IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 1.0 
 
 I.I 
 
 '"' IIIIM 
 - iillM 
 
 m 
 
 f ■< c 
 
 IIIIM 
 \\= 
 IM 
 
 1.8 
 
 
 1.25 
 
 1.4 
 
 1.6 
 
 
 ^ 6" 
 
 
 ► 
 
 v: 
 
 <9 
 
 /}. 
 
 '^A 
 
 e. 
 
 c-l 
 
 v> a 
 
 (Pi 
 
 e- 
 
 -'*. 
 
 O 
 
 A^/ 
 
 A 
 
 Vf 
 
 /A 
 
 Photograpliic 
 
 Sciences 
 Corporation 
 
 ■12 WEST MAIN STREET 
 
 WEBSTER, NY 14580 
 
 716; 872-4503 
 
CIHM/ICMH 
 
 Microfiche 
 
 Series. 
 
 CIHIVI/ICMH 
 Collection de 
 microfiches. 
 
 Canadian Institute for Historical Microreproductions / Institut canadien de microreproductions historiques 
 
Technical and Bibliographic Notes/Notes techniques et bibliographiques 
 
 The Institute has attempted to obtain the best 
 original copy available for filming. Features of this 
 copy which may be bibliographically unique, 
 which may alter any of the images in the 
 reproduction, or which may significantly change 
 the usual method of filming, are checked below. 
 
 L'Institut a microfilm^ le meilleur exemplaire 
 qu'il lui a 6t6 possible de se procurer. Les details 
 de cet exemplaire qui sont peut-dtre uniques du 
 point de vue bibliographique, qui peuvent modifier 
 une image reproduite, ou qui peuvent exiger une 
 modification dans la mdthode normale de filmage 
 sont indiquds ci-dessoufi. 
 
 D 
 
 Coloured covers/ 
 Couverture de couleur 
 
 □ Coloured pages/ 
 Pages de couleur 
 
 □ Covers damaged/ 
 Couverture endommagee 
 
 D 
 
 Pages damaged/ 
 Pages endommag6es 
 
 □ Covers restored and/or laminated/ 
 Couverture restaurde ct/ou pellicul^e 
 
 □ Pages restored and/or laminated/ 
 Pages restaur^es et/ou pelliculdes 
 
 D 
 
 Cover title missing/ 
 
 Le titre de couverture manque 
 
 
 
 Pages discoloured', stained or foxed/ 
 Pages d6colorees, tachet^es ou piqu^es 
 
 n 
 
 Coloured maps/ 
 
 Cartes g^ographiques en couleur 
 
 Coloured ink (i.e. other than blue or black)/ 
 Encre de couleur (i.e. autre que bleue ou noire) 
 
 □Pages detached/ 
 Pages detachees 
 
 Showthrough/ 
 Transparence 
 
 ^ 
 
 D 
 
 Coloured plates and/or illustrations/ 
 Planches et/ou illustrations en couleur 
 
 □ Quality of print varies/ 
 Qualite in^gale de I'impression 
 
 n 
 
 Bound with other material/ 
 Relie avec d'autres documents 
 
 □ Includes supplementary material/ 
 Comprend du materiel supplementaire 
 
 D 
 
 D 
 
 Tight binding may cause shadows or distortion 
 along interior margin/ 
 
 La reliure serree peut causer de I'ombre ou de la 
 distortion le long de la marge int^rieure 
 
 Blank leaves added during restoration may 
 appear within the text. Whenever possible, these 
 have been omitted from filming/ 
 II se peut que certaines pages blanches ajout^es 
 lors dune restauration apparaissent dans le texte, 
 mais, lorsque cela 6tait possible, ces pages n'ont 
 pas et^ film^es. 
 
 n 
 
 Only edition available/ 
 Seule Edition disponible 
 
 Pages wholly or partially obscured by errata 
 slips, tissues, etc., have been refilmed to 
 ensure the best possible image/ 
 Les pages totalement ou partiellement 
 obscurcies par un feuillet d'errata, une pelure, 
 etc., ont 6t6 film^es it nouveau de facon d 
 obtenir la meilleure image possible. 
 
 n 
 
 Additional comments:/ 
 Commentaires suppl^mentaires; 
 
 This item is filmed at the reduction ratio checked below/ 
 
 Ce document est filrn^ au taux de reduction indiqu6 ci-dessous. 
 
 10X 
 
 
 
 
 14X 
 
 
 
 
 18X 
 
 
 
 
 22X 
 
 
 
 
 26X 
 
 
 
 
 30X 
 
 
 
 
 
 
 
 
 
 
 
 
 J 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 12X 
 
 16X 
 
 20X 
 
 24k 
 
 28X 
 
 32X 
 
sils 
 
 du 
 
 difier 
 
 jne 
 
 )age 
 
 The copy filmed here has been reproduced thanks 
 to the generosity of: 
 
 National Library of Canada 
 
 The images appearing here are the best quality 
 possible considering the condition and legibility 
 of the original copy and in keeping with the 
 filming contract specifications. 
 
 Original copies in printed paper covers are fiimed 
 beginning with the front cover and ending on 
 the last page with a printed or illustrated impres- 
 sion, or the back cover when appropriate. All 
 other original copies are filmed beginning on the 
 first page with a printed or illustrated impres- 
 sion, and ending on the last page with a printed 
 or illustrated impression. 
 
 L'exemplaire filmd fut reproduit grSce d la 
 g6n6rosit6 de: 
 
 Bibiioth^que Rationale du Canada 
 
 Las images suivantes ont 6x6 reproduites avec le 
 plus grand soin, compte tenu de la condition et 
 de la nettet6 de l'exemplaire film6, et en 
 conformity avec les conditions du contrat de 
 filmage. 
 
 Les exemplaires originaux dont la couverture en 
 papier est imprim6e sont filmds en commenpant 
 par le premier plat et en terminant soit par la 
 dernidre page qui comporte une empreinte 
 d'impression ou d'illustration, soit par le second 
 plat, selon le cas. Tous les autres exemplaires 
 originaux sont filmds en commenqant par la 
 premidre page qui comporte une empreinte 
 d'impression ou d'illustration et en terminant par 
 la dernidre page qui comporte une telle 
 empreinte. 
 
 The last recorded frame on each microfiche 
 shall contain the symbol — ♦- (meaning "CON- 
 TINUED "), or the symbol V (meaning "END"), 
 whichever applies. 
 
 Maps, plates, charts, etc., may be filmed at 
 different reduction ratios. Those too large to be 
 entirely included in one exposure are 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; 
 
 Un des symboles suivants apparaitra sur la 
 dernidre image de chaque microfiche, seiun le 
 cas: le symbole — ♦- signifie "A SUIVRE", le 
 symbole V signifie "FIN". 
 
 Les cartes, planches, tableaux, etc., peuvent etre 
 film^s d des taux de reduction diffdrents. 
 Lorsque le document est trop grand pour etre 
 reproduit en un seul clich6, il est film6 d partir 
 de Tangle sup6rieur gauche, de gauche d droite, 
 et de haut en bas, en prenant le nombre 
 d'images ndcessaire. Les diagrammes suivants 
 illustrent la mdthode. 
 
 rata 
 o 
 
 >elure, 
 
 1 
 
 2 
 
 3 
 
 32X 
 
 1 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
LABRADOR 
 
 A SKETCH OF 
 
 ITS PEOPLES, ITS INDUSTRIES AND ITS 
 NATURAL HISTORY, 
 
 13 Y 
 
 WINFRID ALDEN STEARNS. 
 
 liOSTON : 
 
 ^^EE AND SIIEl.AKD, 47 FKANKLIN STREET 
 
 NKW YORK. CIIARLK8 T. DILLIXGIIAM. 
 
 1884. 
 
rs39s 
 U 
 
 ST^fifi!^£ bO.fi. 
 
 2,95526 
 
 Entered according to Act of Congress, in the year 1883, 
 
 BY W. A. STEARNS, 
 
 in the Office of the Librarian of Congress, at Washington. 
 
 J 
 
 Printed at Office of Salem Press, 
 Salem, Mass. 
 
PREFACE. 
 
 83. 
 ngton. 
 
 In presenting to the public the following journal sketches of the 
 country of which they treat, the thought that I am writing of a 
 region so new and so little known, though so near to us, together 
 with the pleasure which I have experienced both in my travels and 
 in the preparation of this account of them, will furnish a sufficient 
 excuse for the undertaking. Although so little has been known or 
 written about Labrador, yet it is a region not a thousand miles 
 away from us, and one which bears a most important relation to 
 the fishing interests of this continent. 
 
 The knowledge of this region, which is within reach of the pub- 
 lic, is to be found only in the pages of an old-fashioned document, 
 of two volumes, now out of print, and almost unknown save to 
 large libraries, entitled "Cartwright's Journal ;" in Hind's "Labrador 
 Peninsula," 2 vols., and in occasional articles in some magazine or 
 newspaper If I have added some new and interesting matter to 
 our present knowledge of this subject, my labors will not have 
 been in vain. 
 
 My first trip to Labrador was during the summer of 1875, when 
 I collected largely in all the branches of Natural History, and es- 
 pecially in Botany, finding, with the assistance of the Rev. Mr. 
 Butler, the then missionary to this region, many new and rare 
 plants. This number has since liccn enlarged and a catalogue of 
 them, as of collections in several other branches of Natural History, 
 published in the sixth volume of the Report of the National 
 Museum, at Washington. In September, 1 880, I again visited the 
 
 (iii) 
 
iv 
 
 rREKACE. 
 
 coast and remained studying the region and collecting until Sep- 
 tember, 1 88 1. During this time I had abundant opportunity for 
 investigation. In the summer of 1882 I made my third trip, 
 starting from lioston with a company of about a dozen fellow- 
 voyagers, and i)roceeded nearly to Harrison Inlet. 
 
 I would tender most cordial thanks to several of the officers in 
 charge of various posts of the Hudson's Bay Company where I 
 visited, who gave me much help in my investigations : as also the 
 courteous Magistrate at Bonne Esperance, Mr. \Vm. H. Whitcley, 
 with whom I spent a great part of my time while on the coast. 
 
 Hoping that there may be found enough of value and interest 
 to counterbalance the many too obvious defects, I submit these 
 pages to those who may be interested in such researches. 
 
 Respectfully, 
 
 WiNFRiD A. Stearns. 
 Amherst, July 14, 1884. 
 
CON T K N T S . 
 
 Preface 
 
 Introduction ..'''-■ 
 
 CHAPTER I. 
 
 physk:al geography of kegign 
 
 Surve, or U,..K. - .,_„,r, pH,.cs - .H,.e. ,eo,.p,, 
 
 CHAPTER H. 
 
 T, . , , , STARTING ON THE JOURNEY. 
 
 Trip o Labrador - Arrival at Montreal - Arrival ., n„ , 
 
 description of the city . 'vrrivui at Quel,cc and 
 
 CHAPTER HI. 
 
 •-^'■^^B'^-C TO GREEN ISLAND. 
 
 Ourstay.n Quebec -Starting for Berthier-Berthier Offr r , 
 rador - Bunking in _ Island of Orlean tT "i ' ^'''■ 
 
 -^-Sunday-OtHer islands J^S:-::;^-^^^^^^ 
 
 CHAPTER IV, 
 
 GREEN ISLAND To TONNH KSPEKANCE 
 <1» .logs - Searching f,„ .,Hf, J,, '"*"'' ' ""■'■'^ ^ '-'-- 
 
 IA(,E 
 
 iii 
 1-7 
 
 9-26 
 
 27-38 
 
 39-51 
 
 CHAPTER V. 
 
 52-66 
 
 nONNE ESPERANCE. 
 
 '^'I'nancc — hs, uniiaux r ver andish.1,1 r i 
 En.ori„„ «.,„„e Esporancc Harho ^ V J,*:;'; ,"; '^';'-' " 
 
 67- 
 
 75 
 
VI 
 
 CONTENTS. 
 
 I'AGE 
 
 CHAPTER VI. 
 
 NATURAL HISTORY : TOPOGRAPHY OF COUNTRY. 
 
 Raspberries — Weather — liudsonian chickadee and other bir<ls — 
 Black fly — Topography of country — Old Fort bay, physical 
 features and surroundings — Superstitions concerning the raven 7G-96 
 
 CHAl^ER VII. 
 
 OLD FORT . INVESTIGATIONS. 
 
 Indian tents — New fields for research — Visit to the Indians — 
 Seals' flesh — Dogskin boots — Cattle food in hard winters — 
 Coptis trifolia — Spruce partridge — Inland — Hypothesis of 
 Aurora — Little auk — Signs of a wreck — Ascent of western 
 arm of the bay — Wreck of the Edward Cardwell — Picking up 
 lumber — First snowstorm of the winter .... 
 
 97-118 
 
 CHAFFER VIII. 
 
 A LABRADOR HOME. 
 
 A Labrador home — Houses — Where erected — Stage — Shop — 
 Stable — The house — Papering — Family — Occupation of its 
 
 members — Out-of-door life 
 
 - 119-131 
 
 t 
 
 CHAPTER IX. 
 
 A GALE : KOMATIK, RACKETS, ETC. 
 
 Dinner off fresh meat — Credit and shiftlessness — A Labrador snow- 
 storm — Wind — Preparing for storm — .Storming hard — Fire- 
 wood — Storm increases — Sleepless night — Another day of it 
 — A grand sight — Violence of wind and wave — Destruction 
 of stage — Calmer weather — Beautiful ice scene — End of 
 storm — Thanksgiving day — Komatiks and rackets 
 
 132-151 
 
 CHAPl^ER X. 
 
 WINTER SCENES AND OCCUPATION OF PEOPLE. 
 
 Trip up the river to the mission — Ice pictures — Bad walking — 
 To the old Fort — New scenes and bad walking — Pleasant 
 Sunday — The return — Journal — A komatik ride — Christmas 
 gathering — Wood cutting — Work for the evenings — Making 
 sealskin boots, mittens, and other needful and fancy work 
 
 152-171 
 
CONTKNTS. 
 
 vii 
 
 I'AGE 
 
 CHAPTER XI. 
 
 WINTER SCENES AND AMUSEMENTS. 
 
 New Year's day — How to walk on rackets — "Fish, dogs, and 
 seal," the general topics of conversation — Obtaining skeletons 
 
 — Larch poultices — "Small talk" — Low temperature — Deer 
 stories — Trapping — Indians — Up the river — At the mission 
 
 — Harnessing the puppies — A racket walk - - - 172-190 
 
 CHAPTER XII. 
 
 OLD FORT TO L'aNSE AMOUR. 
 
 80° in winter — Trip eastward — Starting — Esquimaux river and 
 island — Salmon bay — Bradore bay — Caribou island — Five 
 leagues — Middle bay — Belles Amour — Over Bradore hills — 
 L'Anse Dunes — Blanc Sablon — L'Anse Coteau — L'Anse 
 Clair — Forteau — Amour 191-213 
 
 CHAPTER Xni. 
 
