FOURTH REPORT WELLCOME TROPICAL RESEARCH LABORATORIES A T T M E GORDON MEMORIAL COLLEGE KHARTOU M .„..., ANDREW BALFOUR, M.D. ii'ii^ 1 DIRECTOR fell '' ^^bvui ^^Hiwiiiiiiiiiiiiiiiiiiiiii ^ \ '1b> { t^ By kind pemi-.^i^ioa of the Dii-ector, Sui-vey Department, Khaitnum Fig. 1 .— Map of Anglo- Egyptian Sudan showing Provinces ^ y J n^d., ^ y^i 'l^y' FOURTH REPORT ^ vXAi'' OF THE WELLCOME TROPICAL RESEARCH LABORATORIES <^ %1 £n-^, AT THE GORDON MEMORIAL COLLEGE ^ KHARTOUM VOLUME B.^Geiieral Science ANDREW BALFOUR, M.D., B.Sc, F.R.C.P. Edin., D.P.H. Camb. DIRECTOR Fellow of the Eoyal Institute of Public Health, the Society of Tropical Medicine and Hygiene, and the Incorporated Society of Medical Officers of Health ; Member of the Association of Economic Biologists ; Corresponding Member Societe de Pathologic Exotique and American Society of Tropical Medicine ; Medical Officer __ of Health, Khartoum ; etc. ^^T^ViSS^*'*" ^^^^'fujj^ U APR 2 7 1933 <") ^ ,293961 ^^, ^' Published for Department of Education, Sudan Government KHARTOUM BY BAILLIEEE, TINDALL & COX, 8, Henrietta Street, Covent Garden, LONDON 1911 [COPYUIGHT IV-J^A.. THE GORDON MEMORIAL COLLEGE AT KHARTOUM Patron : H.M. THE KING President : Fielu-Maksiial The Viscount Kitchener or Khaktoum, K.P (i C B G.C.M.G., O.M., ETC. Hon. Treasurer : TuE Right Hon. Lord Hillingdon Hon. Secretary/ : Baldwin S. Harvey, Esq., H7, Lombard Street, London, E.C. Committee and Trustees : Field-Marshal The Viscount Kitchener of Khartoum, K.P., G.C.B, G.C.1\LG., O.M., etc.. His Britannic Majesty's Agent and Consul- General in Egypt Lieutenant-General Sir F. Reginald Wingate, K.G.B., K.C.M.G., etc. {ex officio), Governor-General of the Sudan Alfred Clayton Cole, Esq. {ex officio), Governor of the Bank of England The Right Hon. The Earl of Cromer, G.C.B., G.C.M.G., O.M., etc. The Right Hon. Lord Rothschild, G.C.A^O., etc. The Right Hon. Lord Hillingdon The Right Hon. Lord Revelstoke, G.C.Y.O. The Right Hon. Sir Ernest Cassel, G.C.B., G.C.M.G., G.C.V.O., etc. Sir Henry Craik, K.C.B., M.P. Henry S. Wellcome, Esq. The Right Hon. Sir William Mather Hon. Solicitors : Messrs. Norton, Rose, Barrnigton & Co. Hon. Auditors .- Messrs. Gerard Van de Linde & Son Director of Education in tlie Sudan, and Principal of the Collecje : James Currie, F^sq. STAFF OF THE WELLCOME TROPICAL RESEARCH LABORATORIES, 1911 Director . . Research Chemist Pathologist and Assistant Bacteriologist Protocoulof/ist ami Assi,sta)it Bacteriologi Economic Entomologist . . Senior Assistant Chemist Junior Assistant Chejnist Senior Laboratory Assistant Junior Lahoratorij Assistant Cleric Junior Clerk Andrfav Balfour, M.D., etc. William Beam, F.I.C, etc. Captain R. G. Archibald, M.B., R.A.M.C., attached E.A. t Captain W. B. Fry, M.K.C.S., L.K.C.P., R.A.M.C, attached E.A. Harold H. King J. Thompson, Ph.D., etc. J. A. GooDSON, F.I.C. George Buchanan Alex. Marshall J. J. A. VlTALE, B.A. Nicolas Effendi Hakim ACKNOWLEDGMENTS The following have kindly contributed Papers to this Report : — ■ W. M. Aders, Ph.D. Captain R. G. Anderson, M.R.C.S., L.R.C.P., R.A.M.C, attached E.A. Captain L. Bousfield, M.A., M.D., R.A.M.C. A. L. Butler, F.Z.S., M.B.O.U., Superintendent Game Preservation Department, Smlan Government Service H. Curtis, F.R.C.S. S. C. Dunn, Government Geologist, Gordon Memorial College, Khartoum E. S. Edie, M.A., B.Sc, Carnegie Research Fellow in Chemistry, 1907-9 Sir Thomas Fraser, M.D., F.R.S. G. E. Hunt, Assoc. M. Inst. C.E., Gordon College, Khartoum Lieutenant W. E. Marshall, M.B., Ch.B., R.A.M.C., attached E.A. Colonel H. B. Mathias, D.S.O., M.R.C.S., R.A.M.C, P.M.O.E.A. W. H. McLean, Assoc. M. Inst. C.E., Gordon College, Khartoum C G. Seligmann, M.D. A. C. Stevenson, M.B., M.R.C.S., L.R.C.P., D.P.H. F. V. Theobald, M.A., F.E.S., Vice-Principal and Entomologist of the S.E. Agricultural College, Wye. Captain D. S. B. Thomson, M.B., B.Ch., R.A.M.C, attached E.A. Professor F. Werner, The University, Vienna CONTENTS VOLUME B — G E N E E A L SCIENCE PAGE Introduction 17 Effects of Fire — Staff Ohaiigcs — Private Workers — Title of l^aljoratories — Second Keview Supplement — Volume B — Vai-iety of Subjects — Ornithology — Arachnida — Protozoology — Ancient Gold Mining — Ethnology — Anthropology — Tropical Sanitary E)igineering — Entomology — Chemistry — Soil Analyses — Calotropix jiroce)-(i -- Lilirary — distinguished Visitors — Scientific Staff — Future of Scientific Work in the Sudan. Ueport of the Chemical Section. Ey William Beam 23 Introduction — List of ^Vnalyses and Examinations — Scojie of ^\'(lrk. A Test for Hashish— Details of Test — Variations of Method. Khartoum Water-supply — Notes— Physical Characters— Tables of Analyses— Imjirove- ment in Storage — Corrosion of Pipes — Action on Lead. Sobat River Water — Suitability for Irrigation Purposes — Tables of Analyses. ilECHANlCAL ANALYSIS OF Aeid Soils— Various Methods — Tables of Analyses— Centri- fugal Method: Size of Clay Particles — Determination of the "Silt" — Clas.sitication of Soil Particles— Resume of the Method Used. Soils of the Gezira— Native Crops— Rainfall— Classes of Soil— Exiicrimentul Farm- Effect of Baking — Chemical Comjiosition — Tables of ilechanical Analyses — Tallies of Chemical Analyses — Detailed Tables of Analyses. Gypsum as a Fertiliser for Sudan Soils — Lime in Soils — Fertiliser Exiierinients on Cotton Soils — Effects of Application of Gypsum to Soils — Amount to be applied. Gum Research — ^licrobial Origin — Tapping Experiments — Influence of Cold — Method — Varieties of Haxliah Gum — Viscosity — Tai)iiiiig of Talh Trees — Conclusions. Experiments on Gum Production in Kordofan. By E. S. Edie 73 Investigations in Gum Gardens — Tables of Yields from Small Trees — Variations in Yields — Effects of Rain — Sizes of "tears" — Tables of Yields from Large Trees — Maximum and ilinimum Temperatures — Best Time for Tapping — Viscosity and Acidity of Gum from Small and Large Trees — Methods of Determination — Effects of Treatment of llashah Trees on the Yield : Effects of Strong Antiseptics ; Table of Results — Bacterial Origin of Gum ; Transmission of Infection ; " In vitro " Experiments in Gum Production — Analysis of Resulting Bacterial Products — Lalioratoi-y Confirmation of Work in Gum Garden. Preliminary Notes on the Ciiemistkv of the Latkx of Calotkopis puocera. By J. Thompson 85 Native Uses of Ushar Plant — Investigations iinStem-borer — Black or Greasy Cutworm — Asal Fly — Cotton Stem-borer — Melon-Weevil — !Meloii Stem-borer — ^lelon Leaf Beetle — Diamond-back Moth — Cabbage Bug — Kigla Gall Weevil — Fig Stem- l>oring Beetle — White Ants — Black Garden Ant. Plant Diseases Ordinance — Explanatory Note — Contents — Definitions — Articles Exempted — Detailed Explanations of Contents. A Xew Genus and Two New Species of Culii.'im.k from the Sudan. By F. V. Theobald 151 Genus, Diciroini/la : Diceromyid ri/ncnnii, n. sp. — C'ti/cr .mdiiiteiisif:, n. s]i. The Finches and Weaver Birds of the Sudan, BEiN(i Notes on the Group containinc THE Birds Injurious to Grain Crops. By A. L. Butler 157 Introduction — Utility of Birds as Insect Killers. Part I., The Families, Sub-Families and Genera— Family I., Frinyii/kke: Sub- families, Eiiiherizimt ; Friiu/illhut — Family II., Fluceidit : Sub-families, Viduiiuf ; Estrildinui ; Ploceimt. Part II., Notes on the Species — Family, Fi-ingillidw — Sub-family, Endierizime : Genera, Emheriza ; Fringillaria — Sub-family, FrinijUlhio' : Genera, Eri/fhrospiza ; tSerhms ; Poliosjnza ; Passer; Petronia — Family, Ploceidte — Sub-family, Viduiiue : Genera, Hjipochera ; Vidua; Coliuspasser ; Urobrachya ; Pyrumelana ; Quelea — Sub-family, Estrildimt : Genera, Amadina ; Philetairus : Uroloncha ; Ortijijospiza ; Spermestes ; Uraeginthus : Esfrilda ; Lagonosticta ; Pytelia — Sub-family, Ploceimt : Genera, SjieriHospizu ; tiporopipes ; Amhlyosjyiza ; Dinemellia ; Textor ; Ploceijiasser ; Aiiaj)/ectes ; M(diiid)us ; Hyphanturyus ; Sitagra ; HypJiantoi-nis ; Xanthophilus. Additional Note on a Species of Lark, Mdanocoryphu liimaculaia. Scorpions and Allied Annulated Spiders of the Ancslo-Egyptian Sudan. By Professor F. Werner 179 Characteristics of Scorpions — Habits — Poison — Distriljution — Scijrjiions Found in Deposits of the Silurian Period — Virulence of Poison — Symptoms of Poisoning — Twn ilain Divisions of Sudanese Scorpions. Descriptions of Genera — Buthus : B. amoreuxi ; B. (jiu'tKjucstriatus : B. occitaiiiis : 11. acuteairinntus : B. lepitochelys ; B. minaj: — Parabuthus : P. hvnteri — Nanobutlms : X. audersoni — Pandinus : P. pallidus ; P. exitiaiis ; P. imperator — List of Sudanese Scorpions. Characteristics of Pediiialjii — Comparison with Scorpions — Descriptions of Geiuis Ddinun : D. variegatus. Characteristics of Solifugai — Distribution — Habits — Main Divisions of Species — Descriptions of Genera — (ialeodes : G. arahs — Paragaleodes : P. scalai-is — Othoes : 0. floweri — Rhagodes : Rk. melanocepihalus — Solpuga : ti. deutatidens, S. acliweinfurthi — Diesia : D. ilin-nhergi — Gluvidjisis ; G. ru/escens. Notes on 'i'wd Specimens of Spitting-Snakes from Southern Rhodesia. By Henry Curtis 105 Eti'ect of Ejected Fluid — Spitting Cobra — Xain tiigrimllis — Venom — Reproductions of and Extracts from some Recent Literature. A Note on the Sudan Spitting-Snake. By Sih T. R. Phaser 201 Herpetomonas aspongopi. By W. il. Aders -'ol' Hyiier-jiarasitisni — Melon Bug — Alimentary Tract — Method of Examination — Structure and Life-cycle of the Parasite — Resting Stage — Process of Division — Concluding Remarks. CONTENTS 9 PAGE Anoiknt OoLii .A[iN-iN(; IN Till-; Sudan. I!^ S. C. Dunn liOT In Earliest Times— Thu Phienieiaiis — The Egyirtiiins — Turin Papyrus— Wealth of Ancient Egypt — Foreign Domination in Egypt — (Jutic Inscriptions — Method of Mining — .Slave Workers— The Gold Ore— Hewing— Pounding— Milling— Washing— Melting— Grades of Gold — Old Workings — Means of Illumination — Quartz — Preliminary |}reaking — Circular Grinding Mills — Washing Tal)lcs— Melting Pots — Slag — Caini>s— Wells — Buryuig Places — Potsherds. Cult of Nyak.vm; and tuk Divink Kini:s hv rnv. Shilluk. \',\ ( '. G. Sulicmann ... 216 Preliminary Remarks — Shilluk Country — Population — Shilluk Kings — Killing of the King ami Transmission of the Divine Sjiirit — Cult of Nyakang at his Cenota|ih Shrines — Cult of the Shilluk Kings at their Grave Shrines — Reverence for Trees — Apijearance of Shilhik Kings in .Vnimal Form — Concluding Remarks — Some Tribal Customs in tiihiii Kklation to the Medicine and .Mouals oi- TnE Nyam-nyam and Gour Peoples inhabiting the Bahr-El-Ghazal. 1!v Captain R, G. Anderson 239 Introduction — Tribal Distribution — Nyam-nyam People — Gour People — Character and Type Civilisation and Mohammedanism — Morality — PunLshment — Mutilation — Execution : iSy Throttling, By the Knife, By the Spear, By Poison, By Fire — Ordeals — Flogging — Theoiy of Disease — Doctors and their Cures: Salt Cure, Sucking Cure, E.xti-action of Foreign Bodies, Tying Cure, Scarring, Cupping, Heat C!ure, Spitting, Whistling Cure -Oracular Rites: Benga Rite, Devil-dancing, Divination, Divining Board, Other Means of Divination — Evil Spirits : Sahar or Zarr, Blood-brotherhood, Casting out Devils, Sacrifices — Chai'ms : Medico-Magical Charms, Attractive Charms, Whistle Charms, " Gun Medicine," Mourning Charms, Lightning Charms — Maladies : Fever, Chest Complaints, Eye Diseases, Chigger, Guinea- Worm, Snake Bites and Scorpion Stings, Stomach Complaints, Venereal Disease, Headache, Sun-stroke, Mental Disease, Skin Disease, Leprosy, Deformities, Elephantiasis — Drugs — Arrow Poisons — Surgery: Circumcision, Tribal Cosmetic Operations — Extraction and ^lutilation of Teeth — Tribal Markings (Scar Tattooing) : Spurious Face ilarkings, Lip Perforation, Nose and Ear Perforation — ilinor Surgery : Wounds and IHcers, Fractures, Foreign Bodies, Surgical Instruments and Appliances — Personal Hygiene — Cannibalism — Birth — Death — Conclusion. Some .Municipal En(;ineerino Problems in the Tropics. By W. H. McLean and G. E. Hunt 279 Introduction — Meteorological Conditions. Town Planning — Old Khartoum — New Khartoum— Future Development. Streets and Open Spaces — Streets — Rain Water 1 )rainage— Tramways-- Open Spaces— Trees. Buildings — Types of Houses. Water-supply — Distribution — Wells Irrigation. Salutation — Waste Water Dispo.sal : Irrigation, Pei-colation Pits, Waste Water Pits— Latrines and L'rinals — Conservancy. Town Building Eegidations — Contents— Preliminary : Title, Scope, Interpretation, Classifica- tion of Land— General Regulatii)ns— Si)ecial Itc.gulation.s— Building Petition Drawings Lodged \\-ith Municipal Authority— Contra ventiim of Regulations- Dilapidated or Neglected Buildings. Index 313 10 CONTENTS VOLUME A — M E D I C A L PAGE Intkoduction lil Ell'ects of Fire — Staff Changes — Private Workers — Title of Laboratories — Second Review Supplement — Volume A — Papers Contributed — Scope of Work — Acknowledgments — Volume B — Papers dealing with General Science — Acknowledgments — Laboratory Steamer — Museum and Library — Acknowledgments— Scientific Staff — Future of Scientific AVork in the Sudan. Sleeping Sickne.ss in the Anclo-Egvptian Sudan. By Col. H. B. Matiiias, ]).S.0. ... 31 Investigation of Conditions in the Bahr-El-Chazal — Natural Barrier against Sleeping Sickness — Quarantine Ordinance — Cases in Western Bahr-El-Ghazal — Examination of Natives — Cases in the Lado — Infected Area — Summary of Precautionary Measures — Additional Note. Anim.\l Trypanosomiases in the Anglo-Egyptian Sudan. By Captain W. B. Fry ... 41 Present Confusion of Classification — Necessity for Systematic Nomenclature — Review of the Trypanosome Types — Methods of Differentiation —Type 1, T. Iirucei oy pecaudi — Morphology — Curves of Distribution — Animal Inoculaticnis and Reactions — Type 2, T. evansi — ilorjihology — Curves of Distribution — Animal Inoculations and Reactions — Type 3, T. pecoru/a or nannin — Morphology — Curve of Distribution — Animal Inoculations and Reactions — Tyiie -1, T. phxir or cazalboui — Morphology. Some Observations <>n Trypanosoma brucei (pecaudi?) and the Sudan Camel Trypanosome in Cultures, with a Note on Endoglobulak and Develop- mental Forms of T. brucei (pecaudi 0- By G. Buchanan 57 Technique : Culture Media, Inoculations — Ajipearance in Culture : After 24 Hours, After 48 Hours, After more than 48 Hours, Stained Films, Virulence of a 22 Days' Culture on Gerbil — Remarks : Culture of Trypanosomes. Note on Developmental Forms of T. Iirucei (-pecaudi?) in the Internal Organs, Axillary Glands and Bone-Marrow of the Gerbil (Gerhilhis jn/gatytts) — Changes of the Parasite — Morphology and Develojiment in the Spleen : Intra-corpuscular, Extra- corpuscular — Forms Found in Lung Smears... ... ... ... ... ... ... ... 59 Notes on Human Spiroch.ktosis. By Captain L. Bousfield 02 Six Cases of Relapsing Fever — Possible A'ectors of the Di.sea.se — Summary of Clinical Asjiects of 1 )isease -Spirochaites in Blood — Teniiicraturc Charts. The Spirochete of Egyptian Relapsing Fever. , By the Director 07 Morphology — Staining Reactions — Animal Inoculations — Chicken Feeding and Insect Trans- mission Experiments — Bed-bugs — Lice — Ticks — Inferences from Experiments — Tables of Comparison — Probalile Identity of Egyptian Spirochastosis with that due to Sji. hcrhera. Spiroch.etosis of Sudanese Fowls. By the Director 76 Summary of Previous Paper — Brief Account of Recent Literature ; Up. inarckou.ri, Sp. tieveuxi, Sp. anserina — Preliminary Experiments — Dark-field Method — Differential Staining— Experiments in Culture — Action of " C06 "— " Afterphase " Bodies — Differences of the Disease in Chicks and Fowls — Experiments — Possibility of Hereditary Transmission — Close Connection between Intra-corpuscular Bodies and Acvite Spirochastosis — " Graiiules " in A. ])ersicus — Significance of Experiments — 0. aavignyi — Granule Infection — Fowl Lice — Blood Changes in Infected Chicks— Treatment— Effects of "606" on Spirochetes— Favourable Results— Effects of "606" on Intra-corpuscular Forms— Mechanism of the Crisis— The Infective Granule — A cycle of Schizogony — Importance of the Dark-tield Method — Summary and Conclusions — Proposed name for Sudan Fowl Spirochete, Sp. f/rmiulosa (pienetrans). Note on a Method of Obtaining Blood Askpticali.y foi; 'I'tie Culture of H.kmatozoa in the Tropics. By the Director 107 CONTENTS 11 PAGE Fallacies and Puzzles in Blood Examination. By the Director 109 "Adventitious" Fallacies — Yeast Cells, I^ollrn (Jniius, etc.— Spirocliaete Forms— Parasite- like Bodies — Differences between Yeast Cells uiul Parasites— BlottinK-pajier Fallacy — Autogenetic Fallacies — Vacuoles — " Maraglianos '" — (iraiiules — " I'liain, duiub-bell, Hlameiit auil droplet" Phenoiuena — Platelets — Hwnioconia — Free Flagella — Fallacies in Stained l>lootl—,S)>iHr/('/,:e//eH— Alterations in Red Cells— Curious Appearance in Gerbil's Blood — " Howell-JoUy " Bodies— Centrosonies in Bird's Erythrocytes — Unrecognised Bodies — Death and Degeneration of Leucocytes — Bodies in the Extra-nuclear Protoplasm of Certain Mononuclears — Sergentella hui/iiuis — "Howell and Hoii'ocks " Bodies — Unidentified Parasites in Frog's Blood. Hemarks on Kala-azar in the Kassala and Blue Nile Districts of the Sudan. By Captain L. Bousfield 127 Summary of Tour — Table of Patients Treated -Prevalence of Bed-bugs — Duration of 1 >isea.se — Incidence of the Disease — Nationality of Cases — Death-rates of Towns ^Vti'ccted- -Onset of Disease — Physical Condition of Patients — Liver and Spleen Puncture — Atypical Bodies — ] )iag'nosis — Types of Parasites — Blood Films from Animals — Recommendations — Appendix. Kala-azae Commlssion to investigate the Prevalence and Cause of the Disease IN THE E.vstern Sudan. (1) General Report. By Captain D. S. B. Thomson 143 Tours of Inspection — Procedure in Villages — Native Information on Kala-azar — Native Treatment — Infection of Riverain and Inland Villages — Comparison of Number of Cases in Provinces — Method of Transmission — S])leen Puncture — Age, Sex, Race and Seasoual Incidence — Is the Disea.se Curable! — Bilharziosis — Reeoniinendations — Conclusion — Tables of Observations. (2) Pathological Report. By Lieutenant W. E. Marshall 157 Clinical Notes — Treatment — Perijjheral Blood — Presence of Leislimaii-I tcpuovau Bodies- Relative Increase in Large Mononuclear Leucocytes — Other Blood (_!hanges — Relation of Eosinophilia to Kala-azar — Ih-ine and F;eces in Kala-azar — Search for Leisliman- Donovan Bodies in Dogs and other Animals — Tables of Oliservations — Experimental Kala-azar in the Grey Monkey — Infected Cases — Non-infected Cases — Presence of Parasites in Blood of Infected ilonkeys — Tables of Observations — Morphology and Culture of the Parasite — Possibility of Infection by the Bed-bug, Cimex lectularius — Other Protozoa met ^^■ith during the Investigation — Protozoon from post mortem S])lenic Puncture — Flagellated Bodies in Standing Water, in Fieces of Infected Monkey, and in Soil — Conclusions — Possible Methods of Infection. The Alkalinity of the Blood Serum in Kala-azar. By ("aptain R. G. Archibald 173 Technique — Titration of Serum a.iiuinst Standard Acid Solution — Cases 1-4— Conclusions. Herpetomonas lyg.ei. By Captain \\. (i. .\rchibald 179 ^Material and ^Methods Emiiloyed— Pretiagellate Stage — Post-tlagellate Stage .Metliod of Infection — Culture and Animal l^lxpei'inients. Case of Kala-azar treated by "OOG." \\\ the Director and Captain Ai!chibald... 185 Summary of the Case — Diagnosis Confirmed by Spleen Puncture — Injections Commenced — Second Injection — Improvement — Effect on Parasites — Remarks. Two Cases of Non-Ulcerating "Oriental Sore," better termed Leishman Nodules. By the Director, with Clinical Notes by Captain D. S. P.. Thomson ... 191 Introduction — Case I — History — Clinical Examination — Blood Examination Growths — Morphology of Parasites — Histo-])athology of firowths — Recent Literature— Case II. is found to be a Comrade of Case I.— History -Clinical Examination— Blood Examination — Additional Notes — Treatment — Experimental Work. A Cask of Parasitic Granuloma in which Developmental Forms of Leishmania tropica were Present. By ('aptain R. G. Archibald 207 History of Case — Study of Skin Lesions — Examination of Papules — ^Microscopical Ivxamina- tion of Stained Specimens — Ditferential j'lood Count Histo iiatholog>- - Summary. 12 CONTENTS PAGE Two Casks uk \'klut Soke Teeated by Autogenous Vaccines. By the Dieectok ... 213 Vaccine-therapy — Case I. — Case II. — Successful Treatment l)y Autogenous ^'accmes — Use of a Mixed Vaccine. Ulcus teopicum. By the Dieectoe 217 Leucodeema. By the Dieectoe 217 Fevees in the Sudan, Peeliminaey Note. By the Uirectoe 219 Classes of Fevers — Case of Hepatitis — Case of Malta Fever — Plilebotomus Fever or " Sand-tiy " Fever — Dengue Fever — "Toxic Enteritis" — Miliary Tuberculosis — Prolonged Fever of Syphilitic Origin — Helminth.s as a Cause of Fever — Fever due to Old Malarial Infection — Idiopathic Tropical Splenomegaly — Septicajmic Fever — History of Cases — Three Cases of Pyrexia of Unknown Origin — History and S>'miitoms — Classification of Fevers — Temperature Charts of Cases. Four Cases of Pyrexia due to Infection with a Bacillus of the B. cloac.e Type, with Notes on the Vaccine Treatment of Two of these Cases. By Cai'tain II. G. Archibald 231 Blood Cultvire for Early Diagnosis — Clinical Signs — Isolation of the Organism — iloi'iihology — Cultural Characters — Agglutination Tests — Cultural Reactions — Presence of Organism in Excreta — Outcrop of Boils — Success of Vaccine Treatment — Conclusions — Temperature Charts. Diphtheeia in the Tropics. By the Director 239 Diphtheria in the Sudan — A Coccal Form of the Diphtheria Bacillus — Seasonal Incidence — Case in Khartoum — Bacillary Form — Involution Forms — Variation in Bacilli — Additional Note — Associated Organism — Characters of the Organism — Mori)hology — Pathogenicity. Some Aspects of Tropical Sanitation. By the Dieectoe 249 Tfie Sanitary Insjiector in the Tropics — Eftects of Sun and Humidity on Health — Native Populations — Malaria and Mosquitoes — Unexpected Breeding-i)laces — Other Communicable Diseases — Conservancy Method.s — Bacterial Treatment of Sewage — Proposed Sewage Farm — Incinerators — Birth-rate — Milk Supply — Miscellaneous Duties — New Khartoum. Sanitary Notes. By the Dieectoe 263 Meteorological Conditions — Mosquito Work — Statistics — Malarial Outbreak — Irrigation Rules — Influence of Cultivation on Diseases — The Covering of AVater Utensils — Eats in Khartoum — Infectious Diseases — Tables of Statistics — Vital Statistics — Death-rate — Birth-rate — Liability of Sudanese to Tuberculosis. The Watee-Supply of Towns in the Tropics. By the Director 289 Significance of B. roll rommnnh in Water — Bacterial Flora of Tropical Water-supplies — Clemesha's Observations — " Coli " Standard as apjjlied to India — Comparative Value of Tests — B. cloaccv — Varying Powers of Resistance to Sunlight amongst Coliform Organisms — Clemesha's Conclusions — Bacterial Standards in Tropical and Temperate Climates — Shallow Well Water— Classes of Fwcal Organisms— Blue Nile Water— Collection of Samples — Khartoum Water-supply — Geological Conditions — Air-lift Pump — Quantitative Examination — Qualitative Examination — Danger of Crenothrix — Pumping Trials — Report of Mr. Abel and Dr. Todd — Solution of Difficulty — Conclusions — Extracts from Government Report — Tables of Tests — Appendix — Notes on Distriljution. The Presence, Type and Possible Significance of Lactose-feementinc; Bacilli in SuEFACE Waters and in the F.eces of Man and Ceetain Domestic Animals IN the Sudan, to(;ethee with a Consideeation of the Effects of Sunlight and Earth Burial on these Organisms. I'heliminary Note. By Captain E. G. Aechibald 319 Standards of Purity — Observations on Indian Water-supplies — Khartoum Water-supply — Technique — ] differential Sugar Media — Tables of Results — Conclusions — Additional I'^xperimcnts —Lactose Fermenters in Human Fieces — Effect of Sunlight — Effect of Eai'th liiiiial cm the liacterial Content — Clemesha's List of Lactose Fermenters — Tables of Results. CONTENTS 13 PAGE KXPHRIMEXTS l)N THE FiLTKKIXi: I'lIOI'KIJTIKS (IK TIIK /kICI.'. \'>\ CaI'I'AIN I!, 'i. A lirll 1 1; \ Ml 335 Varietii's of Zeers — Action ol' (he Slimy |)c|i(isil ( )|iliiiiiiin ( 'omlit inns t'di- Zcci' Killi;il imi. Human Boteyomycosis. By Captain li. (! Ai!iiiibai,I) 337 l-'reliminary Remarks — Clinical Signs of the Disease — Tendency t(j a Chrunic Course — Fibrotic Changes — Microscopic Appearances — Staining Reactions — Degenerative Changes — Conclusions. Veterinary Notes. By the Director 343 Ejiizootic Lymphangitis of Equines — Coccidiosis in Cattle — Piroplasmosis — Anai)lasniosis — Filaria.sis : in the Horse, in the Camel, and in the Hare. Parasites. Acid-fast Bacilli in the Lt'Nc; of a Camkl. By Captain R. CJ. Auchihald ... 352 Interstitial Pneumonia in a Camel's LuNf;. By Captain H. (i. Archibald 353 Coccidiosis of the Inte.stine in the Goat. By A. C. Stevenson 355 Ca.se for Diagnosis — Post mortem — Other Cases — .ScluzoKony Cycle— SporoKony Cycle — Microgamete Formation — Pathogenicity. A Few Notes on the Protozoa Parasitic in Bufo eeculaeis in Khartoum. 1'>y A. C. Stevenson 359 Protozoa in the Blood : TriijKinomtnn rotntoriii/H, Ha?mogregarines — Protozoa in ihr Tntcstinc : Flagellates, Ciliates. "SI iscellan Eous Notes. H.ematozoa. By the Director 362 Howell-Horrocks Bodies in Human Blood. By the Director 3ri2 Description of Case — Types of Bodies Found — Measurements — Attempts at Culture. My'Cetoma. By the Director 365 Leprosy. By the Director 367 Dysentery. By the Director 367 Peculiar Bodies found in the Intestinal Lvmpiioih I''oi,licles of an E<;yptian. By a. C. Stevenson 369 Routine Work. By the Director 370 Index 371 CONTENTS OF THE SECOND KEVIEW OF RECENT ADVANCES IN TROPICAL MEDICINE, ETC. VOLUME SUPPLEMENTARY TO FOURTH REPORT Air — Ankylostomiasis — Anthra.x — Bacteriologj' — Beri-lieri — Beverages — Blackwater Fever — Blood— liulio— Calabar Swellings— Cancer — Cereliro-Sjiinal Fever— Chicken-i)ox—Chigger — Cholera — Climate — Clothing — Dengaie — Diarrhcea — Diphtheria — Disinfection — Dropsy — Dust — Dysentery — Elephantia.sis — Enteric Fever— Fwces — Fevers — Filariasis — Filters — Flies — Food — Food Poisoning — Goundou — Guinea- Worm — Hsematozoa — Heat Stroke — Hydatid — Hydrophobia — Influenza — Insects — Leishmaniasis — Leprosy — Liver Abscess — JIalaria — Malta Fever — Jleasles — Jlilk — ^Mosquitoes — Mycetoma — Myiasis — Ophthalmia- Oriental Sore — Parasites — Paratyphoid Fever — Pellagra- Phlebotomus Fever- Piroplasmosis — Plague — Pneumonia — Preservatives — Protozoa — Refuse Disposal — Schistosomiasis — Scorpion Sting — Scurvy— Sewage — Skin Diseases — Sleeping Sickness — Small-pox— Snake- bite — Spirochsetosis — Sprue — Staining — Syphilis — Tech nii pie — Tetanus — Ticks — Tropical Medicine— Trypanosomiasis — Tsetse Flies— Tuberculosis- Typhus Fever — Vaccination- ■ Vermin — Verruga — Veterinary Diseases — Water — Whoo|iing Cou^'h — Yaw.s — \ ellow F(!ver. 14 LIST OF ILLUSTRATIONS VOLUME B Fig-. 1. — Ma]i of AiiS'ln-K<;y|itian Sudan, showing I'rovinces ... ... 2 Fig. 2. — Floating Laboratory and S. W. C'ulex 16 Fig. 3. — Mechanical Shaker for Soil tubes ... 35 Fig. 4. — Centrifuge for Soil analysis 39 Fig. 6. — Ajiparatus for the determination of "Silt" 41 Fig. 6. — Map showing mean rainfall in the Anglo- Egyptian Sudan, 1905-9 ...4(;-47 Fig. 7. — Map .showing proposed Gezira Irri- gation Scheme 49 Fig. 8. — Map of the Gezira, showing points at which samples of Soil were collected for Analysis ... ... 51 Fig. 9. — Calofropis jyrocera, showing leaf, flower and seed-j)od ... ... 87 Fig. 10. — Calofropis procera, showing entire plant 87 Fig. 11.— Map of Anglo-Egyptian Sudan ... 94 Plate 1. — (Jhri/sops fuscipennis. Ho'ina- topota abyssinica. Haematopioia hmnnescens. Tnlianus sufis Facing 112 Plate II. — Tabaniis distinctus. Tabanus kingi. Tabanus camelariiis. Tnhanus mordax ... ... Faciuf/ 114 Plate III. — Tahaimx fn^niohi. Tabanus par ... ... ... ... Facimr 116 Plate IV. — Tnhamts difrfniatiis. Gastro- ji/iihi.t ti.iimis. (Esfriis variolosus Facing 118 Fig. 12. — Heads of '/'abainix kiiigi and Taliainis timiola ... ... ... 120 Fig. 13.— View of Khor Arbat 121 Fig. 14. — A rock at i\lior Ai-bat showing sites selecteil l.iy T. kingi for ovipositing ... ... ... ... 122 Pl.^te V. — Tabanus diUvniatiis and kingi 123 Plate VI. — Lyperosia exigiia. Lyperosia miniifa. Lyperosia thirov.vi. Lynchia maura. Echestyp>us sepiaceus. Nycteribosca africatia. Simulium r/riseicollis. Simuliuni daiiinnsum. Culicoides, sp. Facing 126 Fig. 15. — hiirvix oi (Fstrns variolosus ... 127 Fig. 16. — Pupa of CEstrus variolosus 127 Plate VII. — Sphenajitera neglecta ... ... 135 Plate Vlll. — Baris fru-gardlii. .Ijionifri/iin hinilbihi ... ... ... Faring 137 Plate IX. — HinoxyUDi smlanicnm. /laria loratn. Ajjlvvnogaster barbara. Ardacopliora foveicollis. Bagradii picta ... ... ... Facing 140 Yig.ll.-^Diceromyia afi'icana ... ... 152 Plate X. — Diceromyia africana, n.ti'p. ... 153 Fig. 18. — Diceromyia africana, ^ clasper ... 154 Pl.\te XI. — Chdex sudanensis, n.sp 155 Fig. 19. — Culex sudanerisis, $ clRnper ... 156 Plate XII. — Quelea (xthiopica ... Facnig 1()7 Plate XIII. — Hyphantorjiis fmiiojiterus Facing 175 Pl.vte XIV.— Buthidie 178 Fig. 20. — Buthus oca'tajiws, from above ... 179 Fig. 21. — Buthus occifanus, abdomen from below 179 Fig. 22. — Last two segments of cauda of Buflius amm-euxi ... ... ... 183 Fig. 23. — Last two segments of cauda of Bufltiis quiuqnestrintns ... ... 183 Fig. 24.— End of foot of a .S)(Y/i«s 183 Fig. 25. — Cephalothora.x of i??///i«s wwnd.c... 183 Fig. 26. — Movable finger of maxillary palp oi Si Parabuthns ... ... ... 183 Fig. 27. — Stenvxm oi Pandinus 183 Plate XV. — Pandinus imperator. Pan- dinus pallidus. Buthus acute- carinatus. Damon variegatns ... 189 Fig. 28. — Maxillary palp of Damon variegatus 190 Fig. 29.— Mandible of a ffaZeocies 192 Fig. 30.— Mandible of a i?/^n;/orfe« 192 Fig. 31. — Upper mandibular finger of Solpuga schtoeinfurf/d ... ... ... 192 Fig. 32. — Mandible of Gluviojisis rufescens 192 Fig. 33. — Mandible of 7)(e. to 01 nil 28-5 ,, 01 to (12 moderate 28-5 to 31 „ fairly large 02 to 71 trace 31 to 37 „ large i 71 to 80 nil 30 REPORT OF CHEMICAL LABORATORY The analyses as a whole, it was found, seemed to prove that " the best water from a chemical point of view is that of Well No. 7, which is fed mainly from the deeper layers." The recommendations made by Dr. Todd and Mr. Abel, the engineer called in consultation, were, among others, to deepen certain of the wells and to cut off the upper strata by the method detailed by Dr. Balfour in his note on these wells (loc. cit.). The wells in use at the present time are Nos. 1, 6 and 7. Many examinations of these have been made in the last year, the most recent results being as follows. They are from Wells Nos. 1, G and 7, the others having been abandoned as unnecessary. Parts per million Well No. 1 Well No. 6 Well No. 7 Collected Dec. 29, 19in Nov. 9, 1910 Nov. 9, 1910 Total solids 208-00 227-60 195-20 Free ammouia (NH,) 0-50 0-60 0-48 Albuminoid aminoiiia (NH,) 002 none 002 Nitrites (N) none none none Nitrates (N) 001 001 001 Chlorides (CI) 1-44 2-20 1-90 Sulphates (SO,) .3-79 .3-80 3-90 Carbonates (CO,) 109-80 117-60 103-20 Calcium (Ca) 44-80 45-60 40-40 Magnesium (Mg) 12-28 13-84 12-97 Sodium (Na) 5 82 8-18 5-97 Potassium (K) 2-21 4-70 5-48 Iron (Fe) 0-18 0-32 0-12 Manganese (Mil) 018 0-16 0-20 Further improvement The above analyses represent the water not as supplied to the town, but directly it emerges from tlie air-lift. A further very marked improvement is effected by storage in a large cement-lined reservoir. As a result of the very efficient aeration, due to the use of the air-lift, not only is the iron oxidised and precipitated, but, what was entirely unexpected, the manganese is carried down as well. The precipitation of the iron and manganese probably also aids the direct oxidation of the small amount of organic matter present, the effluent water attaining a remarkable degree of organic purity. The following is a comparison of the water as it emerges from the air-lift KHARTOUM WATER-SUPPLY 31 pipes with that taken from the tap in the Laboratories, i.e. after it lias heen allowed to stand in the reservoir for 48 hours or ovei\ The water at the time was the product of Wells Nos. 6 and 7. From air-lift From tap in laboratory Free ammonia (NH,,) 0-54 none Albuminoid ammonia (NH:,) 0-01 none Nitrites (N) none none Nitrates (N) 0-01 0-36 Iron (Fe) 0-22 o-or) to 0-80 Manganese (Mn) 0-18 none or trace The following determinations of iron and manganese in the tap water were made at various times. Date Taken from tap at Iron Manganese Jan., 1909 Medical Meas 0-06 none May, 1909 Military Hospital ... 0-'20 none Nov., 1909 Dr. Balfour's House 0-4,'-) none „ 1909 Sudan Club 0-'22 none Oct. 31, 1910 Chemical Laboratory 0-12 none Not. 1, 1910 ,, ,, 0-80 trace „ 2,1910 ,, ,, 0-20 0-01 „ 3, 1910 ,, ,, O-.-iO none Jan. 4, 1911 ,, ,, 0-06 trace „■ 5,1911 ,, ,, 0-05 none The iron present in the tap water is largely in suspension. It is apparently due to a slight corrosion of the pipes, since the surface water of the reservoir, after only eight hours settling, contains as a rule a smaller amount. There is also the fact that if it were that originally present, it would be associated with a more appreciable proportion of manganese. (iSV« helow, the analysis of sediment from the reservoir.) The cause of the corrosion is doubtless, in part at least, galvanic action, the result of Corrosion the deposition of a small amount of manganese and iron oxide in the pipes. The absence of manganese and presence of iron in the tap water was at first thought to be the result of a slight growth of cmnothrix in the pipes, but no growth of this kind has been detected since the deepening of the wells and the cutting off of the contaiiiiiiated water of the surface strata. An analysis, by Dr. Thompson, of the sediment deposited in tlic storage reservoir gave results as follows : — 32 REPORT OF CHEMICAL LARORATORY Loss of ifrnitiou 17-46 per cent. Iron oxide ... ... (Fo., 0„) 29-40 ,. Manferancsc protosescpiioxide (Mn, OJ 18-117 ., Aliunina (Al., O,) 3-97 „ Lime (CaO) 0-2.-) ,, Maenesin ... (MrO) S-.")? „ Silica (Si O.,) 21-4S ,, Siilpliates ... (SOJ trace Total 99-30 Plumho-.iolvenci/. The solvent action of the water on lead was tested by means of a section of new lead pipe, about 2 metres long and 1-3 centimetres bore. The water of Well No. 7 only was taken for the test, but as the waters of these wells differ but slightly in composition,, the results, I think, may safely be taken as representative of tlie supply as a whole. The lead pipe having been fitted with a stop-cock, it was filled witli the w-ater which was allowed to remain for twenty-four hours. It was then run off and tested, and the Action on lead pipe refilled. The tests were carried out in this w-ay for a month. In the beginning, the amount of lead was approximately 0-7 part per million. The amount dissolved gradually diminished, but at the end of tlie month the proportion found was still considerable, l)eing about 0-44 part per million. Suitability for irrigation purposes SoBAT EiVEE Water In 1908, an attempt was made to carry out an investigation of the composition of the water of the Sobat River, and Mr. Walsh, then of the Sudan Irrigation Service, kindly undertook to collect the samples for the purpose. It was intended to make a series of monthly examinations for an entire year, but Mr. Walsh, by reason of the press of other work, was unable to secure more than nine samples. These were nevertheless well distributed throughout the year and the results probably represent fairly well the usual composition of the water and its variations. The results of the examinations are stated in the table on the next page. As will be seen by comparison with those from the Blue Nile and White Nile, which have already been recorded,' the Sobat water contains, on the whole, lower proportions of dissolved solids than do any of the branches of the Nile examined up to the present. The proportions of sodium carbonate and potassium carbonate are very markedly lower than is found in the other tributaries of the White Nile or in the White Nile itself, and the Sobat water should therefore be superior to these for irrigation purposes. One may also reasonably expect the soil of the district through which this river flows, and which may be subjected to inundation by it from time to time, to be of better quality than that flooded by the White Nile. ' Third Report, Wellcome Tropical Research Laboratories, pp. 386 to 395 SOUAT ]!IVT':i! \V\Ti: 33 O I-- o S t^ CO CJ V o -t w CO 2; C oc X 5i ■^ 1-* w G^ c: 6 c -T CD t^ t 00 o G^ i^ fM iH CO CO >c C o 'M o-l iC G^l 43 X CJ t-^ CO X iC c ^ OC' ■* C' rH 9 r~i t^ -f 9 'T 9 t^ X J} -f 6 6 6 ■^ c O i o T- c: ■^ C^l 0-1 w CO i> I-l o 30 o o ijl X C: CJ 0-1 o t^ CO c t^ lO o CO CD -r 9 9 -r ■T ^ as i^ o 6 ■^■ 6 c ^ o -r t^ ^ ^ X "— . -t cc Ol o X o 0-1 c-i o (M t-- ^ a-. i^ 3 s o rH 'O c5 o o 01 t- t'- -1^ rt 3 cb C iH X o 3 9 5 o Ol -t "^ »C ^ s i-t Cl s Ol "o X' CO -f I-( CO o rH CJ C-1 CO c lO X 9 I-t I-- o X Ol X rt 6 I-* 6 ~ (>i ^ ^ o c; ^ cb c •^ 0-1 CO 0-1 s C-1 o o 4^ GJ o ■CJ o o -f X iC CO ?1 -a 'M Ol c 0-1 CO OJ -T cfl iC t^ +3 -4-3 i^I ^-^ o ^ o 3 7-{ cb o t- X ^ 1-t c: o o o i-H i o >o o o c^ CO O !?1 o o o rH c ^" CO o X ■^1 0-1 o 9 c 9 c: T^ 1 0? 