOLUTE RIMOSE RUGULOSE the spongy part of a phalloid in the form of a net backwardly curved abundantly cracked, “crazed” finely wrinkled SACCATE SCABROUS SECOTIOID SESSILE shaped like an open bag rough and peeling pertaining to S cotium and the Secotiaceae lacking a stalk Page 8 Some Gasteromycetes from Eastern Africa Plate 1 A. Phallus caliendricus, spirit material, habit, c. X 1 (Type); B. P. caliendricus cap ornament c. X 2 (Type); C. P. caliendricus, dried material, cap c. X 4 (Bally); D. P. caliendricus, dried material, habit, c. x 1 (Bally). /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) the pseudostem in Lycoperdaceae of a puffball, the apical pore through which the spores are dis- charged etched with fine lines compact mass of vegetative hyphae bearing fruit bodies (cf. subiculum) of the stipe wall of a phalloid, consisting mainly of a double layer of chambers a sheet of mycelium covering the substrate and bearing the fruit bodies (cf. stroma) grooved covering of soft hairs structure supporting the hymenium cut-off short of the peridiole of Nidulariaceae, the outermost hyaline layer of the stipe-wall of a phalloid, consisting of a single layer of chambers um-shaped the cup-like basal remains of the universal veil after expansion of the fruit-body THE GASTEROMYCETES In a popular sense the term gasteromycete implies a basidiomycete in which the spore-mass or gleba is enclosed in a sac-like peridium. The familiar puff-ball is a typical example. Unfortunately, this conception does not cover the wide range of forms which are conveniently classified as gastero- mycetes. Technically, gasteromycetes are those higher basidiomycetes whose hymenia are enclosed, at least at an early stage of development, and whose basidiospores are not shot-off from the basidium. Thus, some forms very like agarics and boleti are included in the definition of gasteromycete simply because they do not discharge their spores violently from the basidium. Inability to produce ballistospores has resulted in other means of discharging spores. Ingold (1953) has described the gasteromycetes as “an assorted collection of experiments in spore discharge,” and indeed they are. There is, however, at least one other factor which seems to have contributed to this diversity. Since the violent discharge of the ballistospore depends on the hydrostatic pressure within the basidium, the hymenium must be turgid throughout the period of spore-discharge. Gas- teromycetes do not have this limitation and are therefore better adapted to fruit in dry conditions than other basidiomycetes. Once again, this has resulted in diversification of both habitat and habit. In addition to the puffballs and agaricoid forms referred to above there are many other groups. In the birds’ nest fungi (Nidulariaceae) the gleba consists of pellets (peridioles) which are dispersed from the cup-shaped peridium by raindrops. In several groups the differing water-absorbing capacity of the various layers of the peridium is used as an aid to discharge. In the earth-stars (Geastrum) the outer peridium splits into rays which in some species bend backwards on drying, elevating the spore sac to a height more advantageous for spore-discharge. In other species of Geastrum, and in the similar Astraeus, the rays enclose the spore sac when dry, opening and permitting spore discharge when moist conditions prevail. In Mycenastrum the inner peridium behaves in the opposite way, cracking into lobes, bending back and exposing the powdery gleba when dry; closing when wet. In the Clathraceae and Phallaceae (“stinkhorns”), the foetid gleba is exposed to the attentions of flies, often on a flower-like receptacle. In the Hymenogastraceae and similar hypogeal fungi (“false truffles”) the peridium is ruptured and the spores dispersed by burrowing animals. A glance through the illustrations to this paper will give some guide to the range of fungi which go to make up the gasteromycetes. STERILE BASE STOMA STRIATE STROMA SUB-BISERIATE SUBICULUM SULCATE TOMENTUM TRAMAL PLATE TRUNCATE TUNICA UNISERIATE URCEOLATE VOLVA Page 10 Some Gasteromycetes from Eastern Africa Key to the families discussed Gleba becoming powdery (occasionally granular) at maturity 1 Gleba becoming mucilaginous at maturity 3 Gleba neither powdery nor mucilaginous 4 1. Capillitium of ± thread-like hyphae not present in gleba . . Sclerodermataceae, p. \% Capillitium abimdant 2 2. Fruit-body sessile or with a sterile base of ± spongy tissue not orientated along the long axis of the fruit-body Lycoperdaceae, p. 22 Fruit-body with a true stipe of hard, orientated tissue. Columella absent Tulostomataceae, p. 42 Fruit-body with a true, woody stipe prolonged into the gleba as a percurrent columella Podaxaceae, p. 1 1 3. Gleba usually borne on the inside of the receptacle, which may be sessile or stipitate and consist of a globose network or of several columns united at the top, or of spreading arms Clathraceae, p. 15 Receptacle a single, unbranched column supporting gleba near its apex . . Phallaceae, p. 1 1 4. Hymenium borne on ± gill-like plates. Fruit-body usually stipitate . Secotiaceae,p. Hymenium absent, gleba organized into seed-like peridioles. Fruit-body usually cup- shaped, less than c. 1 cm. diam Nidulariaceae, p. 20 SECOTIACEAE The members of the Secotiaceae are easily recognizable by their gross resemblance to agarics. They have a stipe and an apical pileus which at maturity may spread out like that of a mushroom (e.g. in Montagned) or remain ahnost closed (e.g. Galeropsis). The gleba consists of a true hymenium lining persistent tramal plates. These latter are often radially arranged in the same way as the ^lls of an agaric {Montagnea), or more or less anastomosed but retaining some visible radial orientation (Galeropsis), or so anastomosed that their basic orientation is obscured. The spores of the entire group are brown, double-walled, with an apiculus and usually a germ-pore. In the more agaricoid genera they tend to have the same symmetry as a ballistospore; in the more secotioid genera the spores tend to be irregular or globose. They are typically fungi of arid regions. Only two genera are known to us from E. Africa. Montagnea Fr. (=Montagnites Fr.) This genus consists of Coprinus-VAae plants with radial, non-branched gUls and an expanded pileus. It is usually defined as having the apex of the stipe expanded into a small disc, from the margin of which hang the gills. Much material, especially when gathered m an over-ripe condition does appear to fulfil this definition. However, material in good condition clearly shows that the pileus covers the whole of the abaxial edge of the gills but is very thin and splits between them. In older specimens the giUs become twisted and the thin backing of pileal tissue may be obscured, the gills appearing to be attached only where they abut onto the disc. Complete specimens have a well-developed volva. The stipe is hollow in all known Montagnea species, and this character, together with the non-branched gills, clearly separates them from the allied genus Gyrophyragmium Mont. There is but a single well known species, M. arenaria (DC.) Zeller {=Montagnites candollei Fr.)- One of us (R.W.R.) has collected it in the Rift Valley, nr. Suswa volcano and we are informed by Dr. N. Otieno of its occurrence north of Isiolo, Kenya. The sketch (Fig. Id) is based on a photograph of Otieno’s material. Galeropsis Velen. Pileus not expanding at maturity but remaining as a subglobose, subovoid or subconic structure, narrowly open at the base. Gills radial but somewhat anastomosed. Stipe well developed, slender, hollow, without volva at base but often with a well-developed, sometimes marginate bulb. Southern Africa seems to be particularly rich in species of this genus. Two have been recorded for S. Africa: G. mitraeformis (Berk.) Heim, and G. literatus (Kalch.) Heim, both of them from the extreme south, another, G. besseyi (Pk.) Heim var. madagascariensis (Pat.) Heim from several localities in Madagascar. Finally G. paradoxa (Matt.) Heim has been recorded from Ethiopia. Heim (1950) expresses the tentative view that all are forms of a single variable species (and cf. the situation in Podaxis). Indeed distinctions between the species seem difficult to define. For the moment, however, we shall accept the generally held view that separate taxa are involved. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 11 G. aff. paradoxa (Matt.) Heim. (Fig. If-h) Pileus clavate-conic, to about 2x1 cm. pale tan. Gills adnate, frequently anastomosed; basidia 4-spored, waisted, about 25 x Sit., spores amygdaliform, 10-13 x6-7.5(Jt, pale amber, with small but clearly visible germ-pore. Stipe to 8 cm. long with a well-developed basal bulb. HABITAT: On the ground in pasture at 2-3000 m. alt. MATERIAL EXAMINED: KENYA: J. K. Dedan 1233, Forest Dept., Kikuyu, in grass, April, 1965 (K); nr. Limuru, in Kikuyu grass pastures, abundant (R.W.R’s notes). NOTES: The material which we have seen, fresh, dried and preserved in spirit, corresponds with that described by Mattirolo (1924) in habit, habitat (including altitude) and in spore size (the last as given by Heim (1950) for Mattirolo’s material) but the gills are apparently thinner and rather less frequently anastomosed, and the basidia are apparently all tetrasporous whereas Mattirolo noted both 2- and 4-spored basidia. The gills are adnate (Fig. Ig), that is to say that if the pileus were to be opened out like that of an agaric the proximal edges of the gills would be confluent with the tissue of the stipe. This is clearly seen in the spirit material and confirmed in transverse sections of the upper 1/3 of the cap. Contrary to the impression to be gained from study of the literature, adnate insertion seems to be usual in this genus. Kotlaba (personal communication) confirms that it is true for Galeropsis desertoruni Velen. The gills appear to be free in dried material of all species of the genus. Sectioning of the gills shows them to be of the inaequihymeniiferous type (Buller, 1922) with basidia maturing in succession. The basidium elongates considerably just before maturity, and develops four slightly curved sterigmata on which the spores are borne. The result is that just as in an agaric the spores are held clear of the hymenial surface until they become ripe. In an agaric they would then be forcibly cast off from the sterigmata and the basidium would then collapse. In this case, however, they are not discharged but remain attached to the sterigma whilst the basidium collapses and draws them down again on to the hymenium to which they firmly adhere. As a result of the repetition of this process with the successively maturing basidia the hymenium becomes covered with a thick layer of spores. Basidia can be seen in situ only in very thin sections. The cuticle of the pileus is of ordinary hyphae, not cellular as in the Bolbitiaceae, the family of agarics to which this genus is obviously closely related (Singer, 1962). PODAXACEAE This family shares all the characters of the Secotiaceae, except that the tramal plates break down before maturity and are replaced by capillitium. Thus the gleba is pulverulent as in Lycoperdaceae but the spores resemble secotioid spores. There is a single genus, Podaxis Desv. Morse’s (1933) contention that all the forms are referable to a single variable species has not yet been refuted. Podaxis pistillaris (L. ex Pers.) Fr. sensu Morse (Fig. la-c) Sporocarp to 20 cm. high, consisting of an ellipsoid to subcylindric or subconic head supported on a slender stipe. Peridium dirty white to pale fawn, thin, woody, scaly, dehiscing by breaking away from the point of attachment to the stipe, splitting vertically into a small number of rays which bend outward and upward, and finally falling away completely. Stipe tapering upwards, produced into a percurrent columella, bulbous at the base, concolorous with the cap, scaly, longitudinally furrowed, hollow. Gleba copious, ochraceous, blood colour or black, capillitium of spirally thickened, dark hyphae, the spirals sometimes uncoiling to produce ribbons; spores of the basic secotioid type, straw-coloured to dark mahogany, 8-18x7-14(i. HABITAT : On the ground in exposed, dry situations, or on the tops of termitaria. DISTRIBUTION : Widespread in tropics and subtropics. MATERIAL EXAMINED: KENYA: W. J. Dawson, coast nr. Mombasa, rec. 6.8.1914; T. D. Maitland 531, Mombasa, rec. 31.1.1921; P. R. O. Bally B 2185, Garissa, in open country, common, 4.2.1943 ; Bally B 3169, Nairobi-Magadi road, nr. Gill’s Gulch, 2,400 ft., 20.6.1944; Rayner 733, Nairobi-Magadi road, nr. Olorgesailie, sandy soil above termites’ nest, 1944; Bally 7778, idem, 23.4.1950; Watamu, nr. Gede (R. W. R’s notes). PHALLACEAE This family and the closely related Clathraceae are recognizable by their delicate, ephemeral, usually spongy receptacles which develop inside a globose “egg”, bursting from it at maturity, exposing the foetid, mucilaginous gleba to the attentions of insects. In the Phallaceae itself the mature fruit-body consists of the volva, or ruptured remains of the peridium, from which springs a hollow, chambered or spongy stipe holding aloft the gleba. Some Gasteromycetes from Eastern Africa The Scale in the figures represents 10 microns (0.001 m.m.) Fig. 1. Podaxaceae and Secotiaceae. a-c, Podaxis pistillaris: a, habit x i (Rayner 739, from water- colour by E. M. Rayner); b, v.s. upper part; c, spores (Rayner 739); d-e Montagnea arenaria: d, habit x 1; e, spores (from pictures by N. Otieno); f-h Galeropsis aff. paradoxa: f, habit X \\ g, v.s. upper part, x 1 ; h, v.s. hymenium (Dedan 1233). J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 13 In Mutinus the gleba is borne directly on the stipe near its apex, but in most other genera, e.g. Phallus, it is borne on a more or less campanulate cap which fits over the apex of the stipe. An auxil- liary structure, the indusium, is present in some species. It is a conical network suspended from near the apex of the stipe and hanging round it, under the cap. Variations in the nature of the glebiferous part of the receptacle are the main criteria on which the family is divided into genera and species. The structure of the egg of Phallus is shown in Fig. 2b. Mutinus Fr. Receptacle a hollow, fusiform stipe bearing the gleba directly on the upper part. A species (almost certainly M. argentinus Speg., see Dring, 1964) having a red, acute glebiferous part and a slender, paler stipe has been seen in Kenya on several occasions but no material has been preserved. A similar plant from Tanganyika (Suji Mission, Makanyu, 4,000 ft., Jan. 1945, coll. Bally) is preserved in Herb. EA, but paucity of the material precludes accurate determination. Phallus Pers. The receptacle, when expanded, consists of a hollow stipe with a volva at the base and a cam- panulate, glebiferous cap at the apex. The wall of the stipe may consist of a spongy mass or a more or less clearly defined double layer of chambers, or a single layer of large chambers. We shall therefore refer to the chambers as multi- sub-bi- and uniseriate. The chambers may have holes in their walls opening to the exterior of the stipe or into its hollow exterior, or they may be intercommunicating. The tissue of the stipe may be white or some hue of red or orange. The cap consists of a thin basal membrane thickened in places by surface ornament. The orna- ment may be rugulose, papillate, or tuberculate, or consist of a bold network of folds or ridges, in which case it is called reticulate. The cap is white or occasionafiy orange. Cap and stipe are usually perforate at their common apex, sometimes barely perceptibly so, sometimes widely. The perforation is surrounded by wide or narrow ring of rather solid, undifferen- tiated tissue apparently belonging neither to cap nor to stipe. This collar serves as an “egg-tooth”, rupturing the peridium when the fruit-body expands. In some species an indusium hangs down from near the apex of the stipe, under the cap. It is perforated to a greater or a lesser extent, forming a network when fully expanded. It is customary to segregate those species possessing an indusium under two other generic names: Dictyophora, in which the cap ornament is reticulate, and Clautriavia, in which it is usually described as rugulose. However, the indusium alone is inadequate as a criterion on which to base a genus. In P. impudicus, the common European species and type of the genus, an indusium is occasionally present, though other fruit-bodies growing from the same mycelium are without. Closer observation shows that normal fruit-bodies of this species possess the rudiment of an indusium, visible, in microscopic preparations, as a ring of tissue near the apex of the stipe. P. rubicundus (Bose) Fr., (Fig. 2d) Egg globose to obovoid, strongly rooting, often by a single strand; peridium white to pale brown, dehiscing apically. Receptacle to 15 cm. high, stipe hollow, wall spongy, c. 3 mm. thick, multiseriate, with chambers intercommunicating and often perforated to the exterior, flesh colour. Pileus cam- panulate, concolorous with stipe or darker, imperforate, surface rugulose, sometimes almost smooth. Gleba olivaceous, mucilaginous, foetid; spores ovoid-cylindrical 3.5-5 xl.5-2.5p smooth, tinted. HABITAT : on the ground in forest or in the open. DISTRIBUTION : Probably throughout tropics and subtropics. NOTES : The typical form, to which the above description exclusively applie, has a robust, spongy, flesh-coloured stipe. The cap is more or less campanulate and the apex is imperforate. In a previous paper (Dring, 1964) an attempt was made to indicate some of the variation encountered within this so-called species. It seems questionable that such a diversity of plants can profitably be grouped under the same name. As more material is examined a number of well marked types, sometimes of apparently limited geographical distribution, emerge. One of them is dealt with below, in the status of a variety. Orange, truncate forms with thin cap closely applied to the stipe are often separated, with a good deal of justification, as P. aurantiacus Mont. The description suggests that P. armeniacus Patouillard (1924), described from Madagascar, would be referable here, but no material of this has been seen. P. rubiamdus var. gmciUimus Drtag & Rayner, var, nov. (Fig. 2c) spores 3-4 x 2p. Hab. In turf, presumably ephemeral. DISTRIBUTION: S. and E. Africa. MATERIAL EXAMINED: KENYA: Bowker, Kitale, Nov. 1960 (EA); Bally 6377, Soy, alt. 6,000 ft., on a Kikuyu-grass lawn, 30.6.1948 (painting by C. Cripps, EA). NOTES : This variety differs from the type in the extremely slender, long stipe consisting of a single layer of chambers. So far as can be ascertained from the dried material and the coloured drawing at our disposal the cap is thinner than that of P. rubicundus, more narrowly campanulate and not pink but yellowish-brown. The cap is also perhaps less rugulose than that of P. rubicundus but not too much stress should be placed on this rather variable character. P. rubicundus var. gracillimus is similar to P. novae-hollandiae Cda, (=P. gracilis Lloyd, nom. invalid., = Ithyphallus aurantiacus var. gracilis E. Fisch., 1 = P. caleyi Berk.), which, however has a shorter, bii • - . , . . variety of P. novt Phallus caliendricus Dring & Rayner sp. nov. (PI. 1) Typus: Rayner 513, Africa, 15.11.1951 (K). Egg subglobose, dirty white, strongly rooting by a cord-like mycelial strand. Receptacle stipitate; stipe flesh colour, to 10x2 cm. broadly fusiform or columnar, apically attenuate, hollow, the walls about 3 mm. thick, spongy, multiseriate, with up to about 4 layers of more or less intercommunicating chambers. Cap campanulate, 2 cm. long by 1.8 cm. wide at the margin, dull, slightly orangy-red under the olivaceous gleba, surface thrown into very irregular, tom, thin folds, mostly about 1 mm. high but with nodular outgrowths which may attain 2 mm. in length, the whole giving the appearance, after removal of gleba, of a matted wig; margin white, about 1 mm. wide, slightly thickened, formed by the confluence of the folds of the cap; apex at first closed, perforate later; collar prominent, c. 5 mm. diam. Gleba dark olivaceous, mucilaginous, strongly foetid as in P. impudicus, spores 4.5 x2p, ovoid-cylindrical, hyaline, smooth. HABITAT: On the ground. MATERIAL EXAMINED: KENYA: Rayner 513, Hombe distr., above River Research Centre, slopes of Mt. Kenya, 6,500 ft., 15.11.1951 (Type, K, with watercolour); Bally 9722, Nairobi distr., Lukenya, below E. slope, 5,500 ft., 20.5.1954 (EA). NOTES: This species is distinguishable from P. rubicundus, which it resembles in its habit, by the cap ornament being like that of Itajahya, though less well developed. Indeed this species may be a link between that genus and Phallus. P. indusiatus Vent, ex Pers. (Fig. 2a) MATERIAL EXAMINED: SEYCHELLES: C. Jeffrey, s.n., Praslin, Valee de Mai, 1962 (coloured slide only, K); Jeffrey, s n , without data (K). UGANDA: Sir W. Johnston, Bt., no data, ? 1901 (K); C. B. UssHER 78, Mabira Forest, April 1908 (K); Lister, Ishanta R., Kigezi, 31.8.1960 (photo only, K). TANGANYIKA: Sm J. Kirk, Newala, Rovuma R. [S. Tanganyika], 1886 (picture only, K);K. Braun, 691 [Usambaras] 28.6.1905 (EA); Braun 1 554, Kiv, nr. Vuga [Usambaras] (EA); Braun 8624, Amani (E.A.). NOTES : This is the common tropical, white “Dictyophora” and is characterized by its well developed, widely spreading indusium and the small reticulations of the cap ornament (up to about 2 mm, diam.). Ussher 78 has particularly narrow reticulations, mostly about 0.5 mm. diam. The record from Val6e de Mai is of an example with a double stipe and common cap, indusium and volva. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 15 P. duplicatus Bose (Fig. 2b) Egg subglobose, to about 4 cm. diara., dirty white to brownish, strongly rooting, dehiscing apically. Receptacle consisting of stipe, indusium and cap; stipe white, to 15x3 cm. almost cylindrical, hollow, wall multiseriate, white; indusium short, often scarcely protruding below margin of cap, not wiclely spreading, the perforations small especially near the margin where they may be absent or hardly perceptible; cap campanulate, deeply reticulate, the primary reticulations up to about 5 rom. across in medium-sized fruit-bodies, collar narrow, often long-elliptical, apical perforation conspi- cuous, also elliptical. Gleba of the usual phalloid type, smell slightly offensive; spores smooth, elliptical 2.5-3.5x2p. HABITAT : On the ground, usually in woodland. DISTRIBUTION: N. America, E. and S. Africa, W. Europe. MATERIAL EXAMINED: KENYA: Rayner, Upper Kiambu, coffee estate, c. 6,200 ft., 1943 (K). NOTES: This is immediately distinguishable from P. indusiatus by the larger reticulations of the cap, the more massive stipe, and the shorter, less widely perforate and narrower indusium. Coker & Couch (1928) state that the indusium of United States examples is light rosy pink and Smith (1951) says “white to pinkish” but the specimen from Kenya had a white indusium. In the above collection, which is in spirit, the indusium scarcely protrudes below the cap and the perforations, even above, are poorly developed. From the illustrations in Bottomley (1948) and the single specimen in K (J. Medley Wood 667, in bush, Inanda, Natal, leg. W. Haygarth, rec. 27.12.1881, with picture) it would seem probable that this, not P. indusiata, is the correct name for the S. African examples. Records of this species in W. Europe are open to suspicion because of confusion between it and indusiate forms of P. impudicus. It is tentatively suggested that the main distinction between them lies in the structure and shape of the indusium, which is widely spreading, cobwebby, in the latter. General habit and the character of the reticulations of the cap are important additional distinctions. The difference between the two plants is magnificently demonstrated by Pilat (1958, Fig. 16, p. 74). P. hadrianii Vent, ex Pers. (= P. imperialis Schulzer) MATERIAL EXAMINED: SEYCHELLES: Jeffrey, s.n., Mahe Brillant, 7.10.1961 (with coloured slide, K). NOTES: This species differs from P. impudicus mainly in its less disagreeable smell, in the pink colour of its volva, volva gel and, often, of its stipe, in the more broadly based egg, and in the more nearly isodiametric reticulation of the cap. The cap is said to be less abruptly conical and the apical perforation consequently wider but this may be of little diagnostic value (Meulenhoff, 1936). It has been argued with some justification that it is merely a variety of P. impudicus. Jeffrey’s material conforms with the usual concept of this species, except that the stipe is uniformly pale pink whereas it is usually white or coloured only at the base. Its occurrence in the Seychelles is not so strange as would appear at first sight since many records of P. impudicus from Asia are probably referable to this species (Dring & Rayss, 1964). Pearson (1948) records P. hadrianii from S. Africa. CLATHRACEAE This group is closely related to the Phallaceae but differs in that the receptacle is more compli- cated and more variable. It may be sessile or stalked, and consist of a network or of variously united arms or diverging branches. The gleba is borne directly on the adaxial face of the networks, arms, etc., in immediate contrast to the situation in the Phallaceae, where the gleba is always borne exter- nally to the receptacle. The egg (Fig. 2g) differs from that of the phalloids in that the gelatinous “inner peridium” is not a continuous layer but is divided into segments by peridial sutures. These are membranes which connect the outer peridium with the receptacle, passing approximately radially throu^ the gelatinous layer. A peridial suture corresponds to each arm of the receptacle. The lines of fusion between the peridial suture and the peridium can be seen as slight folds running over the outer surface of the unopened egg. The genera of Clathraceae are interdistinguishable by the shape of the receptacle. Unfortunately rather few are known to us from this area. In Clathrus the receptacle is a hollow network, and aU other forms may be considered to be derived from this. Ileodictyon is a close relative of Clathrus in which the receptacle, instead of consisting of chambered tissue, is formed of a continuous, intestine-like tube (Reid & Dring, 1964). /. cibarius Tul. is recorded from SaUsbury (see Bottomley 1948, p. 528). Page 16 Some Gasteromycetes from Eastern Africa Loss of all but the vertical arms of the network has resulted in genera like Linderiella (Fig. 2e) and Blumenavia, both of which have E. African representatives. Another line of evolution has given rise to stipitate forms represented in the E. African flora by Simblum and Kalchbrennera (Fig. 2f, i). Clathrus Mich, ex Pers. Receptacle a hollow, more or less spherical lattice. The gleba may cover the whole inner surface of the receptacle or be restricted to the intersections of the arms of the network. The arms consist of a more or less regular arrangement of chambers though this structure may be obscured in the more massive, spongy species. Only one record is known to us from E. Africa (Maitland, without data, Uganda, photograph only, in Herb. K). The fruit-bodies (Fig. 2h) are white or pale-coloured and have rather characteristic flat meshes as in Mme. Goosens-Fontana’s illustration (in Dissing & Lange, 1963). Dissing and Lange have taken this to be C. baumii P. Henn., with good reason. Though the illustration does not indicate that the gleba is confined to nodular processes at the intersections of the arms as Hennings’s (1903) original description specifies, study of the dried material (BR) shows this to be so. This fungus must be very similar if not identical to that recorded from Jamaica by Dennis (1953) as C. cfr. preussii, from typical examples of which it differs in the flatter arms, more nearly quad- rilateral in section, lacking a fringe along the outer angles (Bring, 1964). Linderiella G. H. Cunn. (,=Linderia Cunn. non Lindera Thunb.) Receptacle of vertical, unbranched (exceptionally forked) columns united at the apex but free at the base. The gleba is borne directly on the inside of the apical part of the colunms. There is only one known species. L. columnata (Bose) Cunn. (= Laternea coliimnata (Bose) Lloyd) (Fig. 2e) Egg subglobose with longitudinal furrows corresponding to the peridial sutures and hence to the columns of the receptacle, to about 5 cm. diam., white to pale brownish, usually with a single thick rooting strand. Receptacle orange to bright red, at first urceolate to ovate, becoming obovate as the top part expands fully, of 2-4 thick columns which are gradually attenuated toward the apex where they are united, abruptly tapering below where they are free, up to about 2 cm. across at the widest part. Gleba borne in a single large mass initially spherical and pendant from the top of the receptacle, as the latter opens dequescing and forming a more or less even coat over the inner surface of the upper part of the columns. Gleba and spores as usual for the family. HABITAT: On the ground. DISTRIBUTION : The Americas, Africa south of the equator, Japan, New Zealand. MATERIAL EXAMINED: KENYA: E. R. Napier, s.n., Kiambu, 5,700 ft., Dec. 1931, coll. Mrs. Armstrong (K); Rayner 736, Scott Laboratories, Nairobi, 6,300 ft. (K); Mrs. P. H. Irwin 613, Peover, Mt. Elgon, 10.6.1962 (watercolours only, K); Mrs. W. Stevens, Nairobi, 30.4.1964 (EA). NOTES : This is very closely related to a section of the genus Clathrus and indeed the tendency of its columns to branch in exceptional cases, makes it very diflBcult to draw a water-tight distinction between the two genera. (See Coker & Couch (1928), PI. 1. for an excellent illustration of a Clathrus- like specimen. These authors, in fact use the original binomial Clathrus columnatus Bose.). Specimens from Japan with only two columns have been segregated as L. bicolumnata Lloyd. Specimens from the E. Rift valley have only two columns though one of the specimens from Mt. Elgon illustrated by Mrs. Irwin has three. Specimens of L. columnata have sometimes been confused with Anthurus but should be immediately recognizable in that the columns of the receptacle are united below into a short stipe in the latter. They are also initially joined at the apex of the receptacle but later become separated and flaring. Hennings (1904) recorded Anthurus sp. from the Usambaras but it was badly damaged. It seems probable that it was A. archeri which is known from S. Africa, though only from oakwoods near the S.W. coast, but it cannot be entirely excluded that it was in fact a Lysurus. Blumenavia Moller Receptacle of vertical columns, united at the apex, free at the base, the sides of each colurmi fringed by irregular, tom wings which bear the gleba. The special glebiferous organ is developed from the large adaxial chamber of the arm. The wall of the chamber is specially thickened except on the adaxial side where it is defective. At maturity it ruptures along this vertex and the thickened walls are free to fold sideways and forwards as irregular flaps carrying the greater part of the gleba with them. In Laternea Turpin, another genus of Clathraceae, special glebifers are also developed from modified receptacular chambers. Since they are similar in habit to Blumenavia a good case might be /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 17 made for uniting the two, though this will not be undertaken here. For an account of two species of Laternea and one of Blumenavia, see Dennis (1953). This series of forms with columns discontinuous at the base, including Linderiella and Blumenavia and ending in Laternea is one of the most complete and beautiful in the Clathraceae. It illustrates the loss of all but the vertical arms of the receptacle, and their reduction in number, the vertical displacement of the gleba, development of a special glebiferous organ (cf. the cap of Phallus) and reduction in size of the receptacle. But one species of Blumenavia is known from E. Africa. B. usambarensis P. Henn. Egg blotched with brown, subglobose, to 4 cm. diam. opening by a series of large irregular apical lobes. Receptacle white, long-ovoid, to about 8x3 cm. consisting of 3-5 vertical columns free below and joined at the apex, to 1 cm. thick at the base, attenuated apically, outer surface strongly trans- versely ridged, sub-triangular or quadrilateral in section, broadest side outermost. Glebiferous wings raggedly dentate, extending along the whole length of the inner angle of the side of the columns. Gleba of the usual type, probably restricted to the upper part of the receptacle. Spores of the usual type, 3-3.5 xl.5p. NOTES ; We have seen no E. African material of this very rare fungus and the description is made up from Hennings’s (1902) original and studies of a W. Indian collection in Herb. K (R. W. G. Dennis, s.n. Noronja, Trinidad, rain forest, 1,800 ft., 2.10.1949, with watercolour). This species would seem to differ from B. rhacodes Moll., the type species, in the position of the glebiferous wings, and in the much more fragile construction of the receptacle, which is white instead of red. Hennings does not mention that the gleba is restricted to the upper part of the receptacle but this seems likely. Laternea angolensis Welwitsch & Currey (1870) may be the same according to the original account. There is no type specimen in herb. BM. Simblum Klotzsch Receptacle consisting of a stipe surmounted by a clathrate, fertile network. The copious gleba s borne on all but the flattened outer face of the arms of the network. Morphologically this may be regarded as a stipitate Clathrus. Although basically globose, the head may be depressed onto the stipe so that it appears hemis- pherical. The colour of the receptacle may be white or some tint of yellow or red. The gleba may vary in the extent to which it covers the sides and exterior of the arms. The fertile network is usually abruptly differentiated from the stipe but occasionally there is a gradual transition. The fertile network varies considerably in the number of meshes (about 5 to about 25) and in their size and shape. Conard (1913) reports finding occasional arms of the network embedded deep in the gleba. Ahmad’s (1952) implication that the irregularity of the network in some specimens brings them very close to Lysurus is interesting and is confirmed by material in K (Varanasi, India, coll. K.B. Khare). The genus does share with Lysurus the tendency of the gleba to migrate towards the outside of the arms, a tendency which is even more marked in the next genus, Kalchbrennera. The number of species in the genus is a matter of dispute. The original one, S. periphragmoides was yellowish in colour and came from Mauritius. Specimens from the Old World are normally placed in this species and are usually some shade of yellow. Reddish (occasionally white) species usually come from the New World and are usually called S. sphaerocephalum (inappropriately, perhaps, since the “hemispherical” head is commoner in New World than in Old World forms. There are, however, notable exceptions to the hypothesis that colour depends on distribution. A yellow form from Texas has been described as S. texense (Atkinson & Long) Long, and Ahmad’s (1952) Pakistani collections are apparently red rather than yellow. We have seen no substantiated record of a Simblum from continental Africa, all supposed examples being referable to Kalchbrennera. S. periphragmoides Klotz. (Fig. 2i) MATERIAL EXAMINED: MAURITIUS: Mrs. A. Telfair, Bois Chery (K) (TYPE). ZANZIBAR: Mrs. H. Faulkner, Massazine, among grass in sandy soil, 19.11.1957 (K), and another collection, in spirit, without date (K). NOTES: The Zanzibar collections comprise rather small fruit-bodies with the “hemispherical” head (Fig. 2i). They are described as having been orange. The stipe wall consists of 2-3 layers of chambers thinning to one layer at the top and bottom. The chambers are arranged in broken vertical columns and many of them communicate with chambers above and below to form vertical tubes. The receptacular network consists of an apparently continuous tube whose walls are rather thicker Page 18 Some Gasteromycetes from Eastern Africa than those of the chambers of the stipe. The veil which lined the inside of the hollow stipe before expansion of the fruit-body persists as a short flaccid cylinder hanging down inside the stipe. The anatomical details correspond closely with those described by Long (1907) for S. texense. Kalchbrennera Berk. Receptacle like that of Simblum except that the fertile network is less well developed and bears simple or forked processes on its outer side at the point of intersection of the arms. The gleba is borne between these processes, that is on the outside of the network, not inside it as in every other clathroid. The position of the gleba is surprising. However, the ends of the processes are external to the gleba. A tendency for the gleba to migrate round the sides of the arms and to take up a position more or less exterior to the receptacle has already been noted for Simblum and Lysurus. Presumably this is an extension of that process. The genus contains a single species. K. corallocephala (Welw. & Curr.) Kalch. (^Simblum clathratum Lloyd) (Fig. 2f) Egg subglobose, becoming obovoid, to about 4 cm. diam. Network of receptacle bright red, hemis- pherical, to 3 cm. diam., meshes very variable in number, arms transversely rugulose, tubular, to 3 mm. diam., appendages bright red, branched or simple, projecting up to 2 cm. from the network, often smaller. Stipe cream-coloured, often tinged with pink above, cylindrical or attenuated down- wards, to 12x3 cm., hollow, walls sub-biseriate, the larger chambers internal, often intercommuni- cating to produce short vertical tubes. Gleba as normal for the Clathraceae, particularly foetid, spores normal. HABITAT: On the ground, in woodland or in the open. DISTRIBUTION : Apparently limited to Africa south of the Sahara. MATERIAL EXAMINED: KENYA: Rayner 738, Bamboo forest Central Prov., c. 7,000 ft., 1947; Bally b6349, Karen Estate, Ngong Distr., leaf mould in forest. May 1948 (EA); b9721. Spring Valley, Nairobi, coll. H. Stone, May 1954 (EA); b11575, Karen Estate, Nairobi Distr., coll. Mrs. J. Pedfield, 19.5.1957 (K, in spirit); L. D. & B. Verdcourt 2066, Muguga, nr. Nairobi, edge of cultivated field and natural woodland, 16.12.1959 (EA); L. D. Verdcourt h201/61, Muguga, nr. Nairobi, in grass, 23.5.1961 (EA). SCLERODERMATACEAE This family is characterized by the possession of a more or less ^obose, sessile or stipitate, usually epigeal fruit-body. The peridium consists of a single layer, being thick or thin, tough or fragile at maturity, continuous with the tramal plates. The gleba is divided into discrete knots of tissue separated by the anastomosing tramal plates. It is not labyrinthine as in the Phallaceae and Lycoperdaceae, nor can it be said to be divided into chambers since there is virtually no hymenium, the basidia and their supporting hyphae forming loosely woven islands of tissue without a central cavity. The basidia apparently become functionless before the spores are fully grown and nutrition of the spore continues through a placenta of hyaline cells which envelopes each spore. The exact nature of the placental cells seems to determine the characteristic ornament of the mature epispore. Mature spores are usually relatively large and strongly ornamented with prominent spines or a deep reticulum, or often a combination of both. The tramal plates are more or less persistent at maturity, dividing the gleba into distinct pockets, often called peridioles. The two best known genera are Scleroderma and Pisolitkus. The latter is not represented in the E. African material at our disposal, though its only widespread speciesP. arrhizus Pers. (=P. tinctorius (Mont.) Fisch.) may occur in the dryer areas. Its fruit-bodies are dark brown, very hard, stipitate, dehiscing by irregular erosion from the apex. The tramal plates are very prominent and split down the midSe so that each peridiole breaks away as a separate grain. The tramal and other tissue yields a yellowish dye. Pseudomycorrhizal association with Euca- lyptus and probably other trees has been demonstrated. Scleroderma Pers. Tramal plates breaking down at maturity resulting in a pulverulent gleba consisting of spores plus large quantities of debris derived from the trama, placentae, etc., and often obscuring the ornamentation of the epispore. The peridium just before it dries out at maturity is always fairly thick (1 mm. to 1 cm.), brittle or tough when dry. Dehiscence is by more or less regular stellate cracking of the peridium and J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Fig 2. Phallaceae and Clathraceae. a, Phallus indusiatus, habit X i and spores (from W. African material); b, P. duplicatus, habit and v.s. egg, x i (Rayner 737, from watercolour by E. M. Rayner); c, P. rubicimdus var. gracillimus, habit X J (from watercolour by C. Cripps); d, P. rubicundus, habit X i and spores (from W. African material) ; e, Linderiella columnata, habit x i (from watercolour by Mrs. Irwin); f, Kalchbreimera corallocephala, habit x \ and spores (from W. African material); g, Clathms sp. v.s. egg X \ (W. Indian material, from watercolour by R. W. G. Dennis); h, Clathrus sp., habit X i (from photo by T. D. Maitland); i, Simblum periphragmoides, habit X 1 (Faulkner). Page 20 Some Gasteromycetes from Eastern Africa recurving of the resulting lobes, or by irregular flaking away of the apex. The most important charac- ters for distinguishing between the species are thickness of the peridium, mode of dehiscence and spore-size and ornament. S. verrucosum Vaill. ex Pers. ssp. verrucosiim (Fig. 3c) Fruit-bodies depressed globose, to about 6 cm. diam., usually less, usually with a short stipe but sometimes practically sessile or with a very long stalk. Peridium when dry, less than 1 mm. thick, fragile, with small, brown, angular, firmly attached scales. Spores globose, strongly echinate 7-12p diam., not including the spines, which are up to 2p long, curved and acute. DISTRIBUTION: Cosmopolitan. MATERIAL EXAMINED: RHODESIA: Eyles Herb. 7222, Salisbury, Feb. 1932 (as S. nitidum Berk.) (K). NOTES: This subspecies must be relatively common in the area though we have seen only one gathering. Bottomley (1948, p. 538) records it for S. Rhodesia and Lloyd (L. 12, p. 1 (Vol. 2) 1906) for the Usambaras. The length of the stipe in some collections is extreme. These have often been described as separate species (see Dring, 1964). S. verrucosum ssp. bovista (Fr.) Sebek is similar to ssp. verrucosum except that specimens with very long stipes are not found, and the ornament on the spores is a coarse reticulum. Intermediates between the two subspecies exist in which the epispore is partly spiny, partly reticulate. Bottomley (1948, p. 540) records ssp. bovista from Rhodesia (as S. bovista). S. flavidum Ellis & Everhart (Fig. 3a, b) Fruit-bodies depressed globose to pyriform, to 6 cm. diam. dehiscing by apical lobes which become partly reflexed, usually with a well-developed stem-like base. Peridium buff, up to 5 mm. thick when fresh, usually more than 1 mm. when dry, smooth, or the upper part deeply cracked or areolate. Gleba usually with a yellowish tinge, powdery, often falling away completely leaving the empty star-shaped peridium. Tramal plates often rather persistent. Spores dark brown, echinate, 10-14|i. DISTRIBUTION : N. America, Australasia, E. and S. Africa. MATERIAL EXAMINED: KENYA: C. Leakey, s.n., Nairobi, Apr. 1964 (K, EA); UGANDA: Maitland 593, Kampala, Sept. 1922 (as S. geaster, K); TANGANYIKA: Gibson, s.n., Golongolo, nr. Lushoto, profuse under P/nMi radiata suggesting mycorrhizal association, 6.7.1959 (as S. bovista, K). NOTES : This is rather a variable species recognizable by its stellate dehiscence, spore and peridial characters and sterile base. It differs from S. geaster in the thinner peridium and in that the spines on the spores are not reticulately arranged. Authentic material in Herb. K (Ellis & Everhart, N. Amer. Fungi Exs., 2 ser., 1698) has spores with very thin sharp spines but other material shows a great deal of variation in this respect. In Leakey’s specimen, the habit of which is absolutely typical of the species, the ornament of the epispore is reduced to warts. Maitland 593 has more typical spores but the fruit-bodies are very depressed in shape and most of them lack a sterile base. The upper part of the peridium is deeply cracked after the manner of some N. American specimens (see Coker & Couch, 1928, pi. 88). Gibson’s specimen is immature and can therefore only be tentatively placed here. S. capense Lloyd (Fig. 3d) Sporocarp small, to almost 1.5 cm. diam., more or less subterranean, sessile, rooting by a mass of fine basal strands. Peridium smooth to minutely cracked, often finely wrinkled on drying, dull ochraceus, very thin and brittle when dry. Gleba reddish-brown; tramal plates yellow, not persistent; spores globose, (7.5)-9-14|i, strongly echinulate with warts grouped together, almost free from debris. DISTRIBUTION: Southern Africa. MATERIAL EXAMINED: KENYA: Thorold t36, Njoro, 7,000 ft., July 1932 (K). NOTES: The specimen corresponds with the description given by Lloyd and Bottomley (1945, p. 540-1) except that the material is caespitose. Neither Lloyd nor Bottomley mention that the spines of the epispore are grouped together in clumps. Bottomley suggests that the species may be the same as S. cepa but the peridium is far too thin for that species, and the colours are wrong. The reddish tinge of the gleba at least in the stage im- mediately before it becomes powdery, is striking. NIDULARIACEAE Members of this family are characterized by the small (never much exceeding 1 cm. diam.) cup- shaped peridium containing a gleba composed of seed-like peridioles, of which there are usually J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 21 more than one per fruit-body. They are held in place in the cup until maturity by mucilage or anchored to its wall by a cord, the funicle. Spores are large, pale, thick-walled, smooth and subspherical to ovoid. The peculiar habit of these fungi gives them the name “birds’-nest” fungi. They have been provi- sionally placed next to the Sclerodermataceae in this paper because of the similarity of organization in which the gleba is divided into discrete peridioles. In addition, a true hymenium is absent and the spores are nourished by placentae as in Scleroderma. The more advanced, funiculate genera are best known, both taxonomically and developmentally, but more detailed study of the simpler members of the group may lead to a better understanding of their relationship to the rest of the gasteromycetes. There is a closely related family, the Sphaerobolaceae, containing only one accepted species Sphaerobolus stellatus Tode ex Pers. The fruit-bodies are up to about 2 mm. diam., cup-shaped, and the single peridiole is discharged explosively by the inner layer of the peridium suddenly turning inside-out. S. stellatus is cosmopolitan, occurring on decayed wood, dung, etc., but it is easily over- looked, and apparently not yet recorded for E. Africa. Cyathus Haller ex Pers. Cup obconical with a compound wall to which the peridioles are attached by a complex funiculus. Periodioles dark-coloured, covered by a dark cortex which is sometimes overlain by a thin translucent tunica. They are dispersed by rain-drops falling asymmetrically into the cup, ejecting the peridiole. As seen in transverse section the cup wall consists of an outer layer of narrow, branched, densely woven hyphae orientated more or less parallel to the surface and bearing the tomentum, if any, on the outside, a middle layer of pseudoparenchymatous tissue, and an inner layer of loosely woven hyphae bounded internally by a cuticle. There has been much corrfusion about the anatomy of the peridiole, which is regrettable since its various features are used as taxonomic criteria. For the moment, following Brodie & Dennis (1954), the appearance of the walls of the peridiole as seen in relatively thick sections such as are usually obtained by cutting the soaked-up peridiole with a razor-blade has been used as a means of separating species. The terminology introduced by Lloyd (1906a) and accepted by Brodie & Dennis (1954) has been used, namely tunica for the outermost, hyaline layer, if any, and cortex for the dark-coloured layer or layers. This system is unsatisfactory, however, and stands in need of drastic revision. C. microsporus Tul. (Fig. 3g) Cups obconical or slightly flaring with very small mycelial emplacement at the base, 5-7 mm. high X up to 6 mm. wide at mouth, externally non-plicate, with a covering of shaggy to adpressed hairs, umber; inside rather greyer, smooth to very faintly plicate. Peridioles 1-2 mm. diam., dark brown, sometimes becoming silvery on drying, tunica thin, cortex single-layered, spores 6.5-5-9.5x4.5-6p. HABITAT : On rotten wood or soil associated with rotten wood. DISTRIBUTION : Florida, W. Indies and tropical S. America, Southern Africa. MATERIAL EXAMINED: KENYA: Gibson 779, Kimothu Nursery, Mt. Elgon, soil close to rotten timber. Sept. 1961. NOTES; The species is recognized by the combination of small spores and non-plicate cups. The published accounts differ in the size and shape given for the spores. Those of the above collections correspond with the figures given by Palmer (1961) for the type and other material. To judge by the various American collections in K, the outer surface of the cups usually has longer hairs than is the case with the African specimen. Mattirolo (1924) records this species from near Meru Mission, Kenya, coll. Rev. Padre G. Balbo. C. poeppigii Tul. (Fig. 3e) Cups dark umber, obconical, to 8 mm. high and 8 mm. wide, fluted to shaggy on exterior, both inner and outer surfaces deeply fluted but fluting sometimes obscured on the outside by the hairs, ridges about 0.5 mm. apart. Peridioles black, shiny, to 2 mm. diam., tunica absent, cortex two layered, dull brown in section; spores ellipsoid, usually about 20-30x30-40(1, but sometimes rather smaller in African material. HABITAT: On wood or soil in contact with wood. DISTRIBUTION : Widespread in tropics and subtropics. MATERIAL EXAMINED: KENYA: Bally 10446, Nairobi, 25.10.1955 (K); Verdcourt 1894, Mrima Hill Forest, 6.9.1957 (K, as C. limbatus); UGANDA: W. Small 143, 1915; R. Duemmer 2112, Kipayo, 4,000 ft., on log in forest, April 1915 (K). NOTES : In addition to the above examples Lloyd (1906b) received it from Braun, Amani, Tanganyika. Some Gasteromycetes from Eastern Africa This species greatly resembles C. limbatus in macroscopic features though the ridges are usually rather closer together. The spores of typical C. poeppigii are, however, twice the size of those of typical C. limbatus, and the cortex of the peridiole is composed of dull brown, not red brown hyphae as in the latter species. Small 143, noted above, lacks mature spores but is probably this species. C. limbatus Tul. (Fig. 3i) Cups obconic, to 10 mm. high and 7 mm. wide, usually with prominent basal emplacement; outer surface dark rust colour, shaggy, fluted, the flutes about 0.75-1 mm. wide; inner surface greyish, fluted. Peridioles dark and shiny, with two-layered cortex, spores usually about 10-15 x 15-20p. DISTRIBUTION : Widespread in the tropics and subtropics. MATERIAL EXAMINED: KENYA: R. M. Nattrass, Nairobi, 5,700 ft., on bamboo trash (K, ex Herb H. J. Brodie 1244); Rayner 734, Karura Forest, nr. Nairobi, 5,700 ft., 1942; Rayner 740, without data. UGANDA: A. French 29, Katonga Forest, Spring 1957 (K). TANGANYIKA: E. Milne-Redhead & P. Taylor 9746, Kitai, Songea District, 880 m., on bare earth, probably associated with rotten wood, 17.4.1956 (K). ZAMBIA: Angus m68, Kawambwa, N. Prov., dead branch on forest floor, 30.10.1952 (K). MAURITIUS: Ayres, Pouce Rouge, 1861 (K). NOTES : This species closely resembles the previous one except in spore-size. The cups are usually rather larger and more elongated with wider fluting. The usual description of the cortex of the peridiole as being “two-layered” needs some qualifi- cation. The outer layer of the cortex is very thin, often consisting of a single layer of red-brown hyphae. In all but the very thinnest sections this layer usually becomes partly detached and tilted so as to appear thicker; in the very thin sections it usually becomes detached altogether. French 29 has imusually narrow spores 14-6x6-7p; Rayner 734 very large ones, up to 30p long. C. triplex Lloyd (Fig. 3f) Cups obconical, not fluted, externally coarsely felted, pale umber; internally light greyish-brown. Peridioles to 2 mm. diam., in dried specimens silvery on the upper surface where the tunica persists, dull grey below where it does not; cortex two-layered; spores narrowly to broadly ellipsoid, 13-20p in major diam., smooth, often very thick-walled. HABITAT: On dead wood. DISTRIBUTION : Central or S. America, Ceylon, Mauritius, W. Africa. MATERIAL EXAMINED: MAURITIUS: Suite des grandes pluies de mars (K, ex Herb. Berk, and Herb. Hooker, as C. intermedius). NOTES: C. triplex differs from C. pallidus Berk. & Curt, in its darker cups with margin hardly or not reflexed, coarser tomentum, double peridiolar cortex and larger spores. C. pallidus is recorded from S. Africa (Bottomley, 1948 and a specimen in K, N. J. G. Smith, Grahamstown, Jan. 1931). C. rudis Pat. (Fig. 3h) Cups campanulate, to 10 mm. high, 8 mm. diam., narrowly fluted, exterior dark reddish brown, strongly tomentose-scaly, the scales obscuring the flutes, interior pale silvery grey, the fluting empha- sized by dark-brown striae on the upper half. Peridioles brown-black, spores ellipsoid 10-12 x5.7p. HABITAT: On manure. DISTRIBUTION: Madagascar, ? New Zealand. MATERIAL EXAMINED: MADAGASCAR: CoUector and locality unknown, Dec. 1932 (K, ex Herb. H. J. Brodie 1118). NOTES : The part of the collection in K consists of only one fruit-body, without peridioles. Brodie has annotated it: “seems to fit very well, few collections known”. The description of the spores is taken from Patouillard’s original description and from New Zealand material in K which Brodie has annoted: “This is what New Zealand mycologists send me as C. novae-zealandiae. To me it is close to C. rudis". It is indeed strikingly similar to the Malagash material, the nature of the tomentum being a most distinctive character. LYCOPERDACEAE Sporocarps more or less spherical, sessile or carried on a pseudostem, that is a basal cylinder or inverted cone of spongy tissue whose elements are not orientated in the direction of the long axis of the fruit-body as is the case in the tissue of a true stipe. Peridium divided into two distinct layers, the exo- and endoperidia. Gleba pulverulent at maturity, consisting of globose or ellipsoid, usually ornamented spores, and, almost always of well-developed capilUtium. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Fig 3. Sclerodermataceae and Nidulariaceae. a-b, Scleroderma flavidum: a, habit x ^ and spore (Leakey); b, spores (Mattland 593); c, S. verrucosum ssp. verrucosum, habit x i and spores, one with debris (Eyles 1222); d, S. capensis habit and v.s., x spores and nurse cells (Thorold); e, Cyathus poeppigii, habit x 2 and spore (Bally 1046); f, C. triplex, habit X 2 (Mauritius); g, C. microsporus, and spores (Gibson 779); h, C. rudis, habit X 2 (Madagascar); i, C. limbatus, habit X 2 and spore (Rayner 734). Some Gastewmycetes from Eastern Africa This family, as defined here, includes the familiar and successful puffballs {Lycoperdon and related genera), and the earth-stars (Geastrum). A number of important organs are characteristic of this family, though none of them is absolutely diagnostic. Perhaps the most important of these is the capillitium. In the narrow sense in which it is used here, the word denotes the thread-like, thick-walled, tinted to dark-coloured mycelium which permeates the gleba. It is admitted that capillitum is not strictly limited to the Lycoperdaceae, and that there are a few species of Lycoperdaceae which lack true capillitium. Nevertheless, the presence of capillitium is the hallmark of the family. Tramal remains which have not developed into true capillitium may also be present in the mature gleba. They are usually hyaline, collapsed, septate hyphae, often clinging together in bundles, or they may be less like ordinary tramal cells and more like immature capillitial hyphae. Such structures comprise the paracapillitium (Kreisel, 1962). In members of the Lycoperdaceae and some other families capillitium is replaced by abundant paracapillitium. In some other cases paracapillitium is mixed with true capillitium in the same gleba. The significance of the various forms of capillitium and the various combinations of true capillitium and paracapillitium both taxonomically and as agents in the process of spore discharge is largely unknown and well worthy of investigation. Knowledge of peridia is equally limited. Little is known about the relationship between the peridia in various families. Certainly wide homologies must not be drawn between the layers of the peridium of different famiUes. The exoperidium of Lycoperdon should not, for example, be assumed to be the homologue of the outer peridium of Tulostoma. In this paper the terms exoperidium and endoperidium have been used only for the peridial layers of Lycoperdaceae. Where a double peridial layer is present in other groups the layers are called inner and outer peridia. The pseudostem is an additional organ frequently found in the puffballs. It consists of glebal tissue which has become sterile and modified to fulfill a supporting function. Its glebal origin is reflected in its usually chambered structure. The absence of the true stipe in the Lycoperdaceae (and in the Phallaceae and Clathraceae) is interesting as it strongly suggests an origin from a sessile, probably hypogeal ancestor. Lycoperdon Pers. Sporocarp sessile or with pseudostem, dehiscing by an apical pore. Capillitium simple or branched, often arising in a central tuft or pseudocolumella and in any case attached to the inner surface of the endoperidium, sometimes undeveloped (paracapillitium). Exoperidium of simple or compound spines or of branny or scurfy particles, often caducous. L. pratense Pers. emend. Quelet (=L. hiemale Vitt., L. depressum Bon.) (Fig. 4b) Exoperidial spines compound, pale. Endoperidium ochraceous to buff or grey, areolate after fall of spines. Subgleba chambered, separated from gleba by a diaphragm confluent with the endo- peridium. Gleba without pseudocolumella, Capillitium almost absent except at periphery of gleba where it is represented by a few straight, occasionally branched and occasionally septate hyphae without pits; paracapillitium of septate, hyaline hyphae; spores 3.5-5.5p diam., almost smooth to finely spiny. HABITAT: On the ground. MATERIAL EXAMINED: KENYA: Verdcourt 3951, 12 miles S. of Mombasa, on path, coastal bush, 19.1.1964 (K). UGANDA: Calder 28, Makerere Hill, exposed mown sward, common after rain, 1964 (K). MAURITIUS: Telfair, Pouce Rouge, Jan. 1863 (K, as L. gemmatum var.). NOTES: This species is recognizable by its diaphragm and capillitium. The latter seems to vary, according to the collection, from almost “mature” capillitium to completely collapsed, almost amorphous paracapillitium; most specimens fall between the two extremes, in fact most show considerable range within the same fruit-body. It is possible that African material will be separable into a number of taxa on capillitial and other characters when more collections are known. Large, caespitose, pyriform to obconical, dark-olivaceous specimens with purpUsh gleba are known from S. Africa, the Congo and Ghana under the name L. djurense sensu Bottomley and are probably specifically distinct. Those species of Lycoperdon with a diaphragm are sometimes placed in a separate genus, Vascellum Smarda. Thus, this species is often called V. pratense (Pers.) Kreisel. On its own the diaphragm is perhaps insufficient as a generic distinction, and other characters, particularly the proportion of capillitium to paracapillitium, are variable. (See also Dissing & Lange, 1962.) L. perlatum Pers. (Fig. 4d) Sporocarp depressed to subturbinate. Pseudostem present. Exoperidium of closely grouped, usually simple warts, often so’arranged that one large wart is surrounded by a ring of smaller ones, the J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 25 Fig. 4. Lycoperdon. a, L. pusillum habit and v.s. X i, spores (on right as seen in air) and capillitium (Buller); b, L. pratense, habit and v.s. X J, details of exoperidium X 5, spores and Para capillitium (Verdcourt 395); c, L. citrinum, habit X \ (Rayner 730, after watercolour by E. M. Rayner), spores and capillitium (Maitland 39); d, L. perlatum, habit and v.s. X exoperidial details x 5, spores and capillitium (Irwin 591); e, L. asperum sensu Dissing & Lange, habit and v.s. x exoperidial details X 5, spores and capilhtium (Irwin 557); f-g, L. fuUginium sensu Dring, f, habit and v.s. x i, spores (Ingold), g, habit x \ (Rayner 720). Page 26 Some Gasteromycetes from Eastern Africa large central wart falling away to leave a marked areolus. Endoperidium buff or umber, dehiscing by an apical, raised, tom mouth. Subgleba grey-brown, of large well-marked chambers. Gleba olivaceous, with large pseudocolumella; capiUitium of long threads, smooth, aseptate, sparingly dichotomous, with a few large and small pits in the walls ; spores globose, 3.5-4.5p, minutely warted. HABITAT : On leaf litter, particularly in conifer plantations. DISTRIBUTION: Widespread. MATERIAL EXAMINED: KENYA: G. F. Scott Elliot, Ruwenzori Exp. 198, Yeria, Ruwenzori, 8-9,000 ft.. May 1894 (BM); Irwin 521, Endebess, Mt. Elgon (K); Irwin 591 and 623, E. Mt. Elgon, cypress plantation, 1.6.1963 and 28.8.1963 (K). NOTES : This is the common European species often called L. gemmatum. The material from Mt. Elgon is very close to L. perlatum var. perlatum, with spines near the top of the fruit-body distinctly divided into two sizes. In all the specimens the large spines have fallen, leaving a distinct areolus surrounded by small, simple, black spines. The spines on the lower part of the fruit-body, including the sterile base, are all small. The sterile base is well-developed and cylindrical in all the fruit-bodies, and the subgleba tan-coloured, large-chambered. L. pusillum Batsch ex Schum. (? aggregate sp.) (Fig. 4a) Sporocarp globose, up to about 2 cm. diam., sterile base absent, strong rooting strand or strands present. Exoperidium of fugaceous, mealy scales; endoperidium membranous, smooth, ochraceous, becoming brown with age, stoma plane, irregular. Gleba yellowish, becoming brown, pseudocolu- mella absent; capiUitial threads honey-coloured to brown, aseptate or occasionally septate, freely branched, up to about 4p diam., irregularly shaped, fragmenting, pits of varying diameter frequent in the walls; spores globose 3.5-5p, very minutely verrucose, with the remains of a pedicel. HABITAT : On the ground in exposed situations. DISTRIBUTION: Cosmopolitan. MATERIAL EXAMINED: UGANDA: Maitland 471, Entebbe, Bot. Garden, dunged flower bed, March 1919 (K). MALAWI: E. J. Buller, Mlanje, Ruo Estate, under tea bushes, Feb.-March 1927 (K). RHODESIA: Eyles Herb. 7220, Salisbury, March 1932 (K). NOTES: This widespread and common fungus needs little additional description. The spores of the E. African specimens appear smooth except when viewed in air-bubbles under oil-immersion. There is a tendency for the spores to be sub-globose, slightly elongated along the axis of the pedicel. In Maitland 471 occasional spores have a longer pedicel than usual, up to twice as long as the spore. L. citrinum Berk. & Br. (Fig. 4c) Fruit-body globose to depressed globose, to 4 cm. diam., without sterile base or almost so, but with stout, branched, white rooting strand. Exoperidium of tiny warts, variable in colour but usually darker than the endoperidium, persistent. Endoperidium bright ochraceous, becoming umber with a metallic sheen. Gleba bright amber, becoming fulvous, without pseudocolumella; spores golden- yellow, globose, 3.5-5.5p diam., smooth to just perceptibly (mounted in water, under oil-immersion) punctate, pedicel up to 20|i long; capiUitium golden yeUow, threads up to about 6p diam., undulating, fragile when young, becoming more elastic in age, walls with minute pits difficult to see even under oil-immersion and often practicaUy disappearing with age. Subgleba absent. HABITAT : On the ground, very common on Kikuyu-grass lawns in Kenya. DISTRIBUTION : Ceylon, southern tropical Africa. MATERIAL EXAMINED: KENYA: Irwin 557, E. Mt. Elgon, 8,000 ft., in Cupressus macrocarpa plantation (K). UGANDA: Maitland 39, Victoria Nyanza region [1914] (K, as L. asperum (Lev.) Speg.). S. RHODESIA: Eyles Herb. 7219, 7221, Salisbury, Feb.-March 1932, Dec. 1931 (K). NOTES: This species is recognizable by the distinctive colour, by the configuration of the peridium and by the pedicellate spores. The Kenya material (Fig. 4c) is placed here only tentatively. It consists of a single fruit-body which differs from aU the others in lacking a persistent rooting strand, its solitary habit, larger size, almost black exoperidial scales, ovoid spores, brown capiUitium and compact subgleba. It corresponds almost exactly with the description and illustration given by Dissing & Lange (1962) for Congolese material attributed to L. asperum (L6v.) de Toni except that the spores are more narrowly ellipsoid, 4-5x3-3.5ji. Rreisel (1964), however, says “Lycoperdon citrinum Berk & Br. 1873 = ? L. asperum sensu Dissing & Lange, 1962”. We are not convinced that the Kenya material, at any rate, is either. Rayner 732, from Kenya (Shield’s garden, Kiambu, on Paspalum notatum lawn, 5,600 ft., 3.12.1944) of which there is only a description, strongly suggests L. capense Cke. & Mass. The specimens must have resembled large (up to 8 cm. diam.) versions of L. citrinum in colour and habit. There is no type specimen of L. capense in Herb. K but it would seem from the description to be quite recognizable if found again. If it were rediscovered it would need renaming, Cooke and Massee’s binominal being a later homonym of L. capense Fr. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) An admirable, though brief, treatment of the puff-balls of the pusiUum and citrinum group is given by Kreisel (1964). In this paper he places all those Lycoperdon in which the subgleba is absent or compact in the genus Bovista, reserving Lycoperdon for the more typical species with chambered subgleba. We have not, for the moment, accepted his line of demarkation between the two genera (see below). L. fuligineum sensu Dring (1964) (Fig. 4f, g) Sporocarp depressed globose to turbinate, occasionally umbonate, to 2.5 cm. diam., solitary to crowded, on rotten wood. Exoperidium fuscous above, usually lighter below, minutely granular, adherent to the endoperidium. Endoperidium pale chestnut, fragile, dehiscing by a tom apical stoma. Gleba amber; capillitium of hyaline, septate, often collapsed hyphae up to about 5p diam., often grouped into fascicles (paracapillitium of Kreisel), spores globose, moderately to strongly echin- nulate, 3-4 diam. Subgleba scanty to well developed, tan, minutely but definitely chambered, cham- bers to c. 0.25 mm. diam. HABITAT : Rotten wood. DISTRIBUTION : Ghana, Sao Tome, Kenya, Uganda. MATERIAL EXAMINED: KENYA: B. F. Scott Elliot 145, Ruwenzori Exp. 1893-4, Ruwenzori 8,000 ft.. May, in forest (BM); Rayner 729, Karura Forest, nr. Ruraka River 5,700 ft. ; 2.12.1945 (K). UGANDA: C. T. Ingold, Upanda Forest, nr. Kampala, Spring 1963 (K). NOTES : This is really a separate, probably undescribed, species in the L. fuligineum group. It differs from the type of that species in the presence of a chambered instead of a compact subgleba and in the more granular exoperidium. The group as a whole forms a natural segregate from Lycoperdon which will be dealt with in another paper (Kreisel & Dring, 1967). It is characterized by lignicolous habit, by the dark colour and small size of the fruit-bodies, and by lack of true capillitium. Bovista Dill, ex Pers. Resembling the sessile Lycoperdons in a general way but with capillitium divided into separate elements with an obvious main axis, dichotomous at the ends and laterally branched, not having permanent connexion with the endoperidium nor confluent into a pseudocolumella. Only one E. African record of Bovista is known to us. B. membranacea H. Lowag was originally described from Kilimanjaro. The type seems to have been lost but from the description it is clear that B. membranacea is a later synonym of B. fusca Lev. Yet another synonym of B. fusca is B. umbrina Bottomley, described from Natal. We are indebted to Dr. H. Kreisel for the above information. The type material of B. umbrina in K (Rev. N. Roberts 5659, Woodbush, Jan. 1913) consists of a single fruit-body, subglobose, 2 cm. diam., with fulvous, granular, exoperidium worn away at the top to expose the greyish-sepia, metallic-looking endoperidium. Spores are dark brown, smooth, 5 x4p, with a pedicel up to 15p long; capillitium hyphae dark brown, to 13p diam. R.W.R. has observed a similar Bovista in the grounds of the French Mission, nr. Nairobi. Calvatia Fr. Sporocarp medium to large, with a strongly rooting sterile base. Exoperidium usually thin, occasionally of two layers, the outer one thick; endoperidium thin, the apical part breaking away in irreg^ar patches to expose the gleba. Gleba copious; paracapillitium seemingly absent; subgleba ranging from small and dense to massive with very large chambers, often poorly differen- tiated from the gleba, even when the latter is mature at the top. C. longicauda (Henn.) Lloyd (=C. agaricoides Dissing & Lange) (Fig. 5b) Fruit-body extravagantly top-shaped to agaricoid, to about 8x8 cm., the flat- to concave-topped head abruptly differentiated from the stipe. Stipe sub-cylindrical, slightly tapering towards the base, 4-7 X li-2 cm. Exoperidium minutely velvety, the tomentum consisting of short columns of bladder- like cells, about 20p diam., umber, adherent to endoperidium. Endoperidium very thin on drying, fragmenting and falling away. Gleba cinnamon buff to hazel, occasionally with a slight purplish tinge; capillitium of pale yellowish, slightly sinuous, much branched (at angles from about 60° to 120°) threads, often swollen at the frequent septa, about 3-4p diam., pits absent; spores subglobose to ovoid or ellipsoid, 3.5-4.5 x4-4.5p, minutely spiny, pale yellowish, with the persistent stump of a pedicel, often with a conspicuous oil-drop. Subgleba brown, chambers i-1 mm. diam. HABITAT: In forest. DISTRIBUTION: Cameroons, Congo, Uganda, Madagascar. Page 28 Some Gasteromycetes from Eastern Africa MATERIAL EXAMINED: UGANDA: Maitland 259, Entebbe, Botanic Gardens, on dry soil, April 1918; Maitland 260a, Entebbe, Botanic Gardens, on clay soil, April 1918; 260b, Busi, on leaf-mould, March 1919; 260c, Entebbe, in forests [no date]. NOTES: This species usually recognizable by its habit; less markedly agaricoid specimens are distinguishable from C. gardneri by the obviously chambered subgleba, umber rather than ochraceous peridium and irregularly branched capillitial hyphae which lack the large holes so characteristic of the capillitium of C. gardneri and its relatives. The gleba is much more persistent in this species than in the C. gardneri group. This is associated with the fact that the hyphae do not fragment nearly so readily, as is quickly noticed when one attempts to tease out the capillitum of both species in the course of making microscopic preparations. C. subtomentosa Dissing & Lange (Fig. 6c) Fruit-body shortly pyriform, with prominent sterile base, rooting by means of white strands. Exo- peridium velvety, with slight tendency to form tufts, isabeUine, splitting (with the endoperidium) into large scales and falling away from the upper part of the fruit-body. Gleba amber to fulvous, spores globose, 3.5^.5p diam., sparsely echinate, light yellow; capiUitium hyphae 3-6p diam., much branched, septate, fragmenting, though not necessarily at the septa, pits frequent, ranging from minute Qust visible under oil immersion) to 4p diam. Subgleba minutely and obscurely cham- bered, lighter in colour than the gleba. HABITAT: On the ground in woodland. DISTRIBUTION: The Congo, Kenya. MATERIAL EXAMINED: B^NYA: Irwin 489, Endebess, Mt. Elgon, Cupressus macrocarpa plantation. Sept. 1962. NOTES: The single fruit-body of this collection is placed here with slight hesitation because of its rather luxuriant exoperidium which cracks into large scales (c. 1 cm2). This may be partly a reflec- tion of the conditions under which it grew. In all other respects it agrees with the type specimen. C. gardneri (Berk.) Lloyd {—Lycoperdon gautieroides Berk. & Br.) (Fig. 5a) Sporocarp turbinate, to 10x10 cm., half the height being occupied by the sterile base. Exoperidium chestnut, minutely velvety below, tinted above, sometimes breaking into small scales. Endoperidium ochraceous, becoming rust colour above, paler below, brittle. Gleba ochraceous, fragile; capiUitium of branched, septate, fairly straight, honey-coloured threads, to 7p diam., usually c. 3p, with occa- sional large pits in the walls; spores globose to subglobose, 3.5-5p diam., slightly rough to minutely and sparsely spiny, concolorous with capUlitium. Subgleba at first ochraceous, becoming brown, minutely and obscurely chambered. HABITAT: On the ground. DISTRIBUTION : Ceylon, Mauritius, E. Indies, tropical Africa. MATERIAL EXAMINED: UGANDA: Maitland 259, Entebbe, Botanical Gardens, on clay soil, AprU 1918. NOTES: This was originally a mixed coUection, some of the fruit-bodies being referable to C. longicauda or they may have strayed into the packet from Maitland 260a. In a previous paper (Dring, 1964) the spores of this species were described as being 5.5-7.5p diam. This was an error; they are in fact much smaUer 3.5-4.5p in African material and 3.5-4.5 x4-5p in the type specimen from Ceylon. The structure of the tufts of hyphae of the exoperidium is interesting. Microscopically they have a central core of elongated ceUs up to 5 x 20p arranged in branched columns. This core is surrounded by rather shorter columns of much stouter cells about 15x20ji. Unfortunately it was not possible to be completely satisfied about the microscopic structure of the exoperidium of the type material. Columns of both narrow and wide cells are present but whether they have the same arrangement in the tuft was not determined. However, in another specimen from Ceylon (Fetch 416, Peradeniya, May 1917 (K), the characteristic structure of the tuft was seen. This species is very similar to C. rubroflava (Cragin) Lloyd. The differences are that the latter lacks the large sterile base, is perhaps generally more brightly coloured, stains yellow on bruising and has a shorter exoperidial tomentum. The spores are apparently always globose, whereas those of C. gardneri tend to be broadly ellipsoid, particularly in Asiatic specimens. The capillitial hyphae of both species have the characteristic large holes in the walls, though in C. rubroflava they are much more variable in number and in size, resembling those of C. subtomentosa. In all three species of this group (C. subtomentosa, C. gardneri and C. rubroflava) the capillitial hyphae tend to fragment at the weak places caused by these holes, rather than at the septa as occurs in the lilac-spored Calvatias (see below). The very persistent gleba of C. longicauda consists of hyphae without pits and which seldom break at the septa. J. E. Afr. nat. Hist. Soc. Vo!. XXVI No. 2 {114) Fig. 5. Calvatia, p.p. a, C. gardneri, habit and v.s. X spores and capillitium (Maitland 259); b, C. longicauda, habit and v.s. X i, spores and capillitium (Maitland 260). Page 30 Some Gasteromycetes from Eastern Africa Fig. 6. Calvatia p.p. a, C. cyathiformis ssp. fragilis, habit and v.s. X spores and capillitium (Mait- land 48); b, C. c. ssp. cyathiformis, habit and v.s. X i (Rayner 728, from watercolour by E. M. Rayner) spores and capillitium; c, C. subtomentosa, habit and v.s. x spores and capillitium (Irwin 489). /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 31 L. gautieroides Berk. & Br. is a synonym of C. gardneri. The type, from Ceylon, and many other collections in Herb. K, consists of small examples of C. gardneri. The type and Fetch 3241, Peradeniya, Oct. 1910 (K) appear to have been growing on or associated with rotten wood, not- withstanding that the original description specifies burnt earth as the substrate. It is evident from variant spellings in Berkely’s manuscripts and the faint preposterousness of “gautieroides” that this is an error. The probable intention was “guatterioides” after the genus Guatteria (Annonaceae). Unfortimately the fungus does not resemble the fruit of Guatteria in the least though it might easily be taken for the fruit of an Annona. C. cyathiformis (Bose) Morgan Sporocarp subglobose to pyriform, to 15 cm. diam., tapering abruptly to a crenulate sterile base. Exoperidium thin, continuous, smooth or floccose, whitish at first, later vinaceous buff, young specimens blackening on bruising. Endoperidium brown at maturity, thin, fragile, often becoming areolate, dehiscing by irregular cracking, often following the areolar margins. Gleba greyish purple; capillitium threads branched, septate, fragmenting, about the same diameter as the spores, with very frequent, minute pits in the walls; spores globose, 4-6p diam., violaceous-brown, shortly pedicellate, moderately to strongly echinulate. C. cyathiformis ssp. cyathiformis (=C. lilacina auett.) (Fig. 6b) Sporocarp large, sterile base large, usually marked-off from the fertile part by a constriction. Sub- gleba of moderately large chambers c. 0.5-1 mm. diam., rather ill-defined, spores usually strongly echinulate. HABITAT : In grassy places where rainfall is high. DISTRIBUTION ; Widespread in the tropics and subtropics, N. America to Alberta, Ontario and Manitoba. MATERIAL EXAMINED: UGANDA: C. B. Ussher, Mabira forest, June 1908; Maitland 261, Entebbe, Botanic Garden, on lawn, April 1918; and in front of Land OflBce, in pasture, Feb. 1919. KENYA: Rayner 782, Kikuyu-grass, Vet. Labs., Kabete, 7,000 ft., 4.11.1946. NOTES: This is recognizable by its large size, large sterile base with chambered subgleba and by the ornamentation of the spores. For a discussion of the choice of the epithet "‘cyathiformis” see Dring, 1964. C. cyathiformis ssp. fragilis (Vitt.) Dring (=C. fragilis Vitt. = C. lilacina (Mont. & Berk.) Henn.) (Fig. 6a) Sporocarp medium sized to about 6 ram. diam., sterile base small, subgleba compact to minutely and obscurely chambered, occasionally with a few larger chambers near the base. Spores usually with small warts. HABITAT : On the ground in open spaces. DISTRIBUTION: Widespread in tropics and subtropics, N. America to Ontario, Central Europe. MATERIAL EXAMINED: UGANDA: Maitland 48, Victoria Nyanza Region (K); Maitland 38, Mabira Forest, 1915 (K). NOTES: Most of the African material (Dring, 1964), including Maitland 48, corresponds well with Vittadini’s specimens of Lycoperdon fragile in K, except that the sterile bases tend to be larger in African collections. Maitland 38, however, has the long-spined spores more characteristic of ssp. cyathiformis. Mycenastrum Desv. Peridium woody, hygroscopic, typically dehiscing in an irre^larly stellate manner. Capillitium of branched and spiny hyphae which are not attached to the peridium nor converging into a centra pseudocolumella. The genus contains but one accepted species. M. corium (Guersent) Desv. (Fig. 7b, c) Fruit-body subglobose, medium-sized. Exoperidium single layered, thin and evanescent. Spores globose, 9-1 3p, with irregularly reticulate ornament. HABITAT : On the ground in dry places. DISTRIBUTION: Widespread. MATERIAL EXAMINED: KENYA: Rayner, Limuru, c. 7,000 ft., bare ground in Acacia planta- tion; Mrs. Joy Bally, Isiolo. TANGANYIKA: H. E. Dingle, Lake Manyara National Park, very dry ground under Acacia, Sept. 1962. UGANDA: Wilson 1333, Moroto, old kraal site. Some Gasteromycetes from Eastern Africa Langemiannia Rostk. (=Lanopila Fr.) Sporocarps medium to large, globose or depressed globose, sterile base absent or vestigial, weakly rooting, usually becoming detached at maturity. Exoperidium thin, smooth to floccose, made up of a single layer. Endoperidium thin, fragile, falling away completely at maturity. Gleba copious; capillitium persisting as a free, more or less naked spongy mass after abscission of the peridium, consisting of branched, interwoven threads, spores globose, smooth to verrucose. The genus has the same characters as Calvatia except for the lack of a significant sterile base, the habit of becoming detached from the ground, and the completely deciduous peridium. The capillitial hyphae do not fragment, unlike those of most Calvatia spp. For a discussion of the synonymy used here see Dring (1964). A single species is recorded for E. Africa. L. wahlbergii (Fr.) Dring (=Lanopila wahlbergii Fr.) (Fig. 7a) Sporocarps to 15 cm. diam., globose to depressed globose, sterile base absent, weakly rooting, Peridium dirty white at first, becoming pinkish rust colour to hazel. Capillitium of septate hyphae, without pits, 2-5p diam., pale brown, spores globose 5-7.5p diam., brown, closely and strongly warted. HABITAT : On the ground, free at maturity. DISTRIBUTION : Africa south of the Sahara, tropical Asia. MATERIAL EXAMINED: KENYA: J. McDonald, Rangai Forest, 6,000 ft., on ground in forest nursery, 22.11.1922 (K); Mrs. H. E. Brown, Rakrup Forest, on ground, common near elephant dung, which it resembles, 26.4.1964 (K); without data, via C.M.I. (K). UGANDA: Wilson 1332, Kadam Mt., 7,000 ft. (K, EA). TANGANYIKA: Braun 1949, Dodwe River, Amani, 16.2.1908 (EA, as Lanopila bicolor). Geastruin Micheli ex Pers. In this genus and its allies the exoperidium is well developed and dehisces radially from the apex, the resulting lobes bending back to give a stellate appearance to the mature fruit-body. The exoperi- dium consists of three layers, the outer mycelial, the middle fibrous and the inner fleshy. Tissue- tension between the two inner layers results in the rays being held at an angle more or less charac- teristic of the species. Some species are strongly hygroscopic, the rays being folded over the endo- peridium when dry, spreading when wet, again due to tension between the two layers. The endo- peridium, which is absent in a few species, dehisces by an apical pore. The characters of the peristome are important in distinguishing species. G. velutinum (aggregate sp.) (Fig. 8a, b) Unexpanded sporocarp globose or ovoid, top rounded or slightly umbonate, epigeal, attached to the substrate by a central, basal, radicating projection or occasionally a sin^e, weU-developed hyphal strand. Exoperidium saccate, split to about the middle into 5-8 expanded, broad, thick, subequal rays; fleshy layer flesh-coloured, becoming umber to dark umber when dry, moderately thick, adnate, usually cracking across the base of the rays to reveal the buff fibrous layer; mycelial layer thick, pliable, typically very finely felted-tomentose, buff-ochraceous, like very fine sand-paper to the touch, or sometimes more coarsely felted-tomentose, umber, but in any case almost free from debris, adnate or sometimes tending to bend away from the fibrous layer, especially at the points of the rays. Endoperidiiun smooth, sessile, globose to depressed globose, to 2 cm. diam., enclosed by the saccate base of the endoperidium; peristome broadly conical, fibrillose, depressed round the edge; concolorous or paler. Gleba umber, pseudocolumella cylindrical to spindle-shaped; spores globose 2.5-4p diam., fuscous, minutely to finely, occasionally moderately, verrucose. HABITAT: On the ground or rotting plant debris, in forest. DISTRIBUTION: Widespread outside Europe. MATERIAL EXAMINED: KENYA: P. J. Greenway, Muguga, in Acacia plantation (K). UGANDA: T. D. Maitland 81, Victoria Nyanza Reg. (K); Duemmer 1419, Kipango, 4,000 ft., on leaf-mould in forest (K); C. G. Hansford 2017, Masaka rd., mile 7, in forest (K); G. N. Calder 10, Mpangu, 4,300 ft., forest litter, 23.3.1964 (K); Calder 44, Mpangu, forest litter in light shade 13.4.1964 (K). NOTES: This group is distin^ished from G. saccatum by its usually larger size, harshly felted, usually paler, thicker, more pliable and continuous mycelial layer which often separates from the fibrous layer especially at the tips of the rays, and by its smaller, more finely verrucose, darker spores. Hansford 2017, Calder 44 and Duemmer 1419 have shortly tomentose, markedly “sand-papery”, greyish ochre, separable mycelial layers. The remainder have a mycelial layer which is more orangy- ochre in colour, and the tangled hyphae of which tend to be woven together into loose bunches. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 33 Fig. 7. Lycoperdaceae p.p. a, Langermannia wahlbergii, habit x i, spores and capillitiuni (Wilson 1332); b-c, Mycenastrum corium: b, habit x spore and capillitial element (Wilson 1333); c, habit, unexpanded, x i (Bally 337). Page 34 Some Gasteromycetes from Eastern Africa often holding a little debris. There is less tendency for the mycelial layer to split away from the fibrous layer, and the fruit-bodies tend to be smaller, with more numerous and pointed rays. There are other fairly well-marked forms of “velutinous” geasters, some of which have been given specific recognition, e.g. G. scleroderma Mont., and it is obvious that a thorough revision of the whole group is desirable. G. schweinitzii (Berk. & Curt.) Zeller {—G. mirabile Mont.) (Fig. 8c) Unexpanded sporocarps small, caespitose, obovoid, attached to a thick buff-coloured subiculum. Exoperidium deeply saccate, split into 5-7 unequal, short rays; fleshy layer flesh-coloured, drying brown, continuous or cracking across the base of the rays, adnate; mycelial layer buff, glabrous but in dried specimens minutely wrinkled. Endoperidium sessile 3-10 mm. diam., subglobose, dark brown, almost hidden in the saccate base of the exoperidium; peristome broadly conical, silky, concolorous or darker, marginally depressed. Gleba dark brown, spores 3-4p diam., fuscous, minutely to finely verrucose. HABITAT : On decaying wood, occasionally other plant debris, in forest. DISTRIBUTION : Africa, America, Australia, Ceylon, Japan, Pacific Is. MATERIAL EXAMINED: UGANDA: Duemmer 1462, Kipayo, on log in forest (K); Maitland 291, Entebbe, in forest (K), Hansford 2017, Musaka rd., mile 7 (K). MAURITIUS: Telfair, ex Herb. W. Bojer (K, as Lycoperdon pusillum). NOTES: This and a small group of similar species, is distinguished by the small size, crowded habit and growth from a subiculum on dead wood. The type specimen of Coilomyces schweinitzii (K), though immature, is unmistakably con- specific with that of G. mirabile (P), and the earUer, though less familiar, epithet must be used. Hansford 2017 was a mixed collection, mainly this species but with other fruit-bodies distributed in G. saccatum and G. velutinum. G. saccatum Fr. (Fig. 8d, e) Unexpanded sporocarp ovoid, umbonate, half-buried, attached by a bunch of hyphal strands which leave a prominent umbilical scar at the base of the sporocarp. Exoperidium splitting to about the middle into 5-7, occasionally up to 10, pliable, thin, subequal, acute rays which may become strongly revolute, or expanded, or remain at an angle of 45° to the vertical; fleshy layer pale rust colour to umber, adnate, frequently rimose; mycelial layer straw colour to ochraceous or rust colour, smooth, thin, radially and irregularly wrinkled at the base of the sporocarp, with adherent debris, flaking off in weathered specimens. Endoperidium sessile, 10-15 mm. diam., subglobose, glabrous, greyish- brown, usually partly enclosed by the saccate base of the exoperidium; peristome fibrillose; almost plane, concolorous with the rest of the endoperidium or paler, depressed at the periphery. Gleba umber; pseudocolumella indistinct, spores globose, 4-6p, usually 4-5, moderately to strongly (occa- sionally finely) verrucose. HABITAT: On soil or vegetable debris in damp places. DISTRIBUTION : Widespread, frequent in warmer areas. MATERIAL EXAMINED: KENYA: C. A. Thorold 38, Elburgon, 8,000 ft., on soil in forest, July 1932 (K); Verdcourt, Muguga, 1954 (K); A. French 8, Mpanya Forest, on ground in shelter of log, 30.3.1957 (K, with picture); Irwin 592, E. Mt. Elgon, 30.5.1960 (K);. Bally b12250, Nairobi, 5.5.1960 (K). UGANDA: Duemmer 1424, Kipango, 4,000 ft., leaf mould in forest, April 1915 (K); Hansford 2017, Masaka rd., mile 7 (K). NOTES : This species is often confused with G. velutinum to which it is not related. Fruit-bodies of G. saccatum tend to be smaller and the spores larger and lighter coloured. The mycelial layer is absolutely different, being thinner, rather dirty and smooth; in dried specimens it is thrown into characteristic wrinkles. It is more difficult to distinguish between G. saccatum and G. lageniforme to which it is closely related and with which it is connected by a few intermediate forms. G. saccatum, however, has shorter wedge-shaped rays and a lighter, thicker often more fragile mycelial layer. The East African sporocarps examined are all smaller and paler than is usual for the species. The mycelial layer is parchment coloured and lacks the characteristic cracks along the rays. G. lageniforme Vitt. (Fig. 8h) Unopened sporocarp ovoid with a strongly pointed apex, dehiscing when about 1.5 cm. broad. Exoperidium splitting to two-thirds or more of the way down into 5-8 (occasionally 10) subequal segments with long acuminate points; mycelial layer very thin, pale fulvous, entire; fleshy layer umber, adnate. Endoperidium sessile, 1-1.5 cm. diam., the base usually enclosed in the saccate J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 35 Fig. 8. Geastrum p.p. a, G. velutinum, felted form, habit x i, mycelial layer X 5, spore (Hansford 2017); b, G. velutinum, velutinous form, habit x i, mycelial layer x 5, spore (Calder 4417); c, G. schweinitzii, habit x 2, spores (Maitland 291); d, G. saccatum habit, egg and expanded specimen, x ^ (from drawing by A. French); e, G. saccatum habit x i, spores (Verdcourt); f, G. triplex, habit x i, spores (Irwin 590); g, G. triplex, habit x ^ (Rayner); h, G. lageniforme, habit, expanded specimen and egg, x spores (Bally 7404). Page 36 Some Gasteromycetes from Eastern Africa base of the exoperidium, umber; peristome fibrillose, often lighter than the surrounding endoperi- dium, conical, marginally depressed. Gleba umber, pseudocolumella clavate; capillitium up to lOp diam., thick-walled, straw-coloured, often encrusted; spores globose, 3.5-4.5p, pale to dark brown with regular or irregular, close, short, often flat-topped warts. HABITAT : On soil or vegetable debris in damp places. DISTRIBUTION: Cosmopolitan. MATERIAL EXAMINED: KENYA: Bally b7404, Lumbwa, Thiele- Winkler Estate, 7,300 ft., on forest floor, 12.9.1949 (K, EA). G. triplex Junghuhn (Fig. 8f, g) Unopened fruit-body 2 cm. diam., umbonate to pointed, with prominent basal mycelial tuft which falls away on drying to leave a marked scar. Exoperidium splitting to two-thirds of the way down into 5-8 broad, recurved or spreading rays ; mycelial layer papery to leathery, pale umber, smooth to squamulose, peeling off in flakes in old specimens, debris-encrusted in patches; fibrous layer leathery, thin, persistent, buff to straw colour on the outside, darker inside; fleshy layer thick, buff to pale rust colour, drying sepia, cracking, particularly across the bases of the rays, the cracked edges tending to curve away from the fibrous layer. Endoperidium sessile, to about 1.5 cm. diam., globose to depressed globose, smooth, pale vinaceous grey to vinaceous buff; peristome broad, often barely defined, silky, usually darker than endoperidium; mouth usually broad with fimbriate margin. Gleba umber, pseudocolumella ovoid; capillitium about equal in diameter to the spores, pale to medium brown, tapering at the ends, often encrusted; spores globose, 4.5-6p diam., regularly, closely and moderately to strongly warted, pale to dark brown. HABITAT: On plant debris, etc., on the forest floor. DISTRIBUTION: Cosmopolitan. MATERIAL EXAMINED: KENYA: Rayner, 5,500-6,000 ft., in Acacia forest; Rayner 740, Limuru Distr., Rift Highlands, 7,000 ft. (K); Irwin 590, E. Mt. Elgon, 8,000 ft., 26.5.1963 (K). TANGANYIKA: K. Pirozynski m593, Kakombe, Kigoma, damp soil, 24.2.1964 (K). NOTES: This species is distinguishable by the characteristic mode of cracking of the fleshy layer around the base of the rays to leave a “cup” enclosing the base of the endoperidium. This feature is not always present but is better seen in fresh than in dried specimens. G. fornicatum (Huds. ex Pers.) Hook. (Fig. 9e) Immature sporocarp subglobose, hypogeal. Exoperidium splitting to two-thirds of the way down into 4-5 rays, which bend backwards causing the mycelial layer to split away except at the tips of the rays and remain in the ground as a cup, whilst the other two layers become completely inverted, carrying the endoperidium upwards. Mycelial layer thickly coated with debris; fleshy layer often falling away in large patches. Endoperidium pedicellate, depressed globose, with an apophysis, hard, fuscous, minutely granular under the hand-lens; peristome silky, at first mammose and sub-definite, soon becoming indefinite and ragged, gleba brown vinaceous; capillitium threads brown, up to 15p diam.; spores globose, finely verrucose, dark brown 3.5-5p. HABITAT : On humus under trees or shrubs. DISTRIBUTION: S. Africa, Mauritius, Europe, Mediterranean region, N. and central America, Australia. MATERIAL EXAMINED : MAURITIUS : H. Bolus, without data, ex Herb. Berkeley (K, photo- graph only). NOTES: The record consists of only a photograph but is certainly referable to this species. The description of the microscopic and colour details is drawn from other material at Kew. G. dissimile Bottomley (Fig. 9f) Resembling G. fornicatum except for the mouth characters. Peristome conical, sub-definite, not tearing readily but becoming radially folded though not truly sulcate. MATERIAL EXAMINED: KENYA: E. A. Herb. 1167, Naivasha, Rongoni Forest, at roots of Acacia (K); Rayner 739, Limuru, E. Rift Highlands, 1942-3 (K). NOTES: This is accepted only provisionally as being specifically different from G. fornicatum. Rayner’s specimens have had most of the peristome eaten by insects, but the little which remains seems to be of the dissimile type. Contrary to Talbot’s claim (in Bottomley, 1948, p. 604) we do not accept that there is any material difference between type material of G. dissimile and the collection in Herb K, labelled by Lloyd "G. fornicatus Huds. forma macowani Kalch” and presumably the same as the specimens described by him as G. macowanii. The real identity of G. macowanii is unknown. It resembled the European /. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 37 form of G. fomieatus but was three times the size: it would have been a very large earth-star indeed. G. dissimile, however, is only about half the size of the European form. The peristome was, unhappily, missing from the type material of G. macowanii at the time when it was described. G. drummondii Berk. (Fig. 9b) Unexpanded fruit-body globose, epigeal, dehiscing when 1-2 cm. diam. Exoperidiuro splitting to the middle into 8-12 acute, subequal, strongly hygroscopic rays; fleshy layer sepia, adnate, conti- nuous; mycelial layer thin, whitish, covered with debris, often becoming detached. Endoperidium sessile or (in dried specimens) occasionally shortly pedicellate, globose to depressed globose, up to 15 mm. diam., dirty white or, less frequently, umber, finely asperulate often becoming smooth with age; peristome boldly and regularly sulcate, often darker than the surrounding parts of the endo- peridium. Gleba rust colour; spores globose, 4-6p diam., moderately and irregularly verrucose, often briefly pedicellate. HABITAT: On the ground in dry places. MATERIAL EXAMINED: BCENYA: Verdcourt 638, Muguga on bare soil after rain, 16.4.1952 (K). NOTES: The truely sulcate peristome distinguishes this species from those noted above and the hygroscopic nature of its rays separates it from the two following species. Southern African specimens of G. drummondii differ from the type in a number of ways, notably in having a thinner fleshy layer and being less hygroscopic. They tend, in fact, towards the European and N. American G. campestre. G. Iloydiattum Rick (Fig. 9c, d) Expanded sporocarps to c. 5 cm. diam. Exoperidium splitting to about the middle into 9-13 narrow unequal acute rays, somewhat hygroscopic; mycelial layer externally encrusted with soil particles, floccose, white to buff, peeling off in sheets, particularly from the tips of the rays; fibrous layer straw colour to buff, thin; fleshy layer reddish brown, to about 1 mm. thick, drying fairly thin, continuous to rimose, often completely lacking in old specimens. Endoperidium globose to depressed globose, sessile, finely felted, mid- to pale umber; peristome concolourous or darker, sulcate. Gleba sepia, capiUitial hyphae pale straw coloured, completely occluded, to 6p diam. ; spores dark brown, globose to irregular moderately and often irregularly warted, 3-5p diam. DISTRIBUTION : South and central America. Tropical Africa. MATERIAL EXAMINED: ZAMBIA: Angus m313, Chilanga, nr. Mt. Makulu, 8.1.1957, on soil in woodland (K). MADAGASCAR: Rev. R. Baron 5318, ?N.-W. Madagascar, 1887 (K). NOTES: This species is distinguishable from G. drummondii by the much less strikingly and less regularly sulcate peristome, the almost smooth endoperidium, and the less hygroscopic rays with less persistent fleshy layer and flocculent mycelial layer. G. nanum Pers. {=G. schmidelii Vitt.) (Fig. 9a) Unopened fruit-body hypogeal to exposed. Exoperidium splitting to about the middle into about 5-8 unequal rays which become recurved; mycelial layer umber to sepia in the dried specimen, persistent, covered with a layer of debris; middle layer pale buff; fleshy layer adnate, entire or occa- sionally cracked. Endoperidium c. i-1 cm. diam., ovoid to um-shaped, often with a slight basal apophysis, shortly pedicellate in the dried specimen, light brown to greyish, the latter colour being due to a farina which covers the endoperidium of imweathered specimens, peristome sulcate, conical, concolorous with or darker than the rest of the endoperidium, surrounded by a groove. Gleba very dark brown, spores globose to subglobose, strongly and somewhat irregularly warted, 3-5.5p diam., usually rather fuscous. HABITAT: Woodland or open places. DISTRIBUTION: Southern Africa, N. and S. America, Australia, India, Pakistan. MATERIAL EXAMINED: KENYA: L. D. Verdcourt, Muguga, Jan. 1956, abundant after rains on soil in woods, etc. (EA). NOTES: Hennings (1%2) records this species for the Usambaras as G. schmidelii. Myriostoma Desv. This genus resembles Geastrum but the spore sac dehisces by several pores and is supported on a number of pedicels. The gleba contains a number of pseudocolumellae. TTiere is a sin^e species. Page 38 Some Gasteromycetes from Eastern Africa M. coliforme (Dickson ex Pers.) Corda (Fig. 9i) Unopened sporocarp to about 6 cm. diam. Exoperidium splitting to about halfway into 5-10 sub- equal triangular spreading or reflexed segments. Fleshy layer soon peeling off to expose the dirty white middle layer; mycelial layer thin, smooth, flaking off in a manner like that of Geastrum triplex. Endoperidium to 4 cm. diam., depressed globose, umber, becoming shining grey, coarsely roughened, with several pedicels and several naked stomata. Gleba with pseudocolumellae corresponding to the pedicels. Spores globose, 3^p diam., excluding the incompletely reticulate ornament which may be up to 2p high; capiUitium hyphae mostly about 4p diam., coloured, branched, spirally twisted, thick- walled, with large crescentic pits in the walls, fragmenting at these pits, occasionally with large, spore- like swellings. HABITAT : On the ground. DISTRIBUTION: Cosmopolitan. MATERIAL EXAMINED: KENYA: R.W.R’s notes. NOTES: The gleba is strikingly different from that of any other earth-star. The capillitial hyphae exhibit hygroscopic movements when first mounted on the slide, the rope-like strands uncoiling. Microscopic data have been taken from S. African specimens in K, leg. MacOwan. Astraeus Morgan This genus differs from Geastrum principally in lacking a true hymenium at any stage of its development. This has been taken by many authors to imply that there is no close relationship between these genera and that any similarity is due to parallel evolution. Astraeus has thus been placed in a separate family. Partly for convenience, however, we have here included it in the Lyco- perdaceae. The single species resembles Geastrum macroscopically to a remarkable degree. There is, however, no pseudocolumella. Microscopically, the spores are much larger than those of most Geastrum species and the capillitium is unlike that of any Geastrum, being hyaline or nearly so, with swollen septa and modified clamp connexions, and strangely reminiscent of the capillitium of Tulostoma. Astraeus hygrometricus (Pers.) Morg. (=Geastrum hygrometricum Pers.) (Fig. 9g,h) Unopened fruit-body 4-6 cm. diam., subglobose. Exoperidium splitting to | of the way down into 5-10 (occasionally more) acute, thick segments, stron^y hygroscopic. Mycelial layer thin, scabrous, not persistent; middle layer hard, smooth brown; fleshy layer light umber or greyish becoming almost black with age, rimose. Endoperidium depressed globose, sessile, membranous, somewhat tomentose and marked with a network corresponding to the rimose cracking of the fleshy layer; mouth torn, indefinite; peristome indefinite. Gleba umber, capillitium threads to 8p diam., branched, thick-walled, often collapsed, almost hyaline, swollen at the septa; spores globose, 6-lOp diam., strongly and very closely verrucose, brown, with the stump of a pedicel. HABITAT : On the ground in woodland. DISTRIBUTION: Worldwide. MATERIAL EXAMINED: ZAMBIA: Angus m2615, nr. Chiwefwe Mine, nr. Mkushi Boma, soil in Brachystegia woodland, 8.4.1964 (K). Broomeia Berk. Sporocarps crowded together in groups of a few to about a thousand, each occupying a depression in the top of a common stroma, the whole said to be covered when young by a common exoperidium, opening to expose the individual endoperidia each of which dehisces by a single apical stoma. This genus shares with the related, monotypic Diplocystis from Cuba the distinction amongst gasteromycetes of being stromatic. The affinities of Broomeia and Diplocystis are unknown ; though they seem close to the Lycoperdaceae, and are here included in that family. Zeller (1949) has placed them, together with Lycogalopsis, which they scarcely resemble, in the Broomeaceae. A number of differences between Broomeia and more typical Lycoperdaceae are indicated below, but whether Broomeia is indeed a member of that family will remain uncertain until much more detailed studies have been carried out. This lack of knowledge extends particularly to the development of the fruit- bodies, but immature material, fresh or preserved in liquid preservative, would be necessary for such a study, and is not available. The character of the exoperidium is usually given as the basis of generic distinction between Broomeia and Diplocystis. In Diplocystis each fruit-body is said to have its own exoperidium while in Broomeia a common exoperidium enshrouds the whole cluster of fruit-bodies although passing between them and being continuous with the walls of the alveoli. Such a generic distinction caimot, however, be logically justified at the moment for a variety of reasons. /. E. Afr. Soc. Vol. XXVI No. 2 {.114) Page 39 Fig. 9. Geastreae p.p. a, Geastrum nanum, habit X 2, spores and capillitium (Verdcourt); b, G. drummondii, habit, dry and wet specimens x 1, spores (Verdcourt); c, G. Uoydianum habit x 1 and spores (Angus); d, G. Uoydianum, habit x 1 (Baron); e, G. fornicatum habit x 1 (Mauritius, from photo); f, G. dissimile habit X 1, spores (E.A. Herb. 1167); g, Astraeus hygrometricus habit, wet specimen X 1, spores and capillitium (Angus); h, A. hygrometricus habit, dry specimen x 1 (Israeli material); i, Myriostoma coliforme, habit x i, spores and capillitium (S. African material). Some Gasteromycetes from Eastern Africa Dissing & Lange (1962) rightly refuse to assume that in B. ellipsospora the exoperidium does in fact cover the fruit-bodies before maturity as it does in B. congregata. The exoperidium has never been seen in B. ellipsospora or D. wrightii, and mature specimens betray not the slightest trace of ever having possessed such a membrane. In any case it is perhaps unjustifiable to call such a membrane an exoperidium; it is an extension of the stroma and its homology must be an object of doubt and study. Finally, an exoperidium which covers each fruit-body separately cannot really be distinguished from one which is common to all, yet passes between each fruit-body and the next, and there becomes confluent with the rim of the alveolus in which each fruit-body is seated. The real difference between the two genera lies in the alveolae themselves. In Broomeia the alveolae are purely depressions in the stromatic surface; in Diplocystis they are not truly alveolae at all but cups, each of which is separate and does not have a wall in common with its neighbours. The two species of Broomeia are distinguished by the thickness of their stromata and the form of the spores. B. congregata Berk. (Fig. lOd-f) Sporocarps in clusters of up to a thousand, borne on a massive corky stroma, which is convex at the top. Spores subglobose to ovoid, 6-8.5p in major diam., finely reticulate. HABITAT : In dry areas, usually at the base of trees, often Acacia, on which it is probably parasitic. DISTRIBUTION: S. Africa, S.W. Africa, Mozambique, Kenya, Nigeria. MATERIAL EXAMINED: KENYA: Rayner, Crater Lake, Naivasha, sandy soil nr. Acacia xanthophloea (K). NOTES: This species has been recently described elsewhere (Bottomley, 1948; Dring, 1964) so only points of special interest, which are many, will be discussed here. The spores are reticulate in the type (K), in Rayner’s specimen and all the other collections to which we have had access. Reticulate spores are unknown in the Lycoperdaceae except in this species and a few species of Disciseda and Abstoma. The gleba of B. congregata, though powdery at maturity, is very compact. Capillitial hyphae often pass directly through the groups of spores, becoming characteristically tortuous, presumably by pressure of the spores against them. Sections of slightly immature gleba show spores nestling in the bends of the capillitium threads. The surface of the endoperidium is inconspicuously verrucose, as mentioned by Berkeley but overlooked subsequently. The curious microscopic structure of the warts is shown in Fig. lOe. They are usually subspherical to ovoid and most of them are only loosely attached to the endo- peridium. In the next species, B. ellipsosporus, they are even better developed, whereas in Diplocystis wrightii the endoperidium is smooth. The endoperidial warts of Geastrum drummondii and related species are very different. They are cylindrical, firmly attached and consist of rather distorted, plate- like, hyaline cells. The stroma consists of interwoven hyaUne hyphae apparently of uniform structure throughout. However, the part immediately beneath the sporocarps is more compact, whereas the centre consists of looser tissue, eventually becoming hollow. (See Pole Evans & Bottomley, 1919, pi. xxi, for a photo of a fresh, hollow stroma.) Dried herbarium material therefore often consists only of the sporocarps plus the upper parts of the stroma, the basal attachment having broken off through the hollow centre. The structure of the mature stroma of B. congregata therefore approaches that of B. ellipsospora. Fresh specimens are said to smell of aniseed, and the flesh of the fresh stroma is said to be red, resembling that of Fistulina hepatica. In herbarium material the red colour is limited to the compact upper layer of the stroma and is considerably less marked than in dried F. hepatica. Both the smell and the pigment are imique in the Lycoperdaceae. B. ellipsospora v. Hohn. {= Diplocystis jmodii Pole Evans & Bottomley) (Fig. lOa-c) Sporocarps in clusters of up to 80 on a thin, saucer-shaped stroma. Spores ovoid, 6-8 x 3.5-4.5jx, smooth or minutely roughened, sterigmatic scar visible at the broader end. HABITAT; On sandy soil. DISTRIBUTION : S. Africa, Mozambique, Angola, S.W. Africa. MATERIAL EXAMINED: MOZAMBIQUE: Rev. H. Junod, Rikatli, Louren?o Marques, 22.5.1919 (Cotype of D. jmodii, K). NOTES: The species has been well described by Bottomley (1948) and Dissing & Lange (1962), though the characteristically warted surface of the endoperidium is not mentioned in either paper nor in von Hohnel’s original description. MacroscopicaUy the warts are brownish and rather promi- nent but partly hidden by the floccose outer covering of the endoperidium. Both warts and flocci tend to wear off. In the case of the warts it is noticeable that their connexion with the underlying layer is tenuous, in fact their whole structure suggests that of sclerotia (Fig. 10b). J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 41 Fig. 10. Broomeia. a-c, B. ellipsospora: a, habit X 1; b, v.s. peridium; c, spores and capillitium (type of “Diplocystis junodii”); d-f, B. congregata: d, habit X 1; e, v.s. peridium Passing through a wart; f, spores and capillitium (Rayner). Page 42 Some Gasteromycetes from Eastern Africa According to the descriptions the spores of the type collection and Dissing & Lange’s (1962) collection from Angola are smooth. Bottomley describes the spores as smooth to rough. Those of the Lourengo Marques collection are minutely verrucose. The arrangement of the warts is rather difficult to see except when the spores are viewed in Melzer’s iodine solution which shows up the ornament, though whether it reacts with it is difficult to establish. The morphology of the stroma is described by Dissing & Lange (1962). As they state, there seems to be a “spurious” mycelial layer under the stroma and separate from it except, perhaps, at points directly under each fruit-body where the stroma protrudes downwards. The specimens at Kew give the impression that the space between the “mycelial layer” and the stroma is created by shrinkage of the tissues similar to that which occurs in B. congregata. Young stages have never been recorded, however, and one must in a genus so aberrant as Broomeia, accept such extrapolations only with extreme caution. TULOSTOMATACEAE This family, though very natural, is difficult to define in few words. Its members have a true stipe and a two-layered peridium, the outer layer of which encloses the immature fruit-body as a universal veil, splitting when the stipe elongates and then often remaining as a volva at the base of the stipe. The stipe is cleanly separable from the inner peridium of the head in one tribe, the Tulos- tomateae, confluent with it in the other, the Phellorineae. The mature gleba is always powdery and lacks any suggestion of columella or pseudocolumella. Nevertheless, the constitution of the gleba differs in the two tribes. In the Phellorineae, the mature gleba is characterized by the persistence of the tramal plates and of fascicles of basidia. True capil- litium is lacking, though the remains of the tramal plates may be mistaken for capillitial hyphae. In the Tulostomateae the gleba contains true capillitium and the tramal plates and basidial fasicles do not persist at maturity except in the transitional genera Battarraea and Schizostoma. The mode of dehiscence of the inner peridium is very variable and is used as a criterion for distinguishing between genera. Phellorinia Berk. Sporocarp consisting of a subglobose head supported on a thick, solid woody stipe. Outer peri- dium covering both head and stipe, warted or scaly, tending to fall away. Inner peridium continuous with the stipe tissue, dehiscing by irregular rupture and erosion of the apical part. For an explanation of the nomenclature applied to the genus and its one accepted species, see Dring (1964). P. herculeana (Pallas ex Pers.) Kreisel Sporocarp to 10 cm. tall; head 3-5 x 2-4 cm.; stem 3-5 x 1-2 cm. Peridium cream coloured. Gleba rust colour, powdery, spores globose, 4.5-7p diam., finely and closely warted, brown; capillitium absent but replaced by hyaline collapsed threads, aggregated into bundles, the remains of the tramal plates. P. herculeana ssp. herculeana (=P. inquians auctt. non Berk.) (Fig. lie) The outer peridium consists of large inbricate scales and the head is usually ellipsoid rather than subglobose. HABITAT : On the ground in dry, open places. DISTRIBUTION: Africa, S. W. United States, Mediterranean region, central Europe, Australia. MATERIAL EXAMINED: SOMALIA: Bally b9591, E. of Ghelinsor, 1,000 ft., open grassland, 5.4.1954 (K). NOTES: In the young, fresh state there is no possibility of confounding this subspecies with the next, though in herbaria much confusion has occurred. The feathery, overlapping scales make this one of the most beautiful of gasteromycetes, but unfortunately they are slightly sticky and tear away or become compacted when the specimen is handled. A good photograph or drawing, made in situ, of a young specimen is much to be desired. P. herculeana ssp. strobilina (Kalch.) Dring {=P. inquians Berk.) (Fig. lid) This differs from ssp. herculeana in that the outer peridium is very thick and splits into large zonate warts as maturity approaches. The habit is usually more squat than that of ssp. herculeana, the stipe being shorter and the head more rounded. DISTRIBUTION: Africa, Asia, Australia. MATERIAL EXAMINED: KENYA: Joy Adamson, nr. Lake Rudolph, Spring 1944 (K); Rayner 741, N. Frontier Distr., nr. Lake Rudolph, c. 1944 (probably part of the same coll., K). J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 43 Battarraea Pers. Head concave below, convex above, firmly fixed to the stipe, dehiscing around the intersection of the two faces. Stipe long, scaly, hollow, with a well developed volva at the base. B. stevenii (Liboschitz) Fr. {=Sphaericeps lignipes Welw. & Curr.) (Fig. llh) Immature sporocarp hypogeal, obovoid to obconical, enclosed in a universal veil. Head thrust above ground at maturity by elongation of the stipe. Stipe 10-50 cm. long, covered with coarse, imbricate scales, volva thick, woody, remaining below ground if stipe be pulled up. Gleba rust colour to sepia; capillitium of two types of hyphae, spirally or annularly thickened, elongated- fusoid threads, usually called “elaters” but constituting the capillitium proper, and collapsed, simple, hyaline threads bound together in fascicles, and really the remains of the trama (cf. paracapillitium); spores brown, subglobose, 4.5-6(-7)p diam., seemingly finely truncate-verrucose. HABITAT : Dry, sandy soil. DISTRIBUTION: Mediterranean area, Hungary, U.S.S.R., Mongolia, Pakistan, the Americas, Africa, Australia. MATERIAL EXAMINED: KENYA: Rayner, nr. Lake Naivasha, sandy ground in Acacia wood- land, May 1945 (K); Bally b10537, Maji ya Chumvi, Coast Prov., 2,700 ft., in Euphorbia thicket on sandy soil, 13.5.1956 (K). SOCOTRA: Prof. Bayley Balfour 1300, 1345, and another collection without no., all collected Feb.-March, 1880 (all K). NOTES: In addition to the above, this species has been recorded for Mozambique (Bottomley, 1948). There is some doubt that B. stevenii is really separable from the European B. phalloides. The latter tends to be smaller, with a less scaly stipe, less persistent gleba (i.e. thinner tramal plates), more heavily and irregularly ornamented spores, and has a volva consisting of two membranous layers said to be separated by a gelatinous layer when fresh. Individuals from the same collection sometimes show a remarkable variation in the first three of these characters. In fact, whatever their geographical provenance, the larger individuals tend to have a more scaly stipe and more persistent gleba. Spores of almost all the extra-European material of Battarraea in K are minutely and re^larly ornamented, while the European material in K (almost all of which is English) has spores with just discemibly less regular and slightly stronger ornament. Schizostoma Ehrenberg ex Leveille Mature sporocarp consisting of a relatively slender stipe fitting into a wide, shallow socket at the base of the head. Stipe hollow, woody, clearly differentiated from head but not cleanly separable from it as in Tulostoma and Queletia. Endoperidium dehiscing in a stellate manner. Universal veil forming a more or less well marked volva at the base of the stipe. Gleba consisting of true capillitium and subglobose, smooth spores. S. laceratim Ehrenb. ex Lev. (Fig. Ilf, g) Sporocarps to about 6 cm. tall, stipe to 0.5 cm. diam., usually not markedly scaly; head pulvinate, to about 3 cm. diam., 2 cm. high, splitting into about 4-6 blunt major lobes. Volva usually reduced to a sand-encrusted swelling at the base of the stipe. Gleba ochraceous, becoming a dark, rich, reddish-brown (near vinaceous of Dade); capillitium of dark, thick-walled, and paler, thin-walled, collapsed hyphae, irregularly shaped, branched, fra^enting into short lengths at the septa, to lOp diam., spores dark, irregularly globose, 3.5-6p diam., absolutely smooth under oil-immersion, apiculate. HABITAT: In semi-arid, sandy places. DISTRIBUTION: Somalia, Sudan, Mali, Persia, W. Pakistan, Khazakstan, N. and S. America, Australia. MATERIAL EXAMINED: SOMALIA: Bally 3034, Burao, 26.1.1944, in sandy soil. NOTES: The above specimen, and the description, refer to S. laceratum in the restricted sense. A number of authors consider S. laceratum to be the only valid species. Tulostoma Pers. Fruit-bodies stipitate, the subglobose head having a well-marked socket below, into which fits the slender, hollow, woody stipe and from which it may be broken away fairly readily and cleanly. Outer peridium taking the form of a universal veil in the immature fruit-body, remaining in the adult as a more or less well marked basal volva and a collar round the socket, and sometimes conti- nuing over the entire head, in which case it is warted. Endoperidium membranous, dehiscing by an apical stroma which may be indefinite or definite, and in the latter case tubular or more or less plane and with a definite peristome or without (in which case it is often called naked). Capillitium of branched and septate hyphae; spores globose to subglobose, smooth or variously ornamented. Page 44 Some Gasteromycetes from Eastern Africa Fig. 11. Tulostomataceae. a, Tulostoma album, habit X 1, spores and capillitium (Bally); b, T. exasperatum, habit X 1, peridial details X 5, spore and capillitium (Greenway); c, T. volvulatum, habit X 1, spores and capillitium; d, Phellorinia herculeana ssp. strobilina, habit X i (W. African material); e, P. h. ssp. herculeana, habit X spores (W. African material); f, Schizostoma laceratum, young specimen without volva, habit x spores (Bally); g, S. laceratum, old specimen, habit x 4 (N. American material); h, Battarraea stevenii, habit, without volva and upper part of peridium, X 4, spores and elater (Rayner). J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 45 T. volvulatum Borshchov (Fig. 1 Ic) Fruit-body usually 7-10 cm. tall, straw-colour throughout. Head depressed globose with plane, naked, circular to elliptical stroma and wide socket. Stipe scaly and longitudinally furrowed, with well marked volva at base. Gleba rust colour; spores subglobose to irregular, 4.5-5.5p diam., smooth and apiculate; capillitium of irregular, short, branched hyphae. HABITAT: Semi-arid ground. DISTRIBUTION : Africa north of the Equator, Central Europe, Asia. MATERIAL EXAMINED : BRITISH SOMALILAND : Bally 7379, on rocky ground, 2.6.1949 (K). SOMALIA: Bally 10401, Berbera-Behendulla road, on rocky ground 17.11.1954. T. album Massee (Fig. 11a) Head globose or depressed-globose, to about 2 cm. diam. Exoperidium impregnated with earth, cracking away late leaving a thick cup at the base. Endoperidium rosy-buff, minutely felted under the lens; mouth shortly tubular, elliptic; collar moderately wide. Stipe to 4 cm. x 3 mm., equal, or more usually attenuate downwards, with a well developed basal bulb, markedly striate, not scaly, concolorous with endoperidium to pale reddish brown. Gleba fuluous ; capillitium hyaline, thick- walled to occluded, to c. 6p diam., swollen and lightly pigmented at the septa; spores irregularly globose, 3.5-5p diam., honey coloured, very irregularly and more or less prominently verrucose, the verrucae being hyaline. HABITAT : On the ground. DISTRIBUTION: Australia, Southern Africa. MATERIAL EXAMINED: KENYA: Bally 2627 [without data] (K). NOTES: The material consists of a single sporocarp and is in some ways intermediate between T. album and T. albicans White. It is closely similar to J. P. H. Alcock’s collection 403, from Kim- berley (K), which Bottomley (1948) doubtfully ascribed to T. albicans. Both differ from the type specimen in having less prominently verrucose spores but these are certainly not smooth as in T. albicans. T. exasperatum Montagne (Fig. lib) Sporocarps in groups on rotten wood. Head depressed globose, to about 1.5 cm. diam.; outer peridium very dark umber, of long and short spines, covering the head, wearing off late to leave an areolate surface reminiscent of that of Lycoperdon perlatum-, inner peridium vinaceous; ostiole fimbriate, peristome definite, conical. Stipe to 5 cm. long, almost concolorous with peridial warts, with short, persistent, suberect scales. Gleba cinnamon buff, spores globose, very pale yellow, with deep hyaline ridges, 6.5-8.5p diam., including ornament. Capillitium hyaline, branched, thick- walled, slightly swollen and not coloured at the frequent, perforated septa. HABITAT : On rotten wood in forest. DISTRIBUTION: Tropical Asia, tropical S. America, tropical Africa. MATERIAL EXAMINED: TANGANYIKA: Greenway 5979, Msalai-Zaria, E. Usambara, 4,500 ft., 4.8.1940 (K, EA). NOTES : This species is easily recognizable by the long (to 2 mm.) peridial spines. The ornamentation of the spores is also distinctive, and the habit of growing on wood in forest remarkable in the genus. REFERENCES Ahmad, S., 1952. Gasteromycetes of West Pakistan. (Punjab Univ. Press), Lahore. Bottomley, A. M., 1948. Gasteromycetes of S. Africa. Bothalia 4: 473-810. Brodie, H. j. & Dennis, R. W. G., 1954. The Nidulariaceae of the W. Indies. Trans. Brit, my col. Soc. 37: 151-160. Buller, a. H. R., 1922. Researches on Fungi, Vol. 2. Coker, W. C. & Couch, J. N., 1928. Gasteromycetes of the Eastern United States and Canada. (Univ. of N. Carolina Press), Chapel Hill. CoNARD, H. S., 1913. The structure of Simblum sphacrocephalum. Mycologia 5: 264-273. Dennis, R. W. G., 1953. Some W. Indian Gasteromycetes. Kew Bull. 8: 307-328. Dissing, H. & Lange, M., 1962. Gasteromycetes of the Congo. Bull. Jard. Bot. Etat. Brux. 32: 325-416. Dissing, H. & Lange, M., 1963. Gasteromycetales I. FI. Icon. Champ. Congo., Fasc. 12: 215-232, pi. 38^0. Dring, D. M., 1964. Gasteromycetes of W. Tropical Africa. Mycol. Pap. No. 98. Dring, D. M. & Rayss, T., 1964. The Gasteromycete Fungi of Israel. Israel J. Bot. 12: 147-178. Heim, R., 1950. Le Genre Galeropsis Velen. Rev. Myc. Paris 15: 1-28. Hennings, P., 1902. Fungi Africae Orientalis, II. Bot. Jb. 33: 34-40. Page 46 Some Gasteromycetes from Eastern Africa Hennings, P., 1903. Fungi, in Warburg, Kunene-Sambesi Exp., p. 164. Berlin. Hennings, P., 1904. Fungi Africae Orientalis, III. Bot. Jb. 34: p. 45. Ingold, C. T., 1953. Dispersal in Fungi. Oxford. Kreisel, H., 1962. Lycoperdaceae der D.D.R Feddes Repert. 64 : 89-201. Kreisel, H., 1964. Vorlaufige tJbersicht der Gattung Bovista Dill, ex Pers. Feddes Repert 69: p. 21 1. Kreisel, H. & Diung, D. M., 1967. The Genus Morganella Zeller (Lycoperdaceae). ibid. 74: 109-122 Lloyd, C. G., 1906a. The Nidulariaceae. (Mycol. Writ., Vol. 2.) Cincinnati. Lloyd, C. G., 1906b. Letter 12, p. 1. (Mycol. Writ., Vol. 2.) Cincinnati. Long, W. H., 1907. PhaUoideae of Texas. /. Myc. 13: 102-114. Mattirolo, O., 1924. Fungi Africani. Atti. R. Acad. ScL, Torino 59: 713-740. Meulenhoff, J. S., 1936. Phallus hadriani Fr. Fungus 7: 38-39 and 53-55. Morse, E. E., 1933. A study of the Genus Podaxis. Mycologia 25: 1-33. Palmer, J. T., 1961. Observations on Gasteromycetes, X. Nidulariaceae in Persoon’s Herbarium. Persoonia 1: 433-451. Patouillard, N., 1924. Basidiomycetes nouveaux de Madagascar (suite). Bull. Mus. Hist. nat. Paris 3. : 526-532. Pearson, A. A., 1948. Phallus imperialis Schulzer J. S. Afric. Bot. 15: p. 176. PiLAT, A., 1958. Fungi, in Novak, Flora CSR, B-1, 36-95 Reid, D. A. & Dring, D. M., 1964. British Records, 71. Trans Brit, mycol. Soc. 47: 293-295. Singer, R., 1962. The Agaricales in Modem Taxonomy, ed. II. (J. Cramer) Weinheim. Smith, A. H., 1951. Puffballs and their Allies in Michigan. (Univ. of Michigan Press), Ann Arbor. Welwitsch, F. & CURREY, F., 1870. Fungi Angolenses. Trans. Linn. Soc. Lond. 26: 279-294. Zeller, S. M., 1949. Keys to the Orders, Families and Genera of the Gasteromycetes. Mycologia 41: 36-58. {Received for publication 25th May, 1966) J. E. Afr. nat. Hist, Soc. Vol. XXVI No. 2 {114) Page 47 BIRD RINGING FOR 1961—1966 By J. Smart The East Africa Natural History Society Bird Ringing Organization started ringing birds at the Eastleigh Sewage Works, Nairobi, in 1960. Brief reports have been published on the birds ringed up to the middle of 1961 (Blencowe, 1960 and 1962). This report covers all activities up to the middle of 1966. Table I which is divided into two lists, palaearctic migrants and African birds, shows the total numbers of each species recorded as ringed on all ringing returns received to date. In most cases the returns cover the period to 30th June, 1966, but. Prof. D. A. Zimmerman’s return up to August, 1966, has been included for the sake of completeness and because of the interesting species recorded. Several returns are still outstanding and those concerned who read this report are asked to send them in as soon as possible. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. Table I BIRDS RINGED BY THE EAST AFRICA NATURAL HISTORY SOCIETY BIRD RINGING ORGANIZATION UP TO 30th JUNE, 1966 Palaearctic Migrants (List 1) Ringed Plover. Charadrius hiaticula Linnaeus Mongolian Sand-Plover. Charadrius mongolus Pallas Great Sand-Plover. Charadrius leschenaultii Lesson Curlew Sandpiper. Calidris testacea (Pallas) Little Stint. Calidris minuta (Leisler) .... Ruff. Philomachus pugnax (Linnaeus) .... Common Sandpiper. Tringa hypoleucos Linnaeus Wood Sandpiper. Tringa glareola Linnaeus Marsh Sandpiper. Tringa stagnatilis (Bechstein) Greenshank. Tringa nebularia (Gunnerus) Grey Wagtail. Motacilla cinerea Tunstall Yellow Wagtails. Budytes species ..... Spotted Flycatcher. Muscicapa striata (Pallas) . Rock Thrush. Monticola saxatilis (Linnaeus) Wheatear. Oenanthe oenanthe (Linnaeus) Isabelline Wheatear. Oenanthe isabellina (Temminck & Langier) Sprosser. Luscinia luscinia (Linnaeus) .... Whitethroat. Sylvia communis Latham .... Garden Warbler. Sylvia borin (Boddaert) Blackcap. Sylvia atricapilla (Linnaeus) .... Upcher’s Warbler. Hippolais languida (Hemprich & Ehrenberg) Olivaceous Warbler. Hippolais pallida (Hemprich & Ehrenberg) Great Reed Warbler. Acrocephalus arundinaceus (Linnaeus) Reed Warbler. Acrocephalus scirpaceus (Hermann) . Sedge Warbler. Acrocephalus schoenobaenus (Linnaeus) Willow Warbler. Phylloscopus trochilus (Linnaeus) Swallow. Hirundo rustica Linnaeus .... Sand Martin. Riparia riparia (Linnaeus) .... House Martin. Delichon urbica (Linnaeus) Red-backed Shrike. Lanins collurio Linnaeus Red-tailed Shrike. Lanius cristatus Linnaeus Total . . . 1,452 African Birds (List 2) 1. SdiCTcd lb\%. Threskiornis aethiopicus (LdXh^Lm) 7 2. African Spoonbill. Platalea alba Scopoli 73 3. Lesser Flamingo. Phoeniconaias minor (Geoffroy) ...... 6 4. Hottentot Teal. Anas punctata Burchell 1 5. Quail. Coturnix coturnix (Linnaeus) ........ 1 6. Chestnut-banded Sand-Plover. Charadrius venustus Fischer & Reichenow . . 100 Page 48 Bird Ringing for 1961 — 1966 7. Kittlitz’s Sand-Plover. Charadrius pecuarius Temminck 8. Blacksmith Plover. Hoplopterus armatus (Burchell) .... 9. Avocet. Recurvirostra avosetta Linnaeus 10. Tambourine Dove. Tympanistria tympanistria (Temminck & Knip) 11. Emerald-spotted Wood-Dove. c^a/co5p//o5 (Wagler) 12. Pigmy Kingfisher. Ispidina picta (Boddaert) ..... 13. Grey-throated Barbet. Gy»7no()Mcco 6o«aparre/ Hartlaub 14. Golden-rumped Tinker-bird. Pogoniulus bilineatus (Sundevall) 15. Yellow-billed Barbet. TracAy/aewMS (Verreaux) . 16. Buflf-spotted Woodpecker. Campethera nivosa (Swainson) . 17. African Pied Wagtail. MoracjV/a ogMJwp Dumont .... 18. Brown Illadopsis. Ma/acocmc/fl/M/veJce«j (Cassin) .... 19. Pale-breasted Illadopsis. Malacocincla rufipennis (Sharpe) . 20. Scaly-breasted Illadopsis. Malacocincla albipectus (Reichenow) 21. Mountain Illadopsis. Malacocincla pyrrhopterus (Reichenow & Neumann) 22. Abyssinia Hill-Babbler. Pseudoalcippe abyssinicus (Ruppell) 23. Dark-capped Bulbul. Pycnonotus tricolor (Hartlaub) .... 24. White-vented Bulbul. Pycnonotus barbatus (Desfontaines) . 25. Bristle-Bill. Bleda syndactyla Swainson ...... 26. Brownbul. Phyllastrephus terrestris Swainson 27. Northern Brownbul. Phyllastrephus strepitans (Reichenow) 28. Smaller Yellow-streaked Greenbul. Phyllastrephus debilis (Sclater) 29. Fischer’s Greenbul. Phyllastrephus fischeri (Reichenow) 30. Toro Olive Greenbul. Phyllastrephus hypochloris (Jackson) 31. Olive-breasted Mountain-Greenbul. Arizelocichla tephrolaema (Gray) 32. Shelley’s Greenbul. Arizelocichla masukuensis (Shelley) 33. Yellow-bellied Greenbul. Chlorocichla flaviventris (Smith) . 34. Zanzibar Sombre Greenbul. Andropadus importunus (Vieillot) 35. Cameroon Sombre Greenbul. Andropadus curvirostris Cassin 36. Little Greenbul. Eurillas virens (Cassin) ...... 37. YeUow-whiskered Greenbul. Stelgidocichla latirostris (Strickland) 38. Ashy Flycatcher. Alseonax cinereus (Cassin) 39. White-eyed Slaty Flycatcher. Dioptrornis fischeri Reichenow 40. Puff-back Flycatcher. Batis capensis (Linnaeus) .... 41. Black-throated Wattle-eye. P/a/yj/e/ra pe/rata Sundevall 42. Chestnut Wattle-eye. Dyaphorophyia castanea (Fraser) 43. Jameson’s Wattle-eye. Dyaphorophyia jamesoni Sharpe 44. White-tailed Crested Flycatcher. Trochocercus albonotatus Sharp 45. Dusky Crested Flycatcher. Trochocercus nigromitratus (Reichenow) 46. Paradise Flycatcher. Tchitrea viridis (Muller) 47. Red-winged Paradise Flycatcher. Tchitrea suahelica (Reichenow) 48. Black-headed Paradise Flycatcher. Tchitrea nigriceps (Hartlaub) 49. Olive Thrush. Turdus olivaceus (Linnaeus) 50. Abyssinian Ground-Thrush. Geokichla piaggiae (Bouvier) 51. White-tailed Ant-Thrush. Afeoco^5ypA«5 poens/s (Strickland) 52. White-browed Robin-Chat. Cossypha heuglini Hartlaub 53. Blue-shouldered Robin-Chat. Cossypha cyanocampter (Bonaparte) 54. Red-capped Robin-Chat. Cossypha natalensis Smith 55. Snowy-headed Robin-Chat. Cossypha niveicapilla (Lafresnaye) . 56. Robin-Chat. Cossypha caffra (Linnaeus) 57. Equatorial Akalat. Sheppardia aequatorialis (Jackson) 58. Brown-chested Alethe. Alethe poliocephala (Bonaparte) 59. Eastern Bearded Scrub-Robin. Erythropygia quadrivirgata (Reichenow) 60. White-starred Bush-Robin. Pognocichla stellata (Vieillot) . 61. African Reed Warbler. y4cracepAa/M5' (Vieillot) . 62. Brown Woodland Warbler. Seicercus umbrovirens (Ruppell) 63. Black-collared Apalis. Apalis pulchra Sharpe .... 64. Grey-capped Warbler. Eminia lepida Hartlaub .... 65. Olive-green Camaroptera. Camaroptera chloronota Reichenow . 66. Grey-backed Camaroptera. Camaroptera brevicaudata (Cretzschmar) . 67. Hunter’s Cisticola. Cisticola hunteri Shelley 68. Banded Prinia. Prinia bairdii (CTassin) ...... 69. Black-faced Rufous Warbler. Bathmocercus rufus (Reichenow) . 70. Angola Swallow. Hirundo angolensis Bocage 71. African Sand Martin. Ripana pa/Mfi?ico/a (Vieillot) .... 3 6 1 5 2 4 1 4 2 1 1 7 13 21 1 3 1 1 7 7 2 3 33 5 15 4 4 3 14 20 195 2 4 2 8 10 20 3 13 1 3 1 15 1 1 6 7 37 2 6 29 27 5 12 1 4 2 1 23 2 1 7 22 4 2 J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) 72. Square-tailed Drongo. Dicrurus ludwigii (Smith) 1 74. Violet-backed Starling. Cinnyricinclus leucogaster (Boddaert) .... 3 75. Green White-eye. Zosterops virens Sundevall ...... 2 76. Kikuyu White-eye. Zosterops kikuyuensis Sharpe ...... 2 77. Bronze Sunbird. Nectarinia kilimensis Shelley 1 78. Eastern Double-collared Sunbird. Cinnyris mediocris Shelley .... 2 79. Olive Sunbird. Cyanomitra olivacea (Smith) ....... 26 80. Collared Sunbird. Anthreptes collaris (Vieillot) 3 81. Green Hylia. praj/na (Cassin) 4 82. Dark-backed Weaver. Symplectes bicolor (Vieillot) 1 83. Black-billed Weaver. Heterhyphantes melanogaster (Shelley) .... 4 84. Reichenow’s Weaver. Othyphantes reichenowi (Fischer) ..... 8 85. Grey-headed Negro-Finch. Nigrita canicapilla (Strickland) .... 2 86. Red-headed Blue-Bill. Spermophaga ruficapilla (Shelley) . . . . 14 87. Abyssinian Crimson-Wing. Cryptospiza salvadorii Reichenow .... 2 88. Peter’s Twin-Spot. Hypargos niveoguttatus (Peters) 1 89. Purple Grenadier. Granatina ianthinogaster (Reichenow) ..... 1 90. Streaky Seed-Eater. Scrims striolatus (Ruppell) 1 91. Thick-billed Seed-Eater. (Gray) 1 Total . . . 947 Grand Total, Lists 1 and 2 . . . 2^^ NOTE: The English and binomial names used in Table 1 are as used by Praed and Grant in their “Birds of Eastern and North Eastern Africa”, 1952. In List 1 it should be noted that no attempt has been made to separate the population of Yellow Wagtails into the various forms which occur in Kenya. In List 2: African Birds, number 5, Quail is in fact C. c. africana Temminck & Schlegel the Cape Quail; and number 70, Angola Swallow is H. a. arcticinta Sharpe the Uganda Swallow. Binomials have been used in the Table for the sake of uniformity. As far as is known all birds were caught in mist nets apart from the African Spoonbill pulli. The bulk of the palaearctic migrants ringed belong to four species : Yellow Wagtails, Swallow, Sand Martin and Little Stint. Most of the Yellow Wagtails were ringed by E. J. Blencowe and A. Carter in 1960 and 1961 at the Eastleigh Sewage Works, Nairobi. Although several birds were retrapped during the same winter, netting was not carried out for long enough over the second winter or in ensuing years to retrap birds after their long trip to their breeding grounds and back. In the 1960 to 1961 season ringing started on 19th November, 1960 and continued until 3rd April, 1961. Table II shows the number of Yellow Wagtails ringed each month and retrapping details. Table II NUMBER OF YELLOW WAGTAILS RINGED AND RETRAPPED AT EASTLEIGH SEWAGE WORKS, NAIROBI 1960/61 (1) (2) (3) (4) (5) Month Number Months in which Number retrapped Number retrapped ringed birds ringed in and expressed as each month and each Column (2) were a percentage of expressed as a month retrapped Column (2) percentage of Column (2) November, 1960 24 None retrapped - - - - December, 1960 148 January to AprU, 1961 5 3i% - _ January, 1961 . . 195 January to March, 1961 4 2% 2 1% February, 1961 . . 67 None retrapped - 2 3% March, 1961 104 None retrapped - - 3 3% April, 1961 48 None retrapped - - 2 4% The retraps are too small in number to be evaluated reliably but it is noticeable that none of the birds ringed in November, February, March and April were retrapped, whereas the percentage of birds caught from January onwards which had been ringed in December and January rose steadily from a figure of 1 % in January to 4 % in April. The figures suggest a population change among the birds using the Eastleigh Sewage Works throughout the winter. Page 50 Bird Ringing for 1961—1966 A further 131 Yellow Wagtails were netted and ringed at the Eastleigh Sewage Works in September and October, 1961 but this area has not been netted again between then and September, 1966. On 7th March, 1964, 36 Yellow Wagtails were netted and ringed as they came into roost in tall elephant grass on the Ruiraka river near Nairobi. The birds dropped from a height into an area of low scrub before entering the elephant grass and it was here that the nets were set up. In late 1965 netting was attempted (unsuccessfully) adjacent to another roost in tall elephant grass at Kabete, the birds flying directly into the elephant grass. Later, this roost was visited at night to attempt to hand-pick sleeping birds off the grass, but they were roosting too high and were too easily disturbed. On 6th November, 1965 one net was set up in a recently planted coffee plantation which was still largely open cultivated ground where Yellow Wagtails were found to flock before proceeding to a nearby roost. Nine birds were caught by driving and it was thought that many would be caught by driving if many nets were set up; however, when this was attempted on a later evening the birds completely avoided this cultivated area. Experience has shown that Yellow Wagtails can most easily be netted at the Eastleigh Sewage Works during the daytime. J. M. Lock has had two successful seasons netting Swallows and Sand Martins at Mweya in the Queen Elizabeth National Park, Uganda. He uses a flick-netting technique which was developed elsewhere. In this method a 60 foot tethered net is set up and fixed at one end only, the other end attached to a pole being held in the hand. He writes as follows (Lock, 1966, personal communication): “One pole must be either fastened to a tree, or else well pushed into the ground and guyed. The other pole is held horizontally by the operator. As the bird flies past the net is quickly jerked up, and immediately lowered to retain the bird if one is caught. The pole is then laid on the ground and the bird removed. Some practice is needed as it is not easy to judge the speed of the bird, and Swallows are extremely agile on the wing. Alternatively the net can be held at both ends, but this requires more practice and a great deal of co-ordination between the two operators. It is very important to clear the ground below the net of all obstructions, pieces of grass, etc., which mi^t catch in the net when it is laid down. The wind direction is important. Swallows using an area will feed slowly upwind. While doing this they are almost impossible to catch as they move too slowly and see the net. They are best caught as they sweep back downwind to the beginning of the feeding area. Elicking is best in a light wind.” Between 4th October, 1964 and 21st January, 1966 he succeeded in catching 210 Swallows, 156 Sand Martins and 1 House Martin. More interesting, three Swallows ringed in December, 1964 were retrapped in the same place in October and November, 1965. As far as is known these are the first birds to the ringed in East Africa and to be retrapped a year later after visiting Europe. This record emphasises the value of continued ringing activities over successive seasons and is an example of the fidelity of individual birds to a particular wintering area in Africa. Little Stints have been ringed largely at Magadi by E. J. Blencowe in 1961 and at Manyara by A. M. Morgan-Davies in 1962 and 1966. No retraps have been reported so far. Unfortunately the 3.0 mm. rings supplied are not monel and may not have a long life in the brackish water frequented by these birds. Although future ringing should probably best be concentrated on Swallows, Sand Martins and Yellow Wagtails which appear to be easiest to catch in large numbers and hence most likely to yield retraps, the ringing of single birds of a species is not necessarily unprofitable. The only Ringed Plover which was ringed (by S. M. Downhill at Mararani near the North Kenya coast on 29th August, 1962) was picked up in a sick condition and later died at Lindi on the South Tanzania coast four months later on 28th January, 1963, about 600 miles to the south. The indiscriminate netting of small passerines by J. B. Smart around Nairobi in an attempt to catch migrants was not very successful and it is clear that the netting of migratory warblers is best attempted only during a passage migration, as was done successfully for a short time by D. J. Pearson from Kampala in April, 1966. An interesting catch at Athi River was an immature White-throated Robin Irania gutturalis (Guerin) which was not ringed. Turning to the African birds, the bulk of these have been netted and ringed by Prof. D. A. Zim- merman during the course of his population studies in various forest areas, mainly at Kakamega, in 1965 and 1966. Mist nets were erected along narrow lines cut through the forest undergrowth with as little disturbance as possible, and were left up for several days before being moved, for as long as the catching rate stayed at a reasonable level. Nets placed parallel to and about 30 to 40 feet away from an open footpath had a greater catching rate than those at right angles to a footpath. D. A. Zimmerman first visited Kakamega forest in 1963 when he used American rings. Ringing details are not known but a number of retraps were made in 1965 and 1966 and East African rings substituted. The 73 African Spoonbill were all ringed as pulli at a nesting colony on Lake Naivasha in June, 1962 by A. Smith. The Chestnut-banded Sand-Plover were ringed at Magadi in 1961 by E. J. Blencowe and in 1964 by A. Smith. The rings may not have a long life as already explained. The only Hottentot Teal reported as being ringed (in the Ngorongoro crater on 3rd June, 1964) J. E. Afr. nat. Hist, Soc. Vol. XXVI No. 2 (114) Page 51 was shot at Lake Naivasha on 13th January, 1965, seven and a half months later and about 180 miles to the north. The only African Pied Wagtail ringed (at Muguga in September, 1961) was caught again at the same place three years later and released. This bird had recently attempted to breed when first caught and can be presumed to have been at least four years old in 1964. The Society has very little information on the recovery in East Africa of birds ringed abroad, and only two cases have come to the writer’s attention although it is certain that more have occurred. If any additional cases are known to readers of this paper the Society would greatly appreciate the details which should be sent to the Director, National Museum, Nairobi. An African Pochard Aythya erythwphthalma (Wied) ringed at Benoni in South Africa on Decem- ber 6th, 1953 was shot in November, 1954 at Lake Naivasha about 1,800 miles to the north. (Ostrich, January 1956). A Lesser Black-backed Gull Lams fuscus Linnaeus ringed at Pernaja Haveror in Finland on June 24th, 1961 was picked up at the Lake Rudolf Fishing Oasis on March 3rd, 1964 about 4,000 miles to the south. (Africana, June 1964.) Although not directly the concern of the Bird Ringing Organization and although it has been recorded elsewhere by L. H. Brown, it is of interest to include here the recoveries of Lesser Flamingos ringed as juveniles at Magadi with B.T.O. rings in October/November 1962. Of the 8,000 birds ringed seven were found dead at Magadi in March, 1963 and two were found at Sodera on the river Awash in Ethiopia in July, 1964 about 800 miles to the north. (Brown, 1966, personal com- munication.) Very httle is known about the movements of birds within East Africa and about the movements of migrants to and from their breeding grounds in Europe and Asia.* In time bird ringing will provide this information provided that recoveries are fully reported. Anyone finding a dead bird with a ring on it may send the ring to the address stamped on it, but should preferably send the ring with a description of the bird and place and date found to the Director, National Museum, Nairobi, who will then inform the Ringing Organization concerned. In this way all available information can be recorded at the National Museum where it can be built up and be available for all those interested. (Revised 1st November 1966) *The author has overlooked the fact that recoveries of European-ringed birds have been published by Eggeling, W. J. The Uganda Journal 15: 17 — 25 up to 1950. Ed. REFERENCES Blencowe, E. j., 1960. Report on Bird Ringing — 1960. J.E. Afr. Nat. Hist. Soc. XXIII: 300-301. Blencowe, E. j., 1962. Bird Ringing for 1960/61. J.E. Afr. Nat. Hist. Soc. XXIV: 74. \m ,1 ;>«sirt-.i n ' .■(it ’t.v* «/ ui'^t;iA'‘**,r '•, ' • ’■ ■■ , ' ■ • ■ ' ? it'J ifvoniT v^ii/ ■iwauptt j<*i .' iV">' 'T-< ' pU-mn "“■ lOVtoK fj ,,I.<5 11.4 ?w-rt$l.o> A{ Iff !• - ff‘lV I«i»#<\f!l*l1«» l>L«5Xtth t )bO /.*3>('jBftia*Mf ■».*«« rfi/ hb?4j!^*fiWW» ivv . (.^^f ijfiuf. .rt^K'.arfs /'jJ wr^ttec fHfsuSftvJr W-^* w- 'M. , , * ■' ,'. w'rAVi j «wst#’.Ua' ' ,:,'n«i ^t Wi /> *«A(38i(«f| »<;(» I •,f4» rt** **»«i> n(!m>fc ♦««ilrt?fiT v'’''f ^ ’fWl'j; ;.<(MUHVoa( ath wra iiiniWi * «JJiM hlfc - - . . . - w)*xiito» ji«/l i; fWw'l^^vi ■w,^»^H'•'^f*^u. >Mf *;h l>>i * >»J «fdhi »fU ‘ “‘ - " ■S r«> liMar-'^tfii fisp^'!D3 «*»u*!’;'»«^jf . ... ■ ^ .1'.-^. ^;..r ■■ ; ■ . .1 .,,lt / ;’,1''.»-.-":,- r M < -i»ft.,.l.-j-. i'. «? 'V- .i^.i lA'30^t^ ' ■• K" "M ■ >wP ^ i4i (/.■.&'"■■■■•■ . »<' f ■' f. '. \i -i.f .(in JJ'fWHi ml J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 53 PRELIMINARY OBSERVATION ON THE EFFECT OF WATER FLOW ON PROTOZOAN POPULATION By Apollo Hezekiah Ogambo-Ongoma {Department of Biology, West Virginia University, Morgantown, West Virginia, U.S.A.) INTRODUCTION Provasoli (1958) has said, “The ecological arena is populated by the products of the continuous challenge of nature to the potentialities of the organisms.” Protozoa are in close contact with their environment and should have a rapid and sensitive response to changes. Since one of the most striking variables in a lotic environment is the rate of flow, it was decided to see what effect two markedly differing rates had on the species composition of the protozona “community”. METHODS AND PROCEDURES This experiment was done at the University of Michigan Biological Station, Pellston, Michigan. The lake water used was from Douglas Lake — the lake on whose shores the station is located. Two troughs, 24"x2"x3", made of galvanized iron were set up. They were arranged so that each was tilting at an angle of approximately 15°, so that water could flow through them. However, both ends of each trough were sealed so that water had to flow in, form a pool, and then overflow. Because of the tilting of the troughs, the pool was deep at the bottom ends and became progressively shallower as one approached the elevated ends. The whole experimental set-up was done in a building and hence irridescent light tubes were set up to provide the necessary radiant energy for photosynthesis. Four stones were collected from Douglas Lake beach, these were washed to get rid of organic matter that might harbour protozoans and two of the stones were put in each trough. The stones were to trap organic matter and hence get the Protozoan community to establish. Table I Slow System Sample Small flagellates Large flagellates Small ciliates Large dilates Sarcodina 2 Days Elevated Area . . 3 Nil 1 Nil Nil Page 54 Protozoan Population --HinvoiriOOm I ro 0^ ”0 rZZZZZZZ ZZZZ ^82:^ „ . ^ Si-)Wi-4Wi-lWi-IWh-l Q Q Q Q ;g 3 sa J Si0>0iJ0iJ0ii0^0 § c/2 K-l J W »-J w h-l w J w h-l I- g,s.llll_ - "SPQOQa J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 {114) Page 55 Small Large Small Large Flagellates Flagellates Ciliates Ciliates Protozoan Population Lake water was then pumped into the two troughs at two different rates. 1. In trough Number I referred to as the Slow System, water was run through at the rate of 4 ml. per second. 2. In trough Number II referred to as the Fast System, water was run through at the rate of 20 ml. per second. The experiment was set up on July 18, 1966, and the first sampling done two days later; this was then carried out every Wednesday and Saturday of each week for three consecutive weeks. One drop of water containing organic matter constituted a sample and two samples were taken from each trough at each sampling. One of the two samples was taken from the pooled water in the area immediately around the stone in the elevated part of the trough. The second sample was taken from the pooled water in the lower area of the trough. Each time the samples were taken from the floor of each trough, and therefore contained as much of the organic sediments available as possible. From these samples counts were made and five categories of protozoans were formulated as a basis for grouping and counting. 1 . Small flagellate, e.g. members of the Orders Phytomonadina, Chrysomonadina, Protomonadina, and other “small-sized” flagellates. 2. Large flagellates, e.g. larger members of the orders Euglenoidina, Dinoflagel- lata, and other “large-sized” flagellates. 3. Small cihates, e.g. members of the order Oligotricha. 4. Large dilates, e.g. members of the orders Hymenostomata, Spirotricha, Holotricha and other “large-sized” cihates. 5. Sarcodina. In order to give a more critical analysis of the results, it was necessary to divide the cihates and flageUates into “small” and “large”, since they occurred more often. This was not necessary for the Sarcodina and hence they were left as one group. RESULTS AND DISCUSSION Results are tabulated in Table I and presented graphically in Figure 2. In course of experimentation, various things were noted: 1. In both systems, members of the Orders Phytomonadina, Protomonadina, and Chrysomonadina were first to establish themselves. Next to establish themselves were the larger flageUates such as members of the Orders DinoflageUata and Euglenoidina. Next in sequence of population establishment were the Sarcodina, then smaUer cihates (such as members of the order Oligotricha) and lastly the larger cihates, e.g. members of the Orders Spirotricha, Hymenostomata and Holotricha). 2. In the Slow System the protozoans were distributed from the lower area of the system right through to the elevated area. However, the lower portion of the system had a much higher concentration of protozoans. The Fast System on the other hand, had a very high concentration of protozoans in the lower area, but almost none in the elevated area of the System. 3. In the Slow System, there was a high concentration of cihates almost evenly distributed throughout the system. On the contrary, the number of cihates in ah parts of the Fast System was almost negligible. See Table I for the actual counts and Figure 2 for graphic representation. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 57 4. There was less organic sediment in the Slow System as compared to the Fast System. Distribution of the sediment was more even and increased gradually from the lower area to the elevated area of the system. On the other hand, in the Fast System organic sediment accumulated heavily in the lower area and almost none in the elevated area. 5. In the Slow System, whenever a dead crustacean or any small dead metazoan was found, there existed a high concentration of ciliates, especially the larger ones like hymenostomes, spirotrichs, and holotrichs. However, in the Fast System this concentration of ciliates never occurred even where dead metazoans were found. This sudden increase in number due to the dead metazoan may be noted in some counts shown in Table 2. Seeing that there existed a significant variation in number of ciliates in the two systems: slow system — 281 ciliates as compared to 10 ciliates in the fast system, yet both systems had everything equal except for variation in water flow, it was thought that water flow had something to do with this variation. It was therefore decided to reverse the rates of water flow and see what effect this would have. The slow system, had its water flow increased from 4 ml/second to 20 ml./second, and the fast system had its water flow decreased from 20 ml./second to 4 ml./second. Two days after this reversal sampling was done. As before, a drop of water containing organic matter constituted a sample and two samples taken from each system, in same areas as in the original set up. The second sampling was done 4 days, and the third and last one 7 days after the reversal. The population set up with regard to other groups of protozoans remained the same as before in the two systems, and hence no records of their counts were made. However, the ciliates were carefully counted and records made. The results of these counts are shown in Table II and graphical representation in Figure 3. *“Slow System” (20 ml./second) 1st Sample (2 days later) . 2nd Sample (4 days later). 3rd Sample (7 days later) Total Table II Small ciliates 3 2 0 5 Large ciliates 6 2 0 8 Total 9 4 0 13 *“Fast System” (4 ml./second) 1st Sample (2 days later). 2nd Sample (4 days later) . 3rd Sample (7 days later) Total .... 0 0 0 4 2 8 13 84 97 17 86 105 *These systems have reversed their water flow, however, to reduce confusion of nomenclature to the readers, the original titles are maintained even though the flow has changed. They are therefore put in quotation marks. On reversal of water flow in the two systems it was noted that the ciliate population in the “slow” system declined very considerably while in the “fast” system no ciliates had established themselves after two days. Four days after the reversal, the “slow” system showed further decline while the “fast” system began to show some ciliates established. Seven days after the reversal, there was a high concentration of ciliates in the “fast” system while the “slow” system had been depleted very considerably of the ciliates. Protozoan Population Number Of Ciliates Key: ■■ “Slow” System M~2 “Fast” System Figure 3. Graphic Representation of the New ciliate Population Set up. It was also noted that where a small dead metazoan occurred, there was excep- tionally high concentration of ciliates in the “fast” system while in the “slow” system, this concentration was not evident even where a dead metazoan occurred. This high concentration accounts for the sudden rise in one of the samples. Results of this experiment were interpreted in the following manner: 1. The sequence in which species become established in the two systems conforms to the generally accepted relationship between “producers” and “consumers”. However, exception to this occurred when some organic material (such as a dead metazoan) was introduced in the systems via the lake water. 2. Uniform distribution of organic sediment in the slow system in the original set- up is due to the slow current which allowed sedimentation to occur throughout the system. The fast system of this set up lacked organic sediment in the elevated area due to the sv^ft water current that swept the organic sediments to the lower area before they had a chance to settle. 3. Sudden population increases in all groups in the Slow System on the 20th day is partly due to the presence of a dead metazoan in the system and secondly, in this system sedimentation of organic matter has been increasing gradually from the lower area to the elevated area of the system. Therefore as the organic sediments increases so does the population increase in the area. In this same slow system, the ciliate population tends to remain approximately the same in the lower area, and the increase seems to occur in the elevated area. This is due to the fact that most of the organic sediment was swept to the lower area, and the accumulation in the elevated area seems to be more gradual and hence, the population increases with increase in the amount of organic sediment. Since all other factors were equal except for the rate of water flow in the two systems, it is apparent that ciliate population distribution was affected by the rate of water flow. This was confirmed when the set-up of the two systems was reversed as shown in Table II and Figure 3. As to why many ciliates should not exist in a system v^th a high rate of water flow, it is possible that the reaction of ciliates to gravity has something to do with it. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 59 Ciliates tend to be just below the surface film as a negative response to gravity. (Jennings 1906) established this fact during his work with Paramecium caudatum. (Kudo 1966: 157-164.) Since the ciliates remain just below the surface film, they were swept out of the system by the high water current before they had a chance to accumulate and reproduce and therefore establish themselves as part of the protozoan community. It is therefore evident that under laboratory conditions, water flow affects proto- zoan population composition and that a high rate of flow more or less eliminates ciliates from the system. Even though this occurred under laboratory conditions, I have no reason not to believe that a similar phenomenon could occur in a stream or any other natural lotic environment. ACKNOWLEDGEMENTS Grateful acknowledgement is made to Dr. John Cairns, Jr., Professor of Zoology, University of Kansas under whose direction the research was done. My thanks to the University of Michigan at Ann Arbor for letting me use their facilities on Douglas Lake. I wish to thank Professor Arnold Benson, Dept, of Biology, West Virginia University for helping in the final correction and reading of the manuscript. 1 wish also to thank Dr. Albert G. Canaris, Professor of Zoology, Dr. Earl L. Core, Chair- man, Dept, of Biology, and Dr. Robert Munn, Acting Dean of Graduate school. West Virginia University for making the necessary arrangements to enable me to be away from the Department while I was doing the Research. Lastly I wish to thank the Agency for International Development for providing the necessary finances while I was doing the work. REFERENCES Kudo, R. K., 1966. “Protozoology”, 5th Edition, Charles C. Thomas, Springfield, Illinois. Manwell, R. D. “Introduction to Protozoology”, St. Martins Press, New York. Provasoli, L., 1958. Ann. Rev. Microbiol, 12, 279. Reid, G. K., 1961. “Ecology of Inland Waters and Estuaries”, Reinhold, New York. (Received 14th November, 1966) UiMiCiWT • fli.'li).' rjiidttw «r.'i#b 'S« i^«,wb IS hati v^>rti inmm dyiilt wii if/ i»«r( 4t5 W’.4(!nt’.»rfli P^WuJho' ’’^^v)»1j^/)K • '■tijiWimit'i swsi x) ^i'.lw 1»o'^ IiC' ,i jr)xr*tia4ffi^7,q«f)0m^iij^rt<' 'tijUrfff rfj i)6 f^i tmtii lisiilj M-(jdj[Wiws?i^ 'dfX^ ('gntrmjU I mtP iti sift *40 l«jew» -jittv/ ^i>rlf ^ 4 ^laSufisuma 0.1 ^ .'immmttvi Di«oxf«o4j5>qi«0pi lip'^Blafm ttm& ^ «wva .iiM>i saaii:’'.^'fcrtsi »oiJWa»v.«k^^ ai>l '^JktWtOp lun., aiW'-i Mjfvi',-iu vijcsfuo^^ »<>yf*1 «hfine5 CHisdfA «0 iisadJ o;o«to vf{ 'i Ujtn' ^ \v}i’'0ytmf I yn i I|v4 *<'¥ ¥ u oan/S .iWrtvn’.o’ 'ufi “ful xt wi %»i _ ■* ^ ^ Jh nM )ii.w l oi jiw » ■• '■■ ‘ '* t-^' ■•>x."--'^- '..f^ 'llVL 14^^ " , ■ ' ‘ ' ..V ^ m i^iSV W^JN 4 ' Mwkwj? , 3f* %^m ' ’ ■ - ,. „.(< .1 . 1 1 /. E. Afr. not. Hist. Soc. Vol. XXVI No. 2 {114) Page 61 THE REEF HERON {Egretta schistacea Ehrenberg,) IS INTERIOR EAST AFRICA By Oscar T. Owre That the Reef Heron is exclusively a bird of the marine littoral has been a long and firmly held impression. Archer & Godman (1937:51), for example, stated categorically that this heron “is confined to the sea-coast, appearing nowhere inland”. More recent pubhcations, by omission of reference to the accumulating inland records, have tended to perpetuate this impression. Mackworth-Praed & Grant (1957:46), although they had previously contributed to delineation of the species’ inland range (see beyond), made no mention of this and Williams (1963) likewise omitted any comment as to occurrence inland. Records from the interior of the Continent are widely distributed. Berlioz (1922 :397 reported a specimen from Lac Abbay, Ethiopia. Chapin (1932:433) called attention to a specimen collected at Lake Albert by Sir Frederick Jackson in 1901. Grant & Mackworth-Praed (1933a:195) gave the range of this species in the interior of Africa as “up the Nile” and again (1933b :245), having examined the specimen referred to by Chapin {loc. cit.), revised the range to include distribution up the Nile as far as Lake Albert and they predicted that the heron would be found to occur all the way up the Nile to Lake Victoria. A specimen of the Reef Heron in the collec- tions of the National Museum of Kenya, which had been collected at Lake Rudolf in March 1947, was cited by North (1966:231) who also gave a recent sight record at Lake Nakuru in May. Forbes- Watson (1966:233) reported that between the period July 1960 and October 1962, he had, on occasions, seen Reef Herons at Ferguson Gulf, Lake Rudolf. Herewith are three additional records of the Reef Heron, all from Lake Rudolf. In 1958, as a member of the R. E. Maytag — University of Miami Expedition, I collected two specimens and made observations of a third. Pertinent data are: a d', Dec. 1, Allia Bay, weight 638 grams a $, Dec. 5, AUia Bay, weight 644 grams a sight record, Nov. 9, El Molo Bay The specimens are a part of the University of Miami Research Collections (UMRC). The male was examined by Dr. Dean Amadon of the American Museum of Natural History. He regarded it a subadult, judging by the amount of brown on the neck, the blackish back, and the lack of occipital plumes. The female resembles the adult grey phase plumage of the species; it lacks, however, plumes on either the back or head. The bird observed at El Molo Bay was similar to the latter, although it was not noted whether plumes were present or not. In these three birds, feathering of the chin, gular area, and anterior area of the jugulum was white. All were recorded as having yellow-orange mandibles, and the irides of the two specimens were yellow. The stomach of one bird contained a partly-digested Tilapia. North {loc. cit.) regarded it “odd” that present Kenya records of the Reef Heron are from inland. The number of such records now at hand suggests that the species is of more than accidental occurrence in interior East Africa. Indeed, considering that ornithological observation at Lake Rudolf has been limited, the number of records now at hand from this area indicate that, at least in small numbers, the species is present there during a considerable portion of the year. Tlie Reef Heron in East Africa It is to be hoped that those concerned with future inland records of the Reef Heron will pay particular attention to certain kinds of information. The concept of polymorphism in birds has been elucidated only recently. It is evident that the “morphs” of a species — in the Reef Heron morphs being evidenced by colour or plumage phases — may vary proportionately within the species’ range. Ratio-clines which thus exist suggest that the morphs may have differences in adaptability to the ecological situations. This is a consideration which has already received attention with respect to some reef herons (see, e.g. Thomson, 1964:658). The plumage patterns of inland visiting birds, therefore, should be noted in detail. At this time practically all of the records for which there is adequate information seem to be those of the grey morph or of those which appear to be intermediate between the grey and the white morph. The number of such records is probably still too small to be treated as significant and bias is no doubt inherent in the accumulation of this information, for white morphs which may occur inland are probably more easily overlooked by virtue of other white-plumaged herons. The record presented by Berlioz {loc. cit.) is apparently one instance of a white morph from inland. There are a number of plumages of the Reef Heron which are those of neither the typical grey nor the typical white morph. To what extent these are intermediate plumages (one assumes of hybrids between the morphs) or are attributable to age or seasonal change is not at this time at all clarified. Certainly these render difficult designation of birds observed in the field as immature or adult. Plumes, particularly occipital ones, should be looked for carefully. The locality, both geographically and ecologically, should be described as accu- rately as possible. “Lake Rudolf”, apparently the only notation as to the locality at which the specimen in the National Museum was taken, is hardly adequate. The three records herein contributed, from El Molo and Allia Bays, were all from areas of sandy, shallow shores, all three birds being encountered on narrow spits of sand. These situations were characterized by a sparsity of aquatic vegetation. Observations of feeding behaviour are of interest. One of the herons I collected, the one incidentally with an empty stomach, was resting in shallow water on its tarsi, its yellow toes extended in front of it. The wings, barely spread, were drooped, their tips dangling in the water. During the half hour or so that I watched this heron from close range, it gave no outward indication of fishing. Its wing tips rocked gently at times by the ripples, its head slightly elevated above the horizontal, the bird gave the appearance of sunning or, at times, of being asleep. Of much interest are sightings which may be indicative of route(s) the herons follow inland. Jackson’s (1938:45) record from Witu, while only fifteen miles or so from the coast, suggests that feeding situations may originally attract birds away from the coast. One may hypothesize that to such birds, river valleys (in this case the Tana River Valley) or other features of the terrain might afford avenues inland. Amadon {pers. com.) regards the birds that ascend the Nile as immature or off- season wanderers. With respect to those Reef Herons which do come into the interior along this route, it is interesting to contemplate the fact that Lake Rudolf was formerly a part of the Nile system. REFERENCES Archer, G. & E. M. Godman, 1937. The birds of British Somaliland and the Gulf of Aden. Vol. 1. Gurney & Jackson, London. Berlioz, M. J., 1922. Etude de la Collection d’Oiseaux rapportee par la Mission Du Bourg-de- Bozas de I’Afrique Tropicale (1902-1903). Bull. Mus. Nat. d'Hist. Nat., Paris. 28: 394-398. Chapin, J. P., 1932. The birds of the Belgian Congo. Part I. Bull. Amer. Mus. Nat. Hist., 65. /. E. Afr. not. Hist. Soc. Vol. XXVI No. 2 (114) Page 63 Forbes-Watson, a., 1966. A further note on reef herons in east Africa. J.E. Afr. Nat. Hist. Soc. 25; 233. Grant, C. H. B. & C. W. Mackworth-Praed, 1933a. On the relationship, status, and range of Egretta garzetta, Demigretta gularis, D. schistacea, D. asha, and D. dimorpha, a new subspecies, and the correct type-locality of Egretta garzetta. Bull. Brit. Ornith. Club. 53: 189-195. — 1933b. Further notes on Demigretta schistacea, Egretta garzetta garzetta, Egretta garzetta dimorpha and some corrections. Bull. Brit. Ornith. Club. 53 : 245-246. Jackson, F. J., 1938. The birds of Kenya Colony and the Uganda Protectorate. Vol. 1. Gurney & Jackson, London. Mackworth-Praed, C. W. & C. FI. B. Grant, 1957. Birds of eastern and north-eastern Africa. Vol. 1, 2nd Ed. Longmans Green and Co., London. North, M. E. W., 1966. A reef heron at Lake Nakuru, Kenya. J.E. Afr. Nat. Hist. Soc. 25: 231-232. Thomson, A. L., 1964. A new dictionary of birds. Thomas Nelson Ltd., London. Williams, J. G., 1963. Field guide to the birds of east and central Africa. Collins, London. (Received 2nd August 1966) / i' v H'' /^w*p i? 'JJif ^ ' .V'm '■ / .'‘'* ' ' •' ...V V.'.^ A S>..v ■ * %'4lW» J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 65 THE IDENTIFICATION OF ALOES IN EAST AFRICA By J. B. Gillett INTRODUCTION Many readers of the Journal of the East African Natural History Society must, no doubt, be already acquainted with Dr. G. W. Reynolds’ splendid book “The Aloes of Tropical Africa and Madagascar” which was published in 1966. Those who have tried to use it to identify Aloes in East Africa will probably have found Dr. Reynold’s key to the groups into which he divides the genus difficult to follow and will have regretted the absence of any quick means of ascertaining which species have been found in any given area. The present paper is an attempt to supply the latter desideratum and to provide a key, which, it is hoped, will be easier to use. It is in no sense an original work and is not based on any detailed study of the genus. It is merely an attempt to reorganize some of the information supplied by Dr. Reynolds so as to make it easier to use. Nobody should try to use the present paper by itself to name Aloes. It should be used simply as an adjunct to Dr. Reynolds’ book and if it helps the reader to arrive more quickly at Dr. Reynolds’ descriptions and illustrations, by reference to which alone can the naming of Aloes be carried out with any approach to confidence, the aim of the author will have been achieved. In the table of geographical distribution the following areas are recognized. WA is Tropical Africa west of the eastern boundary of the former British Cameroons. CA, Central Africa, is the former French Equatorial Africa and former Belgian territory with Spanish and Portuguese enclaves. It is divided into X, the whole area except Rwanda- Burundi and R, Rwanda-Burundi. NE, the North Eastern Area, is divided into SU, the Sudan Republic; ER, Eritrea; AR, Arabia; SC, Socotra; SM, Somalia and ET, Ethiopia apart from Eritrea. EA, East Africa, consists of Uganda, Kenya and Tanzania, U 1-4, K 1-7 and T 1-9 are the provinces of these countries, as recognized in “The Flora of Tropical East Africa” T9 being the islands of Zanzibar and Pemba. STA, South tropical Africa is divided into AN, Angola ; ZA, Zambia ; MA, Malawi, PE, Portu- guese East Africa; RH, Rhodesia (Zimbabwe) and BO, Botswana (Bechuanaland protectorate). SA is South Africa, together with South West Africa, Lesotho and Swaziland. In this column only those species are included which are known also to occur in one or more of the other areas. In the final column a K indicates that the species is dealt with in the key. Empty horizontal lines in the geographical table are used to separate the groups into which Dr. Reynolds divides the genus. These groups are not the same as those used in the present key to East African species. The Key to East African species deals with all species known to occur in Rwanda-Burundi, Uganda, Kenya and Tanzania, together with additional species recorded from adjacent parts of neighbouring countries. The numbers of the species in the key are those used by Dr. Reynolds and can thus be used for quick reference to his work without the need to consult the index. TABLE OF GEOGRAPHICAL DISTRIBUTION OF THE TROPICAL AFRICAN SPECIES OF ALOE 1 myriacantha URRCMT 1234 234567 123456789 .XX.X. XX. ..XX.. ..x.x. K 2 ballii 3 torrei 4 plowesii 5 howmanii 6 wildii 7 musapana 8 inyangensis 9 hazeliana 10 rhodesiana 1 1 buchanahii 12 nuttii X. . 2 X.. . 3 XX. . 4 X. . 5 XX. . 6 X. . 7 X. . 8 XX. . 9 XX. . 10 X... . 11 .X..X.. XXX... . 12 K “The Aloes of Tropical Africa and Madagascar” by G. W. Reynolds (1966). Obtainable from: The Aloes Book Fund, Box 234, Mbabane, Swaziland. Price Shs. 98/-. Page 66 Aloes in East Africa URRCMT 1234 1234567 123456789 naaeho 13 richardsiae 14 bullockii 15 buettneri 16 jucunda 17 hemmingii 18 jacksonii 19 somaliensis 20 erensii 21 peckii 22 mcloughlinii 23 pirottae 24 dorotheae 25 morogoroensis 26 greenwayi 27 amudatensis 28 graminicola 29 kilifiensis 30 greatheadii 31 swynnertonii 32 duckeri 33 saponaria 34 zebrina 35 macrocarpa 36 lateritia 37 hereroensis 38 chabaudii 39 bukobana 40 milne-redheadii 41 mzimbana 42 rivae 43 grata 44 niebuhriana 45 rigens 46 tomentosa 47 doei 48 trichosantha 49 menachensis 50 pubescens 51 eremophila 52 serriyensis 53 dhalensis 54 audhalica 55 barbadensis 56 metallica 57 massawana 58 vacillans 59 officinalis 60 otallensis 61 splendens 62 cremnopliila 63 pendens 64 confusa 65 veseyi 66 mendesii 67 penduliflora xxxxxx X 24 K 25 K 26 K J. E. Afr. mt. Hist. Soc. Vol. XXVI No. 2 (114) 68 venusta 69 macrosiphon 70 compacta 71 cryptopoda 72 crassipes 73 christianii 74 pretoriensis 75 forbesii 76 perryi 77 scobinifolia 78 sinkatana 79 elegans 80 wrefordii 81 sinana 82 camperi 83 adigratana 84 calidophila 85 inermis 86 globuligemma 87 turkanensis 88 leachii 89 guerrai 90 secundiflora 91 ortholopha 92 mawii 93 aculeata 94 rubroviolacea 95 decurva 96 lavranosii 97 ruspoliana 98 classenii 99 sereti 100 mubendiensis 101 wilsonii 102 nkambensis 103 breviscapa 104 tweediae 105 percrassa 106 harlana 107 steudneri 108 berhana 109 monticola 110 schelpei 111 keayi 112 schweinfurthii 113 megalacantha 114 macleayi 115 microdonta 116 marsabitensis 117 medishiana 118 gracilicaulis 119 angolensis 120 gillilandii 121 excelsa 122 littoralis URRCMT 1234 1234567 123456789 naaeho X Page 68 Aloes in East Africa 123 munchii 124 rupicola 125 ballyi 126 volkensii XR URRCMT 1234 1234567 123456789 naaeho K K 127 squarrosa 128 zanzibarica 129 tororoana 130 hendrickxii 131 deserti 132 hildebrandtii 133 yavellana 134 andongensis 135 cameronii 136 palmifomiis 137 retrospiciens 138 babatiensis 139 elgonica 140 flexifolia 141 boscawenii 142 rabaiensis 143 dawei 144 gossweileri 145 catengiana 146 kedongensis 147 ngobitensis 148 nyiriensis 149 arborescens K K K K K K K K K K K K K K 150 sebaea x. . 151 eminens x 150 151 THE NUMBER OF ALOE SPECIES IN EACH AREA West tropical Africa 4, of which 1 endemic Central tropical Africa excluding Rwanda-Burundi 12 Rwanda-Burundi 5 Central tropical Africa including Rwanda-Burundi 15, of which 2 are confined to the area Sudan Republic 8 Eritrea 7 Arabia 18, of which 2 also in Africa Socotra 3, all endemic Somalia 20 Ethiopia (excluding Eritrea) 25 North Eastern Africa and Arabia as a whole 68, of which 53 are confined to the area Uganda 1 (Northern Province) 9 U 2 (Western Province) 4 U 3 (Eastern Province) 2 U 4 (Buganda) 3 Uganda as a whole 14, of which 3 endemic Kenya 1 (North Eastern Province) 9 K 2 (Turkana) 4 K 3 (Rift Valley Province) 7 K 4 (Central Province ) 12 K 5 (Lake Province) 1 K 6 (Masai Province) 5 K 7 (Coast Province) 9 Kenya as a whole 26, of which 6 endemic J. E. Afr. nat. Hist. Soc. Vol XXVI No. 2 {114) Tanzania 1 (Lake Province) 6 T 2 (Northern Province) 8 T 3 (Tanga Province) 9 T 4 (Western Province) 9 T 5 (Central Province) 1 T 6 (Eastern Province) 3 and 1 doubtful T 7 (Southern Highland Province) 8 T 8 (Southern Province) 4 T 9 (Zanzibar and Pemba) 3 all doubtful Tanzania as a whole 30 and 1 doubtful, of which 14 endemic East Africa (Uganda, Kenya and Tanzania) as a whole 54 and 1 doubtful, of which 32 are confined to the area Angola 17 Zambia 15 Malawi 15 Portuguese East Africa 18 Rhodesia 25 Botswana 5 South Tropical Africa as a whole 47, of which 27 are confined to the area South Africa 133, of which 11 occur also in Tropical Africa Key to the species of Aloe occurring in Rwanda-Burundi, Uganda, Kenya and Tanzania anp adjacent parts of neighbouring countries. Based on the account of these species given in G. W. Reynolds “The Aloes of Tropical Africa and Madagascar” (1966). Key to groups (these are artificial groups for the purpose of the key and not the more or less natural groups recognized by Dr. Reynolds in his book). Branches of inflorescence 1-4: Acaulescent, or the stems under 50 cm. long: Teeth on leaves under 5 mm. apart; leaves under 5 cm. wide at the base . . Group 1 Teeth on leaves over 5 mm. apart, leaves often more than 5 cm. wide at the base . Group 2 Stems over 50 cm. long; teeth on leaves over 5 mm. apart: Bracts more than half as long as pedicel Group 3 Bracts less than half as long as pedicel Group 4 Branches of inflorescence 5 or more: Acaulescent, or stems under 50 cm. long: Bracts more than half as long as pedicel: Outer perianth segments united for more than 60% of their length . . Group 5 Outer perianth segments united for less than 60 % of their length . . . Group 6 Bracts less than half as long as pedicel: Outer perianth segments united for 60% or more of their length . . . Group 7 Outer perianth segments united for less than 60 % of their length . . . Group 8 Stems over 50 cm. long: Bracts more than half as long as pedicel Group 9 Bracts less than half as long as pedicel : Outer perianth segments united for 60% or more of their length . . . Group 10 Outer perianth segments united for less than 60% of their length . . . Group 1 1 Group 1 Rootstock not a bulb; leaves with a few spots at the base: Bracts up to 15 mm. long, pedicels over 10 mm. long: Bracts as long as pedicels; perianth 15-20 mm. long, the outer tepals free to the base Bracts i-2/3 as long as pedicels; perianth 38-42 mm. long, the outer tepals 10-75 % united .... Bracts up to 4 mm. long; pedicels 5-7 mm. long; perianth 27 mm. long, the outer tepals 75 % united . Rootstock a bulb; leaves without spots; outer tepals 60-70% united: Bracts 25-30 mm., pedicels 5-7 mm., perianth up to 48 mm. long Bracts 8-10 mm., pedicels 4-5 mm., perianth 30 mm. long . 1 myriacantha (Haw.) R. & S. 12 niutUBak. 1 8 jacksonii Reynolds 1 3 richardsiae Reynolds 14 bullockii Reynolds Aloes in East Africa Group 2 Bracts under 7 mm. long; pedicels under 12 mm. long: Leaves under 5 cm. wide; perianth under 28 mm. long: Leaves under 2 cm. wide Leaves over 2 cm. wide: Bracts more than \ as long as pedicels Bracts less than i as long as pedicels Leaves 5 cm. or more wide : Perianth under 25 mm. long Perianth over 30 mm. long: Bracts shorter than pedicels: Pedicels under 12 mm. long: Bracts 3 mm. long; stamens exserted . Bracts 6 mm. long; stamens not exserted Pedicels over 15 mm. long; bracts 5 mm. long . Bracts longer than pedicels Bracts over 7 mm. long: Pedicels imder 11 mm. long: Teeth on leaves about 8 mm. apart .... Teeth on leaves over 10 mm. apart: Bracts about 7 mm. long ..... Bracts about 12 mm. long Pedicels 14 mm. long, or more: Perianth under 25 mm. long Perianth over 27 mm. long: Bract i as long as pedicel; leaves 3 times as long as wide Bract more than i as long as pedicel; leaves 6-9 times as long as wide: Perianth 35 mm. long Perianth 28-33 mm. long 18 yacA:jom7 Reynolds 128 zanzibarica Milne-Redhead 129 mroroonu Reynolds 129 tororoana Reynolds 24 dorotheae Berger 25 morogoroensis Christian 102 ukambensis Reynolds 57 massawana Reynolds 26 greenwayi Reynolds 57 massawana Reynolds 131 desertiEngX. 27 amudatensis Reynolds 41 mzimbana Christian 70 compacta Reynolds 99 sereti De Wild. Group 3 Bracts shorter than the pedicels: Pedicels 15-20 mm. long; leaves without, or with few dots: Leaves 2.5-4 cm. wide: Infloresecence not pendent; bracts 7 mm. long Inflorescence pendent; bracts 10 mm. long . Leaves 7-8 cm. wide; bracts 13 mm. long Pedicels imder 10 mm. long: Pedicels c. 7 mm. long; leaves with many dots Pedicels 1-2 mm., bracts 1 mm. long Bracts longer than the pedicels: Bracts 12, pedicels 7-8, perianth 32-35 mm. long Bracts up to 30, pedicels 20-25, perianth 38-40 mm. long 64 confusaEa^. 67 penduliflora Bak. 70 compacta Reynolds 128 zanzibarica Milne-Redhead 92 mawii Christian 131 desertiEn^. 139 babatiensis Chvi&W&n Group 4 Stems hanging; leaves 2.5-4 cm. wide; bracts 6-10 mm. long: Perianth 25 mm. long Perianth 30 mm. long Stems not hanging; perianth 33 mm. long, or more: Pedicels 1-2 mm. long; leaves up to 10 cm. wide Pedicels over 12 mm. long: Perianth 40 mm. long; leaves 9 cm. wide; pedicels 20- 25 mm. long Perianth 33-36 mm. long: Pedicels 14 mm. long; leaves 6-9 cm. wide . Pedicels 20-25 mm. long: Leaves about 3.5 cm. wide Leaves about 5 cm. wide 65 veseyi Reynolds 64 confusa Engl. 92 mawii Christian 139 elgonica Bullock 143 dawei Berger 146 kedongensis Reynolds 147 ngobitensis Reynolds J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 71 Group 5 Bracts not above 7 mm. long; pedicels under 11 mm. long: Perianth over 25 mm. long: Perianth 29-30 mm. long; leaves with many spots: Teeth on leaves 4-6 mm. apart Teeth on leaves 10 mm. or more apart Perianth 40-45 mm. long, leaves without spots Perianth 16-20 mm. long; leaves without, or with few spots . Bracts over 9 mm. long; pedicels usually over 11 mm. long: Bracts shorter than the pedicel: Leaves with many dots : Perianth with a pronounced basal swelling, markedly constricted above this : Bracts 2-3 mm. broad; pedicels 20 mm. or more long: Perianth 33 mm. long Perianth 35-38 mm. long Bracts 6 mm. broad; pedicels 16 mm. long Perianth not constricted above the base; bracts 10 mm. broad; pedicels 13 mm. long Leaves with few or no dots: Bracts 10, pedicels 14, perianth 38 mm. long Bracts 5-6, pedicels 8-10, perianth 40-45 mm. long Bracts longer than the pedicel : Leaves spotted, 8 cm. wide Leaves not spotted, 4 cm. wide ..... Group 6 Bracts over 7 mm. long: Leaves up to 9 cm. wide, with many spots: Perianth sharply constricted above the ovary . Perianth not sharply constricted above the ovary : Bracts deflexed, twice as long as the 6-7 mm. long pedicels Bracts erect, shorter than or less than 50 % longer than the pedicels: Bracts c. 1 1 mm. long, 10 mm. broad, as long as the pedicels Bracts 15 mm. long, 8 mm. broad, 50% longer than the pedicels Leaves up to 15 cm. wide, not, or hardly, spotted Bracts under 7 mm. long: Flowers all turned to one side of the inflorescence rhachis (“secund”); teeth on leaves 2 mm. or more long, 10 mm. or more apart : Perianth c. 25 mm. long; leaves with few-many spots Perianth over 29 mm. long; leaves without spots: Leaves c. 6 cm. wide; perianth 30 mm. long Leaves 12-24 cm. wide; perianth 35 mm. long Flowers not “secund”; perianth under 35 mm. long; teeth on leaves small (up to 1 mm. long); 5-8 mm. apart in lower part of leaf : Perianth 16-20 mm. long, the outer segments united for 60% of their length Perianth c. 23 mm. long, the outer segments united for 45 % of their length Group 7 Bracts 10 mm. long or more; perianth 35 mm. long or more, sharply contracted just above the ovary: Bracts less than half as long as pedicels : Pedicels c. 30 mm. long, leaves 10-12 cm. wide Pedicels c. 35 mm. long, leaves 8-9 cm. wide . 20 erensii Christian 23 pirottae Berger 73 Christiana Reynolds 97 nispoliana Bak. 28 graminicola Reynolds 36 lateritia Engl. 29 kilifiensis Christian 68 venusta Reynolds 72 crassipes Bak. 73 Christiana Reynolds 69 macrosiphon Bak. 130 hendrickxii Reynolds 29 kilifiensis Christian 60 otallensis Bak. var. elongata Berger 68 venusta Reynolds 69 macrosiphon Bak. 80 wrefordii Reynolds 87 turkanensis Christian 88 leachii Reynolds 90 secimdifiora Engl. 97 ruspoliana Bak. 115 microdonta Gttov . 32 duckeri Christian 36 a lateritia Engl, var. lateritia Page 72 Aloes in East Africa Group 7 (Continued) Bracts 16 mm. long, equalling the pedicels Bracts under 7 mm. long; perianth not sharply contracted just above the ovary, though sometimes trigonously indented : Perianth markedly trigonously indented above the ovary: Pedicels 20-25 mm. long; perianth 35-40 mm. long; teeth on leaves usually under 10 mm. apart Pedicels under 15 mm. long; perianth not over 35 mm. long; teeth on leaves 10 mm. or more apart: Leaves about 8 cm. wide Leaves up to 17 cm. wide ...... Perianth not markedly trigonously indented above the ovary: Leaves 6-8 cm. wide: Perianth 20-25 mm. long; leaves not spotted Perianth 28-30 mm. long: Spots on leaves few or none; perianth 30 mm. long . Spots on leaves many; perianth 28 mm. long . Leaves 16-18 cm. wide, not spotted .... Group 8 Perianth over 32 mm. long: Perianth markedly trigonously indented above the ovary the outer segments united for 60% of their length Perianth not markedly trigonously indented above the ovary, the outer segments free to the base .... Perianth under 30 mm. long: Pedicels 15 mm., perianth 28 mm. long .... Pedicels under 11 mm. perianth under 26 mm. long: Leaves about 16 cm. wide, the teeth 20-25 mm. apart . Leaves under 14 cm. wide, the teeth up to 16 mm. apart: Leaves many-spotted, c. 13 cm. wide Leaves with few or no spots, under 12 cm. wide: Leaves 7-8 cm. wide; pedicels 8-10 mm. long . Leaves 9-11 cm. wide; pedicels 5-6 mm. long . . Group 9 Bracts over 10 mm. long: Perianth 27-28 mm. long, the outer segments free for half their length Perianth 35 mm. long, the outer segments united for 70% of their length ........ Bracts under 7 mm. long: Pedicels 8 or more mm. long: Perianth 40-45 mm. long ...... Perianth 33-35 mm. long: Stems slender, up to 6 m. tall, free from dead leaves; outer perianth segments united for 1/3 of their length Stems up to 1 m. tall, leafy; outer perianth segments united for 2/3 of their length Pedicels under 7 mm. long; perianth under 26 mm. long: Perianth over 21 mm. long: Leaves with few or many spots all over them Leaves unspotted, or with a few spots at the base only . Perianth 16-20 mm. long 36 b lateritia Engl. var. kitaliensis (Reynol.) Reynolds 38 chabaudii Schonl. 39 bukobana Reynolds 42 rivae Bak. 98 classenii Reynolds 100 mubendiensis Christian 112 b sciiweinfurthi Bak. var. labworana Reynolds 116 marsabitensis Verdoom & Christian 42 rivae Bak. 1 14 macleayi Reynolds 101 wilsonii Reynolds 84 calidophila Reynolds 104 tweediae Christian 98 classenii Reynolds 115 microdonta Ch\o\. 60 otallensis Bak. 70 compacta Reynolds 73 Christiana Reynolds 125 /jn/Zy/ Reynolds 140 flexifolia Christian 87 turkanensis Christian 115 microdonta Chiow. 97 ruspolaina Bak. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Group 10 Perianth under 30 mm. long; bracts under 4 mm. long: Leaves under 9 cm. wide: Perianth 20-25 mm. long, 7 mm. wide across the ovary 98 dassenii Reynolds Perianth 27 mm. long, 5-6 mm. wide across the ovary 1 33 yavellana Reynolds Leaves 16-18 cm. wide Perianth over 30 mm. long: Leaves up to 17 cm. wide Leaves imder 1 1 cm. wide : Teeth on leaves 1-2 mm. long Teeth on leaves 3 mm. or more long : Leaves 5 cm. wide ....... Leaves 6 or more cm. wide : Perianth 40 mm. long Perianth under 36 mm. long : Pedicels 18 mm., bracts 7 mm. long Pedicels 14-15 mm., bracts 4-5 mm. long: Stems stffly erect, simple, or with 1 or 2 branches from the base, up to 4 mm. tall Stems erect or spreading, forming clumps 1 -2 m. tall 116 marsabitensis'VerdooT n & Christian 42 rivae Bak. 140 flexifoUa Christian 147 ngobitensis Reynolds 148 nyeriensis Christian 142 mbaiensis Rendle 126 volkensii Engl. 143 dawei Berger Group 11 Pedicels under 13 mm. long: Perianth over 30 itun. long 42 rivae Bak. Perianth under 28 mm. long: Leaves over 12 cm. wide: Teeth on leaves 20-25 mm. apart, leaves unspotted . 84 calidophila Reynolds Teeth on leaves 10-15 mm. apart; leaves spotted near the base 104 tweediae Christian Leaves under 12 cm. wide: Teeth on leaves up to 5 mm. long .... 98 dassenii Reynolds Teeth on leaves 1-2 mm. long . . . . . 115 microdonta Ch\o\ . Pedicels over 14 mm. long: Leaves over 7 cm. wide : Perianth c. 35 mm. long 126 volkensii Engl. Perianth 28-30 mm. long: Bracts 1-nerved 101 wilsonii Reynolds Bracts 3-nerved 141 boscawenii Christian Leaves under 7 cm. wide 147 ngobitensis Reynolds (Received 21st March, 1967) NOTE : Since this paper was received for publication we have heard with deep regret of the death of Dr. G.W. Reynolds. 1 1 ,iv'A \ ^ AX ^ ’An •’'* l4t.-;''<:j^^ >H\Vf.U' V 'a -.I'.lr:* a;'j>} mV»sA''>''Vvi. ■/ ' «■ ' ''VA ' v<' fill".' >* I ..<-(?! 'iwrt, ,'.t :■ r',r'.:!.'. Itl^',>'■'l<'wivvv^^^ fii')' - vuiV.'Vw e'H Ih»u flXfii.u'iH y.i frtm V ,f Jsm' rijrtRJ'm '/itvo MJ . -c,:a!itol ’ittfft TS , ■' •■()■•...,. iv.'-^ v ;f.. r:ii«x»j’,iiKi^ AT •J5»wj> lUanh-iH ■, ^; , ■.'■ . UI.X ,1 r. .vbw .tnu U Vw/to.} .giRjt w«» !Tv-l7iOvrfO< no riVjftT :a'(ol -»ViW (j/./n no . ' , »ww.i .. , |.^v ' ' .I'.?.' 'r>t}w .ffTTi laoffi to I? , ?,fi» I ,:f««r4 W fllniti’W'f ■ -f/im’V .fm.m 4tl xilowliyil 1 • . HU'I ' !>■ AWJrtCi > 'f ■; I • i t ■ fi I i,i’,. oj r|)i ,> 1.4 ’jd< mo’ii> .. K' ,v . ’ilT I ' ""'.I* 'I.,., 1. . "aH' ■Ml, I i ifh y{,\\u'HyA ‘f- . iff > ■ S' ) '‘‘i < n )Vva>'(!^,vf;'y’V, 'll ' . ri • '■ ,'JS Vi rf)i ’ll r.'(!( 'y - H ■ ■ : ijrui .iXlfA ^.i -oliW) •irfot .mm Of. i»w f|»o« »o‘l .fT(m 8S WOXIU • ,ms £1. T[3''0 «»■'?»>> > .ioi»7 ^ 0£ savaat tio rfJasT, wiiny»t -•'Wl -'Rtr' mrtr'S flo' ftteT I , . ..■• asfid ^ i^w ..>’■ ' . . WIW90'*’ ■V' :mh')} f l lovo siaoito*! • ’ ' • •.7,. 'unrd .-.xfim .'X .•v,.dU»X!i:(-'9^ , - ■' "s^ ■ ' rtiOKha^ ■ ■• '■ '• . ■ ’■' . b;ymn-! ' •■’ ■'.• . ' fewiwififfe ( •xf'i .'AjvwVf VaI£ ... ...;.jii A'-.f. j.-„ . ,' .„-'.iy(iA’jiWtiq I'jt lw.rj.