 A TRAMPING EXPEDITION. 
 
 Canadian porcupine — Picking fall berries in spring — Carrying 
 wood to summer quarters — Anticipating Fourth of July — Sum- 
 mer quarters in winter — Capsized — Fox hunt on rackets — A 
 mile of soft snow without rackets 214-220 
 
 CHAPl^ER XIV. 
 
 MOVING OUT. 
 
 Preparing the summer house to live in — Moving out — A spring 
 rescue — Seals on ice — Larks — A home scene — Spring duck 
 shooting — Repairing the boats — Visit to the Indians — Indian 
 canoes — Netting nets — Labrador mail — Natural scenery of 
 Labrador — Repairing canoes — Visit to Esquimaux graves — 
 
 Ornithological notes 
 
 221-238 
 
 CHAPTER XV. 
 
 FROM BLANC SABLON TO MINGAN. 
 
 Blanc Sablon again — Northern limits of the bittern — Return along 
 the coast of Natashquan — Spring scene in Red bay — Other 
 places — St. Mary islands — Cormoi-ants — At Natashquan — 
 Rambles about the place — Appearance of the birds — The 
 dark day — Arrival at Mingan 239-253 
 
VUl 
 
 CONTENTS. 
 
 CHAPTER XVI. 
 
 MINGAN TO OLD FORT ISLAND .AGAIN. 
 Min^an and surroundings- Hudson's Bay Companies' In-ildings - 
 Mingan river - Indians of this region, their hal.its, religion, etc 
 -Montagnais and Nascopies-The Indian trade at the various 
 places along the north shore- Uomaiue or Olotnanosheebo - 
 Natashquan again - French steamer and salmon freezing - 
 Jewelry peddler - Agwanus, Nahasippi - Terns and gulls-Cod 
 (.sh " schcn.ling "- Es(iuimaux Point - Indian names, etc. : St 
 (.enevieve island, Watchceshoo, Manicouagan, Sa.ldle Hill. 
 Mt. St. John, Washatnagunashka hay, Mushkoniatawee, I'ash- 
 asheeboo, Peashtebai- Shooting at the Fox islands - Mutton 
 I'ay, Great Mecattina islands -Uhl Fort island agam at last 
 
 i'ACB 
 
 254-272 
 
 CHAPTl'lK XVIL 
 
 HOMEWARD MOUND. 
 Aflairs at Old Fort Island -The hshing season - Thun.ler storm 
 -Arrival of vessel which is to take us home - Our trip in her 
 to L'Anse Loup— Scenes at intervening places - Off for home 
 
 — Double reefed fore and mam sails - Island of Anticosti 
 
 — A hurricane— (Quebec and home - . ,,, ,- 
 
 CHAPTER XVni. 
 
 THIRD TRI1> TO LAHRADOR. 
 Third voyage : summer of '82 - Puffin-shooting - Dredging - had 
 weather -Main boom breaks -Chateau and Temple bays - 
 Places of interest - Mines and minerals - Aurora and phospho- 
 rescence - Icebergs - F<;x harbor - Battle island - Indians 
 and Esquimaux - Indian voc- ' .ary - Square island - Dead 
 island - A water garden - Triangle harbor - Homeward 
 bound -Notes on Dutch aiul Es.juimaux settlements. - 280-295 
 
I AGE 
 
 etc. 
 
 lOUS 
 3 — 
 s "*" 
 
 :oci 
 
 St. 
 liU, 
 ish- 
 ton 
 
 INTRODUCTION. 
 
 254-272 
 
 rm 
 lor 
 nir 
 )sti 
 - 273-279 
 
 ad 
 
 io- 
 ns 
 ad 
 rd 
 
 280-295 
 
 LAI'.R.\DOR : ITS DISCOVERY AND ITS LOCATION. 
 
 The coast of Lal)raclor is well known to you all, as it figures on 
 your maps and on your charts. There are probably ^-cry few who 
 do not recollect this little oblong plateau as it appears thus jutdng 
 eastsvard into the Straits of Uellc Isle fron. .n ahnost desolate, and 
 on your maps plainly colored, portion of inhabitable northern North 
 America. If there should be those t whom this ^.cation is as 
 J yet unfamiliar, let them refer to their geogranhy, .nd then, follow- 
 
 J mn Che river St. Lawrence as it fiows to the Oulf, and the Ciulf as 
 
 f It flows through the Straits of Belle Islo, tliey will readily find, just 
 
 before entering the narrowest part of the Stroiis, the coast of New- 
 foundland on the right and that of Labrador on the left of this Dor- 
 tion of the Gulf. ■• 
 
 A section, and that the most easterly, of the Canat'ian Province 
 of Quebec, is usually included with Labrador proper in the term 
 generally applied, of " Labrador." As a part of this Province it has 
 Its mails, though irregular at best ; while its seat of government is 
 HI Quebec. 
 
 Of Labrador a certain writer has said, perhaps as tnily as the 
 times (by which 1 mean the explorations of science and survey) 
 will permit, that -it is an immense peninsula extending over an 
 area of four hundred and fifty thousand superficial miles, and 
 bounded by the Atlantic, the Gulf of St. Lawrence, and Hudson's 
 Bay. 
 
 Of this part of the coast there is a considerable division of 
 opinion, as in fact of other neighboring parts, as to the people who 
 hrst discovered it, as well as c/ the origin of the name. 
 
 (1) 
 
INTRODUCTION. 
 
 It is not the place here to enter into a discussion of this qties- 
 tion nor yet to call up all the points in the arguments in favor of 
 or against any particular hypothesis. The fact that as yet we know 
 little enough of these interesting puints of inquiry, which, while they 
 answer for inJividual effort and research, will hardly be necessary 
 in such a place as this, will therefore limit us to what we do know. 
 
 We do know of the contention among both French and English 
 as to the discovery of Lal^rador by Sebastian Cabot in 1496, and 
 its exi)loration soon after by the Portuguese Corterell, who is said 
 to have named it. We do know that Mr. Samuel Robertson, who 
 has given this matter his most careful attention and who has per- 
 haps .searched more thoroughly than any other man in the country 
 for the real facts bearing upon this subject, in his excellent article 
 read before the Literary and Historical Society of Quebec, dated 
 January 16, 1841, gives us the summing up of what are the results 
 of his labors. From them we learn that "the universal tradition 
 of the coast," and which his in(iuiries seem to verify and establish, 
 is "that one Labrador, a Basque whaler, from the kingdom of 
 Navarre in Spain, did penetrate through the Straits of Belle Isle, as 
 far as Labrador Bay, sometime about the middle of the fifteenth 
 century ; and, eventually, the whole coast took the name from that 
 bay and harbor." 
 
 There is very little doubt, as far as circumstantial evidence goes, 
 that the coast here, as in the neighboring places, was visited by the 
 Norsemen in the tenth century ; but they left no signs of coloniza- 
 tion by which we can prove it. 
 
 There were some remains of buildings, also instruments discov- 
 ered, that were supposed to be of later and Esquimaux origin, but 
 these, Mr. Robertson thinks, can be proven of Basque origin. In 
 all probability, therefore, Labrador was discovered by Basque fisher- 
 men and whalers before the discovery of Cabot, and before Chris- 
 topher Columbus discovered America. We know that the French 
 carried on fisheries along the coast of Newfoundland earlier even 
 than the year 1500. In 1532 Jacques Cartier, 7vith Bretons for 
 pilots, visited this coast, and as early as 1506 a chart of the Gulf 
 
INTRODUCTION. 
 
 ues- 
 )r of 
 :now 
 they 
 ssary 
 :now. 
 iglish 
 I, and 
 s said 
 I, who 
 ,s per- 
 auntry 
 article 
 dated 
 results 
 adition 
 tablish, 
 iom of 
 Isle, as 
 fteenth 
 m that 
 
 fe goes, 
 
 by the 
 
 )loniza- 
 
 had been published. Besides all this the Welsh and Irish, as also 
 the Icelanders, put in the claims of being early visitors. Thus the 
 difficulties of origin increase rather than diminish. 
 
 The first established colony in Labrador was at "Brest," now 
 Bradore. It was founded about 1508, was a trading post, con- 
 tained the residence of the Governor and other officers, and, 
 it is said, contained "200 houses and 1000 inhabitants in winter, 
 which latter were trebled in the summer." Thus you will see 
 that very early this was an important trading post, though it 
 gradually lost its importance, and soon dwindled to a few houses 
 whose inhabitants were settlers, who took what they could find 
 as they found it, and were without law and government, as one 
 might truly say. 
 
 You will understand that this settlement at Bradore was not a 
 permanent one, that is, dependent upon other settlements on the 
 coast ; but was one, so to speak, complete in itself for the purpose 
 of carrying on fisheries at that place, while the greater portion of 
 the coast had neither been much setded nor much explored ; conse- 
 quently, at the time called the Conquest, when the Es(]uimaux were 
 nearly exterminated from these districts, and when the owning foreign 
 monarch began to cut up this almost thriving colony into special 
 grants to his favorites, its prosperity began to decline . Though at first 
 the fish, whale, and fowl were everywhere abundant, a slow but per- 
 ceptible decrease of these productions tended also to dishearten 
 the colony ; and a natural unfavorableness of the coast, being then, 
 as it now is, a mass of granite rock, as also the severity of the cli- 
 mate, combined with a dread of extermination, to scatter them 
 completely. You will now also understand why, if Labrador was 
 once in a way to become a popular fishing station, it broke up 
 rather than increased in its settlements and thus possible fu- 
 ture growth. Having thus broken up, you will see why it has re- 
 mained waiting for the energy and determination of Americans to 
 combine with the sturdy hold on of the English, who have now 
 gained i)ossession of the coast and added it to their already im- 
 portant Canadian and Newfoundland Colonies, to open it again to 
 
INTRODUCTION. 
 
 III 
 
 the world as a fishing post, when centuries shall have restocked 
 its waters. 
 
 As Labrador looks upon the charts, one would easily imagine 
 that it was a vast expanse of lowland, and one almost plain penin- 
 sula extending into the sea as a continuation and part of the Brit- 
 ish Provinces, and auxiliary, perhaps, to the territory of the Hud- 
 son's Bay Company. In point of fact this latter case is partly true 
 and partly false. It is indeed a part of the British possessions, and 
 it is also a part of the Hudson's Bay Company's territory ; at least 
 there are several trading posts of that Company placed occasion- 
 ally along its shores, and the Company own considerable land lo- 
 cated about and around each such post ; yet it forms but a small 
 part of that Company's productive ground, which is much farther 
 north, and extends a considerable way into the interior of the 
 country, thus approaching the great bay which bears the name of 
 that hardy, early pioneer and explorer who discovered it, as also 
 the river of that same name which here takes its rise. At one time 
 Labrador was a small part of these the best fur regions of the 
 world. Then many trading posts were established, of which the 
 most important, perhaps, at least of that portion of the coast of 
 which I shall speak more at length presently, was at St. Augustine, 
 near the St. Augustine river, and only a few miles — so to speak — 
 from the principal settlements in this section about Esquimaux 
 river, and Bonne Esperance the seat of local government. It was 
 only at a late day that this post was abandoned as a trading station, 
 and, unless recently resupplied it is no more a permanent authorized 
 station of the Company. Though the post here has been formed but 
 a few years, new ones are being so constantly made — or rather aban- 
 doned ones resupplied — and old ones broken up, that it is impos- 
 sible to keep an exact and satisfactory account of them all. 
 
 The portions of Labrador which I visited are contained within 
 the easternmost extremity of the Province of Quebec, and the 
 westernmost part of Labrador proper, comprising an included 
 distance of about five hundred miles, in which extent of coastline 
 I have visited nearly every harbor of importance. In describing 
 
INTRODUCTION. 
 
 >cked 
 
 lagine 
 )enin- 
 
 Brit- 
 Hud- 
 iy true 
 s, and 
 .t least 
 ;asion- 
 nd lo- 
 . small 
 farther 
 of the 
 ime of 
 as also 
 [le time 
 
 of the 
 ich the 
 ;oast of 
 istine, 
 
 leak — 
 luimaux 
 It was 
 
 station, 
 
 [horized 
 ed but 
 
 :r aban- 
 impos- 
 
 within 
 bid the 
 
 icluded 
 loastline 
 
 scribing 
 
 
 S 
 
 two or three of these locations a sort of idea may be obtained of what 
 is to be seen elsewhere along the coast, as the simple inhabitants 
 differ litde in customs and manners. 
 
 The line which divides these two portions of territory is estab- 
 lished in a direct northern route from the settlement of Blanc Sa- 
 blon, which is only about twenty miles from Bonne Esperance, the 
 residence of Mr. William H. Whiteley, the magistrate for this part 
 of the coast. 
 
 Of course the trading post of St. Augustine was not the only 
 one on the coast. It was simply the only one which the peo- 
 ple within a radius of fifty miles recognized as available for uses of 
 immediate trade, while yet a licensed station. From St. Augustine 
 almost to the head of the St. Lawrence River, many places such 
 as Tadousac, Port Neuf, Goodbout, Seven Islands, and Mingan ; 
 Natashquan, Musquarro, Romaine, and perhaps others, were regular 
 posts of the Hudson's Bay Company, and, together, formed a chain 
 of the most important localities along the coast, all of which be- 
 longed to the Eastern division of what is known as the Montreal 
 Department, which, with the Northern, the Southern, and the Col- 
 umbian Departments, form the four portions into which that great 
 Company is divided. As the southwesternmost of these posts is 
 not far from Quebec, a sort of central station for most of the trad- 
 ing of this region, one can see quite easily that the settlers along the 
 coast are not so far from communication with civilizaHon, especially 
 in summer, as one would at first imagine ; though in the long winter 
 months when the bays and harbors are frozen over so that no ves- 
 sels can approach, and no boats can sail from post to post, to the 
 stranger at such a time on these ice-bound shores the prospect seems 
 dreary enough. In the winter, travelling is mostly on foot, shod 
 with the racket as it is called, a sort of padded or rather wicker 
 snowpad ; or on komatiks or sledges drawn by dogs which can go 
 tDver the high hills, lining the very coast, with safety and often with 
 great speed. Yet Labrador in as low a latitude as it is, compared 
 to what is beyond, and as near civilization as it is compared to what 
 is beyond, though pleasant for a short time in summer, is suffi- 
 
ii!l 
 
 6 
 
 INTRODUCTION. 
 
 ciently dreary in winter. No wonder that at such a time a stranger 
 feels that he has reached the Hmits of civihzed warmth, so to speak, 
 arising from his own country, home ; while he is surrounded by the 
 icy arms of the far north stretching continually downwards and out- 
 wards ! 
 