0-1 c 9 ^ C-l I^ ■- — 0-1 1— I 1— 1 5 >— . ^ o OS -t fM I-- t>. X' o o t'- C' ^ -r X |~ ■M o fM a X Ol 1 1 9 -r G-l vs >c t^ C. .~ -f 3 o i 5 •b 1 1 -^ A-< 4h tj •M c^ O-l o < 5 1 1 05 a o -e 3 6 1 5 1 1 C^ (5 •i - O s s O ■T. O o r^ C14 CC CO 34 KEPOKT OF CHEMICAL LAHOKATOHY The Mechanical Analysis of Arid Soils Various methods of mechanical analysis Importance ascertaining the total " clay " The mechanical analysis of soils, that is to say the division of their particles into groups according to size, is perhaps of greater importance than any other single determination which the soil analyst is called upon to make. These analyses would undoubtedly be made more frequently than is done at present were the methods suggested more satisfactory and, especially, less laborious. That recommended by Hall,' which appears to be the only one practised in Great Britain, is so tedious and often requires so long a time for its completion as to very seriously impair its usefulness. Its application to arid soils offers a still more serious objection in that it entails a treatment with acid, which is not permissible with arid soils, the larger particles of which may be, and usually are, concretions of calcium carbonate. Hilgard's method, depending upon the use of successive currents of water of velocities adjusted to carry particles of the required size, offers several objections, perhaps the least of which is that, requiring special and expensive apparatus, it is, to quote from Hall (loc. cit.), suited only to laboratories devoted entirely to soil analysis. Osborne's method, which, like that of Hall, is carried out by a series of sedimentations in beakers, proves in our hands, working with arid soils, to be capable of yielding results of the most varying character. It leaves very much to be desired on the score of ease and rapidity, especially when a number of samples are to be treated at the same time. Indeed, the multiplication of vessels alone renders it quite inapplicable in such cases. The method advocated by the U.S. Bureau of Soils appeared to us to be the most promising. It consists in breaking up the soil agglomerations — puddling the clay — by agitation in a mechanical shaker with water and a few drops of ammonia. The separation of the clay from the other groups of particles is then made by means of a centrifuge. Given the appliances suited to the purpose the above method would seem to leave little to be desired on the score either of rapidity or accuracj*. Our experience with it, as applied to arid soils, has been most disappointing. It, as do most of the other methods, fails to separate a very considerable proportion of the clay, which goes therefore to swell of the silt and other fractions.' That the entire amount of clay should be taken into consideration has been abundantly shown by the results of observation on partially water- logged soils, inefficiently drained, and lacking oxygen. Under these conditions sodium carbonate is formed in notable proportion, one of the results being the deflocculation of the clay with its attendant disastrous effect on production. In the analytical work carried out recently it has been noted that even when so little as two per cent, of clay remains associated with the residue of silt and sands, the binding effect on the soil is very marked. Clay aggregates in arid soils are known to be held together much more firmly than in those of humid regions. There is evidence that this condition is associated with the intense baking which these soils receive during the season of drought. (See below.) In the tests which were made by the U.S. Bureau of Soils^ on the effect of length of time of agitation in the mechanical shaker on the proportion of clay found, no marked additional change appeared to take place when the time was increased beyond six hours, except in the case of samples shaken as long as 77 hours. After such prolonged action 1 The Soil, p. 51 - Bure.nu of Soils BiilUHn, No. 24 THE .MECHANICAL ANALYSIS OF AUll) SOILS 35 thcro was usually an inert Fig. 3, — ilechanical Shaker for Soil Tubes 'shaker' iniount of clay and this was taken to mean that thi.' soil was being unduly broken down. Six hours was therefoi-e taken as a sufficient length of time to subject the soil to the action of the shaker. The form of electrically-driven mechanical Mechanical shaker adopted by the LI.H. Bui-eau of Soils is shown in Fig. 3. At the time soil work was begun in these laboratories a machine of this type was not at hand, and the ordinary form of end-over-end rotary shaker was used for the purpose. The following results were obtained on a sample of "cotton soil" known to contain about 54; jier cent, of clay. Five grammes of the soil were, in each case, treated with 75 c.c. of water to which ten drojjs of strong ammonia were added. For the sake of simplicity the proportion of clay only is given, though in many cases complete analyses were made. Length of shaking in rotary shaker Clay 6 hours 7 hours 14 hours 24 hours ... ... 48 hours 24-8 per cent. 26-1 ,, ,, 23-5 „ „ 29-6 ,, ,, 35-1 „ „ Later, a machine of the horizontal type was secured, and a short series of experiments was carried out with it. These were not carried very far since it was soon evident that vyhile the agitation was more violent, the results were not sufficiently improved to render the method practicable. Thus : — Length of shaking Clay found 30 minutes ... ... ... ... 1 hour ... 2 hours... 6 houra 25-3 per cent. 21-4 „ „ 26-5 „ „ . 33-4 „ „ The actual amount of clay present was, as in the former experiments, 54 per cent. The machine was run at the speed prescribed by the U.S. Bureau of Soils, namely, 100 impulses per minute. By shaking more violently the amount of clay separated in six hours was increased to 38' 2, a figure still far below the truth. As will be seen, even after 48 hours' shaking in the rotary machine, tlie proportion of clay separated was only about two-thirds the total amount. 36 KEPORT OF CHEMICAL LAHOKATORY No attempt was made to carry the agitation in the horizontal shaker beyond six hours, since it was considered that this length of shaking is as long as can be conveniently carried out, a longer one entailing either two or more periods of shaking, or its continuance during the absence of the analyst, which always introduces an undesirable uncertainty into the determination. Soils of arid regions usually contain notable proportions of calcium and magnesium compounds, especially calcium carbonate and sulphate. It had already been noted by Briggs, Martin and Pearce' that ammonia, when applied to the treatment of such soils, appears rather to flocculate the clay than to break up the flocculations, and its addition in such cases is therefore not desirable. They further note that certain soils (containing an excess of magnesium and calcium carbonate) and which presumably were shaken with water only, were badly flocculated at the beginning of the mechanical analysis ; but that after two or three decantations had been made, the tendency to flocculate disappeared and the final separation was made without difficulty. These soils were found to contain as much as 25 and 42 per cent, respectively, of carbonate, calculated as calcium carbonate. Although Sudan soils rarely, if ever, contain as much as ten per cent, of carbonates, calcium sulphate is frequently present, more especially in the subsoils, and the two Dertocculation together cause these -soils to flocculate persistently in spite of the absence of ammonia. We have found, however, that trouble of this kind usually disappears on the addition of sodium carbonate. As we have abandoned the needlessly laborious and no more accurate method of determining the clay by direct weighing, the presence of sodium carbonate introduces no difficulty. Deflocculation is so effectually aided by the sodium carbonate that only moderate agitation in the mechanical shaker is i-equired, and either of the two forms of machine mentioned above may be applied to the purpose, the one figured on page 35 being, however, preferable by reason of its far greater convenience in use. The use of sodium carbonate in the analysis of these soils is, it appears to us, all the more indicated in view of the fact that it may be formed in the soil as the result of inefficient drainage. The indications furnished by its use in the laboratory are of direct practical bearing aside from the fact that it appears to be the only agent which will enable us to obtain an idea of the true mechanical constitution of the soil, without the use of acids which, for the reason already given, are wholly inadmissible. A number of experiments was made in order to determine the effect of varying amounts of sodium carbonate. From fifty to one hundred milligrammes to 100 c.c. of water appears to be a suitable proportion. If the amount of sodium carbonate be materially increased, the reverse effect of floceulation results. The yield of clay with the higher proportion of carbonate was nevertheless, in the few experiments made, slightly lower ; which is satisfactory in that it goes to show that there is no danger of solvent action on the fine siliceous silt in the soil. The results were as follows : — the presence of sodium carbonate Form of shaker Time of shaking • Sodium carbonate Clay found Rotary 6 hours 1 gramme 48*6 per cent. ,, .> n-1 ,, 48-9 ,, ,, Horizontal M 1 47-7 ., J) „ „ 0-1 „ 49-1 .. U.S. Bureau of Soils, Bulletin 24, p. 24 THE MECHANICAL ANALYSIS OF AKID SOILS 37 A slightly more effective disintegration of the soil aggregates is had by simple heating with the dilute sodium carbonate solution. The soil tested above gave the results as follows : — Time of heating Sodium carbonate Clay found 15 minutes in steam bath •'2 in 100 c.c. water 50-9 per cent. 30 " '1 >> >) M J) t) 51-4 „ 5 " boiling )» n )» n 51-7 „ 15 .. .. >. >> 51-8 ,, 35 " >» •1 „ ,. ,, 51-8 „ The above results were obtained by the use of the centrifugal machine. If the method of sedimentation in beakers is employed, the boiling method offers the great disadvantage that the number of decantations is very much increased. It is possible that while the clay particles are disintegrated by the boiling, they gradually combine again to a certain extent after the liquid cools. We have found that , by far the most effective agent in bringing about the deflocculation of the clay is an ordinary camel-hair brush. It is employed as follows : The weighed portion of soil is placed in a beaker (or enamelled iron cup) and water cautiously added, sufficient to form a thick paste. The mixture is then stirred with the brush. The clay is rapidly puddled and the mixture becomes so tenacious that a further addition of water is required. Tliis is made very carefully, a few drops at a time, and the puddling continued for from five to ten minutes. There appears to be no objection to carrying out the puddling in a beaker. One of these, used for several successive operations, was found not to have lost even a milligramme. For arid soils, which contain considerable calcium carbonate (and, not infrequently, calcium sulphate) it is essential that all water used should contain sodium carbonate in about the proportion already indicated. The further treatment of the soil, after the above operation, depends upon the method to be employed. If one of sedimentation, the mixture is made up to a height of say ten centimetres and allowed to stand for a length of time depending upon the size of soil particles which is taken as limiting the clay {vide infra). We have adopted what is practically the 8-hour subsidence, which includes, in the clay division, particles up to about 0002 millimetre. After, usually, the third decantation, the residue, from which most of the water should be removed on the water-bath, is again puddled for about a minute, after which no further puddling is required, the residue being simply mixed with successive portions of water (containing sodium carbonate) and allowed to stand for the specified time until the liquid becomes practically clear at the end of that time. As already stated, the brush method of puddling has been found to be more effective than any that we have tried. No matter what be the method of analysis adopted, its use at one or more stages of the process is strongly recommended. Its eflSciency as compared to that of the rubber pestle recommended by Hall and others is most striking. Using the rubber pestle and employing the Hall method (of acid treatment followed by ammonia) we have frequently had to decant as often as 20 to 30 times before the water came away clear. The analysis carried out in this fashion may take more than a month for its completion. By the aid of the brush and the use of sodium carbonate, the wiiole or practically the whole of the clay may usually be removed in five decantations. Use of a camel-hair brush in puddling the clay Great acceleration of analysis by the brush and sodium carbonate method 38 HEPOUT 01'' (JHEMlCAlj LAliOKATOKY The following exporiments were made in order to determine the rate at which the clay is removed : — Clay, per cent, of total present Soil No. 1 Soil No. 2 1st decantatiou 2nd „ 3rd „ 4th ,, 5th ,, 60-5 23-5 9-9 4-1 2-0 7G-0 1'2-G 5-4 3-6 2-4 Total 100-0 100-0 Technique fur using the brush These results were compared with those of the method by boiling with dilute sodium carbonate solution. It was found that when boiling was adopted, not only was more than double the number of decantations required, but the sum total of the clay removed was 51-5 per cent, as against 54 per cent, by the use of the brush. Examination of the soil particles under the microscope shows the particles of sand and silt to be quite clean when the brush is used, which is not the case with the method by boiling. As the camel-hair brush picks up and retains a considerable quantity of the soil, care must be taken to remove this by agitation in successive portions of distilled water in a small beaker, the whole of the separated sand being washed out of the beaker after each washing. Usually four or five washings are sufficient. Tapping the bottom of the beaker with the brush submerged in the water, has been found the most efficient method of dislodging the sand. In order to avoid the frequent washing of the brush, which, while it need only be employed twice in each determination, is nevertheless advantageously applied to bring the clay into suspension after each decantation, it is best to use a separate brush for each soil operated on, and to wash out only once at the end of the process. The brush is conveniently kept in the small beaker in which the operation of washing is carried out at the end. Any form of camel-hair brush may be applied to the above purpose. We have found the flat form about half an inch broad, bound to a wooden handle by a strip of tin, to be very convenient. Paraffin wax should be applied to the tin junction in order to fill the interstices and prevent the entrance of soil. The round form of brush may also be used, a glass rod being inserted into the quill, (which is cut rather short) and fastened by wrapping with thin copper wire. The junction should be protected by wax as noted above. Cleaning of the brush after each puddling is more readily accomplished if the hairs be cut moderately short. Where a great number of soils are to be operated upon, the labour of the preliminary puddling may be avoided by substituting agitation with water and sodium carbonate in the mechanical shaker, as described above. After the fourth decantation the residue (from which some of the water should be removed on the water-bath) need be puddled by the aid of the brush for about one minute onlv. The results obtained in this way have been, THE MECHANICAI, ANALYSIS Ol' Alill) SOILS 39 in our liands, identical witli those in whicli tlie tive or ten minutes preliminary puddling was carried out, but the clay comes away more slowly. The soil need not usually remain in the shaker more tlian about an hour, as will be seen from the results of the following experiments with a soil containing about 54 per cent, of clay. Time of agitation in the mechanical shaker (Water 10 c.c. Na. CO, O'l gramme. Brush not used) Clay found 1 hour 3 hours 5 „ G „ 47'4 per cent. 48-8 ,, ,, ■io-i ,, „ With the majority of soils little appears to be gained by continuing the operation beyond an hour, since the small amount of clay which remains unaffected at the end of that period is easily brought into suspension by the short puddling mentioned. Centrifugal Method While the method of separating the clay, as outlined above, is rapid and generally satisfactory, the operation may be performed still more quickly by means of the centrifugal machine, .^gainst this must be stated the fact that the centrifugal method entails much more of the operator's attention, though the use of an alarm clock which may 40 KEPOKT OF CHEMICAL LAHOKATOKY Centrifugal method Size of the clay particles be set accurately to the minute will enable him to devote a little time to other work during the whirling. We have found an electric alarm the most reliable for the purpose. One of the forms of centrifuges used in America is shown in Fig. 4. The machines are well known and need no special description. It is desired simply to call attention to the method of controlling the results. It is usually stated that this must be carried out by the use of the microscope. While this is, in a measure, true, it would be useless to expect to obtain results exactly comparable with those from sedimentation, in such a manner. The slightest variation in the size of the particles from the standard which has been adopted would, in many soils, make a marked difference in the results. It must be recollected that the particles are never spherical, so that exact measurement by means of a micrometer is practically impossible, and no two observers could be expected to get precisely the same result. The plan we have adopted is to treat several soils of different character by the sedimentation method and then to determine the length of time of whirling in the centrifuge, run at a known speed, to obtain the same result. If the soil has been sufficiently puddled at the start we have found that the number of whirlings required is much lessened, almost the entire amount of clay being removed after the sixth decantation. In the machine at hand, 6 minutes has been fixed upon as the period of whirling, the' number of revolutions being about 840 per minute. The machine figured runs at a much higher speed and the length of whirling would be materially reduced. It has been found that after the 6th or 7th whirling, when nearly all the clay has been removed, the remaining soil packs so loosely at the bottom of the tube that the swirling of the liquid, due to the arresting of the machine, causes some of the fine silt to be drawn up into it. As the result of this the water never becomes quite clear, and the end of the clay removal is difficult to judge. This may be avoided by increasing the time of whirling, after the 6th or 7th, to 10 minutes, so that the soil residue packs more closely. The error introduced by the longer whirling at this stage of the process is negligible. The difficulty mentioned would probably not be experienced with machines adapted to higher speed. The size of the daij particles. Different workers have adopted various limits for the size of the particles taken to represent the group designated " clay." The most common upper limits are '002 mm. and '005 mm. Hall' employs the first mentioned and separates the clay by 24 hours' subsidence from a column of water 8-| centimetres in height. Kilroe, Seymour and Hallisy- employ a method which is attributed to Hall and in which the height taken is 15 centimetres, the particles being stated to have a maximum diameter of -005 mm. Hilgard^ describes a method of subsidence of 24 hours' duration, the colunni of water being 20 centimetres in height "whereby all grain sizes of and above O'Ol mm. diameter are removed from the turbid liquid." Atterberg-* recommends either the limit ■003 mm. obtained by a 4 hours' subsidence from a column of 10 centimetres, or of '002 mm. from a column of the same height but with the time of subsidence increased to 8 hours. It is obvious that the above statements as to the size of the particles are highly discordant. We have made many measurements, all the results of which point to the fact that the figures given by Hilgard and by Kilroe, Seymour and Hallisy and many others, are not even approximately correct. Our measurements agree more nearly with those ' The Soil, 2nd Ed., p. 51. -' The Geological Features and Soils of the Aiiriealliu-nl Slatioii of the Dq)t. of A'jricullare at r.allnhaiiu-, 1910. ■■■ Soih, p. 89. • ijher die KlassiJikatioK der Budenkorner, A. Atterberg, Kaliiuir. THK MECHANICAIy ANALYSIS OF AKll) SOILS 41 of Hall and of Atterberg, but it is uvident that to obtain a separation up to about '002 nun. it is not necessary to carry the time of subsidence as far as prescribed by Hall. Many of influence of our results were obtained by the use of a column of 20 centimetres heieht, but we have '"^'S^t of •' ^ ' column abandoned this method for the reason that the shortening of the number of decantations was not what was expected. It would appear that the longer the column the more the particles which subside carry others with them. A better result is had by employing two vessels with columns one half the height (10 cm.) the liquid being divided equally beween them ; but such a multiplication of vessels does not' appear necessary if the soil be properly puddled as described. An examination of the particles deposited from the 20 centimetre colunms proved them to be but very slightly larger than those of the 10 or 8^ centimetre columns, provided always that the time of subsidence be carried to 24 hours. The following figures show how small is the effect of the height of the liquid on the results obtained from the mechanical analysis of Sudan soils. The duration of subsidence was, as stated, 2'4 hours. HEIGHT OF OOLUMN OF WATER 85 cm. 10 cm. 20 cm. Azaza soil Bardohe soil ^lixture of 10 typical bardohe soils 3G-1 per cent. 52-6 „ „ not determined o3-l per cent, not determined 38-0 per cent. 54-2 „ „ 52-4 ,, Determination of the "Silt" When a number of soils have to be examined at the same time, we have found the use of the apparatus shown in Fig. 5 to effect a great saving of time and trouble. The residues from the clay determinations (made by difference) are washed into the cylindrical beakers (or museum jars) with distilled water containing the usual small amount of sodium carbonate.^ After standing for a length of time de- pendent upon the size of the particles adopted as representing the "silt" division, the liquid is poured away and rejected. This preliminary treat- ment should be made with distilled water, since the soil still contains particles in size bordering on those of the clay division. The subsequent operations may, as a rule, be carried out with clear tap water. ' Since the heating of the soil residue in the oven usually causes It to cake together, it is advisable to add to it a small amount of water containing sodium carbonate and bring almost or quite to boiling. It is then treated as described above. Fig. 5. — Apparatus tor the determiaation of ttUt •Silf 42 ItEPOKT OF CHEMICAL LAliOKATOHY 'I'echnique of determination of ■■Silt" Classification by size The beakers are placed in position and, the vent-cock having been closed, the water tap is turned on very cautiously until the water is seen to appear in the glass tubes. If the level rises equally in all, thus indicating that no air is included, the flow- is allowed to continue until the water reaches the bend in the glass tube. At this moment the full force of the water is quickly turned on. If the operation is not carried out as indicated some of the siphons may begin to act too quickly, with the result that their pull on the water may be so great as to prevent entirely the filling of the others. "When the beakers are about half filled, the pressure is greatly reduced so that the level of the water rises very slowly. By operating in this fashion even very considerable differences in the diameters of the tubes and beakers will cause no trouble, since the back pressure in those in which the flow is more rapid acts as a check on the incoming water. If the filling is performed with care there will be an almost absolute equality in the level throughout. In order to secure the above equality of level the supply tubes should not be too small. Those in the apparatus in use have a bore of 0-5 cm. If the distributing vessel (A) is made as shown, with the top inclined at a slight angle, and the tubes are cut off flush with- its under-surface, the inclusion of air with the water in the tubes very rarely takes place. The same result may be attained by substituting for the distributing vessel A, a metallic tube tilted at a slight angle. During the time the water level is slowly rising in the beakers there is ample opportunity to mix the soil well with the water by stirring with the glass inlet tubes which, as shown in the cut, are connected with the supply tubes by means of flexible rubber tubing. At the end of the period fixed upon for subsidence of the larger particles, the vent- cock is opened and the supernatant water siphoned off. When this operation has been completed, the water is again turned on (the vent-cock being still open) and a small amount of water is run through in order to wash out the last of the turbid water. The vent-cock is then closed and the entire operation repeated, as described, as often as may be necessary. The Classification of Soil Paeticles The want of uniformity in the system of classification of soil particles no less than the diversity in the methods of separation, is greatly to be deplored. Much of the value of the results is lost by reason of the impossibility of comparison with those of other observers. In a paper presented to the International .\gro-geological Congress held at Stockholm in August, 1910, Dr. Atterberg of Kalmar suggested the following limits, with a view to their general adoption : — Cuarse sand (unretentive of moisture) ... Fine sand (water-retaining) Silt Clay ... The times of subsidence (in a 10 centimetre of the two series of silts and clays were ; — 3 to '3 01 2 to '2 ... -3 to -03 (-)• •2 to -02 .. -03 to -003 oi ■02 to -002 ... -003 to — 01 •002 to - column) prescribed for the separation Silts Clays •03 mm. •02 mm. 3J minutes 7') minutes •0U3 mm. •002 mm. 4 hours 8 hours THK MECHANU'Al, ANALYSIS OF AKIl) HOIl.S 43 In the paper referred to, Dr. Atterberg rather favoured the first series of limits from the point of view of the shortening of the times of subsidence. In a more recent connnunieation, however, he states that against this should be counted the fact that the limit of -002 mm. for clay has already been rather widely adopted and, further, that the limit of "2 mm. for the water-retaining sand is more nearly correct than is that of '3 mm. Our own experience with Sudan soils leads us very greatly to prefer the limits 02 and Limits of 02 •002 mm., the latter especially, as it differentiates more sharply the soils with which we ^^^ , "'"'' ^ " ^ ^ preferred have to deal. We have therefore adopted these limits, provisionally, with the hope that a general agreement on these lines may be effected. As already noted the figures given are only approximate, the clay being more accurately defined by the time required for its subsidence. Resume of the Method used fok the Mechanical Analysis The following is a short description of the method of mechanical analysis as carried out in these laboratories. The separation is made, as has been stated, into particles as follows : — (a) Stones and tjrarel above 2 mm. (6) Fine soil, less than 2 nun. The latter is further subdivided into : — Coarse sand (non-retentive of water) ... ... 2 mm. to 0'2 mm. Fine sand (water-retaining) ... . . . . -2 mm. to '02 mm. Silt -02 niiu. to -002 mm. Glai/ ... ... ... ... ... ... -002 nmi. and below The air-dry soil is well mixed, the larger masses being crushed by means of a wooden rolling-pin. One hundred grammes are weighed out on a rough balance and passed through a sieve with round holes, 2 mm. in diameter. The portion, if any, which remains on the sieve is washed with water to separate the adhering sand and soil, dried on the water-bath and weighed. This constitutes the " stones and gravel." The jjortion passing the 2 mm. sieve is passed through a sieve with round holes 1 mm. diameter. (The material which passes is taken for the chemical examination and for the further mechanical analysis.) In the above operation the larger masses of soil are treated with the rolling-pin, and also, if not too hard, by rubbing with a cork bung. When most of the soil has been passed through the sieve the remainder is washed on the sieve by a current of water aided by a camel-hair brush, the small quantity of soil washed away being disregarded. When a number of soils are being examined at the same time, the residues on the sieves are not treated with water at once, but are put aside, in marked beakers, the washings being performed later. In this way the sieves are kept dry during the operations of sifting the soils. The moisture having been removed by several hours' drying in the air-oven at Resume of 110 C, a quantity of the fine dry soil is taken, less than 5 grammes in proportion to the technique amount of particles between 2 mm. and 1 mm. known to be present in the original soil. This plan of excluding the larger particles is adopted in order more safely to secure an average sample when the small quantity (less than 5 grammes) is weighed out. The weighed portion of fine soil is either puddled by the aid of the brush, as described on page 37 or, if a number of soils are to be treated at the same time, it is placed in a shaker bottle, 100 c.c. of water containing -2 gramme of sodium cai'bonate added, and shaken in the machine, for 2 hours, as described on page 35. The soil and water are then 44 BEPOET OF CHEMICAL LAHOKATOKY Entire amount of clay removed in three days " Correction' for soluble salts transferred to a cylindrical beaker marked at 10 centimetres, distilled water added up to the mark, mixed and allowed to stand for not less than eight hours. In the case of heavy clay soils so much clay is brought into suspension by this method that the first sedimentation is allowed to continue over night in order to ensure that the silt may not be held up by the thick liquid. The subsequent subsidence is of eight hours' duration followed by one over night. In this manner the entire amount of clay may usually be removed in three days. The water used for each operation should contain about one half gramme of sodium carbonate per litre. It sometimes happens that clay is present is such large proportion that the usual volume of water is not sufficient to hold it in suspension for the necessary 8 hours. In the case of flocculation due to this cause, it is necessary either to employ a more capacious beaker, or to divide the turbid liquid between two or more beakers of the usual size (8 centimetres diameter). Flocculation, if it takes place, is more usually due to the presence of soluble salts. This is especially the case with subsoils. It may, as a rule, be avoided by using a smaller proportion of sodium carbonate in order to reduce the total amount of salt present. Two tenths of a gramme per litre may be used for the first two decantations, after which it is advisable to increase to the usual proportion. If flocculation has occurred, the excess of water, after decantation, should be removed on the water-bath and the soil puddled with the brush. Once the floccules are formed they do not readily break up again in spite of the fact that the salt may have been almost entirely removed by decanting the more or less clear supernatant liquid. Flocculation due to excess of salts occurs less frequently if the centrifugal method be employed, since it usually takes place very slowly in the presence of the sodium carbonate. After, usually, the third or fourth sedimentation and decantation, the residue is puddled with the camel-hair brush as described on page 37. When almost the entire amount of clay has been removed, the soil is washed into a weighed beaker of about 200 c.c. capacity, using distilled water (in this case without sodium carbonate). After standing for the usual period (i.e. not less than eight hours) or until clear, the water is poured away and the beaker and residue dried by heating in the air-oven for 2 hours at 110 C. From the weight of the soil residue may be calculated the percentage of clay, which is determined by difference. The figure so obtained should be corrected for soluble salts if present in notable quantity. A "correction" for organic matter in the mechanical analysis of a soil is, in the opinion of the writer, neither necessary nor desirable. In any case the amount present in soils even of semi-arid districts, rarely amounts to one per cent. The determination of humus should, of course, be made, but separately, as a part of the chemical analysis. If desired, however, practically the whole of the humus may be removed by using a larger proportion of sodium carbonate in the shaker bottle ('5 to 1 gramme) and, its proportion being determined on another sample by the usual method, the necessary correction may be made. When it is desired to carry out the separation of the clay more rapidly, the centrifugal machine is used, as already described. The residue in the beaker, consisting of sands and silt, is treated as detailed on -page 41. The silt having been removed, the residue is washed back into the original weighed beaker, dried as Viefore and weighed. The silt, like the clay, is determined by difference. THE MECHANICAL ANALYSIS OF ABID SOILS 45 The remaining sand in the beaker is separated into two portions, by means of an appropriate wire sieve.' To the portion remaining on the sieve is added that which was retained by the 1 mm. sieve in order to obtain the entire proportion of coarse sand, 2 mm. to '2 mm. in diameter. The fine sand is determined by difference. All figures should, oViviously, be expressed on the dry soil. In concluding these notes it is desired to call attention to the fact that while the above method is especially recommended for arid soils, it is equally suitable for those of humid regions. Further, it is held that a method of this kind should invariably be used since otherwise the comparison of humid with arid soils becomes impossible. It must be recollected also that the treatment by acid not only results, at times, in the complete solution of the larger particles, but invariably, in the extraction of material from the finer constituents, so that as a method of "mechanical" analysis it cannot but be considered scientifically unsound. Soils