IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 "i 1.0 ^^ m 
 
 ■tt Bii 12.2 
 
 I.I 
 
 ;s 1^ 12.0 
 
 Wmu 
 
 — 6" 
 
 •^ 
 
 
 
 V-v 
 
 HictogFaphic 
 
 uClBOCBS 
 
 Ckflporatian 
 
 23 WKT MAM STRUT 
 
 WmST<n,N.Y. 14ft0 
 (7l*)t73'4M» 
 
CIHM/ICMH 
 
 Microfiche 
 
 Series. 
 
 CIHIVl/ICIVIH 
 Collection de 
 microfiches. 
 
 Canadian Inatituta for Historical Microraproductions / Inatitut Canadian da microraproductions historiquas 
 
T«chnical and Bibliographic Notat/Notos tachniquaa at bibiiographiquaa 
 
 Tha 
 
 toti 
 
 Tlia inatituta haa attamptad to obtain tha baat 
 original copy availabia for filming. Faaturaa of thia 
 copy which may l>a bibliographically uniqua, 
 which may altar any of tha imagaa in tha 
 raproduction, or which may significantly changa 
 tha usual mathod of filming, ara chackad balow. 
 
 D 
 
 D 
 D 
 
 D 
 
 Colourad covara/ 
 Couvartura da coulaur 
 
 I I Covara damagad/ 
 
 Couvartura andommagia 
 
 Covars rastorad and/or laminatad/ 
 Couvartura rastaurte at/ou paliiculAa 
 
 Covar titia miasing/ 
 
 La titra da couvartura manqua 
 
 Colourad mapa/ 
 
 Cartas gtegraphiquas an coulaur 
 
 Colourad ink (i.a. othar than blua or black)/ 
 Encra da coulaur (I.a. autra qua biaua ou noira) 
 
 Colourad plataa and/or illuatrationa/ 
 Planchas at/ou illuatrationa an coulaur 
 
 D 
 
 Bound with othar matarial/ 
 Rali4 avac d'autraa documanta 
 
 Tight binding may cauaa shadowa or diatortlon 
 along intarlor margin/ 
 
 La r% liura aarrAa paut cauaar da I'ombra ou da ia 
 diatortlon la long da la marga IntAriaura 
 
 Blank laavaa addad during raatoration may 
 appaar within tha taxt. Whanavar poasibia, thaaa 
 hava baan omittad from filming/ 
 II aa paut qua cartalnaa pagaa blanchaa ajoutAaa 
 lora d'una raatauratlon apparaiaaant dana la taxta, 
 mala, lorsqua cala 4talt poaalbla. caa pagaa n'ont 
 paa At* filmAaa. 
 
 Additional commanta:/ 
 Commantalraa aupplAmantairaa: 
 
 L'Institut a microfilm* la maillaur axamplaira 
 qu'il lui a *t* poasibia da sa procurar. Las d*taila 
 da cat axamplaira qui sont paut-*tra uniquaa du 
 point da vua bibliographiqua, qui pauvant modifiar 
 una imaga raproduita. ou qui pauvant axigar una 
 modification dans la mAthoda normala da filnnaga 
 sont indiquAs cl-daaaoua. 
 
 r~1 Colourad pagaa/ 
 
 n 
 
 Pagaa da coulaur 
 
 Pagaa damagad/ 
 Pagas andom magmas 
 
 Pagaa rastorad and/oi 
 
 Pagaa raataurtea at/ou pailiculAaa 
 
 Pagaa diacolourad, stainad jr foxat 
 Pagas dteolortes, tachattes ou piquAaa 
 
 Pagaa datachad/ 
 Pagaa d^tachiwi 
 
 Showthroughy 
 Tranaparanca 
 
 Quality of prir 
 
 Quality intipala da I'lmpraaaion 
 
 Includaa supplamantary matarii 
 Comprand du material supplAmantaira 
 
 Only adition availabia/ 
 Saula idition diaponibia 
 
 I — I Pagaa damagad/ 
 
 I — I Pagaa rastorad and/or laminatad/ 
 
 r~l Pagaa diacolourad, stainad jr foxad/ 
 
 I I Pagaa datachad/ 
 
 r~1 Showthrough/ 
 
 r~n Quality of print varlas/ 
 
 I I Includaa supplamantary matarial/ 
 
 I — I Only adition availabia/ 
 
 Tha 
 pos 
 oft 
 film 
 
 Ori( 
 bag 
 tha 
 sior 
 oth( 
 first 
 sior 
 oril 
 
 Tha 
 sha 
 TIN 
 whi 
 
 IMai 
 diff 
 anti 
 bag 
 righ 
 reqi 
 mat 
 
 Pagaa wholly or partially obacurad by arrata 
 allpa, tiaauaa, ate, hava baan rafilmad to 
 ansura tha baat poaalbla imaga/ 
 Las pagaa totalamant ou partlallamant 
 obaeurclaa par un fauillat d'arrata, una palura, 
 ate, ont 4t4 fllmAaa k nouvaau da fapon A 
 obtanir la maillaura imaga poaalbla. 
 
 ThIa Itam ia filmad at tha raduotion ratio chackad balow/ 
 
 Ca documant aat filmi au taux da rMuctton indiqu* cl-daaaoua. 
 
 10X 14X 18X 22X 
 
 26X 
 
 aox 
 
 12X 
 
 16X 
 
 aox 
 
 2IX 
 
 28X 
 
 32X 
 
Th« copy film«d h%n h«* lM«n r«produc«d thanks 
 
 to the gantrosity of: 
 
 Mill! MMnorial Library 
 MoMMttr Unhrtriity 
 
 L'oxomplairo filmA fut raproduit griea A la 
 gAntroait* da: 
 
 Mills Mtmorial Library 
 
 McMMltr Univtriity 
 
 Tha imaga* appaaring hara ara tha baat quality 
 posslbia eonsidaring tha condition and iagibility 
 of tha original copy and in kaaping with tha 
 filming contract spacificatlona. 
 
 Original copies in printad papar covers ara filnnad 
 beginning with tha front covar and ending on 
 the last page with a printed or Illustrated impres- 
 sion, or the back cover when appropriate. All 
 other original copies ara filmed beginning on the 
 first page with a printed or iiiustriited impres- 
 sion, and ending on the last page with a printad 
 or illustrated impression. 
 
 Les images suivantea ont MA reprodultes avac la 
 plus grand soin, compta tenu de la condition at 
 da la nattetA da rexemplaire filmA, et en 
 conformity avac las conditions du contrat da 
 filmage. 
 
 Les exemplaires orlglnaux dont la couverture en 
 papier est ImprimAe sent filmte en commenqant 
 par la premier plat et en terminant soit par la 
 dernldre page qui comporte une empreinte 
 d'Impression ou d'lllustration, soit par la second 
 plat, salon le cas. Tous les autras axempiairea 
 orlglnaux sont filmAs en commenpant par la 
 pramiAre page qui comporte une empreinte 
 d'Impression ou d'lllustration at en terminant par 
 la dernlAre page qui comporte une telle 
 empreinte. 
 
 The last recorded frame on each microfiche 
 shall contain the symbol — ^> (meaning "CON- 
 TINUED"), or the symbol ▼ (meaning "END"), 
 whichever applies. 
 
 Un des symboSes sulvants apparaftra sur la 
 darnlAre image de cheque microfiche, seion le 
 cas: le symbols — »- signifie "A SUIVRE", le 
 symbols ▼ signifie "FIN". 
 
 Maps, plates, charts, etc., may be filmed at 
 different reduction ratios. Those too large to be 
 entirely included in one exposure ara filmed 
 beginning in the upper left hand corner, left to 
 right and top to bottom, as many frames as 
 required. The following diagrams illustrate the 
 method: 
 
 Les cartes, planchaa, tableaux, etc., peuvent Atre 
 fllmAs A des taux da rAduction diff Arents. 
 Lorsque le document est trop grand pour Atra 
 raproduit en un seul clichA, 11 est filmA A partir 
 de Tangle supArleur gauche, de gauche A droite, 
 et de haut en bas, an prenant le nombre 
 d'images nAcessaira. Les diagrammas suivants 
 illustrent la mAthoda. 
 
 1 2 3 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 32X 
 
WANDERINGS IN SOUTH AMEKICA. 
 
w. 
 
"^io^^^K Vtui.p 
 
 -HHnK, 
 
 I mNDEMNGS IN 80UTH AMERICA, 
 
 THE NORTH-WEST OF THE UNITED STATES, 
 
 AND THE ANTILLES, 
 
 IN THE YEARS 1812, 1816, 1820, & 1824. 
 
 Witli Oilfflnal Instructions for the perfect preservation of Birds, Et<;. 
 for Cabinets of Natural History. 
 
 «Y 
 
 CHARLES AVATERTON, Esq. 
 
 NEW EDITION. 
 
 mu\i, bitlj iiograij^ical |ntrobuttioir anlr (f.vplmuiloriT |nh,v, 
 
 BY THK 
 
 REV. J. G. WOOD. 
 
 WITH ONE HUNDRED ILLUSTRATIONS. 
 
 I'onbon : 
 MACMILLAK AND CO. 
 
 1879. 
 
 The Uiijht uf Translation U Iics,:md. 
 
I.OKDOX ; 
 
 K. Ct.AV. PONS, AND TAVI.ni; 
 
 niu:\n STni;i:T iui.l, r.c. 
 
 lAcMASIER UNlVERSklY LIBRARY 
 
PREFACE. 
 
 It 
 
 M.VNY years ago, wliile barely in my " teens," I had the 
 good fortune to fall in with Waterton's Wanderings, then 
 newly placed in the school library. The book fascinated 
 me. Week after week I took it out of the library, and 
 really think that I could have repeated it verbatim from 
 beginning to end. It was a glimpse into an unknown 
 world, wliere I longed to follow the Wanderer, little 
 thinking that I should ever have the privilege of visiting 
 him in his wonderful Yorkf hire home. I looked upon 
 Waterton much as the pagans of old regarded their 
 demi-gods, and not even Sinbad the Sailor was so in- 
 teresting a personage to me as Waterton the Wanderer. 
 
 But there was one drawback to the full enjoyment 
 and comprehension of the book. It mentioned all kinds 
 of animals, birds, and trees, and I did not know what 
 they were, nor was there any one who could tell me. I 
 did not know what a Salempenta was, except that it 
 
 W 
 
VI 
 
 IM{KFA('K. 
 
 was good to cat. It might bo a monkey, a tisli, or u 
 fruit. Neither could T identify the Couanacouclii, 
 Labarri, Camoudi, Duraquava, Hoiitoii, or Karabimiti, 
 except tliat tlie tliree first were snakes and the three last 
 wore birds. 
 
 It was certainly pleasant to learn that tlie traveller 
 in Guiana would be awakened by the crowing of the 
 Hannaquoi, but there was no one who could tell me 
 what kind of a bird the Hannaquoi miglit be. Then, 
 as to trees, I did not know the Siloabali, or tlie Wallaba, 
 or even the Purple-lieart, nor how the last mentioned 
 tree could be made into a AVoodskin. I wanted a guide 
 to the Wanderings, and sucli a guide I liave attempted 
 to supply in the "Explanatory Index." I l)elieve that 
 there is not a single living creature or tree mentioned 
 by Waterton concerning which more or less information 
 cannot be found in tliis Index. 
 
 Tlio IVanderhhf/s I have left untouched as Waterton 
 wrote them, not adding or altering or cancelling a 
 syllable. They constitute, so to speak, the central 
 brilliant of a ring, round which are arranged jewels 
 of inferior value, so as to set off the beauty of the 
 principal gem. 
 
 The plan of arrangement is as follows : First comes 
 a short biography of AVaterton as the Wanderer, and 
 then a memoir of Waterton at home. Next come the 
 Wcmdcritigs, exactly as he wrote them. Tlien there is 
 an Explanatory Index, and lastly a few remarks on the 
 
 V •? 
 
PUKFACI-:. 
 
 VII 
 
 system of Taxidormy wliich lie cieatcd, and in whicli 
 he gave me personal instruction. 
 
 1 Imve much pleasure in recording my obligations to 
 Kdmund Waterton, Esq., who kindly permitted access 
 to tlie old family records, which he is now arranging 
 lor publication. Also to Dr. V. L. Sclater, Secretary of 
 the Zoological Society, for the assistance wliich he 
 rendered in identifying several of the birds; and to 
 J. Britten, Esq., of the British Museum, for the great 
 pains which he took in ascertaining the names of some 
 of the Guianan trees, without which names the work 
 would have been imperfect. 
 
PREFACE TO THE FIRST EDITION. 
 
 I OFFEU this book of Wmuh'riiujs witli ii hesitating' 
 hand. It has little merit, and niiist make its way through 
 the world as well as it can. It will receive many a jostle 
 as it goes along, and perliaps is destined to add one more 
 to the number of slain, in the field of modern criticism. 
 r>ut if it fall, it may still, in death, be useful to me ; for, 
 should some accidental rover take it up, and, in turning 
 over its pages, imbibe the idea of going out to explore 
 Guiana, in order to give the world an enlarged descrip- 
 tion of that noble country, I shall say, " fortem ad fortia 
 misi," and demand the armour ; tliat is, I shall lay claim 
 to a certain portion of the honours he will receive, upon 
 the plea, that I was the first mover of his discoveries; 
 for, as Ulysses sent Achilles to Troy, so I sent him to 
 Guiana. I intended to have written much more at 
 length; but days, and months, and years, have passed 
 away, and nothing has been done. Thinking it very 
 probable that I shall never have patience enough to sit 
 down and write a full account of all I saw and examined 
 
im{i:kack 'I'O tiiI': fii^st edition. 
 
 ill those remote wilds, I give up the intention of doing 
 so, and send forth this account of my Wanderings, just 
 as it was written at the time. 
 
 If critics are displeased with it in its present form, 1 
 beg to observe, that it is not totally devoid of interest, 
 and that it contains something useful. Several of the 
 unfortunate gentlemen who went out to explore the 
 Congo, were thankful for the instructions they found in 
 it ; and Sir Joseph Banks, on sending back the journal, 
 said in his letter, " I return your journal, with abundant 
 thanks for the very instructive lesson you have favoured 
 us with this morning, which far excelled, in real utility, 
 everything I have hitherto seen." And in another letter 
 he says, " I hear with particular pleasure your intention 
 of resuming your interesting travels, to which natural 
 history has already been so much indebted." And again, 
 " I am sorry you did not deposit some part of your last 
 harvest of birds in the British Museum, that your name 
 might become familiar to naturalists, and your unrivalled 
 skill in preserving birds be made known to the public." 
 And again, " You certainly have talents to set forth a 
 book, which will improve and extend materially the 
 bounds of natural science." 
 
 Sir Joseph never read the third adventure. Whilst 
 I was engaged in it, death robbed England of one of her 
 most valuable subjects, and deprived the lloyal Society 
 of its brightest ornament. 
 
C N T E N T 8. 
 
 BIOGRAPHY 
 
 CHAl'TEK I. 
 
 Autobiography of Watorton — Descent from Sir Thomas More— Twenty- 
 soventli Lord of AValton, and sixteenth in descent from John 
 AVaterton-^^Religious faith of the familj' — Persecutions of Roman 
 Catholics and confiscation of the estates — Double taxes and fines — 
 IJirth and early life — Escapades at Tudhoe — The cow and the wash- 
 ing-tub — Removal to Stonyhurst — Birds'-nesting, a chase and a pig- 
 stye — Good advice from one of the fathers — Parting with Stonyhurst 
 — First voyage to Cadiz — The apes at Gibraltar — Habits of tin? 
 animals — Stay in Malaga — Acquirement of Spanish — Projected 
 visit to Malta — Advent of the plague — Seized with the disease and 
 recovery — Closing of the ports — A hazardous and carefully-planned 
 escape — Preparations on board ship— The opportunity seized — 
 Escape successful — Death of an unch- — Discovery of an old friend 
 — Failing health — A'oyag*^ to Dcmerara — Death of his father and 
 succession to the family estates 
 
 rsc.v. 
 
 1 14 
 
 CHAPTER II. 
 
 Journey to Orinoco with despatches — A<lventure with a venomous snake 
 — An involuntary bath— A huge ('ayman— The Labarri snake — 
 Dinner party in Angostura — A too liberal table— The Governor's 
 uniform — Dining in shirt-sleeves — A more sensible uniform — Publi- 
 cation of the li'aiulirinjjs — Reception by the critics— Sydney 
 Smitli — Swainson's criticism upon the cayman — Truth in the garb 
 of fiction — Watertou's style of writing — Quotations— His favourite 
 aithors — Sense of humour— How he answered the critics — Chargt^ 
 of t;-^""<tncity — How he was eccentric — Travels on the continent — 
 Shipwreck— Gallant conduct of Prince Cauino— Iiost by gold — Fall 
 into Dover harbour and narrow escape — The lancet and calomel — 
 .ludgment of the vampire — A bad wound — Rare feet and bad pave- 
 ment — Mode of cure — Accidents at iiome— Gunshot wound -Si'vere 
 fall and dangerous injuries— Crowther the bont^-setter — A i>ainful 
 operation -Ultimate r<!covery- A characteristic warning . . 15 
 
 ;u 
 
XII 
 
 CONTEXTS. 
 
 CHAl'TKll III. 
 
 1 
 Magtiificence and uiouey — Watertoii's mode of life and persoual ex- 
 penses — Sleeping on planks — Hia visits to the cluipel —The " morning 
 gun " — The razor and the lancet — Keduction of tli(! family estates — 
 His work at Walton Hall — Natural advantages of the place — The 
 wall and its cost — Uargees and their guns — Instinct of the herons — 
 Herons and fish-ponds — Drainage of the ponds— The moat extended 
 into a lake—Old Gateway and Ivy-Tower — Siege hy Oliver Crom- 
 well — Tradition of a musket-ball —Drawbridge ai>d gateway in the 
 olden times — Tradition of a canon-ball— IJoth ball and canon dis- 
 covered — Sunken plate and weapons — El-Iio at "Walton Hall — AVest 
 view of lake — How to strengthen a bank — Pike-cutchin^' — Cats and 
 pike — Spot where "Waterton fell 35- 
 
 -48 
 
 CHAPTEK IV. 
 
 Love of trees — Preservation of damaged trees — How trees perish — 
 "Wind and rain — Self-ri!storative powers of the bark — Hidden foes 
 — The fungus and its work — Use of the woodpecker and titmouse — 
 How to utilize tree-stumps — The cole titmouse — Owl-house and 
 seat — Dry-rot — "When to paint timber — Oaken gates of the old 
 tower — Connnand over trees — How to make the holly grow <iuickiy 
 — The holly as a hedge-tree — Pheasant fortresses — iVi-tificial 
 pheasants — The poachers outwitted — "Waterton's power of tree 
 climbing — An aerial study — Ascending and descending trees— 
 C!hurch and Statu trees— The yew — A protection against cold winds 
 — Yew hedge at back of gateway— The Starling Tower — Familiarity 
 of the birds— Th(! Picnic or Grotto — 'Waterton's hospitality—" The 
 Squire " — A decayed mill and abandoned stone — The stone lifted 
 off the ground by a hazel nut 49- 
 
 ■71 
 
 CHAPTER V. 
 
 The Squire't) " dodges"— Tiie " cat-holes "—The dove-cot— Pigeon-shoot- 
 ing matches and mode of supplying the birds — AVaterton's pigeon- 
 house, external and internal — Pigeon-stealers baffled — Anaugemeut 
 of pigeon-holes — Ladders not needed — How to feed pigeons econo- 
 mically — Rats and mice in the garden — The poison-bowl and its 
 safety — Sunken mousetrap — Gates and chains — The carriage-jjond 
 — AVaterton's antipathy to scientific nomenclature — Advantage of 
 such nomenclatm'e as an assistant to science — Popular autl local 
 names — Colonists and their nomenclature — Zoology gone mad — 
 Complimentary nomenclature— The fatal accident in the park — 
 AVaterton's last moments and dealh-The last voyage and funeral 
 — Epitaph written by himself — The new cross, and i)lace of burial, 7-— 8tJ 
 
CCtNTENTS. 
 
 xiu 
 
 FIRST JOURNEY. 
 
 CHArTER I. 
 
 Object, of the JJ'i/iiJcrim/s — Domcrnm IJ.—Snlia — Toiiran — Forest Trocs 
 — rarasitos— Bush-ropo— ] icil monkey —AVild animals— Slotlis- 
 Yenomons snakes — Liznrds — Bell-bird — Houtou — Insects — Dog 
 poisoned with AVourali— Falls— Esscquil)o E.— llnpid de;\ny— Falls 
 of the Essequibo— aiaconshia— Awhitc recluse— Tlu; AVaterinanuna 
 —A savage financier— Tlic Jabiru— Ants' nests— Fort St. Joachini 
 — Lake Pariniii m7_ 
 
 -12.-, 
 
 CHAPTER II. 
 
 The lyiaconshi Indians— Poison vendors— apparent failuri! of poison- 
 Collecting materials for wonrali-poison — Preparing the poison — 
 Superstitions— The blow-pipe gun— The Ourali— The Saniourah— 
 Silk-grass— Acuero fruit— Coucourito palm— Wild cotton— Ar .ws 
 — Quivers— Jaw of Pirai — Packing tlie arrows — Cotton basket — 
 Gun sight made of Aeouri teeth— Poisoned fowl— Suspending the 
 guns — The bow— Ingenious arrows — Small quivers — A wild hog shot 
 — Utilization of indigenous products 12Q- 
 
 130 
 
 CHAPTER III. 
 
 Operatien of the Wourali— Its effects on the Ai, or Three-toed Sloth— 
 Deatli resembling sleep— A poisoned Ox— Poison proportionate to 
 size of animal— Alleged antidotes— An Indian killed by his own 
 arrow- Ligatures and the knife— Descent of the Essequibo— Skill 
 of the boatman— Tiie Buccaneers— Tertian ague— Experiments with 
 "Wouriili— Value of a ligature— Artificial respiration— Long life and 
 cpiet death of Wourali— When good King Arthur ruled tliis land - 
 Return of health 140 _14- 
 
 REMArJKS 
 
 14S— ir>;{ 
 
 SKCOND JOURXEY 
 
 CHAPTER I. 
 
 From Liverpool to Pernambuco— Stormy petrels— Tropical zoology— 
 Flying-fish — Bonito, Albicore, and " Dolphin "- Frigate bird- 
 Arrival at Pernambuco— The expelled Jesuit— Pombal, the Captain- 
 General— Southey's history of Brazil -Botanical garden- Sungredo 
 Buey—Rattle.snake— Narrow escape— Rainy- S.il for Cayenne- 
 Shark -catching ' . _ _ . l,-,4-_l(5S 
 
CONTENTS. 
 
 CHAl'TEK II. 
 
 Arrival at Cayenne — Flamingos — Curlews, i^c. -Vegetable productions 
 of Cayenne — LaGabrielle — Cock of the Rock— Grand Gobe-mouche 
 —Surinam — The Coryntin — New Amsterdam — Stabroek, now George 
 Town — Produce of Demerai'a- Slavery — A traveller's necessaries — 
 AValkiug barefoot — The best costume — Hinnming-birds — Cotinga — 
 Campanero, or Bell-bird — Toucans, or Toucanets — Beak of the 
 Toucan — Evanescence of the colours — The only mode of preserving 
 them 169 
 
 I'Aut: 
 
 184 
 
 CHAPTER III. 
 
 Tlie Houtou — Curious habit of trimming the tail and feathers— its 
 habits — The Guianan Jay — The Boclora — Slight attachment of the 
 feathers — The Cuia — Rice-birds — Cassiques, their habit of mockery 
 — Pendulous nests — Gregarious nesting of diffei-ent species — AVood- 
 peckers of America and England — Kingfishers — Jacamars and their 
 fly-catching habits — Troupiales and their songs — Tangaras — Mani- 
 kins — Tiger-birds — Yawaraciri — Ant Thrushes — Parrot of the Sun 
 — Aras, or Macaws — Bitterns — Egret, Herons, etc. — Goatsuckers — 
 "Whip-poor-Will — Superstitious — Tinamou.s — Powis and Maroudi — 
 Horned Screamer — Trumpeter — King Vulture — Anhinga — Dangers 
 of travel — Quartan ague 185- 
 
 -1'07 
 
 THIRD JOURNEY. 
 
 CHAPTER 1. 
 
 From the Clyde to Demerara—^ Yellow fever — A deserted Plantation — 
 Black John — Medicines for tropical climates — Bats — The lancet — 
 Severe accident and recovery — A primitive spear — History of the 
 Sloth — An inhabitant of the trees -Structure of the limbs -A domes- 
 ticated Sloth — A life of suspense—Structure of the hair — Mark on 
 the back — Capture of a Sloth— Release and escape— Ants — Ant- 
 bears — The great Ant-benr— Its powers of defence— Attitude when 
 standing — How it catches its prey — Glutinous saliva — The Vampire 
 and its habits — Bleeding gratis — Coushie Ants — Armadillo and its 
 habits— Tortoise -Eggs of Tortoise and Turtle .... :.'08 
 
 •i\n 
 
COXTKNTS. 
 
 xv 
 
 CHAl'TEK II. 
 
 The Vttuilla — Meauing of the word— Small (.'ayiiian— Daddy Ciuaslii 
 Wasps — Veiiomous reptiles aiidwijd beasts — I'oisou of the Labarri 
 siuike— Experiment with a Labarri — The Bute-rouge- The C'hegoe 
 — Its iiest — Ticks, and how to get rid of them — The live trilies — 
 Their habitations and mode of life— I'iwarri — The I'ee-ay-man — A 
 nation without a liistory — Runaway negroes— Mr. Edmoustone and 
 his services — Wounded warriors — Valour rewarded — Education — 
 Cliaracter of the native— Skill in hunting— A beatl almanac — 1'he 
 sun as a compass Thinness of population 2^1'- 
 
 ^47 
 
 CHAPTER III. 
 
 Discovery of a large Coulacanara snake— A Bush-master — Stag swal- 
 loweil by a Boa — Negroes and the siuike — Arrangements for the 
 attack — Tlie snake struck — Carrying off the enemy — A snake in a 
 bag — An unquiet night — Dissection of the snake — Daddy Quashi 
 and his dread of snakes — Capture of a Coulacanara — Vultures and 
 their food — Habits of Vulturis — The Aura vulture — Black vultures 
 — Severe blisters — An imiuisitivc Jitguar Fish shooting- Goat- 
 suckers and Campancro '2-iS — 262 
 
 CHAPTER IV. 
 
 Fishing for a Caynuiu— A shark-hook useless— Sting-rays — Turtle and 
 Guana nests — Numbers of eggs— Another failure — Meeting a Jaguar 
 —Guard against fever — More failures — A native hook and way of 
 baiting — The ( -ayman's dinner-bell — Caught at last — How to secure 
 the reptile — Mounting a Cayman— An improvised bridle— Skin and 
 teeth of the Cayman — Embarkmont for England -Collision with 
 the Custom House :2(JiJ 
 
 28i 
 
 FOUliTH JOURXKY. 
 
 CHAPTER I. 
 
 Tliree years in England — Sail for New York — Nomenclature — Altera- 
 tion of scenery — A sprained ankle — Magnificent cure— Feats of 
 climbing — Quebec — Irish emigrants — Ticondtroga — Saratoga — 
 I'hiladelphia — White-headed Eagle — Form and Fashion— C^liuiatt 
 —^Forebodings of the civil war- Sail fcr Antigua . . 2t?5- 
 
 auJ) 
 
XVl 
 
 CUNTKNTS. 
 
 CHAr'J'KK il. 
 
 PAdK 
 
 Arrival at Antigua — Dominica— Frogp and Hunnning-birds — Martinico 
 — Diamond rock — Barbadoes — Quashi and Venus — The Alien Bill — 
 Sail for Dtmcrara — More about tlie Sloth— Scarlet Grosbeak — 
 Crab-eating Owl — Sun-heron— Feet of the Tinaniou — Vampires 
 again — The Karabimiti Humming-bird — The Monkey tribe — The 
 Ked Howler — Koast monkey — Tlu) Nondescript — Altered physi- 
 ognomy — Gold and silver mines — Changes pf Government — Politics 
 —India-rubber — An ingenious deception 310—^24 
 
 ON PRESERVING BIRDS FOR CABINETS OF NATURAL 
 
 HISTORY. 
 
 Faults in bird-stuffing— T(ols required — Knowledge of anatomy- - 
 Attitudes of birds— Flow of the plumage— How to skin a bird — 
 Inserting cotton — Killing wounded birds — Stuffing a hawk — The 
 first incision — The skin to be puslied, not pulled — Arrangement of 
 wings — Modelling the body — Spreading the tail— Constant attention 
 required — Strength and elasticity — Value of corrosive sublimate — 
 Experience and patience 335 — 350 
 
 EXPIj\NATORY INDEX 351-494 
 
 TAXIDERMY 495—510 
 
 INDEX 511-5i'0 
 
 ] 
 
 t 
 
 i 
 
 fi 
 h 
 
 t] 
 
 ai 
 
PA(ir. 
 
 BIOGKAPHY. 
 
 CHAPTER I. 
 
 ""tvSc^/of mt'"--^""* ^^""^ «'^ ^^-- More.-Twe„ty- 
 seventULoidofW niton, and sixteenth in descent from John Waterton 
 -Eehgious fa:th of the family. -Perseeutions of Roman CathoS a "i 
 confiscation of the estates-Double taxes and fines.-Bi land ea" v 
 ife -Escapades atTudhoe.-The cow and the washing.tub.-RemoS 
 
 rom ::;f Tt h^^^^^^^^^ ^ ^'-« -^^ ^ Pigst^-Oood x 
 
 rT2 T. ■'• " ^ ''''*'"S ^'**^ Stonyhurst.-First voya«e to 
 
 Mat;r r '^'' '* Gibraltar. -Habits of the animals. - Stay Jn 
 Ma aga.-Acqun-ement of Spanish-Projected visit to Malta.-Advei^ 
 of the plague.-Se.zed with the disease and recovery.-Clo.in.. of the 
 iTrA.Tl^'"' "'J carefully.planned escape.'-Pr.! a::tiol t 
 an ir^'n ^i^P*'*:*"'"^^ seized. -Escape successful -Death of 
 
 an uncle.-Discovery of an old friend.-Failinc health ~V~ t„ 
 Demerara.-Death of his father and succession to^Llmi'ylS. 
 
 the author has afforded but little account of himself but ^'""^'''y- 
 in the volumes of his Essays, and some of his Letter, 
 he has fortunately given a sufficiency of information to 
 
 hH tf r ^ ""^''^'^ ''''^'''^^'^ '''^ ^"« birth to 
 IS death. His was a very long life, and as he considered 
 that life as a sacred trust, he never wasted an hour of it. 
 
 Waterton was the representative of one of the most 
 ancient English families, and was justly proud of hL ' 
 
 n 
 
 t 
 
15I0GRAPIIY. 
 
 Sir T. 
 More, 
 
 Pediijrcc. 
 
 descent from Sir Thonifis More. A clock which had be- 
 longed to that great ancestor is still in existence, and 
 occupied a place of honour on the upper landing of the 
 central staircase of Walton Hall. It is but a little 
 lock, and has only a single hand, but it keeps time as 
 well as ever, and the sound of its bell is so clear, that it 
 can be heard at a considerable distance from the house. 
 He mentions in his own quaint way, that if his ancestoi's 
 had been as careful of their family records as Arabs are 
 of the pedigrees of their horses, he .:'^.l^^ht have been able 
 to trace his descent up to Adam and Eve. 
 
 The following account of the Waterton family is taken 
 from the Illudrntcd London Axws of June 17, 18G5, and 
 has been revised by a member of the house. 
 
 "The good and amiable old Lord of Walton, Charles 
 Waterton, better known for miles around his ancestral 
 domain as " the squire," was tlie representative of one 
 of our most ancient untitled aristocratic families, and, what 
 is more deserving of record in these days, in the male 
 line. 
 
 " His ancestor, Eeiner, the son of Norman of Normandy, 
 who became Lord of Waterton in 1159, was of Saxon 
 origin. The Watertons of Waterton became extinct in 
 the male line in the fifteenth century, when their vast 
 possessions passed away, through Cecilia, wife of Lord 
 Welles and heiress of her brother, Sir Eobert Waterton, 
 to her four daughters and co-heiresses, who married, 
 respectively, liobert, Lord Willoughby de Eresby, Sir 
 Tiiomas ' Dymoke, Thomas Laurence, Esq., and Sir 
 Thomas Delaware. 
 
 " Sir John Waterton was high sheriff of Lincoln in 1401, 
 
 and master of the horse to Henry V. at Agincourt. Sir 
 
 Lady of Eobert, his brother, whose wife was a lady of the garter, 
 
 "^'^''* vvas governor of Tontefract Castle while Eichard II. was 
 
BIOGRAPHY 
 
 8 
 
 confined there : he had been master of the horse to Henry 
 IV. Sir Hugh, another brother, held high offices of state. 
 Charles Waterton, in whom the representation of his 
 ancient house was vested, was descended from Richard, 
 second son of William Waterton, Lord of Waterton, who 
 died in 1255. In 1435 John AVaterton married the 
 heiress of Sir William Ashenhull, and became Lord of 
 Wdlton and Cawthorne, y«/'c uxoris. 
 
 " Walton formed part of the Honour of Pontefract, of Pontc/ract 
 which Ashenhold, a 2"xon thane, was the Lord, and which 
 was held by his son Ailric, in the reign of S. Edward 
 the Confessor. At the Conquest it was given by William 
 the Norman to one of his followers, Ilbert de Lacy, who 
 granted it back again to Ailric, father of Suein. Adam, 
 the son of Suein, Lord of Brierley, Cawthorne, and Walton, 
 was the founder of the priory of Monk Kretton, and left 
 two daughters and co-heiresses, Amabil and IMatilda. The 
 former had Walton and Cawthorne, and became the wife of 
 William de Nevile. They had one daughter and heiress, 
 who married Thomas, the son of Philip de Burgh. Walton 
 and Cawthorne remained in the possession of the De 
 Burghs for seven generations, and then passed with the 
 co-heiress of Sir John de Burgh to Sir William Ashenhull, 
 whose heiress conveyed it to John Waterton in 1435. 
 
 "Thus Mr. Waterton was twentv-seventh Lord of 
 Walton, and sixteenth from John Waterton, who acquired 
 that lordship. There was a grant of free warren at Walton 
 in the reign of Edward I., and a license to crenellate in 
 1333. Without reference to the numerous distinguished 
 alliances of his ancestors, it may be interesting to state 
 that Mr. Waterton, through distinct sources, traced his 
 descent several times over from S. Matilda, Queen of 
 Germany ; S. Margaret of Scotland, S. Humbert of Savoy, 
 S. Louis of France, S. Ferdinana of Castile, and Wladimir 
 
 B 2 
 
 Distin- 
 gaished 
 ancestors. 
 
BIOGRAPHY. 
 
 Ucfnrma' 
 tion. 
 
 1^^ 
 
 il 
 
 Corrci've 
 Acts. 
 
 the Great, called S. Wladiinir of Russia, and Anne, called 
 S. Aime of lUissia. Through his grandmother he was 
 ninth in descent from Sir Thomas More." 
 
 The Watertons fared but badly in the stormy times of 
 the Reformation, and, preferring conscience to property, 
 tliey retained their ancient faith, but lost heavily in this 
 world's goods. The many coercive acts against the Roman 
 Catholics naturally had their effect, not only on those 
 who actually lived in the time of the Reformation, but 
 upon their successors. A Roman Catholic could not sit in 
 parliament, he could not hold a commission in the army, 
 he could not be a justice of the peace, he had to pay 
 double land-tax, and to think himself fortunate if he had 
 any land left on which taxes could be demanded. He was 
 not allowed to keep a horse worth more than five pounds, 
 and, more irritating than all, he had either to attend the 
 parish church or to pay twenty pounds for every month of 
 absence. In fact, a Roman Catholic was looked upon and 
 treated as a wholly inferior being, and held much the same 
 relative position to his persecutors as Jews held towards 
 the Normans and Saxons in the times of the Crusades. 
 
 Within the memory of many now living, the worst of 
 the oppressive acts have been repealed, and Roman Catholics 
 are now as free to follow their own form of worship as 
 before the days of Henry VIII. They have seats in 
 parliament and on the bench, they hold commissions both 
 in the army and navy, and all the petty but galling inter- 
 ferences with the details of their private life have been 
 abolished. 
 
 Still, Waterton was, during some of his best years, a 
 personal sufferer from these acts, and they rankled too 
 deeply in his mind to be forgotten. Hence, the repeated 
 and mostly irrelevant allusions in his writings to Martin 
 Luther, Henry VIII., Queen Bess, Archbishop Cranmer, 
 
BIOGRAPHY. 
 
 5 
 
 Oliver Cromwell, Charles Stuart, " Dutch William " (mostly 
 associated with the " Hanoverian " rat and the- national 
 debt), and other personages celebrated in history. 
 
 Deeply as he felt the indignities to which ho and his 
 family and co-religionists had been subjected, and fre- 
 quently as he referred to them, both in writing and con- 
 versation, he never used a worse weapon than irony, and 
 even that was tempered by an underlying current of 
 humour. He had felt the wounds, but he could jest at 
 the scars. 
 
 On principle he refused to qualify as Deputy-Lieu- 
 tenant and magistrate, because he had been debarred from 
 doing so previously to the Emancipation Act. His son, 
 however, serves both offices. 
 
 Religious' 
 
 Born in 1782, he spent his childish years in the old Birth. 
 mansion and grounds of the family, and at a very early age 
 displayed those powers of observation, love of nature and 
 enterprise, which enabled him to earn a place among the 
 first order of practical naturalists both at home and abroad. 
 
 At ten years of age he was placed under the Rev. A. Tudhoe. 
 Strong's care, in a school just founded at Tudhoe, a village 
 near Durham. From Waterton's reminiscences, his in- 
 structor seems to have inclined to the severe order of dis- 
 cipline, and to have been rather liberal of the birch, of 
 which instrument Waterton had his full share. His 
 account of storming the larder for the support of hungry 
 inmates; of the anxious glances- which he cast in the 
 morning to judge by the master's wig of the state of his 
 temper ; and of being captured in the very act of getting 
 through a barred window, is exceedingly humorous. 
 
 He also relates two anecdotes, both telling against him- Tmo 
 self, and both prospective, as it were, of the celebrated 
 fact of riding on the back of a cayman and of his ship- 
 
IJTOGRAPIIY. 
 
 The cow 
 and thi; 
 
 Vshaio 
 Colk(jc. 
 
 wreck, lie was *' dared " by his comrades to get on the 
 back of a cow, which he did, but less fortunate than in 
 his cayman adventure, was ignominiously thrown over 
 her burns. lie also took it into his head to get into a 
 wasliing-tub, and take a cruise in the horse-pond ; but lost 
 his balance at the sudden appearance of the master, and 
 was overturned into the muddy water. 
 
 The whole of the account of his Tudhoe school ex- 
 periences is given in a collected volume of his Essays and 
 Letters (F. Warne & Co.), edited by Mr. N. Moore, who 
 had the sad privilege of being with him when he met with 
 liis fatal accident, and by his sofa when he died, about 
 thirty-eight hours afterwards. 
 
 Tudhoe then being only a preliminary school, though it 
 has since developed into Ushaw College, Waterton was re- 
 moved at fourteen years of age to Stonyhurst, where he 
 was one of the first pupils. This establishment, then a 
 comparatively small one, was conducted by the English 
 Jesuits who had been driven from their home at Li(3ge. 
 Of them Waterton always spoke with reverence and 
 affection, and his life at Stonyhurst was a singularly 
 happy one. 
 
 At first, his ingrained propensity for enterprise led him 
 into trouble, and one adventure is too good not to be 
 narrated in his own words. His account of it is another 
 example of the way in which he enjoyed telling an 
 anecdote against himself. 
 
 " At Stonyhurst there are boundaries marked out to the 
 students, which they are not allowed to pass ; and there 
 are prefects always pacing to and fro within the lines to 
 prevent any unlucky boy from straying on the other side 
 of them. 
 
 " Notwithstanding the vigilance of the lynx-eyed guar- 
 dians, I would now and then manage to escape, and would 
 
 Stomj- 
 harat. 
 
 Out of 
 boundf. 
 
nroGnAFiiY. 
 
 bolt into a very extojisive labyrinth of yew unci liolly 
 trees clc^e at Imnd. It was the cliosen place for animatol 
 nature. JJirds, in particuliir, used to frequent the spacious 
 enclosure, both to obtain food and I'lijoy security. IMany 
 a time have I luinted the foumart and the squirrel. I 
 once took a cut through it to a neighbouring wood, 
 where I knew of a carrion-crow's nest. The prefect missed Dlncovrnj 
 nie; and judging that I had gone into the labyrinth, he ""'' *'"***• 
 gave chase without loss of time. After eluding him in 
 cover for nearly half an hour, being hard pressed, I took 
 away down a hedgerow. 
 
 " Here (as I learned afterwards) he got a distant sight 
 of mo ; but it was not sufficiently distinct for him to know 
 to a certainty that I was the fugitive. I luckily succeeded 
 in reaching he outbuildings which abutted on the college, 
 and lay at a considerable distance from the place where 
 I had first started. I had just time to enter the postern 
 gate of a pigsty, when, most opportunely, I found old Joe 
 Bowren, the brewer, bringing straw into the sty. He was Rf/iuie in 
 more attached to me than to any other boy, for I had "^'i/''^^' 
 known him when I was at school in the North, and had 
 made him a present of a very fine terrier. 
 
 " ' I've just saved myself, Joe,* said I ; ' cover me up 
 with litter.' 
 
 "He had hardly complied with my request, when in 
 bounced the prefect by the same gate through which I had 
 entered. 
 
 " ' Have you seen Charles Watcrton ? ' said he, quite out 
 of breath. 
 
 " My trusty guardian answered, in a tone of voice 
 which would have deceived anybody, ' Sir, I have not 
 spoken a word to Charles Waterton these three days, to 
 the best of my knowledge. 
 
 " Upon this, the prefect, having lost all scent of me, 
 
BIOGRAPHY. 
 
 Encapc. 
 
 gave up the pursuit, and went his way. When he had 
 disappearcci, I stole out of cover, as strongly perfumed as 
 was old Falstaff when they had turned him out of the 
 buck basket. 
 
 " Once I had gone into the labyrinth to look into a 
 magpie's nest, which was in a high hollow tree; and 
 hearing the sound of voices near, I managed to get a 
 resting-place in the tree just over the nest, and there I 
 squatted, waiting the event. Immediately die President, 
 two other Jesuits, and the present Mr. Salvin of Croxdale 
 Hall, passed close under the tree without perceiving me. 
 
 " The good fathers were aware of my predominant pro- 
 pensity. Though it was innocent in itself, nevertheless it 
 was productive of harm in its consequences, by causing 
 me to break the college rules, and thus to give a bad 
 example to the community at large. Wherefore, with a 
 magnanimity, and excellent exercise of judgment, which 
 are only the province of those who have acquired a con- 
 summate knowledge of human nature, and who know how 
 to turn to advantage the extraordinary dispositions of those 
 intrusted to their care, they sagaciously managed matters 
 in such a way as to enable me to ride my hobby to a 
 certain extent, and still, at the same time, to prevent me 
 from giving a bad example. 
 
 " As the establishment was very large, and as it con- 
 tained an abundance of prey, the Hanoverian rat, which 
 fattens so well on English food, and which always con- 
 trives to thrust its nose into every man's house when there 
 is anything to be got, swarmed throughout the vast extent 
 of this antiquated mansion. The ability which I showed 
 in curtailing the career of this voracious intruder did not 
 fail to bring me into considerable notice. The cook, the 
 baker, the gardener, and my friend old Bowren, could all 
 bear testimony to my progress in this line. By a mutual 
 
 Insight 
 
 into 
 
 character. 
 
 ■if 
 
 i 
 
BIOGRAPHY 
 
 Com- 
 promise, 
 
 understanding I was made rat-catcher to the establishment, 
 and also fox-taker, foumart-killer, and crossbow-charger at 
 the time when the young rooks were fledged. Moreover, 
 I fulfilled the duties of organ-blower and football-maker 
 with entire satisfaction to the public. 
 
 " I was now at the height of my ambition. I followed 
 up my calling with great success. The vermin disap- 
 peared by the dozen ; the books were moderately well 
 thumbed ; and, according to my notion of things, all went 
 on perfectly right." 
 
 One of those wise teachers did him an inestimable ser- 
 vice. He called the lad into his room, told him that his 
 roving disposition would carry him into distant countries, 
 and asked him to promise that from that time he would not 
 touch either wine or spirits. Waterton gave the promise, 
 and kept it to the hour of his death, more than sixty years 
 afterwards. Once, when returning from one of his foreign 
 expeditions, he took a glass of beer at dinner, but, finding 
 the taste, from long disuse, unpleasantly bitter, he put 
 down the glass and never touched beer again. 
 
 At the age of eighteen he left Stonyhurst with much Departure 
 regret, and after a year spent at Walton Hall amid the siomj- 
 pleasures of the field, he started on the first of his jour- ^"^*^- 
 neys abroad. It was during the Peace of Amiens, and 
 Spain was chosen as the country which he should visit. 
 After staying a short time at Cadiz, he sailed for Malaga, 
 and had the good fortune to visit Gibraltar just in time 
 to see the celebrated apes. 
 
 Gibraltar was the last place in Europe where apes lived 
 wild. How they got there no one knows, but Waterton 
 suggests in one of his Essays that they belonged originally 
 to Africa. 
 
 " Let us imagine that, in times long gone by, the pre- 
 sent Eock of Gibraltar was united to the corresponding 
 
10 
 
 BIOGRAPHY. 
 
 '!■ 
 
 I. I:' 
 
 i 
 
 /ipes of mountain called Ape's Hill, on the coast of Barbary ; and 
 
 Gibraltar, w^q^^ ])y some tremendous convulsion of nature, a channel 
 
 had been made between them, and had thus allowed the 
 
 vast Atlantic Ocean to mix its waves with those of 
 
 the Mediterranean Sea. 
 
 " If apes had been on Gibraltar when the sudden shock 
 occurred, these unlucky mimickers of man would have 
 seen their late intercourse with Africa quite at an end. A 
 rolling ocean, deep and dangerous, would hi've convinced 
 them that there would never again be a highway overland 
 from Europe into Africa at the Straits of Gibraltar. 
 
 "Now as long as trees were allowed to grow on the 
 Eock of Gibraltar, these prisoner-apes would have been 
 pretty well off. But, in the lapse of time and change of 
 circumstances, forced by * necessity's supreme command,' for 
 want of trees, they would be obliged to take to the ground 
 on all-fours, and to adopt a very different kind of life 
 from that which they had hitherto pursued." 
 
 The animal here mentioned is the Barbary Ape, or 
 Magot, a species of Macacque. At Gibraltar it feeds 
 largely on the scorpions that have their liabitations under 
 the loose stones. I do not think that Waterton's sugges- 
 tion as to its altered habits is carried out by facts, for the 
 magot is quite as much at home among rocks or among 
 trees, as are the great baboons of Southern Africa. I 
 The Magot. have seen a number of magots in a large cage, or 
 rather, apartment, in the open air. They were supplied 
 with rock-work and trees, and of the two seemed to 
 prefer the former. Their colours harmonised so completely 
 with that of the rough stones on which they sat, that 
 many persons passed the cage, thinking it to be untenanted, 
 while five or six magots were seated among the rocks, and 
 almost as motionless as the stones themselves. 
 
 Generally, the Gibraltar magots keep themselves so 
 
BIOGRAPHY. 
 
 n 
 
 much aloof, that they cannot be seen without the aid of a 
 telescope, but Waterton was fortunate enough to see the 
 whole colony on the move, they being forced to leave their 
 quarters by a change of wind. He counttd between fifty 
 and sixty of them, some having young on their backs. 
 
 After staying for more than a year in Malaga, and 
 having apparently in the meantime acquired the Spanish 
 language, of which he was totally ignorant when he en- 
 tered Cadiz, but in which he was afterwards a proficient, riag^ie at 
 he projected a visit to Malta, but was checked by a ter- " "^"' 
 rible obstacle. This was the " black-vomit," which broke 
 out with irresistible force, accompanied with cholera and 
 yellow fever. 
 
 The population died by thousands, and so many were 
 the victims of these diseases that graves could not be dug 
 fast enough to keep pace with the mortality. Large pits 
 were dug — much like our plague-pits — and as they could 
 not accommodate the coffins, the bodies of the dead were 
 flung promiscuously into the pits. An uncle of Waterton 
 died of the disease, his body was taken out of its coffin 
 and thrown into the pit, and just beneath him lay the 
 body of a Spanish marquis. No less than fourteen thou- 
 sand people died in Malaga, notwithstanding that fifty 
 thousand persons had fled from the city. 
 
 Waterton did not escape scatheless. He was seized 
 with the black-vomit, but, although it was thought that Seized unth 
 he could not live until the following day, his great strength **<^*"**''- 
 of constitution, aided by his simple mode of life, enabled 
 him to conquer in the struggle. As if to add to the terrors 
 of the time, earthquakes followed the plague, and every one 
 who possessed another home was anxious to leave a spot 
 which had been stricken with such plagues, and among 
 them was Waterton. But the authorities had mean- 
 while laid an embargo on the shipping, and it was next to 
 
12 
 
 BIOGRAPHY. 
 
 S:i! I 
 
 ! I 
 
 impossible to get away. At last, at the risk of imprison- 
 An escape meiit for life, he escaped by the daring and forethought 
 of a Swedish captain. 
 
 He took on board Waterton and his younger brother, 
 the former being entered on the ship's books as a Swedish 
 carpenter, and the latter as a passenger. How carefully 
 the escape was planned, and how skilfully it was executed, 
 must be told in Waterton's own words : — 
 
 " We slept on board for many successive nights, in hopes 
 of a fair wind to carry us through the Straits. At last, a 
 real east wind did come, and it blew with great violence. 
 The captain, whose foresight and precautions were truly 
 admirable, had given the strictest orders to the crew that 
 not a word should be spoken whilst we were preparing to 
 escape. We lay in close tier amongst forty sail of mer- 
 chantmen. The harbour-master having come his usual 
 rounds and found all right, passed on without making any 
 observations. 
 
 " At one o'clock, p.m., just as the governor had gone to 
 the eastward to take an airing in his carriage, as was his 
 custom every day, and the boats of two Spanish brigs-of- 
 war at anchor in the harbour had landed their officers for 
 the afternoon's amusements, our vessel worked out clear 
 of the rest, and instantly became a cloud of canvas. The 
 captain's countenance, which was very manly, exhibited a 
 portrait of cool intrepidity rarely seen : had I possessed 
 the power, I would have made him an admiral on the spot, 
 
 " The vessel drove through the surf with such a press of 
 sail that I expected every moment to see her topmasts 
 carried away. Long before the brigs-of-war had got their 
 officers on board, and had weighed in chase of us, we were 
 Success, far at sea ; and when night had set in we lost sight of them 
 for ever, our vessel passing Gibraltar at the rate of nearly 
 eleven knots an hour." 
 
BIOGRAPHY. 
 
 18 
 
 It was indeed fortunate for Waterton that he succeeded 
 in making his escape, for in the following spring the plague 
 returned with increased violence, and no less than thirty- 
 six thousand more victims perished. Waterton never 
 dwells on the hardships and sufferings which he under- 
 went in his travels, but he remarks that his constitution 
 was much shaken by the Malaga illness, and that in all 
 probability he would not have survived a second attack. Shaken by 
 He had tried to persuade another uncle to take part in the 
 escape, but he declined, and was carried off by the second 
 outbreak of the pestilence. 
 
 So ended Waterton 's first experience of foreign travel. 
 It was not by any means an encouraging tour, for he had 
 lost relatives, friends, and health, while he had gained 
 little except a knowledge of travel, and the sight of 
 flamingos, vultures, and apes at liberty. 
 
 It was characteristic of Waterton that when he found 
 himself at Hull, forty-four years after he started on his 
 travels, he made inquiries about the captain of the ship in 
 which he took his first voyage, discovered that he was 
 alive, sought him out, and renewed the acquaintance 
 begun so many years before. 
 
 His weakened state caused him to take cold as he was 
 sailing up the Channel ; the cold settled On the lungs, and 
 lie was scarcely in less danger in England than he had 
 been in Malaga. However, he again rallied, and was able 
 once more to join the hunting-field. Still, the shock to 
 the system had been very great, and to the end of his 
 life, though he could endure almost any amount of heat, 
 he was painfully sensitive to cold, and especially to cold 
 winds. The chilly climate of England did not agree CUmat': of 
 with his health, and he found himself again obliged "^ "" 
 to gD abroad. He longed, he said, " to bask in a warmer 
 sun." 
 
14 
 
 BIOGRAPHY. 
 
 Voyage to 
 Demerara. 
 
 .1 
 
 
 I i 
 
 ii 
 
 Some estates in Demerara being in possession of the 
 family, Waterton went to superintend them, and in the 
 interval before starting, made the personal acquaintance 
 of Sir Joseph Banks, who at once appreciated the powers 
 which the young traveller was afterwards to develop. 
 He gave Waterton a piece of most excellent advice, 
 namely, to come home for a time at least once in three 
 years. 
 
 He continued to administer the estates for eight years, 
 when, as both his father and uncle, the proprietors of the 
 estates, were dead, he handed over the property to those 
 who had a right to it, and thence began his world-famed 
 Wanderings, the account of which will be given exactly 
 as he wrote it; without the change or omission of a 
 syllable, or the addition of a note. 
 
 'I 
 
CHAPTER II. 
 
 .rouniey to Orinoco with dcspatclies.— Adventure with a venomous snake. 
 —An involuntary bath. — A huge cayman. — The Labarri snake.— 
 Dinner party in Angostura.— A too liberal table.— The Governor's 
 uniform.— Dining in shirt-sleeves.— A more sensible uniform.- Pub- 
 lication of the ?rawrfcn?t(7s.— Reception by the critics.— Sydney Smith. 
 — Swainson's criticism upon the cayman— Truth in the garb of fiction. 
 — Waterton's style of writing.— Quotations.— His favourite authors.— 
 Sense of humour.— How he answered the critics.— Charge of eccen- 
 tricity.— How he was eccentric— Travels on the Continent.— Ship- 
 wreck.— Gallant conduct of Prince Canino.— Lost by gold.— Fall into 
 Dover harbour and narrow escape.— The lancet and calomel.— Judge- 
 ment of the vampire.— A bad wound.— Bare feet and bad pavement.— 
 Mode of cure.— Accidents at home.— Gunshot wound.— Severe fall and 
 dangerous injuries.— Crowther the bone-setter,— A painful operation.— 
 Ultimate recovery.— A characteristic warning. 
 
 During his stay in Demerara, he was selected as the Despatches 
 bearer of despatches to the Spanish Government i^^^OHiwco. 
 Orinoco, and received the first commission which had 
 been held oy any one bearing the name of Waterton 
 since the days of Queen Mary ; the commission being 
 dated August 2, 1808. 
 
 Wliile passing up the Orinoco river in the fulfilment 
 of this mission, an adventure occurred which had well- 
 nigh deprived the world of the Wanderings. 
 
 " During the whole of the passage up the river, there was 
 a grand feast for the eyes and ears of an ornithologist. In 
 the swampy parts of the wooded islands, which abound in 
 this miglity river, we saw waterfowl innumerable; and 
 
K, 
 
 DIOGRArHY. 
 
 Tropical 
 birds. 
 
 I'll' 
 
 Wounded 
 Labarri. 
 
 when we had reached the higher grounds it was quite 
 charming to observe the immense quantities of parrots and 
 scarlet aras which passed over our heads. The loud harsh 
 screams of the bird called the horned screamer were heard 
 far and near ; and I could frequently get a sight of this 
 extraordinary bird as we passed along; but I never 
 managed to bring one down with the gun, on account of 
 the difficulty of approaching it. 
 
 " While we were wending our way up the river, an 
 accident happened of a somewhat singular nature. There 
 was a large labarri snake coiled up in a bush, which was 
 close to us. T fired at it, and wounded it so severely that 
 it could not escape. Being wishful to dissect it, I reached 
 over into the bush, with the intention to seize it by the 
 throat, and convey it aboard. The Spaniard at the tiller, 
 on seeing this, took the alarm, and immediately put his 
 helm aport. This forced the vessel's head to the stream, 
 and I was left hanging to the bush with the snake close to 
 me, not having been able to recover my balance as the vessel 
 veered from the land. I kept firm hold of the branch to 
 which I was clinging, and was three times overhead in 
 the water below, presenting an easy prey to any alligator 
 that might have been on the look-out for a meal. 
 
 " Luckily a man who was standing near the pilot, on 
 seeing what had happened, rushed to the helm, seized hold 
 of it, and put it hard a-starboard, in time to bring the 
 head of the vessel back again. As they were pulling 
 me up, I saw that the snake was evidently too far gone 
 to do mischief ; and so I laid hold of it and brought 
 it aboard with me, to the horror and surprise of the crew. 
 It measured eight feet in length. As soon as I had 
 got a change of clothes, I killed it, and made a dissection 
 of the head. 
 
 " T would sometimes go ashore in the swamps to shoot 
 
 I ' 
 
 ii 
 
BIOGRAPHY. 
 
 17 
 
 the 
 
 ling 
 
 hoot 
 
 maroudies, which are soinewlmt related to the pheasant ; 
 
 but they were very shy, and it required considerable 
 
 address to get within shot of them. In these little 
 
 excursions I now and then smarted for my pains. IMoro 
 
 than once I got among some hungry leeches, whicli made 
 
 pretty free with my legs. The morning alter I had had 
 
 the adventure with the Labarri snake, a cayman slowly Cayman. 
 
 passed our vessel. All on board agreed that this tyrant 
 
 of the fresh waters could not be less than thirty feet 
 
 long." 
 
 I onght to state that the Labarri snake here mentioned 
 is one of the most venomous serpents of Guiana, but 
 as it will be fully described in a subsequent page, I 
 shall say no more about it at present. Waterton never 
 feared snakes, even though knowing that their bite is 
 certain death, but the coxswain of the boat, not having 
 such nerve, might well be excused for taking alarm. 
 
 A rather amusing incident took place when he had 
 reached his destination. 
 
 " On arriving at Angostura, the capital of the Orinoco, Angostura. 
 we were received with great politeness by the Governor. 
 Nothing could surpass the hospitality of the principal 
 inhabitants. They never seemed satisfied unless we were 
 partaking of the dainties which their houses afforded. 
 Indeed, we had feasting, dancing, and music in super- 
 abundance. 
 
 " The Governor, Don Felipe de Ynciarte, was tall and The 
 corpulent. On our first introduction, he told me that he (Governor. 
 expected the pleasure of our company to dinner every 
 day during our stay in Angostura. We had certainly 
 every reason to entertain, very high notions of the 
 plentiful supply of good things which Orinoco afforded ; 
 for, at the first day's dinner, I counted more than forty 
 dishes of fish and flesh. The governor was superbly 
 
 c 
 
18 
 
 HIOGUAPIIY. 
 
 ■t 
 
 Ikmij attired in a full uniform of goltl and blue, the weight 
 uniform. ^£ which alonc, in that hot climate, and at such a 
 repast, was enough to have melted him down. He had 
 not half got through his soup before be began visibly 
 to liquefy. I looked at him, and bethought me of the 
 old saying, * How I sweat ! said the mutton-chop to the 
 gridiron.' 
 
 " He now became exceedingly uneasy ; and I myself 
 had cause for alarm ; but our sensations arose from very 
 different causes. He, no doubt, already felt that the 
 tightness of his uniform, and the weight of the orna- 
 ments upon it, would never allow him to get through 
 that day's dinner with any degree of comfort to him- 
 self; I, on the other hand (who would have been amply 
 satisfied with one dish well done) was horrified at the 
 appalling sight of so many meats before me. Good- 
 breeding whispered to me, and said ; ' Try a little of 
 most of them.' TemperMvce replied, 'Do so at your 
 peril ; and for your ovor-strained courtesy, you shall 
 have yellow-fever before midnight.' 
 
 mat and " At last the Governor said to me, in Spanish, ' Don 
 Carlos, this is more than man can bear. No puedo sufrir 
 tanto. Pray pull off your coat, and tell your companions to 
 do the same ; and I'll show them the example.' On saying 
 this, he stripped to the waistcoat ; and I and my friends 
 and every officer at table did the same. The next day, 
 at dinner-time, we found his Excellency clad in a uniform 
 of blue Salempore, slightly edged with gold lace." 
 
 End of 
 
 IVandcr- 
 
 ings. 
 
 His tropical Wanderings came to an end in 1825, in 
 which year he published the now famous volume. At 
 first, he received from the critics much the same treat- 
 ment as did Bruce and Le Vaillant. Critics would not 
 believe that Bruce ever saw a living ox cut up for food, or 
 
niOGlJAPlIY. 
 
 Mi 
 
 The 
 critics. 
 
 Sjidiiey 
 Smith. 
 
 that the Abyssinians ate Leef raw in preference to cooked. 
 Neither would they believe that Lc Vaillant ever chased 
 a giraffe, because, as they said, there was no such animal, 
 and that therefore, Le Vaillant could not have seen it. 
 
 Similarly, some of Waterton's statements were received 
 with a storm of aerision, more especially his account of 
 the sloth and its strange way of living ; of the mode of 
 liandling deadly serpents, and above all, his ride on tlie 
 back of a cayman. There is however one honourable 
 exception in the person of Sydney Smith, who devoted 
 one of his wittiest and happiest essays to a review of 
 the Wanderings and fully recognized the extraordinary 
 powers of Waterton. 
 
 According to Sydney Smith, Waterton " appears in early 
 life to have been seized with an unconquerable aversion to 
 Piccadilly, and to that train of meteorological questions 
 and answers which forms the great staple of polite con- 
 versation. . . 
 
 "The sun exhausted him by day, the mosquitos bit 
 
 him by night, but on went Mr. Charles Waterton 
 
 happy that he had left his species far away, and is at last 
 in the midst of his blessed baboons," 
 
 Nothing can be better than Sydney Smith's summary of 
 the life of a sloth, who " moves suspended, rests suspended, Suspense. 
 sleeps suspended, and passes his whole life in suspense, 
 like a young clergyman distantly related to a bishop." 
 Or, than his simile of the box-tortoise and the boa, 
 who " swallows him shell and all, and consumes him 
 slowly in the interior, as the Court of Chancery does a Chancery. 
 large estate." 
 
 Or, what can be happier than the turn he gives to 
 Waterton's account of the toucan ? 
 
 " How astonishing are the freaks and fancies of nature ! 
 To what purpose, we say, is a bird placed in the forests of 
 
 c 2 
 
20 
 
 THOOUAPIIV. 
 
 W i 
 
 '. I 
 
 Ciiyoinio, with a bill a yard lorif,', making a iioisn like a 
 Toucan. I'lipjiy dn^', and layiiig ogt,'3 in hollow trees ? The Toueans, 
 to ho sure, might retort — to what purpose were gentlo- 
 men in ISond Street created ? To what i)urpoHO were 
 certain foolish, prating members of Tarliament created? 
 pestering the House of ('ommons with their ignorance and 
 folly, and impeding the business of the country. There 
 is no end of such questions. So we will not enter into the 
 metaphysics of the toucan." 
 
 Perhaps the oddest thing to bo found in criticism is 
 that which is given in Lardner's Cahlnct Cydopalia. 
 AVaterton's statements having been proved to be true, 
 the writer now turns round, and tries to show that after 
 all there was nothing very wonderful in the achieve- 
 ment. 
 
 " The crocodile in fact, is only dangerous when in the 
 water. Upon land it is a slow-paced and even timid 
 animal, so that an active boy armed with a small hatchet 
 might easily despatch one. There is no great prowess 
 therefore required to ride on the back of a poor cayman 
 after it has been secured, or perhaps wounded ; and a 
 modern writer might well have spared the recital of his 
 feats in this way upon the cayman of Guiana, had he not 
 Truth in been influenced in this and numberless other instances by 
 fiction ^^^^ greatest possible love of the marvellous, and a constant 
 propensity to dress truth in the garb of fiction." 
 
 I'utting aside the fact that the writer received some of 
 his earliest instructions from Waterton, who was always 
 ready to impart his knowledge to those who seemed likely 
 to make a good use of it, the assertion is absolutely 
 unaccountable. No man was less influenced by a love of 
 the marvellous, and none less likely to " dress truth in the 
 garb of fiction." 
 
 His knowledge of Nature was almost wholly obtained 
 
UKKJIJAPIIY 
 
 21 
 
 from iipr«onal observation, and not one sini^lc stiitetnent of r<'m>ruil 
 liis lui8 ever been provetl to be cx.igfgerated, much less *"/j^',|"' 
 shown to be false. He nnght sometimes discredit the 
 statements of others. For example, he never couhl 
 believe that any races of men could be cannibals from 
 choice, and not from necessity or superstition. lUit, 
 whether at hoino or abroad, his investigations were so 
 close and patient, and his conclusions so just, that he is 
 now acknowledged to be a guide absolutely safe in any 
 department of Natural History which came within hi^ 
 sco]»e. No one now would think of disputing Wuterton's 
 word. If he denied or even doubted the statements of 
 others, his doubts would have groat weight, and could 
 lead to a closer investigation of the subject. Ihit, if lie 
 asserted anything to be a fact, his assertion would bo 
 accepted without scruple. 
 
 As to the meaning of the sentence about truth and 
 fiction, I fail to understand it, except as a poetical way of 
 rounding a paragraph. In the first place, if truth be truth, 
 it is essentially opposed to fiction, and cannot borrow her 
 garb. In the next place, the writer gives no instance of 
 this remarkable performance, except a reference to the 
 capture of the cayman. Now, nothing can be simpler or 
 more straightforward than Waterton's account of the whole 
 transaction. lie does not glorify himself, nor boast of his 
 courage. He leaped astride the animal, being sure, from a 
 knowledge of its structure, that he could not be reached 
 by the cayman's only weapons, namely, its teeth and its 
 tail, and he never repeated the feat. 
 
 Even the peculiar style in which Waterton wrote, could 
 not justify such a charge as was made by Swainson. 
 
 It was, perliaps unconsciously, formed on that of Sterne, 
 many of whose phrases are employed almost verbatim. 
 Then, his mind was saturated Avith Horace, Virgil, Ovid, 
 
 Style of 
 writiwj. 
 
09 
 
 BIOGHAPIIY. 
 
 Quotation, Cei'vaiites, Washington Irving (himself a disciple of 
 Sterne), Chevy Chase, and literature of a similar character. 
 In tlie days when he first took up the pen, it was the 
 rather pedantic custom to introduce frequent quotations 
 fi'om the classics into writings, speeches, and sermons, 
 and Waterton followed the custom of the day. Moreover, 
 it is an old Stony hurst custom to employ such quotations 
 both in conversation and writing, and Waterton could 
 never shake it off. But, when he came to descriptions of 
 scenes in which he had taken part, nothing could be more 
 simple, terse, and graphic, than his style, especially when his 
 sense of liumour was aroused. Take for example the very 
 scene which Swainson assailed. There is no fine language 
 in it. There are a few of the inevitable quotations, which 
 be omitted with advantage, but all the descrip- 
 couched in the simplest and most forcible 
 without a redundant word. A better word- 
 picture does not exist in our language. We see before 
 us the captured cayman struggling in the water, the 
 mixed assembly of South American savages, African 
 negroes, a Creole, and an Englishman, all puzzled to know 
 how to get the beast ashore without damaging it, or being 
 wounded themselves. 
 
 Then, there is the amusing cowardice of " Daddy Quashi," 
 the negro, who ran away when suspecting danger, hung in 
 tlie rear wlien forced to confront it, and, when it was over, 
 " played a good finger and thumb at breakfast." Water- 
 ton's strong sense of humour prevails throughout the story, 
 but there is not a tinge of vanity. He explains his firm 
 seat on the furious animal's back by mentioning that he 
 
 The hunt' had hunted for several years with Lord Darlington's fox- 
 hounds, but he does not tell the reader that in that cele- 
 brated hunt he was considered, next to Lord Darlington, as 
 tlie best horseman in the field. 
 
 might 
 tion is 
 English, 
 
 Daddji 
 Quashi. 
 
 inij field. 
 
M 
 
 BIOGRAPHY. 
 
 23 
 
 Jg 
 
 rm 
 he 
 
 ■OK- 
 
 It is illustrative of Watcrton's character that when the 
 reviewers impugned his veracity, he trouhlecl liimself very- 
 little about them, saying that tlie creatures whicli he had 
 described would one day find their way to the Zoological 
 Gardens, and then that everybody would see that he had 
 but spoken the truth. So, when the first sloth arrived, 
 Waterton had quite a little triumph over his detractors, 
 Indeed, the probability was, that, after reading one of 
 these reviews, he w^ould invite the assailant to Walton 
 Hall, offer him the good old Englisli hospitality of that 
 place, and settle the point of dispute in friendly controversy. 
 
 But, little as he cared for such attacks, he was deeply Eccevtri- 
 stung by the epithet * eccentric ' which one writer applied "'y- 
 to him, and never could forget it. 
 
 Yet, had he not been eccentric, he could not have been 
 the Charles Waterton so long known and loved. It was 
 perhaps eccentric to have a strong religious faith, and act 
 up to it. It was eccentric, as Thackeray said, to " dine 
 on a crust, live as chastely as a hermit, and give his all to 
 the poor." It was eccentric to come into a large estate as 
 a young man and to have lived to extreme old age with- 
 out having wasted an hour or a shilling. It was eccentric 
 to give bountifully and never allow his name to appear in 
 a subscription-list. It was eccentric to be saturated with 
 the love of nature. It might be eccentric never to give 
 dinner-jiarties, preferring to keep an always open house 
 for his friends ; but it nv.is a very agreeable kind of ec- 
 centricity. It was eccentric to be ever childlike, but never 
 childish. We might multiply instances of his eccentricity 
 to any extent, and may safely say that the world would be 
 much better tlian it is if such eccentricity were more 
 common. 
 
 It formed one of the peculiar charms of his society, and 
 he was utterly unconscious of it. He thought himself the 
 
24 
 
 BIOGIJAPIIV. 
 
 ill 
 
 < ! I nil 
 
 H 
 
 ' J 
 
 ill 
 
 it- 
 
 Thour/ht- 
 fulness. 
 
 most common-place of human beings, and yet no one could 
 be in bis company for five minutes witbout feeling bimself 
 in tbe presence of no ordinary man. He bad no idea tliat 
 be was doing anytbing out of the general course of things 
 if be asked a visitor to accompany him to the top of a 
 lofty tree to look at a hawk's nest ; or if he built bis 
 stables so that the horses might converse with each other 
 after their work was over, or bis kennel so that bis 
 bounds should be able to see everything that was going on. 
 
 Even the pigs came in for their share of bis kindly 
 tboughtfulness. He used to say that in a wild state, 
 swine were not dirty beasts, but that when they are penned 
 into small sties, as is usually the case, they have no op- 
 portunity of being clean. So he bad his sties built of 
 stone, with a stone platform in front, sloping and chan- 
 nelled so as to be easily and thoroughly cleansed, and 
 having a southern aspect so that the pigs might enjoy 
 the beams of that sun which their master loved so much 
 himself. 
 
 On these warm stone slabs they used to lie in a half- 
 dozing state, and Waterton often used to point out tbe 
 multitudinous wasps that came flying into tbe sties and 
 picked off the flies from the bodies of the drowsy pigs. 
 
 If the sties at Tudhoe bad been like those at Walton Hall 
 
 > 
 
 be would not have issued from them in the highly per- 
 fumed state which lie so amusingly describes. See p. 6. 
 Some persons thought that his rooted abhorrence of 
 ifourning. mourning was eccentric. If so, the eccentricity is now 
 shared by many, including myself, who have abandoned 
 on principle tbe black crape, gloves, hat-bands, mutes, 
 black feathers, black-edged writing paper, and other 
 conventional signs of grief. 
 
 Waterton however carried tbe principle still further, and 
 could never be induced to wear even a black coat of any 
 
 1 
 
 
 
T 
 
 T 
 
 mOGKAPIIY. 
 
 25 
 
 )lgS. 
 
 [all 
 
 per- 
 
 6. 
 te of 
 
 now 
 loned 
 Utes, 
 lother 
 
 r, and 
 any 
 
 kind on any occasion. He usually wore a blue body-coat Dress. 
 with gold — not gilt buttons, but at the urgent request of 
 the poHce, who told him that his costl}'' buttons were a 
 perpetual anxiety to them wlienever he went to Wakefield, 
 he at last consented to lay them aside, except at home, 
 and have his buttons covered with bhie cloth. 
 
 This peculiarity once caused him to lose the privilege of 
 an introduction to the Pope (Gregory XVI.). Etiquette 
 demanded that if uniform could not be worn, the presentee 
 must appear in ordinary evening dress. Now, had Water- 
 ton qualified as Deputy-Lieutenant, he could have followed 
 the usual custom and worn that unifornj, but as he had 
 refused to do so, evening-dress was the only alternative. 
 But he would not wear ' frac-nero,' and so lost the 
 presentation. 
 
 On another occasion however, the difficulty was evaded 
 in a very characteristic manner. He bethought himself of 
 his commission in the Demerara militia ; but he had no 
 uniform, and there was no time to make one. Some naval Uniform. 
 friends were with him. Captain ]\Iarryatt being, I believe, 
 one of them, and with Waterton's blue coat and gold buttons, 
 surmounted with a pair of naval epaulettes, and with 
 the addition of a naval captain's cocked hat and sword, 
 they composed an amusingly miscellaneous uniform. One 
 friend wickedly suggested that spurs would have an impos- 
 ing effect in connection with the naval hat and epaulettes, 
 but he was not to be caught in so palpable a snare. 
 
 Of his travels on the Continent, there is but little to say 
 as they are related at some length in the three volumes of 
 Essays. It is remarkable, by the way, that on the Conti- 
 nent, as well as in England, he met with injuries far more 
 severe than any which he received in Guiana. 
 
 Twice he was nearly drowned. 
 
 Essays. 
 
26 
 
 BIOGRAPHY. 
 
 Shipwreck 
 
 m 
 
 I' 
 
 li 
 
 i! 
 
 ill! 
 
 Fall into 
 
 the 
 harbour. 
 
 On one occasion he was on board a vessel named the 
 Pollux, and bound from Civit;i Vecchia to T/eghorn. In 
 tlie night of the same day, an accident befell the Pollux, 
 almost exactly resembling that in which the ill-fated 
 Princess Alice was destroyed. The night was peculiarly 
 calm, the stars were shining brightly, and everything ap- 
 peared to be in security, when all on board were startled 
 from their sleep by a violent shock. A steamer, named 
 the Mongibcllo, from Leghorn to Civita Vecchia, had run 
 into the Pollux, and cut her nearly in two, the cutwater of 
 the Mongihcllo having actually forced its way into Water- 
 ton's cabin. 
 
 Fortunately for the passengers, most of them, including 
 Waterton and his family, were sleeping on deck. As is 
 too often the case under similar circumstances, the officials 
 on board the offending vessel lost their presence of mind, 
 and were actually sheering off from the wreck. Had it 
 not been for the courage and skill of Prince Canino (Charles 
 Bonaparte) the loss of life must have been very great. 
 
 He was a passenger on board the Monrjibcllo, knocked 
 the steersman off the wheel, took the helm himself, and 
 laid the vessel alongside the sinking Pollux. Only one 
 life was lost, that of a man who had a large sum of gold 
 sewed in a belt round his waist, and was drawn under 
 water by the weight. 
 
 In this shipwreck, although Waterton's life was saved, 
 he and his party lost their wardrobes, money in cash, and 
 letters of credit, books, writings, passports, and works of 
 art ; the last mentioned loss being irreparable. Fever and 
 dysentery were the results of the shipwreck, and did not 
 loosen their hold until long afterwards. 
 
 Another time, he fell into Dover harbour while about to 
 embark on board the steamer. Any one who has walked on 
 cliffs on a dark night is aware of the difficulty of distinguish- 
 
 / 
 
 
,1 
 
 BIOGRAPHY. 
 
 27 
 
 avecl, 
 and 
 
 ■ks of 
 and 
 
 d not 
 
 
 luish- 
 
 ingland from water. At Margate I was once within a single Dar^icrs 
 step of falling over the cliff, whose edges corresponded so ^■^ '^ '■"^' 
 exactly in colour with the sea and rocks below, that, had 
 it not been for the information conveyed by a stick, I 
 must have been instantly killed. Several persons, indeed, 
 have lately been killed at the same spot. 
 
 Thinking that he was at the gangway, he stepped over 
 the edge of the quay, and fell fifteen feet into the water 
 sinking under the paddle-box, and only finding support 
 by catching at the wheel itself. Thence he was rescued ; 
 but the cold winds blowing on him as he stood wet and 
 dripping on the deck of the steamer, brought on a violent 
 attack of fever. He had recourse to his usual double 
 remedy, the lancet and calomel, and recovered sufficiently 
 to attend the great religious festival at Bruges, for the sake 
 of which he had left England. 
 
 His reliance on the lancet and calomel was almost in- 
 credible. In these times the former is hardly ever used, 
 a\id the latter has been abandoned by a great number of 
 medical men. But in Waterton's early days these were 
 the principal remedies, and he never lost faith in them. 
 When I last saw him in 18G3, he told me that he had 
 been bled one hundred and sixty times, mostly by his own 
 hand. 
 
 The amount of blood which he would take at a tir ■» 
 from his spare and almost emaciated frame was positive! j' Lancet and 
 horrifying. On this occasion he lost twenty- five ounces '^^ ' 
 of blood, and next morning took twenty grains of jalap, 
 mixed with ten grains of calomel. It was no wonder that 
 the vampire bat of Guiana would never bite him, though 
 he left his foot invitingly out of the hammock in order to 
 attract it. He used to complain tliat the bat never could 
 be induced to bleed liira, though it would attack a man Vampire. 
 lying in the next hammock ; but he might have antici- 
 
r. 
 
 28 
 
 . BI0G1L\PIIY 
 
 Broken 
 glass. 
 
 Barefoot 
 
 walk to 
 
 Rome, 
 
 M !Ri. 
 
 patecl that the vampire would know Letter than to try to 
 suck blood from a man who was constantly bleeding 
 himself. 
 
 Besides these accidents by water, he twice suffered 
 severe injuries when travelling by land. 
 
 In 1818, while returning over Mount Cenis, he fancied 
 'hat the baggage on the top of the carriage was loose, and 
 mounted on the wheel t*:- dxamine it. Unfortunately his left 
 knee broke the window, and two large pieces of glass ran 
 into it just above the knee-joint. In spite of the darkness, 
 he conti'ived to get out the two pieces of glass, bound up 
 the wound with his cravat, cut off his coat pocket, and had 
 it filled with j)oultice at the nearest house, and, although 
 repeatedly attacked with fever, he reached Paris and there 
 gained strength to return to England. The knee remained 
 stiff for two years, but by continual exercise without the 
 aid of a walking-stick, the limb recovered its normal flexi- 
 bility. 
 
 The next accident might have been nearly as seriou3, 
 and is here given in his own words : — 
 
 " I had a little adventure on the road from Baccano to 
 Eome not worth relating, but Avhich I deem necessary to 
 be introduced here in order that some of my friends in 
 the latter city, and others in England, may not give me 
 credit for an affair which deserves no credit at all. These 
 good friends had got it into their heads that I had reached 
 liome after walking barefoot for nearly twenty miles, in 
 order to show my respect and reverence for the sacred 
 capital of the Christian world. AVould that my motive 
 i|iad been as pure as represented. The sanctity of the 
 churches, the remains of holy martyrs which enrich them, 
 the relics of canonised saints placed in such profusion 
 throughout them, might well induce a Catholic traveller 
 to adopt this easy and simple mode of showing his religious 
 
 I ', 
 
BIOGRAPHY. 
 
 29 
 
 IglOllS 
 
 feeling. But, iinfortunatclp, the idea never entered my 
 mind at tlie time. I had no other motives than those of 
 easy walking and self-enjoyment. The affair which caused 
 the talk took place as follows : — 
 
 " We had arrived at Baccano in the evening, and whilst Baccano, 
 we were at tea, I proposed to our excellent friend Mr. 
 Fletcher, who had joined us at Cologne, that we should 
 leave the inn at four the next morning on foot to Rome, 
 and secure lodgings for the ladies, who would follow us 
 in the carriage after a nine-o'clock breakfast. Having 
 been accustomed to go without shoes month after month 
 in the rugged forests of Guiana, I took it for granted that 
 I could do the same on the pavement of his Holiness Pope 
 Gregory the Sixteenth, never once retlecting that some 
 fifteen years had elapsed from the time that I could go 
 barefooted with comfort and impunity ; during the interval, 
 however, the sequel will show that the soles of my feet had 
 undergone a considerable alteration. 
 
 " We rose at three the morning after, and having put a 
 shoe and a sock or half-stocking into each pocket of my 
 coat, we left the inn at Baccano for Eome just as the hands 
 of our watches poi- ted to the hour of four. Mr. Fletcher, 
 having been born in North Britain, ran no risk of injuring 
 his feet by an act of imprudence. The sky was cloudless 
 and the morning frosty, and the planet Venus shone upon 
 us as though she had been a little moon. 
 
 '■' Whether the severity of the frost, which was more than 
 commonly keen, or the hardness of the pavement, or perhaps 
 both conjoined, had deprived my feet of sensibility, I had 
 no means of ascertaining ; but this is certain, I went on 
 merrily for several miles without a suspicion of anything 
 being wrong, until we halted to admire more particularly 
 the transcendent splendour of the morning planet, and then 
 I saw blood on the pavement ; my right foot was bleeding 
 
 Wounded 
 foot. 
 
30 
 
 lUOGHAPIIY. 
 
 ■' ! 
 
 ill 
 
 ! ' 
 
 RfjKiiring apace, and, on turning the sole uppermost, I perceived a 
 
 magca. ^^^^^g ^^ jag^'od ilesli liangiug by a string. Seeing that 
 
 there would be no chance of replacing the damaged part 
 
 with success, I twisted it off, and then took a survey of 
 
 the foot by the light which the stars afforded. 
 
 " Mr. Fletcher, horror-struck at what he saw, proposed 
 immediately that I should sit down by the side of the road, 
 and there wait for the carriage, or take advantage of any 
 vehicle which might come up. Aware that the pain would 
 be excessive so soon as the lacerated parts would become 
 stiff by inaction, I resolved at once to push on to Rome, 
 wherefore, putting one shoe on the sound foot, which, by 
 the way, had two unbroken blisters on it, I forced the 
 wounded one into the other, and off we started for Rome, 
 which we reached after a very uncomfortable walk. The 
 injured foot had two months' confinement to the sofa before 
 the damage was repaired. 
 
 " It was this unfortunate adventure which gave rise to 
 the story of my walking barefooted into Rome, and which 
 gained me a reputation by no means merited on my part." 
 
 Two more serious accidents occurred within his own 
 domains. 
 
 He was out shooting in 1824, when the gun exploded 
 just as he was ramming the wad on the powder. For- 
 tunately the charge of shot had not been put into the gun. 
 As it was, the ramrod was driven completely through the 
 forefinger of the right hand, between the knuckle and first 
 joint, severing the tendons, but not breaking the bone, 
 though the ignited wadding and powder followed the 
 ramrod through the wound. He procured some warm water 
 at a neighbouring house, washed the wound quite clean, 
 replaced the tendons in their proper positions, and bound 
 up the finger, taking care to give it its proper form. 
 
 Of course the lancet was used freely, and by dint of 
 
 Accident 
 
 while 
 shooting. 
 
 I ! 
 
 k 
 
( 
 
 BIOGHAPIIY. 
 
 81 
 
 gun. 
 
 
 poulticinrf ind constant care, the full use of tlio finger was 
 restorcu. 
 
 Tlie other accident might liave caused his death on the 
 spot, and was a far more severe one than that hy which he 
 afterwards lost his life. 
 
 In 1850, he being then in his sixty-ninth year, he was 
 mounted on a ladder for the purpose of pruning tlie branches Fall while 
 of a pear-tree. The ladder, which was merely propped P''"'"'^'"'^- 
 against a machine of his own invention, slipped sideways, 
 and came to the ground, Waterton having fallen nearly 
 twenty feet. 
 
 He had been repeatedly warned that the machine, not 
 having side stays, must fall if the weight were thrown on 
 one side. But he still persisted in using it, although, 
 shortly before the accident, his son had left the spot, 
 saying that he could not be responsible for an accident 
 which he foresaw but could not prevent. He was partially 
 stunned, and his arm greatly injured, the heavy ladder and 
 machine having fallen into the hollow and smashed the 
 elbow-joint. 
 
 His first act on recovering himself was to use his lancet 
 and take away thirty ounces of blood. Unfortunately a 
 second accident happened almost immediately after the 
 first, a servant having thoughtlessly withdrawn a chair 
 just as he was seating himself, and so causing a second 
 shock, and the loss of thirty ounces more blood. 
 
 For some time, he lay insensible and was apparently A second 
 dying fast, but his iron constitution at length prevailed. 
 
 fall. 
 
 and he was restored to life, though not to health. The 
 injured arm was gradually dwindling in size, and gave 
 continual pain, causing loss of sleep and appetite. He 
 had at last resolved on having the arm amputated, when 
 his gamekeeper advised him to try a certain bone-setter 
 living at Wakefield, who was celebrated for his cures. 
 
■B IW.I '.■ ■— '^-K 
 
 i* ^ 
 if' 1'^ 
 
 32 
 
 lUOGRAPIlY. 
 
 The Bone 
 sdkr. 
 
 AVatortoii took liis advice and sent for tlic practitioner, 
 AFr. .]. Crowtlier, who decided that he could cure the in- 
 jured limb, but at the expense of great pain. The wrist 
 was mucli injured, a callus had formed in the elbow-joint, 
 and the shoulder was partially dislocated. After a time 
 spent in rubbing, pulling, and twisting, he got the shoulder 
 and wrist into their places, and then, grasping the arm 
 "just above the elbow with one hand, and below it with 
 the other, he smashed to atoms, by main force, the callus 
 Painful which had formed in the dislocated joint, the elbow itself 
 operation, cracking as though the interior parts of it had consisted of 
 tobacco-pipe shanks." 
 
 The process was rough, and gave inexpressible pain, but 
 it was efl'ectual, sleep and appetite returned, and health was 
 soon restored. 
 
 From this accident Waterton drew a characteristic warn- 
 ing, namely, never to use ladders when climbing trees. 
 
 One, if not the principal reason of his cessation from 
 tropical exi)lorations, was his marriage. In 1829, he 
 married Anne, a daughter of the Charles Edmonstone^ 
 of Demerara, who is often mentioned in the Wanderings 
 as a kind and true friend. 
 
 His marriage has a curiously romantic history. 
 
 Mr. Cha,rles Edmonstone, one of the Edmonstones of 
 Broich in Scotland, had previously gone to Demerara, 
 where he met a fellow-countryman, William Eeid of 
 Banffshire, who had settled there, and had married Minda 
 Princess (generally called Princess IMinda), daughter of an Arowak 
 chief. Charles Edmonstone married Helen, daughter of 
 William Eeid and JMinda, and they had several children, 
 one of whom, Anne Mary, became the wife of Waterton. 
 He met her in Demerara, while she was yet a child, and 
 made up his mind that she should be his wife. 
 
 Mr. Edmonstone afterwards returned with his family to 
 
 Marriage, 
 
 Minda. 
 
 I 
 
I'.lotillAI'IIV. 
 
 m 
 
 13 of 
 Irara, 
 
 of 
 anda 
 )\vak 
 jr of 
 Idren, 
 Irton. 
 
 and 
 
 ilyto 
 
 i 
 
 1 
 
 i 
 
 Scotland, and piircliased Cardruss Tark, an old I'auiily 
 estate tluit had formed a portion <jI' tlie dower of one of 
 his royal ancestors ; Sir Joliii Kihuonstone, who married 
 the Princess Isabel, daughter of liobert II. of Scotland ! 
 and Sir William Ednionstono his son, who married his 
 cousin, the Trincess ^Fary, dau;^ditcr of Piobert III. 
 
 Through this branch, Kdnmnd Waterton, the present head 
 of the family, is descendod lineally from Leofric and Godiva, 
 ■whoso romantic legend is, I regret to say, wholly a myth. 
 It was impossible that she could have ridden through 
 Coventry, for the same reason that, according to the old 
 song, prevented Guy Faux from crossing Vauxhall Bridge 
 on his Avay " to perpetrate his guilt." Coventry was not 
 in existence at the time. 
 
 There is, however, some foundation for the legend. 
 Godiva was a lady possessing vast wealth, with which 
 she determined to found and endow an abbey. This she 
 did, " stripping herself of all that she had," and thence 
 the legend. Coventry gradually arose round the abbey, 
 and had no streets, and consequently no tolls, until Godiva 
 had been dead at least a century. 
 
 On the death of Charles Edmonstone and his wife, their 
 three daughters, Eliza, Anne ^lary, and Helen, were sent 
 to the well-known convent of Bruges, for the pnrpose of 
 completing their education, and, in the Convent Church, 
 Waterton was married to Anne, on ^May 11, 1829, she 
 being then only seventeen, and he forty-eight. There is 
 an old Scotch proverb to the effect that a bride of one 
 May will never see a second. It was but too true in tliis 
 cnse, for Anne Mary Waterton died on April 27, 18*^)0, 
 twenty-one days after giving birth to a son. 
 
 Through him it is to be hoped that a line so interwoven 
 with ancient history, and so prominent in modern times, 
 will not be broken. He married Josephine, second daughter 
 
 D 
 
 C'ai'ilross 
 
 I'nrk. 
 
 Goilivri, 
 
 The eo)i- 
 rent nf 
 Br tiff IS. 
 
 Death of 
 
 Anne 
 Waleiion. 
 
»1 
 
 i5i()(ii;Aiiiv 
 
 K'huiDul of Sir Joliii Kiiiiis. li.iil, of l»alliii;iliowii Court, Co. AVost- 
 
 Wuiainii. j„^,,,ii,^ Iivland. lit! has is.siiL',— Two sons, Charles Kihimiid, 
 
 now ji stuilcut at Stoiiylmrst, and Tlioiiiiis Aforo. Four 
 
 dau^liters, Mary, A^ues, Ainahil (who died a few months 
 
 nftcr licr birtli), and Jos('[)hiiio. 
 
 Watorton could never bear to speak of his wife, but lie 
 needed lu'lp in the care of his infant son. For this purpose, 
 he nsked lier two .sisters, the blisses Eliza and Helen Ed- 
 monstone to take up their aboth; with him. Tliis tliey did 
 to the hour of his death, and he often wrote with affec- 
 tionate gratitude of their devotion to him. 
 
 He yearned to<,'o back again to tlie wilds of Guiana, but 
 considered that his child had prior claims upon him, nnd 
 so, according to his invariable custom, lie sacri^ced in- 
 clination to dutv. 
 
 I 
 
 r 
 
I 
 
 ClIAPTEK III. 
 
 i 
 
 Maguificenco and monoy. — Watertoii's moileof life au'l personal expensiM. 
 -^Slef'piiij^ on idiinks. — His visits to the chiipel. — The *' morning; gnu. " 
 — Tiio razor and the lancet — IteJiK-tion of the family estates.— His 
 work at Walton Hall. — Natnral advantages of the ])laco. — The wall 
 and its cost. — Ikrgoes and tlieirguna. — Instinct of the herons. — Herons 
 and fish-ponds. — Drainage of the ponds. — Tlie moat e.vtended into a 
 lake. — Old Gateway and Ivy-Tower. — Siege by Oliver Cromwell. — 
 Tradition of a musket-ball.-- Draw-bridge and gateway in the olden 
 times. — Tradition of a cannon-ball. —Botli ball and cannon discovered. 
 — Sunken plate and weapons. — Eclio at Walton Hall, — West view of 
 lake. — IIow to strengthen a bank. — Pike-catcliing. -Cats and pike. — 
 Spot where Waterton fell. 
 
 Waterton at home, and, what a home ! 
 
 It was not magnificent in the ordinary sense of the word. 
 Such magnificence may be the result of mere wealth, with- 
 out either taste, imagination, or appreciation. The veriest 
 ll boor in existence, who happens by some turn of fortune to 
 
 be put in possession of enormous wealth, need only give 
 the word, and he may revel in more than royal 
 magnificence. 
 
 As for the house itself, no expenditure could give it the 
 least pretence to beauty or statelmess. It is one of the 
 worst specimens of the worst era of architecture, and is 
 nothing but a stone box perforated with rows of oblong 
 holes by way of windows. 
 
 I tried on all sides to obtain a view of it which would 
 soften down its ugliness, but could not succeed. The 
 
 D 2 
 
 A I hnmc. 
 
 JJ'a'tnn 
 f/nll. 
 
m 
 
 HKKJHAPIIV. 
 
 front of the house is, strange to say, the worst part of it, 
 being a flat, smootli, stone wall, with three rows of ohlong 
 windows, eight in a row. The only speciinerl of architec- 
 ture which could approach it in this respect is a work- 
 house of the same date, those of modem times being 
 infinitely superior in architectural effect. 
 
 Why the grand old house should have been pulled down 
 to make way for such an edifice is ([uite inexplicable. 
 
 i 
 
 P?r! 
 
 ■>^t: 
 
 -=te3 
 
 \V\r.l(lN II Ml., V\ti'\\ llli: I. AUK. 
 
 ,- (i I 
 
 Thi: old 
 honsc. 
 
 Very few houses will be found with an oak-panelled hall 
 ninety feet in length. Yet all this was destroyed ; part of 
 the oak panelling was used in l)uilding a pigeon-house, and 
 the rest was burned. Such was the stale of architecture 
 in the days " when George the Third was king." 
 
 Unfortunately, no paintings or engravings of this most 
 memorable house are in existence, though there are in- 
 numerable plates of the "Seats of the Nobility and 
 
HlOUKAPllV. 
 
 87 
 
 most 
 Ire in- 
 {\nil 
 
 Gentry," most of them in the style satii'i/ed by llof^arth 
 in Ills " Marriage h hi Mode." 
 
 In fact, the architecture of that era is on a par with tlie 
 classical costumes of the stage. I liave possessed for 
 many years a volume of Shakspeare in which there is a 
 portrait of an actor in the part of Troilus. He is 
 classically costumed as a Trojan in a tiglit scale cuirass, 
 a sliort cloak, knee breeches and silk stockings, Eoman 
 buskins, a tie wig, a helmet with a vast plume of ostrich 
 feathers, and he is bi(hling defiance to Diomedes M'ith a 
 toy Moorish sword which would liardly cut off the head 
 of a wax doll. 
 
 So if Waterton had desired architectural magnificence, 
 he could not have obtained it, except by i)ulling the 
 house down, and building another. ]iut, he had no taste 
 for such magnificence, his life being one of rigid, not to say 
 severe, simplicity. 
 
 His personal expenses were such as could have been 
 covered by the wages of one of the labourers on his own 
 estate. His single room had neither bed nor carpet. Ho 
 always lay on the bare boards with a blanket wrapped 
 round him, and with an oaken block by way of a pillow. 
 As has been mentioned, he never touched fermented liquids 
 of any kind, and he took but very little meat. 
 
 "When I knew him, he always retired to his room 
 at 8 r M. Few men of his age would have chosen a 
 room at the very top of a large house; but stairs were 
 nothing to Waterton, whose liml)s were strengthened by 
 perpetual tree climbing. Punctually at three a.m., being 
 roused by the crowing of a huge Cochin China cock, which 
 he called his ' morning gun,' he rose irom his plank couch, 
 lighted his fire, lay down for half an hour, and was always 
 dressed and closely, or as he called it, ' clean ' shaven, by 
 four, when he went into the private chapel which was 
 
 A rch ilcc- 
 tuiv- (tad 
 the stuije. 
 
 Simplicity 
 of lif,: 
 
 The 
 »i(inii)i<j 
 (J tin.' 
 
.38 
 
 UIOGIJAPIIV. 
 
 IK' ' 
 
 (■' ' I '■' < ' 
 
 I'' 
 I- 
 
 llmor nvd 
 lancet. 
 
 next door to liis room, and wliere he usually spent an liour 
 in pi'ayer. 
 
 1 had several friendly altercations with him upon shav- 
 ing, hut he would as soon give up the lancet as the razor. 
 He would not even wear a particle of whisker, and kej)t 
 his thick, snowy hair within half an inch of length, lie 
 had not lost a hair, in spile of his advanced age, and I have 
 often thought that if he had allowed his hair and heard to 
 grow to their full luxuriance, a nobler figure could )iot have 
 visited an artist's dreams. 
 
 Then came reading Latin and Spanish hooks (Don 
 Quixote being always one), and then writing, receiving 
 bailiffs report, &c., until eight, when, at the stroke of Sir 
 Thomas More's clock, breakfast was served. So, he had 
 done a Aiir day's work and finished breakfast at the time 
 when most persons of his position in life had scarcely awoke. 
 
 Tn the next place, he was not a rich man. 
 Diviiiiu- As a rule, the old Yorkshire families are wealthy, and 
 estate. ' the Watertous would have been among the wealtliiest of 
 them, but for the shameful oppressions to which they were 
 subjected. That most accomplished robber, Henry VIII., 
 had confiscated the greater part of the estates, and what 
 with direct robberies, double taxation, fines, and so forth, 
 the estates were terribly reduced when he came into pos- 
 session of them. Even if he had wished it, nuignificence 
 would mjt have been attainable, but he achieved more 
 than magnificence, and with the restricted means at his 
 connnand, converted a Yorkshire valley into a veritable 
 wonder-land. 
 
 In this congenial task he was favoured by circumstances 
 which are not likely to occur again. He posser^s^^d the 
 requisite knowledge, a constitution of iron, and a frame of 
 astonishing endurance and activity. He came into pos- 
 session of the estate as a very voung man, onlv twenty- 
 
 I 
 
1 
 
 I 
 
 BIOGHAPIIY. 
 
 39 
 
 4 
 
 HltV- 
 
 four years of age, and remained absolute master for nearly 
 sixty years. 
 
 It was a pity that he did not bestow as much pains on 
 his estate as on his birds. But lie was no practical 
 agriculturist as his father had been. He could not do 
 anything which looked like oppressing his tenants, and 
 the consequence was, that they were habitually in heavy 
 arrears, and often threw up their farms without paying 
 rent, having iniijoverished the land and einiched them- 
 selves. 
 
 He loved natural history in all its Ibrins, but his chief 
 pursuit was the study of bird-life, and he modified the 
 grounds to the use of the birds, curing much more for their 
 comfort than his own. For this purpose the grounds were 
 admirably adapted by Nature, and he aided her by art. 
 There were a large moat and a succession of ponds for the 
 accommodation of aquatic birds. There weie swampy 
 places where the birds could feed. There were ruined edifices 
 for such birds as chose them for a residence, and the whole 
 of the park was covered with stately trees. Moreover, the 
 house stood on a stone island in the moat, and, as may be 
 seen from the illustration on page 36, permitted the habits 
 of the water-birds to be closely watched. 
 
 The first need was obviously to allow the birds to be un- 
 disturbed by boys and other intruders, and to prohibit the 
 firing of guns — the only sound which birds seem instinc- 
 tively to dread. But, as there was a public pathway run- 
 ning in front of "the house, he had great difficulty in 
 obtaining permission to close it. This object, however, 
 was at last attained, and he then began his wall. It is of 
 a roughly circular form, the house being near the centre. 
 Nowhere is it less than eight feet high, and where it runs 
 along the canal, it is more than double that height, in order 
 to protect the birds fi'om the guns of bargees. 
 
 Birds 
 
 versus 
 rental. 
 
 The Park 
 tvall. 
 
40 
 
 BIOGKAPHY. 
 
 ^ii'U 
 
 ' 
 
 II 
 
 ■li' 
 
 Barcjcs. These men, by the way, used to be most determined 
 poachers, and, on account of their mode of life, even if 
 detected and chased, they could escape by means of their 
 barges. They were chary, however, of venturing inside a 
 sixteen feet wall, and after a while ceased from troubling. 
 Such a work was necessarily very expensive, costing at 
 least ten thousand pounds. It was too large a sum to be 
 paid at once, and Waterton would not run in debt. So, 
 every year, he put aside as much money as could be spared 
 for the wall, went on building until the money was ex- 
 pended, and then stopped the work, and waited until the 
 following year to continue it, The wall was three miles in 
 total length, and inclosed an area of two hundred and 
 fifty-nine acres. 
 
 The value of this wall was shown by the fact that the 
 Heroni'ii. very year after it was finished the herons came and estab- 
 lished themselves within it. At my last visit in 1863, 
 there were nearly forty nests. 
 
 How should they know that a wall could protect them 
 against man ? It was no obstacle to them, and how they 
 could have known, as they evidently did, that it was an 
 obstacle to mankind is one of the yet unsolved problems 
 which puzzle students of zoology. Moreover, they knew 
 that those few specimens of humanity who came within 
 the wall would do them no harm. I have often been in 
 the heronry, with the blue fragments of broken eggs lying 
 on the ground, and seen the herons going to and from their 
 home with perfect unconcern. Even on the ground, the 
 herons had no fear of man. Provided that a man ap- 
 l)roaclied them slowly and quietly, he could come close 
 enough to see their eyes, and even to notice the reflection 
 of the rippling water upon their grey plumage. 
 
 Not only in the heronry, but in other parts of the park 
 near the walci-, the birds would allow themselves to be 
 
 « 
 
 ^^;i 
 
-^' 
 
 BKXinAPlIY. 
 
 41 
 
 approached quite closely, so that their peculiar lm1)its Tmnencss 
 could be watched. I was able to secure slight sketches "-^ iJoons. 
 of the characteristic, and almost grotesque attitudes as- 
 sumed by the heron, and have selected three as examples. 
 
 ang 
 
 their 
 
 the 
 
 Fig. 1. shows the bird in a position which, in common 
 with the flamingo, stork, and other long-legged wading 
 birds, it is fond of assuming. It doubles its legs under 
 the body, thrusts the feet forward, sinks its head upon its 
 shoulders, so as to conceal the long neck, and remains 
 so motionless and so unlike a heron that it might easily 
 be passed without notice. 
 
 Fig. 2 shows the heron standing on one leg at rest. By 
 moving cautiously round the bird, I succeeded in getting a 
 back view, so as to show the perfect balance of the body 
 on the single leg (Fig. 3). 
 
 Waterton had a special love for the heron, and frequently 
 alludes to the services which it renders to the owners of 
 fish-ponds. 
 
 "Formerly we had a range of fish-ponds here, one 
 above the other, covering a space of about three acres of 
 ground. Close by them ran a brook, from which the water- 
 rats made regular passages tlirougli the intervening bank 
 into the ponds. These vermin were engaged in never- 
 ceasing mischief. No sooner was one hole repaired than 
 another was made ; so that we had the mortification to see 
 
 Ohl_fi,sJ,. 
 jionds. 
 
 IVutn- 
 rals. 
 
■■■ 
 
 42 
 
 lUOGI^VPIIY. 
 
 Value of 
 heron. 
 
 ! i 
 
 I ! 
 
 the ponds genomlly eij^lit or ten inches below water-mark. 
 TJiis encouraged the yrowth of weeds to a most incommo- 
 dious extent, which at last put an end to all pleasure in 
 fishing. Finding that the 'green mantle from the standing 
 pool ' was neither useful nor pleasant, I ordered the ponds 
 to be drained, and a plantation to be made in the space of 
 ground which they ha<l occupied. 
 
 " Had I known as much then as I know now of the 
 valuable services of the heron, and had there been a good 
 heronry near the place, I should not have made the change. 
 Tiie draining of the ponds did not seem to lessen the 
 number of rats in the brook ; but soon after the herons 
 had settled here to breed, the rats became extremely 
 scarce ; and now I rarely see one in the place, where 
 formerly I could observe numbers sitting on the stones at 
 the mouth of their holes, as soon as the sun had gone 
 down below the horizon. I often watch the herons on the 
 banks of some other store-ponds with feelings of delight ; 
 and nothing would grieve me more than to see the lives of 
 these valuable and ornamental birds sacrificed to the whims 
 and caprices of man." 
 
 A portion of one of these now dry fish-ponds may be 
 seen in the illustration of the " Grotto," on page 68, On 
 such a rich soil as that afforded by the bsd of an old fish- 
 pond, the trees grew with great rapidity, and the spot is 
 now a singularly picturesque one, with bold effects of 
 light and shade, and shelter from the wind iind sun. 
 
 The next important work was the extension of the moat, 
 a long and costly operation. 
 
 The present house is comparatively modern, standing well 
 clear of the water. But, the original house extended to 
 the M'ater on the south side, and was a fortified building of 
 sufficient strenuth to justifv a siei:e under Cromwell's 
 personal direction. 
 
 
BIOGRAPHY. 
 
 43 
 
 lavk. 
 imo- 
 :e in 
 iding 
 ioud3 
 ,ce of 
 
 f the 
 
 good 
 lange. 
 n the 
 lerons 
 Binely 
 where 
 •lies at 
 I gone 
 
 11 the 
 
 light ; 
 ives of 
 
 whims 
 
 |nay he 
 Oil 
 id tish- 
 [jpot is 
 icts of 
 
 Unfortunately, Waterton's father destroyed this historical 
 huiUling to make room for the present house, and ahnost 
 the only relic of this fortification is tlie old gateway, with 
 its central tower and Hanking turrets, and said to he more 
 than a thousand years of age. 
 
 ;e gate itself is ff very thick oak planking, pierced The gate- 
 \\l oopholes for musu .ry, and bearing tangible evidences ""^ 
 
 
 OAIKWAV AND IVY TDWRH. 
 
 moat, 
 
 11 g well 
 Ided to 
 ling of 
 tiiweU's 
 
 of the siege in the shape of many bullet marks. In the left 
 l)ortion of the gate there is a ball still remaining, which is 
 distinguished by an iron ring round it bearing an inscription 
 to the effect that it was fired by Oliver Cromwell himself. Oliver 
 That he took an active part in the siege is well known, hut ^'■<"""''" ■ 
 it is difficult to identifv anv individual bullet which he 
 
44 
 
 inodiiAPiiY 
 
 I'alshii/ 
 tlic sictjc. 
 
 Drnic- 
 biidye. 
 
 fired. The triulition further states tliat tlie sliot was niiued 
 at the lady of the linnsp, who gallantly conducted the 
 defence herself. The reader may he interested to hear 
 that lier defence was successful. 
 
 The sketch, re[)resenting the Gateway in its present con- 
 dition, was taken on the opposite side of the water, from a 
 spot close to the tali and lightning-shattered poplar-tree, 
 shown on tlie riglit hand of the illustration on p. 3G. 
 Tiie chief interest of this view lies in the gateway itself. 
 Just behind it is an odd-looking tower, which was built by 
 Waterton for the use of starlings, and the place is enclosed 
 on the nortli by a thick and closely-clipped hedge of yew. 
 The heavy masses of ivy which fall in thick clusters from 
 the turrets and which serve as a refuge for many birds, 
 have given to the structure the name of Ivy Tower, by 
 which it is often mentioned in the Essays. 
 
 While still very you ng, I was familiar with the Ivy 
 Tower from Waterton's Essays. They mostly appeared in 
 London's Maffazine of Katiiral Hisfor)/, and as that valu- 
 able publication was taken in at the Ashmolean Society 
 of Oxford, where I lived, I used to watch impatiently for 
 each successive number, in the hope that it might contain 
 an article from Waterton's pen. Thus, the gateway, the 
 lake, the heronry, the starling towers, the fallen millstone, 
 the shattered poplar, the holly hedges and the wooden 
 pheasants, were all known to me, and when at last I had 
 the privilege of visiting Walton Hall, there was not one of 
 those spots that I did not joyfully recognize. 
 
 In the old times, the only approach to the mainland was 
 by a drawbridge, opening oji to the gateway, which was 
 then three stories high. This has long been destroyed, 
 and at present the approach is made by a light iron bridge, 
 rather to the right of the gateway. This bridge is not 
 shown in the sketch. 
 
 I 
 
r.IniiUAPllY. 
 
 45 
 
 A fionil 
 shot. 
 
 As to the siege, tliere are other remiuiscences beside the" 
 gateway itself. 
 
 While the soldiers of Cromwell were occupying the hill 
 nearly opposite the gateway, one of the soldiers started off 
 with a keg on his shoulder to futch beer from the village. 
 Thinking that he would return by the same route, one of 
 the garrison aimed a little gun which was mounted on the 
 topmost story, so as to command the path. Ihe soldier 
 did return by the same way, and was struck down by the 
 ball, which passed through his thigh. 
 
 The tradition of this lucky shot was handed down from 
 father to son, until it reached W'terton's father. He had 
 the curiosity to dig at the spot .vhere the man was said to 
 have fallen, and there he found the ball, a little iron one. The half. 
 This he gave to his son, with a request that it should 
 always remain in the family. 
 
 In 1857, while dredging away the drift mud which had 
 accumulated ronnd the gateway, a small iron cannon was 
 discovered. As the ball fitted it, and it was found exactly 
 below the turret from which the fatal shot had been fired, r/ie 
 tliere could be no doubt that it was the identical gun ''"-"'""'' 
 
 ( ii.\ i:i!iN'. 
 
 mentioned in the tradition ; so VA'aterton had the pleasure 
 of placing the cannon and the ball together in his house, 
 where every visitor could see them. 
 
 Beside the gun, there were found a sword-blade, a spear, Smikm 
 daggers, axe, many coins, keys, and some silver V^^^^- anTplZ. 
 For their presence in the mud Waterton accounts by 
 
I, I 
 
 4() 
 
 HI()(JI{A1'IIV. 
 
 ^il i 
 
 I 
 
 E"ho 
 Stniif. 
 
 The 
 
 uillow,!. 
 
 snj^gosting tliiit tlicy wcii' lliuij,' into tlic moat, wlieii tliu 
 liouso was ransacked for arms after the battle of Ciilloden. 
 Ho told me that he believed that if the lake were completely 
 drained, many more such articles would be recovered. 
 
 The view on pai^e 'Mi is taken from a spot on the northern 
 bank. At some hundreds of yards distance from the house 
 there are a coui)l(^ of splendid sycamores, and close to them 
 is a large block of ironstone, called the Kclio Stone. Any 
 one standing by it, and speaking towards tlie house, will 
 hear every syllabhi returned with wonderful clearness. 
 Sitting on this stone, I made the sketch from which this 
 illustration is taken. On it is engraved the word echo. 
 
 On the western side of tlie gateway there had been a 
 curious old chapel formerly attached to the mansion, 
 Waterton, however, disliked it and took it down, against the 
 remonstrances of the then Duke of Xorfolk, his godfather. 
 
 The lake is widest near the house, and then proceeds 
 almost due west, narrowing as it goes, and taking a turn 
 northwards towards the end, where it passes round a hill, 
 and becomes shallower, allowing the sedges and reeds to 
 appear, and so affording shelter for the aquatic birds. 
 
 Another view of the lake is now given, looking west- 
 ward, and taken from the right-hand first floor window of 
 the house as seen on page 36. 
 
 On the ground-floor may be seen a large window, flanked 
 by a smaller one on either side. These are the west 
 windows of the drawing-room. The central window is a 
 large sheet of plate glass, and behind it is mounted a large 
 telescope, commanding nearly the whole of the lake. 
 
 On the left, before coming to the wood, are a few 
 willows, and between them and the wood is a favourite 
 resort of the herons. The low bank looks as if it would 
 be endnuiiered by the water, but it is perfectly firm, even 
 to the very edge. It is made of large stones, not squared. 
 
IHcHJIIAPIIV 
 
 Ml tll<^ 
 
 )lctoly 
 
 1. 
 
 rthcvn 
 
 house 
 3 tbein 
 . Any 
 3e, will 
 nirness. 
 ch this 
 Clio, 
 been a 
 lansion. 
 linst i\w 
 Kather. 
 :)roceecls 
 r a turn 
 a hill, 
 
 eeds to 
 
 s. 
 
 g west- 
 
 dow of 
 
 but heaped loosely to^'ether. Seeds of various trees, esi)e- 
 cially those of the syejunore, fcill into the water, floated on 
 its surface, and were arrtisted by the bank, where they 
 took root. Tliev were nev(!r allowed to urow into trees, 
 and were constantly cut down. But theii loots twined 
 themselves aniontj the stones, and bound thein tofjether 
 so firmly, that a stronu'cr wall could not be desired. 
 
 •, ' .-^ s, - 
 
 Wiitrr- 
 vail. 
 
 ,AKK. l.dOKISli WKSIW'AUli, 
 
 , flanked 
 
 Ihe west 
 
 low is a 
 
 a large 
 
 Ke. 
 
 |e a few 
 
 favourite 
 
 it would 
 
 •m, even 
 
 squared. 
 
 Tlie holes under these stones are favourite resorts of 
 pike, with which the lake abounds. 
 
 1 am no angler, but I have caught many i)ike near the 
 willows by trolling, using nothing but a willow stick by 
 way of rod, a hank of whipcord for a line, a gorge hook, 
 and a minnow for bait. The largest that 1 ever took there 
 M'eighed rather over ten pounds, and very proud I was of 
 the lish, though it was a heavy and inconvenient article to 
 carry to the house. 
 
 Some of the pike, incliuling the ten -pounder, were for 
 
 I'ih 
 
rr 
 
 -IS 
 
 Vnt 
 
 feviliiiij, 
 
 I'.KMiKAI'IIV 
 
 tlio table, but tho llsh wero ^'(iiieniUy ust'tl for tho purpose 
 of feeding the cats, of which there were nmny about tho 
 stubh'S and cattle-yards, for tlu! purpose of keeping down 
 the rats. It is now well known that a W(!ll-fed cat is the 
 best mouser, seldom eating its jtrcy, but killing it for tho 
 mere sport. 
 
 AVhen the cats were fed, tho fish were chopped up on a 
 wooden block near the stables. It was very amusing to 
 watch the opeiation. Although at lirst not a cat might bo 
 visible, half a dozen blows liad not been struck with the 
 chojjper before impatient cries were heard and cats came 
 swarming round the block, just as they do round a cat's- 
 meat man's barrow in London. 
 
 On the right, just above the tall tree near the edge of 
 
 the lake, a heron is seen Hying in the distance. It was 
 
 The fatal near the bank at the further end of the lake that Waterton 
 
 spot. YCidt with his fatal accident at a spot neaily below the 
 
 Hying heron. 
 
 " 
 
 'I 
 
 i i 
 
CHAITKII IV. 
 
 l.uve of ti'003. — rivsci'vation f duiiiiij^cd tiops, — ilow tivc« jKiinli. — Wiinl 
 mid rain. — Self-restorative powers of *]\o bark.— Hidden foi.s,— The 
 fungus and its work. — Use of the wooditecker and titmouse;.— How to 
 utilize tree-stumps.— The Cole Titmousj. — Owl-house nnd seat.— Dry- 
 rot. — When to ]mint tinibu;-. — Onkc; j^ates of ^lie old tower. — Com- 
 mand over trees.— How to make the holly grov uiekly- The holly as 
 a hedge-tree — Pheasant fortresses. — Artiticia) [-'i nisants. —The ])oacher.s 
 outwitted. — Waterton's power of tree-elimbing. — An >< lial study. — 
 Ascending nnd descending trees. — ( ijuv- h and State Uxv.-,. — The yew. 
 — A protection against cold winds. - Ye.> hedge at back of gateway. — 
 The Starling Tower. — Fannliarity of the birds. — The I'icnic or C!i ttt > 
 — Watcrton's hospitality. — "The S(|uiro"— A decayed mill and aban- 
 iloned stone. — The stone lifted oil the ^-round by a hazel nut. 
 
 Waterton's love of trees alnio-t amounted to veneration. 
 He studied their way.s as minutely and as accurately as he 
 did those of the animal world, and in consequence he could 
 do more with trees than any one else. By patient observa- 
 tion of their modes of growth, he knew how to plant them 
 in the locality best ;:;iiit?d for themselves, how to encourage 
 them, and, if they wtm injured, to reduce their damage to 
 a minimum. 
 
 Many a fine tioe has he shown me which would have 
 been long o-^c condemned by ignorant men, but vhich was 
 then flourishing in full growth, and in such renewed health 
 that scarcely a scar was left in the bark to show the spot 
 on which the injury had occurred. 
 
 One of his triumphs in this art was to be seen by a 
 splendid po])lar situated nearly opi)osite the pictures(jue 
 
 K 
 
 Loce of 
 trees. 
 
I 
 
 s; 
 
 
 50 
 
 Ilcaliiui 
 
 Tree 
 cJiinbhia, 
 
 BIOfn^APHY. 
 
 gateway, and especially favoured by Waterton as having 
 been ■• anted by his father. It was twice struck by light- 
 ning, and the trunk split open for many feet. 
 
 However, Waterton iilled up the breaches, and in course 
 of time the tree recovered itself (see p. 3G). It was in 
 full growth during uiy last visit, but it was blown down 
 by a severe gale in 18G9, having succumbed, not to the 
 lightning, but to age. To heal a tree by filling it with bricks 
 and mortar may appear to be rather a singular method, but 
 it is a very effectual one ; the chief object being to keep 
 rain out of the tree, and so to guard it against rotting. 
 
 How thoroughly Waterton had studied the ways of trees 
 may be seen from the following extract from his essay on 
 the Titmouse and the Woodpecker, in which he combated 
 the popular opinion that these Ijirds were injurious to 
 trees : — 
 
 " Would you inspect the nest of a carrion crow ? Brittle 
 are the living branches of the ash and sycamore; while, 
 on the contrary, those which are dead on the Scotch pine 
 are tough, and ^^'\\\ support your weight. The arms of the 
 oak may sal'ely be relied on ; but, I pray you, trust with 
 extreme caution those of the quick -growing alder. Neither 
 press heavily on the linden tree ; though you may ascend 
 the beech and the elm without any fear of danger. But 
 let us stop here for the present. On some future day, 
 should I be in a right frame for it, I may pen down a few 
 remarks, which Avill possibly be useful to the naturalist 
 wlien roving in quest of ornithological knowledge. I will 
 now confine myself to the misfortunes and diseases of 
 trees ; and I will show that neither the titmouse nor the 
 wood])ecker ever bore into the hard and live wood. 
 
 " 1 rees, in general, are exposed to decay by two different 
 processes, independent of old age. The first is that of a broken 
 branch, which, when neglected, or not cut off close to the 
 
niOGRAPIIY. 
 
 51 
 
 light- 
 
 course 
 was in 
 I down 
 to the 
 1 bricks 
 lod, hut 
 to keep 
 ing. 
 
 of trees 
 3ssay on 
 amhatecl 
 Lrious to 
 
 ■ Brittle 
 ;; while, 
 )tch pine 
 ns of the 
 •ust with 
 
 Neither 
 y ascend 
 tev. But 
 
 ure day, 
 »vn a few 
 
 aturalist 
 I will 
 
 soascs of 
 nor the 
 
 different 
 ■ a broken 
 i)se to the 
 
 
 parent stem, will, in the course of time, bring utter ruin Decay of 
 on the tree. The new wood, wliich is annually formed, '''^^^' 
 cannot grow over the jutting and fractured part, into which 
 the rain enters, and gradually eats deeper and deeper, till 
 at last it reaches the trunk itself. There it makes sad 
 havoc ; and the tree, no longer able to resist the fury of 
 the tempest, is sjilit asunder, and falls in ponderous ruins. 
 But ere it comes to this, the titmouse will enter the cavity 
 in a dry spring, and rear its young ones here. Now, if 
 the diseased or fractured branches were carefully cut off 
 close to the bole, you would see the new accession of wood 
 gradually rolling over the flat surface, which, in time, would 
 be entirely covered by it ; and then the tree would be 
 freed for ever from all dauger in that quarter. The second 
 process towards decay is exceedingly curious, and cannot 
 well be accounted for. If it takes place to a serious ex- 
 tent, no art of man can possibly save the tree ; and sooner 
 or later, according to the magnitude of the disease with 
 which it has been tainted, it will fall before the force of 
 the raging winds. Should this disease be slight, the timely 
 prevention of rain from penetrating the injured part will 
 secure the tree from further mischief. 
 
 •' I must here observe that, in animated nature, the vital 
 functions are internal ; so that, if the part within be 
 mortally wounded, death is the inevitable consequence. 
 With most trees, and with all those of ]3ritain, it is other- 
 wise. Their vitality is at the periphery, connected with FitaJif;/ of 
 the bark, under which an annual increase of wood takes ^^"^ ^"''^'' 
 place, so long as the tree is alive. Should, however, the 
 bark be cut away, the tree will die upwards from the place 
 where all the bark has been destroyed. Not so with its 
 internal parts. You may entirely excavate the interior of 
 a tree ; and provided you leave a sufficient strength of wood 
 by way of wall, in order that it may be able to resist tlie 
 
 E 2 
 
^iii!' 
 
 ii 
 
 !/ 
 
 ■ I 
 
 
 li 
 
 52 
 
 Fatal 
 fungus. 
 
 T5I0GRAPHY. 
 
 fury of the tempest without, taking care ai, the same time 
 to exclude the rain, your tree will remain in vigour from 
 generation to generation. 
 
 " The internal texture of a tree will perish without any 
 notice by which we may be forewarned of the coming ruin. 
 The disease which causes the destruction takes place in 
 the oak ; but more frequently in the sycamore, and most 
 commonly of all in the ash. We will select this last tree 
 by way of elucidation. 
 
 " Often, when arrayed in all the bloom of vegetable 
 beauty, the asli-tree is seen to send forth from its bole, or 
 from some principal branch, a small fungus, which, during 
 the summer, increases to a considerable size. It ripens in 
 the autumn, and falls to the ground when winter's rain 
 sets in. The bark tlirough whicli this fungus sprouted is 
 now completely dead, though it still retains its colour ; and 
 that part of the wood from which it proceeded is entirely 
 changed in its nature, the whole of its vitiated juices 
 having been expended in forming and nourishing the 
 fungus. Nothing remains of its once firm and vigorous 
 texture. It is become what is commonly called touch- 
 wood, as soft and frangible as a piece of cork, which, when 
 set on fire, will burn like tinder. In the meantime, the 
 tree shows no sign of sickness, and its annual increase 
 goes on as usual, till at last the new swelling wood closes 
 over the part from which the fungus had grown, and all 
 appears to go on right again. But ere the slow process 
 arrives at this state the titmouse or tlie woodpecker will 
 have found an entrance and a place of safety for their in- 
 cubation. They quickly perforate the distempered bark, 
 and then the tainted wood beneath it yields to their 
 pointed l>ills, with which they soon efl'ect a spacious 
 cavity. 
 