 We have seen now that Labrador was discovered — no matter 
 by whom — at a time very nearly contemporary with the discovery 
 of Columbus, if we exclude the probable claims of the Norse- 
 men ; that it was once an important fishing station whose develop- 
 ment was in every way hindered rather than helped ; that its relation 
 to the Hudson's Bay Company's trading territory was a very near 
 though not absolutely necessary one, as far as that company was 
 concerned — since the agents of that Company always ran more or 
 less opposition to the Quebec and Halifax traders ; that its re- 
 lation to the Province of Quebec was that of a perfectly natural 
 northeastern dependency or addition ; and that Blanc Sablon, sit- 
 uated as it is at the entrance of the Straits of Belle Isle, was so evi- 
 dent and natural a division between this and the Newfoundland 
 dependency of Labrador proper, that it ought not to be difficult 
 to understand where the one ended and the other began. 
 
 In Anspach's "History of the Island of Newfoundland," 1827, 
 p. 323, the reason for annexing Labrador to Newfoundland is given 
 as follows : — 
 
 " The coast of Labrador, although discovered by Cabot, was 
 very little known until the latter part of the last century, when the 
 progressive increase of the Newfoundland fisheries induced the 
 British Government to extend them to this coast, by annexing it to 
 the government of that island, in the year 1763. The native in- 
 habitants of those parts were included in the regulations which 
 were, at the same time, forwarded to the Governor of the colonies, 
 to prevent the different tribes of Indians from being in anywise 
 molested or disturbed in the possession of such territories as, not 
 having been ceded to or purchased by the Crown of England, 
 were reserved to them as their hunting grounds. All settlements, 
 formed either wilfully or inadvertently upon such lands, were to 
 
 t 
 
 b 
 
INTRODUCTION. 
 
 AX 
 
 latural 
 n, sit- 
 .0 evi- 
 
 1 
 1 
 
 idland 
 
 ■f 
 
 .fficult 
 
 1827, 
 given 
 
 i 
 
 :, was 
 
 fl 
 
 :n the 
 
 41 
 
 d the 
 
 9 
 
 ; it to 
 
 fl 
 
 ^e in- 
 
 '^ 
 
 ,vhich 
 
 ;! 
 
 anies, 
 
 n 
 
 ywise 
 
 
 i, not 
 
 ■'5 
 
 [land, 
 
 'I 
 
 lents, 
 
 '1 
 
 re to 
 
 M 
 
 be immediately given up ; nor were any such lands for the future 
 to be purchased from the said Indians, but in the name of His 
 Ikittannic Majesty, at some public meeting or assembly, of said 
 Indians, to be held for that purpose by the Governor or Comman- 
 der-in-chief of such colony within or near which they should 
 lie. The trade with the said Indians was declared free and open 
 to all British subjects who should take out proper licenses for 
 that purpose. 
 
 This union of the coast of Labrador with Newfoundland, by 
 placing the former under a jurisdiction which could, from local 
 circumstances, more effectually than any other, provide for the 
 maintenance of order and the due administration of justice in those 
 parts, tended materially to increase its importance as a fishery 
 without any injury to the fur trade, both being perfectly compat- 
 ible. When this arrangement was altered in 1774, and the juris- 
 diction of the Governor of Newfoundland was reduced to its former 
 limits, a superintendent of trade, appointed by the Governor-Gen- 
 eral of the Four British Provinces, and responsible to him, used to 
 reside at Labrador. This measure, which appeared to have had 
 for its principal object to encourage the fur trade, must have ])roved 
 very prejudicial to the fishery, and the source of much disorder 
 and irregularity. The re-annexation of the coast of Labrador 
 and adjacent islands to the government of Newfoundland, in the 
 year 1809, was consequently a measure extremely favorable to the 
 interests of the trade and fisheries." 
 
 With a proper understanding, then, of the region visited, I will 
 proceed with the narrative of the expedition, trusting that a greater 
 part of it may be of general interest to the reader. 
 
SS"^ 
 
 ii i 
 
SURVEY OF LABRADOR. 
 
 CHAPTER I. 
 
 I, 
 
 Survey of Labrador — Elementary Physics — Physical Geography. 
 
 The most complete, in fact the only real, survey of Labrador ever 
 made, was in 1832 and 1834, by Capt. H. W. Bayfield, R. N., F. 
 A. S. He was assisted by other parties, though perhaps none of 
 them were -so thorough as was that of Capt. Bayfield — at least he 
 is quoted as authority to this day — and I found that our vessel, by 
 following the directions of his admirable chart, entered and left 
 many an opening of but sufficient size to admit the passage of our 
 vessel, with perfect safety ; showing that the soundings were for the 
 most part taken correctly, and that comparative quiet had remained 
 in regard to subsidence of waters or elevation of land during the lapse 
 of a period of nearly forty years. Though the change of level of the 
 water itself, or the variation in depth, would have been but slight in 
 so short a time, yet the fact that so many of the soundings remained 
 unchanged gave us additional proof against the rising or sinking of 
 the land at that place. Comparing Capt. Bayfield's chart, therefore, 
 with our own soundings, we obtained very favorable and often im- 
 portant results. Previous to a short account of some of the 
 physical phenomena of this region, let us rehearse a few of the 
 elementary princi}>lcs of physical geography necessary for their 
 clear understanding. 
 
 For the purposes of science and convenience, the earth is trav- 
 ersed by imaginary lines running in all directions about and around 
 it. The equator is a great circle passing around the earth from 
 
Ill 
 
 10 
 
 ELEMENTARY PHYSICS. 
 
 :ir 
 
 cast to west and is equidistant from each of the poles. The merid- 
 ians are great circles passing around the globe from north to south, 
 crossing the equator at right angles and meeting at the poles. 
 Tho parallels (of latitude) are smaller circles parallel to the 
 equator. The tropics, marking the highest latitude receiving 
 the vertical rays of the sun, are situated twenty-three and one-half 
 degrees both north and south of the equator. The polar circles 
 are situated (to correspond with the tropics) twenty-three and 
 one-half degrees, the one north of the south pole, and the other south 
 of the north pole, and are the limits of light when the sun is verti- 
 cal at the tropics. The ecliptic marks the apparent path of the 
 sun from tropic to tropic, and as it continues around the earth re- 
 quiring three hundred and sixty-five days for the sun to pass to its 
 end it divides the seasons and days, hence the weeks and months 
 of the year. Latitude is the distance from the equator to the pole, 
 measured in either direction ; thus there are ninety degrees of north 
 latitude and a corresponding number of degrees of south latitude ; 
 the length of a degree of latitude is sixty-nine and one-half miles. 
 Longitude is the distance east or west from any given meridian 
 measured on the equator ; the length of a degree of longitude is 
 also sixty-nine and one-half miles. The relative difference in lon- 
 gitude between two places marks, also, the difference in time 
 between those places ; for since there are 360° degrees around 
 the earth traversed in twenty-four hours, one hour of time must 
 correspond to fifteen degrees, or four minutes to one degree. 
 Wheu, therefore, a place is so many degrees from any given locality, 
 it is easy to see what will be the difference in time between these 
 places ; if east the time is slower, if west faster than the given time 
 at the given place. 
 
 Applying other simple laws we find that the nautical mile is equal 
 to one and one-sixth English miles : or there are sixty of them to 
 a degree of latitude ; it is therefore natural to compute distances 
 in nautical miles by applying the latitude as a measure of distance. 
 
 Though these remarks may not now seem in place, their sugges- 
 tion will become evident at once in measurini^ distances without the 
 
^ 
 
 WAVES. 
 
 11 
 
 ; merid- 
 3 south, 
 ; poles, 
 to the 
 eceiving 
 one -half 
 r circles 
 ree and 
 ler south 
 I is verti- 
 :h of the 
 earth re- 
 ass to its 
 I months 
 the pole, 
 s of north 
 latitude ; 
 alf miles, 
 meridian 
 igitude is 
 :e in lon- 
 
 in time 
 s around 
 [me must 
 
 degree. 
 
 locality, 
 :en these 
 liven time 
 
 is equal 
 
 them to 
 
 [distances 
 
 [distance. 
 
 |r sugges- 
 
 thout the 
 
 use of a regular scale of miles, the degrees and minutes of latitude 
 answering every purpose. 
 
 Waves have been defined as the " alternate rise and fall of suc- 
 cessive ridges of water." They are formed by the action of the 
 wind, or the disturbance of the balance of equilibrium in the sur- 
 face of the water itself. Waves vary in their rapidity of progress, their 
 extent, and their height and breadth, according to the area of water 
 in which they occur, its depth, and the force and direction of the 
 wind ; being smallest in small bodies and largest in large bodies of 
 water. The wave movement primarily is that of simple oscillation, 
 or of repeated risings and fallings without apparent fonvard or 
 backward motion of any kind ; thus a chip thrown ui)on a surface 
 of water affected by such motion retains its position while rising 
 and falling as the w^ater of that area rises and falls. Such is the 
 case if the surface of the water is at apparent rest and there be no 
 wind. Let the wind arise, or the sea be aflectetl by currents, or 
 both, and there arises a con.j)lexity of wave phenomena at once 
 interesting as it is difficult to study with any degree of satisfaction. 
 The other kind of wave motion is called the wave of translation. 
 It is a long, solitary line of crests, such as would be formed by the 
 pushing forward of a mass of water over and above the level of the 
 surrounding water ; it progresses slow or fast according as the depth 
 about is great or small. The typical wave of translation is the tide 
 wave. The phenomena of the rise and fall of the tide are too well 
 known to have escaped the attention of the ordinary individual. 
 They are produced by the action of the sun and moon, primarily 
 of the moon alone, upon the surface of the earth. They occur 
 with regularity in all great bodies of water, and consist of the rising 
 and falling of that body of water effected as a whole, at j^eriods of 
 nearly six hours apart, with a few moments of calm at the end of 
 each period. The period of rising water is called flood tide, and 
 the highest elevation of flood tide high water ; while the period of 
 falling water is called ebb tide, and the lowest ebb tide low water. 
 'I'he i)eriod of time between the highest or lowest tide to the next 
 tide of the same kind, or between one high tide and the next high, 
 
n I 
 
 12 
 
 TIDES. 
 
 or one low and the next low tide is twelve hours and twenty-six 
 minutes. Hence, as there are two such ])eriods in a day, and only 
 tweiiLy-four hours in that period, high tide occurs fifty-two minutes 
 later each successive day. In a month's time this very nearly co- 
 incides with the period of a lunar month, and in fact we find that 
 the highest tides occur at such a time as answers to the new and 
 full moon, and the lowest at the change of first and third quarter. 
 In a table of tides, therefore, the tide, if marked for new moon at 
 a certain place, will answer for every new moon at that place through- 
 out the year and following seasons. Thus in a book of sailing di- 
 rections, a table of high and low water, or rather of spring and neap 
 tides, gives the height of water or time of occurrence of the high- 
 est spring tide (when the moon is exactly at new and full) at any 
 time of any year ; and also the height of water and occurrence of 
 any neap tide (when the moon is at exact first or third (quarter) at 
 any time of any year. 
 
 The interval between is easily determined by taking the time of 
 highest or lowest water, and for each day subtracting fifty-two min- 
 utes from said time. Thus by means of these tables navigators can 
 always tell high or low water at any day or any time of the day for 
 any year, at any place mentioned in these tables ; or they can by 
 marking the spring and neap tides make a table of their own, that, 
 if their reckoning is correct, will answer equally well for other par- 
 ties as for themselves. But a few words of explanation as to the reason 
 of high and low water. I have said that both sun and moon have 
 their influence in the tide of our larger bodies of water throughout 
 the globe. The influence of the moon is about three times that of 
 the sun. The moon attracts both land and water ; one can easily see 
 that the movable particles of the latter will respond to this mag- 
 netic force much more noticeably than the almost immovable 
 particles of the former. The rising of the crest of water, directly 
 under the centre of influence of the moon at full or new, causes 
 of course a depression at the sides, being greatest at the centre of 
 this non-magnetic attraction ; thus the centre of attraction will be 
 distant from the centre of depression by 90°. To restore the equihb- 
 
CURRENTS. 
 
 IS 
 
 2nty-six 
 nd only 
 minutes 
 arly co- 
 ncl that 
 lew and 
 quarter, 
 noon at 
 through- 
 dling di- 
 ind neap 
 he high- 
 1) at any 
 rrence of 
 larter) at 
 
 ; time of 
 two min- 
 ators can 
 e day for 
 y can by 
 wn, that, 
 ther par- 
 le reason 
 3on have 
 roughout 
 ;s that of 
 easily see 
 :his mag- 
 movable 
 directly 
 , causes 
 entre of 
 11 will be 
 equilib- 
 
 
 rium the water must fall to a crest on the opposite side of the earth 
 c(iual to that of attraction beneath the moon. 
 
 The same law will hold good with regard to the sun. Therefore 
 when the moon and sun are in a line with the earth and each other, 
 the greatest elevation will occur, as it does at new and full moon ; 
 and when the moon is farthest from the sun, the attraction of the 
 sun will be exerted on the line of depression 90° from the line of 
 greatest elevation, which is always beneath the moon, and attract 
 that portion to the corresponding depression of the other portion, 
 so that the moon exerts its least influence upon the water beneath 
 it and the tides are the lowest ; these occur at the i st and 3rd (juar- 
 ter. The action of tides may be thus briefly described ; we then 
 come to local variations in height of tide. Thus if the sea beach 
 of any given surrounding is low, and the tide has free access to all 
 parts at once, one can easily see that the tides of that place will be 
 moderate in height and depth and least liable to change, while if 
 the surroundings are narrow and confined and the sea is obliged to 
 force itself against high cliffs and into narrow passages the accu- 
 mulated waters must form an abiding place by raising the level of the 
 water to that of the nearest open basin, which often causes the 
 tides to rise to the extraordinary height of forty, fifty, and even 
 sixty feet above ordinary level. In that narrow arm and clifted 
 channel of the sea, the Bay of Fundy, this remarkable phenomenon 
 actually occurs. At the head of the bay the tides are eighteen to 
 twenty feet high ; here the highest springtides occasionally reach 
 seventy feet. Marine currents are so influenced by local irritation 
 that it would be difficult to describe them, but let us notice a few 
 of the most important ones : — 
 
 It is pretty generally conceded that a constant current sets out of 
 the Gulf between Newfoundland and Cape Breton, in a southeasterly 
 direction. This is changed more to the southward by a current en- 
 tering the Gulf from the Straits of Belle Isle. There are other minor 
 currents due to prevailing winds at the time that affect the waters, 
 but, unfortunately, they have not been studied with sufficient care 
 to define them. There is, however, no doubt about the northern 
 
!'" i 
 
 14 
 
 CUUKKN'rS. 
 
 current, and the huge masses of ice often transported throiigh the 
 Straits would ah^ne prove its existence, coining us tliey do willi 
 their large area of surface directed against a strong opposing south- 
 west wind and even thus reaching the eastern point of Anticosti 
 before finally disappearing. The current bearing these masses is, 
 however, very irregular, being weak at times and very swift at others. 
 In following this most important current I must give the words of 
 the original survey, as simpler than any I can invent, and more 
 truthful than any non-discoverer can give. " After entering the 
 (lulf the current runij along the north orLalirador coast at the dis- 
 tance of two or three miles from the outer islands, leaving a narrow 
 space in-shore in which the streams of the tide, when uninfluenced 
 by winds, are tolerably regular. Passing outside of Mistanorpie, the 
 islands of Great Meccatina and Southmaker's Ledge, it pursues a 
 direction given to it by the trending of the coast till it is turned 
 gradually to the southward by the weak current which is often 
 found coming from the westward between Anticosti and the north 
 coast, during westerly winds, and which is set off to the southward 
 from Natash(iuan point. The united streams continue their south- 
 ern course at a rate diminishing as they become more widely spread, 
 and which seldom exceeds half a knot, and, finally, joining the main 
 downward current out of the St. Lawrence, of which an account 
 will be given immediately, they all pursue a southeast direction to- 
 wards the main entrance of the Gulf, between Cape Ray and the 
 island of St. Paul. It is this current from the northward which is 
 felt by vessels crossing from off the Bird rocks towards Andcosti : 
 and which, together with neglecting to allow for the local attractions 
 of the compass, has been the princijial cause of masters of ves- 
 sels so often finding themselves, unexpectedly, on the south 
 coast. Many shipwrecks have arisen from this cause near Cape 
 Rosier, Gaspe, Mai bay, etc." The same authority adds, further 
 on, as an explanation of the irregularities of current near the north 
 coast of the Gulf, that " both these currents, viz., that from the 
 northward, and the main downward current of the St. Lawrence, 
 are modified by the tides, but in a way directly contrary ; for the 
 
■•f^. 
 