of the Gezira A Preliminary Note The " Gezira " or "island" is that roughly triangular section of the Sudan which lies between the White Nile and Blue Nile, with Khartoum at its apex, at the junction of these two rivers. It has long been held that the main hope of agricultural development of the Sudau lies in the irrigation of this district. The writer takes a more sanguine view, but, in any case, it appears evident from a study of the rainfall map on pages 46 and 47 that the portion south of, say, the 15th parallel, may be capable of profitable development by rainfall cultivation, and there is no little evidence to indicate that even so valuable a product as Egyptian cotton may be successfully grown, as a rain crop, if modern methods of cultivation be api^lied. Very good crops of dura are now produced, even with the primitive methods of the natives, and it appears likely that wheat may displace this largely when the native cultivator has learned to realise the advantage to be gained. The newly established Gezira railway is already having a marked effect on the output of rain-grown grain from the Gezira, and while this is at present largely dura there is also a certain amount of other grain and of cotton, as well. The latter is grown from native seed and therefore of poor quality, but experiments have been made with Egyptian seed, with most promising results. Since many appear to be slow to believe in the possibilities of the Gezira from the point of view of rainfall cultivation, a comparison with similar semi-arid districts in other countries may not be out of place. In these districts of low rainfall a method of so-called " dry soil farming" is practised, the distinguishing features of which are (a) deep ploughing to ensure the entrance of the rain into the soil, and (fe) frequent harrowing in order to preserve a loose mulch on the sui'face and thereby check excessive evaporation. It is obvious that season of rainfall in such dry climates is more important than its actual amount. In the United States (Utah and California), 590 kilogrammes of wheat per acre are stated to have been obtained with a total annual rainfall of about 10 inches. The average annual rainfall over the whole of these wheat-producing districts is only about 14| inches. ' The sieve found to pass particles most nearly approximating 0-2 mm. in diameter was of 80 meshes to the linear inch. This will, of course, depend in part upon the size of wire used. * o - o z: < Hi oi =o s. 5 t^ T^ lO ^ ^ oi =0 "^ r^ =© Co cri t-^ ^ iri di oi II II II II II II O O O O C) o o o o o o o r^ cvj fn ^ lo 5: < ^ ■^ 48 REPORT OF CHEMICAL LABORATOKY In the Gezira, at Wad Medani, the rainfall is ofBcially stated to be 400 mm. (about 16 inches). This is slightly greater than the average in the dry farming districts in the United States. Even half-way between Wad Medani and Kamlin, as reference to the map on pacjes 46 and 47 will show, the rainfall is greater than that prevailing in Utah and California where wheat is successfully grown. That the season of rainfall, more especially in the latitude of Wad Medani and south of it, is suitable for cotton, is proved by the fact that it has been extensively grown in the past, in spite of the unscientific methods practised by the natives. Improvement of seed and more modern methods of cultivation would appear to be all that are required to make this enterprise a financial success. Up to the present but little work has been done on the soils of the Sudan, and especially on those of the Gezira. In 1903 a few samples collected by Sir W. Willcocks' were examined by Mr. Hughes, of the Egyptian Department of Agriculture, but no mechanical analyses were made, and, as Mr. Hughes was not supplied with samples of the subsoil, the results, e.g. as regards soluble salts, are not only incomplete, but even misleading, further, the two samples of which more complete chemical examinations were made and which are described by Sir W. Willcocks as "typical Gezira soil," one collected on the Blue Nile and the other on the White Nile side, were taken only 10 miles and 2 miles, respectively, from Khartoum. A comparison with the results in the tables on page 54 et seq. will show- that these soils are not correctly designated as typical of the Gezira generally, since the soils of that district vary greatly, both their mechanical and chemical characters being dependent upon the location — that is to say, upon the climatic conditions, rainfall especially. At Khartoum, the rainfall is a little over 10 centimetres. At the southern end of the Gezira, it is approximately ten times as great ; and while Sudan soils, like those of Egypt, are chiefly river-borne silt of Abyssinian origin, the deposit has suffered very material change in those places in which the rainfall has been more abundant. There is also evidence of change as a result of the alkaline water of the W'hite Nile, some of the soils on this side of the Gezira having their clay in a more puddled condition, and therefore less permeable. The natives of the Gezira distinguish two main classes of soil. The better of these, from the point of view of suitability for cultivation under ordinary conditions {i.e. rainfall cultivation), is termed by them bardobe. This is the typical so-called cotton soil, characterised by the deep cracks which form when the soil dries. The other chief variety cracks but little, if at all, on drying. It is called azaza and is said to be uncultivable, or, at best, to return inferior yields. Another variety of soil, and the one most esteemed by the native, is termed fud or fuda. This appears to be merely bardobe in a good state of tilth. On drying, it cracks to a less extent than does ordinary bardobe soil, and remains friable and permeable. In view of the proposed irrigation of a section of the Gezira a very great number of soils, with subsoils down to four feet, have been collected for us by Mr. S. C. Dunn, the Government Geologist, and examined in the laboratory. A sketch map prepared by Mr. Dunn, showing the points of collection, is found on page 51. It is evident from the results of the mechanical analyses that while azaza soils are, generally speaking, lighter than bardobe the difference in composition is not nearly so great as estimated by the native. The inferior returns under present conditions are due in great part to scanty rainfall. The porous nature of the soil is also a result of this same condition, so that the effect of low rainfall on crop production is doubly felt. Bardobe soils, I The Nile in 1904, p. 100 32'= 32 3o- S3° 33 3o 34 (1 ( i Map 1 of the (JEZIRA )' '^ CA I'll s LE I.GOO.OOO / //jUKHAHTOU M '5° 30 / '^^==V- ' \\ 1 i )\ ." "^V^^^sS- A"^ / / / / M aharejb'a* ^sVvWad el Turabi 15° ^Mi 1 k^ ^^ ^ N ^ WJ /%! '4gX / "3/ / / / / / / / / \ / ^ VWbu Ushar Yr ^ 1 1 i«ihara/ / / / / 1 / / M \V / 1/ Mesellemiao Vi \\ 14 30' Jj [ ,' / • Um DSJ^iny) »A ( I \ \ I 1 1 1 1 \ \ \ / / \ 1 i 1 ' 1 1 ( ! 1 . ios i^adVMeda 3 V s \ 1 <■ * Managil -n L • ^. 14 13 30' • Maatuk \ \ \ \ \ \ / / / ]1 > \ El Du .W t 1 1 1 1 / / / / / / / / / / / \ \ if ^Tad el Hi ddad^y-5 \' fKawa \ 1 v\ _\ ^ o, t III 1 1(1 1 ' 1 ' lOI I I ' "I §11 S''* 1 1 ; 1 ' \ j|YSennar { w ' 1 1 ' 1 ' 1 // h' \ / ^ Kosti^ / ^Goz Abu Gu \ na 1-f \\ -^^ I 1 Fig. 7. — Map of the Gezira showing approximate line and scope of proposed Irrigation Canal The area coloured green represents that under consideration, with a view to irrigation Contour lines show height in metres above sea level Proposed Canal Railway r* *■* »**« 50 REPOET OF CHEMICAL LABORATORY from their geographical position, not only receive more moisture, Imt, hy reason of tlieii' higher proportion of clay resulting from this condition, are better able to retain the moistiri'e received. Azaza soils, here considered very light, would in a country of normal i-ainfall he held to be rather heavy than otherwise. Near Khartoum, where the rainfall is only a few inches, the prevailing type of soil is a light azaza. This statement does not include the soils directly bordering on the river, which may be subjected to flooding from time to time. These, as a rule, are distinctly heavier. South of Khartoum, as the rainfall increases, the decomposition of the silt and the formation of clay take place to an increasing extent, until, in the neighbourhood of Wad Medani, with an average rainfall of about 16 inches, the proportion of clay amounts to 50 per cent, and even, in some cases, to as much as 60 per cent. Soils in humid regions containing such a high proportion of clay would be considered almost impervious and extremely difficult to work ; but the fact that in arid countries even higher proportions may exist in good arable soils has already been pointed out. Thus Means' calls attention to a sample from Egypt in which the clay (below -005 nun.) amounted to about 75 per cent, and the silt ('005 to '05 mm.) above 12 per cent., making a total of 87 per cent., of heavy material. This "according to the mechanical analysis should be an almost impermeable clay; but in the field the soil was found to be easily drained, perfectly amenable to cultivation and favourable to plant growth." The same remarks apply to Gezira soils and to Sudan soils generally, provided theij are -properly drained. Good drainage is essential to all soils ; but the disastrous eii'ect of inefficient drainage on heavy soils has been especially evident in certain lands, improperly worked, near Khartoum, and in the irrigation basins in Dougola Province, in course of construction. Means {Joe. cit.) explains the ease with which these essentially heavy soils may be worked by the "cementing action of lime and magnesia and iron compounds which join together the fine grains of silt and clay and form larger aggregates, thus giving the soil a lighter appearance than a mechanical analysis would indicate. The fact," he states, " has been very clearly brought out in mechanical analyses of soils from American desert lands where calcium and magnesium carbonates were abundant. In New Mexico, certain soils were classed in the field as sandy loams, but upon subjecting them to mechanical analysis, when water acts upon the soils for several days, the cementing material was dissolved, the aggregates broken down, and the soil was found to contain enough clay to be classed as a loam or clay loam. The field examinations of Egyptian soils show this cementing process to be developed to a high degree, and soils in the field seem lighter than would be indicated by the mechanical analysis." The writer was formerly of the same opinion as to the cause of the above efi'ect, but recent experience has shaken his belief in this respect. It has been found, for instance, that the extent to which the clay particles are held together appears to be independent of the amount of earthy carbonates present. It has already been mentioned that the attempt to apply the United States Bureau of Soils method to Sudan soils had not met with success, disintegration of the clay aggregates not being effected by the prescribed agitation with water, and often still less so when ammonia was added. It was found, however, that subsoils often gave results more nearly approximating the truth, and yet the proportion of earthy carbonates in these was not appreciably less — in some cases, in fact, it was even greater. The following are instances of such a condition : — ' "The rcclamatiou of alk.ali lands in Egyjrt." liiillrtin of I'.S. Jiiirtan of Soils No. 21. Jirj,f. of AgricuUure. 3 3° 3 1° 15' 33 30' KabelGedado ^^^. "^ "''""''"'"""'^s;^ PROPOSKI) j,i8ara,at ^^ (JEZIRA IRRIOATION SCHEMH LATITUDE is' OS'/^ 181 177 ' " 173 \ ^eiGtifz 16B 163 161 ir,7 . Cf} Gemiab x2?>v Scale of Miles 5 4 3 2 10 5 10 15 ' 15 \ ^v 15° ^Debeibat Abdnlla f > 14^ AbJ el MajidiJ/ Line D. '\. LATWubVii'Vs'' *^ oDebeiba ,^^^__ // Abbaso Edeid Abu Ushn, '^Wj •■ + + + ♦ ♦ + "fj 141 137 133 129 12S 121 117 hi ^ ^ 45 Malm •. °lstarahna >\ f^- Line C. ^'^^J^buSeir Sh^e,kl, Karashi ^ '" Rufa'a 14° 45 LATITUDE 14° 45' loi 97^ 93 89 8 IssaheLa'iLJ/'^ \ At/m Sayalct + 77 vi \ oSolfab 'ft^ 73 ^^ oBoseilab VT' 1 "'Ho* \ 14°, 30 iModra Line B. \o LATITUDE 14''35' les ( \ \ Vi Nagaro VSI % + + *-^^%''(^ 1 57 53 049 45 + "Vl Mesellemia "*' ^A A2<^za^ Fadasi^- 37 Yl Line A. Wad 33"7 Hussein *%^ © Kalha ---■»■ + + H 13 9 5 1 \ / .^bu Haraz Medani ij^ LATITUDE 14''25' aJs 21 '>4Z^ 31° 33°15' 33°30' Fig. 8.— Map showing points at which Samples of Soil were collected for Analysis EEPOBT OF CHEMICAL LAHOHATORY Clay separated by 7 hours shaking with water Earthy carbonates expressed as CaCO. Sample No 1 (lat foot) 27'3 per cent. 2-36 per cent. 2(2nd,, ) 47-0 ,, „ 2-26 ,, ., 3 (3rd ,. 1 45-3 „ „ 2-57 ,, .. 4 (4th ., 1 38-9 ,, ,, 1-74 ,, „ 5 (1st ,, ) 25-5 ,, ,, 1'85 „ „ 6 (2nd ,, ) 43-9 „ „ 2-14 „ „ 7(3rd ,. 1 17-9 .. ,, 2-17 ,, ., 8 (4th ,, ) 26-7 „ „ 2-22 „ „ The actual amount of clay in the above soils was approximately : — 1st foot 2nd „ 3rd ,, 4th „ 58 per cent. 56 ,, ,, 61 ,, ,, 64 „ „ It is not held that the above is proof that cementation of the clay particles by calcium and magnesium carbonates does not occur. It doubtless does occur to a certain extent, but it is felt that some more potent cause is active in the majority of cases. In the regions of no rainfall, e.g. Dongola Province, the soils were at first thought to contain practically no clay whatever. Agitation with water, or with water and ammonia, resulted in the bringing of a quantity of material into suspension, but practically the whole of this settled out at the end of a few hours. It was at first assumed that this was due to soluble salts, but the absence of these in the majority of cases was soon proved. It was then concluded that the material present was fine silt, its preservation from decomposition being the result of the total absence of rainfall. Later, these soils were treated with water containing a small amount of sodium carbonate, and were also puddled by the aid of a camel-hair brush as described on page 37, with the result that a large amount of clay was separated. The results of the mechanical analyses carried out in this way at once explained the behaviour of the soil in the field.' It should be stated that the proportion of calcium carbonate in these soils was usually low — in one case only 0'36 per cent. — so that the explanation of the behaviour of the soil to water by its cementing action does not appear to be tenable. In the writer's opinion, the intense baking which these heavy soils receive during the hot season plays a very important part, in that it coagulates the clay and renders the soil more permeable and productive. This is in addition to the well-known fact that the ' These soils puddle very rapidly when not sufficiently drained SOILS OK THK GH/AHX 53 heating of soil evuu to a luodurate degree results in a greater solubility in water ol tlie constituents essential to plant life. The fact that a treatment so far from drastic as the simple puddling of the soil by means of a soft camel-hair brush, in the presence of an extremely dilute cold solution of sodium carbonate (not sufficient to dissolve the humus), suffices at once to disintegrate the clay aggregates and bring the former into suspension, points rather to the conclusion that the clay has simply been flocculated by the heat and dryness, and that, in the soil, it is usually not held together entirely, or even to any very great extent, by cementing material such as calcium carbonate. It is for this reason that the treatment by puddling in the preparation of the sample for mechanical analysis is held to be a sound one and the substitution of the very small amount of sodium carbonate for the ammonia usually employed, entirely unobjectionable. There is further evidence that it is not necessary to assume cementation of the clay in the soil aggregates. It has already been noted (vide page 44) that when subsoils containing rather high proportions of salts are treated for the mechanical analysis, flocoulation of the clay occurs ; and that when the excess of salt is removed, these floccules do not readily break up on the addition of water, even in the presence of sodium carbonate. The floccules may appear to do so at first, but if the liquid be allowed to stand for some hours, precipitation of the clay usually recurs. It is obvious that the intense baking and drying to which arid soils are subjected may cause a similar flocculation of the clay, and the aggregates so formed may be expected to be much more resistant to disintegration than those obtained by precipitation from suspension in water. Chemical coiitposition of Gezira soils. Extended chemical examinations have as yet only been made of the tract in the Gezira which it has been proposed to irrigate, and which is indicated in the sketch map on page 51. The results of examination of individual samples are detailed in the tables on pages 56 to 59. The variations in chemical character are more clearly shown in the table of averages on page 55, which should be studied in connection with the rainfall map. As would be expected, the proportion of water-soluble salts in the surface soil, and especially in the subsoils, is greater towards the northern end of the tract where rainfall and consequent leaching are least. The same is true of the proportion of calcium carbonate. Organic matter, as expressed by the content of humus and of nitrogen, is, on tlie other hand, greater towards the southern end where there has been a greater amount of cultivation and of plant residues left in the soil. Phosphates readily soluble in acid are also less where there has been least rain and wliere, consequently, rock decomposition has taken place to the least extent. Acid-soluble potash compounds are found to vary but slightly, but in the same direction. Speaking generally, the soils, lioth hardohe and a:aza, in tiiis tract may be said to be fairly well supplied with potasli and with phosphates. They are. liowever, like most Egyptian and Sudan soils, markedly deficient in organic matter and in nitrogen, and the results from their cultivation will be largely dependent upon treatment with respect to this deficiency. Rotation with a leguminous crop such as berseem and the ploughing of a fair proportion of the crop into the soil is wliat is especially indicated. Without sucli a rotation, rapidly diminishing crops may be predicted with confidence. Since such leguminous crops have not been grown on most Gezira soils, inoculation of the seed first planted should make a very marked difference in the yield. This experiment was tried on a small scaler lu^ar Khartoum, with results which exceeded all expectations. 54 KEl'OKT OF CHEMICAL LAHOliATOKY GEziiiA Soils. Aveeages. MECHANICAL ANALYSES First foot Second foot Third foot Fourth foot Liue (E) Lat. 1,')° 05' Stones, etc., above 2 mm. Fiiu soil hehjir 2 v/i/zi. Coarse sand 2 to '2 mm. ... Fine sand '2 to '02 mm. ... Silt -02 to -002 mm. Clay •11112 to mm 1-9 1-7 Liue (A) Stones, etc., above 2 mm 3-0 2-1 1-5 0-7 Lat. 14' 25' Fine still hrjon' 2 uini.. Coarse sand 2 to '2 mm. 14-(i 12-0 12-2 9-9 Fine sand •2 to '02 mm. ... 23-4 21-4 22-0 20-7 Silt ■02 to -Oiri mm 13-0 13-3 14-0 14-6 Clay •002 to mm 49-0 b-i-i 50-8 54-8 Line (B) Stones, etc., above 2 mm 3-0 1-6 3-1 1-0 Lat. 14 35' Fine soil bcloii! 2 niin. Coarse sand 2 to "2 mm. ... 13-5 14-0 12-5 15-2 Pine sand '2 to '02 mm 19-7 19-9 19-3 16-0 Silt -02 to -002 mm 14-0 14-2 13-9 13-8 Clay (M -002 too mm 52-8 51-9 53-8 54-4 Liue (C) Stones, etc., above 2 mm 2-5 1-4 1-4 0-9 Lat. 14" 45' Fine soil beloir 2 mm. Coarse sand 2 to "2 mm. ... ... 16-9 20-9 15-0 14-8 Fine sand '2 to '02 mm. ... 24-8 • 21.8 22-8 22-6 Silt -02 to -002 mm 10-7 10-2 13-0 12-G Clay -002 to mm 47-6 47-1 49-2 50-0 Liue (D) Stones, etc., above 2 mm. 2-4 1-4 0-9 1-9 Lat. 14 .55' Fine soil bcloie 2 m.m. Coarse sand 2 to '2 mm. ... 170 18-5 13-6 18-0 Fine sand "2 to '02 mm. ... 25'6 22-7 23-5 26-3 Silt -02 to -002 mm 12-2 13-0 14-6 12-6 Clay -002 too mm 45-2 45-2 46-3 43-1 1-8 21-1 24-4 20-7 19-4 25-0 22-1 22-1 21-5 12-1 10-0 13-1 14-5 41-8 44-1 44-1 44-5 ' Clay separation made by subsidence of 8 hours' duration. If increased to 24 hours the average result would be 2'6 per cent, less clay, the silt being increased in proportion. soils of thk gezika Gbzira Soils. Avekages. CHKMIOAi. axalysks 55 First foot Second foot Third foot Fourth foot Liue (A) Potash, soluble in hydrochloric acid ' . (K,0) 0-53 Lilt. 14° 25' ,, iusolublc ,, ,, Phosphoric acid ' Humus ... Humus nitrogen Total nitrogen Carbonates, expressed as calcium carbonate (P.0„) 0-57 0-19 0-67 0-015 0-036 3-5 Water-soluble salts ... 0-08 ■- 0-14 0-33 0-51 Line (B) Potash, soluble in hydrochloric acid ... . (K,0) 0-46 Lat. 14" 35' insoluble „ „ ■■• Phosi^horie acid Humus ... ... Humus nitrogen Total nitrogen Carbonates, expressed as calcium carbonate 0-68 0-20 0-62 0-012 0-018 3-2 Water-soluble salts 0-08 0-10 0-28 0-eo Line (C) Potash, soluble in hydrochloric acid .. . . (K,0) 0-32 Lat. 14° 45' iusolulile ., , Phosphoric acid Humus Humus nitrogen Total nitrogen Carbonates, expressed as calcium carljonate (P..0,-,) 0-74 0-15 0-59 0-013 0-022 3-4 Water-soluble salts 0-10 0-31 0-58 0-49 Line (D) Potash, soluble in hydrochloric acid ... . (K,0) 0-45 Lat. 14° 55' insoluble „ „ .•■ Phosphoric acid... Hnmus ... .. Humus nitrogen Total nitrogen ... Carbonates, expressed as calcium carliuuate (P.O5) 0-49 0-10 0-52 0-013 0-017 4-9 Water-soluble salts 0-lU 0-43 (1-57 0-54 Line (E) Potash, soluble in hydrochloric acid... . (K„0) 0-42 Lat. 15° 05' „ insoluble Phosphoric acid Hnmus Humus nitrogen Total nitrogen ... Carbonates, expressed as calcium carljonatc. (P..OJ 0-58 0-15 0-51 0-011 0-020 5-7 Water-soluble salts 0-10 0-30 0-57 0-58 10 hours' extractioli with acid of 1-115 Sp. Gr. ■■' Excluding No. 1 which was exceptionally high 5G KEl'OKT 01'' CHEMICAIj LAliOKATOlv V O H 1 «« k rH lO "* rJ^ CO o -§ -«** So ;zi 1-1 CO "t-, o -w eo ^ CO s* CO -t •o T-i (N o V CO 'A rH CO »H o •«. CO S (M (N o CD ^ CO !zi Vin ^ O o CO s hi T-4 o •^ ^ CO ^ Oi i*( O CO s^ rH 1-i Is o CN (M iH CO CO 'A |> !M (M T-I CO CO o !2h lO ^ A ^J W V"" (0 § o «» 05 •fca «H O ?^ :; r. t6 a Ci -f .__ t> t^ -t o iH rH i >C ^1 C-. 1 Ol -t CO (M (N iH CO O rH CD 00 o C^ Oi rH C<) rH ^ t^ ^ O l-^ X> c-1 -M W rH iM rH ^ rH »c CD CO O (M (M t-^ (M rH -* CO to (N CO CO O-l O tH rH 1 C-1 _ I^ r-i t'- r^ iM t^ T-l ^ rH <:d »o a> -+ rH ^ c^ CO rH ■M " CD -M ., rH CO 71 71 I-- iH Tl ^-\ -r 1 C^J -t Ci ,. o O -f ■T. r-t rH >c - s ,s» _; :: rO S id s i--o o 1 O 1 1 o G^ o o o o CO l^ rH X r-i Ol rH 1 -t 7^ CI o t-\ CM CO o 1 O 1 I o (M o o O o X CO ■r~\ X 7^1 CO C3 CO rH 1 O o o »c o 1 o 1 1 o t-t CO lO CO iH 1 o o tH rH o 1 — 1 1 o -f — o O o Ol o o (M y-i o o l^ iC 7q CO o o Ci o c rH o o o — '—^ o ^~" ~' '"" i CO CO CO lO -# >o o ■>-{ o Oi T-i -t o 1 '~ 1 ! ' ' CO o o C' "" o CO -+ o CO CD CO yo -r G-1 o a:- ■X Oi 7-1 G<1 ^ 1 '— ■ 1 1 ~ -f ■^ T-i ^ -" i O o" 2" g" O W p- — _- - o 71 7? ■^ c3 O (H J 5 .2 C! ■« ^ = -i ■§ W E^ d ^ c 1 cS !►, -i3 X> ;^ o ■rt rl y o ^ o M EH - =' SOILS OF TH\i GK/Aii\ 57 ^ CO '5 ^^ CO iH CO o "S> O t- tH 00 ^ s °3 1? o a 'g F: ^ <© O ^ ?S O CO CO 1 ^ CO rH CO o ^ t CO CO ^ I- Tt* CO rH o -* CO a rH CO S 2 s o rH CO i:t. To CO rH rH CD '?; 1? o s s ^^ CO 6 rH CO ^^ s CO -t CO j-i CO o ? CO c<» ^ CO o 1 rH CO -f CO G^ V CO o rH CO '^^ . '=^ « ^ J CQ "< q> ?* Qt ^ O « g f^ cu hc; a sS ■43 s fN sJ .a 1 i-H y IZi >£, o if "S a VJ :u ■^ « ^ CM a a rO ^ 1 I Qd o g ^ ••fi, & > '^ a> ;5i 8 a 58 KEl'OKT OF CHEMICAL, LABOKATORY O CO C -« o CO o ■* CO o rH -* CO o tH CO ;z; ~ Oi ^ iO lO iM 1 ■* CO rH CO iz; o »o CO »o rH ■^ CO O r-* CO !zi iH lO o to tH rH -* CO O iH CO |zi tr- lO » j-i -1* c: -f ^ I^ Oi ■* ^1 -+ 1-H CM rH -* rH 'M ^ t^ CO C5 CO a rH T-i o Zl ^ = a C^ CO CO OS rH C CO X' 1 C-1 cn o t>- ■:-i ■M ^-t co rH Oi CO 03 O rH o l^ O-l GO -** Tt< '^ ,_( rH 3 CO xO O Ol CD .— ( l- ,_, ,-^ Tl 'M r— 1 CO r- a CO 00 ,— , CO c: •o ^-t Q-1 lO ^, CO CD- ■ C' ,_, O) C5 C-l CO -M n CO ^ ~f T-i -J- s : i ■^^ s o Qi —* - •tSi _ 05 ^ ~ •M -M 3 v^ fM ■r „ -B O 3 +3 .* +^ -*^ ■71 s; tM ■M c O ri> r^- (* C •72 CJ t*^. +^ o t^ y: :^ 1 I -M r-1 rM uO 6 -H CO X o Ci (M rH rH -t ^ CO -f lO o O c^ ^-^ -f t- I> o —' o o o o CO '" ■"' o o I t I- t b 6 I I l-- 3 T-( rH rH c<> o o o o I I "* 0 5 O I> o »o t^ ?^ ^ eo o 1-1 CO % SM CO lO 00 S >o rH 05 »n CO rH CO ;zi g^ ?-. Oi 05 s 531 iH CO iz; ^«^ o iC ^ IZi w 1-t >0 CO r-l CO ^ .~ »c o lO 1-1 i-H T^ CO ;2i k T-H (M Oi lO CO iz; W CO OS rH O CO •H CO iz; lO CO CO »c 1-H V CO d rH CO 'A .— t iC o iC '"' d rH CO ^ 'W t- 03 »C -c o ^ 1 ® -«• ' 8 a* o H « CQ 0) f«: - o D S '-i^ 4> H cq 1 s a T-\ := v> ^ ^ i -s a ' o ^ 1 ^ X &> "Jr. 8 8 o o CI CO 7^ cc "M ^ 9 00 GO rH C-1 C-1 00 CM 9 CM .p i to o CM O 6 rH 1 ^ rH -t 00 (M CO CO :o X rH 1 ■M CO 9 o 5^ (-0 o o — O w; rH nH 'M r^ o ri If:. o o b b b h b b I S I I I o I I 2 03 o oo CO >o I-t »o o o o iH t^ 00 O o o o o ~^ l^ o o o o I I o CC 1-H o CO T-t CO tu iC >o u* »c = o »r: c~. OD o o o o o o ^ o o G O I i I I 1 I o ^ o b O O rH \ >< I I O ^f rH o lb o I I O O IJ^ O ^ C) I I I I 05 rH rH O eo rH — -; e O Ii| Ph W Ei 60 KKl'Uli'l' OF I'HKMKAL LAliOKATOKV GrTPSUM AS A FERTILISER FOR SUDAN SOILS The benetieial effect of lime ou soil has long been known and its application in one form or another is a common practice, especially in humid regions. In arid regions like those of the Sudan, the removal of the lime originally present as a soil constituent, by the leaching effect of infiltrating rain water, takes place only to a very limited extent, and the soils in such regions are usually so rich in lime that its application as a fertiliser is rarely if ever practised. There is, however, evidence to show that under certain conditions a soil niay be benefited by lime even when the analysis indicates that there is already present a proportion much above that usually laid down as amply sufficient. Thus in certain heavy, difficultly permeable, soils, lime, existing as small nodules of calcium carbonate, may be so enveloped by clay as not to be available, and the application of lime in a soluble form may be distinctly beneficial. The chief beneficial effect of lime, in the form of calcium carbonate, is usually stated to be in connection with the growth of nitrifying organisms. Nitrification, the conversion of the nitrogen of the soil into nitrates directly available for plant food, will not take place except in slightly alkaline solutions such as that furnished by the action of water on calcium carbonate. There are many other effects of lime equally important, and these are shared by other compounds of calcium, more especially calcium sulphates. It is especially witli a view to call attention to this latter substance and the possible benefits which may result from its use on Sudan soils that the following notes have been gathered together. It is hoped that they may induce agriculturists in the Sudan to make a trial of gypsum on their lands. If the results obtained are sufficiently favourable, the extensive deposits of gypsum which are known to exist in the Bed Sea Province may possibly be worked ou such a scale as will enable the material to be secured at comparatively low cost. Before going into the details of the effects of gypsum in the amelioration of undesirable soil conditions, a glance at the figures in the following table' may not be without interest. They show the results of a very great inmiber of fertiliser experiments on cotton soils carried out between 1887 and 1907 and reported in the official bulletins of the United States Department of Agriculture and of the State Experiment Stations. The actual experiments were performed on one tenth or one twentieth acre plots, but the results are uniformly stated in pounds of cotton per acre. The value assigned to the cotton was only 10 cents {•2 piastres) per pound, but the present value in Egypt is nearly twice that figure. Experiments were carried out both with single fertilisers and witli mixtures. The table given includes the trials made with single fertilisers alone, since it is only in this series that gypsum figures. It will be seen that the total area treated with gypsum was 11 acres. Since the trials were made on one tenth and one twentieth acre plots, the number of different experiments carried out with this fertiliser was between 110 and 220. The proportion of cases of success to cases of failure was as ten to one, and the average gain per acre over the cost of gypsum was #18 (about P.T. 360.) or, at the present value of cotton, nearly double that figure. This is six times greater than the average gain from the use of any other form of single fertiliser. It is fully realised that such a comparison as this is far from scientific; but, as noted by Whitney,'- owing to tlie generally inadequate description of the soils in the rejjorts of experiments, it was not found practicable to group and to analyse the data with relation to ' Taken from " Fertilizers for Cotton Soils," M. Whitney. U.S. De.pt. of AyrieiiUiirc, Burean «f Soils, Bulletin No. 62. - Whitney, Inc. cit. (lYl'SUM AS A KK.HTirilSKK FOK Sl'DAN SOILS 61 'la _^ o X' 'C CO _ ■M 1--. Ci -h CO CO iC i^ t^ CO ^ t-- ^. CO o o tq 00 i.q 1> CI Cl -T q -f ■o q X a q "3 Q *M ^ iC lO CO o iH 1-^ ^ tH r-i ci r^ CO CO SS. 1 1 1 1 1 T-t < 1 1 1 1 ^ o (M CO CO o I^ o l> lO ca CO CO CO o -t u -f CO q CN -t q T-H ^1 -f "-H CI q rH CO q q t-. ■^ o CO o co" aj CO CO CI CO -r CO T-< C7S »o Q I-H rH o .s« ' u ao o rt ^ V V 0. 2 -a c a q (M q q q t^ ■q I> q C3 CO q q CO q q -f q > < -+ CO CJ c-i -t CO rH ■ri -t 1-1 ci a> rH CO ci CO I> -r o CD 'O CO cn> CO CO CI CO CD -t CO O 'C A. 1 ' ' rH . o w o CO 00 (M o CD lO t- I-H CO o CO CO s ^ q q q q i-H q -f •o iH q J>; i—f CD q Cl CO !>. q u o -1^ CO tH CO CO iH >o CI CO i-H -1" -I- -f ci u in k. V .a i! ■a •a o CO "to a to ^ O o CO ca ca c- ,_l CO o o >C tt* u c CO (M CO t^ o o CO o Ol I-H OS CD t-- CO o CI 1 n 3 iH i-H 11 C^l i:^ (M fM lO I-H CI CD^ d co^ Ci C5 rH 1 &. r-T r-T fN -f CO CO o Ol -t o o 05 CO O t^ -t CO O a O CO CO oo in CO o Oi CO Tt" o o lO CI CO CD §^ -f •M CO CM tH CI I-H rH ca CI CI ""• 1 CO -a o O o o o o o o o O o o o o o ^ "SS c 3 +J +i +^ -*a +:• +3 +» +3 += -*^ 4J +3 HJ 1 « u o CSI CO iC o CO X o !>■ o CO 3D l-^ CI t* MC CO o CO 1> cr. -t CI CI CO o f-i -* -f CI c — iH 'M rH 1 ' 1 1 1 1 1 1 1 ! 1 1 1 1 1 O U -> tn --H i-H T-i w i-t iH I-H rH I-H rH fi tH •-S .2 1 Si >C I^ '•p o c CO' CO o O o m q o o o 1 o r^ §1 CO ■^ Ci CO tH ^ « m CO CI iH CI c^ 6 rH -r t— 1 c in 22 4J CO t^ I— ; -f CO ,r> -f CO o o rH CO CI CO CO •r~\ i-H o o £ O I-» iH CO iH "fl c^ CO rH J-i >o H« < r^ rH iC ■a u 3 1. 1 □□ rt O 'S ja •s X U o ja C« fc^ a a QD (S ■a 1 6 a> J3 i u '3 "S c 5 CO 4-« O C3 2 p. (10 M ■s "S J3 ■3 o r-4 O 1 1 ■1 o 1 .2 'So § 0) .1 "5 ao o! -^ O O a 1 o 1^ -4^ 1 s 3 62 BEPORT OF CHEMICAL LABOEATOUY the chemical and [)liysical differences in the soils. The results are nevertheless interestinjj and suggestive. Whether anything like the same benefit will accrue from the application of gypsum to Sudan soils, can only be determined by trials on a practical scale ; but it may be mentioned in passing that a number of pot cultures in the laboratory, with heavy soils, showed marked improvement to result from the addition of moderate amounts of gypsum. The following is a brief account of the various effects which have been noted as resulting from the application of gypsnm to soils, and which may account for the increased yields which have been recorded in the table given on 2'«;/e CI. (rt) The harmful action of sodium carbonate on soils and on plant life is well known. It appears to be the most injurious mineral ingredient found in otherwise good arable soils. Its action is both direct, on the plant, and indirect, on the soil. On the plant it acts by corroding the root, crown or stem and in some cases the latter is girdled, the bark being completely destroyed. This effect is especially liable to take place when the salt accumu- lates near the surface as the result of evaporation of the soil water. The effect on the soil is no less injurious, a proportion less even than one tenth per cent being sufficient to destroy the tilth of the soil and to render it unproductive. The effect is due to the defloceulation (puddling) of the clay,, with the result that the soil becomes almost or quite impermeable to water. In cases like the above, gypsum acts as a veritable specific. The impermeability of the soil renders the simple washing out of the salt impossible ; but the addition of a small amount of gypsum to the soil water suffices to convert the sodium carbonate into compara- tively harmless sodium sulphate, the clay being coagulated and rendered permeable at the same time. The fact that too little attention has commonly been paid to the drainage of irrigated lands is a truism. When this lack of drainage results in the stagnation of the soil water — i.e. insufficient aeration — sodium carbonate is almost invariably formed. Treatment with gypsum should be the first step in the correction of such a condition. (b) The harmful effects of ordinary salt — in fact of practically all soluble salts, even those employed as fertilisers — when present in any but very small proportions, are only too well known. The case of one of these, sodium carbonate, has just been discussed, and its destruction by the use of gypsum noted. The effect of gypsum in this case is usually explained by a consideration of the chemical change which takes place when these two bodies are brought together. It has, however, been discovered that the inhibition of the injurious action extends to other salts as well, a fact all the more remarkable as no adequate explanation can as yet be given for it. In an investigation' of the effect of alkali soils on vegetation carried out in 1902 by Kearney and Cameron, of the United States Dept. of Agriculture, a very important fact was recorded in connection with calcium sulphate. It was found that while certain very small jsroportions of salts, such as common salt, sodium sulphate, magnesium chloride and sulphate, were distinctly injurious to plant life, in the presence of calcium sulphate these proportions may be greatly, in some cases even enormously, exceeded, without apparent injury. Thus, under the conditions mentioned, the maxima of concentration of the salts endurable by the plant were increased as follows : — Magnesium sulphate ... ... ... 480 times chloride 80 ,, Sodium sulphate ... ... .. ... GG chloride ... ... ... ... 10 ,, ' "Snnio nintii.ll relations between alkali soils and vegetation." Bulletin No. 71, 1902 r.YPSUM AS A FKRTTLISER FOR SUDAN SOILS 63 In othor woi'ds, if, in ordinary soil, one part of iiiiigncsinm sulphate was foinid to be suillcient to prevent plant f,'ro\vtli, in the presence of calcium sulphate the proportion of magnesium sulphate could be increased to 480 parts before the same poisonous efi'ect was manifest. The experiments mentioned were carried out with wheat seedlings. The results obtained in 1902 have been confirmed by later investigations. Kearney and Harter in a bulletin' which appeared in 1907 have detailed the results obtained with four varieties of dura, two of oats and two species of cotton (the Jannovitch Egyptian and an American variety). The results establish the fact that different genera and species differ greatly in their power of resistance to the salts of so-called alkali soils, and that in the presence of salts of calcium, especially calcium sulphate, these differences are much less pronounced. They, however, still exist to such a degree as to leave no escape from the conclusion that some species and varieties of these plants are better adapted than others to grow in soils containing a relatively large amount of these salts. The extent to which gypsum neutralises the poisonous effect of salts, thus increasing the resistance of the plant to the same, varies with the nature of the salt. In the case of sodium chloride the resistance of the plant was increased as follows :— Lupine (white) ... ... ... ... 