-n itiif*' ' 'tTitif’d Marine Botany of the Kenya Coast Page 75 MARINE BOTANY OF THE KENYA COAST 1. A First List of Kenya Marine Algae By Wm. Edwyn Isaac University College, Nairobi INTRODUCTION Although there are scattered records of Kenya marine algae in the literature there is only one previous paper known to the author which is specifically on Kenya Marine Algae. This paper was published by Gerloff in 1960 and apart from recording Lyngbya majuscula and citing previous records of Cyanophyta, it deals with a collec- tion of Chlorophyceae only (Gerloff, 1960). The list of species from the East African Herbarium, mostly collected by Greenway and Rawlins, and identified by Gerloff is given in an appendix. At present Gerloff ’s list is published without comment. It is intended that the following list of Kenya marine algae will be followed by papers dealing more fully with the species and their distribution as well as with additional species. The plants listed below were all collected by the author and he is responsible for the identifications unless otherwise stated. Only a minimum of references to literature is included in this paper. ACKNOWLEDGEMENTS I wish to thank the Rockefeller Foundation whose financial support has made this work possible. Thanks are also due to the Botany Dept., British Museum (Natural History) for providing me with working space and facilities in 1966 and for their kindness in providing photo-copies of relevant literature. To the Rijksherbarium, Leiden and to the Kew Herbarium for working space and access to collections also in 1966, thanks are due. To Dr. William Randolph Taylor, I am indebted for identifying material of Turbinaria and describing two new species from the Kenya coast; to Margaret Steentoft for naming material of Galaxaura, and Actinotrichia and to Sophie Ducker for identifying Chlorodesmis caespitosa J. Ag. I also wish to thank Dr. A. B. Cribb for help in identifying some species, confirming my identifications of other species and for helpful discussions while we were both working at the Natural History Museum. Lastly I wish to thank my wife for her constant help during my collecting excur- sions and in dealing with the herbarium material collected. CHLOROPHYCEAE Ulotrichales Ulva fasciata Del. U. lactuca L. U. reticulata Forsk. U. rigida C. Ag. Page 76 Marine Botany of the Kenya Coast Cladophorales Chaetomorpha crassa (C. Ag.) Kuetz. Cladophora patentiramea (Mont.) Kuetz. forma longiarticulata Reinb. (Also known as C. socialis Kuetz) C. pwlifera (Roth) Kuetz. Siphonales Avrainvillea amadelpha (Mont.) A. & E. S. Gepp f. montagneana A. & E. S. Gepp A. amadelpha f. submersa A. & E. S. Gepp A. erecta (Berkeley.) A. & E. S. Gepp A. lacerata J. G. Ag. f. typica A . lacerata var. robustior A. & E. S. Gepp A. obscura J. Ag. Bryopsis hypnoides Lamour. B. pennata Lamour. B. plumosa (Huds.) C. Ag. Caulerpa cupressoides (West) C. Ag. C. fastigiata Mont. C. lanuginosa J. Ag. C. lentillifera J. Ag. C. mexicana (Sond.) J. Ag. C. pickeringii Harv. & Bail. C. racemosa (Forsk.) J. Ag. This is a very variable species. The following varieties have been distinguished on the Kenya coast: — clavifera (Turn.) Web. v. B. clavifera f. macrophysa Web. v. B. gracilis (Zan.) Web. v. B. laetevirens (Mont.) Web. v. B. occidentalis (J. Ag.) Boergs. peltata (Lamour.) Eubank turbinata (J. Ag.) Eubank uvifera (Turn.) Web. v. B. C. scalpelliformis (R. Br.) C. Ag. including var. denticulata (Decne.) Web. v. B. C. serrulata (Forsk.) J. Ag. emend. Boergs. C. sertular hides (Gmel.) Howe C. taxifolia (Vahl) C. Ag. C. verticillata J. Ag. C. vickersiae Boergs. C. webbiana Mont. f. tomentella (Harv.) Web. v. B. Chlorodesmis caespitosa J. Ag. det. Mrs. S. Ducker. For validation of name see Ducker, Williams & Lance (1965). J. E. Afr. nat. Hist. Soc. Vol XXVI No. 2 {114) Page 77 Codium arabicum Kuetz. C. capitatum Silva C. duthiae Silva C. lucasii Setch. subsp. capense Silva C. prostratim Levr. Halimeda discoidea Decne H. incrassata (Ell.) Lamour. H. macroloba Decne H. opuntia (L.) Lamour. H. stuposa Taylor H. tuna (Ell. & Soland.) Lamour., including f. platydisca (Decne) Barton Udotea flabellum (Ell. & Soland.) Howe U. indica A. & E. S. Gepp U. orientalis A. & E. S. Gepp Siphonocladales Anadyomene stellata (Wulf.) C. Ag. A. wrightii Gray Boergesenia forbesii (Harv.) Feldm. Boodlea composita (Harv.) Brand Chamaedoris auriculata Boergs. Cladophoropsis membranacea (C. Ag.) Boergs. Dictyosphaeria cavernosa (Forsk.) Boergs. D. intermedia Web. v. B. D. versluysii Web. v. B., sens. lat. (Valet, 1966) Ernodesmis verticillata (Kuetz.) Boergs. Nereodictyon imitans Gerloff Struvea anastomosans (Harv.) Piccone S. ramosa Dickie Valonia aegagropila C. Ag. V. macrophysa Kuetz. V. utricularis (Roth) C. Ag. V. ventricosa J. Ag. Valoniopsis pachynema (Mart.) Boergs. Neomeris vanbosseae Howe Dasycladales Marine Botany of the Kenya Coast PHAEOPHYTA Ectocarpales Giffordia mitchellae (Harv.) Hamel Dictyotales Dictyopteris delicatula Lamour. Dictyota bartayresiam Lamour. D. ciliolata Kuetz. D. dichotoma (Huds.) Lamour. D. divaricata Lamour. D. pardalis Kuetz. D. pardalis f. pseudohamata Cribb (1954, p. 22, Plate 3, fig. 10), det. Cribb. Padina commersonii Bory P. gymnospora (Kuetz.) Vickers P. tetrastromatica Hauck Pocockiella variegata (Lamour.) Papenf. Stoechospermum marginatum (C. Ag.) Kuetz. Stypopodium zonale (Lamour.) Papenf. Punctariales Colpomenia sinnosa (Roth) Derb. & Sol. Hydroclathrus clathratus (Bory) Howe Fucales Cystophyllum trinode (Forsk.) J. Ag. “Renamed Cystoseira trinodis (Forsk.) C. Ag. by Papenfuss (1967).” Cystoseira myrica (Gmel.) J. Ag. Hormophysa articulata Kuetz. H. triquetra (L.) Kuetz. (H. articulata is often included in this species). Sargassum duplicatum J. Ag. S. latifolium (Turn.) C. Ag. Turbinaria The taxa of Turbinaria listed below have been named or confirmed by Wm. Randolph Taylor. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 79 Turbinaria condensata Send. T. conoides (J. Ag.) Kuetz. T. crateriformis Taylor, 1966 T. kenyaensis Taylor, 1966 T. murrayana Barton T. ormta J. Ag. including f. ecoronata Taylor, f. evesiculosa (Barton) Taylor RHODOPHYTA Nemalionales Actinotrichia fragilis (Forsk.) Boergs., det. Margaret Steentoft Asparagopsis taxiformis (Del.) Coll. & Herv. Galaxaum squalida Kjellm., det. Margaret Steentoft G. subverticillata Kjellm., det. Margaret Steentoft Liagora caenomyce Decne L. ceranoides Lamour. var. leprosa (J. Ag.) Yam. L. ceranoides var. pulverulenta (C. Ag.) Yam. L. mauritiana Boergs. Typically epiphytic on Cymodocea ciliata Ehrenb. ex Asch. L. valida Harv. Scinaia indica Boergs. Cast up. Gelidiales Gelidiella acerosa (Forsk.) Feldm. & Hamel Gelidium caespitosum Kylin G. crinale (Turn.) Lamour. G. pusillum (Stackh.) Le Jolis Pterocladia capiUacea (Gmel.) Bornet & Thur. Cryptonemiales A. Corallinaceae. Amphiroa beauvoisii Lamour., confirmed by A. B. Cribb Corallina mauritiana Boergs. Jania capiUacea Harv., confirmed by A. B. Cribb B. Non-Corallinaceae. Chondrococcus harveyi (J. Ag.) De Toni. On further study this may prove to be a smaller, more delicate form of C. hornemanni. C. hornemanni Kylin Page 80 Marine Botany of the Kenya Coast Grateloupia filicina (Wulf.) C. Ag. Halymenia venusta Boergs. Gigartinales Eucheuma denticulatum (N. L. Burman) Coll. & Herv. (Dixon, 1962) E. horridum (Harv.) J. Ag. E. serra J. Ag. E. striatum Schmitz Gelidiopsis scoparia (Mont. & Mill.) Schmitz Gracilaria cacalia (J. Ag.) Dawson G. crassa (Harv.) J. Ag. G. edulis (Gmel.) Silva G. millardetii (Mont.) J. Ag. G. verrucosa (Huds.) Papenf. Cast up. Hypnea cervicornis J. Ag. H. cornuta (Lamour.) J. Ag. H. harveyi Kuetz. H. musciformis (Wulf.) Lamour. H. rosea Papenf. H. valentiae (Turn.) Mont. Boergesen follows Hauck in that he includes H. cornuta in this species. (Boergesen, 1943, p. 59.) Rhodymeniales Botryocladia chiajeana (Meneghini) Kylin B. kuckuckii (Web. v. B.) Yam. & Tanaka B. leptopoda (J. Ag.) Kylin, det. A. B. Cribb Champia compressa Harv. C. globulifera Boergs. C. parvula (C. Ag.) Harv. C. vieillardii Kuetz. De Toni (1900:561) included this species in C. compressa Harv, but the author concurs with Dawson (1954) in upholding Kuetzing’s species. Coelarthrum boergesenii Web. v. B. Ceramiales Acanthophora delilei Lamour. Included, at least in part, by some authors in A. muscoides A. muscoides (L.) Bory A. spicifera (Vahl) Boergs. Amanzia glomerata C. Ag. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 81 Centroceras clavulatum (C. Ag.) Mont. Chondria armata (Kuetz.) Okam. Also known as Rhodomela crassicaulis Harv. Okamura (1907 ; p. 69, pi. 16) maintains that on a structural basis this plant should be regarded as a Chondria. Dasya scoparia Harv. Dictyurus purpurascens Bory Digenea simplex (Wulf.) C. Ag. Laurencia flexilis Setch. det. A. B. Cribb L. obtusa (Hud.) Lamour., including var. natalensis (Kylin) Boergs. L. papillosa (Forsk.) Grev. L. perforata (Bory) Mont., det. A. B. Cribb Leveillea jungermannioides (Mart. & Herv.) Harv. Martensia elegans Herv. Neurymenia fraxinifoHa (Mert.) J. Ag. Polysiphonia ferulacea Suhr. ex J. Ag. P. variegata (C. Ag.) Zan. Spyridia cupressina Kuetz. S. filamentosa (Wulf.) Harv. S. insignis J. Ag. Vanvoorstia spectabilis Harv. LITERATURE CITED Boergesen, F., 1943. “Some Marine Algae from Mauritius, III, Part 2”, Kgl. Danske Vidensk. Selskkab, Biolog. Meddels, 19, Nr. 1. Cribb, A. B., 1954. “Records of Marine Algae from South-Eastern Queensland 1”, Univ. Queensland Papers, Dept. Bot., 3: 16-37. Dawson, E. Yale., 1954. “Marine Plants in the Vicinity of the Institute Oceanographique de Nha Trang, Viet Nam”, Pacific Sci., 8: 373-469. De Toni, J. B., 1900. Sylloge algaruni . . . 4(2): 387-776. Published by the author, Patavii. Dixon, Peter, S., 1962. “Taxonomic and Nomenclatural Notes on the Floridaeae III”, Bot. Not., 115, relevant pages, 205-251. Ducker, Sopeue, C., Williams, W. T. and Lance, G. N., 1965. “Numerical Classification of the Pacific forms of Chlowdesmis (Chlorophyta)”, Australian J. Bot., 13 : 489-499. Gerloff, Johannes, 1960. “Meeresalgen aus Kenya, 1, Cyanophyta und Chlorophyta”, Willdenowia, 2, Pt. 4: 604-627. Okamura, K., 1907-1909. “leones of Japanese Algae”, 1. Tokyo. Papenfuss, George F. & Jenson, James B., 1967, “The Morphology, Taxonomy and Nomenclature of Cystophyllum trinode (Forsk.) J. Ag. and Cystoseira myrica (Gmcl). C. Ag. (Fucales: Cystoseiraceae) Blumea, 15 pp. 17-24. Taylor, Wm. Randolph, 1966. “Notes on Indo-Pacific Turbinarias”, Hydrobiologia, 28, Fasc. 1 : 91-100. Valet, Gabriel, 1966. “Les Dictyosphaeria du groupe versluysii (Siphonocladales, Valoniacees)”, Phycologia, 5 : 256-260. Marine Botany of the Kenya Coast APPENDIX GERLOFF’S LIST OF SPECIES The species are arranged in order as they appear in his paper but omitting family headings. Cyanophyta : Lyngbya majuscula Harv. ex Gomont Chlorophyta; Ulvales : Ulva lactuca L. U. rigida (C. Ag.) Thuret U. reticulata Forsk. Cladophorales : Chaetomorpha crassa (Ag.) Kuetz. Siphonocladales : Valonia aegagropila Ag. Dictyosphaeria cavernosa (Forsk.) Boergs. D. intermedia Web. v. B. Boergesenia forbesii (Harv.) Feldm. Chamaedoris delphinii (Hariot) Feldm. & Boergs Ernodesmis verticillata (Kuetz.) Boergs. Boodlea composita (Harv.) Brand Spongocladia vaucheriaeformis Aresch. Nereodictyon imitans Gerloff spec. nov. Anadyomene stellata (Wulf.) J. Ag. Dasycladales : Neoineris vanbosseae Howe Caulerpales : Caulerpa webbiana Mont. f. tomentella Web. v. B. C. scalpelliformis (R. Br.) Web. v. B. C. sertularioides (Gmel.) Howe C. serrulata (Forsk.) J. Ag. emend. Boergs. C. racemosa Forsk. var. clavifera (Turn.) Web. v. B. C. racemosa Forsk. var. clavifera f. macrophysa (Kuetz.) Web. v. B. C. racemosa var. uvifera (Turn.) Web. v. B. J. E. Afr. nat. Hist. Soc. Vol. XXVI No. 2 (114) Page 83 Avrainvillea erecta (Berkeley) Gepp A. obscura (Ag.) J. Ag. Chlorodesmis hildebrandtii A. & E. S. Gepp Udotea indica A. & E. S. Gepp U. orientalis A. & E. S. Gepp Halimeda tuna (Ell. & Soland.) Lamour. f. tuna f. platydisca (Decne.) Barton H. cuneata Hering H. opuntia (L.) Lamour. H. opuntia f. triloba (Decne.) Barton H. macroloba Decne. Codium duthieae Silva C. dwarkense Boergs. (Received 1st July, 1967) /. E. Afr. mt. Hist. Soc. Vol. XXVI No. 2 {114) Page 85 NOTES ON A COLLECTION OF AMPHIBIANS FROM ETHIOPIA By Emil K. Urban {Department of Biology, Faculty of Science, Haile Sellassie / University, Addis Ababa, Ethiopia) From 1962 through 1965 various members of the Faculty of Science of Haile Sellassie I University accumulated a series of amphibian specimens from Ethiopia. Except for duplicate specimens which are housed in the Biology Department of Haile Sellassie I University, the collection has been deposited in the Field Museum of Natural History, Chicago, Illinois. Because the species, distribution, and ecology of amphibians of Ethiopia are poorly known, an obvious need exists for papers, large and small, that deal with Ethiopian frogs, toads, and caecilians. Hence, I present below specific comments on this collection. I am indeed grateful to Drs. R. F. Inger and H. Mark of the Field Museum of Natural History for identifying the amphibians in this collection. Moreover, I should like to acknowledge the various collectors mentioned in the text below. Since the literature on amphibian classification is severely limited in the library of Haile Sellassie I University, classification in this paper is based on that supplied by Inger and Marx and where appropriate supplemented by that found in Loveridge (1957, Check list of the reptiles and amphibians of East Africa, Bull. Mus. Comp. ZooL, Vol. 117, No. 2: 1-362). Xenopus clivii Peracca. On 17 October, 1964, 1 collected adults as well as tadpoles in a pond about 200 meters from Gaferssa Reservoir, 18 kilometers west of Addis Ababa along the Ambo Road. The seasonal pond, approximately 2,585 meters in elevation, results from overflow of one stream which supplies the reservoir during the long rains (June-October). Juncus, Cyperus, and Polygonum are the major plants associated with the pond. Bufo regularis Reuss. This ubiquitous toad was obtained in several months of the year at different localities ranging from approximately 535 to 2,585 meters in elevation. On 6 January, 1963, E. W. Beals collected specimens 70 kilometers east of Bati at Waranzo watercourse next to a small pool under a bridge at about kilometer 485 on the Asaab Road (11°20'N, 40°42'E), about 535 meters high; on 25 April, 1965, in a sandy open scrub of Salvadora and Tamarix with a ground cover of Zygo- phyllum some 20 kilometers north of Era Gota, about 1,185 meters in elevation, and on 16 July, 1963, in an Acacia woodland 4 kilometers south of Meki in the Rift Valley, about 1,640 meters high. R. Baxter collected this species along the west shore of Lake Langano, about 1,585 meters high, on 2 November, 1962, while J. M. Prosser and P. Chen caught it along the shore of Lake Awasa, about 1,680 meters high, on 3-4 May, 1963. I collected specimens in semi-desert bush, 48 kilo- meters north of Awash Station, about 830 meters high, on 14 November, 1964; in my garden in Addis Ababa, about 2,420 meters high, on 29 November, 1964; and along the shores of the pond some 20() meters from Gaferssa Reservoir, 18 kilometers west of Addis Ababa, about 2,585 meters high, on 10 October, 1964. Page 86 Amphibians from Ethiopia Bufo dodsoni Boulenger. Beals collected this toad at the Loggia watercourse, which contained Tamarix, 1 kilometers east of Tandaho along the Assab Road (11°44'N, 40°58'E), about 380 meters in elevation, on 6 January, 1963. Leptopelis gramineus (Boulenger). I collected one transforming larva in the pond some 200 meters from Gaferssa Reservoir, 18 kilometers west of Addis Ababa on the Ambo Road, about 2,585 meters high, on 17 October, 1964. Rana abyssinica Peters. Beals obtained a large series of this species from the Danakil desert at Waranzo watercourse next to and in a small pool under a bridge on the Assab Road (70 kilometers east of Bati, at kilometer 485; (1 1°20'N, 40°42'E), about 535 meters high, on 6 January, 1963. Baxter collected this frog along the west shore of Lake Langano, about 1,585 meters high, on 2 November, 1962. Rana angolensis Bocage. I obtained this frog along small streams which cross the Blue Nile Road, 30 kilometers north of Addis Ababa, about 2,600 meters in elevation, on 18 October, 1964; most were captured 1-3 meters above the stream along ploughed banks which contained a new growth of unidentified weeds. Also I collected this species in the pond some 200 meters from Gaferssa Reservoir, 18 kilometers west of Addis Ababa, about 2,585 meters high, on 17 October, 1964. Rana cooperi Parker. Beals obtained this species in Addis Ababa in the garden of Haile Sellassie I University’s Faculty of Science, about 2,420 meters high, on 24 June, 1963. I obtained specimens along small streams which cross the Blue Nile Road, 30 kilometers north of Addis Ababa, about 2,600 meters high, on 18 October, 1964. As with R. angolensis, I found them along ploughed stream banks that had new growth of unidentified weeds. Rana mascareniensis Dumeril & Bibron. Beals and I obtained a large series of this species in the bullrushes halfway down the east shore of Lake Abaya one kilometer south of the ferry landing, about 1,240 meters high, on 28 November, 1964. Earlier, on 17 July, 1963, Beals collected this species in a marsh on the west shore of Lake Zwai, about 1,625 meters in elevation and opposite kilometer 159 on the Addis Ababa-Shashamanne-Awasa Road. Rana ornata (Peters). Specimens of this frog were collected by Beals in a small muddy pond, 4 by 3 meters in size and associated with sen Acacia-Euphorbia woodland, 9 kilometers south of Lake Langano village at kilometer 218 on the Addis Ababa- Shashamanne-Awasa Road, about 1,750 meters high, on 6 July, 1963; by R. B. Wood and Prosser along the shore of Lake Awasa, about 1,680 meters in elevation, on 3 May, 1963; and by Prosser in her garden in Addis Ababa, about 2,420 meters high, on 15 January, 1965. {Received 20th March, 1967) J. E. Afr. not. Hist. Soc. Vol. XXVI No. 2 (114) Page 87 NATURE NOTES LESSER SPOTTED EAGLE ON MIGRATION On November 1 1th this year, while travelling from Mombasa to Nairobi I noticed, a mile or so west of Sultan Hamud, numbers of large birds circling the sky. I thought at first that they were vultures. On approaching the area I noticed numbers of these birds standing on the road and feeding on the verges and adjacent fields. I stopped noting to my surprise that the birds were eagles about the size of the Tauny but a darker brown spotted on the coverts. These, I thought were Wahlbergs, but seeing them in such numbers made me doubtful, never having seen eagles in E. Africa in greater numbers than 4 and 6. The country is open savannah, there had been considerable rain over the previous days and the time was midday. Over an area of 3-4 miles of road there must have been many thousands of these birds, on the wing and on the ground ; they were eating insects, probably termites and undisturbed by passing traffic. In 38 years in E. Africa I have never seen the like so I consulted my friend Leslie Brown who thought they might have been the Lesser Spotted Eagle. On referring to English books and illus- trations I concur; this must have been a vast migration, probably southwards. 23rd December, 1966. Roger V. Bowles, M.D. POSSIBLE OCCURRENCE OF THE WHALE-HEADED STORK IN ETHIOPIA The Whale-headed Stork {Balaeniceps rex Gould) is known from The Sudan, Chad, Congo, Ruanda, Uganda, Tanzania, and Zambia (Burton & Benson, 1961, N. Rhodesia J., 4: 416-418). In the Sudan it is resident in the swamps of the Bahr el Ghazal and upper White Nile and is found as far north as Kodok on the Nile (Cave & Macdonald, 1955, Birds of The Sudan, p. 61). The Whale-headed Stork thus far has not been recorded in Ethiopia (Urban & Brown, 1967, Preliminary Checklist of the Birds of Ethiopia, MS). Recently Mr. Thomas Mattanovich of Addis Ababa informed me that he twice saw this species in western Ethiopia along the Baro River in 1961 or 1962. While travelling in a canoe, he noted the stork standing along side the river approximately 20 kilometers west of Gambela in the rainy season (May-October) and again about 100 kilometers west of Gambela in the dry season (November- April). In both instances he saw the storks standing in grass with marshy areas nearby. The possible occurrence of the Whale-headed Stork in western Ethiopia is not unexpected. The Baro River, with numerous papyrus swamps along its shores, flows westward into the Sobat River which in turn enters the White Nile several kilometers south of Kodok, the northern limit of the stork’s distribution in eastern Sudan. The species is apparently not uncommon in the Bahr el Ghazal and the Sudd, swampy regions of the Nile about 100 kilometers west of where the Sobat flows into the Nile (Burton & Benson, op. cit.). It is reasonable to assume that the Whale-headed Stork occurs in suitable habitats from the Bahr el Ghazal and the Sudd of the White Nile to the Sobat River in The Sudan and the Baro River in Ethiopia. Mattanovich, who has spent several years in this biologically unknown area of western Ethiopia, is reputed to be an excellent observer. Only recently he reported and Mr. John Blower, Senior Game Warden of the Imperial Ethiopian Government Wild Life Conservation Department, collected from the Baro River-Gambela area the first Nile Lechwe [Onotragus megaceros (Fitzinger)] from Ethiopia. Although Page 88 Book Reviews Mattanovich’s identification of the stork is probably correct, the occurrence of this species in Ethiopia needs further confirmation. Expeditions to determine the status of this species in western Ethiopia certainly are now needed. 11 th February, 1967 Emil K. Urban Addis Ababa BOOK REVIEWS FLORA OF TROPICAL EAST AFRICA Berberidaceae by R. M. Polhill, 4 pages, 1 fig., Sh. 1/25; Tamaricaceae by D. R. Hunt, 4 pages, 1 fig., Sh. 1/25; Pittosporaceae by G. Cufodontis, 15 pages, 2 fig., Sh. 2/25; Tecophilaeaceae by S. Carter, 8 pages, 3 fig., Sh. 1/70; Juncaceae by S. Carter, 12 pages, 3 fig., Sh. Ij-. Published under the authority of the Minister for Overseas Development, London, and obtainable from the Government Printers, Box 9124, Dar es Salaam, Box 53, Entebbe and Box 30128, Nairobi, or from the Govt. Bookshop, Box 569, London, S.E.l. These 5 parts which appeared on 11 November, 1966, fully maintain the high standard expected of this Flora. They bring the total number of families now publi- shed up to 46, including in all 221 genera and 885 species. It is pleasing to see that the recently adopted policy of appointing a junior botanist at Kew (Mr. Polhill), who is less preoccupied with administration than are his seniors, as one of the editors has led to a speed-up in publication. He informs us that 3 further families, all large {Caesalpinioideae, Cucurbitaceae and Sapotaceae), are now with the printers, while 8 smaller families are also ready to go to press. It is also most pleasing to report the high standard of efficiency of the Nairobi Government Printer’s sales department which now supplies without delay over the counter, or by post, any part of this flora, thus obviating the delay and increased postal charges involved in ordering from London. The first three families now reviewed are woody Dicotyledons. Berberis (1 sp.) and Pittosporum (6 spp.) are found in upland woodland or forest, Tamarix (2 spp.) by water-courses in the drier parts of Kenya and Tanzania. The other two families are herbaceous Monocotyledons. Juncus (5 spp.) and the related Luzula (3 spp.) occur in wet places in the uplands and mountains, while Cyanastrum (3 spp.) and Walleria (1 sp.), both in the Tecophilaeaceae, are found in various vegetation-types below 1,500 m. in Tanzania. J. B. G. “east AFRICAN WEEDS AND THEIR CONTROL” By G. W. IVENS 243 pages, 113 line drawings, O.U.P., Nairobi, 1967, Sh. 6/50. Some 220 common East African Weeds are described in 110 gioups of one or a few related species each of which is illustrated. These groups are arranged under the headings “Water Weeds”, “Grasses and nutgrasses” (Cyperaceae), “Woody Weeds” and “Herbaceous Weeds”. A concise botanical description of each weed or small group of related weeds is followed by accounts of its distribution and economic significance in East Africa and of the known methods of control. In addition there is a list and chemical classification of the herbicides referred to, a list of those species known, or thought, to be poisonous, a bibliography, a glossary and an index. Although primarily meant for the practical agriculturalist, this work will also be of value to everyone interested in plants in East Africa, especially on account of the pictures. J. B. G. JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. XXVI No. 3 (115) JUNE 1967 Revised Catalogue of the African Sphingidae (Lepidoptera) with Descriptions of the East African Species by R. H. CARCASSON (Published 30/6/68) Price Shs. 30/- EAST AFRICA NATURAL HISTORY SOCIETY NOTICE TO CONTRIBUTORS Contributions. The Committee is pleased to consider contributions on natural history for publica- tion in the Journal on the understanding that these are not also being offered, wholly or partially, to any other Journal. They should be addressed to the Secretary, P.O. Box 4486, Nairobi. Typescript, Articles should be typed on one side of the paper, in double spacing and with wide margins. Illustrations. These should be in a form suitable for reproduction. The Editor cannot be expected to re-draw. Line drawing should be in Indian ink on Bristol board or thick white paper. Reproduction will be better if they are drawn larger than it is intended that they should appear. An indication of the degree of reduction is advisable. Photographs should be printed on glossy paper and a better reproduction is achieved from prints slightly darker in tone than normal. Nomenclature. Where a recent standard work for the area is available (e.g. Praed and Grant for birds) the names given there (both English and scientific) should be used. Initial capitals should be used for specific English names, e.g. Pied Wagtail and small initial letters for group names, e.g. wagtails. Scientific names must always be underlined and followed by the naming authority when mentioned for the first time. Where an English name is used, it must, on first mention be accompanied by the scientific name. References. These are usually abbreviated in the text and listed more fully in alphabetical order of authors at the end of the article. For example, in the text book reference might be (Pinhey 1956 : p. 20). At the bottom of the contribution: Jackson, F. J., 1938. Birds of Kenya and Uganda. Pinhey, E. C. G., 1956. The Emperor Moths of Eastern Africa. J.E. Afr. Nat. Hist. Soc. XXIII No. 1, (98). With short articles it may not be worth making a list of references at the end, but the whole reference in the most abbreviated comprehensible form should then be inserted in the text. Reprints. Provided that they order at the time of submitting their articles, authors may have up to 25 reprints free (other than Nature Notes). Additional copies can be supplied on payment. A list of available back numbers of the Journal and reprints can be obtained from the Librarian, National Museum, P.O. Box 658, Nairobi. JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. XXVI No. 3 (115) JUNE 1967 Revised Catalogue of the African Sphingidae (Lepidoptera) with Descriptions of the East African Species by R. H. CARCASSON (Published 30/6/68) Price Shs. 30/- EAST AFRICA NATURAL HISTORY SOCIETY President: DR. A. D. Q. AGNEW Vice-President: DR. M. J. COE Executive Committee: G. C. BACKHURST, ESQ. C. J. BEECHER, ESQ. R. H. CARCASSON, ESQ. MRS. D. S. FLEMING DR. J. B. FOSTER J. S. KARMALI, ESQ. P. M. OLINDO, ESQ. MISS J. R. OSSENT B. T. PARSONS, ESQ. Hon. Editor: DR. P. J. GREENWAY Hon. Treasurer: P. G. ARCHER, ESQ. Hon. Secretary: MRS. F. NG’WENO All correspondence in connection with this Journal should be addressed to : The Secretary, East Africa Natural History Society, P.O. Box 4486, Nairobi, Kenya. Page 1 REVISED CATALOGUE OF THE AFRICAN SPHINGIDAE (LEPIDOPTERA) with Descriptions of the East African species by R. H. Carcasson. {National Museum, Nairobi) INTRODUCTION The present catalogue consists of a taxonomic revision of the Sphingid fauna of the Ethiopian Region, with descriptions of a number of new genera, species and subspecies as well as detailed descriptions and records of all the species known to occur in Ethiopia, Somalia, Socotra, the Sudan, Kenya, Uganda and Tanzania. For reasons of economy the genitalia and early stages have only been described and figured if previously undescribed, or if of special importance. Illustrations of all adult moths belonging to new species or subspecies and of all species not previously figured have been appended, as well as of many others of special interest. Details of illustrations of African Sphingids in other publications are given in the index. A more natural classification than any available hitherto has been proposed, based mainly on a more complete knowledge and understanding of the genital armatures of both sexes and on the early stages. CLASSIFICATION AND NOMENCLATURE In 1903, when Rothschild and Jordan published their classical revision of the world Sphingidae, they recognised a total of 722 species, 180 of which were African. 35 years later the family had risen to over 1 ,000 known species and in the present work 260 African species are recognised, 160 of which are known to occur in eastern Africa. The majority of the Sphingidae are tropical insects, comparatively few occurring in the temperate regions ; in North America the rather poor resident fauna is boosted by many species of migrants from tropical America; in the Old World only a few species migrate regularly from the tropics, owing to the presence of formidable east- west barriers such as the Sahara, the Mediterranean and the Himalayas. Before 1903 the classification of the Sphingidae had been based on purely superficial characters with the result that many unrelated species were lumped together and many closely related species were placed in widely separated genera. Rothschild and Jordan were the first to adopt a natural, phylogenetic classification in their monumental revision of the family. They used such characters as structure of Page 2 the antennae, spination of the legs and abdomen, structure of the palpi, pilifer and feet, and examined the genitalia of a large number of species. Earlier classification had been based mainly on wing venation and shape, which are characters of very little value in the Sphingidae. Rothschild and Jordan’s classification of 1903 has been accepted, with a few minor changes, by all subsequent authors and is followed in this work. Rothschild and Jordan were pioneers in the practice of dissecting and examining the genital armatures and their rather crude methods (dry preparation, which is particularly unsatisfactory with the females), did not enable them to take these structures into sufficient account when defining genera and following up their relationships. In the Philampelini and Choerocampini this did not matter much as the genitalial structure of these insects is extremely uniform, but it did lead to a number of mis- conceptions in the more advanced Ambulicini. It has been one of the aims of this work to elaborate and perfect the classification of Rothschild and Jordan and to render it more compatible with a critical assessment of the genital armatures of both sexes. The names proposed by R. & J. for supra-generic taxa have been adopted in this work, although older and more appropriate names are available in some cases; however, R. 8c J’s. names have been in use for over 60 years and it would serve no useful purpose to change them. The only names which have had to be rejected are Sesiinae and Sesiicae, because the type genus of these groups {Sesia Fabricius) is not a Sphingid. The other groups have been given the terminations recommended in the “Inter- national Code of Zoological Nomenclature” (1961). Rothschild and Jordan split the family into two “Divisions”, three subfamilies and seven tribes. As the term “Division” has no status in nomenclature, it has not been used in this work, the term “subfamily”, which is the next in rank after family, being used instead ; all subsequent taxa above the rank of genus have been demoted by one step. Thus R. & J’s. “tribes” become sub- tribes, a taxon which is not recognised by the Code of Zoological Nomenclature, but which is nevertheless useful. The alternative would have been to raise the two so-called “divisions” to family rank, a procedure which cannot be justified in view of the great compactness and isolation of the Sphingidae. Family SPHINGIDAE Samouelle, 1819 Type species (designated by R. 8c J., 1903): Smerinthus ocellatus (L.) SubfamUy ASEMANOPHORINAE (R. & J.) 1903 Division ASEMANOPHORAE R. & J. Subfamily ASEMANOPHORAE Janse 1932 No type species or genus has ever been designated for this Subfamily, but since it includes the type species of the family and is therefore the typical or nominate sub- family, its type species must be Smerinthus oceallatus L. (Europe). The correct name of the subfamily should be Smerinthinae Butler 1877. The type species was formerly Page 3 placed in the genus Sphinx L. All the species of this subfamily lack the patch of sensory hairs on the inner surface of the first palpal segment. There is a tendency for the reduction of numerous characters which are well developed in the Semanophorinae, such as the frenulum, which however, is present in all the African species, the proximal pair of tibial spurs, the pulvillus and the paronychium and the modified scales of the male genitalia. On the other hand the genital armatures of the males tend to be much more complex and varied than in the Semanophorinae, the valve being frequently bilobed and provided with a harpe of varying structure and the uncus being sometimes bilobed and even trifid. The larvae are always cylindrical and do not have the fifth segment enlarged, as in some members of the Semanophorinae. Tribe AMBULICINI (Butler) 1877 Subfamily Ambulicinae Butler 1877 Subfamily Ambulicinae R. & J. 1903 Tribe Ambulicini Janse 1932 Type species (designated by R. & J., \903) Protambulyx strigilis L. (America). Smerinthus ocellatus L., which is the type of the family and of the subfamily is included in this tribe and should therefore be its type species. The correct name of the tribe should be Smerinthini Butler 1877. One of the largest groups in Africa. Proboscis usually reduced, often rudimentary, never longer than abdomen. All the African species are strictly nocturnal and none are known to feed on flowers. The females are usually larger than the males, broader winged, comparatively inactive and seldom attracted by light. Antennae never hooked, terminating in a short, densely scaled segment, usually more or less fasciculate, occasionally pectinate in the males (Ceridia, Xenosphingia). Lower margin of eye frequently ciliated. Male genitalia very complex and varied, modified scales small or absent. Outline of wings irregular in many genera. Venation very uniform. Vein 6 of forewing may arise at the end of the discoidal cell, or from a short stalk common to 7 and 8. Veins 6 and 7 of the hindwing may arise at the same point, or have a common stalk of varying length. The larvae are cylindrical, covered in small granules or tubercles, sometimes in short branching or dentate spines. {Lophostethus, Acanthosphinx, Rhadinopasa). Pupae generally naked and subterranean, without free proboscis case. The world distribution of the Ambulicini is as follows: the figures for the African genera and species are based on this work, those for the exotic species and genera on Rothschild and Jordan’s Revision of 1903 and are about 80% of the present figure. This applies to the figures given for the other groups of Sphingida also. Indo-Australian Ethiopian Palaearctic American Total Genera 20 31 9 10 60 Species 59 90 16 27 183 26 genera and 55 species occur in Eastern Africa. Page 4 Tribe ACHERONTHNI (Butler) 1877 Subfamily Acherontiinae Butler 1877 Subfamily Acherontiinae R. & J. 1903 Tribe Acherontiini Janse 1932 Type species Acherontia atropos L. Very similar to previous tribe. Last antennal segment very long, usually hooked. Proboscis short to very long, only exceptionally rudimentary. Outline of wings always entire. Venation very uniform, as in previous tribe. Hindwing cell usually smaller. Sexual dimorphism very slight in most species. Genital armature much more uniform than in previous tribe, simpler; valve always entire, harpe simple, aedeagus unarmed; cornuti present in a few species; modified scales larger, more conspicuous when present. Most species dull coloured, highly cryptic in appearance ; hw not visible when at rest. Females more active and more readily collected than in the Ambulicini. Some species are migratory and very widely distributed. Larvae cylindrical, smooth or granulose, somtimes pubescent, with round heads. Pupae very often with proboscis case prominent, or separated from body. 5 larval instars. Subtribe Acherontiae Tribe Acherontiicae R. & J. 1903 Type species Acherontia atropos L. This group is characterised by having a hollow protected by a dorsal hair tuft on the inner surface of the second palpal segment. World distribution of the Acherontiae. Indo- Australian Ethiopian Palaearctic American Total Genera 3 4 2 1 5 Species 7 6 2 1 13 There are three widely distributed and one endemic species in Eastern Africa, two of which (Herse convolvuli L. and Acherontia atropos L.) are not confined to Africa. Subtribe Sphinges Tribe Sphingini Grote & Robison 1865 Tribe Sphingicae R. & J. 1903 Type species Sphinx ligustris L., Europe. This group includes all the other African species of this tribe. The inner surface of the second palpal segment is not hollowed and there is no hair tuft. Antennae hooked. World distribution of the Sphinges. Indo-Australian Ethiopian Palaearctic American Total Genera 7 13 2 16 36 Species 10 24 9 85 127 9 genera and 16 species occur in Eastern Africa. Subtribe Sphinguli Tribe Sphingulicae R. & J. 1903 Type species Sphingulus mus Standinger, Siberia. A small group of species which link the Acherontiini with the Ambulicini. The second palpal segment is not hollowed, the antennae are not hooked, the proboscis is short, but not rudimentary. The early stages are similar to those of the Ambulicini. All the species are Asiatic or Australian and R. «& J. list 7 genera and 11 species. Subfamily SEMANOPHORINAE Janse 1932 Division SEMANOPHORAE R. & J. 1903 Type species Pholus satellitia Drury, America This subfamily may be readily separated from the Asemanophorinae by the presence of a patch of sensory hairs at the base of the inner surface of the first palpal segment. Generally speaking, this group is more primitive and uniform than the previous subfamily and the tendency towards structural reduction and loss less evident. All the species appear to feed on flowers and all have a well developed proboscis, though never of excessive length as in some species of the Acherontiini. The male genital armature is of a simple and very uniform pattern in all the Chaerocampini and in most of the Philampelini, consisting of an undivided uncus and gnathos, entire valve and ventro-proximal harpe, usually with a few large modified scales on the outer surface of the valve. The only exceptions to this rule are the genus Sphingo- naepiopsis which, although Philampeline in general appearance and structure, has a genital armature which recalls that of the Semanophorinae and the tribe Dilophono- tini which has symmetrically divided uncus and gnathos in the more primitive Ame- rican species, becoming asymmetrical by reduction of the right hand lobes and valves in some of the more advanced genera such as Cephonodes. The larvae of the Semano- phorinae are seldom granular or pilose and never spiny, and have round heads. In the Philampelini and in the Choerocampini the head is often small and the 5th segment enlarged. The pupa usually has a projecting, keeled proboscis case in the Choerocampini and Philampelini, but not in the Dilophonotini. The separation of the three tribes which constitute the subfamily Semanophorinae is not clear cut and there are several genera which share characters common to two or more tribes. Sphingonaepiopsis has already been discussed and should perhaps be placed near the junction of the Asemanophorinae and Semanophorinae. Macroglossum and Leucostrophus, although treated as an extreme development of the Philampelini by R. & J. (1903) and thus widely separated from Cephonodes and from the American Aellopos, the Holoarctic Haemorrhagia and the Oriental Sataspes share many characters with these genera. It is possible that some of the characters shared by these widely separated genera are related to the needs of a diurnal existence and have evolved independently in the two tribes. Page 6 Tribe DILOPHONOTINI Burmeister 1875 Subfamily Sesiinae R. & J. 1903 Tribe Sesiini Janse 1932 Type species : Haemorrhagia (formerly Sesia) tityus L. Europe designated by R. & J. 1 903. Mainly an American group which may be separated from the other tribes of the Semanophorinae by the structure of the male genitalia, which usually have divided uncus and gnathos, or are asymmetrical owing to the loss or reduction of one lobe of the uncus and gnathos and of one valve. All the more primitive species which have symmetrical genitalia and approach the Asemanophorinae in several respects were included by R. & J. (1903) in the subtribe Dilophonotae. Many of the more advanced species are diurnal and some are generalised mimics of such bees as Bombus and Xylocopa; these latter species belong mainly to the Holoarctic genus Haemorrhagia and to the Old World genera Cephonodes and Sataspes, in which the majority of the scales are discarded after emergence, leaving a hyaline wing membrane. As seen earlier, these day flying Dilophonotini have a great number of characters in common with several diurnal genera of Philampelini with which they may be more closely related than R. & J.’s arrangement would suggest. Subtribe Dilophonotae Tribe Dilophonoticae R. & J. 1903 Type species Erinnis ello L., America. A group of primitive species with crested head, double crested thorax, uniseriate abdominal spines and symmetrical genitalia. These insects have some characters in common with the Acherontiini, and can in fact be regarded as a link between the two subfamilies of the Sphingidae. There are 4 genera and 19 species, all American. Subtribe Aellopodes Tribe Sesiicae R. & J. 1903 Type species Haemorrhagia (formerly Sesia) tityus L. (Europe.) Although the type species is not in Sesia Fabricius 1775, the tribe was named by R. & J. after that name which was regarded as the oldest generic name in the group. Since then Sesia has been applied to a genus in the family Sesiidae (formerly Aegeridae) and can no longer be used in the Sphingidae. The next available name for Sesia (Sphingidae) is Aellopos Hubner 1 822, and the name of the subtribe has been derived from it. An alternative name would have been Haemorrhagiae, but this has not been used, as there are doubts about the vahdity of Haemorrhagia Grote «fe Robinson 1865, Haemaris Dalman 1816 having been rejected by Grote & Robinson and later by Rothschild and Jordan on rather dubious grounds. All the more advanced species of the tribe including the highly aberrant day- flying genera with strongly asymmetrical genitalia were placed in this subtribe by Page 7 R. & J. They share a number of characters with the Philampelini and are in some res- pects intermediate between them and the Dilophonotae and Acherontiini. World distribution of the Aellopodes Indo- Australian Ethiopian Palaearctic American Genera 3 1 1 18 Species 14 5 9 76 There is only one species in Eastern Africa. Tribe PHILAMPELINI Janse 1932 Family Philampelidae Burmeister 1878 Subfamily Philampelinae R. & J. 1903 Type species: Pholus satellitia Drury, America. A very variable and ill-defined group. Most of its characters have been described under the heading Semanophorinae. The Philampelini may be readily separated from the Dilophonotini by the undivided uncus and gnathos and by the palpi, which are never laterally angular. The genitalia are never asymmetrical and resemble in in most species those of the Choerocampini. The antennae are very variable and so is the spination of tergites and legs. All the species have frenula and retinacula and there is little reduction of the paronychia and pulvilli. The larvae usually taper towards the head which is small and round, but the enlargement of the 5th segment, so typical of the Choerocampini is not so conspicuous. The Philampelini may be readily separated from the Choerocampini by the structure of the bristly appendage (pilifer) at the base of the proboscis, which is single-lobed, never bilobed as in the Choerocampini. The scaleless area at the inner surface of the palpi is absent or reduced, never as large as in the Choerocampini. All the species are believed to feed on flowers; some are strictly crepuscular and seldom attracted to light. Others are diurnal (Macroglossum and allied genera) and have a number of important structural characters in common with the Dilophonotini. Subtribe Philampeli Tribe Philampelicae R. & J. 1903 Type species Pholus satellitia Drury, America. This subtribe comprises the more primitive species of the group. Apical segment of antenna long; abdominal spines in single series; modified scales small and nume- rous. There is one American genus with 19 species and a second genus with a single species from Hawai. Subtribe Nepheles Tribe Nephelicae R. & J. 1903 Type species: Nephele funebris Fabricius, Africa. Total 20 101 Page 8 A large group, very varied in appearance, with numerous rather specialised side branches. The diurnal genera bear a striking resemblance to the more specialised Aellopodes. Genitalia of simple structure, very uniform in both sexes and of the same structure as those of the Choerocampini. Modified scales present in most genera, large and few. The genus Sphingonaepiopsis is rather aberrant and its male genitalia resemble those of the Asemanophorinae. World distribution of the Nepheles. Indo-A ustralian Ethiopian Palaearctic American Total Genera 18 15 9 9 40 Species 106 91 12 16 221 10 genera and 61 species have been recorded from Eastern Africa. Tribe CHOEROCAMPINI Grote & Robinson 1865 Subfamily Choerocampinae R. & J. 1903 Type species: Pergesa elpenor L. Europe A compact, homogeneous group, very closely related to the Philampelini from which they can be separated by the structure of the pilifer and of the palpi. Pilifer bilobed, the apical part bearing short or vestigial bristles, the proximal part, long bristles. Inner surface and second palpal segment usually naked. Antennae stout, straight, terminally hooked; last segment more of less elongated, armed with 6 or more bristles and dorsally scaleless. Proboscis always well developed. Eyes large, ciliated in some species. Wings elongated, usually with acuminate apices and regular margins. Retinacula and frenula always present. Tibiae always unarmed, paronychia and pulvilli always present. Abdomen elongated and conical. Abdominal spines usually in more than one series at the posterior margin of each somite. Venation uni- form and similar to that of the Philampelini. Male genitalia extremely uniform, of the same pattern as in Nephele, with large modified scales nearly always present on the external surface of the valve. Female genitalia as in Nephele, always with a single elongated longitudinal signum and colliculum. Larva similar to that of the Philampelini, but anterior tapering more pronounced, 4th and 5th segments more enlarged, horn occasionally lacking in final instar. The larvae are usually decorated with large eye spots, especially on the fifth segment. When they are alarmed they withdraw the head and anterior part of the thorax into the enlarged segments and display the prominent eye spots ; they thus acquire a re- markable resemblance to a small snake, an attitude which must be very effective in deterring predators. Pupa with a compressed or keeled proboscis case. Medium to large, fast flying, streamlined moths, many of which are crepuscular. The majority of the species feed on flowers and have a wide distribution. Many species have strong migratory tenden- cies. The more unspecialised species of the group are American. Page 9 World distribution of the Choerocampini. Indo- Australian Ethiopian Palaearctic American Total Genera 1 9 6 3 14 Species 52 43 18 56 157 In Eastern Africa there are 8 genera and 30 species. In the following revision the more advanced genera within each tribe are treated first, followed by decreasingly specialised species and ending with the most generalised. Family trees showing the relationship of the genera within each tribe would be desirable, but cannot be attempted without detailed study of all the world genera. Wing pattern of. a Sphingid Page 10 Abbreviations used in the following text Abbreviations of periodicals and of other publications according to the World List (Third Edition, 1900-1950); abbreviations of works published before 1900 according to Rothschild & Jordan, “A Revision of the family Sphingidae”, Novit.zool. 9 suppL; 1903. Other abbreviations B - BE - BM = CM = comb. nov. — D.C. = ESB = Fw = gen. nov. = GM = H - Ho = Hw = K = L. - L = MB = MC = NM = nom. nov. === P = PM = R = RJ = R. & J. - S = SM - sp. nov. = ssp. nov. = stat. nov. = syn. nov. = K. Brown collection (Uganda). Berio collection, Genoa, Italy. British Museum collection. Carnegie Museum, Pittsburgh, new combination, discoidal cell. E.S. Brown collection, Muguga, Kenya. forewing. new genus. Museo Civico di Storia Naturale, Genoa, Italy. according to Hering in Seitz, “Macrolepidoptera of the World” 14, 1930. C. Howard collection, Rhodesia, hindwing. Kawanda Research Station collection, Uganda. Linnaeus. Lock collection. Queen Elizabeth National Park, Uganda. Berlin Museum collection. McCleery collection, Lindi, Tanzania. National Museum collection, Nairobi, new name. according to Pinhey in “Hawk Moths of Central and Southern Africa”, 1962. Paris Museum collection. Robertson collection, Ilonga, Tanzania, according to Rothschild & Jordan, 1903. Rothschild & Jordan. Sevastopulo collection. Stoneham Museum collection, Kitale. new species. new subspecies. new status. new synonym. NOTE The colours mentioned in the descriptions of new species are according to Ridgway, “Color Standards and Color Nomenclature”, Washington, 1912. ovipositor lobes Sphingid head; labial palpi removed uncus Subfamily ASEMANOPHORINAE Tribe AMBULICINI Page 11 ACANTHOSPHINX Aurivillius 1891 Ent. Tidskr. 12: 228; type species 1879. Head wide, proboscis short, antennae very thick in S- Differs from Polyptychus in the absence of spines on the hindtibia and in the much longer tibial spurs. ACANTHOSPHINX GUESSFELDTI (Dewitz) 1879. (I; 2) Ambulyx guessfeldtiHey/itz Mitt, munch, ent. Ver. 3: 27 (Chinchoxo, Angola, (?). 1891 Acanthosphinxguessfeldtivzx.gigas AuviyiWius Ent. Tidskr. \2'. 219 {Czva&roons, $ ?)syn. nov. 1930 Acanthosphinx guessfeldti cothina Tams Ann. Mag. nat. Hist. (10) 6: 167 (N. Rhodesia, c?) syn. nov. The Type is an abnormally small specimen and var. gigas Aur. is really the normal form. Cothina Tams is based on an aberration. S : fw, 57-65 mm ; apex acute, margin entire, slightly concave, costa convex near apex. Ground colour of wings and body dark purplish grey. Vertex and dorsum of thorax blackish. Two blackish spots before base of abdomen. A small blackish spot on first 3-4 abdominal tergites. Fw with numerous irregular dark lines, and an olive green distal margin. Hw. darker than fw, outer margin olive green. The olive green margins fade to ochreous yellow in old specimens. $: larger, wings more rounded. Fw: 70 mm. HABITAT AND RANGE Forests from Sierra Leone to the Congo, Angola, Zambia, Malawi, Tanzania and Uganda. EAST AFRICAN RECORDS UGANDA NM : Kampala B : Nakawa K : Kawanda Jinja Bukoba, Kigoma BM : Uluguru, Ukami R : Mbimba, near Mbeya Ho : Amani LYCOSPHINGIA R. & J. 1903 Novit. zool. 9, suppl. : 264; type species Smerinthus hamatusTlemtz 1879. Differs from Polyptychus in having a single pair of hindtibial spurs, a very weak proboscis and a strongly produced apex to the fw. Genitalia as in Polyptychus. LYCOSPHINGIA HAMATA (Dewitz) 1879. (I; 1) Smerinthus hanatus Dewitz Mitt, munch, ent. Ver. 3: 28 (Chinchoxo, Angola, (?). »2«ora. Valve elongated, sole-shaped, with a longitudinal ridge near dorsal margin in which 6 large modified scales are set. Harpe smooth, fairly long and slender, apically upcurved and provided with a thin vertical blade connecting the apex to the base. Aedeagus long, slender, slightly curved, armed apically with a long narrow spinose longitudinal ridge which is prolonged into a slender reflexed hook with a few apical spines. Vesica armed with numerous very long spicules. Page 87 FEMALE AND EARLY STAGES: UnknOWn. HOLOTYPE c?: Tanganyika, Mlingano, 20-1-1964, 1. A. D. Robertson, taken at light; to be deposited in British Museum (Natural History). This species is known from the Holotype only, and is dedicated to its discoverer. TEMNORA NEPHELE Clark 1922. Proc. New Engl. zool. C/. 8: 12 (Cameroons) Spanish Guinea to the Cameroons. TEMNORA ELISABETHAE Hering 1930. (V; 21) Rev. Zool. Bot. afr. 17: 406 (Belgian Congo, $). Sexes alike. Fw. 21-24 mm. Apex fairly acute, termen concave at vein 4. Fw, head and body dark grey. Fw with a large dark brown triangular spot at middle of costa, with apex prolonged into a slightly curved line to tornus. A smaller dark brown spot at costa, before apex. Numerous irregular dark narrow transverse lines. Hw uniform very dark brown. Underside bright pinkish buff lined with fuscous ; basal f of fw fuscous. Abdominal tufts pink. $ GENITALIA : postvaginal plate roughly triangular. Anti- vaginal plate not sclerotised, ostium very wide. Colliculum very wide at ostium, tapering, directed to the left; ductus short, narrow, pitted. Bursa very long and narrow, pitted and pleated. 1 large rounded basal signum armed with strong spines. EARLY stages: unknown. HABITAT AND RANGE Forests from the Congo to Uganda. EAST AFRICAN RECORDS UGANDA NM : Kamengo, Entebbe, Mabira, Katera, Makerere, Budongo. BM : Kampala. B : Nakawa. TEMNORA SUBAPICALIS R. & J. 1903. (V; 22) Noyit. zool. 9 suppl. : 572 (Kikuyu Escarpment, Kenya, $). S : fw. 24 mm. Very similar in size, shape and markings to the previous species, but ground colour of both wings rich cinnamon brown, not grey. Diagonal bar from tomus to costa, straight narrow, well defined, not dilated at costa ; a dark dorsal line from head to apex of abdomen. Fore and mid- tibiae whitish, abdominal tufts of S cinnamon. GENITALIA : Very similar to T. griseata; uncus short, evenly downcurved, apically blut. Gnathos almost as long as uncus, apically bisinuate. Subdorsal ridge and modified scales present on valve. Harpe smooth, sharply upcurved, apically pointed. Spinose subapical plate of aedeagus very much broader than in T. griseata. Vesica armed with a single stout apical spine. $: fw. 26 mm. Wings broader and more rounded; diagonal bar dilated distally to form a large dark brown patch with diffuse outer edge. GENITALIA : post-vaginal plate reduced to two narrow lateral arms flanking the ostium. Ante-vaginal plate large, smooth, with anterior margin rounded and slightly sinuate. Colliculum wide and very short. Ductus slender and moderately long. Bursa small, ovoid and rugose, densely covered by small papillae. Basal signum transverse, consisting of two lateral sclerotised irregularly spinose plates con- nected by a lightly sclerotised bridge. Subapical signum a small irregular dentate plate. Page 88 HABITAT AND RANGE Apparently very rare; highland forest in central Kenya and also in Rhodesia, according to Pinhey. EAST AFRICAN RECORDS KENYA NM : Limuru. BM : Kikuyu Escarpment (Type), Uplands. TEMNORA lAPYGOIDES (Holland) 1889. Ocyton iapygoides Holland Trans. Amer. ent. Soc. 16: 60 (Benita, Gabon, c?). 1891 Ocyton preussi Karsch Ent. Nachr. 17: 292 (Barombi, Cameroons). 1894 Pterogon clement si Rothschild Novit. zool. 1: 69 (Sierra Leone). Ssp. iapygoides. (XIV; 12) Sexes alike. Fw. 18-20 mm. Very similar to T. subapicalis R. & J., but rather smaller, more narrow winged. Diagonal bar broader, margins more diffuse, hw a little paler, underside brighter. Abdominal tufts of c? long, very pale. $ genitalia: post- vaginal plate an irregular transverse ridge with deep parallel transverse folds. Ante-vaginal plate slightly sclerotised at the sides only. Colliculum long, tapering towards ductus. Ductus rather short, straight, sclerotised and fluted. Bursa small and rounded, pitted and pleated, sclerotised at base. Signum a small apical spinose rounded plate. HABITAT AND RANGE Forests from Sierra Leone to the Congo and Uganda. EAST AFRICAN RECORDS KENYA NM : Kericho. UGANDA NM : Kawanda, Katera, Fort Portal, Budongo, Bwamba, Kayonza. BM : Sesse Islands, Kampala, Kamengo. Ssp. pernix Kernbach 1962. (V; 25) Mittel. Deutsch. ent. Ges. Jahrg. 21, 4: 53 (Southern Rhodesia). A race of doubtful validity. Slightly larger (fw 20-22 mm.); ground colour brighter, markings heavier, better defined. Amani specimens have the colour and markings of topotypical specimens from Rho- desia, but are smaller. South Congo specimens have the same coloration as Uganda and West African specimens, but agree with pernix in size. range: Rhodesia, Zambia. Malawi and Tanzania to the Kenya Coast. EAST AFRICAN RECORDS KENYA S : Mombasa. TANZANIA NM : Amani. R : Mlingano, Ilonga. TEMNORA ERANGA (Holland) 1889. (V; 23— XIV; 10) Ocyton eranga Holland Trans. Amer. ent. Soc. 16: 61 (Kangwe, Ogowe, 9)- 1931 Temnora heringi Gehlen Ent. Z. 45: 203 (Lake Tanganyika, S) syn. nov. Sexes alike. Fw 16-18 mm. Very similar to previous species, but smaller, ground colour much paler, huffish, markings much more clearly visible, hw with a subraarginal series of angular buff spots. Abdominal tufts of S apparently missing. I Page 89 $ genitalia: post-vaginal plate irregular, with ill-defined posterior margin and with 4 deep longi- tudinal folds. Ante-vaginal plate a small, regular crescent. Colliculum short, tapering abruptly at ductus. Ductus long and narrow. Bursa of moderate size, ovoid, pitted, but not pleated. Signum a narrow, transverse spiny plate at base of bursa. HABITAT AND RANGE Forests from Sierra Leone to the Congo and Uganda and west Kenya. EAST AFRICAN RECORDS KENYA NM : Kakamega. UGANDA NM : Katera, Kalinzu, Budongo. B : Nyabyeya. TEMNORA SCITULA (Holland) 1889. (V;24) Ocyton scitula Holland Trans. Amer. ent. Soc. 16: 60 (Benita, Gabon, S). Sexes alike. Fw. 19-23 mm. Wing margins less crenulated, more regular than previous species. Body brown above; crest very dark brown, a dark brown triangle with pale margins on each tegula. Fw brown, marked and much variegated with dark brown and with several creamy buff spots in the distal part of the wing. Hw reddish brown with a dark margin and a submarginal series of angular pinkish buff spots. Underside spotted with orange brown and cream ; a small ventral creamy spot on each abdominal stemite. Abdominal tufts of male buff. ? GENITALIA : Very similar to T. eranga, but ventro-posterior margin of colliculum very deeply incised at ostium. HABITAT AND RANGE Forests from Gambia to the Congo, Angola and Uganda ; an isolated but typical population in E. Tanzania. EAST AFRICAN RECORDS UGANDA NM : Mabira, Katera, Fort Portal, Kalinzu, Budongo, Bwamba, Kayonza. BM : Kamengo. B : Nyabyeya. TANZANIA NM : Amani, Mbimba. TEMNORA ENGIS Jordan 1933. Novit. zool. 38: 342 (4 1 ■ • ■ ,.r -n Page 67 RECENT BREEDING RECORDS OF STORKS IN EASTERN AFRICA By M. P. Kahl CONTENTS I. Introduction 67 II. Recent breeding records of storks in eastern Africa 67 III. Resume of breeding distribution and seasons 69 IV. Acknowledgements 71 V. Summary 72 VI. References 72 INTRODUCTION Between November 1963 and July 1967 I have been engaged in research on the comparative behaviour and ecology of storks (family Ciconiidae) in Kenya, Uganda, Tanzania, and Ethiopia. A number of previously unreported breeding localities were discovered during this study — some being found by me personally and others being reported to me by various correspondents. This paper is an attempt to compile the recent records, in the hope of giving a better understanding of the breeding distribu- tion and breeding seasons of this family of birds. Storks frequently use the same nesting site for many years, if undisturbed and given the proper climatic conditions, and so perhaps these records will prove useful to future workers studying these birds. I have also included a few comments on the ecological factors involved in the timing of the breeding seasons, as a possible basis for further research along these lines. RECENT BREEDING RECORDS OF STORKS IN EASTERN AFRICA No. Date of Species Location Long. Lat. Nests First Eggs Authority Yellow-billed Stork {Ibis ibis) KENYA, Kisumu (lOkrn SE) 34°47'E 0°10'S 100+ mid-Mar 63 (1) 50-65 early-Mar 64 (2) 25 mid-Mar 65 (2) 50-60 early-Apr 66 (2) 50 late-Mar 67 (2) TANZANIA, Chagana, River Wembere33°55'E 4°30'S 2000+ Jan-Feb 62 (1) Lake Manyara Nat. Park 35°50'E 3'’25'S 150 May 62 (3) Page 68 Open-billed Stork {Anastomus KENYA, lamelligerus) Kisumu (10km SE) 34°47'E OHIO'S 20 + late-Mar 63 (1) 70 early-Mar 64 (2) Garsen (8km S) 40'’10'E 2°20'S 100 + June 56 (4) TANZANIA, Chagana, River Wembere 33°55'E 4°30'S 5000 + Jan-Feb 62 (1) UGANDA, Ishasha, Q.E. Nat. Park 29°40'E 0°35'S 8 mid-Dec 63 (2) Abdim’s Stork {Spheuorhynchus KENYA, abdimii) Busia, Sio River 34°08'E 0°20'N 1 mid-Jan 64 (2) 1 late-Jan 65 (2) 3 late-Jan 66 (5) Kisumu (town centre) 34°45'E 0°07'S 1 Feb or Mar 66 (6) Lokitaung (20km NW) 35°30'E 4°30'N 12 early- Jun 61 (7) ETHIOPIA, Lake Shala 38°30'E 7°30'N 25-30 early-Apr 66 (8) 45 early-Apr 67 (2) Gambella, Baro River 34°00'E 8°10'N ? late-Mar 66 (9) ? Apr 67 (10) Woolly-necked Stork {Dissoura UGANDA, episcopus) Murchison Falls Nat. Pk. 3r35'E 2°15'N 1 Nov-Dec 62 (11) Hoima (20km W) 31°10'E 1°30'N 1 Feb or Mar 64 (12) 1 early Feb 65 (2) Saddle-billed Stork KENYA, (Ephippiorhynchiis Rumuruti (8km NE) 36°35'E 0°20'N 1 about Dec 62 (1) senegalensis) Kisumu (5km E) 34°47'E 0°07'S 1 mid-Mar 66 (2) 1 early-Apr 67 (2) UGANDA, Kazinga Channel, Q.E.Nat. Park 30°00'E OHIO’S 1 mid-May — (13) Mweya, Q.E. Nat. Park 29°55'E OHIO'S 1 about Jun 65 (14) TANZANIA, Seronera River, Serengeti 34°50'E 2°25'S 1 late-Feb 62 (13) Marabou Stork KENYA, (Leptoptilos Kitale (1km W) 35°00'E r02'N ? Oct 62 (1) crumeniferus) 8 early-Oct 63 (1) 15 late-Sep 64 (2) 18 early-Oct 65 (2) 18 mid-Sep 66 (2) Makindu (7km SW) 37°45'E 2°20'S 15 early-Jul 65 (2) TANZANIA, Lake Manyara Nat. Park 35°50'E Page 69 3°25'S 8 Sep 58 (3) ? Jul-Aug 59 (3) UGANDA, Hoima (20km S) 3ri5'E 1°20'N 35 Nov-Dec 63 (2) 80 mid-Nov 66 (2) Chobe, Murchison Falls Nat. Park 32°10'E 2°15'N 5 late-Nov 66 (2) AUTHORITY: (1) M. E. W. North, in Hit. (27 Nov 63) <2) personal observations, M. P. Kahl (8) E. K. Urban, in litt. (8 Jun 66) (9) J. Blower, in litt. (22 Jul 66) (3) A. M. Morgan-Davies, in litt. (10 Aug 65) (10) E. K. Urban, in litt. (31 May 67) RESUME OF BREEDING DISTRIBUTION AND SEASONS YELLOW-BILLED STORK, Ibis ibis (Linne) A wide-spread and commonly occurring species, which probably breeds in many more localities than those listed in the table. At Kisumu, the only habitat where I have studied the ecology of this species in detail, reproductive activity appears to be triggered by heavy rainfall and the resultant flooding of the shallow marshes bordering Lake Victoria. This, in turn, seems corre- lated with food availability. The Yellow-billed Stork is primarily a fish-eater. During the dry season most fish are forced to leave the shallow marshes, which either dry up completely or become too de-oxygenated to support fish, and retreat to the deeper waters of Lake Victoria where they are unavailable to the birds. With the onset of the rains fish move up streams and spread out over the marshes of the Kano Plains to spawn. By nesting at this time the storks are assured a plentiful supply of fish for their young, provided the rains do not end prematurely. The closely related Wood Stork {Mycteria americana Linne) in Florida, U.S.A., also breeds when fish are seasonally abundant. However, owing to different ecological circumstances, this happens to occur there during the dry season {cf. Kahl, 1964). OPEN-BILLED STORK, Anastomus lamelligerus Temminck Another widely occurring species, likely to be found breeding in low-lying areas where their major food, the Pila snail, is plentiful. All the available records from (4) North, 1959 (5) Z. Mwanga, in litt. (25 Jan 66) (6) J. Blencowe, pers. comm. (7) Blencowe, 1962 (11) P. Allen, pers. comm. (12) N. L. Howarth, in litt. (25 Jun 64) (13) Pitman, 1965 (14) I. Ross, pers. comm. Page 70 eastern Africa seem to be from south of the equator, but certain areas in Uganda (such as the Lake Kyoga region) seem likely prospects for northern hemisphere breed- ing. In most areas breeding seems clearly related to flooding and seasonal abundance of snails for food. Pila snails are known to burrow into the mud and aestivate during drought, emerging once again after the area has been re-flooded. ABDIM’S STORK, Sphenorhynchus abdimii (Lichtenstein) This small, insect-eating stork is found, often in great numbers, as a passage mig- rant over much of eastern Africa. Its main breeding grounds lie in the broad belt of semi-arid scrub savanna between 6° and 15° North, extending from Ethiopia and the Sudan across the continent to northern Nigeria and Senegal. In these regions Abdim’s Storks nest commonly, sometimes abundantly, in trees, on rocks, and even on the roof-tops of village huts. In East Africa proper they seem restricted, as breeding birds, to a small area of western Kenya in the vicinity of Kisumu-Busia-Kakamega. North (1940) has also reported their breeding in this region. (The breeding colony reported near Loki- taung, in extreme northern Kenya, is probably best considered a marginal repre- sentative of the main breeding population in the Sudan). Throughout its major breeding range in the north, the Abdim’s Stork nests during the long summer rains. It is everywhere known by the local people as a “harbinger of the rains”, arriving about April or May as the rains begin and leaving once more for the south before the arrival of the dry weather. Their rainy-season breeding is pro- bably related to the “flush” of insect food available for the young at that season. In western Kenya breeding normally starts in January or February, between the short and long rains. This timing results in the young passing the period of greatest food demands and fledging during the long rains. WOOLLY-NECKED STORK, Dissoura episcopus (Boddaert) Rather rare everywhere but most often encountered in eastern Kenya and along the coast, or in the western Rift Valley in Uganda. Nesting is known only from western Uganda, but it may breed near Kilifi, Kenya (where the birds are often seen foraging on exposed reefs at low tide) or in the Tana River region. Little is known of the type of food given to the young. Probably it is similar to that eaten by the adults — i.e. frogs, fish, snakes — and would probably be easier to find during or just after the rains. Woolly-necked Storks in India generally nest just before and during the monsoon. SADDLE-BILLED STORK, Ephippiorhynchus senegalensis (Shaw) This spectacular species is found throughout the region in the vicinity of large marshes or lakes. It is a solitary nester, generally building its nest atop a large tree often some distance from water and usually remote from human habitation. The birds are usually shy at the nest and will often desert if subjected to much dis- turbance by man. Thus, it was surprising to discover an active nest within 50 feet of the main Kisumu-Kericho highway, directly over a cluster of African huts and within easy sight and sound of passing traffic and pedestrians. These birds were amazingly tame and afforded ideal subjects for study during the 1966 and 1967 seasons. In East Africa Saddle-bills breed mostly during the rains. At the Kisumu site the young were fed on large fish, procured in the marshes adjacent to Lake Victoria. As discussed under the Yellow-billed Stork, fish were abundantly available there only during the rainy season. In other areas out of the equatorial belt (e.g. Zambia, Rhodesia, and the Sudan) most breeding occurs at the end of the single, long rainy season (Pitman, 1965). This could be related to the concentrating of fish in drying marshes as the water recedes, but firm ecological evidence on this hypothesis is still lacking. MARABOU STORK, Leptoptilos crumeniferus (Lesson) A discussion of the Marabou’s breeding is presented in detail elsewhere (Kahl, 1966), so only a brief outline will be given here. Most nesting colonies are located near a dual source of food: (1) a supply of carrion to form the bulk of the diet, and (2) a source of fish, frogs, or other small vertebrates to fulfil the calcium requirements of the growing young. Although the Marabou is abundant in many places relatively few breeding sites are known, and these usually comprise only a few nests. Thus, it seems that the Marabou must breed only infrequently, on the average, and live a very long time. Most breeding is confined to the dry season, for a number of possible reasons: (1) carrion is more readily available during drought when animals are concentrated in a few areas of remaining surface water, (2) aquatic vertebrates become concen- trated and easier to catch in many habitats when water is receding, (3) some prey is caught at grass fires, which are common only during dry weather. ACKNOWLEDGEMENTS During 1964-65 these studies were supported by a post-doctoral fellowship from the U.S. National Science Foundation. During 1966-67 support was received from a research grant from the National Geographic Society, Washington, D.C. Additional financial aid came from the Chapman Fund of the American Museum of Natural History. I am grateful to Makerere College, where I was given working facilities during part of my study period, and to the following persons who assisted greatly by Page 72 supplying information : P. Allen, J. Blencowe, J. Blower, L. H. Brown, N. L. Howarth, A. M. Morgan-Davies, Z. Mwanga, M. E. W. North, C. R. S. Pitman, I. Ross, M. Turner, and E. K. Urban. SUMMARY A table of nesting records, most of them previously unpublished, is presented, showing breeding location, number of nests, and approximate date of egg-laying for the six species of storks (Ciconiidae) that nest in Kenya, Uganda, Tanzania, and Ethiopia. A brief analysis of ecological factors related to the timing of the breeding season shows that most storks in this region adapt their date of egg-laying so that the young are being fed during the season of most plentiful food supply. In the breeding localities considered here the Yellow- billed. Open-billed, Abdmin’s, and probably the Woolly- necked Storks rear their young during the rainy season. The Marabou Stork nests primarily during the dry season. The Saddle-billed Stork appears to nest during the rains in the equatorial belt of East Africa but at the end of the rains or during the dry season in Zambia, Rhodesia, and the Sudan. It is hoped that this brief survey of stork breeding will stimulate further contribu- tions to our knowledge of the breeding distribution of these birds and the ecological factors concerned in the regulation of their breeding seasons. REFERENCES Blencowe, E. J. 1962. Abdim’s Storks breeding on cliffs. /. E. Afr. Nat. Hist. Soc. 24: 64. Kahl, M. P. 1964. Food ecology of the Wood Stork {Mycteria americana) in Florida. Ecol. Monogr. 34:97-117. Kahl, M. P. 1966. A contribution to the ecology and reproductive biology of the Marabou Stork (Leptoptilos crumeniferus) in East Africa. J. of zool. 148: 289-311. North, M. E. W. 1940. Field notes on Abdim’s Stork in two Kenya provinces. /. E. Afr. & Uganda Nat. Hist. Soc. 15: 1-5. North, M. E. W. 1959. The great heronry of Garsen on the Tana River. J. E. Afr. Nat. Hist. Soc. 23: 159-160. Pitman, C. R. S. 1965. The nesting, eggs and young of the Saddle-bill Stork, Ephippiorhynchus senegalensis (Shaw). Bull. Brit. orn. Cl. 85: 70-80. {Received 11th July 1967). BOOK REVIEW The Ecology of the Alpine Zone of Mount Kenya by Malcolm James Coe, pub- lished in Monographiae Biologica, Vol. XVII, The Hague, 1967, pp. I-VIII and pp. 1-136, including a bibliography, an index occupying 6pp., 24 plates and 20 text figures. Price Dutch Guilders 25 or U.S. $ 6.95. This account is based on work by Dr. Coe when he was included in the Inter- national Geophysical Year Mount Kenya Expedition under the leadership of Professor I. S. Leupekine. All the field work was carried out on Mount Kenya between December 1957 and January 1963 and was used in part as a Ph.D. thesis of the University of London. Grants were made by the University College, Nairobi, to assist with porter and other expenses when visiting the mountain. The author opens with a brief account of the history of the exploration of the mountain followed by an outline of the physiography, geology and glacial geology, particularly where they are relevant to the study of the ecology of the Alpine Zone. The author’s Alpine Zone of Mount Kenya ranges from 11,480ft. (3,500m.) to the lower limit of the Nival Zone along an ill-defined boundary at 14,760ft. (4,50()m.) in the immediate vicinity of the glaciers and snow line. Excluding the ubiquitous species of tussock grasses he says the dominant genera of plants in the alpine zone are Senecio, Lobelia, Alchemilla and Helichrysum and he uses them as important community indicators and goes on to describe the plants in some detail. Under Senecio he lists three species of Dendrosenecio, four of Eusenecio, two Giant Lobelias, three species of Alchemilla and ten species of Helichrysum. The author divides the vegetation into zones and communities, the latter based on terrain and he recognizes five zones. 1. Ericaceous (Moorland) Zone with three communities, a. Damp boggy ground, b, Raised rocky ridges, or old weathered moraines and c. Stream courses. 2. The Alpine Zone. 3. The Lower Alpine Zone with two communities, a. Flat or gently sloping ground, usually wet and b. Weathered and eroded ridge tops. 4. The Upper Alpine Zone with four communities, a. Valley walls, b. Valley floors, c. Ridge tops, and d. Lakes and Tarns. 5. The Nival Zone. The Nival Zone was distinguished by the two Fries brothers, R. E. and Th. C. E. in their “Phytographical Researches on Mount Kenya and Mount Aberdare, British East Africa” (1948). They considered it a separate zone based on the disappearance of Senecio keniodendron, the lower limit of the zone is approximately 14,929ft. (4,550m.). O. Hedberg in his “Vegetation Belts of the E. African Mountains” (1951), does not distinguish a separate zone at this level but included it in the Alpine Zone. Coe considers the Nival Zone should be retained as a separate region on Mount Kenya as it is an area from which the glaciers have only recently retreated and upon which the earliest stages of vegetation colonisation are clearly discernable. He further states that “The plants forming the main association of this region are small in number, all stunted”, and invariably growing in sheltered positions and in the order in which they occur appears to be quite specific. The alpine climate is then dealt with; this covers temperature, rainfall, wind, climate and the alpine vegetation and he remarks that mountains on or near the Equator do not experience seasons like those mountains in the higher latitudes. Comparable figures of temperature measurements are given for the Great Himalay, Mount Kenya and Kilimanjaro at about the same altitude, 4,000m., in this order, the Mean Maximum being 12.0°C, 5.4°C and 5.17°C., the Mean Minimum -1.5°C., -3.6°C., -0.77°C. and the lowest -10°C., -6.7°C. -4°C. Comparable figures for rainfall, by means of a histograph for Kenya and a table for Kilimanjaro are given and the author draws a comparison to both mountains and says “These figures compare fairly closely with Mount Kenya with the annual precipitation rising sharply to a maximum at 8,000ft., and then falling off rapidly”. Further chapters cover the development and distribution of alpine soils including, 1, Soil generation, 2, The differentiation of alpine soil habitats, 3, The structure and chemistry of alpine soils. Next Colonisation in the Alpine Zone, divided into two groups 1 , Primary coloni- sation in the Alpine Zone, a. Colonisation of the Tyndall Glacier moraine, b. Coloni- sation of the Lewis Glacier moraine, c. Proglacial Tarns, and d. Dating moraine de- posits, then 2, Other Phases of Colonisation. The final chapter covers Biotic Factors in the Alpine Zone and this includes 1, Relations between animals, vegetation and habitats, 2, Herbivores and their relation to vegetation, covering a, Population size and control, and b. Breeding as a control factor. The insects found in the Alpine Zones of Kenya and Kilimanjaro are mentioned. For herbivores in the Alpine Zone cycle of animal-vegetation balance the author lists the Hyrax, Procavia johnstoni mackinderi, the Grooved-toothed Rat, Otomys orestes orestes and the common Duiker, Sylvicapra grimmia altivallis and the author gives the Hyrax habitat in tussock grassland as being rocky and one which has been sufficiently weathered and eroded for them to become established, then below the tussock grassland the Otomys Rats live in the burrows they excavate at the bases of the Giant Senecios or in tussocks. The small number of Duiker occupying valleys in Senecio keniondendron forest with a more or less dense ground cover of Alchemilla argyro- phylla. Mingling of the three communities only takes place when the Hyrax and Duiker leave their own habitats to go to water or when foraging in the Otomys habitat of the valley floor. Food preferences of these three herbivores is discussed and the following avian residents, of which there are six with their food preferences are listed. They are the Scarlet Tufted Malachite Sunbird, Hill Chat, Streaky Seed-eater, Alpine Swift, Black Duck, and the Slender-billed Chestnut Winged Starling. Their territories range from 10,000 to 16,000 ft. except for the Black Duck which is found through the forest and Alpine Zone. Strangely enough there only appears to be one reptile, the Alpine Meadow Lizard, Algyroides alleni. Other birds noted were Mackinders Owl, resident in the Alpine Zone, Augur Buzzard erratically resident, Verreaux Eagle, not common, observed breeding at 13,700 ft. in January 1958 and Lemmergyer resident in small numbers in the Alpine Zone. Finally there is a discussion and the book ends with a summary and conclusions. There are twenty photographs mostly of vegetation, two of Hyrax, and four of birds. The author has given a very interesting account of the ecology of the Alpine Zone of Mount Kenya, its climate, soils, plants, herbivores and birds and one gets an idea of the extremes of temperatures and radiation under which the vegetation and its inhabitants have to survive. This is a book which every ecologist with an interest in the ecology of the plants and animals of our East African mountains should have on his bookshelf. P. J.G. MYLES E. W. NORTH AN APPRECIATION Page 75 Although he had not been in good health for some time, the news of Myles North’s sudden death on 24th November 1967, at the age of 59 came as a great shock. The East African Natural History Society has lost one of its most prominent and active members and his absence will be felt by all those friends who shared his interest in the wealth of bird life East Africa has to offer. Myles was educated at Wellington and Sandhurst, with a view to making the army his career. However, after a brief trial, he decided that army life did not suit him and went to Cambridge. He graduated from Corpus Christi College and joined the Colonial Administration in Kenya. His colonial service followed its normal course, interrupted by the war years when he was a Civil Affairs officer in Abyssinia and Somalia. At the time of his retirement from the Administration he was Director of Co-operative Societies in Kenya. While members of the Natural History Society will remember Myles best for his leadership of rambles, and his lectures, the wider world of naturalists will remember him for his outstanding work on bird songs. In association with the Laboratory of Ornithology at Cornell University he was responsible for producing two remarkable long-playing records of East African bird songs. “Voices of African Birds” and “More Voices of African Birds” will always provide a fitting memorial to this remarkable personality. But having said all this, I feel that I have failed to convey the essential Myles, with his sociability, his kindness to embryo ornithologists, his delight in the company of other naturalists and his meticulous attention to detail in everything he did and planned. His outstanding characteristic was perhaps his readiness to share with all the knowledge he had acquired after years of patient and diligent study and his enthusiasm evoked a corresponding response. I first got to know Myles really well when he was working on the colony of Black- headed Herons breeding in the Nairobi Railway yards. This work made him very interested in photography and we had an arrangement by which he taught me about birds while I advised him on how to photograph them. I suspect that he was a better pupil than I was! And later, when my wife became interested in recording bird songs, Myles once again took infinite trouble to help and advise her. Those of us who knew and loved Myles will always feel a gap in our lives at his going and will always remember him with deep affection. J.S.K. tr/ JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. xxvn No. 2 (117) JULY 1968 CONTENTS Page The Elopoid and Clupeoid Fishes of East African Coastal Waters 77 By G. F. Losse Notes on the Birds Observed in the Vicinity of Tabora, Tanzania, with Special Reference to Breeding Data 117 By J. F. Reynolds Curculionidae (Weevils) of the Alpine Zone of Mount Kenya 141 By I. Jabbal and R. Harmsen Records of Parasitic Nematodes in Kenya 155 By G. D. Schmidt and A. G. Canaris Distribution and Host-Specificity of a Number of Fleas Collected in 157 South and Central Kenya By R. Harmsen and I. Jabbal Nature Note: Singing by the Chiflf-Chaff in Kenya 163 By C. W. Benson Book Reviews 163 Appreciation 167 (Published 31/5/69) Price Shs. 20/- EAST AFRICA NATURAL HISTORY SOCIETY NOTICE TO CONTRIBUTORS Contributions. The Conunittee is pleased to consider contributions on natural history for publication in the Journal on the understanding that these are not also being offered, wholly or partially, to any other Journal. They should >be addressed to the Secretary, P.O. Box 4486, Nairobi. Typescript. Articles should be typed on one side of the paper, in double spacing and with wide margins. Illustrations. These should be in a form suitable for reproduction. The Editor cannot be expected to re-draw. Line drawing should be in Indian ink on Bristol board or thick white paper. Reproduction will be better if they are drawn larger than it is intended than they should appear. An indication of the degree of reduction is advisable. Photographs should be printed on glossy paper and a better reproduction is achieved from prints slightly darker in tone than normal. Nomenclature. Where a recent standard work for the area is available (e.g. Praed and Grant for birds) the names given there (both English and scientific) should be used. Initial capitals should be used for specific English names, e.g. Pied Wagtail and small initial letters for group names, e.g. wagtails. Scientific names must always be underlined and followed by the naming authority when mentioned for the first time. Where an English name is used, it must, on first mention be accompanied by the scientific name. References. These are aisually abbreviated in the text and listed more fully in alpha- betical order of authors at the end of the article. For example, in the text book reference might be (Pinhey 1956: p. 20). At the bottom of the contribution: Jackson, F. J., 1938. Birds of Kenya and Uganda. Pinhey, E. C. G., 1956. The Emperor Moths of Eastern Africa. J.E. Afr. Nat. Hist. Soc. XXIII No. 1, (98). With short articles it may not be worth making a list of references at the end, but the whole reference in the most abbreviated comprehensible form should then be inserted in the text. Reprints. Provided that they order at the time of submitting their articles, authors may have up to 25 reprints free (other than Nature Notes). Additional copies can be supplied on payment. A list of available back numbers of the Journal and reprints can be obtained from the Librarian, National Museum, P.O. Box 658, Nairobi. JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. XX Vn No. 2 (117) JULY 1968 CONTENTS Page The Elopoid and Clupeoid Fishes of East African Coastal Waters 77 By G, F. Losse Notes on the Birds Observed in the Vicinity of Tabora, Tanzania, with Special Reference to Breeding Data 117 By J. F. Reynolds Curculionidae (Weevils) of the Alpine Zone of Mount Kenya 141 By I. Jabbal and R. Harmsen Records of Parasitic Nematodes in Kenya 155 By G. D. Schmidt and A. G. Canaris Distribution and Host-Specificity of a Number of Fleas Collected in 157 South and Central Kenya By R. Harmsen and I. Jabbal Nature Note: Singing by the Chiflf-Chalf in Kenya 163 By C. W. Benson Book Reviews 163 Appreciation 167 (Published 31/5/69) Price Shs. 20/- EAST AFRICA NATURAL HISTORY SOCIETY President: DR. A. D. Q. AGNEW Vice-President: C. J. BEECHER Executive Committee: MISS P. M. ALLEN G. C. BACKHURST ESQ. MRS A. L. CAMPBELL R. H. CARCASSON ESQ. J. S. KARMALI ESQ. DR. T. R. ODHIAMBO MISS J. R. OSSENT Hon. Editor: DR. P. J. GREENWAY Hon. Treasurer: P. G. ARCHER ESQ. Hon. Secretary: MRS. FLEUR NG’WENO All correspondence in connection with this Journal should be addressed to: The Secretary, East Africa Natural History Society, P.O. Box 4486, Nairobi Kenya. Page 77 THE ELOPOID AND CLUPEOID FISHES OF EAST AFRICAN COASTAL WATERS by G. F. Losse {East African Marine Fisheries Research Organization, Zanzibar*) INTRODUCTION Fishery statistics show that the herring-like fishes are among the most important in world fisheries of the present day (FAO, 1966). In East African waters, prior to the introduction of commercial pelagic fishing methods (EAMFRO, 1962; Losse, 1964, 1966), these fishes were exploited in very small quantities by a variety of indi- genous fishing techniques, and virtually nothing was known of the species; their identity, biology or fishing potential. During preliminary biological and fishery investigations of fish caught by introduced purse-seine fisheries in the Zanzibar area of East Africa, it was found that accurate descriptions of species were required before these studies could be accomplished successfully. A large collection of clupeoid fish was therefore made during the routine investigations of the East African Marine Fisheries Research Organization (EAMFRO) from March 1963 to June 1966. Specimens were obtained from commercial purse-seine and stick-held dipnet {bouke-ami) catches taken in the Zanzibar Channel; also from catches made by indigenous fishermen with stake traps, castnets, shore-seines and hook and line, along the entire East African coast. Additional material was obtained from catches made by handnet, stick-held dipnet, seine, demersal trawling in estuaries and hand- lining, during cruises on the Organization’s (EAMFRO) vessels R L CFIERMIN and FRV MANIHINE. This material forms the basis of the descriptions in this paper, in which twenty-three distinct species, including two new distributional records, are described. The biology of the species is still poorly known and little data is available on species of doubtful commercial value. Observations on the biology and fishery of the species are in preparation. The present study could not have been completed without the examination of specimens from the clupeoid collection at the British Museum (Natural Flistory) in London during the summer of 1964. The East African collection described here has now been donated to this museum. MEASUREMENTS Standard length and all other body measurements cited in the descriptions of clupeoid fish were made with dial calipers on preserved material. Standard length (S.L.) is measured from the tip of the snout (jaws closed) to the caudal base. All measurements are expressed in percentages of standard length; depth, the depth of the body measured at the dorsal origin ; head, the length from the snout tip (jaws closed) to the most posterior edge of the operculum, i.e. not necessarily a horizontal line; snout, eye, post-orbital measured in a horizontal line passing through the centre of the eye; interorbit aJ, the bony inter-space between the eyes measured at eye centre; upper jaw, measured from the pre-maxillary symphysis to the posterior tip of the maxilla ; lower jaw, measured from the dentary symphysis to the posterior border of the articular; pectoral fin, pelvic fin, the lengths of these fins measured *Present address; c/o Department of Fisheries, FAO, Rome. Page 79 from the tip of the longest ray to the origin of the respective fins; pre-dorsal, pre- pelvic, pre-anal, these distances measured from the snout tip to the origin of the respective fins ; caudal peduncle, the minimum depth at the base of the caudal fin. \n fin counts simple (unbranched) rays are represented by small (lower case) roman numerals and branched rays by arabic numerals. Spines are shown by large (upper case) roman numerals. In ventral scute counts the pre-pelvic count includes the pelvic scute; the first post-pelvic scute lies between the bases of the pelvic fins and lacks ascending arms (Whitehead, 1965b). Lateral scales or scale pockets are counted at mid-body from the operculum to the caudal base. Vertebral counts include the uro- style. Measurements and counts outside the normal ranges are placed in parenthesis. NOMENCLATURE Classification to family level is based on Whitehead (1963a). In the synonymy reference is made to the original description, other subsequent records from East African waters and the more important literature from adjacent areas (i.e. from the Red Sea to Natal and the Oceanic islands); after the date of publication follow page numbers, plates, figures and localities. Standard common names are introduced here for the first time. These are based on names frequently in use by fishery scientists and fishery officers in East Africa. Vernacular names follow their Swahili rendering and were obtained from the indi- genous fishermen of the coast. East Africa refers throughout only to the coast of Kenya and Tanzania and the offshore islands of Pemba, Zanzibar and Mafia (Fig. 1). Eastern Africa refers to the whole eastern side of the African continent, from the Red Sea to Natal. Sub-order ELOPOIDEI Key to the East African Families 1 . Mouth terminal ; upper jaw bordered by maxillae and pre-maxillae ; (i) Last dorsal ray not filamentous ; pseudobranch exposed . . Family Elopidae (ii) Last dorsal ray filamentous ; pseudobranch not exposed . . Family Megalopidae 2. Mouth inferior; upper jaw bordered by pre-maxillae only . . Family Albulidae Family ELOPIDAE TENPOUNDERS A single genus widely distributed in tropical seas. Genus ELOPS Elops Linnaeus, 1766: 518 (Type: Elops saurus Linn.) Argentina Forskal {part.), 1775: 68. The genus has been revised by Regan (1909), Bertin (1944a) and Whitehead (1962). A single species in East African waters. ELOPS MACHNATA (Forskal) Plate la Argentina machnata Forskal, 1775: 68 (Type locality: Djedda, Red Sea). Elops machnata: Ruppell, 1835: (80) 74 (Red Sea) ; Gunther, 1 866 : 1 21 , fig. (caudal, Zanzibar) ; Sauvage, 1891: 497, pi. 49a, fig. 4 (Madagascar); Gilchrist & Thompson, 1917: 29 (references); Whitehead, 1962: 321 (Indo-Padfic specimens; revision all spp.)\ Losse, 1964: 12 (Zanzibar Channel); Whitehead 1965b: 231 (Red Sea, Gulf of Aden); Losse, 1966a: 89 (Zanzibar); Idem, 1966b: 167 (East Africa; Dar-es-Salaam, Zanzibar, Shimoni, Mombasa, Fundishu, Lamu); Wew, 1966c: 50 (Zanzibar Channel). Elops saurus: Gunther, 1868: 470 (East Africa, Zanzibar); Sauvage, 1891 : 497, pi. 49a, fig. 4 (Mada- gascar); Boulenger, 1909: 25, fig. 17 (Zanzibar, East Africa); Gilchrist & Thompson, 1908-11: 270 (Natal); Gilchrist, 1913: 30, pi. 2 (Algoa Bay, East London, Delagoa Bay); Smith, 1949-1965: 86, fig. 100 (Natal); Baissac, 1951: 124 (Mauritius); Morrow, 1954: 803 (Kenya); Smith, 1955: 306 (Aldabra); Fourmanoir, 1957:5 (Madagascar); Smith, 1958: 131 (Inhaca, Mozambique); Idem, 1963: 8, pi. 44 (Seychelles); Talbot, 1965: 464 (Mafia). Page 80 STANDARD COMMON NAME: Tenpounder. VERNACULAR NAMES: Ganati or Mkami kuoza (Malindi), Munyimbi (Shimoni), Mkizi (Zanzibar). DESCRIPTION: Based on two fishes, 873-890 mm. standard length, from the Mafia Channel, eleven, 169-866 mm., from Zanzibar and one, 410 mm., from Lamu. Dorsal 21-24 (3-4 of which are simple rays), pectoral i 15-17, pelvic i 12-14, anal 13-15 (2-3 of which are simple rays). Abdominal scutes and pelvic scute absent. Gillrakers 6-9 + 12-14, total 20-23 on 1st gill arch. The anterior rakers on the upper and lower arch are often reduced to low (movable) tubercles. Branchiostegal rays 29-33. Scales thin, small, firmly adherent, 86-106 in lateral series, plus 6-7 more on caudal. Pre-dorsal scales 43-44 (Mafia specimens), 13 above lateral line and 12 below. Vertebrae 60-63 (nine Zanzibar specimens). Depth 15.2-20.1, head 18.4-27.2, snout 5.0-6. 6, eye 3.9-5.9, post-orbital 12.5-13.9, inter-orbital 3. 8-4.8, upper jaw 12.2-15.3, lower jaw 11.0-16.3, maxilla 9.3-12.9, pectoral fin 11.6-14.8, pelvic fin 10.2-14.8, pre-dorsal 49.1-55.3, pre-pelvic 44.5-55.7, pre-anal 76.0-81.7, gular plate, length 7.7-10.0, width 1. 5-2.4, pectoral axillary scale 7. 1-9. 5, pelvic axillary scale 6.6-9.2. Body elongate, scarcely compressed. Lower jaw variable; either included within the upper and not covering the anterior pre-maxillary toothband or projecting and covering part or whole of the pre-maxillary toothband ; this character varied with size, the larger specimens possessing the longest jaws. Mouth large and terminal, scarcely inclined. Maxillary reaching beyond the posterior edge of the orbit. Teeth in villiform bands in jaws, on vomer, palatines and tongue. Gular plate large. Pseudobranch exposed. Adipose eye-lids broad. Lateral line straight with simple tubes. Anal fin far behind dorsal fin base ; dorsal and anal fin with scaly sheaths, into which fins can fold completely. Last dorsal ray not elongated. COLOUR: Fresh, dorsal scales bluish, their lateral and posterior borders black. Sides silvery, with a golden sheen or hue in death. Ventral surface white. Top of head dark brown to black with a green- ish hue. A greenish spot in front of nares. Sides of head silver with a golden sheen. Black streak at upper posterior edge of operculum. Caudal fin dark brown to black, violet reflections at base. Anal and pelvic fins yellowish. Pectoral and pelvic axillary scales golden yellow, speckled brown. In alcohol, dorsal surfaces dark brown to black, flanks silvery. Ventral surfaces white. Dorsal and anal flns dark brown, darkest (almost black) at margins, other fins brownish. SIZE: 890 mm. S.L. (9, 13 lb., Mafia Channel), weight 14 lb. ((?, 866 mm. S.L., Zanzibar Channel); average 30 in. and 30 lb. in South African seas (Smith, 1949). DISTRIBUTION: Entire East African coast (recorded from the Mafia Channel, Dar-es-Salaam, Zanzibar, Shimoni, Mombasa, Malindi, Fundishu and Lamu), in estuaries, mangrove areas, lagoons and saline pools (Fundishu, Zanzibar); generally over a muddy or sandy bottom, often together with Megalops cyprinoides. The species was rarely found out to sea and apparently shows a preference for euryhaline conditions. RANGE: Eastern coast of Africa from the Red Sea to Algoa Bay; Seychelles, Aldabra, Comores, Madagascar, Mauritius. Elsewhere, widespread throughout the Indian and Pacific oceans; East Indies, China, Japan and Hawaii (£. hawaiensis is known only from Hawaii and Australia). REMARKS: Whitehead (1962) recognised two Indo-Pacific species, E. machnala (Forsk.) and E. hawaiensis Regan, distinguishing these on vertebral number (63-64 in E. machnala c.f. 68-70 in E. hawaiensis) and length of lower jaw (projecting and covering anterior part of pre-maxillary toothband in E. machnala, and included, with the whole of pre-maxillary toothband exposed in E. hawaiensis). Zanzibar specimens, which were examined, are characterised by a low vertebral count (60-63) and are therefore referable to E. machnala. The lower jaw character is, however, variable and cannot be , relied upon in the identification of the species. Morrow (1954) referred a single specimen (1070 mm.) from Shimoni, Kenya, to E. saums, stating that the ranges of meristic characters showed consider- able overlap and that the specimen could therefore not be referred to any of Regan’s (1909) species, although corresponding closest to E. hawaiensis. Other authors (Fowler, 1940; Smith, 1961) conser- vatively recognised but a single world-wide species, E. saurus Linn. Future work on Indo-Pacific specimens may demonstrate that E. machnala and E. hawaiensis are not specifically distinct, but represent respectively western and eastern components (i.e. sub- species) of a single species. Family MEGALOPIDAE TARPONS Genus MEGALOPS Lacepede, 1803 Megalops Lacepede, 1803: 380 (Type: Megalops filamentosus Lacepede = C/w/jco cyprinoides Brous- sonet). A single species in the Indian Ocean. MEGALOPS CYPRINOIDES (Broussonet) Clupea cyprinoides Broussonet, 1782: (no pagination) pi. 9, (Type locality: oceans between tropics). Megalops filamenlosus: Lacepede, 1803: 289, 290, pi. l3, fig. 3 (Fort Dauphin, Madagascar); Four- manoir, 1957: 5 (Madagascar). Page 81 Megalops indicus: Valenciennes, 1846: 388, pi. 542 (Madagascar, Mauritius). Megalops setipinna: Bleeker, 1863: 345 (Madagascar). Elops eyprinoides: Gunther, 1966: 122 (East Africa); Martens 1869: 143 (Pangani River, Zanzibar). Megalops eyprinoides: Gunther, 1868: 47 (Zanzibar); Sauvage, 1891 : 497, pi. 49a, fig. 3 (Madagascar); Gilchrist & Thompson, 1908-11 : 270 (Natal); Boulenger ,1909: 28, fig. 19 (Zanzibar, Shire River); Gilchrist, 1913: 52 (references); Gilchrist & Thompson, 1917: 292 (references); Barnard, 1925: 104, fig. 11 (larva) (Natal); Smith, 1949-1965: 86, fig. 101 (South Africa); Baissac, 1951 : 123 (Mauritius); Smith, 1955: 306 (Aldabra);/rfew, 1958: 131 (Inhaca,Mozambique);Wew, 1963:9,p.I4E (Seychelles); Losse, 1964: 12 (Zanzibar Channel); Idem, 1966a: 89 (Zanzibar); Idem, 1966b: 168 (East Africa, Dar-es-Salaam, Zanzibar, Mombasa, Malindi); Idem, 1966c: 50 (Zanzibar Channel). STANDARD COMMON NAME: Tarpon. VERNACULAR NAMES: Pawale (general), Kumpangn (Malindi). DESCRIPTION: Based on one fish, 413 mm. standard length, from the Mafia Channel; one, 173 mm., from Zanzibar; one, 480 mm., from Tanga and two, 293-373 mm., from Mombasa. Dorsal iv-v 14-16, total 19-20, pectoral i 14-16, pelvic i 9-10, anal iii-iv 22-23 (25), total 26-27 (28). Gillrakers 14-16+29-33, total 43-+9 on 1st gill arch. Scales firmly adherent, cycloid and very large, 34-36 in lateral series, about 4 more on caudal. Branchiostegal rays 24-27. Depth 26.4-28.6, head 28.2-30.0, snout 4.7-7.7, eye 7. 