 NAVIGATION IN THE ST. LAWRKNCK. 
 
 15 
 
 ongh the 
 do with 
 ng soiith- 
 Anticosti 
 iiasscs is, 
 at others, 
 words of 
 111 (I more 
 tering the 
 It the dis- 
 5 a narrow 
 nfluenced 
 lofjue, the 
 pursues a 
 is turned 
 h is often 
 the north 
 southward 
 leir south- 
 :ly spread, 
 I the main 
 m account 
 :ection to- 
 y and the 
 which is 
 nticosti : 
 ittractions 
 s of ves- 
 Ihe south 
 |ear Cape 
 s, further 
 Ithe north 
 from the 
 Lawrence, 
 for the 
 
 northern current, in through the Straits of Belle Isle, is accelerated 
 by the flood and ( hecketl by the ehl), while the other is accelera- 
 ted by the ebb, ."ad chec:ked by the flood tide. These modifying 
 causes, viz., the tides and winds, give rise to various combinations 
 and conseciuent irregularities, in the direction and strength of these 
 streams which it is extremely difificult at all times to estimate." 
 The idea, then, that in so narrow an area as that occupied by the 
 Straits, and the (iulf and River St. Lawrence, the main currents 
 are constantly diverted by countt '• currents and inconstantly influ- 
 enced by wind and tide, shows the truth of my words when I say 
 that navigation in these waters, especially when imi)eded by 
 adverse winds and weather, and enveloped in the thick fog so char- 
 acteristic of these regions, is anything but easy, even to the experi- 
 rience.-^ seaman ; many are the tales of horror, at which the heart 
 sickens, known only too well to be true, that might be told of An- 
 ticosti, as well as other prominent places along the south shore es- 
 pecially of the St. Lawrence. Among the causes influencing the 
 current, especially in the river are the spring freshets, annually 
 descending the small streams, and pouring their bodies of water 
 into the St. Lawrence. Here, of course, the accumulation of win- 
 ter's snow and ice, thawing, descends in the shape of an accumula- 
 ted mass of fresh water and makes its influence felt far into the 
 river. Then, too, the tide of the sea reaches far up the river and is 
 also often mistaken for a current in the river itself. All these agents, 
 acting in more or less harmony, often produce a current of three and 
 even four knots an hour, decreasing in v^elocity as it approaches the 
 Gulf. There are also great differences in strength and direction of 
 flood and ebb tides, that frequently produce strong ripples on various 
 parts of the coast, and differing according to locality, so that unless 
 a seaman has studied local phenomena carefully, he will be entirely 
 deceived as to the proper course to pursue to keep his vessel from 
 running aground upon the numerous sandbars here present. It is 
 safe to say that the river St. Lawrence, from Quebec to the Gulf, 
 is one of the most difficult of our large American rivers to navigate 
 ^n all seasons, with continually varying wind and weather. The 
 
^^ 
 
 I M 
 
 16 
 
 WINDS. 
 
 
 
 Gulf is sufificiently difficult for large vessels, while the river is per- 
 ha])s still more so for medium sized crafts. 
 
 Next to the currents the winds are perhaps the most varying cause 
 of assistance or hindrance to the navigator in these same portions 
 of the water. During the greater part of the season " the pre- 
 vailing winds," says our best authority upon this subject, " are either 
 directly up or directly down the estuary, following the course of 
 the chains of high lands on either side of the great valley of the St. 
 Lawrence." This appears to be in the main true, though the same 
 author tells us that " westerly winds do not appear to be so n^uch 
 guided in direction by the high lands, excepting along the south 
 coast." AVinds varying from west-northwest to north-northwest 
 frequently blow for days in succession, accompanied by weather in 
 every way fine and beautiful ; then the wind will often turn and 
 blow from corresponding easterly points as long again, bringing cold, 
 wet, foggy weather, and more or less rain. The easterly winds 
 prevail in spring ; the southwest — especially in the Gulf and at the 
 opening of the Straits — in the summer, and the westerly in the 
 autumn. The north and south winds occur only occasionally, in 
 the winter months the northwest and northerly winds prevailing. 
 Strong gales of wind are of frequent occurrence especially in the 
 Gulf, in autumn ; generally speaking, however, all winds subside 
 more or less at dusk, to be followed by light off-shore breezes dur- 
 ing the night. It will be seen, therefore, that the winds are perhaps 
 more regular than might be expected from the nature of the local- 
 ity ; a good navigator will soon learn to depend upon the weather, 
 and even in a measure foretell it with accuracy. Let him be in 
 whatever part he may of the gulf and river St. Lawrence, a care- 
 ful study of the sailing directions will lead him, except in unusual 
 weather, to pretty safe and sure conclusions as to what the morrow 
 will be, and plans laid upon these conclusions seldom fail of real- 
 ization. 
 
 Besides and next in importance to the currents, and the winds, 
 the marine barometer is indispensable in determining the weather, 
 and guiding one's course both by day and night, here as in other 
 
THE BAROMETER. 
 
 17 
 
 : river is per- 
 
 great bodies of water. A few moments only will serve to give a 
 proper idea of this most important instrument. 
 
 A barometer is an instrument used for ascertaining and measur- 
 ing the weight of the air or more strictly speaking of our atmos- 
 phere. It was invented in 1643, when Torricelli was making his 
 examinations as to the reason why water ascended in pumps to the 
 height of thirty-two feet, and there remained. Taking a glass tube 
 some four feet in length and closed above, he filled it with mercury 
 and reversed the open end in a basin of that same substance. The 
 column sank to a level of 27.5 inches and there remained. Com- 
 paring the column of water with that of mercury he found them to 
 be to each other in height in an inverse ratio of the specific gravi- 
 ties of these two substances; that is, a column of mercury 27.5 
 inches high would balance a column of water 32 feet; and both 
 would remain the same, while the pressure of atmosphere from outside 
 upon the open basins of mercury and water in which the cciumns 
 were inverted remained the same. In 1646, at Rouen, one Pascal 
 repeated this experiment, and fiirther proved that while one column 
 of mercury remaining at the bottom of a mountain underwent no 
 change, a similar column taken to the top of the same mountain 
 was reduced in height by several inches by the diminution of pres- 
 sure, while it regained that same height, and corresponded with 
 the one left at the bottom of the mountain when brought down. 
 By this means he proved the possibility of measuring heights by 
 the variations of the barometer, for thus the tube of mercury was 
 named. Boyle, in 1666, discovered that the atmosphere was elas- 
 tic and compressible, and Mariotte that the density of the atmos- 
 phere was in proportion to the weight with which it was com- 
 pressed. The layer of air nearest the earth was heaviest and 
 sustained the weight of all the rest of the atmosphere above it and 
 each succeeding stratum was lighter and lighter. Aqueous vapor is 
 the amount of water evaporated and held in suspension in the at- 
 mosphere. As evaporation is promoted by dry air, wind, a diminu- 
 tion of pressure, and heat, the quantity thus held in suspension 
 depends upon the temperature. Since " heat expands the gaseous 
 2 
 
t 
 
 m i 
 
 ■ ! 1 
 1 
 
 1 1 
 
 ^ 1 
 
 Mtl 
 
 18 
 
 CHANGES OF TEMPERATURE. 
 
 portion of the atmosphere, the spaces between its particles are en- 
 larged and their capacities for containing moisture augmented. 
 Still further we know that "aqueous vapor is highly elastic ; its 
 elasticity, which increases with an increase of temperature, has 
 been determined by Dalton, and its force measured by the height 
 of the mercurial column it is capable of supporting. Acjueous va- 
 por, raised to 32° Fahrenheit, exerts a pressure on the mercury equal 
 to 0.2 of an inch, at 80° to 1.03 of an inch, at 180° to 15.0 inches, 
 and at 212° to 30.0 inches, — a pressure equal to the pressure of the 
 whole atmosphere at the level of the sea." The amount of vapor 
 existing at any time in the air is determined by an instrument called 
 the hygrometer. By means of this instrument we obtain the (feia- 
 point. When the readings of both thermometer and the hygrometer 
 are alike, the temperature of the dew-point is the same as that of the 
 air ; the air is then saturated or full of moisture. Quoting further 
 from our generally accepted theory, we say that " it is chiefly in 
 the nights and early mornings of the winter months, that the atmos- 
 l)here is saturated with vapor, or that vapor is at its maxiininn of 
 elasticity for the temperature. In our climate, vapor never attains 
 its greatest elasticity at a high temperature ; for if in the summer 
 months the atmosphere becomes saturated it is caused by a decli- 
 nation of the heat, which, contracting the spaces between the par- 
 ticles of the air, squeezes the vapor contained in them closer, and 
 thus brings its elasticity to a maximum for the temperature to which 
 the air has fallen. It was upon the changes of temperature in the at- 
 mosphere that Dr. James Hutton founded his theory of rain. He 
 considered rain to be formed by the mixture of two strata of the at- 
 mosphere of different temperatures, and each stratum saturated with 
 moisture. The mean quantity of the vapor contained by the two 
 strata before the mixture being more than the mean heat of the two 
 (after the combination), the excess is precipitated." The princi- 
 pal causes of variance in the weight of the atmosphere are moisture 
 and heat ; the variation is greater in polar than in tropical countries, 
 and in mountainous than in more level regions. We can now un- 
 derstand the importance of the use of the barometer in determin- 
 
BAROMETRICAL RULES. 
 
 19 
 
 :les are en- 
 lugmented. 
 elastic ; its 
 rature, has 
 ■ the height 
 ujueous va- 
 ;rcury equal 
 [5.0 inches, 
 ssurc of the 
 nt of vapor 
 ment called 
 in the dew- 
 hygrometer 
 s that of the 
 )ting further 
 is chiefly in 
 t the atmos- 
 naximum of 
 lever attains 
 he summer 
 by a decU- 
 :en the par- 
 closer, and 
 re to which 
 re in the at- 
 rain. He 
 a of the at- 
 urated with 
 by the two 
 of the two 
 The princi" 
 Ire moisture 
 ,1 countries, 
 n now un- 
 determin- 
 
 ing the weather, especially in such a country as is found along the 
 coast of Labrador. Barometers are of two varieties : the common 
 mercurial, where the air acts directly upon a basin of mercury in 
 which the inverted column of that same substance is placed ; and 
 the aneroid, where the atmosphere acts upon a metallic box, from 
 which the air has been exhausted, and by its pressure communicated 
 by a system of levers acts upon the needle to register the amount 
 of variance ; this is possible, since the column of mercury is 
 counterbalanced by the weight of the atmosphere. For a barom- 
 eter to be of the greatest use it must be read in connection with 
 the direction of the wind, and the temperature of the air, as shown 
 by the thermometer. The hours in the day when it stands highest 
 are at 9 a. m., and 9 p. m. ; and it is usually lowest at 3 a. m., and 
 3 p. M. These hours, therefore, are the best for making observa- 
 tions and are generally used by scientific men generally the world 
 over. 
 
 The following table of rules will be found to apply in nine chances 
 out of ten, for the correct use or reading of the barometric needle. 
 
 Rising barometer with south wind, — fine weather. 
 
 Sudden rise, wind N. or N. N. W., in broken cold weather, 
 — rain or snow and sometimes nightly thaws. 
 
 Rapid rise after S. W. gale and rain, — clear sky and sharp 
 white frost. 
 
 Steady high pressure, wind strong W., — high temperature and 
 very little rain. 
 
 Steady high pressure, wind strong E., — lower temperature and 
 sharp frost. 
 
 Trailing barometer, with N. wind, — cold rain and storms in 
 summer, deep snow and severe frost in winter. 
 
 Falling barometer, wind S., — more or less rain. 
 
 Falling barometer, wind N. W., — cold rain in summer, severe 
 frost in winter. 
 
 Falling barometer, high S. W. wind, — increasing storm. 
 
 Steady and large fall, wind E., — wind S. or heavy snow or rain. 
 
 Sudden and large fall, wind W.,— violent storm N. W., or N. 
 
!! 
 
 I 
 
 I! I 
 
 20 
 
 BAROMETRICAL RULES. 
 
 Great fall during frosty weather, — thaw continued with S. or 
 S. E. wind, and returning frost if S. W. wind. 
 
 Lowest depressions, wind S. or S. E., — much rain and severe 
 gale. 
 
 In general : a rapid rise gives a violent wind ; continued fall, 
 continued wind. 
 
 Great depression in sumn.<ir, — storms, wind and rain, thunder 
 and hail. 
 
 Rise with S. wind, — high temperatures. 
 
 Mercury unsteady, — air in electrical state. 
 
 No great storm sets in with a steady rise. 
 
 N. and S. winds are the origin of our greatest storms. 
 
 W. winds blow mostly at night. 
 
 E. winds calm at night, blow by day. 
 
 There is least wind at sunrise and sunset, and most wind at i or 2 
 p. M. ; wind with the sun fine ; wind against sun mercury falls, bad 
 weather generally. 
 
 Meteors are not common during low temperatures ; the Aurora 
 borealis has been seen at all heights of the barometer. It has 
 been noticed and recorded that, " the finest and most beneficial 
 state of the atmosphere, more especially as regards the health of 
 man, is with a uniform pressure at the mean height of the climate 
 varying from 29.80 to 30.00." 
 