 " Here, tlien, we have the whole mystery unfolded. These 
 
 I 
 
BIOGRAPHY. 
 
 53 
 
 le time 
 ir from 
 
 »iit any 
 ig ruin. 
 )lace in 
 id most 
 ast tree 
 
 egetable 
 bole, or 
 I, during 
 :'ipens in 
 er's rain 
 •outed is 
 3ur; and 
 entirely 
 d juices 
 hing the 
 vigorous 
 toucli- 
 ch, when 
 irae, the 
 increase 
 od closes 
 and all 
 ,v process 
 cker will 
 their in- 
 red bark, 
 to their 
 spacious 
 
 id. These 
 
 JVood- 
 
 peckers. 
 
 birds, which never perforate the live wood, find in this diseased 
 part of the tree, or of the branch, a place suitable to their 
 wantr.. They make a circular hole large enough to admit 
 their bodies, and then they form a cavity within sufficiently 
 spacious to contain their young. Thus does nature kindly 
 smooth the way in order that all her creatures may prosper 
 and be happy. Whenever I see these sylvan carpenters 
 thus employed I say to them, ' Work on, ye pretty birds ; 
 you do no harm in excavating there. I am your friend, 
 and I will tell the owner of the tree that you are not to 
 blame. But his woodman deserves a severe reprimand. 
 He ought to have cut down the tree in the autumn, after 
 the appearance of the fungus.' " 
 
 Even when the tree was hopelessly destroyed by the Tree 
 fungus, Waterton would still find uses for the stump. He 
 would clothe it with ivy so as to render it picturesque, 
 and he would manipulate it so that it should be a home 
 for birds. 
 
 Many of these stumps are in the grounds, and of them 
 1 have selected one or two as examples. 
 
 The first shows the " brick and mortar " svstem which Brick and 
 has already been mentioned. Several habitations for birds """"''"'■• 
 are constructed in it, and the stone in front is intended to 
 aid the observer in looking into the nests. I tried to 
 sketch this stump so as not to make it look like a 
 grotesque human face. But exactness was the first con- 
 sideration, and it is represented precisely as it was in 1863. 
 
 Tlie second sketch was chosen because it represents one 
 of the fungus-visited ash-trees described by Waterton. Old ash. 
 
 The tree has been broken off some ten feet from the 
 ground, at a spot weakened by a fungus. Of th3 tree itself 
 little remains except the broken stump and a few small 
 branches which still retain their leaves. Ivy has ascended 
 
■^ 
 
 ' ^i 
 
 1 1 
 
 I:! 
 
 i! 
 
 '► 
 
 54 
 
 I3T0G1UPIIY. 
 
 it, and is hanging in heavy clusters, so as to give the 
 fast-dying tree a venhue not its own. And, as the reader 
 may observe, two more masses of fungus are projecting 
 from the tree and extracting the life from its fibres. 
 
 The Cole- 
 ti/moHse. 
 
 Dlil AYliD ASH AND liUIl K. 
 
 Just above the upper fungus and on its right is a small 
 door, with a hole near the top, and this little door has 
 rather a curious history. 
 
 In the spot where the door is showr there is a fungus, 
 proving that the wood from which it has sprung was 
 decayed. Now, Waterton had for some time Avanted the 
 Cole-titmouse to breed in his park, and, in accordance with 
 this notion, ]irovided it with a home. First, he separated an 
 oblong piece of wood about an inch in thickness so as to 
 lorni a door. Next, he cut away the soft decayed wood 
 until he had formed a considerable cavity. He then 
 
niOGnAPiiY. 
 
 56 
 
 G the 
 'eader 
 ecting 
 
 replaced the door, fastening it with two little hinges and 
 a hasp, and bored a hole in it about an inch in diameter. 
 
 In fulfilment of his expectations, the very bird which 
 he wanted soon discovered the locality, examined it care- 
 fully, and then built in the chamber so thoughtfully provided 
 
 A 
 
 fni till' red 
 liidcjer. 
 
 liKCAVKlJ ASll AND KlN(ir.S. 
 
 a small 
 loor has 
 
 fungus, 
 Lng was 
 Ited the 
 Lee with 
 Irated an 
 
 so as to 
 led wood 
 
 Le then 
 
 for it. I was never at Walton Hall while the bird was 
 sitting, but have often seen the nest. 
 
 The last of these sketches represents a singularly in- 
 ffenious combination of accommodation for man and bird 
 The trunk of an old oak-tree has been hollowed out, and 
 the interior is divided into two stories. 
 
 In the upper there are nesting-places for birds, (^specially 
 for owls, and in the lower there is a scat where the occu- 
 ])ant can remain unseen. It is placed on the brow of the 
 hill which borders the lake, and is so arranged that not 
 
f M 
 
 !i! 
 
 
 ^ 
 
 
 ii-: 
 
 |i 
 
 ill 
 
 !i 
 
 I I 
 
 m 
 
 BIOORAPIIY. 
 
 only can the observer watcli from its shelter the habits of 
 
 Ou-l-hmsr. the various aquatic beings which frequent the lake, but 
 
 can actually look into the nests built on the tops of lofty 
 
 (iwi- iiousr. Asn skai\ 
 
 trees without the birds suspecting that their movements 
 could be seen. 
 
 With regard to the decay of wood after it had been 
 felled, Waterton was not long in coming to the conclusion 
 that the " dry-rot," as it is oddly named, was caused not 
 so much by external moisture as by the natural sap of the 
 tree which had not been thoroughly expelled. When its 
 Dnj-rot. juices have been completely dried and it is thoroughly 
 " seasoned," wood is as lasting as stone. We have in the 
 British Museum specimens of woodwork which, although 
 more than three thousand years have elapsed since the 
 trees were felled, are as sound as when they were first 
 carved. Waterton used to say that paint was the chief 
 cause of dry rot, especially when it was used to cover the 
 deficiencies of ill-seasoned wood, because it closed the 
 pores and did not allow the sap to escape. As a proof 
 
BIOGHAl'HY. 
 
 bits of 
 :e, but 
 f lofty 
 
 Paint. 
 
 ements 
 
 been 
 iclusion 
 ised not 
 of the 
 len its 
 roughly 
 in the 
 though 
 ■nee the 
 re first 
 le chief 
 ver the 
 sed the 
 a proof 
 
 Seasoned 
 wood. 
 
 that weather does not injure well-seasoned wood, he was 
 wont to point to certain posts, gates, and other articles 
 made of oak, which had never been painted, and which 
 had been in the open air for some seven hundred years, 
 and were perfectly sound. The oak doors of the gateway 
 are fully seven hundred years old. They are pierced and 
 torn with musket-balls, bui are still free from decay. 
 
 When he had new doors made which would be exposed 
 to the weather he used every precaution to keep the wet 
 from lodging in them. No panels were seen on the outer 
 side, which was as smooth as it could be made. The 
 corners were bound with strong iron, painted before it was 
 put on. 
 
 No matter how well-seasoned the wood might be, if the 
 doors were made of deal, three years were allowed to 
 elapse before painting, while, if of oak, it was never 
 painted until six years had passed, and very often was not 
 painted at all. It is also found that if holes were bored 
 transversely into posts, so as to allow free entrance of air, 
 the dry-rot scarcely ever made its appearance. If modern 
 builders would act upon a knowledge of this fact they 
 would render our houses, roofs of buildings, &c., far more 
 enduring than Mi^y are at present. 
 
 Did we wish to show the wonderful command which 
 Waterton had over trees, we need only point to the holly- Th Holly. 
 trees in his park. The holly was a great favourite of his, 
 as it is very hardy when properly planted, possesses a 
 remarkable beauty of its own, affords shelter for birds in 
 winter as well as summer, and can be formed into a. hedge 
 impenetrable to man and beast. 
 
 As to laurel hedges, Waterton never would plant them, z<v?nr?,<'. 
 and he had found by experience that in ordinary hawthorn 
 hedges a bush would often die without any apparent 
 
5fi 
 
 BIOGIiAPIIY. 
 
 Forcing 
 holfi/. 
 
 1 .' 
 
 reason, leaving an unsightly gap which could not be filled 
 up. In most hands the holly is a slow-growing tree, hut 
 Waterton made it grow with astonishing rapidity. 
 
 How he managed to "force " the holly may be seen from 
 ills own words. 
 
 " People generally imagine that the holly is of tardy 
 growth. It may be so in ordinary cases, but means may 
 be adopted to make this plant increase with such effect as 
 to repay u ^ amply for all our labour and expense. 
 
 " Thus, let us dig the ground to a full yard in depth, 
 and plant tlie hollies during the last week of May, taking 
 care to puddle their roots well into the pulverized soil. 
 We shall find by the end of September that many of 
 the plants will have shot nearly a foot in length, and that 
 not one of them has failed, let the summer have been 
 never so dry. 
 
 " Small plants, bought in a nursery, and placed in your 
 own garden for a couple of years, will be admirably 
 adapted for the purpose of transplanting. Had I been 
 aware in early life of this increasing growth of the holly, 
 it should have formed all my fences in lieu of haw- 
 thorn." 
 
 I tried this plan with perfect success upon a stony and 
 ungrateful soil. The rationale of the process is, that 
 Jluots and the youug rootlets, which ought to be carefully spread by 
 branches. ^\^q fingers, are able to draw nourishment rapidly from the 
 earth, and in consequence throw up branches in proportion. 
 Waterton advised me to cut down the young hollies at 
 first, and his advice was most valuable, although it cost 
 some pangs when followed. 
 
 I mentioned just now that a good holly hedge is imper- 
 vious to man and beast. So it is ; and not even the rat, 
 stoat, weasel, or even that worst of poachers, the cat, can 
 get through it. True, they might push their way between 
 
BTOGRAniY. 
 
 [,<:} 
 
 J filled 
 20, but 
 
 11 from 
 
 tardy 
 IS may 
 [tect as 
 
 depth, 
 taking 
 id soil, 
 lany of 
 nd that 
 e been 
 
 in your 
 
 mirably 
 
 I been 
 
 e holly, 
 
 f haw- 
 
 )ny and 
 is, that 
 |read by 
 'om the 
 portion. 
 )llies at 
 it cost 
 
 imper- 
 
 |the rat, 
 
 :at, can 
 
 tietween 
 
 Prkkhi 
 Im irs. 
 
 i'-rx. 
 
 the stems, but there is one obstacle wliich prevents them, 
 namely, that they cannot put their feet to the ground. 
 
 The holly is perpetually shedding its leaves, especially 
 in summer-time, in order to make way for the new leafage. 
 The old leaves fall, become dry, and curl up, with their 
 sharp spikes projecting in all directions. These points, 
 sharp as needles, prick the feet of the prowling animals, 
 and so prevent them from passing. 
 
 Of this property Waterton took advantage. Like many 
 landed gentlemen he had a preserve of pheasants, and was Poiie/tr 
 consequently harassed by poachers. Now he hated prosecu- 
 tion, and always evaded it if possible. On one occasion, 
 for example, when eight men and a boy were captured on 
 Sunday morning, while trespassing in his rookery, he 
 released them on finding that they were tailors, saying that Ninth 2mH 
 he could not think of prosecuting eight-ninths and a half 
 of a man. 
 
 So with the poachers in his preserves. He would not 
 expose them to be shot by keepers, nor would he pro- 
 secute them if he could help himself, but he could circum- 
 vent them, and did so effectually by means of the holly. 
 
 The preserves were situated at some distance from the 
 house, so that the poachers could make a rapid inroad and 
 carry off their booty before they could be seized. So 
 Waterton laid a deep scheme. First he planted near the 
 house, and just opposite his window, a clump of yews, on 
 wliich trees pheasants are fond of perching. Next he sur- 
 rounded them with a thick holly hedge, leaving only one Holly 
 little gap, which could be closed by a strong padlocked /'"'''■^''•'•• 
 gate. Then, leaving the trees to grow, he set about the 
 other preparations. 
 
 He made a number of wooden pheasants, and did it in 
 the simplest manner imaginable. He got some small 
 scaffolding poles and cut them diagonally into pieces about 
 
I 
 
 no 
 
 niOGHAPIIY. 
 
 
 Vf 
 
 
 . 1 1 
 
 iii 
 
 ii'< 
 
 Wnod'-n as loiig m a plieasant's body. A lath fastouecl to one oiid 
 j)kca,iaiit.i. j^jj^Jq j^ capital tiiil, and all that was needed was to trim 
 the shoulder to the neck, and put a head on the other end, 
 a nail doing duty for a bealc. 
 
 ''mm 
 
 T 
 
 ^ 
 
 STni'fTrRK OF WOflDION PHEASANT. 
 
 By the time that the trees had grown sufficiently for his 
 purpose he had made about a couple of hundred of dummy 
 l)heasants. He then threw a few sacks full of beans 
 inside the holly hedge, and laid a train of beans into the 
 preserve. The birds, finding the beans on the ground, 
 naturally followed the trail, and reaching so abundant a 
 supply of food as they saw inside the hedge, flew over it 
 and feasted to their heart's content. Then, not caring to 
 fly, after having gorged themselves, they settled for the 
 night in the yews. 
 
 Meanwhile the wooden pheasants were nailed on the 
 trees in the preserve, and so exactly did they resemble the 
 actual birds that in the dark no one could detect the 
 imposition. Even in daylight the dummy so closely re- 
 presents the bird that a second glance is necessary in 
 order to make sure that it is only an imitation. The ac- 
 companying sketch represents one of these dummies on 
 the outskirts of the preserve. 
 
 •fl 
 
lilOGUAPUV. 
 
 61 
 
 The poachers wei'e completely deceived, sind Watertou poacher^ 
 used to eiijoy the reports of their guns, knowing that they ""'"'' 
 were only wasting their shot upon the wooden images, 
 
 .y for his 
 ■ dummy 
 beans 
 into the 
 ground, 
 mdant a 
 V over it 
 3aring to 
 for the 
 
 on the 
 luible the 
 stect the 
 [osely re- 
 fjssary in 
 
 The ac- 
 kmies on 
 
 WOODEN PHEA.^ANI IN TUEK. 
 
 while the real birds were comfortably asleep under 
 his eye. 
 
 If the reader will refer to. the illustration on page 36, 
 he will see that on the right hand, and near the poplar, is 
 a rather curious circular olyect. This 'represents the 
 pheasant fortress in question, and, although the small size 
 prohibits any detail, the general shape and appearance are 
 sufficiently shown. It will also be seen how close to the 
 house is the fortress, so as to be under the master's eye. 
 
 He made several more of these ingenious refuges, of 
 which other birds besides the pheasants took full ad- 
 vantage. 
 
 There was not n tree in the park that Waterton did not Knowhdgr 
 know, and, if the smallest damage were done, he would "^ '''*^" 
 be sure to find it out. One day 1 found the kee})er much 
 disturbed, having discovered some shot in a tree trunk, 
 and being quite sure that he would be called to account 
 
I 
 
 (;2 
 
 niOOUAPi.Y. 
 
 ^ 
 
 
 1 t 
 
 1 " 
 
 1 > 
 
 l' i 
 
 ^ i 
 
 Aerial 
 study. 
 
 ior it. 1'lie mail was vv^\i 011011,^11, fur Wiitertori found tlio 
 shot, before many liours had passcsd, and the keeper had 
 to undergo a severe cross-examination. 
 
 Not only did lie know the trees individually, and had 
 distinctive names for them, hut there was scarcely one 
 which he had not clinil)ed, and in the tojiinost branches 
 of which he had not sat, pursuing his favourite amuse- 
 ments of watching birds, and reading Horace or Virgil. 
 There are not many men who at the age of sixty would 
 have either the power or nerve to climb a tall tree, but 
 Waterton retained his powers of tree-climbing until his 
 death, and very shortly before his fatal accident had 
 ascended one of the largest trees in the park, ho being 
 then in his eighty-third year. 
 
 Such a spot for study may seem a remarkable one, but 
 Waterton -was never affected by heights, and the man 
 who had scrambled up the cross of St. Peter's at Kome, 
 climbed the lightnii.^ conductor, and stood with one foot 
 on the head of the colossal angel of St. Angelo, was Hot 
 likely to be made giddy by the view from the top of an 
 oak-tree. 
 
 In part of his autobiography, Waterton mentions that 
 Ffctts of ]jg climbed to the top of the conductor, and left his glove 
 
 climbing. ^ ^ 
 
 on it, but he does not tell the sequel of the story. 
 
 All liome rang with the exploit, which reached the ears 
 of the Tope, Pius VII. Knowing that the glove would 
 spoil the conductor, he ordered it to be removed at once. 
 Not a man could be found in Home whose nerves were 
 equal to such a task, and so Waterton had to rei)eat the 
 ascent and fetch his glove down again, to the amusement 
 of his friends, and the delight of the populace. 
 
 No one could have given the advice in tree-climbing 
 which is quoted on page 50, without having experienced 
 the comparative strength of the different trees. Perhaps 
 
 t 
 
 'i 
 
 i 
 
 •i 
 
HIOdllAPIIY. 
 
 68 
 
 md tho 
 
 )Ci' had 
 
 nd had 
 sly one 
 ranches 
 aniusc- 
 Virgil. 
 (T wouUl 
 ree, but 
 util his 
 iiit had 
 being 
 
 one, but 
 ;he man 
 |t liome, 
 ine foot 
 was ilot 
 of an 
 
 )ns that 
 is glove 
 
 the ears 
 would 
 at once. 
 res were 
 )eat the 
 tisement 
 
 tlimbing 
 lerienced 
 Irerhaps 
 
 tho roador may not know that coming down a tree is a far 
 more dillicult tn.sk than ascending it. In tho latter case, 
 the climl)er can see his course, and note beforeiiand whore 
 he shall place his hands and feet, while in descending he 
 has to trust partly to memory, and partly to touch. 
 
 It is easy enough, for example, to spring for a few inches 
 from a lower to a higher brancli, hut to drop those few 
 inches is a very nervous business. I have more than once 
 seen a climber ascend a tree very boldly, and then be so 
 frightened that he couhl not be induced to come down 
 without some one to guide his feet. The same rule holds 
 good with precipices, where a man can always ascend 
 where he has descended without jumping, but not vice 
 i'crsii. 
 
 Even with trees, Waterton must needs have his joke. 
 All the important trees in the park had their names. 
 There were, for example, the Twelve Apostles standing in 
 a group, all starting from one root, the Eight Beatitudes, 
 the Seven Deadly Sins, &c. Then there were an oak and 
 a Scotch fir twined together, and going by the name of 
 Church and State (see p. G4). 
 
 Yew was one of Waterton's favourite trees, and he was 
 accustomed to say that it would be perfect if its leaves 
 were only furnished with spikes sharp enough to keep out 
 the cats, stoats, weasels, and his pet abhorrence, the brown 
 rat, wdiich he always called the Hanoverian rat, and 
 stoutly believed was imported into England by the same 
 ship that brought William of Orange to our shores. I 
 rather fancy that the Hanoverian origin of the brown rat 
 must have been one of Waterton's early jokes, and that he 
 gradually came to consider it as a fact. The yew fur- 
 nishes harborage for many birds, which after all do 
 not seem to suffer much from four-footed enemies. The 
 
 ClliiilniiiJ 
 
 Uilil ill'.' 
 
 siriidiiiij. 
 
 Names of 
 trees. 
 
 The Yen: 
 
64 
 
 HIOGUAPllY. 
 
 well-known yew-hedge in the garden of ^lerton College, 
 Oxford, is full of little birds, though their domiciles are 
 not easily seen through the dense foliage, 
 Uncn of the Waterton made great use of this valuable tree, and 
 formed with it ev^ergreen walls, impermeable to the north 
 wind, the one foe which he dreaded, and which seemed 
 quite to benumb him. I have seen him wi^h his lips so 
 
 yeir. 
 
 
 ■I 
 
 -1,4 
 
 CHURCH AND STATE, 
 
 paralysed by the north wind that he could scarcely frame 
 a word. He spent most of his waking time out of doors, 
 and his yew hedges were a great advantage to him in 
 sheltering liim from the north wind, and forming 
 pleasant nooks which received the cheering rays of th(^ 
 soutliern sun. 
 
 He wrote as follows in his Essay on the Yew-tree : " l( 
 
HIOGRAPHY. 
 
 65 
 
 College, 
 ;iles are 
 
 [•ee, and 
 
 le north 
 
 seemed 
 
 ; lips so 
 
 ^ly frame 
 
 |of doors, 
 
 him in 
 
 forming 
 
 rs of the 
 
 IllXll/l'S', 
 
 B-ee : 
 
 It 
 
 has already repaid me for the pains which I have taken in its 
 cultivation ; and when I resort to my usual evening stand, in 
 order to watch the flocks of sparrows, finches, and starlings, 
 Avhilst they are dropping in upon the neighbouring hollies, 
 I feel not the wintry blast, as the yew-trees, which are 
 close at hand, are to me a shield against its fury ; and in 
 fact, the/ offer me a protection little inferior to that of 
 the house itself." 
 
 There is a magnificent crescent-shaped yew-liedgf , which 
 partly surrounds tlie stables, and shuts them out from 
 sight so effectually, that no one could suspect theii pre- 
 sence unless informed of it. Another yew-hedge forms a 
 sort of wall behind the Ivy Tower, and aids in keeping it 
 quiet for the many birds which breed in it. 
 
 I have given tlie land view of the gateway (sometimes 
 called the " Ivy Tower " in the Essays) because it shows how 
 admirably Waterton adapted existing objects to his chief 
 pursuit at Walton Hall, namely, the cherishing of birds 
 and study of their habits. 
 
 The view is taken from the southern window of the guest- 
 chamber, and is one of the first objects that meets the 
 visitor's eyes on rising in the morning. 
 
 One portion of this illustration requires notice. Just 
 above the yew-hedge may be seen a curious - looking 
 circular tower ; witu a conical roof. This was built ex- 
 pressly for the use of starlings, and is appropriately named Sfnrihig 
 the Starling Tower. Many starlings found a home in the ^""''''*' 
 Ivy Tower, but wishing to accommodate these birds still 
 further, Waterton built this tower for them, and a very 
 interesting structuie it is, uniting several advantages. 
 
 In the first place, it is raised upon a smooth stone 
 pillar, on which rests a large circular, flattened stone, 
 considerably larger than the pillar. The object of tliis 
 arrangement is to keep out rats, the worst foes of the 
 
 F 
 
■::!:f : 
 
 i, i 
 
 06 
 
 15I0GKAP11Y. 
 
 Cats and Starling. Even the most active and sharpest-clawed rat 
 ^'"'*' could hardly climb up the pillar, and if it did, would 
 be stopped by the flat stone. In fact, this pillar and 
 stone are similar in design to the "staddles" on which 
 wheat-stacks ought to be built, if farmers Vv'ish to preserve 
 their grain. Cats are also loes to the starling, but the flat 
 stone is too high for most cats to reach by jumping, and if 
 
 UATKVVAY — BACK VIKW. 
 
 tliey tried to do so, tlie upper surface of the stone is 
 made with a slope, and is so smooth, that the claws could 
 not retain their hold. 
 
 The tower is circular, and is built in regular layers of 
 stones. Eacli alternate stone is loose, and when pulled 
 out, discloses a chamber behind, to which the bird obtains 
 access by means of a channel cut in the corner of the 
 stone. The birds took possession of the tower at once, 
 
 V 'hi 
 
BIOGRAPHY. 
 
 er 
 
 ved rat 
 would 
 lar and 
 , which 
 preserve 
 the flat 
 y, and if 
 
 as well tliey might, and it is very interesting to remove Ta.ncness 
 the stones and see the hirds sitting on their eggs without "^ ^'^^°"^' 
 being in the least alarmed at the intrusion. In con- 
 sequence of the protection which they enjoy, the starlings 
 are to be found in great numbers around the house, and 
 will assemble on the lawn in front of the sitting-room 
 windows, where they feed without fear, notwithstanding 
 
 stone IS 
 iws could 
 
 layers of 
 m pulled 
 I'd obtains 
 ler of the 
 at once, 
 
 STARLING TOV. i:u. 
 
 that they may be \> Hhin a few yards of the window from 
 which they are being watched. A second tower was after- 
 wards built and placed in another portion of the grounds. 
 
 The reader may remember that VVaterton drained some 
 fish-ponds and planted tliem with trees, which grew with 
 great rapidity. By means of the ever-useful yew, various 
 sheltering-places were made in it, and there was a little 
 f5ingle-roomed cottage where Watevton could sit by a fivn 
 
 F 2 
 
il',1 
 
 i :i 
 
 ll 
 
 
 68 
 
 BIOGRAPHY. 
 
 The ill cold weather, and yet be able to watch the birds, for 
 WmscVs ^yiiose benefit the door was always left open. Tliis cottage 
 is not shown in the illustration, but on the right hand 
 is seen a heap of rough stones. These were piled up for 
 the benefit of the Aveasel, which loves such localities, and 
 is tolerably sure to take possession of them. 
 
 IIIK OIllll'KI 
 
 The wliole of this corner of the park goes by the general 
 TiicGruttu. name of the "Crotto," on account of a cave which forms 
 part of it. In this beautiful place Watertou used to allow 
 parties to congregate, on the understanding that they 
 would do no injury, and, for their benefit, he had erected 
 swings, &c., among the trees. For tliis reason, the place 
 was often called the " Picnic," and this name was used 
 indifferently with that of Grotto. 
 
 m 
 
BIOGHAPIIY. 
 
 69 
 
 irds, for 
 i cottage 
 ht hand 
 i up for 
 ties, and 
 
 "The Squire," as he was invariably called, was, after liis rknks. 
 usual fashion, too trustful when he thought that he was 
 doing a kindness to others. During his later years, the 
 privilege became shamefully abused, and when, after his 
 death, a party of picnic-makers set lire to the magnificent 
 yew hedge enclosing" the stables, and destroyed a consider- 
 able portion of it, his son took the opportunity of pro- 
 hibiting picnics for the future. 
 
 By the way, Waterton was scarcely ever mentioned by 
 name, and just as the Duke of Wellington was known as 
 "the Duke," so was Waterton known far and wide as " the 
 Squire." Even his nearest relatives invariably addressed 
 him as "Squire," and it would be perfectly possible for a 
 visitor to be at Walton Hall for a week and never hear the 
 name of Waterton. 
 
 " The 
 
 Squire.' 
 
 \e general 
 
 tell forms 
 
 to allow 
 
 [hat they 
 
 erected 
 
 the place 
 
 Avas used 
 
 To EXHAUST all the objects of interest witliin the park 
 wall would require a large volume, and space is valuable. 
 There are one or two, however, which ought not to be 
 passed over without notice, and one of them is figured in 
 the illustration on page 70. 
 
 In former days there had been a water-mill, but time, MUhtoiie. 
 which, as Waterton quaintly says, is " the great anniliilator 
 of all human inventions saving taxation and the national 
 debt," destroyed the mill, and nothing of it is left except 
 a single millstone, measuring between five and six feet in 
 diameter. " The ground where the mill stood having been 
 converted into meadow, this stone lay there unnoticed and 
 unknown (save by the passing haymaker), from the period 
 of the mill's dissolution to the autumn of the year 1813, 
 when one of our nut-eating wild animals, probably by 
 way of winter store, deposited a few nuts under its 
 protecting cover. 
 
 " In the course of the following summer, a sinule nut, 
 
70 
 
 BIOGRAPHY. 
 
 A young having escaped the teeth of the destroyer, sent up its 
 nut plant, ygrdant shoot through the hole in the centre of the 
 procumbent millstone. 
 
 " One day I pointed out this rising tree to a gentleman 
 who was standing by, and I said ' If this young plant 
 escape destruction, some time or other it will support the 
 millstone, and raise it from the ground. He seemed to 
 doubt this. In order, however, that the plant might have 
 
 m <•„ .y i'i 
 
 ■I 
 
 NUr-TKEK AND MILL-dTONIi. 
 
 Ilaisinrj 
 millstone. 
 
 a fair chance of success, I directed that it should be 
 defended from accident and harm by means of a wooden 
 paling. 
 
 " Year after year it increased in size and beauty, and 
 when its expansion had entirely filled up the hole in the 
 centre of the millstone, it gradually began to raise up the 
 millstone itself from the seat of its long repose. This 
 
 I 
 
 'nltvi 
 
BIOGRAPHY. 
 
 71 
 
 up its 
 of the 
 
 itleraan 
 g plant 
 port the 
 iraed to 
 rht have 
 
 Death of 
 the tree. 
 
 huge mass of stone is now eight inches above the ground 
 and is entirely supported by thd stem of the nut-tree, 
 which has risen to the height of twenty-live feet, and 
 bears ejicellent fruit." 
 
 When I saw it in 1862, however, the tree had been dead 
 for some time, the millstone hitving evidently killed it, not 
 by its weight, but by preventing the flow of sap through 
 the bark. It would, of course, have been more picturesque 
 to have drawn the tree in a living state and of its full 
 height, but I thought it better to give it exactly as I 
 saw it. Mr. Edmund Waterton once told me that when 
 a boy he often climbed the tree in search of nuts, which 
 it then bore plentifully. 
 
 It is a most valuable object, inasmuch as it shows in a 
 striking manner- the tremendous powers of Nature, which are 
 continually being exerted, and which we, as a rule, do not 
 even suspect. The late Charles Kingsley mentions that he 
 has seen a large flat stone raised off the ground in a single '''^'^'^ "'"^ 
 
 ° ° •= grass. 
 
 night by a crop of tiny mushrooms, and I quite lately saw 
 some weighty kerbstones in a crowded London thoroughfare, 
 which had been forced completely out of their places by 
 grass-blades which had grown between their junctions. 
 
 Mush- 
 
 lould be 
 wooden 
 
 ity, and 
 le in the 
 up the 
 . This 
 
 -fS*t1 
 
CHAPTER V. 
 
 The S(iuiie's "do(lj;cs." — The "cut-holes." — The dove-coti'. — Pigeou- 
 shootiuf^ matuhea and mode of supplying the birds, — Watevton's 
 pigeon-house, external and internal. — Pigeon-stealcrs haflled. — Ar- 
 rangement of pigeon-holes. — Ladders not needed. — How to feed 
 pigeons economically. — Rats and mice in the garden. — The poison- 
 bowl and its safety.— Sunken mousetrap. — Gates and chains. — The 
 carriage-pond. — Waterton's antipathy to scientilic nomenclature. — 
 Advantage of such nomenclature as an assistant to science. — Popular 
 and local names — Colonists and their nomenclature. — Zoology gone 
 mad. — Complimentary nomenclature. — The fatal accident in the park. 
 — Waterton's last moments and death. — The last voyage and funeral. 
 — Epitaph written by himself. — The new cross, and place of burial. 
 
 m 
 
 Dudcji's. Now we must cast a glance at some of those ingenious 
 arrangements which I called " the Squire's dodges" on fii;st 
 seeing them. 
 
 There never was a place so full of ' dodges ' as Walton 
 Hall. The Starling Tower, described on page 67, was one 
 of them, being so arranged as to keep out rats and cats. 
 Now, Waterton wished to make a place which cats could 
 enter, but would keep out rats, and he achieved his object 
 by remembering that cats and rats could both climb, but 
 that rats Mere no great jumpers. 
 
 As may be remembered, the flat stone of the starling 
 tower was just out of reach of a cat's jump, which Water- 
 ton calculated at five feet for an ordinarily active cat. So 
 he had a large, smooth. Hat stone let into the wall, and an 
 aperture jnade in it, which he called the " cat's hole." Tt 
 
mOGKAPIIY. 
 
 73 
 
 was just five feet from the ground, so that pussy couM Cars 
 jump into it, while the stone was made so smooth that no 
 rat could climb it. 
 
 Having securely protected the starlings by their towers, 
 and the pheasants by the holly fortress and wooden dum- 
 mies — another dodge — he had to protect his pigeons. 
 
 He found that pigeons were stolen in great numbers, 
 and almost invariably for one purpose, namely, to supply 
 birds for pigeon-shooting matches, many of which took 
 place thirty or forty miles from the spot whence the birds 
 were stolen. Now, "Waterton had a righteous indignation 
 against pigeon-shooting, and had an ingenious mode of 
 thwarting the thieves. 
 
 Their plan was to come at night, when the pigeons were 
 all at home, and throw a net over the 'glover,' i.e. the 
 opening at the top, through which the pigeons enter and 
 leave the cote. Then if they can force an entranc; into 
 the cote they do so, but even if not, they frighten the birds 
 by knocking at the walls, and so drive them into the net. 
 
 Pigeon-houses, however, constructed like those at Walton 
 Hall, can set those nocturnal robbers at defiance. In the 
 first place, the house is so high that thieves could scarcely 
 find a ladder long enough to reach the roof, and then they 
 would need a second ladder to lay on the roof before they 
 could get at the glover. As to gaining admission by the 
 door, it is almost impossible. 
 
 The building is in two storeys, the lower being for the 
 reception of tools, chains, and the other multifarious re- 
 quirements of a farm. The rest of the building is intended 
 for the pigeons, and can only be approached by a door 
 some twenty feet from the ground. The door, which is The Door. 
 very strong, and bound with iron, fits flush into the wall, 
 so that there is no hold for a tool, and moreover, only one 
 man coukl v'c:\ at a time, he having nothing but a ladder 
 
 Piijcon' 
 house. 
 
74 
 
 BIOGRAPHV. 
 
 as a foothold. So much for the outside of the pigeon- 
 house. 
 Interior. If we wi^h to euter the building we must ascend to the 
 door by a ladder and unlock it. We then find ourselves 
 within a large and lofty chamber, well lighted and venti- 
 
 ill 
 
 II tt 
 
 Si 
 
 
 i: 
 
 Rows of 
 mats. 
 
 lated, white-washed, and perfectly clean and neat. The 
 whole place is scraped and wliite-washed at least twice in 
 each year, November and February being recommended 
 for these operations. 
 
 The interior walls are most curiously constructed. Parallel 
 rows of pigeon-holes occupy each wall, and boueath each 
 row is a ledge of brick. There are three rows of bricks 
 between the ledges, which are each one brick in width. 
 Twenty rows of nests occupy each wall. It is easy, there- 
 fore, for a man, without the aid of a ladder, to traverse the 
 whole of the building, and to examine every nest as he 
 goes along the ledges. On an average, to search three rows 
 of holes occupies an hour. Waterton mentions in his Essay 
 on the Do\ecot Pigeon that this single cote furnished in 
 
 I 
 
 t 
 :-S! 
 
 ,i:i 
 
 14L 
 
mOORAPIIY. 
 
 >igeon- 
 
 to the 
 
 rselves 
 
 venti- 
 
 t. The 
 vvice in 
 lended 
 
 •Parallel 
 
 th each 
 
 bricks 
 
 width. 
 
 , there- 
 
 krse the 
 
 as he 
 
 le rows 
 
 Essay 
 
 khed in 
 
 one season suventy-thiee dozens of young pigeons. There 
 is a stove, with flues in the building, but since it caught 
 fire by over-heating, and endangered the whole structure, 
 it has not been lighted. 
 
 There is another ' dodge ' in feeding the pigeons. It 
 was found that when their food was thrown down in ti»-.- 
 yard, the poultry, being stronger than the pigeons, drov,', 
 them away, mil ate the food themselves. So \Va.,«iton 
 had a floor ol ^ ' wattle-work laid under the roof of a 
 shed at some ^ iiL from the ground. The food for the 
 pigeons was always thrown upon this floor, where the 
 pigeons could easily reach it, while it was too high for the 
 poultry. 
 
 Here are two more ' dodges ' intended for the preserva- 
 tion of young peas, beans, &c., from the all-present rats 
 and mice. 
 
 Poisoned food is one effectual plan, but it must be so 
 laid that neither poultry nor game can get at it. This 
 object was attained by means of a poison-bowl, i.e. a 
 rounded bowl of earthenware, or even stone, through the 
 centre of which is a hole just large enough to admit a rat. 
 
 rOISON-BUWl. AND SECTION. 
 
 The use of the poison-bowl was simple enough. A spoon- 
 ful of poisoned meal was laid on the ground, and' the bowl 
 inverted over it. Rats could reach the meal by the hole, 
 and did so, but neither pheasants nor poultry could so 
 much as touch it with the tips of their beaks. 
 
 A section of the poison-bowl is given in the illustration 
 together with a view of its upper surface. 
 
 Feeding 
 Pigeons. 
 
 Poison- 
 bowl. 
 
IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 4 
 
 
 ^ 
 
 ^ 
 
 1.0 ^BtUi 
 
 ■tt JIM 122 
 
 u 
 
 
 6" 
 
 HxitogFaphic 
 
 uClBllC6S 
 
 Corporatian 
 
 ^ 
 
 ^•^ 
 
 
 \^ 
 
 
 •V. 
 
 as vnST MtAM STRHT 
 WltSm,N.r I41N 
 
 (n«)i7a*4MM 
 
 ^^^ 
 ^ 
 

76 
 
 BIOGRAPHY. 
 
 Mouse- 
 trap. 
 
 Next comes a trap of great efficacy, but which answers 
 better for mice than rats. It is simply an earthen or glass 
 jar, having its neck narrower than its shoulder, and buried 
 until its mouth is exactly level with the ground. The in- 
 side of the shoulder is then rubbed with bacon fat — the 
 ranker the better — and the trap is set, with scarcely any 
 expenditure of bait. 
 
 The mice, prowling about in search of fond, soon dis- 
 cover the jar, being attracted by the smell of the bacon. 
 
 SLNKISN JAK. 
 
 !(^ 
 
 They crane over the mouth, try to reach the savoury food, 
 and slip into the jar, from which there is no possibility of 
 escape. 
 
 Yet two more ' dodges,' and we must leave the wonders 
 of Walton Hall. 
 
 Waterton found that when cattle were placed in con- 
 tiguous fields connected by a gate, nothing could keep them 
 Cattle and away from that gate. They came and leant against it on 
 both sides, while they indulged in conversation after their 
 own manner, very much as their masters and mistresses 
 might do. 
 
 Now, the cattle which were in the field towards which 
 the gate opened did no harm to it, because as they leant 
 against it they only pressed against the posts. But the 
 weight of those on the opposite side came heavily upon 
 the catch and hinges, and sometimes even broke them 
 down. 
 
 Gates. 
 
BIOGRAPHY. 
 
 Waterton was much too kind-hearted to convert the 
 open gates into doors, and so to shut out the cattle from 
 social intercourse. So he suspended a stout chain on the 
 weak side, so that the cattle might lean against it and rest 
 upon the gate. As the chain was fastened by a staple at 
 
 77 
 
 Gate and 
 chain. 
 
 OATK AND CHAIN. 
 
 one end, and hitched over a strong hook at the other, it 
 could easily be dropped when the gate had to be opened 
 for traffic, and replaced after it was shut. 
 
 The last ' dodge ' which we can mention is that of the 
 carriage-pond. It was about four feet wider than an ordi- 
 nary carriage, and edged with broad, flat stones. At each 
 end it sloped gradually downwards until in the centre it was 
 just deep enough to allow the water to reach rather above 
 the axletrees. When a carriage or a cart had to be 
 washed after being used it was gently wheeled into the 
 pond at one end, while men with brooms washed the 
 wheels as they turned. In the centre of tJie pond the 
 men had command over every part of the carriage, which 
 was then drawn gently out, the mops and brooms being 
 at work until it came upon level ground. 
 
 Horses could thus be washed as well as carts and car- 
 riages, and very much they seemed to enjoy their bath. 
 
 Cftrriiiqr 
 Pond. 
 
 It is impossible to conclude the memoir of this accom- 
 plished naturalist without allusion to his rooted abhorrence 
 of scientific names. I do not think that this antipathy 
 
*"■••; . 
 
 78 
 
 BIOGRAPHY. 
 
 Scientific 
 names. 
 
 was justified. He was perfectly right in entering his 
 protest against the cumbrous pedantry which bade fair 
 to make science a mere question of memory for names, 
 and the language which he uses is not in the least too 
 strong. 
 
 Perhaps I may be pardoned for inserting a passage from 
 one of my own works, written twenty years ago, and long 
 before I knew Waterton, so that the reader may see how 
 completely I sympathise with him. 
 
 " Owing to the inordinate use of pseudo-classical phraseo- 
 logy, the fascinating study of animal life has been too long 
 considered as a profession or a science restricted to a 
 favoured few, and interdicted to the many until they have 
 undergone a long apprenticeship to its preliminary formulae. 
 So deeply rooted is this idea, that the popular notion of a 
 scientific man is of one who possesses a fund of words, 
 and not of one who has gathered a mass of ideas. There 
 is really not the least reason why any one of ordinary 
 capabilities and moderate memory should not be acquainted 
 with the general outlines of Zoology, and possess some 
 knowledge of the representative animals, which serve as 
 types of each group, tribe, or family ; for when relieved 
 of the cumbersome diction with which it is embarrassed, 
 the study of animal life can be brought within the com- 
 prehension of all who care to examine the myriad varieties 
 of form and colour with which the Almighty clothes His 
 living poems. 
 
 " The true object of Zoology is not, as some appear to 
 fancy, to arrange, to number, and to ticket animals in a 
 formal inventory, but to make the study an inquiry into 
 the life-nature, and not only an investigation of the lifeless 
 organism. I must not, however, be understood to disparage 
 the outward form, thing of clay though it be. For what 
 wondrous clay it is, and how marvellous the continuous 
 

 UIOGRAPHV. 
 
 miracle by which the dust of earth is transmuted into the 
 glowing colours and graceful forms which we most im- 
 perfectly endeavour to preserve after the soul has separated 
 from them." 
 
 But Waterton certainly erred in his persistent rejection Popular 
 of scientific names, which form an universal language, and 
 are needed for the purpose of identifying the creatures whose 
 habits ar^ being described. Even in England, there are 
 thousands of animals which have no popular names, nor are 
 likely to have them, and we are therefore driven to use the 
 names by which they are known to science all over the 
 world. 
 
 It is true that such names are often unintelligible, 
 especially to those who do not know Greek as well as 
 Latin, and the words Lophophorus, Tachipetes, Pachy- 
 cephala, Ptilonorhynchus, PalfEoinis, Meliphagidae, &c., 
 which are selected by Waterton as 'shocking examples, 
 certainly cannot be understood by persons who know 
 nothing of Greek or ornithology. 
 
 But the local words which he himself employs are far 
 less intelligible than the scientific terms. If a very 
 moderate Greek scholar were asked the meaning of these 
 words, he could at all events tell the inquirer that one 
 word referred to swiftness, another to the crest of the bird, 
 another to tlie size of its head, another to its fruit-eatinir 
 habits, and so forth, even though he should know nothing 
 of the birds themselves. But suppose the same inquirer to 
 have read the Wanderings, and to have asked the meaning 
 of Hannaquoi, Camoudi, Salempenta, Maroudi, Coulacanara, 
 Sakawinki, Wallababa, Houtou, Karabimiti, Sawari, &c., I 
 very much doubt whether one scholar in a thousand could 
 have given an answer. Consequent!}; , the otherwise delight- 
 ful Wanderings has been in many points a sealed book. 
 Sometimes the acceptance of local names is misleadinf^' 
 
 Local 
 names. 
 
80 
 
 BIOGUAPHV. 
 
 fi! 
 
 it 
 
 I 
 
 Misused as for example, when the jaguar is called 'tiger' (which 
 
 nainrs. Waterton explains, using either term indifferently), or 
 when he applies the familiar name of chameleon to one of 
 the Anolis lizards, the chameleons being exclusively inha- 
 bitants of the Old World. Similarly, in North America the 
 bison is called the buffalo, and the puma is termed the lion. 
 In Australia the nomenclature of the colonists looks like 
 Zoology gone mad. In that country, the mammalia are, 
 with one doubtful example, marsupials. Yet, we read of 
 the monkey, the bear, the wolf, the tiger, the cat, the rat, 
 the mouse, the badger, the jerboa, the hare, the sloth, the 
 mole, and the hedgehog, not one of these animals really 
 belonging to Australia. So, in these cases, unless we have 
 the scientific as well as the local names, whether native or 
 imported, we shall find ourselves entirely bewildered. 
 
 Compii- With Waterton's views concerning what he happily terms 
 nomew-hi- * Complimentary nomenclature,' I unreservedly agree, and 
 
 ture. have often written strongly on the subject, Waterton, 
 after noticing that in a single volume on birds, no less 
 than thirteen species have the names of men, makes the 
 following witty remarks. 
 
 " In the plates to the first volume of his work, I find 
 that a hawk is called the ' Black Warrior,' and that the 
 Latin name which he has given it is * Falco Harlani.' 
 Pray, who or what is ' Harlani ' ? A man, a mountain, 
 or a mud-flat? Is 'Black Warrior' a Negro of pug- 
 nacious propensities ? " 
 
 Lnsiscenes. Now we come to the last scenes of Waterton's life, for 
 which we are indebted to his friend Mr. N. Moore, who 
 was staying with him at the time. The full account may be 
 seen in the volume of Essays to which reference has been 
 made, and is well worth reading. Briefly, it is as follows. 
 He had gone, accompanied by Mr. Moore, to the further 
 
BIOGHAPIIY. 
 
 i 
 
 end of the park for the purpose of directing some carpen- 
 ters, wlien he caught his foot in an overhanging bramble, 
 and fill, with his side upon a log. lie knew at once the 
 extent of the injury, but contrived to reach the boat. On 
 arriving at the island, he walked to the house, changed his 
 clothes as usual, and, in spite of terrible pain, walked up 
 stairs without help. He would have gone on to his own 
 room at the top of the house, but consented to stop half 
 way, and lie on the sofa of Miss I'Mmonstone's (his sister- 
 in-law's) sitting room, for the sake cf saving trouble to 
 others. 
 
 Here ho died, and I must borrow Mr. Moore's own 
 words. 
 
 " The end was nOw at hand, and he died at twenty- 
 seven minutes past two in t)^e morning of May 27, 18G5. 
 The window was open. The sky was beginning to grow 
 grey, a few rooks had cawed, the swallows were twittering, 
 the landrail was craking from the Ox-close, and a tavourite 
 cock, which he used to call his morning-gun, leaped out 
 from some hollies, and gave his accustomed crow. The 
 ear of his master was deaf to the call. He had obeyed a 
 sublimer summons, and had woke up to the glories of the 
 eternal world." 
 
 So passed away Charles Waterton, a man who was, 
 perhaps, more thoroughly missed and more widely mourned 
 than any other of his time. 
 
 It is much to be regretted that he would never sit for 
 his portrait. As far as the head without the dress goes, 
 Mr. Waterhouse Hawkins's bust gives a good idea of " The 
 Squire," but marble could not give his sweet, kindly smile, 
 or the animated expressions which flitted over his face as 
 he recurred to his former travels, or pointed out the many 
 wonders of the park and lake. A good painter might have 
 succeeded, provided that he knew Waterton thoroughly, 
 
 G 
 
 81 
 
 The fatal 
 fall. 
 
 Death. 
 
 Bust of 
 Watertwi, 
 
H2 
 
 BIOGUAPIIY. 
 
 and, for such a memorial, thousands who never saw him 
 would have been grateful. 
 
 Tlie following account of the funeral is taken from the 
 Illustrated London News of June 17, 1865, 
 FuncraJ. " On the 3rd inst. (the anniversary of Mr. Waterton's 
 birth) his remains were laid amidst the scenes he loved so 
 well. He had selected the spot, and left minute instruc- 
 tions for the funeral ceremony, which commenced at nine 
 o'clock in the morning. 
 
 " The entrance-hall had been converted into a temporary 
 chapel, which was draped with black. Before the door, on 
 a catafalque, rested the coffin, of polished oak, with brass 
 inscription plates. Mr. Waterton was an eminent member 
 of the Roman Catholic Church, and fourteen priests took 
 part at his burial. First, ' the Office for the Dead ' was 
 said, and then requiem high mass, the Bishop of Beverley 
 singing, Canon Walker assisting as Deacon, and the Eev. 
 G. Waterton as Sub-Deacon. 
 
 " The invited guests having arrived soon after ten o'clock, 
 tlie coffin was borne in procession through the ancient 
 portcullis gate, accompanied by mourners and priests, 
 bearing tapers, and followed by Edmund Waterton, Esq., 
 son of the deceased, and chief mourner. At the landing- 
 stairs the body was placed on a floating bier. Thence, 
 preceded by a barge, containing the Bishop and officiating 
 priests, chanting the office, and followed by mourners and 
 friends in attendant boats — Mr. Waterton's own boat, un- 
 occupied, bringing up the rear, — the author of the Wan- 
 derings made his last voyage. The boats were all draped 
 with black, and boats containing the friends of the family 
 flanked the three central ones : the tenantry went by land. 
 " At the head of the lake, beneath the shade of two 
 noble oaks, is the vault, into which the coffin having been 
 lowered, the mourners and others took their place, the 
 
 
BIOGKAPIIY. 
 
 Bishop of Beverley at the head, the blessing of the grave 
 took place, and the ceremony terminated with the canticle 
 ' Benedictus ; ' during which a linnet in one of the oaks 
 overhead joined its song to the chanting of the clergy. 
 And here, * buried in silence, broken only by the cry of 
 the heron, and in a solitude almost as deep as that in 
 which he had lived so long in the swamps of the Orinoco 
 and the forests of the Amazon,' rest the remains of Charles 
 Waterton. 
 
 "Eighty-three being his age, that number of aged 
 persons had been invited to attend at the funeral, and had 
 a dole, each a loaf of bread and sixpence, distributed to 
 them at the park gate." 
 
 The very appropriate spot which received his remains 
 had been long fixed upon by him, but it was not until a 
 year before his death that he caused to be erected the 
 plain stone cross, the base of which bears the inscription 
 written by himself. 
 
 ORATE PRO ANIMA 
 
 CAROLI WATERTON. 
 
 CUJUS FESSA 
 
 JUXTA HANG CRUCEM 
 
 SEPELIUNTUR OSSA. 
 
 Nam 1782. Ohiit 1865. 
 
 83 
 
 • 
 
 Epitaph. 
 
 family 
 laud, 
 of two 
 g been 
 ,ce, the 
 
 The actual day of his birth was not known until June Birthday. 
 3, 1864, when he asked the Misses Edmonstone to come 
 and see a new cross which he had put up between two 
 magnificent oaks near the head of the lake. There are 
 many crosses in different parts of the park. The most con- 
 spicuous one is on the top of the Ivy Tower, and another 
 is on the wall of the Grotto, both of which are shown in 
 illustrations on pages 43, 66, 68. Another is upon a 
 
 Q 2 
 
84 
 
 BIOGRAPHY. 
 
 favourite resting-place nearly opposite the Gateway, but 
 there was a peculiar significance about this last cross. 
 
 " He rowed his sisters-in-law," writes Mr. Moore, " in 
 his boat to the far end of the lake which surrounds Walton 
 The New Hall, and when they arrived at the spot, he told them he 
 intended to be buried there, and put his arms round the 
 cross. ' Squire,' said Miss Edmonstone in Italian, for 
 there was a man at work within hearing, * it is your birth- 
 day.' He smiled and bowed assent." And on the 3rd of 
 June in the following year he was laid in the spot which 
 he had indicated. 
 
 The remains of his grandfather lie beneath an elm-tree 
 in another part of the park. 
 
 It is noteworthy that the spot where he fell is only a 
 Tite fatal stonc's-throw froui the cross, and just half-way between it 
 and the group of trees called the " Twelve Apostles." It 
 is now marked with a cross. 
 
 Originally, his son intended to build a memorial chapel 
 over the grave, but he afterwards, and rightly, relin- 
 quished the intention, thinking that the plain stone cross 
 erected by Waterton himself, engraved with his own 
 simple inscription, was a more fitting memorial than any 
 chapel, however beautiful it might be. 
 
 spot. 
 
 [Note, — The Bust of Waterton has now been reproduced 
 with the dress according to his invariable costume. Copies 
 and Photographs may be procured from F. W. Wilson 
 Westgate House, Kirk-Dale, Sydenham.] 
 
"ill 
 
 WANDERINGS IN SOUTH AMERICA, 
 
Ik 
 
n 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 PmST JOURNEY. 
 
 CHArTEll I. 
 
 " nee hevhn, nee latpus in aapiTis 
 
 Hiulix fefcUit mo loeis." 
 
 Object of the Wariderlvgs. — Demerara R. — Salm. — Toucan — Forest 
 trees. — Parasites. — Bush-ropo. — Hed monkey. — Wild animals. — Sloths 
 — Venomous snakes. — Lizards. — Bell -bird. — Houtou. — Insects.— Dog 
 poisoned with Wourali. — Falls. — Essequibo R.— liapid decay. — Falls 
 of the Essequibo. — Macousliia. — A white recluse. — The ■\Vatprniamma. 
 — A savage financier. — The Jabiru. — Ants' nests, — Fort St. .loacliim. 
 — Lake Parima. 
 
 In the month of April, 1812, 1 left the town of Stabroek, fmsst 
 to travel through the wikls of Demerara and Essequibo, a '' """^^' •^' 
 part of ci-devant Dutch Guiana, in South America. 
 
 The chief objects in view were to collect a quantity of Its object. 
 the strongest Wourali poison; and to reach the inland 
 frontier fort of Portuguese Guiana. 
 
 It would be a tedious journey for him who wishes to 
 travel through these wilds, to set out from Stabroek on 
 foot. The sun would exhaust him in his attempts to 
 wade through the swamps, and the Mosquitos at night 
 would deprive him of every hour of sleep. 
 
 !" 
 
i I 
 
 88 
 
 WANDEUTXGS IN SOUTH AMERICA. 
 
 country. 
 
 ^ii 
 
 III 
 
 i I 
 
 FinsT The road for horses runs parallel to the river, but it 
 jouiiNEY. extends a very little way, and even ends before the 
 cultivation of the plantation ceases. 
 
 The only mode then that remains is to proceed by 
 water ; and when you come to the high lands, you may 
 make your way through the forest on foot, or continue 
 your route on the river. 
 FaceoftJie After passing the third island in the river Demerara 
 there are few plantations to be seen, and those not joining 
 on to one another, but separated by large tracts of wood. 
 
 The Loo is the last where the Sugar-cane is growing. 
 The greater part of its negroes have just been ordered to 
 another estate ; and ere a few months shall have elapsed 
 all signs of cultivation will be lost in underwood. 
 
 Higher up stand tlie sugar-works of Amelia's Waard, 
 solitary and abandoned ! and after passing these there is 
 not a ruin to inform the traveller that either coffee or 
 sugar have ever been cultivated. 
 
 From Amelia's Waard an unbroken range of forest 
 covers each bank of the river, saving here and there 
 where a hut discovers itseir, inhabited by free people of 
 colour, with a rood or two of barren ground about it ; or 
 where the wood-cutter has erected himself a dwelling, and 
 cleared a few acres for pasturage. Sometimes you see 
 level ground on eacli side of you for two or three hours 
 at a stretch ; at other times a gently sloping hill presents 
 itself; and often, on turning a point, the eye is pleased 
 with the contrast of an almost perpendicular height jutting 
 into the water. The trees put you in mind of an eternal 
 spring, with summer and autumn kindly blended into it. 
 
 Here you may see a sloping extent of noble trees, whose 
 foliage displays a charming variety of every shade, from 
 the lightest to the darkest green and purple. The tops of 
 some are crowned with bloom of the loveliest hue; while 
 
WANDERINGS IN SOUTH AMERICA. 
 
 8U 
 
 the boughs of others bend with a profusion of seeds and Finsr 
 
 fruits. JOURNET. 
 
 Those whose heads have been bared by time, or blasted 
 by the thunder-storm, strike the eye, as a mournful sound 
 does the ear in music ; and seem to beckon to the senti- 
 mental traveller to stop a moment or two, and see that the 
 forests which surround him, like men and kingdoms, have 
 their periods of misfortune and decay. 
 
 The first rocks of any considerable size that are ob- Rocks. 
 served on the side of the river are at a place called Saba, 
 from the Indian word, which means a stone. They appear 
 sloping down to the water's edge, not shelvy, but smooth, 
 and their exuberances rounded off, and, in some places, 
 deeply furrowed, as though they had been worn with 
 continual floods of water. 
 
 There are patches of soil up and down, and the huge 
 stones amongst them produce a pleasing and novel effect. 
 You see a few Coffee-trees of a fine luxuriant growth ; and 
 nearly on the top of Saba stands the house of the post- 
 holder. 
 
 He is appointed by government to give in his report to Besidencc 
 the protector of the Indians of what is going on amongst ""^^'o/dcr! * 
 them, and to prevent suspicious people from passing up 
 the river. 
 
 When the Indians assemble here the stranger mav have 
 an opportunity of seeing the Aborigines dancing to the 
 sound of their country music, and painted in their native 
 style. They will shoot their arrows for him with an un- 
 erring aim, and send the poisoned dart from the blow-pipe 
 true to its destination ; and here he may often view all 
 the different shades, from the red savage to the white man, 
 and from the white man to the sootiest son of Africa. 
 
 Beyond this post there are no more habitations of white 
 men, or free people of colour. 
 
1)0 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FIRST 
 JOURNEY. 
 
 Trees, 
 
 The WiU 
 Fig-tree. 
 
 In a country so extensively covered with wood as this 
 is, having every advantage that a tropical sun and the 
 richest mould, in many places, can give to vegetation, it 
 is natural to look for trees of very large dimensions'; but 
 it is rare to meet with them above six yards in circum- 
 ference. If larger have ever existed, they have fallen a 
 sacrifice either to the axe or to fire. 
 
 If, however, they disappoint you in size, they make 
 ample amends in height. Heedless and bankrupt in all 
 curiosity must he be who can journey on without stopping 
 to take a view of the towering Mora. Its topmost branch, 
 when naked with age or dried by accident, is the favourite 
 resort of the Toucan. Many a time has this singular bird 
 felt the shot faintly strike him from the gun of the fowler 
 beneath, and owed his life to the distance betwixt them. 
 
 The trees which form these far-extending wilds are as 
 useful as they are ornamental. It would take a volume 
 of itself to describe them. 
 
 The Green-heart, famous for its hardness and durability; 
 the Hackea, for its toughness ; the Ducalabali, surpassing 
 mahogany; the Ebony and Letter- wood, vying with the 
 choicest woods of the old world ; the Locust-tree, yielding 
 copal ; and the Hayawa and Olou-trees, furnishing a sweet- 
 smelling resin, are all to be met with in the forest, betwixt 
 the plantations and the rock Saba. 
 
 Beyond this rock the country has been little explored ; 
 but it is very probable that ' Uese, and a vast collection of 
 other kinds, and possibly many new species, are scattered 
 up and down, in all directions, through the swamps and 
 hills, and savannas of ci-dcvant Dutch Guiana. 
 
 On viewing the stately trees around him the naturalist 
 will observe many of them bearing leaves, and blossoms, 
 and fruit, not their own. 
 
 The Wild I'ig-tree, as large as a common English apple- 
 
WANDERINGS IN SOUTH AMERICA. 
 
 1)1 
 
 tree, often rears itself from one of the thick branches at first 
 the top of the mora ; and when its fruit is ripe, to it the JQ ^"^''' ^- 
 hirds resort for nourishment. It was to an undigested 
 seed, passing through the body of the bird which had 
 perched on the mora, that the fig-tree first owed its ele- 
 vated station there. The sap of the mora raised it into full 
 bearing ; but now, in its turn, it is doomed to contribute a 
 portion of its own sap and juices towards the growth of 
 different species of vines, the seeds of which, also, the 
 birds deposited on its branches. These soon vegetate, and 
 bear fruit in gTeat quantities ; so what with their usurpa- 
 tion of the resources of the fig-tree, and the fig-tree of the 
 mora, tlie mora, unable to support a charge which nature 
 never intended it should, languishes and dies under its 
 burden ; and then the fig-tree, and its usurping progeny 
 of vines, receiving no more succour from their late foster- 
 parent, droop and perish in their turn. 
 
 A vine, called the Bush-rope by the wood-cutters, on The Busk- 
 account of its use in hauling out the heaviest timber, has ^''^"'' 
 a singular appearance in the forests of Demerara. Some- 
 times you see it nearly as thick as a man's body, twisted 
 like a corkscrew round the tallest trees, and rearing its 
 liead high above their tops. At other times three or four 
 of them, like strands in a cable, join tree and tree and 
 branch and branch together. Other,:, descending from on 
 high, take root as soon as tl Av extremity touches the 
 ground, and appear like shrouds and stays supporting 
 the mainmast of a line-of-battle ship; while otliers, 
 sending out parallel, oblique, iiorizontal, and perpendi- 
 cular sho'ots in all directions, put you in mind of what 
 travellers call a matted forest. Oftentimes a tree, above a 
 hundred feet high, uprooted by the whirlwind, is stopped 
 in its fall by these amazing cables of nature ; and hence 
 it is that you account for the phenomenon of seeing trees 
 
02 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 ni 
 
 Soils, 
 
 III 
 
 FIRST not only vegetating, but sending forth vigorous shoots, 
 JO URNE Y, though far from their perpendicular, and their trunks 
 inclined to every degree from the meridian to the 
 horizon. 
 
 Their heads remain firmly supported by the bush-rope ; 
 many of their roots soon refix themselves in the earth, and 
 frequently a strong shoot will sprout out perpendicularly 
 from near the root of the reclined trunk, and in time 
 become a fine tree. No grass grows under the trees ; and 
 few weeds, except in the swamps. 
 
 The high grounds are pretty clear of underwood, and 
 with a cutlass to sever the small bush-ropes, it is not diffi- 
 cult walking among the trees. 
 
 The soil, chiefly formed by the fallen leaves and decayed 
 trees, is very rich and fertile in the valleys. On the hills, 
 it is little better than sand. The rains seem to have car- 
 ried away, and swept into the valleys, every particle which 
 nature intended to have formed a mould. 
 
 Four-footed animals are scarce, considering how very 
 thinly these forests are inhabited by men. 
 
 Several species of the animal, commonly called Tiger, 
 though in reality it approaches nearer to the leopard, are 
 found here ; and two of their diminutives, named Tiger-cats. 
 The Tapir, the Labba, and Deer, afford excellent food, and 
 chiefly frequent the swamps and low ground, near the sides 
 of the river and creeks. 
 
 In stating that four-footed animals are scarce, the Peccari 
 must be excepted. Three or four hundred of them herd 
 together, and traverse the wilds in all directions, in quest 
 of roots and fallen seeds. The Indians mostly shoot them 
 with poisoned arrows. When wounded, they run about 
 one hundred and fifty paces ; they then drop, and make 
 wholesome food. 
 
 The Red Monkey, erroneously called the baboon, is heard 
 
 Four- 
 
 footed 
 
 animals. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 93 
 
 oftener than it is seen ; while the common Brown Monkey, first 
 the Bisa, and Sacawinki, rove from tree to tree, and amuse JO ""^e y. 
 the stranger as he journeys on. 
 
 A species of the Polecat, and another of the Fox, are 
 destructive to the Indian's poultry; while the Opossum, 
 the Guana, and Salempenta, afford him a delicious morsel. 
 The small Ant-bear, and the large one, remarkable for his 
 long, broad, bushy tail, are sometimes seen on the tops of 
 the wood-ants' nests ; the Armadillas bore in the sand-hills, 
 like rabbits in a warren ; and the Porcupine is now and 
 then discovered in the trees over your head. 
 
 This, too, is the native country of the Sloth. His looks, The Sloth. 
 his gestures, and his cries, all conspire to entreat you to 
 take pity on him. These are the only weapons of defence 
 which nature hath given him. While other animals as- 
 semble in herds, or in pairs range through these boundless 
 wilds, the sloth is solitary, and almost stationary ; he can- 
 not escape from you. It is said, his piteous moans make 
 the tiger relent, and turn out of the way. Do not then 
 level your gun at him, or pierce him with a poisoned arrow ; 
 — he has never hurt one living creature. A few leaves, 
 and those of the commonest and coarsest kind, are all he 
 asks for his support. On comparing him with other ani- 
 mals, you would say that you could perceive deficiency 
 deformity, and superabundance in his composition. He 
 has no cutting teeth, and though four stomachs, he still 
 wants the long intestines of ruminating animals. He has 
 only one inferior aperture, as in birds. He has no soles to 
 his feet, nor has he the power of moving his toes separately. 
 His hair is flat, and puts you in mind of grass withered by 
 the wintry blast. His legs are too sliort ; they appear de- 
 formed by the manner in which they are joined to the body ; 
 and when he is on the ground, they seem as if only calcu- 
 
 lated to be of use in climbing trees. 
 
 He has forty-six ribs. 
 
I! 
 
 1 1 
 
 !> 
 
 1 .11 
 
 II':, 
 
 
 |i!' 
 
 ' '■! 
 
 I I 
 
 94 
 
 FIRST 
 JOUKNEY. 
 
 WANDEKINGS IN SOUTH AMEIilCA. 
 
 Birds. 
 
 The 
 Vulture. 
 
 while the elephant has only forty ; and his claws are 
 disproportionubly long. Were you to mark down, upon a 
 graduated scale, the different claims to superiority amongst 
 the four-footed animals, this poor ill- formed creature's claim 
 would be the last upon the lowest degree. 
 
 Demerara yields to no country in the world in her won- 
 derful and beautiful productions of the feathered race. 
 Here the finest precious stones are far surpassed by the 
 vivid tints which adorn the birds. The naturalist may 
 exclaim, that nature has not known where to stop in form- 
 ing new species, and painting her requisite shades. Almost 
 every one of those singular and elegant birds described by 
 Buffon as belonging to Cayenne, are to be met with in 
 Demerara; but it is only by an indefatigfible naturalist 
 that they are to be found. 
 
 The Scarlet Curlew breeds in innumerable quantities in 
 the muddy islands on the coasts of Pomauron ; the Egrets 
 and Crabiers in the same place. They resort to the mud- 
 flats at ebbing water, while thousands of Sandpipers and 
 Plovers, with here and there a Spoonbill and Flamingo, are 
 Seen amongst them. The Pelicans go farther out to sea, 
 but return at sundown to the Courada-trees. The Humming- 
 birds are chiefly to be found near the flowers at which each 
 of the species of the genus is wont to feed. The pie, the 
 gallinaceous, the columbine, and passerine tribes, resort to 
 the fruit-bearing trees. 
 
 You never fail to see the common Vulture where there is 
 carrion. In passing up the river there was an opportunity 
 of seeing a pair of the King of the Vultures ; they were 
 sitting on the naked branch of a tree, with about a dozen 
 of the common ones with them. A Tiger had killed a 
 Goat the day before ; he had been driven away in the act 
 of sucking the blood, and not finding it safe or prudent to 
 return, tlie goat remained in the same place where he had 
 
WANDERINGS IN SOUTH AMERICA. 
 
 95 
 
 killed it ; it had begun to putrefy, and the vultures had fiust 
 arrived that morning to claim the savoury morsel. jouuney. 
 
 At the close of day, the Vampires leave the hollow trees, The 
 whither they had fled at the morning's dawn, and scour ^^^"P^^^- 
 along the river's banks in quest of prey. On waking 
 from sleep, the astonished traveller finds his hammock 
 all stained with blood. It is the vampire that hath 
 sucked him. Not man alone, but every unprotected 
 animal, is exposed to his depredations : and so gently does 
 this nocturnal surgeon draw the blood, that instead of being 
 roused, the patient is lulled into a still profounder sleep. 
 There are two species of vampire in Demerara, and both 
 suck living animals ; one is rather larger than the common 
 bat ; the other measures above two feet from wing to wing 
 extended. 
 
 Snakes are frequently met with in the woods betwixt the Snakcu. 
 sea-coast and the rock Saba, chiefly near the creeks and on 
 the banks of the river. They are large, beautiful, and for- 
 midable. The Rattlesnake seems partial to a tract of ground 
 known by the name of Canal Number-three ; there the . 
 effects of his poison will be long remembered. 
 
 The Camoudi snake lias been killed from thirty to forty 
 feet long ; though not venomous, his size renders him de- 
 structive to the passing animals. The Spaniards in tlie 
 Oroonoque positively affirm that he grows to the length of 
 seventy or eighty feet, and that he will destroy the strongest 
 and largest bull. His name seems to conflrm this ; there 
 he is called " matatoro," which literally means " bull- 
 killer." Thus he may be ranked amongst the deadly 
 snakes ; for it conies nearly to the same thing in the end 
 whether the victim dies by poison from the fangs, which 
 corrupts his blood and makes it stink horribly, or whether 
 his body be crushed to mummy, and swallowed by this 
 hideous beast. 
 
 I 
 
96 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FIRST 
 JOUHNEY. 
 
 The Whipsnake, of a beautiful changing green, and the 
 Coral, with alternate broad transverse bars of black and 
 red, glide from bush to bush, and may be handled with 
 safety ; tliey are harmless little creatures. 
 
 The Labarri snake is speckled, of a dirty brown colour, 
 and can scarcely be distinguished from the ground or stump 
 on whicli he is coiled up ; he grows to tlie length of about 
 eight feet, and his bite often proves fatal in a few minutes. 
 Unrivalled in his display of every lovely colour of the 
 rainbow, and unmatched in the effects of his deadly poison, 
 tlie Couanacouchi glides undaunted on, sole monarch of these 
 forests ; he is commonly known by the name of the bush- 
 master. Both man and beast lly before him, and allow 
 him to pursue an undisputed path. He sometimes grows 
 to the length of fourteen feet. 
 
 A few small Caimen, from two to twelve feet long, may 
 be observed now and then in passing up and down the 
 river; they just keep their heads above the watei", and a 
 stranger would not know them from a rotten stump. 
 Lizards. Lizards of the finest green, brown, and copper colour, 
 from two inches to two feet and a half long, are ever and 
 anon rustling among the fallen leaves, and crossing the 
 path before you ; whilst the Chameleon is busily employed 
 in chasing insects round the trunks of the neighbouring 
 trees. 
 Fish. The fish are of many different sorts, and well-tasted, but 
 not, generally speaking, very plentiful. It is probable that 
 their numbers are considerably thinned by the Otters 
 which are much larger than those of Europe. In going 
 through the overflowed savannas, which have all a com- 
 munication with the river, you may often see a dozen or 
 two of them sporting amongst the sedges before you. 
 Lisectf. This warm and humid climate seems particularly adapted 
 to the producing of insects ; it gives birth to myriads 
 
WANDEUINGS IN SOUTH AMEIIICA. 
 
 97 
 
 I, but 
 that 
 
 Iters 
 
 111 
 
 beautiful past description in their variety of tints, astonish- first 
 ing in their form and size, and many of them noxious in Jouuney. 
 their qualities. 
 
 He whose eye can distinguish the various beauties of 
 uncultivated nature, and wliose ear is not shut to the wild 
 sounds in the woods, will be delighted in passing up the 
 river Demerara. Every now and tlien, the Maam or 
 Tinamou sends forth one long and plaintive whistle from 
 the depths of the forest, and then stops ; whilst the yelping 
 of the toucan, and the slirill voice of the bird called Pi-pi- 
 yo, is heard during the interval. The Campanero never 
 fails to attract the attention of the passenger : at a distance 
 of nearly three miles, you may hear this snow-white bird 
 tolling every four or five minutes, like the distant convent 
 bell.- From six to nine in the morning, the forests resound 
 with the mingled cries and strains of the feathered race ; 
 after this, they gradually die away. From* eleven to three 
 all nature is hushed as in a midnight silence, and scarce a 
 note is heard, saving that of the campanero and the pi-pi- 
 yo ; it is then that, oppressed by the solar heat, the birds 
 retire to the tliickest shade, and wait for the refreshing cool 
 of evening. 
 
 At sundown the Vampires, Bats, and Goat-suckers dart 
 from their lonely retreat, and skim along the trees on the 
 river's bank. The different kinds of Frogs almost stun the 
 ear with their coarse and hollow- sounding croaking, while 
 the Owls and goat-suckers lament and mourn all night 
 long. 
 
 About two hours before daybreak, you will hear the red 
 monkey moaning as though in deep distress ; the Houtou, a 
 solitary bird, and only found in the thickest recesses of the 
 forest, distinctly articulates, " houtou, houtou," in a low and 
 plaintive tone, an hour before sunrise ; the maam whistles 
 about the same hour ; the Hannaquoi, Tataca, and Maroudi 
 
 II 
 
98 
 
 WAXDKUIXGS IX SOUTH AMERICA. 
 
 l: 
 
 f I 
 
 ! ' < 
 
 FIRST announce his near approach to the eastern horizon, and the 
 
 U ■ Parrots and Parro(|uets conKrni his arrival there. 
 
 The Crickets cliirp from sunset to sunrise, and often 
 durinf,' the day, when the woatlier is cloudy. The Bete- 
 rouge is exceedingly numerous in these extensive wilds, 
 and not oidy man, but beasts and birds, are tormented by 
 it. Mosquitos are very rare after you pass the third island 
 in the Demerara, and Sand-flies but seldom appear. 
 
 Courteous reader, here thou hast the outlines of an 
 amazing landscape given thee ; thou wilt see that the 
 principal parts of it are but faintly traced, some of them 
 scarcely visible at all, and that the shades are wholly 
 wanting. If thy soul partakes of the ardent flame which 
 the persevering Mungo Park's did, these outlines will be 
 enough for thee : they will give thee some idea of what 
 a noble country this is ; and if thou hast but courage to 
 set about giving the world a finished picture of it, neither 
 materials to \^ork on, nor colours to paint it in its true 
 shades, will be wanting to thee. It may appear a difficult 
 task at a distance ; but look close at it, and it is nothing 
 at all ; provided thou hast but a quiet mind, little more is 
 necessary, and the genius which presides over these wilds 
 will kindly help thee through the rest. She will allow 
 thee to slay the fawn, and cut down the Mountain-cabbage 
 for thy support, and to select from every part of her domain 
 whatever may be necessary for the work thou art about ; 
 but having killed a pair of Doves in order to enable thee 
 to give mankind a true and proper description of them, 
 thou must not destroy a third through wantonness, or to 
 show what a good marksman thou art ; that would only 
 blot the picture thou art finishing, not colour it. 
 
 Though retired from the haunts of men, and even with- 
 out a friend with thee, thou wouldst not find it solitary. 
 The crowing of the hannaquoi will sound in thine ears 
 
Dbage 
 
 )main 
 
 libout ; 
 
 thee 
 
 Ithem, 
 
 or to 
 
 only 
 
 with- 
 litary. 
 ears 
 
 WANDKRIXGS IN SOUTH AMERICA. 
 
 09 
 
 FII18T 
 JOITRN'KY. 
 
 The Fire- 
 
 J'y- 
 
 like the daybreak town-clock ; and the Wren and the 
 Thrush will join with thee in thy matin liynin to thy 
 Creator, to thank Him for thy night's res^t. 
 
 At noon thy Genius will lead thee to the Troely, one 
 leaf of which will defend thee from both sun and rain. 
 And if, in the cool of the evening, thou hast been tempted 
 to stray too far from thy place of abode, and art deprived 
 of light to write down the information thou hast collected, 
 the Firefly, which thou wilt see in almost every bush 
 around thee, will l)e thy candle. Hold it over thy pocket- 
 book, in any position which thou knowest will not hurt it, 
 and it will afford thee ample light. And when tliou hast 
 done with it, put it kindly back again on the next branch 
 to thee. It will want no other reward for its services. 
 
 When in thy hammock, should the thought of thy little 
 crosses and disappointments, in thy ups and downs through 
 life, break in upon thee, and throw thee into a pensive 
 mood, the Ow^l will bear thee company. She will tell thee The Owl. 
 that hard has been her fate too ; and at intervals, " Whip- 
 poor- Will," and " Willy come go," will take up the tale of 
 sorrow. Ovid has told thee how the owl once boasted the 
 human form, and lost it for a very small offence ; and were 
 the poet alive now, he would inform thee, that " Whip- 
 poor- Will," and " Willy come go," are the shades of those 
 poor African and Indian slaves, who died worn out and 
 brokenhearted. They wail and cry, " Whip-poor- Will," 
 and " Willy come go," all night long ; and often, when the 
 moon shines, you see them sitting on the green turf, near 
 the houses of those whose ancestors tore them from the 
 bosom of their helpless families, which all probably per- 
 ished through gi'ief and want, after their support was 
 gone. 
 
 About an hour above the rock of Saba, stands the habi- 
 tation of an Indian, called Simon, on the top of a hill. h,u. * 
 
 h2 
 
100 
 
 WANDERINGS IN SOUTH AMEUICA. 
 
 FIRST 
 JOUUNKY. 
 
 Tlie side next tlio rivor is almost perpendicular, and you 
 may easily throw a stone over to the opposite bank. ITero 
 there was an opjwrtunity of seeing man in his rudest 
 state. The Indians wlio frequented this habitation, though 
 livinj: in the midst of woods, bore evident marks of ntten- 
 tion to their persons. Their hair was neatly collected, and 
 tied up in a knot ; their bodies fancifully painted red, and 
 the paint was scented with hayavva. This gave them a 
 gay and animated appearance. Some of them had on 
 necklaces, composed of the teeth of wild boars slain in 
 the chase ; many wore rings, and others had an ornament 
 on the left arm, midway betwixt the shoulder and the 
 elbow. At the close of day, they regularly bathed in 
 the river below ; and the next morning seemed busy in 
 renewing the faded colours of their faces. 
 
 One day there came into the hut a form which literally 
 nii'iht be called the wild man of the woods. On enterin" 
 he laid down a ball of wax, which he had collected in the 
 forest. His hammock was all ragged and torn ; and his 
 bow, though of good wood, was without any ornament or 
 polish ; " erubuit domino, cultior esse suo." His face was 
 meagre, his looks forbidding, and his whole appearance 
 neglected. His long black hair hung from his head in 
 matted confusion ; nor had his body, to all appearance, 
 ever been painted. They gave him some Cassava bread 
 and boiled fish, which he ate voraciously, and soon after 
 left the hut. As he went out you could observe no traces 
 in his countenance or demeanour, which indicated that he 
 was in the least mindful of having been benefited by the 
 society he was just leaving. 
 
 The Indians said that he had neither wife, nor child, 
 nor friend. They had often tried to persuade him to come 
 and live amongst them ; but it was of no avail. He went 
 roving on, plundering the Wild Bees of their honey, and 
 
WAXDEUIXGS IX SOUTH AMKIIICA, 
 
 101 
 
 pickinj:^ up tho fallen nuts and I'luits of llu) ftuest. AVlien fmsT 
 ho fell in with f^anic, ho procuiuel lire from two Mcks, and Joi^^^v 
 cooked it on the spot. When a liut happin'-d to be in his 
 May, he stepped in, and asked for sonuithin*,' to eat, and 
 then months elapsed ere they saw him ii^ain. They did 
 not know what had caused him to he thus unsettled ; he 
 had been so for years ; nor did they Ixilieve that even old 
 ajje itself would chanjje tho habits of this poor, harndess, 
 solitary wanderer. 
 
 From Simon's, the traveller may reach the large fall, 
 with ease, in four days. 
 
 The first falls that he meets are nu'rcly rapids, scarce a 
 stone appearing above the water in the rainy season ; and 
 those in the bed of the river, barely high enough to arrest 
 the water's course, and by causing a bubbling, show that 
 they are there. 
 
 With this small change of a})pearance in the stream, 
 the stranger observes nothing new till he comes within 
 eight or ten miles of the great fall. Each side of the 
 river presents an uninterrupted range of wood, just as 
 it did below. All the productions found betwixt 
 the plantations and the rock Saba, are to be met with 
 here. 
 
 From Simon's to the great fall there are five haliitations 
 of the Indians — two of them close to the river's side ; the 
 other three a little way in the forest. These habitations 
 consist of from four to eight huts, situated on about an 
 acre of ground which they have cleared from the sur- 
 rounding woods. A few I'appaw, Cotton, and mountain 
 cabbage-trees, are scattered round them. 
 
 At one of these habitations a small quantity of the 
 Wourali poison was procured. It was in a little gourd. IVourali 
 The Indian vvho had it said that he had killed a number ^'""''"' 
 of Wild Hogs with it, and two Tapirs. Appearances seemed 
 
 Indian 
 
 hahila- 
 
 tions. 
 
I I. 
 
 I; 
 
 !!J 
 
 
 
 
 i 
 
 1 ,, 
 
 1 
 
 i 
 
 
 102 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FIRST 
 JOURNKY. 
 
 Its 
 strength. 
 
 Th^ great 
 fall. 
 
 to confirm what he said; for on one side it had been nearly 
 taken out to the bottom, at different times, which probably 
 woiild not have been the case had the first or second trial 
 failed. 
 
 Its strength was proved on a middle-sized dog. He was 
 wounded in the thigh, in order that there might be no 
 possibility of touching .a vital part. In three or four 
 minutes he began to be affected, smelt at every little 
 thing on the ground around him, and looked wistfully at 
 the wounded part. Soon after this he staggered, laid him- 
 self down, and never rose more. He barked once, though 
 not as if in pain. His voice was low and weak ; and in 
 a second attempt it quite failed him. He now put his 
 head betwixt his fore-legs, and raising it slowly again, he 
 fell over on his side. His eye immediately became fixed, 
 and though his extremities every now and then shot con- 
 vulsively, he never showed the least desire to raise up his 
 head. His heart fluttered much from the time he lay 
 down, and at intervals beat very strong ; then stopped for 
 a moment or two, and then beat again ; and continued 
 faintly beating several minutes after every other part of 
 his body seemed dead. 
 
 In a quarter of an hour after he had received the poison 
 he was quite motionless. 
 
 A few miles before you reach the great fall, and which, 
 indeed, is the only one which can be called a fall, large 
 balls of froth come floating past you. The river appears 
 beautifully marked with streaks of foam, and on your 
 nearer approach the stream is whitened all over. 
 
 At first, you behold the fall rushing down a bed of 
 rocks, with a tremendous noise, divided into two foamy 
 streams, which at their junction again form a small island 
 covered with wood. Above this island, for a short space 
 there appears but one stream, all white with froth, and 
 
foamy 
 island 
 [space 
 and 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 103 
 
 fretting and boiling amongst the huge rocks which obstruct 
 
 FIllST 
 JOURNEY. 
 
 its course. 
 
 Higher up it is seen dividing itself into a short channel 
 or two, and trees grow on the rocks which caused its sepa- 
 ration. The torrent in many places has eaten deep into 
 the rocks, and split them into large fragments by driving 
 others against them. The trees on the rocks are in bloom 
 and vigour, though their roots are half bared, and many of 
 them bruised and broken by the rushing waters. 
 
 This is the general appearance of the fall from the level 
 of the water below to where the river is smooth and quiet 
 above. It must be remembered that this is during the 
 periodical rains. Probably in the dry season it puts on a 
 very different appearance. There is no perpendicular fall 
 of water of any consequence throughout it, but the dread- 
 ful roaring and rushing of the torrent down a long, rocky, 
 and moderately sloping channel, has a tine effect ; and the 
 stranger returns well pleased with what he has seen. No 
 animal, nor craft of any kind, could stem this downward 
 flood. In a few moments the first would be killed, the 
 second dashed in pieces. 
 
 The Indians have a path alongside of it, through the 
 forest, wher? prodigious Crabwood-trees grow. Up this 
 path they d/ng their canoes, and launch them into the 
 river above; and on their return bring them down the 
 same way. 
 
 About two hours below this fall is the habitation of an HaUtation 
 Acoway chief called Sinkerman. At night you hear the u-ay'\hkf. 
 roaring of the fall from it. It is pleasantly situated on 
 the top of a sand-hill. At this place you have the finest 
 view the river Demerara affords : three tiers of hills rise 
 in slow gradation, one above the other, before you, and 
 present a grand and magnificent scene, especially to him 
 who has been accustomed to a level country. 
 
t Jl 
 
 104 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FinST 
 JOUllNEY. 
 
 Here, a little after midnight on the first of May, was 
 heard a most strange and unaccountable noise ; it seemed 
 as though several regiments were engaged, and musketry 
 firing witli great rapidity. Tlie Indians, terrified beyond 
 description, left their hammocks, and crowded all together, 
 like sheep at the approach of tlie wolf. There were no 
 soldiers within three or four hundred miles. Conjecture 
 was of no avail, and all conversation next morning on the 
 subject was as useless and unsatisfactory as the dead silence 
 wliich succeeded to the noise. 
 
 He who wishes to reach the Macoushi country had 
 better send his canoe over land from Sinkerman's to the 
 Essequibo. 
 
 There is a pretty good path, and meeting a creek about 
 three-quarters of the way, it eases the labour, and twelve 
 Indians will arrive with it in the Essequibo in four days. 
 
 The traveller need not attend his canoe ; there is a 
 shorter and a better way. Half an hour below Sinker- 
 man's he finds a little creek on the western bank of the 
 Demerara. After proceeding about a couple of hundred 
 yards up it, he leaves it, and pursues a west-north-west 
 
 direction by land for the Essequibo. The path is good, 
 though somewhat rugged with the roots of trees, and here 
 and there obstructed by fallen ones ; it extends more over 
 level ground than iiherwise. There are a few steep ascents 
 and descents in ^, with a little brook running at the 
 bottom of them ; but they are easily passed over, and tlie 
 fallen trees serve for a bridge. 
 
 You may reach the Essequibo with ease in a day and a 
 half; and so matted and interwoven are the tops of the 
 trees above you, that the sun is not felt once all the way, 
 saving where the space which a newly-fallen tree occupied 
 lets in his rays upon you. The forest contains an abun- 
 dance of Wild Hogs, Labbas, Acouries, Towisses, Maams, 
 
WANDERINGS IN SOUTH AMERICA. 
 
 105 
 
 land a 
 )f the 
 way, 
 kipied 
 liibun- 
 faams, 
 
 Maroudis, and Waracabas, for your nourishment, and there first 
 are plenty of leaves to cover a shed whenever you are journey- 
 inclined to sleep. 
 
 The soil has three-fourths of sand in it, till you come Tlie Esse- 
 within half -an hour's walk of the Essequibo, where you S'*"*". 
 find a red gravel and rocks. In this retired and solitary 
 tract, nature's garb, to all appearance, has not been, injured 
 by fire, nor her productions broken in upon by the exter- 
 minating hand of man. 
 
 Here the finest Green-heart grows, and Wallaba, Purple- 
 heart, Siloabali, Sawuri, Buletre, Tauronira, and Mora, are 
 met with iu vast abundance, far and near, towering up in 
 mrjestic grandeur, straight as pillars, sixty or seventy feet 
 high, without a knot or branch. 
 
 Traveller, forget for a little while the idea thou hast of 
 wandering farther on, and stop and look at this grand 
 picture of vegetable nature ; it is a reflection of the crowd 
 thou hast lately been in, and though a silent monitor, it is 
 not a less eloquent one on that account. See that noble 
 Purple-heart before thee ! Nature has been kind to it. 
 Not a hole, not the least oozing from its trunk, to show 
 that its best days are past. Vigorous in youthful bloom- 
 ing beauty, it stands the ornament of these sequestered 
 wilds, and tacitly rebukes those base ones of thine own 
 species who have been hardy enough to deny the existence 
 of Him wlio ordered it to flourish here. 
 
 Behold that one next to it ! — Hark ! how the hammer- 
 ings of the Ptcd-headed Woodpecker resound through its 
 distempered boughs ! See what a quantity of lioles he 
 has made iu it, and how its bark is stained with the drops 
 which trickle down from them. The lightning, too, has 
 blasted one side of it. Nature looks pale and wan in its 
 leaves, and her resources are nearly dried up in its ex- 
 tremities ; its sap is tainted ; a mortal sickness, slow as 
 
106 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FIRST a consumption, and as sure in its consequences, has long 
 JO URNE Y. gjjjQg entered its frame, vitiating and destroying i'le 
 wholesome juices there. 
 
 Step a few paces aside, and cast thine eye on that rem- 
 nant of a Mora behind it. Best part of its branches, once 
 so high and ornamental, now lie on the ground in sad 
 confusion one upon the other, all shattered and fungus- 
 grown, and a prey to millions of insects, which are busily 
 employed in destroying them. One branch of it still looks 
 healthy ! Will it recover ? No, it cannot ; nature has 
 already run her course, and that healthy-looking branch is 
 only as a fallacious good symptom in him who is just 
 about to die of a mortification when he feels no more pain, 
 and fancies his distemper has left him ; it is as the mo- 
 mentary gleam of a wintry sun's ray close to the western 
 horizon. — See ! while we are speaking, a gust of wind has 
 brought the tree to the ground, and made room for its 
 successor. 
 
 Come further on, and examine that apparently luxu- 
 riant Tauronira on thy right hand. It boasts a verdure not 
 its own ; they are false ornaments it wears ; the Bush-rope 
 and Bird-vines have clothed it from the root to its topmost 
 branch. The succession of fruit which it hath borne, like 
 good cheer in the houses of the great, has invited the birds 
 to resort to it, and they have disseminated beautiful, though 
 destructive, plants on its branches, which, like the dis- 
 tempers vice brings into the human frame, rob it of all its 
 health and vigour; they have shortened its days, and 
 probably in another year they, will finally kill it, long 
 bafore nature intended that it should die. 
 
 Ere thou leavest this interesting scene, look on the 
 ground around thee, and see what everything here below 
 must come to. 
 
 Behold tliat newly fallen Wallaba ! The whirlwind has 
 
 B 
 
 liil 
 
WANDERINGS IN SOUTH AMERICA. 
 
 107 
 
 long 
 
 the 
 below 
 
 d has 
 
 uprooted it in its prime, and it has brought down to the Finsr 
 ground a dozen small ones in its fall. Its bark has already Jo mtME Y. 
 begun to drop off! And that heart of Mora close by- 
 it is fast yielding, in spite of its firm, tough texture. 
 
 The tree which thou passedst but a little ago, and which 
 perhaps has lain over yonder brook for years, can now hardly 
 support itself", and in a few months more it will have fallen 
 into the water. 
 
 Put thy foot on that large trunk thou seest to the left. 
 It seems entire amid the surrounding fragments. Mere 
 outward appearance, delusive phantom of what it once 
 was ! Tread on it, and like the fuss-ball, it will break into 
 dust. 
 
 Sad and silent mementos to the giddy traveller as he 
 wanders on ! Prostrate remnants of vegetable nature, how 
 incontestably ye provewhat we must all at last come to, 
 and how plain your mouldering ruins show that the 
 firmest texture avails us nought when fleaven wills that 
 we should cease to be ! — 
 
 " The cloud-capt towers, the gorgeous palaces, 
 The solemn temples, the great globe itself, 
 Yea, all which it inhabit, shall dissolve. 
 And, like the baseless fabric of a vision. 
 Leave not a wreck behind." 
 
 Cast thine eye around thee, and see the thousands of 
 nature's productions. Take a view of them from the 
 opening seed on the surface, sending a downward shoot, to 
 the loftiest and the largest trees, rising up and blooming in 
 wild luxuriance ; some side by side, others separate ; some 
 curved and knotty, others straight as lances ; all, in 
 beautiful gradation, fulfilling the mandates they had 
 received from Heaven, and though condemned to die, 
 still never failing to keep up their species till time shall 
 be no more. 
 
 I i 
 
108 
 
 "VVAXDEHINGS IN SOl'TH AMERICA. 
 
 FinsT 
 
 if: 
 
 Header, canst thou not be induced to dedicate a few 
 JO URNE Y. ujQj^ths to the good of tlie public, and examine with thy 
 scientific eye the productions which the vast and well- 
 stored colony of Demerara presents to thee ? 
 
 What an immense range of forest is there from the rock 
 Saba to the great fall ! and what an uninterrupted extent 
 before thee from it to the banks of the Essequibo ! No 
 doubt, there is many a balsam and many a medicinal root 
 yet to be discovered, and many a resin, gum, and oil yet 
 unnoticed. Thy work would be a pleasing one, and thou 
 mightest make several useful observations in it. 
 
 Would it be thought impertinent in thee to hazard a 
 conjecture, that with the resources the government of 
 Demerara has, stones might be conveyed from the rock 
 Saba to Stabroek, to stem the equinoctial tides, which are 
 for ever sweeping away the expensive wooden piles round 
 the mounds of the fort ? Or would the timber-merchant 
 point at thee in passing by, and call thee a descendant of 
 La Mancha's knight, because thou maintainest that the 
 stones which form tlie rapids might be removed with little 
 expense, and thus open the navigation to the woodcutter 
 from Stabroek to the great ftdl ? Or woul'dst tliou be 
 deemed enthusiastic or biassed, because thou givest it as 
 thy opinion that the climate in these higli lands is exceed- 
 ingly wholesome, and the lands themselves capable of 
 nourishing and maintaining any number of settlers ? In 
 thy dissertation on the Indians, thou mightest hint, that 
 possibly they could be induced to help the new settlers a 
 little; and that finding their labours well requited, it 
 would be the means of tlieir keeping up a constant coni- 
 nnuiication with us, which probably might be the means 
 of laying the first ston'^ towards their Christianity. They 
 are a poor, harmless - (offensive set of people, and their 
 wandering and ill-pr( »'idcd way of living seems more to 
 
WANDERINGS IN SOUTH AMERICA. 
 
 109 
 
 ask for pity from us, tli.in to till our heads with thoughts 
 that thoy wouUl ^e hostile to us. 
 
 What a noble fichl, kind reader, for thy experimental 
 philosopliy and speculations, for thy learning, for thy 
 perseverance, for thy kind-heartedness, for everything that 
 isgrcat and good within thee ! 
 
 The accidental traveller who has journeyed on from 
 Stahroek to the rock Saba, and from thence to the banks of 
 the Essequibo, in pursuit of other things, as he told thee at 
 the beginning, with but an indiiferent iiiter[)reter to talk to, 
 no friend to converse with, and tcjtally unlit for that which 
 he washes thee to do, can merely mark the outlines of the 
 path he has trodden, or tell thee the sounds he has heard, 
 or faintly describe what he has seen in the environs of his 
 resting-places ; but if this be enough to induce thee to 
 undertake the journey, and give the world a description of 
 it, he will be amply satisfied. 
 
 It will be two days and a half from the time of entering 
 the path on the western bank of the Demerara till all be 
 ready, and the canoe fairly afloat on the Essequibo. The 
 new rigging it, and putting every little thing to rights 
 and in its proper place, cannot well be done in less than 
 a day. 
 
 After being night and day in the forest impervious to the 
 sun's and moon's rays the sudden traii ition to light has a 
 fine heart-cheering effect. Welcome as a lost friend, the 
 solar beam makes the frame rejoice, and with it a thousand 
 enlivening thoughts rush at once on the soul, and disperse, 
 as a vapour, every sad and sorrowful idea which the deep 
 gloom had helped to collect there. In coming out of the 
 woods, you see the western bank of the Essequibo before 
 you, low and Hat. Here the river is two-thirds as broad as 
 the Demerara at Stabrock. 
 
 To the northward there is a hill higher than any in the 
 
 FIRST 
 JOUltNEY. 
 
 1 
 
•r 
 
 In, > 
 
 :ii 
 
 It 
 
 p 
 
 h- ■. '1 
 
 Ui 
 
 1 ■ I 
 ■ i 
 
 110 
 
 VVANDEHINGS IN SOUTH AMERICA. 
 
 FIK8T 
 oOURNKY. 
 
 Islaiids. 
 
 Falls and 
 rapids. 
 
 W '! 
 
 Demerara ; and in the south-south-west quarter a mountain. 
 It is far away, and appears like a bluish cloud in the 
 horizon. There is not the least opening on either side. 
 Hills, valleys, and lowlands, are all linked together by a 
 chain of forest. Ascend the highest mountain, climb 
 the loftiest tree, as far as the eye can extend, which- 
 ever way it directs itself, all is luxuriant and unbroken 
 forest. 
 
 In about nine or ten hours from this, you get to an 
 Indian habitation of three huts, on the point of an island. 
 It is said that a Dutch post once stood here. But there is 
 not the smallest vestige of it remaining, and, except that 
 the trees appear younger than those on the other islands* 
 which shows that the place has been cleared some time 
 or other, there is no mark left by which you can con- 
 jecture that ever this was a post. 
 
 The many islands which you meet with in the way, 
 enliven and change the scene, by the avenues which they 
 make, which look like the mouths of other rivers, and 
 break that long-extended sameness which is seen in the 
 Demerara. 
 
 Proceeding onwards, you get to the falls and rapids. In 
 the rainy season they are very tedious to pass, and often 
 stop your course. In the dry season, by stepping from 
 rock to rock, the Indians soon manage to get a canoe over 
 them. But when the river is swollen, as it was in May* 
 1812, it is then a difficult task, and often a dangerous one 
 too. At that time many of the islands were overflowed, 
 the rocks covered, and the lower branches of the trees in 
 the water. Sometimes the Indians were obliged to take 
 everything out of the canoe, cut a passage through the 
 branches, which hung over into the river, and then drag 
 up the canoe by main force. 
 
 At one place, the falls form an oblique line quite across 
 
 '^■'^i 
 
WAXDEUINGS IN SOUTH AMEUICA. 
 
 Ill 
 
 FIl*... 
 JOURNEY. 
 
 the river, impassable to tlie asceiidiiij^ canoe, and you are 
 forced to have it dragged four or five hundred yards by 
 land. 
 
 It will take you five days, from the Indian habitation, 
 on the point of the island, to where these falls and rapids 
 terminate. 
 
 There are no huts in the way. You must bring your 
 own cassava-bread along with you, hunt in the forest for 
 your meat, and make the night's shelter for yourself. 
 
 Here is a noble range of hills, all covered with the finest 
 trees, rising majestically one above the other, on the 
 western bank, and presenting as rich a scene as ever the 
 eye would wish to look on. Nothing in vegetable nature 
 can be conceived more charming, grand, and luxuriant. 
 
 How the heart rejoices in viewing this beautiful land- 
 scape ! when the sky is serene, the air cool, and the sun 
 just sunk behind the mountain's top. 
 
 The Hayawa-tree perfumes the woods around ; pairs of 
 Scarlet Aras are continually crossing the river. The Maam 
 sends forth its plaintive note, the Wren chants its evening 
 song. The Capri mulgus wheels in busy flight around the 
 canoe, while " Whip-poor-Will " sits on the broken stump 
 near the water's edge, complaining as the shades of night 
 set in. 
 
 A little before you pass the last of these rapids two Eocks. 
 immense rocks appear, nearly on the summit of one of the 
 many hills which form this far-extending range, where it 
 begins to fall off gradually to the south. 
 
 They look like two ancient stately towers of some 
 Gothic potentate, rearing their heads above the surround- 
 ing trees. What with their situation and their shape 
 together, they strike the beholder with an idea of anti- 
 quated grandeur which he will never forget. He may 
 
 travel far and near and see nothing like them. 
 
 On looking 
 
112 
 
 WAXDEHINGS IN SOUTH AMERICA. 
 
 ■ i 
 
 I 'I '' 
 
 ilt- 
 
 '*!;! 
 
 riusT 
 
 JOURNEY. 
 
 Jtivcr 
 Apoura- 
 
 Macovshi 
 Indians. 
 
 at them thvoujfh a yla.ss, the sunnnit of tlie southern one 
 appeared crowned with buslies. Tlie one to the north was 
 auite bare. The Indians liave it from their ancestors 
 tliat they are the abode of an evil genius, and they pass in 
 the river below with a reverential awe. 
 
 In about seven hours from these stupendous sons of the 
 hill, you leave the Essequibo, and enter the river Apoura- 
 poura, which falls into it from the south. The Apoura- 
 poura is nearly one-third the size of the Demerara at 
 Stabroek. For two days you see nothing but level ground, 
 ricldy clothed in timber. You leave the Siparouni to the 
 right hand, and on the third day come to a little hill. The 
 Indians have cleared about an acre of ground on it, and 
 erected a temporary shed. If it be not intended for pro- 
 vision-ground alone, perliaps the next white man who 
 travels through these remote wilds will find an Indian 
 settlement here. 
 
 Two days after leaving this, you get to a rising ground 
 on the western bank, where stands a single hut ; and about 
 half a mile in the forest there are a few more ; son t of 
 them square, and some round with spiral roots. 
 
 Here the fish called Pacou is very plentiful : it is 
 perhaps the fattest and most delicious fish in Guiana. It 
 does not take the hook, but the Indians decoy it to the 
 surface of the water by means of the seeds of the Crab- 
 wood-tree, and then shoot it with an arrow. 
 
 You are now within the borders of Macoushia, inhabited 
 by a different tribe of people, called Macoushi Indians ; 
 uncommonly dexterous in the use of the blow-pipe, and 
 famous for their skill in preparing the deadly vegetable 
 poison, commonly called Wourali. 
 
 It is from this country that those beautiful parroquets 
 named Kessi-kessi are procured. Here the crystal moun- 
 tains are found ; and here the three different species of the 
 
 lir 
 
WANDEIilNGS IN SOUTH AMERICA. 
 
 Pf» 
 
 Ara ure seen in great abundance. Here, too, grows the tree first 
 
 from which the giim-elnstic is got: it is large, and as tall ^_^' " 
 
 as any in the forest. Tlie wood has much the appearance 
 
 of sycamore. The gum is contained in the bark : when that 
 
 is cut through, it oozes out very freely : it is quite white, 
 
 and looks as rich as cream : it hardens almost immediately 
 
 as it issues from the tree ; so that it is very day to collect 
 
 ii ball, by forming the juice into a globular shape as fast 
 
 MS it comes out : it becomes nearly black by being exposed 
 
 to the air, and is real India-rubber without undergoing 
 
 liny process. 
 
 The elegant crested bird called Cock of the Eock, 
 admirably described by Bufl'on, is a native of the woody 
 mountains of Macoushia. In the daytime, he retires amongst 
 the darkest rocks, and only comes out to feed a little 
 before sunrise, and at sunset : he is of a gloomy disposition, 
 and, like the houtou, never associates with the other birds 
 of the forest. 
 
 The Indians, in the just-mentioned settlement, seemed 
 to depend more on the wourali-poison for killing their 
 game than upon anything else. They had only one gun, 
 and it appeared rusty and neglected ; but their poisoned 
 weapons were in fine order. Their blow-pipes hung from Indian 
 the roof of the hut, carefully suspended by a silk-grass ^^^^'^■'^''^P^'^- 
 cord; and on taking a nearer view of them, no dust 
 .seemed to have collected there, nor had the spider spun 
 tlie smallest web on them; which showed that they were 
 in constant use. The quivers were close by them, with 
 the jaw-bone of the fish Pirai tied by a string to their 
 brim, and a small wicker-basket of wild cotton, which 
 liung down to the centre ; they were nearly full of poisoned 
 arrows. It was with difficulty these Indians could be 
 persuaded to part with any of the wourali-poison, though a 
 good price was offered for it ; they gave me to understand 
 
 1 
 
 I 
 
114 
 
 WAXDEUINQS IS SOUTH AMEIUCA. 
 
 FIRST 
 JOURNEY. 
 
 Lake 
 Parima. 
 
 that it was powder and shot to them, and very difficult 
 to be procured. 
 
 On the second day after leaving the settlement, in 
 passing along, the Indians show you a place where once 
 a white man lived. Ilia retiring so far from those of 
 his own colour and acquaintance seemed to carry some- 
 thing extraordinary along with it, and raised a desire to 
 know what could have induced him to do so. It seems 
 he had been unsuccessful, and that his creditors had 
 treated him with as little mercy as the strong generally 
 show to the weak. Seeing his endeavours daily frustrated, 
 and his best intentions of no avail, and fearing that when 
 they had taken all he had they would probably take his 
 liberty too, he thought the world would not be hard- 
 hearted enough to condemn him for retiring from the evils 
 which pressed so heavily on him, and which he had done 
 all that an honest man could do to ward off. He left his 
 creditors to talk of him as they thought fit, and, bidding 
 adieu for ever to the place in which he had once seen 
 better times, he penetrated thus far into those remote 
 and gloomy wilds, and ended his days here. 
 
 According to the new map of South America, Lake 
 Parima, or the White Sea, ought to be within three or 
 four days' walk from this place. On asking the Indians 
 whether there was such a place or not, and describing that 
 the water was fresh and good to drink, an old Indian, 
 who appeared to be about sixty, said that there was such 
 a place, and that he had been there. This information 
 would have been satisfactory in some degree, had not 
 the Indians carried the point a little too far. It is very 
 large, said another Indian, and ships come to it. Now 
 these unfortunate ships were the very things which were 
 not wanted : had he kept them out, it might have done, 
 but his introducing them was sadly against the lake. 
 
WANORIJINOS IN ROtlTIl AMEIU< A. 
 
 115 
 
 Tims you must either supposo that the ohl savage an<l rnwr 
 his companion had a confused idea of the thin*,', and that ''" ""'^''• ^- 
 probably the Lake I'arinia they talked of was the Atna/.ons, 
 not far from the city f>f Para, or that it was their intention 
 to deceive you. You ought to be cautious in giving credit 
 to their stories, otherwise you will be apt to be led astray. 
 
 Many a ridiculous thing concerning the interior of 
 Guiana has been propagated and received as true, merely 
 because six or seven Indians, questioned separately, have 
 aiireed in their narrative. 
 
 .Ask those who live high up in the Demerara, and they 
 will, every one of them, tell you that there is a nation of 
 Indians with long io'xh ; that they are very malicious, 
 cruel, and ill-natured ; and that the Portuguese have been 
 obliged to stop them off in a certain river, to prevent their 
 depredations. They have also dreadful stories concerning 
 a horrible beast, called the Watermanima, which, when it 
 happens to take a spite against a canoe, rises out of the 
 river, and in the most unrelenting manner possible carries 
 both canoe and Indians down to the bottom with it, and 
 there destroys them. Ludicrous extravagances; pleasing to 
 those fond of the marvellous, and excellent matter for a 
 distempered brain. 
 
 The misinformed and timid court of policy in Demerara Anecdote. 
 was made the dupe of a savage, who came down the Esse- 
 quibo, and gave himself out as king of a mighty tribe. This 
 naked wild man of the woods seemed to hold the said 
 court in tolerable contempt, and demanded immense 
 supplies, all which he got ; and moreover, some time after, 
 an invitation to come down the ensuing year for more, 
 which he took care not to forget. 
 
 This noisy chieftain boasted so much of his dynasty and 
 domain, that the government was induced to send up an 
 expedition into his territories to see if he had spoken the 
 
 I 2 
 
116 
 
 WANDERINGS IN SOUTH AMEKICA. 
 
 :'l:' l| 
 
 FIRST truth, and nothing but the truth. It appeared, however, 
 JO URNE Y, ji^j^^ Yus palace was nothing but a hut, the monarch a needy 
 savage, the heir-apparent nothing to inherit but his 
 father's club, and bow and arrows, and his officers of state 
 wild and uncultivated as the forests through which they 
 strayed. 
 
 There was nothing iu the hut of this savage, saving the 
 presents he had received from government, but what was 
 barely sufficient to support existence ; nothing that indi- 
 cated a power to collect a hostile force ; nothing that 
 showed the least progress towards civilization. All was 
 rude and barbarous in the extreme, expressive of the 
 utmost poverty and a scanty population. 
 
 You may travel six or seven days without seeing a hut, 
 and when you reach a settlement it seldom contains more 
 than ten. 
 
 The farther you advance into the interior the more you 
 are convinced that it is thinly inhabited. 
 
 The day after passing the place where the white man 
 lived you see a creek on the left hand, and shortly after 
 the path to the open country. Here you drag the canoe 
 up into the forest, and leave it there. Your baggage must 
 now be carried by the Indians. The creek you passed in 
 the river intersects the path to the next settlement : a 
 large Mora has fallen across it, and makes an excellent 
 bridge. After walking an hour and a half you come to 
 the edge of the forest, and a savanna unfolds itself to the 
 view. 
 
 The finest park that England boasts falls far short of 
 this delightful scene. There are about two thousand acres 
 of grass, with here and there a clump of trees, and a few 
 bushes and sincjle trees scattered up and down by the hand 
 of Nature. The ground is neither hilly nor level, but 
 diversified with moderate rises and falls, so gently running 
 
 .i'i. 
 
 11 1 
 
WANDERINGS IN SOUTH AMERICA. 
 
 ir 
 
 into one another that the eye cannot distinguish where 
 they begin, nor where they end, while the distant black 
 rocks have the appearance of a herd at rest. Nearly in 
 the middle there is an eminence, which falls off gradually 
 on every side ; and on this the Indians have erected their 
 Imts. 
 
 To the northward of them the foremost forms a circle, 
 as though it had been done by art; to the eastward it 
 hangs in festoons ; and to the south and west it rushes 
 in abruptly, disclosing a new scene behind it at every step 
 as you advance along. 
 
 This beautiful park of nature is quite surrounded by 
 lofty hills, all arrayed in superbest garb of trees ; some in 
 the form of pyramids, others like sugar-loaves towering 
 one above the other, some rounded off, and others as 
 though they had lost their apex. Here two hills rise up 
 in spiral summits, and the wooded line of communication 
 betwixt them sinks so gradually that it forms a crescent ; 
 and there the ridges of others resemble the waves of an 
 agitated sea. Beyond these appear others, and others past 
 them ; and others still farther on, till they can scarcely be 
 distinguished from the clouds. 
 
 There are no sand-flies, nor bete-rouge, nor mosquitos, 
 in this pretty spot. The fire-flies during the night vie in 
 numbers and brightness with the stars in the firmament 
 above; the air is pure, and the north-east breeze blows 
 a refreshing gale throughout the day. Here the White- 
 crested ^laroudi, wliich is never found in the Demerara, 
 is pretty plentiful ; and here grows the tree wliich pro- 
 duces the Moran, sometimes called Balsani-capivi. 
 
 Your route lies south from this place ; and at the ex- 
 tremity of the savanna you enter the forest, and journey 
 along a winding path at the foot of a hill. There is no 
 habitation within this day's walk. The traveller, as usual, 
 
 FIRST 
 .TOURNEY. 
 
 Euiite 
 
118 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FIRST 
 JOURNEY. 
 
 '%'• 
 
 The 
 Jabiru. 
 
 must sleep in the forest; the path is not so good the fol- 
 lowing day. The hills over which it lies are rocky, steep, 
 and rugged, and the spaces betwixt them swampy, and 
 mostly knee-deep in water. After eight hours' walk you 
 find two or three Indian huts, surrounded by the forest ; 
 and in little more than half an hour from these you come 
 to ten or twelve others, where you pass the night. They 
 are prettily situated at the entrance into a savanna. The 
 eastern and western hills are still covered with wood ; but 
 on looking to the south-west quarter you perceive it be- 
 gins to die away. In these forests you may find plenty 
 of the trees which yield the sweet-smelling resin called 
 Acaiari, and which, when pounded and burnt on charcoal, 
 gives a delightful fragrance. 
 
 From hence you proceed, in a south-west direction, 
 through a long swampy savanna. Some of the hills 
 which border on it have nothing but a thin coarse grass 
 and huge stones on them ; others quite wooded ; others 
 with their summits crowned, and their base quite bare ; 
 and others, again, v/ith their summits bare, and their base 
 in thickest wood. 
 
 Half of this day's march is in water, nearly up to the 
 knees. There are four creeks to pass : one of them has a 
 fallen tree across it. You must make your own bridge 
 across the other three. Probably, were the truth knowni 
 these apparently four creeks are only the m'^anders of one. 
 
 The Jabiru, the largest bird in Guiana, feeds in the 
 marshy savanna through which you have just passed. He 
 is wary and shy, and will not allow you to get within 
 gun-shot of him. 
 
 You sleep this night in the forest, and reach an Indian 
 settlement about three o'clock the ne.xt evening, after 
 walking one-third of the way through wet and miry 
 ground. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 119 
 
 Indian 
 after 
 miry 
 
 But, bad as the walking is through it, it is easier than first 
 where you cross over the bare hills, where you have to journey. 
 tread on sharp stones, most of them lying edgewise. 
 
 The ground gone over these two last days seems con- 
 demned to perpetual solitude and silence. There was not 
 one four-footed animal to be seen, nor even the marks of 
 one. It would have been as silent as midnight, and all as 
 still and unmoved as a monument, had not the Jabiru in 
 the marsh, and a few Vultures soaring over the mountain's 
 top, shown that it was not quite deserted by animated na- 
 ture. There were no insects, except one kind of fly, about 
 one -fourth the size of the common house-fly. It bit cruelly, 
 and was much more tormenting than the mosquito on the 
 sea-coast. 
 
 This seems to be the native country of the Arrowroot. Arrowroot 
 Wherever you passed through a patch of wood in a low 
 situation, there you found it growing luxuriantly. 
 
 The Indian place you are now at is not the proper place 
 to have come to in order to reach the Portuguese frontiers. 
 You have advanced too much to the westward. But 
 there was no alternative. The ground betwixt you and 
 another small settlement (which was the right place to 
 have gone to) was overflowed ; and thus, instead of pro- 
 ceeding southward, you were obliged to wind along the 
 foot of the we tern hills, quite out of your way. 
 
 But the grand landscape this place affords makes you 
 ample amends for the time you have spent in reaching it. 
 It would require great descriptive powers to give a proper 
 idea of the situation these people have chosen for their 
 dwelling. 
 
 The hill they are on is steep and high, and full of im- 
 mense rocks. The huts are not all in one place, but 
 dispersed wherever they have found a place level enough 
 for a lodgement. Before you ascend the hill you see at 
 
f 
 
 I -JO 
 
 FIRST 
 J(»URNEY. 
 
 WANDEIilNGS IN SOUTH AMERICA. 
 
 Immense 
 2)lain. 
 
 Creek. 
 
 intervals an acre or two of wood, then an open space, with 
 a few Imts on it ; then wood again, and then an open space, 
 and so on, till the intervening of the western hills, higher 
 and steeper still, and crowned with trees of the loveliest 
 shades, closes the enchanting scene. 
 
 At the base of this hill stretches an immense plain, 
 which appears to the eye, on this elevated spot, as level 
 as a bowling-green. The mountains on the other side are 
 piled one upon the other in romantic forms, and gradually 
 retire, till they are undiscernible from the clouds in which 
 they are involved. To the south-south-west this far- 
 extending plain is lost in the horizon. The trees on it, 
 which look like islands on the ocean, add greatly to the 
 beauty of the landscape ; while the rivulet's course is 
 marked out by the iEta-trees which follow its meanders. 
 
 Not being able to pursue the direct course from hence 
 to the next Indian habitation on account of the floods of 
 water that fall at this time of the year, you take a circuit 
 westerly along the mountain's foot. 
 
 At last a large and deep creek stops your progress : it is 
 wide and rapid, and its banks very steep. There is neither 
 curial nor canoe, nor purple-heart tree in the neighbour- 
 hood to make a Wood-skin to carry you over, so that you 
 are obliged to swim across ; and by the time you have 
 formed a kind of raft, composed of boughs of trees and 
 coarse grass, to ferry over your baggage, the day will be 
 too far spent to think of proceeding. You must be very 
 cautious before you venture to swim across this creek, for 
 the alligators are numerous, and near twenty feet long. On 
 the present occasion the Indians took uncommon precau- 
 tions lest they should lie devoured by this cruel and vora- 
 cious reptile. They cut long sticks, and examined closely 
 the side of the creek for half a mile above and below the 
 place where it was to be crossed ; and as soon as the 
 
WANDERINGS IN SOUTH AMERICA. 
 
 121 
 
 boldest had swum over, lie did the same on the other side, first 
 and then all followed. jouunky. 
 
 After passing the night on the opposite bank, which is 
 well wooded, it is a brisk walk of nine hours before you 
 reach four Indian huts, on a rising ground a few liundred 
 paces from a little brook, whose banks are covered over 
 with Coucourite and ^Eta-trees. 
 
 This is the place you ought to have come to two days 
 ago had the water permitted you. In crossing the plain 
 at the most advantageous place you are ab.ove ankle-d(;ep 
 in water for three hours ; the remainder of the way is dry, 
 the ground gently rising. As the lower parts of this 
 spacious plain put on somewhat the appearance of a lake 
 during the periodical rains, it is improbable but that this 
 is the place which hath given rise to the supposed exist- 
 ence of the famed Lake Parima, or El Dorado ; but this is 
 mere conjecture. 
 
 A few Deer are feeding on the coarse rough grass of this 
 far-extending plain ; they keep at a distance from you, and 
 are continually on the look-out. 
 
 The Spur-winged Plover, and a species of the Curlew, 
 black, with a white bar across the wings, nearly as large 
 again as the scarlet curlew on the sea-coast, frequently 
 rise before you. Here, too, the Moscovy Duck is numer- 
 ous ; and large flocks of two other kinds wheel round you 
 as you pass on, but keep out of gun-shot. The milk-white 
 Egrets and Jabirus are distinguished at a great distance ; 
 and in the ffita and coucourite trees you may observe flocks 
 of scarlet and blue Aras feeding on the seeds. 
 
 It is to these trees that the largest sort of Toucan resorts. Toucan. 
 He is remarkable by a large black sjiot on the point of his 
 fine yellow bill. He is very scarce in Demerara, and never 
 seen except near the sea-coast. 
 
 The Ants' Nests have a singular appearance on this plain. ,,/^J^ 
 
 ■*> , 
 
 ii« 
 
122 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FIRST They are in vast abundance on those parts of it free from 
 
 JO URNE Y, -yvater, and are formed of an exceedingly liard yellow clay. 
 
 They rise eight or ten feet from the ground, in a spiral 
 
 form, impenetrable to the rain, and strong enough to defy 
 
 the severest tornado. 
 
 The wourali-poison, procured in these last-mentioned 
 huts, seemed very good, and proved afterwards to be 
 very strong. 
 Portuguese There are now no more Indian settlements betv.xt you 
 fron lers. ^^^ ^-^q Portuguese frontiers. If you wish to visit their 
 fort, it would be advisable to send an Indian with a letter 
 from hence, and wait his return. On the present occasion 
 a very fortunate circumstance occurred. The Portuguese 
 commander had sent some Indians and soldiers to build a 
 canoe, not far from this settlement ; they had just finished 
 it, and those who did not stay with it had stoj)ped here on 
 their return. 
 
 The soldier who commanded the rest, said, he durst 
 not, upon any account, convey a stranger to the fort ; but 
 he added, as there were two canoes, one of them might be 
 despatched with a letter, and then we could proceed slowly 
 on in the other. 
 
 About three hours from this settlement there is a 
 river called Pirarara; and here the soldiers had left their 
 canoes while they were making the new one. From the 
 Pirarara you get into the river Maou, and then into the 
 Tacatou ; and just where the Tacatou falls into the Rio 
 Branco, there stands the Portuguese frontier fort, called 
 Fort. St. Joachim. From the time of embarking in the 
 river Pirarara, it takes you four days before you reach this 
 fort. 
 
 There was nothing very remarkable in passing down 
 these rivers. It is an open country, producing a coarse 
 grass, and interspersed with clumps of trees. The banks 
 
WANDKRINGS IN SOUTH AMERICA. 
 
 123 
 
 a 
 
 liave some wood on them, but it appears stinted and 
 crooked, like that on the bleak hills in England. 
 
 The Tapir frequently plunged into the river ; he was by 
 no means shy, and it was easy to get a shot at him on 
 land. The Kessi-kessi paroquets were in great abundance ; 
 and the fine scarlet Aras innumerable in the coucourite- 
 trees at a distance from the river's bank. In the Tacatou 
 was seen the Troupiale. It was charming to hear 
 the sweet and plaintive notes of this pretty songster of 
 the wilds. The Portuguese call it the Nightingale of 
 Guiana. 
 
 Towards the close of the fourth evening, the canoe. 
 
 FIRST 
 JOURNEY. 
 
 Message 
 
 which had been sent on with a letter, met us with the Portnymse 
 commander's answer. During its absence, the nights had com- 
 
 uks 
 
 been cold and stormy, the rain had fallen in torrents, 
 the days cloudy, and there was no sun to dry the wet 
 liammocks. Exposed thus, day and night, to the chilling 
 blast and pelting shower, strength of constitution at last 
 failed, and a severe fever came on. The commander's 
 answer was very polite. He remarked, he regretted much 
 to say, that he had received orders to allow no stranger to 
 enter the frontier, and this being the case, he hoped I 
 would not consider him as uncivil : "howevar," continued 
 he, " I have ordered the soldier to land you at a certain 
 distance from the fort, where we can consult together." 
 
 We had now arrived at the place, and the canoe which 
 brought the letter returned to the fort, to tell the com- 
 mander I had fallen sick. 
 
 The sun had not risen above an Ir.our the morning after 
 when the Portuguese officer came to the spot where we 
 had landed the preceding evening. He was tall and 
 spare, and appeared to be from fifty to fifty-five years old ; 
 and though thirty years of service under an equatorial sun 
 had burnt and shrivelled up his face, still there was some- 
 
 mander. 
 
124 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FIRST 
 .T(»rilSEY. 
 
 'I i 
 
 ;K ; 
 
 1 I 
 
 Fort St. 
 Joachim. 
 
 thing in it so inexpressibly affable and kind, that it set 
 you immediately at your ease. He came close up to the 
 hammock, and taking hold of my wrist to feel the pulse, 
 " I am sorry, sir," said he, " to see that the fever has taken 
 such hold of you. You shall go directly with me," con- 
 tinued he, " to the fort ; and though we have no doctor 
 there, I trust," added he, " we shall soon bring you about 
 again. The orders I have received forbidding the ad- 
 mission of strangers were never intended to be put in 
 force against a sick English gentleman." 
 
 As the canoe was proceeding slowly down the river 
 towards the fort, the commander asked, with much more 
 interest than a question in ordinary conversation is asked, 
 where was I on the night of the first of May ? On telling 
 him that I vvas at an Indian settlement a little below the 
 great fall in the Demerara, and that a strange and sudden 
 noise had alarmed all the Indians, he said the same 
 astonishing noise had roused every man in Foi-t St. 
 Joachim, and that they remained under arms till morning. 
 He observed, that he had been quite at a loss to form any 
 idea what could have caused the noise ; but now learning 
 that the same noise had been heard at the same time far 
 away from the Eio Branco, it struck him there must have 
 been an earthquake somewhere or other. 
 
 Good nourishment and rest, and the unwearied attention 
 and kindness of the, Portuguese commander, stopped the 
 progress .of the fever, and enabled me to walk about in six 
 days. 
 