2-7. 9, post-orbital 12.6-14.0, inter-orbital 5. 1-5.7, upper jaw 1T9-15.9, lower jaw 15.2-15.3, gular plate, length 9.6-11.1, width 1.3-1 .8, pectoral fin 17.1-20.2, pelvic fin 11.1-15.4, pre-dorsal 52.3-57.0, pre-pelvic 49.3-55.9, pre-anal 72.4-79.0, last dorsal ray 27.0-30.4. Body elongate, moderately compressed. Mouth very large, terminal. Maxillary reaching beyond posterior border of orbit. Pseudobranch not exposed. Adipose eyelids well developed, almost covering eyes completely. Lateral line well developed, with branched tubes. Pectoral and pelvic axillary scales about two thirds length of fins. Anal fin behind base of dorsal fin, without basal scaly sheath. Last dorsal ray greatly elongated, filamentous, about equal to head or a little more in length. COLOUR: Fresh, dorsal surfaces dark brown, sides silvery to golden; scales with dark borders. Top of head dark brown. A dark patch on posterior edge of operculum. Fins brownish, caudal and dorsal margins dark. Pectoral and pelvic axillary scales speckled light brown. In aleohol, dorsal surface dark brown to black, sides silvery or light brown. Ventral surfaces white. Dorsal, anal and caudal fins dusky to light brown, darker (almost black) at margins. Pectoral and pelvic fins with brownish speckles. SIZE: 480 mm. S.L., weight 5^ lb. ($, Tanga Bay). DISTRIBUTION: Entire East African coast (recorded from the Mafia Channel, Dar-es-Salaam, Zanzibar, Tanga, Mombasa and Malindi), in estuaries, lagoons, bays and mangrove areas, often together with Elops. The species was rarely found out to sea and apparently shows a preference for euryhaline conditions. RANGE: East African coast south of Natal; Seychelles, Madagascar, Mauritius. Elsewhere, widely distributed in the Indian and Pacific oceans; India, Ceylon, East Indies, Philippines, China, Japan, Formosa. Australia. Melanesia. Micronesia and Polynesia. Family ALBULIDAE LADY FISHES Genus ALBULA Scopoli, 1777 Albula Scopoli, 1777: 450 (on Gronow) (Type: Esox vulpes Linnaeus). A monotypic genus widely distributed in tropical seas. ALBULA VULPES (Linnaeus) Esox vulpes Linnaeus, 1758: 313 (on Bone Fish Catesby, 1737: pi. 2, fig. 1; Bahamas). Albula bananas: Valenciennes, 1846 : 345 (Mauritius). Butirinus glossodontus: Gunther, 1866: 120 (Zanzibar). Butirinus glossodontis: Playfair, 1867: 868 (Seychelles). Albula conorhynchus : Gunther, 1868: 468 (Port Natal, Zanzibar, Red Sea). Albula conorhyncus: Gilchrist & Thompson, 1908-11 : 269 (Natal). Albula vulpes: Gilchrist, 1913: 53 (Natal); Gilchrist & Thompson, 1917: 293 (references); Barnard 1925: 106 (Natal); Fowler, 1934: 410 (Durban); Smith, 1949-1965: 85, fig. 99 (Natal to Algoa Bay); Baissac, 1951: 124 (Mauritius); Smith, 1955: 306 (Aldabra);Fourmanoir,1957:4,pl.la(Madagascar); Smith 1958: 131 (Inhaca, Mozambique); Idem, 1963: 8 pi. 4A (Seychelles); Whitehead, 1965b: 232 (Red Sea, Gulf of Aden); Losse, 1966b: 168 (East Africa; Zanzibar, Mombasa). STANDARD COMMON NAME: Bone fish. VERNACULAR NAMES: Mnyimbi (Zanzibar, Shimoni), Mborode (Malindi). DESCRIPTION: Based on a single specimen, 319 mm. standard length, from Zanzibar. Dorsal iv 13, pectoral i 16, pelvic i 8, anal iii 7. Gillrakers poorly developed, reduced to low tubercles, 12 on lower part of 1st gill arch. Lateral line well developed; scales large, silvery and ad- Page 82 herent, from shoulder with three anterior striae, 69 scales in lateral series, 23 pre-dorsal scales, 9} above lateral line, 7^ below. Branchiostegal rays 1 2. Depth 24.5, head 28.0, snout 12.2, eye 4.5, post-orbital 11.5, inter-orbital 7.1, upper jaw 8.4, lower jaw 8.6, projection of snout beyond lower jaw 2.7, pectoral fin 15.6, pelvic fin 11.8, pre-dorsal 50.5, pre-pelvic 59.0, pre-anal 88.1, caudal peduncle 8.0, caudal length 32.4. Body oblong, moderately compressed, abdomen flattened. Snout pointed, projecting beyond tip of lower jaw. Mouth small, inferior. Maxilla short not reaching eye, just surpassing nares in vertical plane. Gular plate not evident externally. Pseudobranch exposed. Eyes almost completely covered by a thick adipose membrane, only a small oval aperture. Dorsal base nearer to caudal than to the snout; pelvic origin beneath 15th dorsal ray. Dorsal, anal and caudal fins with scaly sheaths at bases. Pelvic axillary scale about half length of fin. Pectoral axillary scale poorly developed. Teeth villiform, on pre-maxilla, lower jaw, vomer and palatines. Maxilla edentulous, upper jaw bordered by pre-maxillae only. Broad patches of granulars teeth on parasphenoid, pterygoid and tongue. COLOUR: Fresh, body silvery with about nine dark zigzag lines along dorso-lateral surface. Top of head greenish. A black spot on each side of snout tip. Scattered melanophores on pectoral, pelvic and anal fins. Anterior dorsal rays with black anterior borders; dorsal margins dark. Caudal tips dusky. SIZE: 318 mm. S.L. (Zanzibar). DISTRIBUTION: Entire East African coast (recorded from Zanzibar, Dar-es-Salaam, Mombasa and Malindi), in the vicinity of mangrove areas and in shallow bays over a sandy bottom, where it was often abundant. RANGE: Eastern coast of Africa from the Red Sea to Algoa Bay; Seychelles, Aldabra, Comores, Madagascar, Mauritius and Reunion. Elsewhere, cosmopolitan in tropical seas. Suborder CLUPEOIDEI Key to East African Families 1. Abdominal scutes absent: (i) Body highly compressed; jaw teeth large and fang-like; dorsal origin much nearer to caudal base than to the snout (ii) Body rounder, jaw teeth small, not fang-like; dorsal origin about midway between the snout and the caudal base 2. Abdominal scutes present, keeled or spine-like: (i) Mouth terminal or subterminal, snout not pig-like; upper jaw (maxilla) not extending beyond the posterior border of the eye (ii) Mouth inferior, snout pig-like; upper jaw (maxilla) extending beyond the posterior border of the eye Family Chirocentridae Family Dussumieriidae Family Clupeidae Family Engraulidae Family CHIROCENTRIDAE WOLF-HERRINGS A single genus widespread in the Indo-Pacific area. Genus CHIROCENTRUS Cuvier, 1817 Chirocentrus Cuvier, 1817: 178 (Type: Clupea dorab Forsk.). A single species in East African waters. CHIROCENTRUS DORAB (Forsk.). Clupea dorab Forsk., 1775: 72 (Type locality: Djedda, Red Sea). Chirocentrus dorab: Valenciennes, 1846: 150 (Mauritius, Zanzibar, Red Sea); Peters 1855: 268 (Mozambique); Gunther, 1866: 1 20 (Zanzibar) ; Gunther, 1868:475 (Port Natal, Zanzibar); Peters, 1876: 445 (Mauritius); Gilchrist & Thomspon 1908-11 : 202 (Natal); Barnard, 1925: 120 (Delagoa Bay, Natal); Bonde, 1934: 437 (Zanzibar); Fowler, 1934: 410 (Natal); Smith, 1949- 1965: 87, pi. 5, fig. 104 (South Africa); Baissac, 1951: 130 (Mauritius); Morrow, 1954: 804 (Pemba); Smith, 1955: 306 (Aldabra); Fowler, 1956: 78 (Indo-Pacific specimens); Fourmanoir, 1957: 7, pi. IB (Madagascar); Smith, 1958: 131 (Inhaca, Mozambique); Idem, 1963: 8 pi. 41 (Seychelles); Sanches, 1963: 21, fig. 4 (Inhaca, Mozambique); Losse, 1964: 12 (Zanzibar); Whitehead, 1965b: 233 (Red Sea); Losse, 1966a: 89 (Zanzibar); Idem, 1966b: 169 (East Africa; Dar-es-Salaam, Zanzibar, Mombasa, Malindi, Lamu); Idem, 1966c: 50 (Zanzibar Channel). STANDARD COMMON NAME: Wolf-herring. VERNACULAR NAMES: Mkonge (Zanzibar, Tanzania), Panga (Kenya, general), Bahanafu (Malindi). DESCRIPTION: Based on six fishes, 190-237 mm. standard length, from the Zanzibar Channel; six, 323-341 mm., from Mombasa and one, 319 mm., from Formosa Bay. Dorsal iii-iv 12-15, pectoral i 12-13, pelvic i 5-6, anal iii-iv 29-31. Gillrakers 3 + 12-16 on 1st gill arch. Scales minute thin and very caducous; scale pockets obliterated, no lateral count possible. Branchiostegal rays 8. Depth 13.6-16.0, head 17.7-19.2, snout 4. 1-7.2, eye 3.0-3. 6. post-orbital 7.2-8. 5, inter-orbital 2.4-2.7, upper jaw (8.2) 8. 6-9.4, lower jaw (9.3) 9.6-10.3, pectoral fin 12.1-13.6, pelvic fin 2.5-3.2, pre-dorsal 67.9-73.6, pre-pelvic (45.6) 48.0-52.4, pre-anal 68.5-72.0, pectoral axillary scale 7.8-8.9. Body greatly elongated, strongly compressed. Snout pointed, lower jaw strongly inclined. Mouth moderately large and terminal. Maxilla surpassing the angle of the jaws but not reaching to the an- terior border of pre-operculum. Pseudobranch not exposed. Adipose tissue well developed, often obscuring eyes. Abdominal scutes absent; a single, small, crescentric pelvic scute. Dorsal origin much nearer to caudal base than to the snout. Anal fin long, the origin just in front or under first ray of dorsal fin. Pelvic fins very small, with a small axillary flap, far in front of dorsal fin, much nearer to pectoral origin than to the caudal base. Dorsal and anal fin with well developed scaly sheaths. Pectoral axillary scale well developed; two long scales on caudal base. Teeth well developed. Two large upper canines pointing forward, covered dorsally by a loose fleshy flap extending from snout. Small pointed teeth on maxilla. Lower jaw with caniniform teeth which increase in size posteriorly. Mandibular rami elevated inside mouth. COLOUR: Fresh, back blue, midline grey; sides and belly silvery. A large green patch on shoulder. Supra-orbital part of head bluish, post-orbital dark. Snouth bluish-grey. Iris dusky to silver. Anterior rays of dorsal fin dusky. Minute dark spots on anterior pectoral rays. Caudal dusky, darker, almost black, at tips; ventral fins and anal fins colourless. In alcohol, dorsal sufaces bluish-grey, sides yellowish-white. Snout dark. Sub-orbital part of head and operculum silvery; a dark patch on anterior portion of operculum. Dorsal yellowish at base, dusky, margins darker. First pectoral ray with a black anterior edge, rays 1-5 speckled black. Ventral fins and anal fin colourless. In the single specimen from Formosa Bay the caudal fin and outer half of the pectoral fin are entirely black and anal rays 4-7 are speckled black. SIZE: Specimens (??) of about 800 mm. have been caught in the Zanzibar Channel (by purse-seine); 341 mm. S.L. (cj, Mombasa). Stead (1906) stated “attains a length of fully 12 feet” (Australian seas). Smith (1949) and Fowler stated the same maximum length without reference. DISTRIBUTION: Entire East African coast (recorded from the Mafia Channel, Dar-es-Salaam, Zanzibar, Shimoni, Mombasa, Malindi, Formosa Bay and Lamu), in the shallow waters within the 30 fathom contour. Abudant over shallow banks (Zanzibar Channel), in hays, harbours (Mombasa), estuaries and also quite common in and around mangrove areas. RANGE: Eastern coast of Africa from the Red Sea to Natal; Seychelles, Madagascar, Aldabra and Mauritius. Elsewhere, widely distributed in the Indian and Pacific Oceans; India, East Indies, Philippines, China, Formosa, Japan, Queensland and Melanesia. Family DUSSUMIERIIDAE ROUNDHERRINGS Key to the East African Genera 1 . Branchiostegal rays 14-16; adults large ( > 120 mm. S.L.) (Sub-family Dussumieriinae) DUSSUMIERIA 2. Branchiostegal rays 6-7; adults small (<80mm. S.L.) (Sub-family Spratelloidinae) (i) A single supra-maxilla ; pelvic origin in front of dorsal ; last two anal rays separated from others by a gap .... SPRATELLOMORPHA (ii) Two supra-maxillae; pelvic origin under dorsal base; anal fin entire SPRATELLOIDES Sub-family DUSSUMIERHNAE Genus DUSSUMIERIA Valenciennes, 1847 Dussumieria Valenciennes, 1847: 467 (Type: Diissuinieria acuta Val.) A single species in East African waters. DUSSUMIERRA ACUTA (Valenciennes) Dussumieria acuta Valenciennes, 1847: 467, pi. 606 (Type locality: Bombay); Whitehead, 1963b: 312, fig. 1-5 (revision, synonymy ; Indo-Pacific specimens); Losse, 1964: 12 (Zanzibar Channel); Whitehead 1965b: 234 (Red Sea, Gulf of Aden); Losse, 1966a: 89 (Zanzibar); Idem, 1966b: 170 (East Africa; Dar-es-Salaam, Zanzibar, Kenya); Idem, 1966c: 51 (Zanzibar Channel). Dussumieria hasseltii: Fourmanoir, 1961: 84, fig. 1 (Madagascar). STANDARD COMMON NAME: Round Herring. VERNACULAR NAMES : Dagaa la upapa (Zanzibar), Dagaa (general). Page 84 DESCRIPTION; Based on twenty-five fishes, 70-144 mm. standard length, from Zanzibar and one, 124 mm., from Mombasa. Depth measurements and meristic characters on nine further fishes, 87.5-154 mm., from the Zanzibar Channel. Dorsal iv-v 16-17 (19), pectoral i 12-13, pelvic i 7, anal iii (12) 13-14. Abdominal scutes absent, a single “W” shaped pelvic scute (Whitehead, 1962). Gillrakers, 25-28 on lower part of 1st gill arch. Branchiostegal rays 14—16. Scales caducous, scale pockets virtually all obliterated, no count possible. Vetebrae 54-56 (8 fishes). Depth 18.5-25.1 (increasing with length of fish), head 26.0-28.4, snout 8.7-9.7, eye 5.9-7.0, post-orbital 8.4-10.4, upper jaw 7.9-8.7, lower jaw 12.6-13.8, pectoral fin 11.9-14.8, pelvic fin 8.1- 10.8, pre-dorsal 54.3-58.6, pre-pelvic 58.0-61.9 (62.2), pre-anal (77.4-78.5) 79.0-81.5, anal base 9.2- 11.3. Body oblong, moderately compressed, rounded ventrally. Snout pointed, longer than the eye diameter. Lower jaw projecting slightly in front of upper. Maxilla short, not quite reaching anterior border of orbit. Two supra-maxillae, the second (posterior) not expanded posteriorly. Conical teeth on pre-maxilla, maxilla and mandible. Pelvic origin under dorsal base, nearer to pectoral origin than the caudal base. Dorsal origin nearer to caudal base than to the snout. COLOUR; Fresh, dorsal surface blue-grey. A dark blue mid-dorsal line often with a greenish tinge. Iridescent golden lateral stripe from the operculum to the caudal base. Flanks silver with a golden hue in some fishes. Top of head greenish to golden. Tip of upper jaw black. Tip of mandible dusky. Operculum mainly silvery with a few melanophores on dorsal border; on pre-operculum dark spots form faint lines. A few dark spots on anterior rays of dorsal fin. First ray of pectoral fin dark and a few dark spots on central rays of fin. Caudal dusky, darker at margins, colourless in some (small) specimens. Pelvic fins and anal fin colourless. In alcohol, dorsal surfaces brownish, flanks light brown to grey. A light lateral band visible in some specimens. Snout dark. First ray of pectoral fin and anterior rays of dorsal fin dark. Caudal darkish at tips in some specimens, otherwise yellowish or colourless. SIZE; 154 mm. S.L. (?, Zanzibar Channel), maximum 216 mm. in India (Day). DISTRIBUTION ; Entire East African coast (recorded from Dar-es-Salaam, Zanzibar and Mombasa) in the shallow waters within the 100 fathom contour. A single specimen was taken from the stomach of a sailfish caught in 250 fathoms on longline off Mombasa. Not evident in bays, harbours or estuaries ; the species was abundant over shallow water banks (Zanzibar Channel) in depths of 8-18 fathoms principally during the northeast monsoon (October — February). RANGE; Eastern coast of Africa from the Red Sea to Madagascar. Elsewhere, widely distributed in the Indian and Pacific Oceans, east to Japan. Also as an immigrant species in the Suez Canal and the eastern Mediterranean, from Port Said to Mersin in Turkey (Ben-Tuvia, 1966). Sub-family SPRATELLOIDINAE Genus SPRATELLOMORPHA Berlin 1946 Spratellomorpha Berlin, in Angel, Berlin & Guibe, 1946; 473^ (Type; Sauvagella madagascariensis bianalis Berlin, ex Madagascar. A monot>pic genus; Imown from four fishes (types) from Madagascar and ten juvenile specimens recently discovered by the author at the Mombasa fish market. The species has not been recorded previously from East Africa. SPRATELLOMORPHA BIANALIS (Berlin) Plate lb Sauvegella madagascariensis binalis Berlin, 1940; 300 (ex-Madagascar). Sauvagella bianalis Berlin, 1943; 22, fig. 8; Whitehead, 1936b; 336, fig. 14 (revision; Madagascar). Spratellomorpha bianalis Berlin, 1946; 473^. STANDARD COMMON NAME; Estuarine sprat. VERNACULAR NAMES; Dagaa (Mombasa). DESCRIPTION; Based on seven fishes, 30.2^1.3 mm. standard length from Port Tudor, Mombasa. Dorsal iii 11-12, pectoral i 11-12, pelvic i 7, anal iii (12) 13+2, total (17) 18. Gillrakers 26-28 (three fishes) on lower part of 1st gill arch. Branchiostegal rays 6. Scales caducous, about 41 (pockets) in lateral series. Depth 17.7-21.8, head 26.3-27.7, snout 6.1-6.7, eye 6.7-7.6, post-orbital 7. 3-9.0 (11.3), maxilla 9.2- 10.0. lower jaw 11.0-12.2, pectoral fin 15.8-19.2, pelvic fin 11.6-12.7 (13.2), pre-dorsal 54.7-58.1, pre-pelvic 49.4-54.7, pre-anal 69.6-75.6. Body elongate, laterally compressed, its depth less than the length of the head. Snout pointed, less than eye diameter in length. Abdominal scutes entirely absent, a single pelvic scute with pointed ascending spines. A single supra-maxilla, expanded posteriorly, almost equal to maxilla in depth, with a narrow anterior shaft. A single row of conical teeth on pre-maxilla and maxilla. Maxilla longer than snout, reaching vertical through anterior border of the eye. Dorsal origin much nearer to caudal base than to snout, about equidistant from anterior border of eye and caudal base. Pelvic Page 85 origin in front of dorsal, nearer to the anal base than to pectoral origin. Anal origin much nearer to pelvic origin than to the caudal base. Last two rays of anal fin separated from others by a space which equals two or three rays bases. COLOUR: Fresh, mainly translucent; dorsal and lateral surfaces very light brown, peritoneum white. A faint, poorly defined, silvery to golden lateral band from operculum to caudal base, widest behind dorsal base, narrower than pupil of eye. A few brown spots on head between orbits, post- orbital surface of head brown. Two parallel rows of minute black spots on median dorsal surface from the nape to the dorsal origin and continued from the last dorsal ray to the caudal base. Snout and tip of lower jaw faintly yellowish with a few small dark spots. Lower jaw, sides of head and operculum silvery. Bases of dorsal and caudal fin yellowish, caudal rays with minute orange spots and a few scattered melanophores; two black lines on caudal base and a black streak on the upper border of the caudal peduncle. A row of black spots along base of anal and first two rays of pectoral fin with a few melanophores, rest of fins colourless. In alcohol, light brown with a very faint, poorly defined silvery lateral band from operculum to caudal base, hardly evident in some specimens. Dark markings all retained, rest faded. SIZE; 41.3 mm. S.L. (Mombasa), 45.5 mm. in Madagascar (Bertin). DISTRIBUTION: Port Tudor, Mombasa district. The species is probably more widespread than the present record would indicate. The isolated recorded occurrences of specimens at Port Tudor (Mombasa), essentially an area where estuarine conditions prevail, would indicate that the species may be confined to estuaries in East African waters. RANGE: East Africa (Mombasa). Elsewhere, Madagascar. REMARKS : The first specimen of Spratellomorpha bianalis from the Kenya Coast was discovered by the author at the Mombasa fish market in April 1965. Nine further fishes were obtained from the same locality in December. All were caught in stake traps {uzio) at the entrance to Port Tudor, Mombasa, together with wolf-herrings (Chirocentrus), herrings (/^er/:/o/^^c/^r/^y^), sardines (iS’arr(/«e//a) and anchovies {Thrissina and Thryssa). In addition to these ten specimens from Mombasa the only other examples in existence appear to be the four types, 44.0-45.5 mm. S.L., in the Paris Museum (No. A5174), which Bertin described from Madagascar. Table 1 A comparison of proportional measurements and meristic characters of East African and Malagasy Spratellomorpha bianalis East African material (7 fishes) Standard length In % of S.L. Body depth Head length Snout length Eye diameter Post-orbital distance 30.2^1.4 mm. mean 20.40 27.11 6.40 7.02 8.80 Maxilla length 9.65 Lower jaw length ...... 11.67 Pectoral fin length 17.40 Pelvic fin length 12.28 Pre-dorsal distance 56.71 Pre-pelvic distance ...... 52.48 Pre-anal distance ...... 73.15 Dorsal rays : simple iii branched 11-12 Anal rays : simple branched . Pectoral rays (total) . iii 12-13+2 12-13 Malagasy material (After Whitehead, 1963) (4 fishes) 44.0-45.5 mm. range 17.1- 18.0 25.2- 26.5 6.9- 7.1 7.2- 7.8 8.4- 8.6 9.7-10.0 53.4- 56.5 (62.8) 49.4- 52.2 69.0-72.5 12-13 iii 11-12+2 13 Page 86 Both in meristic characters and body proportions East African S. bianalis agree closely with Malagasy material (Table I). The slight differences which are evident may indicate that East African and Malagasy specimens represent distinct populations; assuming that the species is truly estuarine, isolation is probably complete. Spratellomorpha and other dussumieriid genera placed in the tribe Ehiravini (vide Whitehead, 1963) are of considerable interest as they show close affinities with the true herrings (Clupeidae) in a number of features (Whitehead, loc. cit.). Recently it has been suggested that they may be better placed in the sub-family Pellonulinae of the Clupeidae (Poll et al., 1965). Genus SPRATELLOIDES Bleeker, 1852 Spralelloides Bleeker, 1852: 29 (Type: Clupea argyrotaemata=Clupea gracilis Schlegel). Stoiephonis (non Lacepede) Fowler, 1941 : 561. Two Indo-Pacific species, both of which occur in East African waters. Key to the East African Species 1. Total anal rays 1 1-14; a prominent silver lateral band from operculum to the caudal base S. gracilis 2. Total anal rays 9-11 ; no silver lateral band, whole of flanks silvery . S. delicatiilus SPRATELLOIDES GRACILIS (Schlegel) Clupea gracilis Schlegel, 1846: 238, pi. 108, fig. 2 (Type locality: Japan). Spratelloides japoniciis: Morrow, 1954: 804 (Mkoani harbour, Pemba). Spratelloides gracilis: Whitehead, 1963b: 388, fig. 15-18 (revision, synonymy; Red Sea and Pacific specimens, Whitehead, 1965b: 273, figs. 2b, 3b, (Red Sea); Losse, 1966b: 170 (East Africa; Mafia, Zanzibar). STANDARD COMMON NAME: Silver-striped sprat. VERNACULAR NAME: Dagaa (general). DESCRIPTION : Based on three fishes, 28.0-56.0 mm. standard length, from the Mafia Channel ; seven 40.5-52.0 mm., from Zanzibar and six, 22.1-51.0 mm., from Shimoni. Dorsal iii 9-10, pectoral i 11-12 (14), pelvic i 7, anal ii 9-12, total 11-14. Scales caducous, no accurate count possible (about 40-50 in Red Sea specimens ; Whitehead 1 964b). Branchiostegal rays 6. Depth 12.6-17.4, head 23.8-25.8 (28.0), snout 12-9.5, eye 4. 8-6.9, post-orbital 8.2-10.9, upper jaw 8. 6-9.9, lower jaw 9.8-12.5, pectoral fin 11.4-14.2, pelvic fin 9.7-12.1, pre-dorsal 46.4-50.5, pre-pelvic 53.4-58.8, pre-anal 80.0-84.1. Body moderately compressed. Snout pointed, slightly longer than the eye diameter. Head longer than the maximum body depth. Maxilla moderately long, the posterior border beneath the anterior third of the orbit, not reaching anterior border of the pupil. Dorsal origin a little nearer to snout than to the caudal base, or about equidistant. Pelvic origin under posterior half of dorsal, slightly nearer to the caudal base than to the snout. COLOUR: Fresh, dorsally pale green with a dark-dorsal line. A prominent silver lateral band from operculum to the caudal base, as wide as eye diameter, edged above and below by a fine blue line. Pupil black, iris silver. A small black mark just behind eye. Head silver with small black markings on lower jaw and snout. Bases of caudal rays and posterior border of caudal peduncle with fine black markings. Fins otherwise colourless. hi n/co/(o/, upper and lower surfaces brownish, a silver lateral stripe prominent in some specimens, but faded in others to a dark band: poorly defined in juvenile specimens. All black markings retained. SIZE: 56 mm. S.L. (Mafia Channel), 93 mm. S.L. in Japan (Whitehead, 1963b). DISTRIBUTION: East African coast (recorded from Kilwa, Mafia Channel, Zanzibar Channel, Pemba and Shimoni) in the shallow waters, especially in and around coral reef areas. Not observed in estuaries, harbours, creeks or muddy bays. At times abundant, but great fluctuations in abundance were evident from year to year. RANGE: Eastern coast of Africa from the Red Sea to Kilwa in Tanzania. Elsewhere, widely dis- tributed in the Indo-Pacific, east to Ceylon, Laccadive islands, Japan and Samoa. SPRATELLOIDES DELICATULUS (Bennett) Clupea delicatula Bennett, 1831 : 168 (Type locality: Mauritius). Spratelloides delicatulus: Peters 1876: 445 (Mauritius); Jatjow & Lenz, 1899; 526 (no locality — East African collection); Regan, 1908: 242 (Kosi Bay); Gilchrist & Thompson, 1917: 296 (South Africa); Barnard, 1925; 110 (Zululand); Morrow, 1954; 809 (Mkoani harbour, Pemba); Fourmanoir, 1957; 13 (Madagascar, Comores); Whitehead, 1963b: 345, figs. 16-17, 19, 25, 28a, 30c,31 (references, synonymy, revisions; Zululand, Seychelles, Red Sea, Gulf of Aden); Idem, 1965b: 241, figs. 2a, 3a (Red Sea, Gulf of Aden); Losse, 1966b: 171 (East Africa; Mafia, Zanzibar, Kilifi); Idem, 1966c: 51 (Zanzibar Channel). Stoiephonis delicatulus: Smith 1949-1965 : 89, fig. 107 (South Africa); Baissac, 1951 : 126 (Mauritius); Smith, 1955 : 307 (Aldabra), Idem, 1958 : 131 (Inhaca, Mozambique) ; Idem, 1963 : 8, pi. 4B (Seychelles). STANDARD COMMON NAME: Common sprat. VERNACULAR NAME: Dagaa (general). DESCRIPTION: Based on eight fishes, 43.9-51.1 mm. S.L., from Kilifi and five 23.3^8.0 mm., from Shimoni. Dorsal ii 9-10 (11), pectoral i 10-12, pelvic i 7, anal ii-iii 7-9, total 10-11. Scales moderately caducous, about 32-35 (pockets) in lateral series. Branchiostegal rays 6. Depth 13.8-20.8, head 25.3-28.8, snout 6.3-8.0, eye 5.5-7. 5, post-orbital 9.0-10.4, upper jaw 8.1-10.5, lower jaw 9.8-12.2, pectoral fin 12.5-17.6, pelvic fin 11.0-14.4, pre-dorsal 45.9-417, pre- pelvic 53.8-57.0, pre-anal (78.5) 81.4-84.1. Body moderately compressed. Snout pointed, slightly longer than eye diameter. Maxilla longer than snout, almost reaching pupil. Dorsal origin slightly nearer to snout than to the caudal base. Pelvic origin under posterior half of dorsal fin, slightly nearer to caudal base than to the snout. COLOUR: Fresh, dorsal surface bright blue with darker blue mottlings, sides silvery, belly white. Pupil black, iris silver. Small black marks on upper part of operculum. Tip of snout and lower jaw speckled black. A prominent black mark a little in front and below eye. Two black lines on caudal base and one along the upper border of the caudal peduncle. Fins colourless. In alcohol, upper surfaces bluish, grey or brown, sides white or silvery, belly white. Top of head, snout tip and lower jaw dark brown to black. A small black spot in front of eye. Fins colourless. SIZE: 51 mm. S.L. (Kilifi, Kenya), 77 mm. S.L. in Australian seas (Whitehead, 1963b). DISTRIBUTION: Probably entire East African coast (recorded from Kilifi, Shimoni, Pemba, Zanzibar Channel and Mafia Channel), in the shallower inshore waters; in and around coral reef areas, in bays, inlets, lagoons and estuaries. Often together with S. gracilis but more abundant and widespread. RANGE: Eastern coast of Africa from the Red Sea to Zululand; Seychelles, Aldabras, Comores, Madagascar and Mauritius. Elsewhere, the species ranges widely in the Indo-Pacific, eastwards to Australia. Family CLUPEIDAE HERRINGS Key to the East African Genera 1. Anal fin short, with less than 30 rays; hypomaxillary bone absent: (i) Upper jaw without deep median notch ; tip of lower jaw does not fold completely within the upper (Sub-family Clupeinae) : (a) Few (3-5) fronto-parietal striae ; scale striae complete ; last two anal rays not enlarged HERKLOTSICHTHYS (b) Many (7-15) fronto-parietal striae; scale striae interrupted; last two anal rays markedly enlarged .... SARDINELLA (ii) Upper jaw with a prominent deep median notch ; tip of lower jaw folds completely within the upper (Sub-family Alosinae) . . MILS A 2. Anal fin long, with more than 30 rays; a small toothed hypomaxillary bone present (Sub-family Pristigasterinae) PELLONA The genera Herklotsichthys and Sardinella are frequently confused ; detailed diagnostic features of these two genera have been published by Whitehead (1964c). Hilsa also has been confused in collections with Sardinella, generally with a deep-bodied form such as S. albella. The single species in East African waters, H. kelee, may be distinguished at once by the combination of depth (33-40% in S.L.) and gillraker number (74-177), as all known species of Sardinella from East African waters with a deep body (over 30% in S.L.) have fewer gillrakers (less than 60). Further, in H. kelee the gillrakers are very long and easily visible when the mouth is opened and the scales are firmly adherent; in Sardinella the gillrakers are shorter and the scales are caducous. Sub-famUy CLUPEINAE Genus HERKLOTSICHTHYS Whitley, 1951 Herklotsichthys Whitley, 1951: 67 (proposed to replace Herklotsella Fowler, 1934: 246). Harengula Valenciennes (part., i.e. Indo-Pacific species only), 1847: 201 (Type: Harengula latulus Valenciennes =C/Mpca macrophthalma Ranzani). New world Harengula species are characterised by the presence of a toothed hypomaxilla which is absent in Indo-Pacific fishes formerly placed in this genus (Berry, 1964); Herklotsichthys replaces Harengula for Indo-Pacific species (Whitehead, 1964a). Previously two species of the genus, H. punctatus and H. vittatus, were recognised from the extreme western Indian Ocean and adjacent seas (e.g. Smith, 1949 ; Whitehead, 1965b). The so-called Sardinella- like H. vittatus (Whitehead, 1964c) however, should now be placed in the synonymy of Sardinella melanura (Cuvier) (Whitehead, 1967). Page 88 HERKLOTSICHTHYS PUNCTATUS (Ruppell) Clupea punctata Ruppell, 1837: 78, pi. 21, fig. 2 (Type locality: Red Sea). Ahsa punctata: Gunther, 1866: 123 (Aden, Zanzibar). Clupea venenosa: Gunther, 1868: 449 (Zanzibar). Harengula punctata: Sauvage, 1891 : 493 (Madagascar) ; Regan, 1917 (East Africa) ; Barnard, 1925:114 (Natal); Losse, 1964: 11 (Zanzibar Channel). Sardinella melanura: Smith, 1949-1965: 92, fig. 113 (Natal); Idem, 1963: 8, pi. 4 L (Seychelles). Harengula ovalis: Smith, 1949-1965: 91 (Natal); Baissac, 1951: 126 (Mauritius); Smith, 1955: 307 (Aldabra); Fowler, 1956: 64 (Indo-Pacific specimens); Smith, 1963: 8, pi. 4 G (Seychelles); Sanches, 1963: 20, fig. 3 (Inhaca, Mozambique). Herklotsichthys punctatus: Whitehead, 1965b: 244 (Red Sea, Gulf of Aden); Losse, 1966a: 89 (Zan- zibar); Idem, 1966b: 172 (East Africa ;Dares-Salaam, Zanzibar, Tanga, Mombasa, Malindi, Formosa Bay); Idem, 1966c: 51 (Zanzibar Channel). Two distinct forms, which may prove to be distinct species (see REMARKS), are here described. Key to the East African Forms 1. Body deep, depth 29-33% (mean 31.4%) in S.L. (at 48-85 mm.); a prominent black patch on dorsal fin H. punctatus Form A 2. Body slender, depth 24-29% (mean 26.7%) in S.L. (at 32-120 mm.); dorsal fin dusky, without black patch) H. punctatus Form B HERKLOTSICHTHYS PUNCTATUS Form A STANDARD COMMON NAME: Spotted herring. VERNACULAR NAME: Dagaa (Zanzibar, Tanzanian coast). DESCRIPTION: Based on forty-seven fishes, 47.6-84.8 mm. standard length, from the Zanzibar Channel. Vertebral counts only on six fishes, 76.7-83.6 mm., from Shimoni, Kenya. Dorsal iii-iv (13) 14-15, pectoral i 14-15 (16), pelvic i 7, anal ii-iii (12-13) 14-15 (16), total (14-15) 16-17 (18). Ventral scutes strongly keeled, sharp and exposed, (15) 16-17 pre-pelvic, (10-11) 12-13 post-pelvic, total (27-28) 29-30. Gillrakers, 30-34 (mean 32.39) on lower part of 1st gill arch (at 47.6-84.8 mm. S.L.). Branchiostegal rays 5-6. Scales caducous, about 34-38 (pockets) in lateral series. Vertebrae 41-42 (6 fishes). Depth 29.0-33.3 (mean 31.39), head 27.8-31.6, snout 6.7-8. 3, eye 8.0-10.3, post-orbital 8.9-11.7, upper jaw 12.4-14.0, lower jaw 12.3-13.8 (14.4), pectoral fin (19.0-19.8) 20.1-21.9 (23.0), pelvic fin (11.1-13.9) 14.0-15.8, pre-dorsal 44.8^8.5, pre-pelvic 50.6-55.6, pre-anal 74.5-83.6, caudal peduncle 9.9-1 1 .9. (The means of these values are shown in Table 2.) Body very strongly compressed; ventral profile deeply convex, dorsal profile almost a straight line. Head as long as the maximum body depth or up to 2 % shorter. Snout generally shorter than the eye diameter. Upper jaw much longer than the snout, the maxillary reaching about i to into the eye through the vertical. Operculum about twice as long as broad, the lower margin straight. Sub-operculum rectangular, the posterior margin rounded. Dorsal fin base much nearer to snout than to the caudal base. Pelvic origin almost equidistant from the pectoral base and the anal origin. Scales (from shoulder) with 5-6 complete vertical striae; the posterior scale borders very slightly crenulated or indented, without perforations. Very rarely one or two striae are not complete (as in Sardinella) but interdigitate ; at least four striae are always uninterrrupted. COLOUR: Fresh, scales fallen; dorsal surfaces bluish green, back spotted with numerous small, blue, grey or black blotches which become less distinct at later post-mortem and may fade entirely. A bright yellow or orange humeral patch continued in a lateral stripe to the caudal base, demar- cating dorsal and lateral colouration. Sides, belly and opercular regions silvery or golden. Snout and lower jaw yellowish, tips dusky, occasionally minutely speckled green. Eye silvery with a broad, dorsal yellow or orange band. Dorsal fin orange at base, the anterior rays bright yellow. A prominent black patch on about the first ten dorsal rays. Caudal yellowish at base, dusky or colourless at margins. Other fins colourless. Underwater the yellow or orange markings and the black patch on the dorsal fin were particularly striking and at once distinguished this species from other clupeoids of the area. In alcohol, dorsal surfaces greenish or brown, sides lighter. Black patch on dorsal fin retained and prominent (at all stages of fixation). Caudal yellowish or colourless, margin dusky in some fishes. Snout tip dusky or brown. Rest faded. SIZE: 88.0 mm. S.L. (Zanzibar Channel): 102 mm. F.L., weight 16 gm. (>/)oro/7/ej rypMS Smith ........ O My only record is of one on the outskirts of Tabora on 27th May, 1960. 186 Osprey Pandion haliaetus (\Jixm2iQ\xs) ......... F Family Phasianidae (Game-birds) 191 Coqui Francolin Fra77Co//««^ (Smith) ?? 198 Shelley’s Francolin Franco/mwi 5/!e//ey/ O. Grant . . . . . . ?? Birds seen had the characters of the bird described in Mackworth-Praed and Grant (1955) as F. afer Latham. The various East African forms of this bird are now regarded as races of F. shelleyi (Williams, 1967). 203 Hildebrandt’s Francolin Francolinus hilderbrandti Cabanis . . . . . R + 208 Red-necked Spurfowl Pternistis cranchii (Leach) R + 212 Harlequin Quail Co?M/-n/x T/e/eporpMez Delegorgue ?? 215 Helmeted Guinea-fowl Numida mitrata Pallas F Family Rallidae (Rails, Crakes and Moorhens) 222 Kaffir Rail Rallus caerulescens Gmelin . . . . . . . . ? ? 224 African Crake Crecopsis egregia (Peters) ? ? 225 Black Crake Limnocorax fiavirostra (Swainson) ....... R 228 Lesser Spotted Crake Forzu«a p««7/a (Pallas) ?? 237 Purple Gallinule Porphyrio alba (White) R 238 Allen’s Gallinule Porphyrula alleni (Thomson) R 240 Lesser Moorhen Gallinula angulata Sundevall ....... 242 Red-knobbed Coot Fulica cristata Gmelin Family Balearicidae (Cranes) 245 South African Crowned Crane Balearica regulonim (Bennett) .... Family Otididae (Bustards) 258 Black-bellied Bustard Lissotis melanogaster (Riippell) Family Burhinidiae (Thicknees) 262 Spotted Thicknee Burhimis capensis (Lichtenstein) ...... Family Jacanidae (Jacanas) 264 Jacana Actophilornis africanus (Gmelin) ........ 265 Smaller Jacana Microparra capensis (Smith) ....... Family Charadrudae (Plovers) 266 Ringed Plover Charadrius hiaticula Linnaeus ....... 271 Kittlitz’s Sand-plover Charadrius pecuarius Temminck 272 Three-banded Plover Charadrius tricollaris Vieillot ...... My only record is of one at Kazima Dam on 13th October, 1959. 275 Great Sand-plover Charadrius leschenaultii Lesson ...... 277 Caspian Plover Charadrius asiaticus Pallas ........ 281 Crowned Lapwing Vanellus corona tus (Boddaert) 282 Senegal Plover Vanellus lugubris Lesson ........ 287 Blacksmith Plover Vanellus armatus (Burchell) 291 Wattled Plover Vanellus senegallus (Linnaeus) 293 Long-toed Lapwing Vanellus crassirostris (Hartlaub) ...... 295 Avocet Recurvirostra avosetta Linnaeus My only record is of two at Kazima Dam on 23rd May, 1 962. 296 Black- winged Stilt Himantopus himantopus (Linnaeus) ...... Family Rostratulidae (Painted Snipe) 297 Painted Snipe Rostratula benghalensis (Linnaeus) Family Scolopacidae (Snipes) 298/300 Snipe Capella sp (probably gallinago) 299 Great Snipe Capella media (Latham) 303 Curlew Sandpiper Calidris testacea (Pallas) ....... 305 Little Stint Calidris minuta (Leisler) 307B Broad-billed Sandpiper Limicola falcinellus (Pontopidan) 208 Sanderling Crocethia alba (Pallas) Additional records: one at Kazima Dam on 30th September and 3rd October, 1966. 309 Ruff Philomachus pugnax (Linnaeus) ......... 311 Terek Sandpiper cmerewi (Giildenstadt) Additional record: one at Kazima Dam on 3rd October, 1966. 312 Common Sandpiper Tringa hypoleucos Linnaeus ....... 313 Green Sandpiper Tringa ochropus Linnaeus 314 Wood Sandpiper Tringa glareola Linnaeus ........ 315 Redshank (Linnaeus) ........ 317 Marsh Sandpiper Tringa stagnatilis (Bechstein) 318 Greenshank Tringa nebularia (Gunnerus) ........ 319 Black-tailed Godwit L/'/mom //woM (Linnaeus) ....... 321 Curlew Numenius arquata (Linnaeus) ........ 322 Whimbrel Numenius phaeopus (Linnaeus) ........ Family Phalaropidae (Phalaropes) 323 Red-necked Phalarope Phalaropus lobatus (Linnaeus) Family Glareolidae (Coursers and Pratincoles) 326 Temmink’s Courser Cursorius temminckii Swainson 328 Heuglin’s Courser Hemerodromus cine tus Heuglin ...... 329 Violet-tipped Courser Rhinoptilus chalcopterus (Temminck) ..... 330 Pratincole Glareola pratincola (Linnaeus) Family Laridae (Gulls and Terns) 342 Grey-headed Gull Larus cirrrocephalus Vieillot ....... 361 White-winged Black Tern Chlidonias leucoptera (Temminck) ..... 362 Whiskered Tern Chlidonias hybrida (Pallas) Family Turnicidae (Button-quails) 365 Button-quail Turnix syhatica (Desfontaine) R + O O ?? R + R + R M BM O O O O O O M O O F F M M M M O O M O M OM M OM M M OM OM OM O M M F Page 124 Family Columbidae (Pigeons) 379 Speckled Pigeon Columba guinea Linnaeus .... 386 Red-eyed Dove Streptopelia semitorquata (Riippell) 388 Ring-necked Dove Streptopelia capicola (Sundevall) 392 Laughing Dove Stigmatopelia senegalensis (Linnaeus) 393 Namaqua Dove Oena capensis (Linnaeus) 397 Emerald-spotted Wood-dove Turtur chalcospilos (Wagler) 401 Green Pigeon Treron australis (Linnaeus) Family Cuculidae (Cuckoos) 413 Great Spotted Cuckoo C/awamr (Linnaeus) 41 5 Black-and- White Cuckoo Clamator jacobinus (Boddaert) 417 Didric Cuckoo Chrysococcyx capias (Boddaert) .... 41 8 Klaas’ Cuckoo Chrysococcyx klass (Stephens) .... 423 White-browed Coucal Centropus superciliosus Hemprich & Ehrenberg Family Musophagidae (Turacos) 434 Violet-crested Turaco Gallirex porphyreolophus (Vigors) 441 Bare-faced Go-Away-Bird Gymnoschizorhis personata (Riippell) Family Psittacidae (Parrots) 449 Brown Parrot Foicephalus meyeri (Cretzschmar) .... Family Coraciidae (Rollers) 457 European Roller Coracias garrulus Linnaeus .... 460 Lilac-brested Roller Coracias caudata Linnaeus .... 461 Rufous-crowned Roller Coracias naevia Daudin .... 463 Broad-billed Roller (Muller) Family Alcedinidae (Kingfishers) 465 Pied Kingfisher Ceryle rudis (Linnaeus) ..... 470 Malachite Kingfisher Corythornis cristata (Pallas) 471 Pigmy Kingfisher /ipMwa (Boddaert) .... 476 Brown-hooded Kingfisher Halcyon albiventris (Scopoli) 477 Grey-headed Kingfisher Halcyon leucocephala (Miiller) 479 Striped Kingfisher Halcyon chelicuti (Stanley) .... Family Meropidae (Bee-eaters) 481 Bee-eater Merops apiaster Linnaeus ...... 488 Little Bee-eater Melittophagus pusillus (Muller) .... 496 Swallow-tailed Bee-eater Dicrocercus hirundineus (Lichtenstein) Family 518 Family Family Family 575 584 594 600 Bucerotidae (Hornbills) Grey Hornbill Tockus nasutus (Linnaeus) ........ F Red-billed Hornbill Tockus erythrorhynchus (Temminck) F Von Der Decken’s Hornbill Tockus deckeni (Cabanis) ...... F Crowned Hornbill Tockus alboterminatus (Buttikorfer) F Upupidae (Hoopoes) South African Hoopoe Upupa africana Bechstein BM Phoeniculidae (Wood-hoopoes and Scimitar-bills) Green Wood-hoopoe P/70p«/cm/m5 (Miller) . . . . . .R + Scirmtax-biW Rhinopomastus cyanomelas iy 'iQiWoi) R + Strigidae (Owls) Barn Owl Tyto alba (Scopoli) R Scops Owl Otus scops (Linnaeus) . . . . . . . . . ? ? r'+ R + R+ R+ O White-faced Scops Owl Otus leucotis (Temminck) Spotted Eagle-owl Bubo africanus (Temminck) Caprimulgidae (Nightjars) Fiery-necked Nightjar Caprimulgus fervidus Sharpe Freckled Nightjar Caprimulgus tristigma Riippell . Gabon Nightjar Caprimulgus fosii Hartlaub Pennant-winged Nightjar Semeiophorus vexillarius Gould My only record is of one near Tabora on 28th December, 1959. Capitonidae (Barbets) Black-collared Barbet Lybius torquatus (Dumont) Spotted-flanked Barbet Triclwlaema lacrymosum Cabanis Red-fronted Tinker-bird Pogoniulus pusillus (Dumont) Levaillant’s Barbet Trachyphonus vaillantii Ranzani My only record is of one near Igombe Dam on 12th June, 1960. R + R + M O Page 125 Family Indicatoridae (Honey-Guides) 605 Black-throated Honey-Guide Indicator indicator (Sparrman) R Family Picidae (Woodpeckers) 623 Cardinal Woodpecker Dendropicos fuscescens (Vieillot). . . . . . R Family ApooroAE (Swifts) 643 Little Swift Apus affinis (Gray) . 646 Palm Swift Cypec/ora//5 (Jardine & Selby) . . . F Family Dicruridae (Drongos) 1088 Tixongo Dicrurus adsimilis R-f Family Prionopidae (Helmet-shrikes) 1090 Straight-crested Helmet-shrike Pr/oMopi p/wwam (Shaw) R + 1095 Retz’s Red-billed Shrike Sigmodus retzii (Wahlberg) F Around Tabora the only place where I saw this species was in the Simbo Forest Reserve, fifteen miles from Tabora on the Nzega road, where it appeared to be not uncommon. Family Lanitoae (Shrikes) 1097 White-crowned Shrike Pwrocep/ja/wi Smith ?? Although common on the Wembere this species is absent from the immediate vicinity of Tabora. I saw it in an mbuga between Itaga and Mambali, and in another on the way to Sikonge. Neither of these were visited regularly. 1098/1099 Brubru M/AM5 sp ?? 1103 Lesser Grey Shrike wirtor Gmelin ........ M 1104 Fiscal LflWMi cn//am Linnaeus R I saw this species in two places only — one near the hospital, the other near the Railway Training School — both of which appeared to support one pair each. 1112 Red-backed Shrike PawMs co//M/-io Linnaeus M Earliest date: 1st November, 1965; latest date: 17th April, 1966. 1118 Magpie-shrike t/ra/e5?e.v (Jardine) ?? Seen only on the periphery of the area in the same places as Eurocephalus anguitimens. 1121 Slate-coloured Boubou Paw/onMi /wneftm (Hartlaub) R + 1128 Black-backed Puff-back Dryoscopwi c«6/a (Shaw) R 1133 Black-headed Bush-shrike Pc/zapra (Linnaeus) R-h 1134 Brown-headed Bush-shrike Pc/zapra a{/ifra//s (Smith) . . . . . . R + 1138 Sulphur-breasted Bush-shrike C/z/orap/zozzezzi: iz/Z/z/z-popecmi (Lesson) . . . R 1144 Grey-headed Bush-shrike Mfl/acozzo?Mi 6/anc/zotz Stephens . . . . . R + Family Paridae (Tits) 1155 White-breasted Tit Paz-zzj n/Wvczzm’i Shelley R + 1160 African Penduline Tit ^zzt/zoico/zzzs caz-o/z (Sharpe) ...... F Family Oriolidae (Orioles) 1164 Golden Oriole Ono/«5 orzo/zzs (Linnaeus) . OM 1165 African Golden Oriole Oz-zo/m5 AMz-orz/i Vieillot ?? 1167 Black-headed Oriole Oz-zo/zzi /zzrvato Lichtenstein R + Family Corvidae (Crows) 1172 Pied Crow Coz-vzzi zz//zzz5 Muller . . R + Family Sturnidae (Starlings) 1182 Wattled Starling Cz-eato/z/zoz-a czzzerea (Menschen). ...... M 1183 White-winged Babbling Starling Aeoczc/z/zz ^zzzzzzz-a/zi (Boca ge) . . . . F Seen only in Brachystegia woodland. 1184 Violet-backed Starling Czzzzzyz-zczzzc/zzj /ezzco^zziter (Boddaert) . . . . B? A good deal of local movement certainly occurs but the species is probably resident in the area as a whole. 1188 Blue-eared Glossy Starling Lamprocolius chalybaeus (Hemprich & Ehrenberg) . F Family Zosteropidae (White-eyes) 1219 Yellow White-eye Zoj/ez-o/zi ipzzp^zz/ezziz^ Bonaparte ...... M Page 127 Family Nectariniidae (Sunbirds) 1233 Beautiful Sunbird AfecMW/;/a/»M/c/!e//a (Linneaus). . . . . . . ?? 1245 Mariqua Sunbird O/myw Smith ....... R 1251 Variable Sunbird Cinnyhs venustus (Shaw & Nodder) R + 1263 Scarlet-chested Sunbird (Linnaeus) .... R-t- 1271 Collared Sunbird co//am (Vieillot) . F Family Ploceidae (Weavers, Sparrows, Waxbills and allies) 1286 White-headed Buffalo-weaver i)/neme///a (//Mcme/// (Riippell) . . . . O My only record is of a single bird near Kazima Dam on 27th February, 1960. 1291 Grey-headed Social-weaver P5cw^7o/7/.gr/7fl (Bonaparte) .... R-|- 1 300 Grey-headed Sparrow Passer griseus ( Vieillot) . . . . . . . R -j- 1 306 Chestnut Sparrow Sorella eminibey Hartlaub . M 1311 Speckle-fronted Weaver /ro«?fl/7s (Daudin) R-f 1313 Layard's Black-headed Weaver F/ocet/i- m.gnce/7^ (Layard) . . . . . B 1319 Masked Weaver P/ocet/i (7;/e;7n7T//Mj Riippell ....... B 1359 Red-headed Weaver we/anotK (Lafresnaye) . . . . . R + 1360 Red-billed Quelea Qwe/cfl (Linnaeus) ....... M 1362 Cardinal Quelea cW/rtfl/w (Hartlaub) M 1 365 Black-winged Red Bishop Euplectes hordeacea (Linnaeus) . . . . . R -f 1372 Yellow-mantled Widow-bird Co//MS’po5ie/' (Gmelin) . . . . B? 1373 White-winged Widow-bird Co/fnspflii'cr fl/boHo/flto (Cassin) . . . . . B? 1379 Bronze Mannikin cwcw//ato Swainson . . . . . . R-f 1402 Cut-throat Amadina fasciata (Gmelin) ........ F 1403 Quail-finch atncu/Zw (Vieillot) . . . . . . . ?? 1410 Green-winged Pytilia Pyt/Vm (Linnaeus) ....... R + 1413 Red-billed Fire-finch (Linnaeus) R + 1420 Crimson-rumped Waxbill Estrilda rhodopyga Sundevall . . . . . F 1421 Zebra Waxbill £V7-7V7fa 577fe^flva (Vieillot) ?? 1427 Black-cheeked Waxbill Estrilda erythronotos (Vieillot) ...... R 1431 Red-cheeked Cordon-bleu (/rae^777t/7775 Z7e77c^fl/77i (Linnaeus) . . . . . R-b 1435 Purple Indigo-bird //v/7oc/7C7-a M/t7-fl777a;777a (Gmelin) . . . . . B? 1441 Pin-tailed Whydah F7Vfaa 7?7ac7-OM7'a (Pallas) . . . . . . . B? 1442 Steel-blue Whydah K/t/aa /7ypac/7m‘77a Verreaux . . . . . . . B? 1444 Paradise Whydah 5're^a77777a /7a7-ar/7^aea (Linnaeus) B? Family Fringillidae (Finches) 1448 Yellow-fronted Canary 5'm77M^ 777aza77767CM5 (Muller) . . . . . . R + 1461 Streaky Seed-eater Serinus striolatus (Riippell) ....... O Family Emberizidae (Buntings) 1469 Golden-breasted Bunting E7776e7-7za 77av7Vent77i Stephen . . . . . R-t- 1476 Cinnamon-breasted Rock-bunting P7777g77/a77a ?a/7ap/s7 A. Smith . . . . R + Additional Species recorded from the Ugalla River Game Reserve (U) and the Wembere (W): Family Struthionidae (Ostrich) 1 Ostrich Strutio camelus Linnaeus ......... WU The species is uncommon at the Ugalla where my only record is of two near Senga on 15th November, 1964. Family Ardeidae 38 Yellow-billed Egret Mesophoyx intermedius (Wagler) ...... W Family Falconidae 106 Ruppel’s Griffon Gyps ruppellii (Brehm) U 108 Lappet-faced Vulture PoT-^05 /777c/7e/70t//s (Forster) ...... WU 129 Pygmy Falcon P7-flx ^C7777707g77flto (A. Smith) ...... W 146 Martial Eagle Poleniaetiis bellicosus (Daudin) ....... W Family Phasianidae 195 Crested Francolln pT-a77co/7777/^ 5ep/777e77fl (Smith) . W 209 Grey-breasted Spurfowl Pternistis rufopictus Reichenow . . . . . W Family Rallidae 223 Corn Crake Crex crex (Linnaeus) U Family Heliornithidae (Finfoots) 243 Finfoot Podica senegalensis Hartlaub ........ U Family Otididae 250 Kori Bustard Ardeotis kori (Burchell) ........ W Family Burhinidae 263 Water Thicknee Burhinus vermicidatus (Cabanis) ...... U D Page 128 Family Charadrddae 279 Grey Plover Charadrius squatarola (Linnaeus) ....... W 286 Spur-winged Plover Vanellus spimsus (Linnaeus) ...... W Family Scolopacidae 310 Turnstone Arenaria interpres (Linnaeus) ........ W Family Glareolidae 327 Two-banded Courser Hemerodromus africanus (Temminck) . . . . W Family PTEROCLioroAE 369 Chestnut-bellied Sandgrouse Pterocles exustus Temminck & Tangier . . . W 370 Black-faced Sandgrouse Eremialector decoratus (Cabanis) . . . . . W 372 Yellow-throated Sandgrouse Eremialector guttiiralis (Smith) . . . . W Family CoLUMBroAE 387 Mourning Dove Streptopelia decipiens (Finsch & Hartlaub) . . . . . W Family Cuculidae 422 Senegal Coucal Centropus senegalensis (Linnaeus)* U Family Psittacidae 454 Fischer’s Lovebird Agapornis fisclieri Reichenow ....... W Family Alcedinidae 466 Giant Kingfisher Megaceryle maxima (Pallas) ....... U Family Meroptoae 482 Madagascar Bee-eater Merops superciliosus Linnaeus WU Birds seen had the characters of M.s. persicus Pall., regarded as a separate species, M. persicus, by some authorities, e.g. Williams (1963). Family BucERorroAE 515 Ground HomhWl Bucorvus leadbeateri (yigOTs) WU Family PnoENicULroAE 527 Northern Scimitar-bill Rhinopomastus minor (Ruppell) W Family Strigidae 544 Verreaux’s Eagle-owl Bubo lacteus (Temminck) . U Family CouroAE 568 Blue-naped Mousebird Colius macrourus (Linnaeus) ...... W Family Capitonudae 603 D’Amaud’s Barbet Trachyphonus darnaiidii (Prevost & Des Murs) . . . W Family Apodidae 640 Alpine Swift Apus melba (Linnaeus) ........ W Family Turdidae 879 White-headed Black Chat Thamnolaea arnotti (Tristram) . . . . . U 909 Spotted Morning Warbler Cichladusa guttata (Heuglin) . . . . W Family Sylvudae 946 African Reed Warbler Acrocephalus baeticatus (Vieillot) W 988 Karamoja Apalis Apalis karamojae (Van Someren) ...... W Family Hirundintoae 1066 Grey-rumped Swallow Hirundo griseopyga Sundevall W Family Sturnidae 1199 Ashy StarMug Cosmopsarus unicolor SheWey ....... W 1216 Superb Starling Spreo superbus (Ruppell) W 1218 Red-biWed Oxpecker Bulphagus erytliorhynciws (Stanley) ..... WU Family Ploceidae 1285 Red-billed Buffalo-weaver Bubalornis niger Smith ...... W 1 290 Rufous-tailed Weaver Histurgops ruficauda Reichenow W 1432 Blue-capped Cordon-bleu Uraegintlius cyanocephalus (Richmond) . . . W BREEDING DATA The months in which eggs have been recorded or deduced from records of young birds are shown in Table 3 which is summarized in Table 4. In general, signs of nest building have not been used to deduce the occurrence of eggs though this has been done for two species — Swamp Warbler and Drongo. Page 129 Table 3 MONTHS IN WHICH EGGS HAVE BEEN RECORDED OR DEDUCED Species Months in which eggs recorded or deduced (a) Non-Passerines j F M A M J J A s o N D 1 Little Grebe + + 2 Long-tailed Cormorant + + 3 Darter + + 4 White-headed Vulture + 5 Black-shouldered Kite + 6 Black-chested Harrier-eagle + 7 Fish Eagle + + 8 Gabar Goshawk + 9 Hildebrandt’s Francolin * + + + 10 Red-necked Spurfowl * * + 11 Lesser Moorhen + + + + 12 Spotted Thicknee + + 13 Jacana + + + + + 14 Kittlitz’s Sand-plover * + + + + + 14 Crowned Lapwing + + + + + 16 Senegal Plover + + + 17 Blacksmith Plover + + 18 Wattled Plover + + + + 19 Temminck’s Courser + + + + + 20 Two-banded Courser + + + + + + 21 Heuglin’s Courser + + + + + + 22 Violet-tipped Courser + + + 23 Pratincole + + + 24 Black-faced Sandgrouse + 25 Ring-necked Dove + + + 26 Laughing Dove + + + * + + + + 27 Namaqua Dove + 28 Emerald-spotted Wood-dove + + + + 29 White-browed Coucal + + + + + + 30 Bare-faced-Go-Away-Bird + + 31 Pied Kingfisher + + + 32 Pigmy Kingfisher + 33 Grey-headed Kingfisher + + 34 Little Bee-eater + + 35 African Hoopoe + 36 Green Wood-hoopoe * + 37 Scimitar-bill + 38 Spotted Eagle-owl + + 39 Fiery-necked Nightjar + 40 Freckled Nightjar + 41 Gabon Nightjar + 42 Black-collared Barbet ? + 43 Spotted-flanked Barbet + + 44 Little Swift + + + Table 3 {Continued) Species Months in which eggs recorded or deduced {b) Passerines F M A M J J A S o N D 45 Broadbill + + 46 Fischer’s Sparrow-lark -f + + 47 African Pied Wagtail + + + + 48 Richard’s Pipit + 49 Yellow-throated Longclaw + + 4- 50 Arrow-marked Babbler * -f- + + + 4- 51 Black-lored Babbler + 52 Dark-capped Bulbul + + + + + + 53 Yellow-bellied Greenbul + + + 4- 54 Grey Flycatcher + + + + + + + + 55 Silver-bird + + 56 Chin-spot Puff-back Flycatcher + + 4- 57 Paradise Flycatcher + 4- 4- 58 White-browed Robin-chat + + 59 Red-backed Scrub-robin + + + + 4- 60 Swamp Warbler + 61 Grey-wren Warbler + + + + + 4- 62 Black-breasted Apalis + + 4- 63 Red-faced Crombec + 64 Grey-backed Camaroptera + 4- 65 Zitting Cisticola + 66 Rattling Cisticola + 67 Winding Cisticola + + + + 68 Tabora Cisticola + 69 Tawny-flanked Prinia + + + 70 Pearl-breasted Swallow + + + + + + 71 Mosque Swallow + + + 4- 72 Striped Swallow 4- 73 Drongo + 74 Straight-crested Helmet-shrike + 75 Slate-coloured Boubou + + + + + + 4- 76 Black-headed Bush-shrike + + + + + 4- 77 Brown-headed Bush-shrike + + + -f + + 78 Grey-headed Bush-shrike 4- 79 White-breasted Tit + + 80 Black-headed Oriole + + 4- + 81 Pied Crow + -h 82 Superb Starling * * * * * + 83 Variable Sunbird + -f- + + + + 84 Scarlet-chested Sunbird + 85 Grey-headed Social Weaver + + -h 86 Grey-headed Sparrow + + + + 87 Speckle-fronted Weaver + + + + + 88 Layard’s Black-headed Weaver * -t- + * * 89 Masked Weaver + 90 Red-headed Weaver + + 4- + 91 Black-winged Red Bishop + -h -1- + Page 131 Table 3 {Continued) Species Months in which eggs recorded or deduced {b) Passerines {Continued) JFMAM JJASOND 92 Bronze Mannikin + 93 Green-winged Pytilia + -1- + + -f- + ... + 94 Red-billed Fire-finch + + 95 Red-cheeked Cordon-bleu + + + + + 96 Yellow-fronted Canary ... + + + 97 Golden-breasted Bunting + + + + + + 98 Cinnamon-breasted Rock-bunting + + ... + -t- personal records; * records from Thomas (1960) Table 4 NUMBERS OF SPECIES BREEDING IN EACH MONTH The systematic notes that follow amplify the information given in Table 3, in particular giving full data for those species for which I have few records. LONG-TAILED CORMORANT AND DARTER In 1960 there was a mixed colony of these two species nesting in dead trees in the centre of Igombe Dam. I was not able to visit this colony in subsequent years as no boat was available. WHITE-HEADED VULTURE My sole record is of a bird apparently incubating on 25th October, 1966. BLACK-SHOULDERED KITE Coition followed by the carrying of sticks was seen on 15th May, 1961. The nest was later located in the District Commissioner’s garden. Page 132 BLACK-CHESTED HARRIER-EAGLE Two eyries were found, about 5 miles apart. Birds were seen apparently incubating in October 1961 and 1966. FISH EAGLE One apparently incubating in an eyrie at Igombe Dam in late June and early July 1965. GABAR GOSHAWK A nest found by Mr. I. H. Dillingham contained young in late September 1961. hildebrandt’s francolin Occurs on and around rocky hills. C/8 found on 16th May, 1965 hatched between 3rd and 4th June. A possibly incomplete C/4 found on 12th May, 1966 had been eaten by 14th May. Downy chicks were seen in late July 1966, and young with quills just appearing on 16th September, 1966. RED-NECKED SPURFOWL I failed to find any nests of this very common species. Mr. H. Saidi found what was probably a nest of this species in late April 1961. LESSER MOORHEN A nest containing one egg on 14th March, 1960 had the surrounding reed stems worked into a canopy over the nest. The bases of the stems had been brought together by pieces of rush “tied” in “half-granny” knots. This nest was flooded out. SPOTTED THICKNEE I found three nests: C/2 on 2nd October, 1960 hatched on 17th October; C/1 on 18th October, 1961; C/1 on 10th October, 1965. The latter nest was among tall tussocks of dead grass on the Wembere at least a mile from the nearest scrub. JACANA The latest date on which I saw eggs was 27th October, 1960 at Kazima Dam; at this time of year there is a considerable influx of non-breeding Jacanas. On 5th June, 1962 a walking parent was seen carrying a chick tucked into the flank feathers under one wing from which the chick’s legs dangled down and out. C/3 and C/4 appear to be equally common. KITTLITZ’S SAND-PLOVER The margins of all three Tabora dams have been used for breeding when sufficiently sandy patches were present. It is very common on the Wembere where no nests or small young were seen during October, November and Januaiy visits. All nests found had C/2. Observations from a hide indicated that eggs are covered whenever the bird leaves the nest and that covering is not confined to departures stimulated by the presence of potential enemies. On returning to the nest the bird makes gentle pro- bing movements with its bill as if to locate the exact position of the eggs which are then uncovered by kicking movements with the breast pushed down onto the pile of sand covering the eggs. CROWNED LAPWING I found no nests around Tabora and consider it probable that it does not do so, though there are a few suitable areas. It is very common on the Wembere but considerably less so at the Ugalla where nesting appears to start later, C/2 appears to be the most frequent clutch size. Page 133 SENEGAL PLOVER All breeding records are from the Ugalla where C/3 appears to be commoner than C/2. Most clutches appear to be laid in October. BLACKSMITH PLOVER All my breeding records are from the Wembere where nests were usually sited within 20 yards of water. On 10th July, 1962 I was photographing a bird at a nest with chipping eggs over which ants were swarming. One of the eggs had a large hole through which the ants were entering to attack the chick. The parent showed great agitation at the presence of the ants and eventually picked up the egg which the ants were entering, carried it to the water’s edge and dropped it in the water. Normal incubation was then resumed. WATTLED PLOVER During the rainy season flocks of up to about 50 birds occur in and around paddy fields near Tabora. All my breeding records are from the Ugalla where my latest record is of C/4 on 1st November, 1964. C/4 appears to be the usual clutch, though I have seen several broods of three chicks. temminck’s courser All breeding records are from the Wembere and Ugalla. Both sexes incubate. TWO-BANDED COURSER All records from the Wembere where the species is common. About forty nests, each with C/1, were found in all. Keast and Marshall (1954) and Serventy and Marshall (1957) have produced evidence that the breeding of Australian desert birds is geared to rainfall and that such birds breed only when adequate rain falls. Winterbottom visited a district of the western karoo area of South Africa where a five year drought had ended a month earlier and another district where the drought had continued. He found Two-banded Coursers with eggs and/or young in the first locality but no sign of sexual activity in the birds of the second, though it is not stated whether this species was present in both areas. From these and similar observations Winterbottom and Rowan (1968) concluded that this courser is one of several arid country species whose breeding is stimulated by the onset of rain. Maclean (1967) found that, in the Kalahari, breeding occurred throughout the year with no indications of its onset being stimulated by rain. My Wembere obser- vations refer to the dry season as the area is virtually inaccessible during the rains. I cannot, therefore, say whether breeding takes place in well drained areas kept bare by heavy grazing. However, this possibility is supported by my finding a nest in the rainy season on 16th April, 1965 in Lake Manyara National Park. Although many more observations are needed before any definite conclusions can be drawn, it would appear that Winterbottom’s apparent correlation between rainfall and breeding might well be the result of insufficient sampling during dry conditions. heuglin’s courser Common around Tabora and on the Wembere. A total of fifteen nests were found, all with C/2. In all nests the eggs were about three quarters buried in the substrate. No birds observed from a hide were seen attempting to turn the eggs during incu- bation which is carried out by both sexes. Page 134 At a nest found on 3rd September, 1966 the non-sitting bird could usually be found within 20 yards of the nest accompanied by two fledged young. These remained in the close vicinity of the nest until the eggs hatched on 25th September, after which I lost contact with the group. VIOLET-TIPPED COURSER This species appears to be migratory, arriving in the Tabora area in the second half of May and early June. At this time of year the birds can often be heard calling in flight over the town and, at night, may be seen in gardens and even the streets of the town. My four breeding records are: At the Ugalla, C/3 found on 1st September, 1963 hatched on 20th September; C/2 found on 19th September, 1963 hatched on 23rd September; three young about 2^ weeks old seen on 23rd October, 1964; at Tabora C/3 found on 9th October, 1966 hatched before 16th October. The Ugalla nests were in large open glades though within 20 yards of patches of trees and bushes; the Tabora nest was in open secondary growth such as is often used by Heuglin’s Courser. PRANTINCOLE All breeding records are from the Wembere. BLACK-FACED SANDGROUSE My sole record is of a pair with three fledged young on the Wembere on 12th July, 1961. NAMAQUA DOVE A nest with two well grown young at Kazima Dam on 9th October, 1963 is my only record. BARE-FACED-GO-A WAY-BIRD Two records only; C/2 on 13th August, 1960, and a nest with at least one small young on 20th June, 1965. PIGMY KINGFISHER An occupied nest, probably with eggs, at Tabora on 18th November, 1962. One seen carrying food on 28th November, 1966. GREY-HEADED KINGFISHER Nests with young birds 18th November, 1962 and 8th December, 1965. Nest with eggs 16th October, 1966; eggs hatched on 4th or 5th November. LITTLE BEE-EATER Eggs must be laid in late September and October as young are in the nests throughout November. I have no records that suggest egg laying before late September or after late October and I think in Tabora the species has a very restricted breeding season. SOUTH AFRICAN HOOPOE My only record is of a nest with large young in the hollow branch of an old mango tree in a Tabora garden on 13th October, 1966. GREEN WOOD-HOOPOE A nest with young on 5th November, 1959 is my only record. SCIMITAR-BILL A nest with large young found by Mr. J. A. S. Mackenzie-Grieve in November 1 964 is my only record. Page 135 FIERY-NECKED NIGHTJAR Two young, almost fledged on 20th November, 1966, were almost certainly of this species and constitute my only breeding record. FRECKLED NIGHTJAR C/2 laid on bare rock among burnt grass tussocks and Vellozia on a rocky hill was found on 16th September, 1966. These eggs hatched on the 24th and 25th Sep- tember, but the chicks were attacked and killed by ants as soon as they emerged. GABON NIGHTJAR A male dissected on 2nd June, 1966 had very small testes. C/2 found on 10th Sep- tember, 1966 hatched on 18th September. BLACK-COLLARED BARBET A pair were showing interest in a hole three feet above the ground on 24th April, 1966 but their behaviour did not conclusively indicate breeding. A pair probably had eggs on 25th October, 1966 as the presumed female stayed inside the hole while the mate vigorously chased two Black-throated Honey-guides that were showing great interest in the barbet’s nest. The nesting stump was destroyed by fire. BROADBILL I saw this species only in the Quarry Area where it was remarkably elusive. A nest with C/2 was found on 27th November, 1964, and another being built on 16th November, 1966; this contained one egg on 30th November and C/2 on 2nd December, 1966. BLACK-LORED BABBLER My only record is of a nest found by Mr. I. H. Dillingham on 12th April, 1962 con- taining C/4, all of which had hatched by 18th April. Communal feeding of the young was observed at this nest (Reynolds 1965 b). DARK-CAPPED BULBUL Although the peak of breeding appears to be in late October and November I suspect that smaller numbers may nest throughout the year. On 7th June, 1966 there was an occupied nest in a Jacaranda too high for inspection ; presumably the same pair had built another high nest in a neighbouring tree that was occupied on 15th October, 1966. Two nests found in November 1959 each contained a blue egg presumed to have been laid by a cuckoo {Clamator sp ?) ; both were lost to predators. SILVER-BIRD This species is common on the Wembere; a few pairs are found in mbugas around Tabora where fledged young were seen on 4th February, 1965. On the Wembere a bird was believed to be incubating in the old nest of a Rufous -tailed Weaver on 8th November, 1964, and C/2 was found in another Rufous-tailed Weaver’s nest on 7th January, 1965. PARADISE FLYCATCHER All nests in Tabora were found in the Quarry Area where building usually starts during the second week of November. Egg losses were very high but I failed to obtain evidence for either repeat nests or second broods. On one occasion a female was observed singing near the nest. CAPPED WHEATEAR Territorial behaviour involving, apparently, paired birds can be observed on the open Page 136 country near the abattoir during May and June but I never obtained evidence that the birds were breeding. WHITE-BROWED ROBIN-CHAT This is rather an uncommon bird around Tabora though three or four pairs occur in the Quarry Area where all my nests were found. The first nest found contained C/2 on 17th November 1964. In 1965 a new nest was being built on the remains of this nest on 29th October and contained C/2 on 2nd November. In 1966 the nest in this territory was about two yards from the above site and contained C/2 on 10th November, hatching on 14th November. In another territory C/3 was found on 19th December, 1965. SWAMP WARBLER A pair were carrying nesting material into tall bulrushes at Kazima Dam on 7th April, 1963. BLACK-BREASTED APALIS A nest being built on 7th April, 1963 contained one egg on 13th April and C/2 on 21st April. Young were being fed in a nest about 20' above the ground on 5th January, 1966. Another pair were building at this time but deserted before laying. RED-FACED CROMBEC My only record is of a nest found with C/2 on 13th May, 1962 by Mr. I. H. Dillingham. GREY-BACKED CAMAROPTERA The behaviour of the birds suggests that breeding is confined to the period November to February. Mr. Dillingham showed me a nest with C/2 in late November 1961; I found a nest with C/2 on 2nd January and another with two young on 5th January, 1966. ZITTING CISTICOLA Mr. Dillingham found a nest in March 1962. In 1966 I observed “zitting” from mid- January until late April but failed to find any nests. RATTLING CISTICOLA Nest with three young on 6th March, 1962. C/3 found on 17th Feburary, 1965; the young flew between the 4th and 5th of March. TABORA CISTICOLA Two nests found with young in the first week of March, one in 1965 and the other in 1966. PEARL-BREASTED SWALLOW The only place where I saw this species was in the immediate vicinity of Tabora Boys’ School where there were two nesting sites. While it remains intact the same cup is used for successive broods; it is also used for roosting. The dry season broods are of interest, as Moreau (1964) implies that Hirundo griseopyga is the only species of swallow that nests in the dry season. STRIPED SWALLOW In Tabora, houses are not used for nesting. A few nests are built on the underside of large rocks supported by other boulders but the main nesting site is in drainage culverts passing under the roads. DRONGO This species is fairly common around Tabora but the only evidence that I have for breeding is a pair building on 5th November, 1959. Page 137 STRAIGHT-CRESTED HELMET-SHRIKE The only nest found had C/2 on 8th October, 1966. The eggs hatched on 20th October but the young had been eaten by the 21st. Three birds were in attendance at this nest. GREY-HEADED BUSH-SHRIKE Only two nests found, both in a Strychnos about 9' above the ground. The first was found on 2 1st November, 1964 with three small young and the second on 14th Novem- ber, 1965 also with three small young. SCARLET-CHESTED SUNBIRD Additional records for this species that undoubtedly has a longer season than indicated in Table 3 were unfortunately lost. SPECKLE-FRONTED WEAVER In Mackworth-Praed and Grant (1955) the nest of this weaver is described as “. . . a roughly made sphere of grass with a side entrance usually suspended from the end of a bough ten to fifteen feet from the ground”. I have seen many nests of this species and none of them has corresponded with this description. The nest is a rather coarse domed structure with a long entrance funnel at one side placed along, but not dependent from, a branch of, usually, an Acacia. The chamber is lined with feathers. Around Tabora most nests are between 6 ' and 9 ' above the ground. BLACK-WINGED RED BISHOP The onset of breeding is closely correlated with the availability of suitably long grass for nest sites which in turn is influenced by both the date when the rains start and the amount of rain in November, December and January. C/3 is the commonest clutch size. In 1960 a nest was found in which the eggs were blue with fine black spots, and in 1963 one in which the eggs were white. The young are fed mostly by regurgitation but occasionally small insects (usually larvae) are brought. GOLDEN-BREASTED BUNTING C/2 appears to be about three times as common as C/3. CINNAMON-BREASTED ROCK-BUNTING During the breeding season the bird is confined to rocky hills. Song occurs from Janu- ary to August though no nests have been found before April (however, I have not watched intensively in January, February and March so nesting may well start earlier). Most hills have only enough suitable habitat to support one or two pairs but one near Inara (about 9 miles from Tabora) had five or six pairs. Two nests were only about 20 yards apart. C/3 is the usual clutch; two C/4 have been found but in each case only three eggs hatched. Egg and nestling losses are high, snakes probably being the main predators though fires are also a hazard. Both parents feed the young by regurgitation : I never saw food being brought in the bill. DISCUSSION Moreau (1950), surveying the rather inadequate data then available on the breeding seasons of African birds, concluded that five breeding season patterns could be recog- nised : 1 . the big raptors and scavengers, laying in the middle of the dry season ; 2. some ground birds, such as nightjars, which tend to lay in the rains in areas where grass fires are not an extensive risk, but elsewhere to lay in the dry season after the grass fires are over ; 3. the birds dependent on tall grass which necessarily wait until well after the rains have started, both for nesting sites and seeds; Page 138 4. the water birds which tend to nest when watery habitats have increased in extent as a result of the rains ; 5. most other birds (nearly all dependent wholly or mainly on insects) which tend to nest early, beginning with the pre-rains flush of new foliage and insects. In Moreau (1966) he concludes that the new data published since then in general support the generalisations quoted above. My opinions as to how my own data fit these generalisations are as follows: Category 1 My data are far too scanty to warrant much discussion. The records for Black- shouldered Kite, Fish Eagle and Gabar Goshawk (though two of these hardly count as “big” raptors) conform but the Black-chested Harrier-eagle and White-headed Vulture appeared to lay towards the end of the dry season. Furthermore in Iringa (where I am at present working) I have found Kites and Tawny Eagles starting to nest at the very end of the dry season. Category 2 I think that the individual requirements of ground nesting birds are so diverse that generalisation is apt to obscure too many ecological differences. “Ground nesting” is also a vaguer term than might at first be thought : most workers would agree that it would apply, for example, to both Heuglin’s Courser and the Yellow-throated Long- claw but the siting of the latter’s nest in clumps of grass makes it less vulnerable to flooding than the former’s. As regards plovers, especially Senegal Plover and Crowned Lapwing, and coursers, especially the Violet-tipped and Temminck’s, I would say that suitable conditions for nesting can be created as a result of burning, or of over- grazing, or a combination of both factors. I consider that this can be concluded from a comparison of conditions on the Wembere (heavily overgrazed) and the Ugalla (subject to fires from July onwards); Crowned Lapwings breed from May onwards on the Wembere while nests have only been found in August and September on the Ugalla. Category 3 My experience with the Black-winged Red Bishop and the cisticolas, especially galac- totes is in complete conformity with the generalisation. Category 4 I consider this a rather unsatisfactory category on much the same grounds as Category 2: there are great differences in the nesting sites of water birds. My rather limited data for this group do, however, tend to conform. Category 5 There is no doubt that a number of species start nesting well before the onset of the rains, though I think it equally true that the timing of many of these “early” nests is such that the first heavy storms tend to coincide either with the presence of young in the nest or with recently fledged young. Although the leaf flush is so obvious my (admittedly somewhat subjective) impression is that there is no marked flush of insects, or for that matter arthropods in general, until after at least three or four heavy down- pours. This is certainly true for beetles, termites, millipedes, centipedes and also mollusca. As regards evidence for a widespread pre-rains start of nesting I would conclude that, in general, my data do not support the generalisation. This statement requires amplification as I have recorded 27 species with eggs in October (Table 4). Three of these — Heuglin’s and Violet-tipped Courser and Spotted Thicknee — are dry season nesters hatching their eggs well before the start of the rains, while the bulk of the rest lay Page 139 at the end of the month so that young are likely to coincide with the November storms. The increase to 29 species in February is mainly due to conditions becoming suitable for Category 3 birds. ACKNOWLEDGEMENTS It is a pleasure to thank Messrs I. H. Dillingham, L. A. Haldane and B. W. Stronach for their assistance in problems of identification when I first arrived in Tabora. I should also like to thank the officers in charge of the Bee-keeping Section and the Water Development and Irrigation Department, Tabora for maWng their rainfall records available to me. REFERENCES Bock, W. J. 1958. A generic review of the plover, (Charadriinae, Aves). Bull. Mas. Comp. Zoo!., Harvard 118: 27-97. Burtt, B. D. (ed. Jackson, C. H. N.) 1942. Some East African Vegetation Communities. J. Ecol. 30(1).- 65-146. Handbook of Tanganyika, 2nd Edition 1958. Government Printer, Dares Salaam. Keast, j. a. & Marshall, A. J. 1954. The influence of drought and rainfall on reproduction in Australian desert birds. Proc. Zoo. Soc. London 124: 493-499. Mackworth-Praed, C. W. & Grant, C. H. B. 1955, 1957. Birds of Eastern and North Eastern Africa, 1, 2 (2nd ed.) Longmans, Green & Co., London. Maclean, G. L. 1967. The Breeding Biology and Behaviour of the Double-banded Courser Rhinop- tilus africamis (Temminck). Ibis 109: 556-569. Moreau, R. E. 1950. The breeding seasons of African birds — 1. Land birds. Ibis 92: 223-267. Moreau, R. E. 1964. Breeding season. In “A New Dictionary of Birds” (A. Landsborough Thomson, ed.), pp. 106-108. Nelson, London. Moreau, R. E. 1966. The Bird Faunas of Africa and its Islands. Academic Press, New York, London. Reynolds, J. F. 1965a. On the occurrence of Palaearctic waders in Western Tanzania. E. Afr. Wildl. J. 3: 130-131. Reynolds, J. F. 1965b. Behaviour of Black-lored Babbler. Ibid. 3: 130. Serventy, D. L. & Marshall, A. J. 1957. Breeding periodicity in Western Australian birds. Efim 57: 99-126. Stronach, B. W. H. 1968. The Chagana heronary in western Tanzania. Ibis 110: 345-348. Thomas, D. K. 1960. Birds: notes on breeding in Tanganyika 1958-1959. Tanganyika Notes Rec. 55: 225-243. Thomas, D. K. 1961 . The Ugalla River Controlled Area. Tanganyika Notes Rec. 57: 226-230. W ILLIAMS, J. G. 1963. “A Field Guide to the Birds of East and Central Africa” Collins, London. Williams, J. G. 1967. “A Field Guide to the National Parks of East Africa” Collins, London. WiNTERBOTTOM, J. M. & RowAN, M. K. 1962. Effect of Rainfall on BreedingofBirdsin Arid Areas. Ostrich 33(2); 77-78. {Received 3rd January 1967) CURCULIONIDAE (WEEVILS) OF THE ALPINE ZONE OF MOUNT KENYA Page 141 (Results of the University College Nairobi Mount Kenya Expedition of March 1966 — Publication 1) I. Jabbal {Zoology Department, University College, Nairobi) R. Harmsen *Presently at University of Toronto, Toronto, Ontario, Canada. INTRODUCTION In March 1966, a number of biologists of University College, Nairobi, under the leadership of Dr. Malcolm J. Coe, undertook a research expedition to the alpine zone of Mount Kenya. The main purpose of the expedition was to study the ecology of the relatively dry northern slopes of Mount Kenya which, from a biological standpoint, were virtually unknown. The Base Camp was erected at 12,500 ft. (3,800 m.) in the Kazita West Valley. Most work was carried out in the vicinity of the camp, but a number of collections were made up to the head of the valley (c. 14,000 ft. — 4,300m). The vegetation in this region is fairly typical of the lower alpine zone: consisting mainly of open tussock grassland and patches of Carex monostachya A. Rich, bog on drainage impeded soils. Collections were made in the following situations: 1. Festuca abyssinica St-Yves and F. pilgeri A. Rich, tussocks. 2. Lobelia keniensis R.E. Fr. & Th. Fr. jr. rosettes and inflorescences. 3. Open soil and rocky ground. RESULTS Occurrence and Distribution Appendix 1 includes all the species of Curculionidae so far recorded from the moorland andalpine zonesof Mount Kenya (9,000ft. — 14,900ft. or2,700m. — 4,500m.). This information has been derived from our own collection and the one of the National Museum of Kenya as well as from a survey of the main relevant works in the literature: A. Mustache (1929) and S. Schenkling (1934). Ecology Only five species were found to be of some ecological importance either because of their abundance or as a result of a particular position in a food chain or microha- bitat. Other species may well have such importance in other times of the year or in other microhabitats, as yet undiscovered. The ones of obvious ecological importance as discovered so far are: 1 . Parasystates elongatus Must. 2. Cossonus frigidus Must. 3. Seneciobius basirufus Mshl. 4. Amphitemetis sulcipennis Mshl. 5. Afrotroglorrhynchus (nivalisl) Must. The following ecological data were collected for the above listed species : 1. Parasystates elongatus Mshl. (Plates 1-3) A number of adults of this species were collected (from open soil in Festuca tussock areas) during day time when weather conditions allowed dispersal and mating activity (air temperature 12°C and direct solar radiation). Page 142 During a study of the Festuca tussock as an ecological microhabitat, mature and immature adults as well as larvae and pupae of this species were collected from F. abyssinica tussocks. The specimens were found in chambers produced by the larva in the region of the tussock just above the stem bases. The main matrix of this region consists of semi-decayed, dead Festuca leaves and is penetrated by the living stems of the grass. The beetle larva appears to feed on the lower portions of the live stems, but may derive some nutrition from the dead matter as well. Pupation t^es place in the chambers and the adult emerges and matures there as well. The amount of damage done by Parasystates to the Festuca tussocks is very limited because of the relative rarity of the species. It is, however, potentially a tussock killer because of its mode of feeding. This species was mainly found in F. abyssinica, but appeared to be uncommon in F. pilgeri; whether this reflects a highly selective host specificity or a habitat preference is not known. Within the tussock, the placing of the chamber does reflect a very sensitive microhabitat selection. The chambers are never in the waterlogged region of the tussock below the stem-base and never in the dry upper region. The region in which the chambers are found provides maximum protection against drowning, fire, and predators. This extent of safety to the larvae and pupae makes this species poten- tially an “outbreak” species which, under certain conditions, could severely damage the F. abyssinica cover resulting in considerable changes in the ecomorphology of the alpine zone. 2. Cossouus frigidus Hust. (Plates 4-7) The larvae and pupae of this species were found in very large numbers in the inside of the hollow, recently died, woody regions of the rachis of Lobelia keniensis. Here, they appeared to form a major factor in the disintegration of these structures. They were never encountered in living specimens of Lobelia, nor in the dead “leaf frills” below the living rosette; this, possibly, because of the more or less anoxic conditions of these habitats. The adults emerge in the dead rachis and must then undertake a period of dispersal and mating. No free moving specimens were collected, but a number of mature adults were found in Lobelia inflorescences. It seems probable that the adults are attracted to these in- florescences after dispersal and that they remain stationary here until the flowering period of the Lobelia ends and oviposition takes place. This species obviously does not harm the standing vegetation in any way, and is, therefore, of no influence on the vegetation composition and succession of the alpine zone. The larvae are subject to predation by a number of predatory staphilinid larvae which also inhabit the dead Lobelia rachis. There is a further, and perhaps more severe, predation at the time of adult concentration on the Lobelia inflorescence by a number of insectivorous song birds. In particular the Hillchat {Pinarochroa sordida earnesti Sharpe) and the Scarlet Tufted Malachite Sunbird (Nectarinia i. johnstoni Shelley) were regularly observed feeding on the insects in the Lobelia inflorescence. 3. Seneciobius basirufus Mshl. (Plate 8) Mature, adult specimens of this attractively coloured beetle were collected from tussock grassland and from among rocks on open, frost heaved soil. They appeared quite lethargic when the air tem- perature was low but as soon as it became warm they showed considerable activity. Some of them were caught walking, but no mating was observed. One specimen was collected at 14,000' altitude in a completely moribund state under dead vegetable litter on a cold, cloudy morning (air temp. ±3°C in shady areas). Of this species, neither larvae nor pupae were encountered. Because of its considerable size (16.5 mm.) and its local abundance, this species could form a significant percentage of the food of certain mammals or larger birds. Augur Buzzards (Buteo rufo- fuscus augur Rupp.) were often observed catching beetles and hundreds of beetle elytra were found in buzzard pellets; none of these, however, belonged to S. basirufus. It is possible that the species is poisonous or distasteful, a situation often found in brightly coloured, slow moving animals. 4. Amphitemetus sulcipetmis Hust. (Plate 9) The adults of this species were also collected from among rocks on open soil and from patches of vegetation mainly consisting of Festuca tussocks and Alchemilla cover. Considerable activity, in- cluding mating, was observed as soon as the weather became warm: i.e. air temperature 12°C and direct solar radiation. This beetle was quite common in certain areas and could constitute a major source of protein nutrition for larger birds and such mammals as shrews and insectivorous rodents (e.g. Lophuromys). No direct evidence of such predation is available, and Augur Buzzard pellets did not contain Amphitemetus elytra. This may be because of the excellent camouflage of these slow moving beetles. Another reason could be that the buzzards find it uneconomical to feed on these small beetles when there are a large number of rodents, shrews and larger beetles available. The beetle’s “freezing” behaviour in the presence of human beings suggests that predation is a mortality factor of some significance. 5. Afrotroglorrhynchus (nivalis 1) Hust. Only two specimens of the beetle belonging to the genus Afrotroglorrhynchus were collected while beating Alchemilla Johnstonii Oliv. This species seems to be very near A. nivalis Hust., but according Page 143 to Dr. Edward Voss* it could be a different species. Jabbal (1968) is at present in the process of describing it as a new species, Afrotroglorrhynchus kazitae. Since this beetle is of relatively small size (4.8 mm.) and not very common, it is probably of no quantitative ecological importance. DISCUSSION The unique feature of Afro-montane regions is their remarkable diurnal temperature range, under whose influence an animal may be submitted to sub-zero every night and intense heat and low humidity during the day. Hedberg (1957) called this type of climate “winter every night and summer every day”. The relative humidity, which in these regions fluctuates daily with the temperature and cloud cover, has been described by Coe (1967): “at ground level the relative humidity just before sunrise was 90%, when the sun rose this figure fell within about 90 minutes to below 20%; when during the course of the day the sun became obscured by cloud, the figure rose to 80 %”. In such a climate with large and regular diurnal temperature changes it is not so much the extremes, but rather the speed with which they fluctuate, that is the main controlling factor on insect life. The atmosphere becomes thinner as the altitude increases, thus resulting in lesser heating of the air during the passage of solar radiation. The most important component of this incoming radiation is the ultraviolet. While considering the climate near the ground Geiger (1950) quotes the work of Maurer, who found that the amount by which ground temperature exceeds that of the air increases with altitude, a factor of obviously great significance to microclimate in the equatorial mountains. During the day the outward radiation from the ground and vegetation is not very apparent but at high ^titudes its effect is of great importance as an additional cooling agent at the surface and in consequence this effect may be strongly felt by the invertebrates occupying the microhabitats. The sudden lowering of temperature does not, however, penetrate more than a few inches below the, surface of the soil— 4 to 6 inches below the ground level, the temperature was found to be c6°C. So in large part, the protective insulating mechanisms that have been developed by the vegetation are fully utilized by the invertebrate fauna. The activity of insects in such areas is greatly limited and seems to take place in bursts of short duration whenever circumstances are favourable. At night the intense cold renders them incapable of movement, while during the day, except for a short period after sunrise and just before sunset, the ground temperature is far too high and humidity too low. Thus, not surprisingly, it is due to these two factors that a high percentage of arthropods exhibits sedentary and cryptozoic habits, which keep them within or close to the comparatively constant microclimate of their shelters. Examples of this are Cossonus in the Lobelia rachis and Parasystates in the Festuca tussock. None of the observed species showed any evidence of rhythmic control over activity. Morphological factors seem to play an important role in the adaptation of invertebrate life to the alpine climate. The highly reflective surface of some of the beetles like Parasystates elongatus is probably a means of protection against radiation. The predominantly dark colours of almost all the weevils collected could be important in heat absorption. It is probably an advantage to absorb heat quickly in the early morning, so that the animal can complete its main period of activity before the ground becomes too hot. All the Curculionids collected possess inflated elytra whose enclosed air could have an important insulating function. One point of evolutionary interest is the fusion of the elytra and absence of wings in most of the alpine Curculionids. This aptery or brachyptery is a common phenomenon of high altitude insects and Mani (1962) suggests the obvious selective advantage of such a modification. He points out that a flying insect can easily be carried by up-currents on the barren peak region or swept off the mountain altogether. The winds on East African mountains, however, are not generally as strong as those on the Himalayas. Thus, whether the presence of flightless insects on Mount Kenya represents a selective survival of apterous species which have colonized the mountain or whether it represents the evolu- tion of apterous species from normal, winged species after colonization of the mountain is very much a matter of speculation. During the March 1966 expedition, no attempt was made to collect the smaller species of the family Curculionidae. Yet, of the thirty two previously described species for Mount Kenya four were encountered, and one new species was found {Afrotroglorrhynchus kazitae n. sp.). This in- dicates that a systematic survey of the alpine regions of Mount Kenya will probably uncover a considerable number of weevils as yet undescribed. The purpose of this expedition was to study the ecology of the northern slopes of the alpine zone of Mount Kenya. The work on Curculionidae as reported above throws some light on the role played by the larger and more abundant species of this family in biological interrelationships. The rarer and smaller species must also play some role in the ecology of this region. At present this role seems to be of little quantitative importance, however, it requires a closer study. *Dr. Edward Voss of 4501 Hardenberg, Am Boberg 2, B. R. Deutschland, kindly examined our specimens. E ACKNOWLEDGEMENTS The authors would like to express their gratitude to the Director of the Kenya National Parks and to Mr. Bill Woodley for their co-operation. The work was supported by a Ford Foundation grant for Ecological research to the Zoology Department of University College, Nairobi. BIBLIOGRAPHY 1. Coe, M. J. (1967), The Ecology of the Alpine Zone of Mount Kenya. 134 pp. 20 figs., 24 pi. Junk. Hague. 2. Geiger, R. (1950), The Climate near the Ground, pp. xxi-t-482. Harvard. 3. Hedberg, O. (1957), Afro-Alpine vascular plants. Symb. Bot. Upsal. 15 1 :1-411. 4. Hustache, a. (1929), Voyage de Ch. Alluaud et R. Jeannel en Afrique Orientale, Resultants Scientifiques: Insectes Coleopteres. 19:367-560. 5. Jabbal, I. (1968), A Description of Afrotroglorrhynchus kazitae n. sp. in.ed. — MS. 6. Mani, M. S. (1962), Introduction to High Altitude Entomology. London. 7. SCHENKLiNG, S. (1934), Coleopteronim catalogue auspiciis et auxilio, W. Junk, 28, 29 & 30. Page 145 APPENDIX I CHECKLIST OF CURCULIONIDAE COLLECTED FROM THE MOORLAND AND ALPINE ZONES OF MOUNT KENYA All the names marked * are in the collection of the National Museum of Kenya. Name Authority Collector and Date Altitude References A. Subfamily: Otiorrhinchinae 1. Amphitemetus Alluaud & Jeannel 9,200 ft. and Hustache (1929) griseus Hust. 1912 10,500 ft. pp. 384-385. 2. *A. sulcipennis Hust. Alluaud & Jeannel 9,200 ft. and Hustach (1929) 1912 Joy Peter Bally Nov. 1943 Mrs. Bally Jan. 1964 Jabbal & Harmsen March 1966 10,500 ft. 10,500 ft. 10.500 ft. 12.500 ft. pp. 385-387. 3. Leptospyris Alluaud & Jeannel 9,200 ft. and Hustache (1929) sylvaticus Hust. 1912 10,500 ft. pp. 401-402. 4. L. glacialis Hust. Alluaud & Jeannel 13,100 ft. and Hustache (1929) 1912 14,400 ft. pp. 402-403. 5. L. laevis Hust. Alluaud & Jeannel 13,100 ft. and Hustache (1929) 1912 13,400 ft. pp. 403^04. 6. Parasystates Alluaud & Jeannel 9,200 ft. and Hustache (1929) albovittatus Auriv. 1912 10,500 ft. p. 406. 7. *P. elongatus Hust. Alluaud & Jeannel 7,900 ft. and Hustache (1929) 1912 A. J. F. Gedye Dec. 1943 Museum Staff Jan. 1947 F. C. Delkirk Feb. 1950 Harmsen & Jabbal, 1966 14,400 ft. 13.500 ft. 12,150 ft. 14,850 ft. 12.500 ft. pp. 407-408. 8. P. alternans Hust. Alluaud & Jeannel 10,800 ft- Hustache (1929) 1912 11,500 ft. pp. 408-409. 9. P. nigripennis Hust. Alluaud & Jeannel 7,900 ft. and Hustache (1929) 1912 10,500 ft. pp. 410-411. 10. P. alpinus Hust. Alluaud & Jeannel 7,200 ft.- Hustache (1929) 1912 10,200 ft. pp. 413-414. l\. P. brunneus Hust. Alluaud & Jeannel 9,200ft.- Hustache (1929) 1912 10,500 ft. pp. 414-415. 12. Systates elongatus Hust. Alluaud & Jeannel 7,900 ft. Hustache (1929) 1912 9,200 ft. and 10,500 ft. pp. 424-425. 13. Barypeitlies Alluaud & Jeannel 9,200 ft.- Hustache (1929) microphthalmus Hust. 1912 10,500 ft. pp. 450-451. 14. Omias (Neomias) Alluaud & Jeannel 9,200 ft.- Hustache (1929) kenyae Hust. 1912 10.500 ft. pp. 451-452. 15. O. (Neomias) Alluaud & Jeannel 13,100 ft.- Hustache (1929) kenyae var. 1912 13,400 ft. p. 452. glacialis Hust. 16. O. (Neomias) Alluaud & Jeannel 10,800 ft.