 A poetic barometrical rhyming table which lately came to my 
 notice reads as follows : 
 
 "When rise begins after low, 
 Squalls expect, and clear blow; 
 Long foretold long last, 
 Short notice, soon past, 
 First rise after low 
 Foretells stronger blow." 
 
 Still another comes to mind : 
 
 "When the glass falls low, 
 Prepare for a blow ; 
 When it rises high, 
 Let your kites fly." 
 
CLOUD COMBINx\TIONS. 
 
 21 
 
 ith S. or 
 id severe 
 lued fall, 
 I, thunder 
 
 d at 1 or 2 
 ^ falls, bad 
 
 the Aurora 
 
 ;r. It has 
 
 3eneficial 
 
 health of 
 
 le climate 
 
 me to my 
 
 I 
 
 4 
 ■I 
 
 ■1 
 
 A great many other foretellings of weather are known in one 
 shape or another, but it is not the province or purport of this work 
 to be a prophetic indicator of the weather, but simply to explain 
 some of the most common and important physical phenomena of 
 the region about Labrador : hence a few of the usual combinations 
 of thermometer and barometer readings, which may generally be 
 relied upon with a tolerable degree of accuracy in guiding the 
 mariner along these coasts, where, in summer, travelling is by water 
 and a fair day is predicted as far ahead as possible. 
 
 The combinations of the clouds are also great indicators of 
 weather and affect the barometer more or less indirectly. The 
 cirrus is seen at all seasons of the year and at all heights of the 
 barometer ; it has a slow motion in fair and a rapid one with falling 
 barometer in foul or stormy weather. The cirro-stratus, not unlike 
 the cirrus of which it is a peculiar condition, is the forerunner of a 
 falling barometer with wind or rain. Sometimes it appears after a 
 rapid rise in the mercury ; then also rain generally follows soon. 
 The cirro-cumuius comes with a rising barometer and is a warm 
 weather cloud. The cumulus is seen chiefly in spring and summer ; 
 it is seen in showery weather with the cirro-stratus, and in hot weather 
 alone or with other clouds. 'Tf during a fine morning this cloud sud- 
 denly disappears, and it be followed by the cirro-stratus with the 
 wind backing to the south, the mercury falls, and rain soon follows. 
 The cumulus is a day cloud ; its greatest density keeps off the too 
 scorching rays of the noonday sun ; it usually evaporates an hour 
 or two before sunset. When it increases after sunset, and shines 
 with a ruddy, copper-colored Hght, it denotes a thunder storm." 
 The cumulus is seen with a rismg barometer generally. The cu- 
 mulo-stratus appears much like the cumulus, indicating sudden 
 changes of wind, thunder scjualls, and even hail. It tends to raise 
 the mercury. The stratus, in the words of the same authority 
 above quoted, "is formed from the sudden chill of certain strata of 
 the atmosphere, which, condensing the vapor contained in them, 
 renders it visible in a misty cloud or creeping fog. Calm weather 
 is essential for the formation of the stratus ; it is frecjuent in fine 
 
 
* ,; 
 
 ■ 
 
 , '1 
 
 1 
 
 Mil 
 
 1 ! 
 
 ! 1 
 1 I 
 
 i 
 \ 
 
 ! ■ 
 
 : 
 
 'l i 
 
 - 
 
 
 ; i 
 
 
 
 
 ■^ 1 
 
 
 ; 
 i 
 
 1 ' 
 1 1 
 
 1 
 i 
 
 1 
 
 22 
 
 DEW-POINT. 
 
 autumnal nights and mornings. It obscures the sun until his rays 
 have raised the temperature of ilic u'u sufficiently to evaporate it, 
 when it gradually disappears and leaves a clear, blue sky. The 
 stratus deposits moisture, is called the night cloud and is most fre- 
 quent from September till January. When the temperature, from ra- 
 diation or other causes, sinks below 32°, we find it fettered with icy 
 spiculae upon trees and shrubs and sparkling in exquisite frost work 
 upon all nature." It is not known to affect the barometer much 
 either way. The nimbus is seen during showers : it is not seen with 
 the barometer at great heights. A study of vapor point would be of 
 great interest to one desiring to pursue this subject further ; and 
 there are plenty of books on Physics and Physical Geography for 
 those who are thus inclined, not to waste useless space and time 
 here upon definitions. One point further, however, before we 
 leave lae subject : — 
 
 We see, therefore, that the rising or falling of the barometric 
 needle corresponds with the pressure of the atmosphere. This 
 pressure is due to changes of temperature, moisture and wind. 
 To show that the barometer is more of an instrument than appears 
 upon first sight, and must be studied in itself without reference to 
 the words marked upon its face, a clause from one of our 
 scientific college text-books of natural philosophy tells us that "the 
 practice formerly prevailed of engraving at different points of 
 the barometric scale several words expressive of states of weather : 
 fair, rain, frost, wind, etc., etc. But such indications are worthless, 
 being as often false as true : this is evident from the fact that 
 the height of the column would be changed from one kind of 
 weather to another by simply carrying the instrument to a higher 
 or lower station." The measuring of heights by the barometer will 
 be spoken of in another place. It remains but to speak of 
 the dew-point, and the resultant forms of water manifested. The 
 dew-point is "the temperature at which vapor, in a given case, 
 is precipitated into water in some of its forms." The amount rf 
 vapor in the air is expressed by its tension or elastic power, ana its 
 humidity, or " its quantity present, as compared with the greatest 
 
^^mjf 
 
 USE OF BAROMETER. 
 
 23 
 
 til his rays 
 'aporate it, 
 sky. The 
 s most fre- 
 re, from ra- 
 ed with icy 
 : frost work 
 leter much 
 3t seen with 
 would be of 
 irther; and 
 jgraphy for 
 e and time 
 before we 
 
 barometric 
 here. This 
 and wind, 
 lan appears 
 eference to 
 ne of our 
 IS that "the 
 
 points of 
 if weather : 
 
 worthless, 
 e fact that 
 ine kind of 
 ;o a higher 
 ometer will 
 p speak of 
 Isted. The 
 
 iven case, 
 [amount ^f 
 Jver, and its 
 
 le greatest 
 
 possible amount at that temperature." When the air is colder than 
 the temperature of the dew-point, the surplus amount of vai)or is 
 condensed by the amount of difference of pressure, and the result 
 is dew, frost, fog, etc. The first deposition is dew ; then, according 
 as the radiating body is acted upon with greater or less amount of 
 temperature, we have frost ; then fog, which becomes clouds of dif- 
 ferent kinds, forms, and shapes ; these in turn condense and i)re- 
 cipitate rain, which pressure still forms into a spray called mist ; and 
 this, crystallizing, gives us, first hail, then sleet, and finally snow in 
 all its perfection of microscopic crystals. 
 
 Before closing my remarks on the barometer a few quotations 
 from the Sailing Directions will serve to give you a good idea of 
 the use of this instrument, or rather of the use it may be if carefully 
 and properly studied. "The barometer has a range of from 29 to 
 30.5 inches in the Gulf and River St. Lawrence during the navigable 
 season, and its changes accompany those of the winds and weather 
 with a considerable degree of constancy. The fluctuations of the 
 barometric column are much greater and more frequent there than 
 in lower latitudes, and sudden alterations, which in other climates 
 would be alarming, may occur there without being followed by any 
 corresponding change either in the wind or in the weather." The 
 most practical part is still to follow, and expresses the pith of all 
 thit can be said thus concisely : "But the navigator should not be 
 inattentive to these minor changes, as a constant attention to the 
 instrument can alone enable him to appreciate those decisive indi- 
 cations of the mercury which seldom or never prove deceptive." 
 
 Quoting further from the same source, of the fogs it says : "they 
 may occur at any time during the open or navigable season, but 
 are most frequent in the early part of summer ; they are rare, and 
 never of long continuance during westerly winds, but seldom fail 
 to accompany an easterly wind of any strength or duration. The 
 above general observation is subject, however, to restriction, ac- 
 cording to locality or season. Thus winds between the south and 
 west, which are usually clear weather winds above Anticosti, arc fre- 
 ([uently accompanied with fog in the eastern parts of the gulf. 
 
i'f 
 
 24 
 
 FOGS. 
 
 ! 
 
 h 
 
 " 
 
 Winds between the south and east are most ahvays accompanied 
 with rain and fog in every part. E. N. E. winds above Point de 
 Monts are often E. S. E. or S. E. winds in the Gulf, changed in direc- 
 tion by the high lands of the south coast, and have therefore, in gen- 
 eral, the same foggy character. Winds of considerable strength and 
 duration are meant here, which probably extend over great dis- 
 tances. Moderate or partially fine weather winds may occur with- 
 out fog at any season, and in any locality. In the early part of the 
 navigable season, especially in the months of April and May, clear 
 weather N. E. winds are of frequent occurrence, and they also 
 sometimes occur at other seasons in every part of the Gulf and 
 River St. Lawrence. "The fogs sometimes last several days in 
 succession and to a vessel either running up or beating down, dur- 
 ing their continuance, there is no safe guide but the constant use of 
 the deep-sea lead, with a chart containing correct soundings. 
 
 "The fogs which accompany easterly gales extend higher up into 
 the atmosphere, and cannot be looked over from any part of the 
 rigging of a ship. They, however, are not so thick as those which 
 occur in calms after a strong wind, and which are frequently so 
 dense as to conceal a vessel within hail, while the former often, 
 but not always, adm't the land, or other objects, to be distinguished 
 at the distance of half a mile or more in the daytime. 
 
 "The dense fogs which occur in calms, or even in very light 
 winds, often extend only to small elevations above the sea ; so that 
 it sometimes happens that when objects are hidden at a distance of 
 fifty yards from the deck, they can be plainly seen by a person fifty or 
 sixty feet up the rigging. In the months of October and Novem- 
 ber, the fogs and rain that accompany easterly gales are replaced 
 by thick snow, which causes equal embarrassment to the navigator." 
 
 I have frequently proven the truth of nearly every statement 
 contained within the above quotation. 
 
 One other subject remains to be spoken of, and that but briefly : 
 It is the ice of the Gulf of St. Lawrence and the Straits of Belle 
 Isle. It is very fortunate for us that the admirable work done by 
 Bayfield along this part of the coast remains as a monument to his 
 
DRIFT ICE. 
 
 25 
 
 ompanied 
 : Point de 
 d in direc- 
 ire, in gen- 
 •ength and 
 great dis- 
 iccur with- 
 part of the 
 May, clear 
 . they also 
 ; Gulf and 
 ral days in 
 down, dur- 
 jtant use of 
 lings. 
 
 ;her up into 
 part of the 
 hose which 
 jquently so 
 rmer often, 
 tinguished 
 
 very light 
 ea ; so that 
 distance of 
 rson fifty or 
 id Novem- 
 e replaced 
 navigator." 
 statement 
 
 )ut briefly : 
 ts of Belle 
 (k done by 
 lient to his 
 
 # 
 
 thoroughness of exploration and the study of the phenomena of this 
 region. His well chosen language often supplies to us the lack- 
 ing power of expression in descri])tions of this kind, and we take 
 his words in preference to our own. On this subject, also, he re- 
 marks : "In spring, the entrance and the eastern jxirts of the Gulf 
 of St. Lawrence are frequently covered with drift ice, and vessels 
 are sometimes beset with it for many days." I have seen a vessel 
 in the spring of the year closed in on all sides by ice that formed 
 during the niglit augmented by drift, and obliged to stay there the 
 remainder of the week until the sun broke the surrounding crust 
 and the wind dispersed the drift. Again he says : "lieing unfitted 
 for contending with the danger, they often suffer from it and are 
 frequently lost ; but serious accidents from this cause do not occur 
 frequently, because the ice is generally in a melting state from the 
 powerful effects of the sun in spring. In the fall of the year, acci- 
 dents seldom occur from ice, except when the winter commences 
 suddenly, or when vessels linger imprudently late from the tempta- 
 tion of obtaining high freights." Several instances have occurred 
 where boats and small vessels have thus lingered late and been 
 overtaken by the breaking up of previously formed and solid ice 
 by the sun's rays, on some warm day, and thus been carried off into 
 the Gulf, or wrecked on some of the islands or more dangerous 
 portions of the surrounding coast. In the fall when the ice from 
 above breaks up and is taken by the current through the Straits of 
 Belle Isle, ships often run into large floating bergs that are lying 
 about here and there in the water, and directly in the way of nav- 
 igation, and are either crushed by the falling of the berg upon them, 
 if it be a tall and large one (for it must be remembered that from 
 two-fifths to nine-tenths of an iceberg remains under the water, and 
 that, gradually worn away by abrasion and the warmer waters 
 of the current, it has become so brittle that a jar easily displaces it 
 and therefore causes the remainder to assume a different position 
 — by which I mean that the whole structure falls — and thus doing 
 damage according to its size) ; or, again, a vessel going with her 
 usual speed may run into a solid block of ice, where indeed the cur- 
 
IT' 
 
 26 
 
 DRIFT ICE. 
 
 M 
 
 h! 
 
 ■■' .< 
 
 rent would be otherwise unimpeded, and cause much damage in 
 this way. In spring tlie water is covered with what, in Labrador, is 
 termed shcshe ice ; this is a thin mass of slob (another expression 
 used here) that has formed by frost or snow, or a combination of 
 both, during the night, which in the morning when the sun comes up 
 breaks up and goes floating about in the water, often attaching 
 and carrying away with it still other pieces of stray ice, and 
 becoming generally dangerous. Small boats, that are at this season 
 numerous, since the dog sleds can no longer be* used, often get 
 entangled in these masses, while a turn of the weather brings cold 
 that cements them into the ice which forms an impassable barrier 
 to further progress. The boat cannot urge a way through it, and 
 the foot cannot yet walk on it, the craft drifts at the mercy of the 
 tide, the currents, and the wind, and is c!'ten carried out into the 
 Gulf, or thrown in contact with the shore, while woeful tales of 
 starvation are known to be only too true from this as other like 
 causes. The ice causes another sort of damage here, which is a 
 source of great annoyance. Large numbers of pieces of ice are 
 found at the first breaking up of the bays and waters, until August, 
 and often even later. I have known a piece to strand on some 
 fishing ground, and with a rising tide go on its slow but in a 
 measure almost irresistible way directly through the well laid nets 
 of the fishermen, carrying them off bodily into the sea. Of course 
 this cannot be helped, yet I have witnessed this spectacle, with the 
 unfortunate fellows thus loosing their net, from the shore, unable 
 to do anything with them to recover the lost property. Naviga- 
 tion in the Gulf and River St. Lawrence is usually closed the first 
 of December, and remains so until the first of May ; even after 
 that, drift ice in large quantities is liable to occur for a month more. 
 
TRIP TO LABRADOR. 
 