 Fort St. Joachim was built about five and forty years 
 ago, under the apprehension, it is said, that the Spaniards 
 were coming from the llio Negro to settle there. It has 
 been much neglected; the floods of water have carried 
 away the gate, and desiroyed the wall on each side of it ; 
 but the present commander is putting it into thorough 
 
WANDEHIN(JS IN SOUTH AMEHICA. 
 
 IfT. 
 
 far 
 
 the 
 six 
 
 rears 
 
 lards 
 
 has 
 
 rried 
 
 it; 
 
 [)\igh 
 
 repair. When finished, it v/ill mount six nine and six first 
 twelve pounders. journt^y, 
 
 In a straight line with the fort, and within a few yards 
 of the river, stand the conimauder's house, the barracks, 
 the chapel, the father confessor's house, and two others, 
 all at little intervals from each other ; and these are tlie 
 only buildings at Fort St. Joachim. The neighbouring 
 extensive plains afford good pasturage for a fine breed of 
 cattle, and the Portuguese make enough of butter and 
 cheese for their own consumption. 
 
 On asking the old officer if there were such a place as 
 Lake Parima, or El Dorado, he replied, he looked upon it 
 as imaginary altogether. " I have been above forty years," 
 added he, " in Portuguese Guiana, but have never yet met 
 with anybody who has seen the lake." 
 
 So much for Lake Parima, or El Dorado, or the White 
 Sea. Its existence at best seems doubtful; some affirm 
 that there is such a place, and others deny it. 
 
m 
 
 CHAPTER II. 
 
 "Clmniinatici certant, et adliuc sub juJice lis est." 
 
 The Macoushi Indians— Poison vendors — Apparent failure of poison — 
 Collecting materials for wourtili-poison — Preparing the poison — Super- 
 stitions — The blow-pipe gun — The Ourah — The Samourah — Silk-grass 
 — Acuero fruit — Coucourite palm — Wild Cotton — Arrows— Quivers — 
 Jaw oC Pirai — Packing the arrows — Cotton basket — Gun sight made 
 of Acouri teeth — Poisoned fowl — Suspending the guns — The bow — 
 Ingenious arrows — Small quivers — A wild hog shot — Utilization of 
 indigenous products. 
 
 FIRST 
 JOURNEY 
 
 hi '.! 
 
 Wourali- 
 poison. 
 
 Having now reached the Portuguese inland frontier, and 
 collected a sufficient quantity of the wourali-poison, 
 nothing remains but to give a brief account of its com- 
 position, its effects, its uses, and its supposed antidote^. 
 
 It has been already remarked, that in the extensive M'ilds 
 of Demerara and Essequibo, far away from any European 
 settlement, there is a tribe of Indians who are known by 
 the name of Macoushi. 
 
 Though the wourali-poison is used by all the South 
 American savages betwixt the Amazons and the Oroonoque, 
 still this tribe makes it stronger than any of the rest. 
 The Indians in the vicinity of the Eio Negro are aware of 
 this, and come to the Macoushi country to purchase it. 
 Its effects. Much has been said concerning this "atal and extra- 
 ordinary poison. Some have affirmed that its effects are 
 almost instantaneous, provided the minutest particle of 
 
WANDEUFNdS IN SOUTH AMERICA. 
 
 127 
 
 it mixes vviLli the blood; and others again have maintained fiust 
 that it is not strong enough to kill an animal of the ■"' ^'"-"'^ y- 
 size and strength of a man. The first have erred by 
 lending a too willing ear to the marvellous, and be- 
 lieving assertions without sufficient proof. The follow- 
 ing short story points out the necessity of a cautious 
 examination. 
 
 One day, on asking an Indian if he thought the poison ^Ann'dote. 
 would kill a man, he replied, that they always go to battle 
 with it ; that he was standing by when an Indian was 
 shot with a poisoned arrow, and that he expired almost 
 immediately. Not wishing to dispute this apparently 
 satisfactory information, the subject was dropped. How- 
 ever, about an hour after, having purposely asked him in 
 what part of the body the said Indian was wounded, he 
 answered without hesitation, that the arrow entered 
 betwixt his shoulders, and passed quite through his heart. 
 Was it the weapon, or the strength of the poison, that 
 brought on immediate dissolution in this case ? Of course 
 the weapon. 
 
 The second have been misled by disappointment, 
 caused by neglect in keeping the poisoned arrows, or by 
 not knowing how to use them, or by trying inferior poison. 
 If the arrows are not kept dry, the poison loses its 
 strength ; and in wet or damp weather it turns mouldy, 
 and becomes quite soft. In shooting an arrow in this 
 state, upon examining the place where it has entered, 
 it will be observed that, though the arrow has pene- 
 trated deep into the flesh, still by far the greatest part 
 of the poison has shrunk back, and thus, instead of enter- 
 ing with the arrow, it has remained collected at the 
 mouth of the wound. In this case the arrow might as 
 well have not been poisoned. Probably, it was to this 
 that a gentleman, some time ago, owed his disappointment, 
 
1'2« 
 
 WANKKKINiiS IN Soimi AMKIJICA. 
 
 FMiHT when ho trifd the poison on a horso in tlio town of 
 .iMiMiNKY. Stahrook, th(^ capital of Deniorara; the horso never he- 
 trnyed the least 8ynii)toni of being alleetecl by it. 
 
 Wishful to obtain tiie best information concerning this 
 poison, and as repeated inqniries, in lieu of dissipating the 
 surrounding shade, did but tend more and more to darken 
 the little light that existed, I determined to penetrate 
 into the country where the poisonous ingredients grow, 
 where this pernicious composition is pre[)ared, and where 
 it is constantly used. Success attended the adventure ; 
 and the information acquired made amends for one hun- 
 dred and twenty days passed in the solitudes of Guiana, 
 and attbrded a balm to the wounds and bruises which 
 every traveller must expect to receive who wanders 
 through a thorny and obstructed path. 
 
 Thou must not, courteous reader, expect a dissertation 
 on the manner in which the wourali-poisou operates on 
 the system; a treatise has been already w^ritten on the 
 subject, and after all, there is probably still reason to 
 doubt. It is supposed to affect the nervous system, and 
 thus destroy the vital functions ; it is also said to be 
 perfectly hannless, provided it does not touch the blood. 
 However, this is certain, when a sufficient quantity of it 
 enters the blood, death is the inevitable consequence ; but 
 there is no alteration in the colour of the blood, and both 
 the blood and flesh may be eaten with safety. 
 
 All that thou wilt find here is a concise, unadorned 
 account of the wourali-poison. It may be of service to 
 thee some time or other, shouldst thou ever travel through 
 the wilds where it is used. !N either attribute to cruelty, 
 nor to a want of feeling for the sufferings of the inferior 
 animals, the ensuing experiments The larger animals 
 were destroyed in order to have proof positive of the 
 strength of a uoison which hath hitherto been doubted : 
 
 ;*■•;■ i'l. 
 
WANDERINCIS IN SOUTH AMEUICA. 
 
 129 
 
 and tho sinaller ones wore killed with the liope of sub- fiust 
 stantiating that wliich has conmionly been supposed to be ''" ''"'''''' '^' 
 an antidote. 
 
 It makes a pitying heart ache to see a poor creature in 
 distress and pain ; and too often has the compassionate 
 traveller occasion to hejivo a sigh as he journeys on. 
 However, here, though the kind-hearted will be sorry to 
 read of an unoffending animal doomed to death, in order 
 to satisfy a doubt, still it will be a relief to know that the 
 victim was not tortured. The wourali-i)ois()n destroys 
 life's action so gently, that the victim appears to be in no 
 pain whatever; and probably, were the truth known, it 
 feels none, saving tho momentary smart at the time the 
 arrow enters. 
 
 A day or two before the ^NFacoushi Indian prepares his 
 poison, he goes into the forest in quest of the ingredients, 
 A vine grows in these wilds, which is called "NVourali. It 
 is from this that the poison takes its name, and it is the 
 principal ingredient. When he has procured enough of 
 this, he digs up a root of a very bitter taste, ties them 
 together, and then looks about for two kinds of bulbous 
 plants, which contain a green and glutinous juice. He 
 fills a little quake, which he carries on his bad:, with the 
 stalks of these ; and lastly, ranges up and down till he 
 finds two species of ants. One of them is very large and 
 black, and so venomous, that its sting produces a fever ; 
 it is most commonly to be met with on the ground. The 
 other is a little red ant, which stings like a nettle, and 
 generally has its nest under the leaf of a shrub. After 
 obtaining these, he has no more need to range the forest. 
 
 A quantity of the strongest Indian pepper is used ; but 
 this he has already planted round his hut. The pounded 
 fangs of the Labarri snake, and those of the Couanacouchi, 
 are likewise added. These he commonly has in store ; for 
 
 K 
 
 
ti 
 
 
 ■I'll 
 
 III 
 
 N, 
 
 Fl 
 
 
 130 
 
 Finsr 
 
 JOUKNEY. 
 
 Prepara- 
 tion of the 
 Wourali' 
 poison. 
 
 i'abahou, 
 or evil 
 spirit. 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 when lie kills a snake, lie generally extracts the fangs, and 
 keeps them by him. 
 
 Having thus found the necessary ingredients, he scrapes 
 the wourali vine and bitter root into thin shavings, and 
 puts them into a kind of colander made of leaves : this 
 he holds over an earthen pot, and pours water on the 
 shavings : the liquor which comes through has the appear- 
 ance of coffee. When a sufficient quantity has been 
 procured, the shavings are thrown aside. He then bruises 
 the bulbous stalks, and squeezes a proportionate quantity 
 of their juice through his hands into the pot. Lastly, the 
 snakes' fangs, ants, and pepper are bruised, and thrown 
 into it. It is then placed on a slow fire, and as it boils 
 more of the juice of the wourali is added, according as it 
 may be found necessary, and the scum is taken off with a 
 leaf : it remains on the fire till reduced to a thick syrup of 
 a deep brown colour. As soon as it has arrived at this 
 state, a few arrows are poisoned with it, to try its strength. 
 If it answer the expectations, it is poured out into a 
 calabash, or little pot of Indian manufacture, which is 
 carefully covered with a couple of leaves, and over them 
 a piece of deer's skin, tied round with a cord. They keep 
 it in the most dry part of the hut; and from time to 
 time suspend it over the fire, to counteract the effects of 
 dampness. 
 
 The act of preparing this poison is not considered as a 
 common one : the savage may shape his bow, fasten the 
 barb on the point of his arrow, and make his other imple- 
 ments of destruction, either lying in his hammock, or in the 
 midst of his family ; but, if he has to prepare the wourali- 
 poison, many precautions are supposed to be necessary. 
 
 The women and young girls are not allowed to be 
 present, lest the Yabahou, or evil spirit, should do them 
 harm. The shed under which it has been boiled is 
 
WANDERINGS IN SOUTH AMETUrA. 
 
 pronounced polluted, and abandoned ever after. He who 
 makes the poison must eat nothing that morning, and 
 must continue fasting as long as the operation lasts. The 
 pot in which it is boiled must be a new one, and must 
 never have held anything before, otherwise the poison 
 would be deficient in strength : add to this, that the 
 operator must take particular care not to expose himself 
 to the vapour which arises from it while on the fire. 
 
 Though this and other precautions are taken, such as 
 frequently 'vashing the face and hands, still the Indians 
 think that it affects the health ; and the operator either is, 
 or, what is more probable, supposes himself to be, sick for 
 some days after. 
 
 Thus is appears that the making the wourali-poison is 
 considered as a gloomy and mysterious operation ; and it 
 would seem that they imagine it affects others as well as 
 him who boils it ; for an Indian agreed one evening to 
 make some for me, but the next morning he declined 
 having anything to do with it, alleging that his wife was 
 with child ! 
 
 Here it might be asked, are all the ingredients just 
 mentioned necessary, in order to produce the wourali- 
 poison ? Though our opinions and conjectures may mili- 
 tate against the absolute necessity of some of them, still 
 it would be hardly fair to pronounce them added by the 
 hand of superstition, till proof positive can be obtained. 
 
 We might argue on the subject, and, by bringing forward 
 instances of Indian superstition, draw our conclusion by 
 inference, and still remain in doubt on this head. You 
 know superstition to be the offspring of ignorance, and of 
 course that it takes up its abode amongst the rudest tribes 
 of uncivilized man. It even too often resides with man in 
 his more enlightened state. 
 
 The Augustan age furnishes numerous examples. A 
 
 2 
 
 mi 
 
 FIRST 
 JOUllNKV, 
 
 Indian 
 super- 
 slilion. 
 
 w 
 
I I 
 
 132 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 ; ;, 
 
 i 
 
 H li 
 
 FIRST bone snatched from the jaws of a fasting bitch, and a 
 JOURNEY, feather from the wing of a night owl, — " ossa ab ore rapta 
 jejunse canis, plumamque nocturnse strigis," — were neces- 
 sary for Canidia's incantations. And in aftertimes, parson 
 Evans, the Welshman, was treated most ungenteelly by an 
 enraged spirit, solely because he had forgotten a fumiga- 
 tion in liis witch-work. 
 
 If, then, enlightened man lets his better sense give way, 
 and believes, or allows himself to be persuaded, that cer- 
 tain substances and actions, in reality of no avail, possess 
 a virtue which renders them useful in producing the 
 wished-for effect; may not the wild, untaught, unen- 
 lightened savage of Guiana add an ingredient which, on 
 account of the harm it does him, he fancies ,may be useful 
 to the perfection of his poison, though in fact it be of no 
 use at all ? If a bone snatched from the jaws of a fasting 
 bitch be thought necessary in incantation ; or if witchcraft 
 have recourse to the raiment of the owl, because it resorts 
 to the tombs and mausoleums of the dead, and wails and 
 hovers about at the time that the rest of animated nature 
 sleeps; certainly the savage may imagine that the ants, 
 whose sting causes a fever, and the teeth of the Labarri 
 and Couanacouchi snakes, which convey death in a very 
 short space of time, are essentially necessary in the com- 
 position of his poison ; and being once impressed witli 
 this idea, he will add them every time he makes the 
 poison, and transmit the absolute use of them to his pos- 
 terity. The question to be answered seems not to be, if 
 it is natural for the Indians to mix these ingredients, but, 
 if they are essential to make the poison. 
 
 So mucli for the preparing of this vegetable essence; 
 terrible importer of death, into whatever animal it enters. 
 Let us now see how it is used ; let us examine the weapons 
 which bear it to its destination, and take a view of the 
 
WANDERINGS IN SOUTH AMKRICA. 
 
 133 
 
 japoiis 
 lof the 
 
 till FIRST 
 
 JOURNEY. 
 
 poor victim, from the time he receives his wound 
 
 death comes to his relief. 
 
 When a native of Macoushia goes in quest of feathered Descrip- 
 game or other birds, he seldom carries his bow and arrows. Bim-pipe. 
 It is the blow-pipe he then uses. This extraordinary tube 
 of death is, perhaps, one of the greatest natural curiosities 
 of Guiana. It is not found iu the country of the Ma- 
 coushi. Those Indians tell you that it grows to the south- 
 west of them, in the wilds which extend betwixt them 
 and the Eio Negro. The reed must grow to an amazing 
 length, as the part the Indians use is from ten to eleven 
 feet long, and no tapering can be perceived in it, one end 
 being as thick as the other. It is of a bright yellow 
 colour, perfectly smooth both inside and out. It grows 
 hollow ; 1."^ is there the least appearance of a knot or 
 joint thi'jT' liv t the whole extent. The natives call it 
 Ourah. 1 li', oi itself, is too slender to answer the end 
 of a blow-pipe; but there is a species of palma, larger 
 and stronger, and common in Guiana, and this the Indians 
 make use of as a case, in which they put the ourah. It is 
 brown, susceptible of a fine polish, and appears as if it 
 iiad joints five or six inches from each other. It is called 
 Samourah, and the pulp inside is easily extracted, by 
 steeping it for a few days in water. 
 
 Thus the ourah and samourah, one within the other, 
 form the blow-pipe of Guiana. The end which is applied 
 to the mouth is tied round with a small silk-grass cord, to 
 prevent its splitting ; and the other end, which is apt to 
 strike against the ground, is secured by the seed of the 
 Acuero fruit, cut horizontally through the middle, with a 
 hole made in the end, through which is put the extremity 
 of the blow-pipe. It is fastened on with string on the 
 outside, and the inside is filled up with wild bees'-wax. 
 The arrow is from nine to ten inches long. It is made The arrow. 
 
 
 i'Mi 
 
134 
 
 WANDERINGS IN SOUTH /MERICA. 
 
 FIRST out of the leaf of a species of palm-tree, called Coucourite, 
 JOURNEY. jjjj^j.jj j^jjjj brittle, and pointed as sharp as a needle. About 
 an inch of the pointed end is poisoned. The other end is 
 burnt to make it still harder, and wild cotton is put round 
 it for about an inch and a half. It requires considerable 
 practice to put on this cotton well. It must just be large 
 enough to fit the hollow of the tube, and taper off to 
 nothing downwards. They tie it on with a thread of the 
 silk-grass, to prevent its slipping off the arrow. 
 The quiver. The Indians have shown ingenuity in making a quiver 
 to hold the aiTows. It will contain from five to six hun- 
 dred. It is generally from twelve to fourteen inches long, 
 and in shape resembles a dice-box used at backgammon. 
 The inside is prettily done in basket-work, with wood not 
 unlike bamboo, and the outside has a coat of wax. The 
 cover is all of one piece, formed out of the skin of the 
 tapir. Round the centre there is fastened a loop, large 
 enough to admit the arm and shoulder, from which it 
 hangs when used. To the rim is tied a little bunch of 
 silk-grass, and half of the jaw-bone of the fish called 
 pirai, with which the Indian scrapes the point of his 
 arrow. 
 
 Before he puts the arrows into the quiver, he links 
 them together by two strings of cotton, one string at each 
 end, and then folds them round a stick, which is nearly 
 the length of the quiver. The end of the stick, which is 
 uppermost, is guarded by two little pieces of wood cross- 
 wise, with a hoop round their extremities, which appears 
 something like a v/heel; and this saves the hand from 
 being wounded when the quiver is reversed in order to let 
 the bunch of arrows drop out. 
 
 There is also attached to the quiver a little kind of 
 basket, to hold the wild cotton which is put on the blunt 
 end of the arrow. With a quiver of poisoned arrows 
 
WANDERINGS IN SOUTH AMERICA. 
 
 135 
 
 slung over his shoulder, and with his blow-pipe in liis 
 hand, in the same position as a soldier carries his musket, 
 see the Macoushi Indian advancing towards the forest in 
 quest of powises, maroudis, waracabas, and other feathered 
 game. 
 
 These generally sit high up in the tall and tufted trees, 
 but still are not out of the Indian's reach ; for his blow- 
 pipe, at its greatest elevation, will send an arrow three 
 hundred feet. Silent as midnight he steals under them, 
 and so cautiously does he tread the ground, that the fallen 
 leaves rustle not beneath his feet. His ears are open to 
 the least sound, while his eye, keen as that of the lynx, 
 is employed in finding out the game in the thickest shade. 
 Often he imitates their cry, and decoys them from tree to 
 tree, till they are within range of his tube. Then taking a 
 poisoned arrow from his quiver, he puts it in the blow-pipe, 
 and collects his breath for the fatal puff. 
 
 About two feet from the end through which he blows 
 there are fastened two teeth of the acouri, and these serve 
 him for a sight. Silent and swift the arrow flies, and 
 seldom fails to pierce the object at which it is sent. Some- 
 times the wounded bird remai is in the same tree where it 
 was shot, and in three minutes falls down at the Indian's 
 feet. Should he take wing, his flight is of short duration, 
 and the Indian, following the direction he has gone, is sure 
 to find him dead. 
 
 It is natural to imagine that, when a slight wound only 
 is inflicted, the game will make its escape. Far otherwise ; 
 the wourali-poison almost instantaneously mixes with blood 
 or water, so that if you wet your finger, and dash it along 
 the poisoned arrow in the quickest manner possible, you 
 are sure to carry off some of the poison. Though three 
 minutes generally elapse before tlie convulsions come on 
 in the w^ounded bird, still a stupor evidently takes plact; 
 
 FIRST 
 JOURNEY. 
 
 The Indian 
 in pursuit 
 
 of hia 
 
 game. 
 
 W\ 
 
 Effects of 
 tlie poison 
 
 on the 
 wounded 
 
 bird. 
 
136 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 !ii 
 
 FIRST sooner, and this stupor manifests itself by an apparent un- 
 jo uKNE Y. willingness in the bird to move. This was very visible in 
 a dying fowl. 
 
 Having ];)rocured a healthy full-grown one, a short piece 
 of a poisoned blow-pipe arrow was broken off and run up 
 into its thigh, as near as possible betwixt the skin and the 
 flesh, in order that it might not be incommoded by the 
 wound. For the first minute it walked about, but walked 
 very slowly, and did not appear the least agitated. During 
 the second minute it stood still, and began to peck the 
 ground ; and ere half another had elapsed, it frequently 
 opened and shut its mouth. The tail had now dropped, 
 and the wings almost touched the ground. By the ter- 
 mination of the third minute, it had sat down, scarce able 
 to support its head, which nodded, and then recovered itself, 
 and then nodded again, lower and lower every time, like 
 that of a weary traveller slumbering in an erect position ; 
 the eyes alternately open and shut. The fourth minute 
 brought on convulsions, and life and the fifth terminated 
 together. 
 
 The flesh of the game is not in the least injured by the 
 poison, nor does it appear to corrupt sooner than that killed 
 by the gun or knife. The body of this fowl was kept for 
 sixteen hours, in a climate damp and rainy, and within 
 seven degrees of the equator; at the end of which time 
 it had contracted no bad smell whatever, and there were 
 no symptoms of putrefaction, saving that, just round the 
 wound, the flesh appeared somewhat discoloured. 
 
 The Indian, on his return home, carefully suspends his 
 blow-pipe from the top of his spiral roof; seldom placing 
 it in an oblique position, lest it should receive a cast. 
 
 Here let the blow-pipe remain suspended, while you take 
 a view of the arms which are made to slay the larger 
 beasts of the forest. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 137 
 
 take 
 
 F[KST 
 JOURNEY. 
 
 When the Indian intends to chase the peccari, or 
 surprise the deer, or rouse the tapir from his marshy retreat, 
 he carries his bow and arrows, which are very different from . 
 the weapons already described. 
 
 The bow is generally from six to seven feet long, and The how 
 strung with a cord, spun out of the silk-grass. The forests l^^fj^l^ 
 of Guiana furnish many spp-" i of hard wood, tough and 
 elastic, out of which iauti and excellent !^o\v. are 
 formed. 
 
 The arrows are from four to five feet in length, made of 
 a yellow reed without a knot or joint. It is found in great 
 plenty up and down throughout Guiana. A piece of hard 
 wood, about nine inches long, is inserted into the end of 
 the reed, and fastened with cotton well waxed. A square 
 hole, an inch deep, is then made in the end of this piece of 
 hard wood, done tight round with cotton to keep it from 
 splitting. Into this square hole is fitted a spike of Cou- 
 courite wood, poisoned, and which may be kept there, or 
 taken out at pleasure. A joint of bamboo, about as thick 
 as your finger, is fitted on over the poisoned spike, to 
 prevent accidents and defend it from the rain, and is 
 taken off when the arrow is about to be used. Lastly, two 
 feathers are fastened on the other end of the reed to steady 
 it in its flight. 
 
 Besides his bow and arrows, the Indian carries a little 
 box made of bamboo, which holds a dozen or fifteen 
 poisoned spikes, six inches long. They are poisoned in the Spikes. 
 following manner : a small piece of wood is dipped in the 
 poison, and with this they give the spike a first coat. It 
 is then exposed to the sun or fire. After it is dry, it 
 receives another coat, and is then dried again ; after this 
 a third coat, and sometimes a fourth. 
 
 They take great care to put the poison on thicker at the 
 middle than at the sides, by which means the spike retains 
 
 
1»8 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 
 ■m 
 
 FIRST the shape of a two-edged sword. It is rather a tedious 
 
 jo uRNE T. operation to make one of these arrows complete ; and as 
 
 the Indian is not famed for industry, except when pressed 
 
 by hunger, he has hit upon a plan of preserving his arrows 
 
 which deserves notice. 
 
 About a quarter of an inch above the part where the 
 Coucourite spike is fixed into the square hole, he cuts it 
 half through ; and thus, when it has entered the animal, 
 the weight of the arrow causes it to break off there, by 
 which means the arrow falls to the ground uninjured ; so 
 that, should this be the only arrow he happens to have 
 with him, and should another shot immediately occur, 
 he has only to take another poisoned spike out of his 
 little bamboo box, fit it on his arrow, and send it to its 
 destination. 
 
 Thus armed with deadly poison, and hungry as the 
 hysena, he ranges through the forest in quest of the wild 
 beasts' track. No hound can act a surer part. Without 
 clothes to fetter him, or shoes to bind his feet, he observes 
 the footsteps of the game, where an European eye could 
 not discern the smallest vestige. He pursues it through 
 all its turns and windings, with astonishing perseverance, 
 and success generally crowns his efforts. The animal, after 
 receiving the poisoned arrow, seldom retreats two hundred 
 paces before it drops. 
 
 In passing overland from the Essequibo to the Demerara 
 we feU in with a herd of wild hogs. Though encumbered 
 with baggage, and fatigued with a hard day's walk, an 
 Indian got his bow ready, and let fly a poisoned arrow at 
 one of them. It entered the cheek-bone and broke off. 
 The wild hog was found quite dead about one hundred and 
 seventy paces from the place where he had been shot. He 
 afforded us an excellent and wholesome supper. 
 
 Thus the savage of Guiana, independent of the common 
 
 Kill a 
 wild hog, 
 
WANDERINGS IN SOUTH AMERICA. 
 
 139 
 
 weapons of destruction, has it in his power to prepare a first 
 poison, by which he can generally ensure to himself a JotmNEv, 
 supply of animal food ; and the food so destroyed imbibes 
 no deleterious qualities. Nature has been bountiful to him. 
 She lias not only ordered poisonous herbs and roots to grow 
 in the unbounded forests through which he strays, but has 
 also furnished an excellent reed for his arrows, and another, 
 still more singular, for his blow-pipe ; and planted trees of 
 an amazing hard, tough, and elastic texture, out of which 
 he forms his bows. And in order that nothing might be 
 wanting, she has superadded a tree which yields him 
 a fine wax, and disseminated up and down, a plant not 
 unlike that of the pine-apple, which affords him capitpl 
 bow-strings. 
 
 mon 
 
!■'• 
 
 CHAPTER III. 
 
 operation of the Wourali. — Its effects on the Ai, or three-toetl Sloth. — 
 Death resembling sleep. — A poisoned Ox, — Poison proportionate to 
 size of animal. — Alleged antidotes. — An Indian killed by hia own 
 arrow. — Ligatures and the knife. — Descent of the Essequibo. — Skill 
 of the boatman. — The Buccaneers. — Tertian ague. — Experiments with 
 Wourali. — Value of a ligature. — Artificial respiration. — Long life and 
 quiet death of Wouralia. — When good King Arthur ruled this land. — 
 Return of health. 
 
 ^!! i; 
 
 FIRST 
 JOURNEY. 
 
 Further 
 remarks 
 
 on the 
 virulence 
 
 of the 
 poison. 
 
 Having now followed the Indian in the chase, and de- 
 scribed the poison, let us take a nearer view of its action, 
 and observe a large animal expiring under the weight of 
 its baneful virulence. 
 
 Many have doubted the strength of the wourali-poison. 
 Should they ever by chance read what follows, probably 
 their doubts on that score will be settled for ever. 
 
 In the former experiment on the hog, some faint resist- 
 ance on the part of nature was observed, as if existence 
 struggled for superiority ; but in the following instance of 
 the sloth, life sank in death without the least apparent 
 contention, without a cry, without a struggle, and without 
 a groan. This was an Ai, or three-toed Sloth. It was in 
 the possession of a gentleman who was collecting curiosities. 
 He wished to have it killed, in order to preserve the skin, 
 and the wourali-poison was resorted to as the easiest 
 death. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 141 
 
 Of all animals, not even the toad and tortoise excepted, fiust 
 this poor ill-formed creature is the most tenacious of life. JQ ^'^'^'^ ^''- 
 It exists long after it has received wounds which would have 
 destroyed any other animal ; and it may bo said, on seeing 
 a mortally-wounded sloth, that life disputes with death 
 every inch of flesh in its body. 
 
 The Ai was wounded in the leg, and put down on the 
 floor, about two feet from the table ; it contrived to 
 reach the leg of the table and fastened himself on it, as if 
 wishful to ascend. But this was its last advancing step : 
 life was ebbing fast, though imperceptibly ; nor could this 
 singular production of nature, which has been formed of 
 a texture to resist death in a thousand shapes, make any 
 stand against the wourali-poison. 
 
 First, one fore-leg let go its hold, and dropped down 
 motionless by its side ; the other gradually did the same. 
 The fore-legs having now lost their strength, the sloth 
 slowly doubled its body, and placed its head betwixt its 
 hind-legs, which still adhered to the table ; but when the 
 poison had affected these also, it sank to the ground, but 
 sank so gently, that you could not distinguish the move- 
 ment from an ordinary motion ; and had you been ignorant 
 that it was wounded with a poisoned arrow, you would 
 never have suspected that it was dying. Its mouth was 
 shut, nor had any froth or saliva collected there. 
 
 There was no subsultus tendinum, or any visible altera- 
 tion in its breathing. During the tenth minute from the 
 time it was wounded it stirred, and that was all ; and the 
 minute after, life's last spark went out. From the time the 
 poison began to operate, you would have conjectured that 
 sleep was overpowering it, and you would have exclaimed, 
 " Pressitque jacentem, dulcis et alta quies, placidaeque 
 simillima morti." 
 
 There are now two positive proofs of the effect of this 
 
142 
 
 VVANPRRINGS IN SOUTH AMKRTCA. 
 
 ii :l i 
 
 ■M 
 
 KiiiHT fatal poison ; viz. the death of tlio hoi,', and that of tlie sloth. 
 
 joruNEY. j^jj^ g^m tijesQ animals wero nothing reniarkahle for size; 
 
 and the atrongth of tlie poison in large animals might 
 
 yet be doubted, were it not for what follows. 
 
 Ex}yri- A large well-fed ox, from nine hundred to a thousand 
 
 "^im Ox°"^ pounds weight, was tied to a stake by a roi)e sufliciently 
 
 long to allow him to move to and fro. Having no large 
 
 Coucourito spikes at hand, it was judged necessary, on 
 
 account of his superior size, to put three wild-hog arrows 
 
 into him; one was sent into each thigh just above the 
 
 hock, in order to avoid wounding a vital part, and the 
 
 third was shot traver.sJy into the extremity of the nostril. 
 
 The poison seemed to take effect in four minutes. Con- 
 scious as though he would fall, the ox set himself firmly 
 on his legs, rnd remained quite still in the same place, 
 till about the fourteenth minute, when he sraelled the 
 ground, and appeared as if inclined to walk. He advanced 
 a pace or two, staggered, and fell, and remained extended 
 on his side with his head on the ground. Ilis eye, a few 
 minutes ago so bright and lively, now became fixed and 
 dim, and though you put your hand close to it as if to 
 give him a blow there, he never closed his eyelid. 
 
 His legs were convulsed, ana .his head from time to time 
 started involuntarily ; but he never showed the least desire 
 to raise it from the ground ; he breathed hard, and emitted 
 foam from his mouth. The startings, or subsultus tendi- 
 num, now became gradually weaker and weaker; his 
 hinder parts were fixed in death ; and in a minute or two 
 more his head and fore-legs ceased to stir. 
 
 Nothing now remained to show that life was still within 
 him, except that his heart faintly beat and fluttered at 
 intervals. In five-and-twenty minutes from the time of 
 his being wounded he was quite dead. His flesh was very 
 sweet and savoury at dinner. 
 
 
WAX[)Emxr,s ix south America. 
 
 143 
 
 to 
 
 of 
 My 
 
 rriisT 
 
 JOCUNKY. 
 
 Ooieviil 
 
 obsnrni- 
 
 lions. 
 
 On taking a retrospectivo vunv of the two dillbreut 
 kinds of poisoned arrows, and tlie animals destroyed by 
 them, it wouUl appear that the quantity of poison must 
 he proportioned to the animal, and thus those probably 
 labour under an error who imagine that tlie smallest 
 particle of it introduced into the blood has almost instan- 
 taneous effects. 
 
 Make an estimate of the difference in size betwixt the 
 fowl and the ox, and then weigh a sufficient quantity of 
 poison for a blow-pipe arrow with which the fowl was 
 killed, and weigh also enough poison for three Wild-hog , 
 arrows which destroyed the ox, and it will appear that the 
 fowl received much more poison in proportion than the 
 ox. Hence the cause why the fowl died in five minutes, 
 and the ox in five-and-twenty. 
 
 Indeed, were it the case that the smallest particle of it 
 introduced into the blood has almost instantaneous effects, 
 the Indian would not find it necessary to make the large 
 arrow; that of the blow-pipe is much easier made and 
 requires less poison. 
 
 And now for the antidotes, or rather the supposed anti- 
 dotes. The Indians tell you, that if the wounded animal -Antidotes 
 be held for a considerable time up to the mouth in water, 
 the poison will not prove fatal; also that the juice of the 
 sugar-cane poured down the throat will counteract the 
 effects of it. These antidotes were fairly tried upon full- 
 grown healthy fowls, but they all died, as though no steps 
 had been taken to preserve their lives. Kum was recom- 
 mended and given to another, but with as little success. 
 
 It is supposed by some, that wind intri;duced into the 
 lungs by means of a small pair of bellows would revive 
 the poisoned patient, provided the operation be continued 
 for a sufficient length of time. It may be so ; but this is 
 a difficult and a tedious mode of cure, and he who is 
 
144 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FIRST 
 JOURNKY. 
 
 !! 'H 
 
 m ;Si'! 
 
 wounded in the forest far away from his friends, or in 
 the hut of the savages, stands but a poor chance of being 
 saved by it. 
 
 Had the Indians a sure antido ' -3, it is likely they would 
 carry it about with them, or resort to it immediately after 
 being wounded, if at hand ; and their confidence in its 
 efficacy would greatly diminish the horror they betray 
 when you point a poisoned arrow at them. 
 
 One day, while we were eating a red monkey, erroneously 
 called the baboon in Demerara, an Arowack Indian told 
 an affecting story of what happened to a comrade of his. 
 He was present at his death. As it did not interest this 
 Indian in any point to tell a falsehood, it is very probable 
 that his account was a true one. If so, it appears that 
 there is no certain antidote, or at least an antidote that 
 could be resorted to in a case of urgent need ; for the Indian 
 gave up all thoughts of life as soon as he was wounded. 
 Anecdote. The Arowack Indian said it was but four years ago that 
 he and his companion were ranging in the forest in quest 
 of game. His companion took a poisoned arrow, and sent 
 it at a red monkey in a tree above him. It was nearly a 
 perpendicular shot. The arrow missed the monkey, a^id 
 in the descent, struck him in the arm, a little above the 
 elbow. He was convinced it was all over with him. " I 
 shall never," said he to his companion in a faltering voice 
 and looking at his bow as said it, " I shall never," said he, 
 *' bend this bow again." And having said that, he took 
 off his little bamboo poison-box, which hung across his 
 shoulder, and putting it together with his bow and arrows 
 on the ground, he laid himself down close by them, bid 
 his companion farewell, and never spoke more. 
 
 He who is unfortunate enough to be wounded by a 
 poisoned arrow from Macoushia had better not depend 
 upon the common antidotes for a cura Many who have 
 
 ? ; 
 
WANDKUINGS IN SOUTH AMEPJCA. 
 
 lio 
 
 been in Guiana will recommend immediate immersion in fuist 
 water, or to take the juice of the sugar-cane, or to fill the JQ ^'^^''- ^'- 
 mouth full of salt; and they recommend these antidotes 
 because they have got them from the Indians. But were you 
 to ask them if tliey ever saw these antidotes used with suc- 
 cess, it is ten to one their answer would be in the negative. 
 Wherefore let him reject these antidotes as unprofitable, 
 and of no avail. He has got an active and deadly foe 
 within him, which, like Shakspeare's fell Sergeant Death, 
 is strict in his arrest, and will allow him but little time — 
 very — very little time. In a few minutes he will be 
 numbered with the dead. Life ought, if possible, to be 
 preserved, be the expense ever so great. Should the part 
 affected admit of it, let a ligature be tied tight round the 
 wound, and have immediate recourse to the knife : — 
 
 " Continuo, culpam ferro compesce pnusqiiam, 
 Diva per infaustuin serpant cortagia corpus." 
 
 And now, kind reader, it is time to bid thee farewell. 
 The two ends proposed have been obtained. The Portu- 
 guese inland frontier fort has been reached, and the 
 Macoushi wourali-poison acquired. The account of this 
 excursion through the interior of Guiana has been sub- 
 mitted to thy perusal, in order to induce thy abler genius 
 to undertake a more extensive one. If any difficulties 
 have arisen, or fevers come on, they have been caused by 
 the periodical rains, which fall in torrents as the sun 
 approaches the tropic of Cancer. In dry weather there 
 would be no difficulties or sickness. 
 
 Amongst the many satisfactory conclusions which thou 
 wouldest be able to draw during the journey, there is one 
 which, perhaps, would please thee not a little ; and that 
 is with regard to dogs. Many a time, no doubt, thou hast 
 heard it hotly disputed, that dogs existed in Guiana 
 
U6 
 
 WANDElilXGS IX SOUTH AMEHICA. 
 
 I ;! 
 
 l.n 
 
 1 I 
 
 FIRST 
 JOUr.XLY. 
 
 Politics. 
 
 pvoviously to the arrival of tlic Spaniards in those parts. 
 Wliatever tlie Spaniards introduced, and wliieh bore no re- 
 semUance to anything the Indians liad Lecn accustomed 
 to see, retains its Spanish name to this day. 
 
 Thns the Warow, the Arowack, the Acoway, the IVIa- 
 coushi, and Carib tribes, call a hat, sombrero; a shirt, or 
 any kind of cloth, camiso ; a shoe, zapato ; a letter, carta ; 
 a fowl, galliiia; gunpowder, colvora, (Spanish, polvora;) 
 ammunition, bala ; a cow, vaca ; and a dog, perro. 
 
 This argues strongly against the existence of dogs in 
 Guiana before it was discovered by the Spaniards, and 
 probably may be of use to thee, in thy next canine dispute- 
 In a ])olitical point of view this country presents a 
 large field for speculation. A few years ago there was but 
 little inducement for any Englishman to explore the in- 
 terior of tliese rich and fine colonies, as the liritish 
 Government did not consider them worth holding at the 
 peace of Amiens. Since that period their mother-country 
 has been Idotted out from the list of nations, and America 
 has unfolded a new sheet of politics. On one side, the 
 crown of l>ragan7;a, attacked by an ambitious chieftain, 
 has fied from tlio palace of its ancestors, and now seems 
 fixed on the banks of the Janeiro. Cayenne has yielded 
 to its arms. La Plata lias raised the standard of independ- 
 ence, and thinks itself sulficiently stroncr to obtain a 
 governm;^nt of its own. On the otlier side, the Caraccas 
 are in open revolt, and should Santa Fe join them in good 
 earnest, they may form a powerful association. 
 
 Thus, on each side ol the ci-devant Dutch Guiana, most 
 unexpected and astonishing changes have taken place. 
 AVill they raise or lower it in the scale of estimation at 
 the Court of St. James's ? Will they be of benefit to 
 these grand and extensive colonies ? Colonies enjoying 
 jierpetual summer. Colonies of the ricliest soil. Colonies 
 
WANDKHIXGS IN SOUTH AMERICA. 
 
 147 
 
 containing within themselves everything necessary fi>r first 
 their support. Colonies, in fine, so varied in their quality '^" '^^'"' ^' 
 and situation, as to be capable of bringing to perfection 
 every tropical production; and or»ly want the support of 
 government, and an enlightened governor, to render them 
 as fine as the finest portions of the equatorial regions. 
 Kind reader, fare thee well. 
 
 LKTTER TO TnE PORTUGUESE COMMANDER. 
 
 MuY Senor, 
 Como no tengo el honor, de ser conocido de VM. lo pienso mejor, 
 y mas decoroso, quedarme aqui, liastaque huviere recihido su res- 
 puesta. Haviendo caminado hasta la chozo, adonde estoi, no 
 quisiere volvenne, antes de liaver visto la fortaleza de los P(irtu- 
 giieses ; y pido licencia de VM. para que nie adelaiite. Honradissi- 
 mos son mis motives, ni lengo proyecto ninguno, o de coinercio, o 
 de la soldadesca, no siendo yo, o comerciante, o ofieial. Hidalgo 
 catolico soy, de liacienda in Ynglatierra, y miichos aiios de mi vida 
 he pasado en caminar. Ultimamente, de Demeraria vengo, la qual 
 dex^ el 5 dia de Abril, para ver este hermoso pais, y coger unas 
 ciiriosidades, especialmente, el vcneno, que se llama wourali. Las 
 mas recentes noticias que tenian en Demeraria, antes de mi salida, 
 eran median tristes, medias alegres. Tristes digo, viendo que 
 Valencia ha caido en poder del enemigo comun, y le General Blake, 
 y BUS valientes tropas quedan prisioneros de guerra. Alegres, al 
 contrario, porque Milord Wellington se ha opoderado de Ciudad 
 Kodrigo. A pesar de la caida de Valencia, parece claro al mundo, 
 que las cosas del enemigo, estan andando, de pejor a pojor cada dia. 
 Nosotros debemos dar gracias al AUissimo, por haver sido servido 
 dexarnos castigar ultimamente, a los robadores de sus santns Yglesias. 
 Se vera VM. que yo no escribo Portjjgues ni aun lo hablo, pero, 
 haviendo aprendido el Castellano, no nos faltara medio de coujmu- 
 nicar y tener conversacion. Ruego se escuse esta carta escrita sin 
 tinta, porque un Indio dexo caer mi tintcro y quebrose. Dios le d^ 
 a VM. muchos aiios de salud. Entretanto, tengo el honor de ser 
 
 Su mas obcdeciente servidor, 
 
 Carlos Watertox. 
 I, 2 
 
I: -f!! 
 
 ,! :, \ 
 
 
 ii:l 
 
 H Y 
 
 148 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 IIEMARKS. 
 
 " Incertus, quo fata ferant, ubi sistcre ilctur.' 
 
 li! 
 
 it 
 
 FiiisT Kind and gentle reader, if the journey in quest of the 
 
 jonivEY. -^vourali-poison has engaged thy attention, probably thou 
 
 niayest recollect that the traveller took leave of thee at 
 
 Fort St. Joachim, on the Rio Branco. Shouldest thou 
 
 wish to know what befell him afterwards, excuse the 
 
 following uninteresting narrative. 
 
 Jfhicssat Having had a return of fever, and aware that the 
 
 Fort St. farther he advanced into these wild and lonely regions, 
 
 the less would be the chance -^f regaining his health; he 
 
 lidtcrn to g^ve up all idea of proceeding onwards, and went slowly 
 
 ikmcrara. ]r)r^(.]^ towurds the Dcmerara, nearly by the same route he 
 
 had come. 
 Path of the ^^'^ descending the falls in the Essequibo, whic-> form an 
 Etisi'quibo. oblique line quite across the river, it was resolved to push 
 through them, the downward stream being in the canoe's 
 favour. At a little distance from the place, a large tree 
 had fallen into the river, and in the meantime the canoe 
 was lashed to one of its branches. 
 
 The roaring of the water was dreadful ; it foamed and 
 daslied over the rocks with a tremendous spray, like 
 breakers on a lee-shore, threatening destruction to what- 
 ever approached it. You would have thought, by the 
 confusion it caused in the river, and the whirlpools it 
 made, that Scylhi and Charybdis, and their whole progeny, 
 
WANDERINGS IN SOUTH AMERICA. 
 
 149 
 
 had left the Mediterranean, and come and settled here 
 The channel was barely twelve feet wide, and the torrent 
 in rushing down formed traverse farrows, which showed 
 how near the rocks were to the surface. 
 
 Nothing could surpass the skill of the Indian wlio 
 steered the canoe. He looked steadfastly at it, tlien at 
 the rocks, then cast an eye on the channel, and then 
 looked at the canoe again. It was in vain to speak. The 
 sound was lost in the roar of waters ; but his eye showed 
 that he had already passed it in imagination. He held up 
 his paddle in a position, as much as to say, that he would 
 keep exactly amid channel ; and then made a sign to cut 
 the bush rope that held the canoe to the fallen tree. The 
 canoe drove down the torrent with inconceivable rapidity. 
 It did not touch the rocks once all the way. The Indian 
 proved to a nicety, " medio tutissimus ibis." 
 
 Shortly after this it rained almost day and niglit, the 
 lightning flashing incessantly, and the roar of thunder 
 awful beyond expression. 
 
 The fever returned, and pressed so heavy on him, that 
 to all appearance his last day's march was over. How- 
 ever, it abated ; his spirits rallied, and he marched again • 
 and after delays and inconveniences he reached the house 
 of his worthy friend, Mr. Edmonstone, in Jlibiri Creek, 
 which falls into the Demerara. No words of his can do 
 justice to the hospitality of that gentleman, whose re- 
 peated encounters with the hostile negroes in the forest 
 have been publicly rewarded, and will be remembered in 
 the colony for years to come. 
 
 Here he learned that an eruption had taken place in 8t. 
 Vincent's ; and thus the noise heard in the night of the 
 first of May, which had caused such terror amongst the 
 Indians, and made the garrison at Fort St. Joachim re- 
 main under arras the rest of the night, is accounted for. 
 
 FIUHT 
 JOUllNKY. 
 
 Thiindt'r 
 and litjld 
 
 7IUUJ. 
 
 Fcrcr 
 returned. 
 
 Rcavhi'.s 
 Mililri 
 Cret k. 
 
150 
 
 \VANDERIN(;S IN SOUTH AMERICA. 
 
 FIRST 
 JOURNEY. 
 
 a ! 
 
 Sails f 07' 
 Granada. 
 
 St. 
 
 Thnmas's 
 
 Tovxr. 
 
 Leaves St. 
 
 Thomas's 
 
 and is 
 
 attacked by 
 
 a tertain 
 ague, and 
 
 returns to 
 
 England. 
 
 After experiencing every kindness and attention from 
 Mr. Edmonstone, he sailed for Granada, and from thence 
 to St. Thomas's, a few days before poor Captain 1: ;ake lost 
 his life on his own quarter-deck, bravely lighting for his 
 country on the coast of Guiana. 
 
 At St. Thomas's they show you a tower, a little distance 
 from the to\\'n, which they say formerly belonged to a 
 Bucanier chieftain. Probably the fury of besiegers has 
 reduced it to its present dismantled state. What still 
 remains of it bears testimony of its former strength, and 
 may brave the attack of time for centuries. You cannot 
 view its ruins without calling to mind the exploits of those 
 fierce and hardy hunters, long the terror of the western 
 world. While you admire their undaunted courage, you 
 lament that it was often stained with cruelty ; while you 
 extol their scrupulous justice to each other, jou will find 
 a want of it towards the rest of mankind. Often pos- 
 sessed of enormous wealth, often in extreme poverty, 
 often triumphant on the ocean, and often forced to fly to 
 the forests, their life was an ever-changing scene of ad- 
 vance and retreat, of glory and disorder, of luxury and 
 famine. Spain treated them as outlaws and pirates, while 
 other European powers publicly disowned them. They, on 
 the other hand, maintained that justice on the part of 
 Spain first forced them to take up arms in self-defence ; 
 and that, whilst they kept inviolable the laws which they 
 had framed for their own common benefit and protection, 
 they had a right to consider as foes those who treated 
 them as outlaws. Under this impression they drew the 
 sword, and rushed on as though in lawful war, and 
 divided the spoils of victory in the scale of justice. 
 
 After leaving St. Thomas's a severe tertian ague every 
 now and then kept putting the traveller in mind that his 
 shattered frame, " starting and shivering in the inconstant 
 
WANDERINGS IN SOUTH AMEIilCA. 
 
 151 
 
 blast, meagre and pale — the ghost of what it was" — first 
 wanted repairs. Three years elapsed after arriving in J''^''-'^''^^'- 
 England before the ague took its final leave of him. 
 
 During that time several experiments were made whh Ej-prri- 
 the wourali poison. In Lond(jn an ass was inoculated '"««/« "^ 
 
 ^ Londo)i of 
 
 the 
 Wuurali' 
 
 with it, and died in twelve minutes. The poison was 
 inserted into the leg of another, round which a bandage 
 had been previously tied a little above the place wi.jrc 
 the wourali wa?: introduced, lie walked about as usual, 
 and ate his food as though all were right. After an hour 
 had elapsed the bandage was untied, and ten minutes 
 after death overtook him. 
 
 A she-ass received the wourali poison in the shoulder, 
 and died apparently in ten minutes. An incision was 
 then made in its windpipe, and through it the lungs were 
 regularly inflated for two hours with a pair of bellows. 
 Suspended animation returned. The ass held up her head, 
 and looked around; but the inflating being discontinued, she 
 sunk once more in apparent death. The artificial breath- 
 ing was immediately recommenced, and continued without 
 intermission for two hours. This saved the ass from final 
 dissolution ; she rose up, and walked about ; she seemed 
 neither in agitation nor in pain. The wound, through 
 which the poison entered, was healed without ditUculty. 
 Her constitution, however, was so severely affected that 
 it was long a doubt if ever she would be well again. 
 She looked lean and sickly for above a year, but began 
 to mend the spring after, and by Midsummer became fat 
 and frisky. 
 
 The kind-hearted reader will rejoice on learning that 
 Earl Percy, pitying her misfortunes, sent her clown from 
 London to Walton Hall, near Wakefield. There she goes 
 by the name of Wouralia. Wouralia shall be sheltered 
 from the wintry storm ; and when summer comes she 
 
 2MiS0ll, 
 
^i"i 
 
 i:*2 
 
 VVAXDHIJIXiiS IX SOUTH AMKIiTCA. 
 
 i , 
 
 FinsT 
 
 JOURNKT. 
 
 sliall feed in the finest pasture. No burden shall be 
 placed u])on her, and she shall end lier days in peace.^ 
 
 For three revolving autumns the ague-beaten wanderer 
 never saw, without a sigh, the swallow bend her flight 
 towards warmer regions. Ilf; wished to go too, but could 
 not ; for sickness had enfeebled him, and prudence pointed 
 out the folly of roving again too soon across the northern 
 tropic. To be sure, the continent was now open, and change 
 of air might prove beneficial ; but there was nothing very 
 tempting in a trip across the Channel, and as for a tour 
 through England ! — England has long ceased to be the 
 land for adventures. Indeed, when good King Arthur re- 
 appears to claim his crown he will find things strangely 
 altered here ; and may we not look for his coming ? for 
 there is written upon his grave-stone : — 
 
 "Hie jiicet Artiiras, Rex quomlara Eexquc futurus," 
 " Hern Arthur lies, who formerly 
 Was kiiij^ — ami king again to bo." 
 
 Don Quixote was always of opinion that this famous 
 king did not die, but that he was changed into a raven by 
 enchantment, and that the English are momentarily ex- 
 pecting his return. Be this as it may, it is certain that 
 when he reigned here all was harmony and joy. The 
 browsing herds passed from vale to vale, the swains sang 
 from the bluebell-teeming groves, and nymphs with eglan- 
 tine and roses in their neatly-braided hair went hand in 
 hand to the flowery mead to weave garlands for their lamb- 
 kins. If by chance some rude uncivil fellow dared to 
 molest tlieni, or attempted to throw thorns in tlieir path, 
 there was sure to be a knight-errant not far off ready to 
 rush forward in their defence. But alas I in these de- 
 
 ' Poor Wourali breathed her last on the 15th of February, 1839, having 
 survived the ojoration nearly five-and-twenty years. 
 
WANDEIIIXGS IN SOUTH AMERICA. 
 
 153 
 
 generate days it is not so. Should a harmless cottage first 
 maid wander out of the highway to pluck a primrose or '^^ ^'^^^ ^' 
 two in the neighbouring field tl\e hauglity owner sternly 
 bids her retire ; and if a pitying swain hasten to escort 
 hei back, he is perhaps seized by the gaunt house-dog ere 
 he reach her. 
 
 ^neas's route on the other side of Styx, could not have 
 been much worse than this, though, by his account, when 
 he got back to earth, it appears that he had fallen in with 
 " Bellua Lernce, horrendum stridens, llanimisque, armata 
 Chima3ra." 
 
 Moreover, he had a sibyl to guide his steps ; and as 
 such a conductress nowadays could not be got for love 
 or money, it was judged most prudent to refrain from 
 sauntering through this land of freedom, and wait with 
 patience the return of health. At last this long-looked 
 for, ever welcome stranger came. 
 
\ I 
 
 SECOND JOniNEY. 
 
 I ' 1 
 
 CHAPTER I. 
 
 From Liverpool to rernanibuco.— Stormy petrels. — Tropical zoology, — > 
 Flying-fish. — Bonito, Albicore, and "Dolphin." — Frigate bird. — 
 Arrival at Pernambuco. — The expelled Jesuits. — Pombal, the Captain- 
 General. — Southey's history of Brazil. — Botanical garden. — Sangrcdo 
 Buey. — Eattlesnake. — Narrow escape. — Rainy. — Sail fi)r Cayenne. — 
 Shark -catching. 
 
 SECOND In the year 1816, two clays before the venial equinox, 1 
 jo uxtNE Y. gjjjigd fj.Qj^j Liverpool for Pernambuco, in tlie southern 
 
 Satis for hemisphere, on the coast of Brazil. There is little at this 
 Pernam- '■ , 
 
 buco. time of the year in the European part of the Atlantic to 
 
 engage the attention of the naturalist. As you go down 
 
 the Channel you see a few divers and gannets. The 
 
 middle-sized gulls, with a black spot at the end of the 
 
 wings, attend you a little way into the Bay of Biscay. 
 
 When it blows a hard gale of wind the storiuy petrel makes 
 
 its appearance. While the sea runs mountains high, and 
 
 every wave threatens destruction to the labouring vessel 
 
 this little harbinger of storms is seen enjoying itself, on 
 
 rapid pinion, up and down the roaring billows. When 
 
 the storm is over it appears no more. It is known to 
 
 every English sailor by the name of Mother Carey's 
 
 Chicken. It must have been hatched in -bolus's cave, 
 
 amongst a clutch of squalls and tempests ; for whenever 
 
WANDEHINUS IN SUl'TII AMElilCA. 
 
 IM 
 
 they get out upon the ocean it always contrives to bo of 
 the party. 
 
 Though tlie cahiis and storms, and adverse winds in 
 these kititudes are vexatious, still, when you reach the 
 trade winds you are amply repaid for all disappointments 
 and inconveniences. The trade winds prevail about thirty 
 degrees on each side of the eiiuator. This part of the ocean 
 may be called the Elysian Fields of Neptune's Eii.i^iire ; and 
 the torrid zone, notwithstanding Ovid's remark, " non est 
 habitabilis sestu," is rendered healthy and pleasant by 
 these gently-blowing breezes. The ship glides smoothly 
 on, and you soon find yourself within the northern tropic. 
 When you are on it, Cancer is just over your head, and be- 
 twixt him and Capricorn is the high road of the Zodiac 
 forty-seven degrees wide, famous for Phaeton's misadven- 
 ture. His father begged and entreated him not to take it 
 into his head to drive parallel to the five zones, but to 
 mind and keep on the turni)ike which runs ol^liquely 
 across the equator. " There you will distinctly see," said 
 lie, " the ruts of my chariot wheels, ' manifesta rotie vesti- 
 gia cernes.' " But," added he, " even suppose you keep on 
 it, and avoid the by-roads, nevertheless, my dear boy, be- 
 lieve me, you will be most sadly put to your shifts ; ' ardua 
 prima via est,' the first part of the road is confoundedly 
 steep ! * ultima via prona est,' and after that it is all down 
 hill. Moreover, ' per insidias iter est, lormasque ferarum,' 
 the road is full of nooses and bull-dogs, ' Hoemoniosque 
 arcus,* and spring guns, ' sievaque circuitu, curvantem 
 brachia longo, Scorpio,' and steel traps of uncommon size 
 and shape." These were nothing in the eyes of Phaeton ; 
 go he would, so off he set, full speed, four-in-hand. He 
 had a tough drive of it ; and after doing a prodigious deal 
 of mischief, very luckily for the world, he got thrown out 
 of the box, and tumbled into the river Po. 
 
 SECOND 
 JOUllNEY. 
 
 Trade 
 winds. 
 
15(5 
 
 WANPKniXdS IN SOITII AMEUICA. 
 
 Torrid 
 Zone. 
 
 Fh/ivff- 
 Jinh. 
 
 i 
 
 
 i 
 
 HEcoND Sonic of our modern bloods have been shallow enough to 
 jouuNKY. ^j.y |.Q j^pg ^jjj^ j^Qjj^. empty-headed coachman, on a littlo 
 
 scale, making London their Zodiac. Well for them, if 
 tradesmen's bills, and other trivial perplexities, have not 
 caused them to be thrown into the King's Bench. 
 
 The productions of the torrid zone are uncommonly 
 grand. Its plains, its swamps, its savannas, and forests, 
 abound with the largest serpents and wild beasts ; and its 
 trees are the habitation of the most beautiful of tho 
 feathered race. While the traveller in the old world is 
 astonished at the elephant, the tiger, the lion, and rhino- 
 ceros, he who wanders through the torrid regions of the 
 new, is lost in admiration at the cotingas, the toucans, the 
 humming-birds, and aras. 
 
 The ocean, likewise, swarms with curiosities. Probably 
 the Flying-fish may be considered as one of the most 
 singular. This little scaled inhabitant of water and air 
 seems to have been more favoured than the rest of its 
 finny brethren. It can rise out of the waves, ai.d on wing 
 visit the domain of the birds. 
 
 After flying two or three hundred yards, the intense heat 
 of the sun has dried its pellucid wings, and it is obliged to 
 wet them in order to continue its flight. It just drops into 
 the ocean for a moment, and then rises again and flies on ; 
 and then descends to remoisten them, and then up again 
 into the air ; thus passing its life, sometimes wet, some- 
 times dry, sometimes in sunshine, and sometimes in the pale 
 moon's nightly beam, as pleasure dictates, or as need re- 
 quires. The additional assistance of wings is not thrown 
 away upon it. It has full occupation both for fins and 
 wings, as its life is in perpetual danger. 
 
 The Bonito and Albicore chase it day and night ; but the 
 Dolphin is its worst and swiftest foe. If it escape into the 
 air, the dolphin pushes on with proportional velocity 
 
 J -I 
 
WANDKItlNGS IX SOUTH AMKIITCA. 
 
 167 
 
 Ijencatli, and is ready to snap it up the moment it descends sfcond 
 to wet its winjTH. jq uunk y. 
 
 You will often see al)ovo one liundicd of these little 
 marine aerial fugitives on the wing at once. They appear 
 to use every exertion to prolong their tli;4ht, but vain are 
 all their efforts ; for when the last drop of water on their 
 wings is dried up, their flight is at an end, and they must 
 drop into the occnin. Some are instantly devoured by their 
 merciless pursuer, part escape by swimming, and others get 
 out again as quick as possible, and trust once more to their 
 
 wings. 
 
 It often happens that this unfortunate little creature 
 after alternate dips and llights, finding all its exertions of 
 no avail, at last drops on board the vessel, verifying the old 
 remark, 
 
 " lucidit in Scyllam, cupi(!iis vitaro CharylKUin." 
 
 There, stunned by the fall, it beats the deck with its tail 
 and dies. When eating it, you would take it for a fresh 
 herring. The largest measure from fourteen to fifteen 
 inches in length. The dolphin, after pursuing it to 
 the ship, sometimes forfeits his own life. 
 
 In days of yore, the musician used to play in softest, 
 sweetest strain, and then take an airing amongst the 
 dolphins ; " inter delphinas Arion." But nowadays, our 
 tars have quite capsized the custom ; and instead of riding 
 ashore on the dolphin, they invite the dolphin aboard. 
 While he is darting and playing around the vessel, a sailor 
 goes out to the spritsailyard-arm, and with a long staff, 
 leaded at one end, and armed at the other with five barbed 
 spikes, he heaves it at him. If successful in his aim, there 
 is a fresh mess for all hands. The dying dolphin affords a 
 superb and brilliant sight : 
 
 *♦ Mille traliit iiioiieus, udvcrao sole colorcs." 
 
'iliiii 
 
 111 i 
 
 It 
 
 ■ 1 5 
 1 r|'*« 
 
 iH'f' 
 
 ■I I, 
 
 iiririi 
 
 158 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 SECOND 
 JOURNEV 
 
 Frigate 
 Pelicav, 
 
 All the colours of the rainbow pass and repass in rapid 
 succession over his body, till the dark hand of death closes 
 the scene. 
 
 From the Cape de Verd islands to the coast of Brazil, 
 you see several different kinds of gulls, which probably 
 are bred in the island of St. ]*aul. Sometimes the large 
 bird called the Frigate I'elican soars majestically over the 
 vessel, and the Tropic-Bird comes near enough to let yoa 
 have a fair view of the long feathers in his tail. On the 
 line, when it is calm. Sharks of a tremendous size make 
 their appearance. They are descried from the ship by 
 means of the dorsal fin, which is above the water. 
 
 On entering the bay of Pernambuco, the Frigate Pelican 
 is seen watching the shoals of fish from a prodigious height 
 It seldom descends Mithout a successful attack on its nume- 
 rous prey below. 
 Scenery. As you approach the shore, the view is charming. The 
 hills are clothed with wood, gradually rising towards the 
 interior, none of them of any considerable height. A 
 singular reef of rocks runs parallel to the coast, and forms 
 the harbour of Pernambuco. The vessels are moored be- 
 twixt it and the town, safe from every storm. You enter 
 the harbour through a very narrow passage, close by a fort 
 built on the reef. The hill of Olinda, studded with houses 
 and convents, is on your right hand, and an island thickly 
 planted with cocoa-nut trees, adds considerably to the 
 scene on your left. There are two strong forts on the 
 isthmus, betwixt Olinda and Pernambuco, and a pillar 
 midway to aid the pilot. 
 
 Pernambuco probably contains upwards of fifty thou- 
 sand souls. It stands on a flat, and is divided into three 
 parts — a peninsula, an island, and the continent. Though 
 within a few degrees of the line, its climate is remark- 
 ably salubrious, and rendered almost temperate by the 
 
 Pcrnam 
 buco. 
 
WANDERINGS IN SOUTH AMEKICA. 
 
 159 
 
 refreshing seu "breeze. Had art and judgment contril)uted skcokp 
 their portion to its natural advantages, Pernambuco at this ''ournkv. 
 day, would have been a stately ornament to the coast of 
 Brazil, On viewing it, it will strike you that every one has 
 built his house entirely for himself, and deprived public 
 convenience of the little claim she had a right to put in. 
 You would wish that this city, so famous for its harbour, 
 so happy in its climate, and so well situated for commerce, 
 could have risen under the flag of Dido, in lieu of that of 
 Braganza. 
 
 As you walk down the streets, the appearance of the streets and 
 houses is not nnich in their favour. Some of them are ^°^^^' 
 very high, and some very low ; some r.ewly whitewashed, 
 and others stained, and mouldy, and neglected, as though 
 they had no owner. 
 
 The balconies, too, are of a dark ar d gloomy appearance. 
 
 They are not, in general, open, as in most tropical cities, 
 
 but grated like a farmer's dairy window, though somewhat 
 
 closer. 
 
 There is a lamentable want of cleanliness in the streets. 
 
 Tlie impurities from .'lie houses, and the accumulation of 
 
 litter from the beasts of burden, are unpleasant sights to 
 
 the passing stranger. He laments the want of a police as 
 
 he goes along ; and when the wind begins to blow, his nose 
 
 and eyes are too often exposed to a cloud of very unsavoury 
 
 dust. 
 
 When you view the port of rernanibuco, full of ships of Port of 
 
 all nations, when you know that the richest commodities of luco, 
 
 Europe, Africa, and Asia, are brought to it; when you see 
 
 immense quantities of cotton, dye-wood, and the choicest 
 
 fruits pouring into the town, you are apt to wonder at the 
 
 little attention these people pay to the common comforts 
 
 which one always expects to find in a large and opulent 
 
 city. However, if the inhiibitnnts are satisfied, there is 
 
160 
 
 WANDERINGS IN SOUTH AMKIilCA. 
 
 SECOND 
 
 'i-H. 
 
 Palace of 
 
 tlie 
 Captain- 
 
 nothing more to be said. Sliould tliey ever be convinced 
 jouKNKv. ^Ijq^ij inconveniences exist, and tliat nuisances are too fre- 
 quent, the remedy is in their own hands. At present, cer- 
 tainly, they seem perfectly regardless of them ; and the Cap- 
 tain-General of Pernambuco walks through the streets with 
 as appareuc content and composure as an English statesman 
 would proceed down Charing-cross. Custom reconciles 
 everything. In a week or two the stranger himself begins 
 to feel less the things which annoyed hiui so much upon 
 his first arrival, and after a few months' residence, he thinks 
 no more about them, while he is partaking of the hospi- 
 tality, and enjoying the elegance and splendour within 
 doors in this great city. 
 
 Close by the river-side stands what is called the palace 
 of the Captain-General of I'ernambuco. Its form and ap- 
 Gewrai. pearance altogether strike the traveller that it was never 
 intended for the use it is at present put to. 
 
 Header, throw a veil over thy recollection for a little 
 
 while, and forget the cruel, unjust, and unmerited censures 
 
 tliou hast heard against an unoffending order. This palace 
 
 was once the Jesuits' college, and originally built by those 
 
 D'-sstruc- charitable fathers. Ask the aged and respectable inhabit- 
 
 'f,'" .f ^if ants of rernambuco, and they will tell thee tliat the de- 
 
 K)iici('ty of ' ■' 
 
 struction of the Society of Jesus was a terrible disaster 
 to the public, and its consequences severely felt to the 
 present day, 
 
 When Pombal took the reins of pov/er into his own 
 hands, virtue and learning beamed bright within the col- 
 lege walls. Public catechism to the children, and religious 
 instruction to all, flowed daily from the mouths of its 
 venerable priests. 
 
 They were loved, revered, and respected throughout the 
 whole town. The illuminating philosophers of the day had 
 sworn to exterminate Christian knowledge, and the college ■ 
 
 Jesus. 
 
WANDERINGS IN SOUTH AJ^IEKICA. 
 
 ir.i 
 
 of Pernambuco Avas doomed to founder in the general skcoxd 
 storm. To the long-lasting sorrow and disgrace of I'or- '^^^"^"^'^' 
 tugal, the philosophers blinded her king and flattered her 
 prime minister, Ponibal was exactly the tool tliese sappers 
 of every public and private virtue wanted. He liad the 
 naked sword of power in his own hand, and his heart was as 
 liard as flint. He struck a mortal blow, and the Society of 
 Jesus, throughout the Portuguese dominions, was no more. 
 
 One morning all the fathers of the college in Pernam- 
 buco, some of them very old and feeble, were suddenly 
 ordered into the refectory. They had notice beforeliand of 
 the fatal storm, in pity from the governor, but not one of 
 tliera abandoned liis charge. They had done tlieir duty, 
 and liad nothing to fear. They bowed with resignation to 
 the will of Heaven. As soon as they had all reached the 
 refectory, they were there locked up, and never more did 
 they see their rooms, their friends, their scholars, or ac- 
 quaintance. In the dead of the following night, a strong 
 guard of soldiers literally drove them through the streets 
 to the water's edge. They were then conveyed in boats 
 aboard a ship, and steered for Bahia. Those who sur- 
 vived the barbarous treatment they experienced from 
 Pombal's creatures were at last ordered to Lisb'"!). The 
 college of Pernambuco was plundered, and some t nic after 
 an elephant was kept there. 
 
 Thus the arbitrary hand of power, in one night, smct*". 
 and swept away the sciences ; to which succeeded the low 
 vulgar buffoonery of a showman. Virgil and Cicero made 
 way for a wild beast from Angola ! and now a guard is on 
 duty at the very gate where, in times long past, the poor 
 were daily fed 1 ! ! 
 
 Trust not, kind reader, to the envious remarks which 
 their enemies have scattered far and near ; believe not the 
 stories of those who have had a hand in the sad tragedy 
 
 M 
 
it 
 
 i '-I! i; 
 
 ! I 
 
 r 
 
 
 1 
 
 1 
 
 ^ 
 
 1 
 
 
 1 
 
 
 Bj 
 
 
 in 
 
 
 1 
 
 '1 
 
 162 
 
 WANDEHINCJS IN SOUTH AMEUICA. 
 
 sRcoND Go to liivuil, and soo with thine own eyes the effect of 
 joi^u^,\. p^jmij.Q'g short-sighted policy. Tliere vice reigns triuni- 
 pliant, and learning is at its lowest ebb. Neither is this to 
 be wondered at. Destroy the compass, and will the vessel 
 find her far-distant port ? Will the Hock keep togethei", 
 and escape the wolves, after the shepherds are all slain ? 
 Tlie lira/cilians were told tliat public education would go 
 on just as usual. They might have asked government 
 who so able to instruct our youth as those whose know- 
 ledge is proverbial i who so lit, as those who enjoy our 
 entire conHdeiice ? who so worthy, as those whose lives are 
 irreproachable. 
 
 They soon ibund tliat those who succeeded the Mliers 
 of the Society of Jesus had neither their manners nor their 
 abilities. They liad not made the instruction of youth 
 their i)articular study. ^Moreover, they entered on the 
 field after a defeat, where the olHcers had all been slain ; 
 where the plan of the campaign was lost ; where all was 
 in sorrow and dismay. No exertions of theirs could rally 
 the dispersed, or skill prevent the fatal consequences. At 
 the present day, the seminary of Olinda, in comparison 
 with the former Jesuits' college, is only as the waning 
 moon's beam to the sun's meridian splendour. 
 
 When you visit the places where those learned fathers 
 once flourished, and see with your own eyes the evils their 
 dissolution has caused ; when you hear the inhabitants 
 telling you how good, how clever, how charitable they 
 were; what will you think of our poet laureate for calling 
 them, in his "History of Brazil," " ]Missioners, whose zeal 
 the most fanatical was directed by the coolest policy " ? 
 
 Was it /«?w^im/ to renounce the honours and comforts 
 of this transitory life, in order to gain eternal glory in the 
 next, by denying themselves, and taking up the cross ? 
 Was it fa naUcal to preach salvation to innumerable wild 
 
 L*i^ 
 
WANDEiUNaS IN SOUTH AMERICA. 
 
 ir,3 
 
 hordes of Americans ? to clothe the naked ? to encourage 
 the repenting sinner? to aid tlie dying Christian. The 
 fathers of the Society of Jesus did all this. And for this 
 their zeal is pronounced to be the most fanatical, directed 
 hy the coolest policy. It will puzzle many a clear brain to 
 comprehend how it is possible, in the nature of things, 
 that zeal the most fanatical should be directed by the 
 coded policy. Ah, ]\ir. Laureate, ^Ir. Laureate, that 
 «' quidlibet audendi " of yours, may now and then gild the 
 poet, at the same time that it makes the historian cut a 
 sorry figure ! 
 
 Could Father Nobrega rise from the tomb, he would 
 thus address you : — " Ungrateful Englishman, you have 
 drawn a great part of your information from the writings 
 of the 'Society of Jesus, and in return you attempt to 
 stain its character by telling your countrymen that * we 
 taught the idolatry we believed ! ' In speaking of me, you 
 say, it was my happy fortune to be stationed in a country 
 where none but the good principles of my order were called 
 into action. Ungenerous laureate, the narrow policy of the 
 times has kept your countrymen in the dark with regard to 
 the true character of the Society of Jesus ; and you draw 
 the bandage still tighter over their eyes by a malicious in- 
 sinuation. I lived, and taught, and died in lirazil, where 
 you state tlk.: none but the good principles of my order 
 were called into action, and still, in most absolute. contra- 
 dic'iion to this, you remark we believed the idolatr;/ we 
 taught in Brazil. Thus we brought none but good prin- 
 ciples into action, and still taught idolatry ! 
 
 " Again, you state there is no individual to whose talents 
 Brazil m so greatly and permanently indebted as mine, and 
 that I must be regarded as the founder of that system so 
 successfully pursued by the Jesuits in Paraguay ; a system 
 productive of as much good as is compatible with pious 
 
 M '1 
 
 SKCONI) 
 JorUNKY. 
 
164 
 
 WANDERINGS IN SOUTH AMKRfCA. 
 
 :u 
 
 ! II 
 ii 1 1 
 
 BFXGND fraud. Thus you make me, at one and tlie same time, a 
 J OURNE Y. j;eacher of none but good principles, and a teacher of 
 idolatry, and a believer in idolatry, and still the founder of 
 a system for which Brazil is greatly and permanently in- 
 debted to me, though, by the by, the system was only pro- 
 ductive of as much good as is compatible with pious 
 fraud ! 
 
 " What means all this ? After reading such incom- 
 parable nonsense, should your countrymen wish to be 
 properly informed concerning the Society of Jesus, there 
 010 in Ent'land documents enoudi to show that the svstem 
 <>r he Jesuits was a system of Christian charity towards 
 then" fellow -creatures, administered in a manner which 
 human prudence judged best calculated to ensure success, 
 •dU'\ that the idolatry which you uncharitably affirm they 
 • t&uviit, was really and truly the very same faith which the 
 Catholic Church taught for centuries in England, which she 
 still teaches to those who wish to hear her, and which she 
 will continue to teach, pure and unspotted, till time shall 
 be no more." 
 
 The environs of Pernambuco are very pretty. You see 
 country houses in all directions, a?id the appearance of here 
 and there a sugar plantation enriches the scenery. Palm- 
 trees, Cocoa-nut-trees, Orange and Lemon groves, and 
 all the different fruits peculiar to Brazil, are here in the 
 greatest abuudance. 
 
 At Olinda there is a national botanical garden ; it wants 
 space, produce, and impi remnio. The forests, which are 
 several leagues off, abound with birds, beists, insects, and 
 serpents. Besides a brill isnt plumage, many of the birds 
 have a very fine song. The Trou])iale, noted for its rich 
 colours, sings deliglitlully in the environs of Pernambuco.' 
 The Ked-headed Finch, larger than the European sparrow, 
 pours forth a sweet and varied strain, in comi)any with two 
 
 Environs 
 
 of Pe.nmm 
 
 buco. 
 
WAN'[)KI{IX(iS IN SOUTH AMEUICA. 
 
 1R6 
 
 species of wrens, a little before daylight. There are also skcokp 
 several species of the thrush, which have a song somewhat Jou"^^'- 
 different from that of the European thrush ; and two 
 species of the linnet, whose strain is so soft and sweet that 
 it dooms them to captivity in th^ houses, A bird called 
 here Sangre do Buey, blood of the ox, cannot fail to en- 
 gage your attention : he is of the passerine tribe, and very 
 common about .the houses; the wings and tail are blnck, 
 and every other part of the body ailamingred. In Guiana, 
 there is a species exactly the same as this in shape, note, 
 and economy, but different in colour, its whole body being 
 like black velvet ; on its breast a tinge of red appears 
 through the black. Thus nature has ordered this little 
 Tangara to put on mourning to the north of the line, and 
 wear scarlet to the south of it. 
 
 For three months in the year the environs of Pernam- Scnsons. 
 buco are animated beyond description. From November 
 to March the weather is particularly fine ; then it is that 
 rich and poor, young and old, foreigners find natives, all 
 issue from the city to enjoy the country, till Lent ap- 
 proaches, when back they hie them. Villages and hamlets, 
 where nothing before but rags was seen, now shine in all 
 the elegance of dress ; every house, every room, every shed 
 become eligible places for those whom nothing but extreme 
 necessity could have forced to live there a few weeks ago : 
 some join in the merry dance, others saunter up nnd down 
 the orange-groves ; and towards evening the roads become 
 a moving scene of silk and jewels. The gaming-tables 
 have constant visitors ; there, thousands are daily and 
 nightly lost and won ; parties even sit down to try their 
 luck round the outside of the door as well as in the 
 room : — 
 
 '•Vestibuli'.m ante ipsum primisque in faucibus aulre 
 Luctu'j ;., ul trices, posuere sedilia curie," 
 
 I 
 
 
ir.r, 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 M 
 
 I ill 
 
 KKcoND About six or seven miles from rernambuco stands a 
 
 jo uiiNK Y. p|.gtty little village called Monteiro ; the river runs close 
 
 Monteiro. jjy jf;^ ^nd its rural beauties seem to surpass all others in 
 
 the neighbourhood ; there the Captain-General of Pernam- 
 
 buco resides during this time of merriment and joy. 
 
 The traveller who allots a portion of his time to peep at 
 his fellow-creatures in their relaxations, and accustoms 
 himself to read their several little histories in their looks 
 and gestures as he goes musing on, may have full occupa- 
 tion for an hour or two every day at this season amid the 
 variegated scenes aronnd the pretty village of ]\Ionteiro. 
 In the evening groups sitting at the door, he may some- 
 times see with a sigh how wealth and the prince's favour 
 cause a booby to pass for a Solon, and be reverenced as 
 such, while perhaps a poor neglected Camoens stands silent 
 at a distance, awed by the dazzling glare of wealth and 
 power. Eetired from the public road he may see poor 
 Maria sitting under a palm-tree, with her elbow in her lap, 
 and her head leaning on one side within her hand, weeping 
 over her forbidden bans. And as he moves on "with 
 wandering step and slow," he may hear a broken-hearted 
 nymph ask her faithless swain, — 
 
 " How conlil yon say my fnoo was fair, 
 Aiul yet that face forsake ? 
 How could you win my viifjin licnil. 
 Yet leave that heart to Itieak i"' 
 
 One afternoon, in an inifrcquented part not far from 
 Monteiro, these adventures were near being brought to a 
 speedy and a final close : six or seven blackbirds, with a 
 white spot betwixt the shoulders, were making a noise, and 
 passing to and fro on the lower branches of a tree in an 
 abandoned, weed-grown, orange orchard. In the long grass 
 iniderneath the tree, apparently a pale green grasshopper 
 
 SB 
 
WAXDEinNGS IX SOUTH AMEIIICA. 
 
 I(i7 
 
 was fluttering, as though it had got entaiif';lcd in it. AVhon 
 you once fancy that the thing you are looking at is really 
 ^vhat you take it for, the more you look at it the more you 
 are convinced it is so. In the present case, this was a 
 grasshopper beyond all douht, and nothing more remained 
 to be done but to wait in patience till it had settled, in 
 order that you might run no risk of breaking its legs in 
 attempting to lay hohl of it while it was fluttering — it still 
 kept fluttering ; and having quietly api)roached it, intend- 
 ing to make sure of it — behold, the head of a large rattle- 
 snake appeared in the grass close by : an instantaneous 
 spring backwards prevented fatal consequences. What 
 had been taken for a grasshopper was, in fact, the elevated 
 rattle of the snake in the act of announcing that he was 
 quite prepared, though unwilling, to make a sure and 
 deadly spring. He shortly after passed slowly iVom under 
 the orange-tree to the neighbouring wood on the side of a 
 hill : as he moved over a ]jlace bare of grass and weeds, he 
 ap])earcd to be about eight feet long ; it w.'is he who had 
 enuajied the attention of the birds, and made thein heedless 
 of danger from another ouarter: they Hew away on his re- 
 tiring ; one alone left his little life in the air, destined to 
 become a specimen, mute and motionless, for the inspection 
 of the curious in a far distant clime. 
 
 It was now the rainy season ; the birds were moulting ; 
 fifty-eight specimens of the handsomest of them in the 
 ntnghbourhood of Pernamlmco had been collected; audit 
 was time to proceed elsewhere. Tiie conveyance to the 
 interior was by horses ; and this mode, together with the 
 heavy rains, would expose preserved specimens to almost 
 certain damage. The journey to jNlaranham by land would 
 take at least forty days. The route was not wild enough 
 to engage the attention of an explorer, or civilized enough 
 to afford common comforts to a traveller. V>y sea there 
 
 SKCDNT) 
 .TOIM'.NEY. 
 
 lidiin/ 
 Si ilSOilS, 
 
IM 
 
 WANDEllINGS IN SOUTH AMERICA. 
 
 lii 
 
 :»! i 
 
 fiKcoNi) ^ycvQ no opportuiiitius, except slave ships. As tlie traiis- 
 jo uuNK v. porting poor negroes from port to port for sale pays well in 
 J'ra/il, tlie ships' decks are crowded with them. ThiS 
 would not do. 
 
 Excuse here, benevolent reader, a small tribute of grati- 
 tude to an Irish I'aniily, whose urbanity and goodness have 
 long gained it the esteem and respect of all ranks in 
 Pernambuco. The kindness and attention I receiveil 
 from Dennis Kearney, Esq., and his amiable lady, will 
 be remembered with gratitude to my dying day. 
 
 After wishing fareW''3ll to this hospitable family, I em- 
 barked on board a Tortuguese brig, with poor accommo- 
 dation, for Cayenne in Guiana. The most eligible bed- 
 room was the top of a hen-coop on deck. Even here, an 
 unsavoury little beast, called bug, was neither shy nor 
 deficient in appetite. 
 
 The Portuguese seamen are famed for catching fish. One 
 evening, under the line, four sharks made their appearance 
 in the wake of the vusrool. The sailors caught them all. 
 
 Emhar^'s 
 
 for 
 Cajennc. 
 
 )i 
 
s 
 
 CILVPTER IT. 
 
 Arrival at Cayenne. — Flaniinj^os. — Curlews, &c'.— V cjotaMe productions 
 of Cayenne. — La Galiriellc. — Cock of tlio Hock. 'ul Oolie-niouclie. 
 
 — Snrinan). — The Coryntin. — Xuw Anistenl Stabroek, now 
 
 George Town. — Produce of Donierarii. — Slavery. v travtller's neces- 
 saries — Walking barelbot. — The best costume.- -Humming-birds. — 
 Cotinga. — Caniiianero, or Dell-bird. — Toucans, or Toucanets. — Beak 
 of the Toucan — Evanescence of tlio colours — The only mode of 
 preserving them. 
 
 On the fouvteentli day after leaving Pernaiuhuco, the sKomn 
 brig cast anchor oIT the island of Cayenne. The entrance '^"^''^^"^' 
 is beautiful. To windward, not far off, there are two bold 
 wooded islands, called the Father and JNIother ; and nenr 
 them are others, their children, smaller, though as beautiful 
 as their parents. Another is seen a long way to leeward of 
 the family, and seems as if it had strayed from home, and 
 cannot find its way back. The French call it " I'enfant 
 perdu." As vou pass the islands, the stately hills on the 
 main, ornamented with ever- verdant foliage, show you that 
 this is by far the sublimest scenery on the sea-coast, from 
 the Amazons to the Oroonoquo. On casting your eye to- 
 wards Dutch Guiana, you will see that the mountains 
 become unconnected, and few in number, and long before 
 you reach Surinam the Atlantic wave washes a flat and 
 muddy shore. 
 
 Considerably to windward of Cayenne, and about twelve Constable 
 leagues from land, stands a stately and towering rock, called ^ 
 
 
 V 'A 
 
IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 UilM 125 
 
 1.0 m 
 
 itt l&i 12.2 
 
 I.I 
 
 m 
 
 u 
 
 ^ US, 12.0 
 
 — M 
 
 
 FhotograiJhic 
 
 Somoes 
 
 CarpoFatiQn 
 
 S3 WMT MAIN STRSIT 
 
 WnSTIR,N.Y. 14SM 
 
 (7U)«73'4S0I 
 
 
•>*% 
 
 **#^ 
 
 '^4^ 
 V 
 
 •x 
 
 \ 
 
170 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 ,;i! i, 
 
 l\ { 
 
 SKCOND 
 .TOUltNKY, 
 
 Colony of 
 Cayenne, 
 
 the Constable. As nothing grows on it to tempt greedy ana 
 aspiring man to claim it as his own, the sea-fowl rest and 
 raise their offspring there. The bird called the Frigate is 
 ever soaring round its rugged summit. Hither the Phaeton 
 bends his rapid flight, and flocks of rosy Flamingos liere 
 defy the fowler's cunning. All along the coast, opposite 
 the Constable, and indeed on every uncultivated part of it 
 to windward and leeward, are seen innumerable quantities 
 of Snow-white Egrets, Scarlet Curlews, Spoonbills, and 
 Flamingos. 
 
 Cayenne is capable of being a noble and productive 
 colony. At present it is thought to be the poorest on the 
 coast of Guiana. Its estates are too much separated one 
 from the other by immense tracts of forest ; and the revo- 
 lutionary war, like a cold eastern wind, has chilled their 
 zeal and blasted their best expectations. 
 
 The Clove-tree, the Cinnamon, Pepper and Nutmeg, and 
 many other choice spices and fruits of the eastern and 
 Asiatic regions, produce abundantly in Cayenne. 
 The Town. The town itself is prettily laid out, and was once well 
 fortified. They tell you it might easily have been defended 
 against the invading force of the two united nations ; but 
 Victor Ilugues, its governor, ordered the tri-coloured flag to 
 be struck ; and ever since that day the standard of Braganza 
 has waved on the ramparts of Cayenne. 
 
 He who has received humiliations from the hand, of this 
 haughty, iron-hearted governor, may see him now in Cay- 
 enne, stripped of all his revolutionary honours, broken 
 down and ruined, and under arrest in his own house. He 
 lias four accomplished daughters, respected by the whole 
 town. Towards the close of day, when the sun's rays are 
 no longer oppressive, these much-pitied ladies are seen 
 walking up and down the balcony with their aged parent, 
 trying, by their kind and filial attention, to remove the 
 settled gloom from his too guilty brow. 
 
 Oovcmor 
 
 of 
 Cayenne. 
 
 
WANDEKINGS IX SOUTH AMERICA. 
 
 171 
 
 SECOND 
 JOURNEY. 
 
 Tim In- 
 luihitants. 
 
 This was not the time for a traveller to enjoy Cayenne. 
 The hospitality of the inhabitants was the same as ever, 
 but they had lost their wonted gaiety in public, and the 
 stranger might read in their countenances, as the recollec- 
 tion of recent humiliations and misfortunes every now and 
 then kept breaking in upon them, that they were still in 
 sorrow for their fallen country: the victorious hostile 
 cannon of Waterloo still sounded in their ears: their 
 Emperor was a prisoner amongst the hideous rocks of St. 
 Helena ; and many a Frenchman who had fought and bled 
 for France was now amongst them, begging for a little 
 support to prolong a life which would be forfeited on the 
 parent soil. To add another handful to the cypress and 
 wormwood already scattered amongst these polite colonists, 
 they had just received orders from the court of Janeiro to 
 put on deep mourning for six months, and half-mourning 
 for as many more, on account of the death of the Queen of 
 Portugal. 
 
 About a day's journey in the interior is the celebrated 
 national plantation. This spot was judiciously chosen, for 
 it is out of the reach of enemies' cruisers. It is called La 
 Gabrielle. No plantation in the western world can vie PlantaHvn 
 with La Gabrielle. Its spices are of the choicest kind ; its Gabrielle. 
 soil particularly favourable to them ; its arrangements 
 beautiful ; and its directeur, Monsieur Martin, a botanist 
 of first-rate abilities. This indefatigable naturalist ranged 
 through the East, under a royal commission, in quest of 
 botanical knowledge ; and during his stay in the western 
 regions has sent over to Europe from twenty to twenty- 
 five thousand specimens, in botany and zoology. La 
 Gabrielle is on a far-extending range of woody hills. 
 Figure to yourself a hill in the shape of a bowl reversed, 
 with the buildings on the top of it, and you will have an 
 idea of the appearance of La Gabrielle. You approach the 
 

 172 
 
 WANDERINGS IX SOUTH AMERICA. 
 
 ii 
 
 1 1 
 
 I 
 
 SECOND 
 JOURNEY. 
 
 house through a noble avenue, five hundred toises long, of 
 the choicest tropical fruit-trees, planted with the greatest 
 care and judgment ; and should you chance to stray through 
 it, after sunset, when the clove-trees are in blossom, you 
 would fancy yourself in the Tdalian groves, or near the 
 banks of the Nile, where they were burning the finest in- 
 cense as the Queen of Egypt passed. 
 
 On La Gabrielle there are twenty-two thousand clove- 
 trees in full bearing. They are planted thirty feet asunder- 
 Their lower branches touch the ground. In general the 
 trees are topped at five-and-twenty feet high ; though you 
 will see some here towering up above sixty. The black 
 pepper, the cinnamon, and nutmeg are also in great abun- 
 dance here, and very productive. 
 
 While the stranger views the spicy groves of La Gabrielle, 
 and tastes the most delicious fruits which have been origi- 
 nally imported hither from all parts of the tropical world, 
 he will thank the government which has supported, and 
 admire the talents of the gentleman who has raised to its 
 present grandeur, this noble collection of useful fruits. 
 There is a large nursery attached to La Gabrielle, where 
 plants of all the different species are raised and distri- 
 buted gratis to those colonists who wish to cultivate them. 
 
 Not far from the banks of the river Oyapoc, to windward 
 0/ le oc . ^^ Cayenne, is a mountain which contains an immense 
 cavern. Here the Cock of the Eock is plentiful. He is 
 about the size of a fantail pigeon, his colour a bright 
 orange, and his wings and tail appear as though fringed ; 
 his head is ornamented with a superb double-feathery crest, 
 edged with purple. He passes the day amid gloomy damps 
 and silence, and only issues out for food 13 short time at 
 sunrise and sunset. He is of the gallinaceous tribe. The 
 South-American Spaniards call him " Gallo del Rio Negro,*' 
 (Cock of the Black River,) and suppose that he is only to 
 
 The Cock 
 
WANDERINGS IN SOUTH AMERICA. 
 
 173 
 
 
 be met with in the vicinity of that far-inland stream ; but skcond 
 he is common in the interior of Demerara, amongst the •^ournet. 
 huge rocks in the forests of Macoushia ; and he has been 
 shot south of the line, in the captainship of Para. 
 
 The bird called by Buffon Grand Gobe-mouche has never 
 been found in Demerara, although very common in Cayenne. 
 He is not quite so large as the jackdaw, and is entirely 
 black, except a large spot under the throat, which is a 
 glossy purple. 
 
 You may easily sail from Cayenne to the river Surinam 
 in two days. Its capital, Paramaribo, is handsome, rich Pnmwia- 
 and populous : hitherto it has been considered by far the " ^' 
 finest town in Guiana ; but probably the time is not far 
 off when the capital of Demerara may claim the prize of 
 superiority. You may enter a creek above Paramaribo, 
 and travel through the interior of Surinam, till you come 
 to the Nicari, which is close to the large river Coryntin. 
 When you have passed this river, there is a good public 
 road to 17ew Amsterdam, the capital of Berbice. 
 
 On viewing New Amsterdam, it will immediately strike 
 you that something or other has intervened to prevent its 
 arriving at that state of wealth and consequence for which 
 its original plan shows it was once intended.* What has 
 caused this stop in its progress to the rank of a fine and 
 populous city remains for those to find out who are inter- 
 ested in it ; certain it is, that New Amsterdam has been 
 languid for some years, and now the tide of commerce 
 seems ebbing fast from the shores of Berbice. 
 
 Gay and blooming is the sister colony of Demerara. Demerara. 
 Perhaps, kind reader, thou hast not forgot that it was from 
 Stabroek, the capital of Demerara, that the adventurer set 
 out, some years ago, to reach the Portuguese frontier fort, 
 and collect the wourali-poison. It was not intended, when 
 this second sally was planned in England, to have visited 
 
'■■ I f 
 
 ('« ' 
 
 SECOND Stal)roek again by the route here described. The plan was 
 jouKNEY. ^Q l^fixQ ascended the Amazons from Para and got into tlie 
 Ilio Negro, and from thence to have returned towards the 
 source of the Essequibo, in order to examine the crystal 
 mountains, and look once more for Lake Parinia, or the 
 White Sea ; but on arriving at Cayenne, the current was 
 running with such amazing rapidity to leeward, that a 
 Portuguese sloop, which had been beating up towards Para 
 for four weeks, was then only half way. Finding, therefore, 
 that a beat to the Amazons would be long, tedious, and 
 even uncertain, and aware that the season for procuring 
 birds in fine plumage had already set in, I left Cayenne in 
 an American ship for I'aramavibo, went through the in- 
 terior to the Coryntin, stopped a few days in New Amster- 
 dam, and proceeded to Demerara. If, gentle reader, thy 
 patience be not already worn out, and thy eyes half closed 
 in slumber, by perusing the dull adventures of this second 
 sally, perhaps thou wilt pardon a line or two on Demerara ; 
 and then we will retire to its forests, to collect and examine 
 the economy of its most rare and beautiful birds, and give 
 the world a new mode of preserving them. 
 Stabrock. Stabroek, the capital of Demerara, has been rapidly 
 increasing for some years back ; and if prosperity go hand 
 in hand with the present enterprising spirit, Stabroek, ere 
 long, will be of the first colonial consideration. It stands 
 on the eastern bank at the mouth of the Demerara, and 
 enjoys all the advantages of the refreshing sea-breeze ; the 
 streets are spacious, well bricked, and elevated, the trenches 
 clean, the bridges excellent, and the houses handsome. 
 Almost every commodity and luxury of London may be 
 bought in the shops at Stabroek : its market wants better 
 regulations. The hotels are commodious, clean, and well 
 attended. Demerara boasts as fine and well-disciplined 
 militia as any colony in the western world. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 175 
 
 The court of justice, wliere, in times of old, tiie bandage srcond 
 was easily removed from the eyes of the goddess, and her '^" ""''''^ ^'' 
 scales thrown out of equilibrium, now rises in dignity under ^»urtof 
 the firmness, talents, and urbanity of Mr. President Rough. 
 
 The plantations have an appearance of high cultivation ; Tlicplan- 
 a tolerable idea may be formed of their value when yon 
 know tliat last year Demerara numbered seventy-two 
 thousand nine hundred and ninety-nine slaves. They 
 made above forty-four million pounds of sugar, near two 
 million gallons of rum, above eleven million pounds of 
 coffee, and three million eight hundred and nineteen 
 thousand five hundred and twelve pounds of cotton ; the 
 receipt into the public chest was five hundred and fifty- 
 three thousand nine hundred and fifty-six guilders ; the 
 public ex})enditure, four hundred and fifty-one thousand 
 six hundred and three guilders. 
 
 Slavery can never be defended ; he whose heart is not Slavery. 
 of iron can never wish to be able to defend it : while he 
 heaves a sigh for the poor negro in captivity, he wishes 
 from his soul that the traffic had been stifled in its birth > 
 but, unfortunately, the governments of Europe nourished 
 it, and now that they are exerting themselves to do away 
 the evil, and ensure liberty to the sons of Africa, the situa- 
 tion of the plantation slaves is depicted as truly deplorable, 
 and their condition wretched. It is not so. A Briton's 
 heart, proverbially kind and generous, is not changed 
 by climate, or its strean»" of compassion dried up by the 
 scorching heat of a Demerara sun ; he cheers his negroes 
 in labour, comforts them in sickness, is kind to them in old 
 age, and never forgets that they are his fellow-creatures. 
 
 Instances of cruelty and depravity certainly occur here as 
 well as all the world over ; but the edicts of the colonial 
 government are well calculated to prevent them ; and the 
 British planter, except here and there one, feels for the 
 
) ; 
 
 i 'l)l I I 
 
 I ' ' ' 
 
 la I 
 I; 1 
 
 .1^1 
 
 
 11 
 
 ' ii 
 
 ITfi 
 
 WAXDE1?INGS IN SOUTH AMEiaCA. 
 
 rzcxjsv wrongs done to a poor ill-treated slave, and shows that his 
 jouRNET. i)pjjj.j; grieves for hiui by causing immediate redress, and 
 preventing a repetition. 
 
 Long may ye flourish, peaceful and liberal inhabitants of 
 Denierara 1 Your doors are ever open to harbour the har- 
 hourless ; your purses never shut to the wants of tlie 
 distressed: many a ruined fugitive from the Oroouoque 
 will bless your kindness to him in the hour of need, when 
 tlying from the woes of civil discord, without food or 
 raiment, he begged for shelter underneath your roof. The 
 poor sufferer in Trinidad, who lost his all in the devouring 
 Hames, will remember your charity to his latest moments. 
 The traveller, as he leaves your port, casts a longing, linger- 
 ing look behind ; your attentions, your hospitality, your 
 pleasantry and mirth, are uppermost in his thoughts : 
 your prosperity is close to his heart. Let us now, gentle 
 reader, retire from the busy scenes of man, and journey on 
 towards the wilds in quest of the feathered tribe. 
 
 Leave behind you your high-seasoned dishes, your wines 
 and your delicacies ; carry nothing but what is necessary 
 for your own comfort and the object in view, and depend 
 upon the skill of an Indian, or your own, for fish and game. 
 A sheet, about twelve feet long, ten wide, painted, and with 
 loop-holes on each side, will be of great service ; in a few 
 minutes you can suspend it betwixt two trees in the shape 
 of a roof. Under this, in your hammock, you may defy 
 the pelting shower, and sleep heedless of the dews of night. 
 A hat, a shirt, and a light pair of trousers, will be all the 
 raiment you require. Custom will soon teach you to tread 
 lightly and barefoot on the little inequalities of the ground, 
 and show you how to pass on, unwounded, amid the mantling 
 briers. 
 
 Snakes. Snakes in these wilds are certainly an annoyance, though 
 perhaps more in imagination than reality; for you must 
 
WANDERINGS IN SOUTH AVERICA. 
 
 . i t 
 
 recollect that the serpent is never the first to offenil ; skcond 
 his poisonous fang was not given him for conquest : he ■'"'"^^'■•y- 
 never inflicts a wound with it but to defend existence. 
 Provided you walk cautiously, and do not absolutely touch 
 him, you may pass in safety close by him. As he is often 
 coiled up on the ground, and amongst the branches of the 
 trees above you, a degree of circumspection is necessary, 
 lest you unwarily disturb him. 
 
 Tigers are too few, and too apt to fly before the noble Tiyers. 
 face of man, to require a moment of your attention. 
 
 The bite of the most noxious of the insects, at the very Liscct.i. 
 worst, only causes a transient fever, with a degree of pain 
 more or less. 
 
 Birds in general, with few exceptions, are not common in Binls. 
 the very remote parts of the forest. The sides of rivers, 
 lakes, and creeks, the borders of savannas, the old abandoned 
 habitations of Indians and woodcutters, seem to be their 
 favourite haunts. 
 
 Though least in size, the glittering mantle of the Humming- mmmhg 
 bird entitles it to the first place in the list of the birds of '" " 
 the new world. It may truly be called the Bird of Paradise ; 
 and had it existed in the old world, it would have claimed 
 the title instead of the bird which has no'A the honour to 
 bear it : — see it darting through the air ahnosi. as quick as 
 thought ! — now it is within a yard of your face ! — in an 
 instant gone ! — now it flutters from flower to flower to sip 
 the silver dew — it is now a ruby — now a topaz — now an 
 emerald — now all burnished gold ! It would be arrogant 
 to pretend to describe this winged gem of nature after 
 Buffbn's elegant description of it. 
 
 Cayenne and Demerara produce the same humming-birds. Haunts of 
 
 t/lP 
 
 Perhaps you would wish to know something of their mimmimj 
 haunts. Chiefly in the months of July and August the ^"''^• 
 tree called Bois Immortel, very common in Demerara, bears 
 
 N 
 
178 
 
 VVAXDEUIXGS IN SOUTH AMEUU.'A. 
 
 SKCONI) 
 JOUnXEY. 
 
 •^1 , 
 
 ubuuilanoe of red blossom, whicli stays on the tree sonje 
 weeks; then it is that most of the dilTerent species of 
 Inimniing-birds are very plentiful. The wild red sage is 
 also their favourite shrub, and they buzz like bees round 
 the blossom of the wallaba-treo. Indeed, there is scarce a 
 flower in the interior, or on the soa-coast, but what receives 
 frequent visits from one or other of the species. 
 
 On entering the forests, on the rising land in the interior, 
 the blue and green, the smallest brown, no bigget than the 
 bumblebee, with two long feathers in the tail, and the little 
 forked-tail purple-throated humming-birds, glitter before 
 you in ever-changing attitudes. One species alone never 
 shows his beauty to the sun ; and were it not for his lovely 
 shining colours, you might almost be tempted to class him 
 with the goatsuckers on account of his habits. He is the 
 largest of all the humming-birds, and is all red and changing 
 gold green, except the head, which is black. He has two 
 long feathers in the tail, which cross each other, and these 
 have gained him the name of Karabimiti, or Ara humming- 
 bird, from the Indians. You never find him on the 
 sea-coast, or M'here the river is salt, or in the heart of the 
 forest, unless fresh water be there. He keeps close by the 
 side of woody fresh-water rivers and dark and lonely 
 creeks. He leaves his retreat before sunrise to feed on the 
 insects over the water ; he returns to it as soon as the sun's 
 rays cause a glare of light, is sedentary all day long, and 
 comes out again for a short time after sunset. He builds 
 his nest on a twig over the water in the unfrequented 
 creeks ; it looks like tanned cow-leather. 
 
 As you advance towards the mountains of Demerara, 
 other species of humming-birds present themselves before 
 you. It seems to be an eiToneous opinion that the hum- 
 ming-bird lives entirely on honey-dew. Almost every 
 flower of the tropical climate contains insects of one kind 
 
vVANDERIXnS TN SOUTH AMEIJICA. 
 
 179 
 
 or other; now, the humming-bird is most busy about the 
 flowers an hour or two after sunrise and after a sliowcr of 
 rain, and it is just at this time tliat tlie insects come out to 
 the edge of the llower in order that the sun's rays may 
 dry tlie nocturnal dew and rain whicli they have received. 
 On opening the stomach of the hunmiiug-bird, dead insects 
 are ahnost always found there. 
 
 Next to the ^humming-birds, the Cotingas display the 
 gayest plumage. They arc of the order of Passeres, and 
 you number five species betwixt the sea-coast and the rock 
 Salja. Perhaps the Scarlet Cotinga is the richest of the five, 
 and is one of those birds which are found in the deepest 
 recesses of the forest. His crown is flaming red ; to tliis 
 abruptly succeeds a dark shining brown, reaching lialf way 
 down the back : the remainder of the back, the rump, and 
 tail, the extremity of which is edged witli black, are a 
 lively red ; the belly is a somewhat lighter red ; the breast 
 reddish black ; the wings brown. He has no song, is soli- 
 tary, and utters a monotonous whistle which sounds like 
 " quet." He is fond of the seeds of the hitia-tree, and 
 those of the siloabali and bastard-si] oabali trees, which 
 ripen in December, and continue on the trees for al)ove two 
 months. He is found throughout the year in Demerara ; 
 still nothing is known of his incubation. The Indians all 
 agree in telling you that they have never seen his nest. 
 
 The Purple-breasted Cotinga has the throat and breast of 
 a deep purple, the wings and tail black, and all the rest of 
 the body a most lovely shining blue. 
 
 The Purple-throated Cotinga has black wings and tail^ 
 and every other part a light and glossy blue, save the 
 throat, which is purple. 
 
 The Pompadour Cotinga is entirely purple, except his 
 wings, which are white, their four first feathers tipped with 
 brown. The great coverts of the wings are stiff, narrow, 
 
 n2 
 
 HrroNP 
 
 J0UUN1-,V, 
 
 The 
 
 CotAnijas. 
 
 The 
 
 Purple- 
 
 hrcdfitrd 
 
 Colinyn. 
 
 The Pom- 
 parlour 
 Votinijtt, 
 
180 
 
 WANDKUINOS IN SOUTH AMEUICA. 
 
 i\ii. 
 
 l-i! 
 
 11 
 
 '.i ' »i 
 
 ■*>' 
 
 T/ie Cam 
 paiiero. 
 
 SECOND and pointed, being sliaped quite different from those of any 
 
 joum:Y. Qi-i^gj. jjjj.(j When you are betwixt this bird and the sun 
 
 in his flight, he appears uncommonly brilliant, lie makes 
 
 a hoarse noise, which sounds like " Wallababa," Hence 
 
 his name amongst the Indians. 
 
 None of these three cotingas have a song. They feed on 
 the hitia, siloabali, and bastard-siloabali seeds, the wild 
 guava, the fig, and other fruit trees of the forest. They 
 are easily shot in these trees during the months of Decem- 
 ber, January, and part of February. The greater part of 
 them disappear after this, and probably retire far away to 
 breed. Their nests have never been found in Demerara. 
 
 The fifth species is the celebrated Campanero of the 
 Spaniards, called Dara by the Indians, and Bell-bird by the 
 English. He is about the size of the jay. His plumage is 
 white as snow. On his forehead rises a spiral tube nearly 
 three inches long. It is jet black, dotted all over with 
 small white feathers. It has a communication with the 
 palate, and when filled with air, looks like a spire ; when 
 empty, it becomes pendulous. His note is loud and clear, 
 like the sound of a bell, and may be heard at the Jiatance 
 of three miles. In the midst of these extensive wilds, 
 generally on the dried top of an aged mora, almost out 
 of gun reach, you will see the campanero. No sound or 
 song from any of the winged inhabitants of the forest, not 
 even the clearly pronounced " Whip-poor- Will," from the 
 goatsucker, cause such astonishment as the toll of the 
 campanero. 
 
 With many of the feathered race, he pays the common 
 tribute of a morning and an evening song ; and even when 
 the meridian sun has shut in silence the mouths of almost 
 the whole of animated nature, the campanero still cheers 
 the forest. You hear his toll, and then a pause for a 
 minute, then another toll, and then a pause again, and 
 
WANDEHINfJS IX SOUTH AMERICA. 
 
 181 
 
 The 
 Toucan, 
 
 tlien a toll, and nguiu a pause. Then he is silent for six or hf.cond 
 eight minutes, and then anotlier toll, and so on. Acteon '^" ''"^'''' ^'' 
 would stop in mid chace, Maria would defer her evening 
 song, and Or])heus himself would drop his lute to listen to 
 him, so sweet, so novel, and romantic is the toll of the 
 pretty snow-white campanero. He is never seen to feed 
 with the other cotingas, nor is it known in what part of 
 Guiana he makes his nest. 
 
 "While the cotingas attract your attention by their 
 superior plumage, the singular form of the Toucan makes 
 a lasting impression on your memory. There are three 
 species of toucans in Demerara, and three diminutives, 
 which may he called Toucanets. The largest of the first 
 species frequents the mangrove-trees on the sea-cua^it. He 
 is never seen in the interior till you reach Macoushia, 
 where he is found in the neighbourhood of the river 
 Tacatou. The other two species are very common. They 
 feed entirely on the fruits of the forest, and though of the 
 pie kind, never kill the young of other birds or touch 
 carrion. The larger is called Bouradi by the Indians, 
 (which means Nose,) the other, Scirou. They seem partial 
 to each other's company, and often resort to the same feed- 
 ing tree, and retire together to the same shady noon-day 
 retreat. They are very noisy in rainy weather at all hours 
 of the day, and in fair weather, at morn and eve. The 
 sound which the bouradi makes is like the clear yelping of 
 a puppy dog, and you fancy he says " Pia-po-o-co," and thus 
 the South American Spaniards call him Piapoco. 
 
 All the toucanets feed on the same trees on which the 
 toucan feeds, and every species of this family of enormous 
 bill, lays its eggs in the hollow trees. They are social, but 
 not gregarious. You may sometimes see eight or ten in 
 company, and from tliis you would suppose they are gre- 
 garious ; but, upon a closer examination, you will find it 
 
 I I 
 
ii! ii. 
 
 i 
 
 ! 1 ' 
 
 182 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 pEcoND lias only been a dinner party, which breaks up and dis- 
 
 j ouKNK Y. perses towards roosting-time. 
 
 You will be at a loss to conjecture for what ends nature 
 has overloaded the head of this bird with such an enor- 
 mous bill. It cannot be for the offensive, as it has no need 
 to wage war with any of the tribes of animated nature ; 
 for its food is fruit and seeds, and those are in superabun- 
 dance throughout the whole year in the regions where the 
 toucan is found. It can hardly be for the defensive, as the 
 toucan is preyed upon by no bird in South America, and 
 were it obliged to be at M^ar, the texture of the bill is ill 
 adapted to give or receive blows, as you will see in dissect- 
 ing it. It cannot be for any particular protection to the 
 tongue, as the tongue is a perfect feather. 
 
 Its flight. The flight of the toucan is by jerks ; in the action of 
 flying it seems incommoded by this huge disproportioned 
 feature, and the head seems as if bowed down to the earth 
 by it against its will ; if the extraordinary form and size of 
 the bill expose the toucan to ridicule, its colours make it 
 
 Colours of amends. Were a specimen of each species of the toucan 
 presented to you, you would pronounce the bill of the 
 bouradi the most rich and beautiful ; on the ridge of the 
 upper mandible a broad stripe of most lovely yellow ex- 
 tends from the head to the point ; a stripe of the same 
 breadth, though somewhat deeper yellow, falls from it at 
 right angles next the head down to the edge of the man- 
 dible ; then follows a black stripe, half as broad, falling at 
 right angles from the ridge, and running narrower along the 
 edge to witliin half an inch of the point. The rest of the 
 mandible is a deep bright red. The lower mandible has no 
 yellow: its black and red are distributed in the same 
 manner as on the upper one, with this difference, that 
 there is black about an inch from the point. The stripe 
 corresponding to the deep yellow stripe on the upper 
 
WANDERINGS IN SOUTH AMERICA. 
 
 183 
 
 mandible is sky blue. It is worthy of remark that all second 
 these brilliant colours of the bill are to be found in the •'Q^^^ ^'^ -y- 
 plumage of the body, and the bare skin round the eye. 
 
 All these colours, except the blue, are inherent in the 
 horn ; that part which appears blue is in reality transpar- 
 ent white, and receives its colour from a thin piece of blue 
 skin inside. This superb bill fades in death, and in three 
 or four days' time has quite lost its original colours. 
 
 Till within these few years, no idea of the true colours 
 of the bill could be formed from the stuffed toucans brought 
 to Europe. About eight years ago, while eating a boiled 
 toucan, the thought struck me that the colours in the bill 
 of a preserved specimen might be kept as bright as those 
 in life. A series of experiments proved this beyond a Preserves a 
 doubt. If you take your penknife and cut away the roof Vojtcaji.*'^ 
 of the upper mandible, you will find that the space betwixt 
 it and the outer shell contams a large collection of veins, 
 and small osseous fibres running in all directions through 
 the whole extent of the bill Clear away all these with 
 your knife, and you will come to a substance more firm 
 than skin, but of not so strong a texture as the horn itself ; 
 cut this away also, and behind it is discovered a thin and 
 tender membrane ; yellow, where it has touched the yel- 
 low part of the horn ; blue, where it has touched the red 
 part, and black towards the edge and point ; when dried, 
 this thin and tender membrane becomes nearly black ; as 
 soon as it is cut away, nothing remains but the outer horn, 
 red and yellow, and now become transparent ; the under 
 mandible must undergo the same operation. Great care 
 must be taken, and the knife used very cautiously when 
 you are cutting through the different parts close to where 
 the bill joins on to the head ; if you cut away too much, 
 the bill drops off ; if you press too hard, the knife comes 
 through the horn ; if you leave too great a portion of the 
 
 m 
 
184 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 SEcoxn 
 
 JOURNEY. 
 
 membrane, it appears through the horn, and by becoming 
 black when dried, makes the horn appear black also, and 
 has a bad effect ; judgment, caution, skill, and practice, 
 will ensure success. 
 
 You have now cleared the bill of all those bodies which 
 are the cause of its apparent fading ; for, as has been said 
 before, these bodies dry in death, and become quite dis- 
 coloured, and appear so tlirough the horn ; and reviewing 
 the bill in this state, you conclude that its former bright 
 colours are lost. 
 
 Something still remains to be done. You have ren- 
 dered the bill transparent by the operation, and that 
 transparency must be done away to make it appear per- 
 fectly natural. Pound some clean chalk, and give it 
 enough water till it be of the consistency of tar ; add a 
 proportion of gum-arabic to make it adhesive ; then take a 
 camel-hair brush, and give the inside of both mandibles a 
 coat ; apply a second when the first is dry, then another, 
 and a fourth to finish all. The gum-arabic will prevent 
 the chalk from cracking and falling off. If you remember, 
 there is a little space of transparent white in the lower 
 mandible, which originally appeared blue, but which be- 
 came transparent white as soon as the thin piece of blue 
 skin was cut away; this must be painted blue inside. 
 When all this is completed, the bill will please you ; it 
 will appear in its original colours. Probably your own 
 abilities will suggest a cleverer mode of operating than 
 the one here described. A small gouge would assist the 
 penknife, and render the operation less difficult. 
 
CHAPTEE in. 
 
 The Houtou.— Curious habit of trimming the tail and feathers— its habits. 
 —The Guianan Jay.— The Boclora.— Slight attachment of the feathers. 
 — The Cuia. — Rice-birds. — Cassiques— their habit of mockery. — 
 Pendulous nests. — Gregarious nesting of different species. — "Wood- 
 peckers of America and England. — Kingfishers.— Jacamars and their 
 fly-catching habits. — Troupiales and their songs. — Tangaras. — Mani- 
 kins. — Tiger-birds. — Yawaraciri. — Ant Thrushes.— Parrot of the Sun. 
 — Aras, or Macaws. — Bitterns.— Egret, Herons, etc. — Goatsuckers.— 
 Whip-poor- Will. — Superstitions. — Tiuamous.— Powis and Maroudi. — 
 Horned Screamer. — Truinpeter. — King Vulture. — Anhinga. — Dangers 
 of travel. — Quartan ague. 
 
 Houtou. 
 
 The Houtou ranks high in heauty amongst the birds of second 
 Denierara — his whole body is green, with a bhiish cast in "^'^^'"-'''^^• 
 the wings and tail ; his crown, which he erects at pleasure, 
 consists of black in the centre, surrounded with lovely blue 
 of two different shades : he has a triangular black spot, 
 edged with blue, I. ehind the eye extending to the ear ; and 
 on his breafc a sable tuft, consisting of nine feathers edged 
 also with blue. This bird seems to suppose that its beauty 
 can be increased by trimming the tail, which undergoes the 
 same operation as our hair in a barber's shop, only with 
 this difference, that it uses its own beak, which is serrated, 
 in lieu of a pair of scissors ; as soon as his tail is full 
 grown, he begins about an inch from the extremity of the 
 two longest feathers in it, and cuts away the web on both 
 sides of the shaft, making a gap about an inch long : both 
 
 \\\ 
 
> 
 
 
 186 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 SECOND 
 JOUKNKY, 
 
 W''^ 
 
 Its 
 Haunts. 
 
 male and female Adonise their tails in this manner, which 
 gives them a remarkable appearance amongst all other birds. 
 While we consider the tail of the houtou blemished and 
 defective, were he to come amongst us, he would probably 
 consider our heads, cropped and bald, in no better light. 
 He who wishes to observe this handsome bird in his native 
 haunts, must be in tlie forest at the morning's dawn. The 
 houtou shuns the society of man: the plantations and 
 cultivated parts are too much disturbed to engage it to 
 settle there ; the thick and gloomy forests are the places 
 preferred by the solitary houtou. In those far-extendiug 
 wilds, about daybreak, you hear him articulate, in a dis- 
 tinct and mournful tone, " Houtou, houtou." Move cautious 
 on to where the sound proceeds from, and you will see him 
 sitting in the underwood, about a couple of yards from the 
 ground, his tail moving up and down every time he articu- 
 lates " houtou." He lives on insects and the berries amongst 
 the underwood, and very rarely is seen in the lofty trees, 
 except the bastard-siloabali tree, the fruit of which is grate- 
 ful to him. He makes no nest, but rears his young in a 
 hole in the sand, generally on the side of a hill. 
 
 While in quest of the houtou, you will now and then 
 Tfie Jay of fall in with the Jay of Guiana, called by the Indians 
 Ibibirou. Its forehead is black, the rest of the head white ; 
 the throat and breast like the English magpie : about an 
 inch of the extremity of the tail is white, the other part of 
 it, together with the back and wings, a greyish changing 
 purple ; the belly is white : there are generally six or eight 
 of them in company ; they are shy and garrulous, and tarry 
 a very short time in one place ; they are never seen in the 
 cultivated parts. 
 The Through the whole extent of the forest, chiefly from 
 sunrise till nine o'clock in the morning, you hear a sound 
 of " Wow, wow, wow, wow." This is the bird called Boclora 
 
 IH 
 
WANDERINGS IN SOUTH AMERICA. 
 
 187 
 
 by the Indians. It is smaller than the common pigeon, and sec.>ni» 
 seems, in some measure, to partake of its nature : its head Jo u^>"^ y' 
 and breast are blue ; the back and rump somewhat resemble 
 the colour on the peacock's neck ; its belly is a bright yel- 
 low ; the legs are so very short that it always appears as if 
 sitting on the branch ; it is as ill-adapted for walking as the 
 swallow ; its neck, for above an inch all round, is quite bare 
 of feathers, but tliis deficiency is not seen, for it always sits 
 with its head drawn in upon its shoulders : it sometimes 
 feeds with the cotingas on the guava and hitia trees ; but 
 its chief nutriment seems to be insects, and, like most birds 
 which follow this prey, its chaps are well armed with 
 bristles : it is found in Demerara at all times of the year, 
 and makes a nest resembling that of the stock-dove. This 
 bird never takes long flights, and when it crosses a river or 
 creek it goes by long jerks. 
 
 The boclora is very unsuspicious, appearing quite heed- 
 less of danger : the report of a gun within twenty yards 
 will not cause it to leave the branch on which it is sitting, 
 and you may often approach it so near as almost to touch 
 it with the end of your bow. Perhaps there is no bird 
 known whose feathers are so slightly fixed to the skin as 
 those of the boclora. After shooting it, if it touch a branch 
 in its descent, or if it drop on hard ground, whole heaps of 
 feathers fall off ; on this account it is extremely hard to 
 procure a specimen for preservation. As soon as the skin 
 is dry in the preserved specimen, the feathers become as 
 well fixed as those in any other bird. 
 
 Another species, larger than the boclora, attracts much The Cuia. 
 of your notice in these wilds ; it is called Cuia by the 
 Indians, from the sound of its voice ; its habits are the 
 same as those of the boclora, but its colours different ; its 
 head, breast, back, and rump, are a shining, changing green ; 
 its tail not quite so bright; a black bar runs across the 
 
188 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 > ''li 
 II' 
 
 ' ii 
 
 sKooND tail towards the extremity, and the outside feathers are 
 j ouRKE v. partly white, as in the boclora ; its belly is entirely ver- 
 milion, a bar of white separating it from the green on the 
 breast. 
 
 There are diminutives of both these birds ; they have 
 the same habits, with a somewhat different plumage, and 
 about half the size. Arrayed from head to tail in a robe 
 The of richest sable hue, the bird called Rice-bird loves spots 
 liicc-bird. cultivated by the hand of man. The woodcutter's house 
 on the hills in the interior, and the planter's habitation on 
 the sea-coast, equally attract this songless species of the 
 order of pie, provided the Indian corn be ripe there. He 
 is nearly of the jackdaw's size, and njakes his nest far 
 away from the haunts of man ; he may truly be called a 
 blackbird : independent of his plumage, his beak, inside 
 and out, his legs, his toes, and claws are jet black. 
 
 Mankind, by clearing the ground, and sowing a variety 
 of seeds, induces many kinds of birds to leave their native 
 haunts and come and settle near him ; their little depre- 
 dations on his seeds and fruits prove that it is the property, 
 and not the proprietor, which has the attractions. 
 
 One bird, however, in Demerara is not actuated by selfish 
 motives : this is the Cassique ; in size, he is larger than the 
 starling ; he courts the society of man, but disdains to live 
 by his labours. When nature calls for support, he repairs 
 to the neighbouring forest, and there partakes of the store 
 of fruits and seeds which she has produced in abundance 
 for her aerial tribes. When his repast is over, he returns 
 to man, and pays the little tribute which he owes him for his 
 protection ; he takes his station on a tree close to his house ; 
 and there, for hours together, pours forth a succession of 
 imitative notes. His own song is sweet, but very short. 
 If a toucan be yelping in the neighbourhood, he drops it, 
 and imitates him. Thou he will amuse his protector with 
 
 The 
 Cassique. 
 
WANDERINGS IX SOUTH AMEPJCA. 
 
 189 
 
 the cries of the different species of the woodpecker ; and second 
 when the sheep bleat, he will distinctly answer them. '^ """'^'^ ^' 
 Then comes his own song again ; and if a puppy-dog, or a 
 Guinea-fowl interrupt him, he takes them off admirably, 
 and by his different gestures during the time, you would 
 conclude that he enjoys the sport. 
 
 The cassique is gregarious, and imitates any sound he 
 hears with such exactness, that he goes by no other name 
 than that of Mocking-bird amongst the colonists. 
 
 At breeding time, a number of these pretty choristers 
 resort to a tree near the planter's house, and from its out- 
 side branches weave their pendulous nests. So conscious 
 do they seem that they never give offence, and so little 
 suspicious are they of receiving any injury from man, that 
 they will choose a tree within forty yards from his house, 
 and occupy the branches so low down, that he may peep 
 into the nests. A tree in Waratilla creek affords a proof 
 of this. 
 
 The proportions of the cassique are so fine, that he may 
 be said to be a model of synmietry in ornithology. On 
 each wing he has a bright yellow spot, and his rump, belly, 
 and half the tail, are of the same colour. All the rest of 
 the body is black. His beak is the colour of sulphur, but 
 it fades in death, and requires the same operation as the 
 bill of the toucan to make it keep its colours. Up the 
 rivers, in the interior, there is another cassique, nearly the 
 same size, and of the same habits, though not gifted witli 
 its powers of imitation. Except in breeding time you will 
 see hundreds of tliem retiring to roost, amongst the moca- 
 moca-trees and low shrubs on the banks of the Demerara, 
 after you pass the first island. They are not common on 
 the sea-coast. The rump of this cassique is a flaming scar- 
 let. All the rest of the body is a rich glossy black. His 
 bill is sulphur colour. You may often see numbers of this 
 
 m I 
 
 M 
 
190 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 I I 
 
 8KC0SD 
 JOUHNEY. 
 
 Another 
 
 species weaving their pendulous nests on one side of a tree, 
 while numbers of the other species are busy in forming 
 theirs on the opposite side of the same tree. Though such 
 near neighbours, the females are never observed to kick up 
 a row, or come to blows ! 
 