- Hustache (1929) alpinus Hust. 1912 13,400 ft. pp. 452-457. B. Subfamily: Cleonidae 17. Li XUS nycteropliorus Alluaud & Jeannel 7,900 ft.- Hustache (1929) var. kenyae Hust. 1912 8,900 ft. p. 474. 18. *L. alpinus Hust. Mrs. Bally Jan. 1944 10,500 ft. Hustache (1929) p. 475. Name Authority Collector and Date Altitude References 19. *L. adspersus Boh. Mrs. Bally Boheman(1871) Jan. 1944 Fahrs. Oefrers. Vet. Akad. Fork. 18; 58, 230. C. Subfamily; Rhyparosominae 20. *Oreoscotus Alluaud & Jeannel 10,800 ft. and Hustache (1929) fulvitarsus Hust. 1912 11,500 ft. pp. 465-466. A. J. F. Gedye Dec. 1934 10,000 ft. Museum Staff Jan. 1947 12,150 ft. Mrs. Bally Jan. 1944 10,500 ft. D. Subfamily; Eirrhininae 21. Homoedenodema Alluad & Jeannel 9,200 ft. and Hustache (1929) fulva Hust. 1912 10,500 ft. pp. 483-484. E. Subfamily; Apioninae 22. Apion warendorffi Wagner Alluad & Jeannel Mem. Soc. Ent. 1912 Belg. 19; 41. F. Subfamily; Baridinae 23. Baris kenyae Hust. Alluad & Jeannel 9,200 ft. and Hustache (1929) 1912 10,500 ft. pp. 531-533. G. Subfamily; Cossoninae 24. Mimus glacialis Hust. Alluaud & Jeannel 9,400 ft. Hustache (1929) 1912 pp. 544-545. 25. *Cossonus (or Alluaud & Jeannel 10,800 ft. and Hustache (1929) Afrocossonus) hyperboreus Hust. 1912 11,700 ft. pp. 545-546. Museum Staff 12,150ft. 26. C. dorytomoides Hust. Jan. 1947 Alluaud & Jeannel 10,800 ft. and Hustache (1929) 1912 12,100 ft. pp. 547-548 27. *C. frigidus Hust. Alluaud & Jeannel 13,100ft. and Hustache (1929) 1912 14,400 ft. pp. 549-550. Museum Staff 12,150ft. Rev. Zool. Bot. Jan. 1947 Afr. (1934) 26: 36. A. J. F. Gedye Dec. 1934 13,800 ft. Jabbal & Harmsen March 1966 12,500 ft. 28. *C. (or Pseudo- Alluaud & Jeannel 10,800 ft. and Hustache (1929) mesites) glacialis Hust. 1912 12,100 ft. pp. 552-553. A. J. F. Gedye 13,800 ft. Rev. Zool. Bot. Dec. 1934 Afr. (1934). 26: 36. Museum Staff Jan. 1947 11,000 ft. H. Subfamily; Otiorrhinchinae 29. *Seneciobius basirufus Mshl. Mrs. Bally 12,000 ft. J. E. Afr. Nat. Aug. 1934 Hist. Soc. (1950). 19, 5; 147. Mrs. Bally Jan. 1944 10,500 ft. Museum Staff Jan. 1947 12,150ft. Jabbal & Harmsen 12,500 ft. and March 1966 14,000 ft. Page 147 Name Authority Collector and Date Altitude References 30. *S. semilucens Mshl. A. K. Hading 11,000 ft. J. E.A. Nat. Hist. 1949 Soc. (1950). 19, 5:147 31. *Strictoseneciobius Mrs. Bally 10,500 ft. Ann. Mag. Nat. ebininus Must. 1944 Hist. London (1940). (11) 13:93-98 32. * Seneciobius loveni Aur. A. K. Hading 13,000 ft. Rev. Zool. Bot. or graniilipennis Hust. 1949 Afr. (1923) 11 188. Ann. Mag. Nat. Hist. (1934). (10) 15:503 (Received 30th January 1968) 4 mm. Parasystates elongatus Hust. Page 149 Mouth parts of the larva of Pamsystates elongatus Page 150 Larva of Parasystates elongatus Page 151 Mouth parts of the larva of Cossoms frigidus. L 2 mm. j Larva of Cossoms frigidus. 2 mm. Pupa of Cossonus frigidus. Seneciobius basirufus Mshl. Page 154 Amphitemetus sulcipeimis Hust. Page 155 RECORDS OF PARASITIC NEMATODES IN KENYA Gerald D. Schmidt {Department of Biology, Colorado State College) AND Albert G. Canaris {Department of Zoology, University of Montana, Missoula, Montana) Among the helminth parasites collected by the second author in Kenya from 1962 to 1965 were several species of nematodes, reported herein. Specimens were fixed in AFA and cleared in glycerin for study. Rare species are deposited in the USNM Helminthological Collection, Beltsville, Maryland. The results of this study are reported in Table 1 . Table I NEW RECORDS OE PARASITIC NEMATODES IN KENYA Exam- Infect- Host Location ined ed Locality Parasite FISH Black bass Microptems salmoides Eac6p. intestine 1 1 Lake Naivasha Porrocaecum sp. (Juv.) Tilapia Tilapia sp. intestine 2 1 Lake Naivasha Porrocaecum sp. (Juv.) AMPHIBIANS Bullfrog Pyxicephalus delandii Tschudi cyst on gut 4 1 Mt. Suswa Agamofilaria sp. Toad Biifo regularis Reuss small intestine 4 4 Njoro (Icosiella'l) (Juv.) Aplectana dogieli lungs 4 2 Njoro (Skrjabin, 1916) Rhabdias sp. Grass frog Rana angolensis Bocage intestine 30 24 Njoro Amphibiophiliis acan- intestine 30 9 Njoro thocirratus (Skrjabin. 1916) Falcaustra sp. REPTILES Python Python sebae (Gmelin) 2 0 Lake Solai, (Spironoura) Njoro Mole snake Pseudaspis cana (Linnaeus) stomach & intestine 1 1 Njoro Kalicephalus colubri (Ortlepp, 1923) Purple-glossed snake Chamaetortus aulicus Gunther intestine 1 1 Watamu Thubunaea asymmetricu Puff adder Bitis arietans (Merrem) intestine 1 1 Njoro Baylis, 1930 K. colubri Kenya homed viper Bitis worthingtoni Parker 2 0 Njoro Page 156 Exam- Infect- Host Location ined ed Locality Parasite House gecko Phyllodactylus sp. stomach 2 1 Watamu T. asymmetrica cloaca Chameleon Chameleo dilepis Leach small intestine 14 3 Njoro Pharyngodon mabuien- «s(Mala, 1939) Strongyluris brevicau- Skink Mabiiya homalocephala (Wiegmann) stomach 2 1 Watamu data (Mueller, 1 894) T. asymmetrica BIRDS Black river duck Anas sparsa Egton small intestine 3 1 Njoro Porrocaecum crassum Speckled pigeon Columba guinea Linnaeus 2 0 Njoro (Deslongchamps, 1824) Olive pigeon C. arquatrix Temminck & Knip small Intestine 5 2 Njoro Ascaridia columbae Pink-breasted dove Streptopelia liigens (Riippell) 4 0 Njoro (Gmelin, 1790) Red-eyed dove S. semitorquata (Ruppell) 2 0 Njoro Tambourine dove Tympanistria tympanistria (Temminck & Knip) 1 0 Naivasha Green pigeon Treron australis (Linnaeus) small intestine 12 7 Njoro A. columbae MAMMALS Mole rat Tachyoryctes sp. stomach 2 1 Njoro Ascarops africana Striped grassmouse Rhabdomys pumilio (Sparrman) small intestine 2 2 Njoro (Sandground, 1933) Longistriata impudica Serval cat Felis serval Schreber intestine 1 1 Mau Forest (Baylis, 1928) Ancylostoma paraduo- Elephant Loxodonta africana (Blumenbach) intestine 1 1 (Njoro) Mau Forest denale (Biocca, 1951) Toxocara sp. Quilonia magna Totals 101 56 (Molo) (Neveu-Lemaire, 1928) individuals infected with one or more species of round worms ACKNOWLEDGEMENTS The authors are indebted to Dr. Roy C. Anderson, University of Guelph, for his opinions on the identity of the frog filariid, and to Mr. John Williams, Curator of Birds, Kenya National Museum, who assisted with host determinations. {Received 11th March 1968) DISTRIBUTION AND HOST-SPECIFICITY OF A IVUMBER OF FLEAS COLLECTED IN SOUTH AND CENTRAL KENYA (Including the Collection of the University College, Nairobi, Mount Kenya Expedition, March 1966) by R. Harmsen* *Presently at University of Toronto, Toronto, Ontario, Canada, I. Jabbal {Zoology Department, University College, Nairobi) During the years 1963-1966 a collection of fleas was made by various members of the Department of Zoology of University College, Nairobi (UCN). This collection was mainly based on Dr. J. B. Foster’s trapping of small mammals, and on the fieldwork of Dr. J. B. Sale with hyrax. Some speci- mens, however, were collected occasionally from road kills and other incidental mammalian hosts. The collection culminated with the work of the University College Mount Kenya Expedition in March 1966. The collection was mounted and identified by members of the Division of Insect-Borne Diseases (DIBD), Medical Department, Kenya, and a number of specimens were sent to Mr. F. G. A. M. Smit, Zoological Museum, Tring, England, for confirmation of identification. The present records of distribution and host-specificity are of interest for a variety of reasons. Fleas only spend their adult life in intimate contact with the mammalian host, the larva is a free living insect, feeding on proteinaceous detritus. It is, therefore, to be expected that fleas will be mainly found on those mammals which have permanent dwelling sites, and particularly on nest or den building animals. This is the immediate elfect of the need of a freshly emerged adult flea to contact a new host. Flea eggs deposited in a nest or den have a much higher chance of developing into adults within the immediate reach of a new host than flea eggs deposited at random in the field. One other consequence of this necessity of the flea to locate a host within its life time after larval development away from the host, is a dependence on the climate. It is, thus, to be expected that within the fairly narrow geographical confines of Kenya a number of flea species will be restricted to particular altitudinal levels in response to a climatic adaptation, even when suitable hosts are available at other altitudes. Alternatively, one would expect to find fleas restricted to particular hosts at one altitude, and to other hosts at another altitude, this in response to the combined effect of climate and nesting habits of different hosts. This form of ecologically based isolation could well become the basis of speciation. The Collection The following species were collected and identified : Chimaeropsylla potis potis Ctenocephalides felis strongylus Ctenophthalmus cophurus Ctenophthalmus lycosius Delopsylla crassipes Dinopsyllus longifrons Dinopsyllus lypusus Echidnophaga aethiops Nosopsyllus incisus Procaviopsylla procaviae Xenopsylla cheopis Xiphiopsylla hyparetes Xiphiopsylla levis For each species a number of notes concerning distribution, ecology and host specificity are of significance. In the following account all names of mammalian hosts in the paper follow Allen (1939) and the following symbols refer to location of specimens. *For Rothschild collection, Tring, England. tFor UCN Zoology Dept. Nairobi. $For DIBD collection Nairobi. 1. Delopsylla crassipes Jordan, 1926a; (PuUcidae). This species of flea is only known from the Kenya Highlands with previous records from the Nairobi and Nakuru areas. It is only known from the springhaas, Pedetes surdaster Thomas. In February 1964, Foster collected one hare, Lepiis capensis Linnaeus, and two springhaas in a small area on the Kapiti plains to the S.E. of Nairobi. From the springhaas 2 and 3 of Delopsylla cmssipes were collected, from the hare 1 only. This record of D. crassipes on a hare is not necessarily an indication of this flea selecting the hare as a host. Several other hares have been collected in areas nearby, but not within a springhaas colony; on these hares, fleas were usually encountered, but never D. crassipes. It is therefore likely that this flea is indeed a highly host-specific species, which is merely carried temporarily by other furry hosts such as the hare. 2. Echidnophaga aethiops Jordan & Rothschild, 1906; (Pulicidae). This flea is well known as a bat flea, associated mainly with the bat genus, Nycteris. It has been collected over a wide area of South and Central Africa (South Africa, S.W. Africa, Somalia and the Tsavo region of Kenya! (Jordan & Rothschild, 1906; 1913; Hopkins & Rothschild, 1953), but is not common. In January 1964, Foster collected a bat, Nycteris thebaica Geoffroy,at Olorgesailie, approximately 45 miles S.W. of Nairobi in the Rift Valley. The fleas (8?$)f were restricted to the inside of the ear where they were attached by their proboscis to the membranous skin of the bat. The location and host fit very well within the previously recorded area for this flea. It seems likely, therefore, that E. aethiops is exclusively a bat flea and restricted to the genus Nycteris in the arid regions of Southern and Eastern Africa. On the basis of this single collection, it is impossible to attach any significance to the sex ratio (8$?, no SS). Other collections have, on the whole, also had a very much larger number of ?? than of (de Meillon, Davis & Hardy, 1961). 3. Xenopsylla clieopis (Rothschild), 1903; (Pulicidae). This species is known from the entire world, including Australia and many oceanic islands (Hop- kins & Rothschild, 1953). It parasitizes a large number of rodents including Rattus rattus (Liimaeus), which may well be the reason for it being found on so many islands. It is most likely correct to think of this species as a rodent flea, although it has been collected from such animals as genet, shrews, and man, but these records could easily be interpreted as stray specimens being temporarily harboured by non-host mammals which live in close proximity to one of the real hosts. In February 1964, Foster collected two small rats, Grammomys (Thomas & Wroughton), at the foot of Lukenya Hill (20 miles S.E. of Nairobi) on which one and 2 cJcJt of cheopis were foimd. Among the hosts listed in the literature (Hopkins & Rothschild, 1953) Grammomys is not listed, and is probably a new record. It is, of course, not at all surprising, as X. cheopis is known from a large number of small rodent species. 4. Procaviopsylla procaviae (C. Fox), 1914; (Pulicidae). All records of this species of flea are from the elevated regions of Kenya and Tanzania, and all but one are from Procavia spp. or Heterohyrax spp. (Fox, 1914; Jordan, 1925; 1926b). One isolated record from a dikdik, Rhynchotragus kirki (Gunther), may well be an example of a non-host carrier (Hopkins & Rothschild, 1 953). It is, however, interesting to note that the dikdik is, like the hyrax, a small ungulate. On two occasions (Feb. 1964 and Dec. 1965), Sale captured a number of specimens of Heterohyrax syriacus (Schreber) on Lukenya Hill (20 miles S.E. of Nairobi). On both occasions fleas were found on all or some of the hyrax, and a representative collection was made (26$? and 13>kj(.f« I ' I |J, ‘ < I .» t\n ' ' rP^ttir/ ' ) I. ^ . ., .'•). Mn.f»«p ( -tit .u.A (nt>i -mt.T ■ •■•' ■■ • <:,'■}' «!l- ■.. I, > .ife;tvv*i>N<>iPA* I ' . . vV i,. ' 'vvnir. . ' ■ I )•' i'^' V \ ,V'V:4U ^ftJiCirlT ■' -VP •<' > V J ,!>•• •» '*• . >' ' .. ,r. i .SJ •' •.> X .■ '-W >»■;« id/KM ' -'ll < I^T !)»-{•* . ■«.;•' ’Wv< • ' (Uiuj^4 ■' '' ^ ■'V /<, / '■■’> ■'■ ' i " !' 'i-ii Page 163 NATURE NOTE Singing by the Chiff-Chaff in Kenya I refer to the record of singing by a Chiff-Chaff (Phylloscopus collybita vieillot) on Lengetia Farm, Man Narok, by Sessions (J. E. Afr. Nat. Hist. Soc. XXVI No. 1 (113): p. 40). It is worth noting that, as recorded {Ibis 1945: p. 94), on Mount Elgon at 12,000 feet on 9th February 1941, I heard the song of a Chiff-Chaff many times over, and later saw individuals of a Phylloscopus in the vicinity, presumably P. collybita. I was at this time living at Eldoret, and during a week-end had made a partial ascent of the southeastern slopes of Mt. Elgon, I was already familiar with the song of this species in England, where I had last heard it in the spring of 1939. 30th October, 1967. C. W. BENSON, Dept, of Zoology, Downing Street, Cambridge. BOOK REVIEWS The part of the Flora of Tropical East Africa dealing with the Caesalpinioideae has at last appeared. This group of plants is represented in East Africa by forty genera, mostly trees and shrubs with a few Hanes and climbers. A section of the genus Cassia consists of herbs and Cryptosepalum maraviense, is suffrutescent with a thickened woody rhizomatose rootstock producing tufted erect annual stems. Many are worth a place in a garden on account of their very showy flowers, usually yellow, the most ornamental being Gigasiphon macrosiphon which has a discontinuous distribution including Mrima Hill and Msambweni in Kenya, the East Usambaras and the Ronda Plateau in Tanzania. It is a deciduous tree up to 70ft. tall occurring in forest. Another is Paramacrolobium coeruleum with bluish-mauve flowers. This is recorded in East Africa from the Kwale District in Kenya and the Uzaramo District in Tanzania. Outside East Africa it is found in the Republics of Guinee, Camaroun, Congo and in the Central Africa Republic as well as Sierra Leone. To those with a preconceived idea that the Family Leguminosae consists of plants, mostly herbs, with pinnate leaves butterfly-like flowers and dehiscent pods called legumes, the subfamily Caesalpinioideae is full of surprises. It has genera with simple leaves, Gigiosiphon, Zenkerella, Baphiopsis, Tylosema, Bauhinia and Pilostigma, and others with bipinnate leaves, Erythrophleum, Burkea, Bussea, Caesalpinia, Parkinsonia, Delonix. Pterolobium and Mezoneuron. Some genera are spiny or armed with prickles, Pterolobium, Mezoneuron, Caesalpinia and Parkinsonia, and the leaflets of some genera have pellucid gland dots, to be found in Daniellia, Guibourtia, Oxystigma, Tessmannia and Trachylobium. To add further to the confusion the fruits of some genera are anything but legumes. They can be sausage-shaped like {Swartzia and Tamarindus) or long, woody and cylind- rical as found as in some species of Cassia whilst other genera have dehiscent pods. Genera with woody dehiscent pods include Isoberlinia, Berlinia, Julbernardia, Parama- crolobium and Brachystegia, the fruits of some of which dehisce with a loud explosive sound, scattering their flat round seeds considerable distances. Afzelia has a thick woody pod which opens into two thick valves exposing up to 12 large oblong-ellipsoid black seeds each with a cup-shaped orange, red or vermilion basal aral. The liane Pterolobium is very exceptional as its bright scarlet “pods” have a solitary wing not unlike the samara of a sycamore, whilst the Gum Copel Tree, Trachylobium verrucosum has an ovoid-oblong to obovoid indehiscent thick woody pod which is covered thickly with resinous warts. Page 164 Three other genera are also unusual, Cordyla and Mildbraediodendron with very large plum-like indehiscent so-called pods, and Dialum with small cherry-like indehi- scent globose “pods” containing a pulpy mealy mesocarp which is brown, orange or red when dry and in which 1-2 small seeds are imbedded. The 52 figures illustrate most of the 40 East African genera which compose this curious sub-family of the Leguminosaeae and the author, J.P.M. Brenan, is to be congratulated on a very excellent and painstaking piece of botanical research. The Caesalpinioideae contains timbers, resinous, edible, poisonous and medicinal species, some of which are of economic importance and the genus Brachystegia of which there are fourteen species, these forming the dominant feature of the extensive woodlands over much of Tanzania. Published under the Authority of the Minister for Overseas Development by the Crown Agents for the Overseas Governments and Administrations. The Government Bookshop, P.O. Box 569, London, E.C. 1., price £1.8.0. Also from the Government Printers, Nairobi, Dares Salaam and Entebbe, of the three East African Governments. Pp. 1-230 and 52 text figures with keys to the genera and their species, and a 5p. index. A further part of the Flora of Tropical East Africa, the Cucurbitaceae, is now out. It is by Mr. C. Jeffrey who says it is a panatropical family of about 500 species, poorly represented in the temperate regions. Twenty-eight genera containing a hundred and twenty seven species are represented in our region. They are woody or herbaceous, mostly with climbing or trailing stems bearing tendrils, rarely without, some have tuberous rootstocks. The flowers are usually unisexual, monoecious or dioecious, axillary and variously arranged but the female flowers are more commonly solitary than those of the male flowers. The fruit may be a dry or fleshy capsule, berry or a hard-shelled pepo, indehiscent or dehiscent by valves, or shed their seeds by means of a lid in the fruit, by slits or apical pores, or irregularly, or rarely samaroid. The author has designed the key to the genera so as to be of use even if only the male or female flowers or ripe fruits are available, and also to cover the atypical leafless states, at the same time the author makes it clear that some of the species are not com- pletely known. Of these there are eleven genera containing twenty species that are insufficiently known to be able to describe them fully, or the type specimens are no longer available for examination. The Cucurbitaceae contains one hundred and fifty-six pages, six of them covering the index. There are twenty pages of line drawings of figures of most of the genera des- cribed ; these should be of assistance to any one wanting to name a Curcurbit. The genera and species are not easy to identify because their leaves are very variable, you often do not get the male and female flowers in flower on a plant at the same time, nor do you often get ripe fruits at the same time; if you do they may be very fleshy or so large as to make them difficult to collect. The author is to be congratulated on the results of his study of the very difficult family Cucurbitaceae. It is a family of economic importance, having edible fruits and seeds as well as fibrous fruits and gourds for carrying liquids. Flora of Tropical East Africa, Cucurbitaceae; published under the Authority of the Minister for Overseas Development by the Crown Agents for Overseas Govern- ments and Administrations, price Shs. 18/- from the Government Bookshop, P.O. Box 569, London, S.E. 1. and from the Government Printers, Nairobi, Dar es Salaam and Entebbe, of the three East African Governments. Pp. 1-155 and 26 text figures with keys to the genera and their species, and a 6p. index. Towards the end of January 1968 eight further parts of the Flora of Tropical East Africa were published by the Crown Agents under the Authority of the Minister for Page 165 Overseas Development. The families in this series are the Aralaceae by G. R. Tennant, price Shs. 3/25, Basellaceae by B. Verdcourt, Shs. 1/-, Cactaceae by D. R. Hunt, Shs. 1 /50, Malphighiaceae by E. Lannert, Shs. 3/50, Olacaceae, Shs. 2/50 and Opiliaceae, Shs. 1/50 both by G. LI. Lucas, Salvadoraceae by B. Verdcourt, Shs. 1/75 and the Sonneratiaceae by G. R. Williams Sangai, Shs. 1/-. Most of these are only small families and therefore the booklets only contain a few pages, the largest being Malpighiaceae with six genera and nineteen species, the Olaca- ceae with six genera and ten species, followed by the Aralaceae with three genera and twenty species, the Salvadoraceae with three genera and four species, one of which is Salvadora with one species split into four varieties. The Cactaceae, a family of importance from the plant pest point of view, contains only one indigenous genus, Rhipsalis baccifera, usually found as a succulent pendulous epiphytic shrub on trees, but which is seen sometimes as a terrestrial shrub. The other member of the family is the Prickly Pear, Opuntia; these are natives of the Americas and have become widely naturalized; they are pests in some countries. I do not think we know all we should know about the distribution of Opuntia and its species in our region and Mr. D. R. Hunt, the author, has done the best he could with the material of the genus available for him to study, as only five species are dealt with. Any large plant of the cactus-kind with its very succulent stems and “leaves”, often very bulky and covered with stiff spines or prickles, cannot be described as the ideal of a botanical specimen for a collector. They are very difficult to deal with, take up a vast amount of space in a press, usually take a long time to dry, even if they do not go mouldy and rot before they are dry, and the end product looks like something that has been brought in by the dog. As a result any specimen of this genus is strictly avoided by all botanical collectors ; on this account the genus and other plants like it are very poorly represented in herbaria. The families Basellaceae and Sonneratiaceae have only one genus each, the former Basella with two species ; one of them, Basella alba, is an important spinach-like vege- table. Sonneratia alba in the Sonneratiaceae is a common outer-fringe constituent of mangrove swamps down the East African coast and is easily distinguished from other mangroves by its stout conical finger-like pneumatophores and by its opposite, obovate or oval, leathery yellow-green leaves. These eight parts are well illustrated with full page line drawings showing full botanical details of the plants drawn. Like other parts of this Flora they can be obtained from the Government Book- shop, P.O. Box 569, London, S.E. 1. or from the Government Printers, Nairobi, Dar es Salaam and Entebbe, of the three East African Governments. P. J. G. Page 166 Flora of Tropical East Africa. Edited by E. Milne-Redhead and R. M. Polhill. Orchidaceae (Part 1) by V. S. Summerhayes, O.B.E., B.Sc., F.R.H.S. 1968. Published by the Crown Agents for Oversea Governments and Administrations. Available from Government Printers in Nairobi, Kampala and Dar-es-Salaam. 236 pages, 41 line drawings. Price 28/-. The Orchidaceae is probably one of the most eagerly awaited families to be pro- duced in the F.T.E.A. series. Being a large family, with perhaps over a thousand species in the area covered by the flora, it will be issued in three parts; the first of these became available at the end of 1968. Part 2 is scheduled to be completed by the end of 1970 and Part 3 by the end of 1974. As in other families in this series the aim is to present as complete a picture as possible of the species in the area. For this reason a large amount of revision has been, and will be, necessary in certain genera, which may involve delays in publication, but the accurate and carefully produced result will be well worth waiting for. Part 1 contains a diagnosis of the family Orchidaceae, a key to the tribes occurring in Tropical East Africa, and a detailed study of the tribe Orchidae. Seventeen genera are described and there are keys to and details of about 250 species. A reference to the type material and description, together with references to other publications con- cerning each species precedes a careful description. This is followed by details of cited specimens examined by the author, a note on the distribution and habitat, and con- cludes with a list of known synonyms for each species. There is a page of carefully executed line drawings of at least one species in each genus, and in the larger genera, such as Habenaria, there is a drawing of one representative of each section. The tribe Orchideae contains the following genera which occur in Kenya: Holothrix Brachycorythis, Cynorkis, Habenaria, Bonatea, Platycoryne, Roeperocharis, Disa, Brownleea, Satryrium and Disperis. These genera are also represented in Uganda, with the exception of Roeperocharis and Brownleea. In addition species of Neobolusia, Schwatzkopffia, Schizochilus, Stenoglottis, Centrostigma and Pterygodium, which occur in southern Tanzania, are found in this book. Most of the other ground orchids, together with the large epiphytic genera Polystachya and Bulbophyllum and a few others, will appear in Part 2, and we shall have to wait for Part 3 for the tribe Vandeae (Eulophia and all the epiphytes in the angraecoid group). I have used this part of the Orchidaceae for routine identifications of orchids sent in from Kenya, Tanzania, Rwanda and Burundi since it was published and have found it invaluable. Difficulties arise from time to time in interpreting the keys, when one is required to know details of the leaves and vegetative shoots of the plants. This is not the fault of the key. It is entirely reasonable that simple vegetative characters, such as the number of leaves and their position on the stem, should be used in the construction of an artificial key to species. Unfortunately many collectors (and growers) of orchids seem to think that the flowers, or at times a single flower, will suffice for identification. It often does, but the task would be much easier if a complete specimen could be submitted, or at least an adequate description of the parts which are missing could be sent in with the material. The preparation of this work has involved many years of detailed and critical study. Botanists, naturalists and orchid collectors in East Africa will be grateful to Mr. Summerhayes for many years to come. Page 167 C.J.P. lonides An Appreciation “Iodine”, as he was fondly known to his friends, came to East Africa to join the Game Department in what was then Tanganyika Territory. Before that he had been an Army Officer in India. He chose the profession of a member of the Game Department solely because of his intense love of all forms of animal life, and it was not long before his main interest shifted from mammals to snakes and other reptiles. Nevertheless, almost to the end, he remained specially interested in collecting any specimen of the rarer animals. These he collected not simply as trophies to hang on his walls, but for their scientific interest. In fact he never built up a personal trophy collection, but he loved to feel that he had personally collected and made an important contribution for some museum or institution; and that it was the best that could possibly be obtained. It was because of this particular facet of his life that we first got to know him, at what was then the Coryndon Memorial Museum. Shortly after the second World War onwards, until his untimely death, he collected many excellent specimens, which now form the basis of important habitat groups. It was he who, at his own expense, mounted an expedition to the Congo to get the Mountain Gorilla. It was he who obtained for us the Yellow-backed Duiker. It was he who made the difficult journey to the Southern Sudan and brought back the skin and skeleton of the White Rhino. It was he who obtained for us the Abbott’s Duiker specimen. In more recent years, he gave an increasing amount of his time to collecting snakes, and after he had retired from the Game Department, Tanzania, he lived at Nuala in South Tanzania, and used that as a base for his reptile collecting. In this field too, the National Museum in Nairobi owes an immense amount to him, in particular for the magnificent Water Cobra from Lake Tanganyika, and a number of other rare snakes. He was never happy unless he was doing something in connection with wild life, and one of his most recent exploits was to go with Jonathon Leakey to Thailand, where, between them, they col- lected and brought back sixteen live King Cobras. Because of his inborn shyness, lonides made fewer friends than might have been expected, but those who knew him well, found him one of the most charming, kindly and helpful persons they had ever met. He cared little, himself, for creature comforts, but was always willing to give a hand and help anyone in any way. He was exceedingly generous to those whom he liked and who respected him. “Iodine” never married and so leaves no family to follow his trail. It was not because he disliked women or did not enjoy their company, but simply that he felt that no woman could ever really fit into the sort of life he had chosen for himself. Perhaps he was right; nevertheless, he would have made a kindly husband and an understanding father. The world in general, and East Africa in particular, have lost much through his passing on. L. S. B. L. -ft so C' ^7 JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. xxvn No. 3 (118) JANUARY, 1969 CONTENTS A Check List of Plants Collected or Recorded in the Tsavo East National Park By P. J. Greenway Notes on Some Butterflies of the Ngara District of Tanzania, together with a List of Those Recorded By L. A. Haldane Geocarpy as an Adaptation to Afroalpine Solifluction Soils By A. D. Q. Agnew & O. Hedberg Bird Ringing Report 1967 - 1968 By G. C. Backhurst A Note on the possible Reproductive Structures in Kenyan, Udotes orientalis A. & E. S. Gepp, (CUorophyta) By Shakmtala A. Moorjani A Nineteenth - Century Reference to the Use of Tools by the Egyptian Vulture By R. M. Baxter, E. K. Urban & L. H. Brown The African Pitta, Pitta angolensis Vieillot By V. E. M. Burke Nature Note Book Reviews (Published 31/1/69) Page 169 233 235 235 Price Shs. 25/- EAST AFRICA NATURAL HISTORY SOCIETY NOTICE TO CONTRIBUTORS Contributions. The Conunittee is pleased to consider contributions on natural history for publication in the Journal on the understanding that these are not also being offered, wholly or partially, to any other Journal. They should be addressed to the Secretary, P.O. Box 4486, Nairobi. Typescript. Articles should be typed on one side of the paper, in double spacing and with wide margins. Illustrations. These should be in a form suitable for reproduction. The Editor cannot be expected to re-draw. Line drawing should be in Indian ink on Bristol board or thick white paper. Reproduction will be better if they are drawn larger than it is intended they should appear. An indication of the degree of reduction is advisable. Photographs should be printed on glossy paper and a better reproduction is achieved from prints slightly darker in tone than normal. Nomenclature. Where a recent standard work for the area is available (e.g. Praed and Grant for birds) the names given there (both English and scientific) should be used. Initial capitals should be used for specific English names, e.g. Pied Wagtail and small initial letters for group names, e.g. wagtails. Scientific names must always be underlined and followed by the naming authority when mentioned for the first time. Where an English name is used, it must, on first mention be accompanied by the scientific name. References. These are usually abbreviated in the text and listed more fully in alpha- betical order of authors at the end of the article. For example, in the text book reference might be (Pinhey 1956: p. 20). At the bottom of the contribution: Jackson, F. J., 1938. Birds of Kenya and Uganda. Pinhey, E. C. G., 1956. The Emperor Moths of Eastern Africa. J.E. Afr. Nat. Hist. Soc. XXIII No. 1, (98). With short articles it may not be worth making a list of references at the end, but the whole reference in the most abbreviated comprehensible form should then be inserted in the text. Reprints. Provided that they order at the time of submitting their articles, authors may have up to 25 reprints free (other than Nature Notes). Additional copies can be supplied on payment. A list of available back numbers of the Journal and reprints can be obtained from the Librarian, National Museum, P.O. Box 658, Nairobi. JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. XX vn No. 3 (118) JANUARY, 1969 CONTENTS Page A Check List of Plants Collected or Recorded in the Tsavo East National Park 169 By P. J. Greenway Notes on Some Butterflies of the Ngara District of Tanzania, together with a List of Those Recorded 211 By L. A. Haldane Geocarpy as an Adaptation to Afroalpine Solifluction Soils 215 By A. D. Q. Agnew & O. Hedberg Bird Ringing Report 1967 - 1968 217 By G. C. Backhurst A Note on the possible Reproductive Structures in Kenyan, Udotes orientalis A. & E. S. Gepp, (Chlorophyta) 227 By Shakuntala A. Moorjani A Nineteenth - Century Reference to the Use of Tools hy the Egyptian Vulture 231 By R. M. Baxter, E. K. Urban & L. H. Brown The African Pitta, Pitta angolensis Vieillot 233 By V. E. M. Burke Nature Note 235 Book Reviews 235 (Published 31/1/69) Price Shs. 25/- EAST AFRICA NATURAL HISTORY SOCIETY President: DR. A. D. Q, AGNEW Vice-President: C. J. BEECHER ESQ. Executive Committee: G. C. BACKHURST ESQ. MRS A. L. CAMPBELL J. S. KARMALI ESQ. DR. T. R. ODHIAMBO MISS J. R. OSSENT Hon. Editor: DR. P. J. GREENWAY Hon. Treasurer: P. G. ARCHER ESQ. Hon. Secretary: MISS P. M. ALLEN All correspondence in connection with this Journal should be addressed to: The Secretary, East Africa Natural History Society, P.O. Box 4486, Nairobi, Kenya. A CHECK LIST OF PLANTS RECORDED IN TSAVO NATIONAL PARK, EAST By P. J. Greenway INTRODUCTION A preliminary list of the vascular plants of the Tsavo National Park, Kenya, was prepared by Mr. J. B. Gillett and Dr. D. Wood of the East African Herbarium during 1966. This I found most useful during a two month vegetation survey of Tsavo, East, which I was asked to undertake by the Director of Kenya National Parks, Mr. P. M. Olindo, during “the short rains”, December- January 1966-1967. Mr. Gillett’s list covered both the East and West Tsavo National Parks which are considered by the Trustees of the Kenya National Parks as quite separate entities, each with its own Warden in Charge, their separate staffs and organisations. As a result of my two months’ field work I decided to prepare a Check List of the plants of the Tsavo National Park, East, based on the botanical material collected during the survey and a thorough search through the East African Herbarium for specimens which had been collected previously in Tsavo East or the immediate adjacent areas. This search was started in May, carried out intermittently on account of other work, and was completed in September 1967. BOTANICAL COLLECTORS The first traveller to have collected in the area of what is now the Tsavo National Park, East, was J. M. Hildebrandt who in January 1877 began his journey from Mombasa towards Mount Kenya. He explored Ndara and the Ndei hills in the Taita district, and reached Kitui in the Ukamba district, where he spent three months, returning to Mombasa and Zanzibar in August. Next the famous geologist and explorer J. W. Gregory collected in our area during January to March 1893 reaching Mtito Andei on the 23rd March on his jour- ney to the Rift Valley. Then in November 1893, G. F. Scott Elliott began his journey from Mombasa to Ruwenzori collecting plants on his way. Another collector, T. Kassner, passed through our area on his way from Mombasa to Nairobi in the first half of 1902 and again in 1909. The next collector seems to have been a member of the Agricultural Department, J. MacDonald who obtained a few plants near Voi in 1924. Then between April and June 1931 Miss E. R. Napier, one time botanist of the Coryndon Museum, worked around Voi, and along the roads Voi to Taveta and Voi to Tsavo. It was she who first found a Moringa which is new to science and has still to be described and named. Another lady. Miss J. Ossent, working for a firm of contractors making the pipes for the Mzima Springs- Mombasa pipe line, was stationed at Voi in 1955 and 1956 and by her collections of plants around Voi and along the Mombasa-Nairobi Road between Voi and Mtito Andei increased our knowledge of the flora of Tsavo East. From that period until the present, a whole series of collectors, mostly passing through have collected sporadically between Mtito Andei and Mackinnon Road along the Nairobi-Mombasa Road. Among the many, have been C. G. Macarthur of the Game Department, P.R.O. Bally, botanist of the Coryndon Museum, A. Bogdan, Pasture Research Officer, Agricultural Department, B. Verdcourt of the E. A. Her- barium, Drummond and Hemsley of the Royal Botanic Gardens, Kew, R. Polhill of Kew with Samuel Paulo Kibuwa of the E. A. Herbarium and myself and others in passing when'they saw a plant in flower which attracted their attention. Page 170 Of residents in Tsavo National Park, East, the Chief Warden, D. L. W. Sheldrick and others such as P. Napier Bax, C. Moore and R. Schenkel have been particularly active in collecting plants that are eaten by elephants, rhino and other game animals. Recently Mr. and Mrs. P. Hucks, by collecting and photograping the plants in Tsavo East have contributed greatly to our knowledge of the flora. They camped at Voi within the Park and travelled almost daily over several months in search of plants in flower, to photograph in colour, pressing and drying the plant photographed so that the photo is supported by a voucher specimen. To date they have collected over 1,100 specimens and a duplicate of most are deposited in the E. A. Herbarium as well as at the Voi Headquarters of the Tsavo National Park, East. During my stay at Voi in December and January with the the help of my field assist- ant, Mr. Kanuri Kabuie, we collected over 450 numbers of plants in sets of five which have been named and a set of each deposited in the E. A. Herbarium, Nairobi, the Voi Headquarters of Tsavo East, the Royal Botanic Gardens, Kew and elsewhere. This collection has added considerably to the Cyperaceae and Gramineae as well as to other genera recorded in Tsavo East. The collection of botanical material continues by Dr. A. D. Q. Agnew and others with the aid of students of the University College, Nairobi, and the results of their finds are quoted from time to time in the Check List and I acknowledge with grateful thanks the help and comments of Dr. Agnew who has read through the List. In spite of all this collecting activity a great deal more is necessary before we can say that the flora of Tsavo East is well known. Most of the collecting has been done along the roads and tracks between the Galana River and the Park boundary of the Nairobi-Mombasa Road. More is necessary in the north-west from Manyani-Lugard’s Falls Road up to Mtito Andei. Again the area in the south-east, Ndara-Dika Plains- Aruba Lodge-Buchuma Gate-Mackinnon Road, the south of the Aruba Lodge-Sala Gate Road and the Eastern boundary from Mackinnon Road down to the Sala Gate has hardly been touched. Thence northwards from the right bank of the Galana and the Athi rivers to the north-east, north and north-west boundaries including the Yatta Plateau is for all practical purposes botanically almost unknown. Myself, I have not been in this area of the Park because during my December- January visit these areas were so dry that the rangers in the Northern Area Headquarters had to be withdrawn as the water supplies had dried up. THE VEGETATION If one examines the Ordanance Survey 1 : 250,000 maps covering Tsavo National Park, East, the vegetation is indicated by the wording “Bush” or “Thicket” to which the words “Dense”, “Medium”, “Light” or “Scattered” are prefixed, sometimes “Bush and Scattered Trees”, but there are no indications of Grassland. In places there are blue lines or broken-line stipples, particularly in drainage lines and river courses, the solid lines indicating “Marsh” and the broken ones “Marsh, seasonal”. Most of these “Marsh” areas lie in the northern portion of the Park, an area I have not seen. They may be Flood Plains Grassland or even Riverine Forest as between the Park Air Strip and the 1,600ft. contour line along the Voi River. The main vegetation types met within the area southwards of the Athi-Galana Rivers to the southern side of the Park bounded by the Nairobi-Mombasa Road from Manyani Gate to Voi thence the Voi Gate-Ndara Circuit by way of Aruba Lodge down to the Sala Gate in the east are as follows : 1 Forest. (a) Ground-water Forest: Fringing or riverine forest : Swamp Forest Palm Stands. Page 171 2 Woodland. 3 Wooded Grassland. (a) Grouped-tree grassland. (b) Scattered-tree grassland. (c) Shrub or Dwarf tree grassland. 4 Grassland. The Galana River Grass Fringe. 5 Swamp Vegetation. 6 Bushland. (a) Ever-green Bushland. (b) Thicket. Ever-green thicket. 7 Vegetation of Rocky Elills and Rock Pavements. 1 Forest. (a) Ground-water Forest. The greater part of this is Riverine Forest with a very limited amount of Swamp Forest. In Tsavo East the riverine forest consists of a continuous stand of trees which attain a height of 60ft. (c.lSm.) with crowns touching or intermingling, sometimes freely interlaced with lianes. The canopy is not very dense nor does it consist of several distinct layers. Epiphytes such as orchids and ferns are not evident. The trees have simple or buttressed boles and some of them are in full leaf all the year round. The forest floor is covered with herbs and shrubs where the light penetrates. The Riverine Forest and Swamp Forest can be observed along the banks of the Voi and Galana Rivers. That along the Voi River consists of small stands of Fresh-water Swamp Forest near the Voi-Mombasa Road which thins out soon after passing Ndo- lolo and then becomes a very thin and broken strip on both banks of the river of Fringing Forest. Its chief components are: Trees: Frequent, Dobera glabra, Newtonia hildebrandtii var. hildebrandtii. Common, Acacia sp., Kigelia africana. Occasional, Albizia glaberrima var. glabrescens, A. zimmermannii. Ficus ingens, F. sycomorus, Tamarindus indica and Terminalia kilimandscharica. Shrubs: Frequent, Azima tetracantha, Capparis sepiaria var. fischeri, Pluchea dioscoridis, P. ovalis, Salvadora persica, and Combretum ukambensis, locally frequent. Common, Cordia goetzei. Gardenia jovis-tonantis, Lawsonia inermis, Vernonia hildebrandtii, with Lecaniodiscus fraxinifoUus and Meyna tetraphylla locally common, and Ziziphus mucronata, rare. On the Galana River the fringing forest is of quite a different composition, the fringe of trees thinner and confined strictly to the immediate banks of the river, al- though it is said to have been denser until the great floods of 1961. Here in the riverine forest you have two sub-types. Palm Stands of Hyphaena coria- cea, consisting of a thin line of branched palms up to 50ft. (c.l5m.) tall. Then from place to place down the river there are extensive stands of Ever-green Bushland composed of open to closed stands of a very succulent leaved shrub, Suaeda monoica, up to 20 or more feet tall (c.6m.) which forms an exclusive dominant and is an indicator of a saline soil. The Riverine fringe trees are Acacia elatior ssp. elatior, common, looking very decrepit. Populus ilicifolius in small dense stands of all age grades from juveniles about 1ft. (c.30cm.) high to vigorous trees up to about 30ft. (c.9m.) tall, the older more mature ones very few and far between because most of them swept away by the floods. Tamarindus indica and Adansonia digitata, infrequent. Other trees, mostly rare were Spirostachys africana (G. & K. 12, 923; 13,011) or Excoecaria venifera; Cordia goetzei, Garcinia livingstonei, Mimusops fmticosa with the Hanes Ampelocissus africana, Cordia sp. nov. (G. & K. 13,053) and a Menispermaceae, Tiliacora or Triclisia sp. (G. & K. 13,054). 2 Woodland. This consists of land covered with an open cover of trees, their crowns not forming a thickly interlaced canopy, which are leafless for some periods of the year. Scattered evergreen shrubs are present but not conspicuous. Grasses and herbs form the domi- nant ground cover and consist of perennial and annual species, the former, usually with a tufted habit of growth rarely above 3ft. (30.5cm.) tall. This is not extensive in this part of Tsavo East and has no doubt been considerably reduced by the activities of elephants and by fire. Judging by examples adjacent to the Park its dominants are Commiphora-Lannea-Boswellia with three species of Sterculia, S. africana, S. rhynchocarpa, and S. stenocarpa, these are impossible to identify specifically unless they are in fruit. Other associated trees are Cassia abbreviata ssp. kaessneri, Delonix data, Platycelyphium voense, Melia volkensii. Acacia tortilis ssp. spirocarpa, A. reficiens ssp. misera and A. thomasii with Adansonia digitata as an occasional emergent. When good examples are met with their canopy is generally from 20-25ft. (c.6- 8m.) tall or sometimes 30ft. (c.9m.) with an open ground cover of bushes mainly Acanthaceae, Grewia villosa, some Labiates such as Erythrochlamys specta- bilis and an assortment of interesting Euphorbia spp. Such grasses as are present are ephemerals such as Aristida spp., Brachiaria eruciformis, B. leersioides or wide spaced perennials such as Cenchrus ciliaris, Chloris roxburghiana and Schmidtia bulbosa with stands of Eragrostis caespitosa in areas of impeded drainage on bright red soils. Climbers such as Pergularia daemia, Gerrardanthus lobatus, Kedrostris gijef Ipomoea spp. and Thunbergia guerkeana are frequent. 3 Wooded Grassland. This is land covered usually by perennial grasses and other herbs, with either evergreen or deciduous, grouped or scattered, armed or unarmed, trees and shrubs, that cover less that than 50% of the ground. The grasses dominate the aspect though the trees and shrubs are always conspicuous, the former may be densely procumbent or matted, or may grow in patches, tufts or clumps attaining a height of 4ft. (c. 12 1cm.) rarely more. They may completely cover the surface of the ground or be thinly scat- tered so that the soil is easily visible between the grass clumps or mats. There are three types : — a, Grouped-trees Grassland, b. Scattered-trees Grassland and c. Shrub or Dwarf-trees Grassland to be seen in Tsavo East, but, the most widely distributed are the Scattered-trees and the Shrub or Dwaft-trees Grasslands. (a) Grouped-trees Grassland. This may contain stands of Acacia spp., Melia volkensii, Delonix elata. Commi- phora spp., Dobera glabra and Platycelyphium voense. The grasses being Chloris rox- burghiana, Cenchrus ciliaris, Sporobolus helvolus, Cynodon dactylon, Digitaria spp. not infrequently D. macroblephara and Schmidtia bulbosa. Scattered-trees Grassland and Shrub or Dwarf-trees Grassland merge into one another and are not easy to delimit. Their more open phase deteriorates into Grassland as a result of the activities of elephants or of grass fires. (b) Scattered-trees Grassland. In this type the trees can be Melia volkensii-Platycelyphium voense-Commiphora spp. with a very occasional Euphorbia robecchii or Adansonia digitata with a similar composition of grass species to those found in the Grouped-trees Grassland. Page 173 (c) Shrub or Dwarf-trees Grassland. This type is widespread, its woody components being Boscia coriacea which is found throughout Tsavo East, Dobera glabra frequent. Balanites orbicularis, rare, Cadaba heterotricha, Terminalia parvula, T. spinosa, Platycelyphium voense and Commiphora spp. with the ubiquitous shrubs Sericocomopsis hildebrandtii and Premna resinosa. Other shrubs to be met with in this type are shrubby Acanthaceae, Maerua spp., Thylachium thomasii, Calyptrotheca somalensis, C. taitensis, Cordia ovalis, C. gharaf, Ehretia teitensis, Premna hildebrandtii, P. holstii and the common Combretum aculeatum. 4 Grassland. This is land covered with grasses and other herbs generally perennial, sometimes with evergreen or deciduous trees or shrubs, either very scattered or in small isolated groups, in either case not covering more than 10% of the ground. The grasses may be procumbent or mat-like, forming a dense or thin carpet on the ground or in clumps or tussocks, close or widely spaced or they may be perennial forming a continuous ground cover. They may be from a few inches tall to about 4ft. (c. 121cm.) rarely more. There is hardly any true grassland in Tsavo East that is without any trees or shrubs in its composition. An example is to be seen in the Ndara-Dika Plains area. This had been burnt so it was impossible to tell what species of grass it was composed of at the time of my visit. Along the Pipe-line Road from Worssera Hill towards Irima there is a fairly extensive stand of grassland which originally contained islands of trees or bush groups, now only relics of Acacia bussei-Cadaba heterotricha-Combretum aculeatum on a black cracking clay with grey sand-washes from the slopes of the nearby hills. The grasses collected and noted in this area were : — Brachiaria deflexa, B. leersioides, Brachiaria sp. G. & K. 12,717; Cenchrus ciliaris, Digitaria rnacroblephara, D. rivae, Ischaemum afrum, Latipes senegalensis, Leptochloa obtusiflora, Panicum infestum, P. maximum and Schoenfeldia transiens. It formed an open to closed ground cover with scattered Cyperus obtusiflorus and C. giolii, the latter, a bulbous based annual, was common on the margins of the road. Other herbs were also present. A further example, but in a very open bushland can be seen along the Sobo- Aruba Lodge Road on a low ridge composed of a pale buff sandy and stony loam. The grasses together with the woody vegetation formed a very open cover and the ground between was visible between the different species. The grasses were Aristida adscensionis, A. barbicollis, Cenchrus ciliaris, Chloris roxburghiana, Chrysopogon aucheri var. quinqueplumis, Cymbopogon pospischilii, Digitaria pennata, D. rivae, Enneapogon elegans, Latipes senegalensis, Tetrapogon tenellus and the rare Tetrachaete elionuroides was common on the sides of the road as well as being scattered in the more open parts of the grass cover. The more important components of the woody vegetation were Commiphora spp., Terminalia orbicularis, Boscia coriacea. Acacia tortilis ssp. spirocarpa, Caesal- pinia trothae, Caucanthus albidus. Cassia longiracemosa, Ehretia teitensis and Thy- lachium thomasii. The Galana River Grass Fringe. The Galana River varies greatly in volume as a result of up-country rains ; in the very high floods of 1961 great damage was done to its riverine fringes, the water spreading well beyond the banks. At the height of the dry season it is extremely shallow and it is possible to cross from bank to bank in most places without difficulty. In the riverine fringe the Cyperaceae are represented by three genera and nine species; they are: — Cyperus articulatus, aquatic, frequent; C. alternifolius ?,?vp.flabelli- formis, semi-aquatic, common ; C. compressus, semi-aquatic on sand banks ; on sand banks, ^C. longus ssp. tenuiflorus, locally common and C. maculatus, uncommon. Fimbristylis bis-umbellatus and Lipocarpus chinensis, rare; Cyperus immensus var. taylori, local and C. laevigatus very local. All on the left bank of the Galana. The Gramineae, eighteen genera with twenty-three species are also recorded from the left bank and the following are represented: Aquatic, Diplachne caudata, local. Semi-aquatic, repens, Paspalidium geminatum, both locally dominant, the latter on sand banks; Hemarthria natans and Leersia hexandra, locally common ; Phrag- mites mauritiana, common. Sandbanks, Aristida adscensionis, A. papposa, Eragrostis horizontalis, common; Chloris barbatus and Panicum meyerianum local; Eragrostis exasperata, E. superba, Hyparrhenia filipendula var. pilosa and Perot is patens rare. Cynodon dactylon, frequently dominant, Paspalum vaginatum, locally dominant, Heteropogon contortus, locally common and Cenchrus setigerus and Schmidtia bulbosa both local; Phragmites karka and Sporobolus virginicus rare. On rocks, Sporobolus sp. G. & K. 13,073, local and Eulalia sp. aff. E. ferruginea G. & K. 12,920, rare. Of the above species Cyperus laevigatus, Diplachne caudata, Paspalum vaginatum and Sporo- bolus virginicus are saline-alkaline indicators. 5 Swamp Vegetation. This type can be Permanent Swamp, or Seasonal Swamp Grassland, the former in the southern area of Tsavo East and to be seen on the south bank of the Voi River where it would be classed as Ground-water Forest and is not within the bounds of Tsavo East. Seasonal Swamp Grassland is a flat area where free water accumulates on the sur- face for some periods of the year or which can be temporarily flooded to a shallow depth. The land could be covered with trees when it would be classed as Ground-water Forest or with herbaceous vegetation, the chief components being grasses, rushes, sedges and aquatic plants. The chief areas of interest in Tsavo East are the Kandiri Lake and the Aruba Dam where there is permanent water. Also water-holes, widely scattered, and wallows, most of these of a seasonal nature. It is apparent that Kandiri Lake on the Voi Gate-Aruba Lodge Road was originally an old ox-bow of the Voi River, but has been further deepened and enlarged by the the Park authorities and is fed seasonally by the river. The lake is surrounded by much battered open bushland of Newtonia hildebrandtii, Thylachium thomasii, Salvadora persica and scattered bush clumps of Ecbolium ample- xicaule, E. revolutum, Maerua denhardtiorum, M. subcordatum and the ubiquitous Sericocomopsis; the dominant grasses being Cynodon dactylon, Cenchrus ciliaris and Bothriochloa radicans and in much trodden bare patches the annual herbs Alternan- thera pungens. Euphorbia hirta, Tribulus, and, uncommon Corchorus olitorius or C. trilocularis. In the water are to be found Nymphaea capensis as islands, rafts of Marsilea minuta and dominant Najas graminea submerged, as well as stands of Echinochloa haploclada, rare Panicum sp. nov. G. & K 12,674 and very local Conyza aegyptiaca var. G. & K. 12,673. With the advent of the dry season as other sources of water dry up this lake is considerably reduced hy game, especially elephants and water buck using it for drinking purposes and the aquatic and surrounding vegetation much reduced by grazing, browsing and by trampling of the cover. The Aruba Dam across the Voi River is a source of permanent water. With the alternation of the wet and dry seasons there is a considerable rise and fall in the level of the water in the dam with much exposure of a very trampled grey-black clay loam at the bead of the dam where the river discharges its waters. This area, as the water recedes in the dry season is invaded by the herbs Coldenia procumbens, Eclipta pro- strata, Alternanthera sessilis as co-dominants and scattered Corchorus, Rorippa madagascariensis, Mollugo nudicaule with grasses Echinochloa haploclada and Sporo- bolus helvohis, the latter as a fringe. Towards the river proper are the relics of clumps Page 175 of Cyperus immensus, much damaged through browsing and trampling by elephants and other game. The banks of the dam itself are treeless and Cyperaceae, except for Cyperus articu- latus as a water-side fringe, are conspicuous by their absence. A rare bush clump of Sesbania sericea, heavily browsed, may be seen. The grassland fringe consists of Cyno- don dactylon, Paspalidium geminatum, co-dominant with Sporobolus helvolus and Echinochloa haploclada and rare Eragrostis aethiopica in disturbed ground. The herbs, Grangea maderaspatana, annual, and Phyla nodijiora, perennial, will also be noticed. Scattered water-holes and wallows may have clumps of bushes of Lawsonia inennis, Ziziphus mucronata, or single bushes of Gardenia jovis-tonantis, this latter seems to be indestructable by game, occasionally a Tamarindus indica or Kigelia africana tree, or rarely Acacia stuhlmannii. If the water rapidly evaporates or is quickly dispersed by the game drinking or wallowing the bare ground is first invaded by annuals such as Glinus setiflorus to be followed by the grasses Echinochloa haploclada or Sporobolus helvolus, the latter an indicator of a slightly alkaline soil. Sometimes an occasional scattering of Cyperus distans will also be seen with either of these two grasses. 6 Bushland. This is land covered with more than 50 % cover of shrubs or small trees growing denseley together. The trees or bushes may be evergreen or deciduous, armed or unarmed. The bushes have no clearly defined boles and may be from a few feet to 15ft. (2-5m.) tall, rarely more. Tall trees such as Adansonia digitata, Delonix elata, Melia volkensii and rarely Lonchocarpus-Terminalia, Euphorbia robecchii, may be present, occasionally in clumps, more often as widely scattered individuals. Herbs, ephemeral or succulent or both and grasses not above 3ft. (Im.) tall, mostly annual or short-lived perennial form the ground cover under deciduous bushland. Bushland varies from place to place in the Park and is rich in species, the bulk of it is deciduous but there is (a) Evergreen Bushland, dominated by Suaeda monoica which has already been referred to under Riverine Forest and occurring along the south bank of the Galana River. Several different communities occur in the Park, in most of them emergent trees of Commiphora spp.-Delonix elata-Sterculia spp. -Melia volkensii may be present. One example studied was on a bright orange-red loam adjacent to a granite intrusion. The following shrubs or small trees were present: Dirichletia glaucescens. Euphorbia engleri, E. scheffleri, Hymenodction parvifolium. Commiphora riparia, Strychnos decussata, Combretum aculeatum, Lannea alata, Premna resinosa, Boswellia hildebrandtii, Erythrochlamys spectabilis, Bauhinia taitensis, Terminalia orbicularis, Bridelia taitensis, Sesamothamnus rivae, Calyptrotheca somalensis and Grewia fallax. Associated with them were the climbers Sarcostemma viminale, Gerrardanthus lobatus, Adenia globosa, Thunbergia guerkeana and Vanilla roscheri. A second community, on a brown sandy clay loam, consisted of Combretum grotei, C. aculeatum, Dobera glabra, Cadaba heterotricha, Terminalia spinosa, T. parvula, Caucanthus albidus, Caesalpinia trothae, Ehretia teitensis, Sericocomopsis hildebrandtii, Terminalia orbicularis. Acacia tortilis ssp. spirocarpa and Balanites orbicularis. A third, rather open, on a buff-brown sandy loam, contained Acacia bussei, A. mellifera, A. nilotica ssp. subalata, A. reficiens ssp. misera, Boscia coriacea, Combre- tum aculeatum, Combretum sp. aff. C. grotei, G. EAH. 12,204, Commiphora africana, Cordia ovalis, Grewia tembensis var. kakothamnos and G. villosa with Sericocomopsis. (b) Thicket. As here defined is a close to rather open assemblage of coppicing or virgate deciduous or evergreen bushes with or without scattered trees projecting through the Page 176 main canopy which may be from about 6 to 15ft. (3-5m.) rarely up to 25ft. (c.7m.). Deciduous thicket, some examples observed were : Givotia gosai, this is a virgately branched shrub to 20ft. (7m) with several main branches from ground level. It is very locally dominant forming an open thicket with Melia volkensii. Commiphora spp. and small stands of Platycelyphium voense occur with it. To be seen near Irima Hill and along the Aruba-Sobo Road where it is quite extensive. Bauhinia taitensis thicket, in small patches in open Delonix elata-Boscia coriacea- Platycelyphium voense-Hymenodictyon parvifolium bushland on a rich red loam. The Bauhinia up to about 6ft. (2m.) tall, has very attractive flowers which can be white, yellow or very pale pink. Ochna inermis thicket, a densely branched shrub to 6ft. (2m.) tall with clusters of very fugitive yellow flowers which are followed by small glossy black fruits enveloped in a bright crimson calyx. A co-dominant with Premna resinosa in open to closed thicket to the right of the Pipe-line Track after passing Worsessa Look-out. Evergreen thicket, this is composed of Anisotes parvifolius, a not particularly leafy shrub with green virgate stems up to about 12ft. (3m.) tall. This can be seen in various parts of Tsavo East, but the best examples are to be found down the Voi-Lugard’s Falls Road towards the Mbololo River Drift, and between Voi Gate and Aruba Lodge. 7 Vegetation of Rocky Hills and Rock Pavements. The rocky hills and rock pavements are thinly covered with trees and shrubs, quite a number of which are not found in other habitats. In the course of the survey the following hills and rocky places, Mzinga Hill, Wasessi, Irima Hill, Mudanda Rock and Sobo Rocks, were visited from time to time and the following species were noted : Trees Common: Acacia tortilis ssp. spirocarpa. Delonix data. Euphorbia quinquecostata. Ficus populifolius, F. sonderi. Melia volkensii. Occasional : Adansonia digitata. Albizia anthelmintica. Berchemia discolor. Commiphora sp. G. & K. 12,658 and 12,893. Platycelyphium voense. Sterculia stenocarpa. Terminalia kilimandscharica. Local : Euphorbia kibwezensis. Ficus ingens. Rare: Dalbergia melanoxylon. Diospyros consolata. Holarrhenia febrifuga. Lannea stuhl- mannii. Sterculia africana. Vitex payos. Shrubs Frequent : Boscia coriacea. Combretum aculeatum. Euphorbia scheffleri. Strychnos decussata. Tarenna graveolens. Common: Boswellia hildebrandtii. Boscia sp. Bridelia taitensis. Combretum exalatum. Cordia ovalis. Dirichletia glaucescens. Euphorbia heterochroma. Grewia villosa. Haplocoelum foliolosum. Premna resinosa. Sarcostemma viminale. Thylachium thomasii. Vepriseugeni- ifolia. Locally common : Combretum ukambensis. Dombeya praetermissa. D. umbraculifera. Euphorbia jatrophoides. E. polyantha. Ruttya fruticosa. Sacleuxia newii. Thunbergia holstii. Vello- zia aequatorialis. Occasional : Cassia abbreviata ssp. kaessneri. Crotalaria scassellattii. Croton dichogamus . Grewia nematopus. Strophanthus mirabilis. Vernonia wakefieldii. Rare: Acacia brevispicata. Adenium obesum. Euphorbia sp. G. & K. 12,956. Psychotria sp. G. & K. 12,774 and 12,824. Solanwn taitensis. Strychnos sp. G. & K. 12,660. Tephrosia noctiflora. Tinnea aethiopica. Triaspis erlangeri. Lianes Common: Cissus quadrangularis. Locally common : Cardiosperinwn halicacabum. Cissus rotundifolius. Ipomoea bullata. Rare: Cissus cactiformis. Herbs Frequent: Sansevieria ehrenbergiana. Common : Barleria prionitis. Celosia anthelmintica. Ruellia patula. Locally common : Amorphophallus gallaensis. Centemopsis rubra. Chlorophytum tenuifolium. Cleome stenopetala. Crossandra mucronata. Jatropha spicata. Merremia pinnata. Sansevieria singularis. Stylochiton angustifolius. Urginea sp. G. & K. 12,975 and 13,030. Occasional : Albuca wakefieldii. Anthericum sujfructicosum. Asystasia somalensis. Barleria diffusa. Polygala amboniensis. Zornia glochidiata. Rare: Amaranthus angustifolium ssp. Silvester. Caralluma speciosa. Corchorus sp. aff. C. baldaccii G. & K. 12,957. Cryptostephanus haemanthoides. Drimiopsis sp. G. & K. 12,854. Echidnopsis sp. nov. G. & K. 12,952. Hybanthus danguyanus. Stemodiopsis buchananii, S. humilis. Grasses Frequent : Schmidtia bulbosa. Common: Brachiaria leersioides. Enteropogon macrostachyus. Panicum maximum. Locally common : Andropogon schinzii. Digitaria milanjiana. Tricholaena eichingeri. Occasional : Heteropogon contortus. Oropetium thamaeum. Rhynchelytrum villosum. Sporobolus pellucidus. Tragus berteronianus. Rare: Dactyloctenium sp. G. & K. 12,738. Eragrostis ciliaris. Oropetium sp. nov. G. & K. 12,766. Setaria pallidifusca. Sporobolus festivus. Cyperaceae Common : Kyllinga alba. Locally common: Bulbostylis sp. G. & K. 12,770 and 12,949. Fimbristylis exilis. Page 178 Occasional: Mariscus sp. G. & K. 12,662. Ferns Frequent: Actiniopteris dimorpha. Locally common : Selaginella dregei. Occasional : Actiniopteris semiflabellata. Rare: Actiniopteris radiata. Pellaea adiantoides. Lichens Locally common: Rocella pattensis. Teloschistes validus. Both on dead or dying trees. THE ARRANGEMENT OF THE FAMILIES Only two fungi have been recorded and the arrangment of these follows Clements & Shear. The two lichens follows Smith’s Lichens. For the Ferns and Fern Allies the arrangment as far as possible follows that of Copeland, E. B., Genera Filicum. For the Flowering Plants the arrangement follows that of Hutchinson’s The Fami- lies of Flowering Plants (1926 and 1934), but in each family the arrangement of the genera and species is by alphabetical sequence. Where an account of a family has been published in ‘The Flora of Tropical East Africa’ the family name is followed by F.T.E.A. with the year of publication. For botanists a taxonomic arrangement is ideal but it is quite useless to park wardens and rangers with little or no botanical knowledge and with no botanical books of reference in their park headquarters to which they can refer. To cater for those with no botanical knowledge 1 have provided an index to the families and genera in the hope they will be able to find their way about in the Check List. ACKNOWLEDGEMENTS My thanks are due to Mr. P. M. Olindo, Director of the Kenya National Parks for giving me the opportunity to do a botanical survey of the Tsavo East National Park, and to Dr. O. Starnes, Director, E.A. Agriculture and Forestry Research Organi- sation, for permitting me to undertake the survey and for permission to publish the results. To Mr. D. L. W. Sheldrick, the Park Warden for help and the facilities pro- vided and to Mr. R. Jenkins, the Assistant Park Warden. Also to Mr. J. B. Gillett, Botanist in Charge of the East African Herbarium for the loan of a Field Assistant, Mr. Kanuri, from the Herbarium, whose help enabled me to make a reasonable collection of botanical specimens considering the time of year and the condition of the vegetation in the Tsavo National Park, East. BIBLIOGRAPHY Gillett, J. B. The History of the Botanical Exploration of the Area of the Flora of Tropical East Africa (Uganda, Kenya, Tanganyika and Zanzibar), Comptes Rendus 4, e, 217 (1961), Paris. Greenway, P. J. Second Draft Report on Vegetation Classification for the Approval of the Vegetation Classification Committee, Pasture Research Conference, 1943, (Cyclostyled). Gregory, J. W, The Great Rift Valey (1896), London. Scott Elliot, G. F, A Naturalist in Mid Africa (1896), London. CRYPTOGAMAE FUNGI 74 Polyporaceae Polystictus versicolor Fr. Gregory sn., not seen. 77 Lycoperdaceae Podaxis pistillaris (Pers.) Morse Sr. G. LICHENES 22 Rocellaceae Rocella pattensis Dodge G. &K. 12,859; 12,888. 49 Teloschistaceae Teloschistes validus (Muell. Arg.) Hillm. G. &K. 12,851. FERNS AND FERN ALLIES Equisetaceae Equisetum rainosissimiim Desf. G. & K. 13,007. Selaginellaceae Selaginella dregei (Presl.) Hieron. G. 9615; G. & K. 12,654; Gillett 17,278. 6 Pteridaceae Actiniopteris dimorpha Pic.-Serm. G. & K. 12,656. A. semiflabellata Pic.-Serm. Mucks 991. Pellaea adiantoides (Willd.) Kaulf G. & K. 12,822. A. radiatalAhk. Gregory sn., not seen. P. viridis (Forsk.) Prantl. (Syn. P. hastata Link.) Gregory sn., not seen. 6A Adiantaceae Adiantum incisum Forsk. (Syn. A. caudatum L.) Gregory sn.. not seen. 14 Aspleniaceae Ceterach cordatum (Thunb.) Desv. (Syn. Asplenium ceterach L.) Gregory sn., not seen. Marsileaceae Marsilea diffusa A. Br. G. & K. 12,672. M. minuta L. G. & K. 13,035. DICOTYLEDONES I Archichlamydeae (Families Nos. 15-213). 15 Ranunculaceae F. T. E. A. (1952). Clematis simensis Fres. var. dentata O. Ktze. Gregory sn., not seen. 18 Nymphaeaceae Nymphaea capensis Thunb. G. & K. 13,059. B Page 180 Tiliacora sp. or Triclisia sp. G. & K. 13,054. 23 Menispermaceae F. T. E. A. (1956). 26 Hydnoraceae Hyditora abyssinica Schweinf. H. bogosensis Beccari e desc. Mrs. d. Jong H. 362/62; Venour H. 114/50. Sheldrick H. 177/63. It is recorded that both species are parasitic on the roots of Acacia spp., whilst Adansonia digitata L., is a host plant of H. bogosensis. 32 Papaveraceae F. T. E. A. (1962). 34 Turneraceae F. T. E. A. (1956). 36 Capparidaceae F. T. E. A. (1964). Boscia coriacea Pax G. &K. 12,952; 13,075 Cadaba farinosa Forsk. spp. adenotricha (Gilg & Ben.) R. Grab. G. & K. 12,697. C. heterotricha Hook. C. ruspolii Gilg G. & K. 12,832; Hucks 822. G. & K. EAH. 13,702. C. stenopoda Gilg & Ben. G. &K. 13,081. Capparis sepiaria L. var. subglabra (Oliv.) De Wolf C. tomentosa Lam. G. & K. 12,693; Hucks 889. G. & K. 12,994. Cleome briquetii Polhill C. hirta (Klotsch) Oliv. G. & K. 12,748; 12,850; Hucks 472; 537;. G. & K. 12,901. C. macrophylla (Klotzsch) Briq. C. tenella L. f. Hucks 563. G. & K. 12,755. Gynandropsis gynandra (L.) Briq. (Syn. G. pentaphylla DC.) Gregory sn., not seen. Maerua angolensis DC. M. crassifolia Forsk. Sr. G. Hucks 413; Ossent 66. M. denhardtiorum Gilg M. edulis (Gilg & Ben.) De Wolf G. & K. 13,043; Hucks 833. G. & K. 12,700; 13,079. M. holstii Gilg M. kirkii (Oliv.) F. White Ossent 110. Hucks 826; Ossent 41. M. macrantha Gilg M. ? oblongifolia (Forsk.) A. Rich. Hucks 419. Verdcourt 1594. M. subcordata (Gilg) De Wolf M. triphylla A. Rich. var. calophylla (Gilg) De G. &K. 12,671; 13,080. Wolf Thylachium thomasii Gilg Hucks 469. G. & K. 12,986. 37 Moringaceae 39 Cruciferae Farsetia ? longisiliqua Decne G. & K. 12,722. Rorippa madagascariensis (DC.) Hara G. & K. 13,062. 40 Violaceae Hybanthus danguyanus H. Perrier H. enneaspermus (L.) F. Muell. var. enneaspermus G. & K. 12,916. G. & K. 13,018; Hucks 1,059. Rinorea elliptica (Oliv.) O. Ktze G. &K. 13,018. Page 181 42 Polygalaceae Polygala erioptera DC. P. kilimandscharica Chod. G. & K. 12,909. Napier 1008. P. linifiora Chod. P. petitiana A. Rich. G. & K. 12,917; 12,973. Gregory sn., not seen. P. sphenoptera Fres. Polygala sp. G. & K. 12,863; Hucks 549; 1060. G. & K. 13,069. Polygala ip. Hucks 1018. 45 Crassulaceae. Kalanchoe sp. Hucks 1004. 54 Aizoaceae (Ficoidaceae) F. T. E. A. (1961). Corbichonia decumbens (Forsk.) Exell Hucks 651a; Sheldrick B8567. Gisekia pharnaceoides L. Hucks 468; 786. Glinus setiflorus Forsk. G. & K. 12,973; Hucks 1041; 1056. Hypertelis bowkeram Sond. Hucks 1068. Limeum viscosum (J. Gay) Fenzl var. kenyense Friedr. Napier 1000. Mollugo cerviana (L.) Ser. var. spathulifolia Fenzl M. nudicaulis Lam. Hucks 656; Napier 902. Bally B8797; Gillett 17,268. Triauthema ceratosepala Volkens & Irmsch. T. portulacastnim L. Hucks 440. Hucks 729. Zaleya pentandra (L.) Jeffr. Hucks 60; Napier Bax TNP/E/26. 56 Portulacaceae Calyptrotheca somalensis Gilg G. &K. 12,931. Portulaca oleracea L. Hucks 595; Napier 1044. P. quadrifida L. Hucks 724; 757. Portulaca sp. Napier 951a. Talinum caffrum (Thunb.) Eck. & Zey. G. &. K. 12,734. C. taitensis (Pax & Vatke) Brenan G. &K. 12,821. P. pilosa L. Verdcourt & Polhill 2707. Portulaca sp. Hucks 540. T. portulacifolium (Forsk.) Schweinf Hucks 218; 523; Ossent 81. 57 Polygonaceae F. T. E. A. (1958). Oxygonum atriplicifolium (Meisn.) Martel. O. stuhlmannii Dammer Hucks 774; 775. Hucks 27/A. Polygonum senegalensis Meisn. Hucks 338. 61 Chenopodiaceae F. T. E. A. (1954). Chenopodium ambrosioides L. Hucks 366. Suaeda monoica J. F. Gmel. G. & K. 12,647; Gregory sn., not seen. 63 Amaranthaceae Achyranthes aspera L. Williams 7; Sheldrick TNP/E/100: Hucks 1050. Aerva lanata (L.) Juss. var elegans Susseng. G. 9824; G. & K. 12,941 Alternanthera pungens H. B. K. G. &K. 12,666 Amaranthus ascltersonianus Thellg. Napier 911. Celosia acroprosoides Hochst. G. &K. 12,823; Hucks 793. A. persica (Burm. f.) Merr. Agnew et al. 7332; Hucks 57. A. sessilis (L.) R. Br. G. & K. 13,041. A. graecizans L. G. & K. 12,845; Polhill & Paulo 937. C. trigyna L. Napier 995. Page 182 63 Amaranthaceas {cont.) Centemopsis rubra (Lopr.) Schinz Hucks 685; Napier Bax TNP/Gs/1. Digera mucronata (L.) Mast. G. & K. 12,844; Napier 906. Gomphrena celosioides Mart. G. & K. 12,783; Hucks 353; 1023. Pleuropterantha sp. fide Kew G. 9776. Psilotrichum boivinianum Cavace P. scleranthum Oliv. G. & K. 13,020; Hucks 621. Hucks 816. Pupalia lappacea (L.) Juss. G. & K. 12,684; Hucks 787. Sericocomopsis hildebrandtii (C. B. Cl.) Schinz Agnew et al. 7354; Napier Bax TNP/R/8A; Ossent 26. S. pallida (S. Moore) Schinz G. & K. 12,635; Napier 1006; Sheldrick TNPR. 8. Hucks 956; 267. 66 Zygophyllaceae Tribulus cistoides L. T. terrestris L. G. & K. 12,781 ; Hucks 474. G. & K. 12,817. 67 Geraniaceae Monsonia senegalensis Guill. & Perr. G. & K. 12,947. 72 Lythraceae Ammannia auriculata Willd. Hucks 1042. Lawsonia inermis L. G. & K. 12,649. 77 Onagraceae F. T. E. A. (1953). Ludwigia pubescens (L.) Hara ssp. brevisepala (Brenan) Raven L. stolonifera (Guill. & Perr.) Raven Hucks 593; Napier 999. G. & K. 13,003; 13,034. 81 Thymelaeceae Gnidia latifolia (Oliv.) Gilg Hucks 584; 781. 83 Nyctaginaceae Boerhavia coccinea Mill. B. diffusa L. Hucks 200; 248. Hucks 160; 767. B. erecta L. B. repens L. Hucks 106; 180. Hucks 1094. Commicarpus pedunculosus (A. Rich.) Cuf. C. plumbagineus (Cav.) Standi. Hucks 660. Napier 1051. C. stellatus (Wight) Berhaut G. & K. 12,880; Sheldrick B8571. 101 Passifloraceae Adenia globosa Engl. A. gummifera {HaT\.) Harms Hucks 843 ; Irwin 255. Hucks 5 1 2. A. keramanthus Harms A. scheffleri Engl. & Harms Gardner 3008. G. 9544. Adenia sp. Hucks 1047. Tryphostemma hanningtonianum Mast. T. lanceolatum Engl. Hucks 531 ; 600. Hucks 327; 1072. Tryphostemma sp. G. & K. 12,810. 103 Cucurbitaceae F. T. E. A. (1967). Cephalopentandra ecirrhosa (Cogn.) C. Jeffrey Hucks 971. Citrullus lanatus (Thunb.) Mansf. Napier 1048. Coccinia grandis (L.) Voigt C. microphylla Gilg Hucks 392 ; 1 040 ; 1 080. Hucks 49 1 . 103 Cucurbitaceae {cont.) Coccinia trilobata (Cogn.) C. Jeffrey Hucks 61 ; Verdcourt 3888. Corallocarpus epigaeus (Rottl.) C. B. Cl. Hucks 554 ; Verdcourt 1111. Cucumella engleri (Gilg) C. Jeffrey Drummond & Hemsley 4220. Cucumis aculeatus Cogn. Napier Bax TNP/E/59. C. figarei Naud. Verdcourt 3897. C. sativus L. Hucks 572. Cyclantheropsis parviflora (Cogn.) Harms Rauh Ks 827. Gerrardanthus lobatus (Cogn.) C. Jeffrey Agnew 7335; Hucks 120; 581. Kedrostis foetidissima (Jacq.) Cogn. Hucks 908; Sheldrick 1003; Verdcourt 3887. K. hirtella (Naud.) Cogn. Hucks 560. K. pseudogijef (Gilg) C. Jeffrey Hucks 444; 481. Lagenaria sphaerica (Sond.) Naud. Napier 969; R. Williams & Sheldrick TNP/E/101. Momordica rostrata A. Zimm. Napier 1023; Ossent 45. M. spinosa (Gilg) Chiov. G.10,814; 876- Hucks Peponium vogelii (Hook, f.) Engl. Agnew 7337. Trochomeria sp. ‘A’ = Padwa 186. Hucks 518; 911; G. & K. 12,797. Zehneria pallidinervia (Harms) C. Jeffrey Drummond & Hemsely 4094; Irvin 382. C. dipsaceus Spach Hucks 72; 80. C. prophetarum L. ssp. dissectus (Naud.) C, Jeffrey Greenway EAH. 12,205; Hucks917; 1079. K. gijef (J. F. Gmel.) C. Jeffrey Hucks 891. K. leloja (J. F. Gmel.) C. Jeffrey Ossent 118. Zehneria sp. Hucks 902. 114 Ochnaceae O. ovata F. Hoffm. Hucks 462. 121 Combretaceae Combretum aculeatum Vent. G, & K. 12,666; 12,860. C. grotei Exell Hucks 526, C. ukambensis Engl. G. & K. 12,843; Hucks 439. Combretum sp. Bally B8164; Drummond & Hemsley 4104. Terminalia kilimandscharica Engl. G. & K. 12,779; 13,058. T. parvula Pamp. G. &K. 12,641; 12,919. T. spinosa Engl. Hucks 406. C. exalatum Engl. G. & K, 12,764; Hucks 473; 916. C. mode R. Br. MacArthur B271. Combretum sp. aff. C. grotei Exell Greenway EAH. 12,204; Hucks 857. T. orbicularis Engl. & Diels G. & K. 12,643; Hucks 320. T. prunioides Laws. G. & K. 12, 640: 12,752; 12,837. Corchorus olitorius L. C. trilocularis L. Hucks 143. Greenway EAH. 12,203; Napier 904. Corchorus sp. aff. C. baldaccii Mattei G. &K. 12,951; Hucks 713. Page 184 128 Tiliacaae (cont.) Grewia bicolor Juss. Hucks 734; Sheldrick TNP/E/2. G. forbesii Mast. Agnew et al. 5827; Hucks 763. G. nematopus K. Schum. G. &K. 12,858; Ossent 62. G. tenax (Forsk.) Fiori Hucks 424, G. v/7/oia WiUd. BaUy B8631 ; Hucks 316. Triumfetta flavescens A. Rich. Hucks 237; 350; 1061. G. fallax K. Schum. Greenway EAH. 12,201; G. & K. 12,820; Hucks 460. G. lilacina K. Schum. Hucks 334; Ossent 51. G. tembensis Fres var. kakothamnos (K. Schum.) Burrett A^ew et al. 7353; Ossent 127. G. tristis K. Schum. Hucks 423. 130 Sterculiaceae Dombeya praetermissa Dunkley Napier 1012, ex num. Kew Bull. 1934, p. 183. Hermannia exappendiculata (Mast.) K. Schum. Napier 982; Hucks 188. H. glanduligera K. Schum. G. & K. 12,905; Hucks 176. H. uhligii Engl. Hucks 403; Sheldrick T.N.P/E/107. Melhania ovata (Cav.) Spreng. Napier 926; Moore & Sheldrick TNP/E/159. M. velutina Fosrk. (Syn. M. ferruginea A. Rich.) TNP/E/73. Sterculia africana (Lour.) Fiori G. 9547; Trapnell 2224. S. stenocarpa H. Winkl. G .10,811; G. & K. 12,676; 13,078; Hucks 314. Waltheria indica L. G. & K. 12,715; Napier 963; Sheldrick & Moore TNP/E/152. 131 Bombacaeae D. umbraculifera K. Schum. Agnew et al. 5812. H. fischeri K. Schum. Drummond & Hemsley 4098. H. oliveri K. Schum. G. 10,463; Macdonald 868. 132 Malvaceae Abutilon fruticosum Guill. & Perr. G. 9,502; Hucks B328. A. hirtum (Lam.) Sweet Nattrass 1053. Hibiscus aponeurus Sprague & Hutch. Hucks 279; 821. H. cannabinus L. Napier 1060. H. micranthus L. f. Hucks 358; Sheldrick & Moore TNP/E/137. H. surattensis L. Napier 991. Pavonia arabica Boiss. Bally B8669; G. & K. 12,896. P. grewioides Boiss. Hucks 992. P. zeylanica (L.) Cav. G. & K. 12,906; Hucks 344. Pavonia sp. Napierl032; Ossent 120. Sida ovata Forsk. Drummond & Hemsley 4096; G. & K. 13,022. Thespesia danis Oliv. Burch 62/86. A. grandiflorum Don Napier Bax TNP/E/28. A. mauritianum (Jacq.) Medic. Hucks B313. H. calyphyllus Cav. Hucks 39; 354. H. greenwayi Bak. f. BeU HI 57/56. H. palmatus Forsk. G. & K. 12,944; Hucks 56. H. vitifolius L. (or related species) G. &K. 12,683; Hucks 252. P. elegans Guerke G. &K. 12,819; Hucks 29. P. patens (Andr.) Chiov. G. «&K. 13,083; Hucks 313. P. zeylanica (L.) Cav. var. microphylla Ulbr. McDonald 867. S. rhombifolia L. Gregory sn., not seen. Page 185 133 Malpighiaceac F. T. E. A. (1968). Acridocarpus zanzibaricus A. Juss. Hucks 505;889 Caucanthus albidus (Nied.) Nied. G. &K. 12,885; Hucks 425. Triaspis erlangeri Engl. G. & K. 12,735; Napier 1036. C. auriculatus (Radik.) Nied. G. 9741; Ossent 152. T. niedenzuiana Engl. G. & K. 12,757; Hucks 352; Verdcourt 3892. 136 Euphorbiaceae Acalypha ciliata Forsk. G. & K. 12,841 ; Polhill & Paulo 952. A. indica L. Bally B8044. Bridelia taitensis Pax & Vatke Agnew et al. 7327; G. & K. 12,892. Cephalocroton nudus Pax & K. Hoffm. G. & K. 12,727. Croton confertus Bak. e desc. G. & K. 13,088. Dalechampia scandens L. Hucks 390. D. trifoliata Verde. & Greenway Hucks 1057; Polhill & Paulo 927. Erythrococca atrovirens (Pax) Prain G. 9614. Euphorbia acalyphoides Boiss. G. & K. 12,798. E. cryptospinosa Bally G. & K. 12,665; 13,021; Verdcourt 3225. E. cimeata Vahl var. nov., Gardner B6358 E. espinosa Pax Bally B10,451; Verdcourt 1847 (3). E. grandicornis Goebel. G. &K. 13,071. E. hirta L. G. & K. 12,666; Napier 936. E. kibwezensis N. E. Br. Trump sn. E. ndurumensis (Pax) Bally (Syn. E. taitensis Pax) F. T. Afr. 6, 1 : 571 (1911). Hildebrand! 2859; Kaessner 430; Mildbraed 12 ex num. Greenway EAH. 12,027; G. 9830 and 9831; G. & K. 12,956; Napier 988. E. nyikae Pax E. polyantha Pax Polhill & Paulo 472. G. 9765 ; G. & K. 12,739. E. quinquecostata Volkens E. robecchii Pax G. & K. 12,826; Verdcourt 3891. G. & K. 12,831 ; E. scheffleri Pax E. schinzii Pax, complex fide Kew. G. & K. 12,767; Hucks 156. G. 9830. E. spinescens Pax G. 10,817; Hucks 167; Polhill & Paulo 478 ; Verdcourt 2381. This species according to Carter & Bally is the same as E. cuneata Vahl, but the specimens cited a like it. C. dichogamus Pax Bally B8596; G. & K. 12,825. D. scandens L. var. cordafana Muell. Arg G. & K. 12,803; Hucks 73. E. agowensis Boiss. var. pseudoholstii (Pax) Carter & Bally G. & K. 12,721. E. cuneata Vahl Hucks 842. E. engleri Pax G. & K. 12,678 E. gossypina Pax var. gossypina Hucks 841. E. heterochroma Pax Hucks 220. E. jatrophoides Pax G. & K. 12,762; 12,805. e nothing E. systyloides Pax Napier 1058. E. tirucallii L. Euphorbia sp. aff. E. uhligiana Pax, complex fide Kew Sheldrick TNP/E/108. G. 9831. Euphorbia sp. Euphorbia sp. Hucks 849. Hucks 863. Euphorbia sp, Schenkal 45. Excoecaria venifera Pax Scheldrick TNP/E/93. Page 186 136 Euphorbiaceae (cont.) Givotia gosai A.R. Smith G. 12,521; G. & K. 12,650; Hucks 870; Moore & Napier - Bax TNP/GS/3 type Jatropha dichtar Mildbr. J. parvifolia Chiov. Rauh 12,519. G. & K. 12,756; Hucks 801. J. spicata Pax Jatropha sp. nr. J. stuhlmannii Pax G. & K. 12,657; 12,719; Napier 920. Bally B8664. Monadenium invenustum N. E. Br. G. & K. 13,044. Phyllanthus maderaspatensis L. 1 G. & K. 12,800; Sheldrick & Moore TNP/E/130. Phyllanthus sp. cfr. P. niruri L. Polhill & Paulo 921. Ricinus communis L. Williams & Sheldrick TNP/E/105 Securinega virescens (Willd.) Baill. G. & K. 12,704. Spirostachys africana Sond. G. & K. 12,923; 13,012; it is possible that these two specimens may be Excotcaria venifera Pax, but their flowers are not fully developed. Tragia arabica Baill. T. brevipes Pax Hucks 109; Verdcourt 1597. Bally B7707. T. hildebrandtii Muell. Arg., T. subsessilis Pax G. 9826 fide Kew: G. & K. 12,792. G. & K. 12,791. Tragia sp. Hucks 488 146 Caesalpiniaceae F. T. E. A. (1967). Bauhinia taitensis Taub. B. tomentosa L. G. & K. 12,691 ; 12,728. Tompson B7813. Caesalpinia trothae Harms ssp. erlangeri (Harms) Brenan C. trothae Harms ssp. trothae Hucks 291 ; 414. Bally B8681 ; Verdcourt 1588. Cassia abbreviata Oliv. ssp. kaessneri (Bak. f.) Brenan C. absus L. Ossent 268. C. fallacina Chiov. G. & K. 12,943; Hucks 26. C. occidentalis L. G. & K. 12,961; Napier 971. C. mimosoides L. Group E. Ossent 260. Verdcourt 1592. C. bicapsularis L. G. &K. 12,987; Hucks 398. C. longiracemosa Vatke G. 9744; 10,807; G. & K. 12,945. C. mimosoides L. Hucks 101. Delonix elata (L.) Gamble G. & K. 12,730; Hucks 348. Tamarindus indica L. G. & K. 12,732. Tylosema fassoglensis (Schweinf.) Torre & Hillcoat Hucks 217; Williams EAH. 12,536. 147 Mimosaceae F. T. E. A. (1959). Acacia brevispica Harms A. bussei Sjostedt G. & K. 12,855. Trapnell 2210. A. clavigera E. Mey. ssp. usambarensis (Taub.) Brenan A. elatior Brenan ssp. elatior Hucks 967. Sr. G. A. horrida (L.) Willd. ssp. benadirensis (Chiov.) Brenan A. mellifera (Vahl) Benth. ssp. mellifera Dale 3891 ; Trapnell 2212. Hucks 871 . A. nilotica (L.) Del. ssp. subalata (Vatke) Brenan A. polyacantha Willd. ssp. campylacantha G. & K. 12,958; Sheldrick TNP/E/88. (A. Rich.) Brenan Jeffrey 808. A. stuhlmannii Taub. A. reficiens Wawra ssp. misera (Vatke) Brenan Sr. G. G. & K. 12,939. A. thomasii Harms A. tortilis (Forsk.) Hayne ssp. spirocarpa (A. Rich.) Brenan G. & K. 12,861 ; Hucks 185 Hucks 427; 865. A. zanzibarica (S. Moore) Taub. Hucks 937. Page 187 147 Mimosaceae (coiit.) Albizia anthelmintica (A. Rich.) Brong. n. A. glabenima (Schumach. & Thonn.) Benth. Hacks 831 ; TNP/E/75. var. glabrescens (Oliv.) Brenan G. & K. 12,776; Jeffery 812. A. zimmermannii Harms G. & K. 12,702. Dichrostachys cinerea (L.) Wight & Arn. ssp. africana Brenan & Brumm. Bally 8161. Entada leptostachya Harms Braun 1540; Hucks 704. Neptunia oleracea Lour. Napier Bax TNP/G3/7 Newtonia hildebrandtii (Vatke) Torre var. hildebrandtii G. & K. 12,708; Hucks 224. 148 Papilionaceae Abnis schimperi Bak. ssp. africana (Vatke) Verde. Napier 976; Polhill & Paulo 957. Aeschynomene indica L. G. &K. 12,936; 13,066. Alysicarpus sp. Sr. G. Clitoria ternatea L. G. & K. 13,074; Hucks 326. Craibia brevicaudata (Vatke) Dunn var. brevicaudata G. & K. 13,027; Verdcourt & Polhill 2715. Crotalaria agatiflora Schweinf. ssp. agatiflora Hucks 340. C. laburnifolia L. Hucks 437; 428. C. patula Polhill Bogdan 5327; Drummond & Hemsley 4055. C. scassellatii Chiov. G. & K. 12,740; Polhill & Paulo 936. C. zimmermannii Bak. f. G. 10,436; Hucks 280. Dalbergia melanoxylon Guill. & Perr. G. & K. 13,028. Dolichos uniflorus Lam. var. stenocarpa Brenan Hucks 1025; Napier 1011. Erythrina melanacantha Taub. Sr. G.; Hucks 941. Indigofera arrecta A. Rich. /. G. &K. 13,025; Hucks 705. /. hirsiita L. var. hirsuta Hucks 571 ; Napier 990. I. malindiensis Gillett n. sp. ined. Schenkel 35. I. schimperi Juub & Spach. var. schimperi G. & K. 12639; 12899; Hucks 92. I. schimperi Jaub. & Spach var. baiikeana (Vatke) Gillett Jeffrey 816; Napier 992. I. sesilis Gillett /. spinosa Forsk. Sampson 97; Verdcourt 3875. G. 9780; G. & K. 12,946. /. tanganyikensis Bak. f. forma paiicijtiga Gillett /. trita L. f. var. subulata (Poir.) Ali Drummond & Hemsley 4056 holotype; Gillett 17, 187. Sr. G. I. vohemarensis Baill. I. volkensii Taub. Schenkl 66; Sheldrick & Moore TNP/E/162. Bally B8159. Indigofera sp. I. microcharoides Taub. Napier Bax TNP/R/90. Gillett 16,867. Lonchocarpus eriocalyx Harms Sr. G. Hucks 257. Neorautanenia mitis (A. Rich.) Verde, ined. (Syn. N. pseudopachyrrhizus Harms). Beaton 50. Ormocarpum kirkii S. Moore Thompson B7815. C. glauca Willd. Gregory sn., not seen. C. laburnifolia L. ssp. tenuicarpa Polhill Gillett 17,219; G. 9501 ; G. & K. 12,815. C. polysperma Kotschy Hucks 307; 545. C. tsavoana Polhill G. &K. 12,938; 13,047. C. ukambensis Vatke Hucks 789; Polhill & Paulo 932. costata Guill. & Perr. ssp. goniodes (Bak.) Gillett G. & K. 12,714; Hucks 652A. I. hochstetteri Bak. Hucks 866. ir. latestipulata Gillett ined. Page 188 R. pulchra (Vatke) Harms Greenway EAH. 12,585; Hucks 677. R. sublobata (Schumach.) Meikle Hucks 124; 592; Verdcourt 1108. Papilionaceae (cant.) Ophrestia hedysaroides (Willd.) Verde. (Syn. Paraglycine hedysaroides (Willd.) F. T. Herm.) Ossent sn. Platycelyphium voense (Engl.) Wild G. & K. 12,712; Hucks 191. Rhynchosia minima (L.) DC. Drummond & Hemsley 4136; Napier 996A R. sennarensis Hochst. Hucks 694. R. velutina Wight & Arn. Tweedie 3200. Sesbania quadrata Gillett Hucks 324; 1077. S. sesban (L.) Merr. var nubica Chiov. Williams & Sheldrick TNP/E/102. Spathionema kilimandscharicum Taub. G. 9820; Napier 950; Polhill & Paulo 471 ; Verdcourt 2346. Stylosanthes fruticosa (Retz.) Alston G. & K. 12,813. Tephrosia lortii Bak. f. T. noctiflora Bak. Hucks 994; Napier 928. G. 9742; G. & K. 12,955. T. pentaphylla (Roxb.) G. Don T. pumila (Lam.) Pers. var. pumila G. & K. 12,877. Hucks 148; 76b. T. purpurea (L.) Pers. var. pubescens Bak. T. subtriflora Bak. Gregory sn., not seen. G. 10,461 ; G. & K. 12,812; Napier 929. T. uniflora Pers. T. villosa (L.) Pers. ssp. ehrenbergiana (Schweinf.) Brumm. Hucks 663; Verdcourt 1110. G. 9745; G. & K. 13,014; Sheldrick TNP/E/1. Vatovaea pseudolablab (Harms) Gillett Hucks 85; 586; 1,000. Vigna membranacea A. Rich. ssp. caesia (Chiov.) Verde. V. kirkii (Bak.) Gillett Greenway 10,429; G. & K. 12,643; 12,942; P. Bax TNP/R/69. Napier 994. V. praecox Verde. V. unguiculata (L.) Walp. ssp. cylindrica (L.) van Eseltine Bally 8560; Hucks 449; Polhill & Paulo 469. Sheldrick & Moore TNP/E/1 34. Zornia glochidiata DC. G. &K. 12,771. 156 Salicaceae Populus ilicifolius (Engl.) Rouleau G. & K. 12,760; 12,902. 167 Moraceae Dors tenia ? crispa Engl. Verdcourt 3880. Ficus ingens (Miq.) Miq. F. mucuso Ficalho G. & K. 12,703 Hucks 970. F. populifolia Vahl F. sonderi Miq. G. & K. 12,763; Polhill & Paulo 938. G. & K. 13,031 ; Hucks 37. F. sycomorus L. G. & K. 13,082; Wilhams & Sheldrick TNP/E/103. 173 Celastraceae Cassine aquifolium Fiori G. & K. 12,940; 13,049. Hippocratea africana (Willd.) Loes. G. & K. 13,010. Maytenus senegalensis (Lam.) Exell Hucks 820. 179 Icacinaceae F. T. E. A. (1968). Pyrenacantha malvifolia Engl. Polhill & Paulo 475; Verdcourt 3190; Lucas, Jeffrey & Kirika 269. 180 Salvadoraceae F. T. E. A. (1968). Azima tetracantha Lam. Bally B8742; G. & K. 12,699. Dobera glabra (Forsk.) Poir. D. loranthifolia (Warb.) Harms Napier Bax TNP/GS/6; Agnew et al. 7330. G. & K. 12.698; 12,780. Salvador a per ska L. G. & K. 12,709; Napier Bax TNP/E/48. 182 Olacaceae F. T. E. A. (1968). Ximenia americana L. Sheldrick TNP/E/80 185 Loranthaceae Amyena panganensis (Engl.) Balle Drummond & Hemsley 4103; G. 9832. Erianthemum occulatum (Sprague) Danser G. 9542 fide Kew. Helixanthera kirkii (Oliv.) Danser Rauh Ke648. Loranthus triplinervius Baker & Sprague Hucks 1036. L. schimperi Jaub. & Spach var. schimperi G. & K. 12,639; 12,899; Hucks 92. Oliverella hildebmndtii (Engl.) van Tiegh. Hucks 747; Napier Bax TNP/GS/20. Plicosepalus curviflorus (Benth.) Danser Hucks 402. Tapinanthus sansibarensis (Engl.) Danser Hucks 679. P. sagitlifolius (Sprague) Danser Hucks 959. Tapinanthus sp. Williams & Sheldrick TNP/E/97. Berchemia discolor (Klotzsch) Hemsl. Dale 3676; G. & K. 12,664. Helinus integrifolius (Lam.) Kuntze G. & K. 12,816; Hucks 764. Ziziphus mucronata Willd. Bally B8641; G. & K. 12,694. Ampelocissus africana (Lour. ) Merr. G. &K. 13,019. Cissus aphyllantha Gilg Napper 1339; Verdcourt 1587. C. quadrangularis L. Hucks 721. Cyphostemma adenocaulis A. Rich. Hucks 296; 731. 190 Rhamnaceae 193 Vitaceae (Ampelidaceae) C. cactiformis Gilg G. & K. 12,996; Hucks 1046; Napier 1022. C. rotundifolia (Forsk.) Vahl G. &K. 12,857; Hucks 161. Cyphostemma sp. Agnew et al. 7338. 194 Rutaceae Fagara chalybea (Engl.) Engl. G. & K. 12,908. Vepris eugeniifolia (Engl. ) Verdoorn V. uguenensis Engl. Agnew et al. 5828; Drummond & Hemsley 4063. Gardner 2961 ; Ossent 121. Vepris sp. nov.=Dale & Greenway, Trees and Shrubs of Kenya, p. 493. G. &K. EAH. 13,718. 195 A Balanitaceae Balanites orbicularis Sprague G. & K. 12,889; 12,971 196 Burseraceae Boswellia hildebrandtii Engl. G. 9617; 10,815; G. & K. 12,690; Sheldrick TNP/E/19. Commiphora africana (A. Rich.) Engl. C. boiviniana Engl. Agnew et al. 7323. G. & K. 12,828; Verdcourt 3895; 1113. C. campestris Engl. C. holtziana Engl. Bally B8630; Dale 3895B; PolhiU & Paulo 473. G.&K. 12,633; 12,747; Sheldrick TNP/E/77. C. mildbraedii Engl. C. riparia Engl. Mildbraed 5 type; Verdcourt 3871. G. 10,813 fide Kew; G. & K. 12,682. C. scheffleri Engl. C. trothae Engl. G. 9789; 9827 fide Kew. Ivens 410 Commiphora sp. nr. C. madagascariensis Jacq. and C. merkeri Engl, fide Kew. G. & K. 12,658. Commiphora sp. Commiphora spp. G. in E.A.H. 12,221. Gillett 16,857; 17,189; 17,192. Commiphora sp. Commiphora sp. G. & K. 12,655; 12,896. G. & K. 12,893; 12,894. Commiphora sp. Commiphora sp. = Bally 1667 Verdcourt & PolhiU 2709. G. & K. 12,661 ; Verdcourt 3889. 197 Meliaceae Melia volkensii Guerke Gillett & Burtt 17,048; Hucks 99. Trichilia roka (Forsk.) Chiov. Hucks 878; 290; Sheldrick 1004. 198 Sapindaceae Aphonia senegalensis (Poir.) Radik. G. & K. 12,995; G. & K. EAH. 13,719. Cardiospermum corindum L. C. halicacabum L. Napier 1068. G. & K. 12,848; Hucks 725. DeinboUia borbonica Scheff. Drummond & Hemsley 4277. Haplocoelum foliolosum (Hiern) Bullock Bally B8595; Schenkel 77. Lecaniodiscus fmxinifolius Bak. G. & K. 13,055; G. & K. EAH, 13.720; Sheldrick TNP/E/86. 205 Anacardiaceae Laimea alata (Engl.) Engl. L. stuhlmannii (Engl.) Engl. G. 9617; G. & K. 12,687; Sheldrick TNP/E/7. G. & K. 12, 655. L. triphylla (A. Rich.) Engl. G. 9766; 9787 fide Kew; Sheldrick TNP/E/115. Lannea sp. nov.?=G. 9256 and Makin EAH. 13,721. Lannea sp.=Bally 117. G. &K. 12,746; 13,067. G. 9543. 213 Umbelliferae Berula erecta (Huds.) Coville Bogdan 361 2A. (Tsavo River). Steganotaenia amliacea Hochst. Sr. G. II METACHLAMYDEAE (Families Nos. 221-264) 221 Ebenaceae Diospyros consolatae Chiov. D. cornii Chiov. G. & K. 12,907; 13,057. Sheldrick TNP/E/87; Verdcourt 3886. 222 Sapotaceae F.T.E.A. (1968). Manilkara mochisia (Baker) Dubard G. 9785 fide Kew; Sheldrick TNP/E/112. M. fruticosa DC. Mimusops sp. cfr. M. schliebenii Mildbr. G. & K. 13,017; Hucks 411 G. & K. 13,085 fide Kew. 228 Loganiaceae F.T.E.A. (1960). Strychnos decussata (Pappe) Gilg S. madagascariensis Poir. G. & K. 12,629; 12,663; Hucks 513. G. & K. 12,660; 12,685. 229 Oleaceae F.T.E.A. (1952). Jasminum grahainii Turrill J. parvifolium Knobl. Graham 1658 type. G. & K. 12,705; Hucks 516. 230 Apocynaceae Adenium obesiim (Forsk.) Roem. & Schult. G. 10,810; G. & K. 12,769; Hucks 123. Carissa edulis Vahl Ossent 129; MacArthur B274. *Catharanthus roseus (L.) G. Don Hucks 95. Holarrhena febrifuga Klotzsch G. &K. 12, 653; Hucks 539. Strophanthus mirabilis Gilg G. 9825; G. &. K. 12,871; Hucks 827. Page 191 231 Asclepiadaceae Baseonema gregorii Schlecht. & Rendle Bally B 8745; Hucks 678; Verdcourt & Polhill 2695. Calotwpis procera (Ait.) Ait. G. & K. 12,729; Hucks 192. Caralluma priogoniiim K. Schum. EAH. 95. C. speciosa (N. E. Br.) N. E. Br. Sr. G.; ShirtUffe EAH. 13,596. Cynanchum defoliascens K. Schum. Hucks 983. C. omissum Bullock Drummond & Hemsley 4045 type. C. validum N. E. Br. Sr. G. Diplostigma canescens K. Schum. Greenway EAH. 11,700; G. & K. 12,71 1 ; Hucks 282. Dregea stelostigma (K. Schum.) Bullock Hucks 356; 382; Verdcourt 2395. Echidnopsis dammanniana Sprenger? G. & K. 12,954. Edithcolea grandis N.E. Br. Sr. G. ; Hucks s. n. Glossonema revoilii Franch. Hucks 828. Kamhia laniflora (Forsk.) R. Br. G. &K. 12,921; Hucks 412. Oxystelma bornuense R. Br. G. & K. 13,051; Verdcourt 2396. Pergularia daemia (Forsk.) Chiov. G. 10,462; Hucks 304; 900; Napier Bax TNP/E/43. Sacleuxia newii (Benth.) Bullock Gillett 17,327; G. & K. 12,852. Sarcostemma viminale R. Br. Sarcostemma sp. G. & K. 12,679; Hucks 1037. Ossent 125; Verdcourt 3227. Schlechterella africana (Schlecht.) K. Schum. G. 9546; 10,816; G. & K. 13,091. Secamone pmctulata Decne var. stenophylla (K. Schum.) N. E. Br. G. & K. 12,645; Hucks 497. Stathmostelma pedunculatum (Decne) K. Schum. Leaky B3166. 