5 to 10 times Wheat 5 to 10 ,, Dura ... ... ... ... about 10 ,, Oats „ 9 „ Cotton ((V. harhadense) ... ... ... ,, 32 ,, Beet ,, 8 ,, These experiments were carried out with solutions saturated with gypsum. It was found, further, that when the latter was present even in very small amount, its effect was still marked. Thus, when present to the extent of only one sixteenth part of that required for saturation, the neutralising effect was more than one half that observed for the concentrated solution. It may l)e stated here that very little calcium sulphate is required to form a saturated solution in pure water — approximately two parts in one thousand parts of water. The experiments detailed above were all on a small scale, in the laboratory. A very interesting series of field experiments on a larger scale was made at the Experiment Station of the Hawaiian Sugar Planters' Association.- These were directed to the determination of the effect of salt in the irrigation water of sugar plantations and the extent to which neutralisation miglit be effected by the addition of calcium carbonate (in the form of ground coral) and of gypsum. Very large amounts of salt were used in those experiments (200 grains per gallon of irrigation water, amounting in all to over 30 tons of salt per acre) so that the effect on the yield of cane was very marked. It was found that when gypsum or ground coral were employed, the effect of the salt was to a certain extent neutralised, the yield of sugar being increased by about 46 per cent. The quality of the juice was also slightly improved, the sugar being associated with less impurity. (c) The importance of maintaining a good "tilth" — that loose friable condition of the soil in which the clay exists, not in the puddled state, but agglomerated into small aggregates, thus permitting the free access of moisture and air — is so well known as to ' Bulletin No. 113. Bureau of Plant Industry, U.S. Dept. of Agriculture - Report of the work of the Experiment Station of the Hawaiian Sugar Planters' .\ssociation, Division of -Agriculture and Clicmistry, Bulletin No. 11. 64 KEPOET OF CHEMICAL LABOEATOKY require no special discussion here. A fact Avhich is, however, very generally overlooked in the Sudan, is that the penetrabilitij to roots should extend as far heloio the surface as possible. Gypsum, like other soluble lime compounds, has a marked effect in loosening heavy clayey soil. Being more readily soluble than, e.g., calcium carbonate, it is more quickly removed in the drainage water and the effect is, therefore, less lasting. At the same time it must be remembered that once the soil is made more penetrable and the roots have reached a greater depth, the fact of their presence, and the introduction of more organic matter into the soil, renders the penetrability easier to maintain. The effect of clay in lessening the cohesion of heavy soils was clearly shown by the results of trials made recently in this laboratory. Several soils were made into bricks, in the one case with pure water, and in the other with water containing gypsum. The clay in the soil was, of course, partially puddled in the operation. After these bricks had thoroughly dried, their transverse breaking strain was determined. In three out of four cases the breaking strain of the bricks which contained gypsum was very markedly less than that of those made from the untreated soil. It is evident, therefore, that a plant root would usually lind a gypsum-treated soil more readily penetrable than the original soil. (d) Gypsum shares the effect, exhibited by certain otlier soluble compounds, of rendering the potash of the soil constituents more available. The gypsum, being comparatively soluble, easily penetrates the soil and acts upon the jaotassium compounds, especially the hydrosilicates (zeolites) formed as the result of rock decomposition under atmospheric influences. The result of this action is a partial substitution of calcium for potassiiim, the latter being set free in soluble form, available as plant food. (e) It has already been mentioned that lime (or magnesia), in the form of carbonate, is essential to nitrification, and that these act by neutralising the nitric acid as fast as it is formed, thus preventing its accumulation and the consequent inhibition of the growth of the nitrifying micro-organisms. Calcium sulphate has been found to have a remarkably stimulating effect on the growth and multiplication of these bacteria. In this respect it has been stated by Pichard ' to be more effective than any other substance known. Taking the effect of gypsum as the maximum, he found that, other things being equal, the amounts of nitrates formed were as shown in the table below — Gypsum 100 Sodium sulphate ... ... ... ... ... 47'9 Potassium sulphate. . ... ... ... ... 35-8 Calcium carbonate ... ... ... . ... 13-3 Magnesium carbonate ... ... ... ... 12-5 Hilgard has noted that the above results are confirmed by his observations on the soils of California. The explanation of the increased nitrification in tlie j'l'esenee of gypsum appears to be due to the fact that the latter acts as a food for the micro- organism. The importance of this effect of gypsum in the treatment of many Sudan soils may be very great. There is considerable evidence to show that certain of them are, from one cause or another, markedly deficient in nitrifying power. The application to these of gypsum is well worthy of a trial from this point of view alone. (/) The partial exhaustion of soils by the continuous removal of the same variety of crop, without rotation, cannot be explained simply as the result of abstraction of plant food. Several causes are probably at work and there is evidence to show that among ' Quoted in Soils by Hilgard GYPSUM AS A FERTILISEE FOR SUDAN SOILS 65 them may prohably be included the accumulation of products excreted by the plant, these (like all excretory products) being poisonous to the organism ])ro(luciiig them. The beneficial effects of fertilisers has been thought by some to be due, in part at least, to their direct action on these harmful excretory products. It is not impossible that gypsum may have an action of this kind. Be this as it may, the sterility of some soils has been shown by Shreiner and Shorey of the United States Department of Agriculture to be connected with the presence of certain organic compounds among which is, e.g., dihydroxy- stearic acid. An interesting point in this connection is that while some of the combinations of this acid with bases are as harmful as the acid itself, the potassium and calcium salts are apparently without injurious action. Soils containing this or similar bodies might therefore, it appears to the writer, reasonably be expected to be made more productive by the application of gypsum as a fertiliser. ((/) Lastly, gypsum has long been known to act favourably in promoting the growth of leguminous crops, notably the clovers. The rationale of its action in this case appears not to be fully understood. It may be due to effects not yet discovered or to a combination of those already detailed. Amount to be applied. — The usual dose of gypsum is about 200 pounds to the acre. In special cases, where there is a notable proportion of sodium carbonate present and the soil well puddled in consequence, 500 pounds or even more may have to be employed. It should be remembered that lime either as such or in the form of gypsum is not a plant food in the ordinary sense but that it acts rather as a " tonic.'' It would not be good practice to continirally fertilise with gypsum since the effect might be to impoverish the soil too rapidly. The gypsum should, in most cases, simply be added to aid in bringing the soil into good working condition, and its maintenance in this state can be effected only by having regard to the requirements of each special case. Gum Eeseakch The disadvantages under which the laboratories have been labouring in connection with research on gum jjroducts have already been mentioned. Many lines of investigation have had to be abandoned for lack of efficient aid in the field work. Some advances have nevertheless been made, and it may not be without interest to state, briefly, some of the points which appear to have been established, as well as to indicate the lines along which investigation should be directed in future work. (Jriijin of i/uin. Mr. Edie's researches, both those detailed in the Third Report, and those carried out during the winter of 1908-9 {vide par/e 73), appear to confirm the earlier work of Greig Smith, which indicated that gum production is the result of the Microbial activity of a specific micro-organism. The mode of infection probably varies in different ong'" of g*"" cases, but there is evidence to show that ants, and possibly other insects, frequently act as carriers. During the course of some experimental tappings of talk trees (Acacia seijal) in the Sennar Province, the freshly exuded sap was observed to attract streams of ants to the cut surface, and it seemed extremely likely that this fact was directly connected with the infection of the tree by the gum-producing microbe. The attention of Mr. K. S. Edie. who was at this time at work on hashdb trees (A. senega!) in the Kordofan forests, was called to this point, and his observations {aide page 81) appeared to amply confirm the conclusion arrived at. Such a view would 66 BEPOET OF CHEMICAL LABOEATOEY ••Tapping" experiments explain the results which were obtained from the attempted inoculations of hashnh trees in 1907.' The experiments made at that time were as follows : — Tapping was performed by making a series of gashes with an axe, no bark being stripped off, and (as it was thought that the chances of efficient inoculation might thus be lessened) an attempt was made to ensure the entrance of the microbe by rubbing a moist rag over the bark, in which the microbe was presumed to reside, and subsequently into the cut. A series of trees tapped in the usual native fashion, by stripping the bark, were also treated in the same manner for comparison. The number of trees operated on in each case was twenty-five. The following table exhibits tlie results obtained : — Methods of tapping Method of tapping Treatment Yield of gum per tree per season Ordinary Short gashes ... No attempt to inoculate Attempted inoculation No attempt to inoculate Attempted inoculation 0-9 rotl O-O.'i „ 0-2.S „ 0-14 „ When the results were reported it was suggested that possibly the lower yields were due to the fact that in place of increasing the extent of infection, the procedure adopted had, on the contrary, lessened it, since the sajj exuded was, in great part, removed by the wet cloth. If we accept the view that ants are the chief carriers of infection the results become still more comprehensible. A portion of the sap having been removed (and the amount exuded is usually very small) there was no longer sufficient present to attract the insects, in the usual number, before the cut had healed. Another point which is of interest in this connection is the effect of cold at the time of tapping. No further experiments have been made since those already reported on, but these appear to show conclusively that cold at the time of tapping has a marked retarding effect on gum production. Should the weather remain cold for several successive days after tapping, the operation may even be entirely fruitless. It seems reasonable to conclude from this that the growth of the microbe is inhibited by the low temperature, and that the wound heals before the tree has become suflBciently infected. It should be stated here that the exact determination of the quantitative effect of any given condition on gum production is far from easy by reason of the difficulty in securing a reliable standard for comparison, /. e. a tree of the same age and size, perfectly sound, to the sap of which the microbes are unable to gain access. Small cracks or fissures at one point or another are almost sure to be found during the dry season, and a certain indeterminate amount of infection appears to be inevitable. Methods of tapping. The native method of tapping hash/ib trees is to remove a strip of the bark, from 2 to 3 feet in length and 1 to 3 inches broad, according to the size of the branch operated upon. Trials have been made to improve upon this method by removing shorter strips, or by simply gashing the tree, but up to the present, the native method appears to yield the best results. Tapping of gum trees is only a comparatively recent practice in Kordofan, and is said to date from about 30 years ago. Previous to this, only such gum was collected as was found exuding from natural fissures in the bark. Tapping by the present method was found to increase the yield to something like five times the former amount. Fide Third Report, Wellcome Tropical Research Laboratories, p. 422 GUM EESEARCH 67 Varieties of hashdh (jum. The gum found exuding naturally is called by the natives Varieties of " wady " gum and is usually held by them to be stronger than that resulting from '"*'"* g'"" tapjoing. This is iiot borne out by the result of examination. Indeed, wady gum is, on the whole, of inferior quality, since it is usually darker in colour than the gum of tapped trees. After tapping, the first collection of gum is usually made at the end of several weeks, more or less, depending in part at least irpon the temperature, which has a marked influence upon the rate of exudation. There are usually seven or eight collections in the season, which terminates when the rains begin to fall. The gum of the earliest exudation is usually not completely soluble. On treatment with water, a glairy, mucus-like liquid results, from which there separates, after a time, a greater or less amount of true solution. This effect is usually less marked in weak solutions. Fortunately it has been found that storage of the gum for a short period is sufficient to correct this defect, the gum becoming completely soluble in solutions even of the highest concentration. Senegal gum has been found to behave in a similar manner. The change which takes place in the gum on storage has not yet been worked out, hut it is doubtless connected with the presence of certain enzymes' which have been found to be present. llashdh {A. Senegal) gum is found to be a mixture of two markedly different varieties, one hard, and the other soft. These may be separated by exposing the mixed gum to the sun. The harder variety remains glass-like and transparent, whereas the softer becomes more or less rapidly covered with innumerable small fissures which make it appear almost quite white. This bleaching of the gum is more apparent than real, and if the gum is originally appreciably coloured it will remain so in the interior. A certain amount of true bleaching does, nevertheless, take place, but it is not very marked. The solution of this variety of gum is less viscous than that of the harder variety. The usual range of viscosity of 20 per cent, solutions of the two varieties is as follows, the figures expressing degrees of retardation in the torsion viscosimeter.- Hard gum ... ... ... ... 60 to 70 degrees Soft gum 28 to 33 ,, The two gums exhibit no apparent differences as regards their proportion of ash and acidity. The origin of the hard gum, that is to say, the special conditions which determine its formation, has not yet been determined. A great deal of work has been attempted with this end in view, but the investigation had to be abandoned for a time for the reasons stated above. If tested shortly after exudation, gum of the six or more successive collections in a given season exhibits a more or less regular decrease in apparent viscosity strength of solutions. This is not due to a higher proportion of the hard, transparent, strong gum in the earlier collection but to the presence of the less soluble, constantly changing variety to which reference has already been made. If the gum is stored for some months before testing, the change to complete solubility will usually bo found to have taken place, and all the collections will exhibit approximately the same viscosity strength. The typical hard, strong, gum suffers but little change in viscosity strength, on storing. Effect of coppicing. A considerable proportion of hashdh gum is derived from coppiced Effect of trees. As such gum might show marked difference in quality or quantity from that <=°PP"='"R ' Vide F. Rciuitzer, Zeit f. Physiolog. Clwm. 1909, p. 352 - Vide Second /ieport, JP'eUcome Tropical Research Lnbornlories, p. 232 68 KEPOBT OF CHEMICAL LABORATORY exuded by ordinary trees, the Forestry Department was urged to furnish us with samples of the two, collected at the same time, and from the same district, for purposes of comparison. This req^uest has not been complied with, but a sample from a single coppiced tree, kindly collected for us by Mr. E. S. Edie, was found to weigh only about one-half of that exuded by an ordinary tree of about the same size. As only a single tree has been tested the result must be taken merely as suggestive. The chemical examination of the two gums yielded results as follows : — Gum from ordinary tree Moisture, per cent. Ash, per cent Acidity (mmtfs. KHO required to neutralise one fframme) Viscosity of 20% solution (degrees of retardation in torsion vis- co.simeter) ... 8-41 •2-67 .3-38 17-1 Talk gum Tapping of talk trees There is no marked difference in quality between the two gums, but both were exceptionally weak as regards the viscosity of their solutions. As has been noted in previous Reports, hashah gum is only formed when the tree is reduced in vitality, irsually as a result of the dryness of the soil. It might be expected, therefore, that in the case of coppiced trees the greater root develojjment in proportion to the size of the tree would result in increased vitality during the period of drought, and consequently in lessened gum production. The single instance mentioned above would appear to bear out this assumption, but many collections will have to be made and examined before a reliable conclusion to this effect may be drawn. Talk gum.. This gum, which is the product of two varieties of Acacia seyal, is of much less importance, commercially, than that of A. Senegal, to which it is inferior in several respects ; the solution of the gum is less viscous than that of the hard variety of hashdh gum, and further, on standing exposed to the air, it becomes dark in colour, usually depositing a brownish-black precipitate. The gum is also more acid than is hash/ih gum. A gum of much better quality may be had by selecting only the colourless pieces. These yield a solution less acid and having little or no tendency to darken on exposure to air. Colourless talk gum is found only in very small proportion in the product as ordinarily collected. Tapping of talk trees. Talk trees are not tapped by the native gum collectors. At our request an experimental tapping of these trees was carried out by Mr. S. A. Wood, Inspector of the Woods and Forests Dept., with a view to determine whether the quality or quantity of the gum was naturally improved. From the report made by this official it ajjpeared that the quantity of gum exuded could be very considerably increased in this way ; and, from the sample of such gum sent us, it appeared also that the quality was even more markedly improved. In the Third Report of the Laboratories, page 436, is found a comparison of the results of examination of this sample with that of ordinary gum collected in the same district from untapped trees. Later investigations disclosed the fact that for some unaccountable reason the sample furnished as gum from tapped trees had been picked, all the darker particles being rejected. The comparison of results referred to above was, therefore, rendered not only useless but misleading. GUM BBSEARCH 69 A second attempt was nuule to deterniiiie the effect of ta])ping talk trees hut the forestry official to whom the work of tappiiif^ and collection was confided again failed to carry out the work in a satisfactory manner, and the results were useless except to confirm the conclusion that the sample first furnished was not reliable. A third attempt was made, the operations of tapping being carried out under the supervision of the writer. The trials made and their results are detailed below. Red 'J'aUi Trees. Ttippal \uvf/iiher 18 rpj) a \ m Til ^"^^' trees bERiES A. ien trees tapped by a clean cut with a sharp spoke-shave, removing both bark and liber. The strips removed were about 10 x 1| inches. They were taken from the trunk alone, since the branches were too high to be accessible. Nov. 24. No appearance of gum. Bee. 18. No appearance of gum. The cut appeared to have healed wlien observed on Nov. 24. Series B. Strips, of outer bark only, removed by means of the spoke-shave. Five trees operated upon. Nov. 24. No appearance of gum. Dec. 18. No appearance of gum. Series C. Simple horizontal gashes, to the number of about six on each tree, made with a small native axe. Six trees operated upon. Nov. 24. (luni just beginning to exude. Dec. 18. All the trees but one (C 4) showed exudation of a small amount of gum at each gash. The gum was dark coloured except in the case of tree (C 5) which liad apparently been less affected by fire. Series D. Two horizontal clean cuts, so made as to remove a wedge-shaped piece of the bark and liber, and exposing the wood. Six trees operated upon. Nov. 24. Gum beginning to appear. Dec. 18. Nos. 1, 2 and 3 showed a slight exudation, No. 4 a large blob of gum of rather dark colour. No. 5 tree, markedly less affected by fire, yielded no gum. Series E. Six trees treated as in D. but the bark adjoining the cut bruised by blows with blunt end of the axe. Nov. 24. Gum beginning to appear in most cases, but of dark colour. Dec. IS. In four cases gum was found to be exuding in fair quantity but rather dark in colour, especially where it had passed through the cracks in the bruises. Two of the trees, which were larger and less affected by tire, showed no exudation at all. Series F. Simple vertical gashes made by blows of a small sharp axe. Six trees operated upon. Nov. 24. Gum beginning to exude in some cases. Dec. 18. Two trees showed practically no result from the tapping. One of Tapping them was almost completely shaded by a large tree with heavy foliage. The '^^P""""""^ i-emaining four trees showed gum either at some or at all of the gashes. Series G. Bark and liber removed and an auger hole made in the denuded wood. In this case the removal of the bark was effected by making a horizontal cut and loosening the bark beneath it by bruising slightly with the blunt end of the axe. In this way a strip of the bark may be pulled off by hand, an operation which is otherwise sometimes effected with difficulty. Six trees operated upon. Nov. 24. No appearance of gum. 70 liEPOET OF CHEMICAL LABORATORY Experiments continued Dec. 18. In the case of one fairly large sound tree, no gum was found at the point of tapping. In all the other cases there exuded a fair, and usually quite considerable, amount of gum which was dark in colour, especially in the case of the trees most injured by fire. Seeies H. An auger bole, about ^-inch in diameter, made through the bark and into the wood. Six trees treated. jVoy. 24. Slight exudation of gum, from between the bark and the wood. None from the wood itself. Dec. 18. A small quantity of gum — usually of dark colour — exuding from most of the holes. Series J. Holes made by "pecking" with a sharp-pointed axe. Six trees operated upon. Not'. 24. Gum beginning to exude at each point. Dec. 18. In five out of the six trees, gum of fair colour was found exuding from nearly all the " pecks." In one case, when the exposed roots of a large tree had been much affected by fire, but the rest of the tree practically uninjured, there was an extremely large amount of gum yielded at one of the " pecks." Series K., L. and M., These trees were tapped by removing the bark and liber in the manner detailed for Series G, and the cut surface was treated with strong anti- septics in order to determine whether such treatment would prevent inoculation and gum production. Series K. A solution of corrosive sublimate rubbed into the freshly denuded surface and the neighbouring bark. Three trees operated on. Nov. 24. Gum appearing in greater quantity than in the previously detailed experiments. Dec. 18. Considerable exudation of gum, of rather lighter colour than the average yielded in the former series of experiments. Series L. Diluted formalin applied, under the same conditions, to three trees. {Note. The roots of these trees had been considerably injured by fire.) As in the case of Series K, all three of the trees exuded a large amount of gum. In one case there was not less than a pound, partly hanging from the tree and partly lying on the ground. Series M. Three trees taken. The treatment in this ease was with a solution of sodium metabisuljjhite. The results yielded were in the main similar to those in the case of experiments K and L. One of the trees yielded practically no gum, but the remaining two yielded much larger amounts than in the case of simple tapping without the use of chemicals. The gum, in all three cases of chemical treatment, was evidently exuded much more quickly than is usually the case, and it was spread over the tree to such an extent that it could not be removed without great loss. White talk trees White Talk Trees These trees were tapped on November 19 and the observations made on December 17. Series A. Ten trees. Simple tapping by removing a strip about 8 inches long, including both bark and liber. The removal of the bark was effected as in Series G of the Red Talk trees. When examined on December 17 most of these trees, fairly sound, showed little or no appearance of gum. Several trees which had obviously been injured by fire exuded a fair amount of gum of dark colour. GUM EESEARCH 71 Series B. Eight trees, tapped as above, but, in addition, an auger hole made into the denuded wood. In most cases no gum was found exuding from between the bark and the wood, and in no case was any trace of gum found exuding from the hole in the wood itself. Series C. Six trees treated by boring holes with an auger, through the bark and into the wood. In the case only of one tree was gum found exuding, and from one only of the several auger holes in it. Series D. Five trees treated by removing a wedge-shajjed piece of the bark and liber, and bruising the adjoining bark Isy blows of the blunt end of a small axe. Three of these trees yielded no gum at the point of tajoping. The remaining two yielded a fair amount, but very dark in colour, especially where it exuded from the cracks in the bruised bark. No further reliable observations were made on the above experimental tappings. It was intended when they were instituted that observations be made periodically — once every two or three weeks — in order to watch the course of the exudations. It appears, however, that the forest officer in charge of the district did not visit them until three months after the date mentioned above, when he reported the practical absence of gum from any of the trees except those which had been chemically treated. We were therefore left somewhat in the dark in this connection ; but, in view of the fact that the apparently sound trees, both of White Talk and of Bed Talk, did not yield gum when tapped in the ordinary way, there seems to be strong evidence that, at least in the immediate district in which these experiments were performed, the conditions were not naturally favourable for gum production. Taken altogether the results appear to justify the following conclusions : — (a) In the district in question sound white talh and red talh trees — i.e. trees which have not been injured by fire or which have not suffered marked mechanical injury — are not susceptible to the action of the gum-producing microbe, and will not, therefore, yield gum on simple tapping in October and November. It appears evident Conclusions that at this time the ground is still sufficiently moist, from the lately fallen rains, to keep the trees in a resistant condition. Such trees would probably yield gum if tapped at the end of February or, better, in March, when the weather is hot again and the soil contains less moisture. On the other hand, the soil itself appears to be of rather good texture and fairly rich, which, again, is unfavourable for gum production under ordinary conditions. (6) Trees w'hich have been injured by fire are more susceptible and yield gum fairly readily ; but such gum is very frequently coloured to an objectionable degree. Even in the case of these fire-injured trees, it would seem preferable to perform the tapping as late as possible after the rains, so that the exudation may not be checked by the colder weather of December, January and February. (c) Tapping by cuts with a sharp instrument, such as a spoke-shave, will probably be least effective, since such cuts appear to heal too quickly. This metliod, has, however, not yet been tested on trees known to be in a low state of vitality, and therefore most suitable for gum production. id) Bruising of the bark around a cut allows freer outlet to the gum and increases its production ; but the resulting gum, especially if it exudes through cracks in the bruised bark, is coloured to an objectionable degree. 72 liKPOHT OK CIIEMTCAL LArWHATOHV (i-) rf fduiid feasible it would roduct soluble even in alcohol of a much higher strength, and any of this sugar which had been unchanged by the bacteria would remain in solution when the product of bacterial action was precipitated at first. The same is true of the other constituents of the original medium, and", in addition to this, the amount of asparagin and potassium citi'ate used was so small that the new substance could only be accounted for by having been derived from the levulose, and perhaps to some extent from the glycerine, any products of a protein nature being removed in the treatment of the original solution with acid and afterwards with alcohol. The bacterial product did not reduce Fehling's solution, but on boiling a solution of the former in five per cent, sulphuric acid for twenty minutes and then neutralising this, a good reduction took place, showing another point of resemblance between this bacterial product and the gums exuded from tapped trees. Some of the ordinary gum tests were also tried, such as coagulation with basic lead acetate and ferric chloride, and these gave positive results. From the examination of this bacterial substance, therefore, we must assume that these bacteria, when grown in an artificial medium of suitable composition, produce, among other things, a substance of the same nature as the gum of the tree from which they have originally been isolated. The formation of gum from the sugars of the sap is thus, as Greig Smith pointed out, the result of pathological conditions. In the first place, exudation of gum does not take place from Hashcib trees in those districts where the rainfall is high compared with that met with in the gum-producing districts of Kordofan. In the latter case the tree as a whole is in a state of reduced vitality, and the tissues are therefore more susceptible to the effects of abnormal injuries. In those districts where the proportion of Effect of moisture in the soil is relativelv high, the injury produced when the tree is tapped is "is^<;hanicai "•'•''- ^^ injury to the repaired before the bacteria present have time to produce any appreciable amount of gum. ,rees The wounding of the tree through tapping causes a local weakness in the tissues next to the bark, and the cells of the tree are then unable to prevent the rapid increase of, and increased transformation of the sap by, the bacteria. This appears to go on until sufficient new bark has grown over the wound to form a protection for the exposed surface. In the case of a tree which was quite free from bacteria, it has already been shown how infection can readily be carried by means of ants or flies. Since my return from Taiara, I have confirmed my work there, both in obtaining pure cultures of this bacterium from gum-bearing branches of hashdb trees, and also in producing a substance of the nature of a gum by cultivating the bacteria on artificial media containing sugar as the principal constituent. I regret, however, that I have been unable up to the present to investigate the properties of these bacteria at all fully. They grow equally well on glucose and levulose media, but not so well on maltose and other sugars, and they do not appear to form a very adhesive slime. Greig Smith found his bacteria to produce slime more readily in presence of tannin, but the hashdh bacteria do not grow at all well in media containing tannin. Acting on a suggestion made by Dr. Beam, I obtained, through the kindness of Mr. Tippetts, Inspector at Taiara, some gum-bearing branches of talk trees from Sherkeila, and made cultures from small portions of these in the manner already j^^ij^dQ^ described. It is of great interest to note that by far the largest nuiidier of colonies of the obtained in this case also were those of the same bipolar bacteria isolated from caiacensic ^ bacteria from hashdb branches. In this connection it is well to recall the fact that Greig Smith /■„//, trees 84 liEl'OET OF CHEMICAL LABOKATOEY states that the bacteria found responsible for gum-fliix in Acacia biiiercata also caused production of gum (of a different kind) when made to infect peach trees. In other instances, also, he found that one species of bacterium could produce gum in the tissues of different species of trees. In cases of this kind the differences in the chemical nature of the gums are probably due to the presence of different sugars or other constituents in the sap of the various trees. In concluding this report, I wish again to express my thanks to Dr. Balfour and Dr. Beam for suggestions and advice which they liave kindly given me in connection with mv work. Preliminary Notes on the Chemistry of the Latex of CaLOTROPIS riiOCERA BY James Thompson, Pli.I). (Heidelberg), F.I.C. Senior Assistant Cliemist, Wellcome Tropical Keseurcii Lal)urutories This investigation was nndertaken, primarily, with tlie object of discovering a test which could be relied on in cases of suspected poisoning. Galotropis procera, a shrub belonging to the Nat. Ord. Asclepiadacem, is a common weed throughout the Sudan, and under the name of " ushar " is well known to the natives as a poisonous plant. When wounded, all parts of the plant exude considerable quantities of an acrid milky juice, possessing a nauseating odour, in some degree reminiscent of rhubarb. The leaves are used in the preparation of " merissa," the native beer, and it is possible that the presence of an excessive quantity, whether there by criminal intent or not, might account for cases of apparent poisoning, wdiicli had been investigated in these laboratories with negative results. In India, under the popular name of " madar," or "akh," Calotropis procera has long been famous as a remedy in a variety of ailments, the milky juice, root-bark, and flowers being in use. A detailed description of the plant and its medicinal Previous uses is to be found in the Pharmacographia Iiulica, Vol. II., pages 428-437. The attached work on the photographs (Figs. 9 and 10) give a good idea of the plant, the first showing leaf, flower ^" '^'^ and seed-pod, and the second the entire plant. While considerable work has been done on the chemistry of the root bark,' the literature at my disposal records no similar examination of the latex, nor has communication with several sources revealed that any such examination has been made. In this connection, one is greatly indebted to Dr. Power and Mr. C. J. S. Thompson of the Wellcome Chemical Eesearch Laboratories, London, for their kindness in furnishing valuable extracts of the literature on the subject. So long ago as 1867, in consequence of the numerous cases of infanticide by the milky juice, Dr. McEeddie, Medical Officer of Hurdur, India, suggested that an investigation should be made with a view to the detection of this poison by chemical analysis, but no record that his suggestion was acted on is to hand. Dr. Duncan- (1829) made a chemical examination of the root-bark, and concluded that its medicinal activity was due to an extractive matter which he termed " mudarine." The authors f)f the rharniacognhphia Iiulica,^ repeating the work of Duncan, failed to obtain anything corresponding to his mudarine, but, instead, an acrid resin, with which was associated a bitter principle, probably the active constituent of the plant. Drs. Warden and Waddel,^ after a lengthy research, concluded that a yellow, bitter i-esin, occurring to the extent of 0093 per cent., was the active principle. The resins and bitter principle described by these various investigators being in all probability derived from the dried latex present in the root-bark, a thorough exannnation of ' Pharmaeoi/raphia Indiea, Vol. II., pp. 428-437 - Duucau, Ediiihurijh Malicn/ and Sitryical Jmirnn!, 1829, XXXII., Go •■ Plidrmacoyraphm Intlka, Vol. II., pp. 428-437 ' Wnrdcn and Waddcl, Phnrmnceiilical Journal, 1885, pp. 1C,') I7(i 86 EEPOET OF CHEMICAL LABOEATOBY L'ses of the juice for niaking an intoxicating liquor the latex itself may be expected to furnish valuable evidence as to the precise nature of the toxic principles of the plant. It is worthy of note that experiments carried out in India' indicate that, while the fresh leaves are extremely acrid, producing severe vomiting accompanied by much pain and distress, the cooked leaves are innocuous, and the writer has observed goats and sheep eating witliered parts of the plant, while the natives aflSrm that they will not touch the fresh plant. The milky juice of G. procera is undoubtedly used in the preparation of an intoxicating liquor, both in India and in Central .