 Another species of cassique, as large as a crow, is very 
 fipecics of common in the plantations. In the morning, he generally 
 repairs to a large tree, and there, with his tail spread over 
 his back, and shaking his lowered wings, he produces notes, 
 which though they cannot be said to amount to a song, still 
 have something very sweet and pleasing in them. He 
 makes his nest in the same form as the otlier cassiques. 
 It is above four feet long ; and when you pass under the 
 tree, which often contains fifty or sixty of them, you can- 
 not help stopping to admire them as they wave to and fro, 
 the sport of every storm and breeze. The rump is chest- 
 nut ; ten feathers of tlie tail are a fine yellow, the remaining 
 two, which are the middle ones, are black, and an inch 
 shorter than the others. His bill is sulphur colour ; all 
 the rest of tlie body black, with here and there shades of 
 brown. He has five or six long narrow black feathers on 
 the back of his head, which he erects at pleasure. 
 
 There is one more species of cassique in Demerara, 
 which always prefers the forests to the cultivated parts. 
 His economy is the same as that of the other cassiques. 
 He is rather smaller than the last described bird. His 
 body is greenish, and his tail and rump paler than those of 
 the former. Half of his beak is red. 
 
 You would not be long in the forests of Demerara with- 
 out noticing the Woodpeckers. You meet with them feed- 
 ing at all hours of the day. Well may they do so. Were 
 they to follow the example of most of the other birds, and 
 only feed in .the morning and evening, they would be often 
 on short allowance, for they sometimes have to labour 
 
 Wood- 
 2}cckers. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 191 
 
 three or four hours at the tree before they get to their food, skcond 
 The sound which the largest kind makes in liammering J^ ^^'^nk y. 
 against the bark of the tree, is so loud, that you would 
 never suppose it to proceed from the efforts of a bird. You 
 would take it to be the woodman, with his axe, trying by a 
 sturdy blow, often repeated, whether the tree were sound or 
 not. There are fourteen species here ; the largest the size 
 of a magpie, the smallest no bigger than the wren. They 
 are all beautiful ; and the greater part of them have their 
 heads ornamented with a fine crest, movable at ple.isure. 
 
 It is said, if you once give a dog a bad nam.e, whether 
 innocent or guilty, he never loses it. It Rticks close to 
 him wherever he goes. He has many a kick and many 
 a blow to bear on account of it ; and there is nobody to 
 stand up for him. The woodpecker is little better off. 
 The proprietors of woods, in Europe, have long accused 
 him of injuring their timber, by boring holes in it, and 
 letting in the water, which soon rots it. The colonists in 
 America have the same complaint against him. Had he 
 the power of speech, which Ovid's birds possessed in days 
 of yore, he could soon make a defence. " Mighty lord of 
 the woods," he would say to man, " why do you wrongfully 
 accuse me ? why do you hunt me up and down to death 
 for an imaginary offence ? I have never spoiled a leaf of 
 your property, much less your wood. Your merciless shot 
 strikes me, at the very time I am doing you a service. 
 But your shortsightedness will not let you r?ee it, or your 
 pride is above examining closely the actions of so insignifi- 
 cant a little bird as I am. If there be that spark of feeling 
 in your breast wliicli they say man possesses, or ought to 
 possess, above all other animals, do a poor injured creature 
 a little kindness, and watch me in your woods only for one 
 day. I never wound your healthy trees. I should perisli 
 for want in the attempt. The sound bark would easily 
 

 'I 
 
 I 
 
 192 
 
 WANDERING IN SOUTH AMERICA. 
 
 HKcoND resist the force of my bill : and veie I eveu to i)ierce 
 jouKNKY, through it, there would be nothing inside that I could 
 fancy, or my stomach digest. I often visit them, it is 
 true, but a knock or two convinces me that I must go else- 
 where for sujjport ; and were you to listen attentively to 
 the sound which my bill causes, you would know whether 
 I am upon a healthy or an unhealthy tree. Wood and 
 bark are not my food. I live entirely upon the insects 
 which have already formed a lodgement in the distempered 
 tree. AVhen the sound informs me that my prey is there, 
 I labour for hours together till I get at it ; and by consum- 
 ing it, for my own support, I prevent its further depreda- 
 tions in that part. Thus I discover for you your hidden 
 and unsuspected foe, which has been devouring your wood 
 in such secrecy, that you had not the least suspicion it was 
 there. The hole which I make in order to get at the per- 
 nicious vermin will be seen by you as you pass under the 
 tree. I leave it as a signal to tell you that your tree has 
 already stood too long. It is past its prime. Millions of 
 insects, engendered by disease, are preying upon its vitals. 
 Ere long it will fall a log in useless ruins. Warned by 
 this loss, cut down the rest in time, and spare, spare the 
 unoffending woodpecker ! " 
 The King- In the rivers, and different creeks, you number six 
 '^'"^''' species of the Kingfisher. They make their nest in a hole 
 in the sand on the side of the bank. As there is always 
 plenty of foliage to protect them from the heat of the sun, 
 they feed at all hours of the day. Though their plumage 
 is prettily varied, still it falls far short of the brilliancy dis- 
 played by the English kingfisher. This little native of 
 Britain would outweigh them altogether in the scale of 
 beauty. 
 
 A bird called Jacamar is often taken for a king-fisher, 
 but it has no relationship to that tribe ; it frequently sits in 
 
 The 
 Jacamar, 
 
WANDKHINGS IN SOUTH AMKKICA. 
 
 193 
 
 tlio trees over the water, and as its beak bears some resem- 
 blance to that of the kingfisher, this may probably account 
 for its being taken for one ; it feeds entirely upon insects ; 
 it sits on a brunch in motionless expectation, and as soon 
 as a lly, butterfly, or moth passes by, it darts at it, and re- 
 turns to the brancli it had just left. It seems an indolent, 
 sedentary bird, shunning the society of all others in the 
 forest. It never visits the plantations, but is found at all 
 times of the year in the woods. There are four species of 
 jacamar in Demerara; they are all beautiful; the largest, 
 rich and superb in the extreme. Its plumage is of so line 
 a changing blue and golden green, that it may be ranked 
 with the choicest of the humming-birds. Nature has de- 
 nied it a song, but given a costly garment in lieu of it. 
 The smallest species of jacamar is very common in tlie dry 
 savannas. The second size, all golden green on the back, 
 must be looked for in the wallaba forest. The third is 
 found throughout the whole extent of these wilds ; and the 
 fourth, which is the largest, frequents the interior, where 
 you begin to perceive stones in the ground. 
 
 When you have penetrated far into Macousliia, you liear 
 the pretty songster called Troupiale pour forth a variety 
 of sweet and plaintive notes. This is the bird which the 
 Tortuguese call the Nightingale of Guiana ; its predominant 
 colours are rich orange and shining black, arrayed to great 
 advantage; his delicate and well-shaped frame seems un- 
 able to bear captivity. The Indians sometimes bring down 
 troupiales to Stabroek, but in a few months they languish 
 smd die in a cage. They soon become very familiar ; and 
 if you allow them the liberty of the house, they live longer 
 than in a cage, and appear in better si)irits ; but, when 
 you least expect it, they drop down and die in epilepsy. 
 
 Smaller in size, and of colour not so rich and somewhat 
 differently arranged, another species of troupiale sings 
 
 
 
 SECOND 
 
 lOUr.NKV. 
 
 The 
 
 Trou^nale 
 
 Second 
 S))ccics uf 
 
1 1)4 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 HECOND 
 4oUUNi;V. 
 
 Third 
 
 fpccics of 
 
 2'iou^iaic, 
 
 Fourth 
 
 Slacks of 
 
 Troupiah: 
 
 Tnnfjcira 
 apccics. 
 
 Manikin 
 tqncics. 
 
 melodiously in Demerara. The woodcutter is particularly 
 fiivoured by hiiu ; for while the hen is sitting on her nest, 
 built in the roof of the woodcutter's house, he sings for 
 hours together close by : he prefers the forests to the 
 cultivated parts. 
 
 You would not grudge to stop for a few minutes, as you 
 are walking in the plantations, to observe a third species 
 of troupiale : his wings, tail, and throat are black, all the 
 rest of the body is a bright yellow. There is something 
 very sweet and plaintive in his song, though much shorter 
 than that of the troupiale in the interior. 
 
 A fourth species goes in Hocks from place to place in the 
 cultivated parts at the time the Indian corn is ripe ; he is 
 all black, except the head and throat, which are yellow ; his 
 attempt at song is not worth attending to. 
 
 Wherever there is a wild fig-tree ripe, a numerous 
 species of birds, called Tangara, is sure to be on it. There 
 are eighteen beautiful species here. Their plumage is 
 very rich and diversified ; some of them boast six separate 
 colours ; others have the blue, purple, green, and black so 
 kindly blended into each other, that it would be impossible 
 to mark tlieir boundaries ; while others again exhibit them 
 strong, distinct, and abrupt : many of these tangaras have 
 a fine song. They seem to partake much of the nature of 
 our linnets, sparrows, and finches. Some of them are 
 fund of the plantations; others are never seen there, pre- 
 ferring the wild seeds of the forest to the choicest fruits 
 planted by the hand of man. 
 
 On the same fig-trees to which they repair, and often 
 accidentally up and down the forest, you fall in with four 
 species of Manikin. The largest is white and black, with 
 the feathers on the throat remarkably long : the next in 
 size is half red and half black : the third, black, with a 
 white crown : the fourth, black, with a golden crown, and 
 
WANDEUINTJS IX SOUTH AMKRfCA. 
 
 105 
 
 red feathers at the knee. The half reel and lialt* hUick 
 species is the scai''ost. Tht ro is a creek in the Dciueiiiva 
 called Caniouni. About ten minutes IVoni the mouth, you 
 see a common-sized iig-tiec on your rij^iit hand, as you 
 ascend, han<,'ing over water; it hears a very small i'v^ twice 
 a year. When its fruit is ripe, this manikin is on the tree 
 from morn till eve. 
 
 On all the ripe fig-trees in the i'orest you see the Lird 
 called the small Tiger-bird. Like some of our belles and 
 dandies, it has a gaudy vest to veil an ill-shaped body : the 
 throat, and part of the head, are a bright red ; the breast 
 and belly have black spots on a yellow ground ; the wings 
 are a dark green, black, and white ; and the rump and 
 tail black and green. Like the manikin, it has no song : 
 it depends solely upon a showy garment for admiration. 
 
 Devoid, too, of song, and in a still superber garb, the 
 Yawaraciri comes to feed on the same tree. It has a bar 
 like black velvet from the eyes to the beak ; its legs are 
 yellow ; its throat, wings, and tail black ; all the rest of the 
 body a charming blue. Chiefly in the dry savannas, and 
 here and there accidentally in the forest, you see a songles^ 
 yawaraciri still lovelier than the last: his crown is whitish 
 blue, arrayed like a coat of mail : his tail is black, his wings 
 black and yellow ; legs red ; and the whole body a glossy 
 blue. Whilst roving through the forest, ever and anon 
 you see individuals of the wren species, busy amongst the 
 fallen leaves, or seeking insects at the roots of the trees. 
 
 Here, too, you find six or seven species of small birds, 
 whose backs appear to be oveiloaded with silky plumage. 
 One of these, with a chestnut breast, smoke-coloured back, 
 red tail, white feathers like horns on his head, and white 
 narrow-pointed feather under the jaw, feeds entirely upon 
 ants. When a nest of large, light, brown ants emigrates, 
 one following the other in meandering lines above a mile 
 
 () 2 
 
 NKCONI) 
 JOIUNKV. 
 
 r/ie smnll 
 Tiijcr-hird. 
 
 The Ynva- 
 raciri. 
 
li»G 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 SKCONM) 
 JOUUNKY. 
 
 Ants. 
 
 Parrots 
 
 andFarro- 
 
 quels. 
 
 The 
 Uia-hia. 
 
 The Ara. 
 
 long, you see this bird watching thcra, and every now and 
 then picking them up. When they disappear he is seen no 
 more : perhaps this is the only kind of ant he is fond of ; 
 when these ants are stirring, you are sure to find him near 
 them. You cannot well mistake the ant after you have 
 once been in its company, for its sting is very severe, and 
 you can hardly shoot the bird, and pick it up, without 
 liaving five or six upon you. 
 
 Parrots and Paroquets are very numerous here, and of 
 many different kinds. You will know when they are near 
 you in the forest, not only by the noise they make, but also 
 by the fruits and seeds which tliey let fall while they are 
 ieeding. 
 
 The Hia-hia Parrot, called in England the Parrot of the 
 ►Sun, is very remarkable : he can erect at pleasure a fine 
 radiated circle of tartan feathers quite round the back of 
 his head from jaw to jaw. The fore-part of his head is 
 white ; his back, tail, and wings, green ; and his breast and 
 belly tartan. 
 
 Superior in size and beauty to every parrot of South 
 America, the Ara will force you to take your eyes from the 
 rest of animated nature, and gaze at him: his commanding 
 strength, the fiaming scarlet of his body, the lovely variety 
 of red, yellow, blue, and green in his wings, the extraordinary 
 length of his scarlet and blue tail, seem all to join and de- 
 mand for him the title of *' emperor of all the parrots." He 
 is scarce in Demerara till you reach the confines of the 
 ^lacoushi country ; there he is in vast abundance ; he 
 mostly feeds on trees of the palm species. AVhen the cou- 
 courite-trees have ripe fruit on them, they are covered with 
 this magnificent parrot : he is not shy or wary ; you may 
 take your blow-pipe and quiver of poisoned arrows, and kill 
 more than you are able to carry back to your hut. They 
 arc very vociferous, and, like the common parrots, rise up 
 
WANDERINGS IN SOUTH AMERICA. 
 
 ir.7 
 
 SECOND 
 JOUIINKV. 
 
 ill bodies towards sunset, and fly two and two to their 
 place of rest. It is a j^rand si^dit in ornithology to see 
 thousands of aras flying over your head, low enongli to let 
 you have a full view of their flaming mantle. The Indians 
 find their flesh very good, and the feathers serve for 
 ornaments in their head-dresses. They breed in the holes 
 of trees, are easily reared and tamed, and learn to speak 
 pretty distinctly. 
 
 Another species frequents the low lands of Demerara. 
 He is nearly the size of the scarlet ara, but much inferior 
 in plumage. Blue and yellow are his predominant colours. 
 
 Along the creeks and river sides, and in the wet savannas, 
 six species of the Bittern will engage your attention. They 
 are all handsome. The smallest not so large as the English 
 water-hen. 
 
 The 
 li if tern. 
 
 In the savannas, too, you will sometimes surprise the ^'"' •'^"^"'■ 
 snow-white Egrctte, whose back is adorned with the plumes h'ffrctte. 
 from which it takes its name. Here too the spur-winged 
 Water-hen, the blue and green Water-hen, and tvo other 
 species of ordinary plumage, are found. While in quest 
 of these, the Blue Heron, the large and small Brown 
 Heron, the Boat-bill, and Muscovy Duck, now and then 
 rise up before you. 
 
 When the sun has sunk in the western woods, no longer 
 agitated by the breeze ; when you can only see a straggler 
 or two of the feathered tribe hastening to join its mate, 
 already at its roosting-place, then it is that the Goatsucker The Gnat 
 comes out of the forest, where it has sat all day long in 
 slumbering ease, unmindful of the gay and busy scenes 
 around it. Its eyes are too delicately formed to bear the 
 light, and thus it is forced to shun the flaming face of day, 
 and wait in patience till night invites him to partake of the 
 pleasures her dusky presence brings. 
 
 The harmless, unoffending goatsucker, from the time of 
 
 sucker. 
 
 IM 
 
108 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 SKCOND 
 JOUUNEV. 
 
 The 
 2)luinage. 
 
 Aristotle down to the present day, has been in disgrace with 
 man. Father has handed down to son, and author to author 
 tliat this nocturnal thief subsists by milking the flocks. 
 Poor injured little bird of night, how sadly hast thou 
 suffered, and how foul a stain has inattention to facts put 
 upon thy character ! Thou hast never robbed man of any 
 part of his property, nor deprived the kid of a drop of 
 milk. 
 
 AVhen the moon shines bright, you may have a fair 
 opportunity of examining the goatsucker. You will see it 
 close by the cows, goats, and sheep, jumping up every now 
 and then, under their bellies. Approach a little nearer, — 
 he is not shy, " he fears no danger, for he knows no sin." 
 See how the nocturnal flies are tormenting the herd, and 
 with what dexterity he springs up and catches them, as fast 
 as they alight on the belly, legs, and udder of the animals. 
 Observe how quiet they stand, and how sensible they seem 
 of his good offices, for they neither strike at him, nor hit 
 him with their tail, nor tread on him, nor try to drive him 
 away as an uncivil intruder. Wei'e you to dissect him, 
 and inspect his stomach, you woidd find no milk there. It 
 is full of the flies which have been annoying the herd. 
 The prettily mottled plumage of the goatsucker, like that 
 f the owl, wants the lustre which is observed in the 
 feathers of the birds of day. This at once marks him as 
 a lover of the pale moon's nightly beams. There are nine 
 species here. The largest appears nearly the size of the 
 English wood-owl. Its cry is so remarkable, that having 
 once heard it you will never forget it. When night reigns 
 over these immeasurable wilds, whilst lying in your ham- 
 mock, you will hear this goatsucker lamenting like one in 
 deep distress. A stranger would never conceive it to be 
 the cry of a bird. He would say it was the departing 
 voice of a midnight-murdered victim, or the last wailing of 
 
WAXDEIUNGS IN SOUTH AMERICA. 
 
 190 
 
 Niobe for her poor children, before she was turned into skcond 
 stone. Suppose yourself in hopeless sorrow, begin with a Jo^|^^'- 
 high loud note, and pronounce, "ha, ha, ha, ha, ha, ha, ha," 
 eacli note lower and lower, till the last is scarcely heard, 
 pausing a moment or two betwixt every note, and you will 
 liave some idea of the moaning of the largest goatsucker in 
 Demerara. 
 
 Four other species of the goatsucker articulate some 
 words so distinctly, that they have received tlieir names 
 from the sentences they utter, and absolutely bewilder the 
 stranger on his arrival in these parts. The most common 
 one sits down close by your door, and Hies and alights three 
 or four yards before you, as you walk along the road, cry- 
 ing, " Who-are-you, who-who-who-are-you ? " Another bids 
 you, " Work-away, work-work-work-away," A third cries, 
 mournfully, " Willy-come-go. AVilly-AVilly-Willy-come 
 go." And high up in the country, a fourth tells you to 
 " Whip-poor-Will. Whip-whip-whip-poor- Will." 
 
 You will never persuade the negro to destroy these birds 
 or get the Indian to let fly his arrow at them. They are 
 birds of omen and reverential dread. Jumbo, the demon 
 of Africa, has them under his command ; and they equally 
 obey the Yabahou, or Denierara Indian devil. They are 
 the receptacles for departed souls, who come back again to 
 earth, unable to rest for crimes done in their days of nature ; 
 or they are expressly sent by Jumbo, or Yabahou, to haunt 
 cruel and hard-hearted masters, and retaliate injuries re- 
 ceived from them. If the largest goatsucker chance to 
 cry near the white man's door, sorrow and grief will soon 
 be inside ; and they expect to see the master waste awfiy 
 with a slow consuming sickness. If it be heard close to 
 the negro's or Indian's hut, from that night misfortune sits 
 brooding over it ; and they await the event in terrible 
 suspense. 
 
200 
 
 WANDEiaXGS IN SOUTH AMERICA. 
 
 SECOND You will foi-'dve the poor Indian of Guiana for this. 
 ^S-J. ' He knows no better ; lie has nobody to teach him. But 
 shame it is, tliat in our own civilized country, the black cat 
 and broonistaff should be considered as conductors to and 
 from the regions of departed spirits. 
 jtnecdutr. Man}'' years ago I knew poor harmless Mary ; old age 
 liad marked her strongly, just as he will mark you and me, 
 should we arrive at her years and carry the weight of grief 
 which bent her double. The old men of the village said 
 she had been very pretty in her youth ; and nothing could 
 l>e seen more comely than Mary when she danced on the 
 green. He who had gained her heart, left her for another, 
 less fair, though richer than IVIary. From that time she 
 became sad and pensive ; the rose left her cheek, and she 
 was never more seen to dance round the May-pole on the 
 green : her expectations were blighted ; she became quite 
 indifferent to everything around her, and seemed to think 
 of nothing but how she could best attend her mother, who 
 was lame, and not long for this life. Her mother had 
 begged a black kitten from some boys who were going to 
 drown it, and in her last illness she told Mary to be kind 
 to it for her sake. 
 
 When age and want had destroyed the symmetry of 
 Mary's tine form, the village began to consider her as one 
 who had dealings with spirits ; her cat confirmed the sus- 
 picion. If a cow died, or a villager wasted away with an 
 unknown complaint, Mary and her cat had it to answer for. 
 Her broom sometimes served her for a walking-stick ; and 
 if ever she supported her tottering frame with it as far as 
 the May-pole, where once, in youthful bloom and beauty, 
 she had attracted the eyes of all, the boys would surround 
 her, and make sport of her, while her cat had neither friend 
 lior safety beyond the cottage wnll. IS^'obody considered it 
 cruel or uncharittible to torment a witcli ; and it is probable. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 20t 
 
 long before this, that cruelty, old age, and want, have worn skcond 
 her out, and that both poor Mary and her cat have ceased J^uuney, 
 to be. 
 
 Would you wish to pursue the difTerent species of game, 
 well stored and boundless is your range inDemerara. Here 
 no one dogs you, and afterwards clandestinely inquires if 
 you have a hundred a year in land to entitle you to enjoy 
 such patrician sport. Here no saucy intruder asks if 
 you have taken out a licence, by virtue of which you are 
 allowed to kill the birds which have bred upon your own 
 property. Here 
 
 '* You arc as free as when God first made man, 
 Ere the vile laws of servitude Le^an, 
 And wild in woods the noble savage rau." 
 
 Before the morning's dawn you hear a noise in the forest, 
 which sounds like " duraquaura " often repeated. This is the 
 Partridge, a little smaller, and differing somewhat in colour Thr.^ Part- 
 from the English partridge ; it lives entirely in the forest, ' " ■''^' 
 and probably the young brood very soon leave their parents, 
 as you never flush more than two birds in the same place, 
 and in general only one. 
 
 About the same hour, and sometimes even at midnight, Two specks 
 you hear two species of Maani, or Tinamou, send forth i;'^,^,'/"^^^ 
 their long and plaintive whistle from tlie depth of the Tinamou. 
 forest. The flesh of both is delicious. The largest is 
 plumper, and almost equals in size the black cock of 
 Northumberland. The quail is said to be here, though 
 rare. 
 
 The Hannaquoi, which some have compared to the 
 pheasant, though with little reason, is very common. 
 
 Here are also two species of the Powise, or Hocco, and 
 two of tliesm dl wild turkeyf^' called Maroudi ; they feed on 
 the ripe fruits of the forest, ard are found in all directions 
 
 The. 
 Ihiiina' 
 
 quo). 
 
 The. 
 
 rovisr, or 
 
 Hocco. 
 
 
SECOND 
 JOURNEY. 
 
 Flocks of 
 Waraca- 
 has, or 
 Trum- 
 peters. 
 
 in these extensive wilds. You will admire the Horned 
 Screamer as a stately and majestic bird : he is almost the 
 size of the turkey cock ; on his head is a long slender horn, 
 and each wing is armed with a strong, sharp, triangular 
 spur, an inch long. 
 
 Sometimes you will fall in with flocks of two or three 
 hundred Waracabas, or Trumpeters, called so from the 
 singular noise they produce. Tlieir breast is adorned with 
 beautiful changing blue and purple feathers ; their head 
 and neck like velvet ; their wings and back grey, and belly 
 black. They run with great swiftness, and when domesti- 
 cated, attend their master in his walks with us much ap- 
 parent affection as his dog. They have no spurs, but still, 
 such is their high spirit and activity, that tliey browbeat 
 every dunghill fowl in the yard, and force the Guinea birds, 
 dogs, and turkeys to own their superiority. 
 
 If, kind and gentle reader, thou shouldst ever visit these 
 regions with an intention to examine their productions, 
 perhaps the few observations contained in these Wander- 
 ings may be of service to thee; excuse their brevity: 
 more could have been M-ritten, and each bird more par- 
 ticularly described, but it would have been pressing too 
 hard upon thy time and patience. 
 
 Soon after arriving in these parts, thou wilt find that the 
 species here enumerated are only as a handful from a well- 
 stored g''?tnary. Nothing lias been said of the Eagles, the 
 Falcons, the Hawks, and Shrikes; nothing of the different 
 species of Vultures, the king of which is very handsome, 
 and seems to be the only bird which claims regal honours 
 from a surrounding tribe. It is a fact beyond all dispute, 
 that 'ien the scent of carrion has drawn tog- iJu;. hundreds 
 of t i common vultures, they all retire from the carcass as 
 soon as the King of the Vultures makes his appearance. 
 When his majesty has satisfied the cravings of his royal 
 
WANDERINGS IN SOUTH AMERICA. 
 
 203 
 
 stomach with the choicest bits from the most stinking and second 
 corrupted parts, lie generally retires to a neighbouring tree, J^ ^'''>^'''^ ^^' 
 and then the common vultures return in crowds to gobble 
 down his leavings. The Indians, as well as the whites, 
 have observed this ; for when one of them, who has learned 
 a little English, sees the king, and wishes you to have a 
 proper notion of the bird, he says, " There is the governor 
 of the carrion crows." 
 
 Now, the Indians have never heard of a personage in 
 Demerara higher than that of governor ; and the colonists, 
 through a common mistake, call the vultures carrion crows. 
 Hence the Indian, in order to express the dominion of this 
 bird over the common vultures, tells you he is governor of 
 the carrion crows. The Spaniards have also observed it, 
 for, through all the Spanish Main, he is called Eey de 
 Zamuros, King of the Vultures. The many species of Owls, 
 too, have not been noticed ; and no mention made of the 
 Columbine tribe. The prodigious variety of Water Fowl 
 on the sea-shore has been but barely hinted at. 
 
 There, and on" the borders and surface of the inland 
 waters, in the marshes and creeks, besides the flamingos, 
 scarlet curlews, and spoonbills, already mentioned, will be 
 found Greenish-Brown Curlews, Sandpipers, Eails, Coots, 
 Gulls, Pelicans, Jal a-us, Kandapoas, Crabiers, Snipes, 
 Plovers, Ducks, Geese, Cranes, and Anhingas ; most of 
 them in vast abundance ; some frequenting only the sea- 
 coast, others only the interior, according to their different 
 natures ; all worthy the attention of the naturalist, all 
 worthy of a place in the cabinet of the curious. 
 
 Should thy comprehensive genius not confine itself to 
 birds alone, grand is the appearance of other objects all 
 around. Thou art in a land rich in botany and mineralogy, 
 rich in zoology and entomology. Animation will glow in 
 thy looks, and exercise will brace thy frame in vigour 
 
 I 
 
 
204 
 
 WANDERINGS IN SOUTH A:\IERICA. 
 
 sEcoxn The very time of tliy absence from the tables of hetero- 
 
 jomiNKv. geneous luxury will be profitable to thy stomach, perhaps 
 
 already sorely drenched with Londo-Parisian sauces, and a 
 
 new stock of health will bring thee an appetite to relish 
 
 the wholesome food of the chase ; never-failing sleep will 
 
 wait on thee at the time she comes to soothe the rest of 
 
 animated nature ; and, ere the sun's rays appear in the 
 
 horizon, thou wilt spring from thy hammock fresh as the 
 
 April lark. ]3e convinced also, that the dangers and difh- 
 
 culties which are generally supposed to accompany the 
 
 traveller in his journey through distant regions, are not 
 
 half so numerous or dreadful as they are connnonly 
 
 thought to be. 
 
 Danprr.i fo The youtli wlio incautiously reels into the lobby of 
 
 hm7S'voi Drury-lane, after leaving the table sacred to the god of 
 
 real but wine, is exposed to more certain ruin, sickness, and decay, 
 
 imaginaiy. ^^^^^ j^^ ^^,|^^ wanders a whole year in the wilds of Deme- 
 
 rara. But this will never be believed ; because the disasters 
 arising from dissipation are so connnon and frequent in 
 civilized life, that man becomes quite habituated to them ; 
 and sees daily victims sink into the tomb long before their 
 time, without ever once taking alarm at the causes which 
 precipitated them headlong into it. 
 
 But the dangers which a traveller exposes himself to in 
 foreign parts are novel, out of the way things to a man at 
 home. The remotest apprehension of meeting a tremen- 
 dous tiger, of being carried off l)y a Hying dragon, or hav- 
 ing his bones picked by a famished cannibal; oh, that 
 makes him shudder. It sounds in his ears like the burst- 
 ing of a bomb-shell. Thank Heaven, he is safe by his own 
 fire-side ! 
 
 Prudence and resolution ought to be the traveller's con- 
 stant companions. The first will cause him to avoid a 
 number of snares which he will find in the path as he 
 
WANDERINGS IN SOUTH AMERICA. 
 
 205 
 
 journeys on ; and tlio second will iilways lend a hand to ski.oi.'I) 
 assist Idm, if lie has unavoidably got entangled in them. J"^'"^'';^"- 
 The little distinctions which have been shown him at his 
 own home ought to be forgotten when he travels over the 
 world at large ; for strangers know nothing of his former 
 merits, and it is necessary that they should witness them 
 before they pay him the tribute which he was wont to re- 
 ceive within his own doors. Thus, to be kind and alTable 
 to those we meet, to mix in their amusements, to pay a 
 compliment or two to their manners and customs, to respect 
 their elders, to give a little to their distressed and needy, 
 and to feel, as it were, at home amongst them, is the 
 sure way to enable you to pass merrily on, and to find 
 other comforts as sweet and palatable as those which you 
 were accustomed to partake of amongst your friends and 
 acquaintance in your own native land. 
 
 We will now ascend in fancy on Icarian wing, and 
 take a view of Guiana in general. See an inunense 
 plain! betwixt two of the largest rivers in the world, 
 level as a bowling-green, save at Cayenne, and covered 
 with trees along the coast quite to the Atlantic wave, 
 except where the plantations make a little vacancy 
 amongst the foliage. 
 
 Though nearly in the centre of the torrid zone, the sun's 
 rays are not so intolerable as might be imagined, on account 
 of the perpetual verdure and refreshing north-east breeze. 
 See what numbers of broad and rapid rivers intersect it in 
 their journey to the ocean, and that not a stone or a pebble 
 is to be found on their banks, or in any part of the country, 
 till your eye catches the hills in the interior. How beauti- 
 ful and magnificent are the lakes in the heart of the forests, 
 and how charming the forests themselves, for miles after 
 miles on each side of the rivers ! llow extensive appear 
 the savannas or natural meavlows, teeming w'ith innumer- 
 
 r 
 
 ih 
 
 I 1 
 
2011 
 
 WANDERINGS IN SOUTH AMEIIICA. 
 
 iit'l 
 
 SECOND able herds of cattle, where the Portuguese and Spaniards 
 jounxKY. j^j.g settled, but desert as Saara, where the English and 
 Dutch claim dominion ! How gradually the face of the 
 country rises ! See the sand-hills all clothed in wood first 
 emerging from the level, then hills a little higher, rugged 
 with bold and craggy rocks, peeping out from amongst the 
 most luxuriant timber. Then come plains, and dells, and 
 far-extending valleys, arrayed in richest foliage ; and beyond 
 them, mountains piled on mountains, some bearing pro- 
 digious forests, others of bleak and barren aspect. Thus 
 your eye wanders on, over scenes of varied loveliness and 
 grandeur, till it rests on the stupendous pinnacle?; of the 
 long-continued Cordilleras de los Andes, which rise in 
 towering majesty, and command all America. 
 
 How fertile must the low-lands be, from the accumula- 
 tion of fallen leaves and trees for centuries ! How pro- 
 pitious the swamps and slimy beds of the rivers, heated 
 by a downward sun, to tlie amazing growth of alligators, 
 serpents, and innumerable insects ! How inviting the 
 forests to the feathered tribes, where you see buds, blos- 
 soms, green and ripe fruit, full grown and fading leaves, all 
 on the same tree I How secure the wild beasts may rove 
 in endless mazes ! Perhaps those mountains too, which 
 appear so bleak and naked, as if quite neglected, are, like 
 Potosi, full of precious metals. 
 
 Let us now return the pinions we borrowed from Icarus, 
 and prepare to bid farewell to the wilds. The time allotted 
 to these Wanderings is drawing fast to a close. Every day 
 for the last six months has been employed in paying close 
 attention to natural history in the forests of Demerara. 
 Above two hundred specimens of the finest birds have been 
 collected, and a pretty just knowledge formed of their 
 haunts and economy. From the time of leaving England, 
 in March, 1816, to the present day, nothing has intervened 
 
 Con- 
 clusion. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 to arrest a fine flow of health, saving a quartan ague, which 
 did not tarry, but fled as suddenly as it appeared. 
 
 And now I take leave of tliee, kind and gentle reader. 
 The new mode of preserving birds, heretofore promised 
 thee, shall not be forgotten. The plan is already formed in 
 imagination, and can be penn- d down during the passage 
 across the Atlantic. If the few remarks in these Wander- 
 ings shall have tiny weight in inciting thee to sally forth 
 and explore the vast and well-stored regions of Demerara, 
 I have gained my eixd. Adieu. 
 
 20 
 
 JOUUNEY. 
 
 Charles Waterton. 
 
 j4j>ril(\, IS] 7. 
 
 fi 
 
 !i! 
 
 .ii 
 
 ' :! 
 
 . 1 
 
 I 
 
 It 
 
M!i: 
 
 THIRD 
 
 JOUllNRY 
 
 TIIIllD JOUllNEY. 
 
 ClIAITKll I. 
 
 " Dest'i'tosque vitlerc locos, littiisijue nliL'tiiui." 
 
 From the Clyde to Denierara. — Yellow fever. — A deserteil riantation. — 
 Black John. — Mcdicinea for tropical climates. — Hats. — The Lancet. — 
 Severe accident and recovery. — A primitive Spear. — History of the 
 Sloth — an inluihitant of the trees — structure of the limbs. — A 
 domesticated Sloth — a life of suspense — structure of the hair — 
 mark on the back. — Capture of a Sloth. — Release and escape. — Ants. 
 Ant-bears. — The great Ant-bear — its powers of defence — attitude when 
 standing — how it catches its prey. — Glutinous saliva. — The Vampire 
 Hnd its habits. — Bleeding gratis. — Coushie Ants. — Armatlillo and its 
 habits. — Tortoise. — Eggs of Tortoise and Turtle. 
 
 Gentle reader, after staying a few months iu England, I 
 strayed across the Alps and the Apennines, and returned 
 home, but could not tarry. Guiana still whispered in my 
 ear, and seemed to invite me once more to wander through 
 her distant forests. 
 
 Shouldst thou have a leisure hour to read what follows, 
 I pray thee pardon the frequent use of that unwelcome 
 monosyllable /. It could not well be avoided, as will be 
 seen in the sequel. In February, 1820, 1 sailed from the 
 Clyde, on board the Gknhcrvic, a fine West-Indiaman. 
 She was driven to the north-west of Ireland, and had to 
 contend with a foul and wintry wind for above a fortnight. 
 
\VANUK1UN(JS IN SUUTH AMERICA. 
 
 At last it changed, uii I we Imd a pleasant passage across 
 the Atlantic. 
 
 S:\d and mournful was the story we heard on entering 
 the river Denierara. The yellow fever had swept oil' 
 numbers of the old inhabitants, and the mortal remains of 
 many a new comer were daily passing down the streets, in 
 slow and mute procession to their last resting-place. 
 
 After staying a few days in the town, I went up the 
 Demerara to the former habitation of my worthy friend, 
 Mr. Edmonstone, in Mibiri creek. 
 
 The house had been abandoned for some years. On 
 arriving at the hill, the remembrance of scenes long past 
 and gone naturally broke in upon the mind. All was 
 changed ; the house was in ruins, and gradually sinking 
 under the influence of the sun and rain; the roof had 
 nearly fallen in ; and the room where once governors and 
 generals had caroused, was now dismantled, and tenanted 
 by the vampire. You would have said, 
 
 2(t9 
 
 in I up 
 
 JOl'llNKV. 
 
 Ydloii) 
 
 fever at 
 
 Dcmcraro. 
 
 Ill's! deurc 
 
 at Mibiri 
 
 creek. 
 
 1 '•" 
 
 I 
 
 ii<>ht. 
 
 " 'Tis now the vampire's bleak nboile, 
 'Tis now the ai«vitiiieiit of the toad ; 
 'Tis hero the paiiit'iil Chfgoe feeds, 
 'Tis here the dire Labarri breeds, 
 Couceal'd in ruins, moss, and weeds." 
 
 On the outside of the house, nature had nearly re- 
 assumed her ancient right : a few straggling fruit-trees were 
 still discernible amid the varied hue of the near approach- 
 ing forest ; they seemed like strangers lost, and bewildered, 
 and unpitied, in a foreign land, destined to linger a little 
 longer, and then sink down for ever. 
 
 I hired some negroes from a woodcutter inanother creek Converted 
 to repair the roof ; and then the house, or at least what ^aiuiL^s 
 remained of it, became head-quarters for natural history, dwcliimj. 
 The frogs, and here and there a snake, received that 
 
 V 
 
THIRD attention which the weak in this world generally experience 
 jo iTRNF. Y. i'yq^t^ ^]^q strong, and which the law commonly denominates 
 an ejectment. But here, neither the frogs nor serpents 
 were ill-treated ; they sallied forth, without buffet or rebuke, 
 to choose their place of residence ; the world w^as all before 
 them. The owls went away of their own accord, prefer- 
 ring to retire to a hollow tree rather than to associate with 
 their new landlord. The bats and vampires stayed with 
 me, and went in and out as usual. 
 
 It was upon tliis hill in former days that I first tried to 
 teach John, the black slave of my friend Mr. Edmonstone, 
 the proper way to do birds. But John had poor abilities, 
 and it required much time and patience to drive anything 
 into him. Some years after this his master took him to 
 Scotland, wdiere, becoming free, John left him, and got em- 
 ployed in the Glasgow, and then the Edinburgh museum. 
 Mr. Eobert Edmonstone, nephew to the above gentleman, 
 had a fine mulatto capable of learning anything. He re- 
 quested me to teach him the art. I did so. He was docile 
 and active, and was with me all the time in the forest ; I 
 left him there to keep up this new art of preserving birds, 
 and to communicate it to others. Here then I fixed my 
 head-quarters, in the ruins of this once gay and hospitable 
 house. Close by, in a little hut, which in times long past 
 had served for a store to keep provisions in, there lived a 
 coloured man and his wife, by name Backer. J\Iany a kind 
 turn they did to me ; and I was more than once of service 
 to them and their children, by bringing to their relief in 
 time of sickness what little knowledge I had acquired of 
 medicine. 
 
 I would here, gentle reader, wish to draw thy attention, 
 for a few minutes, to physic, raiment, and diet. Shouldst 
 thou ever wander through these remote and dreary wilds, 
 forget not to carry with thee bark, laudanum, calomel, and 
 
 Raiment 
 and did. 
 
WANDEltlNGS IN SOUTH A^IEKICA. 
 
 211 
 
 jalap, and the lancet. There are no druggist shops here, third 
 nor sons of Galen to apply to in time of need. I never go Journry. 
 encumbered with many clothes. A thin flannel waistcoat 
 under a check shirt, a pair of trousers, and a hat, were all 
 my wardrobe ; shoes and stockings I seldom had on. In 
 dry weather they would have irritated the feet, and retarded 
 me in the chase of wild beasts ; and in the rainy season 
 they would have kept me in a perpetual state of damp and 
 moisture. I eat moderately, and never drink wine, spirits, 
 or fermented liquors in any climate. This abstemiousness 
 has ever proved a faithful friend ; it carried me triumphant 
 through the epidemia at INIalaga, where death made such 
 havoc about the beginning of the present century ; and it 
 has since befriended nie in many a fit of sickness, brought 
 on by exposure to the noon-day sun, to the dews of night, 
 to tlie pelting shower and unwholesome food. 
 
 Perhaps it will V)e as well, here, to mention a fever which 
 came on, and the treatment of it ; it may possibly be of use 
 to thee, shouldst thou turn M'anderer in the tropics : a word 
 or two also of a wound I got in the forest, and then we vvill 
 say no more of the little accidents which sometimes occur, 
 and attend solely to natural history. We shall have an op- 
 portunity of seeing the wild animals in their native haunts^ 
 undisturbed and unbroken in upon by man. AVe shall 
 have time and leisure to look more closely at them, and 
 probably rectify some errors which, for want of proper in- 
 formation OT- a near observance, have crept into their 
 several histories. 
 
 It was in the month of June, when the sun was within 
 a few days of Cancer, that I had a. severe attack of fever. 
 There had been a deluge of rain, accompanied with tre- 
 mendous thunder and lightning, and very little sun. 
 Nothing could exceed the dampness of the atmosphere. 
 Fur two or three days I had been in a kind of twilight 
 
 V '> 
 
 Severe 
 
 attack of 
 
 fcvei'. 
 
 1:1 
 
 ,!,■ 
 
 |i: 
 
21-2 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 THiRi> state of health, neither ill nor what you may call well ; I 
 juuHNEY. ya^yjn3(^| fjjj(j fgjt weary without exercise, and my sleep was 
 
 merely slumber. This was the time to have taken medi- 
 cine ; hut I neglected to do so, though I had just been 
 reading, " navis referent in mare te novi fluctus, O quid 
 agis ? fortiter occupa portum." I awoke at midnight ; a 
 cruel headache, thirst, and pain in the small of the back, 
 informed me what the case was. Had Chiron himself been 
 present, he could not have told me more distinctly that I 
 was going to have a tight brush of it, and that I ought to 
 meet it with becoming fortitude. I dozed, and woke, and 
 startled, and then dozed again, and suddenly awoke, think- 
 ing I was falling down a precipice. 
 
 The return of the bats to their diurnal retreat, which was 
 in the thatch above my hammock, informed me that the 
 sun was now fast approaching to the eastern horizon. I 
 arose, in languor and in pain, the pulse at one hundred 
 and twenty. I took ten grains of calomel and a scruple of 
 jalap, and drank during the day large draughts of tea, weak 
 and warm. The physic did its duty ; but there was no re- 
 mission of fever or headache, though the pain of the back 
 Avas less acute. I was saved the trouble of keeping the 
 room cool, as the wind beat in at every quarter. 
 
 At five in the evening the pulse had risen to one hundred 
 and thirty, and the headache almost insupportable, especi- 
 ally on looking to the right or left. I now opened a vein, 
 and made a large orifice, to allow the blood to rush out 
 rapidly; I closed it after losing sixteen ounces. I then 
 steeped my feet in warm water, and got into the hammock. 
 After bleeding, the pulse fell to ninety, and the head was 
 much relieved ; but during the night, which was very rest- 
 less, the pulse rose again to one hundred and twenty, and 
 at time s the headache was distressing. I relieved the head- 
 ache from time to time by applying cold water to the 
 
WANDERINGS IN SOUTH AMERICA. 
 
 213 
 
 temples, and holding a wet handkerchief there. The next third 
 morning the fever ran very high, and I took five more •"^ ^'''^''^ ^'' 
 grains of calomel and ten of jalap, determined, whatever 
 might be the case, this should be the last dose of calomel. 
 About two o'clock in the afternoon the fever remitted, and 
 a copious perspiration came on ; there was no more head- 
 ache, nor thirst, nor pain in the back, and the following 
 night was comparatively a good one. The next morning I 
 swallowed a large dose of castor-oil : it was genuine, for 
 Louisa Backer had made it from the seeds of the trees 
 which gxew near the door. I was now entirely free from 
 all symptoms of fever, or apprehensions of a return ; and 
 the mornin<5 after I began to take bark, and continued it 
 for a fortnight. This put all to rights. 
 
 The story of the wound I got in the forest, and the mode MeeU u-ith 
 of cure, are very short. — I had pursued a red-headed wood- "f' 
 pecker for above a mile in the forest, without being able to 
 get a shot at it. Thinking more of the woodpecker, as I 
 ran along, thfin of the way before me, I trod upon a little 
 hardwood stump, which was just about an inch or so above 
 the ground ; it entered the hollow part of my foot, making 
 a deep and lacerated wound there. It had brought me to 
 the ground, and there I lay till a transitory fit of sickness 
 went off. I allowed it to bleed freely, and on reacliing head- 
 quarters, washed it well and probed it, to feel if any foreign 
 body was left within it. Being satisfied that there was 
 none, I brought the edges of tlie vround together, and then 
 put a piece of lint on it, and over that a very large poultice, 
 which was changed morning, noon, and night. Luckily, 
 Backer had a cow or two upon the hill : now as heat and 
 moisture are the two principal virtues of a poultice, nothing 
 could produce those two qualities better than fresh cow- 
 dung boiled : had there been no cows there, I could have 
 made it with boiled grass and leaves. I now took 
 
 , Jit 
 
 1 I 
 
WANDERINGS IN SOUTH AMERICA. 
 
 THIRD 
 JOURNFA', 
 
 entirely to the haiiiinock, placing the foot higher than the 
 knee ; this prevented it from throbbing, and was, indeed, 
 the only position in which I could be at ease. When the 
 inflammation was completely subdued, I applied a wet 
 cloth to the wound, and every now and then steeped the 
 foot in cold water during the day, and at night again ap- 
 plied a poultice. The wound was now healing fast, and in 
 three weeks from the time of the accident, nothing but a 
 scar remained ; so that I again sallied forth sound and 
 joyful, and said to myself — 
 
 * i' 
 
 Last Con- 
 versation 
 with Sir 
 Joseph 
 Banks. 
 
 " I, pedes quo to rapiunt et auraj 
 Dum lavet sol, et locus, i secuudo 
 Otuine, et conto latebras, ut olim, 
 
 Itumpe feranun." 
 
 Now, this contus was a tough light pole, eight feet long, on 
 the end of which was fixed an old bayonet. I never went 
 into the canoe without it ; it was of great use in starting 
 the beasts and snakes out of the hollow trees, and, in case 
 of need, was an excellent defence. 
 
 In 181!), I had the last conversation with Sir Joseph 
 Banks. I saw with sorrow that death was going to rob us 
 of him. We talked much of the present mode adopted by 
 all museums in stuffing quadrupeds, and condemned it as 
 being very imperfect : still we could not find out a better 
 way ; and at last concluded, that the lips and nose ought 
 to be cut off, and replaced with wax ; it being impossible 
 to make those parts appear like life, as they shrink to 
 nothing, and render the stuffed specimens in the different 
 museums horrible to look at. The defects in the legs and 
 feet would not be quite so glaring, being covered with 
 hair. 
 
 I had paid great attention to this subject for above 
 fourteen years ; still it would not do ; however, one night 
 
WANDEUINCJS IN SOUTH AMERICA. 
 
 215 
 
 while I was lying in the hammock, and harping on the 
 string on which hung all my solicitude, I hit upon the 
 proper mode by inference ; it appeared clear to me that it 
 was the oidy true way of going to work, and ere I closed 
 my eyes in sleep, I was able to prove to myself that there 
 could not be any other way that would answer. I tried it 
 the next day, and succeeded according to expectation. 
 
 By means of this process, which is very simple, we can 
 now give every feature back again to the animal's face, 
 after io has been skinned ; and when necessary, stamp 
 grief, or pain, or pleasure, or rage, or mildness upon it. 
 But more of this hereafter. 
 
 Let us now turn our attention to the Sloth, whose native 
 haunts have hitherto been so little known, and probably 
 little looked into. Those \\ho have written on this singu- 
 lar animal, have remarked that he is in a perpetual state of 
 pain, that he is proverbially slow in his movements, that he 
 is a prisoner in space, and that as soon as he has consumed 
 all the leaves of the tree upon which he had mounted, he 
 rolls himself up in the form of a ball, and then falls to the 
 ground. This is not the case. 
 
 If the naturalists who have written the history of the 
 sloth had gone into the wilds, in order to examine his 
 haunts and economy, they would not have drawn the fore- 
 going conclusions ; they would have learned, that though 
 all other quadrupeds may be described while resting upon 
 the ground, the sloth is an exception to this rule, and that 
 his history must be written while he is in the tree. 
 
 This singular animal is destined by nature to be pro- 
 duced, to live and to die in the trees ; and to do justice to 
 him, naturalists must examine him in tliis his upper ele- 
 ment. He is a scarce and solitary animal, and being good 
 food, he is never allowed to escape. He inhabits remote 
 and gloomy forests, where snakes take up their abode, and 
 
 TlIIUl) 
 JOUKNEV. 
 
 Stuffing 
 
 hinls aad 
 
 qnadrii' 
 
 yeds. 
 
 T/ic Sloth. 
 
 Lives in 
 glooiny 
 forests. 
 
 \\\ 
 
 m 
 
 H^ 
 
 \l 
 
'210 
 
 WANDERINdS IN SOUTH AMERICA. 
 
 m 
 
 Tiiiui) "wliere cruelly stinging ants .and scorpions, and swamps, 
 jouRNKY. j^j^^j innunievable thorny shrubs and bushes, obstruct 
 the steps of civilised man. Were you to draw your 
 own conclusions from the descriptions which have been 
 given of the sloth, you would probably suspect, that no 
 naturalist has actually gone into the wilds with the fixed 
 detenninUion to find him out and examine his haunts, 
 and see whether nature has com.aitted any blunder in the 
 formation of this extraordinary creature, which appears 
 to us so forlorn and miserable, so ill put together, and so 
 totally unfit to enjoy the blessings which have been so 
 bountifully given to the rest of animated nature ; for, as 
 it has formerly been remarked, he has no soles to his feet, 
 and he is evidently ill at ease when he tries to move on 
 the ground, and it is then that he looks up in your face 
 with a countenance that says, " Have pity on me, for I am 
 in pain and sorrow." 
 
 It mostly happens that Indians and Negroes are the 
 people who catch the sloth, and bring it to the white man : 
 hence it may be conjectured that the erroneous accounts 
 we have hitherto had of the sloth, have not been penned 
 down with the slightest intention to mislead the reader, 
 or give him an exaggerated history, but that these errors 
 have naturally arisen by examining the sloth in those 
 places where nature never intended that he should be 
 exhibited. 
 
 However, we are now in his own domain. Man but 
 little frequents these thick and noble forests, which 
 extend far and wide on every side of us. This, then, 
 is the proper place to go in quest of the sloth. We will 
 first take a near view of him. By obtaining a knowledge 
 Anatomy of his anatomy, we shall be enabled to account for his 
 %()ti^. movements hereafter, when we see him in his proper 
 haunts. His fore-legs, or, more correctly speaking, his 
 
' liis 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 217 
 
 / 
 
 arms, are apparently much too long, while his hind-legs thiud 
 are very short, and look as if they could be bent almost J^^^^^- 
 to the shape of a corkscrew. Both the fore and hind legs, 
 by their form, and by the manner in which they are joined 
 to the body, are quite incapacitated from acting in a 
 perpendicular direction, or in supporting it on the earth, 
 as the bodies of other quadrupeds are sup sorted, by their 
 legs. Hence, when you place him on the floor, his belly 
 touches the ground. Now, granted that he supported 
 himself on his legs like other animals, nevertheless he 
 would be in pain, for he has no soles to his feet, and his 
 claws are very sharp and long, and curved ; so that, were 
 his body supported by his feet it would be by their 
 extremities, just as your body would be, were you to throw 
 yourself on all fours, and try to support it on the ends of 
 your toes and fingers — a trying position. Were the floor 
 of glass, or of a polished surface, the sloth would actually 
 be quite stationary ; but as the grouud is generally rough, 
 with little protuberances upon it> such as stones, or rDots 
 of grass, &c., this just suits the sloth, and he moves his 
 fore-legs in all directions, in order to find something to 
 lay hold of ; and when he Las succeeded, he pulls himself 
 forward, and is thus enabled to travel onwards, but at the 
 same time in so tardy and awkward a manner, as to acquire 
 him the name of Sloth. 
 
 Indeed his looks and his gestures evidently betray his 
 uncomfortable situation ; and as a sigh every now and 
 then escapes him, W3 may be entitled to conclude that he 
 is actually in pain. 
 
 Some years ago I kept a sloth in my room for several 
 months. I often took him out of the house and placed 
 him upon the ground, in order to have an opportunity of 
 observing his motions. If the ground were rough, he 
 would pull himself forwards, by means of his fore-legs, at 
 
218 
 
 WANDEl{IN(5S IN SOUTH AMEIUCA. 
 
 THIUD 
 JOURNKY, 
 
 a pretty good pace ; and he invariably immediately shaped 
 his course towards the nearest tree. JJut if I put him upon 
 a smooth and well-trodden part of the road, he appeared 
 to be in trouble and distress : his favourite abode was 
 the back of a chair : and after getting all his legs in a 
 line upon the topmost ]_)art of it, he would hang there 
 for hours together, and often with a low and inward 
 cry, would seem to invite me to take notice of him. 
 
 The sloth, in its wild state, spends its whole life in trees, 
 and never leaves them but through force or by accident. 
 An all-iuling I'rovidence has ordered man to tread on the 
 surface of the earth, the eagle to soar in the expanse of 
 the skies, and the monkey and squirrel to inhabit the 
 trees : still these may change their relative situations 
 without feeling nmch inconvenience : but the sloth is 
 doomed to spend his whole life in the trees ; and, what is 
 more extraordinary, not njjon the branches, like the squirrel 
 and the monkey, but under them. He moves suspended 
 from the branch, he rests suspended from it, and he sleeps 
 suspended from it. To enable him to do this, he must 
 have a very different formation from that of any other 
 known quadruped. 
 
 Hence, his seemingly bungled conformation is at once 
 accounted for ; and in lieu of the sloth leading a painful 
 life, and entailing a melancholy and miserable existence 
 on its progeny, it is but fair to surmise that it just enjoys 
 life as much as any other animal, and that its extra- 
 ordinary formation and singular habits are but further 
 proofs to engage us to admire the wonderful works of 
 Omnipotence. 
 
 It must be observed, that the sloth does not hang head- 
 downwards like the vampire. "When asleep, he supports 
 himself from a branch parallel to the earth. He first seizes 
 the branch with one arm, and then with the other ; and 
 
WAXDElJiNdS IN SOUTH AMERICA. 
 
 219 
 
 after that, brings w^ ^)otli his legs, one by one, to the sauio third 
 branch ; so that an lOur are in a line : he seems perfectly at JQ '-''^nk y. 
 rest in this jiosition. Now, had he a tail, he would be at a 
 loss to know what to do with it in this position : were he to 
 draw it up within his legs, it would interfere with them ; 
 and were he to let it hang down, it would become the 
 sport of the winds. Thus his deficiency of tail is a benefit 
 to him ; it is merely an apology for a tail, scarcely 
 exceeding an inch and a half in length. 
 
 I observed, when he was climbing, he never used his 
 arms both together, but first one and then the other, and 
 so on alternately. There is a singularity in his hair, 
 different from that of all other animals, and, I believe, 
 hitherto unnoticed by naturalists; his hair is thick and 
 coarse at the extremity, and gradually tapers to the root, 
 where it becomes fine as a spider's web. His fur has so 
 umch the hue of the moss which grows on the branches 
 of the trees, that it is very difficult to make him out when 
 he is at rest. 
 
 The male of the three-toed sloth has a longitudinal bar 
 of very fine black hair on his back, rather lower than the 
 shoulder-blades ; on each side of this black bar there is a 
 space of yellow hair, equally fine ; it has the appearance 
 of being pressed into the body, and looks exactly as if it 
 had been singed. If we examine the anatomy of his fore- 
 legs, we shall immediately perceive by their firm and 
 muscular texture, how very capable they are of supporting 
 the pendent weight of his body, both in climbing and at 
 rest ; and, instead of pronouncing them a bungled com- 
 position, as a celebrated naturalist has done, we shall 
 consider them as remarkably well calculated to perform 
 their extraordinary functions. 
 
 As the sloth is an inhabitant of forests within the 
 tropics, where the trees touch each other in the greatest 
 
 > A 
 
 I D 
 
 n 
 
220 
 
 \VANL)EUIN(JS IN SOUTH AMKRICA. 
 
 TMiRu profusion, there seems to be no reason why he sliould 
 JOURNEY. cQjjfjjjj, himself to one tree alone for food, and entirely 
 strip it of its leaves. During the many years I have 
 ranged the forests, I have never seen a tree in such a 
 state of nudity ; indeed, I would hazard a conjecture, 
 that, by the time the animal had finished the last of the 
 old leaves, there would be a new crop on the part of the 
 tree he had stripped first, ready for him to begin again, so 
 quick is the process of vegetation in these countries. 
 
 There is a saying amongst the Indians, that when the 
 wind blows, the sloth begins to travel. In calm weather 
 he remains tranquil, probably not liking to cling to the 
 brittle extremity of the branches, lest they should break 
 with him in passing from one tree to another ; but as soon 
 as the wind rises, the branches of the neighbouring trees 
 become interwoven, and then the sloth seizes hold of them, 
 and pursues his journey in safety. There is seldom an 
 entire day of calm in these forests. The trade-wind 
 generally sets in about ten o'clock in the morning, and 
 thus the sloth may set off after breakfast, and get a con- 
 siderable way before dinner. He travels at a good round 
 pace ; and were you to see him pass from tree to tree, as I 
 have done, you would never think of calling him a sloth. 
 
 Thus, it would appear that the different histories we 
 have of this quadruped are erroneous on two accounts : 
 first, that the writers of them deterred by difficulties and 
 local annoyances, have not paid sufficient attention to him 
 ' in his native haunts ; and secondly, they have described 
 
 him in a situation in which he was never intended by 
 nature to cut a figure ; I nean on the ground. The sloth 
 is as much at a loss to proceed on his journey upon a 
 smooth and level floor, as a man would be who had to 
 walk a mile in stilts upon a line of feather beds. 
 
 One day, as we were crossing the Essequibo, I saw a 
 
WAXDKULNGS IX SOUTH AMElilCA. 
 
 221 
 
 lurj^e two-toed sloth on tlio ground upon the l>ank ; liow he thikd 
 had got there nobody could tell : the Indian said he had Jt^uj^^Y. 
 never surprised a sloth in such a situation before : he ^'*« "/<"'• 
 would hardly have come there to drink, for both above 
 and below the place, the branches of the trees touched the 
 water, a,id afforded him an easy and safe access to it. Be 
 this as it nuiy, though the trees were not above twenty 
 yards from him, he could not make his way through the 
 sand time enough to escape before we landed. As soon as 
 we got up to him he threw himself upon his back, and 
 defended himself in gallant style with his fore-legs. 
 " Come, poor fellow," said I to him, " if thou hast got into 
 a hobble to Jay, thou shalt not suffer for it : I'll take no 
 advantage of thee in misfortune ; the forest is large enough 
 both for thee and me to rove in : go thy ways up above, 
 and enjoy thyself in these endless wilds ; it is more than 
 probable thou wilt never have another interview with 
 man. So fare thee well." On saying this, I took a long 
 stick which was lying there, held it for him to hook on, 
 and then conveyed him to a high and stately mora. He 
 ascended with wonderful rapidity, and in about a minute 
 he was almost at the top of the tree. He now went off in 
 a side direction, and caught hold of the branch of a neigh- 
 bouring tree ; he then proceeded towards the heart of the 
 forest. I stood looking on, lost in amazement at his 
 singular mode of progress. I followed him with my eye 
 till the intervening branches closed in betwixt us ; and 
 then I lost sight for ever of the two-toed sloth. I was 
 going to add, that I never saw a sloth take to his heels in 
 such earnest ; but the expression will not do, for the sloth 
 has no heels. 
 
 That which naturalists have advanced of his being so 
 tenacious of life is perfectly true. I saw the heart of one 
 beat for half an hour after it was taken out of the body. 
 
1 i1 
 
 
 222 
 
 Tiiinn 
 
 JOURNKY 
 
 Ants, 
 
 WANDKlUNdS IN SOUTH AMKIUC'A. 
 
 Tlie wounili-poisnii seems to bo the only tiling that will 
 ■ kill it quickly. On reference to a fonuer i)nrt of these 
 wanderings, it will be seen that a poisoned arrow killed the 
 sloth in about ten minutes. 
 
 So much for this harmless, unolTending animid. Ho 
 holds a conspicuous place in the catalogue of the animals 
 of the new world. Though naturalists have made no 
 mention of what follows, still it is not less true on that 
 account. The sloth is the only quadruped known, which 
 spends its whole life from the branch of a tree, suspended 
 by his feet. I have paid uncommon attention to him in 
 his native haunts. The monkey and squirrel will seize a 
 branch with their fore-feet, and pull themselves uj), and 
 rest or run upon it; but the sloth, after seizing it, still 
 remains suspended, and suspended moves along under the 
 branch, till he can lay hold of another. Whenever I have 
 seen him in his native woods, whether at rest, or asleej), or 
 on his travels, I have always observed that he was suspended 
 from the branch of a tree. When his form and anatomy 
 are attentively considered, it will appear evident that the 
 sloth cannot be at ease in any situation, where his body 
 is higher, or above his feet. We will now take our leavp 
 of him. 
 
 In the far-extending wilds of Guiana, the traveller will 
 be astonished at the immense quantity of Ants which he 
 perceives on the ground and in the trees. They have nests 
 in the branches, four or five times as large as that of the 
 rook ; and they have a covered way from them to the ground. 
 In this covered way thousands are perpetually passing and 
 repassing ; and if you destroy part of it, they turn to, and 
 immediately repair it. 
 
 Other species of ants again have no covered way ; but 
 travel, exposed to view, upon the surface of the earth. You 
 will sometimes see a string of these ants a mile long, each 
 
WANDKllIXnS IN SOl'TII AMERICA. 
 
 223 
 
 carryin*,' in its mouth to its nest a ^Ycan leaf, the size of a thihk 
 sixpence. It is wonderful to observe tlie order in wliich •'^^^^^'^* 
 thoy move, and Nvith wliat pains and labour they surmount 
 tlie obstructions of tlie path. 
 
 The ants liave their enemies, as well as the rest of Thrcf, 
 animated nature. Amon<.,'st the foremost of these stand '^nc'tlars. 
 the three species of Ant-hears. The smallest is not much 
 larger than a rat ; the next is nearly the size of a fox ; and 
 the tliird a stout and powerful animal, measuring about 
 six feet from the snout to tlie end of the tail. He is the 
 most inoffensive of all animals, and never injures the 
 property of man. He is chiefly found in tlie inmost 
 recesses of the forest, and seems i)artial to the low and 
 swampy parts near creeks, where the troely-tree grows. 
 There he goes up and down in quest of ants, of which 
 there is never the least scarcity ; so that he soon obtains a 
 sufficient supply of food, with very little trouble. He 
 cannot travel fast; man is superior to him in speed. 
 Without swiftness to enable him to escape from his 
 enemies, without teeth, the possession of which would 
 assist him in self-defence, and without the power of 
 burrowing in the ground, by which he might conceal 
 himself from his pursuers, he still is capable of ranging 
 through these wilds in perfect safety ; nor does he fear the 
 fatal pressure of the serpent's fold, or the teeth of the 
 famished jnguar. Nature has formed his fore-legs wonder- 
 fully thick, and strong, and muscular, and armed his feet 
 with three tremendous sharp and crooked claws. Whenever 
 he seizes an animal with these formidable weapons, he hugs 
 it close to his body, and keeps it there till it dies through 
 pressure, or through want of food. Nor does the ant-bear, 
 in the meantime, suffer much from loss of aliment, as it is 
 a well-known fact, that he can go longer without food than, 
 perhaps, any other animal, except the land-tortoise. His 
 
 ! ; 
 
224 
 
 WANDERINGS IN SOUTH AMP]RICA. 
 
 THIKD 
 JOURNEY, 
 
 !lff 
 
 Mr ■" 
 
 skin is of a texture that perfectly resists the bite of a dog ; 
 his hinder parts are protected by thick and shaggy hair, 
 while his immense tail is large enough to cover his whole 
 body. 
 
 The Indians have a great dread of coming in contact 
 with the ant-bear; and after disabling him in the chase, 
 never think of approaching him till he be quite dead. It is 
 perhaps on account of this caution, that naturalists have 
 never yet given to the world a true and correct drawing of 
 this singular animal, or described the peculiar position of 
 his fore-feet when he walks or stands. If, in taking a 
 drawing from a dead ant-bear, you judge of the position in 
 which he stands from that of all other terrestrial animals, 
 the sloth excepted, you will be in error. Examine only a 
 figure of this animal, in books of natural history, or inspect 
 a stuffed specimen in the best museums, and you will see 
 that the fore-claws are just in the' same forward attitude as 
 those of a dog, or a common bear when he walks or stands. 
 J^>ut this is a distorted and unnatural position ; and in life, 
 would be a painful and intolerable attitude for the ant-bear. 
 The length and curve of his claws cannot admit of such a 
 position. When he walks or stands, his feet have somewhat 
 the appearance of a club-hand. He goes entirely on the 
 outer side of his fore-feet, which are quite bent in vards ; 
 the claws collected into a point, and going under the foot. 
 In this position he is quite at ease ; while his long claws 
 are disposed of in a manner to render them harmless to 
 him, and are prevented from becoming dull and worn, like 
 those of the dog, which would inevitably be the case, did 
 their points come in actual contact with the ground ; for 
 his claws have not that retractile power which is given to 
 animals of the feline species by which they are enabled to 
 preserve the sharpness of their claws on the most flinty 
 path. A slight inspection of the fore-feet of the ant-bear. 
 
WANDERINGS IN SOUTIT AMERICA. 
 
 225 
 
 THIRT> 
 JOUUNEY. 
 
 Pcai- 
 
 will immediately convince you of the mistake artists and 
 naturalists have fallen into, by putting his fore-feet in the 
 same position as those of other quadrupeds ; for you will 
 ])erceive that the whole outer side of his foot is not only 
 deprived of hair, but is hard and callous ; proof positive 
 of its being in perpetual contact with the ground. Now, on 
 the contrary, the inner side of the bottom of his foot is 
 soft and rather hairy. 
 
 There is another singularity in the anatomy of the ant- 
 bear, I believe, as yet unnoticed in the page of natural '<^"^^j^? 
 liistory. He has two very large glands situated below the iomy of the 
 root of the tongue. From these is emitted a glutinous ' ' 
 liquid, with which his long tongue is lubricated when he 
 puts it into the ants' nests. These glands are of the same 
 substance as those found in the lower jaw of the wood- 
 pecker. The secretion from them, when wet, is very 
 clammy and adhesive, but on being dried it loses these 
 qualities, and you can pulverize it betwixt your finger and 
 thumb ; so that, in dissection, if any of it has got upon the 
 fur of the animal, or the feathers of tlie bird, allow it to dry 
 tliere, and then it may be removed without leaving the least 
 stain behind. 
 
 The ant-bear is a pacific animal. He is never the first 
 to begin the attack. His motto may be, " Noli me tangere." 
 As his habits and his haunts differ materially from those 
 of every other animal in the forest, their interests never 
 clash, and thus he might live to a good old age, and die at 
 last in peace, were it not that his flesh is good food. On 
 this account the Indian wages perpetual war against him, 
 aud as he cannot escape by flight, he fulls an easy prey to 
 the poisoned arrow, shot from the Indian's bow at a distance. 
 If ever he be closely attacked by dogs, he innnediately 
 throws himself on his back, and if he be fortunate enough 
 to catch hold of his enemy with his tremendous claws, 
 
 Q 
 
22i; 
 
 WANDEULVGS IN SOUTH AMEUICA. 
 
 THIRD 
 JOUUNKY. 
 
 The 
 Vampire. 
 
 the invader is sure to jjay for his rashness with tlie loss 
 of life. 
 
 We will now take a view of the vampire. As there was 
 a free entrance and exit to the vampire in the loft where I 
 slept, I had many a fine opportunity of paying attention to 
 this nocturnal surgeon. He does not always live on blood. 
 When the moon shone bright, and the fruit of the banana- 
 tree was ripe, I could see him approach and eat it. He 
 would also bring into the loft, from the forest, a green round 
 fruit, something like the wild guava, and about the size of a 
 imtmeg. There was something also, in the blossom of the 
 sawarri nut-tree, which was grateful to him ; for on coming 
 up Waratilla creek, in a moonlight night, I saw several 
 vampires fluttering round the top of the sawarri tree, and 
 every now and tlien the blossoms, which they had broken 
 off fell into the water. They certainly did not drop off 
 naturally, for on examining several of them, they appeared, 
 quite fresh ajid blooming. So I concluded the vampires 
 pulled them from the tree, either to get at the incipient 
 fruit, or to catch the insects which often take up their abode 
 in flowers. 
 
 The vampire, in general, measures about twenty-six inches 
 from wing to wing, extended, though I once killed one which 
 measured thirty-two inches. He frequents old abandoned 
 houses and hollow trees ; and sometimes a cluster of them 
 may be seen in the forest hanging head downwards, from 
 the branch of a tree. 
 
 Goldsmith seems to have been aware that the vampire 
 hangs in clusters ; for in the Deserted Villa(/e, speaking of 
 America, he says, — 
 
 "And matted woods, where birds forget to sing, 
 But silent bats in drowsy clusters cling." 
 
 The vampire has a curious membrane, which rises from 
 
WANDERINGS IN SOUTH AMERICA. 
 
 227 
 
 tlie nose, and gives it a very singular appearance. It has third 
 been remarked before, that there are two species of vampii-e JQ ^''^^''- y» 
 in Guiana, a larger and a smaller. The larger sucks men 
 and other animals ; the smaller seems to confine himself 
 chiefly to birds. I learnt from a gentleman, high up in the 
 river I/emerara, that he was completely unsuccessful with 
 his fowls, on account of the small vampire. He showed me 
 some that had been sucked the night before, and they were 
 scarcely able to walk. 
 
 Some years ago I went to the river Paumaron with a Avecdotc. 
 Scotch gentleman, by name Tarbet. We hung our ham- 
 mocks in the thatched loft of a planter's house. Next 
 morning I heard this gentleman muttering in his hammock, 
 and now and then letting fall an imprecation or two, just 
 ibout the time he ought to have been saying his morning 
 prayers. " What is the matter. Sir," said I, softly ; " is 
 any thing amiss ? " " What's the matter ? " answered he, 
 surlily: "why, the vampires have been sucking me to 
 death." As soon as there was liglit enough, I went to his 
 hammock, and saw it much stained with blood. " There," 
 said he, thrusting his foot out of the hammock, " see how 
 these infernal imps have been drawing my life's blood." 
 On examining his foot, I found the vampire had tapped 
 his great toe : there was a wound somewhat less than that 
 made by a leech ; the blood was still oozing from it ; I 
 conjectured he might have lost from ten to twelve ounces 
 of blood. AVhilst examining it, 1 think I put him into a 
 worse humour by remarking, that an European surgeon 
 would not have been so generous as to have blooded him 
 without making a charge. He looked up in my face, but 
 did not say a woid: I saw he was of opinion that I had 
 better have spared this piece of ill-timed levity. 
 
 It was not the last punishment of this good gentleman Species of 
 in the river Paumaron. The next night he was doomed to Avt. 
 
 Q 2 
 
 •ill 
 
 
p 
 
 It j' 
 
 it I 
 
 
 1 1' 
 
 ■A 'i ; 
 
 !!« •: 
 
 |r 
 
 1 ' ^ 
 
 228 
 
 WAlN'DEHINGS IN SOUTH AMERICA. 
 
 THIRD undergo a kind of ordeal unknown in Europe. There ia 
 JOURNEY, g^ species of large red ant in Guiana, sometimes called 
 Ranger, sometimes Coushie. These ants march in millions 
 throiigli the country, in compact order, like a regiment of 
 soldiers ; they eat up every insect in their march ; and if 
 a house obstruct their route, they do not turn out of the 
 way, but go quite through it. Though they sting cruelly 
 when molested, the planter is not sorry to see them in his 
 liouse ; for it is but a passing visit, and tliey destroy every 
 kind of insect vermin that has taken shelter under his roof. 
 Now, in the British plantations of Guiana, as well as in 
 Europe, there is always a little temple dedicated to the 
 goddess Cloacina. Our dinner had chiefly consisted of 
 crabs, dressed in rich and different ways. Paumaron is 
 famous for crabs, and strangers who go thither consider 
 them the greatest luxury. The Scotch gentleman made a 
 very capital dinner on crabs ; but this change of diet was 
 productive of unpleasant circumstances : he awoke in the 
 night in that state in which Yirgil describes Cneleno to 
 have been, viz. " ftrdissima ventris proluvies." Up he got, 
 to verify the remark, 
 
 *' Serins aiit citius, Sederti properaimis ad niiani." 
 
 Now, unluckily for himself, and the nocturnal tranquillity 
 of the planter's house, just at that unfortunate hour, the 
 Coushie Ants were passing across the seat of Cloacina's 
 temple ; he had never dreamed of this ; and so, turning 
 his face to the door, he placed himself in the usual situation 
 which the votaries of the goddess generally take. Had a 
 lighted match dropped upon a pound of gunpowder, as he 
 afterwards remarked, it could not have caused a greater 
 recoil. Up he jumped, and forced his way out, roaring for 
 help and for a light, for he was worried alive by ten thousand 
 devils. The fact is, he had sat down upon an intervening 
 
WAXDEIilNGS IN SOUTH AMERICA. 
 
 220 
 
 body of coushie ants. Many of those which escaped being third 
 crushed to deatli, turned again, and, in revenge, stung the JQ ^junk y. 
 unintentional intruder most severely. The watchman had 
 fallen asleep, and it was some time before a light could be 
 procured, the fire having gone out ; in the mean time, the 
 poor gentleman was suffering an indescribable martyrdom, 
 and would have found himself more at home in the Augean 
 stable than in the planter's house. 
 
 I had often wished to have been once sucked by the 
 vampire, in order that I might have it in my power to say 
 it had really happened to me. There can be no pain in the 
 operation, for the patient is always asleep when the vampire 
 is sucking him ; and as for the loss of a few ounces of blood, 
 that W'Ould be a tritle in the long run. Many a night have 
 I slept with my foot out of the hammock to tempt this 
 winged surgeon, expecting that he would be there ; but it 
 was all in vain ; the vampire never sucked me, and I could 
 never account for his not doing so, for we were inhabitants 
 of the same loft for months together. 
 
 The armadillo is very common in these forests ; he The Amm- 
 burrows in the sand-hills like a rabbit. As it often takes 
 a considerable time to dig him out of his hole, it would be 
 a long and laborious business to attack each hole indis- 
 criminately without knowing whether the animal were 
 there or not. To prevent disappointment, the Indians 
 carefully examine the mouth of the hole, and put a short 
 stick down it. Now if, on introducing the stick, a number 
 of mosquitos come out, the Indians know to a certainty 
 that the armadillo is in it : wherever there are no mosquitos 
 in the hole there is no armadillo. The Indian having 
 satisfied himself that the armadillo is there, bv the mos- 
 quitos which come out, he immediately cuts a long and 
 slender stick, and introduces it into the hole : he carefully 
 observes the line the stick takes, and then sinks a pit in 
 
 dillo. 
 
 
 i!:' 
 

 \\\, 
 
 i 
 
 
 i 
 
 E'. 
 
 Ill 
 
 :' i! 
 
 ;? I 
 
 280 
 
 WAN'DKKIN'iiS IN SOUTH AMKIUCA, 
 
 'I'lIIKD 
 JUUKNEY, 
 
 the sand to catch the cud of it : this done, he puts it 
 farther into the liole, and digs anotlier pit, and so on, till 
 at last he comes up with the armadillo, which had been 
 making itself a passage in the sand till it had exhausted 
 all its strength through pure exertion. I have been some- 
 times three quarters of a day in digging out one armadillo, 
 and obliged to sink half a dozen pits, seven feet deep, 
 before I got up to it. Tlie Indians and negroes are very- 
 fond of the Uesh, but I considered it strong and rank. 
 
 On laying hold of the armadillo you must be cautious 
 not to come in contact with his feet : they are armed with 
 sharp claws, and with them he will inflict a severe wound 
 in self-defence : when not molested, he is very harmless 
 and innocent ; he would put you in mind of the hare in 
 Gay's fables, — 
 
 " Whose care was never to ofl'i'inl, 
 And every creature was her frieiul." 
 
 The armadillo swims well in time of need, but does not 
 go into the water by choice. He is very seldom seen 
 abroad during the day ; and when surprised, he is sure to 
 be near the mouth of his hole. Every part of the arma- 
 dillo is well protected by his shell, except his ears. In 
 life, this shell is very limber, so that the animal is enabled 
 to go at full stretch, or roll himself up into a ball, as 
 occasion may require. 
 
 On inspecting the arrangement of the shell, it puts you 
 very much in mind of a coat of armour j indeed it is a 
 natural coat of armour to the armadillo, and being com- 
 posed both of scale and bone, it affords ample security, and 
 has a pleasing effect. 
 The Land Often, when roving in the wilds, T would fall in with 
 Tortoise. ^]^g Land Tortoise ; he too adds anotlier to the list of un- 
 offending animals ; he subsists on the fallen fruits of the 
 forest. When an enemy approaches he never thinks of 
 
WANDERINGS IN SOUTH AMEIUCA. 
 
 231 
 
 moving, but quietly draws himself under liis shell, and riimD 
 there awaits his doom in patience : he only seems to have '^"^"^'^^' 
 two enemies who can do him any damage ; one of these is 
 the Boa Constrictor: this snake swallows the tortoise alive, 
 shell and all. But a boa large enough to do this is very 
 scarce, and t^- there is not much to apprehend from that 
 quarte the » "^r enemy is man v* > , takes up the tor- 
 toise, and carries him away. Man also is scarce in these 
 never-ending wilds, and the little depredations he may 
 commit upon the tortoise will be nothing, or a mere trifle. 
 The tiger's teeth cannot penetrate its shell, nor can a 
 stroke of his paws do it any damage. It is of so compact 
 and strong a nature, that there is a common saying, a 
 London waggon might roll over it and not break it. 
 
 Ere we proceed, let us take a retrospective view of the 
 five animals just enumerated ; they are all quadrupeds, 
 and have some very particular mark, or mode of existence, 
 different from all other animals. The sloth has four feet, 
 but never can use them, to support his body on the earth ; 
 they want soles, which are a marked feature in the feet of 
 other animals. The ant-bear has not a tooth in his head, 
 still he roves fearless on, in the same forests with the 
 Jaguar and boa constrictor. The vampire does not make 
 use of his feet to walk, but to stretch a membrane, which 
 enables him to go up into an element where no other 
 quadruped is seen. The armadillo has only here and there 
 a straggling hair, and has neither fur, nor wool, nor bristles, 
 but in lieu of them has received a movable shell, on which 
 are scales very much like those of fishes. The tortoise is 
 oviparous, entirely without any appearance of hair, and is 
 obliged to accommodate itself to a shell which is quite hard 
 and inflexible, and in no point of view whatever obedient to 
 the will or pleasure of the bearer. The egg of the tortoise 
 has a very hard shell, while that of the turtle is quite soft. 
 
 J«t ! 
 
 
 i 1 
 
hi 
 
 JE* 
 
 ii 
 
 
 i I 
 
 CHAlTEll IT. 
 
 The Vanilla.— ^Meanins of the word. — Small Cayman. — Daddy Quashi. — 
 Wasps. — Venomous reptiles and wild beasts. — Poison of the Labairi 
 snake. — Experiment with a Laharri. — The IJete-rouj^e. — The Chegoe. 
 — Its nest. — Ticks, and how to get rid of them. — The five tribes — 
 Their habitations and mode of life. — Piwarri. — The Pee-ay-man. — A 
 nation without a history. — Runaway negroes, — Mr. Edmonstone and 
 his services. — Wounded warriors. — Valour rewarded. — Education. — 
 Character of the native. — Skill in hunting. — A bead almanac. — The 
 sun as a compass. — Thinness of population. 
 
 THinn 
 
 JOURNKV. 
 
 The 
 
 Vanilla. 
 
 the Vanilla growing 
 
 In some parts of these forests I saw 
 luxuriantly. It creeps up the trees to the height of 
 thirty or forty feet. I found it difficult to get a ripe pod, 
 as the monkeys are very fond of it, and generally took 
 care to get there before me. The pod hai:qs from the 
 tree in the shape of a little scabbard. Vayna is the 
 Spanish for a scabbard, and Vanilla for a little scabbard. 
 Hence the name. 
 
 In Mibiri creek there was a cayman of the small species, 
 measuring about five feet in length ; I saw it in the same 
 place for months, but could never get a shot at it ; for the 
 moment I thought I was sure of it, it dived under the 
 water before I could pull the trigger. At last I got an 
 Indian with his bow and arrow ; he stood up in the canoe 
 with his bow ready bent, and as we drifted past the place, 
 he sent his arrow into the cayman's eye, and killed it 
 dead. The skin of this little species is much harder and 
 
WANDEIIINGS IX SOUTH AMEKICA. 
 
 233 
 
 A'cr/m 
 scrmnf. 
 
 stronger than that of the h^vge kind : it is good food, and third 
 tastes like veal. jouunkv, 
 
 My friend, ^Er. Edtnonstone, had very kindly let me 
 liave one of his old negroes, and he constantly attended 
 ine ; his name was Daddy Quashi ; he had a hrave stomach 
 for heterogeneous food ; it could digest, and relish too, cay- 
 man, monkies, hawks, and gru1)S. The Daddy made three 
 or four meals on this cayman while it was not absolutely 
 putrid, and salted the rest. I could never get him to face 
 a snake ; the horror he betrayed on seeing one was beyond 
 description ; I asked him why he was so terribly alarmed ; 
 he said it was by seeing so many dogs, fr<»^ > time to time, 
 killed by them. 
 
 Here I had a fine opportunity of examining several Species of 
 species of the caprimulgus. I am fully persuaded that these ^^I'^I'J'/IIJ' 
 innocent little birds never suck the herds ; for when they 
 approach them, and jump up at their udders, it is to catch 
 the flies and insects there. AVhen the moon shone bright, 
 I would frequently go and stand within three yards of a 
 cow, and distinctly see the caprimulgus catch the flies on 
 its udder. On looking for them in the forest, during the 
 day, I either found them on the ground, or else invariably 
 sitting longitudinally on the branch of a tree, not crossicise 
 hke all other birds. 
 
 The Wasps' or Maribuntas, are great plagues in these The Wasj) 
 forests, and require the naturalist to be cautious as he Xuttal-'." 
 wanders up and down. Some make their nests pendent 
 from the branches ; others have them fixed to the under- 
 side of a leaf. Now in passing on, if you happen to dis- 
 turb one of these, they sally forth and punish you severely. 
 The largest kind is blue ; it brings blood where its sting 
 enters, and causes pain and inflammation enough to create 
 a fever. The Indians make a fire under the nest, and 
 after killing or driving away the old ones, they roast the 
 
 i» 
 
 
 ■'\'i' 
 
 I 
 
234 
 
 WAKDElilKGS IN ^:OUTII AMKL'ICA. 
 
 TiiritD 
 
 JOUKNKV. 
 
 ;i 
 
 Snakes and 
 wild beasts. 
 
 Catches a 
 
 live Labar- 
 
 ri snake. 
 
 youiit; gml)S in the comb and eat them. I tried tliem 
 once by way of dessert after dinner, but my stomach was 
 offended at tlieir intrusion ; probably it was more the idea 
 tlian the taste that caused the stomach to rebel. 
 
 Time and experience liave convinced me that there is not 
 much danger in roving amongst snakes and wild beasts, 
 provided only that you have self-command. You must 
 never approach them abruptly ; if so, you are sure to pay 
 for your rashness; because the idea of self-defence is pre- 
 dominant in every animal, and thus the snake, to defend 
 himself from what he considers an attack upon him, makes 
 the intruder feel the deadly effect of his poisonous fangs. 
 The jaguar flies at you and knocks you senseless with a 
 stroke of his paw : whereas, if you had not come upon 
 him too suddenly, it is ten to one but that he had retired, 
 in lieu of disputing the path with you. The labarri snake 
 is very poisonous, and I have often approached within two 
 yards of him without fear. I took care to move very softly 
 and gently without moving my arms, and he always al- 
 lowed me to have a fine view of him, without showing the 
 least inclination to make a spring at me. He would appear 
 to keep his eye fixed on me, as tliough suspicious, but iliat 
 was all. Sometimes I have taken a stick ten feet long, 
 and placed it on the labarri's back. He would then glide 
 away without offering resistance. But when I put the 
 end of the stick abruptly to his head, he immediately 
 opened his mouth, flew at it, and bit it. 
 
 One day, wishful to see how the poison conies out of 
 the fangs of the snake, I caught a labarri alive. He was 
 about eight feet long. I held him by the neck, and my 
 hand was so near his jaw, that he had not room to move 
 his head to bite it. This was the only position I could 
 have held him in v,i*^b safety and effect. To do so, it only 
 required a little resolution and coolness. I then took a 
 
WANDERINGS IN tSOUTII AMKRICA. 
 
 235 
 
 Bly 
 
 small {jieco of stick in the other hiiml, and pressed it 
 aj^'ainst tlie fang, which is invarialtly in the upjjer Jaw. 
 Towards the point of the fan<:(, there is a little oblong 
 aperture on the convex side of it. Through this, there is 
 a communication down the fang to the root, at which lies 
 tt little bag containing the poison. Now, when the point 
 of the fang is pressed, the root of the fang also presses 
 against the bag, and sends up a portion of the poison 
 therein contained. Thus, when I applied a piece of stick 
 to the point of the fang, there came out of the hole ali(pior 
 thick and yellow, like strong camomile tea. This was the 
 poison, which is so dreadful in its effects, as to render the 
 labarri snake one of the most poisonous in the forests of 
 Guiana. I once caught a fine labarri, and made it bite 
 itself. I forced the ])oisonous fang into its belly. In a few 
 minutes I thought it was going to die, for it appeared dull 
 and heavy. However, in half an hour's time, he was as 
 brisk and vigorous as ever, and in the course of the day 
 showed no symptoms of being afl'ccted. Is then the life 
 of the snake proof against its own poison ? This subject is 
 not unworthy of the consideration of the naturalist. 
 
 In Guiana there is a little insect in the grass and on the 
 shrubs, which the French call l>ete-rouge. It is of a beau- 
 tiful scarlet colour, and so minute, that you must bring 
 your eye close to it before you can perceive it. it is most 
 numerous in the rainy season. Its bite causes an intoler- 
 able itching. The best way to get rid of it, is to rub the 
 part affected with oil or rum. You must be careful not to 
 scratch it. If you do so, and break the skin, you expose 
 yourself to a sore. The first year I was in Guiana, the 
 bete-rouge, and my own want of knowledge, and, I may 
 add, the little attention I paid to it, created an ulcer above 
 the ankle, which annoyed me for six months, and if I 
 hobbled out into the grass, a number of bete-rouge would 
 
 TIIIUD 
 JOLUNKY. 
 
 The 
 BUe-rouge. 
 
 
': 
 
 >H t 
 
 He! 
 
 236 
 
 WANDEI.'IXGS IN SOUTH AMKIilCA. 
 
 TiiiKD settle on the edges of the sore, and increase the iiillannua- 
 ^oy^y. tion. 
 
 T/ic Still more inconvenient, painful, and annoying is another 
 
 C/icfjw. 2jj.j.jg ppj^j.^ called the Chegoe. It looks exactly like a very 
 
 small ilea, and a stranger would take it for one. Jfowever, 
 in about four and twenty hours, he would have several 
 broad hints that he had made a mistake in his ideas of the 
 animal. It attacks different parts of the body, but chiefly 
 the feet, betwixt the toe-nails and the llesh. There it 
 buries itself, and at first causes an itching not uni)leasant. 
 In a day or so, after examining the part, yon perceive a 
 place about the size of a pea, somewhat discoloured, rather 
 of a blue appearance. Sometimes it happens that the 
 itching is so trivial, you are not aware that the miner is at 
 ■work. Time, they say, makes great discoveries. The 
 discoloured part turns out to be the nest of the chogoe 
 containing hundreds of eggs, which, if allowed to hatch 
 there, the young ones will soon begin to form other nests, 
 and in timo cause a spreading ulcer. As soon as you 
 perceive thf.t you have got the chegoe in your flesh, you 
 must take a needle, or a sharp-pointed knife, and take it 
 out. If the nest be formed, great care must be taken not 
 to break it, otherwise some of the eggs remain in the flesh, 
 and then you will soon be annoyed with more chegoes. 
 After removing the nest, it is well to drop spirit of tur- 
 pentine into the hole ; that will most effectually destroy 
 any chegoe that may be lurking there. Sometimes I have 
 taken four nests out of my feet in the course of the day. 
 
 Every evening, before sundown, it was part of my 
 toilette to examine my feet, and see that they were clear 
 of chegoes. Now and then a nest would escape the 
 scrutiny, and then I had to smart for it a day or two after. 
 A chegoe once lit upon the back of my hand ; wishful to 
 see how he worked, I allowed him to take possession. He 
 
WAXnKIMNGS IN SOl'TII AMKHICA. 
 
 2:57 
 
 li, 
 
 Ci 
 
 immediately set to work, liead foretnost, and in about tihud 
 hall' an hour he had comjdetely buried himself in the skin. J'^ ''»^'- y- 
 ] tlien let him feel the point of my knife, and exterminated 
 him. 
 
 ^lore than once, after sitting dowii upon a rotten stump, Ticks. 
 I have found myself covered with Ticks. There is a short 
 and easy way to get quit of these unwelcome adlierents. 
 Make a large tire and stand close to it, and if you be 
 covered with ticks, tliey will all f dl off. 
 
 Let us now forgot for a while ihe quadrup.ids, serpents, 
 and insects, and take a transitory view of the native In- 
 dians of these forests. 
 
 There are five principal nations or tribes of In^ -ms in Principal 
 ■decani Dutch Guiana, commonly known by liie name tribp.i,'of 
 of Warow, Arowack, Acoway, Carib, and "'Jacoushi. Tht; Indians. 
 live in small handets, which consist of > ft v.' huts, never 
 exceeding twelve in number. Tliese huts are always in 
 the iurest, near a river or some creek. They are open on 
 all sides, (except those of the !^^acoushi,) and covered with 
 a species of palm leaf. 
 
 Their principal furniture is the hammock. It serves Thrir 
 them both for chair and bed. It is commonly made of " 
 cotton ; though those of the AVarows are formed from the 
 jeta tree. At night they always make a fire close to it. 
 The heat keeps them warm, ;iii>l the smoke di-ives away 
 the mosquitos and sand-flies. "Vou sometimes lind a table 
 in the hut ; but it was not made by the Indians, but by 
 some negro, or mulatto caij)enter. 
 
 They cut down about an acre or two of the trees which Orcupa- 
 surround the huts, and there plant I'epper, I'apaws, Sweet 
 and J)itter Cassava, Plantains, Sweet Potatoes, Yams, 
 Pine-Apples, and Silk-Grass. Besides these, they gene- 
 rally have a few acres in some fertile part of the forest for 
 their cassava, which is as bread to them. They make 
 
238 
 
 VVANDEHINGS IN SOUTH AMERICA. 
 
 !V 
 
 Fermented 
 liquor. 
 
 THIRD eartlieii pots to boil their provisions in ; and they get from 
 JOURNEY. ^Y\e white men flat circular plates of iron, on wliich they 
 hake their cassav" They have to grate the cassava 
 before it is pressed, preparatory to baking ; and those In- 
 dians ^\•\\o are too far in the wilds to procure graters from 
 tiie white men, make use of a flat piece of wood, studded 
 with sharp stones. They have no cows, horses, mules, 
 goats, sheep, or asses. The men hunt and fish, and the 
 ^vomen work in the provision ground, and cook their 
 victuals. 
 
 In each handet there is the trunk of a large tree, 
 hollowed out like a trough. In this, from their cassava, 
 they make an abominable ill-tasted and sour kind of fer- 
 mented liquor, called piwarri. They are very fond of it, 
 and never tail to get drunk after every brewing. The 
 frequency of the brewing depends upon the superabundance 
 of cassava. 
 
 lioth men and women go without clothes. The men 
 have a cotton wrapper, and the w^omen a bead-ornamented 
 square piece of cotton, about the size of your hand, for 
 the fig-leaf. Those far away in the interior, use the bark 
 of a tree for tliis purpose. They are very clean people, 
 and wash in the river, or creek, at least twice every day. 
 They paint themselves with the roucou, sweetly perfumed 
 with hayawa or accaiari. Their hair is black and lank, 
 and never curled. The women braid it up fancifully, 
 something in the shape of Diana's head-dress in ancient 
 pictures. They have very few diseases. Old age and 
 pulmonary complaints seem to be the chief agents for 
 removing them to another world. The pulmonary com- 
 plaints are generally brought on by a severe cold, which 
 they do not know how to arrest in its progress, by the use 
 of the lancet. I never saw an idiot amongst them, nor 
 could 1 perceive any that were deformed from their birth. 
 
 Tlieir 
 habits. 
 
WANDERINGS IN SOUTH AMEllICA. 
 
 239 
 
 Lor 
 th. 
 
 Their women never perish in childbed, owing, no doubt, to third 
 their never wearing fitays. jo urne y. 
 
 They have no public religious ceremony. They acknow- Religious 
 ledge two superior beings, — a good one, and a bad one. ''^"^^^ 
 They pray to the latter not to hurt them, and they are of ceremonies. 
 opinion that tlie former is too good to do them an injury. 
 I suspect, if the truth were known, tlie individuals of tlie 
 village never offer up a single prayer or ejaculation. They 
 have a kind of a priest called a Pee-ay-man, who is an 
 enchanter. He finds out things lost. He mutters prayers 
 to the evil spirit over them and their children when they 
 are sick. If a fever be in the village, the Pee-ay-man 
 goes about all night long, howling, and making dreadful 
 noises, and begs the bad spirit to depart. But he has 
 very seldom to perform this part of his duty, as fevers 
 seldom visit the Indian hamlets. However, when a fever 
 does come, and his incantations are of no avail, which I 
 imagine is most commonly the case, they abandon the 
 place for ever, and make a new settlement elsewhere. 
 They consider the owl and the goatsucker as familiars of 
 the evil spirit, and never destroy them. 
 
 I could find no monuments or marks of antiquity 
 amongst these Indians; so that after penetrating to the 
 Rio Branco, from the shores of the Western Ocean, had 
 any body questioned me on this subject, I should have 
 answered, I have seen nothing amongst these Indians 
 which tells me that they have existed here for a century ; 
 though, for aught I know to the contrary, they may have 
 been here before the Redemption, but their total want of 
 civilization has assimilated them to the forests in which they 
 wander. Thus, an aged tree falls and moulders into dust 
 and you cannot tell what was its appearance, its beauties, 
 or its diseases amongst the neighbouring trees; another 
 has shot up in its place, and after nature has had her 
 
 i^ 
 
 
 LI 
 
 :i Si 
 
240 
 
 WANDEIilXGS IN SOUTH AMERICA. 
 
 THIRD course, it will make way for a successor in its turn. So 
 j.iuRN'EY. ||. jg ^^,^^j^ |.|jg Indian of Guiana; he is now laid low in the 
 
 dust ; he has left no record beliind him, either on parch- 
 ment, or on a stone, or in earthenware, to say what he has 
 done. Perhaps the place where his buried ruins lie was 
 unhealthy, and the survivors have left it long ago, and 
 gone far away into the wilds. All that you can say is, 
 the trees where I stand appear lower and smaller than the 
 rest, and from this 1 conjecture, that some Indians may 
 have had a settlement here formerly. AVere I by chance 
 to meet the son of the father who moulders here, he could 
 tell me that his father was famous for slaying tigers and 
 serpents and caymen, and noted in the chase of the tapir 
 and wild boar, but that he remembers little or nothing of 
 his grandfather. 
 
 They are very jealous of their liberty, and much at- 
 tached to their own mode of living. Though those in the 
 neighbourhood of the European settlements have constant 
 communication with the whites, they have no inclination 
 to become civilized. Some Indians who have accompanied 
 white men to Europe, on returning to their own land, have 
 thrown off their clothes, and gone back into the forests. 
 
 In George-town, the capital of Demerara, there is a large 
 shed, open on all sides, built for them by order of govern- 
 ment. Hither the Indians come witli monkies, parrots, 
 bows and arrows, and pegalls. They sell these to the 
 white men for money, and too often purchase rum with it, 
 to whicli they are wonderfully addicted. 
 
 Government allows them annual presents in order to 
 have their services, when the colony deems it necessary to 
 scour the forests in quest of runaway negroes. Formerly 
 these expeditions were headed by Charles Edmonstone, 
 Esq., now of Cardross Park, near Dumbarton. This brave 
 colonist never returned from the woods without being 
 
 ra- 
 
WANDERINGS IN SOUTH AMERICA. 
 
 241 
 
 victorious. Once, in an attack upon the rebel negroes' tiiii;i) 
 camp, he led the way, and received two balls in his body ; ■J" ^-""^''- '^^ 
 at the same moment that he was wounded, two of }iis 
 Indians fell dead bv his side ; he recovered after his life 
 was despaired of, but the balls could never be extracted. 
 
 Since the above appeared in print, I have had the 
 account of this engagement with the negroes in the forest 
 from Mr. Edmonstoue's own mouth. 
 
 He received four slugs in his body, as will be seen in 
 the sequel. 
 
 The plantations of Demerara and Essequibo are bounded 
 by an almost interminable extent of forest. Hither the 
 runaway negroes repair, and form settlements, from 
 whence they issue to annoy the colonists, as occasion may 
 offer. 
 
 In 1801, the runaway slaves had increased to an alarm- 
 ing extent. The Governor gave orders that an expedition 
 should be immediately organized, and proceed to the 
 woods, under the command of Charles Edmonstone, Esq. 
 General Hislop sent him a corporal, a sergeant, and eleven 
 men, and he was joined by a part of the colonial militia, 
 and by sixty Indians. 
 
 With this force Mr. Edmonstone entered the forest, and 
 proceeded in a direction towards Mahaica. 
 
 He marched for eight days through swamps, and over 
 places obstructed by fallen trees and the bush-rope ; tor- 
 mented by myriads of mosquitos, and ever in fear of 
 treading on the poisonous snakes, which can scarcely be 
 distinguished from the fallen leaves. 
 
 At last he reached a wooded sand-hill, where the 
 Maroons had intrenched themselves in great force. Not 
 expecting to come so soon upon them, JNIr. Edmonstone, 
 his faithful man Coffee, and two Indian chiefs, found 
 themselves considerably ahead of their own party. As 
 
 K 
 
 : \ 
 
242 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 THinn 
 joruxEY, 
 
 yet, they were unpcrceived by tlie enemy, but, unfortu- 
 nately, one of the Indian chiefs firecl a random shot at a 
 distant ]\laroon. Immediately tiie whole negro camp 
 turned out, and formed themselves in a crescent, in front 
 of Mr. Edmonstone. Their chief was an uncommonly fine 
 negro, about six feet in height; and his head-dress was 
 that of an African warrior, ornamented with a profusion 
 of small shells. He advanced undauntedly with his gun 
 in his hand, and, in insulting language, called out to ]\Ir. 
 Edmonstone to come on and fight him. 
 
 JVIr. Edmonstone approached him slowly, in order to 
 give his ovn men time to come up; but they were yet too 
 far off for him to profit by this manoeuvre. Coffee, who 
 carried his master's gun, now stepped up behind him, and 
 put the gun into his hand, which Mr. Edmonstone received, 
 without advancing it to his shoulder. 
 
 He was now within a few yards of the Maroon chief, 
 who seemed to betray some symptoms of uncertainty ; for 
 instead of firing directly at Mr. Edmonstone, he took a step 
 sideways, and rested his gun against a tree ; no doubt with 
 the intention of taking a surer aim. ^Mr. Edmonstone, on 
 perceiving this, immediately cooked his gun, and fired it 
 off, still holding it in the position in which he had received 
 it from Coffee. 
 
 The whole of the contents entered the negro's body, and 
 he dropped dead on his face. 
 
 The negroes, who had formed in a crescent, now in their 
 turn fired a volley, which brought Mr. Edmonstone and 
 his two Indian chiefs to the ground. The Maroons did 
 not stand to reload, but on Mr. Edmonstone's party 
 coming up, they fled precipitately into the surrounding 
 forest. 
 
 Four slugs had entered Mr. Edmonstone's body. After 
 coming to himself, on looking around, he saw one of the 
 
WANDERINGS IN SOUTH AMERICA. 
 
 243 
 
 fiiUen Indian chiefs bleeding by liis side. He accosted third 
 liini by name, and said he hoped he was not much hurt. Jo^^^^^y. 
 The dying Indian had just strength enough to answer, 
 " Oh no," — and then expired. The other chief was lying 
 quite dead. He must have received his mortal wound 
 just as he v^as in the act of cocking his gun to fire on the 
 negroes ; for it appeared that the ball which gave him his 
 death wound, had carried off the first joint of his thumb, 
 and passed through liis forehead. By this time his wife, 
 who had accompanied the expedition, came up. She was 
 a fine young woman, and had her long black hair fancifully 
 braided in a knot on the top of her head, fastened with a 
 silver ornament. She unloosed it, and, falling on her 
 husband's body, covered it with her hair, bewailing his 
 untimely end with the most heartrending cries. 
 
 The blood was now runnin<]j out of Mr. Edmonstone's 
 shoes. On being raised up, he ordered his men to pursue the 
 Hying Maroons, requesting at the same time tliat he might 
 be left where he had fallen, as he felt that he was mortally 
 wounded. They gently placed him on the ground, and 
 after the pursuit of the Maroons had ended, the corporal 
 and sergeant returned to their commander, and formed 
 their men. On his asking what this meant, the sergeant 
 replied, " I had the General's orders, on setting out from 
 town, not to leave you in the forest, happen what might." 
 I>y slow and careful marches, as much as the obstructions 
 in the woods would admit of, the party reached Plantation 
 Alliance, on the bank of the Demerara, and from thence it 
 crossed the river to Plantation Vredestein. 
 
 The news of the rencounter had been spread far and 
 wide by the Indians, and had already reached town. 
 The General, Captains IMacrai and Johnstone, and Doctor 
 Dunkin, proceeded to Vredestein. On examining Mr. 
 Edmonstone's wounds, four slugs were found to have 
 
 u 2 
 
244 
 
 r; 
 
 ipr 
 
 THIRD 
 .JOUHNEY. 
 
 General 
 remarks. 
 
 WANDEllINGS TN SOUTH AMERICA. 
 
 entered the body ; one was extracted, the rest remained 
 tliere till the year 1824, when Lnother was cut out by a 
 professional gentleman of Port Glasgow. The other two 
 still remain in the body ; and it is supposed that either 
 one or both have touched a nerve, as they cause almost 
 continual pain. Mr. Edmonstone has commanded fifteen 
 different expeditions in the forest in quest of the ]Maroons. 
 The Colonial Government has requited his services, by 
 freeing his property from all taxes, and presenting him a 
 handsome sword, and a silver urn, bearing the following 
 inscription : 
 
 " Presented to Ciiarlks Ehmonstoxk, Esq. by tlie Governor and 
 Court of Policy of the Colony of Deniernra, as a token of their 
 esteem, and the deep sense they entertain ot the very great activity 
 and spirit manifested by him on various occasions in his successful 
 exertions for the internal security of the Colony. — January Isf, 
 1809." 
 
 I do not believe that there is a single Indian in ci-devant 
 Dutch Guiana who can read or write, nor am I aware that 
 any white man has reduced their language to the rules 
 of grammar; some may have made a short manuscript 
 vocabulary of the few necessary words, but that is all. 
 Here and there a white man, and some few people of 
 colour, talk the language well. The temper of the Indian 
 of Guiana is mild and gentle, and he is very fond of his 
 children. 
 
 Some ignorant travellers and colonists call these Indians 
 a lazy race. Man in general will not be active without an 
 object. Now when the Indian has caught plenty of fish, 
 and killed game enough to last him for a week, what need 
 has he to range the forest ? He has no idea of making 
 pleasure-grounds. Money is of no use to him, for in these 
 wilds there are no markets for him to frequent, nor mil- 
 liners' shops for his wife and daughters ; he has no taxes to 
 
liese 
 nil- 
 Is to 
 
 WANDERINGS IN SOUTH AMEHICA. 
 
 ^45 
 
 pay, no highways to keep up, no i)oor to maintain, nor third 
 army nor navy to supply ; he lies in his hammock Loth ''" ""''^''' ^' 
 night and day (for he has no chair or bed, neither does he 
 want them), and in it he forms his bow, and makes his 
 arrows, and repairs his fishing tackle. But as soon as he 
 has consumed his provisions, he then rouses himself, and, 
 like the lion, scours the forest in quest of food. He 
 plunges into the river after the deer and tapir, and swims 
 across it ; passes through swamps and quagmires, and 
 never fails to obtain a sufficient supply of food. Should 
 the approach of night stop his career, while he is hunt- 
 ing the wild boar, he stops for the night, and continues the 
 chase the next morning. In my way through the wilds to 
 the Portuguese frontier, I had a proof of this: we were 
 eight in number, six Indians, a negro, and myself. About 
 ten o'clock in the morning, we observed the feet-mark of 
 the wild boars ; we judged by the freshness of the marks 
 that they had passed that way early the same morning. 
 As we were not gifted, like the hound, with scent, and as 
 we had no dog with us, we followed their track by the eye. 
 The Indian after game is as sure with his eye as the dog 
 is with his nose. We followed the herd till three in the 
 afternoon, then gave up tlip chase for the present ; made 
 our fires close to a creek where there was plenty of fish, 
 and then arranged the hammocks. In an hour the Indians 
 shot more fish with their arrows than we could consume. 
 The night was beautifully serene and clear, and the moon 
 shone as bright as day. Next morn we rose at dawn, got 
 breakfast, packed up, each took his burden, and then we 
 put ourselves on the track of the wild boars, which we had 
 been following the day before. AVe supposed that they, 
 too, would sleep that night in the forest, as we had done ; 
 and thus the delay on our part would be no disadvantage 
 to us. This was just the case, for about nine o'clock their 
 
 Hi 
 
 "1 *> 
 
 ii 
 
 
 -I 1i 
 
 I : 
 
 ! 
 
 t 
 
24fi 
 
 THIRD 
 JOUKXKY. 
 
 WANDEUIXr.S IN SOUTH AMEUICA. 
 
 feet-niarks became tVeslier and fresher : we now doubled 
 our pace, but did not give mouth like hounds. We pushed 
 on in silence, and soon came up with them ; there were 
 above one hundred of them ; we killed six, and the rest 
 took off in dififorent directions. But to the point. 
 
 Amongst us the needy man works from light to dark for 
 a maintenance. Should this man chance to acquire a 
 fortune, he soon changes his habits. No longer under 
 " strong necessity's supreme command," he contrives to 
 get out of bed betwixt nine and ten in the morning. His 
 servant helps him to dress, he walks on a soft carpet to 
 his breakfast table, his wife pours out his tea, and his 
 servant hands him his toast. After breakfast, the doctor 
 advises a little gentle exercise in the carriage for an hour 
 or so. At dinner-time he sits dowa to a table groaning 
 beneath the weight of heterogeneous luxury: there he 
 rests upon a chair for three or four hours, eats, drinks, and 
 talks (often unmeaningly) till tea is announced. He pro- 
 ceeds slowly to the drawing-room, and there spends the best 
 part of his time in sitting, till his wife tempts him with 
 something warm for supper. After supper, he still remains 
 on his chair at rest, till he retires to rest for the night. 
 He mounts leisurely up stairs upon a carpet, and enters 
 his bed-room : there, one would hope, that at least he 
 mutters a prayer or two, though perhaps not on bended 
 knee : he then lets himself drop into a soft and downy 
 bed, over which has just passed the comely Jenny's 
 warming-pan. Now, could the Indian in his turn see this, 
 he would call the white men a lazy, indolent set. 
 
 Perhaps then, upon due reflection, you would draw this 
 conclusion ; that men will always be indolent where there 
 is no object to rouse them. 
 
 As the Indian of Guiana has no idea whatever of com- 
 municating his intentions by writing, he has fallen upon 
 
WANDERINGS IN SOUTH AMERICA. 
 
 247 
 
 itli 
 
 a plan of comiuunication sure and simple. When two or third 
 three families have deteriniued to come down the river '^" "'^'''''; ^' 
 and pay you a visit, they send an Indian beforehand with /'tf^''"'- 
 
 . P 1 1 xr 1 1 ■. r.. 1 1 method nj 
 
 a stnng oi beads. You take one bead oil every day; and communi- 
 on the day tliat the string is headless, they arrive at } ' <ur '^"'"'jt. 
 house. 
 
 In finding their way through these pathless wilds, the 
 sun is to them what Ariadne's clue was to Tlieseus. Wiien 
 he is on the meridian, they generally sit down, and rove 
 onwards again as soon as he has sutticiently declined to 
 the west ; they require no other compass. When in chase, 
 they break a twig on the bushes as they pass by every 
 three or four hundi'ed paces, and this often prevents them 
 from losing their way on their return. 
 
 You will not be long in the forests of Guiana, before 
 you perceive how very thinly they are inhabited. You 
 may wander for a week together without seeing a hut. 
 The wild beasts, snakes, the swamps, the trees, the un- 
 curbed luxuriance of everything around you, conspire to 
 "it man has no habitation here — man has 
 
 iniorm you that man J 
 seldom passed this w.iy. 
 
 , y 
 
 I '\ 
 
 In i 
 
 IlllS 
 
 3re 
 
CHAPTEll III. 
 
 V' i 
 
 TIIFKI) 
 JOl'UNl'.V. 
 
 Di.S(7overy of a large Coulauanara snake. — A Bush-master. — Stag swallowed 
 by a Boa. — Negroes and the snake. — Arrangements for the attack. — 
 The snake struck. — Currying ofl" tlie enemy. — A snake in a bag. — An 
 unquiet night. — Dissection of the snake. — Daddy Quashi and his dread 
 of snakes. — Capture of a Coulacanara. — Vultures and their food. — 
 Habits of Vultures. — The Aura vulture. — Black vultures. — Severe 
 blisters. — An imiuisitive Jaguar. — Fioh shooting. — Goatsuckers and 
 Canipanoro. 
 
 Lkt 119 now return to natural history. There was a 
 person making sliingles, with twenty or thirty negroes, 
 not far from Mibiri-hill. I liad offered a reward to any of 
 them wlio would find a good-sized snake in the forest, and 
 come and let me know where it was. Often had these 
 negr(^es looked for a large snake, and as often been dis- 
 appointed. 
 
 One Sunday morning I met one of them in the forest, 
 and asked him which way he was going : he said he was 
 going towards Warratilla creek to hunt an armadillo : and 
 he had his little dog with him. On coming back, about 
 noon, the dog began to bark at the root of a large tree, 
 which Iiad been upset by the whirlwind, and was lying 
 there in a gradual state of decay. The negro said, he 
 tliought his dog was barking at an acouri, which had pro- 
 bably taken refuge under the tree, and he went up with an 
 intention to kill it : he there saw a snake, and hastened 
 back to inform me of it. 
 
ring 
 he 
 
 jro- 
 an 
 
 led 
 
 WANDKUINCiS IN SOUTH AMEIJICA. 
 
 249 
 
 The sun had just passed tlie meridiuu in a cloudless sky ; Tiiini) 
 
 there was scarcely a bird to be seen, for the winged inliabit- '__1 ' 
 
 ants of the forest, as thoujih overcome by heat, had retired ^"7 "' 
 
 ' = . • search nf a 
 
 to the thickest shade : all would have been like midnight snob. 
 silence, were it not for the shrill voice of the I'i-pi-yo, 
 every now and then resounding from a distant tree. I was 
 sitting witii a little Horace in i, y hand, on what hiul once 
 been the steps which formerly led up to the now moulder- 
 ing and dismantled building. The negro and his little 
 dog came down the hill in haste, and I was soon informed 
 that a snake had been discovered; but it was a young 
 one, called the Bush-master, a rare and poisonous snake. 
 
 I instantly rose up, and laying hold of the eight-foot 
 lance, which was close by me, " Well then. Daddy," said I, 
 " we'll go and have a look at the snake." I was barefoot, 
 with an old hat, and check shirt, and trousers on, and a 
 pair of braces to keep them up. The negro had his cutlass, 
 and as we ascended the hill, another negro, armed with a 
 cutlass, joined us, judging, from our pace, that there was 
 something to do. The little dog came along with us, and 
 when we had got about half a mile in the forest, the negro 
 stopped, and pointed to the fallen tree : all was still and 
 silent : I told the negroes not to stir from the place where 
 they were, and keep the little dog in, and that I would go 
 in and reconnoitre. 
 
 I advanced up to the place slow and cautious. The Fimisand 
 snake was well concealed, but at last I made him out ; it enormous 
 was a Coulacanara, not poisonous, but large enough to have C'o«?«t7ni. 
 crushed any of us to death. On measuring him afterwards, 
 he was something more than fourteen feet long. This 
 species of snake is very rare, and much thicker, in pro- 
 portion to his length, than any other snake in the forest. 
 A coulacanara of fourteen feet in length is as thick as a 
 common boa of twenty-four. After skinning this snake I 
 
 • I': *i 
 
 vl 
 
 ! 
 
 I 
 
 
250 
 
 WANDERINGS IN SOUTH AMKUi! A. 
 
 Tiltiil) 
 JOIJKNKY, 
 
 !),L 
 
 could eiisily get my head into his mouth, as tlie singular 
 funiiiitioii of tlio jiiw.s admits of woiidorful extension. 
 
 A Duteh friend of mine, by name IJrouwer, killed a Uoa, 
 twenty-two feet long, with a pair of stag's horns in his 
 mouth : he had swallowed the stag, but could not get the 
 horns down : so ho had to wait in patience with tliat un- 
 comfortable mouthful till his stonuich digesteil the body, 
 and then the horns would drop out. In this plight the 
 l)utchman found him as he was going in his cauoc up the 
 river, and sent a ball through his head. 
 
 On ascertaining the size of the serpent which the negro 
 liad just found, I, retired slowly the way I came, and 
 promised four dollars to the negro who had shown it to me, 
 and one to the other who had joined us. Aware that the 
 day was on the decline, and that the approach of night 
 would be detrimental to the dissection, a thought struck 
 nie that I could take him alive. I imagined if I could 
 strike him with the lance behind the head, and pin him to 
 the ground, I might succeed in capturing him. AVlien 1 
 told this to the negroes, they begged and entreated me to 
 let them go for a gun, and bring more force, as they were 
 sure the snake would kill some of us. 
 
 I had been at the siege of Troy for nine years, and it 
 would not do now to carry back to Greece, " nil decimo 
 nisi dedecus anno." I mean, I had been in search of a 
 large serpent for years, and now having come up with one, 
 it did not become me to turn soft. So, taking a cutlass 
 from one of the negroes, and then ranging both the sable 
 slaves behind me, I told them to follow me, and that I 
 would cut them down if they odered to fly. I smiled as I 
 said this, but they shook their heads in silence, and seemed 
 to have but a bad heart of it. 
 
 When we got up to the place, the serpent had not stirred, 
 but I could see nothing of his head, and 1 judged by the 
 
WAXPEiiiNTis r\ sorrii ameiuca. 
 
 251 
 
 tblcls of his body that it ninst ho at the farthest side of liis 
 den. A species of woodlnne had formed a complete mantle 
 over the hranches of the fallen tree, almost impervious to 
 the rain, or the rays of the sun. Prohahly he had resorted 
 to this sequestered place for a length of time, as it bore 
 marks of an ancient settlement. 
 
 I now took my knife, determininp: to cut away the wood- 
 bine, and break the twigs in the gentlest manner possible, 
 till 1 could get a view of his head. One negro stood guard 
 close behind me with the lance ; and near him the other 
 with a cutlass. The cutlass which I had taken from the 
 first negro was on the ground close by me in case of need. 
 
 After working in dead silence for a quarter of an hour, 
 with one knee all the time on the ground, I had cleared 
 away enough to see his head. It appeared coming out 
 betwixt the first and second coil of his body, and was flat 
 on the ground. This was the very position I wished it to 
 be in. 
 
 I rose in silence and retreated very slowly, making a 
 sign to the negroes to do the same. The dog was sitting at 
 a distance in mute observance. I could now read in the 
 face of the negroes, that they considered this a very un- 
 pleasant affair ; and they made another attempt to persuade 
 me to let them go for a gun. I smiled in a good-natured 
 manner, and made a feint to cut them down with the 
 weapon I had in my hand. This was all the answer I 
 made to their request, and they looked very uneasy. 
 
 It must be observed, we were now about twenty yards 
 from the snake's den. I now ranged the negroes behind 
 me, and told him who stood next to me to lay hold of the 
 lance the moment I struck the snake, and that the other 
 nmst attend my movements. It now only remained to 
 take their cutlasses from them, for I was sure, if I did not 
 disarm them, they would be tempted to strike the snake in 
 
 TIIIIlI) 
 JOirilNKV. 
 
 1!l 
 
 
 » t, 
 
 l3t! ii 
 

 \'. i 
 
 
 .1: 
 
 ti i' 
 
 J; 
 
 IB!'!' ; 
 
 IB: I li 
 
 i; 
 
 26a 
 
 TIHRn 
 JOUKNKY, 
 
 WANDKKINCS IN SOUTH AMERICA. 
 
 time of danger, anJ thus for ever spoil his skin. On tak- 
 ing their cutlasses from them, if I might judge from their 
 physiognomy, they seemed to consider it as a most in- 
 tolerable act of tyranny in me. Probably nothing kept 
 them from bolting, but the consolation that I was to be be- 
 twixt them and the snake. Indeed, my own heart, in spite 
 of all I could do, beat quicker than usual ; and I felt those 
 sensations which one has on board a merchant-vessel in 
 war time, when the captain orders all hands on deck to 
 prepare for action, while a strange vessel is coming down 
 upon us under suspicious colours. 
 
 We went slowly on in silence, without moving our arms 
 or heads, in order to prevent all alarm as much as possible, 
 lest the snake should glide off, or attack us in self-defence. 
 I carried the lance perpendicularly before me, with tlie 
 point about a foot from the p'ound. The snake had not 
 moved ; and on getting up to him, I struck him with the 
 lance on the near side, just behind the neck, and pinned 
 him to the ground. That moment, the negro next to me 
 seized the lance, and held it firm in its place, while 1 
 dashed head foremost into the den to grapple with the 
 snake, and to get hold of his tail before he could do any 
 mischief. 
 
 On pinning hhn to tlie ground with the lance, he gave a 
 tremendous loud hiss, and the little dog ran away, howling 
 as he went. We had a sharp fray in the den, the rotten 
 sticks flying on all sides, and each party struggling for 
 superiority. I called out to the second negro to throw 
 himself upon me, as I found T was not lieavy enough. He 
 did so, and the additional weight was of great service. I 
 liad now got firm hold of his tail ; and after a violent 
 struggle or two, he gave in, finding himself overpowered. 
 This was the moment to secure him. So, while the first 
 negro continued to hold the lance firm to the ground, and 
 
WANDERINGS IN SOUTH AMERICA. 
 
 25.3 
 
 the other was helping me, I contrived to unloose my tihiid 
 braces, and with tliem tied up the snake's mouth. JoruNiiv. 
 
 The snake now finding himself in an unpleasant situa- 
 tion, tried to better liimself, and set resolutely to work, but 
 we overpowered him. We contrived to make him twist 
 himself round the shaft of the lance, and then prepared to 
 convey him out of the forest. I stood at his head and 
 held it firm under my arm, one negro supported the belly, 
 and the other the tail. lu this order we began to move 
 slowly towards home, and reached it after resting ten times ; 
 for the snake was too heavy for us to support him without 
 stopping to recruit our strength. As we proceeded onwards 
 with him, he fought hard for freedom, but it was all in 
 vain. The day was now too far spent to think of dissect- 
 ing him. Had i killed him, a partial putrefaction would 
 have taken place before morning. I had brought with me 
 up into the forest a strong bag, large enough to contain any 
 animal that I slrould want to dissect. I considered this 
 the best mode of keeping live wild animals Avhen I was 
 pressed for dayliglit ; for the bag yielding in every direc- 
 tion to their efforts, they would have nothing solid or fixed 
 to work on, and thus would be prevented from making a 
 hole through it. I say fixed, for after the mouth of the 
 bag was closed, the bag itself was not fastened or tied to 
 anything, but moved about wherever the animal inside 
 caused it to roll. After securing afresh the mouth of the 
 coulacanara, so that he could not open it, he was forced 
 into this bag, and left to his fate till morning. 
 
 I cannot say he allo\\'ed me to have a quiet night. My 
 hammock was in the loft just above him, and the floor 
 betwixt us, half gone to decay, so tliat in parts of it no 
 boards intervened betwixt his lodging-room and mine. He 
 was very restless and fretful ; and had Medusa been my 
 wife, there could not have been more continued and 
 
 , !!■ 
 
 1 1!; 
 
 hi 
 
 ^ 
 
 w 
 
 -; 
 
 Mi 
 
 ; JFf 
 
 : I 
 
 1 
 
254 
 
 WANDEinXGS IN SOUTH AMKIUCA. 
 
 Tiiiiin disagreeable hissing in the bedcJiamber that night. At day- 
 juritxEY. jjrgr^]^^ I gent to borrow ten of the negroes who were cut- 
 ting wood at a distance ; I could have done with half that 
 number, but judged it most prudent to have a good force, 
 in case he should try to escape from the house when we 
 opened the bag. However, nothing serious occurred: 
 Kills and We untied the mouth of the bag, kept him down by 
 dt^seds the j^j^in forcc, and then I cut his throat. He bled like an ox. 
 
 snake. ' 
 
 By six o'clock the same evening, he was completely dis- 
 sected. On examining his teeth, I observed that they were 
 all bent like tenter-hooks, pointing down his throat, and 
 not so large or strong as I expected to have found them ; 
 but they are exactly suited to what they are intended by 
 nature to perform. The snake does not masticate his food, 
 and thus the only service his teeth have to perform is to 
 seize his prey, and hold it till he swallows it whole. 
 