232 Rubiaceae Boneria scabra (Schumach. & Thonn.) K. Schum. s. 1 . G. 9772; 10,431; Hucks 688; Polhill Paulo 930. Dirichletia glaucescens Hiem G. 10,465; G. & K. 12,677; 12,737. Gardenia jovis-tonantis (Welw.) Hiern G. & K. 13,026. Hvmenodictyon parvifolium Oliv. G. 9541; G. & K. 12,680; Hucks 570. Kohautia caespitosa Schinz var. amaniensis (K. Schum.) Brem. G. & K. 13,013; Hucks 1075; Napier 1063. Meyna tetraphylla (Hiem) Robyns G. &K. 12,777; 13,076. Oldenlandia herbacea Roxb. var. holstii (K. Schum.) Brem. O. somala Chiov. Napier 1335. G. & K. 12,935; Hucks 807; O. wiedermannii K. Schum. G. & K. 12,750; Drummond & Hemsley 4099. Paederia pospischilii K. Schum. G. & K. 12,801; Hucks 283; Napier 1050. Pentanisia ouranogyne S. Moore G. & K. 12,786; Hucks 16; Napier 1041. Pent as bussei Krause intermed. P. parvi folia Hiern P. parvifolia Hiern Ossent 6. Bally B4722; Hucks 367. C. msselliana (Brongn.) Cuf. Hucks 1045 C. turneri E. A. Bruce G. & K. 12,652. C. hastifolium N. E. Br. Hucks 111. C. tetraptemm (Turcz) R. A. Dyer Napier 1064. Page 192 232 Rubiaceae (cont.) Pentodon pentander (Schumach.) Vatke var. minor Brem. Napier 997. Psychotria kirkii Hiem P. nairobiensis Brem. ? Ossent 247. G. & K. 12,774; 12,824. Psychotria sp. Schenkel 69 ; this has been named P. nairobiensis Brem. but is quite sterile. Rytigynia sp. nr. R. loranthifolia (K. Schum.) Robyns Hucks 932. Tarenna graveolens (S. Moore) Brem. G. 9625: Hucks 559. Tricalysia ovalifolia Hiern Graham 1615. Xeromphis keniensis Tennant sp. nov. G. 9613; 9740 type; 10,857; Hucks 657. 238 Compositae Acanthospermum hispidum DC. G. & K. 12,962; Hucks 378. Aspilia mossambicensis (Oliv.) Wild G. & K. 13,084; Hucks 611 ; Napier Bax TNP/R/60. Athroisma psyllioides (L.) L. G. & K. 12,965. Bidens incumbens Scherff Hucks 741 ; Napier 1016 type; Verdcourt 3890A. Blepharispermum fruticoswn Klatt & Schinz ssp. lanceolatiini Chiov. B. zanzibaricum Oliv. & Hiern Hucks 420; 639. G. & K. 12,992. Bluniea aurita (L.) DC. Blimiea sp.=Hucks 176; 898, but not B. cajfra (DC.) O. Hoffm. Hucks 869. G. & K. 13,005; 13,042. Conyza aegyptiaca Ait. var. ? G. & K. 12,673 cfr. Drummond & Hemsley 4066. Dicoma tomentosa Cass. Sr. G. Eclipta prostrata (L.) L. G. & K. 12,982; Hucks 32. Erlangea boranensis S. Moore Erlangea marginata (O. & H.) S. Moore Hucks 275. Ossent 31. Erythrocephalum longifolium Benth. Lady Muriel Jex Blake B5078 Galinsoga parviflora Cav. Sheldrick TNP/E/109 Geigeria acaulis Oliv. & Hiern G. & K. 12,948. Grangea maderaspatana Poir. G. & K. 13,063. Gutenbergia polycephala Oliv. & Hiem G. & K. 12,839. Helichrysum glumaceum DC. G. & K. 12,963; Hucks 165; 198. Hirpicium diffusum (Oliv.) Roessl. Shanz & Turner 4243. Kleinia kleinioides (Sch. Bip.) M. R. F. Taylor vel sp. aff. (Taxon 26 of Jeffrey MS.) Sr. G. Rauh 1 ; Napier 1057. Launaea cornuta (Oliv. & Hiern) C. Jeffrey L. intybacea (Jacq.) Beauv. G. & K. 12,862; Napier 925. G. & K. 13,037; Hucks 966. Microglossa oblongifolia F. Hoffm. Napier 935. Notonia sp. Hucks 966. Osteospermum vaillantii (Decne) T. Norl. (Syn. Tripteris vaillanti Decne). Gregory sn., not seen. Pegolettia senegalensis Cass. Hucks 479. Pluchea dioscoridis DC. P. ovalis (Pers.) DC. G. & K. 12.988; Hucks 442; Williams & Sheldrick TNP/E/99 G. & K. 12,984; Hucks 62 Page 193 238 Compositae {cont.) Pluchea sordida (Vatke) Oliv. & Hiem Agnew et al. 7357 Sclewcarpus africanus Murr. Bally B4714; Hucks 377 Senecio discifolius Oliv. Hucks 558 Sphaeranthus napierae Ross-Craig G. & K. 13,002 Spilanthes mauritiana (A. Rich.) DC. Ag. Lab. Nairobi B10226 Tridax procumbens L. G. & K. 13,023; Hucks 90. Vernonia aemulans Vatke Hucks 680; 762. V. cinerea (L.) Less. MacArther B265. V. hildebrandtii Vatke G. & K. 12,692; Hucks 330; 873. V. wakefieldii Oliv. Gillett 16,873; G. & K. 12,706. 239 S. stuhlmannii Klatt MacArthur B270 S. ukambensis O. Hoffm. Hucks 868 V. cinerascens Sch. Bip. Hucks 323; 770; Napier Bax TNP/E/68. V. colorata Drake Hucks 274. V. pauci flora Less. Gregory sn., not seen. Vernonia sp. aff. V. cinerascens Sch. Bip. Hucks 1034. Enicostema hyssopifolia (Willd.) Verdoorn Napier 959. 241 Plumbaginaceae Plumbago zeylanica L. Hucks 309. 244 Lobeliaceae Lobelia anceps L. f. var. anceps Hucks 211. Agnew et £ 249 Boraginaceae Coldenia procumbens L. G. & K. 13,061; Hucks 144; 1076. Cordia gharaf (Forsk.) Aschers. C. ovalis R. Br. Napier Bax TNP/E/R/70. Bally B8629; Hucks 890; Verdcourt 3895B. Cordia sp. nov. aff. C. goetzei Guerke=Bally 12,053; Rawlins 351 ; Verdcourt 1862A. fide Kew. G. &K. 13,053; Hucks 418. Cordia sp. Apew et al. 7349; Polhill & Paulo 474. Ehretia teitensis Guerke G. & K. 12,838; 12,890; Hucks 277. Heliotropium albohispidum Bak. G. & K. 12,814; Hucks 652B H. ? marifolium Retz. Hucks 1052. H. steudneri Vatke G. & K. 12,830; Hucks 532C; 1065. McDonald 880. Heliotropium subulatum (DC.) Martelli G. 10,432; G. & K. 12,842; Hucks 201. Heliotropium sp. nr. H. steudneri Vatke G. & K. 12,713 fide Kew; Hucks 53. Trichodesma zeylanica (L.) R. Br. G. & K. 12,924; Hucks 149; 851 ; Napier 962. Ehretia sp. aff, E. obtiisifolia DC. Hucks 1028. H. indicum L. Hucks 86; 532B; 650. H. ovalifolium Forsk. Hucks 899. H. strigosum Willd. G. & K. 12,637; Hucks 693; Polhill & Paulo 919 H. supinum L. Hucks 128; 1035. Heliotropium sp. Hucks 1064. Capsicum frutescens L. Napier 965. Datura metel L. Hucks 63. Lycium europaeum L. Hucks 405. 250 Solanaceae Page 194 250 Solanaceae {cont.) Solanum dubium Fres. Hucks 482; 1063. S. incanum L. Hucks 626; 720. S. renschii Vatke Hucks 284; 918; Napier Bax TNP/E/41. S. taitense Vatke G. 9822; G. & K. 12,768; Hucks 404; Polhill Withania somnifera (L.) Dunal G. & K. 12,891; Hucks 740; 853. S. hastifolium Dunal Hucks 756; 825; 1055. Solanum nigrum L. Hucks 613; 627. S. somalense Franch. var. planifrons Bitter e desc. Hucks 552; 612A; Ivens 412. & Paulo 480. 251 Convolvulaceae F. T. E. A. (1963). Astripomoea hyoscyamoides (Vatke) Verde. Hucks 576; Napier 993; Napier Bax TNP/E/27. Convolvulus rhyniospermus Chiosy Hucks 808. Evolvulus alsinoides (L.) L. G. & K. 12,964; Hucks 799. Hildebrandtia sepalosa Rendle Ossent 10; 50; 54; G. & K. 12,749; 12,887. Ipomoea aracltnosperma Welw. Bally B8974; Hucks 226. I. cairica (L.) Sweet Hucks 11. /. eriocarpa R. Br. Napier 1049. I. hartmannii Vatke Hucks 269; Verdcourt & Polhill 2763 I. irwinae Verde. Hucks 12. I. longituba Hall. f. Hucks 867. I. ochracea (Lindl.) G. Don var. ochracea G. &K. 12,997; Hucks 10; 88, I. mombassana Vatke Hucks 13; Napier Bax TNP/E/14. I. sinensis (Desr.) Choisy ssp. blepharosepala (A. Rich.) Meeuse Hucks 216; 238. 7. transvaalensis Meeuse ssp. orientalis Verde. Ipomoea sp. nov. = Leipert 6072, from th Verdcourt 1199 type. W. Pare Mtns. Hucks 9944. 7. bullata Oliv. Bally B8613; Hucks 760; Ossent 126. 7. cicatricosa Bak. Hucks 640. 7. garckeana Vatke G. 9788; Hucks 1071 ; Verdcourt 1196. 7. hildebrandtii Vatke Irwin 156. 7. kituiensis Vatke Hucks 742; Napier 1072. 7. obscura (L.) Ker.-Gawl. Hucks 130; 730; 846. 7. oenotherae (Vatke) Hall. f. Ossent sn. 7. pes-tigridis L. var. pes-tigridis Hucks 150; 1031. Jacquemontia tamnifolia (L.) Griseb. Hucks 195; 664; 810. Merremia ampelophylla Hall. f. M. pinnata (Choisy) Hall. f. G. 9777; Hucks 567; 115; Verdcourt 1198. Hucks 214; 618. M. tridentata (L.) Hall. f. ssp. angustifolia (Jacq.) v. Ooststr. Hucks 672. Seddera hirsuta Hall. f. var. gracilis (Chiov.) Verde. G. & K. 12,933; Schenkel 56. Stictocardia incomta (Hall, f.) Hall. f. Sr. G.; Hucks 1074. Turbina stenosiphon (Hall, f.) Meeuse Sr. G. 252 Scrophulariaceae Alectra vogelii Benth. Agri. Officer 796. Craterostigma sp. nov. Hucks 511; 561; Martin 87. Buttonia hildebrandtii Engl. Dale 3861 ; G. 9829; Napier 875; Verdcourt 3895a. Harveya obtusifolia (Benth.) Vatke Bally B6372. Ilysanthes pusilla (Oliv.) Urban Hucks 1044. Page 195 252 Scrophulariaceae (cont.) Pseudosopubia sp. G. & K. 12,875; Hucks 400; 802; Sheldrick B8733. Rhamphicarpa veronicifolia Vatke Hucks 372; 381; 478. Stemodiopsis buchanatni Skan S. Bally B8800; Verdcourt 3896. Striga gesnerioides (Willd.) Engl. S. Polhill & Paulo 950. humilis Skan Gillett & Burn 17,191. latericea Vatke McArthur 268. 257 Bignoniaceae 258 Pedaliaceae F. T. E. A. (1953) Josephinia africana Vatke G. 9768; G. & K. 12,827; Hucks 111. Pedalium murex L. Hucks 184; 635; Polhill & Paulo 922. Pterodiscus ruspolii Engl. Hucks 715. Sesamothamnus busseanus Engl. S. rivae Engl. Sr. G. Dale 3877; 10,812; Hucks 319. 259 Acanthaceae Adhatoda schimperana Nees Lady Muriel Jex Blake CM. 18128. Anisotes parvifolius Oliv. (or related species) Hucks 441 ; 841 ; Napier Bax TNP/E/40; Schenkel 39. Asystasia charmian S. Moore G. & K. 12,967; Hucks 590; 673. A. icA/wpm T. Anders. Hucks 206; 649. Asystasia sp. Hucks 391. Barleria diffusa (Oliv.) Lindau Bally B8686; G. & K. 13,016; 13048; Hucks 225; 243; 322; B. ramulosa C. B. Cl. G. & K. 13046; Hucks 131; 745; Verd- court 3192. B. taitensis S. Moore G. & K. 13070; Hucks 30; 894; Polhill & Paulo 945; Sheldrick TNP/E/142. Barleria sp. sect. Prionitis. Greenway EAH, 12,219; Hucks 36; 169. Barleria sp. sect. Somalia BaUy B8614; Verdcourt 3894. Barleria sp. Hucks 1070. Blepharis linariifolia Pers. Hucks 1051. B. fruticulosa C. B. Cl. Hucks 815. Crabbea velutina S. Moore Hucks 977. Crossandra mucronata Lindau G. 9743; G. & K. 12,659. Dicliptera mossambicensis Klotzsch Hucks 789. Disperma kilimandscharica (Lindau) C. B. Cl. Agnew et al, 5843 ; Hucks, 306 ; 436 ; 805 ; 852 ; Napier 897. Disperma sp. Schenkel 57. Dyschoriste perrotteii (Nees) O. Ktze Bally B866; Natttrass 208A. A. gauge tica (L.) T. Anders. Hucks 823. A. somalensis (Franch.) Gillett G. & K. 12,847; Ossent 35. B. submollis Lindau Hucks 55; 771; 794; Napier 894; Ossent 78; Agnew et al. 5832. Barleria sp. sect. Prionitis. G. & K. 12922; Napier 896; Ossent 164. Barleria sp. sect. Somalia G. & K. 12,980. Barleria sp. sect. Somalia Hucks 206; Polhill & Paulo 929. B. maderaspatensis (L.) Roth Hucks 739. Blepharis sp. Agnew 580. C Page 196 259 Acanthaceae Icont.) Ecbolium amplexicaule S. Moore G. & K. 12,669; Hucks 96; Joanna 2034. E. revolutum (Lindau) C. B. Cl. G. &. K. 12,696; Hucks 239; 791; Napier 1096. Hypoestes hildebrandtii Lindau G. 9823; Hucks 812. Justicia caemlea Forsk. G. & K. 12,968; Hucks 591; 809. J. flava Vahl Hucks 768; Napier Bax TNP/E/50. J. heterocarpa T. Anders. Hucks 83; 119 \ Sheldrick & Moore TNP/E/136. J. sansibarensis Lindau Ossent 138. J. uncinulata Oliv. Hucks 1008. Justicia sp. aff. J. striata (Klotzch) Bullock Bally B8618; Hucks 49; 69; McDonald 862. Justicia sp. Hucks 236. Lepidagathis scariosa Nees G. 9828; Hucks 45; 136; Napier 985; Polhill & Monechma debile (Forsk.) Nees Hucks 240; 654; 767; Napier 905. Neuracanthus ukambensis C. B. Cl. Hucks 337. Neuracanthus sp. G. &K. 12,912. Peristrophe bicalyculata (Retz.) Nees Hucks 33; Napier 916. Rhinacanthus nasutus (L.) Kurz Hucks 339. Ruellia amabilis S. Moore Bally B8735; Hucks 602; Ossent 109. Ruttya fruticosa Lindau Gillett & Burtt 17,185; Hucks 370; Napier 977 Thunbergia holstii Lindau G. & K. 12,773. Thunbergia sp. nov. = Drummond & Hemsley 4104 G. & K. 12,876. E. hamatum C. B. Cl. Hucks 97. E. subcordatum C. B. Cl. Hucks 647. H. verticillaris R. Br. G. & K. 12,785; Hucks 260;675;797;817 J. fischeri Lindau Hucks 651b; 723; Ossent 2. J. glabra Roxb. forma Bally B8736; Hucks 751 ; Napier 1052. J. matammensis Oliv. Hucks 369; 608; 866. J. striata (Klotzch) Bullock G. & K. 12,782; Hucks 784. J. whytei S. Moore Hucks 52; 426; Moore&SheldrickTNP/E/143; Justicia sp. aff. /. odora Vahl Napier 1042. Paulo 944. Monechma sp. ? Hucks 796; 1019. Neuracanthus sp. G. & K. 13,032 = Agnew EAH. 13,723; Hucks 275; 633; 856. R. patula Jacq. G. & K. 12,846 = Napier 891; Hucks 138; 380. T. guerkeana Lindau G. & K. 12,818. 263 Verbenaceae Chascanum hildebrandtii (Vatke) Gillett Gillett 17,208; G. & K. 12,915; Hucks 773A. Clerodendrum eriophyllum Guerke C. makanjanum H. Winkler Bally B7712. G. 9545; Ossent 106; Polhill & Paulo 470. Cyclocheilon eriantherum (Vatke) Engl. G. & K. 12,707; Dale 3647; Ossent 127; Verdcourt 3224. Lantana camara L. G. & K. 13,009; Hucks 346. Lantana sp. G. & K. 12,745; Drummond & Hemsley 4091 named by Kew as L. viburnoides (Forsk.) Vahl but these are certainly not this species, another is Hucks 701 . Phyla nodiflora (L.) Greene G. & K. 12,904; Hucks 33; 110. Premna hildebrandtii Guerke P. oligotricha Bak. Bally B8639; G. & K. 12,990. Agnew et al 5818;Bally B8626;G.&K. 12,807; P. resinosa (Hochst.) Shauer Napier 1066. G. 12,520; G. & K. 12,688; 12,840; Hucks 311; Napier Bax TNP/E/64. Priva cordifolia (L.) Druce var. abyssinica (Jaub. & Spach) Moldenke Hucks 773B; 793; Napier 931 ; Verdcourt 3876. Svensonia laeta (Walp.) Moldenke G. & K. 12,886; Verdcourt 1109. Vitex payos (Lour.) Merr. V. strickeri Vatke & Hildebrandt G. & K. 12,630. Sheldrick TNP/E/R/74. Page 197 264 Labiatae Basilicum polystachion (L.) Moench. G. & K. 12,981 ; Hucks 222; 860. Beciiim sp. Hucks 342; 475; 681 ; 111-, Sheldrick & Moore TNP/E/131. Capitanya otostegioides Guerke Gillett 17,276; G. & K. 12,957; Hucks 637; 780. Coleus, some authorities consider that this genus should be sunk in Plectranthus. C. amboinicus Lour. C. ignarius Schweinf. Bally B8633; B8919. Napier 1024. C. temifloms Vatke Coleus sp. aff. C. teitensis Bak. Napier 919. Napier 1037. Coleus sp. Coleus sp. Bally B8653. Nattrass B1 1 57. Endostemon tenuiflorus (Benth.) Ashby E. tereticauUs (Poir.) Ashby Hucks 988. G. 10,434; G. &. K. 12,681 ; Hucks 485; 728; Napier 1029. Erythrochlamys spectabUis Guerke G. & K. 12,811; 12,829; Hucks 71; 315; Napier 1056. Hemizygia fischeri (Guerke) Greenway (Syn. Ocimum fischeri Guerke.) G. & K. 12,761 ; Hucks 266; 607; Verdcourt 3882. Hoslundia opposita Vahl G. & K. 12,934; 1067; Hucks 811; Rauh K67. Iboza multiflora (Benth.) E. A. Bruce Nattrass 204. Leoiiotis nepetifoUa L.) R. Br. Hucks 379. Leucas neuflizeaiia Courb. L. nubiea Benth. Hucks 749; 806. Hucks 1016. L. oligoeephala Hook.f. L. pratensis Vatke Gregory s.n., not seen. Hucks 484; 502. Leucas sp. G. 10,808; G. & K. 12,998; Hucks 577; 847; Napier 1019; the last was named L. bracteata Guerke by Kew but is not that species. Leucas sp. O. hadiense Forsk. G. & K. 12,864; Hucks 8; 666. Ocimum sp. = Makin 182. Hucks 1058. Orthosiphon sp. Hucks 689. P. cylindraceus Benth. Napier 1071. P. prostratus Guerke Greemsmith CM 16,236. Hucks 417. Ocimum basilicum L. (Syn. O. americanum L.) G. & K. 13,024; 13,068; Napier 1002. Ocimum sp. Hucks 154; 719. Orthosiphon sp. G. & K. 12,970. Plectranthus barbatus Benth. Agnew et al. 5807; Nattrass B1158. P. longipes Bak. Hucks 270; 697; 732. Pycnostachys umbrosa (Vatke) Perkins Hucks 374. Tinnea aethiopica Kotschy & Peyr. G. & K. 13,029; Hucks 515; Rauh 12,506. n MONOCOTYLEDONES I Calyciferae (Families Nos. 266—290). 266 Hydrocharitaceae Lagarosiphon tenuis Rendle Gregory sn. type not seen; Napier Bax TNP/Gs/8. 279 Najadaceae Najas graminea Delile e desc. G. &K. 13,060. 280 Commelinaceae Aneiiema Iwckii De Wild. A. johnstonii K. Schum. Hucks 617B; 221 ; Napier 915. Hucks 519. A. petersii (Hassk.) C. B. Cl. A. rendlei C. B. Cl. Napier 973. Hucks 15; 690; Napier 914; Verdcourt 3901. Page 198 280 Commelinaceae (cont.) Aneilema sp. = Bally B8534. Bally B8534; Hucks 669. Anthericopsis sepalosa (C. B. Cl.) Rendle G. & K. 12,733; Hucks 465; 493. Ballya zebrina (Chiov.) Brenan Verdcourt 3892. Commelina albescens Hassk. C. benghalensis L. Hucks 1069. Agnew et al. 5816; Hucks 646; Lucas et al. 270; Napier 908. C. erecta L. ssp. livingstonii (C. B. Cl.) J. K. Morton Hucks 551. C. forskalaeiWaM G. & K. 12,790; Hucks 715A; 716B; 769; Verdcourt 3899. C. imberbis Hassk. C. latifolia Hochst. Hucks 634 ; 703 . Napier 913. C. Ipetersii Hassk. Hucks 617A; 655; Lucas et al. 271 ; Napier Bax TNP/E/44; Napier 907. C. subulata Roth, FI. Trop. Afr. 8: 38 (1901). Gregory sn., not seen. 290 Zingiberiaceae Kaempferia aethiopica (Solms-Laub.) Benth. G. 9503; Hucks 508. n Corolllferae (Families Nos. 293—326). 293 Liliaceae. Albuca wakefieldii Bak. G. & K. 12,953; Hucks 585; Ossent 22. Aloe deserti Berger Bally B8652; Ossent 99. A. ruspoliana Bak. G. & K. 13,086; Bally B8791. Aloe sp. aff. A. ruspoliana Bak. G. & K. EAH. 13,728. Anthericum brehmerianum Poelln. Hucks 536. A. suffruticosum (Bak.) Milne-Redhead G. &K. 12,918; Hucks 477. Anthrecum sp. G. & K. 12,718. Asparagus asiaticus L. Schenkel 32. A. Inudicaulis Bak. Hucks 450. Chlorophytum gallabatense Bak. Hucks 906. C. tuberosum Bak. Hucks 520. A. lateritia Engl. Sr. G. A. secundiflora Engl. Sr. G. A. falcatus L. Schenkel 34. C. tenui folium Bak. G. & K. 12,736; Hucks 477. C. viridescens Engl. G. & K. 12,966; Hucks 597; Sheldrick B8795. Dasystachys debilis Bak. (Syn. D. gracilis Bak.; Chlorophytum bakeri von Poelln.) G. & K. 12,710; Hucks 546; Ossent 139. Dipcadi viride Moench. Hucks 925. Drimiopsis sp. = G. & K. 11,105 G. & K. 12,854; Hucks 494. Gloriosa abyssinica A. Rich. var. graminifolia Franch. G. simplex L. Jeffrey K814. Bally B8166; Dale K204; Hucks 553 ; Ornithogalum donaldsonii (Rendle) Greenway (Syn. Albuca donaldsonii Rendle) G. & K. 12,895; Hucks 594; Ossent 21. Scilla kirkii Bak. Hucks 523. Urginea sp. aff. U. indica Kunth G. & K. 12,975; 13,030. Page 199 302 Araceae Amorphophallus gallaensis (Engl.) N. E. Br. A. gregoryana Engl. & Gehrm. G. & K. 12,689; Hucks 300; Oxford sn. Gregory sn. type, Mtito Andei, not seen. Stylochiton angustifolius A. Peter e desc. & Fig. Greenway EAH. 12,523; G. & K. 12,725; 12,751 ; Sheldrick B8731, 5. salaamicus N. E. Br. Greenway EAH. 12,710; G. & K. 12,724. Typha domingensis Pers. G. & K. 13,050 fide Kew. 305 Typhaceae 306 Amaryllidaceae Ammocharis tinneana (Kotschy & Peyr.) Milne-Redhead & Schweich. Hucks 463. Crinum kirkii Bak. Crimm sp. Hucks 464. Beecher H345/63/1. Crinum sp. Greenway EAH. 12,200. Cryptostephanus haemanthoides Pax G. & K. 12,853; G. & K. EAH. 13,725; Ossent 265. Haemanthus muhiflorus Martyn Hucks 446. Pancratium trianthum Herb. Hucks 261 ; 648. 307 Iridaccae Acidanthera Candida Rendle Hucks 603. 313 Agavaceae N. E. Br. in Kew Bull. 1915: 198. Sansevieria arborescens Gerome & Labroy S. caulescens N. E. Br. Ic. p. 200. Powell 4. Powell sn., Voi or Taru but exact locality uncertain. S. ehrenbergiana Bak. (Syn. S. robusta N.E. Br. Ic. p. 207). Grenfell 6; 13; 18; Powell sn.; Bally B8651 ; Hucks 706; Napier Bax TNP/E/56. S. intermedia N. E. Br. Ic. p. 211, f. 6. S. powellii N. E. Br. Ic. p. 198, f. 1. Powell 9. Powell 5 ; BaUy B7785 . S. singularis N. E. Br. Ic. p. 222. Powell 2; G. & K. 13,056; Sheldrick B8592; Bally 8592 Sansevieria sp. Hucks sn. 314 Palmae Hyphaene coriacea Gaertn. Sr. G. 319 Velloziaceae Vellozia aequatorialis Rendle Hucks 470; Napier 1342; Verdcourt & Polhill 2697. 326 Orchidaceae F.T.E.A. (1, 1968). Eulophia orthoplectra (Reichb. f.) Summerhayes E. petersii Reichb. f. Hucks 564; Napier 1038. Drummond & Hemsley 4232. E. wakefieldii (Reichb. f. & S. Moore) Summerhayes G. &K. EAH. 13,727. Habenaria ndiana Rendle, FI. Trop. Afr. 7: 239 (1898). Gregory sn. type, not seen. Vanilla rosclieri Reichb. f. G. 10,818; Hucks 456. in Glumiflorae (Families Nos. 327-332) 331 Cyperaceae Kukenthal, G. In Engler, A, & Diels, L. Das Pflanzenreich, IV, 20, 4, Cyperaceae 1935, Leipzig. Napper,D.M. Cyperaceaeof East Africa I-IW- Journ. E.A. Nat His Soc.24:2&5;25’.l ;26:1-1963-1966- Nairobi. Bulbostylis sp=Bagenal|4636. G. &.K. 12,770;il2,949. Page 200 331 Cyperaceae (cont.) Coelochloa setifera (Ridley) Gilly Polhill & Paulo 947. Cyperus alternifolius L. ssp. flabelliformis (Rottb.) Kukenth. Sr. G.; Napier Bax TNP/E/21. C. articulatus L. G. & K. 12,758; Napier Bax TNP/E/20; Sheldrick & Moore TNP/E/148. C. compressus L. C. distans L. f. G. & K. 12,873. G. & K. 13,001. C. giolii Chiov. G. & K. 12,716; Hucks 575. C. gmndibulbosus C. B. Cl. in Engl. A., Pflanzr. Cyperaceae-Scirpoid.— Cyperac. IV, 20: 125 (1935). Scott Elliot 6284 ex num. not_seen._ C. kaessneri C. B. Cl. G. 10,467. C. longus L. ssp. tenuiflorus (Rottb.) Kukenth. G. & K. 13,040. C. obtusiflonis Vahl G. & K. 12,788. F. exilis (H.B.K.) Roem. & Schult. G. & K. 12,978. K. triceps Rottb. var. obtusiflora Boeck. G. & K. 12,804. C. immensus C. Cl. var. taylori C. B. Cl. G. & K. 13,052; Napier Bax TNP/E/23. C. laevigatas L. G. & K. 13,039. C. macula t us Boeck. G. & K. 13,004. Fimbristylis bisumbellata (Forsk.) Bub. G. &K. 13,006; 13,033. Kyllinga alba Nees G. &K. 12,743; Hucks 357. Lipocarpha chinensis (Osb.) Kern G. & K. 13,036. Mariscus aristatus (Rottb.) Cherm. G. 10,469. M. mollipes C. B. Cl. Bogdan 3626. M. taylori C. B. Cl. G. & K. 12,775. Mariscus sp. Agnew et al. 5829; G. & K. 12,662. M. leptophyllus (Hochst.) C. B. Cl. G. & K. 12,802. M. obsoletenervosus (A. Peter & Kukenth.) Greenway G. &K. 12,651. M. taylori C. B. Cl. var. taylori Bogdan 3625 ; Drummond & Hemsley 4097. 332 Gramineae Bogdan A.V., A Revised List of Kenya Grasses, 2nd. ed. (1958), Nairobi. Napper, D.M., Grasses of Tanganyika (1965), Dar-es-Salaam. Acrachne racemosa (Roem. & Schult.) Ohwi Sheldrick TNP/Gs/29. Andropogon schinzii Hack. Greenway 9771; G. & K. 12,976; Napper 1341; Sheldrick 101. Aristida adscensionis L. A. barbicollis Trin. & Rupr. Greenway 9773; Polhill & Paulo 941. G. 9770; G. & K. 12,809; 12,928. A. coerulesceiis Desf. Gregory s.n. not seen. Possibly a mis-identification as this species is not recorded in Kenya. It is, I think, A. adscensionis L. A. mutabilis Trin. & Rupr. A. stenostacliys W. D. Clayton G. & K. 12,926. Bogdan 5322; G. 9769. Bothriochioa radicans (Lehm.) A. Camus G. 9776; G. & K. 12,900; Sheldrick 104. Brachiaria deflexa (Schumach.) Robyns B. eruciformis (J. E. Sm.) Griseb. Sheldrick TNP/E/4. G. 9783 ; G. & K. 12,969; Sheldrick 111. B. lachnantha (Hochst.) Stapf B. leersioides (Hochst.) Stapf Sheldrick 106. Bally B8643; G. & K. 12,795; Napier Bax TNP/E/13a. B. leucacrantha (K. Schum.) Stapf B. nigropedata (Munro) Stapf Bogdan 5679; G. 9774; Sheldrick NP/7. Rauh Ke957. B. (Hochst.) Stapf .0rac/?/ar/a sp.=Greenway 9782. Bogdan 5681 ; Sheldrick TNP/E/120. G. 9782 fide Kew; G. & K. 12,717. Cenchrus ciliaris L. Bally B8645; G. & K. 12,631 ; MacDonald 835; Napier Bax TNP/E/45; Sheldrick 115. C. se tiger us Vahl G. & K. 12,644. Chloris barbata Sw. C. gayana Kunth G. & K. 12,874. Rauh Ke951 ; 951a. Page 201 332 Gramineae (cent.) Chloris roxburghiana Schult. G. & K. 12,636; MacDonald 850; Trapnell 2226. Chrysopogon aucheri (Boiss.) Stapf var. quinqueplumis (A. Rich.) Stapf G. & K. 12,877; Sheldrick 107; Verdcourt 3898. Coix lacryma-jobi L. MacDonald 864; Napier 998. Cymbopogon pospischilii (K. Schum.) C. E. Hubb. G. & K. 12,881; Sheldrick NP/15. Cynodon dactylon (L.) Pers. C. plectostachyus (K. Schum.) Pilger G. & K. 12,667; 13,072; Sheldrick & Moore TNP/E/156. Sheldrick & Moore TNP/E/129. Dactyloctenium aegyptium (L.) Beauv. Bally B8644; Sheldrick TNP/E/11. D. scindicum Boiss. G. & K. 13,045. Digitaria aridicola Napper Sheldrick EAH. 12,507; 110; EAH; 12,506 type. D. milanjiana (Rendle) Stapf G. & K. 12,974. D. rivae (Chiov.) Stapf G. & K. 12,796; 12,879; Sheldrick NP/21. D. giganteum Fischer & Schweickt. G. & K. 12,959; Sheldrick TNP/E/11. Dactyloctenium sp. G. & K. 12,738. D. macroblephara (Hack.) Stapf Bogdan 3527;G.9775; G.&K. 12,787; 13,015. D. pennata (Hochst.) T. Cooke Bogdan 5690; G.&K. 12,884; Verdcourt 3884 D. velutina (Forsk.) Beauv. (Syn. Panicum fenestratum A. Rich.) Gregory sn.. Diplachne caudata K. Schum. G. & K. 12,759. Echinochloa haploclada (Stapf) Stapf G. & K. 12,670; 12,983; 12,999; Sheldrick TNP/E/121; Sheldrick & Moore TNP/E/149. Eleusine indica (L.) Gaertn. Bogdan 5700. Enneapogon cenchroides (Roem. & Schult.) C. E. Hubb. E. elegans (Nees) Stapf G.&K. 12,914. ^ Enteropogon macrostachyus (A. Rich.) Benth. G. & K. 12,632; Sheldrick 109; TNP/E/95; Verdcourt 2698. Emgrostiella bifaria (Vahl) Bor Bogdan 5684; Sheldrick 102; 103. Eragrostis aethiopica Chiov. Greenway 9779; G. & K. 13,064. E. caespitosa Chiov. G. &K. 12,806 ; Sheldrick NP/8 ; Verdcourt 2691 . E. ciliaris (L.) R. Br. G. & K. 12,742; MacDonald 853. E. exasperata A. Peter G. & K. 12,929. E. siiperba Peyr. G. & K. 12,903; MacDonald 833. Eriochloa nubica (Steud.) Thell. G. & K. 12,985. Eulalia sp. alT. E. fenuginea Stapf G. & K. 12,920. Eiistachys paspaloides (Vahl) Lanza & Mattel Sheldrick TNP/Gs/28. Hemarthria natans Stapf G. & K. 12,870. Heteropogon contortus (L.) Roem. & Schult. Bogdan 5685; G. & K. 12,925. Holcolemma canaliculatus (Nees) Stapf & C.E. Hubb. Parker GM99G. Hyparrhenia filipendula (Hochst.) Stapf var. pilosa (Hack.) Stapf G. &K. 13,006. Ischaemum afrum (J. F. Gmel.) Dandy G. & K. 12,794. Latipes senegalensis Kunth Bogdan 3624; G. & K. 12,799; Rauh Ke41 ; Tateoka 3110. Leersia hexandra Sw. G. & K. 12,868. E. rupestris (J. A. Schmidt) A. Chev. (Syn. E. somalensis Chiov.) Bogdan3613;G.&K. 12,833. E. aspera (Jacq.) Nees MacDonald 837. E. cilianensis Lutati Disney 58/7; Sheldrick 116; G. Williams 835. E. ciliaris (L.) R. Br. var. brachystachya Boiss. G. 10,470; G. & K. 12,742. E. horizontalis A. Peter Bogdan 3899; G. & K. 12,872; 12,960. 332 Gramineae (cont.) Leptocarydion vulpiastrum (De Not.) Stapf Bally B8921 ; MacDonald 860. Leptochloa obtiisiflora Hochst. G. &K. 12,723; 12,836. Oropetium thomaeum (L. f.) Trin. G. 9784 fide Kew. Panicum chusqueoides Hack. Parker EAH. 12,796. P. deustum Thunb. Bogdan 5323; Napier Bax TNP/E/32. P. infestum Anderss. Bogdan 5382; G. & K. 12,726; 12,898. P. meyerianum Nees G. & K. 12,866. P. repens L. G. & K. 12,869. Paspalidium geminatum (Forsk.) Stapf G. & K. 12,643. Paspalum vaginatum Sw. G. & K. 12,867. Pennisetum massaicum Stapf Bogdan 3614; G. & K. 12,834. Perotis patens Gand. G. & K. 12,930. Phragmites karka (Retz.) Steud. Heady 1668. Rhynchelytrum repens (Willd.) C. E. Hubb. G. & K. 12,972. Schmidtia bulbosa Stapf G. & K. 12,642; 12,646; 12,849. Schoenefeldia transiens (Pilg.) Chiov. G. 9781; G. & K. 12,793. Setaria pallidefusca (Schumach.) Stapf G. & K. 12,977. Sorghum brevicarinatiim Snowden var. swahelorum Snowden S. verticilliflorum (Steud.) Stapf MacDonald 852. Sheldrick TNP/E/122. Sporobolus festivus A. Rich. S. helvolus (Trin.) Dur. & Schinz G. & K. 12,741. G. & K. 12,675; 12,835 (young stage); 13,000. S. gemiratus W. D. Clayton G. &K. 12,753; 13,073, type. S. pellucidus Hochst. S. spicatus (Vahl) Kunth G. & K. 12,772. G. & K. 12,695. S. virginicus (L.) Kunth G. & K. 13,038. Stipagrostis hirtigluma (Trin. & Rupr.) De Winter (Syn. Aristida hirtigluma Tnn. & Rupr.) G. & K. 12,913. S. uniplumis (Licht.) De Winter (Syn. Aristida papposa Trin. & Rupr.) G. & K. 12,927. Tetrachaete elionuroides Chiov. G. & K. 12,882; Parker GM64G; EAH 12,802. Tetrapogon bidentatus Pilg. T. cenchriformis (A. Rich.) Clayton Bogdan 3903; 5321 ; G. & K. 12,638; 12,778; Sheldrick 118. Sheldrick NP/17. T. tenellus (Roxb.) Chiov. G. 10,430; G. & K. 12,883; Sheldrick & Moore TNP/E/157. Tragus berteronianus Schult. Bally B8642; G. & K. 12,808. Tricholaena eichingeri (Mez) Stapf & C. E. Hubb. G. & K. 12,856. Urochloa sclerochlaena Chiarugi Bogdan 5680; Parker 28. L. Ipanicea (Retz.) Ohwi Sheldrick 112. Oropetium sp. Verdcourt & Polhill 2703; G. & K. 12,766. P. coloratum L. Bogdan 5686; G. & K. 12,787. P. heterostachyum Hack. Napier 938. P. maximum Jacq. G. & K. 12,765; 12,784; Sheldrick TNP/E/98a. Panicum sp. nov.=Phipps 412. G. & K. 12,678. P. mezianum Leeke Sheldrick 105 P. mauritianus Kunth G. & K. 12,910. R. villosum (Pari.) Chiov. Bogdan 3915; G. & K. 12,950; Sheldrick 117. Page 203 NAME CHANGES Hemezygia fischeri (Guerke) Greenway comb, nov., Labiatae, Syn. Ocimum fischeri Guerke in Engl. Bot.Jahrb. 9\: 195 {m A), Fischer 501, type. Mariscus obsoletenervosus (A. Peter & Kukenth.) Greenway comb, nov., Cyperaceae, Syn. Cyperus obsoletenervosus A. Peter & Kukenth. in Engl. & Diels Das Pflanzenreich, Cyperaceae- Sceopoe- deae-Cypereae, 4, 20, 4: 548 (1936), A. Peter 13, 348: 10,723, types. Ornithogalum donaldsonii (Rendle) Greenway comb. nov. Liliaceae, Syn. Albuca donaldsonii Rendle 'mJourn. Bot. 1896: 131, Donaldson Smith sn. type. Floras and Monographs Consulted Clements, F. E. & Shear, C. L. The Genera of Fungi (1931), London. Copeland, E.B. Genera Filicum (1947), Waltham, Mass. Dale, I. & Greenway, P.J. Kenya Trees & Shrubs (1961) Nairobi. F. T. A. Flora of Tropical Africa, 1-10, 1 (1868-1937), London. F.T.E.A. Flora of Tropical East Africa, Ranunculaceae, now sixty-eight families completed — (1952-), London. Hutchinson, J. The Families of Flowering Plants, I Dicotyledons (1926), London. II Monocotyledons (1934), London. Smith, A. L. Lichens (1921), Cambridge. Page 204 INDEX to Families and Genera Numbers are those of the families in the list. Ferns, Fungi and Lichens are at the beginning of the list. ABrus 148 Aristida 332 Abutilon 132 ASclepiadaceae 231 ACacia 147 Asparagus 293 Acalypha 136 Aspilia 238 Acanthaceae 259 Aspleniaceae (Fern) Acanthospermum 238 Astripomoea 251 Achyranthes 63 Asystasia 259 Acidanthera 307 AThroisma 238 Acrachne 332 AZima 180 Acridocarpus 133 Actiniopteris (Fern) BAlanitaceae 195/A ADansonia 131 Balanites 195/A Adenia 101 Ballya 280 Adenium 230 Barleria 259 Adhatoda 259 Baseonema 231 Adiantaceae (Fern) Basilicum 264 Adiantum (Fern) Bauhinia 146 AErva 63 BEcium 264 Aeschynomene 148 Berchemia 190 AGavaceae 313 Berula 213 AIzoaceae 55 Bldens 238 ALbizia 147 Bignoniaceae 257 Albuca 293 BLepharis 259 Alectra 252 Blepharispermum 238 Aloe 293 Blumea 238 Altemanthera 63 BOerhavia 83 Alysicarpus 148 Bombacaceae 131 AMaranthaceae 63 Boraginaceae 249 Amaranthus 63 Borreria 232 Amaryllidaceae 306 Boscia 36 Ammannia 72 Boswellia 196 Ammocharis 306 Bothriochloa 332 Amorphophallus 302 BRachiaria 332 Ampelidaceae 193 Bridelia 136 Ampelocissus 193 BUlbostyleis 331 Amyema 185 Burseracae 196 ANacardiaceae 205 Buttonia 252 Andropogon 332 Aneilema 280 CAdaba 36 Anisotes 259 Caesalpinia 146 Anthericopsis 280 Caesalpiniaceae 146 Anthericum 293 Calotropis 231 APhania 198 Calyptrotheca 56 Apocynaceae 230 Capitanya 264 ARaceae 302 Capparidaceae 36 Argemene 32 Capparis 36 Page 205 Capsicum 250 Crotalaria 148 Caralluma 231 Croton 136 Cardiospermum 198 Cruciferae 39 Carissa 230 Cryptostephanus 306 Cassia 146 CUcumella 103 Cassine 173 Cucumis 103 Catharanthus 230 Cucurbitaceae 103 Caucanthus 133 CYclantheropsis 103 CElastraceae 173 Cyclocheilon 263 Celosia 63 Cymbopogon 332 Cenchrus 332 Cymbosetaria 332 Centemopsis 63 Cynanchum 231 Cephalocroton 136 Cynodon 332 Cephalopentandra 103 Cyperaceae 331 Ceterach (Fem) Cyperus 331 CHascanum 263 Cyphostemma 193 Chenopodiaceae 61 Chenopodium 61 DActyloctenium 332 Chloris 332 Dalbergia 148 Cholorophytum 293 Dalechampia 136 Chrysopogon 332 Dasystachys 293 CIssus 193 Datura 250 Citrullus 103 DEinbollia 198 CLematis 15 Delonix 146 Cleome 36 DIchrostachys 147 Clerodendrum 263 Dicliptera 259 Clitoria 148 Dicoma 238 COccinia 103 Dicotyledons Coelochloa 332 Digera 63 Coix 332 Digitaria 332 Coldenia 249 Diospyros 221 Coleus 264 Dipcadi 293 Combretaceae 121 Diplachne 332 Combretum 121 Diplostigma 231 Commelina 280 Dirichletia 332 Commelinaceae 280 Disperma 259 Commicarpus 83 DObera 180 Commiphora 196 Dolichos 148 Compositae 238 Dombeya 130 Convolvulaceae 251 Dorstenia 167 Convolvulus 251 DRegea 231 Conyza 238 Drimiopsis 293 Corallocarpus 103 DYschoriste 259 Corbichonia 54 Corchorus 128 EBenaceae 221 Cordla 249 ECbolium 259 CRabbea 259 Echidnopsis 231 Craibia 148 Echinochloa 332 Crassulaceae 45 Eclipta 259 Craterostigma 252 EDithcolea 231 Crinum 306 EHretia 249 Crossandra 259 ELeusine 332 Page 206 ENdostemon 264 Haemanthus 306 Enicostema 239 Haplocoelum 198 Enneapogon 332 Harveya 252 Entada 147 HElichrysum 238 Enteropogon 332 Helinus 190 EQuisetaceae (Fern ally) Heliotropium 249 Equisetum (Fern ally) Helixanthera 185 ERagrostiella 332 Hemarthria 332 Eragrostis 332 Hemizygia 264 Erianthemum 185 Flermannia 130 Eriochloa 332 Heteropogon 332 Erlangea 238 Hibiscus 132 Erythrina 148 Hildebrandtia 251 Erythrocephalum 238 Hippocratea 173 Erythrochlamys 264 Hirpicium 238 Erythrococca 136 HOlarrhena 230 EUlalia 332 Holcolemma 332 Eulophia 326 Hoslundia 264 Euphorbia 136 HYbanthus 40 Euphorbiaceae 136 Hydnora 26 Eustachys 332 Hydnoraceae 26 EVolvulus 251 Hydrocharitaceae 266 EXcoecaria 136 Hymenodictyon 232 Hyparrhenia 332 FAgara 194 Hypertelis 54 Farsetia 39 Hyphaene 314 FIcoidaceae 54 Hypoestes 259 Ficus 167 Fimbristyis 331 IBoza 264 ICacinaceae 179 GAlinsoga 238 ILysanthes 252 Garcinia 126 INdigofera 148 Gardenia 232 IPomoea 251 GEigeria 238 IRidaceae 307 Gentianaceae 239 ISchaemum 332 Geraniaceae 67 Gerrardanthus 103 JAcquemontia 251 GIsekia 54 Jasminum 229 Givotia 136 Jatropha 136 GLinus 54 JOsephinia 258 Gloriosa 293 JUsticia 259 Glossonema 231 GNidia 81 KAempferia 290 GOmphrena 63 Kalanchoe 45 GRamineae 332 Kanahia 231 Grangea 238 KEdrostis 103 Grewia 128 KIgelia 257 GUtenbergia 238 KLeinia 238 Guttiferae 126 KOhautia 232 GYnandropsis 36 KYllinga 331 HAbenaria 326 LAbiatae 264 Page 207 Lagarosiphon 266 Lagenaria 103 Laimea 205 Lantana 263 Latipes 332 Launaea 238 Lawsonia 72 LEcaniodiscus 198 Leersia 332 Leonotis 264 Lepidagathis 259 Leptocarydion 332 Leptochloa 332 Leucas 264 Llliaceae 293 Limeum 54 Lipocarpha 331 LObelia 244 Lobeliaceae 244 Loganiaceae 228 Lonchocarpus 148 Loranthaceae 1 85 Loranthus 185 LUdwigia 77 LYcium 250 Lycoperdaceae (Fungus) Lythraceae 72 MAerua 36 Malpighiaceae. 133 Malvaceae 132 Manilkara 222 Mariscus 331 Marsilea (Fem) Marsileaceae (Fern) Maytenus 173 MElhania 130 Melia 197 Meliaceae 197 Menispermaceae 23 Merremia 251 Meyna 232 MIcroglossa 238 Mimosaceae 147 Mirausops 222 MOllugo 54 Momordica 103 Monadenium 136 Monechma 259 Monsonia 67 Moraceae 167 Moringa 37 Moringaceae 37 NAjadaceae. 279 Najas 279 NEorautanenia 148 Neptunia 147 Neuracanthus 259 Newtonia 147 NOtonia 238 NYctaginaceae 83 Nymphaea 18 Nymphaeaceae 18 OChna 114 Ochnaceae 114 Ocimum 264 OLacaceae 1 82 Oldenlandia 232 Oleaceae 229 Oliverella 185 ONagraceae 77 OPhrestia 148 ORchidaceae 326 Ormocarpum 148 Omithogalum 293 Oropetium 332 Orthosiphon 264 OSteospermum 238 OXygonum 57 Oxystelma 231 PAederia 232 Palmae 314 Pancratium 306 Panicum 332 Papaveraceae 32 Papilionaceae 148 Paspalidium 332 Paspalum 332 Passifloraceae 101 Pavonia 132 PEdaliaceae 258 Pedalium 258 Pegolettia 238 Pellaea (Fern) Pennisetum 332 Pentanisia 232 Pentas 232 Pentodon 232 Peponium 103 Pergularia 231 Peristrophe 259 Page 208 Perotis 332 Salvadoraceae 180 PHragmites 332 Sansevieria 313 Phyla 263 Sapindaceae 198 Phyllanthus 136 Sapotaceae 222 PLatycelyphium 148 Sarcostemma 231 Plectranthus 264 SChlechterella 231 Pleuropterantha 63 Schmidtia 332 Plicosepalus 185 Schoenefeldia 332 Pluchea 238 Scilla 293 Plumbaginaceae 241 Sclerocarpus 238 Plumbago 241 Scrophulariaceae 252 POdaxis (Fungus) SEcamone 231 Polygala 42 Securinega 136 Polygalaceae 42 Seddera 251 Polygonaceae 57 Selaginella (fern ally) Polygonum 57 Selaginellaceae (fern ally) Polyporaceae (Fungus) Senecio 238 Polystictus (Fungus) Sericocomopsis 63 Populus 156 Sesamothamnus 258 Portulaca 56 Sesbania 148 Portulacaceae 56 Setaria 332 PRemna 263 SIda 132 Priva 263 SOlanaceae 250 PSeudosopubia 252 Solanum 250 Psilotrichium 63 Sorghum 332 Psychotria 232 SPathionema 148 PTeridaceae (Fern) Sphaeranthus 238 Pterodiscus 258 Spilanthes 238 PUpalia 63 Spirostachys 136 PYcnostachys 264 Sporobolus 332 Pyrenacantha 179 STathmostelma 231 Steganotaenia 213 RAnunculaceae 15 Sterculia 130 RHamnaceae 190 Sterculiaceae 130 Rhamphicarpa 252 Stemodiopsis 252 Rhinacanthus 259 Stictocardia 251 Rhynchelytrum 332 Stipagrostis 332 Rhynchosia 148 Striga 252 RIcinus 136 Strophanthus 230 Rinorea 40 Strychnos 228 ROcella (Lichen) Stylochiton 302 Rocellaceae (Lichen) Stylosanthes 148 Rorippa 39 SUaeda 61 RUbiaceae 232 SVensonia 263 Ruellia 259 Rutaceae 194 TAlinum 56 Ruttya 259 Tamarindus 146 RYtigynia 232 Tapinanthus 185 Tarenna 232 SAcleuxia 231 TEloschistaceae (Lichen) Salicaceae 156 Teloschistes (Lichen) Salvadora 180 Tephrosia 148 Page 209 Terminalia 121 UMbelliferae 213 Tetrachaete 332 URginea 293 Tetrapogon 332 THespesia 132 VAnilla 326 Thunbergia 259 Vatovaea 148 Thylachium 36 VEllozia 319 Thymelaeaceae 81 Velloziaceae 319 Tlliaceae 128 Vepris 194 Tiliacora 23 Verbenaceae 263 Tinnea 264 Vemonia 238 TRagia 136 VIgna 148 Tragus 332 Violaceae 40 Trianthema 54 Vitaceae 193 Triaspis 133 Vitex 263 Tribulus 66 Tricalysia 232 WAltheria 130 Trichilia 197 WIthania 250 Trichodesma 249 WOrmskioldia 34 Tricholaena 332 Triclisia 23 XEromphis 232 Tridax 238 Xlmenia 182 Triumfetta 128 Trochomeria 103 ZAleya 54 Tryphostemma 101 ZIngiberaceae 290 TUrbina 251 Ziziphus 190 Turneraceae 34 ZOmia 148 TYlosema 146 ZYgophyllaceae 66 Typha 305 Typhaceae 305 (Received 3rd February, 1968) Page 211 NOTES ON SOME BUTTERFLIES OF THE NGARA DISTRICT OF TANZANIA WITH A LIST OF THOSE RECORDED By L. A. Haldane The following notes are based on a small collection made in 1948 and 1949 and vegetational and faunal changes may well have occurred since then. Collecting was undertaken only sporadically and as opportunity offered and achieved a far from complete record of butterflies in the area, Lycaenids and Hesperiids in particular being poorly represented. Sufficient species were however taken to give a fair indication of relationships with adjoining areas. A few records are, 1 think, new to the Tanzania list. The Ngara district, with an area of 1,045 sq. miles, lies between fifty and eighty miles to the west of the southern end of Lake Victoria. The altitude varies from 4,200 to 6,000 feet a.s.l. and the rainfall, which occurs mainly between October and April, averages about forty inches a year. The country falls into five main zones: highlands: These lie between 5,500 and 6,000 feet a.s.l., mostly in the form of ridges on a SW/NE axis, outliers to the main body of the Burundi uplands. These ridges are generally open and grass-covered but with patches of stunted thorn and other trees. Anianartia abyssinica Fid., Colias electo L. and Pontia helice johnstoni Crowley occur here. In Bugufi in the north, this zone is heavily populated. Extensive plantations of bananas and coffee intermingled with figs and other trees have given shelter to indigenous vegetation which has been burned off elsewhere. As a result many forest margin species from the next zone are encountered here as well as the usual open country populations. MIDDLE levels: The greater part of the district lies between 4,500 and 5,500 feet and is covered with trees, more or less thickly distributed, with a strong growth of tall grass, annually burned, below. Small pockets of evergreen forest remain here and there and narrow and discontinuous strips of fringing forest in some of the valleys extend through many parts of this zone. While therefore the population is mainly that of open or sylvan country the denser areas support such forest-margin species as Tirumala limniace petiverana Dbl. & Hew., Amauris n. niavius L., Acraea z. zetes L., Acraea a. asboloplintha Karsch, A. s. sotikensis E. Shpe., A. p. perenna Dbl. & Hew., A. johnstoni butleri Aur., Charaxes c. castor Cr., Ch. p. pollux Cr., Ch. dilutus Rths., Ch. etheocles ochracea Roths., Precis natalica Fid., Asterope garega Karsch, Vannessula milca latifasciata Tbt., Pentila p. peucetia Hew., Appias epaphia orbona Bsd., Mylothris poppea Cr., Nepheronia argia varia Trim., Papilio d. dardanus Brown, P. phorcas congoanus Rths., P. bromius chrapkowskoides Storace, P. nireus lyaeus Dbl., and Graphium ridleyanus White. swamps: Many of the broader valleys in both the above zones contain perenially wet grasslands shading into marsh. Here Acraea acerata Hew., A. ventura Hew., Catacroptera cloanthe Cr., Precis ceryne Bsd., Mylothris bernice rubricosta Mab. Etirema hapale Mab. and Metisella midas Btl. occur. LOWLAND forest: The deep gorge of the Ruvuvu river cuts through the district at just over 4,000 feet and along its length, and along the lower reaches of some of its tributaries, small stretches of evergreen forest are to be found where plentiful groundwater supplements the rainfall. The greatest extent of forest occurs at the Russumo Falls in the north-east of the district just below the juction of the Ruvuvu and Kagera rivers. This latter forms part of the northern boundary of the district D Page 212 and flows through a broad belt of papyrus from which the occasional forested island rises a few feet above the level of the swamp. A very brief visit to this zone produced Amauris t. tartarea Mab., Bamatistes quadricolor latifasciata E. Shpe., Bematistes p. poggei Dew., Acraea e. egina Cr., Precis westermanni West., Abisara rogersi Druce and Papilio zoroastres joiceyi Gab. plains: In the south-eastern corner the country falls a few hundred feet, but fairly sharply, to form part of the vast sylvan area which runs south from Lake Victoria through western Tanzania. Among the larger species Charaxes guderiana Dew., Ch. ethalion Bsd., Graphium policenes Cr. and G. antheus Cr. are typical. GEOGRAPHICAL DISTRIBUTION The district forms part of the transitional area between the plains of western Tanzania and the high country which runs north through Burundi and Rwanda to Kabale and the Ruwenzoris in western Uganda. As with the birdsi affinities are predominantly East African. Some 92 out of 146 species identified from Ngara occur in the same races and forms in Rhodesia.2 While 28 of these are fairly ubiquitous Ethiopian species, in the main they are typical of the sylvan area which extends a thousand miles southwards from Lake Victoria. Of the remaining species one group is found particularly on the northern and western edges of this area, distributed from Angola through Burundi, Rwanda, Uganda, Kenya west of the Rift Valley, and as far as Abyssinia in some instances. Ngara representatives of this group include Acraea e. egina Cr., natalica abadima Ribbe, A. s. sotikensis E. Shpe., A. ventura Hew., Charaxes dilutus Rihs., Precis sophia infracta Btl., Bicyclus vulgaris Btl., Neocoenyra cooksoni Ham., Belenois rubrosignata Weym., Colotis eucharis evarne Klug, C. aurigineus Btl., C. hetaera Gerst., and Meti- sella midas Btl. Of more limited distribution within this area, being largely confined to western Kenya, Uganda and north-western Tanzania are Bematistes quadricolor latefasciata E. Shpe., Acraea a. asboloplintha Karsch, Bicyclus saussurei Drury, P. bromius chrapkowskoides Storace, and P. zoroastres joiceyi Gab. West African influences are seen from the following which are found from Senegal to the Congo and Uganda. Amauris n. niavius L., A. t. tartarea Mab., Bematistes p. poggei Dew., Acraea z. zetes L., A. p. perenna Dbl. & Hew., A. johnstoni butleri Aur., Charaxes c. castor Cr., Ch. p. pollux Cr., Precis wetsermanni West., Vanessula milca latifasciata Tbt., Bicyclus campa Karsch, Abisara rogersi Druce, Anthene crawshayi Btl., Mylothris poppea Cr., Belenois theora lortzingi Suflf., Papilio d. dardanus Brown, P. phorcas congoanus Rths., and Graphium ridleyanus White. Again West African, from the Cameroons to Uganda and Abyssinia, are Asterope garega Karsch, Ariande pagenstecheri Suflf., Precis pelarga Fab., Belenois subeida Feld., Belenois solilucis Btl. and Leptosia medusa Cr. The butterfly population of the more open formations is chiefly that of the south- ern sylvan zone^ represented by 67 out of 89 species with, for example, Graphium pylades represented by the eastern and southern race angolanus Goeze rather than by race pylades Fab. which occurs in what are comparatively nearby regions of Uganda. Butterflies associated with forests or forest margins show affinities with the western lowland forest zone^, some 32 out of 47. Many of these occur also on Ukerewe Island oflf the Bukoba shores of Lake Victoria, though Bematistes p. nelsoni Sm. & Kby. which occurs there-* is replaced by the Congolese race poggei Dew. in Ngara. In general the picture seems to be one of a withdrawal of the forests towards the Congo basin with butterflies of the open formations pressing in from the south and east to fill the gap. Meanwhile the western forest species seem likely to have an in- creasingly precarious future to look forward to, reliant as they are on the tenuous thread of forests along the river valleys. Page 213 LIST OF BUTTERFLIES RECORDED IN THE NGARA DISTRICT DANAIDAE Danaus chrysippus L., Tirumala limniace petiverana Dbl. & Hew., Amauris n. niavius L., Amauris t. tartarea Mab. ACRAEIDAE Bematistes quadricolor latifasciata E. Shpe., B.p. poggei Dew., Acraea z. zetes L., A. e. egina Cr. & f. harrisoni E. Shpe., A. natalica abadima Ribbe, A. asboloplintha asboloplintha Karsch, A. encedon L., A. s. sotikensis E. Shpe., A. cabira Hpff. A. acerata Hew., A. eponina Cr., A. ventura YicyN.,A.p.perenna Dbl. & Hew., A.johnstoin butleri Aur. NYMPHALIDAE Charaxes varaties vologeses Mab., Ch. candiope Gdt., Ch. c. castor Cr., Ch. p. pollux Cr., Ch. dilutus Rths., Ch. achaemenes Fid., Ch. guderiana Dew., Ch. viola kirki Btl., Ch. viola vansoni van Som. & J., Ch. etheocles ochracea Roths., Ch. ethalion Bsd., Crenidomimas concordia Hpff., Hamammida dae- dalus Fab., Aterica galene Brown, Neptis saclava marpessa Hpff., N. jordani Neave, N. laeta Over., N. lativittata Over., N. alta Over., Asterope moranti dubiosa Strd., A. boisduvali Wall., A. garega Karsch, Byblia acheloia Wall., Ariadne pagenstecheri Suff., Neptidopsis ophione velleda Mab., Eurytela dryope Cr., Hypolimnas misippus L., Salamis parhassus aethiops de Beauv., Catacroptera cloanthe Cr., Precis artaxia Hew., P. natalica Fid., P. terea elgiva Hew., P. archesia Cr., P. tugela Trim., Precis actia Dist., P. perlarga Fab., P. ceryne Bsd., P. antilope Feist., P. o. octavia Cr., P. sophia in- fracta Btl. & f. albida Suff., P. westermanni West., P. oenone L., P. hierta Fab. P. orithya madagas- cariensis Guen., Vanessula milca latifasciata Tbt., Vanessa cardui L., Antanartia abyssinica Fid., Pbalanta columbina Cr., P. phalantha Dr., SATYRIDAE Melanitis leda Dry., Gnophodes parmeno diversa Btl., Bicyclus safitza West., B. campa Karsch, B. vulgaris Btl., B. angulosus Btl., B. saussurei Dry., Henotesia perspicua Trim., Neocoenyra gregorii Btl., A. Hamltn., YpthimaasteropeKlug, Y. impuraElw. & Edv/., Y.albidaBtL Abisara rogersi Druce. RIODENIDAE LYCAENIDAE Ornipholidotos peucetia peucetia Hew., Lachnonecma bibulus Fab., Virachola antalus Hpff., Hypoly- caena philippus Fab., H. buxtoni Hew., Spindasis mozambica Bert., Axiocerses harpax Fab., Anthene crawshayi Btl., A. definita Btl., Syntaracus telicanus Lang. PIERIDAE Appias epaphia orbona Bsd., Belenois gidica Gdt., B. creona Cr., B. aurota Fab., B. zochalia f. tanga- nyikae Lanz, B. rubrosignata Weym., B. subeida Feld., B. theora lortzingi Suff., B. solilucis Btl., Dixeia orobona vidua Btl., D. pigea Bsd. & f. rubrobasalis Lanz, Mylothris chloris agathina Cr., M. poppea Cr. f. tirikensis Neave, M. bernice rubricosta Mab., Leptosia medusa Cr., L. alcesta Stoll., Pontia helice johnstoni Crowley, Pinacopteryx eriphia Gdt., Colotis aurigineus Btl., C. hetaera Gerst., C. danae Fab., C. eucharis evarne Klug, C. antevippe Bsd., C. evagore antigone Bsd., C. eris Klug, Eronia cleodora f. erxia Hew., E. leda Bsd., Nepheronia thalassina Bsd., A. argia varia Trim., Colias electo L. & f. aurivilliusi Kef., Catopsilia florella Fab., Eurema hecabe L., E. brigitta Cr., E. hapale Mab., E. desjardinsi Bsd. PAPILIONIDAE Papilio d. dardanus Brown, P. phorcas congoanus Rths., P. bromius chrapkowskoides Storace, P. nireus lyaeus Dbl., P. demodocus Esper, P. zoroastres joiceyi Gab., Graphium pylades angolanus Goeze, G. ridleyanus White, G. leonidas Fab., G. policenes Cr., G. antheus Cr., HESPERIIDAE Tagiades flesus Fab., Eretis lugens Rog., Abantis zambesiaca West., Spialia dromus Plotz, Metisella midas Btl., M. orientalis Aur., Ampittia capenas Hew., Kedestes mohozutza Wall,, Borbo mathias Fab. ACKNOWLEDGEMENTS I am most grateful to Mr. T. G. Howarth of the British Museum (Natural History) for considerable help with determinations. Page 214 REFERENCES 1 Haldane, L. A., 1951, Birds of the Ngara District, T.N.R. 30. 2 PiNHEY, E. C. G., 1949, Butterflies of Rhodesia. 3 Carcasson, R. H., 1964. A Preliminary Survey of the Zoogeography of African Butterflies. E. Afr. Wildl. J. 2. 4 Carcasson, R. H., 1961, Acraea Butterflies of E. Africa./.£'. Afr. nat. Hist. Soc. Supp. 8. (Received 1st July, 1968) Page 215 GEOCARPY AS AN ADAPTATION TO AFROALPINE SOLIFLUCTION SOILS By A. D. Q. Agnew & O. Hedberg INTRODUCTION One of us (Hedberg, 1964) has given a comprehensive account of the ecology of Afroalpine plants, without any specific mention of geocarpy as an adaptation to the constant solifluction that may be experienced throughout the year in certain habitats. Since the publication of this account, the present two authors have revisited high altitude vegetation in East Africa and the purpose of this paper is to draw attention to the phenomenon and to remark on the species that show it and their ecology. GEOCARPY AS AN ADAPTATION Hedberg (1964, pp. 29-33 and 64-70) has outlined the features of solifluction soils, and so it is unnecessary to say more here than that they are intermittently moist soils exposed to rapid changes between mild day - and low night-temperatures. The resultant ice formation, and frost-heaving of the soil surface, often occurring every 24 hours, renders the soil surface extremely mobile and unstable. The plants which can grow in such a habitat must be adapted to it, and the adaptations already noted include tough rhizomes and/or roots, the formation of a “bunch” (i.e. tussock or cushion) of strong stems at ground level and the bizarre “free-living” mosses and lichens which are not attached to the soil at all. Clearly, establishment of seedlings must be extremely difficult in such a habitat, for before the young roots have penetrated the mobile surface layer of the soil and got anchored below it they are usually pulled out by frost-heaving. This then, is a critical phase in the life-history of these afroalpine plants and it is not surprising that many of them have developed geocarpy as a means of adapting to it. Geocarpy means burial of the fruit while still attached to the plant, and it is usually accomplished by bending the pedicel or peduncle down into the soil. Geocarpic plants of solifluction soils are low, mostly perennial, with a rosette or creeping habit and tough roots. From our observations it seems that some species are strongly geocarpic, while others show this feature only weakly. Strongly geocarpic afro-alpines so far noted are: Haplocarpha rueppellii (Sch. Bip.) Beauv. Haplosciadiurn abyssinicum Hochst. Limosella africana Gluck L. macrantha R. E. Fr. Ranunculus cryptanthus Milne- Redhead & Turrill R. oreophytus Del. R. stagnalis A. Rich. Several other species tend towards geocarpy in that their pedicels are often re- flexed at maturity bringing the fruits into the region of the soil surface and often in contact with the latter, though they are not actively buried. According to a termi- nology coined by Hylander (1929 p. 211) these may be called depositors — their fruits are deposited on the soil surface. On active solifluction soil such fruits may sometimes get buried by soil movements before they are detached from the plant or the seeds are Page 216 released, so that the distinction between depositors and truly geocarpic plants may become vague in this case. The following species have been found by us to behave usually as depositors; Alchemilla microbetula Th. Fr. jr. A. subnivalis E. G. Bak. Anagallis serpens DC. ssp. meyeri-johannis (Engl.) P. Tayl. Lobelia duriprati Th. Fr. jr. L. lindblomii Mildbr. L. minutula Engl. Oreophyton falcatum (A. Rich.) O. E. Schulz Ranunculus volkensii Engl. Trifolium acaule A. Rich. T. elgonense Gillett Uebelinia crassifolia Th. Fr. jr. Veronica gunae Engl. These are all plants of open soils, sometimes found at roadsides and in muddy hollows between tussocks at lower altitudes on the mountains, and occupying the solifluction soils where these exist. Other plants which grow in such soils are those with a densely tufted habit such as Agrostis sclerophylla C. E. Flubb., Cotula crypto- cephala A. Rich., Myosotis keniensis Th. Fr. jr., Sagina afroalpina Hedb., and Swertia subnivalis Th. Fr. jr. We still do not know how Subularia monticola Schweinf. subsists on its per- manently moist localities with almost nightly frost-heaving (see Hedberg 1964, p. 66), with no fruit burial or other apparent adaptation. The necessity of maintaining active growth under conditions of diurnal alternate freezing and thawing, throughout the year, is responsible for most of the special features of the Afroalpine vegetation. This short note merely records one more. SUMMARY A record is made of 7 strongly geocarpic herbs and of 12 other species which are weakly geocarpic, or depositors, all occurring on solifluction soils in the alpine belt of the high East African mountains. This adaptation to tropical high mountain solifluction has not been noted before. REFERENCES Hedberg, O., (1964). Features of Afroalpine plant ecology. Acta Phytogeogr. Sitecica 49: 1-144. Hylander, N., (1929). Diasporenabtrennung und Diasporen-Transport. Svensk bot. Tidskr. 23: 184-218. (Received 26th June, 1968) BIRD RINGING REPORT 1967-1968 By G. C. Backhurst Page 217 INTRODUCTION This report covers the period 1st July 1967 to 30th June 1968. The number of birds ringed is higher than ever before, unfortunately it is doubtful whether this upward trend can be maintained. One of the most prolific ringers, Dr. D. J. Pearson, left East Africa in June 1968 and his contribution will be missed greatly, although it is probable that some other experienced ringers will be coming out from England towards the end of 1968. Ringing has continued to be centred around Kampala, Nakuru and Nairobi, with smaller amounts in Masindi, the Queen Elizabeth National Park, Kisumu and southern Tanzania. The number of ringers operating in the three countries is still very small indeed and, considering this, the amount of ringing done is highly satis- factory. The full list of birds ringed is given in Table 1 ; birds which are palearctic migrants are printed in bold type, others which are included in the palearctic fauna but which are also ethiopian are not so distinguished. The order is that of Mackworth-Praed & Grant and their numbers are given before the English names, the nomenclature follows these authors for the African species, whereas Vaurie is used for the palearctic birds. SOME NOTES ON RINGING IN EAST AFRICA In the report on the previous season’s activities (Backhurst 1968) I was able to mention only five ringers working in the whole of East Africa; this season the number rose to sixteen. Dr. D. J. Pearson, who left in June 1968, ringed over 2,000 birds, all but 97 of them being palearctic migrants. He ringed most of the waders, warblers, swallows and shrikes — the waders at Lake Nakuru, the others around Kampala. D. J. M. Calfyn ringed a few Yellow Wagtails at Thika, while W. P. Langridge ringed exactly 300 at Kisumu. A. R. E. Sinclair ringed all the Little Swifts as part of a research programme in the Serengeti. R. J. Wheater, Chief Game Warden in Mur- chison Falls National Park, ringed a few migrants and expects to ring many more next season. R. Douthewaite also ringed a few birds in Uganda as part of one of his research programmes. F. J. Thompson at Masindi, Uganda, ringed a number of birds including Redstarts — the first to be ringed in East Africa. A. D. Forbes-Watson, working in Liberia, ringed six Yellow Wagtails with Nairobi rings and hopes to ring many more migrants in the autumn of 1968. D. A. Turner ringed a number of migrants in Kenya and Uganda and will continue to ring next season. Mrs. A. L. Campbell, Dr. E. D. Steel, my wife and I worked mainly in the Nairobi area, either independently or together where 2,001 Yellow Wagtails were caught at Kabete, 1,093 at Eastleigh Sewage Works and 493 at Kariobangi Sewage Works. All but seven of the wagtails caught at Kabete were netted as they came in to roost in Napier Grass, Pennisetum purpureum Schumach. This year there were seldom suitable congregations of wagtails in the fields at Kabete to make the previous season’s catching method practicable (Backhurst, loc. cit.), nevertheless, the roost provided over three times the previous season’s total. Netting was successful on 59 nights with four catches over one hundred (maximum 122), the average catch was 33.8 Yellow Wagtails per night; in addition small numbers of other species, notably Olivaceous Warblers, were caught at the roost. Very few sites have been found where large numbers of migrants can be caught. Passerines have only been worthwhile around Kampala, Kisumu and Nairobi. Page 218 Lake Nakuru has proved to be an excellent wader site, but so far the Coast, despite vast numbers of birds present, has proved unfruitful: J. R. Stjernstalt, working at Mtwara, southern Tanzania, lost all his nets and rings to a sudden tidal wave; the writer tried netting at Gazi, southern Kenya coast in October 1967, but only managed to ring six birds. It seems that, in contrast with the conditions in England, waders on the East African coast have a wide choice of high-water resting places so that, once disturbed by netting activities, they can easily move elsewhere. ACKNOWLEDGEMENTS Ringers gratefully acknowledge the co-operation of the City Engineer, Nairobi, for allowing them to operate at the two sewage works; the Director of the Kenya National Parks for permission to ring at Lake Nakuru; and of the Director of Veteri nary Services, Kenya, for permission to ring at Kabete. Sincere thanks are also due to the following willing helpers, most of whom made ringing much easier at the sewage works and at Kabete: Miss P. Allen, Miss H. Anderson, L. C. and Mrs. R. L. Backhurst, W. Bruce, H. Buck, A. Bygrave, K. Campbell, Miss B. Debbenham, M. D. Ford, J. R. Hudson, Mrs. J. Hyland, A.n Igles, Miss H. Irwin, L. Joyner, E. Lonsdale, Miss U. McCurdy, J. McGhee, T. McNett, Miss C. Parsons, J. H. Phillips, R. Segal and F. Topliff. Very special thanks go to Mr. B. T. Parsons who contributed in many ways to make the season so successful. The assistance of my wife with all the ringing and with the administrative tasks was quite invaluable. The Society is grateful to the Director of the National Museum, Nairobi, for allowing the Museum’s address to appear on the rings. Table 1 BIRDS RINGED BY THE EAST AFRICA NATURAL HISTORY SOCIETY RINGING ORGANIZATION Palearctic Migrants 1967/8 4 Little Grebe Podiceps nificollis (Pallas) ...... 1 63 Sacred Ibis Threskiornis aethiopicus (Latham) .... 0 70 African Spoonbill Platalea alba Scopoli ...... 0 72 Lesser Flamingo Phoeniconaias minor (Geoffroy) .... 0 77 African Pochard Aythya erythrophthalma (Wied) .... 0 83 Yellow-billed Duck Anas undulata Dubois ..... 0 88 Garganey A. querquedula Linnaeus ....... 0 89 Cape Wigeon T. cope/wA Gmelin ....... 16 90 Hottentot Teal T. pwHc/am Burchell ...... 4 91 Red-billed Duck A. erythrorhyncha Gmelin ..... 0 21 1 Quail Cotiirnix coturnix africana Temminck & Schlegel ... 0 242 Red-knobbed Coot Fulica cristata Gmelin ..... 0 266 Ringed Plover Charadrius hiaticula Linnaeus ..... 38 267 Little Ringed Plover Ch. diibius Seopoli 7 270 Chestnut-banded Sand Plover Ch. venustus Fischer & Reichenow . 0 271 Kittlitz’s Sand Plover CA. pecwarhw Temminck .... 92 272 Three-banded Plover Ch. tricollaris Vieillot ..... 3 274 Mongolian Sand Plover Ch. mongolus Pallas ..... 1 275 Great Sand Plover Ch. leschenaidtii Lesson ..... 2 286 Spurwing Plover Hoplopterus spinosus (Linnaeus) .... 4 287 Blacksmith Plover H. armatus (Burchell) ...... 39 295 Avocet Recurvirostra avosetta Linnaeus ...... 5 296 Black-winged Stilt Himantopus himantopus (Linnaeus) ... 2 298 European Snipe Gallinago gallinago (Linnaeus) ..... 1 300 African Snipe G. nigripennis (Bonaparte) 2 303 Curlew Sandpiper Calidris ferruginea Pontoppidan .... 52 305 Little Stint C. minuta (Liesler) 554 Grand total 1 7 73 6 3 31 1 20 34 43 1 15 42 7 100 106 4 2 4 5 45 6 3 75 793 Page 219 306 Temminck’s Stint C. temminckii (Liesler) .... 308 Sanderling C. alba (Pallas) ....... 309 Ruff Philomachus pugiiax (Linnaeus) ..... 311 Terek Sandpiper cwem/s GUldenstadt .... 312 Common Sandpiper Trirtga hypoleucos Linnaeus 314 Wood Sandpiper T. glareola Linnaeus ..... 317 Marsh Sandpiper T. stagnatilis (Bechstein) .... 318 Greenshank T. nebularia (Gunnerus) ..... 329 Violet-tipped Courser Rhinoptihis chalcoptenis (Temminck) 335 Crab Plover Dramas ardeola Paykull ..... 349 Gull-billed Tern Gelochelidon nilotica (Gmelin) 361 White-winged Black Tern Chlidonias leiicoptera (Temminck) 394 Tambourine Dove Tympanistria tympanistria (Temminck & Knip) 397 Emerald-spotted Wood-Dove Tiirtur chalcospilos (Wagler) 471 Pigmy Kingfisher p/c/a (Boddaert) .... 517 Hoopot Vpupa epops epops lAnndieus ..... 560 Gabon Nightjar Caprimidgus fossil Hartlaub .... 586 Grey-throated Barbet Gymnobucco bonapartei Hartlaub 597 Golden-rumped Tinker Bird Pogoniulus bilineatus (Sundevall) 604 Yellow-billed Barbet Trachylaemiis purpiiralus (Verreaux) . 616 Buff-spotted Woodpecker Campethera iiivosa (Swainson) . 630 Grey Woodpecker Mesopicos goertae (Muller) .... 643 Little Swift Apus affnis (Gray) ...... 691 African Pied Wagtail Motacilla aguimp Dumont 694 Grey Wagtail M. cinerea Tunstall ...... — ■ Yellow Wagtail M. fiava Linnaeus ...... 708 Tree Pipit Anthus trivialis (Linnaeus) ..... 713 Red-throated Pipit A. cervinus (Pallas) ..... 735 Brown llladopsis Trichastoma fulvescens (Cassin) 736 Pale-breasted llladopsis T. rufipennis (Sharpe) .... 737 Scaly-breasted llladopsis T. albipectus (Reichenow) . 738 Mountain llladopsis T. pyrrhopterus (Reichenow & Neumann) 740 Abyssinian Hill-Babbler Pseudoalcippe abyssinicus (Riippell) 742 Dark-capped Bulbul Pycnonotus tricolor (Hartlaub) 743 White-vented Bulbul P. barbatus (Desfontaines) 746 Bristle-bill Bleda syndactyla Swainson ..... 753 Brownbul Phyllastrephus terrestris Swainson .... 754 Northern Brownbul Ph. strepitans (Reichenow) 757 Smaller Yellow-streaked Greenbul Ph. debilis (Sclater) 758 Fischer’s Greenbul Ph. fischeri (Reichenow) .... 760 Toro Olive Greenbul Ph. hypochloris (Jackson) 765 Olive-breasted Mountain Greenbul Arizelocichla tephrolaema (Gray) 768 Shelley’s Greenbul (Shelley) .... 769 Yellow-bellied Greenbul Chlorociclila flaviventris (Smith) . 773 Zanzibar Sombre Greenbul Andropadus importmus (Vieillot) 774 Cameroon Sombre Greenbul A. ciirvirostriis Cassin . 755 Little Greenbul Eurillas vireris (Cassin) ..... 776 Yellow-whiskered Greenbul Stelgidocichla latirostris (Strickland) 778 Spotted Flycatcher Mitscicapa striata (Pallas) .... 785 Ashy Flycatcher Alseonax cinereus (Cassin) .... 796 White-eyed Slaty Flycatcher Dioptrornis fischeri Reichenow 805 Yellow Flycatcher Chloropeta natalensis Smith 815 Puff-back Flycatcher Rn/w capewiii (Linnaeus) 822 Wattle-eye Platysteira cyanea (Muller) ..... 823 Black-throated Wattle-eye P. peltata Sundevall 824 Chestnut Wattle-eye Dyaphorophyia castanea (Fraser) 825 Jameson’s Wattle-eye D. jamesoni Sharpe .... 829 White-tailed Crested Flycatcher Trochocercus albonotatus Sharpe 831 Dusky Crested Flycatcher T. nigromitratus (Reichenow) 832 Paradise Flycatcher Tchitrea viridis (Muller) .... 833 Red-winged Paradise Flycatcher T. suahelica (Reichenow) 835 Black-headed Paradise Flycatcher T. nigriceps (Hartlaub) . 840 African Thrush Turdus pelios Bonaparte ..... 841 Olive Thrush T. olivaceus (Linnaeus) ..... 845 Abyssinian Ground Thrush Geokichla piaggiae (Bouvier) 849 White-tailed Ant Thrush Neocossyphus poensis (Strickland) 4 1 366 3 22 32 128 3 1 2 3 89 0 0 0 0 0 0 0 2 161 3 0 4,034 11 3 0 0 0 0 0 39 0 0 0 0 0 0 0 0 0 0 0 0 2 8 1 0 0 1 0 1 0 0 0 0 0 0 0 0 1 0 0 0 5 1 545 3 30 44 185 4 1 2 3 89 5 2 4 Page 220 850 Rock Thrush Monticola saxatilis (Linnaeus) ..... 854 Wheatear Oenanthe oenanthe (Linnaeus) ...... 855 Isabeline Wheatear Oe. isabellina (Temminck & Langier) . 883 Whinchat Saxicola rubetra (Linnaeus) 884 White-browed Robin Chat Cossypha heuglini Hartlaub 887 Grey-winged Robin Chat C. polioptera Reichenow .... 889 Blue-shouldered Robin Chat C. cyanocampter (Bonaparte) . 890 Red-capped Robin Chat C. natalensis Smith 892 Snowy-headed Robin Chat C. niveicapilla (Lafresnaye) 893 Robin Chat C. caffra (Linnaeus) 898 Equatorial Akalat Sheppardia aequatorialis (Jackson) 903 Brown-chested Alethe Alethe poliocephala (Bonaparte) 913 Eastern-bearded Scrub Robin Erythropygia qiiadrivirgata (Reichenow) 914 Brown-backed Scrub Robin E. hartlaubi Reichenow .... 915 White-starred Bush Robin steZ/om (Vieillot) . 917 ReAsiMi Phoenicunis phoenicums ..... 921 Nightingale Lusciiiia megarhynchos Brehm ..... 922 Sprosser L. hiscima (Linnaeus) 924 Whitethroat Sylvia communis Latham ...... 925 Garden Warbler S. borin (Boddaert) 926 Blackcap S. atricapilla (Linnaeus) ....... 933 Barred Warbler S. nisoria (Bechstein) 937 Upcher’s Warbler Hippolais languida (Hemprich & Ehrenberg) . 938 Olivaceous Warbler H. pallida (Hemprich & Ehrenberg) . 942 Great Reed Warbler Acrocephalus arimdinaceus (Linnaeus) 944 Reed Warbler A. scirpaceus (Hermann) ...... 945 Marsh Warbler (Bechstein) ...... 946 African Reed Warbler A. baeticatus (Vieillot) .... 947 Sedge Warbler A. schoenobaenus (Linnaeus) ..... 959 Willow Warbler Phylloscopus trochilus (Linnaeus) .... 964 Brown Woodland Warbler Seicercus umbrovirens (Ruppell) 977 Black-collared Apalis Apalis pulchra Sharpe ..... 993 Grey-capped Warbler Eminia lepida Hartlaub ..... 1010 Olive-green Camaroptera Camaroptera chloronota Reichenow 1011 Grey-backed Camaroptera C. brcv/cawi/atfl (Cretzschmar) 1030 Hunter’s Cisticola Cwt/co/o Ai/«rcd Shelley ..... 1045 Tawny-flanked Prinia Prinia subflava (Gmelin) 1049 Banded Prinia P. bairdii (Cassin) ....... 1053 Black-faced Rufous Warbler Bathmocercus rufus (Reichenow) 1054 Swallow Hinindo mstica Linnaeus ...... 1055 Uganda Swallow H. angolensis Bocage 1062 Red-rumped Swallow H. daurica Linnaeus ..... 1068 Sand Martin R/pada dpadfl (Linnaeus) 1069 African Sand Martin R. paludicola (Vieillot) ..... 1074 House Martin Delichon urbica (Linnaeus) 1089 Square-tailed Drongo Dicntrus ludwigii (Smith) .... 1103 Lesser Grey Shrike Lanius minor Gmelin 1112 Red-backed Shrike L. coZ/hho Linnaeus ...... 1125 Tropical Boubou LawZadMi: ae/^ZopZcMS (Gmelin) .... 1128 Black-backed Puff-back Shrike Z)ryoicopH5 cmWq (Shaw) . 1164 Golden Oriole OrZoZMS orZoZMi (Linnaeus) 1 1 84 Violet-backed Starling Cinnyricinclus leucogaster (Boddaert) 1190 Lesser Blue-eared Glossy Starling Lamprocolius cholropterus (Swainson) 1219 Yellow White-eye 5e«e^aZeMsZs Bonaparte .... 1221 Green White-eye Z. vZrens Sundevall 1 223 Kikuyu White-eye Z. kikuyuensis Sharpe ..... 1230 Bronze Sunbird ZVecron/Hfl /:ZZZme«sZj Shelley ..... 1238 Copper Sunbird CZ/Jwym CMprcMi (Shaw) . ..... 1245 Mariqua Sunbird C. Smith 1254 Eastern Double-collared Sunbird C. mediocris Shelley 1263 Scarlet-chested Sunbird CZjaZcow/tra ic«e^aZeHsZs (Linnaeus) 1266 Green-headed Sunbird Cyanomitra verticalis (Latham) 1269 Olive Sunbird C. olivacea (Smith) ....... 1271 Collared Sunbird ZarZ/rep/ci coZZanj (Vieillot) 1281 Green Hylia TZyZ/a prad«a (Cassin) ...... 1300 Grey-headed Sparrow (Vieillot) 5 7 0 23 0 0 0 0 0 0 0 2 0 1 0 5 2 0 3 254 8 1 0 7 6 168 6 0 83 90 0 0 0 0 1 0 1 0 1 218 0 0 26 0 0 0 2 50 0 1 0 0 2 6 0 0 0 1 2 2 2 1 0 0 1 2 Page 221 1 325 Yellow-backed Weaver Ploceus capitalis (Latham) 1335 Dark-backed Weaver 6/co/or (Vieillot) 1337 Spectacled Weaver ocM/or/5 (A. Smith). 1342 Holub’s Golden Weaver Xanthophilus xanthops (Hartlaub) 1 346 Black-billed Weaver Heterhyphantes melanogaster (Shelley) 1 347 Emin’s Weaver Othyphantes emini (Hartlaub) . 1 348 Reichenow’s Weaver O. reichenowi (Fischer) 1 360 Red-billed Quelea Quelea quelea (Linnaeus) 1361 Red-headed Quelea Q. cryt/iropi' (Hartlaub) 1 375 Red-collared Widowbird Coliuspasser ardens (Boddaert) . 1380 Black and White Mannikin Spermestes poensis (Fraser) 1 386 Grey-headed Negro-Finch Nigrita canicapilla (Strickland) 1389 Brown Twinspot C/y/o.vp/zfl wo«/c/r/ (Hartlaub) 1391 Red-headed Blue-bill 5pcr/7!o/7/!o^o r//^cap/7/a (Shelley) 1 399 Abyssinian Crimson-wing Cryptospiza salvadorii Reichenow 1406 Peter’s Twinspot //yporg^oi /7/vco^«//o/w5' (Peters) 1410 Green-winged Pytilia Pytilia melba (Linnaeus) 141 1 African Firefinch Lagonosticta rubricata (Lichtenstein) 1413 Red-billed Firefinch L. icweg-o/fl (Linnaeus) 1422 Fawn-breasted Waxbill Estrilda paliidicola Heuglin 1425 Black-crowned Waxbill E. nonnula Hartlaub 1431 Red-cheeked Cordon-bleu Uraeginthus bengalus (Linnaeus) 1433 Purple Grenadier Granatina ianthinogaster (Reichenow) 1441 Pin-tailed Whydah Vidua macroura (Pallas) 1448 Yellow-fronted Canary Serinus mozambicus (Muller) 1461 Streaky Seed-eater S. striolatus (Riippell) 1462 Thick-billed Seed-eater S. burtoni (Gray) Total Total Palearctic Migrants . Total Number of Species Total Palearctic Species 30 30 0 1 2 2 2 2 0 4 22 22 0 8 1 1 6 6 1 1 1 1 0 2 3 3 0 14 0 2 0 1 3 3 1 1 6 6 6 6 2 2 1 1 0 1 10 10 3 3 0 1 0 1 6,836 14,112 6,320 12,498 95 186 41 48 Table 2 RECOVERIES AND CONTROLS OF BIRDS RINGED IN EAST AFRICA Key to symbols and terms Ring number : Age : Sex where this is in italics the ring has been returned, f.g. — full grown, age uncertain, ad. — adult, at least one year old. pull. — young, not yet able to fly. juv. — juvenile. 1st W. — 1st winter. S — male. $ — female. recovery: + — ^ shot or killed by man. X — found dead or dying. I II — manner of recovery unknown. V — caught or trapped and released with ring. ( ) — caught or trapped alive and not released, or released but with ring removed. A recovery in the strict sense is a ringed bird found dead, whether by the ringer himself or re- ported by a member of the public; a control is a bird ringed by one ringer and retrapped by another. Red Knobbed Coot FuUca cristata D.0587 ? 6.4.65 Ngorongoro Crater, Tanzania. 3°12'S., 35°30'E. JG -b 16.12.67 Lessos Dam, Kenya. 0°12'N., 36°16'E. Little Ringed Plover Charadrius dubius A.3194 1st W. 14.12.67 Lake Nakuru, Kenya. 0°20'S., 36° 06 'E. DIP X 26.5.68 wear Sukhumi, Georgian SSR, U.S.S.R. 43°00'N., 41 °01 'E. Ruff Philomachus pugnax C.0466 f.g.c? 25.3.68 + 17.5.68 B.0900 f.g.$ 6.4.68 + 31.5.68 Yellow Wagtail Motacilla flava J. 4065 f.g. 1.12.66 X (cat) 25.7.67 J. 14759 f.g. 10.2.68 /?/ 24.4.68 J. 4409 f.g. 20.12.66 V. 21.10.67 J. 4440 Ad.c? 7.1.67 V. 10.12.67 J. 5026 f.g. 14.1.67 V. 12.11.67 J. 8085 f.g. 30.3.67 V. 17.12.67 J. 8390 Ad.c? 8.10.67 V. 12.1.68 J. 8705 f.g. 5.11.67 V. 12.11.67 J. 8759 f.g. 12.11.67 V. 25.11.67 J. 8994 f.g. 1.12.67 V. 27.1.68 J. 12503 f.g. 27.12.67 V. 14.1.68 J. 12450 Ad.(? 29.12.67 V. 23.3.68 J. 11801 f.g. 6.1.68 V. 21.1.68 J. 4466 Ad.cJ 7.1.68 V. 10.12.67 Lake Nakuru, Kenya. DJP Tyumen Region, 200 km. NE of Surgut, U.S.S.R. 62°50'N., 73°00'E. Lake Nakuru, Kenya. GCB Krasnoyarst Region, wear Pirovskoe, U.S.S.R. 57°38'N., 92°15°'E Kabete, Kenya. 1°16'S., 36°43'E. GCB. Oktyabr’sky District, Tatarskaya A.S.S.R., U.S.S.R. 54°24'N., 50°47'E. Kabete (at roost). EDS near Astrakhan, Kalmytskaya A.S.S.R., U.S.S.R. 46°24'N., 48°02'E. Kabete. GCB Eastleigh. ri6'S., 36°51 'E 10 miles ESE. Eastleigh. GCB (M./. lutea). Kariobangi. 1°15'S., 36°53'E. 3 miles ENE. Eastleigh. GCB Kariobangi. 3 miles ENE. Eastleigh. JBS Kariobangi. 3 miles ENE. Eastleigh. GCB (M./. flava). Kabete. lOi miles WNW. Eastleigh. GCB Kariobangi. 3 miles ENE. Kariobangi. GCB Eastleigh. 3 miles WSW. Kariobangi. GCB Kabete. 12i miles W. Dandora Swamp. 1°16'S., 37°00'E. DAT Kabete. c.l5i miles W. Kabete. GCB (M. /. flava). Eastleigh. 8 miles ESE. Eastleigh. LC Kariobangi. 3 miles ENE. Eastleigh. GCB (M. /. flava) Kariobangi. 3 miles ENE. In addition the ringing details of the following bird, recovered in the year 1966-67, have been received : Red Knobbed Coot Fulica cristata D.0503 ? 6.4.65 Ngorongoro Crater, Tanzania. JG + 16.1.67 East side of Lake Naivasha, Kenya. 0°45 'S., 36°23 'E. RETRAPS A retrap is a ringed bird captured and released by the original ringer, or his associates, at or near (within three miles) the locality where it was originally ringed (Spencer, 1965). An encouraging number of birds have been retrapped from previous seasons; in the last report (Backhurst, 1968) there were only seven such retraps, this year there were 116 including four which are listed in Table 2 as controls. The explanation of this high number is twofold: the number of birds ringed in the 1966/67 season was higher than ever before, i.e. there were far more birds available for retrapping in 1967/68; it is also clear that many birds are faithful to the same winter quarters or that they pass through the same area (where they are ringed and retrapped) on their way to more distant winter quarters. Table 3 BIRDS RETRAPPED FROM PREVIOUS SEASONS Little Stint Calidris minuta ........... Ruff Philomachus pugnax ........... Marsh Sandpiper Tringa stagnatilis Yellow Wagtail Motacilla flava .......... Whinchat Saxicola rubetra Garden Warbler Sylvia borin . .......... Great Reed Warbler Acrocephalus anindinaceus ....... Reed Warbler A. scirpaceus ........... Sedge Warbler A. schoenobaenus .......... Apart from the following exceptions all the above were ringed in the 1966/67 season: Yellow Wagtail Motacilla flava A. 0279 f.g. 15.161 V. 7.10.67 Garden Warbler Sylvia borin J. 1426 f.g. 31.3.66 V. 18.12.66 V. 6.4.67 V. 12.11.67 Reed Warbler Acrocephalus scirpaceus J. 1406 f.g. 18.3.66 Gala. DIP V. 31.3.66 ditto. V. 19.1.68 ditto. Gala, Uganda. ri8'N., 31°49'E. DJ ditto. ditto. ditto. Sedge Wabler A. schoenobaenus J. 1401 f.g. 18.3.66 V. 25.2.67 V. 28.12.67 Gala. DJP ditto. ditto. 5 3 1 75 1 5 3 14 9 KEY TO RINGERS’ INITIALS IN LIST OF RECOVERIES GCB G. C. Baclchurst EJB Miss E. J. Blencowe EC Mrs. L. Campbell JG J. Goddard DJP D. J. Pearson IBS J. B. Smart EDS E. D. Steel DAT D. A. Turner D. Caffyn R. Douthwaite D. P. Ebbutt A. D. Forbes-Watson M. D. Ford OTHER RINGERS IN EAST AFRICA W. P. Langridge J. M. Locke A. Sinclair R. Stjemstedt F. J. Thompson R. J. Wheater Table 4 RECOVERIES IN EAST AFRICA OF BIRDS RINGED ABROAD This list contains some birds which were recovered before 1st July 1967; it is hoped eventually to bring Eggeling’s (1951) list up to date by publishing foreign-ringed recoveries in these annual reports. Page 224 The signs and symbols are the same as those used in Table 2. Cattle Egret Bubulcus ibis Pretoria 635/10887 pul. —.12.62 ( ) 20.6.67 Pretoria 553/3272 juv. 13.12.57 + 31.3.60 Pretoria C. 8698 juv. 17.2.57 + 25.8.58 White Stork Cieonia ciconia Pretoria C. 1010 juv. 3.12.61 + 25.3.62 Hiddensee 200 785 juv. 7.7.64 + 25.3.65 Hiddensee 3 569 juv. 13.7.66 + 19.1.67 Hiddensee 3 559 juv. 10.7.66 + 0.5.67 Hiddensee 3 571 juv. 13.7.67 X 27.6.67 Hiddensee 202 251 juv. 24.6.67 + 17.2.68 Hiddensee 200 329 pull. 24.6.64 X (predators) 13.3.65 Varsovia B530.678 pull. 21.6.56 X 16.1.63 Moskva A. 67.216 pull. 28.6.61 X 9.2.63 Moskva B. 67.306 pull. 25.6.59 /?/ 15.3.60 Shoveler Anas clypeata Moskva E. 581.839 f.g. 14.8.61 + 10.3.63 Gumtree Dist., near Ficksburg, O.F.S., S. Africa, 28°51'S. 27°43'E. Pakwach, Uganda. 2°28'N., 31°30'E. Faithful Fountains, S. Africa. 33°40'S., 26°30'E. Tunduru, Tanganyika. 11°20'S., 37°20'E. Westdene Pan Benoni, S. Africa. 26°12'S., 28°18'E. Ngudu, Kwimba, Tanganyika. 2°40'S., 33°30'E. Bredasdorp, S. Africa. 34°35'S., 20°00'E. Tunduma, Tanganyika. 9°00'S, 33°00'E. Bleyen, Seelow, Frankfurt/Oder, East Germany. 52°35'N.. 14°37'E. Aboki, near Lira, Lango Dist., Uganda. 2°22'N., 32°42'E. Fienerode, Genthin, Magdeburg, East Germany. 52°21 'N., 12°10'E. West Kilimanjaro, Tanganyika. c.3°S., 37°20'E. Zitz, Brandenburg, Potsdam, East Germany. 52°20'N., 12“20'E. Acholi District, Uganda. c.3°N., 32°30'E. Fienerode, Genthin, Magdeburg, East Germany. 52°21 'N., 12°10'E. Kitale, Kenya. 1°N., 35°E. Bendelin, Perleberg, Schwerin, East Germany. 52°55'N., 12°10'E. Ligera, Tanganyika. 10°59'S., 36°23'E. near Genthin, East Germany. 52°30'N., 12°12'E. Timau, Kenya. 0“05'N., 37°14'E. Zbietka, wear Wagrowiec, Poland. 52°44'N., 17°18'E. Sotik, Kenya. 0°40'S., 35°08'E. near Chernigov, U.S.S.R. 51°33'N., 31°20'E. Yatta Plateau, north of Ithanga Hills, Kenya. c.l°25'S., 37°30'E. Beloviesha Reserve, U.S.S.R. 52°40'N., 24°00'E. Seronera, Serengeti, Tanganyika. 2°16'S., 34°47'E. Asrakhan Reserve, U.S.S.R. 46°25 'N., 49°05 'E. Lake Naivasha, Kenya. 0°45'S, 36°25'E. African Pochard Aythya erythrophthalma Pretoria C. 5020 ad. 6.12.53 Modder East Dam, Transvaal, S. Africa. c.25°20'S., 30°E. /?/ -.11.54 Lake Naivasha, Kenya. 0°45'S., 36°20'E. Lesser Spotted Eagle Aquila pomarina Tartu 15270 pull. —.7.54 Aegviidu District, Estonian S.S.R., U.S.S.R. c.59°16'N., 25°36'E. + — .3.55 Geita District, Tanganyika. c.2°35'S., 32°56'E. Page 225 Little Stint Calidris mimta Pretoria 601/44213 1.12.66 V. 25.3.68 V. 6.4.68 Swallow Hirundo rustica London HH. 55.749 juv. 12.9.66 () 5.12.66 Pretoria 601/24345 f-g. 8.4.67 15.5.67 Pretoria 601/01716 f-g. 27.2.66 () 15.10.67 Pretoria 601/07278 f-g. 20.3.66 () 18.10.67 Pretoria 662/02868 f-g- 2.2.67 14.10.67 V. 23.10.67 Pretoria 601/47863 f-g- 6.2.66 () 27.10.67 Blue Lagoon, Kafue Flats, Zambia. 15°27'S., 27°24'E. Lake Nakuru, Kenya. (Original ring removed, replaced by Nairobi A. 3376). Lake Nakuru. Bedworth, Nuneaton, England. (At roost) 52°29'N., 1°28'W. Anaka Paromo, near Gulu, Uganda. 2°41 'N., 32°27'E. Vischgewaard, S. Africa. c.26°10'S., 28°E. Tororo, Uganda. 0°42'N., 34°11'E. Rosherville Dam, S. Africa. c.26°S., 28°E. Busia, Kenya. 0°25'N., 34°15'E. Rosherville Dam, S. Africa. Tororo, Uganda. Escom Dam, Kimberley, S, Africa. 28°45'S., 24°46'E. Busia, Kenya. Tororo, Uganda. Rosherville Dam, S. Africa. Amukura, Kenya. 0°36'N., 34°16'E. REFERENCES Backhurst, G. C., 1968. “Report on Bird Ringing 1966-1967.” /. E. Afr. nat. Hist. Soc. 27: 61-67 Eggeling, W. J., 1951. “Ringed Birds recovered in Uganda”. The Uganda Journal 15: 17-25. Mackworth-Praed, C. W. and Grant, C. H. B., 1957. Birds of Eastern and North-Eastern Africa. London, Longmans, Green & Co. Spencer, R., 1965. The Ringers Manual. Tring, The British Trust for Ornithology. Vaurie, C., 1959 & 1965. The Birds of the Palearctic Fauna. Two volumes. London, H. F.& G. Witherby Ltd. {Received 15th August 1968). \tr-' iwH®? • r^\ > Page 227 A NOTE ON THE POSSIBLE REPRODUCTIVE STRUCTURES IN KENYAN UDOTEA ORIENT ALIS A. & E. S. GEPP, (CHLOROPHYTA). by Shakuntala a. Moorjani, University College, Nairobi INTRODUCTION The genus Udotea is pan-tropical with extra-tropical (Natal Coast) extensions. Udotea orientalis A. & E. S. Gepp is an Indo-Pacific species. It has been reported from Kenya (Gerloff, 1960; Isaac, 1967), Mosambique and Natal coast of South Africa (Isaac, 1956). The Siboga Expedition records include Zanizbar in the Indian Ocean list; in the Pacific Ocean it has been reported from Queensland (Australia), China Sea and Japan (A. & E. S. Gepp, 1911). U. orientalis is widespread along the Kenya coast but it is generally less common than U. indica A. & E. S. Gepp. A third species U. flabellum (Ell. & Soland) Howe has been recorded for the Kenya coast (Isaac, 1967), but so far has only been found in the Lamu region (personal communication). U. orientalis is primarily an alga of quiet or protected waters such as those to seaward of mangroves and in lagoons. It is sometimes found in more exposed situations but not so much as U. indica. The morphology of Udotea is well known and has been fully described by Gepp (1911). There is, however, no certain information and few published records of the reproductive structures, and hence it is worthwhile recording the observations made on the Kenya material of U. orientalis. THALLUS STRUCTURE (a) External morphology The thallus of U. orientalis consists of a well diSerentiated stipe and frond (fig. la). The stipe arises from a bulbous rhizoidal mass and may be up to 2.5 cm. long and up to 3 mm. thick. The frond is well developed, broad and cordate in shape. It varies in size according to the habitat. Generally, those growing in quiet waters are larger than those growing in disturbed waters. The frond may be up to 6.5 cm. broad and 4 cm. long. It is concentrically zoned and longitudinally striate due to moderate calcification. The margin of the frond may be entire or lacerate. (b) Internal structures The frond is built up of smooth-walled coenocytic filaments which radiate from the stipe to the margin of the frond. The filaments are repeatedly dichotomously branched and show constrictions at markedly unequal distances above the points of dichotomy (fig. lb). This is a characteristic feature of this species. The filaments vary from 24p to 35p in diameter and are packed with starch granules and chloroplasts. POSSIBLE REPRODUCTIVE STRUCTURES Phillips reported a small ovoid body at the apex of one filament of Atlantic species U. cyathiformis Decaisne, and suggested a similarity between it and the female gametangium of Codium (Phillips, 1957). Phillips observed a wall at the base of the organ. The wall was thinnest in the middle, where he also observed cytoplasmic strands between the filament and the organ. He concluded that the wall was newly formed. He ruled out the possibility that this organ might be a normal growth phe- nomenon, with the other new filament of the dichotomy broken oflT, because the point E Page 228 Fig. 1 . Udotea orentalis A. & E. S. Gepp, a, whole plant (x 1) ; b, filament showing dichotomous bran- ching (x 200); c, ovoid body at apex of filament (x 200); d, ovoid body showing thicker wall (x 200); e, filament with ovoid body, showing slight ingrowths of filament walls (x 200). Page 229 of constriction was above the dichotomy and no branching or trace of branching was evident. The organ was dark green and so he concluded that it was not a vegetative growth but a female gametangium. POSSIBLE REPRODUCTIVE STRUCTURES IN KENYA U. ORIENTALIS In October, 1967, during an algology practical class at the University College Nairobi, when the author was examining teased filaments of U. orient alis, she came across an ovoid body at the apex of one filament (fig. Ic). On careful examination and comparison with Phillips’ drawing of a similar stru- cture in U. cyathiformis, very close similarities were observed except that the cyto- plasmic strands mentioned by Phillips were not observed. An extensive search for similar structures was carried out on herbarium and pre- served material. While examining filaments of preserved material of U. orientalis, Isaac B. 27 and 3221 from Mokowe mud flats Lamu, Kenya, in front of mangroves, the author came across another ovoid body at the apex of a filament (fig. Id). The ovoid body measured 105p in length and the greatest width was 30p. The width of the filament bearing the body was 24p. Examination of the body under oil immersion revealed that the basal wall of the organ was complete and also that the wall surrounding the body was thicker than the filament wall. The thickness of the wall surrounding the ovoid body was 4.5p compared to 3p of the filament wall. The basal wall of the body was of the same thickness as that surrounding the body itself. Thus the organ was completely isolated from the filament and hence the body is very likely a mature gametangium. It was also observed that the basal half of the gametangium was packed densely with roundish bodies. Phillips suggests in his paper on U. cyathiformis that it is probable that the basal wall of the organ arises by ingrowth of filament walls at the organ base. During the course of extensive examinations for similar ovoid shaped bodies at the apex of filaments, a body was observed with slight ingrowths of the filament walls as shown in fig.le. These ingrowths were visible under a phase contrast microscope. This observation is in agreement with Phillips’ suggestion of the origin of the basal wall of the organ. ACKNOWLEDGEMENTS I am grateful to Prof. Wm. Edywn Isaac of the Botany Department, University College Nairobi, for his encouragement and assistance in the preparation of this note. LITERATURE CITED Gepp, a & G. S. The Codiaceae of the Siboga Expendition, Siboga Expedition Monograph L x II Leyden, 1911. Gerloff, Johannes, 1960. Meeresalgen aus Kenya, 1, Cyanophyta & Chlorophyta, Willdenowia, 2, 4: 60U627. Isaac, Wm. Edwyn, 1956. Marine algae of Inhaca Island and Inhaca Peninsula. J. S. Afr. Bot. 22 : 171. 1967. Marine Botany of the Kenya Coast /.£■. Afr. nat.Hist. Soc. 36,(114) : 75-83. Phillips, R. C., 1957. “Notes on gametangia in Udotea”. Quart. !. Florida. Aca. Sc. 20 : 253-254. {Received 26th June, 1968) I't ^ *rjh , ( I >» ■ jtifm •lo'/O'i yftj , o / u-J *? >Uo> \n '■ '!^#i i>m; rt^y'Jt? “rlT Jrt3-l>t> Njtt'w',! lUf m^i") fniJ} ,tci«'\'^ >fi)' in i'M wovo an J.-w.ivi', \ '.iO nO 'fd t>ftiivn)fiyM‘ ,'V9*nJ< oijnaftlq h^f^rji^toijti^ twidlii' U.1 .! uu y* -s, linsi'xa n/\ -.m s/iin’/f):. ^ . -i^jldV’ iMi'j^wm fay'rr>« i rr'j,^'>*\‘i>M^ v^-ojJmK" r'-;'.'>'1|'liJ.'. I>nj1 TS il >iii, ’i 5 1, ti(rA9fj^Li j.l^r^ Mo,'o )w{)i<;(,;i •torfJu£ ydl ■ 'j1*.-;i>' iM ij^'.l! t^viu>'«yt» wtT _ • .’ ,>.v()Mnu,kv-' /Pw^'ii;!Ui,a(')J«nd,m5»fifei;;Ui Orti U i'.JbiV^, *'u-'. I / n, [tu/^ b .'-'“/I Od) Lvfit !o!.bV9‘tf i!.' / N‘,)"J5i' '■''! iin:‘5 iy!i-jbl5 vw vluht <^di' imJtTT)-»oiiiu,a Ojtw ■ .,t '/*• )f/> : im '■!•:. a 'I :. .- j>. !''[■’< >un.<’ '.nil >" vj;-'|d ■' '{Iq tli^w )w,(«J -.mII ii' )'l I I l(>gi '<1 a^d(uu‘» ' -"wV/ (w;,.‘i<)i adi r i .,M, I', j ^»t7 ii .*i':.'jl;;ii;a‘>mi.:^'vr;aiifiV « 'ib:<«»( A .'iJiv/ \-i'. ;ob tfit/i fa^44<»tej3fTij>?i , i » "3 lit ryq/iq hU «d 1 f ?ir.-.'"-i } »sri*wfr 'Jo (iJ-’.'M (t;i vd /£5 jjc *^0 ll»;w b '. L'\ -jmoo ' ' " it i' pH'iU ■ I liii ddw Lv9Vi9«'.k‘ ?.fJ'ft' • .'v.'fi'^l r ;;^y(fl' .p{,i?jn fli -I ■•f.Kj'?i/^05^^i >A ll|5i<.(i'Mf )t ^ ft’ iUlUi^),C’. I'iVA'^V’ .^;swlrj>t..';|K> Js( ■«,.-.V -'■(jja iii'i'i noif^''^vji4<1o ..ni^’vO' odi ‘1<> V/'/x^rAVO-l iP'// Jn fTW^/ ^' ■’■ „ > '.'; b .i; . ,.i ,4 Aikn'r//) n:^’Slf0*> ; u ,ii . ^ ^ ' .!(.^ijb'rt>t;'-|[ vj.h W,Ji) ' ■ ■ .' ■ ,\0'^r ' * ;u ,ii\iiXin-/(nfSfi oo »,•; > : . t''-'. ?\'i : '-'5 ,>( oMU.iiifl «''■*■' '-3 .w>A 'i\i(Vf/te}'iv. A NINETEENTH CENTURY REFERENCE TO THE USE OF TOOLS BY THE EGYPTIAN VULTURE by R. H. Baxter, S. K. Urban and L. H. Brown A recent note by the van Lawick-Goodalls (1966) has described, and illustrated with remarkable photographs, the manner in which the Egyptian Vulture, Neophron perconopterus (Linnaeus), breaks Ostrich eggs by lifting stones in its beak and throw- ing them at the eggs. In this connection the following account (Wood, 1877), published more than ninety years ago, is perhaps of interest : “Two articles of diet which certainly do not seem to fall within the ordinary range of vulture’s food are said to be consumed by this bird. The first is the egg of the ostrich, the shell of which is too hard to be broken by the feeble beak of the Egyptian Vulture. The bird cannot, like the lammergeier, carry the egg into the air and drop it on the ground, because its feet are not large enough to grasp it, and only slip off its round and polished surface. Therefore, instead of raising the egg into the air and dropping it upon a stone, it carries a stone into the air and drops it upon the egg. So at least say the natives of the country which it inhabits, and there is no reason why we should doubt the truth of the statement. The other article of food is a sort of melon . . .” It seems Ukely that the author had heard or read a first or second hand account of the behaviour described by the van Lawick-Goodalls but that, remembering the behaviour of the Lammergeyer, he had assumed that the stone was lifted in the claws rather than in the beak. That he did interpret his information in this way is clearly shown by a later passage (p. 525) in which he says ; “The scattered eggs (of the Ostrich) . . are often eaten not only by beasts, but by birds of prey; the former breaking the shells by knocking them against each other, and the latter by picking up larges stone in their claws, rising above the eggs, and dropping the stones on them”. However, it is interesting that there have been a few reports based on information provided by aborigines, and one on actual observation by a European, that the Australian Black-breasted Buzzard Hawk, Hamirostra melanosternon, breaks Emu eggs in precisely this manner (Chisholm, 1954); perhaps further observations on the Egyptian Vulture may show that it, too, sometimes breaks eggs in this way. REFERENCES Chisholm, A. A., 1954. The use by birds of “tools” or “instruments” Ibis 96 : 380-383. VAN Lawick-Goodall, J., and van Lawick-Goodall, H., 1966. Use of tools by the Egyptian Vulture Neophron percnonptems. Nature 212 (5069): 1468-1469. Wood, J. G., 1877. Wood’s Bible Animals J. W. Lyon & Co., Guelph, Ontario. {Received 3rd June, 1968) ■ : . .. itv^ ». . . ' . ^. ■TKiit'? - ■ " j:>H-i»*a;Hf^aOiiVon iiT»qa5mKi^,is< ' '.« ' ;■ nM i*mw’aHT '/« n^ioofr ^ Mw««i‘.>! ,^4'i . Ki t'fni'ri* ,u*j'M-.i'.<»» ^,^'<^rs''''';‘i: -7-;-- " Mf Vi V6 ua>a^t* ’>mt "wa» u 01 iVv,r !',• “««fSi!S'¥*s.“j;,r£ff ■ ' ■ ■ .fwmthfn ^>1 » t 11 ^ ' ^ ^ iUi£i£ n5j>4i^^< f)H ’<:f4 ij^Vn ' .. ^ ^--.i A . .. Jt:- Page 233 THE AFRICAN PITTA Pitta angolensis, Vieillot by V. E. M. Burke This brilliantly coloured bird is rarely seen. Most of the records are made as the result of it flying into a light at night while on local migrations. As the records are few, the following observation and summary of what Moreau, Praed & Grant have to say may be of interest. An African Pitta was brought to me by the night watchman of the Mwese Hospital, Tanzania at about 8 o’clock in the evening of the 20th November, 1967. The bird had flown against the window of a lighted room. There was a spot of blood at the base of its beak but it was otherwise unhurt. I kept it in a basket overnight and next morning, placed it on open ground and pointed it due north, believing at the time that this was the way it ought to have been going. The bird paused for a moment as if taking its bearings, leaped into the air, swung round and flew off" due south with a a low, direct and fast flight. Mwese lies 6°S of the Equator, about 30 miles east of Lake Tanganyika and mid- way between Kungwe (or Kungu) Mountain and the township of Mpanda. Mwese is a hill about 200 square miles in area which rises to 6,000 ft. like an island surrounded by Brachystegia, Miombo bush. The area is burnt annually and the dominant trees are Combretum and Protea with narrow strips of forest along the rivers. The rainfall probably averages about 40 inches per annum and normally starts in November and ends in March. In 1967 there had been considerable rain by the 20th November when the bird was found. Praed & Grant, in The African Handbook of Birds, Series I. describe the general distribution of the African Pitta as Central Tanzania to the Transvaal and, in the non- breeding season, to the eastern Congo, Uganda and Kenya. They go on to say that the bird seems to breed from the Transvaal up to central Tanzania from September to March and to migrate in the non-breeding season to northern Tanzania, Uganda and southern Kenya. One must suppose from this background that the Pitta seen in central Tanzania flying south on the 21st November, 1967 was returning, a little late, to its breeding area from its sojourn in the north. Moreau, in The Bird Faunas of Africa and its Islands uses the Pitta as an ex- ample of migration within Africa. He indicates that its breeding range is from South- ern Tanzania to Malawi and Rhodesia and that its favourite breeding habitat is dense semi-deciduous thicket at low altitudes. Most of the records he quotes are of birds flying into lighted windows at night and he describes the occasion, when he captured a Pitta in such circumstances, as “One of the thrills of my ornithological life. . . to put my hands on this gorgeously coloured bird, unhurt as it was”. The records he quotes are all of Pitta found between Abercorn in Zambia and the forests of Uganda, that is from 8°S to about 2°N and as there are no breeding records in this area, it is assumed that these are the limits of the migratory range. The most western of the records are from Lake Kivu in eastern Congo Republic and the most eastern are from the Tanzania coastal belt. On the assumption that the migration are on a north-south axis, I have divided these records longitudinally, by the line of latitude of 35°E, i.e. Uganda, Western Tanzania and Zambia to the west of the line, and Kenya and eastern Tanzania to the east. To the west of the line, the records, including my own, follow the Western Rift Valley. Abercorn, 8°S, at the south end of Lake Tanganyika, has six records, all in the month of December. My own record from Mwese, 6°S, is for the month of November. There are two dates mentioned for Lake Kivu, 2°S, in April and May. At the northern end of the range in the Uganda forests, say 2°N, Moreau quotes records in May, June and July. I interpret these records as indicating that breeding is south of 8°S and from De- cember to March, which coincides with the rains in Zambia, Malawi and southern Tanzania; that the birds move north up the line of the Western Rift Valley in April and May and spend June to October in Uganda and return down the same route in November and December. The records in the eastern half of the area are from four places. The birds are recorded as breeding in the extreme south of Tanzania, 10°S; there is a single record of a bird from the Itigi thicket, 8°S, mid-way between Tabora and Dodoma, in the month of December, but it is not revealed whether the bird was breeding or on mi- gration; there are then 10 records all for May and June from north-eastern Tanzania, 5°S, including 5 of Moreau’s own records from Amani; finally there are two records from Nairobi, 1°S, for May and June. My guess is that the Itigi bird, in December was either going south to breed or actually breeding at Itigi, and that the North-Eastern Tanzania birds and the Nairobi birds were all going south, but where they were going I do not know as there are no records from the Kenya highland forests or from the coast. It is possible that this beautiful bird is bestowing its delightful compliment of flying through bed-room windows more frequently than is generally known. More records of its movements would be very interesting. {Received 3rd June, 1968.) Page 235 NATURE NOTE Notes on the life history of Aiteta veluta Hamps (Lep., Nocutuidae) The larva of this species is a dark brown to black over almost the entire surface, with a black head. Segments 2-4 are enlarged to form a rounded lump just behind the head. This lump has a greenish colour over the most expanded part grading to black round the base of the lump. There is a small triangular hump on the last segment above the claspers, rather like that in some of the Notodontidae. Following the shape of this hump, on each side of it, there is a short white line. The larva is about 1^ inches long when full grown. The foodplant is Markhamia hildebrandtii (Baker) Sprague. The chrysalis is formed in a cocoon of dirty yellowish-white silk which is more or less the shape of the pupa of Papilio machaon L. of Europe. It is about 1 inch in length with a prominent pointed bump over the thorax of the chrysalis. The larva started to spin up on 14th March 1969 and emerged on 28th March 1969. 