\frica, but it is impossible to gather whether the juice itself is directly fermented or is simply used as a ferment or bitter. Sir G. Birdwood {Bomh. Prod., 208) says, " the intoxicating liquor bar is prepared from it by the tribes of the Western Ghats," and Lisboa, in Usefnl Plants of Bomharj, asserts that the tribes of C'eiitral Africa prepare from it their ;/n/'v. Details of the reaction A Colour Eeaction for the Latb.x of CALOTRons Peocera When the latex is diluted with four volumes of water, and a few drops of acetic acid added, and the liquid boiled, coagulation readily takes place, and by straining off the coagulated gutta-percha and filtering, a clear, pale yellow liquid, of decidedly acid reaction to litmus, is obtained. To this liquid, and to the residue left on its evaporation, were applied the customary reagents giving colour reactions. Those having strong sulphuric acid as a basis — Erdmann's, Frohde's, Buckingham's, Mandelin's, etc. — were found to give, with the liquid, green to blue colorations ; of the many others tried, none gave any reaction worthy of note. When sulphuric acid, diluted with one half its volume of water to prevent charring, is carefully added to the liquid, a greenish-blue ring is formed, and, on carefully mixing the liquids, a bluish-green solution results. The addition of traces of oxidising agents causes a change of colour to a deep Prussian blue. Ferric chloride has been found the most suitable oxidiser, the blue colour produced in its presence being comparatively stable, lasting for two or three hours, and then changing slowly to a dark brownish-purple. The more energetic action of such oxidisers as hydrogen peroxide, nitric acid, chromic acid, and sodium persulphate, even when these are present only in minute traces, results in a quick change of the blue colour to dark green, then greenish- brown and finally deep yellow. In the presence of nitrates the blue colour rapidly changes, owing to the formation of nitric acid, while the presence of nitrites prevents its production, the deep yellow colour being at once developed. The reagent found to give the most satisfactory results is prepared by diluting sulphuric acid with one half its volume of water, and adding sufficient ferric chloride to impart a faint yellow colour to the liquid. A mere trace of ferric chloride is not sufficient to develop the blue colour ; on the other hand, excess is to be avoided, as any marked yellow colour of the reagent would interfere. The reagent is added carefully in equal volume to the liquid to be tested; a deep blue ring forms at the junction of the liquids, and, on mixing, a deep Prussian blue colour is developed. A decided blue colour is given by the reagent to 1 c.c. of a liquid containing one i)art of latex in 250 parts. Sulphuric acid alone gives no coloration when added to a latex solution of this strength, but the subsequent addition of a drop of ferric chloride solution brings up the blue colour. The substance which gives rise to the blue colour on the addition of sulphuric acid and an oxidising agent, appears to be present in the latex in very minute quantity, and has not yet been isolated. The following facts regarding its properties have, however, ' Harvey, Report on ifalico-lajal Jtdunis, Bengal Presidency, 1870-1872 REPORT OF CHEMICAL LABORATORY Application of the test in cases of suspected poisoning Addition of latex to nati\'e beer is easily detected by the reagent been established. It is not removed from its solution by ether, chloroform, or petrolenni- ether, but is completely precipitated by basic lead acetate in the presence of ammonia, the lead salt being soluble in excess of lead acetate.' Neutral lead acetate, in the presence of acetic acid, does not precipitate the substance, nor does basic lead acetate when added to a solution faintly acid with acetic acid (the free acetic acid is, of course, neutralised by a portion of the basic lead acetate). After reduction with zinc and sulphuric acid, its solution gives no coloration with sulphuric acid ; on reoxidation, the blue colour is produced. The substance may be removed from its solution by boiling with animal charcoal, and is dialysable. As regards the application of the test to cases of suspected poisoning, it is, of course, essential that a solution of the latex free from substances which would give colorations with the reagent, and so obscure the reaction, be obtained. Several experiments have been carried out, which have led to the adoption of the following method. The stomach contents, and, where emesis has occurred, the vomit, are digested in water slightly acidulated with acetic acid. To the strained liquid is added slight excess of neutral lead acetate, to precipitate albuminous matter, and the liquid filtered. The filtrate is carefully treated with basic lead acetate until no further precipitation results, and, after filtration, the liquid is made distinctly animoniacal. Tlie resultant precipitate is filtered off, carefully washed, and decomposed with dilute sulphuric acid. The liquid is filtered from lead sulphate, and the reagent applied to the filtrate, which may be concentrated if necessary. Experiments have shown that, by following this method, 0-5 c.c. of the latex may be readily detected when mixed with stomach contents. In most cases it will be found that the treatment with basic lead acetate and ammonia may be omitted, and the reagent applied direct to the filtrate from the neutral lead acetate, after freeing it from lead by the addition of dilute sulphuric acid. In cases where the volume of liquid to be tested is considerable, the treatment with ammoniacal basic lead acetate is advisable, since concentration of the liquid is thereby rendered unnecessary, the resultant precipitate being decomposed by the addition of a small volume of dilute sulphuric acid. The addition of the latex to merissa (the native beer made from millet) is easily demonstrated by applying the reagent direct to the filtered liquid. Should the latter be coloured, the preliminary treatment with lead acetate, indicated above, would be necessary. Latex, which has been thoroughly dried by exposure in an open vessel, is found to respond to the test after digestion with water. The concentration of dilute latex solutions by boiling does not affect the production of the lilue colour by the reagent, nor does digestion of the latex with pepsin and dilute hydrochloric acid. The action of the stomach juices on the latex being of primary importance as regards the value of the test in cases of suspected poisoning, the following experiment was carried out. An artificial gastric juice was prepared by digesting the fresh stomach of a sheep in water, and adding to the strained liquid sufficient hydrochloric acid to give it an acidity of 0-2 per cent, hydrochloric acid. To 100 c.c. of this liquid was added 10 c.c. latex, and the liquid incubated at 37' C. Portions of 2.5 c.c. each were removed at intervals of 24 hours and the test aiiplied. The treatment with lead acetate, etc., previously described, was employed, a control test being applied to the gastric juice itself. From the liquid, to which latex had been added, positive reactions were obtained in each case, wliile the gastric juice itself gave no coloration with the reagent. 1 When this lead compound [(d.) Table A, seq.] is decomposed by sulphuretted hydrogen or dilute sulphuric acid, a strong solution of the colourable substance of the latex, to which test reagents may be advantageously applied, is obtained. PRELIMINARY NOTES ON THE CHEMISTRY OF THE LATEX 01' CALOTROl'IS TROCERA 89 The possibility that plant juices, other than those of 0. procera, may give a similar reaction, has not been overlooked, and special attention has been devoted to the examination of foodstuffs in common use among the natives, and which might, therefore, Examination be present in the stomach contents of suspected cases. The test has been applied to "' ("othiuns . ^ ^"^ to exclude between twenty and thirty varieties of dura, which is the staple food in the Sudan, and to possibility as many common vegetables and fruits. The foodstuffs were crushed, macerated for °^ <:onhision several hours in 0-2 per cent, hydrochloric acid, and the reagent added to the strained and filtered liquids. The pale-coloured varieties of dura gave no ring or coloration on shaking. In the case of the red varieties, however, a pink ring was developed, and a deep crimson solution obtained on shaking. A strong extract of a red variety of dura, to which a small amount of latex liad been added, gave with the reagent a purple coloration, as was to be expected. Further examination of the dura extract showed that the red colouring matter is entirely removed by basic lead acetate when added to a faintly acid solution. Since the colourable substance of the latex is not so precipitated, the presence of red dura in no way interferes with the test, as several trials on red dura extracts to which small quantities of latex had been added, have proved. Of the other foodstuffs examined, none was found to give the deep Prussian l)lue colour produced by the latex. The majority gave pale yellow or greenish-yellow rings, and pale yellow colorations on shaking. Two, the cucuiuber and sweet potato (Arabic, hamhe) give pale blue rings, and pale blue colorations. The pumpkin (Arabic, yara) gives a purplish-blue colour, and the vegetable marrow a yellowish-green. TJiis latter colour was found to mask the colour given by the latex when the reagent was added to a mixture of a very small quantity of latex and a large quantity of vegetable marrow, a green tint being imparted to the bhie. With the latex in larger proportion, the green tint was not seen. The foodstuff extracts were also treated with lead acetate, etc., as previously described (vide jpar/e 88), and the reagent applied to tlie decomposed ammoniacal basic lead acetate precipitates. Some differences in the colour reactions given were observed. The cucumber gave a pale blue colour, which changed slowly to pale pink, and the pale blue colour given by the sweet potato was rapidly discharged. The vegetable marrow gave no ring or immediate coloration, but a strawberry-pink colour developed on standing. The pumpkin gave a pale heliotrope colour, becoming more pronounced, and then fading. In all cases the colour faded comparatively quickly ; a solution of latex similarly treated gives a Prussian blue colour, which does not fade until after some considerable time. In carrying out these tests the liquids were not cooled after shaking, and the fading of the colours in the case of the vegetables may be due to the heat developed. It is worthy of note, however, that the blue colour of the latex does not readily fade under these conditions. It is to be borne in mind that the degree of dilution of tlie latex, wlieu no intei'fcring Effects of body is present, determines whether the blue coloration can be correctly described as '•''"""" °" ''"" •< ^ ' Prussian lilue Prussian blue. Dilutions of 1 in 5 to 1 in 25 give an unmistakable Prussian blue, 1 in 50 coloration gives a pale Prussian blue, while 1 in 250 gives what could \h'. described as a pale blue, although of exactly the same tint as a much diluted solution of Prussian blue. The depth, purity and permanence of the blue colour given by the latex aie quite distinctive, but a confirmatory reaction has been discovered which eliminates any risk of mistaking these foodstuff colours for tlie colour of the latex, and also enaljles tlu^ latex to be detected with certainty when mixed witli tliese foodstuffs. A very striking colour-reaction, which, so far as is at present known, is characteristic of the latex, is given by sulphuric acid containing a trace of formaldcliydc. The reagent 90 BEPORT OF CHEMICAL LABORATOnY found to give the best results consists of a niixturi! of 75 c.c. concentrated sulpliuric acid and 25 c.c. water, to which is added a drop of 40 per cent, formaldehyde solution. When .\ confirmatory this reagent is added (not too carefully, so as to allow slight mixing) to the coagulated latex, '^^' or to the decomposed ammoniacal basic lead acetate precipitate [(d) Table A], a grass-green I'iiig or band is dc^veloped. Almost iinmcdiatuly a strawberry-pink band begins to form below the green, gradually deepening, and, after allowing to stand for a short time, super-imposed bands of colour are seen in tlie following order, from below upwards : pale strawberry-pink, (Uu^p strawberry-pink, purph^ (nai'i'ow ring), gi'ass-gre(Mi, chrome-yellow. On now cooling in ice, and carefully shaking, a brigiit olive-green solution, of decided depth and permanence of colour, is obtained. This reagent has been applied to the four foodstuff's, which give colorations with the suli)hnric acid reagent simulating in some degree the blue colour given by the latex. Both strong water extracts, and similar extracts treated with lead acetate, etc. [vide Table \ seq.) were tested, and in no single case was any coloration produced. The formalin- sulphuric acid reagent thus furnishes a certain means of distinguishing tlio latex from these foodstuffs, and of detecting it in their presence. The application of this reagent as a test is being extended to other foodstuffs. The solution of the colour-yielding substance of the latex obtained by decomjiosing the ammoniacal basic lead acetate precipitate with sulphuric acid (vidp. footnote, page 88) has also been submitted to the action of reagents in the following manner. After making faintly alkaline with ammonia, small portions were evaporated on crucible lids to dryness. To the dried residues were added a number of the customary colour i-eagents ; the most striking colour reactions obtained are given below. (a) Sulphuric acid : — yellow, changing to orange and finally to rose-pink. (6) Frohde's reagent : — bright greenish-blue, changing at once to grass-green, then slowly to mahogany-brown and finally olive-green. (c) Sulphuric acid and ferric chloride : — electric blue, changing at once to green then brown and finally dark olive-green. (d) Sulphuric acid and trace of formalin: — purplish, then rose-pink. Speaking generally, sulphuric acid with oxidising agents gives blue and green colours, while, with reducing agents, sulphuric acid develops yellow and pink colorations. Analysis of The latex of G. procera is distinctly acid in reaction ; the acidity to phenol-phthalein the latex ^j eight samples, expressed in terms of acetic acid, was found to vary from 0-13 grannne to 0-23 gramme per 100 c.c. One sample, collected from several plants, showed an acidity of 0-46 per cent. These differences are probably due to the age of the plants. Steam distillation of the latex showed this acidity to be fixed. A sample of latex, collectod from several )ilants, gave the following figures on analysis : — Algidity, as lU'ctlt; ncid O'l.'i trnnniiic ])rr liio c.i'. Total solid matter 13-0 \>rr cont. Ash ... ... ... ... ... ... ... 2'4 pur cent. The ash consisted chi(^fly of magnesium, calcium, sulphates, and phosphates, and was distinctly alkaline. The magnesium was present in far larger quantity than the calcium. Exhaustion of the latex, by Adam's fat-extraction method, with alcohol, petroleum- ether, and acetone in succession, yielded a considerable extract to the alcohol, and comparatively little to petroleum-ether and acetone. Tlic petroleum-ether extract was varnish-like, almost colourless, and acrid in taste. The alcoholic extract was only PRELIMINARY NOTES ON THE CHEMISTRY OF THE LATEX OF CALOTROI'IS PROC'ERA 91 partly soluble in cold water, l)ut soluble in liot. This solution gives a deep blue colour with the diluted sulphuviir aciil reagent. If water be added to the alcoholic solution previous to evaporation of the alcohol, a bright yellow, acrid resin is obtained. On reducing the coagulated latex with zinc dust and acetic acid, a liquid is obtained, which, after removal of the zinc by sulphuretted hydrogen, gives no coloration with sulphuric acid, unless previously reoxidised. Concentration of this liquid leaves an amorphous brownish residue, and a considerable crop of silky needle-like crystals of calcium sidphate. Extraction of the licpiid with solvents i-evealed no organic substance of a crystalline nature. The following tables give in sunnuary form Uic results of the experiments carried out to gain some information as to the character of the substance in the latex to which the blue coloration, produced by sulphuric acid and an oxidising agent, is due. From 100 c.c. to 200 c.c. of the latex were taken for each experiment, so that concentrated solutions were obtained. 'l'AI',l,K A .Summary of results Latex Added 4 voUuncs of water .■iiid ;i few drops of acetie .acid, aud heated on the water-batli to eoagulation. Straiaed and filtered. QuTTA-PEncHA, and associated resinous l)od1es, etc. Filtrate, t Added sliKlit i^xeess of neutral lead acetate, and tiltered. Heavy White Pbeoipitate. Washed well with water and decomposed witli dilute sulphuric acid. Solution ° Filtrate (aj.f Added slight excess of basic lead acetate, and tiltered Precipitate. Washed well with water and decomposed with dilute sulphuric acid. Solution ° Filtrate (b).t Added excess of am- monia = bulky yellowish-white ])reeipi- tatc. Filtered Precipitate (d). Washed, and de- composed with dilute sulphuric acid. Solution t Filtrate (e) ° t indli:.ates positive reaction with reafield good results. The average native appears to be of the opinion that a plague of caterpillars or of some other insect is sent by " Eabonah " and that in all probability " Eabouah " will eventually take it away. In the meantime, he hopes that the Government will grant him a remission of his taxes. The Director, Agriculture and Forests Department, has had translated into Arabic and distributed among native cultivators a paper in which are described the life-cycles of three of the more common insect pests of agriculture — viz., the melon fruit-fly — Bacii>< S2)., the melon plant bug — Aspongopus vidiiatus, Fab., and the cotton stem-boring Ijeetle — Sphenoptera neglecfa, King. It is hoped that by this means the native will be induced to take a more intelligent interest in the foes and friends of his crops. Of the grant of £E 700, allotted to be spent in the carrying out of trials to ascertain the value of poisons in the control of locusts in this country, just over £E 500 remains, owing to the non-occurrence of locusts in Berber, the province in which it was intended to make the experiments. This has, in one sense, been very disappointing, as it was hoped to demonstrate the value of the methods employed by the South African Locust Bureau and to induce the Governors of Provinces to adopt these methods throughout the districts in which locusts are in the habit of doing damage. However, as the old adaf'e has it, " everything comes to him who waits," and it is improbable that locusts will prove the exception to the rule. "It is an ill wind that blows nobody good" is another proverb which would not be out of place in this connection. The need of some regulation to prevent the introduction of injurious insects on nursery stock imported from other countries was referred to in my last report. A fumigation chamber has now been erected at Haifa, similar to the one at Port Sudan, and an ordinance enforcing the inspection and fumigation of nursery stock, at the port of entry, has recently been published. A copy of this ordinance is a^jpended. The need for such an ordinance was emphasised by the discovery at Darmali of a scale insect — one of the " mussel " scales, but the specimens were too damaged on arrival at Khartoum for determination to be possible — on a young fig tree, recently imported from Egypt. This tig tree has since died from the attacks of Siuoxylon sudanicum and the scale insect perished with it, but had this boring beetle not been present the scale insect would undoubtedly have spread to the other fig trees growing in the immediate vicinity. Perhaps the more important insects in this country are those which, by sucking ' Mr. King has lately been employed iu mapping out tsetse fly areas in the Lado District of Jlongalla Proviucc — -A.B. REPOKT OF THE ENTOMOLOGICAL SECTION 97 blood or by other means, are responsible for, or aid in, the spread of diseases among man and animals. A number of species of blood-sucking flies, hitherto unrecorded, from the Anglo-Egj-ptian Sudan, has been taken during the past two years, among them being several species new to science, while the distributions of others have been proved to be wider than was originally known. The new species have been, or are being, described by Mr. E. E. Austen, the well known authority on blood-sucking Diptera. As far as possible, all those species which have not previously been figured either in the '^'^"' Eeports of these Laboratories or in Austen's African Blood- sucking Flies, published by fije^ the British Museum (Nat. Hist.) in 1909, are shown in the accompanying plates, as it is thought that an accurate coloured drawing of a fly is of infinitely greater value for purposes of identification — at any rate to the untrained observer — than a detailed technical description. A list is given of all the species of blood-sucking flies — other than mosquitoes — which are recorded from the Anglo-Egyptian Sudan, together with, as far as possible, the localities in which they have been taken. Several species of ticks have also been added to the list of those occurring in this Ticks country, and a synoptic table has been drawn up which it is hoped will be of some aid to collectors. Living specimens of Ornithodoros savignyi have been sent to the London School of Tropical Medicine and to Prof. S. Euge of Kiel for experimental purposes, and a small but excellent collection of cattle and other ticks has been received from Dr. Dschunkowsky of the Zoornabat Anti-Einderpest Station in Transcaucasia. A collection of blood-sucking insects and ticks has also been very kindly presented to tlie laboratories by the Governor of Portuguese East Africa. As many of the inspectors, and other officials stationed in the provinces, expressed their willingness to collect insects if they could be furnished with the means of preserving them, two gross of wooden postal blocks, 7 X I5 X Ij inches, were purchased in 1908. Each of these blocks held two glass tubes, 2:| X ^ inches, in one of which was placed alcohol and in the other entomological pins of various sizes, for pinned dry specimens. Directions for the collection and preservation of blood-sucking insects and ticks were drawn up, and printed in English and Arabic, and a copy of these directions sent with several postal blocks to anyone known to be willing to collect. It was found that while these postal blocks were excellent for the preservation of ticks, fleas, and other specimens in alcohol, the tubes were too small to take the larger blood-sucking flies, so, in consequence of this, a gross of postal blocks, 6J x 2^ x 2^ inches, each containing two tubes, 2x2 inches, Postal blocks was liought this year. These are intended for pinned dry specimens, and both tubes in ^""^ preserving the smaller blocks are now filled with alcohol. To ensure that everyone is asked to forwarding collect, a number of parcels, each containing one large block and two small ones, with '"^'^■^'^ a copy of the directions, has been sent to each of the Governors of the various provinces, with a request that he will distribute them among the officials in his province. Similar parcels have been sent to most of the Medical Ofiicers of the Egyptian Army. It may- give some idea of the spirit which animates those working in this country, when it is stated that no one who has been asked to collect has refused, while many have taken very considerable trouble in obtaining specimens which they conceived to be of interest. Eeports have been issued on some two hundred and twenty-five specimens i-eceived in these postal blocks. In consequence of a suggestion made by the Governor of Mongalla Province, that the fish irimrdinns poscilloides, popularly known as "Millions," which has proved to be of value in the control of mosquitoes in Barbados and elsewhere, should be introduced into the Anglo-Egyptian Sudan, an effort has been made to ascertain whether any of our 98 EEPOET OF THE ENT0MOI,0OI('AL SECTION indigenous fishes would perform similar services. The results obtained from an experiment made with Cyprinodon dispar — a species recommended for trial by the Superintendent of Game Preservation — were very promising, but a second experiment, carried out on a larger scale, has not been so successful. It is, however, yet early to say definitely whether or not this fish can be considered a valuable factor in the control of the members of the Fam. Gulicidse. Control of The Control of injurious birds is perhaps scarcely the work of an entomologist, but in inj^urious ^^^ absence of an official to whom this duty could be assigned, the matter has been given some attention. Almost incredible amounts of grain are lost by cultivators yearly, owing to the ravages of sparrows and other small grain-eating birds. Particularly is this the case in Sennar and some of the other eastern provinces, while, in Dongola Province, the Governor estimates these losses to amount to a quantity equal to one-third of the grain harvested, or over 70,000 ardebs, worth more than £E 70,000. This latter province, situated as it is in an almost rainless region, where grain-eating birds are practically dependent on the cultivations for their food, suggested itself as being best adapted for the carrying-out of an experiment in the control of these small depredators. In June of this year, an estimate was made of the probable cost of such an experiment and a sum of £E 1,255 asked for-, to be spread over two years. It is confidently hoped that this experiment, if carried out, will yield results fully justifying its cost. Mr. A. L. Butler, Superintendent of Game Preservation, having kindly undertaken to identify any birds collected, it is proposed to prosecute some investigations on the food of those species which are suspected of being either beneficial or injurious.* During the winter 1908-9, the sum of £E 35 was voted by the committee of the Sudan Government Museum to be spent on an entomological show-case. A glass-fronted mahogany cabinet containing forty drawers was purchased, and an exhibit jirepared of blood-sucking insects and ticks, and insects injurious to agriculture and to stored goods. Owing to lack of funds, an application for an untrained assistant could not be granted, but the Director of Laboratories has allowed Mr. Marshall, an assistant in the bacteriological section, to give me one hour of his time every day. Mr. Marshall has been of help in cataloguing bulletins and journals and in attending to living occupants of breeding- cages, when I have been out of Khartoum. The Bureau of The Bureau of Entomological Eesearch (Tropical Africa) has been of very great Reserrch°^"^^' assistance in the determination of insects and ticks. Two papers dealing with the bionomics of six species of blood-sucking flies (Fam. Tahanidm) have been published in the bulletin issued by the committee. Mr. Austen has been good enough to contribute some notes on three new species of Tahanidm, and descriptions of several recently described species will be found in this report. Plate VII. is reproduced by kind permission of the Editor of the Journal of Economic Biologi/, and Plates III. and V. from the Bulletin of Entomological Rpsearch. The figures on Plates VIII. and IX. have been drawn by Miss C. M. Beard, with her usual care and skill, while Plates I., II., IV. and VI. aie by the well-known artist Mr. A. J. Engel Terzi. In conclusion I wish to express my indebtedness to those wlio have helped me in my work. Mr. J. Currie, Director of Education in the Sudan, and Dr. A. Balfour, Director of the Wellcome Tropical Eesearch Laboratories, have always given me every • Mr. Butler has recently counselled delay until we are in possession of more complete information regarding the species of birds chiefly responsible for the damage. Prom the information he supplies this would certainly seem to be advisable, but Mr. King's note has served to direct attention to this important question. — A.B. REPORT OF THE ENTOMOIvOCirAL SECTION 99 assistance in their power, and I have benefited greatly from their advice. I am similarly indebted to my colleague, Dr. W. Beam. My thanks are due to Mr. F. V. Theobald, Vice-Principal of the South-Eastern Agricultural College, Wye, Kent, for the determination of mosquitoes ; to Dr. G. H. F. Nuttall and Mr. C. Warburton for the determination of ticks; and to Mr. E. E. Austen, of the British Museum (Nat. Hist.), for the determination of blood-sucking flies and for help in connection with the drawdngs illustrating this report. Others to whom I would express my gratitude are Mr. G. A. K. Marshall, Scientific Secretary to the Entomological Research Committee (Trop. Africa), Sir G. F. Hampson and Mr. G. A. Boulenger of the British Museum (Nat. Hist.). The names of those in the Sudan who have rendered valuable aid by collecting and forwarding specimens are far too numerous to give here, but I cannot conclude without mentioning in this respect Captain Hills, A.S.C., Mr. Landon, Sudan Irrigation Service, Major Percival, Captains Mackenzie, Cummins, Drew and Anderson, R.A.M.C., and Mr. H. A. MacMichael, Junior Inspector, Kordofan. To these, and to many others, I would here express my sincere gratitude. Anim.\ls In.iurious to Man and Animals Mosquitoes Gulicidse Among the mosquitoes collected during three months spent on the White Nile in 1909 were representatives of five new species and a new sub-species. One of the new species has been constituted the type of a new genus. Descriptions of all these appeared in Volume V. of Mr. F. V. Theobald's Monograph of the Culicidie of the World, published by the British Museum (Nat. Hist.), but for the convenience of those in the Sudan who do not possess a copy of that work, these descriptions are given here. Other species taken during the last two years which had not previously been recorded from this country are Grabhama wUlcncJcsii, Theob., found at Zeidab and Kingia Inteocephala, Newstead, bred from larvas taken from a hole in a tree near Bor, Mongalla Province. Qiiasit:teyom>jia dnhiii, Theobald Monn. Callcil. V., p. 13^! (1910) Head black with a median snow-white area and a small white patch on each side ; Qu„sisiego- proboscis black ; palpi nearly as long as the proboscis, acuminate, no hair tufts, black, the "V" «'«*'o two last segments with basal snowy spots, a broad median white band and a narrow basal one ; antennse with deep brown plumes. Thorax black, with a median patch of flat silvery-white scales in front near head, a large patch of snow-white large broad curved scales on each side ; a smaller patch in front of the wings, a patch of flat w-hite scales l)ehind on each side of the bare space in front of the scutellum ; scutellnm witli flat, silvery-white scales. Abdomen black, with basal white bauds. Legs black, mid femora witli a silvery spot at apex and on apical half; metatarsi of fore and mid legs, and first tarsal with basal white bands, hind femora also with spots, and a white basal band on second tarsal. t Head clothed with flat black scales over most of its surface, two rows of flat silvery-white median scales, narrow dark line between, flat white scales at the sides, dusky upright forked scales at the back, black chsetae projecting forwards and inwards over the eyes; clypeus and proboscis black, basal lobe of antennae black, with flat silvery-wliite scales on the inside, joints banded black and brow-n, plume hairs deep blackish-brown ; 100 BEPOBT OF THE ENTOMOLOGICAL SECTION palpi nearly as long as the proboscis, bluntly acuminate, black, the two apical segments nearly equal, each with a basal snow-white spot, both with a few long black chaetae, and some on the apes of the antepenultimate segment, which has a broad snow-white band near middle, and a narrow one basally. Thorax black, clothed with narrow-curved bronzy-black scales, in front a large snow- white median spot of flat scales, on each side behind a large snow-white patch of broad curved scales, and a small one just at the roots of the wings ; surrounding the bare space in front of the scutellum are flat silvery-white scales, with some broad curved ones in front showing mauve reflections ; chaetae thick, black ; scutellum black with flat snow-white scales ; metanotum black ; pleurae black, with four patches of flat silvery-white scales. Abdomen black, with basal silvery-white bands, the last two segments with lateral white patches. Legs black, ornamented as follows : — Fore femora with a few white scales, mid witli a snow-white apical spot, and another on the shaft with one or two white scales more basally, hind femora white ventrally on the basal two-thirds, then a silvery spot and another at the apex ; tihiie all black ; metatarsi with a basal white band, broadest and whitest on the hind pair, fore and mid first tarsals with narrow white band, and in the hind legs the second tarsal also ; fore and mid ungues unequal, the larger simple, the smaller uniserrate ; hind equal and simple. Wings with broad scales on the sub-costal and first long vein, twniorhynchus-Wke ones on the second, thinner lateral ones on the other veins ; fork-cells short, the first longer and narrower than the second, their bases about level, stem of the first more than half the length of the cell ; stem of the second nearly as long as the cell ; posterior cross-vein longer than the mid, about twice its own length distant from it. Genitalia with rather small, narrow claspers, with dark blunt apical spine. Length. — d'S to 5 mm. Habitat. — Bor, Sudan (H. H. King). Time of Capture.— June 8, 1909. Genus Genus Kingia, Theobald fCifina Mono. Cnlicid. V., p. 13.5 (1910) Head clothed with flat scales and upright forked scales, with traces of a few narrow- curved ones behind. Palpi of $ short. Thorax v.