 In general, the skins of snakes are sent to museums 
 without the head : for when the Indians and Negroes kill 
 a snake, they seldom fail to cut off the head, and tlien they 
 run no risk from its teeth. When the skin is stuffed in 
 the museum, a wooden head is substituted, armed with 
 teeth which are large enough to suit a tiger's jaw; and 
 this tends to mislead the spectator, and give him erroneous 
 ideas. 
 
 During this fray with the serpent, the old negro, Daddy 
 Quashi, was in Georgetown procuring provisions, and just 
 returned in time to help to take the skin oft'. He had 
 spent best part of liis life in the forest with his old master, 
 Mr. Edmonstone, and amused me much in recounting their 
 many adventures amongst the wild beasts. The Daddy 
 had a particular horror of snakes, and frankly declared he 
 could never have faced the one in question. 
 
 The week following, his courage was put to the test, and 
 
 It was a curious contlict, and 
 
 A f lucks 
 snake. 
 
 he made good his words. 
 
list 
 lad 
 ter, 
 leir 
 
 AVANDERINGS IN SOFTH AMERICA. 
 
 255 
 
 took place near the spot where I had captured the large Tinun 
 snake. In the morning I had been following a new species J"''"^"'''^^- 
 of paroquet, and the day being rainy, I had taken an um- 
 brella to keep the gun dry, and had left it under a tree ; 
 in the afternoon I took Daddy Quashi with me to look for 
 it. Whilst he was searching about, curiosity took me to- 
 wards the place of the late scene of action. There was a 
 path where timber had formerly been dragged along. Here 
 I observed a young coulacanara, ten feet long, slowly mov- 
 ing onwards ; I saw he was not thick enough to break my 
 arm in case he got twisted round it. There was not a 
 moment to be lost. I laid hold of his tail witli the left 
 hand; one knee being on the ground ; with the ri'^htl took 
 my hat, and held it as you would liold a shield for defence. 
 
 The snake instantly turned, and came on at me, with his 
 head about a yard from the ground, as if to ask me, what 
 business I had to take liberties with his tail. I let hiin 
 come, hissing and open-mouthed, within two feet of my 
 face, and then, with all the force I was master of, I drove 
 my fist, shielded by my hat, full in his jaws. He was 
 stunned and confounded by the blow, and ere he coiiLl re- 
 cover himself, I had seized his thrml with both hand^, in 
 such a position that he coula not Iiivc me ; I then allowed 
 liini to coil himself round my body, and marched off with 
 him as my lawful prize. He pressed nie hard, but not 
 alarmingly so. 
 
 In the meantime. Daddy v^uashi having found the um- 
 brella, and having heard tbe noise Mdiich the fray occa- 
 sioned, was corning cautiously up. As soon as he saw me, 
 and in what company I was, he turned about and ran off 
 liome, I after him, and shouting to increase liis fear. On 
 scolding him for his cowardice, the old rogue begged that I 
 would forgive him, for that the sight of the snake had 
 positively turned him sick at stomach. 
 
 t ' 
 
 M';;' 
 
 :, I ■ 
 
 ■ \ 
 
 hi 
 
 1 
 
 = ? 
 
 If! 
 
 m" 
 
 '>\i 
 
 P'f 
 
 ^31 :) 
 
 
 ft 
 
256 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 THIRD 
 JOtKNKY. 
 
 The King 
 
 of the 
 
 Valures. 
 
 When I had done with the carcass of the large snake, it 
 was conveyed into the forest, as I expected that it would 
 attract the king of the vultures, as soon as time should 
 have rendered it sufficiently savoury. In a few days it 
 sent forth tliat odour which a carcass should send forth, 
 and about twenty of the common vultures came and 
 perched on the neighbouring trees ; the king of the 
 vultures came too ; and I observed that none of the 
 common ones seemed inclined to begin Ijreakfast till his 
 majesty had finished. When he had consumed as much 
 snake as nature informed him would do him good, he 
 retired to the top of a high mora-tree, and then all the 
 common vultures fell to, and made a hearty meal. 
 
 Tlie head and neck of the king of the vultures are bare 
 of feathers ; but the beautiful appearance they exhibit 
 fades in death. The throat and the back of tlie neck are 
 of a fine lemon colour ; both sides of the neck, from the 
 ears downwards, of a rich scarlet j behind the corrugated 
 part tliere is a white spot. The crown of the head is 
 scarlet ; betwixt the lower mandible and the eye, and close 
 by the ear, there is a part which has a fine silvery blue 
 appearance ; the corrugated part is of a dirty light brown ; 
 behind it, and just above the white spot, a portion of the 
 skin is blue, and the rest scarlet ; the skin wliicli juts out 
 behind the neck, and appears like an oblong caruncle, is 
 blue in part, and part orange. 
 
 The bill is orange and black, the caruncles on his fore- 
 head orange, and tlie cere orange ; the orbits scarlet, and 
 the irides wiiite. Below the bare part of the neck there is 
 a cinereous ruff. The bag of the stomach, whicli is only 
 seen when distended with food, is of a most delicate 
 white, intersected with bine veins, which appear on it just 
 like the blue veins on the arm of a fair-complexioned 
 person. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 257 
 
 The tail and long wing-featliers are black, the belly third 
 white, and the rest of the body a fine satin colour. jouknky. 
 
 I cannot be persuaded that the vultures ever feed upon 
 live i^nimals, not even upon lizards, rats, mice, or frogs ; I 
 have watched them for hours together, but never t;ould S':g ,j 
 
 them touch any living animals, though innumerable li;^ards, 
 frogs, and small birds swarmed all around them. I have 
 killed lizards and frogs, and put them in a proper place for 
 observation; as soon as they began to stink, the Aura 
 vulture invariably came and took them off. I have fre- 
 quently observed, that the day after the planter had burnt 
 the trash in a cane-field, the aura vulture was sure to be 
 there, feeding on the snakes, lizards, and frogs which had 
 suffered in the conflagration. I often saw a large bird 
 (very much like the common gregaidous vulture at a dis- 
 tance) catch and devour lizards ; after shooting one, it 
 turned out to be not a vulture, but a hawk, with a tail 
 squarer and shorter than hawks have in general. The 
 vultures, like the goatsucker and woodpecker, seem to be 
 in disgrace with man. They are generally termed a 
 voracious, stinking, cruel, and ignoble tribe. Under these 
 impressions, the fowler discharges his gun at them, and 
 probably thinks he has done well in ridding the earth of 
 such vermin. 
 
 Some governments impose a fine on him who kills a 
 vulture. This is a salutary law, and it were to be wished 
 tliat other governments would follow so good an example. 
 I would fain here say a word or two in favour of this 
 valuable scavenger. 
 
 Kind Providence has conferred a blessing on hot 
 countries in giving them the vulture ; he has ordered it to 
 consume that which, if left to dissolve in putrefaction, 
 would infect tlie air, and produce a pestilence. When full 
 of food, the vulture certainly appears an indolent bird ; he 
 
 *»t, 
 
 [111 
 
 '». 
 
 !!; 
 
 M if 
 
 11 
 
 fi ' 
 
 IP 
 
 '? i. ' 
 
 I V' 
 
 ' i 
 
 i: 
 
 .; 'i 
 
 
 '' 
 
 1 
 
 f 
 
25'? 
 
 WANDERINGS TN SOUTH AMERICA. 
 
 Tiirnn 
 
 JOUUXEF. 
 
 n-: 
 
 will stand for hours together on the branch of a tree, or on 
 the top of a house, with his wings drooping, and, after rain, 
 with them spread and elevated to catch the rays of the 
 sun. It has been remarked by naturalists, that the flight 
 of this bird is laborious. I have paid attention to the 
 vulture in Andalusia, and to those in Guiana, Brazil, and 
 the West Indies, and conclude that tliey are birds of long, 
 even, and lofty ilight. Indeed, whoever has observed the 
 aura vulture, will be satisfied that his flight is wonderfully 
 majestic, and of long continuance. 
 
 The bird is above five feet from wing to wing extended. 
 You will see it soaring aloft in the aerial expanse on 
 pinions which never flutter, and which at the same time 
 cai'ry him through the fields of ether with a rapidity equal 
 to that of the golden eagle. In Paramaribo the laws pro- 
 tect the vulture, an(i the Spaniards of Angustura never 
 think of molesting him. In 1808, I saw the vultures in 
 that city as tame as domestic fowls ; a person who had 
 never seen a vulture would have taken them for turkeys. 
 They were very useful to the Spaniards ; and had it not 
 been for them, the refuse of the slaughter-houses in 
 Angustura would have caused an intolerable nuisance. 
 
 The common black, short, square-tailed vulture, is gre- 
 garious ; but the aura vulture is not so ; for, though you 
 may see fifteen or twenty of them feeding on the dead 
 vermin in a cane-field, after the trash has been set fire to, 
 still, if you have paid attention to their arrival, you will 
 have obsev. d tiiat they came singly and retired singly, 
 and tlius tlioir being all 1 -getlier in the same field was 
 merely acciciental, and caused by each one smelling the 
 effluvia as he was soaring through the sky to look out 
 for food. I have watched twenty come into a cane-field ; 
 they arrived one by one, and from different parts of 
 the heavens. Hence we may conclude, that though the 
 
WANDERINGS IN SOUTH AMERICA. 
 
 259 
 
 other species of vulture are gregarious, the aura vulture tiiiud 
 
 is not. JOUKNEY. 
 
 If you dissect a vulture that has jusi: bepn feeding on 
 carrion, you must expect that your olfactory -lerves will be 
 somewhat offended with the rank eftluvia from his craw ; 
 just as tliey would he were you to dissect a citizen after 
 the Lord Mayor's dinner. If, on the contrary, the vulture 
 be empty at the time you commence the operation, there 
 will be no offensive smell, but a strong scent of musk. 
 
 I had long wished to examine the native haunts of the 
 Cayman ; but as the river Demerara did not afford a 
 specimen of the large kind, I was obliged to go to the 
 river Essequibo to look for one. 
 
 I got the canoe ready, and went down in it to George- 
 town ; where, having put in the necessary articles for the 
 expedition, not forgetting a couple of large shark-hooks, 
 with chains attached to them, and a coil of strong new 
 rope, I hoisted a little sail, which I had got made on 
 purpose, and at six o'clock in the morning shaped our 
 course for the river Essequibo. I had put a pair of shoes 
 on to prevent tlie tar at the bottom of the canoe from 
 oticking to my feet. The sun was flaming hot, and from 
 eleven o'clock till two beat perpendicularly upon the top 
 of my feet, betwixt the shoes and the trousers. Not 
 feeling it disagreeable, or being in the least aware of 
 painful consequences, as I had been barefoot for months, I 
 neglected to put on a pair of short stockings which I had 
 with me. I did not reflect, that sitting still in one i)lace, 
 witli your feet exposed to the sun, was very different from 
 being exposed to the sun while in motion. 
 
 We went ashore in the Essequibo, about three o'clock in Sufcrs 
 the afternoon, to choose a place for the night's residence, ''Inthffcet 
 
 to C'/Ilect firewood, and to set the fish-hooks. It was then/''^''^ <'^'<^^«- 
 
 sivc Jicctt 
 that I first began to find my legs very painful : they soon 
 
 s 2 
 
 M.Mii' 
 
 'V'vl 
 
2r)0 
 
 WANDEKINGS IN SOUTH AMERICA. 
 
 THinn 
 
 JOURNEY 
 
 Visited in 
 
 became much intljimed and red and blistered ; and it 
 required considerable caution not to burst the blisters, 
 otherwise sores would have ensued. I immediately got 
 into the hammock, and there passed a painful and sleep- 
 less night, and lor two days after I was disabled from 
 walking. 
 
 About midnight, as I was Ij'ing awake, and in great 
 
 ^ %ua '^ P^-iij I heard the Indian say, " Massa, massa, you no hear 
 
 Jaguar tiger ? " I listened attentively, and heard the softly sound- 
 
 '^^^ ' ing tread of his feet as he approached us. The moon had 
 
 gone down ; but every now and then we could get a glance 
 
 of him by tlie light of our fire ; he was the jaguar, for I 
 
 could see the spots on his body. Had I wished to have 
 
 fired at liira, I was not able to take a sure aim, for I was in 
 
 such pain that I could not turn myself in my hammock. 
 
 The Indian would have fired, but I would not allow him to 
 
 do so, as I wanted to see a little more of our new visitor ; 
 
 for it is not every day or night that the traveller is favoured 
 
 with an undisturbed sight of the jaguar in his own forests. 
 
 Whenever the fire got low, the jaguar came a little 
 nearer, and when the Indian renewed it, he retired 
 abruptly ; sometimes he would come within twenty yards, 
 and then we had a view of him, sitting on his hind legs 
 like a dog ; sometimes he moved slowly to and fro, and 
 at other times we could hear him mend his pace, as if 
 impatient. At last the Indian, not relishing the idea of 
 having such comx>any in the neighbourhood, could contain 
 himself no longer, and set up a most tremendous yell. 
 The jaguar bounded off like a racehorse, and returned no 
 more ; it appeared by the print of his feet the next morn- 
 ing that he was a full-grown jaguar. 
 Jteachesthe In two days after this we got to the first falls in the Es- 
 iJequikK sequibo. There was a superb barrier of rocks quite across 
 the river. In the rainy season these rocks are for the most 
 
PANDERINGS IN SOUTH AMERICA. 
 
 2r,i 
 
 part under water ; but it being now dry weatlier, we had a third 
 fine view of them, wliile tlie water from the river above '^'^' ^"^''; ^' 
 them rushed tlirough the different openings in nuijestic 
 grandeur. Here, on a little hill, jutting out into the river, 
 stands the house of Mrs. Peterson, the last house of people 
 of colour up this river ; I hired a negro from her, and a 
 coloured man, who pretended tliat they knew the haunts of 
 the cayman, and understood everything about taking him. 
 We were a day in passing these falls and rapids, celebrated 
 for the Pacou, the richest and most delicious fish in Guiana. 
 The coloured man was now in his element ; he stood in the 
 head of the canoe, and with his bow and arrow shot the 
 pacou as they were swimming in the stream. The arrow 
 had scarcely left the bow before he had plunged headlong 
 into the river, and seized the fish as it was struggling with 
 it. He dived and swam like an otter, and rarely missed the 
 fish he aimed at. 
 
 Did my pen, gentle reader, possess descriptive powers, I 
 would here give thee an idea of the enchanting scenery of 
 the Essequibo ; but that not being the case, thou must be 
 contented with a moderate and well-intended attempt. 
 
 Nothing could be more lovely than the appearance of Scnunj. 
 the forest on each side of this noble river. Hills rose on 
 liills in fine gradation, all covered with trees of gigantic 
 height and size. Here their leaves were of a lively purple, 
 and there of the deepest green. Sometimes the Caracara 
 extended its scarlet Wossoms from branch to branch, and 
 gave the tree the appearance as though it had been hung 
 with garlands. 
 
 This delightful scenery of the Essequibo made the soul 
 overflow with joy, and caused you to rove in fancy through 
 fairy-land ; till, on turning an angle of the river, you were 
 recalled to more sober reflections on seeing the once 
 grand and towering mora, now dead and ragged in its top- 
 
 ' 
 
 11 
 
 ( 
 
 i 
 
 tr:i; 
 
 ' > 
 
 ':; i 
 
 !?■ 
 
 R: 
 
 IL 
 
 Hi 
 
 
 ! ii 
 
 "ill 
 
 
 m 
 
ip 
 
 
 262 
 
 WANDERINGS IN SOUTH AMEUICA. 
 
 THIRD most branches, while its aged trunk, undermined by tlie 
 
 JO URNE Y, rugiijng torrent, hung as tliough in sorrow over the river, 
 
 which, eie long, would receive it, and sweep it away ibr ever. 
 
 During the day, the trade-wind blew a gentle and re- 
 freshing breeze, which tlied away as the ni-ht set in, and 
 then the river was as smooth as glass. 
 
 The moon was within three days of being full, so that 
 we did not regret the loss of the sun, which set in all its 
 splendour. Scarce had he sunk behind the western hills, 
 when the Goatsuckers sent forth their soft and plaintive 
 cries ; some often repeating, ' ' Who are you — who, who, 
 who ure you ? " and others, " Willy, Willy, Willy come go." 
 
 The Indian and Daddy Quashi often shook their heads at 
 this, and said they were bringing talk from Yabahou, who 
 is the evil spirit of the Essequibo. It was delightful to sit 
 on the branch of a fallen tree, near the water's edge, and 
 listen to these harmless birds as they repeated their even- 
 ing song ; and watch the owls and vampires as they every 
 now and then passed up and down tlie river. 
 
 The next day, about noon, as we were proceeding on- 
 wards, we heard the Campanero tolling in the depth of the 
 forest. Though I should not then have stopped to dissect 
 even a rare bird, having a greater object in view, still I 
 could not resist the opportunity offered of acquiring the 
 campanero. The place where he was tolling was low and 
 swampy, and my legs not having quite recovered from the 
 effects of the sun, I sent the Indian to shoot the campanero. 
 He got up to the tree, which he described as very high, 
 with a naked top, and situated in a swamp. He fired at 
 the bird, but either missed it, or did not wound it suffi- 
 ciently to bring it down. This was the only opportunity I 
 had of getting a campanero during this expedition. We 
 had never heard one toll before this morning, and never 
 heard one after. 
 
 Tfte C'-tm- 
 pancru. 
 
the 
 ssect 
 ill I 
 the 
 and 
 the 
 
 CHAT'TEE IV. 
 
 Fishing for a Cayman. — A slin .ook useless. — Sting-rays.— Tnrtlo and 
 Guana nests. — Numbers of eggs. — Anotlier failure, — fleeting a Jaguar. 
 — Guard against fever. — More failures. — A native hook and way of 
 baiting. — The Cayman's dinner-bell. — Caught at last. — How to scture 
 the reptile. — Mounting a Cayman. — An improvised bridle. — Skin and 
 teeth of the Cayman. — Embarkment for England. — Collision with the 
 Custom House. 
 
 About an hour before sunset, we reached the place Avhich TiiiRn 
 the two men who had joined its at the falls pointed out '^" ^"^^^ ^' 
 as a proper one to find a Cayman. There was a large 
 creek close by, and a sand-bank gently sloping to the 
 water. Just within the forest on this bank, we cleared a 
 place of brushwood, suspended the hammocks from the 
 trees, and then picked up enough of decayed wood for 
 fuel. 
 
 The Indian found a large land tortoise, and this, with 
 plenty of fresh fish which we had in the canoe, afforded a 
 supper not to be despised. 
 
 The tigers had kept up a continued roaring every night Roarhui of 
 since we had entered the Essequibo. The sound was '^ '^^'** 
 awfully fine. Sometimes it was in the immediate neigh- 
 bourhood; at other times it was far off, and echoed 
 amongst the hills like distant thunder. 
 
 It may, perhaps, not be amiss to observe here, that when 
 the word tiger is used, it does not mean the Bengal tiger. 
 
 'iii 
 
 '1' I 
 
 l« 
 
 ■I 
 
 ¥ 
 
 
 k 
 
 1 [! 
 
 ' i\ 
 
 ''■' ' 
 
 'l-.i 
 
 1 , A 
 ■ ■ ' 
 
 |; 
 
 ! ■ 
 
 fl i^ 
 
 »(• 
 
 P 
 
 r 
 
 , ■ 
 
 1 
 
 ■ ■, M- 
 
 \{. 
 
 I' : 
 
 : ; i 1' 
 
 ! I 
 
 I 
 
 

 ■■■' V 
 
 IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 1.0 
 
 1.1 
 
 m ^^" ■■■ 
 
 I2t 1^ 1122 
 2.0 
 
 IM 
 
 
 Hiotographic 
 
 Sdaices 
 
 Corporatian 
 
 
 13 WMT MAM STRHT 
 
 WtUTIR,N.Y. USM 
 (7U)«7a-4MS 
 
 

 ^ 
 
 
204 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 THIRD 
 JOITIINKY. 
 
 Noise of 
 
 the Cay' 
 
 men. 
 
 It means the jaguar, whose skin is beautifully spotted, and 
 not striped like that of the tiger in the East. It is, in fact, 
 the tiger of the new world, and receiving the name of tiger 
 from the discoverers of South America, it has kept it ever 
 since. It is a cruel, strong, and dangerous beast, but not 
 so courageous as the Bengal tiger. 
 
 We now baited a shark-hook with a large fish, and put 
 it upon a board about a yard long, and one foot broad, 
 which we had brought on purpose. This board was carried 
 out in the canoe, about forty yards into the river. By 
 means of a string, long enough to reach the bottom of the 
 river, and at the end of which string was fastened a stone, 
 the board was kept, as it were, at anchor. One end of the 
 new rope I had bought in town was reeved through the 
 chain of the shark-hook, and the other end fastened to a 
 tree on the sand-bank. 
 
 It was now an hour after sunset. The sky was cloud- 
 less, and the moon shone beautifully bright. There was 
 not a breath of wind in the heavens, and the river seemed 
 like a large plain of quicksilver. Every now and then a 
 huge Ush would strike and plunge in the water ; then the 
 owls and goatsuckers would continue their lamentations, 
 and the sound of these was lost in the prowling tiger's 
 growl. Then all was still again, and silent as midnight. 
 
 The caymen were now upon the stir, and at intervals 
 their noise could be distinguished amid that of the jaguar, 
 the owls, the goatsuckers, and frogs. It was a singular and 
 awful sound. It was like a suppressed sigh, bursting forth 
 all of a sudden, and so loud that you might hear it above 
 a mile off. First one emitted this horrible noise, and then 
 another answered him ; and on looking at the countenances 
 of the people around me, I could plainly see that they 
 expected to have a cayman that night. 
 
 We were at supper, when the Indian, who seemed to 
 
WANDERINGS IN SOUTH AMERICA. 
 
 265 
 
 have had one eye on the turtle-pot and the other on the third 
 bait in the river, said he saw the cayman coming. journey. 
 
 Upon looking towards the place, there appeared some- 
 thing on the water like a black log of wood. . It was so 
 unlike anything alive, that I doubted if it were a cay- 
 man ; but the Indian smiled, and said he was sure it was 
 one, for he remembered seeing a cayman, some years ago, 
 when he was in the Essequibo.. 
 
 At last it gradually approached the bait, and the board 
 began to move. The moon shone so bright that we could 
 distinctly see him open his huge jaws, and take in the 
 bait. We pulled the rope. He immediately let drop the 
 bait ; and then we saw his black head retreating from the 
 board, to the distance of a few yards, and there it remained 
 quite motionless. 
 
 He did not seem inclined to advance again ; and so we 
 finished our supper. In about an hour's time he again 
 put himself in motion, and took hold of the bait. But, 
 probably, suspecting that he had to deal with knaves and 
 cheats, he held it in his mouth but did not swallow it. 
 We pulled the rope again, but with no better success than 
 the first time. 
 
 He retreated as usual, and came back again in about an 
 hour. We paid him every attention till three o'clock in 
 the morning ; when, worn out with disappointment, we 
 went to the hammocks, turned in, and fell asleep. 
 
 When day broke, we found that he had contrived to get 
 the bait from the hook, though we had tied it on with string. 
 We had now no more hopes of taking a cayman till the 
 return of night. The Indian took off into the woods, and 
 brought back a noble supply of game. The rest of us 
 went into the canoe, and proceeded up the river to shoot 
 fish. We got even more than we could use. 
 
 As we approached the shallows, we could see the large 
 
 M 
 
 L ,,,, 
 
206 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 Birds. 
 
 W 
 
 l»> 
 
 THi:*D Sting-rays moving at the bottom. The coloured man 
 JO URNE Y, never failed to hit them with his arrow. The weather 
 was delightful. There was scarcely a cloud to intercept 
 the sun'p rays. 
 
 I saw several scarlet aras, anihingas, and ducks, but 
 could not get a shot at them. The parrots crossed the 
 river in innumerable quantities, always flying in pairs. 
 Here, too, I saw the Sun-bird, called Tirana by the 
 Spaniards in the Oroonoque, and shot one of them. The 
 black and white Scarlet-headed Finch was very common 
 here. I could never see this bird in the Demerara, nor 
 hear of its being there. 
 
 We at last came to a large sand-bank, probably two 
 miles in circumference. As we approached it we could 
 see two or three hundred Fresh-water Turtle on the edge 
 of the bank. Ere we could get near enough to let fly an 
 arrow at them, they had all sunk into the river and 
 appeared no more. 
 
 We went on the sand-bank to look for their nests, as 
 this was the breeding season. The coloured man showed 
 us how to find them. Wherever a portion of the sand 
 seemed smoother than the rest, there was sure to be a 
 turtle's nest. On digging down with our hands, about nine 
 inches deep, we found from twenty to thirty white eggs ; 
 in less than an hour we got above two hundred. Those 
 which had a little black spot or two on the shell we ate 
 the same day, as it was a sign that they were not fresh, 
 and of course would not keep : those which had no speck 
 were put into dry sand, and -were good some weeks after. 
 
 At miduight, two of our people went to this sand-bank, 
 while the rest stayed to watch the cayman. The turtle had 
 advanced on to the sand to lay their eggs, and the men 
 got betwixt them and the water ; they brought off half a 
 dozen very fine and well-fed turtle. The egg-shell of the 
 
 Turtle^ 
 lusts. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 267 
 
 fresh-water turtle is not hard like that of the land tor- thikd 
 toise, but appears like white parchment and gives way to '^*^ ^"^^' ^' 
 the pressure of the fingers; but it is very tough, anr" 
 does not break. On this sand-bank, close to the forest, 
 we found several Guana's nests ; but they had never more 
 than fourteen eggs a-piece. Thus passed the day in 
 exercise and knowledge, till the sun's declining orb 
 reminded us it was time to return to the place from 
 whence we had set out. 
 
 The second night's attempt upon the cayman was a 
 repetition of the first, quite unsuccessful. We went a 
 fishing the day after, had excellent sport, and returned to 
 experience a third night's disappointment. On the fourth 
 evening, about four o'clock, we began to erect a stage 
 amongst the trees, close to the water's edge. From this 
 we intended to shoot an arrow into the cayman : at the 
 end of this arrow was to be attached a string, which would 
 be tied to the rope, and as soon as the cayman was struck 
 we were to have the canoe ready and pursue him in the 
 river. 
 
 While we were busy in preparing the stage, a tiger 
 began to roar. We judged by the sound that he was not 
 above a quarter of a mile from us, and that he was close 
 to the side of the river. Unfortunately, the Indian said 
 it was not a jaguar that was roaring, but a couguar. The Couguar. 
 couguar is of a pale, brownish red colour, and not as large 
 as the jaguar. As there was nothing particular in this 
 animal, I thought it better to attend to the apparatus for 
 catching the cayman than to go in quest of the couguar. 
 The people, however, went in the canoe to the place 
 where the couguar was roaring. On arriving near the 
 
 I! 
 
 spot, they saw it was not a couguar, but an immense 
 jaguar, standing on the trunk of an aged mora-tree, which 
 bended over the river ; he growled and showed his teeth 
 
 1.;,! j 
 
2G8 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 Ill'' 
 III" 
 
 Tn:-'D as they approached; the coloured man fired at him with 
 jouKNET. ^ j^^ji^ i^y^ probably missed him, and the tiger instantly 
 descended, and took off into the woods. I .vent to the 
 place before dark, and we searched the forest for about 
 lialf a mile in the direction he had fled, but we could see 
 no traces of him, or any marks of blood ; so I concluded 
 that fear had prevented the man from taking steady aim. 
 
 We spent best part of the fourth night in trying for the 
 cayman, but all to no purpose. I was now convinced that 
 something was materially wrong. We ought to have been 
 successful, considering our vigilance and attention, and 
 that we had repeatedly seen the cayman. It was useless 
 to tarry here any longer ; moreover, the coloured man began 
 to take airs, and fancied that I could not do without him. 
 Discharges I never admit of this in any expedition where I am com- 
 ' coXmt.^ ™^^^*^^^» ^^^ ®® ^ convinced the man, to his sorrow, that 
 I could do without him ; for I paid him what I had agreed 
 to give him, which amounted to eight dollars, and ordered 
 him back in his own curial to Mrs. Peterson's, on the 
 hill at the first falls. I then asked the negro if there 
 were any Indian settlements in the neighbourhood; he 
 said he knew of one, a day and a half off. We went in 
 quest of it, and about one o'clock the next day the negro 
 showed us the creek where it was. 
 
 The entrance was so concealed by thick bushes that a 
 stranger would have passed it without knowing it to be 
 a creek. In going up it we found it dark, winding, and 
 intricate beyond any creek that I had ever seen before 
 When Orpheus came back with his young wife from Styx, 
 his path must have been similar to this, for Ovid says 
 it was 
 
 " Arduus, oTdiquus, caligine densus opaca," 
 
 and this creek was exactly so. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 269 
 
 Indion 
 'f nner. 
 
 When we had got about two-thirds up it, we met the thiud 
 Indians going a fisning. I saw, by the way their things •'**^[]^y. 
 were packed in the curial, that they did not intend to 
 return for some days. However, on telling them what we 
 wanted, and by promising handsome presents of powder, 
 shot, and hooks, they dropped their expedition, and invited 
 us up to the settlement they had just left, and where we 
 laid in a provision of cassava. 
 
 They gave us for dinner boiled ant-bear and red 
 monkey; two dishes unknown even at Beauvilliers in 
 Paris, or at a London city feast. The monkey was very 
 good indeed, but the ant-bear had been kept beyond its 
 time ; it stunk, as our venison does in England ; and so, 
 after tasting it, I preferred dining entirely on monkey. 
 After resting here, we went back to the river. The 
 Indians, three in number, accompanied us in their own 
 curial, and, on entering the river, pointed to a place a 
 little way above, well calculated to harbour a cayman. 
 The water was deep and still, and flanked by an immense 
 sand-bank ; there was also a little shallow creek close by. 
 
 On this sand-bank, near the forest, the people made a 
 shelter for the night. My own was already made ; for I 
 always take with me a painted sheet, about twelve feet 
 by ten. This, thrown over a pole, supported betwixt two 
 trees, makes you a capital roof with very little trouble. 
 
 We showed one of the Indians the shark-hook. He 
 sh-'.^k his head and laughed at it, and said it would not 
 do. When he was a boy, he had seen his father catch the 
 caymen, and on the morrow he would make something 
 that would answer. 
 
 In the meantime, we set the shark-hook, but it availed 
 us nought; a cayman came and took it, but would not 
 swallow it. 
 
 Seeing it was useless to attend the shark-hook any 
 
I 
 
 
 270 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 tif. 
 
 THinD longer, we left it for the night, and returned to our 
 jouKNEY. hammocks. 
 
 Ere I fell asleep, a reflection or two broke in upon me. 
 I considered, that as far as the judgment of civilized man 
 went, everything had been procured and done to ensure 
 success. We had hooks, and lines, and baits, and patience ; 
 we had spent nights in watching, had seen the cayman 
 come and take the bait, and after our expectations had 
 been wound up to the highest pitch, all ended in disap- 
 pointment. Probably this poor wild man of the woods 
 would succeed by means of a very simple process ; and 
 thus prove to his more civilized brother that, notwith- 
 standing books and schools, there is a vast deal of know- 
 ledge to be picked up at every step, whichever way we 
 turn ourselves. 
 
 In the morning, as usual, we found the bait gone from 
 the shark-hook. The Indians went into the forest to 
 liunt, and we took the canoe to shoot fish and get another 
 supply of turtles' eggs, which we found in great abundance 
 on this large sand-bank. 
 
 We went to the little shallow creek, and shot some 
 young caymen, about two feet long. It was astonishing 
 to see what spite and rage these little things showed when 
 the arrow struck them ; they turned round and bit it, and 
 snapped at us when we went into the water to taKC them 
 up. Daddy Quashi boiled one of them for his dinner, and 
 found it very sweet and tender. I do not see why it 
 should not be as good as frog or veal. 
 
 The day was now declining apace, and the Indian had 
 made his instrument to take the cayman. It was very 
 simple. There were four pieces of tough hard wood, a foot 
 long, and about as thick as your little finger, and barbed 
 at both ends ; they were tied round the end of the rope, 
 in such a manner, that if you conceive the rope to be an 
 
WAND15RINGS IN SOUTH AMERICA. 
 
 271 
 
 arrow, these four sticks would form the arrow's head ; so thikd 
 that one end of the four united sticks answered to the -^o^^^'' 
 point of the arrow-head, while the other ends of the 
 sticks expanded at equal distances round the rope, thus — 
 
 rr^rrrrrrr^iTf rrri^t 
 
 Now it is evident, that if the cayman swallowed this, (the 
 other end of the rope, which was thirty yards long, being 
 fastened to a tree,) the more he pulled, the faster the barbs 
 would stick into his stomach. This wooden hook, if you 
 may so call it, was well baited with the flesh of the 
 acouri, and the entrails were twisted round the rope for 
 about a foot above it. 
 
 Nearly a mile from where we had our hammocks, the 
 sand-bank was steep and abrupt, and the river very still 
 and deep ; there the Indian pricked a stick into the sand. 
 It was two feet long, and on its extremity was fixed the 
 machine ; it hung suspended about a foot from the water, 
 and the end of the ropo was made fast to a stake driven 
 well into the sand. 
 
 The Indian then took the empty shell of a land tortoise 
 and gave it some heavy blows with an axe. I asked him 
 
 1 
 
 Hi 
 
 n 
 
 l| 
 
 Ml 
 
 ! 
 
 i 
 
 Ifif 
 
 
 IK' 
 
 t; 
 
 iH'^l 
 
 
 il -';■ 
 
 'til 
 
 111 
 
 '1- 
 
272 
 
 WANDERIXGS IN SOUTH AMERICA. 
 
 THIRD 
 JOUKNEY. 
 
 
 Succeed in 
 hooking a 
 Cayman. 
 
 why he did that. He said it was to let the cayman hear 
 that something was going on. In fact the Indian meant 
 it as the cayman's dinner-bell. 
 
 Having done this, we went back to the hammocks, not 
 intending to visit it again till morning. During the night, 
 the jaguars roared and grumbled in the forest, as though 
 the world was going wrong with them, and at intervals we 
 could hear the distant cayman. The roaring of the jaguars 
 was awful ; but it was music to the dismal noise of these 
 hideous and malicious reptiles. 
 
 About half-past five in the morning, the Indian stole off 
 silently to take a look at the bait. On arriving at the 
 place he set up a tremendous shout. We all jumped out 
 of our hammocks, and ran to him. The Indians got there 
 before me, for they had no clothes to put on, and I lost 
 two minutes in looking for my trousers and in slipping 
 into them. 
 
 We found a cayman, ten feet and a half long, fast to 
 the end of the rope. Nothing now remained to do, but 
 to get him out of the water without injuring his scales, 
 " hoc opus, hie labor." We mustered strong : there were 
 three Indians from the creek, there was my own Indian 
 Yan, Daddy Quaslii, the negro from Mrs. Peterson's, James, 
 Mr. R. Edmonstone's man, whom I was instructing to 
 preserve birds, and lastly, myself. 
 
 I informed the Indians that it was my intention to draw 
 him quietly out of the water, and tlien secure him. They 
 looked and stared at each other, and said I might do it 
 myself ; but they would have no hand in it ; the cayman 
 would worry some of us. On saying this, " consedere 
 duces," they squatted on their hams with the most perfect 
 indifference. 
 
 The Indians of these wilds have never been subject to 
 
 the least restraint : and I knew 
 
 enough 
 
 of them to be 
 
W.\\I)EUI\(iS IN Sul'TII AMBllir.A. 
 
 273 
 
 aware, that if I tried to force tliem against their will, they 
 would take off, and leave nie and my presents unheeded, 
 and never return. 
 
 Daddy Quashi was for applying to our guns, as usual, con- 
 sidering them our best and safest friends. I immediately 
 offered to knock him down for his cowardice, and he 
 shrunk back, begging that I would be cautious, and not 
 get myself worried ; and apologizing for his own want of 
 resolution. My Indian was now in conversation with the 
 others, and they asked if I would allow them to shoot a 
 dozen arrows into him, and thus disable him. This would 
 have ruined all. I had come above three hundred miles 
 on purpose to get a cayman uninjured, and not to carry 
 back a mutilated specimen. I rejected their proposition 
 with firmness, and darted a disdainful eye upon the 
 Indians. 
 
 Daddy Quashi was again beginning to remonstrate, and 
 I chased him on the sand-bank for a quarter of a mile. 
 He told me afterwards, he thought he should have dropped 
 down dead with fright, for he was firmly persuaded, if I 
 had caught him, I should have bundled him into the cay- 
 man's jaws. Here then we stood, in silence, like a calm 
 before a thunder-storm. " Hoc res summa loco. Scinditur 
 in contraria vulgus." They wanted to kill him, and I 
 wanted to take him alive. 
 
 I now walked up and down the sand, revolving a dozen 
 projects in my head. The canoe was at a considerable 
 distance, and I ordered the people to bring it round to the 
 place where we were. The mast was eight feet long, and 
 not much thicker than my wrist. I took it out of the 
 canoe, and wrapped the sail round the end of it. Now it 
 appeared clear to me, that if I went down upon one knee, 
 and held the mast in the same position as the soldier holds 
 his bayonet when rushing to the charge, I could force it 
 
 T 
 
 THIHD 
 JOL'llNEV. 
 
 1!" 
 
?74 
 
 WAXDEniXGS IN SOUTH AMERICA. 
 
 * I 
 
 Tiiinn 
 
 JOiniNEY, 
 
 Prepare to 
 
 Ittb'. (he 
 
 (Jinnnan 
 
 alive. 
 
 down tlio cnymnii's throat, should he come open-mouthed 
 at luo. WluMi this was tokl to the Indians, they brightened 
 up, and said they would help me to pull him out of the 
 river. 
 
 " Brave squad ! " said I to myself, " * Audax omnia 
 pcrpeti,' now that you have got me betwixt yourselves 
 and danger." I then mustered all liands for the last 
 time before the battle. We were, four South American 
 savages, two negroes from Africa, a Creole from Trinidad, 
 and myself a wliito man from Yorkshire. In fact, a little 
 tower of Babel group, in dress, no dress, address, and 
 language. 
 
 Daddy Quasi li hung in the rear; I showed him a large 
 Spanish knife, which I always carried in the waistband of 
 my trousers : it spoke volumes to him, and he shrugged 
 up his shoulders in absolute despair. The sun was just 
 peeping over the high forests on the eastern hills, as if 
 coming to look on, and bid us act with becoming fortitude. 
 I placed all the people at the end of the rope, and ordered 
 them to pull till the cayman appeared on the surface of 
 the water ; and then, should hp i^lunge, to slacken the rope 
 and let bin; go again into the deep. 
 
 I now took the mast of the canoe in my hand (the sail 
 being tied round the end of the mast) and sunk down upon 
 one knee, about four yards from the water's edge, deter- 
 mining to thrust it down his throat, in case he gave mo 
 an opportunity. I certainly felt somewhat uncomfortable 
 in this situation, and I thought of Cerberus on the other 
 .side of the Styx ferry. The people pulled the cayman to 
 the surface ; he plunged furiously as soon as he arrived in 
 these upper regions, and immediately went below again 
 on their slackening the rope. ■ I saw enough not to fall 
 in love at first sight. I now told them we would run all 
 risks, and have him on land immediately. They pulled 
 
WANDERINGS IN SOUTH AMERICA. 
 
 875 
 
 to 
 in 
 tin 
 all 
 all 
 ed 
 
 again, and out ho catno, — " monstrunvhonenduni, informe." Tinnn 
 This was an interesting moment. I kept my position •"•' ^'"'^'''■ '^"- 
 firmly, with my eye fixed steadfast on him. 
 
 By this time the cayman was within two yards of me. 
 I saw he was in a state of fear and perturbation ; I in- 
 stantly dropped the mast, sprang up, and jumped on his 
 back, turning half round as I vaulted, so that I gained my 
 seat with my face in a right position. I immediately 
 seized his fore legs, and, by main force, twisted them on 
 his back ; thus they served me for a bridle. 
 
 He now seemed to have recovered from his surprise, and 
 probably fancying himself in liostile company, he began to 
 plunge furiously, and lashed tlie sand with his long and 
 powerful tail. I was out of reach of the strokes of it, by 
 being near his head. He continued to plunge and strike, 
 and made my seat very uncomfortable. It must have been 
 a fine sight for an unoccupied spectator. 
 
 The people roared out in triumph, and were so vociferous, 
 that it was some time before they heard me tell them to 
 pull me and my beast of burden farther inland. I was 
 apprehensive the rope m.ight break, and then there would 
 have been every chance of going down to the regions under 
 water with the cayman. That would have been more 
 perilous than Arion's marine morning ride : — 
 
 " Delphini insidens vada cserula sulcat Arion." 
 
 The people now dragged us above forty yards on the 
 sand : it was the first and last time I was ever on a 
 cayman's back. Should it be asked, how I managed to 
 keep my seat, I would answer, — I hunted some years with 
 Lord Darlington's fox-hounds. 
 
 After repeated attempts to regain his liberty, the cayman 
 gave in, and became tranquil through exhaustion. I now 
 managed to tie up his jaws, and firmly secured his foie- 
 
 T 2 
 
 i 
 
 
 { 
 
 '' ' 
 
 bI 
 
 1 
 
 
 ' ': 
 
 1 
 
 
 ili 
 
276 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 THIRD foet in the position I had held them. We had now another 
 JO URNE Y, severe struggle for superiority, but he was soon overcome 
 and again remained quiet. While some of the people 
 were pressing upon his head and shoulders, I threw myself 
 on his tail, and by keeping it down to the sand, prevented 
 him from kicking up another dust. He was finally con- 
 veyed to the canoe, and then to the place where we had 
 suspended our hammocks. There I cut his throat ; and, 
 after breakfast was over, commenced the dissection. 
 
 Now that the affray had ceased, Daddy Quashi played a 
 good finger and thumb at breakfast; he said he found 
 himself much revived, and became very talkative and 
 useful, as there was no longer any danger. He was a 
 faithful, honest negro. His master, my worthy friend 
 Mr. Edmonstone, had been so obliging as to send out 
 particular orders to the colony, that the Daddy should 
 attend me all the time I was in the forest. He had lived 
 in tlie wilds of Demerara with Mr. Edmonstone for many 
 years ; and often amused me with the account of the frays 
 his master had had in the woods with snakes, wild beasts, 
 and runaway negroes. Old age was now coming fast upon 
 liim ; he had been an able fellow in his younger days, and 
 a gallant one too, for he had a large scar over his eyebrow, 
 caused by the stroke of a cutlass, from another negro, 
 while the Daddy was engaged in an intrigue. 
 The hack of ^^^^ hoxik of the cayman may be said to be almost im- 
 p'jnetrable to a musket-ball, but his sides are not near so 
 st/ong, and are easily pierced with an arrow ; indeed, were 
 they a? strong as the back and the belly, there would be 
 no part of the cayman's body soft and elastic enough to 
 admit of expansion after taking in a supply of food. 
 
 The cayman has no grinders ; his teeth are entirely made 
 for snatch i.nd swallow ; there are thirty-two in each jaw. 
 Perhaps no animal in existence bears more decided marks 
 
 tlie 
 Cayman. 
 
 Its teeth. 
 
WANDERINGS IN SOUTH AMEHK'A. 
 
 277 
 
 n- 
 
 iso 
 pe 
 be 
 
 Ito 
 
 de 
 
 Iw. 
 ks 
 
 in his countenance of cruelty and malice than the cayman, third 
 He is the scourge and terror of all the large rivers in 'journey. 
 South America near the line. 
 
 One Sunday evening, some years ago, as I was walking Anecdote. 
 with Don Felipe de Ynciarte, governor of Angustura, on 
 the bank of the Oroonoque, " Stop here a minute or two, 
 Don Carlos," said he to me, " while I recount a sad acci- 
 dent. One iine evening last year, as the people of Angus- 
 tura were sauntering up and down here, in the Alameda, 
 I was within twenty yards of this place, when I saw a 
 large cayman rush out of the river, seize a man, and carry 
 him down, before anybody had it in his power to assist 
 him. The screams of the poor fellow were terrible as the 
 cayman was running off with him. He plunged into the 
 river with his prey ; we instantly lost sigiit of him, and 
 never saw or heard him more." 
 
 I was a day and a half in dissecting our cayman, and 
 then we all got ready to return to Demerara. 
 
 It was much more perilous to descend than to ascend 
 the falls in the Essequibo. 
 
 The place we had to pass had proved fatal to four Indians Great 
 about a month before. The water foamed, and dashed and descending 
 boiled, amongst the steep and craggy rocks, and seemed to <^c falls of 
 warn us to be careful how we ventured there. 
 
 I was for all hands to get out of the canoe, and then, 
 after lashing a long rope ahead and astern, we might have 
 climbed from rock to rock, and tempered her in her passage 
 down, and our getting out would have lightened her much. 
 But the negro who had joined us at Mrs. Peterson's said 
 he was sure it would be safer to stay in the canoe while 
 she went down the fall. I was loath to give way to him ; 
 but I did so this time against my better judgment, as he 
 assured me that he was accustomed to pass and repass 
 these falls. 
 
 quibo. 
 
 ' I r I j 
 
278 
 
 AVANDERIXGS IN SOUTH AMEllICA. 
 
 TiriiiD Accordingly we determined to push down : I was at the 
 j oruNE Y. helm, the rest at their paddles. But before we got half 
 way through, the rushing waters deprived the canoe of all 
 power of steerage, and she became the sport of the torrent ; 
 in a second she was half full of water, and I cannot com- 
 prehend to this day why she did not go down ; luckily the 
 people exerted themselves to the utmost, — she got headway, 
 and they pulled through the whirlpool ; I being quite in 
 the stern of the canoe, part of a wave struck me, and 
 nearly knocked me overboard. 
 
 We now paddled to some rocks at a distance, got out, 
 unloaded the canoe, and dried the cargo in the sun, which 
 was very hot and powerful. Had it been the wet season, 
 almost everything would have been spoiled. 
 
 After this, the voyage down the Essequibo was quick 
 and pleasant till we reached the sea-coast ; there we had 
 a trying day of it ; the wind was dead against us, and the 
 sun remarkably hot ; we got twice aground upon a mud- 
 flat, and were twice obliged to get out, up to the middle 
 in mud, to shove the canoe through it. Half way betwixt 
 the Essequibo and Demerara the tide of flood caught us ; 
 and after the utmost exertions, it was half-past six in the 
 evening before we got to Georgetown. 
 
 We had been out from six in the morning in an open 
 canoe on the sea-coast, without umbrella or awning, ex- 
 posed all day to the fiery rays of a tropical sun. My face 
 smarted so that I could get no sleep during the night, and 
 the next morning my lips were all in blisters. The Indian 
 Yan went down to the Essequibo a copper colour, but the 
 reflection of the sun from the sea, and from the sand- 
 banks in the river, had turned him nearly black. He 
 laughed at himself, and said that the Indians in the 
 Demerara would not know him again. I stayed one day 
 in Georgetown, and then set off the next morning for 
 
 George- 
 town, 
 
WANDERINGS IN SOUTH AMERICA. 
 
 head-quarters in Mibiri creek, where I fiiiisihed the 
 cavman. 
 
 Here the remaining time was spent in collecting Lird?:, 
 and in paying particular attention to their haunts and 
 economy. The rainy season having set in, the weather 
 became bad and stormy ; the lightning and thunder were 
 incessant : the days cloudy, and the nights cold and misty. 
 I had now been eleven months in the forests, and collected 
 some rare insects, two hundred and thirty birds, two land 
 tortoises, five armadillas, two large serpents, a sloth, an 
 ant-bear, and a cayman. 
 
 I left the wilds arid repaired to Georgetown to spend a 
 few days with Mr. E. Edmonstone previous to embarking 
 for Ei;rope. I must here return my sincerest thanks to 
 this worthy gentleman for his many kiuuiiesses to me ; 
 his friendship was of the utmost service to mo, and ho 
 never failed to send me supplies up into the forest by 
 every opportunity. 
 
 I embarked for England, on board the Dee West-India- 
 man, commanded by Captain Grey. 
 
 Sir Joseph Banks had often told me, he hoped that I 
 would give a lecture in public, on the new mode I had 
 discovered of preparing specimens in natural history for 
 museums. I always declined to do so, as I despaired of 
 ever being able to hit upon a proper method of doing 
 quadrupeds ; and I was aware that it would have been 
 an imperfect lecture to treat of birds only. I imparted 
 what little knowledge I was master of, at Sir Joseph's, to 
 the unfortunate gentlemen who went to Africa to explore 
 the Congo ; and that was all that took place in the shape 
 of a lecture. Now that 1 had hit upon the way of doing 
 quadrupeds, I drew up a little plan on board the Dee, 
 which I trusted would have been of service to naturalists ; 
 and by proving to them the superiority of tlie new plan, 
 
 279 
 
 Til IK I) 
 JOUUNKV, 
 
 Embarka 
 
 for 
 England. 
 
 i! JH 
 
 'ill 
 
280 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 Liverpool. 
 
 THIRD they would probably be induced to abandon the old and 
 jouRNKY. common way, which is a disgrace to the present age, and 
 renders hideous every specimen in every museum that I 
 have as yet visited. I intended to have given three lec- 
 tures ; one on insects and serpents ; one on birds ; and 
 one on quadrupeds. But, as it will be shortly seen, this 
 little plan was doomed not to be unfolded to public view. 
 lUiberality blasted it in tlie bud. 
 
 We liad a pleasant passage across the Atlantic, and 
 arrived in tlie Mersey in fine trim and good spirits. Great 
 was the attention I received from the commander of the 
 Dee. He and his mate, Mr. Spence, took every care of 
 my collection. 
 Arrives at On our landing, the gentlemen of the Liverpool Custom- 
 house received me as an old friend and acquaintance, and 
 obligingly offered their services. 
 
 Twice before had I landed in Liverpool, and twice had 
 I reason to admire their conduct and liberality. They 
 knew I was incapable of trying to introduce anything 
 contraband, and they were aware that I never dreamed of 
 turning to profit the specimens I had procured. They 
 considered that I had left a comfortable home in quest 
 of science ; and that I had wandered into far-distant 
 climes, and gone barefooted, ill-clothed, and ill-fed, 
 through swamps and woods, to procure specimens, some 
 of which had never been seen in Europe. They con- 
 sidered that it would be difficult to fix a price upon speci- 
 mens which had never been bought or sold, and which 
 never were to be, as they Avere intended to ornament 
 my own house. It was hard, they said, to have exposed 
 myself, for years, to danger, and then be obliged to pay 
 on returning to my native land. Under these considera- 
 tions, they fixed a moderate duty, which satisfied all 
 parties. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 281 
 
 It taught THIRD 
 JOURNEY. 
 
 However, this last expedition ended not so. 
 me how hard it is to learn the grand lesson, " requam 
 memento rebus in arduis servare mentem." 
 
 But my good friends in tlie Custom-house of Liverpool 
 were not to blame. On the contrary, they did all in their 
 power to procure balm for me instead of rue. But it 
 would not answer. 
 
 They appointed a very civil officer to attend me to the 
 ship. While we were looking into some of the boxes, to 
 see that the specimens were properly stowed, previous to 
 their being conveyed to the king's depot, another officer 
 entered the cabin. He was an entire stranger to me, and 
 seemed wonderfully aware of his own consequence. With- 
 out preface or apology, he thrust his head over my shoulder, 
 and said, we had no business to have opened a single box 
 without his permission, I answered, they had been opened 
 almost every day since they had come on board, and that 
 I considered there was no harm in doing so. 
 
 He then left the cabin, and I said to myself as he went 
 out, " I suspect I shall see that man again at Philippi." 
 The boxes, ten in number, were conveyed in safety from 
 the ship to the depot. I then proceeded to the Custom- 
 house. The necessary forms were gone through, and 
 a proportionate duty,- according to circumstances, was 
 paid. 
 
 This done, we returned from the Custom-house to the 
 depot, accompanied by several gentlemen who wished to 
 see the collection. They expressed themselves highly 
 gratified. The boxes were closed, and nothing now re- 
 mained but to convey them to the cart, which was in 
 attendance at the door of the depot. Just as one of the 
 inferior officers was carrying a box thither, in stepped the 
 man whom I suspected I should see again at Philippi. 
 He abruptly declared himself dissatisfied with the valua- 
 
282 
 
 THIRD 
 JOUKNKV. 
 
 I 
 
 |l 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 tioii wliich the gentlemen of the Customs had put upon 
 the collection, and said he must detain it. I remonstrated, 
 but it was all in vain. 
 
 After this pitiful stretch of power, and bad compliment 
 to the other officers of the Customs, who had been satisfied 
 with the valuation, this man had the folly to take me 
 aside, and after assuring me that he had a great regard for 
 the arts and sciences, he lamented that conscience obliged 
 him to do what he had done, and he wished he had been 
 fifty miles from Liverpool at the time that it fell to his 
 lot to detain the collection. Had he looked in my face as 
 he said this, he would have seen no marks of credulity 
 there. 
 
 I now returned to the Custom-house, and after ex- 
 pressing my opinion of the officer's conduct at the depot, 
 I pulled a bunch of keys (which belonged to the detained 
 boxes) out of my pocket, laid them on the table, took my 
 leave of the gentlemen present, and soon after set off for 
 Yorkshire. 
 
 I saved nothing from the grasp of the stranger officer 
 but a pair of live Malay fowls, which a gentleman in 
 Georgetown had made me a present of. I had collected 
 in the forest several eggs of curious birds, in hopes of 
 introducing the breed into England, and had taken great 
 pains in doing them over with gum-arabic, and in packing 
 them in charcoal, according to a receipt I had seen in the 
 gazette, from the Edinhurgh Philosophical Journal. But 
 these were detained in the depot, instead of being placed 
 under a hen ; which utterly ruined all my hopes of rearing 
 a new species of birds in England. Titled personages in 
 London interested themselves in behalf of the collection, 
 but all in vain. And vain also were the public and 
 private representations of the first officer of the Liverpool 
 Custom-house in my favour. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 283 
 
 At last tli&re came an order from the Treasury to say, third 
 that any specimens Mr. Waterton intended to present to Jo^^y. 
 public institutions might pass duty free ; but those which 
 he intended to keep for himself n. ast pay the duty ! 
 
 A friend now wrote to me from Liverpool, requesting 
 that I would come over and pay the duty, in order to save 
 the collection, which had just been detained there six 
 weeks. 1 did so. On paying an additional duty, (for the 
 moderate duty first imposed had already been paid,) the 
 man who had detained the collection delivered it up to 
 me, assuring me that it had been well taken care of, and 
 that a fire had been frequently made in the room. It is 
 but justice to add, that on opening the boxes, there was 
 nothing injured. 
 
 I could never get a clue to these harsh and unexpected 
 measures, except that there had been some recent smug- 
 gling discovered in Liverpool ; and that the man in question 
 had been sent down from London to act the part of Argus. 
 If so, I landed in an evil hour ; " nefasto die ; " making 
 good the Spanish proverb, " Pagan a las veces, justos por 
 pecadores ; " at times the innocent suffer for the guilty. 
 After all, a little encouragement, in the shape of exemption 
 from paying the duty on this collection, might have been 
 expected ; but it turned out otherwise ; and after ex- 
 pending large sums in pursuit of natural history, on my 
 return home I was doomed to pay for my success : — 
 
 " Hie finis, Caroli fatorum, hie exitus ilium, 
 Sorte tulit ! " 
 
 Thus, my fleece, already ragged and torn with the thorns 
 and briers, which one must naturally expect to find in 
 distant and untrodden wilds, was shorn, I may say, on its 
 return to England. 
 
 However, this is nothing new : Sancho Panza must have p<»}' 
 heard of similar cases ; for he says, " Muchos van por 
 
 r > '' I 
 
 M /'A 
 
284 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 IF ! 
 
 THIRD lana, y vuelven, trasquiladps ; " many go for wool, and 
 JO URNE Y, come home shorn. In order to pick up matter for natural 
 history, I liave wandered through the wildest parts of 
 South America's equatorial regions. I have attacked and 
 slain a modern Python, and rode on the hack of a cay^nan 
 close to the water's edge ; a very different situation from 
 that of a Hyde-park dandy on his Sunday prancei* before 
 the ladies. Alone and barefoot I have pulled poisonous 
 snakes out of their lurking-places ; climbed up trees to 
 peep into holes for bats and vampires, and for days together 
 liastened through sun and rain to the thickest parts of the 
 forest to procure specimens I had never got before. In 
 fine, I have pursued the wild beasts over hill and dale, 
 through swamps and quagmires, now scorched by the 
 noon-day sun, now drenched by the pelting shower, and 
 returned to the hammock, to satisfy the cravings of 
 hunger, often on a poor and scanty supper. 
 
 These vicissitudes have turned to chestnut hue a once 
 English complexion, and changed the colour of my hair, 
 before father Time had meddled with it. The detention of 
 the collection after it had fairly passed the Customs, and 
 the subsequent order from the Treasury that I should pay 
 duty for the specimens, unless they were presented to 
 some public institution, have cast a damp upon ray energy, 
 and forced, as it were, the cup of Lethe to my lips, by 
 drinking which I have forgot my former intention of 
 giving a lecture in public on preparing specimens to 
 adorn museums. In fine, it is this ungenerous treatment 
 that has paralyzed my plans, and caused me to give up 
 the idea I once had of inserting here the newly-discovered 
 mode of preparing quadrupeds and serpents ; and without 
 it, the account of this last expedition to the wilds of 
 Guiana is nothing but a— fragment. 
 
 Farewell, Gentle Eeader. 
 
 ■; 
 
rOURTn JOURNEY. 
 
 CHArTER I. 
 
 "Nunc hue, nunc illuc et utrinque sine ordino curro." 
 
 Three years in England. — Sail for New York. — Nomenclature, — Altera- 
 tion of scenery. — A sprained ankle. — Magnificent cure. — Feats of 
 climbing. — Quebec. — Irish emigrants. — Ticonderago. — Saratoga. — 
 Philadelphia. — White-headed Eagle. — Form and fashion. — Climate. — 
 Forebodings of the civil war. — Sail for Antigua. 
 
 Courteous reader, when I bade"' thee last farewell, I fourth 
 thought these Wanderings were brought to a final close ; '^" ^^^'' ^' 
 afterwards I often roved in imagination through distant 
 countries famous for natural history, but felt no strong 
 inclination to go thither, as the last adventure had ter- 
 minated in such unexpected vexation. Tlie departure of 
 the Cuckoo and Swallow, and summer birds of passage, 
 for warmer regions, once so interesting to me, now scarcely 
 caused me to turn my face to the south ; and I continued 
 in this cold and dreary climate for three years. During 
 this period, I seldom or ever mounted my hobby-horse ; 
 indeed it may be said, with the old song — 
 
 " The saddle and bridle were laid on the shelf," 
 
 and only taken down once, on the night that I was 
 induced to give a lecture in the philosophical hall of 
 
286 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 Formit Leeds. A little after this, Wilson's Ornithology of the 
 jouuNKv. United States ^q\\ mio my \yAni\&. 
 
 Sails for The dcsiic I had of seeiiif' that country, together with 
 
 ew none. |.|jQ animated description which Wilson had given of the 
 birds, fanned up the almost expiring flame. I forgot the 
 vexations already alluded to, and set off for New York, in 
 the beautiful packet John Wells, commanded by Captain 
 Harris. The passage was long and cold ; but the elegant 
 accommodations on board, and the polite .attention of the 
 commander, rendered it very agreeable ; and I landed, in 
 health and merriment, in the stately capital of the new 
 world. 
 
 We will soon pen down a few remarks on this magnifi- 
 cent city, but not just now. I want to venture into the 
 north-west country, and get to their great canal, which the 
 world talks so much about, though I fear it will be hard 
 work to make one's way through bugs, bears, brutes, and 
 buffaloes, which we Europeans imagine are so frequent 
 and ferocious in these never-ending western wilds. 
 
 Lf.av€n I l^^t N^w York on a fine morning in July, without 
 i\cM> York one letter of introduction, for the city of Albany, some 
 
 "fOT 
 
 Albany, hundred and eighty miles up the celebrated Hudson. I 
 seldom care about letters of introduction, for I am one of 
 those who depend much upon an accidental acquaintance. 
 Full many a face do I see, as I go wandering up and down 
 the world, whose mild eye, and sweet and placid features, 
 seem to beckon to me, and say, as it were, " Speak but 
 civilly to me, and I will do what I can for you." Such a 
 face as this is worth more than a dozen letters of intro- 
 duction ; and such a face, gentle reader, I found on board 
 the steam-boat from New York to the city of Albany. 
 
 There was a great number of well-dressed ladies and 
 gentlemen in the vessel, all entire strangers to me. I 
 fancied I could see several whose countenances invited an 
 
WANDERINGS IN SOUTH AMKUICA. 
 
 287 
 
 unknown wanderer to come and take a seat beside them ; fouutii 
 but there was one who encouraged me more than the rest. '^"""^'''^' 
 I saw clearly that he was an American, and I judged, by 
 his manners and appearance, that he had not spent ^U his 
 time upon his native soil. I was right in tliis conjecture, 
 for he afterwards told me that he had been in France and 
 England. I saluted him as one stranger gentleman ought 
 to salute another when he wants a little information ; and 
 soon after, I dropped in a word or two by which he might 
 conjecture that I was a foreigner; but^I did not tell him 
 so ; I wished him to make the discovery himself. 
 
 He entered into conversation with the openness and 
 candour which is so remarkable in the American ; and in 
 a little time observed that he presumed 1 was from the 
 old country. I told him that I was, and added, that I was 
 an entire stranger on board. I saw his eye brighten up at 
 the prospect he had of doing a fellow-creature a kind turn 
 or two, and he completely won my regard by an affability 
 which I shall never forget. This obliging gentleman 
 pointed out everything that was grand and interesting as 
 the steam-boat plied her course up the majestic Hudson, 
 Here the Catskill mountains raised their lofty summit; 
 and there the hills came sloping down to the water's edge. 
 Here he pointed to an aged and venerable oak, which 
 having escaped the levelling axe of man, seemed almost 
 to defy the blasting storm and desolating hand of time ; 
 and there, he bade me observe an extended tract of wood, 
 by which I might form an idea how rich and grand the 
 face of the country had once been. Here it was that, in 
 the great and momentous struggle, the colonists lost the 
 day ; and there, they carried all before them : — 
 
 " They closed full fast, on every side 
 No slackness there was found ; 
 And many a gallant gentleman 
 liay gasping on the ground," 
 
 ; 
 
2!^^ 
 
 WANl»KKIN'(JS IN' Sol'TII AMKUICA. 
 
 II I 
 
 The, gnat 
 canal. 
 
 roiTUTH Here, in fine, stood a notod ie;^'iinent ; there, moved their 
 jouHNKY. njreat captain ; here, the fhfots fired tlieir broadsides ; and 
 there, the whole force rushed on to battle : — 
 
 "Hie Dolnpum in.'inus, hie maj^ims teiiiloliiit Acliillcs, 
 ('la8Htbii.s hit! loeiiH, liic iieies certiiru solcbat." 
 
 At tea-time we took our tea together, and tlie next 
 morning tliis worthy American walked up with me to tiio 
 inn in Albany, shook me by the hand, and then went his 
 way. I bade him farewell, and again farewell, and hoped that 
 fortune might bring us together again once more. I'ossibly 
 she may yet do so ; and should it be in England, I will 
 take him to my house, as an old friend and acquaintance, 
 and offer him my choicest cheer. 
 
 It is at Albany that the great canal opens into the 
 Hudson, and joins the waters of this river to those of 
 Lake Erie. The Hudson, at the city of Albany, is distant 
 from Lake Erie about three luindred and sixty miles. The 
 level of the lake is five hundred and sixty-four feet higher 
 than the Hudson, and there ai'e eighty-one locks on the 
 canal. It is to the genius and perseverance of De Witt 
 Clinton that the United States owe the almost incalculable 
 advantages of this inland navigation. " Exegit monumen- 
 tum sere perennius." You may cither go along it all the 
 way to Buffalo, on Lake Erie, or by the stage;. or some- 
 times on one and then in the other, just as you think fit. 
 Scenery, Grand, indeed, is the scenery by either route, and capital 
 the accommodations. Cold and phlegmatic must he be 
 who is not warmed into admiration by the surrounding 
 scenery, and charmed with the affability of the travellers he 
 meets on the way. 
 
 This is now tiie season of roving, and joy and merriment 
 for the gentry of this happy country. Thousands are on 
 the move from different parts of the Union for the springs 
 
WANDEUINOS IN SOUTH AMERICA. 
 
 989 
 
 he 
 
 nt 
 Ion 
 
 gs 
 
 und lakes, and tlie falls of Niagara. There is nothing Fouunr 
 hauyhfy or forbidding in the Americans; and wherever "'"^"^'^'• 
 you meet tliem, they appear to be quite at liome. This is 
 exactly what it ought to ])e, and very much in favour of 
 the foreigner wlio journeys amongst them. The immense 
 number of highly polished females who go in the stages to 
 visit the different places of amusement, and see the stupen- 
 dous natural curiosities of this extensive country, incon- 
 testably proves that safety and convenience are ensured 
 to them, and that the most distant attempt at rudeness 
 would, by common consent, be immediately put down. 
 
 liy the time I had got to Schenectady, I began strongly 
 to suspect that I had come into the wrong country to look 
 for bugs, bears, brutes, and buffaloes. It is an enchanting 
 journey from Albany to Schenectady, and from thence to 
 Lake Erie. The situation of the city of Utica is particu- 
 larly attractive; the Mohawk running close by it, the 
 fertile fields and woody mountains, and the falls of Tren- 
 ton, forcibly press the stranger to stop a day or two hero 
 before he proceeds onward to the lake. 
 
 At some far-distant period, when it will not be possible 
 to find the place where many of the celebrated cities of the 
 East once stood, the world will have to thank the United 
 States of America for bringing their names into the western 
 regions. It is, indeed, a pretty thought of these people to 
 give to their rising towns the names of places so famous 
 and conspicuous in former times. 
 
 As I was sitting one evening under an oak, in the high 
 grounds behind Utica, I could not look down upon the 
 city without thinking of Cato and his misfortunes. Had 
 the town been called Crofton, or Warmfield, or Lews- 
 bury, there would have been nothing remarkable in 
 it ; but Utica at once revived the scenes at school long 
 past and half forgotten, and carried me with full speed 
 
 u 
 
 ■i 
 
 uK- i 
 
 \'4\ 
 
290 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 Nii 
 'III 
 
 I I 
 
 FOURTH back again to Ital}', and from thence to Africa. I crossed 
 JO URNE Y, ^i^g Rubicon with Caesar ; fought at Pharsalia ; saw poor 
 Pompey into Larissa, and tried to wrest the fatal sword 
 from Cato's hand in Utica. When I perceived he was no 
 more, I mourned over the noble-minded man who took 
 that part which he thought would most benefit his coun- 
 try. There is something magnificent in the idea of a man 
 taking by choice the conquered side. The Roman gods 
 themselves did otherwise. 
 
 " Victrix causa Diis ])lacuit, sed vida Catoni." 
 
 " In this (lid Cato with the Gods divide, 
 They chose the conquering, he the conquer'd side." 
 
 Faceoftlic The whole of the country from Utica to Buffalo is 
 countnj. pleasing ; and the intervening of the inland lakes, large 
 and deep and clear, adds considerably to the effect. The 
 spacious size of the inns, their excellent provisions, and 
 the attention which the traveller receives in going from 
 Albany to Buffalo, must at once convince him that this 
 country is very much visited by strangers ; and he will 
 draw the conclusion that there must be something in it 
 uncommonly interesting to cause so many traveller? to 
 pass to and fro. 
 
 Nature is losing fast her ancient garb, and putting on a 
 new dress in these extensive regions. Most of the stately 
 timber has been carried away ; thousands of trees are 
 lying prostrate on the ground ; while meadows, corn-fields, 
 villages, and pastures are ever and anon bursting upon the 
 traveller's view as he journeys on through the remaining 
 tracts of wood. I wish I could say a word or two for the 
 fine timber which is yet standing. Spare it, gentle inha- 
 bitants, for your country's sake ; these noble sons of the 
 forest beautify your landscapes beyond all description; 
 when they are gone, a century will not replace their loss 
 
WANDERINGS IN SOUTH AMERICA. 
 
 291 
 
 ields, 
 the 
 fining 
 the 
 linha- 
 k the 
 )tiou ; 
 loss 
 
 they cannot, they must not fall ; their vernal bloom, their fourth 
 summer richness and autumnal tints, please and refresh •'ouiinbt. 
 the eye of man ; and even when the days of joy and 
 warmth are fled, the wintry blast soothes the listening ear 
 with a sublime and pleasing melancholy as it howls through 
 their naked branches. 
 
 *' Around mc trees unniimber'd rise, 
 Reautiful in various dyes : 
 The gloomy pine, the poplar blue, 
 The yellow beech, the sable yew ; 
 The slender fir, that taper gi-ows, 
 The sturdy oak, with broad-spread boughs." 
 
 A few miles before you reach Buffalo, the road is low 
 and bad, and in stepping out of the stage I sprained my 
 foot very severely ; it swelled to a great size, and caused 
 me many a day of pain and mortification, as will be seen 
 in the sequel. 
 
 Buffalo looks down on Lake Erie, and possesses a fine Buffalo. 
 and commodious inn. At a little distance is the Black 
 Eock, and there you pass over to the Canada side. A 
 stage is in waiting to convey you some sixteen or twenty 
 miles down to the falls. Long before you reach the spot 
 you hear the mighty roar of waters, and see the spray of 
 the far-famed falls of Niagara, rising up like a column to 
 the heavens, and mingling with the passing clouds. 
 
 At this stupendous cascade of nature, the waters of the Falls of 
 lake fall one hundred and seventy-six feet perpendicular. ^^^'J^^^^- 
 It has been calculated, I forget by whom, that the quantity 
 of water discharged down this mighty fall, is six hundred 
 and seventy thousand two hundred and fifty-five tons per 
 minute. There are two large inns on the Canada side ; 
 but, after you have satisfied your curiosity in viewing the 
 falls, and in seeing the rainbow in the foam far below 
 where you are standing, do not, I pray yon, tarry long at 
 
 u 2 
 
 !''' ll 
 
 <*^:: 
 
 .it- 
 
292 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FOUKTII 
 JOiniNKY. 
 
 I ! 
 
 American 
 ladies. 
 
 either of them. Cross over to the American side, and 
 there you will find a spacious inn, which has nearly all the 
 attractions: there you meet with great attention, and 
 every accommodation. 
 
 Tiie day is passed in looking at the falls, and in saun- 
 tering up and down the wooded and rocky environs of the 
 Niagara ; and the evening is often enlivened by the merry 
 dance. 
 
 Words can hardly do justice to the unaffected ease and 
 elegance of the American ladies who visit the falls of 
 Niagara. The traveller need not rove in imagination 
 through Circassia in search of fine forms, or through 
 England, France, and Spain, to meet with 'polished 
 females. The numbers who are continually arriving here 
 from all parts of the Union confirm the justness of this 
 remark. 
 
 I was looking one evening at a dance, being unable to 
 join in it on account of the accident I had received near 
 Buffalo, when a young American entered the ball-room 
 with such a becoming air and grace, that it was impossible 
 not to have been struck with her appearance. 
 
 " Her bloom was like the springing llower 
 
 That sips the silver dew, 
 ' The rose was budded in her cheek, 
 
 Just opening to the view." 
 
 I could nut help feeling a wish to know where she had 
 
 " Into such beaut J' spread, and blown so fair." 
 
 Upon incpiiry, I found tliat she was from the city of 
 Albany. The more I looked at the fair Albanese, the 
 more I was convinced, that in the United States of America 
 may be found grace and beauty and symmetry equal to 
 iiiiything in the old world. 
 
of 
 
 the 
 
 srica 
 
 ,1 to 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 293 
 
 I now for good and all (and well I might) gave up the fourth 
 idea of finding bugs, bears, brutes, and buffaloes in this '^Q ^"^^ ^"- 
 country, and was thoroughly satisfied that I had laboured 
 under a great mistake in suspecting that I should ever 
 meet with them. 
 
 I wished to join in the dance where the fair Albanese 
 was " to brisk notes in cadence beating," but the state of 
 my unlucky foot rendered it impossible ; and as I sat with it 
 reclined upon a sofa, full many a passing gentleman stopped 
 to inquire the cause of my misfortune, presuming at the 
 same time that I had got an attack of gout. Now this 
 surmise of theirs always mortified me; for I never had a 
 fit of gout in my life, and moreover, never expect to have 
 one. 
 
 In many of the inns in the United States, there is an 
 album on the table, in which travellers insert their arrival 
 and departure, and now and then indulge. in a little flash 
 or two of wit. 
 
 I thought, under existing circumstances, that there 
 would be no harm in briefly telling my misadventure ; and 
 so, taking up the pen, I wrote what follows ; and was never 
 after asked a single question about the gout. 
 
 " C. Waterton, of Walton-Hall, in tlie county of York, 
 England, arrived at the Falls of Niagara in July, 1824, 
 and begs leave to pen down the following dreadful acci- 
 dent : — 
 
 " He sprained his foot, and hurt his toe, 
 On tlie rough road near Buffalo. 
 It quite distresses him to stagger a- 
 Long the sharp rocks of famed Niagara. 
 So thus he's doomed to drink the measure 
 Of pain, in lieu of that of pleasure. 
 On Hope's delusive pinions borne, 
 He came for wool, and goes back shorn, 
 N.B. — Here he alludes to nothing but 
 Th' adventure of his toe and foot ; 
 
 
 ■J, it 
 
 
 
 <ll tli I 
 
 m 
 
294 
 
 Four.Tir 
 
 JOURXEY. 
 
 H 
 
 i H 
 
 I I! 
 
 ' I 
 
 t 'i 
 
 WANDEUINGS IN SOUTH AMERICA. 
 
 Save this —lie sees all tliat which can 
 Delight and charm the sonl of man, 
 But feels it not, — because his toe 
 And foot together plague him so." 
 
 I remember once to have sprained my ankle very 
 violently, many years ago, and that the doctor ordered me 
 to hold it under the pump two or three times a day. 
 Now, in the United States of America, all is upon a 
 grand scale, except taxation ; and I am convinced that 
 the traveller's ideas become much more enlarged as he 
 journeys through tlie country. This being the case, I can 
 easily account for the desire I felt to hold my sprained 
 foot under the fall of Niagara. I descended the winding 
 staircase which has been made for the accommodation of 
 travellers, and then hobbled on to the scene of action. 
 As I held my leg under the fall, I tried to meditate on the 
 immense difference there was betwixt a house-pump and 
 this tremendous cascade of nature, and what effect it 
 might have upon the sprain.; jut tlie magnitude of the 
 subject was too overwhelming, and I was obliged to 
 drop it. 
 
 Perhaps, indeed, there was an unwarrantable tincture of 
 vanity in an unknown wanderer wishing to have it in his 
 power to tell the world, that he had held his sprained foot 
 under a fall of water which discharges six hundred and 
 seventy thousand two hundred and fifty-five tons per 
 minute. A gentle purling stream would have suited 
 better. Now, it would have become Washington to have 
 quenched his battle-thirst in the fall of Niagara ; and 
 there was something royal in the idea of Cleopatra drink- 
 ing pearl-vinegar, made from the grandest pearl in Egypt ; 
 and it became Caius Marius to send word, that he was 
 sitting upon the ruins of Carthage. Here, we have the 
 person suited to the thing, and the thing to the person. 
 
WANDERINGS IN SOUTH AMEPJCA. 
 
 295 
 
 If, gentle reader, thou wouldst allow me to indulge fouiitii 
 a little longer in this harmless pen-errantry, I would tell Jo ^^^^ '"'- 
 thee, that I have had my ups and downs in life, as well 
 as other people ; for I have climbed to the point of the 
 conductor above the cross on the top of St. Peter's, in 
 Kome, and left my glove there. I have stood on one foot, 
 upon the Guardian Angel's head, on the castle of St. 
 Angelo ; and, as I have just told thee, I have been low 
 down under the fall of Niagara. But this is neither here 
 nor there ; let us proceed to something else. 
 
 When the pain of my foot had become less violent, and 
 the swelling somewhat abated, I could not resist the 
 inclination I felt to go down Ontario, and so on to 
 Montreal and Quebec, and take Lakes Champlain and 
 George in my way back to Albany. 
 
 Just as I had made up my mind to it, a family from 
 the Bowling-green, in New York, who was going the same 
 route, politely invited me to join their party. Nothing 
 could be more fortunate. They were highly accomplished. 
 The young ladies sang delightfully ; and all contributed 
 their portion, to render the tour pleasant and amusing. 
 
 Travellers had already filled the world with descriptions 
 
 of the bold and sublime scenery from Lake Erie to 
 
 Quebec : — 
 
 " The fountain's fall, the river's flow, 
 The woody valleys, warm and low ; 
 The windy summit, wild and high, 
 Roughly rushing to the sky." 
 
 And there is scarce one of them who has not described 
 the achievements of former and latter times, on the 
 different battle-grounds. Here, great Wolfe expired. 
 Brave Montcalm was carried, mortally wounded, through 
 yonder gate. Here fell the gallant Brock; and there 
 General Sheaffee captured all the invaders. And in yonder 
 
 'Ml: 
 
296 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 .. *i 
 
 I 
 
 FOURTH harbour may be seen the mouldering remnants of British 
 JOURNEY, vessels. Their hour of misfortune has long passed away. 
 The victors have now no use for them in an inland lake. 
 Some have already sunk, while others, dismantled and 
 half-dismasted, are just above the water, waiting, in 
 shattered state, that destiny which must sooner or later 
 destroy the fairest works of man. 
 
 The excellence and despatch of the steam-boats, together 
 with the company which the traveller is sure to meet with 
 at this time of the year, render the trip down to Montreal 
 and Quebec very agreeable. 
 
 The Canadians are a quiet, and apparently a happy 
 people. They are very courteous and affable to strangers. 
 On comparing them with the character which a certain 
 female traveller, a journalist, has thought fit to give them, 
 the stranger might have great doubts whether or not he 
 were amongst the Canadians. 
 
 Montreal, Quebec, and the falls of Montmorency, are 
 well worth going to see. They are making tremendous 
 fortifications at Quebec. It will be the Gibraltar of the 
 new world. When one considers its distance from Europe, 
 and takes a view of its powerful and enterprising neigh- 
 bour, Virgil's remark at once rushes into the mind. 
 
 The Cana 
 dimis. 
 
 Fortificn- 
 lions at 
 Quebec. 
 
 "Sic vos non vobis nidificatis aves," 
 
 i'l 
 
 I left Montreal with regret. I had the good fortune to 
 be introduced to the Professors of the College. These 
 fathers are a very learned and worthy set of gentlemen ; 
 and on my taking leave of them, I felt a heaviness at 
 heart, in reflecting that I had not more time to cultivate 
 their acquaintance. 
 
 In all the way from Buffalo to Quebec, I only met with 
 one bug ; and I cannot even swear that it l)elonged to the 
 United States. In going down the St. Lawrence, in the 
 
 1 
 
WANDERINGS IN SOUTH AMERICA. 
 
 297 
 
 !Mi!. 
 
 in 
 
 igh- 
 
 
 steam-boat, I felt something crossing over my neck ; and fourth 
 on laying hold of it with my finger and thumb, it turned Jo u"^^ ^^ 
 out to be a little half-grown, ill-conditioned bug. Now, 
 whether it were going from the American to the Canada 
 side, or from the Canada to the American, and had taken 
 the advantage of my shoulders to ferry itself across, I 
 could not tell. Be this as it may, I thought of my uncle 
 Toby and the fly ; and so, in lieu of placing it upon the 
 deck, and then putting my thumb-nail vertically upon it, 
 I quietly chucked it amongst some baggage that was 
 close by, and recommended it to get ashore by the first 
 opportunity. 
 
 When we had seen all that was worth seeing in Quebec 
 and at the falls of Montmorency, and had been on board 
 the enormous ship Columhus, we returned for a day or two 
 to Montreal, and then proceeded to Saratoga by Lakes 
 Cham^lain and George. 
 
 The steam-boat from Quebec to Montreal had above five 
 hundred Irish emigrants on board. They were going 
 " they hardly knew whither," far away from dear Ireland. 
 It made one's heart ache to see them aU huddled together, 
 without any expectation of ever revisiting their native soil. 
 We feared that the sorrow of leaving home for ever, the 
 miso.rable accommodation on board the ship which had 
 brought them away, and the tossing of the angry ocean, in 
 a long and dreary voyage, would have rendered them 
 callous to good behaviour. But it was quite otherwise. 
 They conducted themselves with great propriety. Every 
 American on board seemed to feel for them. And then 
 " they were so full of wretchedness. Need and oppression 
 starved in their eyes. Upon their backs hung ragged 
 misery. The world was not their friend." Poor dear Ire- 
 land, exclaimed an aged female, as I was talking to her, I 
 shall never see it any more ! and then her tears began to 
 
 Irish 
 emigrants. 
 
 (llUi 
 
298 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 1 ' 
 
 ? I 
 
 l\ i 
 
 t I 
 
 II ... 
 
 FouKTH flow. Probably the scenery on the banks of the St. 
 ju uRNE Y. Lawrence recalled to her mind the remembrance of spots 
 once interesting to her : 
 
 " The lovely daughter, — lovelier in her tears, 
 The fond companion of her father's years, 
 Here silent stood, — neglectful of her charms, 
 And left her lover's for her father's arms. 
 With louder plaints the mother spoke her v oes, 
 And blessed the cot where every pleasure rose ; 
 And pressed her thoughtless babes with many a tear, 
 And clasped them close, in son-ow doubly dear, 
 While the fond husband strove to lend relief, 
 In all the silent manliness of grief." 
 
 We went a few miles out of our route to take a look at 
 the once formidable fortress of Ticonderoga. It has long 
 been in ruins, and seems as if it were doomed to moulder 
 quite away. 
 
 •* Ever and anon there falls 
 Hugo heaps of hoary moulder'd walls. 
 But time has seen, that lifts the low 
 And level lays the lofty brow, 
 Has seen this ruin'd pile complete. 
 Big with the vanity of state ; 
 But transient is the smile of fate." 
 
 The scenery of Lake George is superb; the inn re- 
 markably spacious and well attended ; and the conveyance 
 Saratoga, from thencB to Saragota very good. He must be sorely 
 afflicted with spleen and jaundice, who, on his arrival at 
 Saratoga, remarks, there is nothing here worth coming to 
 see. It is a gay and fashionable place ; has four uncom- 
 monly fine hotels ; its waters, for medicinal virtues, are 
 surpassed by none in the known world ; and it is resorted 
 to, throughout the whole of the summer, by foreigners 
 and natives of the first consideration. Saratoga pleased 
 me much; and afforded a fair opportunity of forming a 
 pretty correct idea of the gentry of the United States. 
 
WANDERINGS IN SOUTH AMERICA. 
 
 299 
 
 Mi^ 
 
 There is a pleasing frankness, and ease and becoming Fouu\n 
 dignity, in the American ladies ; and the good humour, ^" ^"^" ^^ 
 and absence of all haughtiness and puppyism in the 
 gentlemen, must, no doubt, impress the traveller with 
 elevated notions of the company who visit this famous 
 spa. 
 
 During my stay here, all was joy, and affability, and 
 mirth. In the mornings the ladies played and sang 
 for us ; and the evenings were generally enlivened with 
 the merry dance. Here I bade farewell to the charm- 
 ing family, in whose company I had passed so many 
 happy days, and proceeded to Albany. 
 
 The stage stopped a little while in the town of Troy. Troy. 
 The name alone was quite sufficient to recall to the mind 
 scenes long past and gone. Poor king Priam ! Napoleon's 
 sorrows, sad and piercing as they were, did not come 
 up to those of this ill-fated monarch. The Greeks first 
 set his town on fire, and then began to bully : — 
 
 " Incensd Danai dominantnr in urbe." 
 
 One of his sons was slain before his face ; " ante ora 
 parentum, concidit." Another was crushed to mummy 
 by boa-constrictors; "immensis orbibus angues." His 
 city was rased to the ground, "jacet Ilion ingens." And 
 Pyrrhus ran him through with his sword, " capulo tenus 
 abdidit ensem." This last may be considered as a for- 
 tunate stroke for the poor old king. Had his life been 
 spared at this juncture he could not have lived long. 
 He must have died broken-hearted. He would have seen 
 his son-in-law, once master of a noble stud, now, for want 
 of a horse, obliged to carry off his father, up hill, on 
 his own back, " cessi et sublato, montem genitore petivi." 
 He would have heard of his grandson being thrown 
 neck and heels from a high tower, " mittitur Astyanax 
 
 l:'' 
 
 {;!;• \ 
 
800 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 I r 
 
 : I 
 
 J!' 
 I' 
 
 ■I r 
 
 I 'll 
 
 FOURTH illis tie turribus." He would have been informed of his 
 jo uKNR Y. ^i^'g tearing out the eyes of king Odrysius with her finger 
 nails, " digitos in perfida lumina condit." Soon after 
 this, losing all appearance of woman, she became a bitch, 
 
 " Pcrdidit infelix, hominis post omnia formnm," 
 
 and rent the heavens with her bowlings, 
 
 Exteriiasquo novo latratu terruit auras." 
 
 Then, becoming distracted with the remembrance of her 
 misfortunes, " veterum memor ilia malorum," she took off 
 howling into the fields of Thrace, — 
 
 '* Turn quofiue Sithonioj, riUilavit mcesta per agros." 
 
 Juno, Jove's wife and sister, was heard to declare, that 
 poor Hecuba did not deserve so terrible a fate, — 
 
 •* Ipsa Jovis conjuxquo soroniuc, 
 Eventus Hecubam memisse ncgaverit illos," 
 
 Had poor I'riaip escaped from Troy, one thing, and only 
 one thing, would have given him a small ray of satisfac- 
 tion, viz., he would have heard of one of his daughters 
 nobly preferring to leave this world, rather than live to 
 become servant-maid to old Grecian ladies : — 
 
 *' Non ego Myrmidonlim scdes, Dolopumve superbas, 
 Adspiciam, aut Graiis servitura matribus ibo. " 
 
 .^ 
 
 At some future period, should a foreign armed force, or in- 
 testine broils, (all which heaven avert,) raise Troy to the 
 dignity of a fortified city, Virgil's prophecy may then be 
 fulfiUed, 
 
 " Atque iteram ad Trojam magniis mittetuv Achilles." 
 
 After leaving Troy, I passed through a fine country to 
 Albany ; and then proceeded by steam down the Hudson 
 to New York. 
 
 
WANDERINGS IN SOUTH AMEI '<J1|^ 
 
 301 
 
 Its 
 
 Museum. 
 
 Travellers hesitate whether to give the pn'f'ret »? tc fourth 
 Philadelphia or to New York. Philadelphia is certainly a J^ ^jbhe y. 
 noble city, and its environs beautiful ; but there is a degree miadeU 
 of quiet and sedateness in it, which, though no doubt very 
 agreeable to the man of calm and domestic habits, is not so 
 attractive to one of speedy movements. The quantity of 
 white marble which is used in the buildings, gives to 
 Philadelphia a gay and lively appearance ; but the same- 
 ness of the streets, and their crossing each other at right 
 angles, are somewhat tiresome. The water-works which 
 supply the city, are a proud monument of the skill and 
 enterprise of its inhabitants ; and the market is well worth 
 the attention of the stranger. 
 
 When you go to Philadelphia, be sure not to forget to 
 visit the Museum. It will afford you a great treat. Some 
 of Mr. Peale's family are constantly in it, and are ever 
 ready to show the curiosities to strangers, and to give them 
 every necessary information. Mr. Peale has now passed 
 his eightieth year, and appears to possess the vivacity, and, 
 I may almost add, the activity of youth. 
 
 To the indefatigable exertions of this gentleman is the 
 western world indebted for the possession of this splendid 
 museum. Mr. Peale is, moreover, an excellent artist. 
 Look attentively, I pray you, at the portrait he has taken 
 • of himself, by desire of the State of Pennsylvania. On 
 entering the room he appears in the act of holding up a 
 curtain, to show you his curiosities. The effect of the 
 light upon his head is infinitely striking. I have never 
 seen anything finer in the way of light and shade. The 
 skeleton of the mammoth is a national treasure. I could 
 form but a faint idea of it by description, until I had seen 
 it. It is the most magnificent skeleton in the world. The 
 city ought never to forget the great expense Mr. Peale was 
 put to, and the skill and energy he showed, during the 
 
 m 
 
 /'■ 
 
 fi 
 
 :\ 
 
 t 
 
 m 
 
 I . 
 
 .■ ii 
 
 
 
^ 
 
 
 III 
 
 ,;l| 
 »t'ii 
 
 I C: 
 
 302 
 
 WANDEUING8 IN SOUTH AMERICA. 
 
 .Ill: 
 
 
 ;'l ' 
 
 jl 
 
 White- 
 lieaded 
 Eagles. 
 
 FOURTH many mouths ho spent in searching the swamps, where 
 jouRNKY. tinjgo enormous bones had been concealed from the eyes of 
 the world for centuries. 
 
 Tlie extensive squares of this city arc ornamented witli 
 well-grown and luxuriant trees. Its unremitting attention 
 American to literature miglit cause it to bo styled the Athens of tlie 
 United States. Here, learning and science have taken up 
 their abode. The literary and philosophical associations, 
 the enthusiasm of individuals, the activity of the press, 
 and the cheapness of the publications, ought to raise the 
 name of Philadelphia to an elevated situation in the temple 
 of knowledge. 
 
 From the press of this city came Wilson's famous 
 "Ornithology." IJy observing tlio birds in their native 
 haunts, he has been enabled to purge their history of 
 numberless absurdities, which inexperienced theorists had 
 introduced into it. It is a pleasing and a brilliant work. 
 We have no description of birds in any European publica- 
 tion that can come up to this. By perusing Wilson's 
 " Ornithology " attentively before I left England, I knew 
 where to look for the birds, and immediately reco^n:«^ed 
 tliem in their native laud. 
 
 Since his time, I fear the White-headed Eagles have 
 been much thinned. I was perpetually looking out for 
 them, but saw very few. One or two came now and then, 
 and soared in lofty flight over the falls of Niagara. The 
 Americans are proud of this bird in efiigy, and their hearts 
 rejoice when its banner is unfurled. Could they not then 
 be persuaded to protect the white-headed eagle, and allow 
 it to glide in safety over its own native forests ? Were I 
 an American, I should think I had committed a kind of 
 sacrilege in killing the white-headed eagle. The Ibis was 
 held sacred by the Egyptians ; the Hollanders protect the 
 Stork ; the Vulture sits unmolested on the top of the 
 
WANDEUINOS IN SOUTH AMEUICA. 
 
 303 
 
 liouscs in the city of Aiigustura ; and Uobin-rcd-brcast, kouuth 
 loi liis charity, is cherished by the English : — jq uunk v. 
 
 " No liuriiil these pretty babos 
 Of any mnn recoivcH, 
 Till robin-ret' .breast [lainfuUy 
 Dill cover thorn with Uiavf's." ' 
 
 have 
 >ut for 
 then, 
 The 
 learts 
 ; then 
 allow 
 iVere I 
 ind of 
 is was 
 ct the 
 )f the 
 
 
 Poor Wilson was smote by the hand of death, before ho 
 had finished his work. I'rinco Charles liuonaparte, nephew 
 to the late emperor Napoleon, aided by some of thj most 
 scientific gentlemen of Pennsylvania, is continuing this 
 valuable and interesting puldication. 
 
 Now York, with great propriety, may be called the com- New York. 
 mereial capital of the now world ; — 
 
 •' Urb« angnstft potons, nulli cessura." 
 
 Kre long, it will be on the coast of North America what 
 Tyre once was on that of Syria, In her port are the ships 
 of all nations ; and in licr streets is displayed merchandise 
 from all parts of the known world. And then the approach 
 to it is so enchanting ! The verdant fields, the woody hills, 
 the farms, and country houses, form a beautiful landscape 
 as you sail up to tlie city of New York. 
 
 Broadway is the principal street. It is three miles and The streets, 
 a half long. I am at a loss to know where to look for a 
 street, in any part of the world, which has so many attrac- 
 tions as this. There are no steam-engines to annoy you 
 by filling the atmosphere full of soot and smoke; the 
 houses have a stately appearance ; while the eye is relieved 
 from the perpetual sameness, which is common in most 
 streets, by lofty and luxuriant trees. 
 
 Nothing can surpass the appearance of the American a 
 ladies, when they take their morning walk, from twelve to 
 
 ' The fault against grammar is lost in the beauty of the idea. 
 
 houses, «tr. 
 
 mericnn 
 ladies. 
 
 
 Li 
 
 
 I'll 
 
 I- : ! 
 
 I .1- 
 
 . 'li , i 
 
 IV 
 
 I' t 
 
304 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 FOURTH 
 
 5: ^11 1-' 
 
 r ■^^\'. 
 
 '" .ill 111 
 
 I ■•!■- 
 
 it ! 
 
 three, in Broadway. The stranger will at once see that 
 JO URNE Y, ^j^gy \^Q;yQ rejected the extravagant superfluities which ap- 
 pear in the London and Parisian fashions ; and have only 
 retained as much of those costumes, as is becoming to the 
 female form. This, joined to their own just notions of 
 dress, is what renders the New York ladies so elegant in 
 their attire. The way they wear the Leghorn hat deserves 
 a remark or two. With us, the formal hand of the mil- 
 liner binds down the brim to one fixed shape, and that 
 none of the handsomest. The wearer is obliged to turn her 
 head full ninety degrees before she can see the person who 
 is standing by her side. But in New York the ladies have 
 the brim of the hat not fettered with wire, or tape, or 
 ribbon, but quite free and undulating ; and by applying 
 the hand to it, they can conceal or expose as much of the 
 face as circumstances require. This hiding and exposing 
 of the face, by the bye, is certainly a dangerous movement, 
 and often fatal to the passing swain. I am convinced in 
 my own mind, that many a determined and unsuspecting 
 bachelor, has been shot down by this sudden manoeuvre, 
 before he was aware that he was within reach of the 
 battery. 
 
 The American ladies seem to have an abhorrence (and a 
 very just one too) of wearing caps. When one considers 
 for a moment, that women wear the hair long, which nature 
 has given them both for an ornament and to keep the head 
 warm, one is apt to wonder, by what perversion of good 
 taste they can be induced to enclose it in a cap. A mob 
 cap, a lace cap, a low cap, a high cap, a flat cap, a cap with 
 ribbons dangling loose, a cap with ribbons tied under the 
 chin, a peak cap, an angular cap, a round cap, and a 
 pyramid cap ! How would Canova's Venus look in a mob 
 cap ? If there be any ornament to the head in wearing 
 a cap, it must surely be a false ornament. The American 
 
\VAXI)KIIIX(;S I\ SOITII AMKIIK'A. 
 
 H05 
 
 A. mob 
 with 
 icr the 
 and a 
 a mob 
 earinj:? 
 ericaii 
 
 ladies are persuaded that the head can be ornamented fouhth 
 without a cap. A rose-bud or two, a woodbine, or a si)riff •''»^[^''>'- 
 of eglantine, look well in the braided hair ; and it' there 
 be raven locks, a lily or a snowdrop may be interwoven 
 with effect. 
 
 Now tliat the packets are so safe, and make sucli quick 
 passages to the United States, it would be as well if some 
 of our head milliners would go on board of them, in lieu 
 of getting into the Diligence for Paris. They would bring 
 back more taste, and less caricature. And if they could 
 persuade a dozen or two of the farmers' servant girls to 
 return with them, w'e should soon liave proof positive, 
 that as good butter and cheese may be made with the hair 
 braided up, and a daisy or prinn'ose in it, as butter and 
 cheese made in a cap of barbarous shape ; washed, perliaps, 
 in soap-suds last new moon. 
 
 New York has very good hotels, and genteel boarding- Hotels and 
 ] louses. All charges included, you do not pay above two j^^^^^f' 
 dollars a day. Little enough, when you consider the 
 capital accommodations, and the abundance of food. 
 
 In this city, as well as in otliers which I. visited, every 
 body seemed to walk at his ease. I could see no inclina- 
 tion for jostling ; no impertinent staring at you ; nor 
 attpmpts to create a row in order to pick your pocket. I 
 would stand for an hour together in liroadway, to observe 
 the passing multitude. There is certainly a gentleness in 
 these people, both to be admired and imitated. I could 
 see very few dogs, still fewer cats, and but a very small 
 proportion of fat women in the streets of New York. The 
 climate was the only thing that I had really to find fault 
 with ; and as the autumn was now ap])roaching, I began to 
 think of preparing for warmer regions. 
 
 Strangers are apt to get violent cold, on account of the Vlunnk, 
 sudden change of tl: atmosphere. The noon would often 
 
 X 
 
 m 
 
 i\: 
 
 m{ 
 
no'J 
 
 WANDEKINGS IN SOUTH AMEllICA. 
 
 i-oi-init I'G as Avarm as tropical weather, and the close of day cold 
 '" "'"•'' ''' ^' and chilly. This must sometimes act with severity upon 
 the newly-arrived stranger ; and it requires more care and 
 circumspection than I am master of to guard against it. 
 I contracted a bad and obstinate cough, which did not 
 quite leave me till I had got under the regular heat of the 
 sun, near the equator. 
 Ji8 society. I niay be asked, was it all good fellowship and civility 
 during my stay in the United States? Did no forward 
 person cause offence ? was there no exhibition of drunken- 
 ness, or swearing, or rudeness ; or display of conduct which 
 disgraces civilized man in other countries ? I answer, very 
 few indeed : scarce any worth remembering, and none 
 worth noticing. These are a gentle and a civil people. 
 Should a traveller, now and then in the long run, witness 
 a few of the scenes alluded to, he ought not, on his return 
 Lome, to adduce a solitary instance or two, as the custom 
 of the country. In roving through the wilds of Guiana, I 
 have sometimes seen a tree hollow at heart, shattered and 
 leafless ; but I did not on that account condenni its 
 vigorous neighboiu's, and put down a memorandum that 
 the woods were bad ; on the contrary, I made allowances : 
 u thunder-storm, the whirlwind, a blight from heaven 
 might have robbed it of its bloom, and caused its present 
 forbidding appearance. And, in leaving the forest, I car- 
 ried away the impression, that though some few of the 
 trees were defective, the rest were an ornament to the 
 wilds, full of uses and virtues, and capable of benefiting 
 the world in a superior degree. 
 
 A man generally travels into foreign countries for his 
 own ends ; and I suspect there is scarcely an instance to be 
 found of a person leaving his own home solely with the 
 intention of benefiting those amongst whom he is about to 
 travel. A commercial speculation, curiosity, a wish for 
 
f /ill 
 
 t I 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 307 
 
 information, a desire to reap benefit from an acquaintance fouhth 
 with our distant fellow-creatures, are the general induce- Jo^^^y. 
 ments for a man to leave his own lire-side. This ought 
 never to be forgotten ; and then the traveller will journey- 
 on under the persuasion that it rather becomes him to 
 court than expect to be courted, as his own interest is the 
 chief object of his travels. With this in view, he Avill 
 always render himself pleasant to the natives ; and they 
 are sure to repay his little acts of courtesy with ample 
 interest, and with a fund of information which will be of 
 great service to him. 
 
 While in the United States, I found our western brother 
 a very pleasant fellow ; but his portrait has been drawn in 
 such different shades, by different travellers who have been 
 through his territory, that it requires a personal interview 
 before a correct idea can be formed of his true colours. He 
 is very inquisitive ; but it is quite wrong on that account 
 to tax him with being of an impertinent turn. He merely 
 interrogates you for information ; and when you have 
 satisfied him on that score, only ask him in your turn for 
 an account of what is going on in his own country, and he 
 will tell you everything about it with great good humour, 
 and in excellent language. He has certainly hit upon the 
 way (but I could not make out by what means) of speak- 
 ing a much purer English language than that which is in 
 general spoken on the parent soil. This astonished me 
 much ; but it is really the case. Amongst his many, good 
 qualities, he has one unenviable, and, I may add, a bad 
 propensity : he is immoderately fond of smoking. He may 
 say, that he learned it from his nurse, with whom it was 
 once much in vogue. In Dutch William's time (he was a 
 man of bad taste), the English gentleman could not do 
 without his pipe. During the short space of time that 
 corporal Trim was at the inn inquiring after poor Lefevre's 
 
 X 2 
 
 ;!i 
 
 
 ff 
 
308 
 
 WANDEHINGS IX SOUTH AMElilCA. 
 
 
 
 
 ■\i 
 
 P 
 
 FOU RTII 
 JOURNEY. 
 
 health, my uncle Toby had knocked the ashes out of tliree 
 pipes. " It was not till my uncle Toby had knocked the 
 ashes out of his tLltd pipe," &c. Now these times have 
 luckily gone by, and the custom of smoking amongst 
 genteel Englishmen has nearly died away with them ; it is 
 a foul custom ; it makes a foul mouth, and a foul place 
 where the smoker stands ; however, every nation has its 
 whims. John Bull relishes stinking venison ; a French- 
 man depopulates whole swamps in quest of frogs; a 
 Dutchman's pipe is never out of his mouth ; a llussian 
 will eat tallow candles ; and the American indulges in the 
 cigar. " De gustibus non est disputandum." 
 
 Our western brother is in possession of a country replete 
 The lavs with everything that can contribute to the happiness and 
 govern- comfort of mankind. His code of laws, purified by ex- 
 ment. perience and common sense, has fully answered the expec- 
 tations of the public. By acting up to the true spirit of 
 this code, he has reaped immense advantages from it. His 
 advancement, as a nation, has been rapid beyond all calcu- 
 lation ; and, young as he is, it may be remarked, without 
 any impropriety, that he is now actually reading a salutary 
 lesson to the rest of the civilized world. 
 
 It is but some forty years ago, that he had the dispute 
 with his nurse about a aish of tea. She wanted to force 
 the boy to drink it r'.;cording to her own receipt. He said, 
 he did not like it, and that it absolutely made him ill. 
 After a good deal of sparring, she took up the birch rod, 
 and began to whip him with an uncommon degree of 
 asperity. When the poor lad found that he must either 
 drink the nauseous dish of tea or be flogged to death, he 
 turned upon her in self-defence ; showed her to the outside 
 of the nursery door, and never more allowed her to meddle 
 with his affairs. 
 
 Since the independence, the population has increased 
 
WANDKRINTiS IX SOITIl AMERICA. 
 
 309 
 
 .1 tlie 
 have 
 oiigst 
 it is 
 place 
 as its 
 ench- 
 
 gs; 
 
 a 
 
 iissiaii 
 in the 
 
 ceplete 
 iss and 
 by ex- 
 expec- 
 pirit of 
 ;. His 
 , calcvi- 
 ithout 
 lutary 
 
 ^lispute 
 :o force 
 Le said, 
 Uim ill. 
 xh rod, 
 jree of 
 either 
 jath, he 
 1 outside 
 meddle 
 
 ere 
 
 ased 
 
 from three to ten millions. A tine navy lias heen built ; fourth 
 and everything attended to that could ensure prosperity J'^^^R^k"^'- 
 at home, and respect abroad. 
 
 The former wilds of North America bear ample testi- 
 mony to the achievements of this enterprising people. 
 Forests have been cleared away, swamps drained, canals 
 dug, and nourishing settlements established. From the 
 shores of the Atlantic an immense column of knowledge 
 has rolled into the interior. The INIississippi, the Ohio, 
 the Missouri, and their tributary streams, have been 
 wonderfully benefited by it. It now seems as if it w^ere 
 advancing towards the stony mountains ; and probably 
 will not become stationary till it reaches the Pacific 
 Ocean. This almost immeasurable territory affords a 
 shelter and a home to mankind in general: Jew or Gentile, 
 king's-nian or repulilican, he meets with a friendly recep- 
 tion in the United States. His opinions, his persecutions, 
 his errors, or mistakes, however they may have injured 
 him in other countries, are dead, and of no avail on his 
 arrival here. I'rovided he keeps the peace, he is sure to 
 be at rest. 
 
 Politicians of other countries imagine that intestine 
 feuds will cause a division in this commonwealth ; at 
 present there certainly appears to be no reason for such 
 a conjecture. Heaven forbid that it should happen ! The 
 world at large would suffer by it. For ages yc' to come, 
 may this great commonwealth continue to be the United 
 States of North America. 
 
 The sun was now within a week or two of passing into 
 the southern hemisphere, and the mornings and evenings 
 were too cold to be comfortable. I embarked for the Embarks 
 island of Antigua, with the intention of calling at the ,,/P*] 
 different islands in the Caribbean sen, on my way once 
 more towards the wilds of (Juiaiiii. 
 
 Anlvjna. 
 
 m 
 
 \ ; 
 
 III 
 
H 
 
 ii 
 
 .,1 
 
 , I i! 
 I !' 
 
 ■ 1 7; : t 
 
 Mtli 
 
 FornTii 
 
 JOURNEY, 
 
 CHAPTER II. 
 
 Arrival at Antigiia. — Dominica. — Frogs and Humming Birds. — Martinico. 
 — Diamond rock. — Barbadoes. — Quaslii and Venus. — The Alien Bill. — 
 Sail for Demerara. — More about tho Sloth. — Scarlet Grosbeak. — 
 Crab-eating Owl. — Sun-heron. — Feet of the Tinamou. — ^|Vampires 
 again. — The Karabimiti Humming-bird. — The Monkey tribe. — The 
 Red Howler. — Roast monkey. — The Nondescript. — Altered physi- 
 ognomy. — Gold and silver mines. — Changes of government. — Politics. 
 — India-mbber. — An ingenious deception. 
 
 We were thirty doys in making Antigua, and thanked 
 Providence for ordering us so long a passage. A tre- 
 mendous gale of wind, approaching to a hurricane, had 
 done much damage in the West Indies. Had our passage 
 been of ordinary length, we should inevitably have been 
 caught in the gale. 
 St. Johns. St. John's is the capital of Antigua. In better times it 
 may have had its gaieties and amusements. At present, it 
 appears sad and woe-begone. The houses, which are 
 chiefly of wood, seem as if they had not had a coat of 
 paint for many years ; the streets are uneven and ill- 
 paved; and as the stranger wanders through them, he 
 might fancy that they would afford a congenial promenade 
 to the man who is about to take his last leave of surround- 
 ing worldly misery, before he hangs himself. There had 
 been no rain for some time, so that the parched and barren 
 pasture near the town might, with great truth, be called 
 
WANDEllLNGS IN SolTlI AMEKICA. 
 
 311 
 
 Eosinante's own. The mules feedintf oil them, put you in fouutu 
 mind of Ovid's description of famine : — jh itrnk v. 
 
 "Dura cutis, per quam spoctart viscira possent." 
 
 Tt is somewlmt singuh^r, that there is not a single river or 
 brook in the whole island of Antigua. In this it differs 
 from Tartary in the other world ; which, according to old 
 writers, has five rivers ; "iz. Acheron, Phlegethon , Cocytus, 
 Styx, and Lethe. 
 
 In this island I found the Eed Start, described in Wilson's 
 " Ornithology of the United States." I wished to learn 
 whether any of these birds remain the whole year in 
 Antigua, and breed there ; or whether they all leave it for 
 the north when the sun comes out of the southern hemi- 
 sphere ; but, upon inquiry, I could get no information 
 whatever. 
 
 After passing a dull week here, I sailed for Guadaloupe, Island of 
 whose bold and cloud-capped mountains have a grand 
 appearance as you approach the island. Basseterre, the 
 capital, is a neat town, with a handsome public walk 
 in the middle of it, well shaded by a row of fine tamarind 
 trees on each side. Behind the town. La Souffriere raises 
 its high romantic summit ; and on a clear day, you may 
 see the volcanic smoke which issues from it. 
 
 Nearly midway, betwixt Guadaloupe and Dominica, you 
 descry the Saintes. Though high, and bold, and rocky, 
 they have still a diminutive appearance when compared 
 with their two gigantic neighbours. You just see Mari- 
 gaiante to windward of them, some leagues off, about a 
 yard high in the horizon. 
 
 Dominica is majestic in high and rugged mountains, island of 
 As you sail along it, you cannot help admiring its beauti- -^^""'"''"• 
 ful coffee plantations, in places so abrupt and steep, that 
 you would pronounce them almost inaccessible. Roseau, Roseau, 
 
 Gnada- 
 loupe. 
 
 : If: 
 
 \ ■•'] 
 
 *'i,\ 
 
 W. I: J 
 
 1 
 
312 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 [! i 
 
 1 1 
 
 f".,ii 
 
 ij 
 
 FODHTir the capital, is but a small town, and lias nothing attractive 
 j ouBNE Y. gj^pgp^; j;]jg ^vell-kno^vn hospitality of the present harbour- 
 master, who is particularly attentive to strangers, and 
 furnishes them with a world of information concerning the 
 West Indies. Roseau has seen better days ; and you can 
 trace good taste and judgment in the way in which the 
 town has originally been laid out. 
 
 Some years ago it was visited by a succession of mis- 
 • fortunes, which smote it so severely, that it has never 
 recovered its former appearance. A strong French fleet 
 bombarded it; while a raging fire destroyed its finest 
 buildings. Some time after, an overwhelming flood rolled 
 down the gullies and fissures of the adjacent mountains, 
 and carried all before it. Men, women, and children, 
 houses, and property, were all swept away by this mighty 
 torrent. The terrible scene was said to beggar all descrip- 
 tion, and the loss was immense. 
 
 Dominica is famous for a large species of Frog, which 
 the inhabitants keep in readiness to slaughter for the table. 
 In the woods of this island, the large Ehinoceros Beetle 
 is very common ; it measures above six inches in length. 
 In the same woods is found the beautiful Humming-bird, 
 the breast and throat of which are of a brilliant chanoinii; 
 purple. I have searched for this bird in Brazil, and 
 through the whole of the wilds from the Rio Branco, 
 which is a branch of the Amazons, to the river Paumaron, 
 but never could find it. I was told by a man in the 
 Egyptian-Hall, in Piccadilly, that this humming-bird i.s 
 found in Mexico ; but upon questioning him more about 
 it, his information seemed to have been acquired by 
 hearsay ; and so I concluded that it does not appear in 
 Mexico. I suspect that it is never found out of the 
 Antilles. 
 Martinico. After leaving Dominica, you soon reach the grand 
 
WANDEIlINdS IX SOUTH AMKIUCA. 
 
 3i;i 
 
 '6' 
 
 -bird, 
 
 jar 111 
 If the 
 
 and magnificent island of Martinico. St. Pierre, its Koum-ir 
 lapital, is a fine town, and pos-^osses every comfort. •'" ""'^'' ^' 
 The inhabitants seem to pay considerable attention to 
 the cultivation of the tropical fruits. A stream of water 
 runs down the streets with great rapidity, producing a 
 pleasing effect as you pass along. 
 
 Here I had an opportunity of examining a Cuckoo, 
 which had just been shot. It was exactly the same as 
 the Metallic Cuckoo in Wilson's " Ornithology." They 
 told me it is a migratory bird in Martinico. It probably 
 repairs to this island after its departure from the United 
 States. 
 
 At a little distance from Martinico, the celebrated 
 Diamond Itock rises in insulated majesty out of the 
 sea. It was fortifed during the last war with France, 
 and bravely defended by an English captain. 
 
 In a few hours from Martinico, you are at St. Lucie, St. Lucie. 
 whose rough and towering mountains fill vou with sub- 
 lime ideas, as you approach its rocky shore. The town 
 Castries is quite embayed. It was literally blown to Castries. 
 pieces by the fatal hurricane, in which the unfortunate 
 governor and his lady lost their lives. Its present for- 
 lorn and gloomy appearance, and the grass which is grown 
 up in the streets, too plainly show that its hour of joy is 
 passed away ; and that it is in mourning, as it were, with 
 the rest of the British West Indies. 
 
 From St. Lucie, I proceeded to Barbadoes in quest of a 
 conveyance to the island of Trinidad. 
 
 Near Bridgetown, the capital of Barbadoes, I saw the 
 metallic cuckoo, already alluded to. 
 
 Barbadoes is no longer the merry island it was when I Barkuhcs. 
 visited it some years ago : — 
 
 u 
 
 'Ml 
 
 "rand 
 
 Inl'elix Imbitum, tomiioiis hujus habet." 
 
 ■"■ it 
 
314 
 
 WAXDERTXdS IX SOrill AMEIUCA. 
 
 ! I 
 
 FOURTH There is an old song, to the tune of La Belle Catliarino, 
 J OURNE Y, yvhich must evidently have been composed in brighter 
 times : — 
 
 " Come lot us (knee and sing, 
 AVliilo Biiibadoos bells do ring ; 
 Quaslii sn'ai>es the fiddle-string, 
 And VeuUH plays the lute." 
 
 (Jaaslii's fiddle was silent ; and mute was the lute of 
 Venus during my stay in IJarbadoes. The difference 
 betwixt the French and British islands was very striking. 
 The first appeared happy and content ; the second were 
 filled with murmurs and complaints. The late proceedings 
 Slavery, in England, concerning slavery, and the insurrection in 
 Demerara, had evidently caused the gloom. The abolition 
 of slavery is a question full of benevolence and fine 
 feelings, difficulties and danger : — 
 
 " Tautum ne noceas, dum vis prodcsso videto." 
 
 It requires consummate prudence, and a vast fund of true 
 information, in order to draw just conclusions on this 
 important subject. Phaeton, by awkward driving, set the 
 world on fire : " Sylvte cum montibus ardent." Daedalus 
 gave his son a pair of wings without considering the con- 
 sequence ; the boy flew out of all bounds, Lst his wings, 
 and tumbled into the sea : — 
 
 "Icarus, Icariis nomina fecit aquis." 
 
 When the old man saw what had happened, he damned his 
 own handicraft in wing-making ; " devovitque suas artes." 
 Prudence is a cardinal virtue : — 
 
 " Omnia consulta mentc gercnda tegens." 
 
 Foresight is half the battle. " Hombre apercebido, medio 
 combatido," says Don Quixote, or Sancho, I do not 
 
WANDERINGS IN SOUTH AMERICA. 
 
 SLI 
 
 were 
 
 artes. 
 
 medio 
 o not 
 
 u 
 
 remember which. Had queen Bess weighed well in her own fouhtii 
 mind the probable consequences of this lamentable traffic, '^ """^^' ^' 
 it is likely she would not have been owner of two vessels 
 in Sir John Hawkins's squadron, which committed the 
 first robbery in negro flesh on the coast of Africa. As 
 philanthropy is the very life and soul of this momentous 
 question on slavery, which is certainly fraught with great 
 difficulties and danger, perhaps it would be as well at 
 present for the n. tion to turn its thoughts to poor ill-fated 
 Ireland, where oppression, poverty, and rags make a heart- 
 rending appeal to the feelings of the benevolent. 
 
 But to proceed. There was another thing which added 
 to the dulness of Barbadoes, and which seemed to have 
 considerable effect in keeping away strangers from the 
 island. The legislature had passed a most extraordinary 
 bill, by virtue of which every person who arrives at 
 Barbadoes is obliged to pay two dollars, and two dollars 
 more on his departure from it. It is called the alien bill ; 
 and every Barbadian who leaves or returns to the island, 
 and every Englishman too, pays the tax ! 
 
 Finding no vessel here for Trinidad, I embarked in Embarks 
 a schooner for Demerara, landed there after being nearly uemerara. 
 stranded on a sand-bank, and proceeded without loss of 
 time to the forests in the interior. It was the dry 
 season, which renders a residence in the woods very 
 delightful. 
 
 There are three species of Jacamar to be found on the 
 different sand-hills and dry savannas of Demerara; but 
 there is another much larger and far more beautiful to be 
 seen when you arrive in that part of the country where 
 there are rocks. The jacamar has no affinity to the wood- 
 pecker or kingfisher, (notwithstanding what travellers 
 affirm,) either in its haunts or anatomy. The jacamar 
 lives entirely on insects, but never goes in search of them. 
 
 The 
 Jacamar, 
 
 t . ' 
 
 
 ■,li' 
 
 n 
 
 i*'\ 
 
 
 z 
 
 I IH 
 
 M 
 
\\}\ 
 
 ! 
 
 
 1 
 
 1 
 
 1 
 
 1 : 
 
 , j 
 
 1 
 
 
 HI 
 
 1 ' 
 
 lii 
 
 ;:' 
 
 n 
 
 i 
 
 j 
 
 r.ir, WAXDlCUINdS in SoI'TII AMKRIfA. 
 
 Foi'itrii It .sits palieiitly for liours tooether on the brancli of c, troo, 
 •" "'"^''' ^'' iitid when the incautious insect approaches, it ilies nt it 
 with the rapidity of an arrow, seizes it, and generally 
 returns to eat it on the branch which it h;id just (piitted. 
 Jt has not the least attempt at song, is very solitary, and 
 so tame that you may get within three or four yards of it 
 hefore it takes llight. The males of all the dilferent species 
 which I have examined have white feathers on the throat. 
 I suspect that all the male jacamars hitherto discovered 
 have this distinctive mark. I could learn nothing of its 
 incubation. The Indians informed mo that one species of 
 jacamar lays its eggs in the Wood- Ants' nests, which are 
 so frequent in the trees of Guiana, and appear like huge 
 black balls. I wish there had been proof positive of this ; 
 but the breeding time was over; and in the ants' nests 
 which I examined, I could find no marks of birds having 
 ever been in them. Early in January the jacamar is in 
 fine plumage for the cabinet of the naturalist. The largest 
 species measures ten inches and a half from the point of 
 the beak to the end of the tail ; its name amongst the 
 Indians is Una-waya-adoucati, that is, grandfather of the 
 jacamar. It is certainly a splendid bird ; and in brilliancy 
 and changeableness of its metallic colours, it yields to 
 none of the Asiatic and African feathered tribe. The 
 colours of the female are nearly as bright as those of the 
 male, but she wants the white feathers on the throat. The 
 large jacamar is pretty common about two hundred miles 
 up the river Demerara. 
 The three- Here I had a fine opportunity once more of examining 
 toeddoih. j.|^g Three-toed Sloth. He was in the house with me for a 
 day or two. Had I taken a description of him as he lay 
 sprawling on the floor, I should have misled the world, and 
 injured natural history. On the ground he api)eared really 
 a bungled composition, and fiiulty at all points ; awkward- 
 
 
WANDKItlNiJS IN SolTH A.MbMIU.N. 
 
 ;ii- 
 
 II 1' 
 
 IS 111 
 
 I 
 
 11U.S.S lui'l iiiiHcry wcic depicted on his counteiiiiiicu ; and i-nnMii 
 when I niiide liiiii iidvaiicu he sighinl us tliuu;^di in pain. •"'i^'^^_^^'- 
 reiiiapH it was, that by neeiiij,' liiiii thns out of hin element 
 ns it were, tliat the (jumt de JUiHoii, in liis history of the 
 shjth, asks tlie question — " Why sliouhl not some animals 
 be created for misery, since, in the human npecies, the 
 •greatest nundjer of individuals are devoted to pain from 
 the moment of their existence ?" Were the question put 
 to me, I would answer, 1 cainiot conceive that any of them 
 are created for mis(!ry, Tluit thousands live in misery 
 there can be no doubt; but then, misery has overtaken 
 them in their path throtij^h life, and wherever man has 
 c(tme up with them, 1 should supi>use they have seldom 
 escaped from e\i)eiieii(;inf,' a certain proportion of misery. 
 
 After fully satisfyinj,' myself that it only leads the world 
 into error to descrilu! the sloth while he is on the ground, 
 or in any ]»lace except in a tree, 1 carried the one I had 
 in my possession to his native haunts. As soon as he 
 came in contact with the branch of a tree, all went right 
 with him. 1 could see as he climbed up into his own 
 country, that he was on the right road to happiness ; and 
 felt ])ersuaded more than ever, that the world has hitherto 
 (MTcd in itH conjectures concerning the sloth, on account 
 of naturalists, not having given a description of him when 
 he was in the only ]»osition in which he ought to have 
 been described, namely, clinging to the branch of a tree. 
 
 As the api)earance of this part of the country bears 
 great resemblance to Cayenne, and is so near to it, I was 
 in hopes to have found the Grande (lobe ^louche of Buflon, 
 and the septicoloured Tangara, both of which were common 
 in Cayenne ; but after many diligent searches, I did not 
 succeed; nor could I learn from the Indians that they had 
 ever seen those two species of birds in these parts. 
 
 II (Me 1 procured the (Irosbeak with a rich scarlet Qrosbeak 
 
318 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 '^'i 
 
 , t II 
 
 r '■''I 
 
 FOURTH body, and black head and throat. Buffon mentions it as 
 
 j ouKNE Y. coming from America. I had been in quest of it for years, 
 
 but could never see it, and concluded that it was not to 
 
 be found in Demerara. The bird is of a greenish brown 
 
 before it acquires its rich plumage. 
 
 Amongst the bare roots of the trees, alongside of this 
 
 Procures a part of the river, a red crab sometimes makes its appear- 
 
 sp^iesof ance, as you are passing up and down. It is preyed upon 
 
 Owl. by a large species of Owl, which I was fortunate enough 
 
 to procure. Its head, back, wings, and tail, are of so dark 
 
 a brown as almost to appear black. The breast is of a 
 
 somewhat lighter brown. The belly and thighs are of a 
 
 dirty yellow white. The feathers round the eyes are of 
 
 the same dark brown as the rest of the body ; and then 
 
 comes a circle of white, which has the appearance of a 
 
 large pair of spectacles. I strongly suspect that the dirty 
 
 yellow white of the belly and thighs has originally been 
 
 pure white ; and that it has come to its present colour by 
 
 means of the bird darting down upon its prey in the mud. 
 
 But this is mere conjecture. 
 
 The Here too, close to the river, I frequently saw the bird 
 
 Sun-bird, called Sun-bird by the English colonists, and Tirana by the 
 
 Spaniards in the Oroonoque. It is very elegant ; and in 
 
 its outward appearance approaches near to the heron tribe ; 
 
 still it does not live upon fish. Flies and insects are its 
 
 food; and it takes them just as a heron takes fish, by 
 
 approaching near and then striking with its beak at its 
 
 prey so quick, that it has no chance to escape. The 
 
 beautiful mixture of grey, yellow, green, black, white, and 
 
 chestnut in the plumage of this bird, baffles any attempt 
 
 to give a description of the distribution of them which 
 
 would be satisfactory to the reader. 
 