 27 
 
 mage in 
 rador, is 
 press ion 
 lation of 
 omes up 
 ittaching 
 ice, and 
 is season 
 aften get 
 ings cold 
 le barrier 
 ^h it, and 
 cy of the 
 t into the 
 il tales of 
 )ther like 
 ,vhich is a 
 )f ice are 
 ;il August, 
 on some 
 but in a 
 laid nets 
 f course 
 :, with the 
 •e, unable 
 Naviga- 
 the first 
 ;ven after 
 mth more. 
 
 CHAPTER II. 
 
 Trip to Labrador — Arrival at Montreal — Arrival at nuebec, and Description 
 of the City. 
 
 It was in the early part of September that I first formed the idea 
 of a trip to Labrador, where I hoped to remain during the winter 
 months, and the following season. I had been working hard 
 during the previous year, and forming an idea that a trip to 
 Labrador and a study of the natural history of that region would 
 be of great use in the determination of the variety of species and 
 their geographical distribution in migration in the study of New 
 England Natural History, I determined to profit by an opportunity 
 offered and start for the coast. 
 
 Knowing that the last vessel left Quebec for Labrador some- 
 time during the above mentioned month, a letter and a telegram 
 apprised us (myself and two friends who also wished to go) that 
 we must start for Quebec at once if we wished to reach this means 
 of conveyance before it was too late. My letters reached me on 
 Thursday the 9th of the month, and this gave but two days to pre- 
 pare for a journey of over a thousand miles, and an absence of at 
 least a year in the cold region of the North, since I must leave by 
 the late train on Saturday night to be in time to reach the vessel 
 which the letter said would probably start Tuesday from Quebec. 
 
 The suddenness of the decision which I had thus formed will be 
 seen when I say that many of my friends saw the news first in the 
 papers before even my letters reached them, though I had written 
 as soon as it was possible after deciding. Intending to purchase 
 
 Hi 
 
TT 
 
 28 
 
 LEAVING HOME. 
 
 the greater part of my supplies in Quebec, I put in my trunk only 
 such things as I needed, adding a few simple medicines, since I 
 was going to a place where there were no doctors, and, bidding 
 my home and friends the usual form of parting, was at the station 
 promptly for the train. 
 
 Travelling by night is at no time pleasant to the majority of peo- 
 ple whose occupation takes them from place to place, and it is often 
 especially unpleasant where the nine o'clock evening train is a slow 
 mixture of a very few passenger and a great many freight cars, whose 
 fastest gait is an amble, and its stops fre(|uent and of long duration ; 
 while the interior o' the most inhabitable of the cars upon the pas- 
 senger list is at best a dim, cloudy, close atmosphere of condensed 
 and accumulated dirt and soot and bad breath, — said interior look- 
 ing as if it had seen neither pure air nor water since the day of its 
 first appearance on the road up to the present. Such as it was we 
 were all soon seated, as comfortably as possible, having waved our 
 hands at our friends on the platform, feeling that the car was 
 really in motion and we were being surely drawn towards the north 
 and away from home. It was then that 1 had my first opportunity 
 for reflecting : first, that I was leaving home for a new and compara- 
 tively little-known region, second, that I might have left the very 
 things that I should most want and need and taken things that would 
 be of comparatively little or no use — though there was really little 
 to fear on that score; — and third, as to the prospect of outliving 
 the fierce winter of such a region as that to which we were going. 
 
 I did not long trouble myself with these reflections, however, as 
 the hurry, excitement, and constant bustle of the past two days had 
 left me considerably fatigued, and so turning over an empty seat 
 behind, and arranging mysolf iu the most comfortable position 
 possible where the seats are hard ..t the best, I composed myself to 
 sleep. I say to sleep, but I should much better have said to try to 
 sleep ; no doubt the excitement of the previous days rendered it a 
 more difficult matter than it would have been had the journey been 
 longer contemplated. Thus I dozed away and caught an occasional 
 nap of short duration throughout the night, until we reached St. 
 
 
 
 ^ 
 
4 
 
 ARRIVAL AT MONTREAL. 
 
 20 
 
 All)ans somewhere in the small hours of the morning. After a httle 
 refreshment, and a short stay here, not long enough, however, to 
 enable me to get much of a glimpse of this qu»;er old Canadian 
 and American town, yet with [.lenty of time to secure the latest 
 New York and lioston newsi)ai)ers, the train which was *o take 
 us to Montreal arrived ; and soon we were whirling off in the di- 
 rection of that place, comfortably seated in a car whose atmosphere 
 was a pleasing variation from that of the car we had so lately 
 left, in that its purity was its main attraction. We had now daylight 
 before us, and though st'll rather tired and sleei)y, could at least 
 vary the monotony of - ir trip with the natural objects that i)re- 
 sented themselves to our gaze through the car windows. An 
 officer soon awakened those disposed to slumber with the in- 
 telligence that we were crossing the line and must soon submit 
 to having our baggage inspected. I thought it very kind of him 1 1 
 give us notice beforehand of this most important event, and also 
 most kind of the inspector when he appeared, and most indicative 
 of his ability to judge of human character by the face, that he 
 gave us so little trouble with our bundles. These contemplations, 
 however, did not long deter me from another attempt at a nap, and 
 once more settling myself down I tried to sleep. 
 
 As our train went along slowly it was sometime before we reached 
 Victoria bridge spanning the St. Lawrence below Montreal ; after 
 a long rumbling through this covered bridge, whose little skylights 
 far above the head on the side of the building shed but a feeble 
 occasional gleam, we emerged into the light again only to pass along 
 through low fiats of land or near high walls of buildings, both of 
 which seemed equally the abodes of poverty and dirt, and reached 
 Montreal two hours late since we were due at 8 and it was 
 then lo A. M. Here our trunks were inspected with the same 
 kindness that our bundles had been in the cars — mine not being 
 opened at all — and we were soon rumbling over the stones to the 
 hotel where an abundance of good warm substantial food soon 
 revived us again. It is quite unnecessary to enter into a description 
 of this old-fashioned place, as that has been too often done 
 
 4 
 
 \i\ 
 
I 
 
 'Si;: 
 
 30 
 
 EN ROUTE FOR QUEBEC. 
 
 f 
 
 and to a much better advantage than anything I could say with 
 my present knowledge of the city, confined as it is to a day only, 
 and that too a Sunday, when none of the places of interest to the 
 visitor were open, and the long lines of huge, gloomy stone buildings 
 frowned upon pavements, and churches, especially that massive 
 structure, the Notre-Dame Cathedral alone, were the scene of flock- 
 ing multitudes. 
 
 Early the next day, at about seven o'clock, we were aroused 
 from our sleep and told to get ready at once as the train for Quebec, 
 and the one on which we intended leaving, started from the other 
 end of the city at about half-past seven, a truly cheering bit of news. 
 Hastily dressmg and descending to the hotel parlor I found my 
 friends just hurrying down fiom their room, and together we went 
 to the clerk to inquire if breakfast was ready ; to our surprise it 
 was not, but soon would be. Here was a long delay of five or ten 
 minutes before the door to the breakfast-room was opened, and we 
 were summoned to table just as the coach drove up to the door 
 bringing us word that we were probably already too late to catch 
 the train ; but breakfast must be eaten, and some six minutes were 
 lost again which made the driver sure we were too late, while it 
 assured us that we had had just enough hot steak and coffee to 
 make us wish for more. 
 
 Nothing could be done, however, but to pay an exorbitant hotel 
 bill and urge the coach off at its fastest gait for the depot at the 
 other end of the town. We had six minutes, I think, in which to 
 accomplish our journey of about a mile over the hardest of pave- 
 ments, and in a vehicle that tossed about from side to side threaten- 
 ing every moment to upset as we fairly steamed down the streets, 
 rattling as if our life depended on the amount of noise made, and 
 bouncing about in a manner calculated to digest what little nourish- 
 ment we had taken before our ride was half completed. 
 
 We reached the door of the station on the minute, purchased 
 our tickets, received our checks from the driver, who had left his 
 team at eminent risk of its running away to procure them, and 
 stepped on the train as it was just moving from the depot. 
 
 J 
 
STOr FOR LUNCH. 
 
 31 
 
 Travelling by rail is at no time pleasant when the train is a slow 
 one, and the prospect of an all day's trip a pretty safe one. I do 
 not mean to enter into the subject of conveniences or inconve- 
 niences of such travelling, but true it is, that one can often make 
 one's self comfortable or uncomfortable, as one desires, when placed 
 under such circumstances. Desiring the former, as by all means 
 to be preferred to the latter, I took off my tall hat and, much to 
 the apparent envy (I flatter myself judging from their looks) of 
 several persons around me, drew from my pocket, and adjusted on 
 my head, a soft, brimless, smoking cap, while I lay back in the 
 seat, head to the window and feet on the extreme end of the cush- 
 ion and against the handle bars, and read my book. Oblivious to 
 all around, I continued to peruse the before mentioned volume 
 till our noon stop at a long-platformed, spacious-roomed station. 
 Finding, here, that it would be an hour or more before the train 
 was made up that was to take us on to Quebec, or rather Point 
 Levis, as we were on the Montreal side of the river yet, I took a 
 hasty lunch in the refreshment room and then started off in the 
 direction of, the river which here flowed gently by not far off, at the 
 right of the railroad track. The water was quite shallow at this 
 place, and long banks of sand extended here and there in points as 
 far as one could see. I thus walked easily half way across the 
 stream, stepping cautiously over rills and damp and soft places, 
 until I came to the main channel of the water. As far as I could 
 see, both up and down the stream, these low sand flats extended 
 fully half way across the bed of the river ; exposed now, though 
 probably a part of the year several feet under water. 
 
 Dead shells of various species of Unio, and Anodon, common to 
 fresh water rivers fardier south and west, were everywhere in abun- 
 dance. I gathered several distinct species, all of which probably 
 live in the river in great numbers. 
 
 But time progressing, I unwillingly relinquish^ i my pleasai.t 
 aci^uaintance with these attractive natural surroundings of a place, 
 -^ ^ne too well known natural-history-wise, and j)ocketing my spec- 
 imens started for the station. To the amusement of several gen- 
 
 !:i 
 
 if 
 
 f 
 
 Jli 
 
 I I. 
 
 i 
 
 ■ 'IF 
 
 w 
 
 \ i! 
 
Tjr 
 
 1 II 
 
 II 
 
 
 82 
 
 ARRIVAL AT TORT LEVIS. 
 
 J 
 
 
 I I 
 i 
 
 
 i i! 
 
 II 
 
 tlemen of the, perhaps, loafing profession, — though I may be clo- 
 .ing said gentlemen a great injustice, for which I beg many pardons, 
 not intending my remark as an insult, but rather using the word in 
 place of any knowledge of the occupation of said persons who 
 were strangers to me, — whose inquisitiveness led them to assemble 
 about nie and ask various questions, I emptied my pockets on the 
 sill of a window of the station house and proceeded to do the 
 specimens up carefully in papers that I might the more easily cairy 
 them, and having placed them in a satchel, I began to look about me 
 for the train. Before long it approached the station, and after stand- 
 ing still in front of the door for half an hour or so, during whii^h 
 time we transferred our portable luggage and ourselves to a 
 comfortaiile seat near the rear of the last car, it slowly started off. 
 and again we were on the move. 
 
 The train was a slow one, and after occupying half of the aftei - 
 noon reading I spent the greater part of the remainder of the trip 
 on the rear platform watching the nature of the country through 
 which we were passing. The air was cool and clear, and veg- 
 etation and foliage still wore their summer dress, but in place of 
 the usual moist feeling which New England air possesses, the words 
 dry and crisp seem to express, to me at least, the condition of 
 earth and air in this "cold temperate " region of the globe. Stand- 
 ing thus and enjoying the comfort and ease with which the car 
 slowly proceeded through the varied scenes along the passage, the 
 shades of evening approached, and as darkness came on, passing 
 through the outskirts of what was apparently a manufacturing vil- 
 lage, by day the scene of industry, but now a confusion of shops, 
 sheds, wharves and buildings, as seen thus in the darkness of night, 
 we came at last to the celebrated Port Levis. 
 
 It was night when we reached this ancient military station, and 
 b' ing in great haste to see to the transporting of our luggage to the 
 ferry boat which was then waiting for us and the other passengers 
 who were to be transported to the Quebec side, there was no op- 
 portunity of going about the place at all with the view of exam- 
 ining its huge fortresses of military fame, and of which we had read, 
 
CROSSING THE FERRY. 
 
 88 
 
 ! ■ 
 
 Y be do- 
 pardons, 
 ; word in 
 ons who 
 assemble 
 ts on the 
 do the 
 sily cairy 
 about me 
 ter stand- 
 ing whi('h 
 ves to a 
 tarted off. 
 
 the aftei - 
 
 )f the tr^p 
 
 y through 
 
 and veg- 
 
 place of 
 
 the words 
 
 idition of 
 
 Stand- 
 
 li the car 
 
 .sage, the 
 
 p, passing 
 
 ;uring vil- 
 
 |of shops, 
 
 of night, 
 
 as well as its other attractions of merit ; we could simply reflect 
 on the immense cost of warfare, and of the reported expenditure 
 of some millions of dollars in the construction of the three forts 
 here represented, as well as their maintenance, and which, defend- 
 ing the lower harbor and city of Quebec, rendered the place fa- 
 mous. If my memory serves me, it was in 1759 that the lower 
 town of Quebec was bombarded from the heights of this town by 
 General Monckton, at the time the English captured the former 
 place and reduced it to an English Province. Hastily recalling 
 these events the boat is quite ready to leave, and we are soon set- 
 tled and pursuing our way accompanied by the dash and splash 
 of the paddles as heard from outside, and the nearer plunge and 
 spit of the engine as it drove the machinery, while the fires below 
 roared and crackled, and the boat shivered from stem to stern as 
 she urged her way forward in the dark waters which sent billows 
 of foam on either side of the prow, to reflect an instant in some 
 light either from the vessels near by, or the lamps on the opposite 
 shore. 
 
 The scene that presented itself to us as we stood thus on deck 
 looking out into the twilight about and beyond was one of strange 
 fantasy, you might almost say : above, on our left, dark frown- 
 ing heights were illuminated with lamps that sent their rays 
 in many directions penetrating the darkness, while all about us the 
 S(^litary lights of vessels lying at anchor glimmered from a darkness, 
 vvhich occasionally reflected a darker outline of the vessel to whose 
 ligb we were nearest. We wove in and out of these until, rounding 
 the cliff, the full blaze of the lower town shone upon us, and 
 si owed the confusion of a low-roofed housed city on the left, and 
 a li .\ i- full of boats, vessels, and ships of all kinds and descriptions 
 on the right ; thus we approached the landing. 
 