-ith narrow-curved scales to the mesonotum and with flat scales in the anterior median line and a large lateral patch on each side of similar flat scales and others at the sides ; scutellum with flat scales. Venter of abdomen slightly tufted. Wings very densely scaled with long txniorhynchuf-Yike lateral scales, the scales on the first, second and third veins all overlapping. This genus is very near Quasistegomyia, Theobald, but can be told at once by the flat white lateral scales. A", /uieo- Kingia liiteocephala, Newstead (1907) Stegomyia luteocephala, Newstead (1907) Annals Tropical Mrdirinc and Paraxiloloyy I., No. 1, I.t U'*"7) " Head yellow. Palpi black with white tips. Thorax brown, with two large, anterior, lateral silvery spots, a median yellow stripe and posterior lateral yellow spots ; scutellum white. Abdomen black with pale narrow bands, terminal segments silvery. Legs black with silvery spots and white banded tarsi. KEPORT OF THE ENTOJIOLOGIOAL SECTION 101 " 1^ Head with large central area thickly clothed with large, loose, flat yellow scales, gradually merging into smoky-yellow in front ; a narrow silvery-white, line to the anterior half of the eyes, formed of a single series of broad, flat, closely appressed scales ; between the marginal line and the central yellow patch is a broad band of brownish-black scales from which anteriorly arise several upright forked scales ; nape with a few long, thick, straight or slightly curved pale golden scales, on either side of w-hich is a group of upright forked scales, intermixed black and yellow ; lower basal portion with flat dusky- white scales, the marginal ones forming two dull silvery spots, sharply divided by a dense black spot. Antennae black, nodes white ; hairs black, pubescence grey. "Palpi black; tips with long silvery-white scales. Thorax: prothoracic lobes with flat silvery-white scales ; mesothorax with a well-defined median line and two lateral spots of narrow-curved, golden-yellow scales ; anteriorly there are also two large spots of flat silvery-white scales, and a few silvery-white scales on the lower margin of the posterior yellow spots ; the rest of the mesothorax with rich dark brown scales ; scutellum with flat silvery-white scales ; pleurae dark brown with two large patches of silvery scales. "Abdomen rich bronzy-brown; segments 1 to 6 each with a well-defined, narrow basal band of smoky-yellow scales ; penultimate segment with a large lateral patch and the terminal segment almost covered with brilliant metallic-silvery scales ; venter with well-defined, more or less triangular patches of metallic-silvery scales narrowing towards the apex, where they appear as two divergent lines ; the scales, forming the outer lateral angles of the spots projecting at the sides of the abdomen, appearing as ouUtandinq scales. "Legs bronzy-blackish-brown; coxte and trochanters, ochreous ; anterior and mid femora with scattered metallic-silvery scales ; hind femora with a central anterior band and an apical group of silvery scales ; anterior and mid tarsi with narrow dull-white basal bands to the first three segments, metatarsal band broadest ; hind tarsi with a broad white basal band to the first, a narrow one to the second, and the third segment almost entirely white above, basally it is not so. " Wings uniformly pale brown, rather densely scaled, first sub-marginal cell nuich longer and slightly narrower than the second posterior." The wings show the following : — First fork-cell much longer and narrower than the second fork-cell, its base nearer the base of the wing, its stem a little more than one-third the length of the cell ; stem of the second fork-cell not quite as long as the cell, posterior cross-vein longer than the mid, not quite twice its own length distant from it. J* Head with rather loose ochreous yellow scales becoming dusky and then black except in the middle line, a patch of silvery-white scales on each side at the eye border, numerous ochraceous forked scales along the nape becoming dusky in the middle. Antennae black, with dull grey bands, dense black plumes, showing grey reflections at the tip ; proboscis long, thin and black ; palpi as long as the proboscis, thin, black, a very narrow white baud at the base of the apical and penultimate segments, a broad one in the middle of the antepenultimate, last two segments nearly equal, a few black hairs on each side of the penultimate, fewer on the apical segment, and a few on the apex of the antepenultimate. Thorax as in ^, but the median yellow line and the lateral yellow-scaled spots more pronounced. Abdomen unhanded, the segments with prominent silvery-white basal lateral spots ; hairs black with golden sheen ; claspers apparently simple, basal lobes of genitalia long and narrow. Fore and mid ungues unequal, simple ; hind equal and simple. 102 KEPOKT OF THE ENTOMOLOGICAL SECTION Wings with short fork-cells, the first longer and narrower than the second, its base if anything a little nearer the base of the wing, its stem a little less than half the length of the cell ; stem of the second fork-cell about two-thirds the length of the cell ; posterior cross-vein about two and half times its own length distant from the mid ; wing scales mostly rounded apically. Length, 4'5 to 5 mm. Hahitat. — -Kimba, Congo Free State (Newstead) ; Sudan (H. H. King) ; Mpuma, Uganda (Sir David Bruce). Stegomyia Stegomyla ijebeleinensis, Theobald gebeUinensis ■ Mono. CiilicU. V., p. Ibl (1910) Head as in Stegomyia Scutellaria:, Walker. Thorax rich brown with a rather broad median white line in front, and where it terminates, a small white spot on each side, pleurae white scaled, scutellum white scaled. Abdomen black with narrow basal silvery-white bands on the dorsum and venter. Legs black with narrow basal white bands on some of the joints. 1^ Head as in Stegomyia scutellaris. Walker. Thorax black, clothed with small narrow- curved bronzy-brown scales, a rather broad median white scaled line, running from the front to about the middle, narrowing posteriorly, where it ends is a small white scaled spot on each side of similar scales, and just anterior to these run two indistinct paler scaled lines back to the scutellum, paler scales also before the scutellum, which is covered with flat white scales and has black border-bristles ; metanotuin dark brown ; pleurae dark with dense rather loose flat white scales, especially above ; prothoracie lobes with flat white scales. Legs damaged ; the femora white scaled at the base ; the mid pair with a narrow pale basal band to the metatarsi and first tarsal. Wings with typical stegomyia scales ; fork-cells short, the first longer and slightly narrower than the second, their bases nearly level, stem of the first nearly as long as the cell ; stem of the second as long as the cell ; posterior cross-vein not quite three times its own length distant from the mid. Abdomen black, with narrow basal snow-white bands which spread out laterally, basal segment all dark-scaled ; border-bristles pale golden and brown ; venter dark with basal white bands. Length, 35 mm. Habitat. — -Gebelein, Sudan (H. H. King). stemmvia Stegomyia lilii, Theobald Mono. CiiUcUl. v., p. 160 (I'JIO) Head black with snow-white median area and white at the sides ; palpi black in $ with snow-white apices ; proboscis black. Thorax black with a large lateral snow-white spot in front, and a smaller one behind before roots of wing ; two narrow median yellow and two sub-median white lines run from the middle of mesonotum to the scutellum ; scutellum white ; pleurae dark with white puncta. Abdomen black with basal silvery-white bands, last segment with two white s2Dots, no band. Legs black with basal white bands. UKl'Oli'l' l)K THK KXTOMOLOGICAL SKCTION 103 !^ Head clothed with flat Ijlack scalos, a double row of Hat snow-wliite ones in the middle and white at the sides, a narrow white border due to reflection around the eyes, thick black chiBtte project over the eyes inwards ; palpi black with snow-white apices ; proboscis and antennae black. Thorax black, clothed with small narrow-curved black scales, in front on each side a large silvery-white triangular patch of broader scales, behind this a smaller patcli just in front of the roots of the w-ings ; from the middle of the mesothorax run backwards two parallel thin yellow lines and two sub-median snow-white ones, the latter quite reaching the scutellura ; chaetae jet black ; soutellum clothed with flat silvery-white scales, dusky in some lights ; metanotum brown ; pleurae black with three prominent silvery-white spots of flat scales. Abdomen black, the second segment with trace of white basal band, the third, fourth, fifth and sixth with prominent silvery-white basal bands, the seventh with two white basal spots, eighth dark, also the basal segment ; chiBtae and border-bristles black ; venter with broad white basal bands. Legs, black, the fore and mid metatarsals and first tarsals with Inisal white bands, rest of tarsals black ; in the hind legs the femora are pale on the basal half, the black apical half with two silvery-white spots, one apical, the metatarsi and first and second tarsals with basal snow-white bands, the third all dark, the fourth white. Wings with dark scales, especially along the outer costal border; fork-cells rather short, the first longer and slightly narrower than the second, its base slightly nearer the base of the wing, its stem about half the length of the cell, stem of the second posterior not quite as long as the cell ; posterior cross-vein about the same length as the mid, about its own length distant from it. Length, 4:'5 mm. ,^ Head as in the !(? ; palpi as long as the proboscis, blunt, no hair tufts, a few black bristles, black, a ventral white area at the base of the apical segment, a broad median white band and a narrow basal one ; antennte banded black and white with black plume hairs. Thorax as in the 5^, but the median thin yellow lines curve around the bare space in front of the scutellum and reach it, the last part being composed of broader curved scales.* Abdomen and legs as in the !^ ; fore and mid ungues unequal and simple ; hind equal and simple. Length, 4-5 mm. Hahitat.^Boi (H. H. King). Time of capture.— May 26, 1909. Genus Mispidimyia, Theobald Genus Hispidim via .Uuiio. Citlkid. v., 1). 245 (1910) Head clothed with flat scales, except for a small group of minute luirrow-curved ones on the nape, and a larger group of upright forked scales. Proboscis swollen on the apical third, composed of two segments, the joint at the apical third ; labella acuminate. Antennae of $ densely plumose with two very long terminal segments pilose only ; other segments thick, basal one globular ; palpi of male a little longer than the proboscis, clavate at the apex ; of two segments the apical one short and clavate ? with ventral chaetae and two large and some small apical ones. Palpi of $ about one-fourth the length of the • This probably also occurs iu the 9 as there are traces of these larger scales.— F.V.T. lOi KEl'OKT OK THE ENTOMOLOGICAL SECTION proboscis. Prothoracic lobes with large curved spines in ?, smaller in If! ; mesothorax and scutellinn with narrow-curved scales, the whole thorax with long dense-curved backwardly projecting chaetiB. Abdomen densely hairy in the $, especially ventrally, also hairy in !^. Wings with only median vein scales, spatulate in the $, some lateral broad short scales on the second vein, etc., iu the $. Fork-cells short in both sexes, the first with its base nearer the apex of the wing than the second. Hispidimyia HUpidimnia hispida, Theobald Mi>m,. ridicid. v., p. 245 (191U) Head brilliant creamy-white ; proboscis ochreous, dark at apex ; palpi of ^ clavate, a little longer than proboscis, ochreous, dark brown on clavate area, which is spiny below, apex spiny ; antennae flaxen-brown. Thorax rich brown with two median pale 23arallel shiny lines, which seem to continue around the bare space before the scutellum, a darkened area on each side of the parallel lines in front ; pleurae pale ochreous. Abdomen deep brown with basal lateral yellow spots which form almost a continuous line on each side, densely hairy. Legs unhanded, deep brown, except base and under side of femora, with brassy reflections ;' tibiae spinose. Wings with scanty brown scales. $ Head clothed with flat creamy-white scales with ochreous and iridescent reflections, a small area near the nape of narrow-curved golden scales, a larger area of black upright forked scales and some almost golden along the nape ; four long black chaetae on each side on the ocular border projecting inwards, a few small short golden chiEtae in front between the eyes and two long black chaetae ; there is a median parting of the flat scales. Clypeus apparently very small ; proboscis ochreous with some dark scales above, dark at the apex, which is slightly swollen ; palpi a little longer than the proboscis, clavate and blunt at the apices, ochreous below, some dark scales above, clavate area black with violet reflections ; clavate area with black chaetae below, two large black apical chaeta and some smaller ones. Antennae with the segments thick and short, except the last two, which are long, thin and pilose, rest with dense flaxen-brown plume hairs ; basal segment very large, globular and dark. Thorax rich ochreous brown, clothed with black narrow-curved scales, showing bronzy- metallic reflections, leaving two median bare parallel lines ; bare space in front of the scutellum, pale, especially at the sides, the area looking like a continuation of the two bare pale median lines ; numerous black, backwardly curved chaetae, a patch of small golden ones before the base of the wing ; scutellum shiny, with pale reflections like the bare space before it, clothed with narrow-curved bronzy scales ; six posterior border-bristles, three on each side of the mid lobe ; metanotum ochreous brown, darker in the middle ; prothoracic lobes ochreous with numerous long curved brown chietae, with bright golden reflections, pleura3 pale ochreous, with grey reflections, some dusky and white semi-transparent flat scales. Abdomen clothed with black scales with dull violet reflections, each segment with large basal lateral yellowish spots, which on some of the segments extend to their apical borders, last segment with many ochreous scales ; posterior border-bristles golden, dense lateral and apical bristles brown, with golden reflections ; venter densely hairy. Legs blackish-brown, with dull violet reflections ; femora pale-creamy at the base and below, fore femora spinose and all the tibiae, especially of the hind legs ; fore ungues unequal, the larger biserrate, the smaller simple. EEPOK'l' OF 'I'HK KNTOIIOLOGICAL SECTION 105 Wings witli short I'ork-Cflls ; the first a little longer and narrower tlian the second, its base nearer the apex of tlie wing, its stem longer than the cell, stem of the second fork-cell longer than the cell ; mid-cross vein longer than the supernumerai-y, posterior as long as the mid, curved, about half its own length distant from it ; the second to the sixth veins with only a single row of median vein scales, no lateral scales. Genitalia with small narrow basal lobes; claspers as long as the lobes, simple, with a dark terminal spine. Length, 5 mm. $ Head very similar to the f, but the scales not quite so pale and the few scanty dark fork-scales spreading more over the occiput ; three black chaHse only on each side on the ocular line ; proboscis ochreous in the middle, dark at base and apex, with numerous short pale hairs; palpi about one-fourth the length of the proboscis, brown, with short fine hairs. Antennae brown, pilose. Thorax black with narrow-curved bronzy-black scales ; paler behind almost ochreous, but dark brown through the bare space before the scutellum, with dull golden curved scales around it and dark ones at the sides, the golden scales broader than the dark ; chsette long, black in the middle and in front, golden over the roots of the wings ; scutellum ochreous, darker on mid lobe, scales narrow-curved (?) ; six golden- brow-n posterior border-bristles, three on each side of the mid lobe ; metanotum ochreous dark in the middle ; pleurae bright ochreous with grey sheen and pale hairs, pale golden ones also between the dark mesonotum and pale pleur®. Abdomen black with violet reflections and basal lateral yellow spots and many yellow- scales on the apical segment, posterior border-bristles pale golden ; venter pale, dull ochreous. Legs as in the J, but the femora paler, almost ochreous in the hind legs, fore femora spinose and all the tibiie ; ungues equal and simple. Wings with median vein scales only to most of the veins, if lateral ones of pyriform shape ; fork-cells small, the first about the same length as the second and narrower, its base near the apex of the wing, its stem a little longer than the cell ; stem of the second about the same length as the cell ; mid cross-vein longer than the supernumerary, posterior about the same length as the mid, not quite its own length distant from it. Length, 4-5 mm. ifafeiiaf.— Near Gebel Ahmed Aga and Bor, Sudan (H. H. King) Time of capture.— Ma.}' 2 and 20 and June 20, 1909 (triihhama willcocksii, Theobald Grabhamn wilkockui Mum. Cidicid. IV., p. 296 (1907) Thorax clothed with bright golden-brown and ochreous or creamy scales; in the middle there are two lines of a pale creamy hue ; pleurae pale with grey scales. Abdomen mostly pale scaled, but with two more or less distinct dark areas on each segment. Legs with apical and basal pale bands; all ungues are equal and uniserrate in the $. Wings mostly pale scaled, a few scattered dusky ones, but variable. Ungues of male, fore and mid unequal and uniserrate ; hind equal and simple. $ Head deep brown, clothed witli pale creamy narrow-curved scales and pale creamy upright forked scales on the occiput, becoming bright ochreous laterally, and a few deep black ones behind on each side, flat creamy and dusky lateral scales. Palpi mottled 106 REPORT OF THE ENTOMOLOGICAL, SECTION with dark and creamy scales ; proboscis mostly ochreous scaled, witli a few scattered dark scales, the apical region black ; antennis brown. Thorax black, densely clothed with bright golden-brown narrow-curved scales, except for tw'o creamy lines running nearly the whole length of the thorax and widest in front ; scales in front of the scutellum much paler; traces of brighter scaled lines also laterally in some specimens ; scutellum with pale creamy scales ; chsette bright reddish to golden- brown ; metanotum bright brown ; pleurie brown with flat white scales. Abdomen brown, mostly clothed with ochreous and white scales, the latter form a median line and a large patch either median or basal laterally, the ochreous ones more confined to the apical borders of the segments, the dark scales form more or less marked dorsal lateral dark areas, but pale scales may be dotted about over these ; basal segment with two tufts of flat white scales ; venter mostly creamy scaled. Legs with the femora ochreous, with a few scattered black scales ; femora dark brown with scattered white scales above, mainly ochreous with a few dusky scales below ; first tarsals with more dark scales, in the fore and mid legs there is a jjale band involving both sides of the first and second tarsals, and another involving the joint between the second and third tarsals, rest dark scaled ; in the hind legs the bands are wider and extend to all the joints, the last tarsal being pure white ; ungues on all the legs equal and uniserrate. Wings with the majority of the scales pale creamy, some bright ochreous ones on the costa and first long vein and some scattered black ones here and there on the other veins ; fork-cells short, the first sub-marginal longer and narrower than the second posterior cell, its base nearer the apex of the wing than that of the second posterior cell, its stem about three-fourths the length of the cell ; stem of the second posterior about two-thirds the length of the cell ; mid cross-vein longer than the supernumerary or posterior, the latter about its own length distant from the mid ; halteres with slightly fuscous apex clothed with grey scales and a dark line on the ochreous stem. Length, 3 to 4 mm. $ Palpi straw-coloured, a narrow dusky band at the apex of the penultimate and antepenultimate segments and some dusky scales at the apex, apical segment shorter than the penultimate ; hair-tufts moderate, flaxen. Antennae with flaxen hairs. Abdomen with basal grey bands and most of the apical segments grey scaled. Genitalia with rather long, narrow basal lobes, long claspers curved apically with long terminal spine ; harpes prominent, curved, broadened at the middle. Ungues of fore and mid legs unequal, uniserrate ; hind ungues equal and simple. Length, 4-5 to 5 mm. Time of capture. — -June. Hahitat. — Kafr-el-Dawar, Egypt (F. Wilcocks) ; Zeidab, Sudan (H. H. King). Ciilex ti^iripes Cidex tigripes, Grandpre, var. bimacnlatn, Theobald M07io. Citlicid. v., p. 393 (1910) Legs with marked lines of creamy spots as in type. Thorax with two pronounced median small pale spots, a pale scaled line from each, running backwards, also a pale scaled median spot near the head, an indistinct one on each side and some pale scales over the wings. Abdomen with the second to fifth segments dark with narrow yellow apical bands, the second to fourth each with two nearly median yellow spots ; remaining apical area golden-yellow. ffafci'fa^— Meshra-el-Zeraf (H. H. King). var. iiiinatJuiat* liKrOHr OK THE ENTOMOLOGICAL SECTION 107 Uranotsenia alboabdominadi.!' , Theobald .yrono. Ciilkid. v., p. 508 (1910) Head pale blue ; palpi and proboscis black. Thorax rich bright brown, a silvery-white Uraiwi,tnia line running up to base of wing for about half the length of the thorax, and another „i,j^,„,i„„iit parallel one on the pleurte, which are ochreous. Abdomen with the first four segments almost entirely creamy-white scaled, rest dark with basal median creamy-white patches. Legs dark brown, unhanded. ^ Head clothed with pale blue and some scattered creamy flat scales, deep violet in the middle at the back, dusky at the extreme sides, with four black upright forked scales and a large tuft of long creamy scales projecting between the eyes ; few chaetrn, black ; palpi and proboscis black ; antennae brown, basal segment bright testaceous. Thorax bright brown, clothed with scanty dai-k and dull golden narrow-curved scales and a white line of spindle-shaped scales running from the base of the wings to about two-thirds the length of the mesonotum forwards, chsetae deep brown ; scutellum clothed with small flat dusky scales ; metauotum deep brov,-n ; pleura pale ochreous with greyish areas, a broken line of flat white scales running from the prothoracic lobes, parallel with the line to the roots of the wings ; both with very pale blue reflections ; prothoracic lobes covered with very pale blue, almost white flat scales. Abdomen with the 2nd to 5th segments almost entirely creamy-white scaled, a few brown ones at their bases ; 6th, 7th and 8th segments with median creamy basal scales, not forming bands ; basal segment very small brown ; border-bristles pale golden. Legs uniformly deep brown with bronzy reflections ; ungues small, equal and simple. Wings with brown scales, but with a white line at the base continuous with the white thoracic line ; scales on ends of veins large and lanceolate, on most of third long vein, pale compared to median vein scales ; first fork-cell shorter and slightly narrower than the second, its base much nearer the apex of the wing, its stem about twice as long as the cell ; stem of the second fork-cell about one and a quarter as long as the cell ; posterior cross- vein longer than the mid, about one and a-third times its own length distant ; apex of second long vein close to the first. Length, 3 mm. ^ Head more white scaled than the ^, with more dark upright forked scales in the middle ; antennas with pale internodes, plume hairs brown. Thorax and abdomen as in '^. Fore and mid ungues unequal, curved, simple; Iriud equal and simple. Wing very much like the ^. Length, 3 mm. Habitat. — Bor to Mongalla, Sudan (H. H. King) Time of capture. —M-dy 19 and 20, 1909. Uranotsenia abnonnalin, Theobald Uranottcnia ahnormalis .]f,„„.. Cnticid. v., p. 513 (1910) $ Head as in the '^ ; antenna; brown with dusky plume hairs, pale at the tips ; proboscis and palpi black. Thorax as in the ^, also the abdomen. Legs unhanded ; fore legs with short thick metatarsi nearly one-half the length of the first tarsal, last tarsal very small ; tibia with claw-like apical spine, ungues equal and simple, curved almost at a right angle, broad, spines above ; the segments following 108 HEPOKT OF THP: ENTOMOLOGICAL SECTION metatarsi, thin and bent ; mid legs with short metatarsi, second tarsal normal, third bent under the fourth and projecting with two dark unequal claws, like ungues and with two lateral leaf-like processes on each side of the claw's, ungues on terminal segment equal and simple ; hind legs with the tibiae with five basal bristles in a row, contracted at one-third their length, where there is a tuft of five long orange to brown spines of nearly equal length, three together bent apically and also a sixth much longer spine, near the middle of tibia is a single long spine ; metatarsi and tarsi thin, ungues equal and simple. Wings much as in the ^ ; the first fork-cell much smaller than the second, its base much nearer the apex of the wing, its stem nearly three times the length of the cell, stem of second fork-cell about one-and-a-half times the length of the cell ; posterior cross-vein longer than the mid, about one-and-a-half times its own length distant ; second long vein close to first. Genitalia small and hidden ; claspers thick and blunt, basal lobes triangular. Length, 3 mm. Habitat. — Sudan, Bor to Mongalla, 21 specimens including 3 J''s (H. H. King). The female of this species was described by Mr. Theobald in the Third Report of these Laboratories under the name of Uranotienia paltidocephala sub. sp. cieruleus, but later, on examining the male, he raised it to the rank of a species. The Conteol of Mosquitoes Pump irriga- tion more dangerous than sakias The control of mosquitoes on artificially irrigated farms is a matter of considerable importance, bearing as it does directly on the subject of the health of those living in the neighbourhood. Without mosquitoes the northern provinces are — except for the heat — healthy, but if anophelines are allowed to exist in any numbers, malaria at once becomes rife. The ordinary native-owned cultivation is dependent for its water-supply on a sakia — a cumbrous, wooden water-wheel, turned by cattle-— and very few mosquitoes breed out on these sakia-lands. The volume of water raised is so small that the owner cannot afford to allow any of it to go to waste. The floors of the gudwals — as the water-channels are termed— are usually above the level of the surrounding land, and consequently when the sakia ceases to turn, the gudwals run dry. The only breeding-places provided for mosquitoes under such conditions are occasional puddles among the grass growing on the sides of the gudwals, due to unsuspected leaks. The water used in irrigating a farm of from a few hundred to several thousand feddans is raised by means of pumps, worked by engines, and in this case a comparatively large volume of water has to be dealt with. The water is conveyed from the pumps by a main canal from which open smaller canals. From these canals the water passes along large gudwals, and from them by smaller ones directly on to the land. Each canal and gudwal is, roughly, at right angles to those with which it connects. There may be many canals and gudwals of various sizes, and when a heavy weight of water has to be supported by earthen banks, leakages are bound to occur. The floors of the gudwals — which are always much larger than any used on sakia-land — are generally below rather than above the level of the surrounding land, and these water channels cannot therefore be emptied. Consequently, when the gudwal is not in use, the water in it remains stagnant until it either soaks away and evaporates — a process which may occupy several weeks — or the gudwal is used again. In some localities, owing to the chemical composition of the soil, water is continually soaking through the canal and gudwal banks to form little pools on the other side. A bank of this nature is known as a " weeping" bank. The canals invariably contain numbers of small fish, and, moreover, the water in them is usually moving, so there EEPOET OF THE ENTOMOLOGICAL SECTION 109 is not much cliaiiee for mosquitoes to breed out in them, but the gudwals, and especially the smaller ones, may not be used for weeks at a time, and a more ideal nursery for mosquitoes than stagnant water in an overgrown gudwal can scarcely be imagined. Each gudwal terminates in a " dead " or " blind " end, that is to say, there is no opening from a gudwal " Weeping" at its extreme end. As a result of this, the water in these " blind " ends is always more or ..^"l.^f"' , ' •' "blind ends less stagnant, and generally bears a covering of green slime. These "blind" ends are necessary to prevent the flow of the water from wearing away the banks. Anyone who has had experience in the control of mosquitoes in tropical or sub-tropical countries will be able to understand wliat a source of danger a farm of this kind can be to the inhabitants of the neighbourhood if suitable precautions are not taken. In one instance I attributed all the mosquitoes, which had caused a serious outbreak of malaria in a native village, to a single " blind " end. It is generally thought that mosquitoes cannot be controlled on a large artificially irrigated farm except at a very considerable — and frequently in the owner's opinion, an unnecessary — expense. If the matter is gone into carefully, however, it will be found that not only can the mosquitoes be kept in check at a comparatively small cost, but also that it will pay the owner to do so. It must be remembered that when from one to two-thirds of the staff of labourers employed on a farm are suffering from intermittent attacks of fever, they are absent from work for varying periods during the month, and are also often physically unfit for hard and protracted toil during the times when they are supposed to be working. The life-cycles of the species of mosquitoes found in the northern provinces occupy about ten days, consequently if all stagnant water on any given area is examined, and, if found to contain mosquito larvae, treated with paraffin or some other larvicide at intervals of a week, no mosquitoes can breed out in that area. It is not always easy to detect mosquito larvas in stagnant water bearing a covering of slime, nor is it possible to approach many of the pools in a gudwal owing to the softness of the mud surrounding them. For Mosquito practical purposes, therefore, it is better to assume that all stagnant water requires to P'''='^*""°ns "°' ■c^ Jr r ' » o X necessarily be treated with parafiin once every week — a great saving of time will be thus effected, expensive A man armed with a stick with a bundle of rag tied at one end, and accompanied by a donkey bearing a tin of pai-aflfin, can walk a long way in a week, paraffining all the stagnant pools he finds on his way, and an intelligent native at a monthly wage of from £E3-4, w'ith a little supervision from the British overseers on the farm, can be trusted to carry out this work. The upkeep of the gudwals is, of course, the duty of the owner — or the tenant, if the land is let out to tenants — and if this is well done, the work of the " mosquito men " will be very considerably lightened. It should be the duty of these men to report to their employer the condition of any gudwal which is either leaky or overgrow-n. Paraffin is the larvicide most generally in use against mosquitoes, for it is effective, moderately cheap, and obtainable almost everywhere. A sufficient quantity should be applied to form a film over the entire surface of the water. Another larvicide which has i.arvicides been used with great success by Mr. J. A. le Prince, Chief Sanitary Inspector, Panama Canal Zone,' is made from carbolic acid, resin, and caustic soda prepared in the following way : — "Crude carbolic acid containing about 15 per cent, phenol is heated to 212' F., finely pulverised resin is added, and the mixture kept boiling until the resin is all dissolved. Caustic soda solution is then added and the mixture kept at 212 F. for about ten minutes, or until a perfectly dark emulsion without sediment is obtained. The mixture is thoroughly ' Ros.s, R. 11910), 77«; Prevenliun of Malaria, p. 357 tion IS not dangerous 110 REPORT OF THE ENTOMOLOGICAL SECTION stirred from the time the resiii was added imtil the end." Mr. le Prince states that as the composition of crude carbolic acid varies greatly, the proportion of ingredients of the larvicide will vary, and it is necessary to have small experimental lots made in the laboratory and tested before the batch of larvicide for use in the field is manufactured. One part of this mixture placed in five thousand parts of water containing mosquito larvae is said to kill them all within five minutes. If it is used in the proportion of one to eight thousand the larvae are killed in thirty minutes. Basin irriga- A method of irrigation, practised in some districts, notably in the province of Dongola, is that known as " basin irrigation." Where the land-levels admit of its being done, a canal is dug, along which water flows from the river when it is in flood on to land situated generally behind the sakia-lands. There is, I think, little fear of mosquitoes breeding out in the pools left by these basins when the Nile falls, as the myriads of fish, which are brought in from the river with the water, are constrained by hunger to devour any mosquito larvae they can find, even though under more normal conditions, such larvae would not be very palatable to them. During some three months spent on the White Nile in 1909, I endeavoured to find a species of fish which would emulate the example set by Girardinus pxcilloides. This Barbados fish has gained for itself a world-wide reputation as a mosquito- destroyer, under the name of " Millions," and it was suggested by the Director of the Laboratories that possibly some species of fish existed in this country which would be of like value. Such a fish, to be capable of yielding the best results, must possess several characteristics. It must be capable of living and breeding in stagnant as well as in running water, it must be small, preferably not more than 10 cm. in length, a prolific breeder,! hardy enough to bear transportation and, last but not least, so fond of mosquito larvae that it will take them in preference to anything else. It is sometimes said that any fresh water fish will eat mosquito larvae, and this may possibly be the case when no other food is available, but it certainly is not so with Nile fish living in their natural habitat. In some of the marshes bordering the White Nile, shoals of tiny fish can be seen in water literally alive with mosquito larvffi and yet making no effort to catch them. Many species of fish exist in these marshes and all which could be caught were given a trial. Some would not live in glass jars — no other and larger vessels were Fish as available — while others, though they throve in the jars, refused to eat mosquito larvse larvicides . ' & J J ' i until they had been without food for periods varying from one to three or four days. A single species, — later determined by Mr. G. A. Boulenger of the British Museum (Natural Hist.) to be Ophiocephalits obscurus — taken in a khor situated between Gebel Ahmed Aga and the river, appeared to look on mosquito larvtB as its natural and only food. The specimens caught were small fry — in which stage they rather resembled tadpoles — and though only 1-5 — 2 cm. in length would tackle large Stegomyia and Scufomi/ia larvEB with the keenness of terriers catching rats. They remained healthy although the water in their jar was sometimes unchanged for several days and allowed to become green and slimy. Later, when mosquito larvae could not be obtained, they were fed on earthworms. This fish, therefore, possesses all the characteristics needed for a mosquito controller, except one, or perhaps two — when full grown U. ohscunis attains a length of 35 cm. and possibly it does not breed freely. In swamps and similar places, where the fry after feeding in the shallows — and it is in shallow water that mosquitoes prefer to breed — could as they increased in size regain the deeper parts, I believe this fish, wlien once established, would yield the verj' best results, but its size unfits it for use on artificially irrigated farms. On