 The great There is something remarkable in the great Tinamou, 
 
 immou. ^.jji^jj^ J suspect has hitherto escaped notice. It invariably 
 
iWy 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 319 
 
 roosts in trees ; but the feet are so small in proportion rouRrn 
 to the hody of this Lulky bird, that they can be of no •'o^^-y. 
 use to it in grasping the branch ; and, moreover, the hind 
 toe is so short, that it does not touch the ground when the 
 bird is walking. The back part of the leg, just below the 
 knee, is quite flat, and somewhat concave. On it are 
 strong pointed scales, which are very rough, and catch 
 your finger as you move it along from the knee to the toe. 
 Now, by means of these scales, and the particular flatness 
 of that part of the leg, the bird is enabled to sleep in safety 
 upon the branch of a tree. 
 
 At the close of day, the great tinamou gives a loud, 
 monotonous, plaintive whistle, and then immediately 
 springs into the tree. By the light of the full moon, the 
 vigilant and cautious naturalist may see him sitting in the 
 position already described. 
 
 The small Tinamou lias nothing that can be called a The small 
 tail. It never lays more than one egg, which is of a ' 
 chocolate colour. It makes no nests, but merely scratches 
 a little hollow in the sand, generally at the foot of a tree. 
 
 Here we have an instance of a bird, the size of a partridge, 
 and of the same tribe, laying only one egg, while the rest 
 of the family, from the peahen to the quail, are known to 
 lay a considerable number. The foot of this bird is very 
 small in proportion, but the back part of the leg bears no 
 resemblance to that of the larger tinamou; hence one 
 might conclude that it sleeps upon the ground. 
 
 Independent of the hollow trees, the Vampires have 
 another hiding-place. They clear out the inside of the 
 large ants' nests, and then take possession of the shell. 
 I had goue about half a day down the river, to a part 
 of the forest where the wallaba-trees were in great 
 plenty. The seeds had ripened, and I was in hopes to have 
 got the large scarlet ara, which feeds on them. But, 
 
 ! 
 
 F> 
 
 
 i 
 
 If 1 
 
 i| 9 1 
 
 )( 1 III 
 
 ;;^5^ 
 
a": 
 
 m 
 
 320 
 
 WAXDRUTNTtS in SOUTir AMR1?ICA. 
 
 The 
 1 'iniiptrc. 
 
 Kouinii unfortunately, the time had passed away, and the seeds 
 
 While ranging here in the forest, we stopped under an 
 ants' nest; and, by the dirt below, conjectured that it 
 had got new tenants. Thinking it no harm to dislodge 
 them, " vi et armis," an Indian boy ascended the tree ; 
 but, before he reached the nest, out flew above a dozen 
 Vampires. 
 
 I have formerly remarked, that I wished to havd it in 
 my power to say, that I had been sucked by the vr.mpire. 
 I gave them many an opportunity, but they always fought 
 shy ; and though they now sucked a young man of the 
 Indian breed very severely, as he was sleeping in his 
 liammock in tlie shed next to mine, they would ^lave 
 nothing to do with me. His great toe seemed to ha^'e ^1 
 the attractions. I examined it minutely as he was ba' ..i'/ r 
 it in the river at daybreak. The midnight surgeon had 
 made a hole in it, almost of a triangular shape, and the blood 
 was then running from it apace. His hammock w\t,s so 
 defiled and stained with clotted blood, that he was obliged 
 to beg an old black woman to wash it. As she was taking 
 it down to the river side, she spread it out before me, and 
 shook her head. I remarked, that I supposed her own toe 
 was too old and tough to invite the Vampire-doctor to get 
 his supper out of it ; and she answered, with a grin, that 
 doctors generally preferred young people. 
 
 Nobody has yet been able to inform me how it is that 
 the vampire manages to draw such a large quantity of 
 blood, generally from the toe, and the patient, all the time, 
 remain in a profound sleep. I have never heard of an 
 instance of a man waking under the operation. On the 
 contrary, he continues in a sound sleep, and at the time of 
 rising, his eyes first inform him that there has been a 
 thirsty tliief on his toe. 
 
WANl)EUIX(iS IX SOl'TII AMERICA. 
 
 821 
 
 The teeth of the vampire are very sharp, and not unlike fouutii 
 those of the rat. If it be that he inflicts the wounds Joi^^. 
 with his teeth, (and he seems to have no other iiistru- -^'s ^'''^^^^ 
 nients,) one would suppose that the acuteness of the pain 
 would cause the person who is sucked, to awake. We 
 are in darkness in this matter ; and I know of no means 
 by which one might be enabled to throw light upon it. 
 It is to be hoped that some future wanderer through the 
 wilds of Guiana, will be more fortunate than I have been, 
 and catch this nocturnal depredator in the fact. I have 
 once before mentioned tliat I killed a vampire which 
 measured thirty-two inches from wing to wing extended ; 
 but others, which I have since examined, have generally 
 been from twenty to twenty-six inches in dimension. 
 
 The large humming-bird, called by the Indians Kara- 
 bimiti, invariably builds its nest in the slender branches 
 of the trees which hang over the rivers and creeks. In 
 appearance, it is like brown tanned leather, and without a 
 particle of lining. The rim of the nest is double inwards, 
 and I always conjectured that it had taken this shape on 
 account of the body of the bird pressing against it while 
 she was laying her eggs. But this is (j[uite a wrong 
 conjecture. Instinct has taught the bird to give it this 
 shape, in order that the eggs may be prevented from 
 rolling out. 
 
 The trees on the river's bank are particularly exposed 
 to violent gusts of wind, and while I have been sitting in 
 the canoe, and looking on, I have seen the slender branch 
 of the tree which held the hunmiing-bird's nest so violently 
 shaken, that the bottom of the inside of the nest has 
 appeared, and had there been nothing at the rim to 
 stop the eggs, they must inevitably have been jerked 
 into the water. I suspect the humming-bird never lays 
 
 I never found more than two in anv 
 
 Y 
 
 .iihiij; 
 
 
 more than two eggs. 
 
322 
 
 WANDERINGS IN SOUTH AMERK'A. 
 
 1 f ■ ■ !•■•; 
 
 1 i:i: 
 
 
 1 
 
 i ;■. 
 
 [^' 
 
 1 
 
 
 Moilnis, 
 
 Thm 
 i'lassra of 
 Monki'ijs. 
 
 KuiTirrii of ilie many nests wliich Im^'e come in my way. Tlio eggs 
 "'^-'I'^L!:^" were always M'liite, without any spots on them. 
 
 Trohably travellers have erred in asserting tliat the 
 monkeys of South America throw sticks and fruit at their 
 pursuers. I have had fine o])portunities of narrowly 
 watching the different species of monkeys which are found 
 in the wilds, betwixt the Amazons and the Oroonoque* 
 I entirely acquit them of acting on the offensive. When 
 the monkeys are in the high trees over your head, the 
 dead branches will now and then fall down upon you, 
 having been broken off as the monkeys pass along them ; 
 but they are never hurled from their hands. 
 
 Monkeys, commonly so called, both in the old and new 
 continent, may be classed into three grand divisions ; 
 namely, the ape, which has no tail whatever; the baboon, 
 which has only a short tail ; and the monkey, which has 
 a long tail. There are no apes, and no baboons, -as yet 
 discovered in the new world. Its monkeys may be very 
 Avell and very briefly ranged under two heads ; namely, 
 those with hairy and bushy tails; and those whose tails 
 are bare of hair underneath, about six inches from the 
 extremity. Those with hairy and bushy tails climb just 
 like the S(juirrel, and make no u&o of the tail to help them 
 from branch to branch. Those which have the tail bare 
 underneath towards the end, find it of infinite advantage 
 to them, in their ascent and descent. They apply it to the 
 branch of the tree, as though it were a supple finger, and 
 frequently swing by it from the branch like the pendulum 
 of a clock. It answers all the purposes of a fifth hand to 
 the monkey, as naturalists nave already observed. 
 The large The large red monkey of Demerara is not a baboon, 
 iiionkni f/thougli it goes by that name, having a long prensile tail.^ 
 
 eiitnaui. j j bfllieve prens'de is a new-coined word, i have seen it, but do not 
 leniciiibcr wliere. 
 
WAXDEKlXrJS IN SOI^TIT AMERICA. 
 
 a23 
 
 Nothing can sound more dreadful than its nocturnal fouuth 
 liowlings. AVhile lying in your hammock in tliese gloomy •'^j^^^'* 
 and immeasurable wilds, you hear him howling at in- 
 tervals, from eleven o'clock at night till daybreak. You 
 would suppose that half tliu wild beasts of the forest were 
 collecting for tlie work of carnage. Now, it is the tremen- 
 dous roar of the jaguar, as he springs on his prey : now, 
 it changes to his terrible and deep-toned growlings, 
 as he is pressed on all sides by superior force ; and 
 now, you hear his last dying moan, beneath a mortal 
 wound. 
 
 Some naturalists have supposed that these awful sounds, 
 which you would fancy are tliose of enraged and dying 
 wild l)easts, proceed from a number of the red monkeys 
 howling in concert. One of them ulone is capable of 
 producing all these sounds ; and the anatomists, on an 
 inspection of his trachea, will be fully satisfied that this 
 is the case. When you look at him, as he is sitting on 
 the branch of a tree, you will see a lump in his throat, 
 the size of a large hen's egg. In dark and cloudy- 
 weather, and just before a squall of rain, this monkey will 
 often howl in the daytime ; and if you advance cautiously, 
 and get under the high and tufted tree where he is sitting, 
 you may have a capital opportunity of witnessing his 
 wonderful powers of producing these dreadful and dis- 
 cordant sounds. 
 
 His fiesli is good food; but when skinned, his appear- i''^fs/(^/'///« 
 ance is so like that of a young one of our own species, 
 that a delicate stomach might possibly revolt at the idea of 
 putting a knife and fork into it. However, I can affirm, 
 from experience, that after a long and dreary march 
 through these remote forests, the flesh of this monkey is 
 not to be sneezed at, when boiled in Cayenne pepper, or 
 
 A young one tastes 
 \' 2 
 
 . monkey. 
 
 roasted on a stick over a good lire 
 
 ^ff 
 
 1 i 
 
 \ 
 
 
 m 
 
 I 
 
 II 
 
 m 
 
 
 it 
 
 Til 
 
 "< 
 
 
 'i I 
 
 
J 
 
 t ' I 
 
 |iH 
 
 324 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 m 
 
 
 ■rfc; 
 
 FOURTH not unlike a kid, and the old ones have somewhat the 
 jouuNEY. flg^^our of he-goat. 
 
 I mentioned, in a former adventure, that I had hit upon 
 an entirely new plan of making the skins of quadrupeds 
 retain their exact form and feature. Intense application 
 to the subject has since that period enabled me to shorten 
 the process, and hit the character of an animal to a very- 
 great nicety, even to the preservation of the pouting lip, 
 dimples, warts, and wrinkles on the face. I got a fine 
 specimen of the howling monkey ; and took some pains 
 with it, in order to show the immense difference that 
 exists betwixt the features of this monkey, and those 
 • of man. 
 
 I also procured an animal which has caused not a little 
 speculation and astonishment. In my opinion, his thick 
 coat of hair, and great length of tail, put his species out 
 of all question ; but then his face and head cause the 
 inspector to pause for a moment, before he ventures to 
 pronounce his opinion of the classification. He was a 
 large animal, and as I was pressed for daylight, and more- 
 over, felt no inclination to have the whole weight of his 
 body upon my back, I contented myself with his head 
 and shoulders, which I cut off : and have brought them 
 with me to Europe.^ I have since found, that I acted 
 quite right in doing so, having had enough to answer for 
 the head alone, without saying anything of his hands and 
 feet, and of his tail, which appendage, Lord Karnes asserts, 
 belongs to us. 
 
 The features of this animal are quite of the Grecian 
 cast ; and he has a placidity of countenance which shows 
 
 ^ My young friend, Mr. J. H. Foljambe, eldest son of Thomas Foljambe, 
 Esq. of Wakefield, has made a drawing of the head and shoulders of this 
 animal, (see Frontispiece,) and it is certainly a most correct and striking 
 likeness of the original. 
 
Ijambe, 
 of thia 
 itrikiiig 
 
 I 
 
 WANDERINGS IN SOUTH AMERICA. 
 
 325 
 
 that things went well with him when in life. Some fourth 
 gentlemen of great skill and talent, on inspecting his Journky. 
 head, were convinced that the whole series of its features 
 has been changed. Others again have hesitated, and 
 betrayed doubts, not being able to make up their minds, 
 whether it be possible, that the brute features of the 
 monkey can be changed into the noble countenance of 
 man. — "Scind.'iu algus." One might argue at consider- 
 able length on this novel subject ; and perhaps, after all, 
 produce little more than prolix pedantry. " Vox et 
 prteterea nihil." 
 
 Let us suppose for an instant, that it is a new species. 
 Well ; " Una golondrina no hace verano ; " one swallow 
 does not make summer, as Sancho Panza says. Still, for 
 all that, it would be well worth while going out to search 
 for it ; and these times of Pasco-Peruvian enterprise are 
 favourable to the undertaking. Perhaps, gentle readers, 
 you would wish me to go in quest of another. I would 
 beg leave respectfully to answer, that the way is dubious, 
 long, and dreary ; and though, unfortunately, I cannot 
 allege the excuse of " me pia cbnjux detinet," still I 
 would fain crave a little repose. I have already been 
 a long while errant : — 
 
 " Longa mihi exilia, et vastuni maris fpquor aravi, 
 Ne mandate mihi, nam ego sum defessus agendo." 
 
 Should anybody be induced to go, great and innumerable 
 are the discoveries yet to be made in those remote wilds ; 
 and should he succeed in bringing home, even a head alone, 
 with features as perfect as those of that which I have 
 brought, far from being envious of him, I should consider 
 him a modern Alcides, fully entitled to register a thir- 
 teenth labour. Now if, on the other hand, we argue, that 
 this head in question has had all its original features 
 
 m 
 ^ n 
 
 4 1 
 
 I 
 
 i' 
 
 '\< ■ 
 
 m 
 
pp 
 
 i 
 
 826 
 
 WAXDEinXGS IN SOUTH AMEia^A. 
 
 
 :r;i 
 
 k.! ; :n; : S 
 
 Hf ^ iiil 
 
 Mm 
 
 I ' ' 
 
 HH 1 
 
 FouuTii destroyed, and a set of new ones given to it, by wliat 
 j oiuxK Y. jiieans has this hitherto unheard-of cliange been effected ? 
 Nobody in our museums has as yet been able to restore 
 the natural features to stuffed anhnals. ; and he who has 
 any doubts of this, let him take a living cat or dog, and 
 compare tliem with a stuffed cat or dog in any of the 
 first-rate museums. A momentary glance of the eye 
 would soon settle his doubts on this head. 
 
 If I have succeeded in effacing the features of a brute, 
 and putting those of a man in their place, we might be 
 entitled to say, that the sun of I'roteus has risen to our 
 museums : — 
 
 " I'uiu.s liic ruciuiii, f'fuiis traiisfoniuit in oniiies ; 
 Nunc lioiiio, luuii' tiyris ; uuiK,' e([ua, nunc nuilk'r." 
 
 If I have effected this, we can now give to one side of 
 the skin of a man's face the appearance of eighty years, 
 and to the other side that of blooming seventeen. We 
 could make the forehead and eyes serene in youthful 
 beauty, and shape the mouth and jaws to the features of 
 a malicious old ape. Here is a new field opened to the 
 adventui'ous and experimental naturalist : I have trodden 
 it up and down till I am almost weary. To get at it 
 myself I have groped through an alley, which may be 
 styled, in the words of Ovid, — 
 
 " Arduus, obliquus, caligiuc deusiis oi)acii." 
 
 I pray thee, gentle reader, let me out a while. Time 
 passes on apace ; and I want to take thee to have a pee}) 
 at the spots where mines are supposed to exist in Guiana. 
 As the story of this singular head has, probably, not been 
 made out to thy satisfaction, perhaps, (I may say it nearly 
 in Corporal Trim's words,) on some long and dismal winter's 
 evening, but not now, I may tell thee more about it; 
 
WANDERIXCJS IN SOT'TII AMEliirA. 
 
 n-i: 
 
 togetlier witli tliat of another head, wliich is e<iually vuuuiir 
 striking. juuunkv. 
 
 It is commonly reported, and I tliink there is no reason to 
 doubt tlie fact, that when Demerara and Essequibo were 
 under the Dutch flag, there were mines of gold and silver 
 opened near to the river Essequiho. 'J'he miners were not 
 succiessful in their undertaking, and it i^j generally con- 
 jectured that their failure proceeded from inexperience. 
 
 Now, when you ascend the Essequibo, some hundretl 
 miles above the place where these niiiics are said to be 
 found, you get into a high, rocky, and mountainous 
 country. Here many of the moiiiitiiins have a very 
 barren aspect, producin ,' only a lew stinted shrubs, and 
 here and there a tuft • ' coarse grass. I could not learn 
 that they have ever been explored, and at this day their 
 mineralogy is totally unknown to us. The Indians are so 
 thinly scattered in this part of the country, tliat there 
 would be no impropriety in calling it uninhabited : — 
 
 "Apparent rari errajites in gnrgite vtisto." 
 
 It remains to be yet learnt, whether this portion of 
 Guiana be worth looking after, with respect to its supposed 
 mines. The raining speculations at present are flowing 
 down another channel. The rage in England for working 
 the mines of other states has now risen to such a pitch, 
 that it would require a considerable degree of caution in 
 a mere wanderer of the woods, in stepping forward to say 
 anything that might tend to raise or depress the spirits 
 of the speculators. 
 
 A question or two, however, might be asked. AVlien 
 tlie revolted colonies shall have repaired in some measure 
 the ravages of war, and settled their own political eco- 
 nomy upon a firm foundation, will they quietly submit to 
 see foreigners carrying away those treasures which are 
 
 I I; .' 
 
 ^m 
 
 1 i^i 
 
328 
 
 WANDERINGS IN SOLTII AMEHICA. 
 
 FouRTit absolutely i)arL of their own soil, and which necessity 
 j oi-HXK Y. (necessity has no law) forced them to barter away in their 
 hour of need? Now, if it should so happen that the 
 masters of the country begin to repent of their bargain, 
 and become envious of the riches which foreigners carry 
 off, many a teasing law might be made, and many a vexa- 
 tious enaction might be put in force, that would, in all 
 probability, bring the speculators into trouble and 
 disappointment. 
 
 I'csides this consideration, there is another circumstance 
 which onght not to be overlooked. I allude to the change 
 of masters nearly thronghout the whole of America. It 
 is a curious subject for the European philosopher to 
 moralize upon, and for the politician to examine. The 
 more they consider it, the more they will be astonished. 
 If we may judge by what has already taken place, we are 
 entitled to predict, that in a very few years more, no 
 European banner will be seen to float in any part of the 
 new world. Let us take a cursory view of it. 
 
 England some years ago possessed a large portion of 
 the present United States. France had Louisiana ; Spain 
 held the Floridas, Mexico, Darien, Ten-a Firma, Buenos 
 Ayros, Paraguay, Chili, Peru, and California ; and Portugal 
 ruled the whole of Brazil. All these immense regions are 
 now independent states. England, to be sure, still has 
 Canada, Nova Scotia, and a few creeks on the coast of 
 Labrador ; also a small settlement in Honduras, and the 
 wilds of Demerara and Essequibo ; and these are all. 
 France has not a foot of ground except the forests of 
 Cayenne. Portugal has lost every province ; Spain is 
 blockaded in nearly her last citadel ; and the Dutch flag 
 is only seen in Surinam. Nothing more now remains to 
 Europe of this immense continent, where, but a very few 
 years ago, she reigned triumphant. 
 
WAXDEHTXGS IN SOUTH AMERICA. 
 
 329 
 
 "With regard to the West India Ishinds, they may be foitrth 
 considered as the mere outposts of tliis mammoth domain. •'''^[^^'Y* 
 St. Domingo has already shaken off lier old masters, and 
 become a star of observation to the rest of the sable 
 brethren. The anti-slavery associations of England, full 
 of benevolence and activity, have opened a tremendous 
 battery upon the last remaining forts, whicli the lords of 
 the old continent still hold in the new world ; and, in 
 all probability, will not cease firing till thoy shall have 
 caused the last flag to be struck, of Europe's late mighty 
 empire in the Transatlantic regions. It cannot well be 
 doubted, but that the sable hordes in the West Indies will 
 like to follow good example, whenever they shall have 
 it in their power to do so. 
 
 Now with St. Domingo as ah example before them, how 
 long will it be before they try to raise themselves into 
 independent states ? And if they should succeed in 
 crushing us in these our last remaining tenements, I would 
 bet ten to one that none of the new governments will put 
 on mourning for our departure out of the new world. AVe 
 must well remember, that our own government was taxed 
 with injustice and oppression by the United States during 
 their great struggle ; and the British press for years past 
 has, and is still teeming with every kind of abuse and 
 unbecoming satire against Spain and Portugal for their 
 conduct towards the now revolted colonies. 
 
 France also comes in for her share of obloquy. Now, 
 this being the case, will not America at large wish most 
 devoutly for the day to come when Europe shall have no 
 more dominion over her ? Will she not say to us, Our new 
 forms of government are very different from your old 
 ones ? We will trade with you, but we shall always I e 
 very suspicious of you as long as you retain possession of 
 the West Indies, which are, as we may say, close to our 
 
 
 if 
 
 I- ii^ 
 
 ^f 
 
, I .1 
 
 8no 
 
 WANDKriXflS IN SOrill AlMKIJirA. 
 
 
 m 
 
 M 
 
 is 
 
 I oritur 
 
 JnlUNKV, 
 
 Cods of 
 the Rack: 
 
 (luor-tjtt'iuls. Yuu iiiiist lie very cuiitious liow you iiitur- 
 IcrcMvitli our jjolitics ; lor, if wo liud you lucdtUiu^' with 
 tlieiu, and l>y thai lueauH cause us to coino to logiJiorheuiis, 
 wo yiuill he obliged to send you Ijuek to your own homes, 
 tliree or lour thousand miles aeross the Atlantie; and then, 
 with that great ditch betwixt us, we may hojjo we shall bo 
 good friends, lie who easts his eye on the East Indies^ 
 will there see quite a different state of things. The con- 
 quered districts have merely changed one European master 
 for another; and I believe there is no instance of any por- 
 tion of the East Indies throwing off the yoke of the Euro- 
 peans and establishing a government of their own. 
 
 Ye who are versed in polities, and study the rise and 
 fall of empires, and know what is good for civilized man, 
 and what is bad for him, or in other words, what will 
 make him happy and what will make him niiserablo — tell 
 lis how comes it that Europe has lost almost her last acre 
 in tlie boundless expanse of territory which she so lately 
 possessed in the west, and still contrives to hold her vast 
 property in the extensive regions of the east ? 
 
 But whither am I going { I lind myself on a new aiid 
 dangerous path. Pardon, gentle reader, this sudden devia- 
 tion. Methinks I hear thee saying to me, — 
 
 " Traniite quo tcmlis, ninjora([«c viiibus amies." 
 
 T grant that I have erred, but I will do so no more. In 
 general I avoid politics ; they are too heavy for me, and I 
 am aware that they have caused the fall of many a strong 
 and able man; they require the shoulders of Atlas to 
 support their weight. 
 
 When I was in the rocky mountains of Macoushia, in 
 the month of June, 1812, 1 saw four young Cocks of the 
 Rock in an Indian's hut ; they liad been taken out of the 
 nest that week. They were of a uniform dirty brown 
 
Bti'ong 
 Us tu 
 
 iia, 111 
 ^f the 
 kf the 
 orowii 
 
 WAN'I»i:UIXGS IN' SuL'TII AMKUICA. 
 
 r,;u 
 
 colour, tiud l»y thu position ol' the young I'uuther.s u[ioii the koimiiii 
 lieiid, you might hvq tliiit tlieie would be a crest there J"^^^^';^"- 
 wiieu the bird arrived at maturity. J»y seeing young ones 
 in the month of June, I immediately concluded that the old 
 cock of the rock would be in line plumage from the end 
 of Novend)er to the beginning of May : and that the 
 naturalist, who was in quest of specimens for his museum, 
 ought to ai'range his pLms in such a manner as to be able 
 to get into Macoushia during these months. However, T 
 lind now, that no xact period ca.i be fixed; for, in I)ecem- 
 l»er, 1824, an Indian, in the river Uemerara, gave me ii 
 young cock of the rock not !i month old, and it had j"st 
 l)een brought from the jM: '■oushi Duutry. By having a 
 young specinu'ii at this time of lb year, it puts it out of 
 one's power to say at what precis.^! time tlic old birds are in 
 full plumage. 1 took \i .n board a biup with me for 
 l^ngland, but it was so very susceptible of cold that it 
 shiveretl and died, three days after we had passed 
 Antigua. 
 
 If ever there should be a great demand for large supplies Lnii"- 
 of gum-elastic, commonly called India-rubber, it may be *'"^''^'- 
 procured in abundance far away in the wilds of Demerara 
 and Esse<piibo. 
 
 Some years ago, when I was in the Macoushi country, A>i TiaUmi 
 there was a capital trick played upon me about India- 
 rubber. It is >di(iost too good to be left out of these Wan- 
 derings, and it sliows that the wild and uneducated Imlian 
 is not without abilities. Weary and sick, and feeble 
 through loss of blood, I arrived at some Indian huts, 
 whicli were about two hours distant from the })lace where the 
 gum-elastic trees grew. After a day and a night's rest I 
 went to then), and with my own hands made a fine ball 
 of pure India-rubber ; it hardened immediately it became 
 exposed to the air, and its elasticity was almost incredible. 
 
 trkk. 
 
 (!: 
 
 I 
 
 
 1 [. ' .; 
 ' Hit 
 
 \l 
 
 ^1 
 
 t 
 
 j" i 
 
 iiji 
 
 4 
 
 1 :.i 
 
 2 |:,i 
 
 
 

 :i| 
 
 h 
 
 332 
 
 WANDEiaXGS IN SOUTH AMEIUCA. 
 
 FOURTH 
 JOURNEY. 
 
 While procuiiijg it, exposure to the rain, which fell in 
 torrents, brought on a return of inflammation in the 
 stomach, and I was obliged to have recourse again to the 
 lancet, and to use it with an unsparing hand. I wanted 
 another ball, but was not in a state the next morning to 
 proceed to the trees. A fine interesting young Indian 
 observing my eagerness to have it, tendered his services, 
 and asked me two handsful of fish-hooks for his trouble. 
 
 Off he went, and to my great surprise returned in a 
 very short time. Bearing in mind the trouble and time 
 it had cost me to make a ball, I could account for this 
 Indian's expedition in no other way except that, being an 
 inhabitant of the forest, he knew how to go about his 
 work in a much shorter way than I did. His ball, to be 
 sure, had very little elasticity in it. I tried it repeatedly, 
 but it never rebounded a yard high. The young Indian 
 watciied me with great gravity, and when I made him 
 understand that I expected the ball would dance better, 
 he called another Indian, who knew a little English, to 
 assure me that I might be quite easy on that score. The 
 young rogue, in order to render me a complete dupe, 
 brought the new mooi to his aid. He gave me to under- 
 stand that the ball was like the little moon, which he 
 pointed to, and by the time it grew big and old, the ball 
 would bounce beautifully. This satisfied me, and I gave 
 liim the fish-hooks, which he received without the least 
 change of countenance. 
 
 I bounced the ball repeatedly for two months after, but 
 I found that it still remained in its infancy. At last I 
 suspected that the savage (to use a vulgar phrase) had 
 come Yorkshire over me ; and so I determined to find out 
 how he had managed to take me in. I cut the ball in 
 two, and then saw what a taught trick he had played me. 
 It seems he had chewed some leaves into a lump, the size 
 
gave 
 least 
 
 WANDEELIGS IN SOUTH AMEHTCA. 
 
 333 
 
 imj 
 remarkn. 
 
 of a walnut, and then dipped them in the liquid gum- fourth 
 elastic. It immediately received a coat about as thick as •'^uuney. 
 a sixpence. He tiien rolled some more leaves round it, 
 and gave it another coat. He seems to have continued 
 this process till he made the ball considerably larger than 
 the one I had procured ; and in order to put his roguery 
 out of all chance of detection, he made the last and outer 
 coat thicker th . a dollar. This Indian would, no doubt, 
 have thriven well in some of our great towns. 
 
 Finding that the rainy season was coming on, I left the Returns 
 wilds of Demerara and Essequibo with regret, towards the England. 
 close of December, 1824) ; and reached once more the 
 shores of England, after a long and unpleasant passage. 
 
 Ere we part, kind reader, I could wish to draw a little Condml- 
 of thy attention to the instructions which are to be found 
 at the end of this book. Twenty years have now rolled 
 away since I first began to examine the specimens of 
 zoology in our museums. As the system of preparation 
 is founded in error, nothing but deformity, distortion, and 
 disproportion, will be the result of the best intentions and 
 utmost exertions of the workman. Canova's education, 
 taste, and genius enabled him to present to the world 
 statues so correct and beautiful that they are worthy of 
 universal admiration. Had a common stonecutter tried 
 his hand upon the block oui, of which these statues were 
 sculptured, what a lamentable want of symmetry and 
 fine countenance there would have been. Now, when we 
 reflect that the preserved specimens in our museums and 
 private collections are always done upon a wrong prin- 
 ciple, and generally by low and illiterate people, whose 
 daily bread depends upon the shortness of time in which 
 they can get through their work, and whose opposition to 
 the true way of preparing specimens can only be surpassed 
 by their obstinacy in adhering to the old method ; can we 
 
 li;:| 
 
 4- t; 
 
 It :^ 
 
 ii 
 
 
 
 'I 
 
 
 
 M 
 
 t- 
 
 v'i 
 
 'A, 
 
 m 
 
 1 
 
 I \ 
 
 '• I 
 
 m 
 
 f 
 
 I 
 
 ' I ^ 
 
 ' \ 
 
 ? 
 
 
"i it' 
 
 ii! 
 
 I'* 
 
 834 
 
 AVANDEUfNGS TN SOUTH AMRRfCA. 
 
 I'Mt'iiTM auy loiioer wonder at their want of success ; or hope to 
 jmukxkv. ggg g^ single speciiueii produced that will be worth looking 
 at ? With this 1 conclude, hoping that thou hast received 
 some information, and occasionally had a smile upon thy 
 (countenance, while perusing these " Wanderings ; " and 
 begging, at the same time, to add that 
 
 Well I know thy penetration 
 
 Many a stain and blot will s -e, 
 In the languid, long narration, 
 
 Of my sylvan errantly. 
 
 l''(ir the ])i!n too oft was weary 
 
 111 the wandering writer's hand, 
 Ah he roved through deep and dreary 
 
 Forests, in a distant land. 
 
 Sliow thy meri'y, gentle reader, 
 
 l,et him not entreat in vain ; 
 It will be his strength's best feeder, 
 
 Should he ever go again. 
 
 And who knows how scon, comphuiiiiig 
 
 Of a eold and wifeles-s home, 
 lie may leave it, and again in 
 
 liijuatorial regions roani I 
 
 C. W 
 
 ■' » 
 
 i 
 
 i ' I 
 
 1*1 : ' 
 
ox rilESERVIXG BIRDS FOR CABIXETS OF 
 NATURAL HISTORY. 
 
 I'UKSKPV- 
 I NO BIIIDS. 
 
 FiiuUs ill Ini'd-studiiig. — Tools rtMiuircJ. — Knowleclgi; of anatomy. — 
 AttitiiLles of birds. — Flow of the ])liiinng('.— How to skin a bird. — 
 Inserting eottoii. — Killing womuled birds. — Stnfling a l)awk. — Tiie 
 tirst inciision. — The skin to be pu.shed, not pnlled. — Arrangement of 
 wings. — Modelling the body. — Spreading the tail. — Constant attention 
 re([nired. — Strength and elasticity. — Valne of «;orrosive sublimate. — 
 Experience and patience. 
 
 AVeee you to pay as muoli attention to birds as the sculp- 
 tor does to the human frame, you would immediately see, 
 on entering a nmseum, that the specimens are not well 
 done. 
 
 Tliis remark will not Le thought severe vhen you 
 reflect tluit that which once was a bird has probably 
 been stretched, stuffed, stiffened, and wired by the hand 
 of a common clown. Consider likewise how the plumage 
 must have been disordered by too much stretching or 
 drying, and perhaps sullied, or at least deranged, by the 
 ])ressure of a coarse and heavy hand — ])lumage which, ere 
 life liad fled from within it, wns accustomed to be touched 
 by nothing rougher than the dew of heaven, and the pure 
 and gentle breath uf air. 
 
 In dissecting, three things are necessary to ensure Dlmrti in/. 
 success, viz., a ])enknife, a hand not coarse or clumsy, 
 and practice. The flist will furnish you with the means; 
 
 • i Hi 
 
 'lii 
 
 1 W \\ ' 
 
33(> 
 
 ON PUESEHVTNd IHRDS. 
 
 'r ■ I 
 
 PRKSEUV- 
 ING BIRDS. 
 
 Slttffuig. 
 
 Requisite 
 
 to have a 
 
 thorough 
 
 knowledge 
 
 of Ornitlw- 
 
 logical 
 Anatomy. 
 
 Examine 
 tlie eco- 
 nomy of 
 the orders 
 of birds. 
 
 the second will enable you to dissect ; and the third cause 
 you to dissect well. These may be called the mere 
 mechanical requisites. 
 
 In stuffing, you require cotton, a needle and thread, a 
 little stick the size of a common knitting-needle, glass 
 eyes, a solution of corrosive sul ximate, and any kind of a 
 common temporary box to hold the specimen. These also 
 may go under the same denomination as the former. But 
 if you wish to excel in the art, if you wish to be in orni- 
 thology what Angelo was in sculpture, you must apply to 
 profound study and your own genius to assist you. And 
 these may be called the scientific requisites. 
 
 You must have a complete knowledge of ornithological 
 anatomy. You nmst pay close attention to the form and 
 attitude of the bird, and know exactly the proportion 
 each curve, or extension, or contraction, or expansion of 
 any particular part bears to the rest of the body. In a 
 word, you must possess Promethean boldness, and bring 
 down fire and animation, as it were, into your preserved 
 specimen. 
 
 Repair to the haunts of birds on plains and mountains, 
 forests, swamps, and lakes, and give up your time to 
 examine the economy of the different orders of birds. 
 
 Then you will place your eagle in attitude commanding, 
 the same as Nelson stood in, in the d.ay of battle, on the 
 Victory's quarter-deck. Your pie will seem crafty, and 
 just ready to take flight, as though fearful of being sur- 
 prised in some mischievous plunder. Your sparrow will 
 retain its wonted pertness by means of placing his tail a 
 little elevated, and giving a moderate arch to the neck. 
 Your vulture will show his sluggish habits by having his 
 body nearly parallel to the earth, his wings somewhat 
 drooping, and their extremities under the tail instead of 
 above it — expressive of ignoble indolence. 
 
 * . ! 
 
ON PKESEKVING BIKDS. 
 
 337 
 
 3n the 
 and 
 sur- 
 will 
 
 tail a 
 neck. 
 
 g his 
 ewhat 
 
 sad of 
 
 The. 
 featkers. 
 
 Your dove will be in artless, fearless innocence, looking presehv- 
 mildly at you, with its neck, not too much stretched, as ^^ ^ °"tP '*' 
 if uneasy in its situation, or drawn too close into the 
 shoulders, like one wishing to avoid a discovery ; but in 
 moderate, perpendicular length, supporting the head hori- 
 zontally, which will set off the breast to the best advan- 
 tage. And the breast ought to be conspicuous, and have 
 this attention paid to it ; for when a young lady is sweet 
 and gentle in her manners, kind and affable to those around 
 her ; when her eyes stand in tears of pity for the woes of 
 others, and she puts a small portion of what Providence 
 lias blessed her with into the hand of imploring poverty and 
 hunger — then we say she has the breast -of a turtle-dove. 
 
 You will observe how beautifully the feathers of a bird 
 are arranged, one falling over the other in nicest c rder ; 
 and that, where this charming harmony is interrupted, the 
 defect, though not noticed by an ordinary spectator, will 
 appear immediately to the eye of a naturalist. Thus, a 
 bird not wounded and in perfect feather must be procured 
 if possible, for the loss of feathers can seldom be made 
 good; and where the deficiency is great, all the skill of 
 the artist will avail him little in his attempt to conceal 
 the defect, because, in order to hide it, he must contract 
 the skin, bring down the upper feathers, and shove in the 
 lower ones, which would throw all the surrounding parts 
 into contortion. 
 
 You will also observe that the whole of the skin does 
 not produce feathers, and that it is very tender where the 
 feathers do not grow. The bare parts are admirably 
 formed for expansion about the throat and stomach, and 
 they tit into the different cavities of the body at the wings, 
 shoulders, rump, t'nd thighs with wonderful exactness; so 
 that in stuffing the bird, if you make an even rotund 
 surface of the skin where these cavities existed, in lieu of 
 
 z 
 
 M't ! 
 
 s m 
 
338 
 
 ON PRESERVING BIRDS. 
 
 !l 
 
 pRKsEuv- reforming them, all symmetry, order, and proportion are 
 
 You must lay it down as an absolute rule that the bird 
 is to be entirely skinned, otherwise you can never succeed 
 in forming a true and pleasing specimen. 
 
 You will allow this to be just, after reflecting a moment 
 on the nature of the fleshy parts and tendons, which are 
 often left in : 1st, they require to be well seasoned with 
 aromatic spices ; 2dly, they must be put into an oven to 
 dry ; 3dly, the heat of the fire and the natural tendency all 
 cured flesh has to shrink and become hard renders the 
 specimen withered, distorted, and too small; 4thly, the 
 inside then becomes like a ham or any other dried meat. 
 Ere long the insects claim it as their own; the feathers 
 begin to drop off", and you have the hideous spectacle of 
 death in ragged plumage. 
 
 Wire is of no manner of use, but, on the contrary, a 
 great nuisance; for where it is introduced, a disagreeable 
 stiffness and derangement of symmetry follow. 
 
 The head and neck can be placed in any attitude, the 
 body supported, the wings closed, extended or elevated, the 
 tail depressed, raised or expanded, the thighs set horizontal 
 or oblique, without any aid from wire. Cotton will effect 
 all this. 
 
 A very small proportion of the skull bone, say, from the 
 forepart of the eyes to the bill, is to be left in; though 
 even this is not absolutely necessary. Part of the wing- 
 bones, the jaw-bones, and half of the thigh-bones, remain. 
 Everything else, flesh, fat, eyes, bones, brains, and tendons, 
 are all to be taken away. 
 General While dissccting, it will be of use to keep in mind, — 
 
 yom fingers and a little knife, you must try to shove it, in 
 lieu of pulling it, lest you stretch it. 
 
ON PRESERVING BIRDS. 
 
 330 
 
 lOUgh 
 
 That, yon must press as lightly as possible on the bird, preseuv- 
 and every now and then take a view of it, to see that the JNo^niuu.s. 
 feathers, &c., are all right. 
 
 That, when you come to the head, you must take care 
 that the body of the skin rests on your knee ; for if you 
 allow it to dangle from your hand, its own weight will 
 stretch it too much. 
 
 That, throughout the whole operation, as fast as you 
 detach the skin from the body, you must put cotton im- 
 mediately betwixt the body and it ; and this will effectually 
 prevent any fat, blood, or moisture from coming in contact 
 with the plumage. Here it may be observed that, on the 
 belly you find an inner skin, which keeps the bowels in 
 their place. By a nice operation with the knife, you can 
 cut through the outer skin, and leave the inner skin whole. 
 Attention to this will render your work very clean; so 
 that, with a little care in other parts, you may skin a bird 
 without even soiling your finger ends. 
 
 As you can seldom get a bird without shooting it, a line 
 or two on this head will be necessary. If the bird be still 
 alive, press it hard with your finger and thumb, just behind 
 tJie wings, and it will soon expire. Carry it by the legs, 
 and then, the body being reversed, the blood cannot escape 
 down the plumage through the shot-holes. As blood will 
 often have issued out before you have laid hold of the 
 bird, find out the shot-holes, by dividing the feathers with 
 your fingers, and blowing on them, and then, with your 
 penknife, or the leaf of a tree, carefully remove the 
 clotted blood, and put a little cotton on the hole. If, 
 after all, the plumage has not escaped the marks of blood ; 
 or if if has imbibed slime from the ground, wash the 
 part in water, without soap, and keep gently agitating 
 the feathers, with your fingers, till they are quite dry. 
 Were you to wash them, and leave them to dry by 
 
 z 2 
 
 If,: 
 
 r 
 
 lU 
 
 ;:l 
 
340 
 
 OX PKESKUVING BIRDS. 
 
 i1 
 
 if: I 
 
 I'KKsERv- themselves, they would have a very meau and shrivelled 
 
 INU BIUD8. 
 
 . • appearance. 
 
 In the act of skinning a bird, you must either have it 
 sk^nil^ a "P*^" ^ table, or upon your knee. Probably, you will 
 
 bird, prefer your knee ; because when you cross one knee over 
 the other, and have the bird upon the uppermost, you can 
 raise it to your eye, or lower it, at pleasure, by means of 
 the foot on the ground, and then your knee will always 
 move in unison with your body, by which much stooping 
 will be avoided and lassitude prevented. 
 
 With these precautionary hints in mind, we will now 
 proceed to dissect a bird. Suppose we take a hawk. The 
 little birds will thank us, with a song for his death, for he 
 has oppressed them sorely; and in size he is just the 
 thing. His skin is also pretty tough, and the feathers 
 adhere to it. 
 
 AVe will put close by us a little bottle of the solution of 
 corrosive sublimate in alcohol ; also a stick like a common 
 knitting-needle, and a handful or two of cotton. Now till 
 the mouth and nostrils of the bird with cotton, and place 
 it upon your knee on its back, with its head pointing to 
 your left shoulder. Take hold of the knife with your two 
 first fingers and thumb, the edge upwards. You must not 
 keep the point of the knife perpendicular to the body of 
 the bird ; because, were you to hold it so, you would cut 
 the inner skin of the belly, and thus let the bowels out. 
 To avoid this, let your knife be parallel to the body, and 
 then you will divide the outer skin with great ease. 
 
 Begin on the belly below the breast-bone, and cut down 
 the middle, quite to the vent. This done, put the bird iu 
 any convenient position, and separate the skin from the 
 body, till you get at the middle joint of the thigh. Cut it 
 through, and do nothing more there at present, except 
 introducing cotton all the way on that side, from the vent 
 
 I 
 
 
ON PRESERVING BIRDS. 
 
 841 
 
 to the breast-bone. Do exactly the same on the opposite preskuv- 
 
 side. '^"J!!!"^^- 
 
 Now place the bird perpendicular, its breast resting on 
 your knee, with its back towards you. Separate the skin 
 from the body on each side at the vent, and never mind at 
 present the part from the vent to the root of the tail. 
 Bend the tail gently down to the back, and while your 
 finger and thumb are keeping down the detached parts of 
 the skin on each side of the vent, cut quite across, and 
 deep, till you see the back-bone, near the oil-gland at the 
 root of the tail. Sever the back-bone at the joint, and 
 then you have all the root of the tail, together with the 
 oil-gland, dissected from the body. Apply plenty of 
 cotton. 
 
 After this, seize the end of the back-bone with your 
 finger and thumb : and now you can hold up the bird 
 clear of your knee, and turn it round and round, as 
 occasion requires. While you are holding it thus, contrive, 
 with the help of your other hand and knife, by cutting and 
 shoving, to get the skin pushed up till you come to where 
 tlie wing joins on to the body. 
 
 Forget not to apply cotton ; cut this joint through ; do 
 the same at the other wing, add cotton, and gently push 
 the skin over the head ; cut out the roots of the ears, 
 which lie very deep in the head, and continue skinning til\ 
 you reach the middle of the eye ; cut the nictitating^ 
 membrane quite through, otherwise you would tear the 
 orbit of the eye ; and after this, nothing difficult intervenes 
 to prevent your arriving at the root of the bill. 
 
 When this is effected, cut away the body,* leaving a 
 little bit of skull, just as much as will reach to the fore- 
 part of the eye ; clean well the jaw-bones, fasten a little 
 cotton at the end of your stick, dip it into the solution, 
 and touch the skull and corresponding part of the skin, as 
 
 'tin I (ill 
 
 \fn 
 
 J 
 
F 
 
 :l 
 
 'j^ 
 
 m i 
 
 '■\'i 
 
 11 1 
 
 84;^ 
 
 ON PJ^ESKIiVIXa niRDS. 
 
 puKKKitv. you cannot well ^et to these places afterwards. From the 
 i.N o HiKD s. j.jjjj^ qJ" pi,j.]jii,g l\^^, g]^jn Qver the head, you are supposed 
 
 to have had the bird resting upon your knee ; keep it 
 there still, and vith great caution and tenderness return 
 the head through the inverted skin, and when you see the 
 beak appearing, pull it very gently till the head comes out 
 unruflled and unstained. 
 
 You may now take the cotton out of the mouth ; cut 
 away all the remaining llesh at the palate, and whatever 
 may have remained at the under jaw. 
 
 Here is now before you the skin, without loss of any 
 feathers, and all the flesh, fat, and uncleaned bones out of 
 it, except the middle joint of the wings, one bone of the 
 thighs, and the fleshy root of the tail. The extreme point 
 of the wing is very small, and has no flesh on it, com- 
 paratively speaking, so that it requires no attention, except 
 touching it with the solution from the outside. Take all 
 the flesh from the remaining joint of the wing, and tie a 
 thread about four inches long to the end of it ; touch all 
 with the solution, and put the wing -bone back into its 
 place. In baring this bone you must by no means pull 
 the skin ; you would tear it to pieces beyond all doubt, for 
 the ends of the long feathers are attached to the bone 
 itself; you must push off the skin with your thumb-nail 
 and forefinger. Now skin the thigh quite to the knee ; cut 
 away all flesh and tendons, and leave the bone : form an 
 artificial thigh round it with cotton ; apyjly the solution, 
 and draw back the skin over the artificial thigh : the same 
 to the other thigh. 
 
 Lastly, proceed to the tail ; take out the inside of the 
 oil-gland, remove all the remaining flesh from the root, 
 till you see the ends of the tail-feathers ; give it the solu- 
 tion, and replace it. Now take out all the cotton which 
 you have been putting into the body from time to time to 
 
ON PRESEHVING lUliDS. 
 
 B43 
 
 presoi've the featliers from ^roase and stains. Placo the ruKSEuv- 
 bii'd \\\)C)n your knee on its back ; tie togetlier the '^ ".""" "^' 
 two tlireatls wliieh yon liad fastened to the end of the 
 wing-joints, leaving exactly the same space betwixt them 
 as your knowledge in anatomy informs you existed there 
 when the bird was entire ; hold the skin open with your 
 finger and thumb, and apply the solution to every part of 
 the inside. iNuglect the head and neck at present; i\wy 
 are to receive it afterwards. 
 
 Fill the body moderately with cotton, lest the feathers 
 on the belly should be injured whilst you are about tlie 
 following operation. You must recollect that half of tlie 
 thigh, or in other words, one joint of the thigh-bone, has 
 been cut away. Now, as this bone never moved perpen- 
 dicular to the body, but, on the contrary, in an oblique 
 direction, of course, as soon as it is cut off, the remaining 
 part of the thigh and leg, having nothing now to support 
 them obliquely, must naturally fall to their perpendicular. 
 Hence tlie reason why the legs appear considerably too 
 long. To correct this, take your needle and thread, fasten 
 the end round the bone inside, and then push the needle 
 through the skin just opposite to it. Look on the outside,- 
 and after finding the needle amongst the feathers, tack up 
 the tlii^h under the winu' with sevei'al strom? stitches. 
 This \\ ill shorten the thigh, and render it quite capable of 
 supporting the weight of the body without the help of 
 wire. This done, take out every bit of cotton, except the 
 artificial thighs, and adjust the wing-bones (which are con- 
 nected by the thread) in the most even manner possible, so 
 that one joint does not appear to lie lower than the other ; 
 for unless they are quite equal, the Avings themselves will 
 be unequal when you come to put them in their proper 
 attitude. Here then rests the shell of the poor hawk, 
 ready to receive, from your skill and judgment, the size, 
 
 
 I 
 
 11 
 
; 
 
 i !i 
 
 U' 
 
 
 
 f !!' 
 
 , I' 
 
 i> ir I, 
 
 i, f 
 
 I 
 
 
 ^it 
 
 Ui 
 
 ON PRES, ;t> VTNG HIUDS. 
 
 rnKsKnv- the shape, tlin foatiiiv.s ami expression it had, ore dcalli, 
 iN o niRP H. j^^^^i yQjj^. (]ig<5(%(.t|„g liaiul, hrouj^ht it to its present still 
 
 and fonnless state. The cold liand of death stamps deep 
 its mark upon the prostrate victim, "When tlie lieart 
 ceases to beat, and the blood no lon;,'or courses through the 
 veins, the features collapse, and the whole frame seems to 
 shrink M'ithin itself. If then you have formed your idea 
 of the real appearance of the bird from a dead specimen, 
 you will be in error. With this in mind, and at the same 
 time forming your specimen a trille larger than life, to 
 make up for what it will lose in drying, you will reproduce 
 a bird that will please you. 
 
 It is now time to introduce the cotton for an artificial 
 body, by means of a little stick like a knitting-needle ; and 
 without any other aid or substance than that of this little 
 stick and cotton, your own genius must produce those 
 swellings and cavities, that just proportion, that elegance 
 and harmony of the whole, so much admired in animated 
 nature, so little i^ttended to in preserved specimens. After 
 you have introduced the cotton, sew up the orifice you 
 originally made in the belly, beginning at the vent. And 
 from time to time, till your arrive at the last stitch, keep 
 adding a little cotton, in order that there may be no defi- 
 ciency there. Lastly, dip your stick into the solution, and 
 put it down the throat three or four times, in order that 
 every part may receive it. 
 
 When the head and neck are tilled with cotton quite to 
 your liking, close the bill as in nature. A little bit of 
 bees' wax, at the point of it, will keep the mandibles in 
 their proper place. A needle must be stuck into the 
 lower mandible perpendicularly. You will shortly see the 
 use of it. Bring also the feet together by a pin, and then 
 run a thread through the knees, by which you may draw 
 them to each other, as near as you judge proper. Nothing 
 
0\ PnERKUVTXO niRDR. 
 
 845 
 
 now roniains to 1)0 aildod l)ut tlio eyes. Witli your little rnrsmv- 
 stick niako a hollow in tlin cotton within the orbit, and '^ " """ "^' 
 introduce the ^la.ss eyes thvou^di the orbit. Adjust tlin 
 orbit to tlicni, as in nature, and that requires no other 
 fastener. 
 
 Your close insjtection of the eyes of animals will 
 already have informed you, that the orbit is ca])able of 
 receiving a much larf,'er body than that part of the eye 
 which a])]»ears M'iihin it when in life. So that, were you 
 to proportion your eye to the size the orbit is capable of 
 receivin;:,', it would be far too lar^'O. Inattention to this 
 has caused the eyes of every specimen, in the best cabi- 
 nets of natural history, to be out of .ill proportion. To 
 prevent this, contract the orbit, by means of a very small 
 delicate needle and thread, at that part of it farthest from 
 the beak. This may be done with such nicety, that the 
 stitch cannot be obscn-ved ; and thus you have the artificial 
 eye in true proportion. 
 
 After this, touch the bill, orbits, feet, and former oil- 
 gland at the root of the tail, with the solution, and then 
 you have given to the hawk everything necessary, except 
 attitude, and a proper degree of elasticity, two qualities 
 very essential. 
 
 Procure any common ordinary box, fdl one end of it, 
 about three-fourths up to the top, with cotton, forming a 
 sloping plane. Make a moderate hollow in it to receive 
 the bird. Now take the hawk in your hands, and, after 
 putting the wings in order, place it in the cotton, with its 
 legs in a sitting posture. The head will fall down. Xever 
 mind. Get a cork, and run three pins into the end, just 
 like a three-legged stool. Place it under the bird's bill, 
 and run the needle, which you formerly fixed there, into 
 the head of the cork. This will support the bird's head 
 admirably. If you wish to lengthen the neck, raise the 
 
 I ;n 
 
346 
 
 ON PRESERVING BIRDS. 
 
 Hi 
 
 If ^ ' 
 
 puESKuv- cork, by putting more cctton under it. If the head is to he 
 iNn HiiiDs. iji-ought forward, bring the cork nearer to the end of the 
 box. If it requires to be set backwards on the shoulders, 
 move back the cork. 
 
 As in drying, the back-part of the neck will shrink more 
 than the fore-part, and thus throw the beak higlier than 
 you wish it to be, putting you in mind of a stargazing 
 horse, prevent this fault, by tying a thread to the beak, 
 and fastening it to the end of the box with a pin or needle. 
 If you choose to elevate the wings, do so, and support 
 them with cotton ; and should you wish to have them par- 
 ticularly high, apply a little stick under each wing, and 
 fasten the end of them to the side of the box with a little 
 bees' wax. 
 
 If you would have the tail expanded, reverse the order 
 of the feathers, beginning from the two middle ones. 
 When dry, replace them in their true order, and the tail 
 will preserve for ever the expansion you have given it. Is 
 the crest to be erect? move the feathers in a contrary 
 direction to that in which they lie, for a day or two, and 
 it will never fall down after. 
 
 Place the box anywhere in your room, out of the 
 influence of the sun, wind, and fire ; for the specimen must 
 dry very slowly, if you wisn to reproduce every feature. 
 On this account the solution of corrosive sublimate is un- 
 commonly serviceable ; for at the same time that it totally 
 prevents putrefaction, it renders the skin moist and flexible 
 for many days. While the bird is drying, take it out and 
 replace it in its position once every day. Then, if you see 
 that any part begins to shrink into disproportion, you can 
 easily remedy it. 
 
 The small covert feathers of the wings are apt to rise a 
 little, because the skin will come in contact with the bone 
 which remains in the wing. Pull gently the part that 
 
ON PllESERVING BIRDS. 
 
 347 
 
 rise a 
 
 bone 
 
 that 
 
 rises, with your finger and thumb, for a day or two. Press pukskuv- 
 the feathers down. The skin will adhere no more to the i^'^J^J's- 
 bone, and they will cease to rise. 
 
 Every now and then touch and retouch all the different 
 parts of the features, in order to render them distinct and 
 visible, correcting at the same time any harshness, or 
 unnatural risings or sinkings, flatness or rotundity. This 
 is putting the last finishing hand to it. 
 
 In three or four days the feet lose the ir natural elas- 
 ticity, and the knees begin to stiffen. Wlien you observe 
 this, it is time to give the legs any angk you wish, and 
 arrange the toes for a standing position, or curve them to 
 your finger. If you wish to set the bird on a branch, bore 
 a little hole under each foot, a little way up the leg ; and 
 having fixed two proportional spikes on the branch, you 
 can, in a moment, transfer the bird from your finger to it, 
 and from it to your finger, at pleasure. 
 
 Wbon the bird is quite dry, pull the tlircad out of the 
 knees, take away the needle, &c,, from under the bill, and 
 all is done. In lieu of being stiff with wires, the cotton 
 will have given a considerable elasticity to every part of 
 your bird ; so that, when perching on your finger, if you 
 press it down with the other hand, it will rise again. You 
 need not fear that your hawk will alter, or its colours fade. 
 The alcohol has introduced the sublimate into every part 
 and pore of the skin, quite to the roots of the feathers 
 Its use is twofold. 1st. It has totally prevented all ten- 
 dency to putrefaction ; and thus a sound skin has attached 
 itself to the roots of the feathers. You may take hold of 
 a single one, and from it suspend five times the weight of 
 the bird. You maj ^ jrk it ; it will still adhere to the skin, 
 and, after repeated trials, often break short. 2dly. As no 
 part of the skin has escaped receiving particles of sub- 
 limate contained in the alcohol, there is not a spot exposed 
 
 I'l'!^!' 
 
 'i 
 
 :.' Uj 
 
\tv 
 
 ''M 
 
 
 |::. 
 
 
 t 
 
 
 
 
 
 
 1 P . 
 
 B : 
 
 '* 1 ■ 
 
 
 • 
 
 34« 
 
 ON PRESERVING RTRDS. 
 
 I'liKiiKRv- to the clcprediition of insects, for they Mill never ventuic 
 I NO HI It ns. ^^ attack any substance wliicli has received corrosive 
 suLlimate. 
 
 You are aware that corrosive sublimate is the most fatal 
 poison to insects that is known. It is antiputrescent ; so 
 is alcohol ; and they are both colourless ; of course they 
 cannot leave a stain behind them. Tlie spirit penetrates 
 the pores of the skin with wonderful velocity, deposits 
 invisible particles of the sublimate, and flies off. The sub- 
 limate will not injure the skin, and nothing can .detach it 
 from the parts where the alcohol has left it.^ 
 
 Furs of animals, immersed in this solution, will retain 
 their pristine brightness and durability in any climate. 
 
 Take the finest curled feather from a lady's head, dip it 
 in the solution, and shake it gently till it be dry ; you will 
 find that the spirit will fly off in a few minutes, not a curl 
 in the feather will be injured, and the sublimate will pre- 
 serve it from the depredation of the insect. 
 
 Perhaps it may be satisfactory to add here, that, some 
 years ago, I did a bird upon this plan in Demerara. It 
 remained there two years. It was then conveyed to 
 England, where it stayed five months, and returned to 
 Demerara. After being four years more there, it was 
 conveyed back again through the West Indies to England, 
 where it has now been near five years, unfaded and 
 unchanged. 
 
 On reflecting that this bird has been twice in the tem- 
 perate and torrid zone, uwd remained some years in the hot 
 and humid climate of Demerara, only six degrees from the 
 line, and where almost everything becomes a prey to the 
 
 * All the feathers rerinire to be touched with tlie solution, iu order that 
 they may be preserved from tlie depredation of the moth. The surest way 
 of proceeding is, to immerse the bird in the solution of corrosive sublimate, 
 and then dry it before you begin to dissect it. 
 
ON PKESEllVING BIKDS. 
 
 349 
 
 insect, and that it is still as sound and bright as when it i-reseuv- 
 was first done, it will not be thought extravagant to ^^^ b"'>"^- 
 surmise that this specimen will retain its pristine form 
 and colours for years after the hand that stuffed it has 
 mouldered into dust. 
 
 I have shown this art to the naturalists in Brazil, 
 Cayenne, Demerara, Oroonoque, and Home, and to the 
 royal cabinets of Turin and Florence. A severe accident 
 prevented me from communicating it to the cabinet of 
 Paris, according to my promise. A word or two more, 
 and then we will conclude. ' 
 
 A little time and experience will enable you to produce 
 a finished specimen. "Mox similis volucri, mox vera 
 volucris." If your early performance should not corre- 
 spond with your expectations, do not let that cast you 
 down. You cannot become an adept all at once. The 
 poor hawk itself, which you have just been dissecting, 
 waited to be fiedged before it durst rise on expanded 
 pinion ; and had parental aid and frequent practice ere it 
 could soar with safety and ease beyond the sight of man. 
 
 Little more remains to be added, except tliat what has 
 been penned down with regard to birds may be applied, 
 in some measruic, to serpents, insects, and four-footed 
 animals. 
 
 Should ycp. find these instructions too tedious, let the 
 wish to p'ivg you every information plead in their defence. 
 They nught have been shorter : but Horace says, by 
 labouring to be brief you become obscure. 
 
 If, by their means, you should be enabled to procure 
 specimens from foreign parts in better preservation than 
 usual, so that the naturalist may have it in his power to 
 give a more perfect description of them than has hitherto 
 been the case ; sliould they cause any unknown species to 
 be brought into public \'h.\v, and thus add a little more to 
 
 
 iimrii 
 
 ■I -1 
 
 mi > ' 
 
 
 t^*^ 
 
iw 
 
 ill r I 
 
 N I 
 
 350 
 
 ON PRESERVING BIRDS. 
 
 pRESEuv- the page of natural history, it will please me much. But 
 iN Q BIRD S, siiould they, unfortunately, tend to cause a wanton expense 
 of life ; should they tempt you to shoot the pretty song- 
 ster warbling near your door, or destroy the mother, as she 
 is sitting on the nest to warm her little ones ; or kill the 
 father, as he is bringing a mouthful of food for their sup- 
 port ; — oh, then ! — deep indeed will be the regret that I 
 ever wrote them. 
 
 Adieu, 
 
 Chakles Waterton. 
 
 I I; t' 
 
 B ■; I' 
 
 r: 
 
 ' I: 
 
 r 
 
 I' 
 
IP 
 
 1. ' .'1 
 
 Mi 
 
 EXPLANATORY INDEX. 
 
 
 llM 
 
 ll. 
 
 ft i 
 
 ' 1 
 
 i-j* 
 

 ■if 
 
 
EXPLANATORY INDEX. 
 
 x\. 
 
 ' i! 
 
 ( 
 
 
 AcAiAiiT.— This is the resinous guiu of the Hayawa, or 
 Hiawa, tree, Idea heptaphijUa. 
 
 It is reddish brown, rather translucent, and brittle. When 
 placed on the fire, or, even bettor, laid on a piece of iron 
 nearly red-hot, it melts and then forms bubbles, which on 
 breaking throw out puffs of a highly perfumed smoke, very 
 much I'esembling that of the purest incense. Mr. Waterton 
 gave me a large piece of it, and when describing the manners 
 and customs of Guiana to my guests I have often gmtitiod 
 them with the fragrant smoke of the Acaiari. The natives 
 sometimes burn lumps of it as torches. 
 
 In his Essays, Waterton frequently gives disconnected 
 pieces of information which are extremely useful when 
 employed as additions to the Wanderings. For example, 
 when treating of the trade winds, ho makes the following 
 remai'ks on the acaiari gum : — - 
 
 " In Guiana there is a tree called Hayawa ; it produces a 
 deliciously-smelling resin, fit for incense. When the Indians 
 stop on the banks of a river for the night, they are much In 
 the habit of burning this resin for its fine and wholesome 
 scent. It is found in a hard and lumpy state, all down tho 
 side of the tree from which it has oozed. It is also seen on 
 the foot of the tree, and incorporated with the sand. 
 
 A A 
 
 h\'i 
 
 mi 
 
 :i I 
 
 .isiiifi 
 
 ^ .'^•■' 
 
364 
 
 KXIM-AXATOIJY INDKX. 
 
 Ill 
 
 " When we had taken up our niglitly quaitors on Uuj bank 
 of the Essequibo, many a time we perceived this delightful 
 fragrance of the Ilayawa, which came down the bed of the 
 river to the place where we were, in a direction (|uite opposite to 
 the trade wind. My Indians knew by this that other Indians 
 weie encamped for the night on the river side above us." 
 
 In appearance the Acaiari much resembles the kaurie or 
 cowdie gum of New Zealand, which exudes from a species of 
 pine, and, like the Acaiari, is often found in large lumps im- 
 bedded in the sand. It is imported in great quantities, as an 
 excellent varnish is made of it. 
 
 The Hayawa ti-ee grows plentifully on the Essequibo in 
 loose f-andy soil. The wood is not miich used because it 
 decays on exposure to weather, but, on account of its per- 
 fume, it is recommended for naturalists boxes and cabinets. 
 It runs to about fifty feet in height. The gum is sometimes 
 known as Resin of Konima. 
 
 AcouRT {Dasijprocta accnc/ii). — This is one of the ('a vies, of 
 which there are .several species. The teeth, which are used 
 as sights for the blow-gun, are the two incisors of the lower 
 jaw. A lump of "kurumanni" wax is heated and fixed to 
 the tube. The teeth are then laid upon it nearly parallel 
 to each other, with the convex side upwards, and pressed into 
 the wax. When the weapon is bi ought to this country, the 
 ■wax becomes brittle, and the teeth are apt to be broken off. 
 In travelling, the only hope of pre.'-erving this ingenious back- 
 sight, is by wrapping it in a piece of linen, and then covering 
 it with a thick layer of tow or cotton wool. 
 
 The size of the Acouri is about equal to. that of a large 
 rabbit, and its flesh can be eaten, although it is vei'y dry and 
 has scarcely a particle of fat. 
 
 AcuEHO. — Sometimes called Aquiro (Astrocarjjon acideatiim). 
 A species of palm. The fruit is used in lieu of butter, and 
 the seeds, which are spherical, very hard, and as black as 
 ebony, are cut into table napkin ring.s, kc, just like the nut 
 of the voL'etablo ivorv. 
 
EXPLANATOHY INDKX. 
 
 Xjo 
 
 'g 
 
 id 
 
 uit 
 
 ^TA {Mauritia Jlexuosa). — This is a palm, which, in favour- 
 able situations, attains an enormous size, sometimes being 
 upwards of a hundred feet in height before the branches, or 
 rather leaves, are reached. As this great palm is widely 
 spread and is conspicuous, it has received various names. 
 Wateiiion calls it -^ta, spelt by others Ita or Itah. Cut 
 the name by which it is most generally known is Moriche, or 
 Murichi. Of this splendid palm, Kingsley writes as follows 
 m At Last : — 
 
 "The noble Moriche palm delights in wet, at least in 
 Trinidad and on the lower Orinoco; but Schomburgk describes 
 forests of them — if, indeed, it be the same species — as growing 
 in the mountains of Cuiana up to an altitude of four thousand 
 feet. 
 
 " The soil in which thoy grow here is half pitch pavement, 
 half loose brown earth, and over both, shallow pools of water, 
 which will become nuich deeper in the wet season ; and all 
 about float or lie their pretty fruit, the size of an apple, and 
 scaled like a fir cone. They are last year's, empty and de- 
 cayed. The ripe fruit contains first a rich pulpy nut, and at 
 last a hard cone, something like that of the vegetable ivory 
 palm {Ph ytclephas macrocarpa) which grows in the mainland, 
 but not here. Delicious they are, and precious to monkeys 
 and parrots, as well as to the Orinoco Indians, among wliom 
 the Jamunacs, according to Humboldt, say, that when a man 
 and woman survived that great deluge, which the Mexicans 
 call the age of water, they cast behind them, over their heads, 
 the fruits of the Moriche palm, as Deucalion and Pyrrha cast 
 stones, and saw the seeds in them produce men and women, 
 who re-peopled the eaith. No wonder, indeed that certain 
 tribes look on this tree as sacred, or that the missionaries 
 should have named it the tree of life." 
 
 Humboldt gives the following eloquent account of this 
 palm in his Personal Xarratke ; — 
 
 " In the season of inundations these clumps of Mauritia, 
 with their leaves in the form of a fan, have the appearance 
 
 A A -1 
 
 
 i rv!', ' 
 
;j5G 
 
 KXI'I.ANATOIIY INDEX. 
 
 ,.<l' : I 
 
 I : 
 
 of II foiGst rising from tlio bosom of tlie wiitern. Tlio navi- 
 gator in procot'iling along tho channels of the delta of the 
 (Orinoco at night, sees with s,iuM-rise the summit of the palm- 
 trees illumined by largo liros. These are the Iiabitations of 
 the Guaraons (Tivitivas and W;.iawoties of Raleigh), which 
 are suspended from the trunks of tho treos. These tribes hang 
 up mats in tho air, which i liey fill with earth, and kindle 
 on a layer of moist clay the lire necessary for their household 
 wants. They have owed tlnir liberty and their political in- 
 dependence for ages to the (piaking and swampy soil, which 
 they pass over in the time of drought, and on which they 
 alone know how to walk in security to their solitude in the 
 delta of the Orinoco, to their abode on tho trees, where 
 religious enthusiasm will prol)al)ly never lead any American 
 Stylites. 
 
 " The Mauritia palm tree, the tree of fife of tlie missionaries, 
 not only affords the (hiaraons a safe dwelling during the 
 risings of the Orinoco, but its shelly fruit, its farinaceous 
 pith, its juice, abounding in sa(charine mattei', and the fibres 
 of its petioles, furni.sh them with food, wine, and thread 
 proper for making cords and weaving hammocks. These 
 customs of tho Indians of the delta of the Orinoco were found 
 formerly in the Gulf of J^aiien (Uraba), and in the greater 
 part of the inundated lands between the Guerapiche and the 
 mouths of the Amazon. It is curious to observe in the lowest 
 degree of human civilization the existence of a whole tribe 
 depending on one single species of palm-tree, similar to those 
 insects which feed on one and the same flower, or on one and 
 the same part of a plant." 
 
 The word Guaraon, here used by Humboldt, is another 
 rendei'ing of the word Warow, one of the native tribes, a 
 branch of which chooses this curious life. 
 
 Travellers in the country where the Moriche grows, are in 
 the habit of using sandals made of the basal part of the leaf- 
 stalk. They do not la.-t long, and have to be renewed every 
 third day. 13ut, as the Moriche is always plentiful, and any 
 
Tho nivvi- 
 jlta of tho 
 p the pal in- 
 vitations of 
 sigh), which 
 tribes hang 
 
 and kindle 
 r household 
 political in- 
 f soil, which 
 
 which they 
 itude in the 
 trees, where 
 tiy American 
 
 missionaries, 
 
 during tho 
 ■* farinaceous 
 md the fibres 
 
 and thread 
 3cks. These 
 o were found 
 
 the greater 
 )iche and the 
 in the lowest 
 
 whole tribe 
 uilar to those 
 )!• on one and 
 
 It, is another 
 tive tribes, a 
 
 grows, are in 
 it of the leaf- 
 renewed every 
 itiful, and any 
 
 
 KXPLAXATOUY T\l FA. 
 
 native can make those sandals as well as twist the thongs 
 from the outer fibres of the same leaf, there is no ditllculty 
 in keeping up tho supply. 
 
 Ai. — See Hloth, Three-toed. 
 
 Alijicouk [Thjpinvs jiacijicns). — As its scientific name im- 
 plies, it is closely related to the celebrated tunny. It is a 
 voracious fish, and is easily taken by a hook when there is 
 a smart breeze, and the ship is going swiftly thr* iv.'! tho 
 water, but in a calm, the Albicoi'e becomes suspicious, (<nd 
 will no' 'me near the ship. 
 
 ALi.Ki —See Cayman. 
 
 Aniii . {^J'lofus (uiMiKja). — This very remarkable bird 
 belongs to the important group of the Pelicans, and is a 
 mighty catcher of fish. It pursues its prey in a most singului" 
 
 fashion. It ci'awls along some rock or tree-stump, whicii 
 overhangs the water, and sits so motionless that it is 
 scarcely distinguishable. When it sees a fish within reach, it 
 
 ,lii' 
 
 n 
 
 m 
 
 J 
 
 I ,f 
 

 N> 
 
 IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 Wo 
 
 ^ 
 
 ^<r 
 
 
 ^ 
 
 <^ 
 
 ^ 
 
 ^ 
 
 
 ''Z 
 
 ^>' 
 
 jr 
 
 V 
 
 '^^ 
 
 // 
 
 1.0 
 
 I.I 
 
 11.25 
 
 itt m 
 
 m 
 u 
 
 25 
 2.2 
 
 Itt 
 u 
 u 
 
 lAO 
 
 U ||.6 
 
 Hiotographic 
 
 Sdmoes 
 
 GarpoiBtian 
 
 ^ 
 
 ^ ^ 
 
 <- 
 
 ^. 
 
 
 23 VnST MAIN STIMT 
 
 WltSTU,N.Y. 14S00 
 
 (7U)I73-4SM 
 
 
 ;\ 
 
[4" 
 
 
;55H 
 
 HXPI.ANATOUY IN'DKX, 
 
 drops into the water so silently that the river seems as if it 
 were made of oil. It is an oxcollont swiininei", keeping its 
 body submerged, and writhing its U)ng and pliant neck in so 
 serpentine a manner that it has often been mistaken for a 
 water- snake. 
 
 In consequence of a habit of darting its long neck and 
 pointed beak through the foliage among which it is resting, 
 it has been called the Darter. Homotimes it is known by the 
 name of Duoklar. 
 
 Ant, Black [Ponera graudh). - It is one of the largest and 
 most venomous of the Ants, and is here given of its natural 
 size. 
 
 The sting of this Ant is not only [).iinful, but absolutely 
 dangerous. Mr. Brown mentions that one of his men nearly 
 lost his life from the stings of three of these terrible Ants. 
 
 (IIIKAT rU.AlK ANT IMI'SMII) 
 
 Indeed, so much are they dreaded, that travellers have been 
 fairly turned out of their houses by the entrance of only a 
 few Muniris, as they are called. 
 
 Ant, ('oushie {(Ecodoma ce.jihalntufi). (V)U^hfes are veritable 
 plfigues in tropical America. They niake burrows in all 
 directions, sometimes entering houses and giving no notice of 
 their prasence until their tunnel is broken, in which case, the 
 inhabitants have to vacate the premises for a time. I knew 
 of an instance where the Coushies ruined a gold mine for a 
 

 EXPI.ANATORY IXDKX. 
 
 .']5!,t 
 
 time, breaking into it with a tunnel some eighty yard.s in 
 length, and letting in a torrent of water, which broke down 
 the machinery, and washed away all the supports, so that the 
 mine had to be dug afresh. 
 
 Moreover, the stronghold of the Conchies had to be dis- 
 covered and destroyed, a long and costly task, only to be 
 undei'taken by men who make it their special business. 
 
 r^' -^s. 
 
 IIIL'SHIK AN'I 
 
 When a tunnel has been traced to the nest, a large dome is 
 built over it, filled with wood and sulphur, and closed except 
 a few openings for the admission of air. 
 
 The wood is then lighted, bellows are inserted into the 
 holes, and negi'o slaves work the bellows day and night until 
 all the fuel is exhausted. Perhaps there may be a dozen 
 burrows radiating from the nest. These are discovered by 
 the smoke rising from their entrances, which are stopped up 
 and marked. 
 
 When the fire has ceas^ed, the holes for the bellows are 
 stopped, and the nest is left for sevei*al days, so that not one 
 Ant is left in it alive. Then the tunnels are laid open from 
 beginning to end, and filled up with clay rammed into them. 
 The tropical sun soon bakes the clay a;^ hard as brick, and 
 not until then is the locality considered as safe. 
 
 Mr. C. B. F3rown met with rather a ludicrous incident in 
 which the Coushies were caught in their own trap : — 
 
 " I had a small tin can, shaped like a sandwich box, with a 
 layer of cork in the bottom, in which I pinned any curious 
 insects met with. One night our camp was not far from the 
 
 ■ 
 
 ;J. ■ i; 
 
 I it , ( . 
 
 ■'I :i' 
 
 I 
 
 \i' 
 
360 
 
 KXPLAXATOUY INDKX. 
 
 large earth-mounds of the Cushi ant {(Ecodoina cephalutes), 
 and a band of these creatures got into the insect case, de- 
 posited their loads of earth, and then set to work to destroy 
 all my small but choice collection in the most heartless 
 manner. They were unable to carry off the insects, after 
 cutting them up, owing to the curved shape of the sides of the 
 box preventing their egress, and were thus neatly trapped." 
 
 This insect is sometimes called by the name of Saiiba. 
 
 Ant, Small Eed {Myrniica scevissima). — This is popularly 
 called the Fire Ant, on account of the sharpness of its sting. 
 The same traveller, who trapped the coushies, himself fell a 
 victim to the Fire Ant : — 
 
 " At this place I observed a tall, straight sapling, with a 
 regularly ai'ranged stjle of branches in one or two sets near 
 its top. Taking a machete, or cutlass, I gave the tree a blow 
 with the intention of cutting it down, and was in the act of 
 raising my hand to deliver another, when I received two or 
 three sharp stings on the back of the neck, which felt like 
 sparks of tire, and produced such a demoralizing effect upon 
 my nerves, that I dropped the cutlass and fled. 
 
 " I hastily raised my hand to my neck, and seized two or 
 three long- bodied, amber- coloured Ants, which had been dis- 
 lodged from the tree by the jar produced by the blow, and 
 had fallen upon me. These Ants always irJiabit this kind of 
 tree, living at tlie base of its leaf-stalks. The pain did not 
 leave ine for over an hour afterwards, and made me feel 
 exceedingly rulHed and wroth." 
 
 Several travellers have told me that each sting feels like a 
 red hot needle thrust into the skin, the only difference being 
 that the pain of thn needle would soon die away, while that 
 of the sting remains. 
 
 There is much more to be said about the Ants of Guiana, 
 but space is too limited for further description. 
 
 Ants' nests in trees. — These are evidently Termites, so 
 generally miscalled White Ants. One of tlieir nests is well 
 described by C. Kingsloy : — 
 
lOXPLANATUliY INDEX. 
 
 \m 
 
 "We passed too, in the path, an object curious enough, if 
 not beautiful. Up a smooth stem ran a little rib, seemingly 
 of -earth and dead wood, almost straight, and about half an 
 inch aci'oss, leading to a great brown lump among the branches, 
 as big as a bushel basket. We broke it open, and found it 
 a covered gallery, swarming with life. Brown, ant-like crea- 
 tures, white, maggot like creatures, of several shapes and 
 sizes, were hurrying up and down, as busy as human beings 
 in Cheapside. They were Termites, "white ants" — of which 
 of the many species I know not — and the lump above was 
 their nest. But why they should find it wisest to pack their 
 nest aloft is as ditiScult to guess, as to guess why they take 
 the trouble to build this gallery up to it, instead of walking 
 up the stem in the open air. It may be that they are afraid 
 of birds. It may be, too, that they actually dislike the 
 light. At all events, the majority of them — the workers and 
 soldiers, I believe, without exception — are blind, and do all 
 their work by an intensely developed sense of touch, and it 
 may be of smell and hearing also. Be that as it may, we 
 should have seen them, had we had time to wait, repair the 
 breach in the gallery, with as much discipline and division of 
 labour as average human workmen in a manufactory." 
 Ants' nests on ground. Also Termites. 
 Ant-Bear, oh Great Ant-eater, {Myrmeco^jJiaga jtibata)- 
 — Waterton's statement that it could be a dangei'ous foe was 
 long discredited. Now, as Waterton has rescued so many 
 animals fi'om the evil report that they were dangerous, he 
 might have been believed when he said that the Ant-bear, 
 though it never attacks without provocation, is a terrible 
 antagonist when irritated. As usual, Waterton was right and 
 the critics wrong. In Brown's Canoe Life in Guiana there is 
 a story of a doubly fatal fight between an Ant-bear and a 
 nn.Uve, which completely corroborates Waterton's statement. 
 
 " We had not gone many miles before the guide lost the 
 path, and we all scattered to look for it. In doing so, I 
 walked almost on the top of a sleeping Ant bear, which. 
 
 i''i 
 
 
 
 
 \\\ 
 
 hJ 
 
 1 
 
 
 ih 
 
 
 :.!i 
 
 ■ii i \ 
 
 i 
 
 I 
 
•M'l KXPLANATORY INDEX. 
 
 sprlngini? up, sat on its hind legs, and gra>ptd at me with 
 its huge fore-claws. I sprang quickly to one side, and thus 
 escaped. Thinking that it was good eating, I shot it, but, 
 the Indians said that it was not wholesome food, although, 
 from the great interest they took in seeing it killed, I thought 
 it was. [Waterton says that its flesh is good eating.] 
 
 " These large Ant-eaters are very dangei'ous customers, 
 and have been known to kill men. William told me that an 
 
 "M^ 
 
 v'v'SS'-.' 
 
 IIRK.M' AM MKAII. 
 
 I ' 
 
 Indian, living near Roraima, was hunting in the forest to the 
 north of that mountain with some others, armed with his long 
 blow-pipe. In returning home, considerably in advance of 
 the rest of the party, it is supposed that he saw a young 
 Ant-eater, and, taking it up in his arms, w.is carrying it 
 home, when its mother gave chase, overtook, and killed him ; 
 for, when his companions came up, they found him lying 
 dead on his face in the embrace of the Ant-be.ir, one of its 
 large claws having entered his heart. In the struggle ho 
 had managed to stick his knife behind his back into the 
 animal, which bled to.death, but not before the poor fellow 
 had succumbed to its terrible hug. 
 
KXPLANATOKY IXDKX. 
 
 ;i(i;j 
 
 " It was evident thit ho hnJ only heard the Ant-eatiu' 
 coming when it was clo^e upon hiiu, and in turning round 
 to look, his blow pipe got caught across the path in front of 
 him ; then, as he turned to run, it formed a bar to his pro- 
 
 of 
 
 its 
 ho 
 he 
 ow 
 
 
 l.lTl'l.K AM-ltKAl:. 
 
 gress, and he fell over it as the animal seized him. So firmly 
 had the animal grappled him, that to separate it from the 
 corpse the Indians had to cut off its fore-legs." 
 Ant-Bear (Smaller.)— (" Cyduthurus il{dact>/lHs.") 
 
 i^i 
 
 ii'iMiil 
 
 \'W 
 
•MU 
 
 KXPI.ANATOIIY IXDKX. 
 
 This is a very much smaller animal than the Great Ant- 
 bear, and while the one feeds on insects which it finds in the 
 ground, so does the other subsist on the ants and termites 
 which live in trees. Its tail is long and prehensile, and when 
 dissected its ribs are seen to be so Hat, so thin, and so wide 
 that this part of the skeleton reminds the observer of the 
 bands of an armadillo. The very characteristic sketch of a 
 Little Ant eater in repose was drawn by a lady, and lent to 
 the late C. Kingsley, from whose delightful At Last it is 
 taken. 
 
 Ant-biuds. — These are also called Ant-Thrushes, and belong; 
 to the genus Pitta. They are all thick-bod: 'd birds, with largo 
 heads, long legs, short tails, and very strou^' bills, as perhaps 
 is needful, considering the food on which they almost entirely 
 subsist. The largest of them is called the Giant Pitta, and 
 is about equal in size to our English rook. It is a native of 
 Hurinam, 
 
 Several species of Ant-Thrush are to be found in India 
 They find an English representative in the well-known 
 Dipper of our streams and lakes. 
 
 Apron. — This ornament, for it can hardly be called a dress, 
 is made of various sizes, the average being that of the beautiful 
 example shown in the illustration, and drawn from a specimen 
 in my collection. It is eight inches in width, and four in 
 depth, and the colours are most artistically arranged, so as 
 to produce definite patterns in blue, yellow, green, carmine 
 chalk-white, and opaque vermilion. Other specimens are 
 rather larger, and some are not half the size. According to 
 Mr. C. B. Brown, the Accoway tribe seem to be the best 
 Apx'on makers. 
 
 " I started from the VVoynamou landing and walked to tho 
 village, which consisted of four or five palm-thatched houses 
 with open sides, situated in a clearing, in which dwelt some 
 Ackawoise Indians. 
 
 *' I was struck with the manner in which the men of the 
 ])lace wore their hair, allowing it to grow long and fall far 
 
m 
 
 KXILANATOUY INDEX. 
 
 m'j 
 
 below their shoulders, giving them a very feminine appear- 
 ance. The heail man had his hair bound round at the end 
 with cord in such a manner that it stuck out at a right angle 
 behind. The old fellow's corporation projected in front quite 
 as far as his pig- tail did behind, and gave him a most ridicu- 
 lous appearance. Neither men nor women had eyebrows, 
 having, according to Indian custom, plucked them out. Their 
 upper lips and both sides of the mouth were tattooed with 
 blue curved lines. The dress of the women was exceedingly 
 simple, consisting only of a small cotton and bead-work 
 apron, made fast to a cotton cord round the body above the 
 
 APRON. 
 
 hips. The apron, called a * .^ueyou,' is manufactured by the 
 wearer in a most ingenious manner, each bead being slipped 
 on the cotton thread in its proper place as it is being woven. 
 The patterns of these dresses are marked out with red, blue, 
 and white beads, and they generally have an ornamented 
 border." 
 
 AuAs. — These magnificent birds aie popularly known as 
 Macaws. The Blue and Yellow Macaw is scientifically termed 
 Ara ararauna, and the Scarlet and Blue Macaw is Ara macao. 
 Both birds are tamed by the natives, and kept about their 
 houses, in company with spider monkeys and other pets. 
 Their long and splendid tail feathers are much used in the 
 
 I'l 
 
 ! 1 1 
 
 iM' 
 
 I ' : Mil 
 
 •mi 
 
 m 
 
 
 N 
 
 wi 
 
 i- ^ 
 
 if 
 
 \l 
 
 H 
 
 n 
 
 mi 
 

 
 866 
 
 KXPLANATOUY INDKX. 
 
 manufacture of he id chesaes, some tine examples of which are 
 in Waterton's must utn. 
 
 AuM.VDii-L() {DasypuH se.rcinctHs). — There are several species 
 of Armadillo, hut this is the one mentioned by Waterton. 
 The Nine banded Armadillo is called Cuchicime. 
 
 Arrows, Elow-gun. — These wonderful little Arrows are 
 about nine or ten inches in length, and not thicker than a 
 lady's steel knitting-ntedlo. Indeed, these Arrows (without 
 
 I'lT- I - — • i^^^-^ 
 
 f 
 
 ltl.l>W-(il'N" AI(U()W6 STia.NC; 
 
 the poison) would answer very well for knitting. The leaf- 
 rib of the coucourite palm is, when dry, very hard and elastic, 
 although like steel, it combines brittleness with hardness and 
 elasticity. It is also a heavy wood, for Arrows made of light 
 
KXPLANATOKY INDKX. 
 
 ;i(H 
 
 woods never Hy true. I have made imitations of tliese Arrows 
 in many kinds of wood, and found yew to be the best. Tlio 
 end which is destined to receive the poison is beautifully 
 tapered by means of tlio pirai tootli sharpener, and can, in 
 consequence of its hardness, ba brought to a wonderfully tine 
 point. Owing to their very small size, a hundred or more can 
 l;e carried in the quiver, and they are arranged as shown in 
 the illustration, taken from my own specimens. 
 
 The first illustration re{>resents a few of these Arrows linked 
 together as described. 
 
 1 may mention that the .string.s at the ends of the central 
 .stick are double The Arrows are laid one by one between the 
 strings, a single turn being made between each Arrow, and 
 they are kept in their places by a coujde of sliding knots, 
 which can be moved up and down the strings. The Arrows, 
 when in their places, bear some resouiblanco to the peculiai- 
 mats used for cream chee.'«es, and perhaps a .still greater le- 
 semblance to the parallel straws once employed by milliners 
 in the operation of " gauft'ring " muslin, crape, and other line 
 textures. 
 
 lll.iiW-iaN AIlKuWS lioI.I.KLi AND lIKh. 
 
 It will be .seen that the Arrows can be rolled round the 
 stick and tied together, and if one of the Ai-rows be wanted, it 
 can easily be pushed between the spokes of the wheel without 
 disturbing the roll. 
 
 By some tribes, the wheel is not employed, but the stick 
 projects sufficiently to protect the hand, and is generally 
 forked at the top. 
 
 Now for the Arrow as piejiared for the Idowgun. The 
 wild cotton, to v hich Waterton alludes, is soft, yellow in 
 
 i s, I, 
 
:]>\A 
 
 KXr'LAXAT<»l{V 1X1 (KX. 
 
 
 I 
 
 i 
 
 colour, and fino in (ihro, but tlio fibres are too short to fit it 
 for nianufiicturo into thi'ead, and for tliis pin'poso the native 
 has recourse to another plant belonging to the genus (ioAHypiuui, 
 or to the silk-grass. 
 
 The accompanying illustration shows the Arrow when com- 
 plete. In order to make the cotton adhere better to the 
 Arrow, the latter is slightly rubbed with Kurumanni wax, 
 and some of the cotton moulded i*ound it in a sort of spindle 
 
 ni,()W(.rs AitUDW rojii'iKTi;. 
 
 shape. It is, in the thickest part, slightly wider than the 
 tube of the blow-gun, so as to le.vve no windage when pro- 
 pelled by the breath. 
 
 One of the chief dilhculties in putting in the cotton is to 
 make it exactly symmetrical, for if one side be in the least 
 longer than the other, it cannot fly straight. The cotton is 
 tied on in a sort of chequer patten, with a very fine fibre of 
 silk grass, and soiuo time is occupied in doing it properly. 
 But, a native cares nothing about time, especially when a 
 faulty Arrow may cost him a meal. 
 
 He generally prepares about ten or twelve for the blow- 
 gun, leisui'ely fitting and tying the cotton while lying in his 
 hammock. These are kept in the quiver for present use, the 
 others forming a reserve. Should he miss his aim while shoot- 
 ing, he does not search for his Arrow, but takes another. 
 
 The force with which these tiny weapons can be propelled 
 is really wonderful. They can kill birds on the tops of lofty 
 ti'ees where they are far beyond the reach of a shot gun. I have 
 sent them a hundred yards, and have no doubt that a Macoushi 
 Indian could project them much farther. They are perfectly 
 noiseless, and next to invisible, so that when one bird is shot, 
 another on the same tree will not take alarm. 
 
 Arrows, Fish. — These are of considei'able length, some of 
 
KXIM.ANATOUY IN'DKX. 
 
 809 
 
 thorn boinj» six feet long. How they avo made and used ia 
 shown by Mr. C. li. Brown in his work on (Jiiiana. 
 
 ♦'Our men frequently shot deop}»odiod, silvery ociiled fish, 
 called ciirtahac {TelrajonQplcrna /(tfiis), and another somewhat 
 similar fish called pacu (J///A'/»'."» pw u). The male of the 
 latter has a large deep body of a dull goldtish colour, while 
 the female is of a dull bluish brown. Their flesh is firm, and 
 of a peculiarly pleasant nutty flavour. They have even rows 
 of teeth shaped like the incisors of mammals. These fish 
 browse upon the aquatic plants in the swift running water, 
 and are easily shot, when feeding near the surface, by those 
 skilled in the use of the bow and arrow. 
 
 "The bows that our men used for this purpose had been 
 procured from the Indians of the interior. They were very 
 lonff and straight, and made of hard nd wood. The shafts 
 
 Hsn AKIIiiW 
 
 of the Arrows were made of long straight reeds— the flower- 
 stalks of the wild cane— tightly bound to a short piece of 
 extremely tough wood inserted at one end, called a 'shimara- 
 sherie' or 'wire-cash.' The end of the latter fits into an iron 
 arrow-head, to which it is attached by a piece of cord in such 
 a way, that when the point pierces the fish the barbed head 
 comes off the arrow end, but still remains attached to it by 
 the cord. This allows the shaft freer play, so that during the 
 struggles of the fish it is not broken. Sometimes a light 
 cotton cord of about fifteen feet in length is made fast to the 
 wire-cash, one end being attached to the left hand of the 
 fisherman. When the fish is within a short distance of the 
 man the latter lets fiy the Arrow, and aiming a few inches 
 
 II It 
 
370 
 
 EXPLANATORY INDEX. 
 
 below the fish, to allow for refraction, strikes it in the body. 
 The fish, pierced by the barbed point whieli remains in it, 
 being now in direct communication with the man's hand by 
 means of the cord, is easily landed." 
 
 The illustration of the head of one of these arrows is taken 
 from a specimen in my collection. The * shimara-sherie ' is 
 made of letter- wood, q.c, which is weighty as well as hard, so 
 that, together with the iron head, the fore part of the arrow 
 is heavy enough to enable the weapon to penetrate the water. 
 Sometimes it is used for catching turtle, and then is shot high 
 into the air so as to fall perpendicularly on the turtle's back. 
 
 Arrow, Hog. — This weapon is quite as large as the tish 
 arrow which has just been described, and, with the exception 
 of the head, is made in much the same manner. It is used 
 for shooting wild hogs, capybara, tapirs, and the larger 
 monkeys. One of the principal objects in putting the head 
 loosely into a square, or rather, oblong hole, 's, that the shaft 
 falls to the ground, and can be picked up by the hunter and 
 used again with a fresh point. Thus, he need not trouble 
 himself to carry more than a couple of shafts, and, as a 
 Guianan native never takes any trouble that he can avoid, 
 this arrangement suits him admirably. 
 
 I once happened to say to Waterton that I supposed a 
 native could easily sit down and make a fresh Arrow, and was 
 
 ■MM 
 
 llOG-ARROW. 
 
 immediately 'pulled up' as having shown ignorance of the 
 native customs. For, as Waterton graphically put it, an 
 Indian never runs if he can walk, never walks if he can 
 sit, and never sits if he can lie down. All Arrow making 
 is done while lying in the hammock. 
 
EXPLANATORY INDEX. 
 
 371 
 
 ing 
 
 The illustration repi'esents tlie head of one of the Arrows 
 in my collection, one {ignre showing it as ready to be placed 
 on the bow, and the other as covered with its bamboo guard. 
 The native is so cautious about the vvourali, that he never 
 u-;e3 any cutting instrument when the i)oison is uncovered, 
 Icst he might scratch himself, and some of the wourali find its 
 way into the wound. He never carries the hog-arrow without 
 its guard, and if an uncovered arrow be pointed at him, he is 
 as much frightened as we should be if aiuied at with a loaded 
 riHe. 
 
 The feathers are generally those of the wild turkey. 
 
 Akkow, Eappu or Poison wood. — I cannot tind that this 
 weapon is anywhere mentioned by Watorton, although he 
 brought home many spteimens, and gave me sexeial of them, 
 describing their properties, which are very nearly identical 
 with those of wourali. IMr. (J. B. Brown gives an amusing 
 account of a controversy between three hunters. One named 
 Parmu was well ahead of the others, t>hot a wild hog with a 
 Rappu Arrow, and, knowing that it must fall, went after the 
 rest of the herd, leaving the poisoned head in the wound. 
 Presently, another hunter came up, fired at the hog, and as it 
 did not run away, thought that he had inflicted a fatal wound. 
 Hunter number three now came up, and al-o fired with the 
 same re.'^iilt. 
 
 "On the return of the men, this hog was knocked on the 
 head, the poisoned Arrow pulled out, and the aniiual brought 
 to the boat, when Ben claimed it as the result of his prowess. 
 Griffiths was astounded at this, declaring thtit he had islain it 
 himself. When, liowever, it had been .'■calded and sciaped, it 
 was found that not a single t^hot had touched it, and the 
 Arrow head formed its only wound. The subject was a sore 
 one for both men after their warm discussion, and it was fre- 
 quently brought up by the rest of the party, when the topic 
 of conversation turned on shooting.' 
 
 The .same traveller thus describe.' the manufiicture of the 
 Arrow. 
 
 \i It 1^ 
 
 • iirl 
 
 m 
 
 t 1. 
 
372 
 
 EXPr.AXATOKY IXDKX. 
 
 I 
 
 " We arrived at a smooth, ojion river beyond the Eappu 
 Rapids. The islands in these rapids and a river near by ai-e 
 80 called from the existence of a peculiar species of tall and 
 graceful bamboo which flourishes there, not being found 
 further north. Pieces of the stem of this bamboo are dried 
 and used by the Indians as Arrowheads, which are said to 
 possess similar properties to the far-famed wourali poison. 
 They split up the stem, and dry the pieces over a fire, and 
 then shape them into lance-heads, whi(!'i they fasten on the 
 ends of Arrows. Wild animals wounded by these Arrows are 
 at once completely paralyzed, and in that condition easily 
 despatched. This bamboo is tall, growing singly, and not in 
 clumps, from a mass of matted roots, like the common 
 bamboo." 
 
 The heads ai'e about six inches in length, and are shaped 
 like willow leaves. 
 
 For the quivers employed for the blow-gun and hog-arrows, 
 see ''Quivei'." 
 
 Arrow-reed {Gynueciuni Haccharivuni ,y \i is described on 
 page 369. 
 
 Arrow-root {.Uaranta arundiuaced.) This is a sort of 
 cane, with broad, branching leaves und white flowers. 
 Several species are cultivated for food. 
 
 AwMNG, Canvas. — Later travellers have proved the value 
 of Waterton's experience. Nothing is so certain to bring on 
 the dreaded fever than the drenching rain which is so common 
 in (jruiana, especially when the rain storm comes on at night. 
 As the average temperature at night rarely varies more than 
 from 7'!" to 8.5° Fahr. the traveller can sleop in the open air 
 without inconvenience, so far as regards heat and cold. Against 
 rain, however, he must guai*d, and such a waterproofed canvas 
 awning as Waterton describes is found to answer that 
 purpose admirably. 
 
lilt 
 
 EXPLANATORY INDEX. 
 
 B. 
 
 Bamboo [Baoihusa arundinacea.) -The many uses to which 
 this gigantic grass is put are too well kriovvn to require notice. 
 
 Ban'ANA {Musd sapientium.) — One of the strangest of plants. 
 It scarcely ever grows wild, and has been so long propagated 
 by cuttings that it has lo&t the power of seeding, except in 
 the Andaman Islands, whei'e it does produce seeds, some of 
 which have been sent to England by the late C Kingsley, 
 who thus writes of it : — 
 
 " Most beautiful it is. The lush, fat, green stem ; the 
 crown of huge leaves falling over in curves like those of 
 
 H\NANA, 
 
 human limbs ; and below the whorls of green or golden 
 fruit, with the purple heart of flowers dangling behind them; 
 and all so full of life that this splendid object is the product of 
 nfev) months. 
 
 ; ! 
 
 
 
 JV 
 
 ifi 
 
 
 m 
 
 jii 
 
 i 
 
 if 
 
 I 
 
374 
 
 EXPLANATORY INDEX. 
 
 \^ 
 
 " I am told that if you cut the stem off at certain seasons, 
 you may see the young leaf actually move upwards from within, 
 and grow before your eyes ; and that each stem of plantain 
 will bear from thirty to sixty pounds of rich food during the 
 year of its short life." 
 
 Another plantain (Jfusa j)<^^'<^(J'isa{ca) bears large fruits 
 which require to be cooked, while the fruit of the banana is 
 eaten raw. 
 
 Bkll-hihd. — See " C.impanero." 
 
 BKTK-iiour.i:. — This horrible little pest much resembles our 
 harvest bug in colour, size and habits. It is a minute species 
 of tick, belonging to the genus Leptus, and causes the most 
 violent irritation when it attacks a human being. As its 
 principal haunts are in grassy spots, the feet are generally 
 exposed to its onslaught, but it will make its way between 
 the collar and neck, or on the wrists, and, being so small, 
 scarcely larger than the dot of the letter i, it is seldom dis- 
 covered until the mischief is done. 
 
 Indeed, there seems to be no mode of evading it. Mi'. C. 
 ]^. Brown mentions that he once thought that he had found a 
 new species of iSlaam, both sides of the head being marked 
 with a scarlet [)atch. But, on examining the bird closer, he 
 found that the scu-let patches were nothing but clusters of 
 the Brte-rouge. 
 
 BiSA MuNKKV {lirachyurns satanns.) — 
 
 Sometimes this monkey is called the Bearded Saki, or 
 Cuxio. The male is a much deeper black than the female. 
 It seems to be more solitary in its habits than the generality 
 of its tribe, and to live rather in pairs than companies. It 
 seems fearful of wetting its beard, and, when it drinks, scoops 
 up the water in the hollow of its hand. 
 
 The name is sometimes spelled Beeshu. 
 
 Bl(»w GUV. Sometimes called Pucuna. The beautifully 
 perfect specimen given to me by "Waterton is exactly eleven 
 feet in length, and is yet so light that it is only a little 
 more than a pound and a half in weight. For the various 
 
EXPLANATORY INDEX. 
 
 375 
 
 
 materials from which it is made see " Wourali," "Ourah," 
 "Samourah," " Silk-graas," "Aeuero," itc. 
 
 The buck-sight is made in a very ingenious manner. *Se« 
 *' Acouri." 
 
 Slight as is the pucuna it is much stronger than it looks, 
 and when held to the lips scarcely curves at all. The mode 
 
 ''\'.i' 
 
 nrSA MOS'KF.V. 
 
 e 
 
 S 
 
 of holding it is rather curious, and was taiight me by 
 Waterton. 
 
 Ai'tists generally represent it as held to the mouth with 
 the right hand, and resting upon the left hand, the arm being 
 stretched to its furthest extent. This is all wrong. When 
 taking aim the native hunter places his left elbow against 
 his side, with the palm of the hand turned upwards, just as 
 if he were handling a violin, and grasps the tube about eight 
 inches from the end. The right hand is next placed on it, 
 with the palm downwards, so that when the tube is grasped, 
 about two inches project from the hands. 
 
 Tt will then be found that merely by bending the body 
 backwai'ds the tube can bo raised or lowered with scarcely 
 
 ./ 
 
 ^ »i 
 
;{7r» 
 
 KXPI.ANATORY INDEX. 
 
 any exertion, the left elbow supporting its weight, and the 
 muscles of the loins supplying the power. The Macoushies 
 are very fastidious as to the straightness of 
 the pucuna, which thay never allow to lean 
 against a tree, but always suspend by a 
 loop just below the foresight. When on foot, 
 they never carry the pucuna horizontally, as 
 we 'trail' arms. ])ut hold it perpendicularly, 
 and keep it as upright as the vegetation will 
 ])ermit. In fact, the position of tiie holder is 
 almost identical with that of a soldier when 
 ' ordering ai*ms.' 
 
 There is a shorter and much heavier Blow- 
 gun used by some tribes. It has no ourah, 
 and is made of a young palm, first split, then 
 hollowed, then put together and bound spirally 
 with vegetable fibre. 
 
 Boat-Bill. — See " Crabier." 
 
 BocLORA {Ih'ogonmelanopterus). — This species 
 
 of Trogon seems to have a wider range than 
 
 the generality of its kind. Mr. C. B. Brown 
 
 was fortunate enough to have an opportunity 
 
 I J of watching the bird and her young. 
 
 mL " Close to my tent at that place there was 
 
 ..^^ the nest of a common dark-coloured trogon, 
 
 ^V called Bowclora, which merely consisted of a 
 
 Wf hole scooped out of a large, round termite's 
 
 f nest, built on the stem of a tree. In it were 
 
 two young ones which kept up a most doleful 
 
 whistling all day, while the mother sat on a 
 
 tree-bough near by, being afraid, from our 
 
 near proximity, to approach them." 
 
 From examination of the stomach of the 
 l3oclora, its food seems to be of a rather mixed nature, portions 
 of fruit and the remains of insects, chiefly the mantis, being 
 found in it. 
 
 llI.O\V-(iLN 
 
KXPLANATOUY INDEX. 
 
 Bois Immortelle {Erythriaa umhrosa). — TLe fvjlluvvJiig de- 
 scription of this splendid tree is given by C. Kiiighley ; — 
 
 " Among the young cacao-trees, at some twenty yards 
 apart, are the stems of a tree looking much like an ash, 
 save that it is inclined to throw out broad spurs like a ceiba 
 (cotton -tree). You look up, and see that they are Bois Im- 
 mortelles, fifty or sixty feet high, one blaze of vermilion 
 against the blue sky. 
 
 "Those who have stood under a Lombardy poplar in early 
 spring, and looked up at its buds and twigs, showing like 
 pink coral against the blue sky, and have felt the beauty of 
 the sight, can imagine faintly — but only faintly — the beauty 
 of this ' Madres de Cacao ' or ( 'acao-mothers, as they call 
 them here, because thei' shade is supposed to shelter the 
 cacao-trees, while the dew collected by their leaves keeps the 
 ground below always damp." 
 
 Both scientific names of this gorgeous tree are appropriate. 
 The first is derived from a Greek word, signifying red, and 
 the second is Latin and means shady. The magnificent 
 flowers are greatly frequented by humming birds. 
 
 BoNiTO {I'elaDii/s sarda). — This well known fish seldom 
 reaches a yard in length. Its fiesh is eaten, but does not 
 agree with many persons. One of my friends told me that he 
 once had several attacks of virulent nettlerash, before he found 
 out that he had been eating Bonito fies-h. At the time, I was 
 suffering from violent rash caused by handling the beautiful 
 " canker-worm," i.e. the caterpillar of the gold-tailed moth, 
 and on seeing my face and hands, my friend thought that I 
 had been eating Bonito to find out what it was like, and was 
 paying the price of curiosity. I never saw the flesh, but 
 am told by those who have often eaten it, that it is very red 
 and has much the look of fresh meat. The fish maybe recognised 
 by the longitudinal brown streaks on the belly. 
 
 BoLRADi. ISee " Toucan." 
 
 BiLKTUE {Mbnvsops Balata). — Sometimes called Burueh. 
 One of the giant-tree.s of (Juiana, with a reddish brown trunk, 
 
378 
 
 EXPLANATORY INDEX. 
 
 and rising to the li 'iglit of a Imndred and fifty feet. From 
 its bark, when cut, exudes a milky white sap, which, wlien 
 congealed, producoH a gum partaking of the properties of 
 caoutchouc and gutta percha. Tuis gum appears not to 
 have reached the English market. 
 
 Once in every five years it beirs an abundant store of ex- 
 cellent fruit, much like an Knglish plum, and the tree is so 
 gigantic that a single trunk will produce a log of a yard 
 square and a hundred feet in length. The wood is superior 
 to oak except for shipbuilding, sea-water injuring its other- 
 wise incorruptible texture. Like the cotton-tree (q.v.), the 
 trunk throws out spurs of great size. 
 
 The timber merchants call it Bullet, or Bully-tree, and as 
 the wood is not injured by weather, it is used for house- 
 frames, posts, and shingles. Another kind of Bullet- wood 
 is procured from the Sapota IMulleri. 
 
 BusiLMASTKR. — See " Couanacouchi." 
 
 Busii-RoPK. — This is a general name applied to a vast 
 number of climbing plants, which grow in the remarkable 
 way described by Waterton. They are also known by the 
 popular name of Liana. Schomburgk gives the following 
 description of one species of Bush Hope : — 
 
 *' As we forced our way through the wood, we were greeted, 
 from time to time, by the finest perfume, which we traced to 
 a liana, or creeper, and one of the Bush-ropes of the colonists. 
 Its sweet-smelling plant was Schndla brachystachya, with 
 white flowers, of which the largest patch was spotted with 
 pink, growing in voluminous clusters, its stem twisted and 
 contorted in so remarkable a manner, as well to deserve the 
 name of Bush rope. 
 
 " To describe the various ways in which these twists and 
 contortions take place would be difficult. Sometimes the stem 
 is as delicate as a ribbon, while at others it presents a bundle 
 of stems so closely twined together, as to make it no easy 
 matter to separate them with an axe." 
 
 Home of these Bush-ropc's are very pliant, will bear any 
 
EXPLANATORY IXDEX. 
 
 379 
 
 amount of twisting, and are much used for h')u.so building. 
 They go by the name of Mainouril. Others are so brittle 
 that they snap if tied in knots or twisted. Stedman compares 
 the ai)poaranee of the tropical forest with its tall tree trunks 
 and interlacing bush ropes to that of a fleet at anchor, a 
 comparison afterwards employed by Waterton. 
 
 Cadhagi:, Mountain [Orcodoxa oleracea). — One of the most 
 beavitiful of the Palms. The topmost shoot is popularly 
 Called the " cabbage," and is a very excellent vegetable for 
 the table. As the palm dies when the central shoot is de- 
 stroyed, the usual i>l in is to cut it down, knowing that theie 
 are plenty of others ready to take the place of those which 
 are destroyed. 
 
 Kingsley well described his first sight of the Cabbage 
 Palm :— 
 
 " Grey pillars, which seemed taller than the tallest poplars, 
 smooth and cylindrical as those of a Doric temple, each 
 carrying a ilat head of darkest green, were ranged along 
 roadsides and round fields, or stood in groups or singly, near 
 engine works, or towered above rich shrubberies which 
 shrouded comfortable country houses. It was not easy, as 
 I have said, to believe that these strange and noble things 
 ■were trees ; but such they were. At list we beheld, with 
 wonder and delight, the pride of the West Indies, the 
 Cabbage palm — Palmistes of the French settlers — which 
 botanists have well named Oreodoxa, the 'glory of the 
 mountain. ' 
 
 '* We saw them afterwards a hundred times in their own 
 native forests, and when they rose through tangled masses of 
 richest vegetation mixed with other and smaller species of 
 palms, their form, fantastic though it was, harmonized well 
 with hundreds of forms equally fantastic. But here they 
 seensed, at first sight, out of place, incongruous, and artificial, 
 standing amid no kindred forms, and towering over a cultiva- 
 tion and (uvilisation which might have been mistaken, seen 
 from tlie sen, for wealthy farms along some English shore. 
 
 A 
 
■ 
 
 380 
 
 KXPLANATOUY INDKX. 
 
 (fladly would we have gone on shore, were it but to have 
 stood a while under those Palmistes." 
 
 It is the custom that when a spot has been cleared for the 
 purpose of building a house, a few Cabbage palms are left 
 standing round it. 
 
 Camoudi {Ennedes niari'min). — One of the giants among 
 snakes, more generally known by the title of Anaconda. It 
 is sometimes called Iluillia. The snake is generally found 
 near water, and is apt to be dangerous when large. Mr. C. B. 
 Brown remarks that whenever a Camoudi is killed, two king 
 vultures, q.v., will come and take possession of it. Waterton 
 noticed the same fact. When gorged, it is in the habit of 
 lying coiled up near the water until it has digested its meal. 
 
 In this state it is so motionless that it might easily be 
 mistaken for a log of wood. Indeed, Waterton mentions in 
 one of his essays, that a negro committed the error of sitting 
 down on one of these snakes, taking it for a fallen tree, and 
 was only undeceived by the snake moving away. 
 
 Kingsley mentions a case where four young ladies were 
 bathing in a lagoon, and one was seized by a Camoudi from 
 behind. Thinking that one of her sisters had caught her 
 dress in play, she felt no alarm until she saw her three sisters 
 on the bank, and found that she had been seized by a snake. 
 The three girls courageously dashed into the water and tried 
 to drag her away from the snake. Fortunately, the reptile 
 had only caught her bathing dress, which, being made of thin 
 cotton, gave way, and she escaped safely to land. 
 
 Campanero {Arajnniga alba). — No one has described the 
 singular cry of this bird as Waterton has done. 
 
 It is one of the great tribe of the Chatterers The horn like 
 projection from the base of its beak has in all probability 
 much to do with the resonance of the cry. The accompanying 
 illustrations show the two positions of the 'horn.' The head 
 of the bird gives the horn as it appears while the bii'd is 
 tolling its wonderful bell, and the full figure depicts the 
 flaccid state of the horn while the l)ird is at rest. 
 

 KXPI.ANATOUY INDKX. 
 
 3rtl 
 
 Capivi, or CoPAiUA Gim. — This important gum is produced 
 from several species of Copal/era {see " Purple heart"), and is 
 obtained by making incisions in the branches. 
 
 CAMPANIiHll TOI.I.INCi. 
 
 C'APRiMULfjrs. — The Latin term for Goatsucker. This is 
 an example f)f Waterton's lax nomenclature, the Whip-poor- 
 Will (q.v.) mentioned in the same sentence being also a 
 (Joatsucker. 
 
 lAMl'A.NKHll «! I.K.N r. 
 
 Caracara. — Evidently a Liantasse, probably belonging lo 
 the genus SchneMa. 
 
 Cassava {Jatropha manihot). — This most useful plant is 
 sometimes called ^Manioc, and is to the Guianan natives what 
 
 m 11 
 
 
 m 
 
882 
 
 KXPLANATUUY INDKX. 
 
 m 
 
 corn is to uh. It is a tall, unbranclied plant, grow- 
 ing irregularly and knotted at intervals, and having 
 leaves with a purple gloss. 
 
 The root \n the portion that is caton, and it is 
 scraped down on a hoard stuck full of sharp (lint or 
 other stones, and called by the name of Tuniarrie. 
 It then looks just like horseradish as brought to 
 our tables, but is filled with a poisonous juice. In 
 order to extract this juice, the scraped cassava is 
 forced into a long, narrow basket called a matappi, 
 and made exactly on the principle of the ' Siamese 
 links ' which were once popular as toys. The ma- 
 terial of which it is made is a species of Calathea. 
 
 When the matuppi is full, it is scarcely half its 
 length when empty, but is more than double its 
 thickness. It is then hung to a branch of a tree or 
 to a beam of a house, an earthen pot is placed under 
 it, and a heavy weight is tied to the lower end. The 
 weight of the stone causes the matappi to increase 
 in length, but to diminish in thickness, thus exert- 
 ing a powerful pressure on the cassava, and squeez- 
 ing out the juice, which runs through the interstices, 
 and so down the matappi into the pot. 
 
 The dry CJas.-iava is then removed from the matappi, 
 rubbed through a basket-work sieve, formed into 
 Hat circular cakes about two feet in diameter, and a 
 quarter of an inch in thicknes.s, .and bsiked upon a 
 
 Hat, heated stone or 
 plate of iron. 
 
 INIeanwhile, the poi- 
 sonous juice has been 
 kept out of reach of 
 children, poultry, &c., 
 and, on being boiled, 
 and favoured with 
 red pepper or capsicum becomes the well-known cassa- 
 
 CASSAVA I'KI^S. 
 
 <'AKSAVA liliWI,. 
 
KXPL.WATtJUY IN'DKX. 
 
 m\ 
 
 rt'pp or pepper not of the AN'e.st Jmlies. Tlio pot is nevor 
 cleaned, mo that, as" it i.s yvvy thick, very woft, iind very 
 parous, it iil).sorl»H the juices. When ( "as.sava bread is eaten, it 
 is generally dipped in the cassareep, which often contains 
 pieces of meat, Ac, and which, when the palate has become 
 accustomed to the inordinate amount of red pepper, is not 
 only nourishing, but appetizing. A new comer, however, will 
 run the risk of starving altogether, for the native cook is so 
 very heavy-handed with her red popjjer, tliat the lips and 
 tongue of a novice are scorched as if with red-hot iron. 
 
 The illustration of the matappi and cassareep pot is taken 
 from si)ecimens in my collection. A considerable amount of 
 Cassava is consumed in this country under the names of 
 tapioca and semolina. 
 
 Cassiques. — These birds, of which there are several species, 
 all belonging to the genus Cassicus, are popularly called 
 Merles in Jamaica. The word is evidently the same as our 
 Merle (Lat. Menda), which distinguishes the common black- 
 bird. 
 
 According to (.'. Kingsley, the Merles are "birds the size 
 of a jackdaw, brown and yellow, and mocking-birds, too, of 
 no mean ability. The pouches (nests) two feet long and more, 
 swayed in the breeze, fastened to the end of the boughs with 
 a few small threads. Each had, about half way down, an 
 opening into the round sac below, in and out of which the 
 merles crept and fluttered, talking all the while in twenty 
 different notes. 
 
 " Most tropic birds hide their nests carefully in the bush : 
 the merles hang theirs fearlessly in the most exposed situa- 
 tions. They tind, I pre.sume, that they are protected enough 
 from monkeys, wild cats, and gato-melaos (a sort of ferret), 
 by being hung at the extremity of the bough. So thinks 
 M. Lestaud, the accomplished describer of the birds of 
 
 Trinidad." 
 
 Some writers call the Merles by the name of mocking-birds. 
 
 (Jastok-oii- {h'iebins co)>ntiunis). — Another of the plants 
 
 ''H-: 
 
■M^ 
 
 EXPLAXATOIIY IX DEX. 
 
 imported from the Old World. It belongs to the great tribe 
 of the Euphorbise or Spurge^ of which our common milky- 
 weed is a familiar examjde. 
 
 From its seeds is obtained the Castor-oil of commerce. The 
 best, called ' cold-drawn ' oil, is procured by simple pressure. 
 Tlie plant is sometimes called Paliiia Christi, because its 
 leaves look something like widely-spread fingers. 
 
 The Romans gave it the name of llicinus, because the seeds 
 bear some resemblance in shape to sheep-ticks or 'ricini.' 
 They procured it originally from Egypt, where seeds may 
 still be found in the tombs. 
 
 Cayman {Chiimpsa nigra). This is the animal which 
 Waterton so brilliantly captured. There are many of tho 
 crocodiles and alligators in North and South America, and 
 in habits they seem to be much alike. 
 
 All have the peculiar way of attacking animals on land by 
 knocking them into the water with a blow of the tail, smd 
 
 I'AVMAN. 
 
 carrying them oH before they can recover from the effects of 
 the blow. Sometimes thoy have been known to attack canoes 
 in this manner. 
 
 They all possess a most abominablo musky smell, " floating," 
 jis Mr. C. B. Brown says, " like a deadly miasma round our 
 camp, and finding its way even to our palates." Then, all of 
 them are in the habit of emitting loud, bellowing noises, 
 especially at night, so that they make themselves as objection- 
 able to tlip ears as to the nostrils. 
 
 /• 
 
 
lour 
 I of 
 Ises, 
 ion- 
 
 /! 
 
 / 
 
 EXrr.ANATORY INDKX. 
 
 385 
 
 The hook which was used by Waterton was engraved in the 
 original edition of the WanderiiKjn, but F am sure that the 
 
 CAVMAN HOOK. 
 
 draughtsman who drew it could not have seen it. The instru- 
 ment itself is in Waterton' s museum, and I have here sub- 
 stituted my own sketch of it. 
 
 The four prongs are not barbed, but are shai'ply pointed, as 
 seen in the illustration. They are fiattish, and very tough, 
 as they need be, for they are bitten and cracked all over by 
 the teeth of the cayman. The prongs aie kept in their 
 diverging position by wooden pegs driven between them, and 
 the whole instrument is thus made so elastic that it can be 
 compressed by a strong grasp of the hand, and then springs 
 back again to its original form. So, when compressed by the 
 entrails of the aconri, which were Avrapped round thein, the 
 instrument would slip easily down the cayman's throat, and 
 then expand on being swallowed. 
 
 Chameleon. — One of the Anolis Lizards, probably the 
 Red-throated Anolis (Anolius hnllatis), which are active, 
 chase flies upon trees, and are changeable in colour when 
 excited. The true Chameleons exclusively inhabit the Old 
 World, and are much too sluggish to chase insects. There 
 are several species of Anolis in Cuiana. 
 
 CnicJOE, sometimes spelled JificiEU, or TsciiiKO (Pulex pene- 
 trans.) — Tiny as it may be, the little Chigoe is one of the most 
 detested plagues of the West Indies. To all appearances it 
 resembles our common domestic flea, but it has fortunately not 
 yet become acclimatised in any part of l"]ur()jte. There is 
 
 ii: 
 
 I «i 
 
 I! 
 
 ^il 
 
 '[ 
 
 4. 
 
 '0. 
 
 . m 
 
li ' 
 
 386 
 
 EXPLANATORY INDEX. 
 
 scarcely a traveller in the West Indies who writes with the 
 least patience about this more than troublesjome insect. There 
 is some credit in facing a rattlosnako, killing a jaguar, or in 
 braving the many dangers of tropical travel, but there is 
 none in becoming the victim of a Ilea, though the apparently 
 insignificant enemy may, unless its attaciks be propei'ly re- 
 pelled, cause the loss of a limb, or oven of life. 
 
 In one of his essays Waterton has extondod the information 
 given in his Wanderings. 
 
 " In the plantations of Guiana tliere is generally an old ne- 
 gress known by the name of Granny, a kind of Junonis anus, 
 who loiters about the negro yard, and is supposed to take 
 charge of the little negroes who are too young to v> jrk. 
 Towards the close of day you will sometimes hear the most 
 dismal cries of woe coming from that quarter. Old Granny 
 is then at work grubbing the Chigoe nests out of the feet of 
 the sable urchins, and filling the holes with lime juice and 
 cayenne pepper. This searching comi)ound has two duties to 
 perform ; firstly, it causes death to any remaining Chigoe in 
 the hole ; and secondly, it acts as a kind of a birch rod to 
 the unruly brats, by which they are warned, to their cost, 
 not to conceal their Chigoes in future ; for, afraid of en- 
 countering old Granny's tomahawk, many of them prefer to 
 let the Chigoes riot in their Uosh, rather than come under 
 her di.ssecting hand. 
 
 "A knowing eye may always perceive wlion the feet of ne- 
 groes are the abode of the Chigoe. They dare not place their 
 feet firmly on the ground, on account of the pain which such 
 a position would give them ; but they hobble along with their 
 toes turned up ; and by this you know that thoy are not 
 suffering from tubboes (a remnant of the yaws), but from the 
 actual depredations of the Chigoes, which have penetrated 
 under the nails of the toes, and then formed sores, which, if 
 not attended to, would, ere long, bt^conie foul and corroding 
 ulcers. As I seldom had a shoe or stocking on my foot from 
 the time that I finally left tlio sea const in I HI 2, the Chigoe 
 
EXPLANATORY INDEX. 
 
 3«7 
 
 was a source of perpetual disquietude to me. I found it 
 necessary to examine my feet every evening, in order to 
 counteract the career of this extraordinary insect. Occasion- 
 ally, at one overhauling, I have broken up no less than four 
 of its establishments under the toe- nails, 
 
 " In 1825, a day or two before I left Guiana, wishful to try 
 how this puny creature and myself would agree during a sea 
 voyage, I purposely went to a place where it abounded, not 
 doubting but that some needy individual of its tribe would 
 attempt to better its condition. Eie long a pleasant and 
 agreeable kind of itching under the bend of the great toe 
 informed me that a Chigoe had bored for a settlement. In 
 three days after we had sailed a change of colour took 
 place in the skin, just at the spot where tlie Ch'goe had 
 entered, appearing somewhat like a blue pea. By the time 
 we were in the latitude of Antigua my guest had become 
 insupportable, and I saw there was an immediate necessity 
 for his discharge. Wherefore, I turned him and his numerous 
 family adrift, and poured spirits of turpentine into the cavity 
 which they had occupied, in order to prevent the remotest 
 chance of a regeneration. 
 
 " The Indian and negro wenches perform the operation of 
 extracting Chigoes with surprising skill. They take a pin, 
 and, by a very slow process, they lay the part bai'e, and con- 
 trive to work quite round the bag which contains the Chigoe 
 and its offspring. As soon as this has been effected, they 
 turn the bag out, whole and uninjured, by which means none 
 are left in the hole to form a new colony. For my own pai"t, 
 I never trouble these gentle operators, although I have looked 
 on many a time, and admired their exquisite skill, while they 
 were fingering the toes of my acquaintance. 
 
 " Once, however, I had it not in my power to be my own 
 surgeon, and on that occasion a faithful old negro performed 
 the friendly office. I was descending the Demerara, with an 
 inveterate tertian ague ; and I was so much exhausted by 
 sitting upright in the canoe that I had no sooner got ashore 
 
 c c 2 
 
 li' V' 
 
 i u . 
 