 I had chosen thv„ deck of this little ferry steamer for my outlook 
 upon the city, as I could then have plenty of air, and be free from 
 the crowd below. The boat was small and substantially built, 
 rather cramped for standing or sitting room, and inconvenient ; 
 besides, the body of the lower part was an open receptacle for 
 3 
 
 ? t 
 
iii,' 
 
 Ml 
 
 34 
 
 ARRIVAL IN QUEBEC. 
 
 i ;llii 
 
 i: 
 
 Il: 
 
 I M 
 
 li 
 
 bales, boxes, and bags of luggage which nearly filled it, and the 
 passengers' rooms small and full of angles and corners ; every- 
 where a strong smell of closeness overladen with tobacco smoke 
 filled the air, and moving about was attended with a great deal 
 of difficulty. Congratulating myself that I had secured as airy 
 a place as anybody, and viewing the scene as I have attempted to 
 describe it, we slowly made the landing and were in Quebec. 
 
 The lower city of Quebec })resents very few attractions by night 
 to a stranger ; the streets and the sidewalks are narrow, while both 
 are muddy and slippery at this season of the year, especially 
 if it has lat«:ly rained. The cooped-up feeling that it gives you 
 to land in t '>f these narrow alley streets and find that you 
 know nobody a. nody knows you, that you don't know where to 
 go and a dozen cu... drivers and hotel porters all say that you are 
 going with them, is a peculiar one and best appreciated by anybody 
 that has been in such a place and obliged to decide immediately 
 what to do, or have it decided for him by having his baggage 
 and himself suddenly ushered into a cab and the door shut, — this 
 is anything but pleasant. Though I did not undergo this last ex- 
 perience literally, it was so near becoming such that the cab started 
 off with the luggage ; while, after a few inquiries and some compli- 
 cated directions, I started off to find the wharf at which the vessel 
 destined to convey me away was moored. My companions and 
 myself went together and after travelling through various streets and 
 turning down many side alleys, all of which presented the same 
 dirty, narrow, and contracted general appearance, bestrewn with 
 plenty of mud and nearly solitary, we approached a shed-like open- 
 ing on the opposite side of a very dirty street, through which we 
 passed and emerged on a long wharf, on either side of which were 
 vessels closely packed and in process some of lading and some 
 of unlading. Not knowing which one contained the object of our 
 search, — the captain, with whom we had not yet even taken pas- 
 sage, — we hailed a man standing by and inquired of him ; from his 
 reply, only given after the question had been repeated in French, we 
 found the vessel, aroused some of the crew, and put to them our 
 
 f-: 
 
STROLLING ABOUT THE CITY. 
 
 86 
 
 n- 
 
 and the 
 , ; every- 
 
 smoke 
 ;reat deal 
 
 1 as airy 
 mpted to 
 )ec. 
 
 by night 
 k^hile both 
 especially 
 gives you 
 that you 
 V where to 
 Lt you are 
 y anybody 
 ^mediately 
 s baggage 
 lut,— this 
 is last ex- 
 ab started 
 lie compli- 
 the vessel 
 .nions and 
 Istreets and 
 the same 
 ;revvn with 
 •like open- 
 which we 
 hich were 
 and some 
 ect of our 
 taken pas- 
 , from his 
 rench, we 
 them our 
 
 inquiries for the captain and learned that he was probably at the hotel 
 to which our luggage had been taken. There being nothing to do liut 
 to go back again and find the hotel, we started off in no very pleas- 
 ant spirits at the failure of our expedition. Pursuing the way in 
 silence we again reached our stopping place, found the captain, en- 
 gaged our passage, settled all the preliminaries, and were soon 
 washed and sitting down to a spread table with the prospect of a 
 comfortable, hot meal of the best the place afforded, in which we 
 were not deceived. The grossness of describing a meal, unless it 
 is an exceptionally good or an extremely poor one, is so obvious 
 that I will not enter into the details of ours, but, simply using the 
 novelist's phrase say, "we ate as only hungry persons eat with a sa- 
 vory and hot meal before them." I retired quite early after supper 
 being rather tired with the excitement of the previous day's journey, 
 and determined now that we had reached Quebec, and knowing 
 that we had all the next day before us, to make up for my lost 
 quota of sleep of the previous nights. 
 
 Morning came at last and with it I sprang from my bed, quickly 
 dressed myself, and descended to the hotel office. The night had 
 been a good one for sleeping, the air was cool and the temperature 
 outside such that windows could be safely kept open all night. I 
 had slept soundly, not waking once to the best of my recollection 
 till morning ; and now, refreshed and enthusiastic for a day's pleasure 
 trip about the city, I took my cap and went out for a short stroll 
 before breakfast which was not yet ready. 
 
 As a rule I do not believe in introducing borrowed matter into a 
 work that pretends to be an account of original journey and explo- 
 ration narrative, but as occasionally the already published accounts 
 of well known places visited on such excursions, from the pens of 
 accurate writers, are so much more complete than could be given 
 by any passing stranger, that, considering myself such a stranger, 
 and finding in a little volume of guidance through the city of 
 Quebec a most excellent account of the jjrogress and growth of 
 that city, I will venture to give the same in extract, as it may contain 
 matter new to many and interesting to all ; — 
 
 *"' 
 
r,ff\ 
 
 
 86 
 
 HISTORY OF QUEBEC. 
 
 li • !ii:l! 
 
 (f li 
 
 ! I 
 
 \ !| 
 
 ! 
 
 "In 1534 Canada was discovered by Jacques Cartier, of St. Malo, 
 in France. The name is derived from 'Kanata', an Indian word 
 signifying 'a collection of huts.' In 1535 Jacques Cartier made 
 a second voyage and made friends with Uonnacona, the chief of 
 Stadacona, where Quebec now stands. Stadacona is Algonquin, 
 while Tiontirili is Huron, both meaning 'the narrowing of the 
 river ; ' the St. Lawrence beinp; less than a mile wide opposite the 
 city. Jacques Cartier wintered in the river St. Charles and called 
 it St. Croix. His winter quarters were near the present residence 
 of Mr. Park Ringfield. In 1540 he made a third voyage and built 
 a fort at Cape Rouge and also visited Hochelaga, now Montreal. 
 In 1608 Champlain arrived at Stadacona, and landing his followers 
 founded the city of Quebec. No satisfactory explanation can be 
 given of the meaning of the word. The city has been besieged five 
 different times. In 1629 Champlain was obliged to deliver up 
 everything to Sir David Kerkt ; but by the treaty of St. Germain- 
 en- Laye, Canada was restored to France, and Champlain returned 
 as Governor of the colony. In October, 1690, Sir \Villiam Phipps 
 appeared before the city and demanded its surrender of Count de 
 Frontenac who refused ; after a harmless bombardment the English 
 retired. In 1 71 1 another English fleet under Sir Harendon Walker 
 sailed for Quebec, but was nearly destroyed by a storm in the Gulf 
 of St. Lawrence ; xbr these two last deliverances the little church 
 in the lower town was called Notre Dame des Victoires. 
 
 "On the 25th of June, 1759, Admiral Saunders anchored his 
 fleet and transports, with General Wolfe and the English army on 
 board, otf the Island of Orleans, then called Isle de Bacchus. 
 The troops landed on the Island on the following day, near the 
 church of St. Laurent and marched to the west end from which 
 position they could view Quebec ; the French army under Mont- 
 calm, consisting of about 13,000 men, was encamped on the oppo- 
 site shore of Beauport. General Monckton, with four battalions, 
 occupied the heights of Levis from which position he bombarded 
 the city and laid it in rums. General Wolfe then crossed to the 
 mainland, to the east of the river Montmorenci, and on the 31st 
 
HISTORY OF QUEBEC. 
 
 37 
 
 :. Malo, 
 n word 
 r maile 
 ;hief of 
 onquin, 
 of the 
 site the 
 \ called 
 :sidence 
 ,nd built 
 [ontreal. 
 oUowers 
 can be 
 iged five 
 iliver up 
 iermain- 
 returned 
 n Phipps 
 lount de 
 English 
 Walker 
 the Gulf 
 church 
 
 I 
 
 1 
 
 of July attacked the French and was defeated with a loss of 1S2 
 killed, 650 wounded, and 15 missing. After some delay the Eng- 
 lish fleet sailed past the city and on the 14th of Septemlx'r landed 
 his troops at Wolfe's Cove, scaled the famous heights of Abra- 
 ham, met the astonished Montcalm and defeated him. Wolfe died 
 on the field in the moment of victory, and Montcalm, killed also, 
 was buried in the Ursuline Convent. The city was surrendered to 
 the English on tlie i8th of the month and General Murray with 
 6000 men was left as a garrison, — the former was also Governor. 
 The fleet with Wolfe's body sailed for England in October. On 
 the 28th of April the next year a French army of 10,000 men, 
 under De Levis, appeared on the Plains of Abraham and met 
 Murray with 3,000, sickness and death having reduced the number. 
 The English were obliged to retire behind the fortifications of the 
 city, but on the 15th of May, an English fleet under Commodore 
 Saunders, arrived with reinforcements, and compelled the French 
 to retreat. At this time Quebec became an English colony. 
 
 "In 1775 Quebec was again threatened. General Arnold, with 
 a small American army, arrived on the heights of Levis by the 
 Chaudiere valley and on the 14th November landed at Wolfe's 
 Cove ; soon General Montgomery took command ; the attack was 
 unsuccessful though much property was destroyed just outside the 
 town, while Montgomery was killed and Arnold wounded. 
 
 "Li 1837 Quebec suffered a rebellion within its own walls. The 
 militia were called out and the city placed under military rule but 
 nothing of consequence occurred. One night, however, there was 
 heard a loud ringing of befls, and it was said that the rebels had 
 risen and would sack the place. The cause of all this alarm 
 was, nevertheless, very simple-— the singeing of a pig in the Hotel 
 Dieu Nunnery Yard. In the following year Messrs. Teller and 
 Dodge, two American sympathizers imprisoned with three others in 
 the Citadel, escaped ; four of them let themselves down from the 
 flagstaff" bastion, and Teller and Dodge passing the city gates 
 reached the U. S. 
 
?/ 
 
 w 
 
 I 
 
 I' I 
 
 1,1 ; 
 
 38 
 
 HISTORY OF QUEBEC. 
 
 
 1 IM I 
 
 "In 1832 and '34 Quebec was visited by the Asiatic Cholera. 
 In 1834 the Castle St. Louis was destroyed by fire. On May 28, 
 1845, the whole of St. Roche was burnt ; on June 28, the suburbs 
 of St. John and St. Louis were also burnt ; the loss by these two 
 fires was over ^2,000,000, the insurance and subscriptions of aid 
 amounting to $900,00 ^. In 1846, in the month of June, the Thea- 
 tre, formerly the Riding school attached to the Castle of St. Louis, 
 was destroyed by fire during a performance when the building was 
 crowded, — fifty-five persons lost their lives. In 1853 the Parlia- 
 ment houses were burnt down, when a large library and museum 
 were destroyed. The sittings of the House were transferred to 
 the Church of the Grey Sisters near Gallows Hill (which had not 
 then been consecrated) ; this also burnt down, when the sittings 
 were held in Music Hall in Louis Street. In 1867 the Province of 
 Quebec was granted a colonial government, with the seat of the 
 Province at Quebec." 
 
 I 1 
 
 }i 
 
 : ''1, 
 
 :l-.l 
 
i 1 
 1 1 
 
 SIGHT-SEEING. 
 
 89 
 
 Cholera. 
 May 28, 
 : suburbs 
 ;hese two 
 ns of aid 
 he Thea- 
 3t. Louis, 
 Iding was 
 le Parlia- 
 [ museum 
 iferred to 
 I had not 
 ,e sittings 
 rovince of 
 ;at of the 
 
 CHAPTER III. 
 
 Our stay in Quel)cc — Starting for Berthier — Rerthier — Ofl" for Labrador- 
 Bunking in — Island of Orleans — Islands and Channels — Sunday ■ 
 Other Islands — -The Saguenay — Fog again. 
 
 I WILL not attempt to give a description of my feelings at being in 
 this ancient city, nor yet will I give an account, which would be 
 much inferior to many former similar ones, of the sights I saw and 
 the impressions they left upon me ; all this has been done so often by 
 others that the charm of novelty would be lost, and it would prove 
 only a waste of time and words thus to attempt to describe, from 
 a stay of only a few days, that which would need weeks to see 
 properly. I will therefore hasten to say, that the brief stay here 
 was spent in the continual pleasure of evening rambles about town, 
 combined with the necessity and in truth unpleasantness of mak- 
 ing daily purchases for a year's sojourn in a country about which I 
 knew very little ; yet the pleasure, counterbalancing this unpleas- 
 antness, so far exceeded it, that I remember the former while forget- 
 ting the latter. The walks about the Terrace, of a calm, clear 
 evening, both before and after the lamps were lighted, displayed 
 the lower city and harbor — the former with its rows of roofs, for 
 they were the only parts of the houses visible, and the latter with its 
 countless masts and water-vehicles from the boat to the man-of-war, 
 of which three lay swinging at the end of their cables just near the 
 channel over against the opposite shore, the whole scene present- 
 ing a silent witness to the industry of the day ; then the view of 
 Point Levis from the walls of the upper city ; and a trip around the 
 city itself, — all these are pleasures to be remembered ; and I shall 
 
 «i 
 
 ,^i 
 
 ■■■t 
 
i-TTT- 
 
 ill 
 
 i'i 
 
 1 
 
 ll 
 
 11 
 
 Si 
 
 1 
 
 ! 1 
 
 
 i 
 
 
 ! 
 { 
 
 i 
 I. 
 
 1 
 
 [l 
 
 \ 
 
 u 
 
 V 
 
 'jl! 
 
 \ ii I 
 ill 
 
 40 
 
 PREPARING FOR DEPARTURE. 
 
 remember them with a fresh delight whenever I recall the circum- 
 stances that caused me to start upon this Labrador trip, for the 
 invigorating air of new scenes and a new climate. 
 
 Friday, in the evening, we started from the . Dtel, with our bag- 
 gage, for the vessel. Arriving there we deposited our effects on 
 the whorf, whence they were speedily transported to the deck; 
 then the trunks were taken to the hold and the bags and boxes to the 
 cabin of the vessel, and we stationed ourselves to wait for the captain, 
 as we expected him to start that night. After waiting several hours, 
 and some after dark and consequently past supper time, — which 
 fact I mention since our dinner had been taken early and it was now 
 fast getting late, — he arrived in a great hurry, and then we learned 
 that the vessel must wait until high tide (it was then low and we 
 could see the muddy bottom near the vessel), which would not be 
 until about three next morning, before leaving ; the ladies of the 
 party not caring to sleep in so exposed a situation, the captain 
 kindly invited us to the hotel, where we all went with him, partook 
 of an oyster supper, and passed the night quite comfortably. The 
 vessel started in the morning for Berthier, where we were to meet 
 her the next day noon by a little steamer that runs from Quebec 
 and makes that place one of its stopping stations. The night's rest 
 had refreshed us and the next morning, quite disconsolately, we 
 wandered about the streets of the lower city, and the wharf whence 
 the steamer was to leave, waiting for the time of departure, which 
 came slowly. It was market-day, and the streets, especially the 
 square about the market, were crowded with all sorts and nationali- 
 ties of people buying their week's provisions ; we watched them for 
 a long time, and were especially amused in noticing the people 
 who failed to catch the ferry boat that landed not far from us. 
 Some ran for it and, leaping, caught it, but others running for it 
 failed to catch it ; and as the boats alternated with each other every 
 fifteen minutes, there was always some one in a state of frenzy 
 with the retreating ferry for a fault for which he alone was to blame ; 
 there were several old women with large baskets, evidently returning 
 home from market ; two or three priests ; and a number of other peo- 
 
BERTH I ER. 
 