a farm with a system of water channels as described KEPOUT OF THE ENTOMOLOGICAIj SECTION 111 above, the fry \Youlfl be carried into the gudvvals and there do good work, but they would be unable to return to the deep canals and so would eventually be captured by some native who had an eye for the "pot." Only those fish which spent their entire life in the canals would be able to perpetuate their species. At the suggestion of Mr. A. L. Butler, Superintendent of Game Preservation, the fish Gyprinodon dispar was, later in the year, made the subject of an experiment. This fish, when mature, attains a length of 8 cm. and exists in Khor Arbat in untold myriads. Khor Arbat is a stream of slightly brackish water running in a gorge in the rocky hills about twenty-two miles N.N.W. of Port Sudan. When in flood I believe it is in places seven or eight feet deep, but at other times it is a shallow stream rippling over and around Cyprinodon stones, and eventually losing itself in the desert. A visit was paid to this Khor early "'*"'' in September, 1909, and one hundred living specimens brought to Khartoum. Considerable difficulty was experienced in persuading them to live in captivity until it was found that while, if placed in jars nearly full of water, most of them died within twelve hours, yet if given only about two inches of water over a layer of sand, they could be transported fairly easily. The jars containing them were carried to Port Sudan by hand and from there to Khartoum by rail. No mosquito larvae were obtainable at Khor Arbat, but a number, offered them on their arrival at Khartoum, were readily taken. This, however, is not conclusive proof that G. dispar feeds on mosquito larvae in its natural state, as the fish had then been without food for two days. As they did not thrive in captivity, advantage was taken of the permission very kindly granted by Mr. A. Maclntyre, Acting Manager of the estate at Zeidab belonging to the Sudan Plantation Syndicate, Ltd., to place them in a gudwal on the Company's Estate. A short length of gudwal — about fifty yards — was utilised for the purpose, and the pipe connecting it with the canal, netted to prevent other fish gaining access. This gudwal was not needed for irrigation purposes so the water was allowed to stagnate, fresh water being occasionally added to prevent it from drying out. Owing to various reasons — leave and duty — I was unable to visit Zeidab again until March, 1910, when I was told by Mr. Maclntyre that the fish had lived in the gudwal until the end of January — that is, for over four months — when, as the water had become very overgrown with weeds and slime, he had transferred them to one of the canals, fearing that though still apparently healthy they would die if left longer in the filthy water. During the four months that they were in the gudwal he had been unable to find any mosquito larvae in it. It is almost certain that mosquitoes laid eggs in this gudwal during that period, for in March there were numbers of the larvae of Pyretophorus costalis and a Gulex sp. in another gudwal containing water not far distant. The results of this experiment were so promising that it was decided to make an attempt to establish G. dUpar in some place in or near Khartoum, from whence supplies might be sent to anyone requiring them. H. E. the Governor-General very kindly granted permission for a large reservoir in the Palace gardens to be used for the purpose, and Mr. C. J. Slade, Towm Engineer, Khartoum, and Mr. F. S. Sillitoe, Experiment.s Superintendent of Gardens, rendered invaluable aid to the scheme by having all the ";"^. fish already in the reservoir removed — no small task — and by giving general assistance. This reservoir is some eight feet deep, has brick sides, and is fed from the river by a pump, the water being conveyed from the pump to the i-eservoir by a wooden gutter. This gutter was fitted with a wire net to prevent river fish from finding their way into the reservoir. Early in April over two thousand living fish of various sizes, mostly immature, were brought from Khor Arbat and of these about sixteen hundred were placed in the 112 REPOBT OF THE ENTOMOLOGICAL SECTION reservoir. The remainder were put in a small tank in the Gordon College garden for the use of which I was indebted to Mr. M. F. Simpson, Assistant Director of Education. Mr. Addison, Assistant Engineer in the Gordon College Workshops, kindly undertook to see that this tank was kept supplied with water. All those in the tank died within a few days, apparently from the attacks of a fungus, for prior to death, the fish assumed a light furry appearance on the shoulders, and swam aimlessly about. At the same time a number of those in the reservoir died, showing similar symptoms, but how many could not be determined as several kingfishers were in the habit of fishing in this water and they may have removed some of the sickly fish. On my return from Dongola at the end of May no G. dispar could be found by searching the edges of the reservoir, nor have any been noticed since, so it is feared that they have all perished. It is possible, though, that there are still some hundreds remaining, for the volume of water contained in the reservoir is very considerable. It is difficult to understand why these fish brought up in April should liave perished, while those caught in the previous September flourished for over four months in a stagnant gudwal. Possibly some simple precaution, which should have been taken to ensure their remaining in good health throughout their rather trying journey, was overlooked, and in any case the species seems deserving of another trial for it is one of the few " small " fish which occur in this country, and, judging from the numbers which exist in Klior Arbat, it is a most prolific breeder. Blood-sucking flies other than Mosquitoes Blood-sucking As will be seen from the list given some forty-nine species of blood-sucking THptera flies other than o^j^gj. than mosquitoes are now known to occur in the Sudan. Of these, two tabanids. Mosquitoes one from Eagaa and Dem Zubeir, Bahr-El-Ghazal Province, and the other from the Lado District, Mongalla Province, are as yet nameless, as the specimens taken were not in sufficiently good condition to allow of an accurate description being drawn up. A third nameless tabanid, from Malakal, is also possibly representative of a new species near Tabanus fiisciqje^, Ric, but, up to the present, only very much rubbed specimens have been captured. Descriptions by Mr. E. E. Austen of three new seroots, all from the same locality, viz. Khor Arbat, Eed Sea Province, are embodied in this rejaort for the convenience of those who may not have seen tlie publications in which they have appeared. Of the small seroot, Tabanus xiifis, Jaenn., two forms occur, specimens taken at Khor Arbat being invariably slightly larger and uniformly paler than those from other localities (Plate I., figs. 4 and 5). Three of the species figured — Chri/aops fuscipennis, Ric. (Plate I., fig. 1), Tahanus diftinctna, Ric. (Plate II., fig. 1) and Hiematopota ahysniiica, Surcouf (Plate I., fig. 2) — do not, so far as is known at present, occur within the bounds of the Sudan, but have been taken just over the borders. It has been thought advisable, however, to illustrate them. Since the last Report of these Laboratories was issued, the name Tahanus focius, Walker, has been sunk as a synonym of Tabanus tseniola, Pal. de Beauv. It will probably be noticed that the closely-allied species, Tabanus variatns, Walker, has not been included in the list of Sudanese blood-sucking Diptera. This is because I am convinced that it is not a valid species. Mr. Austen, in his African Blood-sucking Flies, alludes to the possibilities of these two species proving to be one and the same, and the examination PLATE 1. CInyso/is /mcipeniii.':, Ricardo 9 -■ l/,eiiinto/tola n/;v.tsinka, Surcniif 9 3. lla-mato/'Ota hrutmesceiis, Ricardo 9 1. I'ltl'mius mji!. J.icnnicke ,5 5. rnl'amis si/Jis. J.'iennicke 9 Small illtistfatioHS im/hatr ntittiru/ .rizt-s i;i;roi;i' ()!•' 'I'lii'; ioN'r().M(ii,()(a('Ai. suction I i.'i ot some InuuU'uds ot spucimons of the various {onus of this labaniil has Iccl me to Ijclicvu that the name T. v(;()(1ICAL SMC'I'ION III buried themselves in the sand prior to pupating. A careful search, however, revealed the T.ianioia fact that they had disappeared, and it was not until later that mice were discovered to 'f"'* , , •' '^ '^ destroyed by bo the cause of the loss. Tin- two remaining larvaa were then killed and preserved, mice It is possible, therefore, that the larva described below is not quite mature. 7'. Ui'uiolih larvic are more active and ferocious than those of T. jiar., vigorously attacking any other lai'va with which they may come in contact. They have not, however, the power possessed bj' 7'. far of lying dormant in the soil for at least fifty-s(!V(!ii days if the conditions are unfavourable for their development. Dencripfionx — The <■;/;/ (Plate III., fig. 7), is spindle-shaped, about I'lT) nun. in length and, when first laid, white in colour. It becomes darker as the embryo within develops. The larva (Plate III., fig. 3), when fully extended, measures about 29 mm. Colour white to greyish-white. Mandibles black. On the anterior third of each abdominal segment, except the eighth, is a ring of pseudopods, eight in eacli ring -two dorsal, two latei'al, four venti'al — except on the first abdominal segment, where the dorsal pair are wanting. On the second abdominal segment the dorsal pair are very poorly developed. The ventral pseudopods are always larger than the dorsal. Each pseudopod bears a crown of colourless spines or hooks, and between the prolegs there are also spines or hooks, often darker in colour, and forming a continuous ring. The anus is situated ventrally at the base of the eighth abdominal segment and is edged with dark hairs. On either side of the anus is a patch of dark hair, roughly kidney-shaped, and beyond each patch, laterally placed on the segment, are two small round spots of dark hair. The siphon tube consists of two segments and when exserted is shorter than the eighth segment. The whole surface of the larva is more or less shiny, with varying longitudinal striation, the areas bearing very fine stria; being markedly duller than the rest. The prothorax has the dorsal area smooth in the anterior two-thirds and rather coarsely striate posteriorly ; the ventral area is almost entirely smooth and divided in two by a median furrow ; the two lateral areas are finely striated in the basal third and more coarsely so in the anterior parts. The mesothorax has the dorsal and ventral areas smooth and shining in the anterior two-thirds, and rather coarsely striate posteriorly, the ventral area having no furrow ; the lateral areas are a little more finely striate than those of the prothorax, and there is a rather broad dull non-sti'iatod band at both the anterior and posterior margins. Similar dull bands occur on the metathorax and the abdominal segments, but comi:)letely encircling the segment. The abdominal segments 1 to 7 have the dorsal and ventral areas moderately shiny, and the striation is rather coarser and irregular ; the lateral areas appear much duller, owing to the extreme fineness of the striation. On the eighth abdominal segment the stria' ai'e moderately well-marked and of similar appearance on all the faces. A natural ciienii/. From an egg-mass of 7'. hi'iiiala, taken at Gebelein, numljors of a A para.sitic small Hymenopteron were bred.' This has not yet been identified, but is figurcnl in the |."„"'of accompanying plate, together with the parasitised egg-mass show-ing the exit hole of the '/'. in-nhla parasites. Taliannx difienialiis, Macq. Plate IV., figs. 1 and 2; Plate V., figs. 1 10 The distribution of this tabanid as given by Austen" is a very wide one. In Africa it occurs from the Transvaal in the south to Egypt in the north, while outside the bounds of ' This insect belongs to the family (!linh:iiliihr, and luis «in(;c been (U'sf'rilied nndci' the n.'une of Tchnomns bmcfnetor, Crawford — Guy A. K. Marshall. "■ Austen, K. E. (lOnll), Afrinm l!l. -S O — c/^ -b .S '55 u t^ ^ CD t^ ■^^ >^ ^ ct CD XI c c CD -^ ;g CD T3 jii > S ^ TO ^ "p: o E c jn O CD >^ C\j u 'ex O o — o C C >-, ■- ^ 'B :§ CD .b ..£ XI ^ -a CO c\3 o txO _ o -^ CO ^ O Q, ^ M- C. O X CD C C O C '■<-' O O U (D -J— • r- CD .£ 'O CD & X ■+-• c o tj CD c c o (J i£ to CO -t^ c ^ X T3 CO ri J— -i— CO CD o u :3 % .y . GO _co Ki CD 3 X E _ c >% CD O CD CD CX CO _U) (J c^ CD c^ H := CX , , -4—* > '^ g CD X -o u CD CO _ C CO , , CX ^ no o CO Xi ^ S =2 CD CO CO c^ O CO CO o T3 .-&^ cd CD E O CO -4—' bO ° S CO CD CO C cd > C o c o >, CJ = 1- CD 0) ^ c +-■ CD 5 CD ct3 >_ t« C C •- o _ 'XI k! cd CJ CD CX ~ rd CD CX tj > C CD C O -O CD H _ CO CD >- o o H H CX O v« o i3 CD O X a O (73 K E a •Q '^^ V, TIKI'OUT Ol' Till'. I'.NTOIMOI.OGICAI, SUCTION 127 C I M I C I D .1; When the last Eeport of these Laboratories was issued, it was beheved that bed-bugs amhiJi,- were responsible for the spread of the usually fatal disease known as kala-azar. Since then, however, the researches of Capt. D. S. B. Thomson and Lieut. W. E. Marshall, E. A.M.C., in this country, and of others woi'king elsewhere, seem to jn'ove that in this case, at any rate, these obnoxious insects may possibly have been unjustly accused. Of the two species found in tiie Sudan, Gimex lectiilarins, Linn., occurs tliuughuut the northern provinces, and has also been taken at Bor and other stations in the south. Gimex rofiindatufi, Sig., has twice been taken — at Port Sudan and at Suakin — on pilgrims from Yemen. Quite recently Captain Percival sent specimens from the Lado District. OESTKID.E Comparatively little is known at present of the (Estrldm of the Sudan. (Kflriia oris, L., Oisiridu: occurs but does not appear to be conuuon. Mr. G. H. Storrar, Sudan Government Eailways, bred out one specimen from a larva expelled from a goat at Eabak in 1909, and occasionally it is noticed in Khartoum. Almost every hartebeeste which one kills on the White Nile harbours the larvic of (Ustnis rariolosns, Lw. (Plate IV., fig. 4 and figs. 15 and 16), and numbers have been received here, taken from hartebeeste shot on the Blue Nile. The larva of another species of GephaJomijia — C. macnlata, Wied. — is a common parasite of camels. A single adult specimen of dush-o- pliihi^- flavipes, Oliv., was taken in Khartoum in 1909 by El-Bimb. Williams, Veterinary Department, and an adult female of I laslrophilvs at^iiiuf, Brauer (Plate IV., fig. 3), was captured near Eenk in June, 1909, attempting to oviposit on the legs of a donkey. It was interesting to notice how panic- stricken a donkey became on hearing the distinctive hiun made by this fly when on the wing. Three loose donkeys, on hearing it, stampeded, and when the cestrid transferred its attentions to another donkey tethered near by, this unfortunate animal, unable to get away, rolled on the ground, kicking as if in tlie last stages of fright. These same donkeys would show little or no excitement when numbers of seroots were sucking, or attempting to suck, blood from them. ^Ir. J. C. Walker, Suchin Government Eailways, has sent to these laboratories a number of cestrid larvie taken from under the skin of an ariel gazelle shot at Khor Aral); the larva of another species has been taken by Mr. MacMichael from a similar situation in a dorcas gazelle killed at Gebel Sungur, N.W. Kordofan, and Captain Spencer has sent yet another specimen from gazelles shot in the Eed Sea Province. Larvifi of a different species living under the skin of a bohor reed-buck, shot on the Meshra-Wau road in May, 1909, have been received from Col. Mathias, D.S.O., P.M.O. Egyptian Army. Ui — (Jistrtts var/o/tfsits, pupa, X -3 iil'aw. A. J. Enoki : tr.zi I'ifi. l.'>. — ClistmA vnria/osus 128 kerout of tue entomological section Ticks Ixudoidca ^"■^■' There are now twonty-ono species of ticks known to occur in tliis country. Of these, the following species have been placed on record sineu the last Eeport of these Laboratories was issued : — Aryas bruinpti. — Nine immature specimens were taken among the rocks at the foot of one of the hills or gebels at Gebelein. This tick was originally described from specimens taken by Dr. Brumpt in somewhat similar situations in Somaliland. Judging from the effects produced by a nymph which attached itself to my forearm one evening, its bite is a thing to be carefully avoided. At the time, the pain was hardly noticeable and the tick was allowed to remain sucking blood for several minutes until circumstances compelled me to remove it. As I am usually very resistant to the bites and stings of ticks and insects, I was considerably surprised to find the following morning that the site of the puncture was marked by a small hard red lump around whicli was a slightly sw'ollen and very much discoloured area, resembling a bruise, some three inches long and two inches wide. The discoloration lasted for several days, while the hard red lump remained for weeks and was sometimes the seat of a certain amount of pruritis. A. hrampti probably attacks wild animals as well as man — from the fact that the Arabs living around Gebelein were unaware of the presence of " Khaim " (as they term ticks of the genera Aryas and Ornithodoros) in the gebels, it seems probable that it relies almost entirely on creatures other than human beings for its food. These gebels harbour leopards, warthogs, hysenas, honey-badgers, conies, etc., besides untold myriads of bats. Ai-yas vi-apcrtilionis. — Eight larvffi were taken from an undetermined species of bat, in Khartoum. They differ sliglitly in appearance from English specimens. Hiemaphijaalis leaclii. — This species has been taken on the dog and hare, on the White Nile, and on a stoat (? possibly a species of mongoose) at Erkowit (Dr. E. V. Crispin, Sudan Medical Department). Bermacentor rhinoccrotis . — A large number of both sexes was taken on grass and bushes in the vicinity of Azzar, a few miles south of Bor. Specimens of this tick were also collected by Major Brakenridge, E.A.M.C., from a rhinoceros shot by H.E.H. the Duke of Connaught in British East Africa. Ixodes sp. — A single female was taken from a dog, at Benk wood-station, and a nymph of what may be the same species, also from a dog, at Kaka wood-station. The adult specimen differs from I. cavipalpus in possessing a longer scutum but in other respects closely resembles that species. Owing to its mouth-parts being broken it cannot be definitely determined. Aponomrna exornatnm. — This has been taken a few miles south of Meshra-el-Zeraf on the White Nile, and at Kassala (Captain Bousfield, E.A.M.C.), in both cases from a large lizard. Apouomtna sp. incert. — A nymph of a species of Apunonnna which could not be determined, was found with its rostrum buried in the quill of a partially-formed feather on a wild guinea-fowl at Azzar. This is the first instance recorded of a member of this genus attacking a bird. Anihhjomma tholloni. — ^A single female was taken on an elephant. Aviblyomma sp. incert. For the determinations of these ticks I am indebted to Professor G. H. F. Nuttall, F.E.S., who very kindly examined a quantity of material collected on the White Nile in 1909. REPORT OF THE ENTOMOLOGICAL SECTION 129 Tlie tick referred to in the last Eeport, issued by these Laboratories, as Manidnrpiif: anniilatinf var. decolorata, has been determined by Professor Nuttall to be identical witii Boophilus an.itrali.<. This species appears to be widely distributed throughout the southern provinces, and a number of specimens has been received from Portuguese East Africa. Ornithodoron aaviyniji has been taken by Mr. H. A. MacMichael in several parts of Kordofan, and no fewer than three hundred and seventy specimens were collected, in the space of two hours, from under a single tree by a well, a few miles N.N.E. of Khartoum North in August, 1909. This tick has been the subject of some investigations carried out by the Bacteriological Section of these Laboratories. In the following synoptic table of Sudanese ticks, as far as possible, only those characters which can readily be made out are utilised. Many of these characters are those used by Salmon and Stiles in their work on the cattle ticks of the United States.* This table is incomplete as regards some of the species of Amhlyomma owing to lack of available material, but all the recorded members of this genus are nevertheless mentioned. Ornithodoros moiibata, the transmitter of relapsing fever in Uganda and elsewhere, is included, as it is more than possible that it occurs in the Bahr-El-Ghazal and other southern provinces, and in any case it is useful to be able to distinguish between it and the closely allied 0. ^avif/ni/i. It must be remembered that although this table deals with all the species known at present to occur in this country — with the exception of the undetermined species of Aponomma from a guinea-fowl — it is almost certain that the names of others will have to be added to the list as time goes on. Synoptic T.\i)le op Sitd.^nese Ticks Synoptic Table (Scutum absent ... ... ... ... ... ... ... ... ... ... 2 I Scutum present ... ... ... ... ... ... ... ... ... ... 7 (Margin of body difl'ering in structure from general integument ... ... ... 3 LMargin of body not differing in structure from general mtegument ... ... 5 ^Dorsum pitted by symmetrically arranged polygonal depressed areas bounded by 3 j rugose ridges ... ... ... ... ,.. ... ... ... Anjan brnDipti ^Dorsum not pitted by such depressed areas ... ... ... ... ... ... 4 fBody longer than broad ... ... ... ... ... ... . . Argas per.''icus \ Body broader than long ... ... ... ... ... ... Argas vespertilionis J. (Body panduriform ; found in ear ... ... ... ... Ornithodoros megnini iBody oval G (Eyes present on supra-coxal fold ... ... ... ... Ornithodoros savignyi LEyes absent ... ... ... ... ... ... ... ... Ornithodoros: monbata ("Palpi short, subtriaugular, not or only slightly longer than broad; capitiihnii sliort 8 [Palpi longer than broad; capitulum long ... ... ... ... ... ... IS C Dorsal surface of capitulum hexagonal ... ... ... ... ... . . 9 o - [Dorsal surface of capitulum rectangular ... ... ... ... ... 17 (Stigmata comma-shaped ... ... ... ... ... ... ... ... ... 10 I Stigmata nearlv round ... ... ... ... ... ... Boophilim ain'trali." lo|^^l^ " 11 tlemale . . ... ... ... ... ... ... ... ... ... ... 14 (Punctations on scutum unequal (large and small) ... ... ... ... ... 12 (Punctations equal or nearly so ... ... ... ... ... ... ... . l.'j • Salmon and Stiles, (1902), Cattle Ticks of the United States, United States Dept. of Agriculture i; 130 liEPOKT OF THK ENTOIWOLOOICAL SECTION Synoptic Table [contiiuied] 12 13 14 15 IG 17 18 19 20 21 22 23 24 25 26 27 28 f Puiictations distributed regularly iPunctations distributed irregularly rPunctations separated (distinct) [Punctations contiguous, making tlie scutum rough (shagreened) f Scutum elongate oval I Scutum short oval, or as long as broad I Scutum with unequal (large and small) punctations l_Scutum with equal punctations [Punctations separated (distinct) JUiipiri'plitihis ^(1)11/ 1! /•liens liltipiceplmlus punctafissimus lihipicephalus sinius lihipicephaliis evertsi 15 16 icephalns saiig ii inens Uhipicephalus pmicfaMssimns SMpicephahis siniits [Punctations contiguous, making scutum rough (shagreened) Tthipicephalus evertsi (Eyes absent; coxae 1 not bidentate ... ... ... Hn'maphi/saliit leachi I Byes present; eoxte 1 bidentate ... ... ... ... Tiermacentor rhinocerolig (Anal groove surrounds anus anteriorly and open posteriorly ... Ixodes sp. ineert. I Anal groove surrounds anus posteriorly and opens anteriorly ... ... 19 (Eyes absent; anal plates absent ... ... .. ... ... ... ... 20 lEyes present ... ... ... ... ... ... ... ... ... . . 21 (Scutum marked with green metallic spots I Scutum not marked with green metallic spots C Scutum with metallic reflections 1 Scutum without metallic reflections (Male ... ^Female (Eyes flat [Eyes hemispherical, in sockets (Marginal groove present ... I Marginal groove absent On either side of scutum a without a light mark On either side of scutum a dark triangular-shaped area containing a light mark [Eyes flat [Eyes hemispherical, in sockets ... dark triangular-shaped area Apono inma, exornatuiii Aponomma Iseve ... 22 TTiiaJ omnia legijptium, 23 26 ... 24 ... 25 ^1 mhli/omnia marnwreii m Aiiihl i/0)nii>a tholloni AmhJyoin.ma sp. ineert. l ^4 mbh/onima variegtifiiin — hehrfenin 27 28 Amhlyomma tholloni ... Amhli/omma marmoreuin A nibl ijomma variegafnm — marmoreum ... A inhliioiiima sp. ineert. Dura Stem-boicr Animals injurious to Farm and Garden Grots The Dura STEM-noKER Sesaiiiid cretica The larva, pupa and adult of this moth were figured and an account of its life history given in the Third Report, of these Laboratories.! It is a difficult pest to control as it breeds throughout the year on dura, maize [Dura shamt), and sugar-cane, and, moreover, * Neither spcciraeus nor descriptions of these ticks arailable. — H. H. K. ' Thint Report, Wellcome Tropical Tlexr.arch Lnhorntorics, 1908, jmyc 222 ijioi'oiri' OK rill'; kntoiiological section 131 exce))t when in tlie adult sta-fic it ])asses the wliolo of its existence hidden within th(t stem or the sheathing leaves of the )ilaiit it attacks. Some investigations liave been carried out in Dongola Province witli a view to ascertaining whether it would not be possible, in that region at any rate, for the farmers to lessen its numbers by concerted action. The seasons in Dongola Province are roughly as follows : — The main crop of the year is tlie flood or demera. crop, sown at high Nile, and consists mainly of dura and maize. It is harvested some four months later. The wheat crop is put in in December. The winter or shitn-e and the summer or i-aifie crops are comparatively small and miscellaneous but include a proportion of dura. When the demera crop is harvested, the second brood of borers is in the larval stage, The larvne most of these larvte being within the stalks. A few, however, are in the roots. The stalks, ^,":^':", '" ® ' ' ' dried t/unr which are known as (/ii.inaJi, and which constitute the staple diet of the farm live-stock stalks afiiT throughout the greater part of the year, are removed from the cultivations and stacked, ^^'^'^^ usually on the roofs of houses, but the stubble is left in the ground. The houses are situated among the cultivations and the stock are fed close to the houses. Qiiagab is not very palatable fodder and, consequently, stock prefer to eat only the tops and leaf, and to reject the hard stems — those very parts, in fact, in which the caterpillars like to take up their winter quarters. The i/iisaab from a demevd crop frequently lasts until after the next demera crop has appeared above the ground. Occasionally, a strip of land is left with the stalks of the demera crop standing to act as a wind-break to protect other crops. As many as seventy larvje, together with a few pupae, have been taken within tlie short space of half an hour from a wind-break of this kind. The dura stem-borer will sometimes attack wheat, and the larva? have also been found wintering in d.n.kn stubble. A number of stalks containing larvae, from a demera crop was collected in Dongola Province in February and placed in a breeding-cage in Khartoum. One moth emerged in March, four in April, thirty-five in May and June, eleven early in July, and the last had made its appearance by August 6, the total number of moths obtained from these stalks being fifty-two. The natives assert that while the demera crop is always heavily infested, there are comparatively few caterpillars to be found in the !nis are froquontly not eaten — these should be collected every day and burnt. 132 KEPORT OF THE EN'TOMOLOGICAIi SECTION Black or Greasy Cutworm The basin system of irrigation (d) When a young crop of dura, maize, sugar-cane or wheat is about a foot in height, all plants containing caterpillars should be pulled up, and the caterj)illars cut out and squashed. This operation should be carried out again two or three weeks later. Plants containing caterpillars can be known by their young centre leaves being withered and dying or dead, while their outer lower leaves are still green and living. (e) Stray plants of dura, maize, sugar-cane or wheat should not be allowed to grow among other crops or in odd places such as the edges of gudwals (water channels). The Black ob Gbeasy CuTwoiur Agrotis i/psilon, Rott. A very considerable amount of damage was suffered by crops on Nuri and Gureir basins in Dongola Province, owing to the ravages of this cutworm, in November and December of 1909. A. ypsilon is a well-known pest in many other countries, including the United States, Ceylon and Egypt, attacking tobacco, tea, cotton, berseem and a host of other plants. Life hisfori/ and habits. — The female moth deposits her eggs on the lower parts of the food plants, and, on hatching, the larvaB at first feed freely on the foliage. After a few days spent in this way they acquire the habit of feeding only by night and hiding by day in cracks and holes in the soil. As these older larvae are poor climbers they usually cut off young seedling plants close to the ground to enable them to feed more easily on the tender leaves. They will also sometimes attack the plant below the ground. When mature they are about two inches in length and of a dirty green colour with darker spots or tubercles. If alarmed they at once curl up and feign death. They pupate in cells hollowed out in the soil and after about fourteen days give rise to dull greyish-coloured moths. In Dongola Province, up to the present, this pest has confined its attentions to crops grown on land irrigated on the basin system and has not attacked the sakia crops. Sakia land is irrigated by means of waterwheels — sakias — which lift the water from the river or from wells. The basin system of irrigation is briefly as follows. — Where the land levels permit of its being done, canals are dug, along which water flows from the river while it is in flood on to land situated generally behind the river-sakia land. These basins are therefore flooded each year for periods varying with the height to which the river attains, and the crops are sown when the land has become sufficiently dry to enable men to walk about on it. The moths of A. ypnilon lay their eggs on the weeds which spring up as soon as the water has gone and are particularly fond of terroba — probably a GhrozopJwra sp. — a weed which is very plentiful on the basins. The cultural methods practised by the cultivators of these flooded lands are of the simplest. No effort is made to break up or clean the soil prior to sowing the seed by hand in holes made witli a stick — or seluka as it is termed — and as might naturally be expected, the land is very foul. When the crop of wheat, barley, or whatever it happens to be, appears above the ground the land may be lightly hoed to give it a chance of getting ahead of the weeds, and it is then left to itself — unless possibly an effort is made at bird-scaring — until it is ready for harvesting. The cutworms, which are probably already present on the weeds when the seed is put in, attack the crops either from choice, or from necessity if the weeds on which they started life are uprooted just when the crops are available for food. In 1909 the crops on Nuri basin were badly attacked, those on Gureir basin to a less extent, while those on the new basin on Kerma plain escaped altogether. This KEl'OK-r OK I-IIB ENTOMOLOGICAL SECTION 133 cutworm has appeared on Nuri basin in previous seasons, increasing in nuinlici-s every year. Some four years ago the area of land iu this basin, available for cultivation, was considerably increased by the construction of a canal, and the occurrence of A. ypMon in noticeable numbers appears to date fi-oni then. In 1909, the barley and other crops were entirely destroyed when they were only a few inches in height and the land was then thoroughly hoed. Numbers of birds followed the men engaged in this work and B'r'i enemies devoured the larvte which were exposed, a bird known locally as yairdouu — a species of wagtail — being particularly beneficial iu this way. The crops were then resown and were not again attacked by cutworms. The basin was searched for cutworms in the latter end of May, 1910, and although there was a considerable area of land under crops watered by well-sakias, and the whole basin was more or less covered with terroba and other weeds, the only traces of the pest which could be found were empty pupal cases iu the soil. It is believed that the moths were in hiding in the deep cracks in the ground — these cracks may extend to two metres in depth — or in similar places. Preventive and remedial meamiret:. — In other countries ^1. ypsilon usually produces several generations in the course of the year, but in Dongola Province it appears to be single-brooded. Should it continue to be single-brooded in the future its control should be effected by purely cultural methods, and even were it to acquire the habit of breeding throughout the year it should never attain the importance, as a pest, of the dura stem- borer — Seiiaima cretica — if measures are taken to keep it in check. The cultural methods recommended are as follows. When the water soaks away or recedes from the basins, and as soon as the land is sufficiently dry to permit of its being done, the soil should be well hoed, or ploughed, and cleared of all weeds. This operation will enable birds to devour numbers of the larvae, others will be crushed and the remainder will be hard put to it to iiiul sufficient food to enable them to comjalete their life-cycles. This hoeing should also yield good results in other ways — it will tend to conserve the supply of w-ater in the soil by preventing undue evaporation, and the soil itself will be sweetened by the action of the sun and air. If there is no second breed of cutworms, crops sown on land treated in Prevention of this way will be safe from attack, but should a second brood appear it will be necessary to 'he Greasy ^^ •' Cutworm have recourse to the use of poisoned baits or to hand-picking. This latter process is somewhat costly and needs to be very thoroughly done to be effective, moreover sufficient labour for the purpose may not be available, and consequently it cannot be recommended except for small areas. The destruction of the larvae by means of poisoned baits has been found in other countries — notably in the United States — to be the most practical and efficient method of dealing with cutworms over large areas, the substances used as baits including bran, meal, etc., and green forage. The latter has several advantages over bran and meal, and is used in the following way. Any available green-stuff — e.g. herseem., grass, barley, young succulent weeds, etc. — is collected, out up into small pieces of about half-an-inch to an inch in length and moistened with a mixture made up of 1 lb. arsenite of soda, 8 lb. treacle and 10 gallons water. This poisoned forage is then thinly scattered over the infested land. This should be done a few days before the croij appears above the ground and, if necessary, again if injury is noticed after the plants have begun to grow. The land should be thoroughly cleared of weeds before the first application. Cutworms will take green forage treated with this mixture in preference to untreated growing plants, owing to their love for sweet things. By applying the poison in this way there is very little danger of cattle, sheep, donkeys, etc., being poisoned, for the pieces of green forage are too small to be readily picked up. Care should be taken, nevertheless, that live-stock, including poultry, do not have access to land which has recently been Asal FIv 131 KEI'OKT OF THE ENTOirOLOCICAri SECTION treated. Poisoned forage should always be applied in the evening, as it will then lie fresher and more attractive to the cutworms than if it has been allowed to dry and harden in the sun. There is evidence to prove that in bygone centuries large areas of land in Dongola Province were irrigated by means of Hood canals during high Nile, and by well-sakias during the remainder of the year. On Kerma plain there still exist the remains of sakia- wells showing that at some time the water-level in the soil was sufficiently high to enable cultivation to be carried on. During recent years there has been but a small area of land irrigated in this way, but considerable sums of money have been, during the last four years, and are now being spent in the construction of canals which will result in many thousand feddans of laud being rendered available for cultivation. It is highly important, therefore, that the natives cultivating these basins should realise the necessity of doing something more to the land than merely dibWing in the seed. The As.'Vl Fly AiiliM aorijld, Theobald As stated earlier in this report, a considerable amount of time has been spent in endeavouring to trace, the life history of the so-called asal fly and to devise some method for controlling it. It is to bo regretted that up to the present these endeavours have not been crowned with success. Mr. Chas. P. Lounsbury, Government Entomologist, Cape of Good Hope, and Mr. Claude Fuller, Chief of the Division of Entomology and Horticulture, Natal, have very kindly furnished me with what information concerning the bionomics of this aphis they possessed, and my thanks are due to them. The work in this direction will be continued in the hope of discovering some effective and practical remedy for dealing with this pest. The Cotton Stem-isohek lipheuoptera neylecta, Klug Plate VII. The Cotton This beetle has been found attacking cotton and garden hibiscus throughout the stem-borer northern Nile provinces, and it appears probable that unless measures are taken to keep it in check it will have to be regarded as a serious pest when the area of land under cotton in the Sudan increases. Mr. C. O. Waterhouse, of the British Museum (Nat. Hist.), has determined it to be a member of the genus Sjjhenupfera, probably N. negleota, Klug. A member of the same genus — S. gonfi/pii — is recorded by Lefroy' to attack cotton in Bombay, Central Provinces and the Punjab — possibly the species here noted is identical with the Indian cotton stem-borer. lJescriptiou>i — %V (H- !)