1st April, 1969. M. P. Clifton, Entomologist. BOOK REVIEWS BACKGROUND TO EVOLUTION IN AFRICA Edited by Walter W. Bishop and J. Desmond Clark, published by the University of Chicago Press, 1967, pp. x+935. U.K. price 247/- This large book is the proceedings of a symposium: “Systematic Investigation of the African later Tertiary and Quarternary” held at Burg Wartenstein Castle, Austria in July and August 1965 and financed by the Wenner-Gren Foundation for Anthropological Research, New York. The implicit claim of the title to cover all evolution is of course misleading; instead the book’s declared object is to review Quarternary and later Tertiary research in Africa over the past decade, and to ini- tiate the revision of existing geological and archaeological sequences, terminologies and methods. One subject which is not dealt with is the evolution of man himself, but all else which forms the relevant background to this topic is covered. The book is divided into three parts: 184 pages of palaeontology, 220 of stratigraphy, and 464 of archaeology, followed by a list of recommendations in English and French and a post-conference appraisal. Eight of the forty-three papers are in French with English summaries, the remainder are in EngUsh. Each paper is followed by a discussion in Enghsh. The philosophy of the book is stated by R. F. Flint on p. 189: that “fol- lowing a code forces closer examination and more accurate description of the strati- graphy at any place, and of changes from place to place. Hence it improves correlation and further illuminates the inferred historical events”. The precept applies as much to palaeontology and archaeology as to stratigraphy. Advances in knowledge and understanding can come from such codification as well as from entirely new dis- coveries, and in fact new discoveries presented in isolation can sometimes increase confusion. Some of the authors were asked to draw up formal provisional charts for the stratigraphy of the more important site areas. These charts separate clearly the climatic, chronological and faunal aspects of stratigraphic nomenclature from the more basic rock units themselves, and will be one of the factors ensuring the book a longer period of usefulness than might be expected for the proceedings of a symposium. Archaeology in particular suffers from the confusion of cultural-stratigraphic with time-stratigraphic terms, and the archaeologists brought a selection of material to the conference, so as better to reach agreement on a unified nomenclature of forms and types of stone tools and on a terminology of the techniques used in their manu- facture. Their conclusions have been included in the recommendations, which should Page 236 have a beneficial effect if followed by future authors, especially as the archaeologists deliberately stopped short of trying to impose one overall system on their subject. One may hope that the necessary money and enthusiasm will be forthcoming for future work. Mammalian palaeontology is not sufficiently advanced for the const- ruction of faunal zones with their attendant problems of nomenclature and pan- African correlations, but one paper draws up a sequence for the Lake Chad area based on elephants alone, and two papers consider the problems in southern Africa. The printing of the main points arising in discussion of each paper is an excellent feature of the book, helping to show the untidy miasma of one-sided conceptions through which factual data are interpreted, which opposes as well as fosters the processes of codification, and which contrasts with the disciplined presentation of good scientific papers. Such discussions are often unavoidably disjointed, but contain many interesting and even amusing scraps of information. This is not a book for the general reader, but for anyone who wishes to be more than a general reader in the subject it is as good and up-to-date a starting point as any available. The post-conference appraisal would be a good place for such a reader to make his approach. A. W. G. on the FLORA OF TROPICAL EAST AFRICA A further series of this Flora has now been received from the publishers. The Crown Agents for Overseas Governments and Administrations. Unless otherwise stated these were published on the 28th June 1968, U.K. prices are quoted. The families are: — Aquifoliaceae by B. Verdcourt, 4 pp.-f-l fig. .Srcx/accoc by B. Verdcourt, 3 pp. + l fig. Caprifoliaceae by B. Verdcourt, 3 pp. + l fig. Cucurbitaceae by C. Jeffrey, 156 pp.-l-26 figs, published on the 28th June 1967. Dilleniaceae by G. LI. Lucas, 6pp.-t-l fig. . Dipsaceae by D. M. Napper, 11 pp.+4 figs Elatinaceae by B. Verdcourt, 5 pp.-fl fig. Icacinaceae by G. LI. Lucas, 17pp.-f-6 figs. Monimiaceae by B. Verdcourt, 33 pp.-f 1 fig. • Orchidaceae, Part 1 by V. S. Summerhayes, 235 pp.+41 figs, published c 29th July, 1968 by B. Verdcourt, 8 pp.+ 3 figs. . Sapotaceae by J. H. Hemsley, 78 pp.-bl3 figs. . Scytopetalaceae by B. Verdcourt, 3 pp. + l fig. . Sphenocleaceae by FI. K. Airy Shaw, 3 pp. + l fig. Valerianaceae J. O. Kokwaro, 9 pp.+2 figs. ... The Cucurbitaceae, Orchidaceae and the Sapotaceae are large families, the rest of the parts are composed of small families containing only one or two genera with few species. The Cucurbitaceae deals with twenty-eight genera with one hundred and twenty- six species, most are herbaceous climbers and provide gourds, edible fruits and sponges, the loofah; some are extremely poisonous. An interesting one is Cucumis humifructus Stent, which actually buries its fruits in the ground and has been recorded from Nakuru, near Nairobi, then from the S.E. Congo, Zambia, Rhodesia, Angola and South Africa. The Sapotaceae are mostly trees, some of economic importance from the timber point of view but the genus Butryospermum has an edible fat of economic importance 1/- 1/- 1/- 18/- 1/6 1/6 1/- 2/6 1/- 28/- 1/6 8/6 1/- 1/- 1/6 Page 237 in West Africa and to a lesser extent in Uganda and is called Shea butter which is extracted from the ripe seeds. In East Africa this family is represented by thirteen genera and forty-five species. The Orchidaceae (Part 1) covers all the terrestrial orchids recorded in East Africa. It has seventeen genera with four hundred and forty-three species, the largest genus being Habenaria with one hundred and thirteen species. All the parts have keys to the genera and species and are well illustrated with line drawings and the high standard of production is maintained and the eight authors are to be congratulated on the completion of these fifteen families of the Flora of Tropical East Africa. The parts are obtainable from the Government Printers in East Africa or from the Government Bookshops in the Provinces or London in the United Kingdom. P. J. G. A GUIDE TO SOME OF EAST AFRICA'S UPLAND FLOWERS Kenya Shell Ltd. are to be congratulated on the publication of another Guide, this one on some of East Africa’s Upland Flowers, a thirty-four page book of well produced coloured photographs of some of our very attractive flowers. There are sixty of them taken by Mr. R. Fulton who also supplied the native tribal lore about them and they cover trees, shrubs and herbs. The photographs were identified by the East African Herbarium (E.A.A.F.R.O.) and Mrs. Tweedie, and each is labelled with its correct scientific name, its native names, as well as the English name if one is available, and a brief description about it and its native uses and in many cases the altitude at which it grows. The cover is a full page photograph of a Gloriosa flower and the back one has three flowers of the “Black-eyed Susan,” Thmbergia alata. The reproduction of the photographs is very good and it is not easy to make a selection from them ; I particularly like the Clematis and Disa photographs on page 18 and opposite page 22 there is a very fine one of a water-lily flower, Nymphaea caerulea in its natural habitat. It is on sale for Shs. 6/- and is well worth the money. It is to be hoped that further guides to the wild flowers of other areas of East Africa will appear and give to those interested an idea of the very interesting and rich flora by which we are surrounded. P. J. G Page 238 ORCHIDS OF EAST AFRICA By Frank Piers, Verlag Von M.D., F.L.S., J. Cramer 1968, 304 pages with 116 black and white photographs, a map and two coloured plates. Price £8. 5. 0 in paper covers. This is the second and fully revised and enlarged edition of a book first produced in 1959 in cyclostyled form and of very limited distribution. It is clearly stated in the preface that this new printed version is still intended as a practical guide for the naturalist, the orchid lover in particular, and even the intelligent tourist and makes no claim to be a scientific treatise. Those who want something more comprehensive can, for many of the ground orchids, already refer to the first volume of the orchids for the new Flora of Tropical East Africa and continue to look forward in hope or possibly despair to the subsequent volumes which will complete the family. Those interested in orchids in Uganda, and even in parts of Western Kenya and Western Tanzania will also be able to refer to Volume III Part 1 of the new edition of the Flora of West Tropical Africa for accounts of several species not to be found in this new book. A great value of the new edition of Piers’ book lies in the fact that there is nothing else that one can buy that covers the same ground. The amateur botanist will be glad to have descriptions of the plants that he can understand and use without recourse to a microscope. Nine times out of ten however orchids discovered in the field are not in flower and how useful it would have been to have some kind of field key based on environment and vegetative characters to enable provisional naming before flowering, though, if one already recognises the genus, the species descriptions will often make this possible. Plant hunters will sooner or later find species they are unable to identify from this book and one sad omission from the second edition is the listing of additional species and even a few genera not described in the text which often, using the earlier edition, enable one to run down less common plants such as Zeuxine elongata or Microcoelia microglossa which could both easily be identified from the Flora of West Tropical Africa once one had a lead to the names. One suspects that the coverage of this book is very much more thorough for Kenya than for the other two countries. In Kenya there are even descriptions of several forms awaiting names at Kew. In contrast to this treatment of Kenya orchids, several quite well known Uganda species such as Cyrtorchis injoloensis which is illustrated, however 1 believe wrongly, on page 261 above the name Cyrtorchis ringens, and Diaphananthe ugandensis from Kigezi, are not mentioned at all, while otW recently discovered rarities are alas given incorrect localities, e.g. Rangaeris rhipsalisocia, which is in fact only known in Uganda from a single collection in Budongo Forest and Podangis dactyloceras which so far as I know has never been found at or indeed anywhere near Entebbe and Genyorchis pumila which has only been found from a single tree, but not in Mpanga Forest, by Mr. Hermann Meyer. The first East African finding of Poly- stachya affinis was not by Mr. Leakey in fact but by Mr. Kamya of the Uganda Forestry Department. A number of well known Kenya orchids which are quite common in Uganda on the three eastern volcanic mountains Elgon, Kadam and Moroto, including Rangaeris amaniensis, Diaphananthe quintasii, and Polystachya transvaalensis receive no mention as occurring in Eastern Uganda. A comparison of the species described in Piers’ book from Uganda and those listed as having known Uganda localities in the Flora of West Tropical Africa, for example the genus Ancistroryn- chus (A. recurvus, A. metteniae, A. capitatus, A. straussii are all recorded for Uganda in addition to the well known A. ovatus) will make the relatively scant treatment of this area apparent, though one of the finest of these western species found in Uganda Tridactyle gentilii (de Wild.) Schltr. escapes mention for East Africa in the Flora of West Tropical Africa. Page 239 The account of Nephrangis filiformis appears possibly to embody a confusion between two quite unrelated species. I am not closely familiar with the distribution of this orchid in Kenya, but in Uganda it occurs only in the West, being nowhere com- mon. It is foundin the branches of the large trees in the canopy of Budongo and Kalinzuh Forests and occurs quite commonly at about 7,000-8,000 ft. on Cupressus trees beside a main road in Kigezi. On the other hand Tridactyle tridentata (var. subulifolial) which^has a superficially rather similar vegetative morphology and pendulous habit occurs very commonly and often in large clumps on large trees in areas near Lake Victoria in Kenya and Uganda in just the way that Piers describes for Nephrangis. Incidentally another terete leaved Tridactyle, T. teretifolia, is also quite common near the Western shores of Lake Victoria both North and South of the Tanzania border and is known to extend as far West as Kikagati near the Uganda-Rwanda border. The author has based his book on his first hand knowledge as far as possible and this makes the coverage satisfactory only in the areas where th^e author is familiar with the orchid flora. It would seem that either more personal visits or more recourse to herbaria or more consultation with local botanists, or all of these, could have consider- ably improved the value of the book for areas beyond Dr. Piers’s normal safari circuit. Nevertheless even at £8 .5 .0 anyone desiring to become familiar with East African orchids in the field should buy this book and by the time he senses its inadequacies too strongly it will already have achieved its purpose and the owner will then be using the Floras and Herbaria and not having to rely too much on this very useful introductory manual. On page 14 there is a selective list of East African orchids recommended for cultiva- tion and suggestions for their growth in Tropical, Intermediate or Cool conditions, with an added class of “difficult” species. The choice of the species is of course a personal one, though why the very drab Bulbophyllum cochleatum should be included is hard to understand. The arrangement of the classes of growing conditions has some peculiarities and bears little very obvious relation to the comparative ecology of many of the species in the wild nor in fact to their needs in cultivation. Eor example Eury- chone rothchildiana, which grows in deep shade near water, in middle altitude forests on very few host species, if given suitable conditions for say Aerides odorata or Angraecum giryamae would assuredly die within days if not hours. While from a single tree near Entebbe I have collected growing together species from all four lists i.e. Angraecum infundibular e, Rangaeris brachyceras Anragecopsis gracillima and Cirropetalum umbel- latuml It is very difficult to classify orchids satisfactorily by growing conditions. Some species will stand or adapt with care almost any conditions while others are far more demanding, particularly of humid conditions. Piers sometimes gives the altitude of known wild habitats of the plants — this is perhaps the most useful guidance to conditions for cultivation, provided that the information is read with adequate geographical background knowledge. The book is nicely produced and printed and the photographs are useful and quite good, as indeed are the descriptions and accounts of most of the 271 spceies covered. In bringing this book to print and sharing his knowledge with us all Dr. Piers has made a notable contribution and earned our gratitude. C.L.A.L. Printed by The English Press Ltd., P.O. Box 30127, Reata Road, Nairobi lOURNIL OF THE Em AFRICi NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM CONTENTS Page Shallow soils and their Vegetation in the Region of Nairobi i By Kishor L. Modha The Distribution of the Genets, Gemtta genettUy G. servalina and G. tigrina in East Africa 7 By M. E. Taylor A New Four-toed Mongoose from Kenya, Bdeogale crassicauda nigrescent, ssp. nov. lO By John B. Sale & M. E. Taylor Bird Ringing Report 196S-1969 16 By G. C. Backhurst Migration of the butterflies Glycesthia aurota, Catopsilia florella and Crenis occidentalum in East Africa in 1967-68 27 By John E. C. Flux Observations on Butterfly Migration at Entebbe, Uganda. 30 By S. D.K. Sempala An Analysis of the Features of Sardinella gibbosa (Bleeker) Scales, with Special Reference to the Problem of Age Determination 35 By W. Okera Notes on Kenya Acetabularia Lamouroux (Chlorophyta) 47 By Shakuntala A. Moorjani The Lugard Plant Collection 53 By E. M. Tzoeedie & A. D.Q. Agnezo Observation on Colonial Breeding in the Black-headed Weaver and Vieillot’s Black Weaver 56 By J. R. Hall Data from a Colony of Vieillot’s Black Weaver, Melanopteryx (Ploceus) Nigerrimus Vieillot 62 By J. R. Hall Nature notes: Crowned Hawk-Eagle Raids Hombill Nest 64 By Allan R. Keith Book Reviews 65 Errata to Vol. XXVII No. 3 (118) January 1969 70 (Published October, 1970) ^ T Price Shs. 25/-. - JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. XXVIII No. 1 (119) JULY, 1970 CONTENTS Page Shallow soils and their Vegetation in the Region of Nairobi i By Kishor L. Modha The Distribution of the Genets, Genetta genetta, G. servalina and G. tigrina in East Africa 7 By M. E. Taylor A New Four-toed Mongoose from Kenya, Bdeogale crassicauda nigrescens, ssp. nov. lo By John B. Sale & M. E. Taylor Bird Ringing Report 1968-1969 16 By G. C. Backhurst Migration of the butterflies Glycesthia aurota, Catopsilia florella and Crenis occidentalum in East Africa in 1967-68 27 By John E. C. Flux Observations on Butterfly Migration at Entebbe, Uganda. 30 By S. D.K. Sempala An Analysis of the Features of Sardinella gibbosa (Bleeker) Scales, with Special Reference to the Problem of Age Determination 35 By W. Okera Notes on Kenya Acetabularia Lamouroux (Chlorophyta) 47 By Shakuntala A. Moorjani The Lugard Plant Collection 53 By E. M. Tweedie & A. D.Q. Agnezo Observation on Colonial Breeding in the Black-headed Weaver and Vieillot’s Black Weaver 56 ByJ. R. Hall Data from a Colony of Vieillot’s Black Weaver, Melanopteryx (Ploceus) Nigerrimus Vieillot 62 By J. R. Hall Nature notes: Crowned Hawk-Eagle Raids Hombill Nest 64 By Allan R. Keith Book Reviews 65 Errata to Vol. XXVII No. 3 (118) January 1969 70 (Published October, 1970) Price Shs. 25/-. EAST AFRICA NATURAL HISTORY SOCIETY President: Mr. J. S. Karmali Vice-President: Mrs. a. L. Cambell Executive Committee: G. C. Backhurst M. P. Clifton Miss A. J. Evans A. Duff-Mackay Mrs. j. Hayes R. A. Leakey Dr. F. a. Mutere Dr. T. R. Odhiambo Miss J. R. Ossent Hon. Editor: Dr. P. J. Greenway Hon. Treasurer: P. G. Archer Hon. Secretary: Miss P. M. Allen Asst. Hon. Secretary Miss B. L. Debenham All correspondence in connection with this journal should be addressed to: The Secretary, East Africa Natural History Society. P.O. Box 4486, Nairobi, Kenya. Page 1 SHALLOW SOILS AND THEIR VEGETATION IN THE REGION OF NAIROBI, KENYA By Kishor L. Modha Uganda National Parks f Nuffield Unit of Tropical Animal Ecology, Queen Elizabeth Park, Uganda. INTRODUCTION Shallow-soil areas are here defined as those where up to 20 cm. of soil covers the underlying rocks (fig. i), and eleven such areas were studied in the Nairobi National Park. The vegetation of these shallow-soil areas is not much disturbed by human ac- tivities such as cultivation and road-making, and is very sensitive to drought; it only grows after the rains, remaining dormant during the dry months of the year. The present study was therefore started soon after the beginning of the short rains in October, 1967. Elongated areas of shallow soils are usually found on the side of a valley or the shoulder of a profile following the contour in the position of maximum slope. In the Nairobi National Park such areas are found mainly in the western wooded part which has many river valleys, but some occur near the Observation Point and near Hippo Pool. Shallow- soil areas are bordered by grassland and woody vegetation on one side or on both upper and lower edges. Sampling Procedure From a random throw within a shallow-soil area two transects were considered, one down the slope and the other up the slope. Along these transects the position of each sample was obtained by stratified random sampling within each consecutive 10 m. length. The number of quadrats in a transect varied from three to nine, depending on the length of the transect. Having found the position of the sample a quadrant of i X i m. was marked. A sample of soil was taken from each quadrant from a depth of about 2-3 cm. for pH determination and the occurrence of every plant species was recorded. The depth of the soil was determined by pushing a metal rod down into the soil until it reached the underlying rock. The depth to which it sunk was measured. In the deeper soils, if the rocky base was not struck by the metal rod the depth was entered as “more than 70 cm.” Identification of plants was aided by the use of Heriz-Smith (1962). Soil Acidity It was found that the soil in general is slightly acidic— pH ranging from 6.4 to 7.0 — with the majority of the sites having pH 6.6 or 6.8. There is apparently no relation between the pH and the soil depth and therefore none between the pH and the vegetation, since the vegetation depends on the soil depth. Page 2 Page 3 Fi^re 2. — Species diagram showing the relationships between species found on shallow and adjacent soils. Page 4 Comparison of samples To find out if there is any significance of correlation among the species, the method adopted by Agnew (1961) was used. The species in each sample are treated on a pre- sence or absence basis which consists of testing the correlation (positive or negative) of every species with every other by means of Chi (X^) square. In this study only positive correlations were taken into account, because of the small number of samples finally obtained. First of all, the data were reduced by eliminating all those species which occurred less than five times. This left 42 species. A table was prepared showing the occurrence of each species with every other. Then, by using 2x2 contingency tables, for each pair a chi-square value was calculated. The probability values were calculated from Fisher & Yates (1953). For visual examination and appreciation of these correlations they were arranged as shown in Fig. 2, which is referred to as the “species diagram” (Agnew, 1961). It is a two-dimensional representation of positive correlations, the best possible arrange- ment being obtained by trial and error and it demonstrates some of the ecological rela- tionships between species. It is clear that some typical plants of the shallow-soil area show high positive corre- lations with each other, e.g. Eragrostis hispida, Sporobolus discosporus, S. stapfianus, Kyllinga leucocephala, K. erecta. Euphorbia rivae, Evolvulus alsinoides, Craterostigma spp., Eriochloa nubica. Cassia mimusoides, and Harpachne schimperi, these plants are ecologically related and always occur together on the shallow-soil area. Woody shrubs and trees show higher positive correlations with each other and so form a group of their own as can be seen from the diagram. Ochna ovata, Grewia similis, Croton dichogamus and Gnidia subcordata are small trees which are found among shrubs like Lippia javanica, Fuerstia africana, Nesaea erecta, Psiadia arabica and grasses like Eragrostis braunii, Hyparrhenia collina and Panicum maximum. These two groups are not connected on the diagram by a common link, so they are apparently ecologically separated groups of plants, and in nature they hardly, if ever, intermingle with each other. Discussion The following is a list of plants generally found in the shallow-soil area. The flora of the shallow soil is peculiar in that it contains a large proportion of poikilohydric species whose leaves dry up but do not die in dry months and become green as soon as they get wet; they flower soon after rain and again dry at the end of the rainy season. This property is an effective adaptation to the environment in these areas, and the poikilohydric species probably evolved in this habitat. They are marked with an asterisk in the list. Anthericum gregorianum Cassia mimusoides Coleus caninus * Craterostigma hirsutum * Craterostigma spp. Cyperus spp. Eragrostis hispida * Eriochloa nubica Euphorbia rivae Evolvulus alsinoides Fimbristylis spp. Harpachne schimperi Ilysanthes pusilla Kyllinga erecta Kyllinga leucocephala Oldenlandia herbacea * Sporobolus discosporus S. stapfianus Trachyandra saltii Page 5 The woodland vegetation consists of small trees from 5-10 m. high. The most characteristic species are Croton dichogamus, Dombeya burgessiae, Erythrococca bongensis, Gnidia subcordata, Grewia similis, Turraea mombassana, Strychnos usambarensis and occasionally Olea africana. Of these Croton dichogamus, Gnidia subcordata, Grewia similis and Turraea mombassana are typically forest-edge trees. The dense ground flora up to about 50 cm. high usually consists of grasses, herbs Mk^Justicia whytei, at times climbers like Rhynchosia elegans, and other dicotyledonous plants. Woodland in some places is closed; in others it is more open. The grassland type of vegetation consists of the following species amongst others : The three types of vegetation often occurred in one transect, which was orientated simply to pass through the shallow-soil area (fig. i), and no attempt was made to sample the vegetation t}rpes separately. The species diagram (fig. 2) shows that there are high positive correlations among the species of the shallow-soil and that these species form a “ring” which is quite separate from the other group of woody plants on the deeper soil. There are no intermediates which link the two types and this suggests that they are ecologically fundamentally separated. The shallow soils studied here were presumably formed by the filling-in of depressions and crevices in bare rock, much of which remains in these area. Further weathering of the exposed rocks should add to the volume of soil, and one would expect more soil to accumulate as pioneer plant species colonize and consolidate it. In a normal succession, as soil depth increases, the pioneer species are gradually and continuously replaced by woody perennials, but in our shallow-soil areas there is apparently a discontinuity between successional stages. They have developed a specia- lized and characteristic flora, whose species are different from those of the surrounding woodland and/or grassland, and which are apparently a permanent and regular feature of East African vegetation. As mentioned above, shallow-soil areas usually occupy positions on the shoulders or sides of valleys, and apparently here erosion is capable of maintaining the soil depth at under about 25 cm. Throughout the study virtually no interaction was observed between the bigger animals and the shallow-soil vegetation. In one place shoots of Anthericum gregorianum and Trachyandra saltii had been nipped off as if eaten by some herbivore, probably a bushbuck, Tragelaphus scriptus (Callas), which was seen grazing in the vicinity. Other plants like Craterostigma spp.. Cassia mimusoides, Evolvulus alsinoides, llysanthes pusilla. Euphorbia rivae and Oldenlandia herbacea were not touched by the animals, because of either their small size or some unpalatable constituent. For reasons not known, some dwarf grasses like Eragrostis hispida, Eriochloa nubica, Sporobolus discosporus, and sedges like Fimbfristylis spp. and Cyperus spp., were not touched either. The small size of these plants is perhaps an advantage to them, for herbivores may find it difficult to feed on small plants. In the forest, the advantage of these shallow-soil areas could be that they would act as firebreaks as they would be devoid of combustible matter even in the dry seasons when the carpet of dried plants is too thin for fire to pass through. Aristida adoensis Fuerstia africana Hyparrhenia collina FI. filipendula Justicia whytei Leonotis nepetifolia Lippia javanica Panicum maximum Pennisetum spp. Psiadia arabica Rhynchosia elegans Rhynchelytrum repens Senecio discifolius Setaria verticillata Themeda triandra Page 6 SUMMARY A study of shallow soils round Nairobi shows that the vegetation of these areas is not simply the first stage in a successional series, as one would expect, but a separate, specialized vegetation type discontinuous with neighbouring and successional com- munities. The vegetation of these areas is described and its biological significance discussed briefly. ACKNOWLEDGEMENT I am grateful to Dr. A. D. Q. Agnew for directing the study and for going through the manuscript; to the Director, Kenya National Parks for allowing supervised access to Nairobi National Park for this investigation; to the Botany Department, University College, Nairobi where this investigation was carried out; and to Mr. M. L. Modha for much discussion and advice. REFERENCES Agnew, A.D.Q. 1961. The ecology of Jnncus effusus L. in North Wales. J. Ecol. 49:83-102 Fisher, R.A. & Yates F. 1953, Statistical tables for biological, agricultural and medical re- search Oliver & Boyd, Edinburgh. Heriz-Smith, S., 1962. The wild flowers of the Nairobi Royal National Park, Nairobi. D. A. Hawkins Ltd., Nairobi. (Received October 1968) Page 7 THE DISTRIBUTION OF THE GENETS, GENETTA GENETTA, G. SERVALINA and G. TIGRINA IN EAST AFRICA By M. E. Taylor, Department of Zoology, University of Toronto, Toronto, Canada INTRODUCTION There are three species of genet in Kenya: Genetta genetta (Matschie), G. servalina (Thomas) and G. tigrina (Matschie). Of these G. tigrina is by far the commonest and is composed of two indistinct subspecies, G. tigrina erlangeri (Matschie) and G. tigrina stuhlmanni (Matschie). In this account the genets are treated at species level. G. genetta is a light coloured animal, the background colour being a dirty white, while G. tigrina has a much darker coloration; G. servalina has a yellowish-brown background colour with far more numerous markings. It is rare in Kenya, living in the remaining thick forests of western Kenya. The most recent record is from the Kakamega forest, collected in 1955. It is commoner in the forests of Uganda, several animals being caught in the Budongo forest in western Uganda in the last three years. Records Most of the data have been collected from trapping results, though additional information has been accumulated from road deaths and occasional sightings. The material in the National Museum, Nairobi, has been consulted and all the information to date includes about 200 records. There is virtually no information on the distribution of genets in the outlying districts of Kenya. The scatter of the collecting points reflects the road network, and also the number of collectors in the proximity of Nairobi. Fig. i shows the collecting results up to December 1968, with the approximate 63.5 cm.-i62.5 cm. (25 and 65 in.) rainfall contours. Most of the G. genetta records are from areas with less than 63.5 cm. (25 in.) of rain a year. They may be found in a few localities with a slightly higher rainfall, there being an overlap with G. tigrina in some places. Examples of this possibly sympatric asso- ciation have been noted at Ulu, Sultan Hamud and Voi; these areas have high and low rainfall regions. G. tigrina occurs in the wetter areas though these genets are not found in thick forests or at altitudes over 3048 m. (10,000 ft.) G. servalina occurs in wet forests with a rainfall of over 162.5 cm. (65 in.) a year. Walker (1964) refers to the frequency of melanistic genets, and they are certainly common in parts of Kenya. There are many records from the vicinity of Nairobi, though this reflects the number of collectors in the Nairobi region. Melanistic individuals are also recorded from parts of the Kinangop, Naivasha and the Mau Escarpment. All these Pc^e 8 Fig. I. — Distribution of the genets in East Africa. G. genetta G. servalina □ and G.tigrina O Melanistic animals are indicated (m). The symbols indicate one or more animals from an area. The approximate 63.5 cm (25 in.) and 165 cm. (65 in.) rainfall contours are estimated from the Mean Annual Rainfall Map for East Africa produced by the Overseas Survey, 1955. animals are G. tigrina and are indicated (m) in Fig. i. There are no records of melanistic G. genetta. Summary The distribution of the three species of genet is related to rainfall. G. genetta is essentially a dry country animal, occuring in regions with less than 63.5 cm. (25 in.) of rain a year. G. tigrina occurs in wetter regions, with a rainfall between 63.5 cm. and 162.5 cm. (25 and 65 in.) and G. servalina is found in regions with rainfall exceeding 162.5 cm. (65 in.) a year. ACKNOWLEDGEMENT I would like to thank all those people who have trapped animals for me and brought in corpses from the roads. I am also very grateful to Dr. L. S. B. Leakey for his continual help with this collecting programme. REFERENCES Walker, E. P., 1964. Mammals of the World, 2, The John Hopkins Press, Baltimore. {Received 11th December 1968) Page 10 A NEW FOUR-TOED MONGOOSE FROM KENYA, BDEOGALE CRASSICAUDA NIGRESCENS ssp. nov. By John B. Sale {Department of Zoology, University College, Nairobi) and Mark E. Taylor {Department of Osteology, Centre for Prehistory and Palaeontology, National Museum, Nairobi*) INTRODUCTION The four-toed mongooses belong to the African genus Bdeogale Peters. The genotype, Bdeogale crassicauda, was described by Peters in 1852 and currently contains four sub-species. They are distributed through central Mozambique, Malawi and Zambia {Bx. crassicauda Peters 1852); northern Mozambique and southern Tanzania {Bx. puisa Peters 1852); Zanzibar Island {Bx. tenuis Thomas & Wroughton 1908); and northern coastal Tanzania and Kenya {B.c. omnivora Heller 1913), according to Coetzee (1967). In addition, most recent authorities (Walker, 1964; Coetzee, 1967) regard Galeriscus Thomas as a sub-genus of Bdeogale, containing the species B. nigripes Pucheran (1855) and B.jacksoni Thomas (1894). A new sub-species of B. crassicauda is described in this paper and the sub-specific name nigrescens proposed. During a period of mammal trapping by the senior author in February, 1965, a number of specimens of an unfamiliar type of B. crassicauda were live-trapped at Lukenya, Kenya. Two of these animals were kept under observation in captivity in Nairobi for several months. Investigations showed that this form of Bdeogale did not correspond to existing descriptions of sub-species of the genus but further comparison and description were precluded when the animals escaped from captivity. A further seven specimens were trapped in the same locality during November/December 1967. One animal was kept alive in captivity until April 1968 and three, which were prepared as museum specimens (skull and skin), form the basis of the present description of a new sub-species. Unfortimately the type specimens of B.c. crassicauda and B.c. puisa were destroyed by bombs in 1945 while in the Berlin Museum, therefore impossible to compare the new material with these. The holotype of the new sub-species was, however, compared with British Museum examples of Bdeogale, viz three specimens of B.c. crassicauda, four of B.c. tenuis and one of B.c. puisa. It was also compared with the type specimen of B.c. omnivora (a female) in the United States National Museum. In addition, comparison was made with three specimens of B.c. omnivora in the National (formerly Coryndon) Museum (C.M.M.), Nairobi and a further three specimens collected by the junior author from the Sokoke Forest, near Gedi on the Kenya coast. Present address: Department of Zoology, University of Toronto, Ontario, Canada Page 11 TYPE LOCALITY The holotype and both paratypes used in the present description were trapped around the base of Lukenya Hill 37 km south-east of Nairobi, immediately to the north of the Nairobi-Mombasa road. Lukenya is a 3 km long ridge of metamorphic Basement System rock running north-east to south-west and rising 60 m. above the surrounding dry savanna (altitude 1840 m.). The vegetation around the base of the hill is grassland with scattered shrubs and trees, especially Acacia spp. and Commiphora africana (A. Rich.) Engl. There are also grasses on the hill itself, many raiubling herbs such as Cissus quad- rangularis L. and Sarcostemma viminale R. Br. and a variety of shrubs but only an occasional tree. In particular Ficus spp. are found growing among the rocks. Lukenya is a typical Klipspringer {Oreotragus oreotragus Neumann) habitat and a number of pairs of this koppie-dwelling antelope are found there. Numerous bare outcrops protrude from the sides of the hill and around the bases of these and the many boulders which are strewn around, boles and crevices of a variety of sizes provide excellent shelter for a host of small mammals. Of particular note is the fact that tihe new sub-species of B. crassicauda is found sharing a habitat with four other similar sized carnivores (one mustelid and three viverrids), viz Zorilla {Ictonyx striatus Perry); Genet {Genetta tigrina Matschie, see Taylor, 1969); Black-tipped Mongoose {Herpestes sanguineus Riippell) and White-tailed Mongoose {Ichneumia albicauda G. Cuvier). During a study of viverrids, the junior author has foimd this type of situation to be quite common. Attempts to find specimens of B.c. nigrescens in other localities up to 80 km from Lukenya have so far been unsuccessful. One report exists- of a similar mongoose being seen at Athi River about 8 km from Lukenya. The type locality of B.c. omnivora is Mazeras, approximately 450 km south-east of Lukenya near the Kenya coast. There appear to be no records of either B.c. omnivora or B.c. nigrescens between the two type localities but extensive trapping of this large area has yet to be carried out. The new sub-species appears to be geographically isolated from other known locaUties of the species. This marked isolation provides good additional evidence for designating the Lukenya form as a distinct sub-species. DESCRIPTION OF THE NEW SUB-SPECIES The holotype is an adult male, number BM 68.1103, iw British Museum (Natural History). It was collected by the authors on 24 November 1967. Two paratypes are being designated as follows : Paratype i : an adult male, number C.M.M. 7512, in the National Museum, Nairobi, collected by the authors on 6 December 1967. Paratype 2 : an adult male, number C.M.M. 7513, in the National Museum, Nairobi, collected by the authors on 7 December 1967. The name proposed for the new sub-species is ]Bdeogale crassicauda nigrescens^ on account of the very dark, almost black, coat that distinguishes it from other members of the species. General features The new sub-species is a medium-sized, four-toed mongoose with a very dark, almost black, glossy coat and a short bushy tail. The head is rather roimded for a viverrid, giving the appearance of a relatively short muzzle (Fig. i). Like other members of the species this animal is markedly docile and lacks the aggressive nature of the majority of viverrids when first caught. Page 12 Weight and external measurements (see Table i) Unfortunately the weights of the older museum specimens of the genus were not taken. The wei^t of the new form is much greater than that of recently collected B.c.omnivora from the Kenya coast (average 907 g.) although the difference between the body lengths of these two genera is not very great (see below). The head and body length of B.c. nigrescens is intermediate between B.c.puisa, which is longer, and B.c. crassicauda, B.c. tenuis, and B.c. omnivora which are shorter. Of the four existing sub-species, the new form is nearest to B.c. omnivora. The tail length of nigrescens is shorter than all the others except tenuis but it is not clear how the tail lengths of the earher type specimens were obtained. Our own tail measurements for the new sub-species were measured dorsally between the last sacral vertebra and the tip of the last caudal vertebra. The ratio of tail length to head and body length is less in the new form than that of the types of crassicauda and omnivora and similar to those of puisa and tenuis. Skull measurements (see Table i) The skull of B.c. nigrescens is greater than that of B.c. tenuis in all measurements recorded in Table i. Conversely, it is less than B.c. puisa in all dimensions except the distance between the orbit and the ant-orbital foramen, in which respect the new form is greater. In some dimensions, especially the condylo-basal leng^, the new form is similar to B.c. omnivora. However, greatest similarity is shown with B.c. crassicauda, where the only significant difference is in the greater condylo-basal length of the latter. Coat characteristics (see Table 2) As seen from the summary in Table 2, B.c. nigrescens is distinguished by having a generally darker coat than the other sub-species. In particular, the back and tail (which is completely black dorsally) are very dark due to a preponderance of long guard hairs, of which at least the distal portion is a shiny black. The coat of the living animal presents a much blacker appearance than the museum skin, due to the fact that in the former the Table i Measurements (in g. and mm) of the five ssp. of Bdeogale crassicauda puisa crassicauda tenuis omnivora nigrescens Holotype Paratype i Paratype 2 Weight Head and body ~ ~ ~ ~ 1580 1575 1500 length 500 400 410 420 432 443 473 Tail length 250 300 180 245 232 200 210 Hind foot 75 84 70 81 74 82 75 Ear Condylobasal ~ ~ 20 34 32 30 32 length 87.5 87.9 80.8 85 85.4 82.2 83.4 Zygomatic width Interorbital 48.5 47-4 44.1 46 46.4 44-6 48.3 width Post-orbital 20.6 18.4 16.4 18 19.4 18.2 19.2 constriction 16.6 16.5 14-5 14 16.3 15-7 15.6 Palate length Palate width 48.9 49.2 45.0 50 49-5 47.8 47.8 at Ml Upper tooth 27. S 26.9 25.2 27.6 26.5 28.0 row length 38.2 38.5 35-2 40 37-9 36.6 38.2 Ml width Distance between 7.0 7.0 5-9 7 6.8 6.4 6.9 orbit & ant- orbital foramen 3.6 4.5 2.3 — 3-9 4-3 4.0 a,.™,-.-. iJSSv&£ Xbdomen "°‘ ||SksaS'z?’ I. Uniform light yellow to light S- |ilS5S StS#5 II Pf4 ill lai Pill 1 §“isrs '”■ i«S~ Lfiss:“ 3;„'?S5?.'T'’‘''‘“’'“°”-'‘“’' Sis? fc5?= pn}£ Si,e.« g^aiav E5'ss,75“rw?'??oi*a‘ at. %ssr pESrS: ?°F^e^wtr"curly hairs, uniformly 2!'^kckXo dark brown, (44). ^:lSte-rown. X^en ::sas::<„, nisriscens darker than others. 5?? 4, SpSCit^h“i.ri^ ipliis ptsis SSlSi kSSS- Hank of mgnW not so dark as Feet I. Mid brown tip h light yellow fcsur®-- £!»><«'• '”■* is^- pp-airtizs ?°u"Xndylightyellowtomid rg^g^it brown, rest light ims Whm Ess?*- SlE^= Page 15 guard hairs are at least partially erected, their black tips tending to hide their lighter bases and the yellowish brown underfur. The flanks and belly are progressively less dark due to the relatively lower density of black guard hairs in these regions. The tail of B.c. nigrescens is very distinctive, being broad compared with its length. It has a very dense covering of black hairs, particularly dorsally, which give it a “bottle- brush” appearance when the hairs are erected in the live animal. Biology From trapping records and the fact that hardly anyone seems to have seen them in the wild, B.c. nigrescens would appear to be a nocturnal animal. In captivity it may become diurnal, though on the whole it remains nocturnal. It seems likely from the nature of the type locality, that it shelters in well-hidden holes in the rocks during daytime. Its narrow, only slightly curved claws do not suggest that the new sub-species is adapted for extensive digging, though it may well scrape up insect larvae from the surface layers of the soil. Moreover, its timid, docile nature indicates a shy animal that spends the daylight hours in strict seclusion. All museum specimens obtained to date have been live-trapped and killed later, so that there is no information on diet from stomach content analyses. However, the animals kept in captivity did show certain food preferences. They would not eat bananas, mangoes or avocado pears and found difficulty in breaking open a chicken’s egg. When the egg was previously broken, however, they lapped up the contents quickly. Their basic food was raw meat which was cut up into chunks and mixed with bone meal and multi-vitamins. One of the captive animals was also fed with wild young rats, live-trapped. It was very slow in catching them and did not appear to know how to kill them, seizing these rodents by the tail or back and then finally killing them by biting the back of the skull. One of the captive nigrescens was fed a loo cm long stripe-bellied sand snake {Psammophis subtaeniatus Werner) which it soon killed and ate. This may indicate that the new sub-species feeds more on lizards and snakes than rodents, though its main food is probably insects. ACKNOWLEDGEMENTS Especial thanks are due to Dr. G. Corbet of the British Museum and Dr. Henry W. Setzer of the U.S. National Museum, both of whom kindly compared our specimens with material in their respective institutions. Mr. B. Lawrence-Brown, on whose property at Lukenya the animals were trapped, has been most helpful and we wish to record our gratitude to him. The senior author is grateful for funds provided by the Ford Foundation. Thanks are also due to Dr. L. S. B. Leakey for encouragement with this work and for providing funds to enable the junior author to examine material in Berlin and London. REFERENCES COETZEE, C. G. 1967. Preliminary Identification Manual for African Mammals. 7. Carnivora. Ed. J. Meester. Smithsonian Institution, Washington. Heller, E. 1913. New Antelopes and Carnivores. Smithsonian Misc. Coll. 61(13) :i2, 13. Peters, W. 1852. Monatsb. K. Preuss. Akad. Wiss. Berlin, p. 81. ,, 1852. Reise nach Mossambique, Saugeth. p. 1 19-125. Pucheran, 1855. Rev. et Mag. deZool. (2)7:111. Taylor, M. E. 1969. The Distribution of the Genets in East Africa. Jl E. Africa nat. Hist. Soc. (In press). Thomas, O. 1894. Am. Mag. Nat. Hist. (6)13:522. Thomas, O. and Wroughton, R. C. 1908. The Rudd Exploration of South Africa. IX. List of Mammals obtained by Mr. Grant on the Gorongoza Mountains, Portuguese S. E. Africa. Proc. Zool. Soc. Land. (2): 164- 173. Walker, E. P. 1964. Mammals of the World. II. Johns Hopkins Press. Baltimore. {Received 17th April, 1969). Page 16 BIRD RINGING REPORT 1968-1969 By G. C. Backhurst INTRODUCTION This report covers the period i July, 1968 to 30 June, 1969. It is pleasant to report that, once again, the number of birds ringed is higher than ever before. As was hinted in last year’s report (Backhurst, 1969&) some experienced ringers joined the scheme towards the end of 1968 while another came out early in 1969. The full list of birds ringed is given in Table i ; the birds which are palearctic migrants are printed in bold type, others which are included in the palearctic fauna but which are also ethiopian are not so distinguished. The nomenclature used follows White (i960, 1961, 1962, 1963, 1965), a departure from previous reports but one that is consider- ed desirable since White’s lists reflect modern ideas more accurately than do the north- eastern parts of Mackworth-Praed & Grant’s Handbook; moreover White’s lists are to be the basis of the forthcoming AFRING list of birds (C. C. H. Elliott, pers. comm.). The number of recoveries. Table 2, shows a satisfying increase. Birds ringed in previous seasons and retrapped at or near their original ringing sites are listed in Table 3. SOME NOTES ON RINGING IN EAST AFRICA Most birds were ringed in Kenya, a few hundred in Uganda and none at all in Tanzania. Fifteen ringers were operating in the two countries this season; most of the ethiopian species were ringed in western Kenya by P.L. & Mrs. H. A. Britton and J. F. & Mrs. L. M. Harper; smaller numbers were ringed in western Uganda by Dr, M. P. L. Fogden. Most of the ducks and waders were ringed at Lake Nakuru and at Naivasha by the Brittons, Harpers, Dr, E. D. Steel and G. C. & Mrs. D. E. G. Backhurst. Hirundines were ringed at Nairobi, Nakuru and in western Kenya. The bulk of the Yellow Wagtails Motacilla flava were ringed at Nairobi and Kabete by the combined efforts of the Misses J. & D. Angwin, Mrs. L. Campbell, Dr. Steel and the Backhursts; work is progressing on a separate report on this species (Backhurst, in prep.). One palearctic species, the Pintail Snipe Gallinago stenura ringed at Naivasha, was new for the Kenya list (Backhurst, 1969a); another palearctic wader, the Long-toed Stint Calidris subminuta caught at the same marsh at Naivasha during ringing operations, although not ringed, was also new for the national list (Backhurst & Britton, 1969). An example of the European Wryneck Jynx torquilla was ringed at Ng’iya in western Kenya, another first record for the country (Britton & Harper, 1969). Britton & Harper {loc. cit.) record the occurrence of eight ethiopian species and one subspecies new to Kenya, three of these species have been recorded before in Kenya but with no supporting evidence. The new birds are: Blue-breasted Bee-eater Merops variegatus. Yellow-fronted Tinker-Bird Pogoniulus chrysoconus. White- winged Warbler Bradypterus graueri. Yellow Swamp-Warbler Chloropeta gracilirostris. Northern Brown- throated Weaver Ploceus castanops. Black Bishop (Uganda race) Euplectes gierowii ansorgei. Brown Twin-Spot Clytospiza monteiri, Black-rumped Waxbill Estrilda troglodytes, and Bar-breasted Fire-Finch Lagonosticta rufopicta. ACKNOWLEDGEMENTS Ringers gratefully acknowledge the co-operation of the City Engineer, Nairobi, for allowing them to operate at the two sewage works ; the Director of the Kenya National Parks for permission to ring at Lake Nakuru; the Director of Veterinary Services, Kenya for permission to ring at Kabete; and to Mr. Douros for allowing ringers to operate on his land at Kabete. My wife gave me invaluable help in ringing and administrative tasks. The Society is grateful to the Director of the National Museum, Nairobi, for allow- ing the Museum’s address to appear on the rings. Page 17 Table i BIRDS RINGED BY THE EAST AFRICA NATURAL HISTORY SOCIETY RINGING ORGANISATION Palearctic migrants in bold type. Podiceps ruficollis (Pallas), Little Grebe Ardeola ibis (L.), Cattle Egret ..... A. ralloides (Scopoli), Squacco Heron Phoenicoptems minor Geoffrey, Lesser Flamingo Platalea alba Scopoli, African Spoonbill Threskiornis aethiopica (Latham), Sacred Ibis Anas capensis Gmelin, Cape Wigeon .... A. erythrorhynchos Gmelin, Red-billed Duck A. hottentota Eyton, Hottentot Teal A. querquedula L., Garganey .... A. undulata Dubois, Yellow-billed Duck . Netta erythropthalma (Wied), African Pochard . Accipiter badius (Gmelin), Shikra .... Milvus migrans migrans (Boddaert), Black Kite Falco biarmicus Temminck, Lanner .... F. subbuteo L., Hobby Coturnix coturnix africana Temminck & Schlegel, Quail C. delegorguei Delegorgue, Harlequin Quail Fulica cristata Gmelin, Crested Coot Gallinula chloropus (L.), Moorhen .... Porphyria porphyria (L.), Purple Gallinule Charadrius asiaticus Pallas, Caspian Plover . C. dubius Scopoli, Little Ringed Plover . C. hiaticula L., Ringed Plover .... C. leschenaultii Lesson, Great Sand Plover C. mongolus Pallas, Mongolian Sand Plover C. pallidus Strickland, Chesmut-banded Sand Plover . C. pecuarius Temminck, Kittlitz’s Sand Plover . C. tricollaris Vieillot, Three-banded Plover Vanellus armatus (Burchell), Blacksmith Plover . V. spinosus (L.), Spurwing Plover .... Dramas ardeola Paykull, Crab Plover Cursorius chalcopterus Temminck, Violet-tipped Courser Glareola pratincola (L.), Pratincole .... Larus cirrocephalus Viellot, Grey-headed Gull Sterna leucoptera Temminck, White- winged Black Tern S. nilotica (Gmelin), Gull-billed Tern Himantopus himantopus (L.), Black-winged Stilt Recurvirostra avosetta L., Avocet .... Rostratula benghalensis (L)., Painted Snipe Calidris alba (Pallas), Sanderling .... C. ferruginea Pontoppidan, Curlew Sandpiper C. minuta (Leisler), Little Stint .... C. temminckii (Leisler), Temminck’s Stint Gallinago gallinago (L.), Snipe .... G. nigripennis (Bonaparte), African Snipe . G. stenura (Bonaparte), Pintail Snipe Philomachus pugnax (L.), Ruff .... Tringa glareola L., Wood Sandpiper T. hypoleucos L., Common Sandpiper . T. nebularia (Gunnerus), Greenshank T. ochropus L., Green Sandpiper .... T. stagnatilis (Bechstein), Marsh Sandpiper T. terek Latham, Terek Sandpiper .... Streptopelia capicola (Sundevall), Ring-necked Dove . S. decipiens (Finsch & Hartlaub), Mourning Dove S. senegalensis (L.), Laughing Dove .... Tartar afer (L.), Blue-spotted Wood Dove T. chalcospilos (Wagler), Emerald-spotted Wood Dove T. tympanistria (Temminck & Knip), Tambourine Dove Grand Total II 73 7 II5 43 91 7 36 4 3 17 74 9 4 lOO 136 19 97 II I 18 6 15 Page 18, Centropus superciliosus Hemprich &: fihr^nberg, White-browed Coucal Chrysococcyx caprius (Boddaert), Didric Cuckoo .... C. cupreus (Shaw), Emerald Cuckoo ..... C. Maas (Stephens), Klaas’ Cuckoo ...... Glaucidum tephronotum Sharpe, Red-chested Owlet Caprimulgus climacurus Vieillot, Long-tailed Nightjar . , C. fossil Hartlaub, Gabon Nightjar ...... C. poliocephalus Riippell, Abyssinian Nightjar . . . Macrodipteryx longipennis (Shaw), Standard-wing Nightjar . Apus affinis (Grey), Little Swift ...... A. coffer (Lichtenstein), White-rumped Swift .... Colius macrourus (L.), Blue-naped Mousebird .... C. striata Gmelin, Speckled Mousebird ..... Alcedo cristata Pallas, Malachite Kingfisher .... Ceryle rudis (L.), Pied Kingfisher . . . . Ceyx picta (Boddaert), Pigmy Kingfisher . . . . Halcyon chelicuti (Stanley), Striped Kingfisher .... H. leucocephala (Muller), Grey-headed Kingfisher H. senegalensis (L.), Woodland Kingfisher .... Merops albicollis Vieillot, White-throated Bee-eater M. apiaster L., Bee-eater ....... M. bulocki bullockoides Smith, White-fronted Bee-eater M. pusillus Muller, Little Bee-eater ...... M. superciliosus L., Blue-cheeked Bee-eater .... M. variegatus Vieillot, Blue-breasted Bee-eater . . . Upupa epops L., Hoopoe Gymnobucco bonapartei Hartlaub, Grey-throated Barbet Lybius bidentatus (Shaw), Double-toothed Barbet Lyguifsobalito Hermann, Black-billed Barbet . . . . L. lacrymosus (Cabanis), Spotted-flanked Barbet ... L. leucocephalus (Defilippi), White-headed Barbet Pogoniulus bilineatus (Sundevall), Golden-rumped Tinker-Bird P. chrysoconus (Temminck), Yellow-fronted Tinker-Bird P. leucomystax (Sharpe), Moustached Green Tinker-Bird P. pusillus (Dumont), Red-fronted Tinker-Bird .... Verreaux, Yellow-billed Barbet Indicator indicator (Sparrman), Greater Honey-guide I. minor Stephens, Lesser Honey-guide . . I. variegatus Lesson, Scaly- throated Honey-guide Campethera nivosa (Swainson), Buff-spotted Woodpecker C. nubica (Boddaert), Nubian Woodpecker .... Dendropicos fuscescens (Vieillot), Cardinal Woodpecker Jynx torquilla L., Wryneck ....... Mesopicos goertae (Muller), Grey Woodpecker . . . Mirafra africana A. Smith, Rufous-naped Lark . Campephaga phoenicea (Latham), Black Cuckoo Shrike Dicrurus adsimilis (Bechstein), Drongo ..... D. ludwigii (Smith), Square-tailed Drongo .... Emberiza flaviventris Stephens, Golden-breasted Bunting Amandava subflava (Vieillot), Zebra Wax-bill .... Clytospiza monteiri (Hartlaub), Brown Twinspot . . Cryptospiza jacksoni Sharpe, Dusky Crimson-wing C. salvadorii Reichenow, Abyssinian Crimson-wing C. shelleyi Sharpe, Shelley’s Crimson-wing .... Estrilda astrild (L.), Wax-bill .....•• E. bengala (L.), Red-cheeked Cordon-bleu .... E. erythronotos (Vieillot), Black-cheeked Wax-bill E. ianthinogaster (L.), Purple Grenadier ..... E. melanotis (Temminck), Yellow-bellied Wax-bill E. nonnula Hartlaub, Black-crowned Wax-bill .... E. paludicola Heuglin, Fawn-breasted Wax-bill . . E. rhodopyga Sundevall, Crimson-rumped Wax-bill E. troglodytes (Lichtenstein), Black-rumped Wax-bill . Hypargos niveoguttatus (Peters), Peters’ Twinspot Page 19 Lagonosticta rubricata (Lichtenstein), African Firefinch L. rufopicta (Fraser), Brown Firefinch .... L. senegala (L.), Red-billed Firefinch .. .. Lonchura bicolor (Fraser), Black & White Mannikin L. cucullata (Swainson), Bronze Mannikin Nigrita canicapilla (Strickland), Grey-headed Negro-Finch . Pytelia melba (L.), Green-winged Ptylia .... Spermophaga ruficapilla (Shelley), Red-headed Blue-bill Vidua chalybeata (Muller), Indigo Bird .... V. macroura (Pallas), Pin-tailed Whydah .... Serinus atrogularis (A. Smith), Yellow-rumped Seed-eater . S. burtoni (Gray), Thick-billed Seed-eater 5. citrineUoides Riippell, African Citril S', dorsostriatus (Reichenow), White-bellied Canary S. mozambicus (Muller), Yellow-fronted Canary . S. striolatus (Rtippell), Streaky Seed-eater .... S. sulphuratus (L.), Brimstone Canary .... Delichon urbica (L.), House Martin .... Hirundo abyssinica Guerin, Striped Swallow H. angolensis Bocage, Angola Swallow H. daurica L., Red-rumped Swallow .... H. fuligula Lichtenstein, African Rock Martin . H. rustica L., Swallow ....... H. semirufa Sundevall, Rufous-chested Swallow . H. senegalensis L., Mosque Swallow .... H. smithii Leach, Wire-tailed Swallow .... Psalidoprocne albiceps Sclater, White-headed Rough-wing P. pristoptera (Ruppell), Black Rough-wing Riparia cincta (Boddaert), Banded Martin R. paludicola (Vieillot), African Sand Martin R. riparia (L.), Sand Martin ..... Eurocephalus anguitimens Smith, White-crowned Shrike Dryoscopus ciibla (Shaw), Black-backed Puff-back Shrike D. gambensis (Lichtenstein), Puff-back Shrike Laniarius barbarus (L.), Black-headed Gonelek . L./errM^/neMS (Gmelin), Tropical Boubou L. funebris (Hartlaub), Slate-coloured Boubou L. IweMeri Reichenow, Liihder’s Bush Shrike Lanius collaris L., Fiscal . . . . . L. collurio L., Red-backed Shrike ..... L. excubitorius Prevost & Des Murs, Grey-backed Fiscal L. minor Gmelin, Lesser Grey Shrike .... Prionops plumata (Shaw), Curly-crested Helmet Shrike Tchagra australis (Smith), Brown-headed Bush-Shrike T. minuta (Hartlaub), Black-cap Bush-Shrike T. senegala (L.), Black-headed Bush-Shrike Malaconotus dohertyi (Rothschild), Doherty’s Bush-Shrike . M. sulfureopectus (Lesson), Sulphur-breasted Bush-Shrike . Nilaus afer (Latham), Northern Brubru .... Anthus cervinus (Pallas), Red-throated Pipit . A. leucophrys Vieillot, Plain-backed Pipit .... A. novaeseelandiae Gmelin, Richard’s Pipit A. trivialis (L.), Tree Pipit Macronyx croceus (Vieillot), Yellow-throated Longclaw Motacilla alba alba L., White Wagtail .... M. alba vidua (Sundevall), African Pied Wagtail M. capensis L., Cape Wagtail ...... M. cinerea Tunstall, Grey Wagtail .... M. flava L., Yellow Wagtail ...... Batis capensis (L.), Puff-back Flycatcher .... B. minor Erlanger, Black-headed Puff-back Flycatcher B. molitor (Halin & Kiister), Chin-spot Puff-back Flycatcher Bradornis pallidus (Muller), Pale Flycatcher Melaenornis chocolatina (Ruppell), White-eyed Slaty Flycatcher M. edoloides (Swainson), Black Flycatcher .... 1968/9 Grand Total 5 5 5 5 I I 4 8 10 10 Page 20 Muscicapa adusta (Boie), Dusky Flycatcher .... M. aquatica Heuglin, Swamp Flycatcher ..... M. caerulescens (Hartlaub), Ashy Flycatcher .... M. striata (Pallas), Spotted Flycatcher ..... Myioparus plumbeus (Hartlaub), Grey Tit Flycatcher Platysteira blissetti (Sharpe), Jameson’s Watde-eye P. castenea (Fraser), Chestnut Wattle-eye ..... P. cyanea (Muller), Wattle-eye ...... P. peltata Sundevall, Black-throated Wattle-eye Terpsiphone rufiventer (Swainson), Black-headed Paradise Flycatcher T. viridis (Muller), Paradise Flycatcher ..... Trochocercus albonotatus Sharpe, White-tailed Crested Flycatcher T. longicauda (Swainson), Blue Flycatcher .... T. nigromitratus (Reichenow), Dusky-crested Flycatcher Acrocephalus arundinaceus (L.), Great Reed Warbler A. boeticatus (Vieillot), African Reed Warbler .... /I. gra«7/Voirm (Hartlaub), Lesser Swamp Warbler A. palustris (Bechstein), Marsh Warbler .... A. rufescens (Sharpe & Bouvier), Greater Swamp Warbler . A. schoenabaenus (L.), Sedge Warbler ..... A. scirpaceus (Hermann), Reed Warbler ..... Apalis cinerea (Sharpe), Grey Apalis ..... (Strickland), Black-breasted Apalis .... A. pulchra Sharpe, Black-collared Apallis ..... Bathmocercus cerviniventris (Sharpe), Black-faced Rufous Warbler Bradypterus cinnamomeus (Riippell), Cinnamon Bracken Warbler . Camaroptera brachyura (Vieillot), Grey-backed Camaroptera C. chloronota Reichenow, Olive-green Camaroptera Chloropeta natalensis Smith, Yellow Flycatcher Warbler C. similis Richmond, Mountain Yellow Flycatcher Warbler . Cisticola brunnescens Heuglin, Pectoral-patch Cisticola C. carruthersi O. Grant, Carruther’s Cisticola .... C. chiniana (A. Smith), Rattling Cisticola ..... C. erythrops Hartlaub, Red-faced Cisticola .... C. galactotes (Temminck), Winding Cisticola .... C. humeri Shelley, Hunter’s Cisticola ..... C. natalensis (Smith), Croaking Cisticola C. robusta (Riippell), Stout Cisticola ...... C. woosnami O. Grant, Trilling Cisticola ..... Eminia lepida Hartlaub, Grey-capped Warbler .... Eremomela icteropygialis (Lafresnaye), Yellow-bellied Eremomela . Hippolais icterina (Vieillot), Icterine Warbler .... H. languida (Hemprich & Ehrenberg), Upcher’s Warbler H. pallida (Hemprich & Ehrenberg), Olivaceous Warbler . Hylia prasina (Cassin), Green Hylia ..... Phylloscopus budongoensis (Seth Smith), Uganda Woodland Warbler P. trochilus (L.), Willow Warbler P. umbrovirens (Riippell), Brown Woodland Warbler . Prinia bairdii (Cassin), Banded Prinia ..... P. leucopogon (Cabanis), White-chinned Prinia .... P. subfiava (Gmelin), Tawny-flanked Prinia .... Sylvia atricapilla (L.), Blackcap S. borin (Boddaert), Garden Warbler ..... S. communis Latham, Whitethroat ..... S. nisoria (Bechstein), Barred Warbler ..... Sylvietta brachyura Lafresnaye, Crombec ..... S’, leucophrys Sharpe, White-browed Crombec .... S. whytii Shelley, Red-faced Crombec ..... Alcippe abyssinica (Riippell), Abyssinian Hill Babbler Trichastoma albipecta (Reichenow), Scaly-breasted Illadopsis T. fulvescens (Cassin), Brown Illadopsis . . _ . T. pyrrhoptera (Reichenow & Neumann), Mountain Illadopsis T. rufipennis (Sharpe), Pale-breasted Illadopsis .... Turdoides jardinei (Smith), Arrow-marked Babbler T. melanops (Hartlaub), Black-lored Babbler .... 1968/9 Grand Total T. plebejus (Kretzchmar), Brown Babbler ..... Alethe archeri (Sharpe), Archer’s Robin Chat .... (Bonaparte), Brown-chested Alethe A. poKophrys Sharpe, Red-throated Alethe .... Cercotrichas hartlaubi (Reichenow), Brown-backed Scrub Robin . C. leiicophrys (Vieillot), Red-backed Scrub Robin C. quadrivirgata (Reichenow), Eastern Bearded Scrub Robin Cossypha c^ra (L.), Robin Chat ...... C. cyanocampter (Bonaparte), Blue-shouldered Robin Chat C. heuglini Hardaub, White-browed Robin Chat C. natalensis Smith, Red-capped Robin Chat .... C. niveicapilla (Lafresnaye), Snovvy-headed Robin Chat C. poliopte g Reichenow, Grey-winged Robin Chat Luscinia luscinia (L.), Sprosser ...... L. megarhynchos Brehm, Nightingale Monticola saxatillis (L.), Rock Thrush ..... Myrmecocichla nigra (Vieillot), Sooty Chat .... Neocossyphus poensis (Strickland), White-tailed Ant Thrush Oenanthe isabellina (Temminck & Tangier), Isabeline Wheatear O. oenanthe (L.), Wheatear ....... O. plesehanka Lepechin, Pied Wheatear .... Phoenieurus phoenicurus (L.), Redstart .... Pogonocichla stellata (Vieillot), White-starred Bush Robin Saxicola rubetra (L.), Whinchat ...... S', torquata (L.), Stonechat ....... Sheppardia aequatorialis (Jackson), Equatorial Akalat Turdus abyssinicus (L.), Olive Trush ...... T. pelios Bonaparte, African Thrush ...... T. piaggiae Bouvier, Abyssinian Ground Thrush Anthreptes collaris (Vieillot), Collared Sunbird .... Nectarinia alinae (Jackson), Blue-headed Sunbird N. bifasciata (Shaw), Little Purple-banded Sunbird N. cuprea (Shaw), Copper Sunbird ...... N. erythroceria Hartlaub, Red-chested Sunbird .... N. kilimensis Shelley, Bronze Sunbird ..... N. mariquensis (Smith), Mariqua Sunbird ..... N. mediocris (Shelley), Eastern Double-collared Sunbird N. olivacea (Smith), Olive Sunbird ...... N. preussi (Reichenow), Northern Double-collared Sunbird . N. pulchella (L.), Beautiful Sunbird ..... N. regia (Reichenow), Regal Sunbird ..... N. reichenowi (Fischer), Golden-winged Simbird N. senegalensis (L.), Scarlet-chested Sunbird .... N. tacazza (Stanley), Tacazze Simbird ..... N. venusta (Shaw & Hodder), Variable Sunbird N. verticalis (L.), Green-headed Sunbird ..... Oriolus larvatus Lichtenstein, Black-headed Oriole O. oriolus (L.), Golden Oriole ...... Pams albiventris Shelley, White-bellied Tit Amblyospiza albifrons (Vigors), Grosbeak Weaver Anomalospiza imberbis (Cabanis), Parasitic Weaver Euplectes albonotatus (Cassin), White-winged Widow Bird . E. ardens (Boddaert), Red-collared Widow Bird E. axillaris (Smith), Fan-tailed Widow Bird .... E. gierowii Cabanis, Black Bishop ...... E. hordeaceus (L.), Black-winged Red Bishop .... E.7ac^50w/ Sharpe, Jackson’s Widow Bird . .... E. macrourus (Gmelin), Yellow-mantled Widow Bird . E. orix (L.), Red Bishop ....... Malimbus rubriceps (Simdevall), Red-headed Weaver . Passer eminibey (Hartlaub), Chesmut Sparrow .... P. griseus (Vieillot), Grey-headed Sparrow .... P. iagoensis (Gould), Kenya Rufous Sparrow .... Plocepasser mahali Smith, Stripe-breasted Sparrow Weaver . Ploceus alienus (Sharpe), Strange Weaver ..... 1968/9 25 I 10 9 5 42 8 Grand Total 25 I 39 I 3 20 5 15 12 37 10 6 10 1 3 3 16 2 18 24 47 2 37 27 6 6 7 13 65 81 Page 22 1968/9 P. baglafecht (Daudin), Emin’s & Reichenow’s Weavers P. bicolor Vieillot, Dark-backed Weaver ..... P. bojeri (Cabanis), Golden Palm Weaver ..... P. castanops Shelley, Northern Brown-throated Weaver P. cucullatus (Muller), Black-headed Weaver .... P. intermedius RUppell, Masked Weaver ..... P. jacksoni Shelley, Golden-backed Weaver .... P. luteolus (Lichtenstein), Little Weaver ..... P. melanocephalus (L.), Yellow-backed Weaver .... P. melanogaster Shelley, Black-billed Weaver .... P. nigricollis (Vieillot), Black-necked Weaver .... P. ocularis Smith, Spectacled Weaver ..... P. pelzelni (Hartlaub), Slender-billed Weaver .... P. superciliosus (Shelley), Compact Weaver .... P. velatus VieiUot, Vitelline Masked Weaver .... P. xanthops (Hartlaub), Holub’s Golden Weaver Quelea cardinalis (Hartlaub), Cardinal Quelea .... Q. erythrops (Hartlaub), Red-headed Quelea .... Q. quelea (L.), Red-billed Quelea ...... Andropadus curvirostris Cassin, Cameroon Sombre Greenbul A. importunus (Vieillot), Zanzibar Sombre Greenbul . A. latirostris Strickland, Yellow- whiskered Greenbul . A. montanus Reichenow, Shelley’s Greenbul .... A. tephrolaemus (Grey), Olive-breasted Moimtain Greenbul A. virens Cassin, Litde Greenbul ...... Bleda syndactyla Swainson, Bristle Bill ..... Chlorocichla flavicollis Swainson, Yellow-throated Leaf-Love C. flaviventris (Smith), Yellow-bellied Greenbul Phyllastrephus baumanni Reichenow, Toro Olive Greenbul . P. debilis (Sclater), Smaller Yellow-streaked Greenbul P. fischeri (Reichenow), Fischer’s Greenbul .... P. strepitans (Reichenow), Northern Brownbul .... P. terrestris Swainson, Brownbul ...... Pycnonotus barbatus (Desfontaines), Dark-capped Bulbul Buphagus erythorhynchus (Stanley), Red-billed Oxpecker Cinnyricinclus leucogaster (Boddaert), Violet-backed Starling Creatophora cinerea (Menschen), Wattled Starling Lamprotornis chloropterus Swainson, Lesser Blue-eared Glossy Starling L. caudatus (Muller), Riippell’s Long-tailed Glossy Starling Spreo superbus (Ruppell), Superb Starling .... Zosterops abyssinica flavilateralis (Reichenow), Yellow White-eye . Z. senegalensis jacksoni (Neumann), Green White-eye . Z. s. kikuyuensis (Sharpe), Kikuyu White-eye .... Total Total Palearctic Migrants Total Number of Species . Total Palearctic Species . Table 2 RECOVERIES AND CONTROLS OF BIRDS RINGED IN EAST AFRICA Key to symbols and terms Ring number Age Sex Manner of recovery — Where this is in italics the ring has been returned, f.g. — full grown, age uncertain; ad. — adult, at least one year old; pull. — yoimg, not able to fly freely; juv. — juvenile, able to fly freely; 1st W. — first winter. S — male. $ — female. + — shot or killed by man; X — foimd dead or dying; Grand Total 75 I 4 178 77 90 2 425 5 3 47 lOI 3 42 49 73 19 10 322 7 19 24 13 17 5 7 3 45 7 7 420 2 33 5 3 II 14 5 33 985 27 376 360 57 Page 23 xA — found long dead; / ?/ — manner of recovery unknown; V — caught or trapped and released with ring; 0 — caught or trapped alive and not released, or released but with ring removed. Date of recovery : — Where this is unknown, the date of the reporting letter is given in brackets. A recovery in the strict sense is a ringed bird found dead, whether by the ringer himself or reported by a member of the public; a control is a bird ringed by one ringer and retrapped by another, or a bird retrapped by the original ringer at a point more than three miles from the locality where it was first ringed. Charadrius pecuarius Kittlitz’s Sand Plover A. 3416 f-g- /?/ 20.4.68 6.4.69 Lake Nakuru, Kenya. o°20'S., 36°o6'E. DJP. Lake Elmenteita, Kenya. o°2fS., 36°i5'E. 23 km SE. (Paul Kihika) Calidris minuta Little Stint A. 0773 f-g. 3.3.67 1.1.69 ake Nakuru, Kenya. JBS. LNaivasha, Kenya. o°46'S., 36°2o'E. 58 km SE. (GCB) Tringa glareola Wood Sandpiper B. 5638 ad. /?/ 12.4.69 0.9.69 Lake Nakuru, Kenya. EDS. near Mezen’, Arkhangelsk’ Region, U.S.S.R. 65°50'N., 44°i7'E. (Ringing Centre, Moscow) Tringa hypoleucos Common Sandpiper A. 4997 f.g. + 14.4.68 28.5.68 Lake Nakuru, Kenya. GCB. near Chaikovskii, Perm’ Region, U.S.S.R. 56°46'N., 54°o8'E. (Ringing Centre, Moscow) Philomachus pugnax Rulf B. 0617 f.g.? /?/ 6. 5.68 0.10.68 Lake Nakuru, Kenya. DJP. Roberts’ Camp, Lake Baringo, Kenya. o°4o'N., 36°oo'E. 105 km N. (Mrs. B. Roberts). B. 5553 ad? + 8.2.69 20.5.69 Lake Nakuru, Kenya. EDS. near Berdigestyakh, Gornyi District. Yakutian A.S.S.R., U.S.S.R. 62°07'N., i26°39'E. (Ringing Centre, Moscow) Hirundo abyssinica Striped Swallow J. 22240 f.g. 11.2.69 Kabete, Kenya (at roost). i°i4'S., 36°4s'E. EDS. V 25.2.69 Kabete — flew into house, caught, then released with ring. J. 2294? 20.3.69 Kabete, Kenya (at roost). EDS. 17.5.69 Nairobi (St. Andrew’s Church) i°i6'S., 36°48'E. 13 km SE. (Alfred Mathu). The finder did not note the final digit of the number! Hirundo rustica Swallow J. 19928 juv. /?/ 31.10.68 12. 6.69 Kariobangi, Nairobi. i°i5'S., 36°53'E. GCB. near Chukhloma, Kostroma Region, U.S.S.R. 58°44'N., 42°42'E. (Ringing Centre, Moscow) J. 21474 juv. + 1. 11.68 14.10.69 Kariobangi Nairobi. GCB. Busia, Uganda. o°28'N., 34°05'E. 365 km NW. (P.A.A. Malingu) J. 25893 X 12.12.69 19. 2.68 Lake Nakuru, Kenya. PLB & JFH. Nakuru Sewage Farm. (Njoroge Kiarie) Page 24 J. 25897 juv. /?/ 12.12.68 3. 5.69 Lake Nakuru, Kenya. PLB & JFH. near Tjuratam, Kzyl-orda District, Kazakh S.S.R., U.S.S.R. 45°40'N., 63°i6'E. (Ringing Centre, Moscow) J. 27056 ad. /?/ 28.12.68 26. 7.69 Lake Nakuru, Kenya. PLB. near Tskhaltubo, Georgian S.S.R., U.S.S.R. 55°o8'N., 48°oi'E. (Ringing Centre, Moscow) Riparia paludicola African Sand Martin J. 26119 ad. 26.12.68 Lake Nakuru Kenya. PLB. V 15. 8.69 Kipsigis Tugen Farm, near Nakuru. (David arap Towett) J. 28230 juv. V 28.6.69 15.8.69 Lake Nakuru, Kenya. PLB. as J. 26119: both these birds flew into a house and were caught together. Riparia riparia J. 7066 f.g. /?/ Sand Martin 16.4.67 Entebbe, Uganda. o°05'N., 32°30'E. DJP. 20.6.69 Sorochinsk, Orenburg Region, U.S.S.R. 52°26'N., 53°ii'E. (Ringing Centre, Moscow) Motacilla flava J. 5653 ad. $ lutea + Yellow Wagtail 8.12.68 Kariobangi, Nairobi. GCB. 23.8.69 near Raevskii, Bashkirian A.S.S.R., U.S.S.R. 54°04'N., 54°ii'E. (Ringing Centre, Moscow) J. 10171 f.g. ? flava 12.2.69 Kabete, Kenya (at roost). GCB. (window) 19.2.69 Nairobi, Kenya c. 12 km S. (I. Wahlstrom) J. 14853 ad.(? flava xA 13.2.68 Kabete, Kenya (at roost). EDS. 0.5.69 near Sechenovo, Gorkii Region, U.S.S.R. 55°i2'N., 45°54'E. (Ringing Centre, Moscow) J. 10735 ad. ? /?/ 23.2.69 12.8.69 Kariobangi, Nairobi, Kenya. GCB. near Narym, Parabel District, Tomsk Region, U.S.S.R. 58° 53'N., 81° 36’E. (Ringing Centre, Moscow) J. 29046 ad. ? + 13.3.69 20.6.69 Kabete, Kenya (at roost). EDS. near Yalchiki, Chuvashian A.S.S.R., U.S.S.R. 55° o8'N.,, 48°oi'E. (Ringing Centre, Moscow) J. 18236 ad. $ 111 26.4.69 22.5.69 Kariobangi, Nairobi, Kenya. LC. near Shemysheika, Penza Region, U.S.S.R. 52°54'N., 45°24'E (Ringing Centre, Moscow) Table 3 BIRDS RETRAPPED FROM PREVIOUS SEASONS Charadrius hiaticula ............ 2 Calidris minuta ............. 8 Gallinago gallinago ............. i Philomachus pugnax ............ 10 Tringa glareola ............. i T. stagnatilis ............. 2 Hirundo rustica ............. i Motacilla flava ............. 195 Acrocephalus scirpaceus ............ i Except for some of the Motacilla flava, all the above were retrapped where ringed; some of the M. flava showed movements of a few kilometres between roosts, and roosts and feeding sites. Page 25 KEY TO RINGERS’ INITIALS IN LIST OF RECOVERIES GCB G.C. & D. E. G. Backhurst PLB. P. L. & H. A. Britton PLB & JFH. P. L. Britton & J. F. Harper LC. Mrs. L. Campbell DJP. D. J. Pearson JBS. J. B. Smart EDS E. D. Steel OTHER RINGERS IN EAST AFRICA Miss D. Angwin G. Harrington Miss J. Angwin M. Reid Miss E. Blundell D. A. Turner R. Cheke R. Waterhouse M. P. L. Fogden R. J. Wheater Table 4 RECOVERIES IN EAST AFRICA OF BIRDS RINGED ABROAD This list includes some further additions to the recoveries of birds ringed abroad prior to I July, 1967 (Backhurst, 19696.) The signs and symbols are the same as those used in Table 2. Ciconia ciconia White Stork Zagreb D. 102368 pull. /?/ 15.6.55 18.9.56 Balimbegova, Skopje, Yugoslavia. 42°oo'N, 2i°3o'E. Itobo, Nzega District, Tanzania. 4°59'S., 32°48'E. Zagreb D. 106069 pull. 4. 6.62 5.11.62 Aradac, Zrenjanin, Yugoslavia. 45°25'N., 20°24'E. Karamoja District, Uganda, c. o°45'N., 3i°58'E. Larus fuscus Lesser Black-backed Gull Goteborg D. 45945 Stockholm 8.009.210 pull. X pull. -t- 26.6.60 1. 10.60 5. 7.62 24.9.62 Spjutso, Brakne Hoby, Blekinge, Sweden. 56°o8' N., I5°02'E. Musoma, Tanzania. i°3i'S., 33°48'E. Laggrund, Oregrundsgrepen, Uppland, Sweden, c. 6o°27'N., i8°i8'E. Lake Salisbury, Soroti, Teso District, Uganda. i°40'N., 34°oo'E. Helsinki H-55.916 pull. X 3. 7-66 22.8.68 Gamlakarleby, (Vasa), Finland. 63°5o'N., 23°o6'E. Agwatta, Uganda. i°59'N., 33°oo'E. Helsinki H-93.014 pull. X 13. 7.68 25.H.68 Vasterskar, Tjock (Vasa), Finland. Nyamazugo, Geita District, Tanzania. o°30'S., 32°30'E. REFERENCES Backhurst, G. C., 1969a. “A Record of Gallinago stenura from Kenya.” Bull. Br. Orn. Club 89: 95-96. Backhurst, G. C., 19696. “Report on Bird Ringing 1967-1968.” J. E. Afr. nat. Hist. Soc. 27 :2i7-225 Britton, P. L. & Harper, J. F., 1969. “Some new distributional records for Kenya.” Bull. Br. Orn. Club. 89 : 162-165. Pag^ 26 Mackworth-Praed, C. W. & Grant, C. H. B., 1957; Birds of Eastern and North-Eastern Africa. London, Longmans, Green & Co. White, C. M. N., i960. “A Check List of the Ethiopian Muscicapidae (Sylviinae) Part i.” Occ. Pap. natn. Mus. Sth. Rhod. 3 :24B : 399-430. White, C. M. N., 1961. A Revised Check List of African Broadbills, Pittas, Larks, Swallows, Wagtails and Pipits. Lusaka, The Government Printer. White, C. M. N., 1962. “A Check List of the Ethiopian Muscicapidae (Sylviinae) Parts II & III.” Occ. Pap. natn. Mus. Sth. Rhod. 3 :26B: 653-738. White, C. M. N., 1962. A Revised Check List of African Shrikes, Orioles, Drongos, Starlings, Crows, Waxwings, Cuckoo-Shrikes, Bulbuls, Accentors, Thrushes and Babblers. Lusaka, The Government Printer. White, C. M. N., 1963. A Revised Check List of African Flycatchers, Tits, Tree Creepers, Sunbirds, White-Eyes, Honey Eaters, Buntings, Finches, Weavers and Waxbills. Lusaka, The Govern- ment Printer. White, C. M. N., 1965. A Revised Check List of African Non-Passerine Birds. Lusaka, The Government Printer. {Received 21st January, 1970) Page 27 MIGRATIONS OF THE BUTTERFLIES GLYCESTHIA AUROTA, CATOPSILIA FLORELLA AND CRENIS OCCIDENTALUM IN EAST AFRICA IN 1967-68 By John E. C. Flux Zoology Department, University Cblle^d^ Nairobi and Ecology Division, D.S.I.R.y l^ew, Zealand The complex patterns of insect migration in Africa are not well understood for even the two commonest pierid migrants Glycesthia diirota Fab. and Catopsilia florella L. For a review of existing records and illustrations of these two species see Williams (1958), and for the most recent discussion of the problems involved, Johnson (1969). From July, 1967 to July, 1968 all butterflies which appeared to be miigrating (most fly below 20 ft. and are very conspicuous) were counted for five minutes (Table i) as they crossed an open area 50 yards wide in front of my office at the Botany Department, University College, Nairobi. Field trips away from Nairobi generally took less than one week and, as major migrations usually take longer ;than this, probably few were missed during the year. No attempt was made to select days or times of day of special abundance for the counts, which were made whenever butterflies were sufficiently numerous to be noticed. Wind direction varied and did not seem to affect orientation at all, but most movement occurred in sunshine. The butterflies were identified from collections in the National Museum, Nairobi. ACCOUNTS OF SPECIES Glycesthia aurota , . There was a strong easterly migration from 17 July, to 4 August, 1967 or a little later (I was absent from 5-15 August and migration had ended on my return). Species actively migrating in the same direction included many G. creona Cramer and Precis sp., but several other species seemed to get caught up in this movement occasionally. The average intensity was 64 G. aurota per five minutes across the 50-yard front, and at least the whole of Nairobi (four miles wide) was affected. From mid August till the end of the year few G. aurota were seen, although there was a weak westward migration averaging 15 butterflies per five minutes from 9 to 17 December, 1967 after the short rains. A major movement started on 30 January, 1968 and reached a peak on 3-5 February when over 400 butterflies were passing each five minutes. On 5 February, 1968 I drove from Nairobi to the north side of Mt. Kenya and saw vast nmnbers of G. aurota moving west over the whole route. Mr. A. Walker told me they were just as abundant and moving in the same direction, for at least 20 miles down the Nairobi-Mombasa road. This gives a north-south front of no miles, and assuming a six-hour day (most activity was between 1000 hrs and 1600 hrs), a total of in million butterflies per day. A few G. aurota were moving west over alpine moorland at 13,000 ft. on the north side of Mt. Kenya, and eight dead ones (six male and two female) were picked up which had presumably been killed by frost at night. The movement had almost ended by 9 February when I returned to Nairobi. In the second half of March 1968, there was a small migration to the east or east north-east, but no others before I left Nairobi on 31 July, 1968. Page 28 Table i FIVE MINUTE COUNTS OF BUTTERFLIES CROSSING TO RIGHT OR LEFT OVER A 50 YARD LINE OF SIGHT AT RIGHT ANGLES TO THE DIRECTION OF MOVEMENT, AT CHIROMO, NAIROBI Date No. Butterflies and direction Glycesthia aurota 17. 7.67 79 flying E 3 flying W 18. 7.67 56 „ „ I 28. 7.67 30 » « I yy 31. 7 67 66 ,, ,, 2 I. 8.67 60 „ „ 2 4. 9.67 9 » » I fy 9.12.67 2 ,, ,, 16 12.12.67 1 » » 4 17.12.67 6 „ „ 14 30. 1.68 2 „ „ III yy 31. 1.68 0 ,, ,, 2 I. 2.68 2 „ „ 26 yy yy 2. 2.68 142 yy 3. 2.68 2 „ „ 362 yy yy 4. 2.68 5 3> 5) 493 yy yy 13. 2.68 9 flying SSE 7 flyinj jW 18. 2.68 7 » S 46 yy 16. 3.68 23 „ ENE 2 yy yy 26. 3.68 15 » E I yy 27. 3.68 13 » » 0 yy 31- 3-68 32 » » 0 >y Remarks Several other species also Very few for next month Very few for next 3 months Several other species New emergence. None all January Direaion rather confused New emergence over past few days Includes 2 Precis Includes i precis. Few for next 4 months Catopsilia florella 3. 1.68 13 flying NE 6. 1.68 10 „ „ 14. 1.68 19 „ „ 15. 1.68 17 „ „ 16. 1.68 24 „ „ 17. 1.68 16 „ „ 30. 1.68 2 „ „ 31. 1.68 21 „ „ 1. 2.68 I „ „ 2. 2.68 3 „ „ 2 flying SW Catopsilia florella Only one migration of this species occurred during the year, from 3 January to 2 February, 1968, averaging 13 (max. 24) butterflies per five minutes. Two features were noted: the butterflies preferred to fly up and over the three storey building across their path rather than detour round it; and from 30 January to 2 February their north-easterly path crossed the far stronger westerly one of G. aurota. This provided the fascinating spectacle of completely independent streams of insects flying in different directions, as described by Wilhams (1958) for C. florella and Terias senegalensis Boisd. Crenis occidentalum Mabille (Nymphalidae) On 30 June, 1968 from 40 miles north to seven miles west of Fort Portal, Uganda, C. occidentalum was flying south-east against a moderate wind. The butterflies kept within five feet of the ground and at their thickest 205 were counted in five minutes crossing a 50 yard space in front of the car. Since we drove from Hoima to Queen Elizabeth Park this day, the approximate width of the migration (40 miles) was known, but not its duration. Three female and four male butterflies were picked off the car radiator, so both sexes were involved. Page 29 ACKNOWLEDGEMENTS I am grateful for the New Zealand National Research Fellowship which enabled me to work in East Africa; to Professor D. S. Kettle for providing facilities; and to Drs. M. C. Crawley and J. A. Gibb for commenting on the manuscript. REFERENCES Johnson, C. G., 1969. Migration and dispersal of insects by flight. Muthuen, London 766pp. Williams, C. B., 1958. Insect Migration. Collins, London. 237 pp. {Received 17th February, 1969) Page 30 OBSERVATIONS ON BUTTERFLY MIGRATION AT ENTEBBE, UGANDA By S. D. K. Sempala {East African virus Research Institute^, Entebbe Uganda) INTRODUCTION Bird migration has been recognized as a to-and-fro movement for a long time, but this type of movement is recorded for relatively few species of butterflies. The best known of these are the Monarch {Danaus plexippus L.) of North America and the Red Admiral {Vanessa atalanta L.) and Painted Lady {V. cardui L.) of the Old World. Records of migratory movements in these and many others, amounting to a total of 214 species, are summarized by Williams (1930) who defines insect migration as follows: “It is a periodic, more or less unidirectional continued movement assisted by the efforts of the animal and in a direction over which it exerts a control, which results in the animal passing away from its previous daily field of activity.” It should be noted that Williams’ definition does not state that migration is necessarily a two-way movement, although he also discusses the importance of observations on return flights which, in contrast to gregarious outgoing flights, are often very thinly spread and therefore likely to be overlooked. The butterfly movements reported below at Entebbe were largely unidirectional, the insects flying consistently against the wind. They represent examples of trans- migration. (Williams, ibid.) OBSERVATIONS Sites: All observations were made in the area of Entebbe town and its surroundings within a distance of five to ten miles north of the equator. Period of observations: Observations during the time of migration were conducted from the beginning of June to the end of the first week of July, 1967. Daily casual observa- tions were made up to the end of the year and throughout 1968.* The detailed counts were all made on the compound of the East African Virus Re- search Institute on an extensive open area of mown lawn. Other occasional observations were made on the golf course, the airstrip, Kigungu beach which faces south into the open lake and on the lake about three hundred yards from the shore at Kigungu. Methods: At the principal site a front of 35 yards running East-West was established, and insects flying across this line were counted and their direction noted. This front lay more or less at right angles to the main direction of flight. RESULTS Species represented: Specimens were taken of the following migrating species : — Belenois creona Cr. (Pieridae) Catopsilia florella Fab. (Pieridae) Papilio demodocus Esper (Papilionidae) *No further migration has been apparent at Entebbe up to July 1970. Page 31 FIG. 1 Belenois creona FIG. 3 Papilio spp. Page 32 Records of these and of the conspicuous Papilio bromiusjnireus group were made. Towards the end of June and in early July some large yellow Pierids were evidently migrating, but as these flew high and evaded identification, records are not given. B. creona v;as by far the commonest migrant. It is possible that closely similar Belenois species may have been represented, but out of a total of 70 specimens collected over six occasions with at least one sample per week through June (apart from the third week when overall numbers were very low), all were of the single species B. creona. Well over 50 per cent observed were males. C. florella was also fairly common, particularly towards the end of the period of observations. Other species recorded were relatively scarce, though definitely on the move. Resident species of Eurema, Neptis, small Lycaenids and others, though quite commonly seen, showed no consistently directional flight. Other butteries could have been migrating elsewhere on the Entebbe peninsula, but none was seen at the sites used for regular observations. Migratory movements: Overall period: Considerable migration of butterflies had been in progress at Entebbe for at least two weeks before detailed observations began. Figures i, 2 and 3 do not therefore represent the full period of migration, but they do show the relative intensity of migration during the period of the observations. There was no butterfly migration across Entebbe peninsula during the extended period of observation. Time of day of flight: In order to obtain an indication of the period of commencement, peak and cessation of flight activity, counts lasting for five minutes each were conducted every hour throughout the day (on 14 June) which was calm and sunny. The results are given in the table. Table RESULTS OF A 5-MINUTE COUNT IN EACH HOUR THROUGHOUT THE DAY ON THE 14th JUNE (ON A 35-yd FRONT) TIME (Solar time) BUTTERFLY SPi s CIES TOTAL BELENOIS CREONA CATOPSILIA FLORELLA PAPILIO SPP. 0700 — 0705 0 0 0 0 0800 — 0805 0 0 0 0 0900 — 0905 0 0 0 0 1000 — 1005 9 0 0 9 1100 — 1105 13 2 2 17 1200 — 1205 17 4 I 22 1300 — 1305 67 3 I 71 1400 — 1405 128 0 0 128 1500 — 1505 63 0 0 63 1600 — 1605 0 0 12 1700 — 1705 I 0 0 I 1800 — 1805 0 0 0 ° TOTAL .. 