 V :::!hi, 
 

 liij 
 
 IEh' 
 
 BHi' 
 
 illll 
 
 if Hi 
 
 
 388 
 
 EX PI. AX ATOR Y INDEX. 
 
 at the Indian hut than I i y down on the ground at full 
 length. Sickness had pressed so heavily on me that I was 
 callous to the well-known feeling which the Chigoe causes. I 
 was quite unconscious that there were nine thriving nests of 
 Chigoes in my back, until one was accidentally observed by the 
 old negro ; and this led to the discovery of the rest. I handed 
 him my penknife, and told him to start the intruders. Sick 
 as I was, I wished an ai-tist were present at the operation. 
 The Indian's hut, with its scanty furniture, and bows and 
 arrows hanging round ; the deep verdure of the adjoining 
 forest ; the river flowing rapidly by ; myself wasted to a 
 shadow ; and the negro grinning with exultation as he showed 
 me the Chigoes' nests which he had grubbed out, would have 
 formed a scene of no ordinary variety. 
 
 '* Dogs are often soi'ely tormented by the Chigoe ; and they 
 get rid of them by an extremely painful operation. They 
 gradually gnaw into their own toes, whining piteously as 
 they do it, until they get at the Chigoe's nest. Were it not 
 for this singular mode of freeing themselves from the latent 
 enemy, dogs would absolutely be cripples m Guiana." 
 
 In Mr. Brown's Camp and Canoe Life in Guiana there is 
 an anecdote which well expresses the terrors which this tiny 
 plague can inspire into Europeans as well as natives. 
 
 " Just then Ben returned from a voyage of discovery along 
 the path leading away from our camp, with a joyful expres- 
 sion of countenance betokening good news. ' Why, sir,' he 
 said, ' there is a good house not far off, and it ain't got no 
 one in it.' I immediately went with him to see it, and, plod- 
 ding through mud and water, climbed a slight eminence to a 
 large open shed, situated in the midst of a clearing. I halted 
 close to it, and proposed that he should take a look at the 
 floor for insects. He walked boldly in, hesitated, looked 
 down at his bare feet, and, exclaiming ' Jiggers and dog 
 fleas full da here I ' fled precipitately. I may here mention 
 that the jigger or Chigoe is a small flea that burrows under 
 the toe-nails, where it forms an egg-bag almost the size of a 
 
L-e IS 
 i tiny 
 
 along 
 pres- 
 he 
 ot no 
 plod- 
 to a 
 lalted 
 the 
 oked 
 dog 
 ntion 
 nder 
 of a 
 
 RXPLAN ATO 1{ Y IN DEX. 
 
 880 
 
 small pea, wliich has to be extracted with a needle, an opera- 
 tion giving a considerable amount of pain. 
 
 " I remarked to Ben in a casual way, * Surely the natives 
 have some way of getting rid of these insects 1 just think 
 the matter over in your mind, and don't lose sight of the 
 fact that it depends upon you whether we sleep in the moist 
 swamp to-night, or under this good roof.' 
 
 " He proposed two ways of performing the operation, one 
 was to ' bun tliem,' as he called it, and the other was to 
 sweep them out. I seized upon both ideas ; we would burn 
 them first, and sweep out their roasted carcases afterwards. 
 
 ** Fortunately there was a heap of dry palm-leaves in the 
 house, of which a few bunches were made into brooms, and 
 the rest scattered over the floor. If any one had been looking 
 at us from a short distance when performing this operation, 
 he would have come to the conclusion that we were both de- 
 mented, from the way in which we skipped about, darting 
 out every now and then to brush ofE our enemies from our 
 clothes. Then we set fire to the leaves, and had the whole 
 place covered with flames a foot in height — a grand and satis- 
 factory though short-lived flare-up. At this stage of the 
 proceedings we were reinforced by Pedro and Vincente, who, 
 on the subsiding of the flames, swept out the residuary mat- 
 ter, composed of carcases, ashes, and dust. After the crusade 
 was over, a few hundreds of the enemy were left, but these we 
 did not mind. 
 
 "It is a curious and sad fact, as I have always found to 
 my cost, than when Indians leave their houses for a time the 
 dust on the floors becomes infested with jiggers and fleas. The 
 former come from the eggs dropped from the jigger egg-bags 
 in the toes of both Indians and their dogs, and the latter 
 from eggs dropped by the fleas of the dogs," 
 
 The " nest " of the Chigoe is in reality the swollen body of 
 the female, which contains the eggs, and becomes as large in 
 proportion to the head, thorax, and limbs, as does that of the 
 queen termite. In common with most noxious insects, such. 
 

 IIhI 
 
 
 
 1 
 
 ■Eli' 
 
 
 1 
 
 
 
 1 
 
 
 1 
 
 1 
 
 
 1 
 i 
 
 390 
 
 EXPLANATORY INDEX. 
 
 for example, as the mosquito and the wasp, the females are 
 the only aggressors. The male mosquito cannot bite, nor the 
 male wasp sting, nor the male Chigoe form its irritating 
 nest. 
 
 There is now befoi-e me one of those female Chigoes, with 
 the abdomen fully swollen and full of eg^s. It was sent to 
 me by a resident in the West Indies, who was kind enough 
 to allow it to inhabit his toe until it was sufficiently deve- 
 loped, and then removed it and preserved it for me. The 
 skin of tlio swollen abdomen is ve.*y tough, so that it can 
 easily be turned oub of the hollow which it has formed. 
 The size of my spe^im^n is as neirly as possible equal to 
 that of an ordinary sweeb-pei. In some places the Chigoe is 
 called Chicorine. 
 
 Cinnamon (Clnnamomum ZeyJanicum). — Only the bark of 
 the young shoots is used. Ceylon is the ch'ef country of tha 
 cinnamon. It is allied to the common laurei. 
 
 Clove. — The cloves of commerce are the unexpanded 
 flowers of a plant, known scientifically as Garyophyllus 
 aromaticus. It is one of the myrtle tribe. The name clove 
 is a corruption of the French name clou, the dried flowers 
 bearing some resemblance to a nail. 
 
 Cocoa-nut Palm (Cocos nucifera). — Essentially a sea-side 
 ti'ee, and apparently not a native of the West Indies, but im- 
 ported, either by man or by the wind and waves. It grows 
 freely in India and the South Sea Islands, as well as in the 
 West Indies, and leaohes a height of a hundred feet. 
 
 The nut grows in a very curious manner. When it has 
 fallen, one of the three holes gives way to a shoot, which strikes 
 out a root, piercing into the ground, while it still retains its 
 connection by a sort of cord, with the nut from which it draws 
 its nourishment, until it is strong enough to obtain the whole 
 of its sustenance from the ground. See tiie 'alien nuts in the 
 foreground of the illustration. 
 
 It has a " cabbage " like that of the cabbage palm, and is 
 sometimes cut down for the sake of obtaining this vegetable. 
 
side 
 
 im- 
 
 rows 
 
 the 
 
 d is 
 Lblo. 
 
 EXI'LANATOKY IXDKX. 
 
 3U1 
 
 
 C'O(!0A PALM AND POCAl.. 
 
 m 
 
 'M 
 
 in 
 
 i 1 ' 
 
 
 
 
 
 
 , 
 
 •1 
 
 , i 
 
 ill 
 
392 
 
 EXPLANATORY INDEX. 
 
 Toddy is simply the fermented juice of the tree ; coir is made 
 from the husk, which is also cut up into scrubbing brushes for 
 household use ; and the " porcupine wood " of the cabinet- 
 maker is obtained from the dense and heavy wood near the 
 root. 
 
 The long groves of Cocoa palms are called cecals, and some- 
 times extend for several miles along the shore. The Cocoa 
 palm should not be confounded with the cacao, of which 
 chocolate and cocoa are made, and which is a totally different 
 plant, belonging to the genus 'Theobroma. 
 
 Cock of the Kock {liupicola aurantia). — This fine bird is 
 the largest of the JNIanakins, q.v., and on account of the beauty 
 
 ' 
 
 
 (•i)rK OK llli: l!ii( K. 
 
 of its plumage is in great request with bird stuffers. Un- 
 fortunately, the brilliant orange of its feathers is very 
 fugitive, and a stuffed bird is sure to fade unless protected 
 from the light. 
 
 I have before me a specimen of a stuffed Cock of the Rock 
 which has been exposed to daylight for sevei'al years, and the 
 colour of which has so completely faded, that the bird can 
 
 
 .■L'r.-'n'ainiiiiiw 
 
EXPLANATORY INDEX. 
 
 \v,m 
 
 only be recognized by its shape, ihe feathers having changed 
 from brilliant orange to a dull yellowish brown. 
 
 The following account of this bird is taken from Mr. C. B. 
 Brown's work on Guiana : — 
 
 " Cocks of the llock were numerous in the surrounding 
 thickets, where their sharp disagreeable cry was frequently 
 heard. 
 
 " They are so restless, jumping and flying from bough to 
 bough, and tree to tree, that Paulie, who was trying to shoot 
 them, started five, but only succeeded in obtaining one 
 specimen. He found a dancing place of these birds in a 
 thicket, the ground being beaten down quite smooth by their 
 feet ; and on visiting it early in the morning with Ben, the two 
 together succeeded in shooting two cocks and a hen bird. They 
 told me that there were numbei'S around the dancing place, and 
 that the two cocks they shot were strutting about with their 
 feathers distended, showing themselves off before the rest." 
 
 Only the adult males possess the i*ich orange plumage, the 
 females and immature males being of a dull yellow green, and 
 having but a small crest. The bird, though the largest of the 
 Manakins, is but a small one, being scarcely equal in size to a 
 Tumbler pigeon. 
 
 Coffee. — It is rather remarkable that the two chief pro- 
 ducts of the West Indies, namely Coffee and Sugar, are both 
 natives of the Old World, and have been acclimatized in the 
 New. 
 
 Coffee, as its name imparts, (Joffua Arabica, is indigenous to 
 Northern Africa, and was imported into Europe as a curiosity- 
 Not much more than a hundred and fifty years ago a single 
 layer of two slips was taken from Holland to Martinique, and 
 it throve so well that it furnished a supply for the whole 
 of the West Indies. 
 
 There is a romantic story connected with its introduction, 
 A Frenchman, named Desclieux, had charge of the plant. 
 On the voyage the vessel fell in with a series of storms, and 
 all on board were put on short allowance of water. The 
 
394 
 
 EXl'J.AXATOKY INL)p:X. 
 
 M M 
 
 heroic Frenchman divided his share of water with the (Joffee- 
 
 phmt, 
 
 " Ami Miutiuii!o loads her sliips 
 With pioiliU'e from tliose dear-saved slips." 
 
 It belongs to the useful group of Cinchonaceai. Even the 
 leaves possess many of the qualities which make the seeds so 
 useful. 
 
 Copal. See ** Locust Tree." 
 
 OouAL Snake {Tortrix Scytale). — In some parts of the 
 country this snake is made a pet, being twisted round the 
 neck like a gold and black ** torque." It is but a small snake, 
 averaging twenty-six inches in length. 
 
 CoTiNGAs. — These all belong to the group of the Ampel- 
 inse, or Chatterers. 
 
 The Pompadour Cotinga (Cot'mja pompadcara) is a singularly 
 beautiful bird, its plumage being mostly of the beautiful hue 
 which is known as pompadour, and which used to be very 
 fashionable at the beginning of the present century. The 
 feathers are splashed and sti*eaked with white, and the wings 
 are tipped with black. In size it rather surpasses our 
 starling. 
 
 The Purple-throated Cotinga is known to science as Cotinga 
 cayenna. For other remarkable Chatterers see " Bell-bird ' ' and 
 " Cock of the Rock." 
 
 Cotton. — The cotton which is used for thread and string, is 
 procured from several species of Gossypium, one of which 
 produces the cotton so largely used by ourselves. The natives 
 always have some of these bushes planted near their houses, 
 and cotton spinning goes on almost as interminably as knitting 
 or crochet among English ladies. 
 
 The mode of spinning is that which is prevalent all over 
 the world, and even in England has only lately been super- 
 seded by machinery. A wooden spindle is passed through a 
 whorl of a heavy wood, bone, or sometimes stone, and the 
 fibres attached to t. The spindle is then made to revolve, 
 thus spinning the fibres into thread. In savage countries, the 
 
KXIM.ANAT(JKY INDEX. I^l 
 
 invariable practice is, to roll tho thread ou tl\« l.u tbifH, 
 and somotimes this labour in carried on so unceasingly as to 
 cause sores. 
 
 Spindles such as have been described were found in the 
 tombs. of the lueas, and several of them came into my posses- 
 sion, together with pieces of the fabrics made from tho 
 threads spun by them. 
 
 The species which is so largely cultivated for commerce is 
 Cossifimim tricusjnduhim . 
 
 CoTTox-TiiEE, {Doiiiha,io celha). — This magnificent tree, some- 
 times called th f Silk Oottou, is among the many wonders of 
 the West Indies, and is admirably described by C. Kingsley 
 in his joyous book At Last : — 
 
 "These latter (the Ceibas) are useless as timber ; and their 
 roots are, of course, hurtful to tho sugar canes. But the ne^'ro 
 is shy of felling the Ceiba. It is a magic tree, haunted by 
 spirits. There are ' too much jumbies in him,' the negro says ; 
 and of those who dare to cut him down some one will die, or 
 come to harm, within the year. 
 
 *"In Jamaica,' says my friend Mr. Gosse, 'they believe 
 that if a person throws a stone at the trunk, he will be visited 
 with sickness, or other misfortune. When they intend to cut 
 one down, they first pour rum at the root as a propitiatory 
 offering.' The Jamaica negro, however, fells them for canoes, 
 the wood being soft, and easily hollowed. 
 
 " But here, as in Demerara, the trees are left standing about 
 in cane-pieces and pastures to decay into awful and fantastic 
 shapes, with prickly spurs and board- walls of roots, high 
 enough to make a house among them simply by roofing them 
 in ; and a flat crown of boughs, some seventy or eighty feet 
 above the ground, each bough as big as an average English 
 tree, from which dangles a whole world of lianas, matapolos, 
 orchids, wild pines with long air-i'oots or gi'ey beards ; and 
 last, but not least, that strange and lovely parasite the 
 lihvpsalis cassytha, which you mistake first for a plume of 
 green sea- weed, or a tress of mermaid's hair which has got up 
 
 'i-i 
 
 ■HilH* 1 
 

 :vjc, 
 
 KXPLAXAToIlY IXDKX. 
 
 there by mischance, and then for some delicate kind of pen- 
 dent mistletoe ; till you iiro ti)M, to your astonislimont, that it 
 is an abnormal form of cactus — a family which it resembles, 
 save in its tiny (lowers and fruit, no more than it resembles 
 the Ceiba-troo on which it ^rows ; and told, too, that, strangely 
 enough, it has been discovered in Angola— the only species of 
 the cactus tribe in the 0U\ World." 
 
 SIl.K-IOTroX-TRKll. 
 
 The Cotton-tree was a never-failing wonder to Kingsley, 
 who again writes of it in the same work : — 
 
 '* If you are all safe, your next steps probably, as you 
 struggle through the bush, between tree trunks of every 
 possible size, will bring you face to face Avith huge upright 
 walls of seeming boards, whose rounded edges slope upward, 
 till, as your eye follows them, you find them enter an enormous 
 stem, perhaps round, like one of the Norman pillars of 
 
KNPF.AXATOUY IMUCX. 
 
 IW 
 
 
 iJurbam niivo, ami just as liti^'c ; perhaps (luted, like oim of 
 William of Wykchniii's coluums at Winchester. 
 
 •'There is the stem, but where is the tree? Al)()ve the 
 green cloud. You struggle up to it, between two of the 
 board walls, but tind it not so easy to roach. IJitwcon you 
 and it, are half a dozen tough strings which you had not 
 noticed at first — the eye cannot focus itself I'apidly enough in 
 the confusion of distances— which have to bo cut through ero 
 you can pass. Sonio of thoni are rooted in the ground, 
 straight and tense j some of them dangle and wave . . the 
 wind at every height. 
 
 " What are they 1 Air-roots of wild pinos {tlUawhia), 
 or of matapolos, or of 6g.s, or of .'^^oguiues (ji/ilMemlronf 
 aiithnrtiDti, itc.) or of some other parisite? Probably: but 
 you cannot see. All you can see is, as you put your chin 
 close against the trunk of the tree and look up, as if you 
 were looking up against the side of a great ship set on end ; 
 that some sixty or eighty feet up in the green cloud, arms as 
 big as English forest trees branch off ; and that out of their 
 forks a whole green garden of vegetation has tumbled down 
 twenty i r thirty feet, and half climbed up again. You 
 scramble round the tree to find whence tlie ac/ial garden has 
 sprung : you cannot tell. The tree trunk is smooth and free 
 from climbers ; and that mass of verdure may belong possibly 
 to the very cables which you met ascending into the gieen 
 cloud twenty or thirty yards back, or to that impenetrable 
 tangle, a dozen yards on, which has climbed a small tree, and 
 then a taller one again, and then a taller still, till it has 
 climbed out of sight, and possibly into the lower branches of 
 the big tree. And what are their species ] What are their 
 families'? Who knows? Not even the most experienced 
 woodman or botanist can tell you the names of plants of 
 which he sees only the stems." 
 
 I'rom this tree is procured the Wild Cotton which has 
 already been mentioned on page 131. I believe that yet no use 
 has been found for this delicate and short yellow libre, except 
 
 ill J 
 
 li- 
 
 n * ri 
 
 .til 
 
 i!; i 
 
 a 
 
 I \ ! («, 
 
r 
 
 398 
 
 ii 
 
 EXPLANATOKY INDEX. 
 
 as stuffing for pillows and couches. The native never troubles 
 himself to remove the seeds, which are hard, spherical, nearly 
 black, and about as large as peas. 
 
 CouANACouciii (Lachesis mutus). — Popularly called Bush- 
 master, a name originally given to it by the Dutch. Some- 
 times it is called Curucuru. When living, it is as beautiful 
 as it is deadly, but the lovely prismatic colours which play 
 over the body during life are extinguished in death, and not 
 even Waterton could restore to the skin the beauty of the 
 living serpent. It is found both in trees and on the ground. 
 
 Coucourite-Palm. — There are several species of this palm, 
 all belonging to the genus Maximiliana. The most beautiful 
 
 
 COUCOl-RITE. 
 
 of them is the species mentioned by Waterton, and appro- 
 priately named Maximiliana regia. " In this plant," writes 
 Kingsley, " the pinme are set on all at the same distance 
 apart, and all in the same planes in opposite sides of the 
 stalk, giving to the whole foliage a grand simplicity ; and 
 producing, when the curving leaf points toss in the breeze, 
 
EXPLANATORY INDEX, 
 
 309 
 
 ill-''' 
 
 that curious appearance which I mentioned in an earlier 
 chapter, of green glass wheels with rapidly revolving spokes." 
 
 The leaves are sometimes twenty-five feet or more in length, 
 and their stems are triangular. When dried, they are won- 
 derfully light, strong and elastic, and are often cut into lengths 
 and imported to England as walking-sticks. When young, 
 the tree has scarcely any stem, the leaves springing almost 
 directly from the ground. These trees flourish best on sand 
 or gravel. 
 
 Stedman describes the young leaves as diverging from each 
 other like the flaming fuse of a shell. 
 
 CouGUAR (Leopardus concolor). — It is sometimes misnamed 
 the American lion, and sometimes the panther, just as the 
 jaguar goes by the name of tiger. In some places it is called 
 the deer tiger. Mr. C. B. Brown had a curious adventure 
 with one of these animals : — ■ 
 
 " One morning, whilst returning to camp along the portage 
 path that we were cutting at Wonobobo falls, I walked faster 
 than the men, and got some two hundred yards in advance. 
 As I rose the slope of an uneven piece of ground, I saw a 
 large puma [Felis concolor) advancing along the other side of 
 the rise towards me, with its no.'e down on the ground. The 
 moment I saw it I stopped ; and at the same instant it tossed 
 up its head and seeing me also came to a stand. With its 
 body half ciouched, its head erect, and its eyes round and 
 black, from its pupils having expanded in the dusky light, it 
 looked at once a noble .and an appalling sight. I glanced 
 back along our wide path to see if any of my men weie 
 coming, as at the moment I felt that it was not well to be 
 alone without some weapon of defence, and I knew that one 
 of them had a gun ; but nothing could I see. As long as I 
 did not move the puma remained motionless also, and thus Ave 
 stood, some fifteen yards apart, eying one another curiously. 
 I had heard that the human voice is potent in scaring most 
 wild beasts, and feeling that the time had arrived to do some- 
 thing desperate, I waved my arms in the air and shoiited 
 
 
400 
 
 KXPLAXATOIIY IXDKX. 
 
 
 loudly. The effect on tho tiger was ole<!trical; it turned 
 quickly on one side, and in two boundM was lost in the forest, 
 I waited until my men came up, however, before passing the 
 place at which it disappeared, in case it might only be lying 
 in ambush there ; but we saw nothing more of it. 
 
 " When returning down the portage and dragging our boats 
 over, wo saw a jaguar sitting on a log near the same spot, 
 watching our movements with evident curiosity, and although 
 the men were singing as they hauled the boats along, it did 
 not seem to mind the noise. As soon as it saw that it was 
 observed, it jumped off the log, and with a low growl made 
 off. From this I infer that the (light of my puma must have 
 been owing more to the windmill-like motion of my arms than 
 to my voice." 
 
 CouLACANAUA. — Watcrtoti dooH not give sullicient descriji- 
 tion of this snake for identilication. It is almost certainly 
 JJoa imjierator. 
 
 It is characteristic of Waterton that ho should have sat 
 down immediately after his battle with tho snake to send an 
 account of it in Latin hexameters to his old friends at Stony- 
 hurst. Mr. Edmund Waterton only lately discovered the 
 document in pencil among his father's papers. 
 
 CouRADA (Avicennia n'dida). — Sometimes spelled Courida. 
 The W^hito Mangrove of the Colonists. 
 
 The trees grow in profusion on tho shore, reaching a height 
 of fifty or sixty feet, and from their U[»per branches dangle 
 innumerable air-roots, which, when they reach the ground, 
 will strike into it, and become tlie stems of future trees. The 
 manatee is fond of browsing on the loaves of the Courada. 
 
 Crauwood {Caraj)a G'uianeimn). — Tho tree is very useful on 
 account of an oil, called 'crab oil,' which is expressed from- its 
 seeds. It serves various purposes, and gives a good light 
 when burned in a lamp. Th(5 timber of the Crabwood is in 
 great request for masts and spars. 
 
 CuiA. — A species of Trogon, similar in hize to the bocloia,, 
 buli brighter in colours, as mentioned by VVatert'»*<. 
 

 EXPLANATORY INDEX. 
 
 401 
 
 Crabier. — The Boat bill (Cancroma cancrophaja).— It is a 
 Fmall heron, having an oddly-shaped beak, much resenabling a 
 boat turned upside down. It frequents the shoi'es, and feeds 
 chiefly on the smaller Crustacea. Sometimes it is called the 
 Wallaba Bird. 
 
 Cricket. — Waterton uses the collo(|uial term. The so-called 
 Crickets are in fact Cicadas, several species of which are 
 common in the southern parts of Europe. Only one Britit>h 
 species is known, and may be found in the New Forest. In 
 Guiana, the Cicadas attain a very large size, and their cry has 
 been compared to the whistle of a railway engine. 
 
 Cuckoo, IMetalijc (Coccygus Americanus). 
 
 Curlew, Scarlet. — Waterton here employs the colloquial 
 name for the Scarlet Ibis {Tantalus ruber). 
 
 There are several species of Ibis, the most familiar being 
 the sacred Ibis of Egypt, which figures so largely in the 
 ancient Egyptian .sculptures and paintings. This is but a dull 
 bird in colour, being only white and black, and with a neck 
 quite bareof feathers, and looking as if made of a very eld 
 and very crumpled black kid glove. 
 
 The Scailet Ibis, however, is far more beautiful than any of 
 its congeners, being of a mosst brilliant scarlet, with a few 
 patches of jet black. 
 
 SCAllI.Er CLllLKW. 
 
 The accompanying illu^tlation represents one of theije biids 
 at it appeared when running it.self in the Zoological Gardens. 
 It bowed its head until the tip of the curved beak nearly 
 
 n n 
 
40'J 
 
 EXPLANATORY INDEX. 
 
 touched the ground, half ojjoned its wings, thrust them 
 forward, and kept up a tremulous movement of the feathers 
 over the whole body. The bird is mostly found along the 
 coasts. 
 
 Another species, the lliver Ibis {Ibis iv/uscatus), is found, 
 as its name imports, on the rivers of Guiana. It is larger 
 than its scarlet relative, but is not nearly so handsome, being 
 dull olive green, with light green legs and beak. It goes by 
 the popular name of Curi-curi, in consequence of its cry, which 
 is said to resemble these words. 
 
 Deer. — Several species of Deer inhabit Guiana, but that 
 mentioned by Waterton is probably the Forest Deer {Cervus 
 humUis). It is a small animal, reddish fawn in colour, spotted 
 with white on the flanks, and with little straight hox*ns. The 
 native name is Wiriebiserie. 
 
 Diamond Rock. — This is a very singularly-shaped rock, 
 close to Pointe du Diamante, and having only a narrow 
 channel between itself and Martinique. The shape is roughly 
 conical, and is said to I'esemble the gi'eat Pyramids of Egypt, 
 but to be twice as large as the largest of them. 
 
 "In the end of 1803," writes the late C. Kingsley, "Sir 
 Samuel Hood saw that French ships passing to Fort Royal 
 harbour in Martinique escaped him by running through the 
 deep channel between Pointe du Diamante and this same 
 rock, which rises sheer out of the water GtJO feet, and is 
 about a mile round, and only accessible at a point to the 
 leeward, and even then only when there is no surf. He who 
 lands, it is said, has then to creep through crannies and 
 dangerous steeps, round to the windward side, when the eye is 
 suddenly relieved by a sloping grove of wild fig-trees, cling- 
 ing by innumerable air roots to the cracks of the stone. 
 
 " So Hood, with that inspiration of genius so common then 
 among sailors, laid his seventy-four, the Centaur, close along- 
 
KXI'LAXATUUY IXDKX. 
 
 40;} 
 
 the 
 same 
 id is 
 
 the 
 iwho 
 
 and 
 
 i^e is 
 lling- 
 
 Ithen 
 
 ong- 
 
 side the Diamond ; made a hawser, with a traveller on it, fast 
 to the ship and to the top of the rock ; and in January, 1804, 
 got three long 2-t'8 and two 18's hauled up far above his mast- 
 head by sailors who, as they * hung like clusters,' appealed 
 * like mice hauling a little sausage. Scarcely could we hear 
 the governor on the top directing them with his trumpet ; the 
 Centaur lying, close under, like a cocoa nut shell, to which 
 the hawsers are affixed' {A\ival Clirouides) , 
 
 "In this strange fortress Lieutenant James Wilkie Maurice 
 (let his name be recollected as one of England's forgotten 
 worthies) was established with 120 men and boys, and am- 
 munition, provisions, and water for four months ; and the 
 rock was borne on the books of the Admiralty as his IVla- 
 jesty's ship Diamond Rock, and swept the seas with her guns 
 till the 1st of June, 1805, when she had to surrender, for 
 want of powder, to a French squadron of two Ti's, a frigate, 
 a corvette, a schooner, and eleven gunboats, after killing and 
 wounding some seventy men on the rock alone, and destroy- 
 ing three gunboats, with a loss to herself of two men killed 
 and one wounded. 
 
 " Remembering which story, who will blame the traveller 
 if he takes off his hat to his Majesty's quondam corvette, as 
 he sees for the first time its pink and yellow ^ides shining in 
 the sun above the sparkling seas over which it domineered of 
 old?" 
 
 I may add that the sailors were greatly amused at their 
 very remai'kable sloop, and invariably used nautical +erms 
 when speaking of it, or " her." For example, when the;; had 
 by extreme ingenuity " parbuckled " a couple of guns nearly 
 to the summit of the rock, they named them the " topgallant 
 battery." Everything went on exactly as on board a man-of- 
 war. Watches were set and relieved according to naval dis- 
 cipline, observations were daily taken, and, except that there 
 was no man at the wheel, and no sails had to be looked after, 
 the crew had just the same duties as if the rock had been a 
 ship afloat. 
 
 u D 2 
 
^ii 
 
 404 
 
 I<:XPLANATORY INDEX. 
 
 Wheti at last they did surrender, they were allowed to 
 leave their "sloop" with all the honours of war, and to 
 row themselves in their own boats to the French ships. 
 
 Dolphin. — The creature which Watei'ton here describes is 
 not the true dolphin, which is one of the mammalia, and not a 
 fish ; but is the Ooryphene {Corijphene hippuris), a splendid 
 iihh, which sailors will insist on calling by the name of dolphin. 
 The weapon which is thrown at, or rather dropped upon the 
 fish is called by the name o* "grains," and much resembles 
 Neptune's trident, except that it has five points instead of 
 three. The colours of the dying Coryphene are singularly 
 beautiful ; but it is evident that the thick, tough, blubber- 
 lined hide of the true dolphin could undergo no change. 
 
 DucALABALi. — Ouo of the many useful species of West 
 Indian Laurus, the value of which is not sufficiently recog- 
 nised in this country. The wood is very hard, close grained, 
 heavy, durable, and deep-red in colour. Only the heart-wood 
 is u.'-ed. As it is spotted with blackish-brown, like the 
 jaguar's skin, it is often called Tiger-wood. The tree is rare, 
 and is mostly found near the Essequibo. It grows chiefly in 
 clay and sandy soils, and reaches about one hundred and 
 twenty feet in height. The heart-Avood, however, is seldom 
 more than two feet in diameter. 
 
 DuRAQUAURA {Odoutophonis Gmameitsls). — This bird, like 
 the English partridge, builds a nest on the ground, but is 
 a better architect. The nest is made of leaves and twigs, 
 and is placed under the shelter of a bush. It is roofed, and 
 has a small entrance in front. The eggs are eight or nine 
 in number, white in colour, and are about as large as those 
 of the bantam. 
 
 E. 
 
 Eagle, White Headed {Ilaliaelus leucocephalus.) 
 
 Ebony Tree. — There are several trees which furnish the 
 
 well known ebony of commerce, but the best kind is obtained 
 
 from the heart- wood of Diospyros ehcninn. 
 
 i^. 
 
EXPLANATORY INDKX. 
 
 40o 
 
 Egret.— This beautiful little heron {drdea eyretta), is 
 pure white in colour except the delicate train feathers, which 
 have a slight creamy tinge. They are long enough to hang 
 over the tail, which they conceal when the bird is in repose. 
 
 ' 1 
 
 '! .; 
 
 
 ;i'!l 
 
 
 Owing to their Ijauty, these plumes are in gi^eat request for 
 the ornamentation of head dresses, not only for the natives, 
 but for ladies in Europe. The natives, who take great pride 
 in their head-dresses of ceremony, and arrange the feathers 
 with marvellous taste, make much use of the plumes, which 
 they mingle with the feathers of macaws and parrots. 
 
 The second sketch was taken from a fine specimen in the 
 possession of J. F. Jackson, Es^q., of Bexley. As evening had 
 come on the bird had retired to a sort of small grotto on the 
 edge of a pond, and had sunk to sleep, as represented (p. 406). 
 Owing to the darkness of the evening I had nearly passed the 
 bird without seeing it, and but for the snowy whiteness of 
 its plumage, it would have escaped unnoticed. 
 
 -Hi 
 
! 
 
 403 
 
 KXrr.AXATOHY INDEX. 
 
 In many juivts of Ciuiana tho Egret is called by the name of 
 Wliito (Jmildin. 
 
 F. 
 
 FiC!, Wii-T) (Clvsia alh<i). 
 
 Finch, IIed-iikadku (Probably Passerina gvlarin). 
 
 V\\iVA''\A'. — In Tropical America there are so many lumi- 
 nous insects Avhich go by tho popular name of firefly that 
 no particular species can be hero designated. Most of these, 
 however, are beetles belonging to the group of elaters. Many 
 species exist in England, and are well-known as skip jack 
 bee<:los, spring-beetles, or click- beetles, because if laid on their 
 backs, they spring up in the air with a smart click, and take 
 
 KlillKT ASI.KKP. 
 
 their chance of falling on their legs. The tei'rible wire-worm 
 is the larva of one of these beetles. 
 
 The species which is most common is the Cucujo (Pvro- 
 phorus luriihiosus.) As is the case with our glow-worm, the 
 light disappears after de.ath, and, indeed, how it is produced 
 in life is an unsolved mystery. Even the spectroscope yields 
 no information, giving onl}^ a ' continuous ' spectrum, i.e. one 
 which is not crossed by lines, either dark or luminous. 
 
 Still, the light which the insect emits is so powerful that, 
 as Watertou mentions, it will, if held over a paper at night. 
 
KXPLAXATOUY INDEX. 
 
 407 
 
 permit the holder to read or write by its natural lantern. It 
 is sppn at it.s best when on the wing, as it then discloses four 
 l^-.iinous spots, of which only the upper pair are visible 
 when the insect is at rest. 
 
 The upper pair are oval in shape, and arc seen at t}>e base 
 of the thorax, one on cither side, and after death fade into a 
 brownish yellow. When the beetle expands its wings for 
 flight, two more spots are f-een, which had been concealed by 
 the wing-cases, so that the appeax'ance of a single Firefly on 
 the wing is gorgeous in the extreme, and, considering that they 
 fly in countless thousands, their coruscating lights can be 
 compared to nothing but starry showers of fireworks. Even 
 in the warmer countries of Europe the Fireflies are very 
 beautiful, but they are completely surpassed in splendour by 
 those of the tropics. 
 
 In one of his essays, 'vVaterton mentions an absurd state- 
 ment, that there are certain birds which fasten Fireflies on 
 their nests in order to keep off the bats which might devour 
 their young. He very curtly disposes of the matter by 
 saying that bats do not eat young birds, but that they do 
 eat Fireflies, and would in consequence rather be attracted 
 than frightened by them. 
 
 Flamingo {Phfjenicopterus ruler). — Waterton devotes part 
 of one essay to a congenial task, that of demolishing an 
 error in natural history. This particular error concerns the 
 nesting of this bird, which is said to raise conical mounds of 
 mud in order to keep its nest out of the water, to lay its 
 eggs on the top of the heap, and to sit upon them with its 
 legs hanging down into the water. 
 
 This attitude, as Waterton showed, by reference to his own 
 herons, is not necessai'y, the long legs of the Flamingo and 
 the heron being as easily bent under the body as those of the 
 short-legged birds. In consequence of a conversation with 
 Waterton on the subject, I went to the Zoological Gardens 
 in order to watch the attitudes of the Flamingo, and was greatly 
 struck with the lithe activity of the bird. I made a number 
 
 lit 
 
 •iii 
 
 ■i' 
 
 ii, 
 
 ' ' I'' 
 
 li 
 
 • ■ ! i: 
 
 \m 
 
I 
 
 1 
 
 4U8 
 
 KXl'LANAToKY INDEX. 
 
 of sketches from thorn, some of which are here given to the 
 reader. 
 
 Judging from stuffed specimens, the Fhimingo seems to be 
 rather an awkward bird, but in reality, it is peculiarly easy and 
 graceful in its movements. The long neck can be twined and 
 turned in all dii'ections, very much like the coils of a serpent, 
 
 or the lithe proboscis of the elephant. Some of the curious 
 attitudes which it assumes are shown in the illustraiiona. 
 
 As may be seen by them, the leg.s, awkward as they may 
 seem, are as much under command as the neck, and I have 
 seen the Flamingo scratch its head with its claws as easily as if 
 it were a parrot or a canary. Any one who had not studied 
 the living bird, would have thought that when it required 
 
 ■"^^^y 
 
 M.EKFINO. 
 
 l-nKPARINO FOR SI.EKP. 
 
 repose, it would squat on the ground, so as to rest its stilt like 
 legs. But, in common with many other birds, it sleeps while 
 
to 
 
 )e 
 id 
 id 
 t, 
 
 EXl^LANATORY INDKX. 
 
 4(t9 
 
 standing on one log, which is kopt perfectly straight, while 
 the neck lies in folds upon the !<houldor, with the head nestling 
 among the feathers. 
 
 This position of the head during sleep or illness is universal 
 among birds, and is popularly called " putting the head under 
 the wing." How such a phrase could have been invented or 
 perpetuated is really wonderful, as any one who has kept 
 poultry or pet birds of any kind must have noticed that the 
 
 l'\ I i. 
 
 
 us 
 
 ve 
 
 if 
 ed 
 ed 
 
 ke 
 
 head is not hidden under the wing, but rests among the 
 plumage of the shoulders. So, we must bid farewell to our 
 pretty nursery rhyme : - 
 
 " Tlic north wind doth blow, 
 And we shall have snow, 
 And what will poor Kobin do then, 
 Poor thing ? 
 
 He will sit in a baiii 
 To keep himself warm. 
 And hide his liead under his wing, 
 Toor thing ! " 
 
 During life the beak of the Flamingo is pinkish yellow at 
 the base, becoming black at the tip. After death, however, 
 although the tip retains its blackness, it loses its polish, 
 and fades into dingy yellowish brown. Similarly, the legs, 
 which ai*e red in life, become brown at death, and the only 
 mode of restoring these colours has been by employing paint, 
 which has really a ghastly appearance. I have not had the 
 opportxinity of treating the beak of a Flamingo as Waterton 
 managed to colour the beak of a toucan, but I have little 
 doubt that the process would be as effectual with one bird as 
 it has been with the other. 
 
 !':■ 
 
 Mj 
 
 ^i' 
 
410 
 
 KXIM.ANAToUY INDKX. 
 
 Tho fj'athers uro mostly wliito with u 8li{,'Iit pink tinge, 
 but tho wing-fciithors aro mostly brilliiint scarlet, mid liavo 
 earned for tho bird tho generic title of Pba'nicopterua, i.e. 
 ' scarlet winged. ' 
 
 The curving bend, or angle in the beak of the Flamingo, is 
 given to tho bird in order to enalde it to feed after its own 
 peenliiir fashion. Like the duck, the Flamingo feeds by 
 dabbling with its beak in muddy water, and filtering out tho 
 nutritive mutter by means of an arrangement which mucli 
 re.sembU'8 a pair of very fine, soft, and flexible combs, fitting 
 
 into each other in strainer fashion. The duck, having short 
 legs, can push its l)ill into tho water without difficulty, but 
 owing to its very long neck and legs, the Flamingo could not 
 do so, except for the peculiar curve in its bill, which enables 
 it to place its head upside down when it feeds, and in this 
 rather odd position to extract nourishment from the water. 
 
 Any one who wishes to see the mechanism by which a 
 Flamingo separates the nutritive substances from the water, 
 has only to take the head of a duck and examine the beak 
 and tongue. In order that the structure should be seen in 
 ]:)erfection, the head ought to be held under water during 
 examination, the beak widely opened and closed, and the 
 finger be used as well as the eyes. 
 
 The accompanying illustrations are taken from sketches 
 made at the Zoological Gardens, in order to show the extra- 
 ordinary attitudes into which this bird throws itself, and 
 especially the power of balance on one leg when sleeping, and 
 the extraordinary contortions into which the bird can twist its 
 neck while the body is resting on the ground. 
 
 Flying-Fish {Exocuitns volitans). — Waterton, while men- 
 tioning the flight of this curious fish, does not touch on the 
 
KXPTwWAToHY INDEX. 
 
 411 
 
 tlieputed point of its ciipiicity to alter its coiirso in tlic air, or 
 to extend its tli^'lit by flapping its wing like lins. V. Kingslcy, 
 howovor, b<'li('V(>s from personal observations, that it can do 
 botli. 
 
 '* The Hying tisli now began to bo a source of continual 
 amusement, as they scuttled away fi'om under the bows of 
 the ship, mistaking her, probably, for some hii. e devouring 
 whale. 80 strange are they when lirst seen, though long 
 read of and long looked for, that it is diflicidt to recollect that 
 they are actually fish. The lirst littlo one was mistaken for 
 a dragon-fly, the first big one for a grey plover, 
 
 " The flight is almost exactly like that of a quail or part- 
 ridge-flight, I must say ; for, in spite of all that has been 
 learnedly written to the contrary, it was too dithcult as yet 
 for the English sportsmen on board to believe that their 
 motion was not a true flight, aided by the vibration of the 
 wings, and not a mere impulse given (as in the leap of the 
 Halnion) by a rush under water. 
 
 "That they can change their course at will is plain to one 
 who looks down on them from the lofty deck, and still more 
 from the paddle box. The length of the flight seems too great 
 to be attributed to a few strokes of the tail j while the plain 
 fact that they renew their flight after touching, and only 
 touching, the sin'face, would seem to show that it was not due 
 only to the orig'nal impetus, for that would be retarded, 
 instead of being quickened, every time they touched. Such 
 were our first impressions ; and they were confirmed by what 
 we saw on the voyage home." 
 
 I think that this statement from a good naturalist and 
 keen sportsman, such as was Kingsley, settles the question of 
 flight. I doubt, however, whether the fish dip into the sea for 
 the purpose of wetting their wings, as suggested by Waterton. 
 Perhaps their flight may be like that of the short-winged 
 birds, and consist of alternate dippings and risings. 
 
 Fkioate Bird {Attagen aquUa). — This really wonderful 
 bird seems mostly to bo composed of feathers, the body 
 
 
 ! 
 
 M 
 
 'Hi' i: 
 
 i 
 
 
 ':* 
 
 lib 
 
 H 
 
 w. 
 
:f t 
 
 412 
 
 EXPLANATORY INDEX. 
 
 being quite small, and the expanded wings measuring about 
 eight feet from tip to tip. The total length of the bird, 
 including the long tail, is about three feet. The muscles of 
 the breast, by which the wings are moved, are themselves 
 one quarter the weight of the entire body. Thus, like the 
 albatross, it can pass the greater part of its time on the win 
 
 g' 
 
 KKKiATE Blltn. 
 
 not seeming to feel fatigue. As it wheels its flight over ships, 
 the scarlet throat-pouch is very conspicuous, the light shining 
 through its skin when expanded. 
 
 It is a very voracious bird, and Dr. Bennett remarks that 
 he has repeatedly seen it sweep upon the top of the mast and 
 carry away the coloured vane. 
 
 Being no diver, and a very poor swimmer, it is apt to 
 pounce upon gannets and other diving birds, and rob them of 
 their prey. Should the gannet pluck up sufl[icient spirit to 
 resist, the Frigate bird is sure to be worsted. Both birds 
 necessarily fall into the sea, where the gannet is quite at 
 home, and escapes by diving, while the Frigate bird is 
 helplessly flapping about in the water. 
 
 The colour of the adult male Frigate bird is black, with a 
 gloss of green, the female being of much duller black above 
 without any green gloss, and having the under surface nearly 
 white, A\ ith splashes of reddish brown. 
 
 Fox (Vtilpes cancrivora). — This animal is generally called 
 Savannah Fox by the colonists, and Mikang by the natives. 
 It is very swift, and in fair chase can outrun most dogs. It 
 is very small, grey in colour, and has a large bushy tail. 
 
EXPLANATORY INDEX. 
 
 413 
 
 G 
 
 Goatsuckers.- -All these bircLs, of which there are many 
 in Guiana, belong to the Caprimulgida^, and some, from the 
 singular cry which they utter, are objects of superstitious 
 dread, both to natives and negroes. Kingsley gives an 
 aiausing and graphic description of the alarm caused by one of 
 these birds. After narrating how he tried to sleep, and was 
 kept awake, first by the romping of his companions, who 
 broke down a four-post bedstead in their play, and then by the 
 wind, which blew all the clothes off the bed, he proceeds as 
 follows : — • 
 
 *' Then the dogs exploded outside, probably at some hen- 
 roost robbing opossum, and had a chevy through the cocos 
 till they treed their game, and bayed it to their hearts' con- 
 tent. Then something else exploded — and I do not deny it 
 set me more aghast than I had been for many a day — ex- 
 ploded I say, under the window, with a shriek of hut-tut-tut- 
 tut, hut tut, such as I hope never to hear again. After which, 
 dead silence ; save of the surf to the east and the toads to 
 the west. I fell asleep, wondering what animal could own so 
 detestable a voice ; and in half an hour was awoke again by 
 another explosion ; after which, happily, the thing, I suppose, 
 went its wicked way, for I heard it no more. 
 
 " I found out the next morning that the obnoxious bird was 
 not an owl, but a large Goatsucker, a Nycteribius, I believe, 
 who goes by the name of jumby-bird among the English 
 negroes ; and no wonder ; for most ghostly and hon'ible is his 
 cry. But w^rne ; he has but one eye, and a glance from that 
 glaring eye, as from the basilisk of old, is certain death ; and 
 worse still, he can turn oil its light as a policeman does his 
 lantern, and become instantly invisible ; opinions which, if 
 verified b;, experiment, are not nl.vays found to be in accord- 
 ance with facts. But that is uc reason why they t-hould not 
 be believed. 
 
 " In St. Vincent, for instance, the negroes one evening 
 
414 
 
 HXPLAXATOUY INDKX. 
 
 (:: 
 
 in if I 
 
 :M 
 
 rushed shrieking out of ix boiling-house. ' Oii I INlassa Robert, 
 we all killed. Dar one great jumby-bird come in a hole a top 
 a roof. Oh 1 Massa Kobert, you no go in ; you killed, we 
 killed,' &e. &c. Massa Robert went in, and could see no bird. 
 * Ah ! Massa Robert, him darky him eye, but him see you all 
 da same. You killed, we killed,' Ac, <la atjio. 
 
 " Massa Robert was not killed ; but lives still to the great 
 benefit of his fellow-creatures, negroes enpecially. Neverthe- 
 less, the negroes hold to their opinion. Ho might, could, 
 would, or should h.ave been killed j and was not that clear 
 proof that they were right 1 
 
 " After this, who can deny that the negro is a man and a 
 brother, possessing the same reasoning faculties, and exercising 
 them in exactly the same way, as throe out of four white 
 persons ? " 
 
 This is evidently the " largest doatsuckcr," mentioned by 
 "Water ton on page 19'J. 
 
 GiiEENHEAiiT {Xectcmdra rndm), called by the natives 
 Bibiri. — Here we have another of the neglected products 
 of Guiana. Kingsley thus mentions it in his At Last, ; - 
 
 " The carapo is not the only tree of Houth America whose 
 bark may be used as a substitute for quinine. They may 
 be counted possibly by dozens. A glance at the excellent 
 enumerations of the uses of vegetable products to be found in 
 Lindley's Vegetable Kinydom (a monument of learning), 
 will show how God provides, and how man neglects and 
 wastes. 
 
 " As a single instance, the laurels alone are known already 
 to contain several valuable febrifuges, among which the De- 
 merara Greenheart, or Bibiri, claims perhaps the highest rank. 
 ' Dr. Maclagan has shown,' says Dr. Lindloy, that ' sulphate 
 of Bibiri acts with rapid and complete success in averting 
 ague.' This tree spreads from Jamaicra to tlio Hpanish Main. 
 It is plentiful in Trinidad, still more plentiful in Guiana; yo!; 
 all of it that reaches Europe is a little of its hard, beautiful 
 wood for the use of cabinet-makers ; while in Demerara, I am 
 
EXPLANATORY INDEX. 
 
 415 
 
 assured by an eye-witness, many tons of the precious Green - 
 heart bark are thrown away year by year. So goes the 
 world, and man meanwhile at once boasts of his civilizatic i 
 and complains of the niggardliness of nature." 
 
 This is just the complaint made by Waterton in more than 
 one passage of the Wanderings. He had some furniture 
 at Walton Hall made of the Greenheart, and very excellent 
 furniture it was, and probably is still. 
 
 Greenheart is one of the eight A 1 timbei's at Lloyds'. It 
 is employed for kelsons, planking, and ' knees,' when these are 
 not made of wrought iron. It is time that legal protection 
 should be given to this tree, for the timber-merchants and 
 chai'coal-burners have made great havoc with it. There are 
 three varieties, called respectively, the Yellow, Black, and 
 Mainop Greenheart. Itogs are sometimes seventy feet long 
 and two feet square. 
 
 Every part of the Greenheart is useful, bark, juice, and 
 timber being equally valuable in the service of man. 
 
 Grosbeak, Scarlet {Cardimdis Vrnjinianvf). — This bird is 
 called by various names, such as Virginian nightingale, Car- 
 dinal bird, &c. In consequence of the beauty of its plumage, it 
 is often brought to England, where it t'n*ives well, but is apt 
 to lose its magnificent scarlet plumage, and become whitish- 
 red. In its native country it builds in some thick evergreen, 
 and lays five pale grey eggs splashed with greenish brown. 
 
 Guana or Iguana (Iguana tuberctdata). — This is a very odd- 
 looking, and not very handsome lizard, which inhabits the 
 West Indies, and is mostly found on the branches of trees 
 which overhang the water. It is rather large, an average sized 
 adult being about four feet long, while some attain a length 
 of six feet. The body however, is not very large, but the 
 tail is very long, and can be lashed from side to side so 
 sharply, that a stroke will cause much pain. 
 
 In M)'. Brown's work on Guiana are some interesting 
 remarks on this lizard, of which, by the way, there are 
 several species : — 
 
 tli 
 
 I 
 
 ;; » , 
 
 k'tl'll 
 
/i( 
 
 ■¥ 
 
 416 
 
 EXPLANATORY INDEX. 
 
 " We were frequently amused with the manner in which 
 the Iguanas, ahirmed by the noise produced by the boat's 
 paddles, threw themselves from the overlianging branches of 
 trees into the river, many coming down broadside on the 
 water. Their flesh resembles that of a chicken in flavour and 
 quality, and is very good when properly cooked. 
 
 " They are exceedingly numerous on the river (the Cuyuni), 
 where they dig long underground chambers in the sandbanks 
 at the ends of islands, in which they deposit their eggs. Near 
 the foot of a cataract, where there was a high beach, our men 
 dug up the sand in order to ti'ace out some of these chambers 
 for eggs, and succeeded in capturing four Iguanas, which they 
 dragged out by their tails and then seized by the back of the 
 neck and secured. They had to be pretty quick, for the 
 Iguana can turn round suddenly and give a fearful bite. An 
 Iguana must be drawn from its hole with rapidity, for, if it 
 has time to think, it lets go its tail at the base, leaving it in 
 the hands of the would be capturer, and thus escapes. 
 
 " One of these reptiles, captured at its burrow, when killed 
 and cut up for cooking, was found to contain ten eggs of an 
 ellipsoidal form, shell less, and midway in size between a 
 pigeon's and a hen's egg. These are good eating when boiled 
 for about five minutes and then allowed to get quite cold. 
 They then require some manipulation. A hole is made in one 
 end of the skin, and the albuminous part, v;hich never 
 coagulates, is squeezed out ; then the skin is stripped off, and 
 the semi-hardened yelk, of the consistency of butter, is eaten 
 with salt. 
 
 " Our men and the Indians had a most cruel way of pre- 
 venting captured Iguanas from escaping, by slitting down the 
 sides of two fore and two hind toes, on opposite feet, and 
 passing the toes between the bone and sinew in such a manner 
 that they could not be disengaged by the struggles of the 
 poor animal. The manner in which they were slaughtered, 
 viz., by inserting a hard-pointed stick up one nostril into the 
 brain, was also a very cruel proceeding." 
 
EXPLAXATORY INDEX. 
 
 417 
 
 Perhaps this may he a cruel way of killing the CInana, but, 
 like nearly all reptiles, it is little sensitive to pain, and 
 wonderfully tenacious of life, and, as it must be destroyed for 
 food, perhaps the pointed stick is the quickest mode of killing 
 it. Not that the natives trouble themselves about the in- 
 fliction of pain, for, besides the mode of securing the Guana 
 as above mentioned, they sew its lips together, in order to 
 prevent it from biting, and keep it without food until they 
 want it. Here, again, they are not more ci'uel than our rat- 
 catchers, who used to sew together the lips of their ferrets, or 
 our fishermen, who used to disable their lobsters by " pegging " 
 their claws as soon as caught. 
 
 GuAVA {Psidium pomi/cruin). — The tree which bears this 
 well known fruit, is quite a, little one, scarcely larger than a 
 
 privet bush, and the fruit is small, round and green. It can 
 be oaten without any preparation, but is mostly made into 
 
 K E 
 
 1,'i' 
 
 )}' 
 
 ;l 
 
 1 1' 
 
418 
 
 EXri.AXATollY IXDKX. 
 
 ii 
 
 \ 
 
 \\\ 
 
 I 
 
 iii 
 
 jolly. It is remavkablo, howevei', that few ti'opical fruits, 
 except the orange, the cocoa-nut, the pine-apple, and the 
 (lurian, can be compared with our plums, peaches, straw- 
 borrios, &c. The durian indeed, which is said by its lovers to 
 be the king of fruits, requires an education before it can be 
 appreciated ; but when it is once tasted, the eater feels as if 
 ho can never have enough of it. 
 
 There are several species of CUiava, among which is the 
 little, stunted-looking Water Guiivn { J ',s! J hi mjluviatih). The 
 mos-t valued of them is the Perfumed Ouava (P. ])omiJ'erimi), 
 which grows abundantly on the banks of the Cuyuni river, its 
 roots insinuating themselves into the cracks among the stones. 
 
 At least seven species or varieties are cultivated. 
 
 H. 
 
 growing in 
 
 ITAf'KEA {Siderodeiulnm frijlornm). — A tree 
 sandy soil, and flowering in November. The flowers are 
 bright yellow, and when the trees are seen against the dark 
 foliage of the forest, they look like masses of gold. Hence 
 the name Siderodendron, or Star-tree. 
 
 The wood is not used for general purposes, on account of 
 its extreme hardness, but is exceedingly valuable for cogs, 
 shafts, iVc. The tree is about sixty feet in height, and the 
 wood is brown in colour. It squares about foui-teen inches. 
 
 Hataurie.— aScc " Wourali." 
 
 Hammock. — These ingenious contrivances are now so well 
 known in England that little description is needed. Here 
 they are a luxury, and a very great one ; there, they ai-e an 
 absolute necessity, 
 
 They are sometimes formed of silk-grass fibre, and some- 
 times of cotton, the latter being the more pliant, and the 
 former the more lasting. The best hammocks are not formed 
 like nets, and knotted, but the strings are intertwined so that 
 they adapt themselves to every movement of the body. A 
 
?:xri,.\XATouY ixdex. 
 
 4i:i 
 
 nativo nover j»oe.s on a jnnrnoy without his lianiiiiock, wliich 
 ho rolls into a sort of rope, and passes it over one shoulder 
 and under the other, just as oflicers in the army wear their 
 overcoats on the march. 
 
 In such a moist country as fJuiana, where to sleep wet 
 means to ensure a fever, and where a traveller uijl sometimes 
 have to Avalk day after day up to his knee-* in water, the 
 immock is a nece; ■ '^i life. When a traveller wishes to 
 rest, ho has only to nanj^ his hammock bftween a couple of 
 trees, and ho can then damher into it. lub liis wet legs dry, 
 nnd lie down in comfort and security. 
 
 TTammocks are of various sizes, some being very small and 
 made for children, while olhers aie birge fnough to hold two 
 or more people. As a rule, however, each person has a 
 hammock to himself. 
 
 1 li 
 
 rvell 
 ere 
 an 
 
 il \NNAi;l 11. 
 
 Hanxaqi'OI {OTlalhlii uK>t)>H>l).—'n\(^v(^ are several spccie.s of 
 the ^rotmota, as these birds are called from their note, but 
 that wliich is given is probably that to which Wateiton refers. 
 
 i; i: 
 
 ^^i' 
 
 U 
 
I. I 
 
 i 1 
 
 420 
 
 EXPLAXAToUY INDEX. 
 
 As the bill is large in proportion to the size of the bird, and 
 deeply notched along the edges, it has been by some persons 
 thought to belong to the toucan tribe. It is a solitary bird, 
 and not to be seen except by those who, like Waterton, travel 
 into the interior of the forests. 
 
 ]Mi*. C B. Brown found a nest of the Ilannaquoi situated in 
 a low tree. It was made of sticks, and contained four 
 speckled eggs, about equal in size to those of the common fowl. 
 The plumage of the Ilannaquoi is very beautiful, being almost 
 entirely blue and green glossed with crimson, and there is a 
 black velvety spot on the breast and head. 
 
 The bird is easily tamed, and soon becomes domesticated 
 with the poultry. 
 
 Hayawa Trek. See " Acaiari." 
 
 Heuon, Blue {Ardea lencixjastar.) 
 
 Biiow.v {Ardea cirescens.) Popularly called the Grey 
 
 Gauldin, 
 
 HiA-iiiA. - >Se« "Parrot, Sun." 
 
 HiTiA. — As this tree is mentioned in conjunction with the 
 siboaLilis, it is probably an Kporua. 
 
 Hoc'co. — See " Powise." 
 
 Hoo Wild. — See "Peccary." 
 
 HouTou {Priolelus temnurus). — Thie beautiful bird is one 
 of the splendid group of the Trogons, whose plumage literally 
 blazes with scarlet and green, contrasted with velvet black 
 and snowy white. This particular species is also called the 
 Cuba Trogon, and it is easily recognised by means of the 
 singular shape of the tail-feathors. They have the central 
 shaft of the feather extending for some little distance beyond 
 the web, and look exactly as if they had been cut with a pair 
 of scissors. The name of ' tennmruft,' or 'clip-tail,' is given 
 to the bird in consequence of this poculiarity. 
 
 Waterton says that the Iloutou clips its tail-feathers with 
 its beak, but he nowhere states that he had ever seen it do 
 so, and there is little doubt that the feathers owe their curious 
 formation to nature alone. Mr. Could says, that the Houtou 
 

 EXri.AXAToifY JNDKX. 
 
 421 
 
 behaves much like the woodiieckers, running about the trunks 
 and branches of trees and pecking insects from their hiding- 
 places under the bark. 
 
 inn Tou. 
 
 HowLEii, PiED {Mycetes nrslnus). — This fine monkey, of 
 which so many strange stoiies are told, is sometimes men- 
 tioned under the title of Amguato. It sometimes attains a 
 total length exceeding six feet, the tail occupying more than 
 half of that measurement. The following account of the 
 lied Howler is taken fiom Mr. C. B. Brown's work on 
 Guiana : — 
 
 " At early morning, at dusk, and through the night, at all 
 our camping places, we were accustomed to hear the Howlers 
 serenading. To my mind the sounds produced by these 
 monkeys more nearly resembled a roar than a howl, and when 
 sufficiently far off are not unpleasant to the ear. When 
 heard from a distance of half a mile or so they seem to begin 
 with low notes, swelling gradually into louder and longer 
 
 ■•i i' .1 
 
 if 
 
 .< 
 
i;XlM,A.\AT<>i:V INDKX. 
 
 ones till they merge into a prolonged roar, which dies gradu- 
 ally away with a mournful cadence. 
 
 " When not more than one or two hundred yards away, 
 and consequently plainly heard, they commence with a series 
 of short howls, which hreak oil' into grunts, and, at every 
 repetition, become longer and longer till their voices have got 
 fairly in tune, when they give their tinal roar, which dies as 
 gradually away. Then, alter an instant's ailence, a few deep 
 grunts are given, as if the remains of the compressed air in 
 their throat diuius were being got rid of. Listening carefully 
 
 to the performance, one can detect a voice at a much higher 
 key than the others, especially in the dyingaway portion. 
 The Indians say this is made by a dwarf monkey of the same 
 family which accompanies every troop. I was of the opinion 
 that it wiis the voice of a female Howler, but the Indians, 
 who are very careful observers, said it was not. With their 
 black negro-like faces, and long red beards covering the bony 
 howling apparatus below their throats, they are curious- 
 looking creatures." 
 
 Home specimens of the lied Howler which were brought to 
 
lOXPI-AXATOltY INDEX. 
 
 423 
 
 KiiglunJ, exhibited tlio peculiar stnicture of tlio throiit nio.>t 
 admijiibly. When the monkoy opens its mouth, the do- 
 velopment of the ' hyoid ' bone is easily seen. It forms a 
 kind of bony drum communicatin<,' with the wind-pipe, and 
 enabling the animal to emit those horrid yells which have 
 earned for it the popular name of Howler. 
 
 The reader is here requested to look at the note upon the 
 " Nondescript." 
 
 ^Ii^ 
 
 
 KAKAlll.Ml ! 
 
 Pi I 
 
 HuM.MiNLi-DiJii), Aha, or Kauauimiti (Topaut jiella), some- 
 times called the Crimson Topaz. — This is one of two hum- 
 ming-birds which are distinguishable by the length of two 
 tail-feathers, Avhich are supposed to resemble tlie elongated 
 tails of the aras or macaws. The other species is the tiery 
 topaz {2'opar.a jryra), so called because the general colour of 
 the bird is flaming scarlet instead of the rich crimson which 
 distinguishes the Karabimiti. 
 
 Tlie food of the humming bir'is was long a matter of 
 
 'i\ 
 
 m 
 
 ^■M 
 
 t 
 
 0: 
 
 i.\:'%' 
 
42t 
 
 KXI'LANATOlfY IN'DEX. 
 
 uncertainty, but has boon dofinitivoly settled by Wuterton 
 and other pructiciil naturalists who camo after him. 
 
 •* Neither the monkey nor the humming bird, on account of 
 the formation of the feet in this, and of the hands in that, 
 can labour on the ground for their food. Yet, when they are 
 in the right region to acquire it, there is a visible difference 
 in their mode of proceeding. Thus, the monkey sits on the 
 branch, and in that position supplies its wants with what 
 the tree produces. I'.ut thn humming- bii*d must be on the 
 wing whilst it extracts food fivmi the tlowers, and never can 
 it possibly be seen to take nourishment whilst perching on 
 a tu'ig. 
 
 " This rule is absolute for the Inrnming-bird. 
 " The vault of heaven oilers a large supply of food to these 
 birds. It is interesting to see how they satistios the call of 
 hunger, by invading the columns of insects which frequent 
 the circumambient atmosphere. Darting from the shade with 
 the rapidity of a meteor, the humming-bird stops short at the 
 column, and then, apparently motionless, it regales itself, and 
 then departs as swiftly as it had approached. 
 
 "Authors are divided as to the exact kind of food which 
 hummir.g-birds requiie. In all the species which I have in- 
 spected (and I have inspected not a few) I have found insects, 
 or fragments of insects, in the o'sophagus ; and occasionally, 
 by applying my tongue to the contents of the stomach, I have 
 experienced a sweet taste, as though of sugar and water. Still, 
 were I asked if I considered that the nectar in flowers consti- 
 tuted the principal food of humming-birds, I should answer 
 in the negative. Insects form their principal food. The 
 robust frames of these birds seem to require something more 
 solid to support life than the nectarcous dew abstracted from 
 flowers ; and I don't exactly see, if these birds do principally 
 exist on this kind of nutriment, how it is that they continue 
 to keep it pure in their own hot stomachs ; and then, by a 
 process unknown to us, convey it to the stomachs of their 
 gaping little ones." 
 
KXIT.AXATOKY IN'DKX. 
 
 ■i'2[> 
 
 111, perfectly right in his 
 
 Wiiterton was, as 
 Soveral experiments have been made since he was in CJuinna, 
 and they have proved the soundness of his n Msoning. 
 
 Mr. Webber succeeded in taming a rul»y-throatod hum- 
 ming-bird, which would come from any part of the room, and 
 eagerly drink sugar and water and hcaey from a china cup, 
 perching on the brim, and thrusting its long beak into the 
 sweet mixture as fai' as the very base. 
 
 The same naturalist after'"irds reared a pair of ruby- 
 throats, which weiu so completely domesticated that, 
 although they went way with theii comrades at thi^ usual 
 time of migration, they returned to the house in the follow- 
 ing season and brought their i lutes with them, so that i:« 
 time there was quite a fami'./ of hi. iiimin^' birds perched 
 upon the cups and drinking. 
 
 He found, however, that they reqviivd other food besides 
 nectar, and discovered that t).<n' were in the I it'nt of poising 
 themselves before spider-web, , aud delicately picking out the 
 spider from the centre of its web. Mr. Gosse, who (litiifid 
 several humming birds during his stay in Jamaica, found that 
 they required insects as well as syrup, and were especially 
 fond of small ants, that use 1 to crawl into the syrup, and 
 nearly cover its surface with their bodies. 
 
 I may here mention that Bimiti is the native name for 
 Humming-birds in general. Thus K'Ara-Bimiti is 'The Ara 
 Humming-bird,' as stated by Waterton. 
 
 I. 
 
 Ibibourou. — 1 can only find one Jay which in the least 
 agrees with Watertou's description ; it is Cyanocorax cristaUis. 
 But it is a North American bird, and cou.'d not live in 
 Guiana. 
 
 lauAXA. — See " Guana." 
 
 India Rubber Tree {Ficus e^«s.'/ca).~ Sometimes called 
 the Caoutchouc Tree. 
 
 M 
 
426 
 
 EXPLANATORY INDEX. 
 
 J. 
 
 Jabiru {Mr '.eria Americana). — The Jabirus, of which there 
 are several species, are closely related to the well-known 
 adjutant birds of India, and belong to the family of storks. 
 Large as are several of the storks, the Jabirus are still larger, 
 and come next to the ostriches in point of size. They may be 
 easily distinguished from other storks by the shape of their 
 very large beaks, which turn slightly upward at the point. 
 
 One species inhabits Australia, and from the account of 
 Dr. Bennett, who kept one in confinement for a long time, it 
 almost exactly resembles the South American species, especi- 
 ally in its extreme wariness. It is so cautious that even the 
 natives find great difficulty in procui'ing a specimen. 
 
 The bird became very tame, and was allowed to wander 
 about the house and yard. It made friends with the cook, 
 and when hungry, used to search for her. It never tried to 
 steal the meat, but waited patiently until it was cut up. Dr. 
 Bennett mentions that it would catch flies on the wing, and 
 that he has seen it, while squatting on the ground, with its 
 long legs doubled up beneath its body, dart its 
 bill into the ground, and drag out the larvie of 
 insects, mostly those of the great cicada. 
 
 On account of the black, bare head, the Dutch 
 of Surinam call it negro-cope, or black-head. 
 
 Both illustrations here given were sketched in 
 the Zoological Gardens from the same bird, in 
 order to show how completely the attitude alters 
 the aspect. The first shows it as it was viewed 
 from behind. The bird was then standing 
 quietly, resting the whole weight of its body on 
 one leg, and having the other slightly raised. 
 Its attention was roused by my movements as 
 I shifted about to procure a favourable view, and it kept a 
 wat'ihful eye upon me the while. 
 
 I VlUllU 
 
 (.ii.Nii:.Mi>i.ATivi;. 
 
( >n 
 
 KXPI ANATOUY INDKX. 
 
 427 
 
 The second view was taken from the front, when the Jabiru 
 was enjoying itself in the sun. While thus employed, it 
 kept up a continual shivering movement of the feathers, just 
 as has been related of the ibis. In its native state it is to be 
 
 .lAlllUL SUSNINi; ITM;1,I''. 
 
 found on the banks of lakes, marshes, and rivers, and feeds 
 upon the aquatic reptiles, tish, &c., which it finds in such 
 localities. 
 
 Every traveller in Guiana is sure to see this splendid bird, 
 and Mr. C. B. Brown often met Avith it. lie gives the 
 following description of the nestling : — 
 
 " I was greatly amused with the appearance of two young 
 but fully fledged Jabirus, which stood on their large, flat nest, 
 composed of sticks entwined together, on the branch of a, 
 large isolated tree, growing on the river's bank. They looked 
 like two shipwrecked mariners on a rock in mid-ocean, 
 waiting to be delivered from their lonely watch by a passing 
 ship. They stood there as if scanning the horizon, apparently 
 deep in thought, shifting their i)Osition now and then from 
 one leg to the other, or taking a solemn or stately stroll 
 round the confines of their nest. Thus we left them, to await 
 the time when their powers of flight would be sufiiciently 
 developed to enable them to go forth into the world and 
 forage for themselves. They were fully feathered with a 
 grey plumage, which on moulting would change to pure 
 white." 
 
 The ^ame author mentions, on another page, the habits of 
 
 1 (i 
 
 I ' I' 
 
 I 
 
 
 
 I 
 
 Hi 
 
 in 
 
 Ki 
 
4i»8 
 
 EXPLANATORY FN'DKX. 
 
 a domesticated Jabiru, wliich lieliavitd very much after the 
 fasliion of the bird described by Dr. Mennett. 
 
 "The people had a tamo Jabiru, which stalked about in a 
 most deliberate and stately way, as if every step it took was a 
 matter of deep deliberation. In spite of its nage looks there 
 was some fun in it, as it seemed to enjoy playing a practical 
 joke now and then on the village dogs j for whenever it had 
 a chance it stalked stealthily behind one and delivered a well 
 directed prod with its pointed beak, full on the dog's back, 
 which caused the dog to fly and the old Jabiru to 'smile 
 inwardly.' ' 
 
 If the Jal)iru can find a tall, pillar like rock, it is sui'e to 
 build its nest on it just as does the stork. 
 
 Schomburgk gives a touching account of a Jabiru which had 
 built its nest on the top of a tall basaltic pillar named Piere- 
 piapa, about fifty feet in he'ght. One young bird was in the 
 nest, and the mother, on setting the travellers, Hew to its 
 help, and stood guard over it. One of Schomburgk's men 
 saw the bird standing over its young, and shot it before he 
 could be checked. 
 
 The Jabiru is called by the natives 'J'ararama. Its flesh is 
 eatable, and is said to resemble beef, 
 
 Jacamars. — The species here described is the Great Jacamar 
 — {Jaca merops grand is ) . 
 
 The mode of feeding employed by this bird, and described 
 by C. Kingsley in At Last, exactly corroborates Waterton's 
 account. He had been startled by seeing a bat Hy past 
 at mid-day, the shade of the tropical forest being so deep and 
 dark. 
 
 " And there is another I No ; as it turns, a blaze of 
 metallic azure off the upper side of the wings proves this one 
 to be no bat, but a morpho, a moth as big as a bat. And 
 what was that second largo flash of golden green, which 
 dashed at the moths, and back to yonder branch not ten feet 
 off 'J A Jacamar — kingfisher, as they miscall htr here, sitting 
 fearless of man, with the moth in her long beak. Her throat 
 
EXPLANATORY INDEX 
 
 4-Jl) 
 
 is snowy white, her under parts a rich brown. Her breast 
 and all her upper plumage and long tail, glitter with golden 
 green. There is light enough it seems, in the forest." 
 The morpho, by the way, is a butterfly, and not a moth. 
 
 c;IU.Al' .(Ai AMAR, 
 
 Jaguar {Leopardus oncn). — Waterton did not see much of 
 the jaguars (which he sometimes calls tigers), and certainly 
 was not afraid of them. Mv. Brown gives the following 
 account of the mode in which the jaguar is successfully 
 hunted ; — 
 
 " On returning to the head of the New River for provisions, 
 we were followed for many miles by a Tiger, for on going 
 back we saw its huge tracks in the swampy places on our 
 path. 
 
 " With good hunting-dogs fine Jaguar and Puma hunting 
 might be obtained on the banks of this river, where without 
 doubt they are exceedingly numerous. 
 
 *' Many of the Indian hunting-dogs, trained for deer or 
 tapir will hunt Tigers. When on the track of either of those 
 animals, should they come across the scent of a Tiger, their 
 
 IT V 
 P ' V 
 
 5 :'■ 
 
 ;J:! 
 
 1^^ 
 
 il 
 
 -^'■ 
 
 m 
 
 I 
 
 I'll 
 
 
 
 l! 
 
430 
 
 EXPLANATORY IXDKX. 
 
 eagei* and confident mannei' of pressing on after the game is 
 immediately changed, and with the hair on their backs erect 
 they become cautious and nervous to a degree, jumping jit 
 ©ven the snapping of a twig. Abandoning the hunt they take 
 up the Tiger's track and follow it. But should the huntsman 
 call them from it, or not cheer them on with hie voice from 
 time to time, they exhibit great fear, and keeping close to his 
 heels cannot be induced to hunt any more in that district for 
 that day. 
 
 " On the contrary, if allowed to follow the Tiger, they track 
 it up with caution, being fully aware of the cunning dodge 
 practised by that animal ; which is, when the dog is close at 
 hand, to spring to one side and lie in ambush until it passes, 
 when with one spring the dog is seized. 
 
 "Ordinary dogs would fall a prey to this trap, but not the 
 self-taught Tiger dogs. Their fine powers of scent warn them 
 of their near approach to the quarry, when they advance with 
 great caution, never failing to detect the Tiger in time, and 
 when once their eye is upon their enemy it has no chance of 
 escape. 
 
 " In its pride of strength, the Jaguar scorns the dogs, and 
 with a rush like a ball from a cannon springs madly at one of 
 them, feeling sure that it cannot escape. It has reckoned, 
 however, without its host, for tiie dog eludes the spring with 
 ease, and with great quickness flies on the Tiger's flank, giving 
 it a severe nip. As the Tiger turns with a growl of pain and 
 disappointment, the dog is off to a little distance, yelping 
 lustily, and never remaining still an instant, but dnrting first 
 on one side and then on the other. After one or two in- 
 effectual charges the Tiger gives it up, and on the approach of 
 the hunter springs into the nearest suitable tree, which it 
 seldom leaves alive." 
 
 It is to be wished that Waterton had secui-ed an uninjured 
 specimen of a Jaguar .skin, on which he might have exercised 
 his unrivalled powers of taxidermy. 
 
 JrMno. — The negroes still retain tliis term for their chief 
 
EXPLANATORY INDEX. 
 
 431 
 
 deity, which, of course, is a malignant one. They croucli in 
 abject terroi' before Jumbo, who, according to them, takes 
 possession of birds {see Goatsucker), desolate swamps, and ho 
 forth. 
 
 The I'eally terrible Obeah or Fetish -worship of the negroes 
 is connected with Jumbies and Duppies (the spirits of the 
 dead), and the negroes are absolutely subservient to the 
 Obeah men and women, who can put Obeah into anything they 
 choose. Even the whites are not free from the fear of them, 
 and with reason, for they have an intimate knowledge of many 
 poisons, and will use them on either black or white. Stedman 
 mentions several inst;inces of such crimes, and it was nearly 
 certain that his faithful Joanna fell a victim to the insidious 
 poisons of Obeah. 
 
 K. 
 
 Kauartmiti. — «%e " Humming Bird. Ara." 
 
 Kessi-kessi. — A vast number of parrots are called Kessi- 
 kessis, and are very plentiful. They all may be referred to 
 the genus Comirus, and between forty or fifty species are 
 known. 
 
 They are always to be found among the forests bordering 
 on rivers, and aie very noisy among the upper branches, the 
 more so as th^y are sociable in their habits. Some species 
 make their nests in the hollows of trees, upon the decaying 
 wood, and several pairs of birds may be found in the same 
 cavity. Others build in trees. Their nests are large, toler- 
 ably globular in shape, made of thorny branches, and have 
 the entrance by an aperture at the side. 
 
 The best-known species of kessi-kessi parrots is known 
 scientifically as Conurus sohtHlaJh, and is one of the most 
 plentiful of the parrakeets. It is a very pretty creature, its 
 plumage being a mixture of orange, red, and green. 
 
 King of the Vultures. — See " Vulture, King." 
 
 KuRUMANXi Wax. — This is composed vi the wax of a wild bee 
 
 > I) 
 
 
 
 ' I 
 
 \\\ 
 
432 
 
 KXPLANATORY INDEX. 
 
 {Ceroxijlon andicola), mixed with a pitch-like substance ob- 
 tained from several trees, chiefly the Maara-tree. It is pitchy 
 black in colour, and when exposed to heat, looks almost 
 exactly like pitch. It is used for many purposes, and is 
 equally useful as a waterproof material or a cement. The 
 seams of canoes are always " payed " with Kurumanni wax. 
 
 L. 
 
 Labarri [Elaps lemniscatus). — As is the case with most 
 serpents, the Labarri is a beautiful creature during life, but 
 shortly after death the various colours which play over its 
 skin begin to fade, and soon perish entirely. 
 
 Labba, sometimes spelled Lapo. — One of the Cavies (t'tdo- 
 genys paca) of which the common Guinea-pig is so familiar an 
 example. The name Calogenys, or Hollow-cheek, is given to 
 it on account of its enormously developed cheek-bones, and 
 lai'ge cheek-pouches. Its flesh is very delicate, being as ricl^ 
 in fat as that of the Acouri is destitute of it. The mode of 
 hunting the Labba is described bv Mr. C. B. Brown in his 
 work on Guiana. 
 
 " The flesh of the Labba is considered the most delicate of 
 all bush animals, and is therefore much sought after. This 
 animal is a rodent about the size of a hare, but with a stouter 
 body, and more rat-like head. It is nocturnal in its habits, 
 spending the day in its burrow under tree-roots near the edges 
 of rivers. These sleeping-chambers are not very long, and 
 have two entrances. The huntsman rams a thin pole in at one 
 end, which startles tlie occupant, causing it to fly precipitately 
 from the other, like a ball from a cannon, closely followed by 
 the dog. 
 
 " The Labba, when close pressed, takes to the water, and 
 while swimming is hunted down by the Indian sportsman in 
 his wood skin. Sometimes the animal escapes by diving, and 
 hiding amongst fallen brushwood on the river's edge, but more 
 frequently falls a victim to the arrow of the Indian." 
 
^'H 
 
 EXPLANATORY INDEX. 
 
 4B;-} 
 
 
 Its colour is dark brown, with four rows of white spots on 
 each side. Th(3 fur is considered useless. 
 
 Lemon [Citrus liinonum.) — This tree belongs to the same 
 genus as the orange, and there are nearly as many varieties of 
 Lemons as there are of oranges. 
 
 Letteu-wood {Broshnum aubletii). — The wood of this small 
 tree is intensely hard and very heavy. It derives its popular 
 name from the colour f the wood, which is deep brown 
 scribbled over with reddish marks looking something like 
 Persian or Arabic letters. In consequence of its weight and 
 hardness it is much used for the ** shimara sherie " of the 
 long fish and hog arrows, q. r. 
 
 In this tree, only the heart-wood produces the beautiful 
 markings, and in a tree of twenty inches in diameter the 
 heart-wood measures barely seven inches. It runs to about 
 eighty feet in height, and is rather scarce. The natives call it 
 by the name of Buro-koro, or Paira. It may be here men- 
 tioned that the native name for all heart- wood is Tacouba. 
 
 Locust-Tree [llymenaa courharil). — For its use in canoe 
 making see Purple-heart. 
 
 The bark of this tree has but one defect. It is rai'ely more 
 than a quarter of an inch in thickness, and so is easily bent 
 to the required form. But it is heavier than water, so that if 
 the canoe be upset, it immediately sinks, and cannot be raised 
 without difficulty. 
 
 It is from the Locust-tree that the gum animi of commerce 
 is obtained. Very often, when the tree is dead and decayed, 
 enormous quantities of the gum may be found on the spot 
 which it had occupied. The wood makes excellent furniture, 
 being hard, heavy, brown streaked with veins, and taking a 
 fine polish. The tree is also known by the names of Simiri 
 and K'wanarri. 
 
 n> 
 
 m 
 
 1,1 
 
 r F 
 
434 
 
 KXIT.ANATORY IXDKX. 
 
 M 
 
 I 
 
 Maam. See " Tinamou." 
 
 IMaho.sany {SimctoHiamahiKjoni.) — Tho tree which furnishes 
 the wt'li-kiiown mahogany wooil is a vory fine one, some sjieci- 
 mens being nearly a Inmdred feet in heiglit. It belongs to the 
 group of Cedraceio. Tho timber is espi'cially valuable, because, 
 no matter how largo or long it may be, it is almost invariably 
 sound throughout, and free from 'shakes.' A single log has 
 been sold for a thousand pounds. Its uses were discovered 
 by Sir W. Kaleigh. 
 
 Rosewood, which comes from the same country is a Mimosa, 
 several species of which furnish tho wood. It derives its 
 name fi'om the rose-like smell of the freshly-cut timber. 
 
 Manakins. — These are nearly all small birds, the Cock of 
 the Rock, q.v., being the largest of them. In England, we 
 have an example of the Manakins, namely, the well-known 
 Waxen Chatterer {Ampelw garruJa), so celebrated for the wax- 
 like appendages to several of the feathers. All the manakins 
 are comprised in the group of Piprinsc. 
 
 Maribunta. — This is merely a Portuguese word signifying a 
 Wasp, and is applied to all wasps indiscriminately. 
 
 Maroudi. — There are several species of Maroudis, those 
 which are best known being the common maroudi (Penelope 
 cristata), and the white-headed maroudi (Penelope ]>{pile). Of 
 these birds, Mr. C. B. Brown writes as follows : — 
 
 *' The white-headed maroudi makes an extraordinary rattling 
 noise with its wings in early morning and late in the evening, 
 evidently amusing itself, or following a custom of its kind, for 
 when it likes, it can fly noiselessly enough. 
 
 " I examined their wings, and found that the males have four 
 curiously shaped feathers at the tip of the wing, with which 
 they make this noise. The end portion of these feathers is 
 stiff, with very short pennules. The white-headed females 
 have only thiee of these feathers in each wing, which are not 
 
KXPLANAToliY 1X1 MA. 
 
 435 
 
 so intensely modilied as in the male ; while the male of the 
 common kind lias only two of those feathers in each winjr, 
 which are modified in a lo.-s degree than tlio.si' in the females 
 of the white headed species." 
 
 'J ■': ' '1 
 
 1.4 
 
 wiini; iri;.\iii:ii MAiioriii. 
 
 If the reader will refer to the note on Pee ay-n:an, lie will see 
 Ihat the Marondi is thought to be a bird of good spiritual 
 influence. 
 
 M( ;<.'<) MOCo TiiEE {CuUxidratti arhoreKceiix.) 
 
 Monkey, ^vown {Cehis Apella). — This monkey is well 
 known in England, and is very intelligent. Some years ago 
 there was one in the Zoological Gardens who had been taught 
 to crack nuts with a stone, if he found them too hard for his 
 teeth. He taught a companion the same art, and it was most 
 absurd to see the two sitting side by side, and cracking nuts 
 jilternately. Tltey never seemed to quarrel about the pos- 
 Fes.sion of the stone, but as soon as one had cracked a nut, ho 
 put down the stor.e nnd the other took it up. They had worn 
 the stone quite smooth by continual u.«c. 
 
 F V -1 
 
 \'V' 
 
 i\''' 
 
;;r, 
 
 KXI'r,ANATuUV INDKX. 
 
 Another posse.ssed the rather d ingorous fiiculty of ligliting 
 liicifor-matchos, and knox-king them about with his liands 
 while blazing. Nothing pUiased him so much as a (laming 
 "fnsee," as it blazed more tiercely than an ordinary match, 
 and lemained alight much longer. The sides and back of his 
 cage were covered with the marks of the matches which he 
 had lighted. 
 
 Other monkeys employ artificial methods of nut cracking 
 as is shown by Mr. C. J J. Brown. 
 
 He had found a number of Brazil nuts on the ground en- 
 closed in their hard, shelly cases. " My men used to open 
 
 I 
 
 I 
 
 1. 
 
 ilKUVN .MiiNKl:v. 
 
 them by chopping oil' their ends with a cutlass, which, owing 
 to their hardne.ss, was no easy operation. The quatas, or 
 large black spider monkeys, spent a good deal of their time in 
 trying to open them by beating them against the bnanches of 
 trees, or on hard logs upon the ground ; and as we passed a 
 grove of Brazil-nut-trees it was amusing to hear the hammer- 
 ing sounds produced by these fellows at their self-imposed 
 tasks. Where a single monkey was thus employed the blows 
 were most laughably • * few and far between,' the creature 
 showing its trxie indolent character by the slow way in which 
 it performed its work, resting for a few minutes between every 
 blow. It also showed an amount of perse v-e ranee, however. 
 
EXI'LANATORV INDEX. 
 
 4:il 
 
 that one would nut look for in a monkey, and a knowledge 
 that it would eventually reap a reward for its hard labour. 
 
 "Cloodnoss knows how long it takes one of the>e monkeys 
 to break a nut-case ; but the time mu.st be great, for on one 
 occasion, we got quiotly aiiiongst .i. lot of the nut breakers, 
 and secured a nut case which one in it.s luu-ry had left upon a 
 log, and which was worn smooth by tho friction of the mon- 
 key's hands. This ii.id evidently been pounded for a length of 
 time, but showed no signs of cracking. Its natural up rturt' 
 was large enough to allow tho monkey's linger to touch t'u"' 
 ends of the nuts inside, which were picked and worn by its 
 nails. Near the same place wo saw a nut case split in two, 
 on the flat surface of a large granite rock, that had evidently 
 been broken by a monkey, for there weie no Brazil nut trees 
 from which it could have fallen, overhanging the i<pot." 
 
 There are several species belonging to the genus Celfiin, and 
 they are called by the general name of Capucins. 
 
 jMoNKiiVs AM) MissiLKS. — Watcrtou, as will hero bo seen, 
 entirely denies that any monkey can use a missile, and recurs 
 to the subject in one of his essays. In this article he otters to 
 accompany any one to any collection of monkeys, and to supply 
 the animals with stones, tiles, lead, pewter-pots and sawdust, 
 all of which articles he has been told had been used as missiles 
 against human beings by monkeys. He otters, in his own 
 amusingly trenchant style, to give the monkeys every oppor- 
 tunity v\' iiurling these objects at him, and that if one of 
 them does so, he will admit that the knowledge which he had 
 acquired " duiung a long sojourn in the forests of Guiana {the 
 native haunts of monkeys) is rotten, and not worth one single 
 farthing." 
 
 There can be no doubt that Waterton is perfectly right in 
 denying that a monkey eitlier would or could, throw stones or 
 any missiles by way of defence. INlost monkeys can catch a 
 missile, and many are adroit enough to catch flies on the wing. 
 But, although thoy can catch, they cannot throw. Still, they 
 may be able to drop branches, i^c. from the tops of trees, not 
 
 I'r 
 
 |:i'| 
 
 ; :'! ■ 
 
 m 
 
 II: 
 
 M 
 
HA 
 
 KXI'l.AXATtiUV I\|)i:\, 
 
 beciuKso tlio branches luo ikiul and brittle, and havo been 
 accidentally broken by the passage of tho monkeys anionj,' 
 them, but with the deliberuto intention of driving away a 
 supposed foe. 
 
 Waterton never saw such a feat performed, and ho is right 
 to Kay so. r.ut there are other travellers quite as worthy of 
 credence as Waterton, who detinitely state that they have been 
 eye-witnesses to sut-h a proceeding. No one, I would presume, 
 would impugn a direct as.>-ertion of Mr. Alfred L. Wallace. 
 Vet in his well known work on tho iAlalay Archipelago, among 
 the islands of which ho was continually travelling for moro 
 than seven yenrs, he has the following observations : — 
 
 "I afterwards shot two adult females and two young ones 
 of different ago.s, all of which 1 preserved. 
 
 "One of the females, witii .several young ones, wa.s feeding 
 on a Durian tree with unrijo fruit ; and as soon as she saw 
 us she began breaking ott' branches, and the great spiny 
 branches with every appearance of rage, causing such a shower 
 of missiles as effectually kepv us from approaching too near 
 the tree. 
 
 " This habit of throwing down branchea when irritated has 
 been doubted, but I have, as here narrated, observed it my- 
 self on at least three separate occasions. It was, however, 
 always the female Mias who behaved in this way, and it may 
 be that the male, trusting more to his great strength and his 
 powerful canine teeth, is not afraid of any other animal, and 
 does not want to drive them away, while tho parental instinct 
 of the female leads her to adopt this mode of defending her- 
 self and her young ones." 
 
 Perhaps it may be said that Wallace travelled in the Malay 
 Archipelago, and wrote of the orang-outan, while Waterton 
 travelled in Guiana, and wrote of the Coaita which inhabits 
 that counti'y. Still, his sweeping assertion included all 
 members of the monkey race, and moreover, a traveller and 
 naturalist, who spent much time in (Juiana, writes as follows 
 of tho Coaiti, or Quata : - 
 
 L 
 
i:\PI, ANATUliV iM)i:x. 
 
 489 
 
 *' When eiif,'figeil in tho forest, cutting our portage, we were 
 frequently visited by bands of large Coiata monkeys, which wero 
 very numerous on the bunks of tl\e upper part of this river 
 (the Ess('(piibo). They were the black bodied, red faced kind 
 of large spider monkey, and uttered a sort of barking grunt. 
 Their cry, when calling to each other, has a wailing sound, 
 and is very loud. 
 
 "On seeing us, they used frequently to hurl down large 
 dead branches, some of which came rather too close to our 
 heads at times to be comfortable. The manner in which 
 they performed this was singular ; they held on by tail and 
 hind-foot to a live bough in a tree top, alongside of a dead 
 one, and pushing with their hands with all their force against 
 the latter, generally succeeding in breaking it oft", when down 
 it came." (0. B. Brown, i'unoe and Vamp Lift In (luiaiia.) 
 
 Mr. Brown's statement appears to be too definite to be re- 
 jected as unworthy of belief. Putting it by the side of the 
 observations made independently in the Malay Arcliipelago, 
 it seems to prove beyond doubt, that monkeys can intention- 
 ally drop missiles in order to annoy intruders, though they 
 cannot throw them. 
 
 MouA {Mora exctha). — One of the many giants of the 
 vegetable kingdom which are found in (Juiana, sometimes 
 attaining a height of two hundred feet. 
 
 Mr. C. B. Brown gives the following description of this fine 
 tree : — 
 
 "The band of INIora-trees lining both banks of the Esse- 
 quibo continued along it as far up as we went, but did not 
 grow on the banks of the llupununi up to Pirara landing. The 
 Mora grows only on a moist soil along the borders of the river, 
 forming a band on both sides of varying width, according to the 
 breadth of the band of flooded country during the rainy sea- 
 son. In places it is often not more than 100 yards in width. 
 Up many small side streams the i\Iora has marched to their 
 sources, while along others it has not attempted to spread. 
 
 1 1: 
 
 
440 
 
 EXPLANATORY INDEX. 
 
 By their solid-looking tops, composed of massive dai'k greon 
 leaves, and their great height, they attract attention and 
 excite admiration; and when portions of their tops are 
 sprouting new leaves, which are of a light liver colour at first, 
 turning red, and then gi-een, a forest of them presents many 
 varied tints. Each branch throws out shoots and gets new 
 leaves at all times of the year, independent of other portions 
 of the same tree." 
 
 The seeds of the ]\Lora are rather pink inside, and used by 
 the natives as food, beiiig boiled, grated, and then mixed with 
 cassava me?l, giving it a brown colour, but a pleasant and 
 sweetish tas"e. Three varieties of this tree are recognised, 
 namely, the Red Mo"a, White Mora, and Mora Bucquia. The 
 timber is excellent for ship-building, and is one of the eight 
 fii'st class woods at Lloyd's. 
 
 Mosquito {CtiJex jjipiens). — Several species of Gnats are 
 called by the name of Mo r^uito, which signifies a little fly. 
 They are found all over the world, and infest hot and cold 
 countries equally, though their bite seems to be fiercer in 
 the former than in the latter. 
 
 Even in England the Mosquito can be more than unpleasant. 
 I have had my right hand laid up for a long time by the bite 
 of a single Mosquito just at the base of the thumb. The 
 insect settled on my hand in broad daylight, and I killed it 
 as soon as I felt the prick of its beak ; but the mischief was 
 done, and I had to cari'y my arm in a sling and have the 
 hand covered Avith ice for sev'eral days. As to the hand, it 
 looked more like a discoloured boxing-glove than a hand. 
 
 It is only the female Mosquito which bites, the male being 
 perfectly harmless. 
 
 There is a larger and longer legged species of JNIosquito in 
 Guiana. Its scientific name is Cuhx ji^'licidaris, and it is 
 popularly known as Gally-nippor. 
 
 Musk Duck {('alrina moschata). — Popularly, but wrongly 
 called the Muscovy Duck. 
 
EXPLANATORY INDEX. 
 
 441 
 
 N 
 
 Nandapoa. — Sometimes called Negi'ocopo, i.e. Black-head 
 (Tcmtalna loculator), a species of Ibis. The Jabiru, q. v., is 
 also called Negrocope. 
 
 Mr. C. B. Brown's observations on this bird agree exactly 
 with Waterton's passing description. He mentions that 
 *' they are white, with black wings, and are frequently set n 
 soaring high in the heavens in circles, mounting ligher and 
 higher till they appear like mere specks." Th'' head and 
 neck have no feathers, and are covered with a black, wrinkled 
 skin. 
 
 NoNDEsc'RirT. — This wonderful specimen of Waterton's 
 skill in taxidermy is formed from the head .md shoulders of 
 the Red Howler monkey, q. v. Jn manipulating it, Wateiton 
 has so rijodelled the skin that he has discharged from the face 
 every vestige of the original features, and has substituted 
 those of a .nan, grotesque enough, but still human. As bare 
 ski 1 always becomes black when dry, the contrast of the 
 black face with the tiery red hair has a very striking eil'ect 
 and adds to the resemblance. 
 
 In his witty review of the Wanderings, to which reference 
 has already been made, Sydney Smith says of tlie frontispiece, 
 which represents the Nondesci-ipt, " Upon stutting animals wo 
 have a word to say. Mr. Waterton has placed at the head of 
 his book the picture of what he is pleased to consider a non- 
 descript species of monkey. In this exliibition our author is 
 surely abusing his stufllng talents, and laughing at the public. 
 It is clearly the head of a ^Master in Chancery — whom we 
 have often seen backing in the House of Commons after he 
 has delivered his message. It is foolish thus to tritle with 
 science and natural history." 
 
 The principal difficulty in preparing this grotesque head 
 lay in the change of tlie facial angle from that of the monkey 
 to that of tlie man. This could not liavo been done if the 
 
 
 ■^N, 
 
 I'..-:r 
 
442 
 
 EXPLANATOUV INDEX. 
 
 ! 
 
 IJ 
 
 skull, or any part of it, had beeti allowed to remain, and the 
 really wonderful feat could only be porfoniied by Waterton's 
 system of removing the whole of the bf)neH, and paring down 
 all the bare skin until it was not thicker than ordinary writing- 
 pfiper. 
 
 The drawing of the head, howevisr, Hcarcely does justice to 
 the original, for Waterton made the nose much more acpiiline 
 and thinner than is shown in the engraving, which also 
 makes the face look as if it were hal:'«/ whereas it is abso- 
 lutely bare. If any visitor to L'shaw (J(dh'go can manage to 
 obtain a profile view of the Nondesr-ript, hi; will bo greatly 
 struck with the ingenuity which lias changed the flattened 
 nose of the monkey into the acjuilijio nose of a lunnan being. 
 
 Many persons indeed, on seeing th(j Nondescript, really 
 thought that it was human, and said that Waterton ought not 
 to have been allowed to kill n:itives in order to show his skill 
 in preserving their skins. 
 
 KuTMWi {}fjjriHt}c(i ninschatd.) — ]\Iuce is the scarlet envelope 
 which surrounds the seed. It becomes reddish-yellow when 
 dry. 
 
 () 
 
 ill! 
 
 Ol(ju, sometimes written Oolu, — The tree is a lai'ge one, 
 running to eighty or ninety feet in height, and found in loose 
 sandy soil on the li^ssequibo. It produces s(!ented gum much 
 like that of the hayawah, and stiems to bo also a species of 
 Ic'ica. The wood is reconnuended for wardrobes, entomological 
 cabinets, and tlie like. 
 
 Opossum {Chaironectas laywc/;).- -Sometimes, on account of 
 its predatoi'y habits and its semi-a(piatic life, it is called 
 the Demerara Ottei'. Bufl'on also terms it an otter. 
 
 It is really a curious being, and has greatly puzzled syste- 
 matic naturalists. Most of tlu; opossium tribe inhabit trees, 
 but the Yupock pas.ses the greater piul of its time in the 
 water, and is never found far from rivers. The feet are 
 
EXPLANATORY INDEX. 
 
 im 
 
 wobbecl, the cheeks jiio fiiniislied with large pouches, uixd the 
 fore-feet look exiictly as if they htid six toes instead of 
 five. 
 
 It is rather pi'ettily marked, the ground colour being grey, 
 on which are bold black marks. In the young, the grey is 
 Ighter and the black dai-ker than in the adult animal. 
 
 ^fit^S: 
 
 :i' 
 
 
 ol'o.-irtUM. 
 
 Orange (Citrus arrant I ion). — Those who have travelled in 
 the West Indies say that the Orange cannot be appreciated 
 unless eaten oft' tho "^ vee, and that the best "China" Orange 
 of our shops is but ''iul brown sugar and water to the Orange 
 Avhen fresh. 'VI, e on] v ilrawback is, that the aromatic oil of 
 the peel* is so powe'fui that the lips are blistered, unless 
 every particle of p< f>l be removed. 
 
 Otter {Pferr.inra Sanhaa.'ri^. — The following account of 
 this animal is f^Iveii by Mr. C. B. 13rown : — 
 
 "We frequently came up with parties of Otters, composed 
 of from live to eight individuals, which tried to dispute the 
 passage of the river with us. They used to approach the 
 canoes, and all pojiping their heads out of water together, 
 would snort and snr.rl as if trying to intimidate us. Seeing 
 that our advance was not tj be checked, ihey took long dives 
 away, and (piickly disappeared i^eyond a bend in the river, to 
 l)e seen no more. They wcne of a brownish colour, with a 
 
 1;;; ,' "^ 
 
 1' ' * ■ 
 
 \\\m 
 
44i 
 
 KXPLANATUllY INDEX. 
 
 i 
 
 patch of white under the tlii'oat. I coukl not resist the temp- 
 tation of having a shot at one occasionally, but never could 
 tell whether it took effect, for, as they always disappeared 
 beneath the surface at the instant of the report, it was difh- 
 •cult to say whether they had dived at the flash or wore killed 
 and sunk to the bottom." 
 
 The length of one which was shot by jNIr. Brown was five 
 feet four inches, nearly two feet moie than the average 
 length of our English otter. The buri'ows seem to have two 
 entrances, one opening into the river and the other leading to 
 the bank. 
 
 OuiiAil (Arumlinaria tScIiomlnmjIiH). — Waterton did not 
 know the reed called Ourah, neither did the Macoushies, 
 who were in the habit of purchasing it. It is a very local 
 plant, and according to Schomburgk, is only to be found on a 
 sandstone ridge of the Upper Orinoco River. In some respects 
 it resembles the bamboo, and, like that plant, grows in thick 
 clusters, with long slender branches that wave in the wind 
 like magnified tufts of grass. 
 
 Nowhere is the stem more than half an inch in diameter, 
 and the first joint, which is used for the blow-gun, is fifteen 
 or sixteen feet in length, without a single knot, hollow, and 
 polished Avithin. Though its walls are very thin, they are 
 of great strength, owing to the tubular form of the reed, and 
 are thus able to uphold the slender branches, which sometimes 
 I'each forty feet in length. 
 
 Pacou {Myletes />rtc??.).— How this fish is taken by means 
 of poisoning the water is described under " Wourali.'' 
 Poison, however, is not absolutely necessaiy, for the Pacous 
 have a fashion of coming to the surface of the water, 
 showing their heads and parts of tbeir backs, and then dis- 
 appeai'ing. Waterton used to shoot pike in the same manner 
 as they came to bask on tlie surface of the water, after tlie 
 
>'l 
 
 EXPJ.AXATURV INDEX. 
 
 Ai'> 
 
 custom of their kind. The fle.sh of the Pacou affords excellent 
 food. 
 
 Paddle. — The Paddles are really curloua objects. They are 
 made from the wood of the Paddle-tree {Asj)idosj)ermuni excel- 
 suin), a most strange-looking tree. It runs to a considerable 
 height, and the outline of the trunk is most rt narkrible. 
 
 The reader will remember that che Ceiba (see p. 395) has 
 the lower part of the trunk modified into buttresses, but 
 the Paddle-tree seems to bo all buttress, and bears a curious 
 lesemblance to the clustered pillars found in some of our old 
 cathedrals. Indeed, the section of the tree looks very much 
 like a piece of one of those intricate puzzle-maps and pictures 
 which used to be found in the toyshops. 
 
 As the wood is soft while fresh, an Indian, when he has to 
 make a new paddle, splits off one of the "flutes," as these 
 buttresses are called, trims it carefully into shape, and then 
 hands it over to the women, who paint it in divers patterns 
 of black and red. 
 
 The Paddle wood tree is called by the natives Yarari or 
 Massara. When dry, the wood is very light, very elastic, very 
 hard, and very strong. The oddly-shaped tree averages sixty 
 or seventy feet in heiglit, and tive feet in diameter. A good 
 section of it is in the Technological Museum of the Crystal 
 Palace. 
 
 Pai'I'aa\ or Pai'aw Thee (Carica j^apaya). — This tree is 
 planted by the natives near all their permanent settlements, 
 and is seen in company with the cotton and red pepper. It 
 not only furnishes an edible fruit, but possesses the singular 
 property of making tough meat tender when rubbed with the 
 acrid juice of the unripe fruit, or even with the leaf. In fact, 
 as Tom Cringle says, it can convert a piece of bull's hide into 
 a tender beef-steak. 
 
 Paint, Red, used by Natives. — The natives are fond of 
 decorating their bodies with paints during feasting times. Eed 
 and black are the two chief colours. Red is obtained from the 
 seeds of the Arnotto plant {Bixa oreJlarm), and the black from 
 
 i;;t. 
 
 t i;ii 
 
 I \ 
 
 v ' 
 
lil I 
 
 i:: I 
 
 ;f? 
 
 
 "' 
 
 f: 
 
 440 
 
 EXPLANAToliV INDKX. 
 
 the juico of the fruit of the Lana tree (fienipa Americana). 
 The Bixa is the phmt that furnishes the annate dye, witli 
 which Ave colour our clieeses. The natives call tlie paint by 
 the name of Rucu. 
 
 
 -li^^i- 
 
 
 ^ 
 A 
 
 w^ 
 
 ^m 
 
 
 Parima, Lake. — Ever since the time of the great adven- 
 tiiier, Rileigh, there has been great discussion as to the 
 Lake Pu-inia and El Dorado, or the Golden City. Some 
 geograpi.i.rs have denied the existence of either lake or 
 city, wI;;Io others hive b^en so certain as to the former, that 
 a map la my possession places it about forty miles north-east 
 of Fort St. Joachim, in the loop of the Tacatu River, 
 between Lat. 4 and 5 N. 
 
 Waterton could find no traces of it, and the inliabltajits of 
 Fort St. Joac'h'm had not been more successful. Schomburgk, 
 
 
EXPLANATORY INDEX. 
 
 447 
 
 wlio went much by the ti'acks of Waterton, reached the very 
 f-pot, and found that Waterton' s cautious suggestion was more 
 than correct. He has given a beauciful drawing of the spot, 
 which would have been the centre of the lake, had it existed, 
 but on which was a small settlement of native huts. Of the 
 lake he writes as follows : — 
 
 " The vast savannahs upon vhich St. Pirara is situated, aro 
 encompassed by the Pacaraiina mountains to the north, the 
 Canoku and Carawaimi mountains to the south, the thick 
 forests of the Essequibo and isolated mountains, to the east, 
 and the mountains of the Mocajahi, and branches of the Sierra 
 Parima to the west; and, according to a superficial computa- 
 tion, cover a space of fourteen thousand four hundred square 
 miles. The geological structure of this rogion leaves but little 
 doubt that it was once the bed of an inland lake, which, by 
 one of those catastrophes of which even later times give us 
 examples, broke its barriers and forced a path for its waters 
 to the Atlantic. May we not connect with this inland sea 
 that fable of El Dorado and Lake Parinia 1 " 
 
 Thus Schomburgk has proved that Waterton's conjecture 
 was correct, and that we may erase El Dorado and Lake 
 Parima from our maps and memories. 
 
 Parrot, Six, or Hia-hia {Drrof>/pis corovatiis). — It derives 
 its popular and approi)riate name from the way in whioh 
 the feathers of its neck can be spread into a sort of flaming 
 glory round its head. There is a splendid specimen in 
 Waterton's museum. 
 
 Pataca. — There seems to be no possibility of identifying 
 this bird. No description is given, and its name is only once 
 casually mentioned. As it is named in connection with the 
 maroudi, it may belong to the same group of birds. iSce 
 " Maroudi." 
 
 Peccary, sometimes called Wild Hog (Bicofelcs tajacu). 
 — This is not a largo animal, weighing only some fifty or 
 sixty pounds, but it is greatly dreaded, not only by the 
 n.atives, but by Europenns. The tusks of the boars are so 
 
 :. i 
 
 W 
 
 
 liW 
 
t! 'i 
 
 i 
 
 418 
 
 EXIM.ANATOUY IXDKX. 
 
 small that they can scarcely be seen beyond the lips, but they 
 ai'e teiTlbly sharp, cut like lancets, and can be wielded with 
 singular swiftness and force. 
 
 Moreover, the Peccaries, until they have learned respect for 
 man from long experience, seem to bo entirely without the 
 sense of fear, and to bo in the habit of taking offence at the 
 least sound, and charging at the spot. There is scarcely a 
 hunter who has not been forced to climb into the branches of 
 
 trees in order to escape a herd of Peccuriea, and even when 
 they have driven him into the tree, they will sit round it, 
 gnashing their tusks in anger. The sound of the clashing 
 tusks is well known to hunters, and warns them to prepare 
 for a cluu'ge. 
 
 The jaguar preys upon them, but is afraid of them. Tliey 
 always go in herds, and if one be attacked, the rest are sure 
 to come to the rescue. So, the jaguar silently follows the 
 herd, strikes down a straggling Peccary and makes for the 
 nearest tree. Thei'c it remains until the Pectaries aie tired 
 

 i:XI>LAXATOI{Y IXDKX. 
 
 -II 'J 
 
 of waiting for it, and it then descends the tree, finishes the 
 Peccary, and then follows the herd in search of another. In- 
 stances have been however known where the jaguar has not 
 had time to escape, and has been cut to pieces by the lancet - 
 like tusks of the Peccaries. 
 
 Two species are known in Guiana, one the Collared Peccary, 
 having a white band over the neck, and the other the White- 
 lipped Peccary, which has a white streak upon the jaws. Both 
 
 species have on the back n fetid open gland, which must be 
 cut out as soon as the animal is killed, as if it were allowed 
 to remain, the flesh would be uneatable. The White-lipped 
 Peccary is larger than its relative, fiercer, and more dreaded. 
 The natives use the white tnsks for necklaces and other 
 ornaments. 
 
 Pee-av-max. — The word is spelt variously by different 
 travellers, some using the word Piaiman. The pronunciation 
 however is the same in both cases. 
 
 The Pee-ay-man, or sorcerer, is a very great man indeed, 
 like the medicine man of North America or the prophet of 
 South Africa, and has to endure a curiously similar ordeal of 
 bodily torture and exhaustion before he can be admitted to 
 the coveted rank. Tlie mode in which he exercises his art is 
 narrated by Mr. ('. B. Brown, lie had procured a guide who 
 
 '''tit 
 
 m 
 
 
M. 
 
 hi 
 
 
 I' : 
 
 1 1;|| 
 
 450 
 
 KXPLANAToltV IXDKX. 
 
 was full of wild legends relating to the locality, and would 
 insist on telling them. 
 
 " He turned out to bo a famed sorcerer n- Piaiman ; and at 
 a village called Itabay, where we stopped one night in return- 
 ing, ho left the house in which we had put up our hiimmocks, 
 telling the interpreter to inform me that his absence for tho 
 night was unavoidable, owing to his li > ing to go up amongst 
 the mountains to roam about for the night, whilst his good 
 
 i'i:i: AV MAN. 
 
 spirit remained in one of the houses to cure a sick man, 
 who had demanded his good offices. 
 
 " In tv/o minutes after he had left us his powerful voice was 
 heard making the most discordant sounds imaginable, chanting, 
 howling, coughing, and many other diabolical noises, to these 
 were added the shishing sound of an instrument called a shak- 
 shak, made of a small round calabash filled with seeds, and 
 [>laced on a handle, by which it is shaken. 
 
 i:li 
 

 KXl'f.AN'AToIlY 1\I»K.\. If.l 
 
 "All the llrcs in tlio house had heeii put out iu anticipation 
 of his arrival, and tlio \Anvv, was pitch dark, so that the illu- 
 sion that he was not there in the flesh might he kept up. 
 Owing to the noise \ did not get a wink of sleep until about 
 two o'clock in the morning, and often thought, during the 
 night, what a good thing it would have heen if his spii'it had 
 only accompanied his body to the mountains. 
 
 " The house being close to the one ] was in, the interprtttr 
 could hear all If said, and nt my reijuest, but with o.idcnf 
 reluctance, told i vhat it was. It seems that he entered the 
 hou.se silently, then commenced the sort of din above 
 
 mentioned, beating with a palm bianch on the floor. He then 
 asked in a deep sepulchral voice what it was that ailed the 
 patient, to which the sick man's wife responded that some evil 
 spirits or kanaimas had ' done him bad.' The sorcerer then 
 .said, * Well, T don't know whether I can cure him, but I will 
 do my best.' 
 
 "He then called to his aid the good spirit of a bird, called 
 the maroudi, the descent of which from the roof was made 
 known by the shaking of the palm leaf raised up and graduftlly 
 lowered to the floor. 
 
 " On the arrival of this good spirit, it at first complained of 
 having had a long journey, and that it was much heated by 
 the haste it had been obliged to make when so suddenly .sum- 
 moned. It made the usual amount of noise, interlarded with 
 whistlings (in imitation of the maroudie's call), and promised 
 to do its best. When its best, which consisted of discordant 
 sounds, had been done, its egress through the roof was made 
 known by the shakings of the palm- leaf, and a dead silence of 
 a minute's duration succeeded, when the fluttei'ing of the leaf 
 announced the arrival of the spirit of an alligator, which in 
 its turn was replaced by a duraquara, and so on." 
 
 Peoall. — A native basket, very light, elastic, and strong. 
 
 Pelican (Pefecavns nnocrotahis). — Several Pelicans are 
 known, such, for example, as the frigate bii'd, and all of 
 them have the under part of the skin of the neck and throat 
 
 (1 (i '1 
 
 ii I 
 
 
 m 
 
 n* 
 

 -> 
 
 IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 v- 
 
 
 
 z 
 
 1^ 
 
 ^ 
 
 1.0 
 
 I.I 
 
 I&WIA |25 
 
 |jo "^^ JHIH 
 
 1.25 
 
 U |K6 
 
 FtiotQgrdphic 
 
 Sdmces 
 
 CorpQratiQn 
 
 ^ 
 
 as WIST MAIN STIHT 
 
 VVIUTIR,N.Y. t4SM 
 
 (7l*)t73-4S09 
 
 
fc iH 
 
 If; 
 
 ■111 ■ ■■ 
 
 i'Wi 
 
 IS 
 
 452 
 
 KXPLAXATUllY IXDKX. 
 
 modified into a pouch, which can be greatly d'stended. It is 
 chiefly nsed for carrying fish to its young, and being largo 
 enough, when fully distended, to hold two gallons of water, it 
 can carry a plentiful supply. 
 
 Keepers of travelling menageries are in the habit of exhibit- 
 ing the capacity of the pouch by thrusting their bent arms 
 into it. Yet when it is not needed for use, it can hardly be 
 seen, so elastic are the membranes of which it is composed, 
 and the pelican only looks like a bird with a long, straight 
 beak. 
 
 As far as can be judged from its behaviour in captivity, it 
 is a playful bird, but rather quick tempered. While watching 
 the pelicans at the Zoological Gardens, I was able to secure a 
 few of their characteristic attitudes. 
 
 Fig. 1 represents a l)lrd that had just been fed, and was 
 resting after its dinner. 
 
 Fig. 2 is the same bird as it appeared at feeding time, when 
 it was scolding another for eating a fish which it wanted for 
 itself. 
 
 Fig. 3 shows the attitude of a pelican while sunning itself, 
 with the pouch slightly expanded. 
 
 Fig. 4 gives the attitude when the bird is startled by an 
 unexpected noise. 
 
 Fig. 5 represents the bird in the act of preening the feathers 
 of the breast. Tliis attitude gave rise to the fable of the 
 
KXI'I,.\NAT()J{V JNItKX. 
 
 pelican feeding itH young with its own blood drawn from its 
 breast, the rod tip of the beak having a blood like look against 
 the white feathern. 
 
 J^'ig. G shows the bird with the mouth half-open, and Fgs. 
 7, H, {) aro ordinary attitudes. 
 
 :t>v 
 
 The reader will see that the poueh is hcarcely bhown at all 
 in these tigurns, the whole of which were f^ketched from the 
 
 i|! 
 
 
 
 tf!''' 
 
 living birds. 'J'ho colour of the pelican is white with a delicate 
 rosy tinge. 
 
 
 llAi;:'" 
 
'. ; 
 
 454 
 
 l-:XPLANATUliY INDKX. 
 
 C.jl ! 
 
 Hi' 
 
 m 
 
 i 
 
 Peppeu-Bi.al'K. {Piper niyrum.) 
 
 PKPPf;K-lvKi) (Capsicum annuum). — Few pliints are more 
 valued both abroad and in Englanu than the Capsicum. 
 In its small state it is known as "chili," and is sometimes 
 gathered before it has obtained the red stage of existence. 
 When ground, it is sold as Cayenne pepper, the species 
 usually employed for this purpose being Caj)sicinn/rutescens. 
 
 It grows well in England, if care be taken of it, and is a 
 very pretty and ornamental plant, especially when the pods 
 uttain their brilliant red hue. 
 
 Pkhai {SerrimihiiHs pmii/n). — This fish is quite as much 
 dreaded as the alligator or cayman in the rivers, or the 
 shark in tlie sea. Indeed, its teeth, sharply pointed and 
 razor edged, very much resemble those of the shark. Its 
 voracity seems to know no bounds. It will attack other fish, 
 such as the gilbacker, lowlow, paraima, &c., bite large pieces 
 out of their fins and tails. Even when scarcely larger than 
 an English gudgeon, the Perai, which is sometimes called the 
 Blood-fish of the Orinoco, can make fatal attacks on human 
 beings, its numbers compensating for its small size. 
 
 The following account of it is taken from Mr. C. B. Brown's 
 work to which reference has already been made : — 
 
 " The Corentyne and its branches were literally teeming with 
 fish of various kinds, the greater number being haimara and 
 Perai. The latter were so abundant and ferocious that at 
 times it was dangerous, when bathing, to go into the water to 
 a greater depth than up to one's knees. Even then small 
 bodies of these hungry creatures would swim in and make a 
 dash close up to our legs, and then reweat to a short distance. 
 They actually bit the steering paddles as they were drawn 
 through the water astern of the boats. A tapir which I shot 
 swimming across the water had its nose eaten off by them 
 whilst we were towing it to the shore. 
 
 "Of an evening the men used to catch some of them for 
 sport, and in taking the hook from their mouths produce a 
 wound from which the blood ran freelv. (.^n throwing them 
 
EXPLANATORY INDEX. 
 
 455 
 
 back into the water in this injured condition, they were im- 
 mediately set upon and devoured by their companion.!. Even 
 as one was being hauled in on the line, its comrades, seeing 
 that it was in difficulties, attacked it at onco. One day, when 
 the boat was hauled in to some rocks, a ftsw of the men wer(3 
 engaged shooting tish near by, and in so doing wounded a 
 large haimara. 
 
 " Having escaped from its human tormentors, it made for 
 the open river, but was instantly attacked by Perai attracted 
 by the blood escaping from its wound, and was driven back to 
 the shelter of the rocks close to the boat, from which I had a 
 
 » :f\ 
 
 
 good view of the chase. The large fish followed by its savage 
 enemies reminded me of a parallel case on land — a stricken 
 deer pursued by wolves, 
 
 " The Perai, fortunately, lie only off sand-beaches and in 
 quiet pools, not frequenting the cataracts, where their presence 
 would be anything but acceptable to the men while woi'king 
 in the water, I was fortunate enough to find the sjmwning 
 place of some Perai on the matted clusters of fibrous roots of 
 some lianes, which hung from the branches of a tree into the 
 water, amongst which much earthy sediment had collected, 
 and many small aquatic plants had grown. The sediment 
 gave weight to the roots which kept the clusters under water, 
 and the force of the current made them buoyant, giving the 
 lianes a slope when the river was high, which kept them not 
 far from its surface. 
 
 I il;l 
 
 I- 
 
 H 
 
 l;l 
 
 m 
 
 
 
\ 'I 
 
 ' i 
 
 450 
 
 KXIT.AXATOIIY INDEX. 
 
 :•-' ; 
 
 lii 
 
 t.;,l 
 
 *' My attention \va,s atti'ucted to thoni by two Pevai lying 
 close to them, with their heads up stream, as the men said, 
 engaged in watching tlieir eggs. Procuring one of the roots, 
 I examined it, and found amongst it numbers of single eggs 
 and clusters of small, jelly-like young, which had been already 
 hatched. The eggs were white, and of about one-eighth of an 
 inch in diameter, with a hard exterior. The young were very 
 little larger, and had a glutinous surface, which caused them 
 to adhere together on being taken from the watei\ They had 
 not acquired any jwwers of locomotiop, but could just wriggle 
 tlieir tails like tadpoles. 
 
 " Under a lens they resembled the egg devoid of its covei'ing, 
 with a gelatinous ridge around three-quarters of its circum- 
 ference, one end of which expanded into a knob (probably the 
 head), while the other termination was flattened and tail- 
 like. I could not detect any eyes or mouth in them, but 
 their bodies were speckled with grey markings of colouring 
 matter.", 
 
 Stedman mentions the fish under the title of Peery. 
 
 The Waraw palm-dwellers {see ^ta) dispose of their dead 
 simply by dropping the bodies into the lake, where the flesh 
 is rapidly stripped from the bones by the Perai, just as our 
 ants can do with frogs, mice, and small birds. If the deceased 
 should happen to be a man of importance, the body is lowered 
 in a net, and the skeleton, when quite cleaned, is drawn up, 
 dried, painted red, and hung iji the hut. 
 
 Petrel, Stor.my {Thalassidroma pelayica). — This bird is too 
 well known to need description. 
 
 Phaeton. /See "Tropic Bird." 
 
 Pi-pi-YO. Nothing is known of the bird, except that its 
 cry is shrill, and is thought to resemble the syllables forming 
 its name. 
 
 PiXE Apple {Pitcairnia sp.). 
 
 PiWARRi. — It is singular that a drink prepared as isPiwarr^ 
 should not only have been invented and di'unk at all. but 
 that it should have been emjiloyed by nations who could have 
 
EXPLANATORY INDKX. 
 
 457 
 
 hud no connection with each other. Piwani (which is also 
 spelled Piuri or Paiworie) is prepMii^d as follows. 
 
 Cassava bread is chewed, and then placed in a large pot, in 
 wluch is some of the expressed cassava juice already mentioned. 
 It is boiled for twelve hours, and then put into jars, where 
 fermentation takes place, and it is ready for use. It has an 
 intoxicating quality of a very feeble character, but the natives 
 compensate for lack of quality by quantity, and, by dint of 
 drinking successive bowls of the liquid, they do succeed in 
 intoxicating themselves. 
 
 When the natives can get brandy, they prefer it to any 
 other liquid, because it makes them drunk so soon, but next to 
 brandy they like piwarri. 
 
 Not only is it a favourite beverage, but it has a kind of 
 sanctity at+ached to it, and is drunk at their solemn feasts. 
 Here we have a most singular resemblance to the kava 
 drinking of Polynesia. The Mexicans prepare a drink called 
 " mudai " in a similar manner, except that they employ apples 
 instead of cassava or kava. Those who have been obliged to 
 drink Piwarri for reasons of policy, say that it rather resembles 
 very weak table beer. In some of the larger settlements, 
 they have enormous Piwarri bowls, shaped like canoes, of 
 about the same size, and called by the same name. 
 
 Plantain (Mtisa jHcradisaica). 
 
 Plovers. There are many species of Plover in Guiana. 
 The most common are the Black-breasted {Charadrhis Virgiui- 
 aims), the Ring-neck (C. sehiipafviatns), and the Sandy 
 Plover (S'frepsi/i's irderjn'i <). 
 
 Porcupine. — The species mentioned by Waterton is the Tree- 
 Porcupine, or Coendoo (Cercolabes jn'ehnmUs), which, like the 
 sloths, finds its nutriment in the trees and not on the ground. 
 In Northern America there is an allied species, also living in 
 trees and called the urson, cawquaw, or Canadian porcupine, 
 from which the " Indians," as they are called, procure the quills 
 with which they decorate their dresses and other articles, pre- 
 viously staining tliem with dyes extracted from various herbs. 
 
 ■''fA 
 
 ili'iJ 
 
Si liBI! 
 
 I! I 
 
 ii 
 
 I 
 
 ilf 
 IIP 
 
 458 
 
 EXPLANATOllY INDKX. 
 
 The Coendoo, however, may bo distinguished by its long 
 prehensile tail, which can be coiled round the branches like 
 that of the spider-monkey and the little ant eater, which has 
 been already described. 
 
 Its food consists of leaves, flowers, young twigs, and similar 
 substances, and its flesh is said to be delicate and tender. If 
 so, it very much belies the odour which proceeds from its 
 body, and which is thus described by Charlt's Kingsley : — 
 
 HOROUHLNK. 
 
 " More than once we became aware of a keen and dreadful 
 scent, as of a concentrated essence of unwashed tropic 
 humanity, which proceeded from that stiange animal, the 
 Porcupine with a prehensile tail, who prowls in the tree tops 
 all night, and sleeps in them all day, spending his idle hours 
 in making this hideous smell. Probably he or his ancestors 
 have found it pay as a protection ; for no jaguar or tiger-cat, 
 it is to be presumed, would care to n.eddle with any thing so 
 exquisitely nasty, especially when it is all over sharp prickles." 
 
 As to the theory that any animal, own the skunk itself, is 
 
KXPI.ANATUUY INDEX. 
 
 469 
 
 IS 
 
 protected by an evil odour, Waterton always treats it with 
 contempt. In an essay on the weasel, he has the following 
 remarks : — 
 
 " Many of the weasel tribe have the power of emitting a 
 very disagreeable odour from the posterior pert of the body. 
 We are gravely informed in the American ]ilo(jrapliy of Birds, 
 that the pole cat has this faculty 'given him by nature as a 
 defence.' And, pray, at what old giannys tireside in the 
 United States has the writer of this picked up such an im- 
 portant piece of information ? How comes the polecat to be 
 aware that the emitted contents of a gland (I use gland in 
 the singular number, for the sake of brevity, but the animal 
 has two glands), inoiiensive to itself, should be offensive to all 
 its pursuers ? I say, inoffensice to itself, because I cannot 
 believe that our Creitor would condemn an unott'endiug animal 
 to produce its own punishment by means of a smell which 
 never leaves it — whether it roam up and down as a solitary 
 animal, or whether it have a partner and a family of young 
 ones to i>rovide for. 
 