■ ijtiugth, 1-25 -1-50 mm. The endjryo is enveloped in a thin, transparent to whitish menibrane, which is covered by a dull, greenish-blue, scale-like shell, in shape roughly oval, and bearing a number of irregular crinkles or ridges. Larca (fig. 2). Length up to 29 ram. Colour, head brown, mandibles black, thorax and abdomen yellowish-white. The larva is of the typical Buprestid shape, the small head being retracted into the broad, flattened, first thoracic segment, and the abdomen being long and comparatively slender. The first thoracic segment bears Ilulletiu Imperial IndUiUe, 1907, Vol. V., No. 2, payc 164 PLATE Vli Cotton Stem- borer {S/>/ie nop/cm ncgkcta, Klug) 1. Egg 2. Larva 3. Pupa 4. Adult n. Portion of stem of cotton plant, showing egg {a) and exit hole of adult (/•) fi. Longitudinal section of stem of cotton plant showing tunnels niade by borers Suiaii illustrations indicate natural sizes X36 EEPOBT OF THE ENTOMOLOGICAL SECTION both a dorsal and a ventral shield, the former being cut by a longitudinal median groove not quite extending to the anterior border, and the latter by a Y-shaped, forwardly pointing groove, also barely reaching to the anterior border. The remainder of the body is flattened and wrinkled. The anus is situated at the apex of the terminal segment. The whole body bears a few scattered short pale hairs. Pupa (fig. 3). Colour, yellowish-white, eyes dull purple. AduU (fig. 4). Length, 9-10-5 mm. Colour, greenish to reddish-bronze. When first it emerges from the pupal cell the beetle is covered with a fine yellowish meal, especially on the frons, sides of the pronotum, prosternum, metasternum, and venter generally. Head, pronotum, scutellum and venter, reddish-bronze, irregularly punctured and bearing a few scattered, short, pale hairs. Mesonotum and metanotum, bright metallic green, punctured. Metanotum bears a V-shaped, backwardly pointing groove or gutter, in which is a median ridge, black. Dorsum of abdomen, bright metallic green, punctured, and bearing a few scattered, short, pale hairs, with the exception of the basal margins of the segments, and a longitudinal median ridge, which are smooth, and have a bronzy tinge. Antenna of twelve segments — first, small, globular; second, elongated, swollen : third and fourth, smaller ; remainder of antenna serrate. Elytron reddish-bronze, with punctures arranged in longitudinal rows, ridged towards the apex, and terminating in three short spines. Hind wing slightly clouded, especially towards the apex ; eosta terminates in a short spine ; apical margin irregular. Tibiae with longitudinal rows of short, sharp spines ; fore tibiae bear at the apices one longer spine ; mid and hind tibiae two similar spines. Habits and Habits and life history. — The eggs are deposited singly on the bark of the plant, life history q,^ either the main stem or the branches — usually the former — and preferably in a crevice or wound. As many as nine eggs have been found on a stump of caravonica cotton, scarcely twelve inches high, but these had probably been laid by several beetles. On hatching, the larva burrows into the stem, without rupturing the external shell of the egg, and commences a tortuous tunnel in the wood. Freqirently this tunnel runs immediately under the bark for some distance, but it may go deeper in the wood, especially as the larva grows older, and may even, in the case of smaller branches, follow the course of the pith. It may also extend below the level of the ground. As the larva proceeds it packs the tunnel behind it with frass^ and wood chips, which, at first light in colour, become dark brown in course of time. It invariably lies in its tunnel in a doubled-up position. On attaining maturity the larva hollows out for itself a little chamber, usually near the bark, and pupates with its head end pointing towards the bark. The adult eventually gnaws a circular hole through the bark and makes its exit. Dnratioit of the life-cycle. — The observations made on the life history of this pest indicate that there are at least two, and probably three, broods in the course of the year. The season for planting cotton varies in different provinces from February to July, and the crop usually remains on the ground for about eight months. The eggs of the first brood are laid on the young seedling plants while they are only a few inches high, and breeding continues until the cotton wood is dead. If an infested plant of any size is pulled up and thrown aside, the larva can complete its development in the dead wood. Damage done hij the borers. — Young plants are killed outright, but older plants, unless they harbour several of the borers, usually live to the end of the season, though obviously unhealthy and stunted in growth. White ants — termites — if present in the soil, at once ' Frtms — The term applied to the exereta of insects PLATE VIII \«^^t:r-^ 0. M. Bkabu Melon Weevil. Han's trtr^arfih i A u r i v . 1. I.arva 3. Adult 4. Pupa: cells 6. Pupa Melon Stem Borer. Aponu-ryna bin uHia , P a s c . 2. Larva n. Adult 7. Pupa Illustrations are natural size unless other^vise indicated KBPOKT OF THE ENTDMOLOdlCAL SECTION 137 attack a plant which has been weakened by the work of the borer, and the entire damage is then not infrequently attributed to these smaller insects. Preventive and remedial measure!'. — To control the various insect pests of cotton, it is I'icvention of absolutely essential that the remains of the old crop should be collected and burnt as soon yt^n,°borer as the cotton has been gathered, and before the new crop is sown. The effect of tliis measure on S. neylecta will be to lessen its numbers very considerably, as it is usually present in all its forms in the old cotton wood at the end of the season. When the young crop appears, all infested plants should he collected and burnt to prevent the occurrence of a second brood. The Melon Wekvil Baris trfeyardlu, Auriv. Plate VIII., figs. 1, 3, 4 and 6 Eeference to this pest of sweet melons was made in the Third Report of these The Melon Laboratories.^ It does not appear to be nearly as common as the melon fruit-fly — ^'^'" Dacuts sp. — but is nevertheless responsible for a certain amount of damage. Description — Adult (fig. 3). — Length, 4-6'5 mm. Colour uniformly black, elytra sometimes faintly tinged with brown. Head bears comparatively few punctures, suialler than those on the thorax. Proboscis curved, about equal in length to the prothorax, densely punctured, grooved on either side to take the folded antennas. Antenna of nine segments, first segment long, finely punctured, apically swollen ; second segment one-third the length of first segment ; third to eighth segments small ; ninth segment swollen, consisting of four rings, and clothed with greyish pubescence ; second to eighth segments inclusive bear black spines or bristles, shortest on second segment and longest on eighth. Prothorax as broad as long, densely punctured, anterior margin constricted to form a collar. Elytra densely punctured and longitudinally and deeply grooved, the grooves converging towards the apex and a marginal groove extending around the whole elytron except the basal margin ; a small hump is situated in the exterior basal angle of each elytron. Pygidium and venter densely pimctured. Besides these deeper punctures the whole surface of the beetle is very finely punctured. Legs densely punctured, femora bear a few short curved grey hairs ; tibiae bear greyish or black spines or bristles and a cluster of strong black spines at the apices, and terminate in a brown claw-like process; tarsi four-jointed; under-surfaces of apices of tibi* and under-surfaces of tarsi clothed with yellow hair. Preventive and remedial measures. — All infested melons should be collected and burnt. They should on no account be thrown into the river, as there would be a danger of their floating down-stream and becoming a source of infection to other cultivations. The Melon Stem-bohek Apomecijna binnbila, Pasc. Plate VIII., tigs. -2, 5 and 7 Complaints are occasionally received of damage done by this longicorn which was Melon referred to in the Third Report of these Laboratories.- stem-borer ' Third Report, Wellcome Tropical Jiesecrch Lnhoraforif.i, 1908, pnge 2.'5'2 - Third licpurl, H'ellcoiiie Tropical Research Laboratories, 1908, page 233 13H KEPOKT OF THE ENTOIMOLOGICAL SECTION lieiiicJidl measure. — The only methocl whicli can be reconiuiendefl of dualiny with an attack, is that of cutting out and destroying the larvte when the plant shows signs of being infested. The Melon Leaf Beetle Aulacopiwra fureicollt!^, Kiistei' Plate IX., fig. 6 Melon Leaf This beetle is frequently found, in company with the plant-eating lady-bird, Epilaclina chri/someUua, Fab., devouring the leaves of melon plants. The larval stage has not been seen. Preventions and remedies. — Clean cultivation will do a lot towards lessening the numbers of both these pests, and where plants are attacked a dressing of a mixture of Paris green and flour, in the proportion of one to ten, dusted over the foliage, will either poison the beetles or drive them away. DlAMOND-lSACK MoTH Plutella maciiiipennis, Curtis This pest has a world-wide distribution, occurring in Great Britain, New Zealand and many other countries. In the northern provinces of the Sudan it is frequently found attacking cabbages. Diamond- back Description and life history. — The eggs, white, cylindrical bodies, are deposited on the under-surface of the leaf of some plant belonging to the Nat. Order Gruciferse. The larvae, on emerging, feed on the soft tissue of the leaf, generally leaving the upper epidermis intact. These caterpillars are green in colour, spindle-shaped, and when full fed about 13 nun. in length. When alarmed they drop from the leaves and hang suspended by silken threads which enable them to regain their places when the danger is past. The pupie are white to j'ellowish, with darker markings as the imago within develops, and are enclosed in a whitish silken cocoon attached to the leaf, usually in the angle at the junction of two veins. The moth is about 7 mm. in length and has a wing expanse of 13-14 mm. In colour it is a greyish-brown with a light irregular stripe on the posterior margin of the forewing. When at rest, the wings are closed, and these light stripes together form two or three diamond-shaped pale areas, to the presence of which the insect owes its popular name of " Diamond-back Moth." Plants attacked by this pest present a peculiar and characteristic bleached appearance, owing to the habit of the larviE of leaving uneaten the upper epidermis of the leaf. Preventive and remedial measures. — As this caterpiller feeds only on plants belonging to the Nat. Order Cruciferse, no wild plants belonging to this order should be allowed to grow in or near a garden in which it is proposed to cultivate cabbages, turnips, radishes, etc. In gardens, where only small areas of cabbages, etc., are cultivated, the best method of dealing with an infestation of these caterpillars is to sprinkle the plants with boiling water by means of a watering can, fitted with a fine rose. This remedy is recommended by Mr. W. W. Froggatt, Government Entomologist, New South Wales. On large areas where this may not be practicable the young ])lants should be sprayed with paraffin emulsion. Moth kepokt of tiik entomological section 139 The Caui'.aiW'; Bug lliii/rdda picla, F. Plate IX., fi^. 6 • This brilliantly coloui'ed plant-lnij; occurred in large niuuhers on cabbages in some Caljliage of the gardens in Khartoum during the winter 1909-10. It can be distinguished from the "" closely allied B. picta hilar in, Burm., by the markings on the soutellum — in B. picta the scutellum bears a median longitudinal yellowish-white stripe extending from the basal margin to the apex, while in B. picta hilar is this stripe is present on the apical half of the scutellum only. Preventive and remedial measaren. — Gardens should be kejJt free from weeds, especially those belonging to the Nat. Order Crncifent', and any bugs noticed attacking cabbages or other plants should be collected and destroyed without delay to pi'event them depositing eggs. The Eigla Gall Weevil Barii- lurata, Marshall, sp. nov. Plate IX., fig. 2 This small weevil has been noticed in Khartoum attacking Portulaca oleracea, I^inn. RigiaGaii — Arabic name liiijla — which is cultivated as a vegetable and for use in salad. The Weevii larva lives in the stem, its presence causing a gall to form, and frequently a plant may be seen bearing upwards of thirty galls. An undetermined species of Hymenopteron is parasitic upon the larval stage of this beetle. Descriptions — Larva — head, chestnut-brown; body, yellowish-white, curved and transversely wrinkled. Pupa — yellow, eyes black, mandibles brown. The body bears a few scattered long pale hairs. AJiiJf — Length, 3 mm. Colour brown, with a darker brown to black median lateral area on each elytrum. Head densely and finely punctured, and also sparsely and coarsely punctured. Proboscis curved, about equal in length to the prothorax, punctured, grooved to take the folded antennae. Antenna of nine segments, first segment long, slender, a])ically swollen, second to eighth segments small, bearing narrow white scales, ninth segment swollen, consisting of four rings, clothed with greyish pubescence. Prothorax densely and deeply punctured except the anterior margin which is densely but finely punctured ; a lateral patch on either side of mixed yellowish-white and yellowish-brow'n, almost circular, scales extending from the posterior margin three-fourths of its length. Each elytrum bears a basal lateral patch and an apical patch of mixed yellowish-white and yellowish- brown scales, while smaller brown scales are scattered over the median dorsal areas, and the remaining surface bears longitudinal rows of narrow dark brown scales, and the whole surface of the elytrum is longitudinally grooved. Pygidium clothed with yellowish-white and yellowish-brown scales. Venter punctured and bearing patches of white to yellowisli- white scales. Legs punctured and clothed with narrow w-hite to yellowish-white scales, tibiffi terminate in a black, claw-like process, tarsi four-jointed. The arrangement of the scales varies considerably in different specimens. 140 KErOliT OF THE ENTOMOLOGICAL SECTION Imported tig trees attacked by the Fig Stem- boring Beetle Transvaal experiments in prevention of damage by White Ants The Fig Stem-boeing Beetle Sinoxylon sudanicam, Leslie Plate IX., tig. 1 A number of young fig trees imported from Egypt by Mr. Geoffrey Nevile and planted in his garden at Darmali were severely damaged by this beetle early in 1910. When first planted out, the trees were naturally in a weak condition and the beetles — for Mr. Nevile told me that they were usually in pairs — bored into the twigs immediately either above or below a bud, and then circled the twig within the bark in such a way that at the slightest strain it would break at that point. This had the eifect, on all but the main stems, of completely cutting off the flow of sap and so killing the twig. If left undisturbed, the beetles continued to tunnel in the dead and dying wood and presumably had the intention of depositing eggs, though no larvae were taken from twigs collected in March. A single adult emerged from one of these twigs in July. Healthy, strongly growing trees were not attacked. 8. sndaniottm can be distinguished from the more common S. senegalenne by its coloration — the latter species is uniformly dark brown to black, but S. sitdamciim has the basal halves of the elytra tinged with yellow. Preventive and remedial measures. — As this beetle confines its attention to trees which are for some reason in a sickly condition, it is onlj- necessary to keep trees healthy to avoid an attack. When trees are first transplanted and until they have established themselves the vicinities of the buds should be smeared with some substance — such as tar — which is ofl'ensive to tlie beetles. Newly transplanted trees should also be examined every day or two, and all beetles found attempting to bore into twigs, collected and destroyed. If a beetle has succeeded in making a tunnel, in a small twig, of sufficient length to conceal itself, that twig should be cut off below the tunnel and burnt. Mr. Nevile saved most of his trees by picking the beetles out of their tunnels with a pin and removing the twigs which were either dead or dying. When seen in July, these trees had established themselves and made a quantity of new growth, which the beetles had not attacked. White Ants Tei-mitida; In view of the enormous amount of damage done yearly by certain species of the Fam. Teriiiitidce, or as they are popularly termed, "White Ants," the selection of some chemical with which to treat wood used in the construction of houses, etc., or for telegraph poles, to protect it from the attacks of the insect is a matter of considerable importance. No series of experiments has as yet been started in this country to test the respective values of the various chemicals recommended for the purpose, though it is proposed to do so during the coming year. A very comprehensive series of experiments in this direction was started in the Transvaal, in 1906, by Mr. C. W''. Howard, who then held the post of Government Entomologist to that Colony, and has been continued by his successor, Mr. F. Thomson. As the results obtained up to the present from these experiments cannot but be of interest to many in the Sudan, extracts from the articles published by Mr. F. Thomson giving these results are quoted here : — '"The tests were carried on in the following manner. About 800 pieces of wood of a uniform size, 3 x 3 x 12 inches, were obtained. These were mostly of deal, but Transvaal AyricuUural Journal, No. 27, 1909, p. 513 PLATE IX ^ :j I. Fi« Stem-bori,,;; IScc-tlc. .S,«.Kiyh,i SH.tamcum, Lesne 2. Rigia Uall Weevil. BaHs hraln, Mshl •i .-.".I i. Black Carde,, Ant, ..l//„,„„^«/rr tariora. L. 5. Melon Leaf Beelle, Aulaco/./..,ra Mcicias. Ku..er 6. Cabbage Bug, £a^-a(fa fitcta, F, Small illuxtfailo'is inHicaii- natural sizes KEPOKT OF THE ENTOMOLOGICAL SECTION 14] some blue Ki'™. boekenliout, and lemon wood (Xijmalo.i mono^pora) wei'e employed ; all of whieh woods are readil>' eaten ijy wliite ants. In selecting the material witli which to treat the woods, we tried to test such substances as the farmer would have at lumd or could easily obtain, also giving attention to patent mixtures, for sale on the South African Market, which are recommended for this purpose. "The methods for treating the wood were four in number, i.e. (1) soaking in the mixture for 24 hours ; (2) soaking for 24 hours and then bringing to the boiling point over a fire, and boiling for from one to two hours ; (3) painting the surface with one or two coats of the mixture ; (4) boring a hole J inch in diameter down the middle, filling this with the liquid or dry chemicals and plugging up the opening, leaving the chemical to soak through the wood. A certain number of pieces of deal and blue gum were left untreated, to act as a check on the experiment. '■ When the pieces of wood were all ready they were thoroughly mixed up, so that any two i)ieces that had been similarly treated should not be together, and were then planted in the soil about 12 feeJt apart each way, with the upper ends above the surface. Care was taken to choose a place to bury the wood where termites were very abundant. For this purpose Mr. E. F. Bourke, of Pretoria, kindly allowed us to make use of his farm Kalkheuvel, near Pienaar's Eiver Station, District of Pretoria. Parts of this farm are badly infested with white ants, and the worst parts were employed for this experiment. The wood was buried on March 27 and April 15, 1906. ' The first examination was made on June 4, 1907, the second examination was made on August 21, 1908, and the third examination took place on July 15, 1909. "Termites were present on the day of examination on or near the pieces of wood. "Arsenical compounds. — -It was found that mostly all the pieces of wood treated with the arsenical compound were still untouched, and it is very plain that these arsenical mixtures are very effective. The following table will show this: — Greatefficiency Arsenite of soda, 10 per cent. : All wood treated not eaten at third examination. preparations Atlas Preservative, full strength: All wood treated not eaten at third examination. against White Atlas Preservative, 10 per cent. : All wood treated not eaten at third examination. Demuth's Dip, 10 per cent. : All wood treated not eaten at third examination. Cooper's Dip, 1 pint to 9 gallons water : All wood treated not eaten at third examination. Alderson's Cattle Dip, 4 lb. to 8 gallons water: All wood treated not eaten at third examination. Street's White Ant Cure was not effective ; all those pieces of wood which had been treated with this compound were eaten. " Tar and creoKote compound. — -Of all the pieces of wood treated with any of the tar compounds only two pieces, painted twice with coal-tar, withstood the ravages of the termites for three years ; all those treated with Stockholm tar, creosote, carbolic acid, Jeyes' fluid, carbolineum, solignum, asphanite, etc., were more or less attacked and eaten. " Tobacco extracts'. — At the third examination all the pieces of wood treated with these extracts were found to have been eaten. " OiU and paints. — These pieces were also destroyed, with the exception of two pieces of deal painted with raw^ linseed oil, white lead and arsenic mixed ; two pieces of deal painted with raw linseed oil, red lead, and arsenite of soda mixed ; two pieces of lemonwood painted with raw linseed oil, red lead, and arsenic mixed. '■ Transvaal Agricultural Journal, No. 29, 1909, p. 86 .\nts 142 EEPOKT OF THE ENTOjrOLOGICAL SECTION Technique for poisoning with carbon bisulphide " These pieces of wood had not been touched at the time of the third examination, and were as \v sc;iles :{. (»eniiali.-i of ^ {a) forked claspcr (/') inner scale- luft (i) apical tuft ii. Wing of a]pus ■_*. Head of rf 4. Wing of 9 fi. Head of 9 154 A NEW GENUS AND TWO NEW SPECIES OF CULICID.E FBOM THE SUDAN type, others spatulate, lateral linear ones on the apical areas of the veins ; fork-cells short, the first submarginal longer and n,arrower than the second posterior, its base slightly nearer the apex of the wing, its stem nearly two- thirds the length of the cell, —f1ic,ro„iyin a/rica,,a,>^ <.p. i c^a^^v stem of the second posterior also about two-thirds the length of tlie cell, posterior cross-vein nearly twice its own length distant from the mid cross-vein. Length : 3"5 to 4 mm. Habitat : Nasser, Sobat Elver. Time of capture : July, 1907. Observations: Described from three $'s and three ,?'s, three specimens being nearly perfect. Collected by Dr. Wenyon. A very marked species clearly coming in a new genus. It bears some resemblance, however, to M(iiiso}iia (.'') nigra, Theob. [vide Second Report, p. 80, 190G), which was taken on the Blue Nile. It differs, however, in (i) the palpi of the $ not having two white bands ; (ii) the abdomen not having irregular apical white bands ; (iii) the tarsals of the first and second pairs of legs only having three, not four, bands ; (iv) the ungues of fore and mid legs being uniserrate and not simple as in M. nigra. There are also (v) differences in wing venation and the wing scales, the sixth vein not having large heart-shaped scales. Types in collection of the writer. Cidex sndanensis, nov. sp. (Plate XI.) '^"'''■*' Thorax tessellated with golden-brown and deep bi'own, showing traces of linear arrangement. Proboscis unhanded. Abdomen black, with broad basal creamy-white bands, biit two segments with yellow apical bands ; very hairy in the male. Legs brown, with narrow basal pale bands. Wings normal. ij! Head black, with pale narrow curved scales, pale upright forked scales in the middle, dark ones at the sides, flat lateral scales mostly dark, some creamy ones bordering the eyes and at the extreme sides, median chaitse golden, lateral ones black ; proboscis, palpi and antennae brown, the palpi with dull creamy apical scales. Thorax black, clothed with pale golden brown narrow-curved scales arranged somewhat in lines, the spaces between with scanty scales, thus giving a darker lined appearance, the scales becoming pale, almost creamy before the scutellum, which is also clothed with narrow-curved pale scales ; chsetse brown, with golden-brown apical reflections ; prothoracic lobes dark, with narrow curved pale scales ; metanotum brown ; pleura; dark, with patches of flat white scales. Abdomen black, with prominent creamy-white basal bands, the last two segments with apical yellow bands as well ; posterior border-bristles j)ale golden ; genital lobes black and long, hairy. Legs dark brown, femora and tibiae with some scattered pale scales and pale golden hairs, metatarsi and first two tarsals of fore legs, first three of the mid and all the hind with narrow pale basal bands ; knee spots yellow ; fore and mid ungues equal and uniserrate, hind '? PLATE XI Ctt/i'.v suifnni'iisis, n. sp. 1. Head of 9 3. Wing of 9 ft. (jenitalia of (5 2. Apex of 9 abdomen 4. Wing of S, 6. Genitalia of ^ under pressure, the spine being hidden 156 A NEW GENUS AND TWO NEW SPECIES OF ClTLTCin.T; FROIW THE SUDAN Wings with short fork-cells, the first slightly longer and narrower than the second, its stem more than half the length of the cell, its base slightly nearer the apex of the wing, stem of the second fork-cell about three-fourths the length of the cell ; posterior cross-vein not quite twice its own length distant fi'oni the mid. Length : 3'5 mm. $ Similar to %, but the abdomen very hairy. Tlie basal lobes of the genitalia are long and hairy and the long claspers are minutely bifid, the true termination of the clasper is hairy on the inside, and the other arm is formed by a stout chffita longer than the other projection. Fore and mid ungues unequal and uniserrate, hind equal and simple. Wings with short fork-cells, the first longer and narrower tlian the second, its stem about two-thirds the length of the cell, its base very slightly nearer the apex of the wing, stem of the second fork-cell nearly as long as the cell ; posterior cross-vein nearly twice its own length distant froui the mid. Length. : 3-5 mm. Habitat : Sudan (Harold H. King). Observations: Described from 3 $'s and 1 ^. The male genitalia are very marked ; no basally banded legged species of this appearance has before been found in the Sudan. One $ showed the abdominal basal bands narrower than in the type specimen and there are evidently basal white lateral spots which are hidden when the insects are dried up. Types in the collection of the writer. Cuh-.v stu/afif-iisis, nup 'I'lLi'; FiN(_:iiEs AN II \\'i;avki! J>iuns oi-' the Sl'J)AN BEING XOTES ON THE ClltOUl' (U)NTA[NINU THE BlUDS In.H-KIOUS TO (iKAIN (hlOl'S r.Y A, L. lluTLEK, F.Z.S., Jl.];.().U. Superintendent, Sudan Game Preservation Department Intkoduction A good deal of attention lias lately been called to the damage done to growing grain crops in the Sudan by vast flights of small birds, but we have little or no information as to the relative abundance, distribution, habits, and destructiveness of the different species by which the damage is done in different districts. To obtain this, some knowledge of the group of birds containing the destructive species is required, and, as there is little easily accessible literature on the local avifauna, a general sketch of the grain and seed-eating finches of the Sudan may be useful at the present time. This paper includes all the finches that I know of as occurring witliin our limits, though I believe that only a very small proportion of them will be found to be appreciably destructive to agriculture. As far as my knowledge goes the damage seems to be done entirely by the Sparrows (Pat^ser), and the extremely abundant Weaver Birds of the genera Hi/phantornis, XaiithopMlus, Quelea, and, in a smaller degree, Pyronelana. The full list here given may perhaps, however, lead to the recognition of others as criminals, especially in the remoter districts. It should not be forgotten that the birds which congregate on the crops, and do so rtiiiivas much damage at harvest time, feed their young almost entirely upon insects during the '"sect killers breeding season, and their immense numbers must make them an important factor in keeping down insect life. Their utility in this respect, however, seems to be somew-hat discounted by the fact that, while they visit the cultivated districts to do damage, they retire to forest and marsh country away from the cultivation at the period when they become useful as insect killers. The whole question of the relation of birds to agriculture in the Sudan would repay study, and in this it is hoped that the present paper may be of assistance. Paet I The Families, Suu-Families, and Geneka All the destructive species belong to the Fringilline, or Finch-like section of the Passerine Birds. In this group the bill, though varying in shape and propoi'tionate size, is strong and conical, with smooth unnotched cutting edges to the mandibles, and with the nostrils placed close to the forehead ; the wing has nine or ten primaries ; the tail consists of twelve feathers ; the legs and feet are strong, with the tarsus (the part of the leg between the " knee " and the foot) covered with scales in front and plain behind. The group falls naturally into two families : I. F ring ill idle, represented in the Sudan by Buntings, Sparrows, Serin Finches, and some others. 158 THE FINCHES AND WEAVER HIKUS OF THE SUDAN II. Ploceicke, containing the Weaver Birds, Waxbills, Widow Finches, and Bishop Finches, etc. In the Fringillidif tlie wing has nine primaries only, the tirst (outer) feather being about the same length as the second. In the Flocekhe there are fen primaries in the wing, the tirst of which is very small, not nearly so long as the second. The base of the bill extends back like a wedge into the feathers of the forehead. A glance at the outermost wing feather of a bird, or a count of the primaries, will therefore show to which of these families it belongs — the first step to " running it down." [A'ofe.--The word "primaries" may here require explaining. The wing of a bird folds like a Z. The "upper arm" (a strong single bone, the huinerii:<}, is directed backwards; next is the "forearm," a long joint composed of two bones (the radins and the ulna) forming the stem of the Z; then comes the "hand," formed of several small bones, and directed backwards again — the bottom of the Z. The large flight feathers which spring from the hand are the primaries, and those which spring from the forearm are the secondaries. They form a continuous series, but the joint between the forearm and the hand is the dividing point between them.] Most of the Fringillidse (Finches) found in the Sudan are resident, only a small proportion of the species being cold weather visitors. Most are of gregarious habits, and feed principally on grain and seed ; in the breeding season, however, they feed their young mostly upon insects, though partly on semi-digested seed. All except the Sparrows and Eock-Sparrows make open cup-shaped nests. They moult once in the year, in autumn, and have a partial change of colour in spring, due to the falling off of worn feather tips. (A good instance of this style of change is afforded by the cock of the common Sparrow ; in winter the black throat feathers are edged with whitish ; in the spring these edges are worn off, leaving the throat mark pure black.) The eggs of the Frmyillidie vary greatly, but are usually coloured and spotted or streaked. The Floceidw (Weavers) are all residents. All are gregarious, especially those of the genera Hyphantomis and Qiielea, some of which congregate in flocks literally of millions. They build covered nests, and their eggs vary greatly, being white, uniformly coloured, or spotted. They feed on grain or seeds and insects. Most of the sub-family Ploceinee breed in colonies, making beautifully w'oven nests suspended from trees or reeds. These are often retort-shaped, with the entrance at the bottom of a hanging tunnel. The Estrildinx pair off in the breeding season, making small domed nests in grass or bushes, and not breeding in colonies. The VidiiinH} build woven nests, often placed in grass or I'eeds. Many of them are polygamous, a small proportion of brightly plumaged males attending to a colony of nesting females. Family I. FringiUidie The locally occurring birds of this family are divided into two groups or sub-families: — 1. Fmberizinie (Buntings). 2. Fringillinas (Sparrows, Serin Finches, etc.). The Fmherizinee, or Buntings, of which the English Yellow-hammer is a well-known type, may be i-eadily recognised by the form of the bill, in which, when closed, the cutting edges of the mandibles do not quite meet together in the middle. There is also a sharp angle on the under surface of the lower mandible, close to the chin. The Buntings build open nests on or near the ground, and their eggs are generally marked with irregularly scribbled lines. THE FtNCHKS AND VVEAVEIt BIRDS OF THE SUDAN 159 Tlie FringUliuie are typical Finches, in which, unlike the Buntings, the cutting edges of the mandibles fit closely together when the bill is closed, and the pronounced angle on the under edge of the lower mandible, characteristic of that sub-family, is absent or slight. The very destructive Sparrows come in this grou]). The Finches of this sub-family make open nests, with the exception of the Sparrows, which build domed ones. The eggs are generally speckled. 1. Sub-family: Einhprizliuv Sub-family: hmberizirit^ The African Buntings are divided into two genera : — BDihei-iza (Typical Buntings). In these there is a white pattern on the outer tail feathers. Frimjillaria (Eock Buntings), in which there is no white on the outer tail feathers. Sub-fanuly : 2. Sub-family : b\-iii