310 9 4 323 The difference in the periods of peak activity in the species under consideration are hardly significant, as the number of C. florella and Papilio spp. were too small. The results nevertheless demonstrate that flight activity was distinctly diurnal leading up to, and from, midday to early afternoon (solar time) maximum in B. creona. DISCUSSION It is noteworthy that all species of migrating butterflies observed showed a marked southerly bias in flight direction which would have taken them out into the open lake. However, the method of choosing a front which lies more or less at right angles to the main direction of flight, and then counting butterflies flying across this line is likely to yield quantitatively biased results. This was not realized at the time and the results given here should be viewed accordingly. When movement is entirely unidirectional it would seem obvious that to assess flight activity one would follow the above method. On the other hand, for insects which diverge from the main directional axis, the real front is reduced as their direction of flight approaches that of the front. Thus, if one considers insects flying at random across an East- West line, then none of those flying due East or West would be recorded at all, and the resultant directional chart would be shaped like a figure-of-eight. To eliminate such bias it is suggested that an area should be staked which subtends 6o° of a circle. The observer should stand at the centre of the circle, and insects counted and their direction noted at the points at which they enter the area. Such a method would be feasible when conspicuous butterflies are under observation. The radius of the area may be varied depending on observability. If the tendency to migrate southwards is genetically determined, then one might assume that migration must be of considerable advantage to persist in such a spectacular form as indicated by the numbers observed. These butterflies were migrating not only at Entebbe during this period; on 27 and 28 May Belenois butterflies were seen in very great numbers flying across the Kampala to Mbarara road, a distance of 165 miles, in a south and south-eastwards direction (J. Kingdon, pers. comm.) and the same movement was observed on 3 June and it was estimated that more than 95 per cent were flying towards the lake (A. W. R. McCrae, pers. comm.) By ii June numbers appeared to be much fewer and direction less consistent (idem). The observation that flight activity was distinctly diurnal with a midday to early afternoon maximum in B. creona raises the question of whether the butterflies would rest or attempt to rest while over the lake. There are numerous examples of butterflies and other insects being seen at great distances out to sea. R. H. Carcasson (pers. comm.) has personally seen swarms of V. cardui in the Mediterranean and C. florella in the Arabian sea, some five hundred miles from the nearest land. Tliis suggests either that they are capable of resting on and taking off from the sea* or that over open water they do not cease flying. This may be a question of assistance by winds and by endurance in terms of energy reserves. Hocking (1953) calculated that owing to the high efficiency of a food reserve in the form of fat, many lepidoptera are capable of flying several hundred miles without feeding. A proportion of the butterflies which head over the lake from Entebbe and other places may nevertheless perish before they reach land, and such migrations could be suicidal. It would be of great interest to have observations from the southern and south-eastern shores of Lake Victoria, e.g. from Mwanza. The causes and objectives of such migrations still remain a matter for speculation. B. creona is a very widespread and highly mobile species inhabiting dry country, as numerous records of migratory flights show (Williams, 1930). The food plant of this f wllich^oc£?m nda ?gdy in^ry e north-west of the country. Pitman ( 1928, in Williams, 1930) I : in the lowlands of West I j the ^ s myriads of b Fage34 shimmer of white just above the surface of the ground he did not observe direc- tional movements at the time. It was clear that this population was extremely crowded and food for the ensuing larvae would have been limited, therefore necessitating either very high mortaUty or dispersal. Food supply in such dry areas is seasonal, and such locally abundant species may only survive periods of food shortage either by a period of dor- mancy or by migration. A closer study of the biology of migratory Belenois should illuminate these extremely interesting problems. SUMMARY Observations on migrant butterflies at Entebbe are described. B. creona, C. florella, P. demodocus and the Papilio bromiusinireus group were the species represented during the course of migration and they all showed a marked southerly bias in flight direction. ACKNOWLEDGEMENT The author is greatly indebted to Dr. M. C. Williams, formerly Director of the East African Virus Research Institute, Entebbe, for suggesting this study. I would also hke to thank Mr. A. W. R. McCrae for his encouraging advice and assistance in preparing this manuscript and Mr. C. Mawejje for his valuable field assistance. REFERENCES Hocking, B., 1953. The intrinsic range and speed of flight of insects, Trans. R. ent. Soc. Land. 104, 223-345. Pitman, C. R. S., 1928. “The area in the West Nile Province of Uganda from which start the great southward migrations of Belonois mesentina Cram, in Uganda and Kenya.” Proc. ent. Soc. Land. 3, 45-46. Williams, C. B., 1930. Migration of Butterflies. Edinburgh, Oliver and Boyd. (Received 12th May, 1969) Page 35 AN ANALYSIS OF THE FEATURES OF SARDINELLA GIBBOSA (BLEEKER) SCALES, WITH SPECIAL REFERENCE TO THE PROBLEM OF AGE DETERMINATION By W. Okera Department of Zoology & Marine Biology, University College, Dar es Salaam INTRODUCTION Five species of Sardinella have been recorded in East African waters (Losse, 1968), S. gibbosa and 5. albella (Valenciennes) being the most common in inshore waters of the Tanzanian coast. This investigation on the scales of S. gibbosa was a product of an attempt to find out whether there was any possibility of reading from the scale of the age of this species of sardine from its scales. In temperate waters where the seasons are well marked, many of the piscine struc- tures, e.g. scales, otoliths, vertebrae, dorsal spines, pectoral spines and opercular bone, show well defined differences in their growth. The seasonal growth zones in scales and otoliths have provided a relatively easy and quick way of assessing the absolute ages of commercially important fishes such as herring, cod, plaice, and salmon with a satis- factory degree of accuracy. Beside these growth rings which are formed as a result of the accelerated and retarded growth processes during spring and winter respectively, “the scales of many fishes show spawning rings and marks which are the result of the cessation of feeding and exhaustion during the spawning period”, (Nikolsky, 1963, page 194). Therefore if the fish spawns once a year and at the same time leaves a spawning ring on its scales, the number of these rings can be used to read the age of the fish since it first started spawning. The biology of East African S. gibbosa and other local sardines is very poorly known. For example, it is not known at what len^ or at what age the local S. gibbosa attains sexual maturity. Its maturation and spawning habits are also not known. For the Indian S. gibbosa, Nair (1959) mentions that Chacko (1946) found S. gibbosa to attain sexual maturity in the Palk Bay and Gulf of Manar regions when it reaches a length of 14 cm. whereas Sekharan (1955; quoted from Nair’s paper) “showed that the fish below 8.5 cm. are immature and a proportionate increase of mature fish was observed in the higher size groups, with most of the sardines of the 10.7 cm. size group being mature.” Sekharan also concluded that maturity is attained at this size at the end of first year. Ganapati and Rao (1957; quoted from Nair’s paper) working on S. gibbosa of Lawson’s Bay, Waltair, found that the species attains maturity at an average length of ii to 12 cm. As for the maturation and spawning habits of this species of sardine, Dharmamba (1959) showed that S. gibbosa of Lawson’s Bay has “a single spawning season a year” and “spawning is restricted to a definite short period with individuals spawning only once in each season.” Losse (unpublished data) says that, in the Zanzibar Channel, the spawning season of 5. gibbosa extends from April to October, with the smaller fish being first spawners and spawning early in the year (April/May) and the largest spawning in August to October, after the main spawning peak of June/July. From Losse’s observation, it is perhaps justifiable to say that the local S. gibbosa has a single spawning season a year and that generally, individuals belonging to different size groups have a restricted spawn- ing period at different times during the spawning season. Therefore if S. gibbosa spawns once a year, it is worth investigating whether the annuli seen on its scales result from the spawning activity of the fish or are caused by other unknown factor /facortors. Page 36 At first it was found necessary to investigate and ascertain the nature of other conspicuous features of the scale, namely the striae and the fissures, so as to eliminate any possibility that they may be associated in some way with spawning. MATERIAL AND METHOD For all examinations, the specimens were obtained from a local, Japanese-built, sardine-fishing vessel operating the Japanese stick-held dipnet, the “Bouk-ami”. Material for histological examination was fixed in Allen’s B15 (1500 ml. saturated aqueous picric acid -f 500 ml. 40 per cent formalin + too ml. glacial acetic acid + 40 gm. urea -j 30 gm. chromic acid) immediately after the fish were caught. Fixation was allowed to take place for about twenty-four hours and the fixed material was then washed thoroughly with tap water. Pieces of the skin with a little underlying muscle were then cut — each about 2 cm. long, 3 cm. deep and 2 to 3 mm. wide — and processed as follows : — Dehydration to 70 per cent alcohol — decalcified in a solution of 5 per cent HNO3 in 70 per cent ethanol for 12 hours — dehydrated to too per cent alcohol — left overnight in 8p methyl benzoate/celloidin solution — blocked in wax and sectioned at 8p thickness. The sections were stained with Masson’s trichrome method, using iron Haematoxylin, Xylidine ponceau — acid fuchsin mixture and light green solution. Scales for whole mount examination were first stained with carmine and mounted with Canada Balsam, but this method proved faulty as some of the features were masked with the stain and the mounting fluid. The method was abandoned in preference to mounting the scales dry between two glass slides taped at the ends. The latter method was also proved useful for the Northern achovy, Engraulis mordax and the Pacific sardine Sardinops caerulea (Girard) (Miller, 1955). THE SCALE Examination of S. gibbosa scale (fig. i) shows several features whose nature has received a sparse and inadequate mention in the literature: (i) The circuli or striae: — These are the very fine transverse lines that run from dorsal to ventral* sides of die scale, nearly parallel to each other. Sometimes, one or a few may end abruptly or one stria may fuse with another. In longitudinal section (fig. 2) the striae can be seen to form very fine ridges in the bony layer. The posterior region of the scale which is not embedded in the scale pocket lacks the surface sculpture of striae; instead there are a number of round or oval perforations and growth patterns unique to this region. The posterior margin of the scale is crenulate. As to the function of these striae. Van Oosten (1957) says “they are probably func- tioning in anchoring the scale in its pocket”. He makes no suggestion as to how this may be achieved. Kerr (1952) makes the following comment about the longitudinal calcified ridges of Amia calva L. scale: “Possibly along with anchoring strands of connective tissue from the rear of the scale, they serve to maintain the position of the scale, to allow flexibility under normal circumstances and yet to permit the tearing out of a scale by a predator without excessive damage to the skin. Such a general explanation might apply also to the concentric striations of the teleost scale . ...” In the light of modern studies on the reflectivity in silvery teleosts (Denton & Nicol, 1966) one may also postulate that the striae may play a part in the visibility of the fish. For instance, it would be relevant to investigate the light transmitting properties of the striated and non-striated parts of the scale, and also the relationship of the reflecting layers (stratum argenteum and layers of * The lateral sides of the scale are referred to as the dorsal and ventral with respect to the scale’s orientation on the body of the fish. Page 37 orientated reflecting platelets) to the two distinct regions of the scale. All these suggestions need to be subjected to the rigours of experiments, before the functions of the striae can be fully understood. (ii) The posterior continuous fissure and the dorsal and ventral fissures: — These fissures, the dorsals, ventrals, and the posterior are represented by thick lines in fig. I, and labelled dvf and pf in fig. 2. No work to my knowledge exists on the nature and formation of these fissures. Photographs of the scale of Caspialosa kessleri Grimm, (in Nikolasky’s The Ecology of Fishes, 1963, page 194) and Engraulis mordax, (in Miller’s 1955 paper) show structures which appear to be similar but not identical to the fissures under consideration; no mention of these structures is made in the text.. Eighty scales from twenty-five individuals of 5. gibbosa were removed from the lateral body wall, in the region behind the dorsal fin, about 1/3 to 1/2 from the dorsal profile of the body. All of them showed one general pattern; a posterior continuous fissure and several dorsal and ventral fissures (fig. i). Often, the scales have an equal number of the dorsal and ventral fissures which are symmetrically arranged. For example, of the 80 scales examined, 43 had an equal number of the dorsal and ventral fissures and the remaining 37 had an unequal number of dorsal — ventral fissures. In another fish, 43 scales were removed from a single scale row (hatched line, fig. 5), begin- ning with the scale behind the humeral spot and ending with the one preceding the Table i, showing the number of dorsal and ventral fissures of 43 scales removed from a single scale row, (hatched line, fig. 5) Scale Number of fissures Length of scale row occupied by consecutive five Scale Number of fissures Length of scale row occupied by consecutive five Dorsal- Ventral ces measured from humeral spot. Dorsal-Ventral ces measured from humeral spot. I 2 — 2 21 5—5 ''l 2 I — I 22 6—5 1 3 3—3 yo — icm 23 5—4 V4.85-6.2cm 4 5—3 24 6-4 I 5 4—3 J 25 6-5 J 6 3—4 1 26 5—5 1 7 3—3 I 27 6-5 1 8 4—3 >I 2.2cm 28 6—5 V6.2-7.4cm 9 4—3 [ 29 7—6 1 10 5—3 J 30 7—5 J II 4—3 1 31 7—5 12 5—3 1 32 7—6 1 13 4—3 !»2.2-3.65cm 33 5—5 V7.4-8.5cm 14 5—3 1 34 5—5 15 4—4 J 35 5-6 J 16 5—5 1 36 6—4 1 17 4—4 37 6—6 18 3—5 ^3.65-4.85cm 38 5—4 J-8.5-9.75cm 19 6—4 1 39 7—7 20 6—4 J 40 6-4 J 41 5—5 1 42 5—3 J-9.75-10.55cm 43 4—4 J Page 38 Table 2 Number of annuli Length of fish in mm 0 117(1), ii8(i), 119(1), 123(1), 125(2), 126(3), 127(1), 128(1), and 131(1). I ii8(i), 122(1) Table 3 gives the result of annulus reading of samples of fish taken during the months of January, March, April, and May, 1969. For convenience, the individual lengths of fish have been omitted and instead the fish have been put into 5mm size groups. irregular scales of the caudal peduncle. Again all of them showed the basic pattern — a posterior continuous fissure and several dorsal and ventral fissures. Of the 43 scales, 15 had an equal number of the dorsal — ventral fissures, while the remaining 28 had an unequal number of dorsal — ventral fissures, there being not more than 2 extra fissures on either the dorsal or ventral side of the scale (see Table i). Scales from the caudal peduncle, from the region immediately behind the opercle and from the dorsal profile showed an irregular pattern of these fissures. After examination of all these scales, it became apparent that these fissures had nothing to do with spawning, simply because spawning marks are left as circular scars rather than transverse marks. Further, longitudinal sections of the scale (fig. 2) revealed that the fissures were filled with a connective tissue material that was staining green, just like the walls of the scale pocket, suggesting that the fissures were probably passages for strands of connective tissue that helped to hold the scale in its pocket. Perhaps they also serve to increase the scale’s flexibility. An explanation as to how these fissures may be formed was sought by studying the scale in relation to its surrounding features. The imbrication of the scales (fig. 3, 4) seemed to be a possible cause for the fissures. Fig. 4 shows the pattern of the anterior edges of the scale pockets, and it is from these anterior edges that the dorsal walls of the scale pockets emerge from the body wall. Considering the scale M (fig. 4), it seems that as this scale grows outwards and overlaps part of the anterior edges, li, I2, ri, r2, (shown by thick lines in fig. 4) of the scale pockets flanking it dorsally and ventrally, connective tissue from li, I2, ri, and r2, grows through scale M which overlies li, I2, ri, and r2. This explanation is based on the observation that the only structure on the body wall that can be likened with the form (i.e. orientation, thickness and length) of the dorsal and ventral fissures are the overlain portions of the anterior edges of the scale pockets, (see fig. 4). However, there are several difficulties involved in this hypothesis. Firstly, a scale usually overlies the portions of anterior edges of two pockets on its dorsal and ventral sides, whereas the number of dorsal and ventral fissures found on a scale varies from two to eight on either side. That is, a scale may have two dorsals and two ventrals or four dorsals and five ventrals or eight dorsals and six ventrals and so on. Secondly, although the formation of dorsal and ventral fissmres may be explained by considering the bordering scale pockets, the formation of the posterior continuous fissure does not seem to be exphcable on the same basis. Thirdly, it is difficult to explain the pattern of fissures of the scale of E. mordax (assuming that the lines and splits seen in the photographs of E. mordax scales are similar to the fissures imder consideration) or of the caudal peduncle scales of S. gibbosa by the growth — imbrication theory. Table 3 Page 39 - - 1 - - - - - - - - - V - - - g 1 - VO - - s § hj - c/o p-l p 0 0 - - - - b w N c/5 s - - - 8 0 - - - 0 0 - - NUMBER 1 OF ANNULI 0 1 i-doubtful 1 i-clear 0 I i-clear 1 i-clearj 2nd doubtful 2-clear 0 I I -doubtful 1 I-clear 1 i-cleatj 2nd doubtful 0 1 I -doubtful [ I-clear I-clear, 2nd doubtful! 1 MONTH AND NUMBER OF FISH EXAMINED JANUARY, 1969 24 fish MARCH, 1969 25 fish APRIL, 1969 tc MAY, 1969 25 fish 2-cIear Page 40 Whitear (pers. comm.) points out that the fissures might correspond to the canals penetrating the scales of more primitive actinopterygians which usually carry blood capillaries and nerves. Further histochemistry is needed to ascertain the presence of these structures in the fissures of S. gibbosa. Thus the problem of explaining the formation of these fissures still remains unsolved and a solution could perhaps emerge by studying the development of these scales. Before leaving the fissures, it may be mentioned that Losse (1966, 1968) refers to these fissures as striae, whereas Van Oosten (1957) defines striae as “the reheved^* smrface ridges that are continuous and homogenous with the bony layer of the scale”. As mentioned earlier, the striae are the very fine and numerous transverse lines that are seen in a whole mount of the scale and in L. S. as fine ridges in the bony layer. Van Oosten (1957) does not mention the fissures^* although they are of importance in classi- fication of the Clupeidae (Losse, 1966, 1968). (iii) Annuli In fig. I, these are shown by the two narrow and light coloured bands that are concentric with most of the periphery. Miller (1955) refers to these rings in E. mordax scales as “typical clupeoid annual checks”. He uses the definition of an aimulus as given for the Pacific sardine, Sardinops caerulea (Giaard), by Walford & Mosher (1943b) as a basis for annulus interpretation in E. mordax. It seems that this definition which is as follows, also applies to annulus recognition in S. gibbosa. “An annulus is concentric with the margin of the scale. It is not always a sharp or unbroken line; nor are the segments of an interrupted annulus always perfectly co- circular (if the shape of a scale may be called circular in this discussion). But the course of an annulus, continuous or broken as it may be, can usually be traced, by careful scrutiny if necessary, entirely around the sculptured part of the scale from left hand to right hand margins. Sometimes they can be followed even around the unsculptured part. Annuli are clearly separated from each other and do not ordinarily meet at any point. If an annulus has formed it is present in all the normal scales of an individual”, (quoted from Miller, 1955)* Table 4 Month Percentage c d omposition of fish in the sample witl ifferent number of annuli 1 o-annulus I -doubtful I -clear i-clear 2nd doubtful 2-clear 2-doubtful January 1969 75 12.5 12.5 0 0 0 March 1969 72 0 16 4 8 0 April 1969 76 12 8 4 0 0 May 1969 36 16 24 4 16 4 *1 = relief. His usage is not correct. Surface ridges would have been sufficient. *2 = unless his radii, defined as “open channels cut completely through the bony surface”, are equivalent to what I have called fissures. Page 41 It may be mentioned that the annual growth rings on scales of temperate fishes such as salmon and trout are also called annuli. The annuli in these fishes are formed by the circuli being closer together at certain times of the year than at others. De Bont (1967) mentions that Holden (1955) described the annuli of Tilapia variabilis Boulenger and r. esculenta Graham of Lake Victoria as rings which are formed by irregular circuli. In S. gibbosa, the annulus is formed by the striae being either interrupted, indented or bent in the annular region and the spaces between the striae are “lighter” in the annular region compared to the rest of the striated region of the scale. In order to find out whether the Walford and Mosher definition of an annulus also holds for S. gibbosa scales, three individuals of the sardine were examined in detail for annuli marks on their scales. 259 scales were removed from regions i to 12 (fig. 5) of a 125 mm. Standard Length (S. L.) fish, and nearly all of them had no annulus. In a second individual of 132 mm. S. L., 151 scales were removed from regions i to 12, and nearly 90 per cent of the scales showed a clear formation of a single annulus. In the third specimen 129 mm. S. L., 153 scales from regions i to 12 were examined and 70 percent of the scales showed a clear formation of two annuli. Fourteen other individuals (a very small part of the catch taken on the night of 26 November, 1968) were examined for aimulus marks. In these cases about 12 scales only were removed from regions 5 and 8 (fig. 5) of every fish and the scales were then examined for annuli. Table 2 shows the result obtained; the figures in parentheses give the number of fish in that size group. Table 4 gives the percentage of fish in the sample with different number of annuli for the same months viz. January, March, April and May 1969. Table 5 has been included to show that it is also possible to read the number of annuli from S. albella scales. Sixteen individuals were obtained from a mixed subsample of about a thousand fish, the latter being part of the catch taken on the night of 17 February, 69. Table 5 Number of annuli Length of fish in mm. 0 72(1), 91(1), 98(1), 99(3), loi(i), 102(1) 103(2). I doubtful 104(1) I clear 105(1) I clear, 2nd doubtful 97(1), 129(1) 2 clear 102(1), 103(1). Anterior Fig.l Dorsal Ventral Fig. I — Surface view of 5. gibbosa scale. The directions indicate the orientation of the scale relative to the orientation of the fish. Fissures are shown in thick lines and the two annuli by the two light incomplete circular bands. X 30. -Posterior- Fig. 2 — Longitudinal section of S. gibbosa scale, showing the posterior fissure and one set of the lateral fissures. Key: pf — posterior fissure; If — lateral fissures; bl — bony layer of the scale; fl — fibrillary layer of the scale, x 6o. | 3 Dorsal Fig. 3 — Diagrammetic representation of the surface view of the arrangement of the scales from the middle of the lateral body wall of the fish. Scale i overlaps scale i R and i L (dotted areas). Scale I itself is overlapped by 2 R and 2 L (areas with crosses), the latter two scales being overlapped by scale 2 which is itself overlapped by 3 R and 3 L and so on. Posterior Anterior Fig. 4. Dorsal Ventral Fig. 4 — Diagrammatic representation of the surface view of the anterior edges of scale pockets from middle of the lateral body wall of the fish. Four scales are shown with dotted outline. For 11j h> rij rgj 2nd M, see text. Posterior Fig. 5 Fig. 5 — Division of the lateral body wall into twelve regions from which scales were removed for annuli studies. The hatched lines represents the scale row from which 43 scales were removed for fissure investigation, x i. This preliminary investigation on the annuli gives the following indications : — 1. The definition of an annulus as given by Walford and Mosher for 5. caerulea can also be used to identify annuli in S. gibbosa and S. albella. 2. In S. gibbosa fish showing no annulus varied in length from 86 mm. to 131 mm., a range which nearly overlaps that of those showing one annulus (Table 3). On the other hand, the same table shows that fish with two aimuli on their scales have so far been found only in the higher size groups, none of them being less than 123 mm. in length. 3. For the months of January, March, April and May, 1969, the percentage of fish with no annulus is higher than of those with one annulus and the proportion of the latter is greater than of those with two annuli (doubtful cases have been omitted). Until the causation of the annulus is established, it will be perhaps hazardous to extract any information from these figures. 4. If annuli on S. gibbosa scales are yearly registered marks, then the few year-groups ( ?) observed probably fits in with the other properties, viz. early maturity and annual spawning, of fish with a short life cycle (Nikolasky, 1963, p. 217). Several workers have tried to investigate what induces the formation of annuli in some other fishes. For E. mordax. Miller (1955) says “that these rings are not spawning checks is shown by the presence of an aimulus on young immature fish and by the pre- sence of new rings already forming on the scales of most fish collected just before and dining the spring spawning.” Further, he establishes that these checks result from the retarded growth during late summer and fall, whereas most of the spawning is confined to winter, spring and early summer, the latter two periods being also the time when maximum growth of the fish and its scales takes place. Newell (1957) mentions that F. Talbot found that “in the fishes of East African waters, a zoning is often present in the scales of many species. In Lethrinus nebulosus (Forskal) a check in scale growth has been found in September, the month of minimum temperatures”. On the other hand. Van Someren’s work (1950; quoted from De Bont, 1967) on rainbow trout, Salmo gairdneri, living in the rivers of the foothills of Mount Kenya, showed that the annuli are spawning marks. In Tilapia macrochir Boulenger, De Bont (1950; quoted from his 1967 paper) found no correlation between spawning and the laying of annuli on the fish’s scale. For S. gibbosa work is in progress to find out whether annuli are spawning rings or whether they develop as a result of some other cause of unknown periodicity. Only Page 46 when this latter point is cleared, would it be possible to say whether or not annuli can be used as age indicators. For the moment, it seems that they are probably the best structures yet available as tentative indicators of age. SUMMARY 1. Scales of S. gibbosa have been investigated in an attempt to find an ageing method for the fish from these structures. 2. The nature of the striae, posterior, dorsal and ventral fissures is ascertained. A suggestion is made as to how the fissures may be formed, although several difficulties are involved in the hypothesis. 3. Annuli of the clupeoid type have been observed and the validity of their use in age determination is the subject of further investigation. ACKNOWLEDGEMENTS I am greatly indebted to Dr. Mary Whitear, University College, London, for her assistance in going through the manuscript and in making quite a number of useful comments. I also wish to express my thanks to Mr. G. F. Losse for having allowed me to use some of his unpublished biological data. Prof. I. Griffiths for many helpful sugges- tions and for lending me literature. I am grateful to the Research and Publications Committee, University College, Dar es Salaam, for giving me a grant to pursue the work on the local sardines. REFERENCES De Bont, a. F., 1967. Some aspects of age and growth of fish in temperate and tropical waters; p. 67-88. In Gerking’s (Ed.) The Biological Basis of Freshwater Fish Production. Blackwell Scientific Publications, Oxford and Edinburgh. Denton, E. J., and Nicol, J. A., 1966. A survey of reflectivity in silvery teleosts. J. mar. biol. Ass. U.K. 46, 3 :685-722. Dharmamba M., 1959. Studies on the maturation and spawning habits of some common clupeoids of Lawson’s Bay, Waltair, Indian J. Fish. 6, 2:374-88. Kerr, T., 1952. The scales of primitive living actionpterygians. Proc. zool. Soc. Land. 122, Part 1:55-78. Losse, G. F., 1966. Check list of elopoid & clupeoid fishes in East African coastal waters. J. E. Africa nat. Hist. Soc. 25, No. 3(112) : 166-178. , 1968. The elopoid & clupeoid fishes of East African coastal waters. J. E. Africa nat. Hist. Soc. 27, No. 2(117) : 77 - 115. Miller, D. J., 1955. Age determination of the Northern anchovy Engraulis mordax. State of Calif . Fish Game, Bull. No. loi. Nair, R. V., 1959. Synopsis on the biology & fishery of the Indian sardines. Proc. World Sci. Meeting on Biology of Sardines & Related Species. FAO Rome. Nikolasky, G. V., 1963. The Ecology of Fishes. Academic Press (English translation). Van Oosten, J., 1957. The skin and scales; in Brown’s Physiology of Fishes, i :207-243. New York. {Received 8th July, 1969) NOTES ON KENYA ACETABULARIA LAMOUROUX, (CHLOROPHYTA) By Shakuntala a. Moorjani University College, Nairobi Acetahularia is a pan-tropical genus with extra-tropical extensions. It comprises about twenty species which are all marine. The mature thallus is calcified (in varying degrees according to the species) and has an erect, unbranched siphonous stipe terminating in one or more whorls of gametangial rays. The rays may be free or fused along their lateral margins and contain numerous spherical cysts at maturity. On the upper surface and near the base, each ray bears a coronal knob, which knobs jointly comprise the corona superior. Some species also have a corona inferior below the game- tangial rays. The corona superior bears delicate, decidous hairs, which in certain species are rudimentary. On the east coast of Africa, Acetahularia has been recorded in South Africa (Levring, 1938; Papenfuss & Egerod, 1957) and in Mozambique at Inhaca Island and Peninsula (Isaac & Chamberlain, 1958). As far as the author has been able to ascertain from the records available to her, the genus Acetahularia has not been previously recorded for the Kenya coast. In this paper three species of Acetahularia are reported for the Kenya coast. The first of these Acetahularia was recorded in April, 1968. A piece of old coral on which Udotea orientalis A. & E. S. Gepp, were growing was collected on the coral reef at Diani Beach and brought to Nairobi to make further observations on this species. On this piece of old coral there appeared an Acetahularia which was identified as A. moebii Solms-Laubach by Prof. Wm. E. Isaac. Again, earlier this year, some Acetahularia were observed on old coral in culture of U. orientalis. The laboratory culture consisted of three pieces of old coral which had been collected on the reef at Diani Beach on 4 April, 1969 and transported to Nairobi in a Polythene jar filled with sea water, from which they were removed a week later and transferred to a glass aquarium (12 in X 7^ in x 7I in) which was filled with filtered sea water brought from the coast. Two petri dishes, full of clean sand from the coast, in which individual Udotea had been planted, were placed in the aquarium. The level of water in the aquarium was marked. The aquarium was placed on a bench near a window. Daily from 8.30 a.m. to 4.30 p.m. the aquarium was exposed to illumination provided by a 100 watt Bench lamp which was placed on one corner of the aquarium. The heat from the lamp also helped to raise the temperature of the water from 18° to 24°Cand maintain it for some time. Aeration was provided by means of an aerator with pol3rthene tubing ending in an air stone-diffuser. Care of the culture involved only the adding of distilled water from time to time to replace the water lost by evaporation and thus maintain a constant volume and salinity. Algae growing on the walls of the aquarium were removed. The first Acetahularia were seen growing a week after the culture had been set up in the laboratory. In the following weeks unbranched axes with terminal whorls of sterile hairs were noticed which later developed the characteristic disks of gametangial rays. The developmental sequence was followed and was found to be similar to the one describ- ed by Egerod (1952). Later three species were identified from the culture material. There may be a fourth species but at present this is uncertain. The Kenya Acetahularia found are all small species. When mature, the thalli are between 3.0-10.0 mm. tall and the disks are between i. 5-4.0 mm. in diameter. Following is a key to the Kenya species found. Page 48 KEY TO THE KENYAN SPECIES 1. Gametangial rays entirely free A. clavata. Gametangial rays not entirely free 2. 2. Gametangial rays in contact throughout; broadly rounded or emarginate apices A. moebii. Gametangial rays partly adhering, mammillate apices A. exigua. Acetabularia clavata Yamada, 1934, p. 57, figs. 24 and 25, (fig. i) Fig. I Acetabularia clavata Yamada. a. A mature reproductive thallus ( X 10). b. Part of a gametangial disk, as seen from above ( x 20). c. A plant with two gametangial disks ( X 10). Egerod: Univ. Calif. Publ. Bot., vol. 25, No. 5, p. 413. Thallus slightly calcified, 4.0-6.5 mm. high, stipe slightly rugose, bearing a single apical gametangial disk, i. 5-2.0 mm. in diameter, composed of 6-9 gametangial rays; the rays free and widely spaced, each ray clavate in shape with smooth rounded distal margin and containing numerous spherical cysts, 50-60^ in diameter; each coronal knob of corona superior bearing tv\ 0 or three sterile hairs ; corona inferior lacking. The Kenya plants are in general agreement with the description of A. clavata given by Egerod (1952) but are, however, larger in size. They grow up to 6.5 mm. in height. Egerod mentions that A. clavata frequently occurs intermingled with A. moebif to which it bears some resemblance and that although both species are of approximately the same size, they can be distinguished by the arrangement of the gametangial rays. Whilst it is true that the Kenya A. clavata also occurs intermingled with A. moebii, the two are not of the same size in culture. A. clavata is much smaller than A. moebii and it can be easily distinguished from it by the arrangement of the gametangial rays. The rays of A. moebii form a solid disk while those of A. clavata are free from each other (fig. ib). Another distinctive character of A. clavata observed in culture is its Page 49 dark green disk. Egerod also mentions that the rays of Hawaiian A.clavata are equal in width throughout their length, but this is not the case in the Kenya plants in which the rays are clavate in shape. Another observation made on the Kenya specimens is that there is a variation in the number (two or three) of hair scars on the coronal knobs of the same gametangial disk. Egerod has reported that a number of Hawaiian plants of A. clavata had two game- tangial disks at the apex. In the culture all the plants, except one, had a single apical disk. In the plant wiA two gametangial disks, the two disks were clearly alongside one another (fig. ic) and not in two series as illustrated by Egerod. Type locality'. Ryukyu Islands. Geographical distribution: PACIFIC OCEAN. Ryukyu Islands, Hawaiian Islands. INDIAN OCEAN. Kenya. Diani Beach, reef opposite Jadini. Acetabularia moebii Solms-Laubach, 1895, p. 30, pi. 4, fig. i. (fig. 2) A. minutissima Okamura, 1912, p. 184 pi. 100. A. wettsteinii Schussnig, 19306, p. 338 (cf. Feldmann, J. & G., 1947, p. 81, fig. i & 2.) Borgesen: '‘''Marine Algae from Mauritus”, 1950, p. 6, fig. i. Fig. 2. Acetabularia moebii Solms-Laubach. a. A mature reproductive thallus ( X 10). b. Part of a gametangial disk, as seen from above ( X 15). c. Coronal Imobs with thin walled hair scars ( x 200). Egerod: Univ. Calif. Publ. Bot., vol. 25, No. 5, p. 411. Thallus slightly calcified, 8-10 mm. high, stipe slightly rugose, bearing a single flat apical gametangial disk, 3-4 mm. in diameter, composed of 14-19 rays, the rays cemented together laterally through calcification, each ray cuneate with smooA broadly rounded or emarginate distal margin and containing numerous spherical cysts 79-158^ in diameter; each coronal knob of corona superior bearing 3-5 sterile hairs; corona inferior lacking. The Kenya plants examined agree well with the description given by Egerod (1952) and Borgesen (1951). However, the size of the Kenya plants examined exceeds those described from Hawaii and Mauritius. Papenfuss (1957) and Isaac & Chamberlain (1958) have also made similar observations for A. moebii found on the east coast of S. Africa and Inhaca Island and Peninsula respectively. The Kenya plants reach a height of up to 10 mm and the disk has a diameter of up to 4 mm. The gametangial rays are in close contact laterally through slight to moderate calcification (fig. 2b). No disks with free rays were observed as have been observed by Egerod for the Hawaiian plants and by Borgesen for the Mauritius plants. Borgesen (1951) quotes Solms-Laubach as stating that there are only five hair scars per coronal knob. In the Kenya plants 3-5 scars were observed. This is in agreement with Egerod and Borgesen who have observed similar numbers. In addition, it was observed that in the Kenya plants there is a variation in the number of hair scars on the coronal knobs of the same gametangial disk. This was also observed by Egerod for the Hawaiian plants. The hair scars of the Kenya plants (fig. 2c) are thin walled in agree- ment with Solms-Laubach and not thick walled as reported for the Mauritius plants. Type locality: Mauritius. Geographical distribution: ATLANTIC OCEAN. Mediterranean Sea. INDIAN OCEAN. Red Sea, Mauritius, east coast of Southern Africa, Kenya, PACIFIC OCEAN. China Sea, Japan, Hawaiian Islands, Southern Marshall Islands. Kenya: Diani Beach, reef opposite Jadini. Acetabularia exigua Solms-Laubach, 1895; p. 28, pi. 2. fig. i, 4. (fig. 3) E. Y. Dawson: Pacific Sci., vol. X, No. 1956, p. 42. Thallus slightly calcified, 3-6 mm. high, stipe slightly rugose, bearing a single apical disk, 1.5-3.0 mm. in diameter; disk flat or cup-shaped, composed of (6)-8-i2 gametangial rays, rays slightly cemented together laterally for part of the length; each ray ovoid with mammillate apex and containing numerous spherical cysts, 79-95 p in diameter; each knob of corona superior bearing two or three sterile hairs ; corona inferior lacking. This is a small plant like A. clavata but can be easily distinguished from the latter by its mamillate ray tips and also by its often cup-shaped disk (fig. 3a) in contrast to the flat disk of A. clavata. The rays in A. exigua are cemented together laterally for part of the length (fig. 36) while they are entirely free in A. clavata. In many plants of A. exigua it was observed that the apices had broken off (fig. 3c). There is a variation in the number of hair scars on the coronal knobs of the same gametangial disk. In the laboratory culture, a plant was observed which had two gametangial disks in series but the lower disk was incomplete and had the gametangial rays in twos as shown in fig. 3•• • , I '*■- . ' .rtyjrfi*. : ■> / - ; ' :>-»'! ' >: ■■• . . - . . . 1' .,•//> Vi" . ■’ * : » '•: rr.‘V -ii't ..■r. (*•■. ■ S^'0' ‘.n «• -f ''i i^ir/ ■■ *J<- , ' •>*' .'I > r> ,' -, '• r I n !■•' ^ jKlxwfl'i '■■■'} ' 1. • ' • . »t r .. > . 'V-. • ,v.. ,„• V /.'S' ..;.i T-V I-.V ■' • - ' ■■ '■'T'" r 'i' - "■ ■'.■■■'*.. ■•■ St ;'>-■' ' ■', '■ • y. ' w *••.? Printed by Kenya Litho Ltd., P.O. Box 40775, Cardiff Road, Nairobi. JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM .b7 Vol. 29 5th NOVEMBER 1971 No. 128 A NEW HYPEROLIUS (AMPHIBIA ANURA) FROM KENYA By A. Duff-MacKay The National Museum of Kenya and Arne Schi0tz Danmarks Akvarium, Char lot tenlimd, Denmark In the course of recent studies of the amphibia of East Africa, the first author has collected a form of Hyperoliits in south-eastern Kenya which appears to be undescribed. A brief description is given here as the species is to be treated further in two coming papers by Schi0tz. Hyperolius sheldricki sp. n. Holotype No. A/366/3 An adult male from north of Aruba Dam, Tsavo National Park (East) 3“ 18' S, 38° 54' E. April 1967. Paratypes 6 (Jo? A/366/2, 4-8 collected together with the holotype. 8 A/929/1-8 from 15 km west of Kakoneni 3" 14' S, 39 40' E. April 1969. The holotype and 6 paratypes are to be kept in The National Museum, Nairobi, 6 paratypes in the Zoological Museum, Copenhagen, and 2 paratypes in the British Museum (Nat. Hist.). Diagnosis: A small member of the Hyperolius viridiflavus superspecies, the female phase* with a characteristic, apparently constant, pattern in dark and light brown as in Fig. i ; in life the inner aspects of the thigh and tibia dark maroon. The upper eyelid is slightly pointed in profile. Description of the holotype: The snout is short; the gular sac and protective flap very large; webbing of the hind foot extensive; the upper eyelid bears a fairly conspicuous conical protuberance. The colour pattern is illustrated in Fig. i, and consists of dark brown stripes on a very light brown backgroimd; dark brown spots are scattered fairly evenly over the back and upper parts of the arms and legs, with a concentration on the thighs and also on the gular flap. The belly is white, and the underside of hands, feet, femur, and tibia a dark maroon in life — pink in spirit. Some dimensions are given in Table i. *In most members of the genus Hyperolius all juveniles and some adult males have a cryptically coloured “juvenile phase”, while all adult females and some of the males have what is often an entirely different “female phase”. The juvenile phase is often identical or similar in different species, but the appearance of the female phase is often diagnostic. Page 2 Description of paratypes: All the paratypes'are adult males. Those of series A/366 from near Aruba Dam are very uniform and similar to the holotype. Apart from A/929/1 which was kept in captivity for some time and is similar to the holotype, the series from west of Kakoneni differs somewhat as only three show, rather indistinctly, the diagnostic pattern; the remaining four having a more or less uniform, whitish, dorsum which is undoubtedly the juvenile phase. One of these, A/929/4, shows faint traces of lumbar spots, which seem to be the remains of an hourglass pattern. The combination of an hourglass pattern in the juvenile phase and a striped female phase would be tmique in the H. viridiflavus superspecies. Habitat: H. sheldricki appears to be confined to very temporary rain pools in a hot arid area with an erratic rainfall averaging 250 mm or less a year. This is in contrast to the neighbouring H. viridiflavus ferniquei (Mocquard) which seems to require breeding sites with permanent or nearly permanent water and a good reed cover. The first population (A/366) was found a few days after the first heavy rain following a long dry period. They were in a shallow pan in which stood coarse dry grass and which was of a more temporary nature than the majority of other waterholes in the vicinity. The pan was practically unoccupied by other amphibia characteristic of the area. The frogs were very numerous and were calling while clinging to long grass stems protruding from the water. The second population (A/929) was calling in sparse grass bordering a shallow roadside ditch immediately after the first heavy storm of the rainy season. The water in the ditch had completely disappeared a few hours after the rain ceased. A number of nearby waterholes were investigated and, although densely populated with other amphibia, were not occupied by H. sheldricki. Systematic remarks: By the shape of the head, the large gular pouch, and the voice, H. sheldricki is a typical m.ember of the H. viridiflavus superspecies. X Page 3 Is 16 h /8 A/929/r /2 13 Loc. near Aruba W. of Kakoneni Snout-vent Tibia 24.1 12,4 23,8 12.3 22.2 12,7 23.4 12.2 24.0 12.2 23.2 12.6 23.2 12. I 22.2 10.5 18.2 9.5 21.4 10.6 21.7 10.6 19.2 9.5 21.3 10.4 21.3 10.0 20.7 9-5 width of Foot prot. flap 6.9 10.7 7.9 10.9 7.2 10.7 6.5 10.6 6.8 10.7 6.9 10.9 6.2 10.7 6.1 9.4 4.6 8.6 6.7 8.4 6.0 9.7 6.0 8.9 6.5 8.6 Table i Measurements of H. sheldricki in millimetres It is regarded as having full specific ranl< as it does not show any great relationship in pattern to the adjacent forms of that group : H. v. ferniquei, H. v. glandicolor (Peters), and H. mariae Barbour and Loveridge. H. v. ferniquei has, furthermore, been collected at Kenani less than 80 km from the type locality. The nature of the known breeding localities would suggest a strong ecological separation. It is a pleasure for us to name this new species after David Sheldrick, warden of Tsavo National Park (East), in appreciation of his invaluable help and hospitahty over many years on occasions when we, more often one of us (A.D.-M.), have been at Tsavo. ACKNOWLEDGEMENTS We are grateful to the Director and Trustees of the Kenya National Parks for permis- sion to work in the Tsavo Park, to A. D. Forbes-Watson for his help in the field, and to the Museums Trustees of Kenya for permission to publish. REFERENCES SCHI0TZ, A. in press: The Superspecies Hyperolius viridiflavus. Vid. Medd. Dansk Naturh. Foren. 134. — in press: The Treefrogs of Eastern Africa. {Received 10th August, 1971). Published by The East Africa Natural History Society, Box 44486, Nairobi, Kenya and printed by Kenya Litho Ltd., Box 40775, Nairobi. JOURNAL OF THE EAST AFRICA NATURAL HISTORY SOCIETY AND NATIONAL MUSEUM VOL. 28 25th JANUARY 1972 No. 129 AN ORNITHOLOGICAL SURVEY OF THE KIDEPO NATIONAL PARK, NORTHERN UGANDA By C. C. H. Elliott Percy Fit z Patrick Institute of African Ornithology, University of Cape Town INTRODUCTION This paper summarises the results of the Oxford University expedition to the Kidepo Valley, in the long vacation of 1966 (July 20th-September loth). The expedition was undertaken by the author and R. L. Rolfe (and, although this paper is the former’s responsibility, the field work was a combined effort in which Rolfe did a lion’s share). The main aims of the expedition were to study the birds of this very unspoilt region and to satisfy the conditions laid down by Uganda National Parks, when permission was given to work in the Kidepo, by making a complete list of the species occurring in the park and preparing a small collection of sldns sufficient for the identification purposes of future researchers and interested tourists. In the event, the previous provisional list of 200 species was increased to almost 400 and a collection of 240 skins of about 170 species was made. Most of the skins are now available at the Kidepo H.Q., while about 30 of the exceptional ones have been presented to the British Museum. The main method used was to set strings of mist-nets in suitable trapping sites around each camping place. The surrounding country was then covered on foot by one man while the other operated the nets. Several species were never seen except when caught in the nets. Shooting was used to collect a few difficult species, such as the nightjars, for which nets were of no use. The vegetation of the area was already well known from botanical surveys carried out principally by Thomas (1943) and Wilson (1962), the latter being the more detailed for the Kidepo. This, combined with the two weeks of assistance to our expedition by J. M. Lock, a botanist working for NUTAE at the Queen Elizabeth Park, provided a reasonable knowledge of at least the broader aspects of the vegetation. Thus, we were able to correlate the distribution of birds with the vegetation, as described in the first part of the paper. General observations and the collection of skins, the more difficult to identify being taken back to the U.K. for further study, produced some interesting records which are discussed below in the two sections — Distribution and Races, and Breeding and Moult. A short survey was made on three species of Tockus (Hornbills) and results are given in the third section. The fourth (and last) includes incidental observations on the Park birds. I f Page 2 Map I __ KIDEPO PARK, UGANDA - — thatched line; ca.5000' contour ^ — solid black ; forest ^ — Scale i": 6 miles (i cm:3.8 km.) O MORONGOL Page 3 Throughout this paper, nomenclature follows C. M. N. White (1960-1965), while the English names are those of Praed & Grant. General Description of the Park The Kidepo Park occupies 500 sq. miles (1300 km^) in the north-east corner of Karamoja Province of Uganda (around 4°N), In the north, it borders the Sudan for 30 miles (50 km) and its eastern boundary is about 15 miles (25 km) from the Kenya border. It consists mainly of the basins of the rivers Narus and il^depo which flow north- west to meet in the Sudan. Both flow only during the rains, being for most of the year dry, except for pools trapped behind rocky catchments. The Kidepo valley (average altitude 3200 ft (975 m) is surrounded on all sides (see Map i), except for the north- north-west, by moimtain ranges (composed of rocks of the Basement complex), mostly about 5000 ft (1500 m) high, but rising to 9020 ft (2749 m) at Mt. Morongole and slightly lower at Mt. Lotuke, in the Sudan. The floor of the valley is almost flat, but is broken up by the numerous tributaries of the river and the occasional kopje (inselbergs) protruding a few hundred feet above the plains. The Narus valley is less open and flattens out only as it nears the Sudan border. The Narus river itself is about a quarter the width of the Kidepo at its widest, but has areas of flood-plain on either side, which have a thick growth of “elephant-grass” and acacia. The uplands west of the Narus rise to the 7400 ft (2250 m) mountain, Lonyili. The Kidepo National Park was created in 1962, when it had for some years been a game reserve, in which shooting was allowed but controlled. The whole area is just north of a belt cleared of bush in the anti-tsetse campaign, but tsetse is rife in the Park itself. As a National Park, the Kidepo is quite different from the other two parks in Uganda, having plains species, like cheetah and Bright’s Gazelle. The birdlife is also very varied, as indicated by the number of species, which is larger than that of Murchison Falls National Park and almost as large as that of the Queen Elizabeth National Park, both of which have large migrant populations attracted to their permanent water supplies. Climate A rain-gauge has been operating only during the past few months in the Kidepo and then only at the Park headquarters. The records from Moroto station suggest that the average annual rainfall for most of the Park would be 30-35 ins (760-890 mm), precipitat- ed in the months of May to October. An unbroken dry season of at least five months is usual and it may extend to eight months (e.g. 1965). The vegetation indicates that the driest area is in the extreme north-east corner of the Park, SST3 (see Map II), the rainfall being as low as 15-20 ins (380-510 mm) per annum (Wilson, Lock). This may be due to its being in the rainshadow of Mt. Lotuke. The two montane forest areas are the wettest in the park, with an estimated rainfall of at least 45 in (1150 mm) per annum. (Wilson). On Morongole, in particular, this rainfall is due in part to precipitation from dank mists which periodically sweep over its upper slopes. The maximum temperatures occur during the dry season, December and January. Avifauna In the following descriptions of the vegetation areas, only those species restricted to a given area are mentioned. Greater detail is given in the section on riparian “forest” and the distribution of all species is included in the appendix to this paper. VEGETATION From the forest of Lonyili and Morongole northwards, there is a decrease in rainfall and altitude accompanied by vegetation changes (Map II). The differences between the Page 4 Page 5 two sets of contours can mostly be related to the initial differences in the heights of the two mountains and also to the much gentler decrease in altitude occurring in the more hilly Lonyili half than in the quite dramatic plunge to the flat plain of the Morongole half. The difficulty about such a broad physiognomic description, correct as it may be about the important aspects of the vegetation, is to relate the bird distribution to it. Each segment is broken up by atypical vegetation — whether, for example, it be riparian forest of varying thickness, or swamps fed by a sulphurous spring on the edge of the driest area of the Park. It should be mentioned that this account is heavily based on Wilson (1962) for much of the vegetational description and his classification is used for vegetation divisions. Savannah I. Kananarok hotspring (A on map) SST3 (2): ST6 To the north of Camp A is found the Park’s driest part, classified as SST3 (2) (Succulent Shrub Thicket) characterised by the occurrence of numerous succulent and woody shrubs, small trees and many succulent herbs, found in dense clumps 20- 120 ft. (6-36 m.) in diameter, with intervals of bare ground. SST3 (2) was the only part of the Kidepo where is to be found the Lesser Kudu and where the duiker is at all common. Very little other game occurred in this habitat, except for that attracted by the hotspring. At the campsite, a three-acre (i . 2 ha) area of swamp, caused by a sulphurous hot- spring, lies about 10 miles (16 km) from the nearest permanent water, a recently-con- structed dam. To the south of Camp A, the vegetation becomes more open and is classified as ST6 — Tree and Shrub Steppe. It is characterised by an open-to-fairly-dense assemblage of small trees and shrubs 6-12 ft (2-4 m) high, with an aerial coverage of 40-70 per cent. The thickets are much smaller than those of SST3, shrubs are fairly common but the ground layer is poorly developed, with a preponderance of herbs over grasses. The area was rich in bird species, including Amadina fasciata and Passer iagoensis, which were not found in any other zone. II. Kidepo River (B on map) R4; SI/S7 This is the largest area of Borassus — dominated riparian woodland in the Park, about three miles (4.8 km) at its widest. It is classified as R4. The Borassus palms are often as high as 60 ft (18 m), with the result that a dense undergrowth of tall grasses, dominated by Phragmites communis Trin., can develop underneath. The shrubs Lawsonia inermis L. and Pluchea dioscoridis (L.) DC. are fairly common, right on the river banks and in the open areas of the community. Other than a herd of waterbuck and occasional visiting herds of elephants, no game was recorded within the Borassus woodland. Because these palms normally occur in Uganda in lines along river banks, it was suggested that a Borassus “forest” might have an unusual avifaima. This was not borne out by our observations, which showed that a few species were especially common in the area, but that none was restricted to it. South of the area immediately affected by the Kidepo was a region of fiat savannah shown as S1/S7, where the soil was heavy-clay type, rapidly softened by rain. The vegetation consisted of a mosaic of small trees 6 to 8 ft (i . 8-2 . 4 m) high mixed with taller trees 15-20 ft (4.5-6 m). Shrubs are only occasional, these being Cadaha farinosa Forsk and Harrisonia abyssinica Oliv. Scattered Balanites aegyptiaca (L.) Del. are often associated with this community and a considerable variety of herbs is distributed among well-defined tufts of the dominant grass. During our visit, the greatest concentration of game was found here. Lion, Bat-eared Page 6 Fox, and rhino were occasionally seen and plainsgame, such as zebra, hartebeeste, eland and Bright’s Gazelle, were quite common. III. Lorupei camp (C) and Apoka Hq. (D) S2 & S2/S12 The area S2 was the most thoroughly-explored in the Park, since it included the base to which we returned after each week’s sortie. Where S2 touched the Park’s eastern and western boundaries, especially in S2/S12, the ground became much more hilly, and the riparian thickets along the dried-up streams contributed more to the vegetation. On the lower parts, the vegetation consisted of tree savannah and savannah woodland, interspers- ed with much better-defined riparian forest. The savannah trees were mostly deciduous, averaging about 15-20 ft (4.5- 6 m) in height. The ground layer was dominated by perennial grasses such as Setaria incrassata (Hochst) Hack. Shrubs were rare. In the hilly areas, klipspringer. Rock Hyrax and, occasionally. Roan Antelope were seen. The savannah had a small population of hartebeeste, oribi and giraffe. The thickly-distributed stands of trees and the well-developed ground layer contained many birds, some occupying the canopy (which often included multi-species bird parties) and others, chiefly ploceids, areas of grass. Dendropicos obsoletus (Wagler) found here, was the only species not found elsewhere. IV. Narus River (Worked from D) R2 Unlike the previous vegetation type, the Narus River riparian woodland contained few species of birds, mostly ploceid seedeaters. The area was swampy in places and had two small man-made dams, which encouraged water birds. The woodlands contained trees up to 25 ft (7.5 m) tall, with a lush, perennial grass layer from 5 to 8 ft (1.5 to 2.5 m) in height and sometimes more, giving ground cover up to 90 per cent. The whole area was liable to flooding. In the dry season, this area is said to be full of game, but apart from occasional herds of buffalo and elephant passing through, the only large animals seen by us were two crocodiles in one of the dams. The only birds not found elsewhere were aquatic species. V. Kopem Kopje (E on map). S3 This region was, next to SST3, perhaps the driest in the Park. The vegetation was classified as S3 — a savannah woodland community of small trees, often regularly-spaced, with a dominantly perennial layer, mainly dominated by tufted grasses. Shrubs were rare, except where protection from &e existed such as in the crown of small rock outcrops. Only a few days were spent at Kopem, from which most sorties were directed at the foothills of Morongole in order to get some idea of the altitude zonation of birds on that mountain. Conclusion The two most impressive things about the Kidepo were the total number of savannah birds and the variety within each habitat. Various experts made “off-the-cuff” estimates of about 200 as the total number of species to be expected from the region, on the grounds of its semi-aridity and its geographical position. The final total reached nearly 330 (excluding forest, montane and palearctic birds) which, for so small an area, compares dramatically with the total of 418 for all West Africa and of 497 for the whole of the Sudan (Moreau, 1966). Of the 16 species unrecorded in the Sudan, 14 were savannah species and included such birds as Apalis karamojae. Vidua hypocherina, Turdoides jardinei, Mirafra africanoides and Pterocles gutturalis. By comparing the lists of typical species for each area, it is Page 7 apparent that, although certain species occur throughout the savannah, the basic popula- tion changes considerably over comparatively short distances. This can be related to similarly-rapid changes in vegetation, of coiurse. Forest and Upland This section describes the vegetation in the Park above 5000 ft (1525 m), a region containing the two areas of montane forest and the surrounding upland vegetation, which has a bird population much influenced and overlapping with that of the forest. It then shows how the forest species extend their range down the tongues of riparian gallery forest, almost to the floor of the Kidepo valley. VI. Lonyili Forest camp. (F on the map) M6 and HE5 In appearance, the relict forest of Lonyili, at 5000-6500 ft (1525-2000 m), was the only “real” forest in the whole Park. The community exists on deep soils and is perhaps best described by Macdonald & Cave’s term : bowl-forest. Although the rainfall is probably similar to that of Morongole, it is much more effective in producing lush vegetation, because water collects in the depression, or bowl, on the side of the mountain. The forest is only about three miles (5 km) long and about a mile (1.5 km) wide. The botany of the forest is not really known. Wilson did not investigate it, though he points out that, in composition, it was unlike any other in Karamoja and that its closest afiinity was probably with the flora of the Imatong and Dongotona mountains of the Sudan. The last important point is that the bulk of the forest was at 5500-6500 ft (1700- 2000 m) and therefore can be considered lowland forest — quite distinct from the montane forest on Morongole, which commenced above approximately 7000 ft (2100 m). The vegetation is classifled by Wilson as M6 — Tree Savannah; but, as stated, this refers mainly to the open areas and not to the forest patches. He gives three species of Acacia which reach a height of 60 ft (18 m) and are relict survivors of forest, so may occur within the forest patches. Lock, the expedition’s botanist, found the variety of trees in the forest of extreme interest and unlike the species in southern Uganda. He suggested that some of the larger trees may be a species of Ficus. They were over 100 ft (30 m) tall with broad buttressed trunks (often more than 20 ft (6 m) in diameter), some carrying fruit attractive to many species of birds. A ferocious nettle, probably Laportea alatipes Hook f. grew in profusion in deep shade associated with the following: Forest: Acacia abyssinica Hochst. Ficus spp. A. lahai Benth. Dracaena spp. Albizia gummifera (Gmel.) C.A. Sm. Impatiens sp. possibly a Coffea. Peperomia sp. Lock was unable to identify the tall 9-1 1 ft (2 . 7-3 . 3 m) plant of which the marsh in tlie centre of the forest was composed. The forest had a population of about 25 Uganda blue monkeys and, during the dry season, it probably holds a small number of elephant and buffalo, of which there were many signs in the forest, such as well-worn paths. The tree savannah consisted of an open meadow extending from 6500 ft (1980 m) to the summit at 7400 ft (2255 m), in which the trees gradually became smaller and were more dominated by Protea gaguedi Gmel. Areas of bare rock also became more frequent and the summit consisted of large boulders, with a steep cliff falling away into the Sudan, and the bowl-forest, beginning on one of the gentler slopes on the Uganda side. The lower vegetation merged into HE5 type, where httle time was spent on birds. The latter also applies to ST6. Page 8 Birds found only in M6 were : Accipiter tachiro; Cisticola brachyptera and C. aberrans. Those found only in the forest were: Columba delegorguei; Merops lafresnayi; Lybius leucocephalus; Lybius bidentatus; Phyllastrephus fischeri; Alethe poliocephala; Apalis cinerea; Camaroptera chloronota; Nectarinia verticalis; Anthreptes collaris and Ploceus ocularis. Bearing in mind the forest’s small size and low altitude, it is interesting to compare Lonyili with its nearest equivalent in the Sudan. This is the Lotti forest (about 50 miles (80 km) to the north, in the Imatong mountains), which is at a lower altitude, much larger and well-developed. The Lonyili forest would be expected to have a very impoverished version of the bird population of the Lotti forest and the brevity of the list shows this. Of the 15 species recorded in the Kidepo and unrecorded in the Sudan, only one, Camaroptera chloronota, is a forest bird. More than 65 species are listed for the southern Sudan (excluding the dry country species) many of which are forest species recorded in the southern ranges like Lotti. Notable absences from the Lonyili forest are six species of Warbler including two Apalis; seven species of thrush including three Cossypha; five Muscicapinae; seven Pycnonotidae; and all Trichastoma. VII. Morongole Mountain: Camps at 6500 ft (1980 m). M3, M5 and HE5 and 10,200 ft (2500 m). (G on the map). The vegetation of Morongole consists of: (a) “mist” forest and forest on steep slopes, equivalent to M3, both above 7000 ft (2135 m); (b) Highland tree savannah (i.e. scattered trees) M5 7000 ft (2135 m) to the summit, 9020 ft (2749 m) replaced below 7000 ft by savannah woodland, with HE5 (closer tree clumping). Wilson describes HE5 as a savannah woodland of deciduous and compound-leaved trees, with a well-developed perennial grass layer. The tree heights average 10-20 ft (3-6 m). Unlike the forest birds of M3, the savannah species had a wide altitudinal range and many species occurring in HE5, were also found at 9000 ft (2750 m). Found only in HE5 was the species Parisoma lugens. Although the general nature of the vegetation above HE5 is clear enough, with two forms of forest and an upland grassland, there seems to be a difference of opinion between Thomas (1943) Wilson, on the actual composition and even on such important matters as to which were the dominant trees. Thomas describes the meadow as “shrubby moorland”, passing into “grassy moor- land” (he does mention Erica spp. as occurring), while Wilson makes a point of saying that it is not moorland at all (no Erica), but should be described as “tree savannah” passing into “grass savannah”. Only four of the 79 species listed by Wilson are given amongst the 15 quoted by Thomas. Thomas also draws attention to the affinities of the vegetation with that of the Imatongs. Coleus grandicalyx E.A. Bruce, with violet-coloured flowers, from the high meadow and the grass, Setaria splendida Stapf, are found only on the two ranges. Birds found only in M5 (in which are included the summit cliffs) were: Gypaetus barbatus; Columba unicincta; Caprimulgus poliocephala and Psalidoprocne pristoptera. Thomas gives no details in his description of the forest vegetation, except to mention that Dombeya goetzenii K. Schum. is the dominant tree in the forest and that Cussonia specata Thunb. is common. Neither species is mentioned by Wilson, who classifies the area as “M3 — Dry Montane Forest.” He describes it as being “evergreen forest with an inter- mittent canopy, usually in large clumps, but occasionally contiguous with a marginal shrub layer. Depending on the density of the canopy, a herb/grass layer occiurs quite frequently on the forest floor”. Wilson goes on: “The forest margin shrub layer varies greatly according to the successional state of transition to forest or fire climax grassland. Under the forest canopy the shrubs are not common except for Senecio petitianus A. Rich, which is a common liane. Ferns, lichens, mosses are common, particularly on the branches of the largest forest trees and Bracken Pteridium aquilinum Kuhn, is an occasional constituent”. Page 9 Francolinus squamatus Columba arquatrix Chrysococcyx cupreus Tauraco leucolophus .. Turdoides rubiginosus Alcippe abyssinica Pycnonotus barbatus .. Muscicapa adusta Melaenornis chocolatina Turdus olivaceus Cossypha caffra ... Bradypterus cinnamomei Cisticola cantans Prinia subflava Chloropeta similis Hirxindo fuligula Psalidoprocne albiceps Parus albiventris Tchagra australis 1 Corvus rhipidurus Onycognathus morio .. Zosterops senegalensis Nectarinia tacazze N. venusta N. preussi Ploceus baglafecht Euplectes capensis Estrilda melanotis Serinus canicollis Quelea cardinalis Altitude above Level TABLE I ALTITUDINAL ZONATION OF BIRDS ON MORONGOLE MOUNTAIN Page 10 The M3 community develops best on the deep soils of plateau sites. On the steeper slopes, the composition changes considerably with Juniperus procera A. Rich., Olea hochstetterii Baker, and Teclea nobilis Del., becoming the dominant trees. A quite large population of the Karamoja race of mountain bushbuck living on the mountain, was found browsing in the meadows at dusk and dawn. A few klipspringer haunted the summit rocks. Leopard was the chief predator and one frequently roared from a rock a few hundred feet below our camp. Birds found only in the montane forest were: Accipiter melanoleucus; Columba arquatrix; Chrysococcyx cupreus; Musophaga rossae; Chloropeta similis; Phylloscopus umbrovirens and Cryptospiza salvadori. The division of bird populations between the meadow and forest was not great, because many of the small passerines like Phylloscopus umbrovirens fed in the forest margins, while others like the sunbirds, although feeding mostly in the grassland, also fed in the forest canopy and presumably nested amongst the trailing Usnea spp. Others, like the Red-wing Starlings, were usually visible doing noisy aerobatics above the grass- land, but descended to fruit trees in the forest to feed. The Morongole forest birds were remarkable for their extreme paucity. Whereas the Lonyili forest appeared noisy and full of barbets, turacos and blue monkeys, the Morongole forest was almost silent, except for the flapping of sunbird wings high in the canopy and the reedy singing of Chloropeta similis. In the highest forest, there were no turacos, except for the isolated record of Musophaga rossae. Perhaps the most conspicuous absentees from both forest areas were the Casqued Hornbills {Bycanistes spp.). The difference in numbers of forest species (36 spp. for Lonyili, and 25 for Morongole) is probably due to the different type of forest, attributable (at least partly) to altitude. Macdonald & Cave point out that “bowl-forest” is the nearest approach to true rain forest, hke those of the Congo, at this latitude; cloud-forest is much poorer and subject to cold dank mists, which are not generally conducive to the production of food-plants. It seems that the change from one type of forest to the other occurs at 7000 ft (2135 m) altitude. If we had had time to investigate the avifauna of some of the broad-gully gallery forest on Morongole at lower altitudes, we would probably have found a species composi- tion much more hke that of Lonyih. A similar paucity of birds was found by Tennent (1964) in the Kitui District mountains of Kenya which he explained not only by the poorness of a forest similar to Morongole, but also thought the effects of isolation important. Isolation cannot be said to have played a part in reducing the number of species on Morongole. As indicated above, the forests of the southern Sudan (particularly the Imatongs) have a much richer forest avifaima although they are further from the presumed centre of dispersal of the forest species in the Kenya highlands. Another point, which emerged from our observations, was the clear altitudinal zonation of certain birds on Morongole and Table I shows the altitude range of these 30 species. Lastly, we come to the extensions of the forest species down to lower levels where these montane forest species (e.g., Tauraco leucolophus) met species typical of riparian forests (e.g., Laniarius erythrogaster) and those that feed in the surrounding savannah, seeking refuge back in the forest (e.g., Turdoides jardinei). I have tried to consider the riparian birds of four areas in the Park, i.e. the substantial gallery forest of the Lorupei (the branch of the Narus from Lonyili); the upper Kidepo; the narrower thicket/forest of the Kananarok rivers; and the minor branches of the Narus, the Kakel and Losigiria. The avifauna shows a gradation in the proportion of forest species as the distance from the forest increases. The vegetation in these areas varies in size, is similar in appearance, but is of extremely varied constitution. Lock (pers. comm.) describes the Narus branches as being nearest Wilson’s T2 and T3. But in places, particularly Lorupei, the vegetation is more developed than these (T2/T3) would indicate, with many large trees draped with lianas, including species of Ficus which attract the forest barbets and turacos. Page II In the following lists of birds, those also recorded in the two montane forests are marked F. Only lo species were recorded in all or three out of four of the areas. Next, a list is given of the species occtirring in two of the four areas, most of which are typical of riparian vegetation. Last is given a list of the species, individual to the riparian vegetation of each area, which begins to suggest that the bird population becomes impoverished the fmther it recedes from a forest source. Lorupei is the nearest (to Lonyili), Kidepo next (to Morongole), while Kananarok (to Lotuke) and Narus (to Morongole) are about the same distance from forest. The Kananarok species are, however, all thicket species. Birds common to three or four areas : Turtur tympanistria F Terpsiphone viridis Tauraco leucolophus F Cichladusa guttata Tockus erythrorhynchus Cossypha heuglini F Indicator indicator F Camaroptera brachyura Campethera nubica Tchagra senegala Glaucidium perlatum Laniarius barbatus Oriolus monacha Turdoides rubiginosus Pycnonotus barbatus Birds common to two areas : Streptopelia semitorquata F Colius striatus Chrysococcyx caprius Lybius lacrymosus Psittacula krameri Thripias namaquus Poicephalus meyeri Batis molitor Halcyon senegalensis Sylvietta whytii Corythaixoides leucogaster Prionops plumata Otus scops Nectarinia mariquensis Birds individual to each area: Lorupei: Kananarok: Francolinus squamatus F Numida meleagris Turtur abyssinicus Tockus deckeni T. afer Bradornis pallidus Centropus superciliosus Phoeniculus minor Crinifer zonurus Pogoniulus minor Merops lafresnayii F Laniarius funebris Indicator variegatus Nectarinia senegalensis Turdoides jardinei Estrilda erythronotos Platysteira cyanea F Petronia xanthocollis Turdus pelios Pytelia melba Hypargos nitidulus F Ploceus luteolus Coracina pectoralis Laniarius ferrugineus F Narus: Malaconotus sul £ ureopectus Pytelia afra Zosterops senegalensis F Cuculus clamosus Z. virens Kidepo: Treron waalia F Dicrurus adsimilis Halcyon chelicuti Emberiza forbesi (only high alt.) Phoeniculus purpureus Serinus dorsostriatus Bubo lacteus Lybius rolled Apalis pulchella Eremomela icteropygialis Page 12 Compare the avifauna of the four riparian types by Jaifard’s Coefficient of Community (C.C.) or Simpson’s Coefficient (S.C.)j Hagmeier & Stults (1964). If two populations of equal size have a C.C. of 68 . 5, 75 per cent of the species is common to both; or if when the S.C. value is 75, more than 75 per cent of the smaller of the two is found in the larger, then in either case the populations are considered faunistically identical. Using these methods, the only two populations shown to be faunistically identical are those of the Lorupei and Narus. Since one is a branch of the other, this would be expected. Below, (Table II), are the other S.C. values. They reinforce the point that the avifauna of the Kidepo does change substantially over a very small area (Kidepo 48 Kananarok, a distance of only about 10 miles). ^ Table II Lorupei Lorupei Narus 91 Kidepo 58 Kananarok 55 Narus 91 67 67 Kidepo 58 67 — 48 Kananarok 55 67 48 BREEDmG AND MOULT. The semi-arid climate of the Kidepo means that the whole Park, except for the small areas of the two mountain ranges above about 6500 ft (1980 m) is subject to great but irregular seasonal variation. In such circumstances, the ultimate factor controlling breeding seasons is the occurrence of tlie rains, with the insectivores mostly breeding at the start and the seed-eaters towards the end. Each thereby feeds its young dining a food abun- dance. The breeding season of a species can be determined from (i) observation of nests, eggs or recently-fledged young; (ii) the condition of the gonads; (iii) extrapolation from moult, which in most species at this latitude follows breeding. The assumption is made that food is short during about eight months of most years and that this is not countered by the migration of relatively few birds. The deduction is less valid for species that live in deep forest or high mountains, where seasonal variation is less marked and the rigours of moult can be spread over a greater length of time. However, only three species on which we have data fall into this category, namely Cisticola cantans, Phylloscopus umbrovirens and Cossypha caffra, for which we did not obtain support from either of the other lines of evidence (i.e., observation or gonads). But in no case is the breeding of any species deduced from moult, contrary to what might have been expected from previous knowledge. The 40 species breeding during the seven weeks of our visit are listed below, with the abbreviations: 0= Observation; G= Gonads; M= Moult. Aegypius tracheliotus 0. Prinia subflava M. Coturnix delegorguei O.G.M. Cisticola chiniana 0. Streptopelia capicola G. C. cantans M. Streptopelia senegalensis O.G.M. Sphenoeacus mentalis G. Streptopelia decipiens 0. Prionops plumata 0. Ceyx picta M. Eurocephalus anguitimens M. Lybius leucomelas M. Lanius excubitorius M. Indicator indicator O.M. Anthreptes collaris G. Indicator variegatus M. Nectarinia venusta G. Campethera nubica M. Bubalornis albirostris O.G. Turdoides jardinei G. Ploceus cucullatus O.G.M. Turdoides rubiginosus M. P. jacksoni G. Alcippe abyssinica O.M. P. baglafecht emini O.G.M. Page 13 Platysteira cyanea G. P. luteolus M. Batis molitor M. Euplectes gierowii G. Terpsiphone viridis M. E. hordeaceus G. Cossypha cajfra M. E. albonotatus G. Sylvietta whytii O.G. Quelea quelea G. Camaroptera brachyura O.G. Q. cardinalis 0. Phylloscopus umbrovirens M. Petronia xanthocollis M. By far the most conspicuous and numerous of the 40 breeding species were the seedeaters, the large areas of long grass in the Narus valley being ahve with breeding Quelea, Euplectes and some other ploceids. The area around Kananarok hotspring was also full of nests of Coturnix delegorguei and Streptopelia senegalensis in various stages of development. On the one hand, only five of the 15 seed-eating species were found in moult, indicating that most were still breeding during August and September, whilst of the 22 insectivorous/ carnivorous species, 17 were either moulting or (those marked O) recently-fledged chicks or juveniles were seen, suggesting breeding in June or July. Thus, the evidence supports the presumption that the insectivorous species breed before the seedeaters. DISTRIBUTION AND RACES The Kidepo is interesting geographically, because it is only about 50 miles (80 km) from the Imatong and Dongotona and about 10 miles (16 km) from the Didinga ranges in the southern Sudan. These three massifs, with Morongole, are only 200 miles (320 km) from the Ethiopian highlands, with the Boma hills as an intermediate stepping stone. Nevertheless, although some of the races of our species were the Ethiopian ones, the montane birds, at least, are very much those of the Kenya highlands (Moreau, 1966), the rich avifaima of the Imatong-Lotti forest being a particularly good example. The proximity of the Kidepo to the Sudan and Ethiopia, however, combined with the past distribution of ornithologists (especially the fact that the nearest thorough collecting expedition south of the Kidepo has been on Mt. Elgon), produced some interesting records. These are given below under the categories of species and races new to Uganda, undescribed races and extensions of known range. Where races are mentioned, these are based on skins. 1. Species new to Uganda Merops orientalis. Little Green Bee-eater. Most probably, it is of the race cleopatra as this has been identified as far south as Torit in the Sudan. The Kidepo is probably the southernmost limit of the species. Typical habitat: SI/S7. Cercotrichas leucophrys leucoptera. White-winged Scrub Robin. The record represents the western limit of the species in East Africa. The typical habitat: SST3 and S3. 2. Races new to Uganda Caprimulgus poliocephalus poliocephalus, Abyssinian Nightjar. Extension from western Uganda (Jackson), or 50 miles (80 km) south from Imatongs (White). Dendropicos fuscescens hemprichii. Cardinal Woodpecker. An extension of range from the east bank of Lake Rudolf (150 miles (240 km) west) between areas of D.f. lepidus (30 miles (48 km), north) in the Sudan and 100 miles (160 km) south in Moroto. Habitat: 6500 ft (1980 m) HE5. Cisticola aberrans petrophila, Rock-loving Cisticola. A small extension, 40 miles (64 km) south from the Dongotona Mts. brings this into Uganda. Habitat: 6500 ft. (1980 m). M6. Cisticola chiniana bodessa, Rattling Cisticola. Cave & MacDonald, and White, record Page 14 C. c. simplex for the southern Sudan and Karamoja respectively. C.c. bodessa is the Ethiopian race, previously with its southern limit on the Boma hills. Habitat: R4. Tchagra senegala habessinica. Black-headed Bush Shrike. An extension, 50 miles (80 km) south across the border from the Imatongs. Widespread below 5500 ft (1675 m). Corvus rhipidurus, Fantail Raven. This is mentioned neither by Praed & Grant, nor by White for Uganda, despite having been collected near Morongole by Stoneham (1926). Kopje above 3000 ft (900 m). Estrilda melanotis quartinia, Yellow-belhed Waxbill. An extension across the border from the Imatongs. 8300 ft (2525 m). Meadow M5. Estrilda erythronotos charmosygna. Black-cheeked Waxbill. An extension of the Ethiopian race 250 miles (400 km) west. E.e. delamerei is found only in southern Uganda. 3. Extensions of Range within Uganda These include interesting records of species which previously have been found only in the large areas of forest on the Congo border. Larus fuscus fuscus, Lesser Black-backed Gull. Praed & Grant say that it visits larger inland lakes, but is rarely seen far from the coast. Lake Rudolf is the nearest large Me 190 miles (300 km) east and the nearest coast is 700 miles (1125 m) S.E. The condition of the bird was poor enough to elicit attack from a Nubian Vulture, Aegypius tracheliotus, and a Bateleur, Terathopius ecaudatus, before being rescued by us (to die two days later). Musophaga rossae, Ross’s Turaco. The consensus of opinion (Jackson, White, Cave) is that this Turaco occurs all over Uganda and not only south of the Lango swamp, 150 miles (240 km) south-east as indicated by Praed & Grant. Eurystomus glaucurus suahelicus. Broad-billed Roller. The expected race for the Kidepo would be E.g. aethiopicus. Our specimen indicates an extension 250 miles (400 km) north from Elgon. Camaroptera chloronota, Olive-green Camaroptera. Forest species ; well outside its normal range of south-western Uganda forests. Apalis karamojae, Karamoja Apahs. This record extends the peculiar distribution (the isolated areas of northern Uganda (near Moroto) and Nzega in Tanzania) too miles (160 km) northwards from the Mt. Moroto region. Little is knov/n of its habits. Our single record was of a party of five birds, feeding low in dwarf Acacias, mostly A. drepanolobium and small Combretum stands. Hypargos nitidulus schlegeli. Green-backed Twinspot. Previously recorded only from the forests and adjacent areas to the Congo forests. Macdonald & Cave found H.n. chubbi in the nearby Dongotonas. Typical habitat: gallery forest within Si 6, where very difficult to see; only recorded after capture in mist-nets. 4. Possible New Races Three species, collected above 8000 ft (2440 m) on Morongole, are represented there by birds that appeared to differ subspecifically from other populations and might be worthy of formal description. Bradypterus cinnamomeus. Cinnamon Bracken Warbler. Four collected. Closest to B.c. cavei from the Imatongs, but differ in being whiter on the chest and belly and dingey- brown on the upper surfaces instead of a rich chestnut. They are also, on average, a little smaller than the Imatong race. Phylloscopus umbrovirens. Brown Woodland Warbler. Three collected. The Morongole birds appear to be intermediate between mackenziana and omoensis in that it is mainly whitish on the underside, like the former, and yet more greenish-brown above, like the latter. Turdus abyssinicus, Olive Thrush. Two collected. Nearest to T.a. abyssinicus, particularly those from the Imatongs and those from Elgon, previously known as elgonensis. It is. Page 15 however, much greyer on the chest and paler tawny on the flanks than any of the typical abyssinicus birds in the British Museum, including specimens from Elgon. FEEDING HABITS OF HORNBILLS Of the four species of the genus Tockus which inhabit the park, only three, T. nasutus, Grey Hornbill; T. erythrorhynchus, Red-billed Hornbill; and T. deckeni, Jackson’s Hornbill were more or less common. All are about the same size and shape (beak differences are not initially striking) and all have been recorded in the literature as occupying the same sort of habitat and as feeding on the same kinds of food, with no hint of ecological separation. Our observations were directed to seeing if there was any habitat or food preference. The records necessarily apply only to the seven weeks of the expedition; but, since very little has yet been published on the subject, it seems worth while to give tentative conclu- sions based on some 300 records. (a) T. erythrorhynchus (191 records) had a strong association with substantial riparian vegetation and was only occasionally found in dryer, open-plain areas of S3 and SST3. 55 per cent of the records were in area R4 (Borassus woodland). (b) T. nasutus (52 records) was the widest-ranging of the three species, with 60 per cent of the records almost equally divided between areas A1/S7, S3 and S2. The birds were not permanently-based in any of these areas except S3, but tended to occupy one of them for about 10 days before vacating it for one of the other two. S3 savannah was the only area which exclusively contained nasutus. (c) T. deckeni (28 records) was the least common of the three and was found typically in dense riparian thickets of smaller rivers and gullies. One pair always to be found in a single place, at the edge of R4. Food: Nine erythrorhynchus, three deckeni and one nasutus stomachs analysed, showed, not surprisingly, that erythrorhynchus eat the greatest variety of food classified as follows : T. erythrorhynchus: Yellow seeds, 1-3 mm. greatest diameter; Black seeds, 2-5 mm. g.d. ; Orthoptera, grasshoppers ; Isoptera, termites ; Formicoidae, black ants ; Lepidoptera, green/yellow caterpillars; Quelea cardinalis eggs; Commelina fruit (creeping monocot. herb); and Capparidaceae fruit (Shrub). Common to all the stomachs were the yellow seeds ; all but two stomachs contained black ants, and half contained termites. Opportunity had obviously been taken by the bird whose stomach contained 25 caterpillars. Egg-eating must be common; apart from the above record of Q. cardinalis eggs being eaten, T. nasutus was seen raiding a Bubalornis albirostrisIPlocepasser mahali colony. Nesting ploceids were very hostile to both T. erythrorhynchus and T. nasutus. T. deckeni: In general appearance, the stomach contents were very similar to those of erythrorhynchus, except for the absence of several items of animal food. One stomach contained only about too small snails: Yellow seeds 1-7 mm. g.d.; Black ants; Coleoptera. Beetle elytra; Snails, 2-4 mm. g.d.; Small green fruit. T. nasutus: The one stomach examined contained a base of yellow seeds, 1-3 mm., g.d., together with a higher proportion of pieces of stout beetle cuticle than was found in deckeni. From the foregoing, it may be concluded that there is no sharp difference between the species in the habitats at the time of year encountered, although there was some preference by erythrorhynchus for R4, deckeni for ST6/S16 and nasutus for S3. But on at least four occasions, mixed flocks of nasutus and erythrorhynchus and, on one occasion, all three, were observed together. It is possible that the time of observation in the Kidepo was one of superabundant food being just after the end of the rains when there should be plenty, at least, of animal food. That each species showed some preference for a particular habitat may indicate the Page 16 habitat in which the species more or less confines itself at other times of the year when food shortage occurs (Cain, pers. conun.). It is nevertheless very probable that the coexistence of the three species is accounted for by food differences. This, though unproven by available data, is suggested by differences in physical structure, particularly that of bill shape. While the bill of erythrorhynchus is slender and long, that of nasutus is about the same length, with a slightly greater depth, but has tooth-like ridges on the outer edges ; that of T. deckeni is shorter and much deeper than either of the other two. The nasutus ridges are probably adapted for holding and crushing and may possibly be specialised for heavily-armoured insects and hard-shelled fruit. The food of erythrorhynchus was shown to be very varied. Its long sharp bill suggests a more insecti- vorous or carnivorous diet than that of deckeni which, with its blunter shorter bill is possibly chiefly frugivorous. Since the above was completed, A. Kemp (pers. comm.), working nearly 2000 miles (3200 km.) south in the Kruger National Park in South Africa, has informed me of his studies on the three hornbills T. erythrorhynchus, T. fiavirostris and T. nasutus. His results so far make an interesting comparison with mine and he has kindly allowed me to quote them. On habitat, his results are almost diametrically opposed to mine, as he finds that the typical habitat of nasutus is mainly riparian and secondarily tail-tree Acacia nigrescens woodland and that of erythrorhynchus is open — usually overgrazed — ^grassland. He correlates this with local behaviour, nasutus feeding in the trees, erythrorhynchus mostly on the ground, digging the earth and sifting piles of dung for insects. He does, however, suggest that the foraging behaviour becomes generalised in the wet season and that erythrorhynchus in particular becomes a more active forager, often pursuing free-moving insects such as Orthoptera. This discrepancy between our ideas on habitat, at least for erythrorhynchus, may perhaps be due to the fact that observations in the Kidepo were made only in the wet season. It does seem odd, however, that on no occasion in seven weeks in the Kidepo did I see nasutus feeding in riparian vegetation, although erythrorhynchus was common there. Kemp mentions no change in the behaviour of nasutus in the wet season. In the diet, we agree. He finds that nasutus is the main fruit-eater, especially on species with tough skins (Diospyros and Pseudocadia), and its main animal items are large hard Coleoptera and tree frogs. T. erythrorhynchus is confirmed as mostly carnivorous, 87 per cent of its food items being animal, but he includes a particular liking for Coprina beetles foimd in dung. Egg-eating has not been recorded, but opportimism occurs in all three species. For example, nasutus was recorded taking mice during a recent plague. In both diet and habitat, Kemp found fiavirostris less specialised than the other two species, as I found with T. deckeni in the Kidepo. MISCELLANEOUS (a) On one occasion, the following mixed flying-ant swarm, Formicoidea : Halcyon chelicuti Merops bulocki *Campethera nubica Apus coffer Apus aequatorialis Cypsiurus parvus *Terpsiphone viridis Hirundo aethiopica OBSERVATIONS concourse of birds was observed feeding on a Hirundo senegalensis Psalidoprocne albiceps Coracina pectoralis *Dicrurus adsimilis Corvinella corvina *Oriolus monacha Cinnyricinclus leucogaster Euplectes hordeaceus Also present, but not actually observed feeding: Quelea cardinalis; Streptopelia capicola Page 17 Of the 1 8 species recorded, only four (marked with *) are at all typical of mixed bird flocks of the area. These could be called “nucleus” species, while the others remain “circumference” species (Winterbottom, 1949), but it seems much more likely that they were just a chance collection of birds taWng advantage of an obvious food supply. (b) On another occasion, a Black Kite, Milvus migrans, was seen to fly into a tree where it was attacked with great commotion by three shrikes, Eurocephalus anguitimens. One clung to the kite’s mantle with both claws and bill as the kite flew away 200 yds. (180 m.) before it released its grip. I imagine that this ferocious attack by the much smaller bird was more territorial than anti-predator, since there were no shrikes nesting in the area. (c) On several occasions. Broad-billed Rollers, Eurystomus glaucurus, were seen hawking insects until nearly dark, in open woodland, and then collecting in a roost of 20 or 30 birds in one large tree. Roosting behaviour does not seem to have been recorded before. ACKNOWLEDGEMENTS The expedition was made possible by many people and my gratitude to them has been duly recorded in the Oxford Exploration Club Bulletin. In the preparation of this paper, my thanks are owed to the late R. E. Moreau; Professor A. J. Cain; Richard Rolfe; Airs. M. K. Rowan; Professor J. M. Winterbottom; and A. D. Forbes-Watson, for helpful criticism of the manuscript. I am also grateful to my father. Sir Hugh Elliott, for vetting the manuscript and for help in the subspecific identification of our skins which were lodged at the Natural History Museum. The paper’s form was initially based on that by Elliott & Fuggles- Couchman (1948). SUMMARY The results of an ornithological survey of the Kidepo National Park, northern Uganda, carried out by an Oxford expedition in the long vacation of 1966, are recorded. The bird- plant communities of five principal t5Tjes of lowland savannah, of two areas of upland forest and savannah, and of the riverine areas within the Park are described, details being given in the Appen^x at the end. The 40 species breeding and the details of the form of the information are given. The more interesting records of species new to Uganda, imdescribed races, races new to Uganda and extension of range within Uganda, are described. The problems of speciation of three closely similar species of the genus Tockus (Hornbills) are discussed. It is thought that the differences in the physical structure of the beaks of the birds allows different food and feeding habits which, in turn, allows considerable overlap and coexistence within the same general habitat. Incidental observa- tions made by the expedition are recorded at the end. REFERENCES Cain, A. J. (in prep, and pers. comm.). Cave, F. O. & Macdonald, J. D. (1955), Birds of the Sudan. Oliver and Boyd, Edinburgh & London. Elliott, C. C. H. & Rolfe, R. L. (1967), Oxford University Expedition to Uganda, 1966. Bull. Oxf. Univ. Explor. Club, 15; 14-16: 43-55. Elliott, H. F. I. & Fuggles-Couchman, N. R. (1948). An Ecological Survey of the birds of the Crater Highlands and Rift Lakes, Northern Tanganyika Territory. Ibis, 90: 394-425. Hagmeier, E. M. & Stults, C. D. (1964). A Numerical Analysis of the Distributional Patterns of North American Mammals. Syst. Zool. 13; 125-155. Iackson, F. j. (1938). The Birds of Kenya Colony and the Uganda Protectorate. 3 Vols. Gurney & Jackson. London. Kemp, A. (in prep, and pers. comm.). Mackworth-Praed, C. W. & Grant, C. H. B. (1952-60). African Handbook of Birds. 2nd. cd. Vol I, 1957, Vol. 2, i960. Longmans Green. London. Page 18 Moreau, R. E. (1966). The Bird Faunas of Africa and its Islands. Academic Press, London & New York. Stoneham, H. F. (1926). Field Notes on a Collection of Birds from Uganda. Ibis, 2; 26-91. Tennent, J. R. M. (1964). The birds of Endau Mountain in the Kitui District of Kenya. Ibis, 106: 1-6. Thomas, A. S. (1943). The vegetation of the Karamoja District. Uganda. J. Ecol. 31: 149-177. White, C. M. N. (i960). A Check-list of the Ethiopian Muscicapidae (Sylviinae). Occ. Pap. Nat. Mus. S. Rhod. 24B: 399-430. White, C. M. N. (1961). A Revised Check-list of African broadbills, pittas, larks, wagtails and pipits. Govt. Printer, Lusaka. White, C. M. N. (1962a). A Check-list of the Ethiopian Muscicapidae (Sylviinae). Occ. Pap. Nat. Mus. S. Rhod. 26B: 653-738. White, C. M. N. (1962b). A Revised Check-list of the African shrikes, orioles, drongos, starlings, crows, waxwings, cuckoo-shrikes, bulbuls, accentors, thrushes and babblers. Govt. Printer, Lusaka. White, C. M. N. (1963). A Revised Check-list of African flycatchers, tits, treecreepers, sunbirds, white-eyes, honey-eaters, buntings, finches, weavers and waxbills. Govt. Printer, Lusaka. White, C. M. N. (1965). A Revised Check-list of African nonpasserine birds. Govt. Printer, Lusaka. Wilson, J. G. (1962). The vegetation of Karamoja District, Northern Province of Uganda. Memoirs res. div., Dept, of Agric. Uganda Protectorate. Series 2, No. 5. Winterbottom, j. M. (1949). Mixed bird parties in the tropics, with special reference to Northern Rhodesia. Auk, 66: 258-264. In the above descriptions of the vegetational areas of the Kidepo, only the bird species restricted to each area have been mentioned. Belotv is a summary of all records made from 20.7.66 to 10.9.66. This serves as a check-list of the Kidepo National Park but it should be remembered that only a small part of the annual cycle was covered and observations at other times of the year may give wider distributions for some birds. Some species were recorded in the very incomplete list of birds existing before our arrival, and were not seen by us. These have almost all been included and where the distribution was recorded, it has been marked in, where not, it has been left blank. The zone marked as SW refers to the hotspring swamp at Kananarok, which formed an oasis in the middle of the otherwise uniform regions SST3 and ST6. Key to vegetation zones : Savannah: SST3 = dry shrub/thicket Upland: HE5 = savannah woodland {Received 18th January, 1971). APPENDIX I Birds recorded in the Kidepo National Park, Uganda with an indication of their habitat preference SW = swamp ST6 = dry shrub/thicket S1/S7 = open plain M6 = tree savannah including Lonyili forest S2/S12 = long grass/woodland S3 = dry woodland M5 = highland tree savannah M3 = dry montane forest Riparian : R4 = Kidepo R. area R2 = Narus R. area T2/T3 = Upper Narus branches Page 19 SAVANNAH Stmthio camelus Linnaeus Ostrich .... Phalacrocorax africanus (Gmelin) Long-tailed Cormorant Anhinga rufa (Daudin) Darter Pelecanus onocrotalus Liimaeus White Pelican P. rufescens Gmelin Pink-backed Pelican Ardea cinerea Liimaeus Grey Heron A. melanocephala Vigors & Childr( Black-headed Heron Egretta alba (Linnaeus) Great White Egret E. intermedia (Wagler) Yellow-billed Egret E. garzetta (Linnaeus) Little Egret Ardeola ibis (Linnaeus) Buff-backed Heron A. ralloides (Scopoli) Squacco Heron Butorides striatus (Linnaeus) Green-backed Heron Scopus umbretta Gmelin Hamerkop . Ciconia ciconia (Linnaeus) White Stork C. abdimii Lichtenstein Abdim’s Stork C. episcopus (Boddaert) Bishop Stork Anastomus lamelligerus Temmincl^ Openbill Leptoptilos crumenifems (Lesson) Marabou Ibis ibis (Linnaeus) Wood Ibis . Threskiornis aethiopica (Latham) Sacred Ibis . Sarkidiornis melanotos (Pennant) Knob-billed Goose Dendrocygna viduata (Linnaeus) White-faced Tree-Duck D. bicolor (Vieillot) Fulvous Tree-Duck Sagittarius serpentarius (Miller) Secretary Bird Gyps ruppellii (Brehm) Riippell’s Griffon G. bengalensis (Gmelin) White-backed Vulture . Trigonoceps occipitalis (Burchell) White-headed Vulture . Aegypius tracheliotus (Forster) Lappet-faced Vulture Neophron monachus (Temminck) Hooded Vulture . j i Page 20 SPECIES SAVANNAH RIPARIAN UPLAND R CO Co CO Co Co > R s 1 s N. percnopterus (Linnaeus) Egyptian Vulture .... Falco peregrinus Tunstall X X Peregrine Falcon .... F. biarmicus Temminck X Lanner ..... X F. chicquera Daudin Red-necked Falcon F. ardosiaceus Bonnaterre & Vieillot X X X X Grey Kestrel .... F. cuvieri Smith X African Hobby .... F. tinnunculus Linnaeus X X Kestrel F. alopex Heuglin Fox Kestrel .... Polihierax semitorquatus Smith X X X X X Pygmy Falcon .... Milvus migrans (Boddaert) X Black Kite . . . Elanus caeruleus (Desfontaines) X X X X X Black-shouldered Kite . X X X X X Aquila verreauxii Lesson Verreauxs’ Eagle .... A. rapax Temminck X X Tawny Eagle .... A. wahlbergi Sundevall X X X X X Wahlberg’s Eagle .... X X X Hieraaetus spilogaster (Bonaparte) African Hawk-Eagle Polemaetus bellicosus (Daudin) X X X Martial Eagle Lophaetus occipitalis (Daudin) X X Long-crested Hawk-Eagle Kaupifalco monogrammicus Temminck X X X X X Lizzard Buzzard . Circaetus cinereus Vieillot X X Brown Harrier-Eagle C. pectoralis Smith Black-chested Harrier-Eagle . Butastur rufipennis (Sundevall) X X X X Grasshopper Buzzard Temthopius ecaudatus (Daudin) X X X Bateleur . . . Haliaeetus vocifer (Daudin) X X X X X X X Fish Eagle . Gypohierax angolensis (Gmelin) Palm-nut Vulture Gypaetus barbatus Linnaeus X Lammergeyer .... Butco oreophilus Hartert & Neumann X X Mountain Buzzard B. rufofuscus (Forster) X X X X Jackal Buzzard .... Buteo buteo (Linnaeus) Steppe Buzzard .... Accipiter tachiro (Daudin) X X X X 1 African Goshawk .... A. melanoleucus Smith X X „ 1 Great Sparrowhawk X 1 Page 21 SPECIES A. badius (Gmelin) Shikra .... A. nisus Linnaeus Sparrowhawk A. minullus (Daudin) Little Sparrowhawk Melierax gabar (Daudin) Gabar Goshawk M. poliopterus Cabanis j Pale Chanting Goshawk M. metabates Heuglin Dark Chanting Goshawk Circus macrourus (Gmelin) i Pallid Harrier ‘ C. aeruginosus (Linnaeus) Marsh Harrier Polyboroides typus (Scopoli) Harrier-Hawk Francolinus squamatus Cassin Scaly Francolin j F. cJappertoni Children Clapperton’s Francolin . F. sephaena (Smith) Crested Francolin : F. leucoscepus Gray Yellow-necked Spurfowl F. icterorhynchus Heuglin Heuglin’s Francolin i F. afer (Muller) J Grey-wing . . j Coturnix delegorguei Delegorgue Harlequin Quail . ' C. chinensis (J. & E. Verreaux) ; Blue Quail .... I Ptilopachus petrosus (Gmelin) [I Stone Partridge Numida meleagris (Linnaeus) i Tufted Guinea-fowl ( Gallimda chloropus (Linnaeus) Moorhen .... Crex egregia (Peters) J African Crake 1 Limnocorax flavirostra (Swainson) ! Black Crake Eupodotis senegalensis (Vieillot) White-bellied Bustard . E. melanogaster (Riippell) Black-bellied Bustard E. hartlaubii Heuglin Hartlaub’s Bustard Otis kori Burchell , Kori Bustard I Neotis denhami (Children) Denham’s Bustard Burhinus capensis (Lichtenstein) Spotted Thick-knee Actophilornis africanus (Gmelin) Jacana .... Vanellus senegallus (Linnaeus) Wattled Plover SPECIES SAVANNAH RIPARIAN UPLAND -o 5 CO CO = w ± _ INc^TITIITIOn"' NOliniliSNI NVINOSHillNS SBiavaaiT libraries SMITHSONIAN institutic lINbMIUMUIN^ 2 ^ w 2 - ^ Z ’s3 I ava a LI B RAR l Es‘^SMITHSONIAN^lNSTITUTION‘^NOIiniliSNI^NVINOSHilWs'"'s3 I avaS dSsl d 5 'institution ^NOIinillSNI^NVINOSHimS^S3 lavaan^LIBRARI ES^SMITHSONIAN^INSTiTUTK S3iavaan~^'®^^’^'^^(y,^^'''’^soNiAN institution NoiiniiiSNi nvinoshii/js^S3 i avaa \ i INSTITUTION f^O>J-rUllSNI_NVINOSHillNS S3 I a Va a I1_LI B RAR 1 ES SMITHSONIAN_INSTITUTIC '' o iil ■' o ^ -j z -I 2 S3 1 ava a n^Li BRAR i es ^Smithsonian ^institution ^NoiifuiiSNi^NViNOSHiiws^sa i avaa INSTITUTION NOlinillSNI NVlNOSHilWS S3iavaaiT LIBRARIES SMITHSONIAN INSTITUTI liliSNi NViNosHiii^s S3iavyan libraries Smithsonian institution NoiifuiiSNi n\ - z !I “ "■ =) -o ^ ' [\osv^}>^ i I ^ ( ^ — CO _ CO « CO lAR I ES SMITHSONIAN INSTITUTION NOliDiliSNI NVINOSHilWS S3iavaaiT LIBRARIES S^