 " Although this odour from individuals of the weasel tribe 
 is very distressing to our own nasal sensibilities, it by no 
 means follows that the scent should have a similar effect upon 
 those of all other animals. For example, the smell from 
 purulent carrion is certainly very disagreeable • j us bipeds; 
 still it cannot prove so to the dog — for, in lieu of avoiding it, 
 this quadruped never loses an opportunity of rolling in it. 
 If the polecat has had the fetid gland ' given him by nature 
 as a defence,' then must nature have given a sweet one to the 
 civet for its destruction ; seeing that, whilst we shun the first 
 on account of its insupportable stench, we pursue and kill the 
 last in order to obtain its perfume. Now, as both these 
 animals are of the same family, I cannot help remarking, with 
 Sterne, in the case of the * poor negro girl,' that nature has 
 put one of this tribe sadly over the head of the others, if the 
 North American theory be sound. 
 
 "■Again, if nature has givoti tiiis abomiiiablo stench to 
 
 Ml;,, I 
 
 U 
 
 & 
 
 t 
 
 ■t.- 
 
 ■I 
 
 \A4 
 
 ym 
 
460 
 
 KXPLANATOIIV IXDKX. 
 
 niiiny of the polecat tribe 'as a defence,' she has cruelly 
 neglected our former invader, the Hanoverian rat. The pole-^ 
 cat is not much exposed to destruction, as its movements are 
 chiefly nocturnal, and, in general, it is apt to shun the haunts 
 of men. But our ITanov(>rian, having a most inordinate 
 appetite for the good things of this world, is ever on the stir, 
 in the very midst of its enemies, to satisfy the cravings of its 
 capacious stomach; and it will eater for itself the four-and- 
 twenty hours tlu'oughout. Itenci' your housekeeper complains 
 that it will try its tooth on prime.st Stilton in broad daylight, 
 and that it will have its whiskeis in the creambowl, even 
 whilst the dairymaid is gone up stairs with butter for the 
 breakfast table. Still my darling Hanoverian has nothing 
 but an ordinary set of teeth wherewith to protect itself, 
 although exposed to ten times moie danger than the foumart, 
 which last has a fetid gland given it l)y nature * as a defence,' 
 — in addition, I may add, to vast nuxscular strength, and to 
 two full rows of shai'p and well assorted teeth. 
 
 " This being the case, let us reject the Transatlantic theory 
 as a thing of emptiness ; and if we are called upon for an 
 opinion as to the real uses of the fetid gland in polecats, let 
 us frankly own that we have it not in our power to give any- 
 thing satisfactory on the subject." 
 
 Polecat {GaJcra harbara). — Called Tayra by natives. It 
 is an active, lively little creature, black, with a white patch 
 on the throat. Colonists are apt to call the opossum by the 
 name of polecat. See " Opossum." 
 
 Potato, 8\veet {Convolvtdus batatas). 
 
 PowiSE {Crax aJectw). — This fine bird is one of the Curas- 
 sows, of which thei'e are several species. It seems to bear 
 our climate very well, and as it is easily tamed, and will 
 take its place in the poultry yard, attemi)ts have been made 
 to acclimatize it like the turkey, the guinea-fowl, and the 
 peacock. As yet, however, the enterprise has met with small 
 success, which is the more to be regretted as the great size of 
 the bird, nearly equalling that of the turkey, and the delicacy 
 
KXPLAXATOIIY INDKX. 
 
 401 
 
 of its llosli, would lUiiko it a welcomo addition to our 
 poultry. 
 
 Within the tropics, however, it is a very common inhabitant 
 of tho poultry yard, where its only drawback is, that it 
 assumes autliority over the turkeys and guinea fowls, and 
 pecks them to death if they do not oboy. In many parts of 
 
 1 ',i 
 
 riiwisi:. 
 
 Guiana there is a belief that if a dog licks up the blood of a 
 wounded Powise, or mumbles its bleeding feathers as dogs 
 love to do, it will go mad. 
 
 Puiu'LE-iiEART {Copciifera puhNjf ova). — This most useful 
 tree derives its name from the purple colour of its wood, 
 which is very hard, close grained, durable, and tough. 
 
 It is, however, chiefly valued for its bark, which is used in 
 making canoes. Mr. C. B. Brown gives the following account 
 of the manufacture : — 
 
 "The bark canoes used by Indians are called ' Woodskins' 
 by the Creoles, and are made of one piece of bark, stripped 
 from a tree called the Purple-heart {Cojmi/era pnhli/fi^ra). 
 
 \n 
 
•l*)-. 
 
 i:XI'l,ANAT(»|{V INKKX. 
 
 *M 
 
 The bark of the locust- tree or siiniri {Ifyuieuaa covrhnril), is 
 also sometimes used. This hark is from one -eighth to ono- 
 fourth of an inch in thickness and very heavy, being of 
 greater specific gravity than water ; so that in the event of 
 the canoe sliipping water to any extent, it sinks immediately, 
 le.iving its occuj>ants to find their way ashore. 
 
 "They are ticklish things to travel in at first, and to stand 
 np in one, until acquainted with its freaks, is a dangerous 
 experiment, ending in an undignified exit over one side, the 
 canoe shooting away in the opposite direction. 
 
 " To make one of these woodskins, a largo Purple-heart tree 
 is cut down, and the bark of the requisite length taken off. 
 A wedge-shaped piece is then cut out of tlie trough-shaped 
 bark, from the top downwards, at a distance of three feet or 
 so from both ends on each side. The ends are then raised till 
 the edges of the cuts meet, when holes are pierced on either 
 hand, at a distance of six inches from the cut, and numbers of 
 turns of a strong withe or liana, called Mamurie, passed through 
 them and made fast in a neat manner to a small roimd stick 
 placed along the inside. Two strong pieces of wood fastened 
 across at the splits prevent the sides from closing in. The 
 ends are then trimmed down level with the sides, and a tick- 
 lish but serviceable little craft is tvirned out. The seats are 
 made of curved pieces of the same bark, and are very low. 
 
 "The whole process of making one of these woodskins, 
 including the drying of the bark, occupies a space of three 
 weeks. In this period must, I think, be included some loss 
 of time from laziness on the part of the Indian canoe-builder. 
 Woodskins vary in size, but usually are from fifteen to twenty- 
 five feet in length, and an ordinary sized one will carry 
 three or four people with their hammocks and provisions." 
 
 The tree when full grown, rather exceeds one hundred 
 and fifty feet in height, and the useful bark is smooth, and 
 of a dark brown colour. 
 
 Two varieties are known by the natives as Koorooboorelli 
 and Marawinaroo, 
 
ml 
 
 KXri-AXAT()l!Y IXDKX. 
 
 4«>:{ 
 
 Q. 
 
 Quail {Ortyx rirgmnna), — Partly migratory. 
 Quake. — A basket of opon work and very elastic and ex- 
 pansive. The drawing is from my own specimen. 
 
 liiji; 
 
 ijr vKn. 
 
 Quivers. — These are of two kinds, the one for holding the 
 arrows for the blow-gun, and the other for containing the 
 poisoned heads of the hog-arrows. A very tine specimen of 
 one of the former was given to me by Waterton, and is hero 
 represented. 
 
 The cover is not of peccary but of capybara skin, and has 
 the hair inside, so that by giving it a kind of screw when put 
 on, it remains firmly in its place. 
 
 The layer of kurumanni wax, with which the quiver is 
 wholly covered, is of considerable thickness, so that if the 
 hunter were to allow the quiver to fall into the water, it 
 might float for hours without any moisture penetrating to the 
 interior, and so weakening the wourali on the aiTows. 
 
 The coil of silk-grass, and the scraper of perai-teeth are 
 seen hanging to the quiver. When the scraper is used, the 
 arrow-point is placed between two of the teeth, which are flat, 
 pointed, and edged like razors. The arrow is then drawn 
 
 :|i 
 
Wi ts 
 
 M 
 
 'i 
 
 lit 
 
 I; 
 
 J? 
 
 4G4 
 
 EXPL AN ATOUY IN DE X. 
 
 between two of the teeth, and is thus shaved down to the 
 needle-like point which is required, 
 
 A small basket shaped like a flask, but slightly flattened, is 
 mostly carried with the quiver. This basket is used to hold 
 the wild cotton with which the arrows are fitted to the bore 
 of the blow-gun. Thus, with an equipment weighing al- 
 together little more than three pounds, the native hunter can 
 
 CjrlVKR, BI,OW-(ilIN. 
 
 QUlVr.R, IlOd-APROW. 
 
 carry with him a powerful gun and about three or four 
 hundi'ed deadly missiles. 
 
 The Quiver for the heads of hog-arrows is very mucli 
 smaller, and is only a piece of bamboo about seven or eight 
 inches in length, and having a cover to guard it from damp. 
 The Indian generally ornaments it by wrapping cotton-strings 
 round it in patterns, as is seen in the illustration, which is 
 taken from one of my specimens. 
 
EXPLANATORY INDEX. 
 
 466 
 
 R. 
 
 13 
 
 Rattlesnake {Uropsophus durissus). — This is so familiar 
 a reptile, that I shall not occupy valuable space by describing 
 it. 
 
 I have already mentioned that when the sloth arrived in 
 England, Waterton was able to point to it as a proof that he 
 had been right in his description of that animal. Similarly, 
 when a box of live Rattlesnakes, twenty-seven in number, 
 were exhibited at Leeds, he had an opportunity of proving 
 that he was not romancing when he dei>cribed his encounters 
 with deadly snakes. 
 
 He invited a large party of friends and scientific men to 
 meet him at Leeds, whither he had sent a large glass case, so 
 that the snakes might be properly seen. After cautioning 
 the visitors to refrain from moving or speaking, he, with his 
 bare hands, transferred the snakes from the box to the glass 
 case and back again. 
 
 He explained his mode of handling a venomous serpent by 
 the following illustration. He assumed my hand, wrist, and 
 arm to be part of a Rattlesnake, the hand being the head, the 
 wrist the neck, and the arm part of the body. He then 
 approached his hand silently, slowly, and quietly to my 
 wrist, which he by degrees clasped, but did not picss. He 
 next lifted the arm gently from the table on which it was 
 resting, and put it down again in the same slow and cautious 
 manner. He trusted for his immunity to the sluggish nature 
 of the serpents generally, which if disturbed, will glide gently 
 away, but if they are trodden upon or hurt, will, in self- 
 defence, strike at their assailant. 
 
 He said that after that lesson, I should be as able to carry 
 living Rattlesnakes as he was. Perhaps so, but I should 
 feel rather nervous about trying it. 
 
 Red Howler. See "Howler, Red." 
 
 H H 
 
 lit 
 
 Pi 
 
 !i 
 
 m 
 
4rit'i 
 
 KXPI,A\AT()HY IXI'KX. 
 
 Keustar r {Se(o})h<i(ja or Musciaipa rnticlUu). — It hiis 
 nothing to do with tho English Iveclstart, 
 
 RniNocKUos Bektle {Meyasonm 7'/Vrt/?), popiihirly called the 
 Groat J3ro\vn Sawyoi'. 
 
 Kouo.u.— ,sVe "Paint." 
 
 S. 
 
 SacawinivI, — One of the Squirrel jNIonkeys. 
 
 This tiny monkey, several times mentioned by Waterto::, 
 is plentiful in Ciuiana, and is noticed by Schomburgk. 
 
 " A troop of little Sacowinkis, or squirrel monkeys, some 
 of tlie most beautiful and active of their kind, leaped Vt'ith 
 tlie agility of a taine squirrel from branch to branch, and, 
 
 !::!, 
 
 \: 
 
 • \r.\\WsK\. 
 
 Ill 
 
 alarmed by our apjjcarance, uttered their painful call, re- 
 sembling nuicli more the cry of a biid than that of an animal, 
 and then, hastening away, were soon hidden among the thick 
 foliage of the large forest trees." 
 
 Several species of squirrel -monkey appear to be included in 
 the general name of Sakawinki. Stedman figures the Marmoset 
 (Jacclnts vuff/arifi) \inder the name of Sakawinki, while 
 Waterton, who mentions it as being black, and scarcely six 
 inches in length, evidently refers to the White-whiskered 
 Tamarin {Jacchns hucogeiiys), wliich exactly accords with his 
 desci'iption and has uocordingly Ijecn figured. 
 
ni 
 
 KXI'LANATOliY IN'DKX. 
 
 40 
 
 S/.LKMi'KNTA (TiilvH 7V;y7/ra/v«).— Tliis line lizard is soine- 
 tiiiies ciiUod tlu! Hufo^'uanl, heciinsc it is supposed to give 
 notice of tlu* iipproacli of the allif,'at()i', just as tlio African 
 monitors are iKjiioved to warn travellers against the crocodile. 
 
 Jt is a large animal, stoutly made, and reaching ilv(! feet 
 or more in length, it is handsomelv <'oloured with hlack. 
 
 \i 
 
 
 •''■'\ 
 
 I 
 
 SAM.MI'KSIA. 
 
 yeHow, green, and white, an-anged so variously tiiat hardly 
 any two specimens are alike. For tiiis reason it is called the 
 Variegated Lizanl. 
 
 Like the iguana, the Salempenta allords very delicjite food, 
 which is thought to resemble tiie Mesh of a very delicate 
 young chicken. New comers are at first averse to eating a, 
 lizard of any description, but tlu^y very soon find out their 
 mistake, and would oven prefer an iguana or Salempenta 
 cutlet to a chic! (in, 
 
 Samoliiaii, 'The palm whi(;h is known by tliis name is 
 called Ircnrtia Hcflf/cra, and like the ourah, is of very smnll 
 diameter in pro[»ortion to its length, 
 
 Sand Fi,v {SlniiiHit l*ertuia.r), called by the natives Mapire, 
 and mostly haunting the sea shore. What the mosquito is by 
 night, the Sand Fly is by day, and is a still gr( ater pest. ]t Hies 
 veiy rapidly, settlis, bitcH, and tills itsidf with blood almost 
 in.stantaneously, having nothing of the deliberate action of 
 the mos([uito. The biles afterwards beonu^ exceedingly 
 painful. 
 
 11 H :.' 
 
 
 41 
 
 
 <m 
 
 
 i 
 
 r 
 
468 
 
 EXPLANATORY INDEX. 
 
 Sanobe do Buey {Pyranga rubra). — Sometimes called 
 the Sciirlot Tanagci'. As is the case with many brilliant 
 birds, the splendid hues of tlie plumage belong only to the 
 adult male, the female having the body dull green and yellow 
 instead of scarlet, and the wings and tail brown instead of 
 black. It is about six inches in length. 
 
 Sawaui or SouARi {Cari/ocar foiueiifosum.) — This is the tree 
 which furnishes the well-known butter nuts. The nuts grow 
 in the interior of large fruits, tilled with a whitish meal. The 
 roots are used for making Hoors, and in the futtocks of ships. 
 It thrives best on the hills. The natives are so fond of the 
 nuts, that the po.ssessiou of the finest trees is said to be an 
 ackuowledg(Hl cause of war. When th ■ nut is cracked, the 
 kernel is found to be filled with a white almond-like jelly. 
 
 S(.'auli:t (JiiosHKAK. — SW "Grosl)eak. Scarlet." 
 
 Si'AIU.l;'! ciliOSHI'.AK. 
 
 SciHOi; (Ii/tantph(ifitos toco). — See "Toucans." 
 
 S("RF,AMKU, HoKNEi) (Palaimdcfi cornnta). — This fine bird 
 is about as large as a tuikey, and derives its popular name 
 from ii small horn like appendage situate upon the upper 
 part of (he h(>a<l. The f)I)j(H't of tliis ajipendage is absolutely 
 
KXl'I.AXAK; :Y index. 
 
 Wj 
 
 unknown. Some zoologists believed it to be employed ius a 
 weapon, but such an opinion is simply absurd, as no bird 
 strikes with any part of its head exeept the beak, and even 
 if the Screamer did so, the liorn would be quite inadequate as 
 a weapon of oii'ence. 
 
 :ii i; 
 
 1 bird 
 
 name 
 
 ihiUNKii sci:kami;|{. 
 
 When it wishes to attack nny foe, it uses the sharj) spurs, 
 which are attached to the wings, are very strong, and about 
 an inch in length. 
 
 The natives call it by tl\e name of Kamichi. 
 
 Silk-Grass. — This peculiarly useful ilbre is obtained from 
 the leaf of one of the Aloe tribe (Afjare vivipava). 
 
 The fibi'es run nearly parallel to each other tliroughout the 
 whole length of the leaf, and are extracted by steeping the 
 leaves in water until they are quite soft, and then drawing them 
 through a loop of cord, so as to clear oft the decayed pulp 
 and preserve the fibres. These are then dried in the sun and 
 tied in bundles. 
 
 The Silk-grass is of incalculable value to the natives, who 
 always have a supply on hand. The strength of the fibre is 
 really wonderful, and though so tine that it seems as if it 
 would .snap at a touch, it seems more like steel wire than a 
 vegetable fabric. A coil of it is always attached to the 
 quiver. Why it is not largely used in this country I cannot 
 
 -a 
 
 l! 
 
 y 
 
 !;1 
 
 iiV 
 
 II 
 
 •i"i' 
 
470 
 
 KXPLAXATolJV INDEX. 
 
 ''■ 
 
 i inagine. It grows in the greatest abundance, can be easily 
 extracteil from thi^ le.if, and is not only strong, but light. 
 
 ••^^' 
 
 Sir.K-UKASS. 
 
 SiLoAUAi.i. This name is .sometimes .sjudt as Ciroubali, or 
 Siruabali. There are several trees going by this name, all be- 
 longing to the genus Eperna, and therefore allied to the 
 wallaba tree, <j.r. The wood is excellent for ship building, 
 and Mr. ^IcTurk, who collected specimens of Guianan woods 
 for the Palis Exhibition of 1878, says that it is even better 
 than greenheart, q.v., and ought to be classed at Lloyds among 
 the A 1 wo()<ls. As the wood is light, and Hoats well, it is 
 much used for boat building. 
 
 When freshly cut, it has a strong aromatic scent, and the 
 juice is bitter. The shij)-worm (TereJo) will not touch it, 
 and it is therefore useful for planking boats. 
 
 Sloth, Three tokd, or At {lh'o<h/jnis trhlucfyJnH). — 
 Waterton's account of this animal is so full and accurate 
 that littl».> needs to be added. The mark on the back of the 
 
KXPLAXATORY INDEX. 
 
 471 
 
 male is most singular, and looks exactly as if it had been 
 stamped with an oval-sha|)ed iron, very hot along the centre, 
 so as to burn away the hair nearly to the skin, and blacken 
 it. Then, if the iron be not quite so hot on either part of 
 the centre, it would press down the fur and turn it brown. 
 Lastly, supposing the iron to bo only hot enough towards the 
 edges to scorch the fur slightly, we shall produce a fair 
 imitation of the mark impressed by nature upon the other- 
 wise long, jind hay-like fur. 
 
 i 
 
 The central line then is black, and very decidedly markc.l. 
 On either side, the fur is still short, but of a creamy colour, 
 and it fades by short bands of brown into the grey hue of 
 the fur. 
 
 Stedman says that the animal is popularly called the 
 Loggurree or Sheep-Sloth, and that the names Ai, or Heeay, 
 are given to it on account of its plaintive cry. The natives 
 seemed to be very much afraid of its claws, and the first 
 specimen which was brought to him had been deprived of all 
 its feet, so as to render it harmless. 
 
 The illustrations are taken from a s{i<>('imen in Waterton's 
 
 
472 
 
 KXPLAXATOKY INDEX 
 
 museum, and are given in oriler to show one of the charac- 
 teristic attitudes assumed by the animal, as well as the 
 position and shape of the mark on the back of the male. 
 
 A I. 
 
 Sloth, Two-toed {Bnidyjma didactylus). — It is well known 
 that Waterton was the original discoverer of the real habits 
 of the Sloth, and showed how entirely false were the accounts 
 of Bud'on and others. So full is his description that no 
 traveller has been able to add any important particulars which 
 Waterton has not mentioned. 
 
 In Surinam, the popidar name for this animal is Dog-Sloth. 
 When motionless, it clings to the branches in such a manner, 
 that even if seen, it w(mld be mistaken by most persons 
 for a mere excrescence on the bark. The specimen which I, 
 saw in the Zoological Gardens was remarkable for the ex- 
 tremely wet state of the muzzle. 
 
 Spoonbill {Plaialea leucorodia). 
 
 Spue-wing.— »See "Water Hen, Spur-winged." 
 
 Stabroek. — The reader will bear in mind that the 
 country now known as British Guiana, was originally 
 Dutch Guiana. The capital of Demerara then was called 
 Stabroek, but since Guiana has passed into the possession of 
 
L'X- 
 
 tho 
 ally 
 lied 
 of 
 
 EXPLANATOllV INDEX. 
 
 473 
 
 England, the name of the capital was changed, and it is now 
 called Georgetown. 
 
 Sting Ray {Tryrjon vustlnaca). — Many species of Sting l^ay 
 are known, and they all belong to the Skate family. They 
 are well known by their long whip-like tails, and the sharp, 
 doubly barbed bone which pi'oceeds from the base of the tail, 
 and becomes a weapon which can be wielded with terrible 
 eifect. 
 
 As a rule, the Guianan natives are not a warlike race, but 
 when war is forced upon them, their weapons are vei'y 
 formidable. War arrows, for example, are sometimes headed 
 with the bone of the Sting Ray, and ai*e nearly sure to cause 
 death, as, when the arrow strikes the enemy, the force of the 
 blow causes its very brittle head to snap asunder, and to 
 leave several barbs imbedded in the body. The Polynesian 
 warriors probably took the bone of the Sting Ray as a model 
 for their many-barbed spears. 
 
 SuoAR-fANE. — -The pi'incipal plant which is used for the 
 production of sugar is the gigantic grass known as Sugar-cane 
 (Saccharinum officinarum). It was originally brought from 
 the E. Indies to Southern Europe by the Crusaders, and so 
 made its way into the colonies. There are many species of 
 saccharinum, or sugar-grasses. Some of our own grasses have 
 i\ perceptible saccharine flavour. 
 
 SuN-BiUD. — A kind of Heron (l/elias eurypyga). — In Bar- 
 badoes it is called Tigri-Fowlo. 
 
 The following account of the bird is from C. Kingsley's At 
 Last : 
 
 "These are strange birds too. One, whom you may f^ee in 
 the Zoological Gardens, like a plover with a straight beak and 
 bittern's plumage, whose business it is to walk about the table 
 at meals uttering sad metallic noises and catching flies. 
 
 " Its name is Sun-Bird, * sun-fowlo,' of the Surinam negroes, 
 according to dear old Stedman, ' because, when it extends its 
 wings, which it often docs, there appears on the interior part 
 of carh wing a most beautiful representation of a sun. This 
 
 it 
 
 \h\ 
 
 
 m 
 
 3'!' 
 
 
r 
 
 ; ,)' 
 
 It 
 
 474 
 
 EXl'LANATOKY INDEX. 
 
 bird,' ho continues very truly, ' might bo styk'il tho jiorpotuul 
 motion, its body making a continual movement, and its tail 
 keeping time like the iiendulum of a clock.' " 
 
 Stedman also calls it by the name of Flycatcher, one of 
 the many instances of misleading nomenclature Mhich are to 
 be found in books of travel. The colour of tho sun-bird is 
 reddish, spotted with black, and there is a round black spot 
 on the small head. The beak and legs are gx*een. 
 
 T. 
 
 -Sen 
 
 "■ Sang re 
 
 lo 
 
 Tanoara. — Mis spelling for Tanager. 
 IJuey." 
 
 Tapiu (I'ajdriis terrestrin). — The Tapirs form one of tho 
 links connecting the elephants with the swine, and tho 
 only two species which survive, one in Tropical America, and 
 the other in Malacca and Sumatra, really look as if they were 
 extinct pachyderms that had been restored to the world. 
 
 The Tapir is never found far from tho water, in which it 
 disports itself as if it belonged naturally to that element. If 
 alarmed on land, it always makes for the river, into which it 
 plunges, and then can escape from any foe except man. It is 
 .11 vegetable eater, feeding mostly on wild gourds, melons, and 
 similar fruits. It is easily tamed, and soon becomes us 
 familiar as a dog. 
 
 Tauronira (IIumiriuiH flvribundum). — Found plentifully 
 in sandy ground near swamps, but not in them. 
 
 The natives perfume their hair with a minute fungus which 
 always gi'ows in the newly-cut wood of the Taui'onira. The 
 timber is chiefly used for house-frames and wheel-spokes. 
 The tree is about ninety feet in height. 
 
 Tiger. — See "Jaguar." 
 
 Tiger-Bird. — There are several species of these birds, 
 which derive their popular name from tho peculiar cry 
 which they utter. This call resembles tho cry of tho jaguar, 
 papiilarly called the tiger, iind, like tho JMguar, the Tiger-Bii-d 
 
 I I: 
 
KXI'LAXAToUV INDKX. 
 
 47.'> 
 
 H 
 
 do 
 
 utti'1'8 its ci'y in tlu^ t*urly iii()ruin<,' iind Into in tlic evening'. 
 Two sjH'i'icH are t()loriii)iy cnininon. One {Ti(ji'lnoiiia Jirazll- 
 ieiini') i.s HJinjtly brown in oolour, hut another {'/'it/rinomn 
 fiiji'lniiin) i.s (hill yellow, with ii niunher of short black bars 
 across the feathers. 
 
 TuiKii-CAT. One of the siiiall Leopards known as Mar^'ays 
 ( fjenjjanfnn (ifjrliuin). Several speeies of leopard are called hy 
 this name. Watorton tamed one of them, brou^'ht it home, 
 and actually trained it to run with the foxhounds. It was 
 very u.seful as a rat-catcher. 
 
 TiNAMui:. — The.se birds all ludonj,' to the genus Tinamotls, 
 and are all natives of South America. 'I'hey are on an 
 average about the size of a grou.se. They have very short 
 tails, which gives them rather a lumjash appearance. They 
 seem to be rather stupid birds, and can be caught by a 
 noo.se Hxed to the end of a long stick. The .sjjccies which is 
 mentioned by Waterton under the name of JSInani is probably 
 Tiinimotit ehynns. 
 
 TiiiANA.— A'*;« " Sun bird." 
 
 Tortoise. — This is the l>ox-Tortoise {('istuda Carolina), 
 so called because it can not only draw its limbs and head 
 
 i: 
 
 i 
 i 
 
 ,!j 
 
 -••Ki . 
 
 Tiir.lMISK 
 
 within the shell, but can fold the .shell together, so that it 
 is quite invulnerable. — See Sydney Sinith'.s simile on [wige 19, 
 The negroes call it by the name of Cooter. 
 
11 
 
 i 
 
 ■4 
 
 i 
 
 I7<» KXI'r,A\AT()i:Y IN'DKX. 
 
 It Js always to ht* foiuiil in dry Hi'tiiatlona, prcforrinj? th(> 
 piiio forosts, hccaiiso tlicy also lovo a dry soil. It is of .siiiiill 
 size, and as it is very prettily colountd, it is soiu 'tiiiuis niauii- 
 facturod into a suull'-hox, boin<:f mount i>d in silver, and having 
 the niovabh< plates tiited with hinges, and s[)rint(-elaspa. 
 The colour is most varialilo, but black and yellow are the 
 most eonspi(!uous hues. 
 
 Toi'CAN. — Tropical America produces an inexhaustible 
 variety of livini^ creatures, and there is perhaps none more 
 remarkable than that extraordinary },'roup of birds known as 
 Toucans, from the native name Toco. 
 
 TullAS. 
 
 Kingsley seems to have been greatly impressed with the 
 appearance of the toucan, and agrees with Waterton that it is 
 essentially a fruit-eating bird : — 
 
 "A Toucan out of the primaival fore.st, as gorgeous in 
 colour as ho is ridiculous in shape. His general plumage i.s 
 black, set off by a snow-white gorget fringed with crimson ; 
 crimson and green tail coverts, and a crimson and green beak, 
 with blue cere about hi.s face and throat. 
 
EXI'l.AXATDHY TNPKX. 
 
 " Ills PHormouH iiiid woiik bill MOt^nis iiiatlr for i]u> j)iir|>o,s(< 
 of 8\viill»)\ving It.Mimnas wli. .!•• ; how h(i fi't'tls himsolf witli it 
 in tho forest i.H (litlitiilt to j^mess (wc " I'liiiiiiui") ; niitl wlu-n lu* 
 liopH up aiul clown on his ^/icat cliitteriuf,' foil two toes turn«((l 
 forward and two hack — twisting liead and h»'ak ri^dit and loft 
 (for ho cannot well soo straij^dit before him) to soo wlienco the 
 bananas are coniin;^ ; or wiien ajjain, after gor^'in^' a ('ouj)lc, 
 lie sits gulping and winking, digesting them in serene satis- 
 faction, he is as good a specimen as can be seen of tho 
 ludicrous -dare I say tho intentionally ludicrous? — element in 
 nature." 
 
 It is true that in confinement tho Toucan will eat little 
 birds, just as a monkey will eat a mouse, but I very much 
 doubt whether either toucan or monkey touches animal food 
 in its native woods. Toucans make great inroads on tho 
 ]»lantations of oranges, guavas. and other fruit, but, as they 
 are shot in considerable nund)ers. the delicacy of their flesh 
 makes amends for their depredations. Waterton seems to 
 iiave <'aten his Toucans boiled. 
 
 There are many sjiecies of Toucan. The liouradi of Water- 
 ton is J\'/iaitijt/utstos eri/f/irorfii/nc/iiis, and the Scirou is 
 Kluunphastos Arivl, as shown in the illustration. The 
 Toucanot is lkh(niij)l/(istos riteUinns. It is an active little 
 bird, uttering its yelping cry either by day or in the evening, 
 and jerking its head about in a mo.st ludicrous fashion. 
 
 TuiBKs, Nativk. — The live great tribes are given on p. 'I'M. 
 Taking them in their order, the Waraws, or Guaraons, as 
 Humboldt calls them, are more exclusively acpiatic than the 
 others, one great divisicn of them living wholly in lake 
 dwellings built in the stems of the ita or nioriche palm. ^See 
 " ^]ta." They are also tlie best canoe makers, and can carve 
 canoes out of ti*ee trunks some of which arc able to carry 
 very large numbers of men They sell these canoes to other 
 tribes. 
 
 The Arowacks are taller and fairer than the other tribes, 
 averaging from D ft. 4 in. to r> ft. •) in. in hc'ght. and the mixed 
 
 m 
 
 i ii •' 
 
 : 
 
I ■'. 
 
 478 
 
 KXPLAXATUHY INDEX. 
 
 progeny of Arowack and European is said to bo remarkable 
 for its beauty and intellect. The Acoways wear scarcely 
 any clothing, and indeed, when no white men are near, do 
 not trouble themselves about clothes at all. They wear their 
 coarse black hair very long, and are accustomed to carry a 
 piece of wood in the cartilage of the nose. 
 
 lIKAt. AMI IIIKAI.. 
 
 The Caribs are also nude, and wear ornaments in their 
 under lips. Lastly, come the Macoushies, who aie the best 
 wounili makers, and sell it to other tribes. 
 
 In the accompanying illu.stration I hiive brought together 
 three figures, in order to show the difference between the 
 European and Guianan types of the human frame. The 
 " Painter's Ideal " is coj)ied from an imaginative drawing 
 of a young Indian girl, as depicted by a well-known artist. 
 
KXPLAXATOKY INDEX. 
 
 47;» 
 
 The model is clearly of an English type, with tlie abundant 
 wavy hair, sloping shoulders, ample brow, full eyes> and 
 straight delicate nose. Substitute an English archery dress 
 for tho skin robe, and a bow for the spear, and there is a 
 handsome English girl at an ai'chery meeting. 
 
 The Sculptor's Ideal of the same subject is wholly Greek. 
 Substitute a crescent diadem for the plumed circlet, a light 
 classical chiton for the feather apron, and there is Diana. 
 
 How different is the real figure of the Guianan type of female 
 beauty, copied exactly from a photograph in my possession ! 
 Note the difference of size, tho coarse, straiglit hair, the little 
 eyes, the high cheek-bones, tlio short nock, the scpiare shoulders, 
 tlie in-turned feet, and the spreading toes. Diess her as you 
 will, or even change her complexion, she can never look 
 English or Greek. 
 
 Truely. — This is a palm with very long and wide lefives, 
 which are much used in the construction of houses. Stedman 
 mentions the palm under the name of Trooly, and states that 
 the leaves diverge directly from tlie ground, like those of tho 
 young coucoiirite. The leaves are known by the name of tas. 
 It is probably one of the fan-palms (T/irinar), the leaves of 
 which are more than twelve feet in wddth, and pi'oportionately 
 long. When young, it has no stem, so that it answers both 
 to Stedman's and Waterton's descriptions. 
 
 Tuoi'K^ Bird {Phaeton o'thercns). — This beautiful bird is 
 one of the Pelican family, and derives its popular name from 
 the fact that it is never seen outside the tropics unless blown 
 by gales too violent to be resisted, even by its powerful wings. 
 Like its relative, the frigate l)ird, it can remain on tlie wing 
 during the whole day, and has been met iit sea at least a 
 thousand miles from the nearest land. It can, however, rest 
 its wings by sitting on tlie water. 
 
 There are several species of tliose birds, one named tho 
 lioseate Tropic Bird, being in great request among the natives 
 of the South Seas, who use their long tail-plumes as ornaments 
 in their liead-dresses of ceremony. In this species, the tail 
 
 f 
 
 " 
 
 
480 
 
 EXPLANATOKY INDEX. 
 
 feathers are scarlet, and are fastened into a i)laited fillet 
 which surrounds the head, very much as the Guianan natives 
 make their feather-crowns with the tail-feathers of macaws. 
 
 On account of the aei'ial habits of the bird, it is not easy to 
 procure these feathers, and these islanders manage to supply 
 themselves by watching their nesting-places, crawling up to 
 them, and jerking out the feathers as they sit on their eggs. 
 
 Waterton had great difficulty in obtaining one of these 
 birds, but at last shot one, in a voyage across the Atlantic. 
 
 TtlOPIf BlliD. 
 
 He offered a guinea for the recovery of the bird, when a 
 Danish sailor plunged into the sea after it. He nearly lost 
 not only the reward, but his life, for no boat could be launched, 
 and the ship was going so fast through the water, that in trying 
 to back, she missed stays, and they had to wear lier. How- 
 ever, the man kept himself afloat, and delivered the bird to 
 
KXPLANATOHY fXDKX. 
 
 4MI 
 
 Waterton, wl.<, Jm.l a Hpocial alTecti.,u fo.- it ever afterwards 
 IJ.e illustration is taken from a sketch of this very specimen 
 as it stood in his museum. 
 
 Trouimals -There are many of tliese birds, the best known 
 of which IS the Itice Troupial {Dolh-honyx, orizivorm). 
 
 It obtains its name from its extreme fondness for rice and 
 .« sometimes called the rice-bunting. As it has a very 'wide 
 range, is extremely plentiful, and possesses a I,eautiful voice it 
 has received a variety of popular names, such as butter-bird, 
 because it b(,comes so fat in harvest time ; reed-bird, because it 
 IS fond of settling in vast numl,ers among seeds; bob o-link, 
 bob Imkum, At, 
 
 Trumpeter, s.m.etimes called Waracaba (/Vo^./,/« n-cintn,>,). 
 
 ,M 
 
 ■~>t 
 
 Mil Mci;t. 
 
 —This bird is allied to the crane. Waterton mentions that 
 It can be domesticated, and in such cases, it usually considers 
 the hou.e as its own, and is madly jealous of any other pet 
 which It fancies may deprive it of its master's affections As 
 for cats, it cannot lear them, and always turns them out of 
 
 I I 
 
i 
 
 482 
 
 KXIM.AXATOHY INDEX. 
 
 tlio room if possilile. Even dogs yield to it unless they are 
 very conragooiis, for the Ti'umpeter attacks them after a very 
 peculiar fasliinn, jumping on its foo'a back, and kicking and 
 pecking him until he is quite bewildered. 
 
 It is usually a beautiful bird, especially in the breast, where 
 the black feathers are richly glossed with gold and purple 
 upon their edges. 
 
 TiRTi-K, Freshwater.— Several reptiles are known by this 
 name. That to which "Waterton here refers is probably the 
 larger of the two {Podoenemis expansa), called Sachapana by 
 the natives. Its eggs are large, spherical, and with white 
 shells. The smaller species {Emys tracaxa), called by the 
 natives Taracai, lays soft oval eggs. The eggs of both 
 species are valued as food, and only the yolk is eaten. It ia 
 allied to the well-known chicken tortoise of North America. 
 
 V. 
 
 A'"amimre {Vampirus spectrum). — As to the Vampire bats 
 which would not bite Waterton, who did his best to allure 
 them, and would insist on biting everybody else, who did not 
 wi.sh to be bitten, C. Kingsley has the following remarks : — 
 
 "Then we inspected a coolie's great toe, which had been 
 severely bitten by a Vampire in the night. 
 
 " And here let me say that the popular disbelief of Vampire 
 stories is only owing to English ignoi'ance, and disinclination 
 to believe any of the many quaint doings which John Bull 
 has not seen, because he does not care to see them. If ho 
 comes to these parts, he must be careful not to leave his feet 
 or hands out of bed without mosquito curtains ; if he has 
 good horses, he ought not to leave them exposed at night 
 without wire-gauze round the stable shed — a plan which, to 
 my surjn'ise, I never saw used in the West Indies. 
 
 *' Otherwise, he will be but too likely to find in the morning 
 a triangular bit cut out of hi.s own Jlesli, or even wor;^e, out of 
 
EXPLANATORY INDEX. 
 
 483 
 
 been 
 
 mpive 
 lation 
 Bull 
 If ho 
 feet 
 has 
 night 
 ch, to 
 
 his horse's withers or throat, whose twisting and lashing cannot 
 shake the tormentor off ; and must be content to have himself 
 lamed, or his horses weakened to staggering and thrown out 
 of collar-work for a week, as I have seen happen more than 
 once or twice. The only method of keeping off the Yampire 
 yet employed in stables is light ; and a lamp is usually kept 
 burning there. But the negro — not the most cai'eful of men 
 — is apt not to fill and trim it ; and if it goes out in the small 
 hours, the horses are pretty sure to be sucked, if there is a 
 forest near. 
 
 " So numerous and troublesome, indeed, are the Vampires, 
 that there are pastures in Trinidad in which, at least till the 
 adjoining woods were cleared, the cattle would not fatten, or 
 even thrive ; being found, morning after morning, weak and 
 sick from the bleeding which they had endured at night." 
 
 The coolie in question — a Hindoo — made very light of the 
 injury, although the toe bled considerably, and the wound 
 could scarcely be healed. Why the Vampire should always 
 select the withers of the horse Jind the toes of the human 
 being is a problem as yet unsolved. Neither do we know 
 how the Vampires lived when they had neither horses, cattle, 
 nor human beings to suck, any more than we know how count- 
 less millions of mosquitos, fleas, land-leeches, and other 
 noxious creatures contrive to exist without the blood of man 
 or beast. 
 
 Vanilla {VauUla planifoVm). — The plant which furnishes 
 the well known Vanilla, so much used by pastry-cooks, is a 
 curious parasitic orchid. 
 
 It originally starts from the ground, and then ascends the 
 trunk of a tree. The stem is square, and throws out a number 
 of little rootlets, which make their way into the bark and 
 drain the tree of its sap. 
 
 Indeed, so entirely does it depend upon the juices of the 
 tree for its life, +hat if it be cut away below, so as to sever its 
 connection with tlie ground, it will continue to grow as well 
 as ever. Our ivv. if cut, alwavs dies, bocuuso it derives its 
 
 1 I -1 
 
 W 
 
 PI 
 
il! ' 
 
 I 
 
 484 
 
 EXPLANATOKY INDEX. 
 
 whole subsistence from the ground, and none from the tree, 
 and, as is shown by the sketches of the Ivy Tower on pages 
 43 and 66, flourishes as well on stone walls as on trees. 
 
 The flowers grow in spikes, and are richly perfumed, and 
 the pods are very flat, and open along the side. Very little 
 genuine vanilla finds its way into England, not more than a 
 few hundred-weight being imported. Its place is supplied by 
 the common and much cheaper balsam of Peru. 
 
 Vulture, Aura {Catharistes mira). — The name of this bird 
 will always call to mind the memorable controversy between 
 
 VI LTIUK. AI HA, 
 
 Waterton and Audubon, in whicli the former came off the 
 victor. The whole controversy is printed in the volumes of 
 Fssays, and it is too long to be inserted in this book, and too 
 valuable to "^ judged by means of extracts. I st:uiw'.y 
 recommend < '•■ ) reader to procure these essays, and admire the 
 masterly ma iner in which Waterton handled the subject, and 
 the completeness with which he cut away the gi'ound from 
 under the feet of his antagonist. 
 
EXPLANATOUY INDEX. 
 
 4Hr» 
 
 The Aura Vulture, jjopularly known us the Turkey Buzzard, 
 is very common in its own country, and in civilized places i.s 
 protected by law, on account of its value as a scavenger. 
 Waterton states, in corroboration of his own view of the case, 
 that during the dreadful plague iii ISIaluga, which has already 
 been mentioned on p. 11, the V^ulture did inestimable service. 
 
 " So great was the daily havoc of death, that no private 
 burials could be allowed, and many a corpse lay exposed in 
 the open air until the dead carts made their rounds at night- 
 fall to take them away to their last resting-place, which was 
 a large pit, prepared by the convicts in the day time. 
 
 " During this long- continued scene of woe and son-ow, which 
 I saw and felt, I never could learn that the Vultures preyed 
 upon the dead bodies which had not had time enough to 
 putrefy. Uut, when the wind blew in from the Mediterranean, 
 and washed ashore the corrupted bodies of those who died of 
 the pestilence, and had been thrown overboai'd from the 
 shipping, then it was that the vultures came from the neigh- 
 bouring hills to satii^fy their hunger." 
 
 The popular name of turkey buzzard is given to the bird 
 on account of its resemblance to the common turkey, and 
 many a new comer has found himself an object of derision 
 because he has shot an Aura Vulture, taking it for a 
 turkey. 
 
 I ::; I 
 
 VII.ITUK .->l NMNd nsKI.K. 
 
 The smaller illustration is taken from a sketch at the 
 Zoological Gardens, representing the Vulture in a favouiite 
 attitude, sitting on its ankles, with its wings half spread, 
 rejoicing in the sun. 
 
 Vri/ri RF, King {SarcorJaimjihnu jxi/xi). — Being one of the 
 
 l,« 
 
4H<1 
 
 KXI'LAXATOIJV INDKX. 
 
 longest and most powerful of the Vulture tribe, this bird has 
 dei'ived its name from the tyranny which it exercises over the 
 aura and black vultui*e.s, not allowing them to eat until its 
 o»vn hunger is satisfied. The beautiful colours which adorn 
 the beak are well and accurately described by Waterton, and 
 it is on account of these fleshy excrescences that the bird is 
 called by the scientific name of Sarcorhamphus, or "flesh- 
 beaked." These colours, however, are not to be found in the 
 young bird, and do not make their appearance until the fourth 
 year. The name of ^ot/j^ is equivalent to that of pope, and 
 signifies that the bird is a sort of pope among the other vul- 
 tures. Its curious facility for discovering dead carrion and 
 snakes has already been mentioned on p. 380. 
 
 n 
 
 w. 
 
 Wallaba Tree {Epernafalcata). — So called from the shape 
 of the pods, which resemble a sickle, Lat, falx. " I passed 
 . . . the Wallaba Tree, with its thin curved pods dangling 
 from innumerable bootlaces six feet long." — 0. Kingsley. 
 The wood of the wallaba splits fi'eely, and is used for staves 
 and shingles. The oily resin exudes from the tree when 
 wounded. Also, see " Siloabali." 
 
 Wallababa. — See " Cotinga, Pompadour." 
 
 Waracaba. — See " Trumpeter." 
 
 Water-Hen, Splr-wincjed {Parra Jdcana). — Several species 
 of Jacana are known, and are at once recognizable by the 
 enormous proportionate length and slenderness of the toes. 
 
 This provision of nature enables them to walk with safety 
 upon the floating leaves of water-plants, and so to procure its 
 food, which consists of aquatic insects, kc. It is a good 
 swimmer and diver, but its powers of wing are trifling. The 
 magnificent leaf of the Victoria regia is a great favourite with 
 the Jacana, which finds a firm footing on a leaf which is able to 
 sustain the weight of a little girl some six or seven years old. 
 
 1 
 
KXPLANATOUY IXDICX. 
 
 -IHI 
 
 On each wing is a sharp and ratlier stron« spur, which has 
 earned for the bird the popular name of Spur-wing. It is 
 also remarkable for a curious leathery excrescence, which rises 
 from the base of the beak, both above and below. 
 
 til'l tt-WlNliKlJ WATKIi I1I;N. 
 
 Wateumamma. — Throughout the whole of Guiana there is 
 a superstitious dread of some strange being which, like the 
 nippen of Norway, the gnomes, goblins, and other malignant 
 semi-spiritual beings of Europe, the jinns of Asia, and the 
 jumbies of Africa, take a demoniacal delight in waylaying and 
 murdering travellers. 
 
 The Watermamma seems to be able to assume various 
 forms, though it generally prefers that of a human being. 
 Mr. Brown mentions it under the name of water-child. 
 
 " On our way we passed a deep pool, where there was an 
 eddy, in which the guide informed us there lived a " water- 
 child " covered with long hair. A woodskin (i.e. canoe) with 
 Indians was passing the spot one day, when the water-child 
 came to the surface, caught hold of, and upset the canoe. One 
 
it 
 
 488 
 
 KXPLANATORY INDEX. 
 
 of its occupants Miink with it, and, being seizetl by the chilil, 
 never ciiine up to the surface again.' ' 
 
 This account, it will bo seen, tallies exactly with that of 
 Water ton. 
 
 The same writer mentions this somi-supernatural being 
 under another name, i.e. "didi." 
 
 "Thetir.st night after leaving Peiir;iah we heard a long, 
 loud, and most melancholy whistle, proceeding from the 
 direction of the depths of the forest, at which some of the 
 men exclaimed, in an awed tone of voice, 'The didi! ' Two 
 or three times the whistle was repeated, sounding like thi*; 
 made by a human being, beginning in a high key, and dyirt' 
 slowly and gradually away in a low one. There were con- 
 flicting opinions amongst the men regarding the origin of 
 the-s-e sounds. Some said they proceeded frojn the wild hairy 
 man, or ' didi,' of the Indians ; others that they were pro- 
 duced by a large and poisonous .snako which lives in one tree 
 from its youth up, when it attains a great size, living upon 
 birds which are so unfortunate as to alight mix it, and thus 
 become victims to its power of fascination. The 'didi' is 
 said by the Indians to be a sliort, thick- set. and powerful 
 wild man, whose body is covered with hair, and who lives in 
 the forest. A belief in the existence of this fabulous creature 
 is univer.sal over the whole of British, Venezuelan, and 
 Bi'azilian Guiana. On the Demerara river, some years after 
 this, I met a half-bred woodcutter, who related an encounter 
 that he had with two didi 3 — a male and female —in which he 
 successfully resisted th-^ir attacks with his axe. In the fray, 
 he stated that lie wa^ a good deal scratched. His story 
 requires to be taken with a. very large grain of salt." 
 
 Wax of Wild Bee. — S'ee "Kurumanni." 
 
 AVhip-poor-Will. — See " Goatsuckers." 
 
 Whtp-S.vakp: {Philodryas viridissimus). — Several Snakes are 
 called by this name, because their long, slender bodies look 
 very much like the plaited thong of a hunting whip. Indeed, 
 more thivu once a Whipsnake has been grasped under the 
 
EXPLANATORY INDEX. 
 
 4>^'J 
 
 impression that it was a whip-thong, which had been dropped 
 in the path. 
 
 Some of them are venomous, but that which is mentioned 
 by Watorton is perfectly harmless. It is about two feet in 
 length, of a lovely green colour, and very active, darting 
 among the leaves and brancnes with such rapid agility, that 
 the slightest twig scarcely bends under its weight. 
 
 Wiio-ARE-vou. — A Goatsucker ; species uncertain. 
 
 Willy come-go. — Ditto. 
 
 Woodpecker, Red-headed {Melaner})e8 erythrocephaluB) . 
 
 WooDSKiN. — Popular name for a Bark Canoe. See 
 " Purple Heart." 
 
 Work-away. — A Goatsucker ; species uncertain. 
 
 WouRALL — The information acquired by Watorton is most 
 interesting, and it is to his energy and perseverance that we 
 owe our knowledge of the most wondeiful poison ever invented 
 by savage races. As may be seen from Waterton's account, 
 its manufacture is evidently the result of experience. More 
 than one substance is used in it, and as tlie Wourali made by 
 the Macoushis, one of the five grpat tribes, is so superior in 
 quality that the other tribes are glad to buy it, there is no 
 doubt that very great skill is required for its prejxiration. 
 Even among the Macoushis there are some families which are 
 celebrated for the strength of the Wourali made by them, and 
 the secret of its manufacture is said to be handed down from 
 father to sou. the pi-ay men, or conjurors, being the chief 
 manufacturers. 
 
 Several instances are known of experiments made in Europe 
 with Wourali, and which have failed. I have little doubt that 
 the poison with which they were made was not genuine, or that 
 it had been neglected. The savages of Guiana are quite as 
 expert at adulteration as the civilized niilknian, publican, or 
 grocer, and are perfectly cnpuble of making Wourali for the 
 market, but not for u.se. 
 
 Watorton never failed with the ex])eriments which he made 
 with the poison -which he himself obtained from tlie Maeoushi.'i 
 
 
VM 
 
 KXPLANATollY INKKX. 
 
 I 
 
 i:t 
 
 while dwelling amonrf them. Nor, as fiii* as 1 know, hi\8 
 any one failed with Wiitorton's poison. I made a few ex- 
 periments with some of it, and found that his account of it 
 was literally true. 
 
 Death was not instantaneous, but the creature which was 
 wounded seemed to be immediately deprived of all wish to 
 move. On the spot it was wounded, there it remained, its 
 eyes giving no indication of sensitiveness when touclied, and 
 its limbs gradually relaxing as if in sleep. Yet the poison 
 which I used had been preserved nearly forty years at Wal- 
 ton Hall, but it had been carefully kept from damp, which 
 injures, even if it does not destroy its powers. 
 
 In his essay on the Monkey family, Waterton makes a 
 passing, but valuable remark on the Wonrali : — 
 
 " If you are in want of a tender monkey, a month old or so, 
 to boil for broth or to educate as a pet, your only chance of 
 success is to shoot the poor mother, but not with a fowling- 
 piece. Nine times out of ten the wounded mother would 
 stay in the clefts of the trees, whex-e she would ultimately 
 peiish with her progeny. An arrow, poisoned with Woui-ali, 
 is your surest weapon. 
 
 '* Take a good aim, and in a few minutes the monkey will 
 be lying dead at your feet. The Wourali poison totally 
 destroys all tension in the muscles. Now, a gun-shot wound, 
 even though it be mortal, has not such an immediate effect. 
 
 " Knowing this to be the case, whenever a monkey was 
 wanted, recourse was had to poisoned arrows. By this pre- 
 caution, the ill-fated animal's existence was not prolonged 
 under the painful anguish of a deadly wound. The Wourali 
 poison would act as a balmy soporific, and the victim would 
 be dead at your feet in .a very short space of time." 
 
 The necessity for some such poison is evident from the 
 fact that on account of the exuberant luxuriance of tropical 
 vegetation, a mortally wounded monkey, if only able to 
 traverse a couple of hundred yards, would be hopelessly lost, 
 and whether the body remained wedged among the boughs or 
 
 |i 
 
KXI'LAN'ATOUY IN'DKX. 
 
 VM 
 
 if 
 
 fell to the ground would be a socrot only known to the 
 vultures and carrion-feeding insects. 
 
 'he peculiar " Quvke " or basket into whicli the Wourali- 
 maker puts his materials, is shown on pa^o Ml.'). 
 
 Chief among the materials is the Wourali Vine {Sfri/r/ttmn 
 tox{fern). This, as its name imports, is allied to the plant 
 which furnishes the well-known strychnine poison. 
 
 When full grown, its vine-liko stem is about throe inches in 
 diameter, covered with a rough greyish bark, like that of the 
 vine. The daik-green leaves are oval in .•shape, and ai*e set 
 opposite to each other. It is a tolerably common plant in 
 certain places, but it is very local, and among the abundance 
 of herbage is not readily distinguishable. Jt bears a round 
 fruit, shaped like an apple, and containing steeds imbedded in 
 a very bitter pulp. 
 
 The "root of a very bitter taste " is evidently the hyarri, 
 or haiarri, a plant belonging to the genus Lonrhncarpus, 
 chiefly used in poisoning fish. Of this plant the following 
 description is given by the Rev. J. H. Bernuu, in his Mission- 
 ary Labours in British Guiana \ — 
 
 "The haiarri is a papilionaceous vine, bearing a small 
 bluish cluster of blossoms, producing a pod abcut two inches 
 in length, containing some small grey seeds. The root itself 
 is stronger in its effects than the vine, and is always px'ef erred 
 by the Indians. A solid cubic foot of this root will poison 
 an acre of water, even in the rapids. In creeks and standing 
 water, its effects are still more extensive." 
 
 When used, the soft yellow roots are pounded with a 
 stone or mallet and steeped in water, which is then thrown 
 into the river. Heaps of the pounded roots are often found 
 on the stones on the river bank, showing that the fish have 
 been lately poisoned. They appear to be stupefied by the 
 poi.'^on, and float on tbe surface, when they are either shot 
 with arrows or simply lifted out by hand. 
 
 No injurious effect on the flesh is produced by the poison, 
 which has been conjectured to paralyse the gills, and so to 
 
 ill! 
 
f?l 
 
 i! 
 
 m 
 
 11 ! 
 
 I 
 
 If- 
 
 402 
 
 EXPLANATORY INDKX. 
 
 kill the fish, or at least to render it senseless, by asphyxiation. 
 Perhaps the paralysing effect of the Wourali may be due to 
 the haiarri juice. 
 
 In Mr. C. B. Brown's work on Guiana, thei'e is an interest- 
 ing account of this mode of fish killing : — 
 
 " I set out at an early hour one morning, with Ben, Eruma, 
 and Yackarawa, in a wood-skin, for a place where the Cowen- 
 amon Indians were going to poison a pool so as to obtain its 
 fish. After about two hours hard paddling we arrived at a 
 large cataract, called Cartoweire, and, taking our canoe into 
 smooth water above, found the Indians, eleven in number, 
 busily engaged in beating bundles of a soft yellow root 
 with sticks, fhese ha'ai'ie roots were each about two inches 
 in diameter, and of a light yellow colour, containing a yellow 
 creamy juice, having a disagreeable raw smell. Each bundle 
 was abo\it a foot in diameter and two feet in length. When 
 thoroughly pounded into pulp, they were thrown into canoes, 
 in which a little water had been previously placed, and then 
 the juice was wrung from tliem. The enclosure to be acted 
 upon was of an irregular shape, occupying about two acres of 
 river, and formed by dams of rock, built into the spaces be- 
 tween rocky areas and small islands. In building this the 
 Indians had left two large gaps open, one being whei*e the 
 greatest body of water ran in, and the other where it flowed 
 out. When wo arrived they had closed these gjips with a 
 wattle arrangement, so that all chances of escape for the fish 
 were cut off. Three canoes, containing the juice of six bundles 
 of haiarie, were then taken to the u[)per end of the enclosure, 
 and the subtle poison discharged from them. It was borne 
 down by the slight current, and mingled rapidly with the 
 pure dai'k water. JNIost of the Indians then got into tlie 
 canoes and pushed out, bows and arrows in hand, into the 
 middle -.f the enclosure, whilst the remainder, with my men 
 also furnished with the same weapons, stood upon the rocks 
 at the edge. In ten minutes time numbers of small fish caiiio 
 to the surfjice, ntid swam uneasily Jibout, trying to rise above 
 
EXPI.ANAT(.)KY IXDKX. 
 
 4'J3 
 
 water ; then soon were floating about quite dead. After an 
 interval of five minutes more, a single paeu showed its back 
 fin, and also tried to raise its head above water. An instant 
 more, and the whole place swarmed alive with lai'ge fish, pacu 
 and cartabac, all struggling and flapping at the surface or 
 whirling round and round. Many tried to force themselves 
 out of the water up the sloping surface of the rocks, and two 
 were successful in this, dying on the strand. From the 
 excited manner in which thoy struggled, it seemed to me as if 
 the poison had an intoxicating effect upon them. It might 
 have been that the contact of the poison with their gills had 
 produced a feeling of suffocation — hence their endeavours to 
 escape from their native element. 
 
 " It was a most exciting scene for a time, as the Indians 
 shot arrow after arrow into the bewildered dying fish and 
 hunted them ashore or into the canoe. In about an hour tho 
 murderous work was over, and 150 fine pacu find cartabac 
 were lying dead upon the rocks around the pool, the victims 
 of Indian prowess and poison. During tlie whole proceedings 
 I stood on the rocks at the upper end of the pool, and had a 
 fine view of the scene, the finest part of which was to see the 
 naked savage, in all his glory, drawing liis bow with strength 
 and ease and letting fly his arrows with unerring aim." 
 
 Another plant, called Konamie, is used for the same pur- 
 pose. It belongs to the ( 'omposifa\ 
 
 The " two bulbous plants" which supply the glutinous matter 
 I cannot identify. 
 
 As to the red pepper, ants, and snake-fangs, I do not believe 
 that they have any effect in strengthening the poison. When 
 rightly prepared, it has about the consistence of treacle, and 
 possesses a fragrant and penetiuting odour peculiarly its own. 
 Although so deadly when it directly enters the blood, it is, 
 like the poison of snakes, harmless when swallowed. I have 
 tasted some with which Waterton furnished me, and found it 
 to be intensely bitter, with somewhat of a quassia-like aroma, 
 and that its taste was as unique as its j^mell. 
 
 iiji 
 
 •ir 
 ii 
 
 h'a 
 
404 
 
 EXPLANATORY IXDKX. 
 
 Waterton not only succeeded in procuring the manufactured 
 poison, but also the whole of the materials of which it is made. 
 Unfortunately, the canoe was capsized in the falls of the 
 Essequibo river, the precious parcel was lost, and there was 
 no second opportunity of replacing it. As to the abandon- 
 ment of the temporary hut after the Wourali has been made, 
 subsequent travellers have said that the hut is not only aban- 
 doned, but burned to the ground, so as not to leave a trace of 
 the mode by which the poison was prepai*ed. 
 
 Wren {Troglodytes eudon). 
 
 Yam (Dioscorea sp.). 
 
 Yawaraciris. — All these little birds are Manakins, q.v. 
 and belong to the genus Fijn'a. 
 
TAXIDERMY. 
 
 i!;:"il 
 ,'i,. 
 
rt^ 
 
 rAXIDKKMY. 
 
 The late Arthur Strickland, of whom Waterton speaks in 
 such high praise as to say that " he had more orthodox orni- 
 thology in his little finger than most of our mighty closet 
 naturalists have in their entire carcass," and that his collec- 
 tion of British birds was without a rival ; never would have 
 his birds stuffed, keeping the skins in drawers. 
 
 Many years ago, when he was showing me his collection 
 I rather wondered to find that the skins had only just enough 
 cotton wool in them to keep them slightly open, but soon found 
 that he was right. Could he have put them all under Water- 
 ton's magic fingers he would have been glad enough that they 
 should be - set up," but he was too good a naturalist to entrust 
 them to ordinary bird stuft'ers. 
 
 In the chapter added to the Wanderimjs, and to be found on 
 page 335, Waterton gives the first account of the art which ho 
 created, and even with the aid of those few pages, any one 
 with ordinary intellect, a thorough knowledge of the animal, 
 and a fair acquaintance with anatomy, especially that of the' 
 skeleton, would be able to produce specimens of taxidermy 
 far superior to those which are ordinaiily to be found in 
 museums. But art must be progressive, and as he gained ex- 
 perience Waterton was perpetually discovering improvements 
 m taxidermy. Most of these improvements are given in one 
 or two portions of hj.s /'Jsmi/,, the principal part of wliich 
 will presently bo extracted. 
 
 K K 
 
 •Uti 
 
498 
 
 TAXILf.RMY. 
 
 I 
 
 i ; 
 
 
 First, we come to his mode of preparing birds' eggs. When 
 I was a boy, I used to be terribly annoyed, to iind that 
 valuable eggs, which had been obtained almost at the risk of 
 my life, were putrid, and swarming with maggots, owing to 
 the decay of the lining membrane. Waterton's gradual im- 
 provement in egg preservation was very intei'esting, as 
 showing the gr nvtli of the art. 
 
 Of course, he introduced the solution oi corrosive sublimate, 
 but he did so by sucking it into the egg by a hole at one end, 
 while the lips were applied to the other. Now, he soon found 
 that making a hole at the end of the egg spoiled its contour, 
 and he then made the two holes at one side. Afterwards, he 
 only employed one rather large hole, and emptied the egg by 
 shaking it in water. 
 
 I have found, however, that a single small hole on the side 
 will suffice. Get a piece of ordinary glass tu jing, about 
 eighteen inches long, and not more than a quarter of an inch 
 in diameter, heat it red-hot in v. spirit lamp or a gas flame, 
 and then draw it out in the shape of an elongated hour-glass. 
 r»reak it asunder, and you have a tube with a point about as 
 large as a darning needle. 
 
 Now break up the yolk of the egg with your needle, intro- 
 duce the glass point, and you can suck out the contents of the 
 egg in the tube. Inject water once or twice so as to clean it 
 completely, and then, by the same means, introduce the poison. 
 
 For soft -shell eggs like those of the snake or turtle, the 
 best plan is to empty them, poison them, fill them with sand, 
 and let them dry, hanging them separately in pieces of net or 
 muslin to keep them from flattening. When quite dr}'-, the 
 sand can be shaken out and the egg will ever afterwards re- 
 tain its shape. If the egg be " bard set," the best plan is to cut 
 out a small piece of the shell, dry it on a curved surface, lest 
 it should lose its shape, and extract the contents through the 
 opening. When it is dry, the piece of shell can be replaced 
 with a little thin paper and diamond cement, and if neatly 
 done, the junction will scarcely be visible. 
 
rAXIDEUMV 
 
 4'JO 
 
 When 
 d that 
 risk of 
 7\T\g to 
 xal im- 
 Ing, as 
 
 •liinate, 
 ne end, 
 n found 
 iontour, 
 irds, he 
 > egg by 
 
 the side 
 ■, about 
 an inch 
 s flame, 
 ir-glass. 
 ibout as 
 
 intro- 
 of the 
 clean it 
 poison, 
 tie, the 
 th sand, 
 net or 
 ry, the 
 ards re- 
 s to cut 
 ace, lest 
 )ugh the 
 eplaced 
 neatly 
 
 i 
 
 There are some eggs, notably those of the kiiiglishoi", which 
 alter their colour when empty. The kijiglishor's egg, when 
 fresh, is almost translucent, and a beautiful rosy white, but 
 when blown, it becomes as opaque and us dead white as the egg 
 of a fowl. I hit upon a plan of restoring the colour which 
 answered admirably, and is veiy easy. Mix Ciirmino powder 
 with melted white wax, and make it much larker than the 
 required hue, as the shell absorbs much of the colour. 
 
 Empty and poison the eggs as before, and got the wax to 
 boiling point. Heat the egg and gl.iss tube as much as thoy 
 will bear, draw a few drops of the wax into the tube, blow it 
 into the egg, and keep turning the egg over and over in front 
 of a fire until every part of it is equally coloured, llemove it 
 from the fire and continue to turn it until it is cold, and the 
 colour will then be so perfectly restored that if it be placed 
 among a number of fresh eggs, the keenest eye will not be able 
 to distinguish it. If the egg be laid down before it is cold, 
 the wax will run downwards and make the colour streaky 
 instead of regular. 
 
 Here is Waterton's account of the improvements which he 
 made in taxidermy after he published the Wioaltrhajs. 
 
 " Those who preserve quadrupeds for cabinets of natural 
 history seem not to be aware that, after the skin of the 
 animal has been taken off, there is a necessity for some part.s 
 of it to be pared down from within. These parts are chiefly 
 the nose, the lips, and the soles of the feet. Unless they bo 
 rendered thin by the operation of the knife, there will be no 
 possibility of restoring to them that natural appeai-ance whicli 
 they were seen to posii3ss in life. The inner skin of the ear.s, 
 too, must be separated from the outer one, until you come 
 close to the extreme edge. Nothing short of this operation 
 can save the ear from becoming a deformity. 
 
 " Every bone in the skin, to the last joint of the toe, next 
 the claw, must be taken out, in order to allow the operator 
 an opportunity of restoring the skin to its former just 
 proportion. 
 
 K K 2 
 
' I 
 
 M 
 
 ., I 
 
 500 
 
 TAXIDERMY. 
 
 " The mouth must be sewed up from the inside (the skin 
 being inside out when you sew it), beginning exactly in the 
 front, and continuing the operation each way to the end of the 
 gape. When the skin is taken out of the solution, it must be 
 tilled quite full of chaff or saw-dust (but I prefer chaff), not 
 minding whether the fur be wet or dry. When this has been 
 done, the skin has almost the appearance of an inflated bag, 
 quite deficient in feature and in muscular appeai'ance. 
 
 " There are now no obstacles either from without or from 
 within, to irapeue the artist's progress. The skin is perfectly 
 free from all chance of putrefaction, is quite supple, and will 
 remain so as long as required. There is no hard body inside 
 to obstruct the transit of a working iron ; there is not any- 
 thing in the shape of wires to prevent him from lengthening 
 or shortening the neck, body, thigh, and legs, according to his 
 own judgments. 
 
 " Now we proceed to support the skin in any attitude the 
 artist may wish to place it in. 
 
 " Join two pieces of wood in the shape of a carpenter's 
 gimlet, and of size corresponding to the size of the animal. 
 When you have nearly filled the abdomen with chaff, introduce 
 this machine, and let the shank hang down outside of the 
 skin, just as though it were a fifth leg in the centre of the body, 
 equi-distant fi-om the fore and hind legs. This fifth leg, or 
 what may be called the shank of the gimlet, is of any suffi- 
 cient length, and is passed through a hole in the table before 
 you, and then fastened with a couple of wedges. By this 
 contrivance you can raise the animal as high as you wish, or 
 you can lower it at your pleasure, and the feet will just touch 
 the table, without I'equiring any wire inside to support them. 
 
 " Everything is now ready for the artist to exercise his 
 abilities. 
 
 "With a piece of iron, from the size of a large darning 
 needle to that of a ramrod (or larger and thicker still, if the 
 bulk of the animal require it), and shaped at one end like a 
 carpenter's pricker, he will push out eveiy part of the skin 
 
 I 
 
TAXIDEliMY. 
 
 501 
 
 • 
 
 which ouyht to Iv /;ished out, and then reduce with the euJ of 
 his finger any piirt that may be too prominent ; having al- 
 ready mad(! divers small holes in the skin with his penknife, 
 in order to afford entrance to the working iron. Thus, a small 
 hole on the top of the head will enable him to reach the nose, 
 upper lip, and cheeks ; another behind the loot of each car ; 
 another under the jaws ; others, again, on the back, that ho 
 may reach the legs, and the remaining parts of the body. 
 Under each foot there will also be a hole, to give him the op- 
 portunity of getting at the toes. The lips are by far the most 
 ditficult part to manage. The operator must have a working 
 iron in both hands. One of these will do the work wuthin the 
 head, and the other that without, for the lips require to be 
 reformed with a beautiful rotundity, and this can only be 
 effected by means of the inner and the outer irons working in 
 opposite directions. During the actual operation, the animal 
 need not be kept in its original position. A smaller animal 
 may be placed on the operator's lap, the larger may be thrown 
 on the ground, or on the table. Every day the nose, and lips, 
 and orbits, ought to be touched with the oil and turpentine, in 
 order to keep them moist. At first, after you have used the 
 working iron in every quarter where it is required, there will 
 be no appearance of a re-formation of the features. Never- 
 theless, in the due course of time, as the skin stiffens, the 
 artist will see the features gradually appear ; and every day he 
 will be more and moi'e content with his work. At last, the 
 skin will retain the slightest impression communicated to 
 it by the touch of the working iron. Then the artist will 
 have it fully in his power to reproduce wrinkles, or warts, or 
 hollows, or a smooth surface, just as occasion may require. 
 
 "The fur will be equally under his command. He will 
 raise it, or depress it, according to circumstances, and it will 
 retain the position ever after. Thus a stuffed cat in anger 
 will exhibit a tail of the same extraordinary bulk which it 
 does when a dog threatens its existence. 
 
 "As there are parts of a quadruped's skin which are bound 
 
 ^:1 
 
 l.i 
 
 !f1 
 
602 
 
 TAXIDERMY. 
 
 li I 
 
 111 ' 
 
 I! 
 
 l! I 
 
 down, as it were, to the bone (at the oyos for example), it will 
 be necessary to pass a throiid, with a sufficient knot at one 
 end, through these parts, and to let the end without a knot 
 hang loose after it has been drawn out at the opposite quarter. 
 Thus, there must be a thread in the extremities at the gape of 
 the mouth, and one at the corners of the eyes ; and others in 
 different parts of the body, according to the operator's judg- 
 ment. By pulling these at the end which hangs out, he will 
 be enabled to depress the parts into their natural shape. 
 
 " The artificial eyes must be put in on the first day of the 
 operation, and taken out and put back again every time the 
 head of the specimen is modelled. 
 
 " When all is completed, and tiie skin has become perfectly 
 dry, the artist takes out the chaff or saw-dust, and he finds 
 that the si)ecimen is quite firm enough to stand without any 
 support from wires. He cuts three sides of a square hole 
 imder the feet, to let out the chaff, and when this is done, he 
 returns the skin to its place. 
 
 " A slit must be made in the crown of the head, or under 
 the jaws, to allow him to fit the artificial eyes with a little 
 putty or wax. The slit, if properly done, will leave no mark 
 on the fur. 
 
 " If the quadruped be stuffed in distant countries, with an 
 intention to be sent home, it may be cut up, when finished, 
 into three or four different pieces, and this will facilitate the 
 carriage. When dividing it, the operator must take care to 
 hold his knife so as to humour the angle which the fur forms 
 with the skin. Thus, were I to cut a preserved skin in two 
 parts, the blade of my knife would point to the head, and the 
 haft to the tail of the animal. By attention to this, not a 
 hair of the fur will be cut during the operation. 
 
 "I will ju.st add here (although it be a digre.ssion), that 
 there is no difficulty in making the legs and feet of eagles, 
 turkeys, and other large birds, retain their natural size. 
 You may go through every known museum, and you will find 
 that the legs of these, and of all large birds, are dried and 
 
 iL. 
 
TAXIDERMY 
 
 603 
 
 shrivelled, as though they belonged to the mummiea of 
 ancient days. In order to give the legs of birds a natural 
 appearance, and a natural size, the skin from tlio very claws 
 to the top of the leg must bo separated from tho bone by 
 running a working-iron betwixt it and the bone, and then 
 modelling the skin with the working iron. 
 
 "The wattles of fowl, the caruncles of turkeys, and the 
 combs of cocks, by the simplo proce s of internal modelling, 
 may be made to retain their natural size." 
 
 All improvements are gained by experience, and, when 
 I first knew Waterton, he had abandoned the box of cot- 
 ton wool, and employed a far superior mode of fixing his 
 subject. 
 
 Instead of a box, he had made a simple framework, as here 
 shown, both ends being open and the upper part projecting 
 considerably. This was bored full of holes at irregular inter- 
 
All-l 
 
 'A\II>KI!MY 
 
 ) I 
 
 r' I 
 
 I 
 
 vain, 80 that an upi'l^'ht rod iniyht bo inwerteil into any of 
 (hotn. An onlinary wine cork wns bored and passed over tho 
 rod, 8() as to slide rather stitHy up and down. A stont pin was 
 stuck into the cork, a piece of twine tied to it, uud the simple 
 apparatus was complete. 
 
 The sketch was taken while Waterton was preparing a phea- 
 sant. A livrge lump of wax is pressed on the beuk and from 
 it projects a sharp needle point, which can be thrust into the 
 cork. It will be seen that tho bird can be tlius supported in 
 any position, and tho wings raised or depressed at pleasure. 
 For convenience sake, Waterton usually i)laced his bird-stand 
 on a little oak table, which he called "Peggy," and which 
 could be turned about so as to enable him to reach every 
 part of the bird. The piece of twine attached to the pin 
 was Ui-ed for giving the proper position of the head, the twine 
 being hitched into notches cut in the side of the stand. 
 
 On such a stand I have prepared birds in the act of stand- 
 ing, flying, swimming or feeding, and found it to be in- 
 valuable. It also answers well for the smaller animals, but 
 the larger kinds must necessai'ily have stands of coi-respond- 
 ing size. 
 
 I found it better to wash all birds and animals in soft, warm 
 water and soap, then rinse them thoroughly, and then, if birds, 
 soak them in the poison. I used generally to dry them by 
 tying a piece of strong tape to their legs and swinging them 
 backwards and forwards, so that not even the most delicate 
 down was matted. It was wonderful to see how beautifully 
 their colours came out after the washing. ' 
 
 It is better to make absolutely sure of the various parts 
 of the skin to be noted, such as the position of the joints, A'c. 
 A very simple way of doing so is by fixing the animal in the 
 required attitude before skinning it, and passing a fine white 
 thread through the skin at the different points, letting the 
 ends hang down. They will come out easily when the opera- 
 tion is completed. Measurements, and if possible, a sketch 
 to scale should be taken, so as to guard against the usual 
 
 I 
 
 
'I'AXIDKllMY 
 
 [)UJ 
 
 ' 
 
 error of muking the body too long, logs uik'(|Uii1, unci, if ii 
 bird, tho wings too wide apart or too clohO togothcr. 
 
 Now for the poison. 
 
 Corrosive sublimate is not easy to procure, and the shortest 
 way is to get an order from a medical man to purchase a 
 pound or two at Apothecaries' Ifall. Chemists may not sell 
 it without such an order. Being very diilicult to dissolve, it 
 must be reduced to an almost impal])ablo powder. It is 
 better to have this done by a chemist, as tho i)owder is apt to 
 fly about and is very dangerous. 
 
 Alcohol, even when methylated, is rather costly. Waterton 
 suggested to me to try water instead of spirit. I did so, and 
 found it answer for iill kinds of fur and most kinds of feathers. 
 But, owing to the closeness of the plumage of water-birds, 
 spirit is absolutely necessary for them, and they must not 
 only be thoroughly steeped, but continually turned and the 
 feathers repeatedly raised and pressed. As long as a single 
 bubble of air is left among the feathers, there will be a spot 
 which the poison has not reached, and which the moth or 
 mite is sure to find out. 
 
 Take great care not to dip the hands in the solution. 
 Should there be a scratch on them, they will suffer intense 
 pain for a long while, and the nails will always be stained a 
 deep brown which cannot be removed but by the gradual 
 growth of a new nail. Good India-rubber gloves are very 
 useful. 
 
 The right strength of the poison is of very great import- 
 ance, and can be secured by Waterton's plan of making it too 
 strong at first, and then adding spirit until it leaves no white 
 deposit on a black feather which has been dipped in it and 
 drjed. 
 
 For moths and butterflies, spirit is also necessary, as water 
 glides off their scale covered wings, leaving no poison behind 
 it. They should, when "set," be dipped in the solution, and 
 placed to dry just inside the window, with their heads inwards, 
 so that a draught of air shall blow up the hairy down with 
 
506 
 
 TAXIDERMY. 
 
 \l ' 
 
 
 which so many are adornerl, and which will be matted together 
 if this precaution be not taken. A pair of small bellows 
 will be found very useful, especially if the nozzle be kept 
 heated with a spirit lamp. A drop of the solution is sure to 
 form at the end of the wing, and should be removed with 
 blotting paper, as it is apt to bend the tip downwards and 
 spoil the shajDo. 
 
 Large-bodied moths should have the abdomen removed, 
 the contents extracted through the wound, the inside painted 
 with poison, and stuffed with cotton -wool. When body and 
 abdomen are both dry, they can be joined by diamond cement. 
 Not a trace of the junction will be visible. The insect will 
 not only be im])ervious to mites, but it will always preserve 
 its shape, and it will never be subject to " grease," tliat 
 terror of all entomologists. 
 
 Eyes will be found exceedingly troublef-onie, and the 
 very greatest care is needed, especially in the management 
 of the skin at the corners. Always let it be remembered 
 that eyes in life never start spherically out of circular sockets. 
 Care must be taken to have the eyes of the right colour. 
 Most eyes can be procured from the shops, but there aie 
 some, notably those of the toads and snakes, which must be 
 made l)y the operator. The best plan is to have a number of 
 little hollow glass globes, like those used for doll's eyes. 
 Paint them o)i the ini^ide, fill them with melted wax, push a 
 piece of iron wire into the wax, and they will answer ad- 
 mirably, giving a life-like aspect which can never be obtained 
 by the glass eyes of the shops. 
 
 Two golden maxims for the followers of Waterton, are, 
 first, never be di.scouraged ; second, never be in a hurry. 
 
 There is need f r them both, especially the former. 
 Nothing can be more discouraging to a beginner than the look 
 of a bird newly skinned, and partly filled with cotton-wool. 
 It is more like a bundle of miscellaneous feathei's than a bird. 
 The upper and lower beaks are quite independent of each 
 other, the neck looks like a sausage, the wings are turned 
 
TAXIDERMY. 
 
 507 
 
 
 in different directions, and the legs and tail dangle about 
 helplessly. N"o vestige of an attitude can be seen, and 
 the operator feels very much inclined to give it up in 
 despair. 
 
 Day after day he works at it according to instructions, and 
 seems to make no progress whatever. Presently, however, he 
 is encouraged by finding that the skin begins to respond to 
 his touch, and before very long, it becomes as plastic as clay 
 in the hands of the sculptor. 
 
 Now comes the time for the second maxim. The skin, as 
 Dickens's butclier said of his meat, " must be humoured, not 
 drove," and if any attenijjt is made at hurraing, it will be 
 totally spoiled. 
 
 Waterton's deliberation while preparing a bird or animal 
 was almost exasperating. He would give it a touch here and 
 a touch tliere, smooth down a starting feather with the instru- 
 ment whicli he mostly used, a blow-gun arrow from which 
 the poison had been removed ; or, he Avould slightly alter the 
 pose of tlie liend, or mould afresh a piece of skin which was 
 beginning to shrivel. He thus kept every feather and hair 
 under command, and put in touch after touch to the skin just 
 as a painter does to his canvas. The result was absolute 
 perfection, but tlie means nppeared strangely inadequate. 
 
 No one could prepare a humming-bird like Waterton. 
 Excei^t in his collection, it is next to impos.sible to find a 
 stuffed hunmiing-bird in whicli tlie glittering gorget is not 
 disfigured by little dark spots. Each sucli sjwt shows that 
 a feather is inissinir. 
 
 Now Waterton found that such missing feathers had 
 rarely been pulled out of their sock<^ts and lost, but had 
 been dragged under the otlier feathers by the contraction 
 of the skin. He always searched for them, found them, 
 drew them from their concealment and laid them in their 
 places. So, the breasts of his hummisig-birds simply blazed 
 with gold, ruby, azure, or emerald, according to their species, 
 as they did in life. 
 
508 
 
 TAXIDEKMV. 
 
 The same patient care enabled him to give to all feathers 
 and furs the "flow" which they possessed in life, and which 
 no method except his own has been able to restore. 
 
 Waterton also found that two pieces of skin if properly 
 moulded together while wet, would adhere to each other firmly, 
 and that a little fine glue would cause them to unite as one 
 piece. It was through the knowledge of this fact that he was 
 able to produco the ludicrous combinations of different crea- 
 tures which he placed in his museum and ticketed with all 
 kinds of quaint names. 
 
 There was for example, "Noctifer," or the Spirit of Night, 
 made of portions of a bittern and an eagle owl, both nocturnal 
 birds. 
 
 Then he had an absurd grouj) of John Bull surrounded by 
 difficulties. John Bull was a tortoise with the head of an 
 
 Mil Tlli:h. 
 
 I 
 
 exceedingly stout but exceedingly worried man. Ife was sup- 
 porting the eight hundred millions of national debt, to which 
 such frequent reference is made both in the Wamlprimj.^ 
 and Essays. Clinging to his back, and driving its claws 
 into him, is perched " Di.abolua bellicosus," a sort of grin- 
 ning lizard all over .'ibnormal spikes and horns. Before him 
 
'I'AXIIiEiaiV. 
 
 500 
 
 j,'(tL'8 " iJiabidns aiiibitiosus,'" with outspread wings. " Dia 
 bolus illudoriH " is guiding him on liis path, and " Diabolus 
 cn'i'ulcus," with its open mouth and sharp toetli, is bringing 
 up tho roar. 
 
 Thn museum was fidl of these taxiderinal jokes, and not 
 long l)eforo hiu deatli I procured for him a quantity of the 
 exuvin- of the wjrpents in thi- Zoological (Jardens, so that he 
 might work them into new combinations. Whether he did 
 so I do not know. 
 
 Not least of tho excellences of Ids system was the porta- 
 bility of tho objects j^repared by it. We know how heavy is 
 
 Pdl.KCAT. 
 
 even a stuffed dog of ordinary size, with all its internal para- 
 phernalia of iron bars, wood, tow, &c. But Waterton's 
 specimens are absolutely empty, the skin depending wholly on 
 itself for support, and being as light and elastic as thin 
 horn. 
 
 Moreover, his plan of cutting the prepared skins to pieces 
 which could bo packed within a small compass, reduced their 
 volum<( a;4 well as weight. They could be taken to pieces 
 and put tog<(ther again in a few minutes, without showing 
 tho least sign of a junction. Huch, for example, is the speci- 
 men here given. No one would ever suspect that it was not 
 an unbroken skin, so admirably has it been prepared, and so 
 perfectly are t!ie junctions concealed by the fur. 
 
 iSrnrvollous, however, as were Waterton's achievements in 
 taxidermy, there was omt pro])lem which he never succeeded 
 in solving, ujimely, hnw to jtrevonl bare skin from tiu-ning 
 
 F 
 
610 
 
 TAXIDEiniY. 
 
 black. IIo could model the form of a Laud, a lip, or an ear, 
 to the minutest wrinkle, but, do what he would, the skin 
 always turned black. It is much the same with insects, such 
 as dragon flies, the colour of Avhich depends on living tissues 
 under the skin. He suggested that the body might be taken 
 to i)ieces, cleaned, painted on the inside, and put together 
 again. 1 tried this plan, and it answered very well as far as 
 the abdomen went, but utterly failed in the thorax and head, 
 which persisted in turning black. For the solution of this 
 and other problems in taxidermy, we want another Waterton. 
 Where shall we find him ? 
 
 i 
 
\ 
 
 INDEX. 
 
 Aborigines, 89. 
 Acaiari, 118, 238, 353. 
 Accident at Dover, 26. 
 
 Mount Cenis, 28. 
 
 Fatal, 81. 
 
 on " Pollux," 26. 
 
 to foot, 213. 
 
 with gun, 30. 
 
 with ladder, 31. 
 Aconri, 104, 3o4. 
 
 Teeth of, 135. 
 Acowiiys, 237, 477. 
 Acuero, 133, 354. 
 Adventure near lionie, 28. 
 
 on Orinoco, 15. 
 
 with Labarri, 15. 
 iEta, 120, 121, 355. 
 Agave vivipara, 469. 
 Ague, Tertian, 150. 
 Ai, 140, 357, 470. 
 Ai killed by Wourali, 141. 
 Albicore, 156, 857. 
 Alligator, 120, 357. 
 American hulics, 292, 299, 303. 
 Anaconda, 380. 
 Anhingas, 203, 266, 357. 
 Anolis', 385. 
 Anolius huUaris, 385 
 Ants, 129, 222, 358. 
 
 Black, 358. 
 
 Ooushie, 228, 358. 
 
 Nests of, 121. 
 
 Itiinger, 228, 358. 
 
 Small red, 360. 
 
 \Vood, 93, 
 Ant-bears, 93, 223, 231, 269, 361. 
 
 Large, 93, 361. 
 
 Ant-bear, Small, 93, 363. 
 Ant-birds, 364. 
 Antidotes to Wourah, 143. 
 Antigua, 310. 
 Apes, 13arbary, 10. 
 
 of Gibraltar, 9. 
 Apoura-poura, 112. 
 Apron, Indian, 238, 364. 
 Aquiro, 354. 
 
 Aras, 113, 121, 156, 196, 365. 
 Ara ararauva, 365. 
 
 macao, 365. 
 Ara humming-bird, 178, 423. 
 Ara, Scarlet, 111, 123, 266, 
 
 319, 365. 
 Arapunga alba, 380. 
 Ardea cgrcttn, 405. 
 
 Icucogaster, 420. 
 
 vircsccns, 420. 
 Armadillo, 93, 229, 231, 366. 
 Arnotto, 445. 
 Arowacks, 237, 477. 
 Arrow reed, 372. 
 Arrows, 137, 240, 366. 
 
 Blow-gun, 366. 
 
 Fish, .-jeS. 
 
 Hog, 370. 
 
 Poison-wond, 371. 
 
 Eappu, 371. 
 
 Wourali, 133. 
 Arrowroot, 119, 372. 
 ArundiHaria Sc/tomburgkii, 444. 
 Ash, decayed, 53. 
 Aspidospcnmim nxrlsum, 445. 
 ytstrocarijoii acuf>:.dum, 354. 
 Attagcn aquila, 411. 
 Aura Vulture, 4S4. 
 Aricoinia nitida, 400. 
 Awning, Canvas, 372. 
 
 i 
 
^- 
 
 /■■ 
 
 ■V 
 
 r 
 
 IN'DKX. 
 
 !: 
 
 B. 
 
 Halsam-cai'ivt, 117. 
 
 Unmboo, 373. 
 
 Diiinha.sii arinidiniicni, 373, 
 
 Uiiniiiia, 22(5, 373. 
 
 liiinlv!', Sir Josi'iih, 214. 
 
 Hiul null IPS, 313. 
 
 Harlmvv A[>(>, 10. 
 
 15atH, 97. 
 
 Bearded Saki, 374. 
 
 Heeslm monkey, 374. 
 
 Beetle, Kliiiioornis, 312, 4t>(j. 
 
 Bell-binl, LSO, 374. 
 
 Bete-rou^% 98, 117, 235, 374. 
 
 Bil.iri, 414. 
 
 Bill of Toucan, 182. 
 
 Bio(;rai'I1Y of Watkuton, 1. 
 
 Bird's et^gs, Mode of preparing, 498. 
 
 Birds of Deinerara, 94, 177. 
 
 Bird-stuffing, 214, 335, 497. 
 
 Bird-vine, 106. 
 
 Bisa, 93, 374. 
 
 Bittern, 197. 
 
 Bixa orellana, 445. 
 
 Blackbirds, 1(J6. 
 
 Black-breasted Plover, 457. 
 
 Black Pepper, 454. 
 
 " Black Vomit" in Spain, 11. 
 
 Blow-pipe, 113, 133, 374. 
 
 Arrows of, 366. 
 Blue Heron, 197, 420. 
 Boa Constrictor, 231. 
 Boa impcrator, 400. 
 Boat-bill, 197, 376, 401. 
 Bob-o-link, 481. 
 lioclora, 186, 370. 
 Bois Immortelles, 177, 377. 
 Bomhax cciki, 395. 
 Bonito, 156, 377. 
 Bouradi, 181, 377, 477. 
 Bows, 137, 240. 
 Box-tortoise, 475. 
 Bradnjuriis satuiias, 374. 
 Bradi/piis Jidactylus, 472, 
 
 tridadijhis, 470. 
 Broadway, 303. 
 Brosimum nuhldil, 433. 
 Brown Heron, 420. 
 
 Mcnkev, 93, 435. 
 BufTalo Town, 291. 
 Pxiletre, 10,-;, 377. 
 P.iiro-koro, 433. 
 
 Burueh, 377 
 
 iJush-master, 96, 378, 398. 
 Bush-rope, 91, 92, 106, 378. 
 liuzzard, Turkey, 485. 
 
 C. 
 
 CAi!»A(iK-TiM;i:, 101, 370. 
 
 Vairina munc/tatu, 410. 
 
 t'dlhidiinn a rbort'Hi-enii, 435. 
 
 Camoudi, 95, 3S0. 
 
 Camouni, ]{., 195. 
 
 Campanero, 97, 180, 262, 38(». 
 
 Canadians, 296. 
 
 Cancruma cuiicrophaya, 4(il. 
 
 Canoe, 120. 
 
 Canvas awning, 372. 
 
 Caoutchouc, 425. 
 
 Capivi, 381. 
 
 Ccijiskum aiiniiuni, 454. 
 
 frittf'sccns, 454. 
 Caprimiihjus, 111, 233, 381, 413. 
 Capture of Cayman, 272. 
 Caracara, 261, 381. 
 Carapa Gainncnds, 400. 
 Cardinal bird, 415. 
 Cufdinalis Vinfinianm, 415. 
 Caribs, 237, 477. 
 Carica 2iapai/a, 445. 
 Carriage pond, 77. 
 Cartabac, 369. 
 L'urijoi-ar tomciitomnn, 468. 
 CaryopIujUns arotnaliciis, 390. 
 Cassava, 381. 
 
 Bitter, 237. 
 
 Sweet, 237. 
 Cassareeit, 382, 
 < 'a.ssiipies, 188, 383. 
 Castor-oil, 213, 383. 
 Castries, 313. 
 Caihan'.'iffs aura, 484. 
 Cat.s, 48. 
 Cayenne, 1 70. 
 
 Cayman, 17, 90. 232, 2G4, 270, 
 384. 
 
 Back of, 270. 
 
 Capture of, 272. 
 
 Hook, 270, 385. 
 
 Noise of, 264. 
 
 Teeth of, 270. 
 Obus (ipella, 435. 
 Crrcofalxs pir/irnsi/is, 457. 
 Ccro.rij/aii. uiiilii-oln, 432. 
 

 INDEX. 
 
 ;"iI.T 
 
 ■ 390. 
 
 licum. 
 
 390. 
 
 Cerviis humilis, 402. 
 
 Cliiimeleon, 9b', 38r». 
 
 C'ha7npsa nigra, 384. 
 
 Char adrius semi pal malm, 
 Vmjinianus, 457. 
 
 Chcironedcs Yapnck, 442. 
 
 Chigoe, 236, 385. 
 
 Childhood, 5. 
 
 Cicadas, 401. 
 
 Ciniiai , IVu, 
 
 C'mna,n^..um Zcj. 
 
 Cirouabali, 470. 
 
 Cistuda Carolina, 47.'). 
 
 Citrus auranlium, 443. 
 limomim, 433. 
 
 Climate of New York, 305. 
 
 Clock of Sir Thomas More, 2. 
 
 Clove-tree, 170, 172, 390, 
 
 Clusia alba, 406, 
 
 Coeal, 392. 
 
 Coccygus Americanus, 401, 
 
 Cocoa-nut, 1G4, 390, 
 Cock of the Rock, 113, 172, 330, 392, 
 Corns nucifera, 390. 
 Coelogcnys pr"-a, 432. 
 Coendoo, 457. 
 Cof/'cea Arabica, 393. 
 Colfee-trees, 89, 393. 
 Collared Teccary, 449. 
 Conurus solstitialis, 431. 
 Convolvulus batatas, 4G0. 
 Coot, 203. 
 
 Cootcr, 475, ' 
 
 Copaiba, 381. 
 Copaifera publifora, 461. 
 Copaifera sp., 381. 
 Copal, 90, 394. 
 Coral snake, 96, 394. 
 Corrosive sublimate, 340. 
 Coryjihene, 404. 
 Coryphenc himmris, 404. 
 Cotton, 101, 396. 
 Cotingas, 156, 179. 394. 
 Pompadour, 179, 394. 
 Purple-breasted, 179 
 Purple-throated, 179, S94, 
 Scarlet, 179 
 Ciitinga cayenna, 394, 
 2}om])adoura, 394. 
 Cotton-treo, 395. 
 Couanacouchi, 96, 129, 132, 398 
 Coucourite, 121, 123, 13 1, 398. 
 Couguar, 267, 399. 
 
 Coulacanara, 249, 255, 400, 
 
 Courada-tree, 94, 400, 
 
 Coushie ant, 228, 
 
 Crab, 318, 
 
 Cral)ier, 94, 203, 401. 
 
 Crabwood, 103, 
 
 Cranes, 203. 
 
 Crux alcdor, 460. ' 
 
 Crickets, 98, 401, 
 
 Ci k( ( 313, 401, 
 
 Cuc.jO, 406, 
 
 <'uia, 187, 400. 
 
 Culcx pulicularis, 440. 
 
 pi/jie>is, 440, 
 Culverin f(jund in lake, 45 
 Curial, 120. 
 Curlews, 208, 
 Curlew, Plack, 121. 
 
 Scarlet, 94, 121, 170,203, 4ol. 
 Curucuru, 398, 
 Custom-house duties, 281. 
 Cutlass or Machete, 92. 
 Cuxio, 374. 
 
 Cyanocorax crisfatus, 425. 
 Cydothwrus didadylus, 363. 
 
 D. 
 
 Dardv Quashi, 233. 
 
 Dura, 180. 
 
 Darter, 358. 
 
 Ikmjproda acouchi, 354, 
 
 Dasypus sexcindus, 366. 
 Death of Waterton, 81, 
 
 Deer, 92, 121, 402, 
 Demerara, 173. 
 Otter, 442. 
 Voyage to, 13. 
 Dcrntypus coromdus, 4i7. 
 Diisceiit of Waterton, 2. ' 
 Diamoiul Rock, 313, 402. 
 Dicotclcs tujacu, 447. 
 Didi, 487, 
 
 Diet and raiment, 210, 
 JHoscorea sp., 4lt4. 
 Diospyros ebniam, 4(i4. 
 Diss(!cting, Kcquisites for, 335. 
 Divers, 154. 
 
 Dodges, The "Squire's," 72. 
 Dog, Kxperiment on with Wourali. 
 102, 
 in Guiana, 146, 
 Sloiii, 472. 
 
614 INDEX. 
 
 Ihl ichonyx oriziuorus, 4Sl. 
 
 Dolphin, 15«, 4f)4. 
 
 I)i)iiiiiucii, iill. 
 
 J)ry-n>t, 50. 
 
 Dii.'iilabiili, 90, 401. 
 
 Diuks, 203, 21)0, 440. 
 
 Duck, Muscovy, 121, 197, 440. 
 
 Musk, 440. 
 Ducklar, 358. 
 DurjKluara, 201, 404. 
 Duty oil specimens, 2S1. 
 
 E. 
 
 Eaoles, 202, 302, 404. 
 E;ij,'le, Wliite-headed, 302, 404. 
 Ebony, 90, 404. 
 Kriio stone, 46. 
 Ivhnonstono, Charles, 240. 
 Ef,Met, 94, 121, 170, 197, 40.^ 
 J'JIaps lemniscatiiti, 452. 
 VA Dorido, rj.*;,- 440. 
 Kniigrauts, Irish, 297. 
 Emys tracaxa, 432. 
 Ki)ernafalcala, 486. 
 J-Jperna, sp., 470. 
 Erythrina unihrosa, 377. 
 Escape from Spain, 12. 
 Esseipiibo, Falls of, 110, 277. 
 Eunectes viarinaa, 380. 
 Exi'LANATOKY InDEX, 3.^53. 
 
 EjcococIus vol'Uuns, 410. 
 Eyes, False, 345, 600. 
 
 F. 
 
 Face of Couxtuy, 290. 
 Falcons, 202. 
 False eyes, 345, 506. 
 Falls, 102, 112. 
 
 Essequibo, 110, 277. 
 Fever at Demerara, Yellow, 209. 
 
 Severe attack of, 211. 
 F'lcui elastica, 426. 
 Fiery Topar, 423. 
 Fig, Wild, 90, 91, 180, 194, 4m. 
 Finch, Red-headed, 104, 200, 400. 
 Firelly, 99, 117, 406. 
 FiivsT Journey, 87. 
 Fish, 96. 
 
 Fish, Flying, 150, 410. 
 Fish arrow, 308. 
 Flamingo, 94, 170, 203, 407. 
 Fly, 119. 
 
 Sand, 98, 117, 467. 
 Flying Fish, 156, 410. 
 Forest on banks of Demerara, 88. 
 Fort St. Joachim, 124. 
 Fourth Journkv, 285. 
 Fowls, Malay, 282. 
 Fox, 93, 412. 
 Fox, Savannnh, 412 
 Fresh-water Turtle, 482. 
 Frigate F'elican, 158, 170, 411. 
 Frogs, 97, 264,312. 
 Fungus, Trees killed by, 52. 
 
 O. 
 
 G'alera harhara, 400. 
 Gally-nipper, 440. 
 Gannets, 154. 
 Gatewaj', 43. 
 Gauldin, Grey, 420. 
 
 White, 406. 
 Geeso, 203. 
 
 Genipa Americana, 446. 
 Georgetown, 240, 473. 
 Giant Pitta, 304. 
 Gibraltar apes, 9. 
 Goat, 94. 
 Goatsuckers, 97, 197, 262, 264, 381, 
 
 413. 
 Gobe-niouche, 173. 
 Godiva, Lady, 33. 
 Gussjiphim, a/)., 3< 8, 394. 
 
 tricuspidalum, 395. 
 Green-heart, 90, 105, 414. 
 CJrey Gauldin, 420. 
 Grosbeak, .scarlet, 317, 415, 468. 
 Grotto, 07. 
 Guadaloupe, 311. 
 Guana, 93, 207, 415, 42fi. 
 
 Nest of, 267. 
 Guaraons, 237, 477. 
 Guava, Perfumed, 418. 
 
 Water, 418. 
 
 Wild, 180, 417. 
 Gulls, 154, 158, 203. 
 Cyiiaiciuni sacchariiiuvi, 372. 
 
TX1)F.X. 
 
 I, 88. 
 
 1. 
 
 \, 381, 
 
 68. 
 
 H. 
 
 IlAniTATioNS, Indian, 10]. 
 Hackca, 90, 418. 
 Haiani, 418, 491. 
 IJaliautiia leu(:occ2)hahis, 404. 
 Hammock, 418. 
 Hannaquoi, 97, 98, 201, 419. 
 Hawks, 202, 257. 
 Hawtliorn hedrjes, r>7. 
 Hayawa, 90, 100, 111, 238, 42<i. 
 Ilrlias eurypyga, 473. 
 Herons, 40, 420. 
 IJIue, 197, 420. 
 
 Brown, 420. 
 Hia-hia Parrot, 196, 420, 447. 
 Hiawa, .353. 
 Hills, 111. 
 Hitia, 179, 180, 420. 
 Hocco, 201, 420. 
 Hog arrow, 370. 
 
 Hof,', Wild, 92, 101, 104, 420, 417. 
 Holly-trees, 57. 
 Hook, Cayman, 270, 385. 
 Horned Screamer, 202, 4fiS. 
 Hotels in New Ycirk, 305. 
 Honse, Uf,'liness of, 35. 
 Hontou, 97, 113, 185, 420. 
 Howler Monkey, 92, 97, 269, 322, 
 
 421, 441. 
 Huillia, 380. 
 
 Uumirimn forihmdum, 474. 
 Humming-birds, 94, 156, 177, 312, 
 o2l. 
 
 Ara, 178, 423. 
 
 Nests, 321. 
 Huts, Indian, 101. 
 Hyarri, 491. 
 Hyawa, 353. 
 Jlymeiuca courbaril, 433, 461. 
 
 I. 
 
 IiiiBinou, 186, 425. 
 Ibis infascatis, 402. 
 Ibis, liiver, 402. 
 Scarlet, 401. 
 Idea hcptaphylla, 353. 
 Iguana, 93, 267, 415, 425. 
 Iguana Iv.hcrcuJata, 415. 
 Indux, Explanatory, 353. 
 India-rubber, 113, 331, 425. 
 Indians, Acoway, 237, 478. 
 
 Indians, Arowack, 237, 477. 
 
 C'aril), 237, 478. 
 
 Macoushi, 112, 126, 237, 178. 
 
 Waraw, 237, 477. 
 
 L'eligion of, 237. 
 Indian apron, 238. 
 
 Huts, 101. 
 
 Tribes, 2;i7, 477. 
 Insects, 97, 177. 
 
 Instructions to adventurers, 17(!. 
 ■ Ireartia sctigcra, 467. 
 Irish emigrants, 297. 
 Ita palm, 355. 
 ivy tower, 44. 
 
 J. 
 
 JABintT, 1 o, 11 9, 121, 203, 425. 
 Jacamar, i:)2, 315, 428. 
 Jacamcrojis grandis, 428. 
 Jacchus leucogcnys, 466. 
 
 vulgaris, 466. 
 Jaguar, 260, ^64, 26", 272, 429. 
 Jatropha manihot, 381. 
 Jay, 186, 425. 
 Jesuits in rernanibuco, 16ii. 
 Jigger, 385. 
 John Bull in diQiculties, 5t)S. 
 
 JOUJINEY FIRST, 87. 
 SECOND, 154. 
 Tiiiui), 208. 
 
 FOURTH, 285. 
 
 Jumbo, 199, 430. 
 
 K. 
 
 KajMiciii, 469. 
 
 Karabimiti, 178, 821, 423, 431. 
 
 Kessi-kessi, 112, 123, 431. 
 
 King, Savage, 115. 
 
 King Vulture, 94, 431, 485. 
 
 Kingfisher, 192. 
 
 Konamie, 493. 
 
 Koorooboorelli, 462. 
 
 Kurumanni wax, 431. 
 
 K'wanarri, 433. 
 
 I.-AHARRr, 15, 96, 129, 132, 23J, 
 432. 
 Adventure with, 15. 
 

 oil) INDEX. 
 
 l.iibiini, poison of, 235 
 
 Labbn, 92, 104, -.32. 
 
 Lachiiis mufun, 31)8. 
 
 Ladies, American, 292, 299, 30;i. 
 
 La Gabriello, 170. 
 
 La SoudVi^re, 311, 
 
 Lake, 4(5. 
 
 Luke Pavima, 440. 
 
 Lana-tieo, 410. 
 
 Lapo, 432, 
 
 Liinivl hedges, 57. 
 
 Leopnrdus, voncolor, 399. 
 
 onai, 429, 
 
 tiijrinus, 475. 
 Leeches, 17. 
 Lemon-trees, 1G4, 4.33, 
 Letter-wood, 90, 433. 
 Leofric, 33. 
 Liana, 378. 
 liinnets, lt)5. 
 I-izards, 96, 
 
 l,ocust-tree, 90, 433, 461. 
 LoHclwcurpus m>., 491. 
 
 M, 
 
 Maam, 97, 104, 111, 201, 434, 475, 
 
 Macacqne, 10. 
 
 Macaws, 365. 
 
 Macoushi Indians, 112, 126. 
 
 Magot, 10. 
 
 Maliogany, 434, 
 
 Mahiy fowls, 282. 
 
 Iklamonril, 379. 
 
 Manakins, 194, 435. 
 
 Maou, K„ 122. 
 
 Mapire, 467. 
 
 Maranta anincHnacca, 371. 
 
 Marawinaroo, 462, 
 
 Maribuntas, 233, 132. 
 
 Marigalante, 311, 
 
 ^Largays, 474. 
 
 Marmoset, 466. 
 
 Maroudi, 17, 97, 105, 201, 135, 434. 
 
 Maroudi, "White-crested, 117. 
 
 White-headed, 434. 
 Martinico, 313. 
 Massara, 445. 
 Matatoro, 95. 
 Maurida fleumosa, 355, 
 Maximiliana regia, 398, 
 Megasoma Titan, 466. 
 Mtlancrpes erythrocephalus, 489. 
 
 Merles, 883, 
 
 Metallic Cnckoo, 401, 
 
 Mikang, 412. 
 
 Mill-stono and nut-tree, 69, 
 
 Mimvsops Ha/atit, 377, 
 
 Missiles, Monkeys and, 437. 
 
 Moat, 42, 
 
 Moca-nioca, 189, 435, " 
 
 Monkeys, 240, 822, 415. 
 
 Monkey, Misa, 374, 
 
 Ihown, 93, 415, 
 
 Flesh of, 323. 
 
 Howler, 441. 
 
 Itcd, 92, 97, 269, 322, 411. 
 Monkeys and missiles, 437, 
 Monteiro, 166. 
 Mora, 90, 91, 105, 116, 256, 261, 
 
 439. 
 Mora cxwJsa, 439. 
 ^loran, 117. 
 More, Sir Thoma.o, 2. 
 Mos(iiiitos, 98, 117, 440. 
 Mother Carey's chicken, 154, 
 Motmot, 419, 
 Mountain Cabbage, 379. 
 Mouse-traps, 76. 
 Musa paradisaica, 374, 457. 
 
 sapicntum, 373. 
 Muscicapa ruticilla, 466. 
 Muscovy duck, 121, 197, 440. 
 Musk duck, 440. 
 Mycetes ursinus, 421. 
 Mycteria Americana, 426. 
 AJyhlcs pacu, 369, 444, 
 Myristica moschata, 442. 
 A/yrmecophaga jubata, 361, 
 Myrmica sccvimma, 360. 
 
 N. 
 
 Nanoapoa, 203, 441. 
 Native tribes, 477. 
 Ncctandra rodoii, 414 
 Negroeope, 441, 
 Negroes, runaway, 240, 
 New York, 303, 
 
 Climate of, 305. 
 
 Hotels, 305. 
 
 Streets of, 303. 
 Niagara, 291, 294. 
 Nightingale, Virginian, 415. 
 Noctifer, 508, 
 Nomenclature, 77. 
 
INDEX. 
 
 ;i7 
 
 Nondescript, 324, 441. 
 Nut-tree anil niill-stoiie, 69. 
 Nutmeg, 170, 172, 442. 
 
 0. 
 
 Obkah, 431. 
 
 Objects of Wanderings, 87. 
 Ociontophnriis (Juianensis, 404, 
 (Ecodoma ccphalotea, 358. 
 Olou, 90, 442. 
 Oolo, 442. 
 Opossum, 93, 442. 
 Orange-trees, 1(J4, 443. 
 Oreodoxa olcracea, 379. 
 Ortalida mntmot, 419, 
 Ortyx Virqiniana, 463. 
 Ottens, 96," 443. 
 
 Demcrara, 442, 
 Ourah, 133, 444. 
 Owl, 55, 97, 99, 264, 318. 
 Ox killed by Wourali, 142. 
 
 Pacou, 112, 261, 369, 444. 
 Puddles, 445. 
 I'addle-tree, 445. 
 Paint, 445. 
 Paira, 433. 
 Paiworie, 238, 457. 
 Palamcdea cornuta, 468. 
 Palms, 164. 
 Palm, Cocoa-nut, 390. 
 
 Coucourite, 398, 
 Papaw, 101, 237, 445. 
 Paramaribo, 173, 
 Parima, 114, 121, 125, 446. 
 Ibarra Jacana, 486, 
 Parroqnets, 98, 112, 123, 196, 255. 
 Parrots, 98, 196, 240, 266, 
 Parrot, Hia-hia, 196. 
 
 Sun, 447. 
 Partridge, 201 
 Passerinu uidaris, 405. 
 Pataca, 97, 447. 
 Peccary, 92, 447, 
 
 Collared, 449, 
 
 "VVhite-lipped, 449. 
 Pedigree of Waterton, 2. 
 Pce-ay-man, 239, 449, 
 Pegalis, 240, 451. 
 Peggy, 603. 
 Pclamys sarda, 377. 
 
 Pelccanus onocrotaf iin, 401. 
 Pelican, 94, 203, 451. 
 
 Frigate, 158, 170, 411. 
 Penelope cristata, 434. 
 
 pipilc, 434. 
 Pepper, 129, 170, 172, 237, 45 1. 
 
 Black, 454. 
 
 Ked, 454 
 Perfumed Guava, 418. 
 Pernambuco, 158. 
 
 Jesuits in, 160, 
 
 Port of, 159. 
 Pestilence in Spain, 11. 
 Petrel, Stonnv, 155, 456. 
 Phiioton, 170,' 454, 
 Phaeton cetheirus, 479. 
 Pheasants, Wooden, 69, 
 Philadelphia, 301, 
 
 Museum at, 301, 
 Philodryas virldismnvn, 488. 
 Phcenicoptcrus ruber, 407. 
 Piaiman, 449. 
 Piapoco, 181. 
 Piari, 457. 
 Picnic, 68. 
 Pigeon-houses, 73. 
 Pigeons, Mode of feeding, 75. 
 Pigstye, Adventure in, 6. 
 
 Arrangement of, 24. 
 Pike, 47. 
 
 Pine-apples, 237, 456. 
 Piper nigrum, 454. 
 Pi-pi-yo," 97, 249, 456. 
 Pi2)ra sp., 493. 
 Pirai, 113, 134, 454. 
 Pirarara, 11. 122. 
 Pitta, Giant, 364. 
 Pitcairnia sp., 45Q, 
 Piwarri, 238, 456, 
 Plague in Spain, 11. 
 Plantains, 237, 457. 
 Plantations of Stabrock, 175. 
 Platalea levcorodia, 472, 
 Plotus aniiinga, 367. 
 Plover, 94, 203, 457. 
 
 Black-breasted, 457. 
 
 Ring-neck, 457. 
 
 Sandy, 457. 
 
 Spur-winged, 121. 
 Poachers, 69. 
 Podocnemis expansa, 482. 
 Poison-bowl, 75, 
 Poison- wood arrow, 371, 
 
\ 
 
 618 
 
 TXI»KX. 
 
 rol.'.Mt, 03, 4f)0, 
 roiLimdour ('otii|i;ii, 394, 
 J'liinrK (jnnnlis, ;J68. 
 l'o|tc, I'l't'sciitiitioii to, 25. 
 Pdiruiiino, lt3, 457. 
 rotato, Hwcft, 237, 4rtO. 
 I'owiscH, 104, 135, 2(H, 4(50. 
 J'riofclus tniiHiinis, 4l'l>. 
 I'siiUHiH Jlni'iiUilr, 418. 
 
 pmaifcriun, 417. 
 rsii/ihid cir/illmis, 481 
 I'tiriinura Sanbaxhii, 443. 
 I'ucuiiii, 374. 
 Pii/rx pcnefnois, 385. 
 riiri)k'-l)mi''tu(l Cotiii^a, 170. 
 i'uri>lo-tliroiitea Cotinj^a, 170, .'504. 
 riiri.li'-ht'urt, 105, 120, 4G1. 
 ]'liritii(iit ruhni, 468. 
 rijroplioriiii, luininoam, 40G. 
 
 Q. 
 
 QrADRrrKPs, 02. 
 Quail, 201, 4(5:!. 
 (.tiuikc, 120, 403. 
 Quebec, 207. 
 Queyou, 365. 
 Quiver, 134, 4G3. 
 
 R. 
 
 Ratt-s, 203, 
 Raiment and diet, 210, 
 Ra]>pu arrow, 371. 
 Rattlesnake, 0.5, iS7, 465, 
 
 Adventure with, 167. 
 Rapids of Essequibo, 110. 
 Rat, Hanoverian, 8, 
 Ray, Sting, 266, 473, 
 Recluse, AVhito, 114, 
 Red-Headed Finch, 164, 406. 
 Red-Headed Woodpecker, 213, 480. 
 Red Howler, 421, 441. 
 Red Monkey, 92, 97, 260, 322. 
 Red Pepper, 454. 
 Red Start, 311, 466 
 Reed Arrow, 372. 
 Reformation, the Watertons in, 4. 
 L'emarks on First Journey, 148. 
 llhamphastos Ariel, 477. 
 
 eri/t/uwhynchiis, 477. 
 
 fAK'o, 468. 
 
 vitcllinus, 477. 
 Rhinoceros beetle, 312, 466. 
 Rice bird, 188. 
 
 Rice Trouiiial, 481. 
 Jiiciiiiis fiimiininis, 383, 
 lliiig-neok I'lover, 457. 
 Ifiver li)is, 4o2. 
 Rome, adventure near, 28. 
 Roseau, 31 1. 
 Roucou, 238, 466. 
 liuuaway ncj,'roes, 240. 
 lliipkola aiirunlidy 302. 
 
 S. 
 SAn.', 80. 
 
 Si(ccli<iriniini officiiinruni, 473. 
 Saeawinki, 23, 466. 
 Snge, red, 178. 
 SaiuteH, 311. 
 Saki, Bearded, 374. 
 Salempentnj 03, 4(i7. 
 Sumourah, 133, 467, 
 Sandllies, 08, 117, 467. 
 Sandpiiier, 04, 203. 
 Sandy I'lover, 457. 
 Sangre do Buey, 165, 468, 
 Saratoga, 208. 
 Sarcorkamphus pajm, 485. 
 Saiiba, 360. 
 Savage King, 115. 
 Savanna, 116. 
 Savannah Fox, 412. 
 Sawari, 105, 226, 468. 
 Scarlet Ara, 111, 2()6, 365. 
 
 Curlews, 203, 401. 
 
 Grosbeak, 317, 415, 468. 
 
 Ibis, 401. 
 
 Tanager, 468. 
 Scenery, 261, 287. 
 Sclinclla, sp. 381. 
 
 hrachystncliya, 378. 
 School life at Stonyhurst, 7. 
 
 Tudhoe, 6. 
 Scientific names. Hatred of, 77. 
 Scirou, 181, 468, 477. 
 Screamer, Horned, 202, 468. 
 Skcond JOUIINEY, 154. 
 Scrrasalmo 2nra!ja, 464. 
 Sharks, 158, 168. 
 Shiniara-sherie, 369. 
 Shrikes, 202. 
 
 tSiilerodendron triflorum, 418. 
 Silk-grass, 133, 134, 237, 460. 
 Siloabali, 105, 179, ISO, 470. 
 
 Bastard, 179, 180, 186. 
 Simiri, 433, 461. 
 
TXDF.X. 
 
 510 
 
 Siiiiidiii /irrtiiuu, 407. 
 
 Si|iar(jiiiii, 112. 
 
 Sir Joseph Hiiiks, 214. 
 
 Sliivfiy, 175, ;J14. 
 
 Sloth, !);J, 17/5, -nr,, 231, 31 I. 
 
 Iliihitsof, 21.'5. 
 
 Jvillud hy Woiira];, 1 U. 
 
 Two-toed, 'J-JO, 472. 
 
 Thive-toed, .'{lO, 470. 
 Snakes, 9.'!, 170, 231, '2 la. 
 Hiiipcs, 2U3. 
 Soil, !>2. 
 Sounds, Strange, 104. 
 
 Kxidanutiou ol", 14!). 
 Sounri, 408. 
 Spain, Stay in, 9. 
 Sjtikes for arrows, 137. 
 Spoonbill, 94, 17(», 2(»2, 472. 
 Spur-wing, 472, 485. 
 Spur-winged plover, 121. 
 Stabroek, 173, 472. 
 Starling towoi-, 65. 
 Sting-ray, 200, 473. 
 St. Joachim, Fort, 122, 124. 
 St. John's, Antigua, 31u. 
 St. Jjueia, 313. 
 Storm, Mi). 
 
 Stormy Tetrel, 154, 450. 
 Strange sounds, 104. 
 
 Exi)lanation of, 1 49. 
 Strepsifus interprcs, 457. 
 Slrychnos toxifem,491. 
 St. Thomas's Tower, 150. 
 Stalling, Bird, 214, 3.35, 4!>7. 
 
 Materials for, 330. 
 Sugar Cane, 473. 
 Suu-bird, 260, 313, 473. 
 
 parrot, 447. 
 Sioietonia mahagonl, 434. 
 
 T. 
 
 Tacatoit, R., 122. 
 
 Tacouba, 433. 
 
 Tamarin, White-whiskered, 406. 
 
 Tanager, Scarlet, 468. 
 
 Tangara, 165. 
 
 Tantalus localator, 441. 
 
 ruber, 401. 
 Tapir, 92, 101, 123, 474. 
 Tapirus tr.rrcstvis, 474. 
 Tiiracai, 482. 
 Turarama, 428. 
 
 Tauronira, li).5, 10', 174. 
 Taxidernial jokes, 5iJ8. 
 Taxidkumv, 335. 
 
 Ndtks to, 4U5„ 
 Tayra, 400. 
 Ti'ius Ti'ifii-.rin^ 4G/. 
 Termites', 300. 
 Tertian ague, 150. 
 Trfntf/oiiii/t/ri Hs lni.ii^, 300. 
 TkalaHsiilrdiiia ii,'/i/fjirn, 450. 
 Tiiini) JoiriiNKV, 2U8. 
 Thniut.i: sj>, 47!). 
 Thn h, !)!>, 165. 
 Thi/niiits paeijlcus, 357. 
 Ticks, 237. 
 Ticond.'rag", 298. 
 Tiger, !»2, 94, 177, 203, 474. 
 
 IJird, 195, -U. 
 
 Cats, 92, 475. , 
 
 Tigri-fowlo, 473. 
 Tijrisoma, ilnr.illenne, 475. 
 
 tiijriiinm, i.'". 
 Tln-nnotiH rJi'if,i,ii.i, 47.). 
 Tiuamous, it/, 104, 111, 2('J, 475. 
 
 (ireat, 318. 
 
 Small, 319. 
 Tirana, 200, 475, 
 Titniousi! and Woodiiecker. 50. 
 J oeo, 4(5. 
 Topaz, Fiery, 123. 
 Topaxti jiel/a, 423. 
 
 pyra, 423. 
 Tortoise, Land, 230, 203, 475. 
 TortrU ,Si-i/fak, 394. 
 Toucan, [>6, 97, 121, 150, 181, 475. 
 
 Hill of, 182. 
 Toncauet, 477. 
 i'.Mver, Ivy, 44. 
 
 Starling, 05. 
 Traps, J\louse, 70. 
 Tribes, Indian, 237, 477. 
 Troely, 99, 233, 478. 
 Troglodytes ci(do)i, 493, 
 Trognn mclano/derics, 370. 
 Tropic bird, 158, 479. 
 Troupials, 123, 104, 193, 481. 
 Troy, 299. 
 
 Trumpeter, 202, 481, 
 Trygiiu vti.sttnaca, 473. 
 Tschiko, 385. 
 Turkey Buzzard, 485. 
 Turtle, 266, 482. 
 
 Nest of, 206. 
 
521") 
 
 INDEX. 
 
 U. 
 
 Una-waya-adoucati, 316. 
 Uropsophus durissus, VZ. 
 
 V. 
 
 Vampire, 95, 97, 226, 229, 319, 482. 
 
 Teeth of, 321. 
 Vampirus spectrum, 482. 
 Vanilla, 232, 483. 
 Vanilla planifoliay 483. 
 Vines, 91. 
 A'iiie, Wourali, 129. 
 Virginian nightingale, 415. 
 Voyage to Demerara, 13. 
 Viilpes cancrivora, 412. 
 Vulture, Aura, 257, 484. 
 
 Common, 94, 119, 202, 25(^. 
 
 King, 94, 2.56, 485. 
 
 Beak of, 256. 
 
 W. 
 
 Wallaha, 105, 106, 319, 466. 
 
 Bird, 401. 
 AVallabaha, 180, 486. 
 Wall round park, 39. 
 "Walton Hall, ugliness of, 35. 
 "Wanderings, objects of, 87. 
 "Waracaba, 105, 'l35, 202, 481, 486. 
 Waraws, 237, 456, 477. 
 "Wasps, 233. 
 AVater Guava, 418. 
 "Water-hen, Spur-winged, 197, 486. 
 Waterton at home, 35. 
 
 Death of, 81. 
 
 Eccentricity, 23. 
 
 Funeral of, 83. 
 
 Love of trees, 49. 
 
 ^larriageof, 32. 
 
 Mode of life, 36. 
 
 Tree-climbing, 62. 
 "Watermamma, 115, 487. 
 "W^ax, 133, 134. 
 
 Kurumanni, 431. 
 
 "W^eeds, 92. 
 
 "Whip-poor- Will," 99, 111, 199, 
 
 488. 
 Whipsnake, 96, 488. 
 White Gauldin, 406. 
 White-headed Eagle, 302. 404. 
 White-lipped Peccary, 449. 
 White recluse, 114. 
 "Who-are-you," 199, 262, 489. 
 Wild Beasts, 234. 
 Fig, 406. 
 
 Hogs, 138, 245, 420, 447. 
 Man, 160, 
 "Willy-come-go," 99, 199, 262, 
 
 489. 
 Wire-cash, 369. 
 Wiriebiserie, 402. 
 Wood ants, 93. 
 Wooden pheasants, 59. 
 Woodpeckers, 190. 
 
 Ked-headed, 105, 213, 489. 
 and Titmice, 50. 
 Woodskins, 120, 461, 489. 
 "Work-away," 199, 4S9. 
 Wourali, 101, 112, 113, 122, 126, 
 140, 489. 
 Antidotes, 143. 
 Experiment on Ai, 141. 
 Ass, 151. 
 Dog, 102. 
 Ox, 142. 
 Sloth, 141. 
 Preparation, of, 129. 
 Strength of, 102. 
 Vine, 129, 491. 
 Wouralia, 151. 
 AVren, 99, 111, 165, 494. 
 
 y. 
 
 YABAHOtT, 130, 199. 
 Yams, 237, 494. 
 Yarari, 445. 
 Yawaraciris, 195, 494. 
 Yellow-fever at Demerara, 209. 
 Yew-tree, 63. 
 
 LoNbO.V : It. CLAY, BOWS, A.ND TAVLOK, BUF.AD STREF.T UILL. 
 
 
 
 1 
 
)9, 111, 19P, 
 
 302. 404. 
 449. 
 
 262, 489. 
 
 20, 447. 
 >, 1&9, 26-J, 
 
 213, 489. 
 
 t89. 
 
 i9. 
 
 i, 122, 12rt, 
 
 , 141. 
 151. 
 , 102. 
 142. 
 1, 141. 
 
 ■a, 209.