 41 
 
 pie who were succcssivtly left by the retreating boat, — some exj^ress- 
 ing their indignation while others, like ourselves, contented themselves 
 to wait their turn. At last, twelve o'clock, and our captain, came ; 
 and the steamer started. We slowly passed along the channel, 
 watching the scenes on either bank, and, leaving the city with its 
 confusion and the harbor with its abundance of crafts of all kinds, 
 we steamed into pure waters, clearer atmosphere, and the rocky 
 borders of the mighty St. Lawrence. On our left we presently 
 passed, if I remember rightly, the famous falls of Montmorency — 
 at least we saw a precipitous mass of dashing, struggling waters, 
 that looked the mighty cascade that it probably was — while a short 
 turn soon brought us to the Berthier wharf. Our vessel was at the 
 wharf receiving her last cargo, in the shape of potatoes and sev- 
 eral kinds of fresh vegetables, for her voyage, and as it would take 
 several hours to complete the loading we accepted the captain's 
 kind invitation to visit his home and dine there with him. Berthier 
 is a small P'rench Canadian village, situated on the southern side of 
 the St. Lawrence, and very nearly opposite the eastern extremity 
 of the island of Orleans ; it is about fifteen miles from Quebec. 
 The long wharf built for the accommodation of freight and coal, 
 extending far into the water, had several vessels lying at each side, 
 either being or waiting to be loaded, and though the coal dust 
 was everywhere under foot, and in the air flying in our faces, we 
 forced ourselves through it and soon reached a cleaner footing 
 and much purer air. A walk up a rather long slightly sloping 
 hill brought us to the principal street, along which low roofed, yet 
 small and cosey looking houses, for the most part clean with 
 white paint, not yet ugly from exposure, with their correspond- 
 ing barns — mostly unpainted — extended on either side where 
 the open fields on the right near the water, and high hillocks of 
 granite on the left, back of the houses, had not establish.' " prior 
 claim. As neat a house as appeared, on this apparently one 
 streeted township of about fifty houses — if the town was really no 
 larger than it seemed — was that of our Captain, and we 
 passed a quiet time both before and after a hearty, homely meal, 
 
1 
 
 ill 
 
 49 
 
 IJUNKING IN. 
 
 U 
 
 .j;; . 1 
 
 »!? 
 
 i 
 
 
 Mi 
 
 until about dusk, when we started for the wharf again taking a cut 
 across the meadows in the direction of the water, having seen very 
 little of the town itseli', though enough to commend it as an old 
 fashioned, curious place, well worth visiting and investigating. 
 Soon the wharf was gained ; a boat waiting bore us to the vessel, 
 which by this time had finished her loading and was lying at anchor 
 out in the channel, and with a dim twilight at our backs, a lip' 
 house in front of us, and a bonny breeze to shape our course, 
 hoisted sail and bade farewell to the last landing this side of "The 
 Labrador," whither we shaped our course. We were at last really 
 moving. The vessel was really gliding along under the pressure of 
 the wind towards that region so prominent on the eastern part of 
 northern North America, yet so little known, called the plateau of 
 Labrador, or, as the people themselves call it, as I have above 
 quoted, "The Labrador." 
 
 It is by no means an easy matter to arrange four persons, let us 
 say for example two men and two ladies — though it makes very little 
 difference as to the number and sex when even a single person is, so 
 to speak, unceremoniously deposited in a small cabin of a sm- 
 sailing vessel, and shown a small bunk in which to sleep, scar* 
 big enough for one yet the usual abode of two al>/e bodied ( ?) sea- 
 men — in such contracted compartments as those we were about to oc- 
 cupy : how the affair was brought to a happy termination I cannot 
 tell. Our voyage would last about a week, our accommodations 
 for that time multiim in parvo, and with a more literal meaning 
 than we had ever before imagined that the words could possibly con- 
 vey. Our trunks were expelled from the cabin and confined to the 
 hold ; our bags were insufficient at best to meet our wants ; the mat- 
 ter was thus ludicrous as well as provokingly uncomfortable and in- 
 convenient. While we were enjoying ourselves on the outside of 
 the cabin, watching the stars, the dim outlines of the shores in the 
 darkness around, and the darker yet sparkling water — sparkling 
 from the reflection of the stars — the ladies were somehow prepar- 
 ing affairs below ; soon they joined us, and we together sat watching, 
 we could hardly tell what (the custom usually seems to be when 
 
ISLAND OF ORLEANS. 
 
 48 
 
 starting on a sea trip, to spend the first evening on deck), until the 
 lateness and chilliness of the hour reminded us of our berths below, 
 and descending the narrow stairs or steps of the cabin we sought 
 them and were soon wrapped in slumber in spite of the uncom- 
 fortableness of our contracted and narrow box-beds. 
 
 I will not describe either the ship or the accommodations provided 
 for us while we traversed the St. Lawrence to our destination ; the lat- 
 ter would be too personal a matter, — though it was the best the place 
 afforded, — and the former, hardly diffe -ing from the thousand water 
 conveyances, of similar shape and size, too trivial and uninterest- 
 ing to be worth mention here. Four of us, who were together bound 
 for the same place, made the best and freest use of both ship and 
 accommodations, while the sociality usual on shipboard prevailed 
 with gTeat harmony. In so small a vessel we could not but take an 
 interest in all that occurred, and in all the crew, and they, no doubt, 
 took full notice of all we said or did ; while the wonder expressed 
 in their faces whenever I at least met them seemed plainly to say, 
 why are you going to such a place as I-abrador? to which mute 
 question I ^reserved the probable and veracious answer — I hardly 
 know myself, why. Now that we were fully started upon our 
 journey, the chart, the directions for sailing, and what we did and 
 saw occupied our chief attention, and as we pursued our way we 
 studied these diligently. 
 
 In going to Berthier by the steamer we had passed the Island of 
 Orleans, that curious, oval island only three miles outside of Quebec 
 that so nearly fills the river at this point. It lies like an egg in the 
 very centre of the stream with only a narrow pass on either side, and 
 while its length is full eighteen miles, its width is scarcely five. The 
 north shore being rather flat and muddy with more or less rocky out- 
 line, the south shore with its sandy beach and few rocky points pre- 
 sents the best and most used channel for vessels going to and from 
 Quebec. From a central elevation of some three hundred and 
 seventy-five feet the land slopes to the rather steep banks around it. 
 We did not see much of the beautiful gardens and places which are 
 said to occupy the southern slope of the island, neither did we see 
 
 1 
 
 I 
 
 J.lii 
 
 f 
 
 I ' 
 
 
'ill : 
 
 ^nei 
 
 44 
 
 CHANNELS AND ISLANDS. 
 
 I, 
 ir'' 
 
 in 
 
 1 
 
 ■ill 
 ii 
 
 ! !! 
 
 1 1 1 ' 
 
 I ' . 
 
 the little churches of St. Lawrence and St. John though standing 
 near the shore, since we had no one to point them out to us. The 
 two lighthouses, that of St. John village with its white revolving 
 flash every half-minute, which is visible ten miles, and that of St. 
 Lawrence village with its fixed white light visible eight miles, would 
 also have escaped our notice, as we passed them in the daytime and 
 not the night, but that the white towers seen from the wharf attracted 
 our attention. On leaving Berthier a small fixed light on a little 
 island, known as Belle Chasse Island, guided us towards the entrance 
 of that part of the channel known as the Middle Traverse. Leaving 
 the North Channel, separated from our route by a shoal of rocks 
 and reefs on the north, and the South Traverse, similarly separated 
 on the south, we pursued our way, by this time guided by the fixed 
 light on the western end of Crane Island, which with Goose Island, 
 forms a very narrow, meadowy, and muddy strip of lantl, some fif- 
 teen miles long between us and South Traverse, until we had safely 
 passed the dangerous shoal in the centre of the Traverse ; then, 
 coming into clear open water, we saw the light boats of St. Roque 
 on our right, and Isle aux Coudres on our left as we sailed easily 
 and pleasantly along the now safe passage before us. Soon the 
 captain with the chart showed us our position, and a hasty glance 
 as we passed this litde island, about twice as long as it is broad (5^ 
 miles by 2^), lying snugly tucked away in a baylike enlargement 
 of the north shore, revealed but a glimpse of Notre Dame Church, 
 steeple as we passed to the open water beyond. Here we came 
 for the first time upon the open river, a distance of about eleven 
 miles across from Mai Bay (which is ninety miles from Quebec and 
 noted for its salt-water bathing and trout fishing, the sports of its 
 summer visitors, who are often quite numerous) , with its bold, rocky 
 point on which a light is situated while the shoals are only a quar- 
 ter of a mile from shore on the north, to Riviore Quelle, with its 
 summer pleasure grounds and hotel only ninety-two miles from 
 Quebec on the south bank of the St. Lawrence ; only a little below 
 which point, this hilly region of earti ^.lakes bears the name of Les 
 Eboulements. But it is now Sunday, and the ladies have fixed their 
 
 
 ^ 
 
SUNDAY ON THE SEA. 
 
 46 
 
 abodes, and with fair winds and a good run we compose ourselves 
 to the first real rest that we have had since leaving home eight 
 (lays before. We read and doze in the brigiit, warm sun like dogs, 
 or cats, or insects, insensible to all save the inspiration of rest and 
 the enjoyment of ease. 
 
 At sea on board a vessel there are two counteracting influences 
 at work to render Sunday — especially if it come as ours did on 
 the next day after setting sail — either a day of rest or a day of un- 
 rest. In the latter case, to one who would not attend any place of 
 divine worship, even were he at home and on shore, the desire to 
 be quiet is drowned by a deluge of worldly activity that waits and 
 longs for the morrow that his conscience may be easy in pursuing 
 his own pleasures, and so the hours pass heavily, and " the longest 
 day I have known for a perfect age," as it is often aptly styled, comes 
 to its close, with no refreshment to the mind, and thus a con- 
 tinued increase rather than decrease of the bodily infirmities ; since 
 the real health of the body is, in a certain measure and to a greater 
 extent than perhaps we imagine, dependent upon that refreshment 
 of mind, from a spiritual source. In the former case, it is differ- 
 ent. A day of rest means, in every sense of the v. ord a day of 
 rest, mentally and spiritually, hence, more or less directly, bodily. 
 While many thousands enjoyed this rest in attending church at their 
 homes this beautiful Sabbath day, we enjoyed it in indulging our 
 own thoughts of rest, quiet, and all good things, while bathing in a 
 flood of bright, warm sunlight ; while watching the almost calm 
 surface of the now sun-reflecting water ; and while reading or con- 
 versing with each other on pleasant topics, yet in no mere sensual 
 manner. 
 
 Night came soon enough, and if the day had been quiet and 
 peaceful, the evening, with its moon full at yesternight, was much 
 more so. After an almost perfect clay we retired to rest. The 
 clear sky remained throughout but part of the night and then gave 
 place to clouds ; the calm, warm, almost hot weather turned them 
 to mist, fog and then rain ; the west wind to east southeast. 
 
 We awoke Monday morning with an uncomfortable feeUng 
 
 n 
 
 71 
 1 
 
 
 lil 
 
 
 
 1 ' 
 
 
 
 't 
 
 J! 
 
 !^ 
 
 ft 
 
 \m 
 
 ■i 
 
 
 ul 
 
 i 
 
w 
 
 yr rw i^gi*!*'" 
 
 
 h 
 
 ;i 
 
 46 
 
 THE NEIGHBORHOOD OF GREEN ISLAND. 
 
 'I 
 
 ■■ill 
 
 m 
 
 If' 
 
 of dampness all about us, of chilliness within us, and an atmos- 
 phere of thick, sticky saltwater — and especially St. Lawrence salt- 
 water noted for its nature as described — above, around, and on all 
 sides, enveloping the vessel and shutting out earth and sky and 
 water. We were in the miserable arms of a light attempt at New- 
 foundland fog, to us a most admirable display of those qualities so 
 inherent to the atmosphere about and around all parts of the 
 Gulf of St. Lawrence. If this was a taste of what was to come, 
 no wonder that they were noted the world over ; and one could 
 readily believe the old saying that " the inhabitant.-; of N""^ound- 
 land use their fog for cloth, cut it with scissors, and n^ e shirts 
 and breeches of it." The vessel, of course, was obliged to anchor. 
 Once during the day we tried to proceed, but, unable to do so, we 
 anchored again, this time off Green Island, to which place several 
 of rs went to examine the hunting, while the crew searched for 
 dry wood — a hard thing to find — to burn. We had passed, up to 
 this time, several noticeable places both on shore and in the shape 
 of islands close to us ; and as it is well to know the important items 
 of interest connected with places thus visited or passed by the tour- 
 ist, let us see what has escaped us. On our right, directly opposite 
 Mai Bay, the group of Kamouraska islands just hides the village 
 (with its little church, hotel, and other buildings) by the same name. 
 As a pleasure resort it is noted in Quebec and the neighboring 
 regions for its fresh, country-like appearance, as are most of 
 the villages on the south shore of the river, and its excellent bath- 
 ing. A little farther on the same side, the Pilgrims — whence the 
 name I cannot tell — another smaller group (five small islands), 
 about six miles further eastward, are quite interesting in a scientific 
 point of view. They are quite near the land and connected by 
 reefs dry at low water. Though small at best, the largest of these 
 islands, nearly two hundred and ninety feet high, is well wooded. 
 Still but a short distance farther east are the three small, steep 
 islands called the Brandy Pots. The most northern and largest, 
 being half the height of the largest of the Pilgrim group, and also 
 partially wooded, is often visited at the fine spring of water on its 
 
CACOUNA. 
 
 47 
 
 southwestern extremity, which spring dries only in the warmest 
 weather. About and around nearly all of these islands reefs and 
 rocks are everywhere visible. I am not confident that we passed 
 through the narrows between the Pilgrims and Hare Island 
 Bank, but think that we did ; at any rate, when about opposite 
 Riviere du Loup we had Hare Island on our left. The two oblong 
 islands called Hare and White, the latter just east of the former and 
 very much smaller, with their reefs, which are quite extensive, are 
 about twenty miles long ; of their nature I was unable to judge. I was 
 much interested in spying the distant horizon of Riviere du Loup, 
 formerly so much frequented by the fashionable of Canada — as 
 well as elsewhere — a village of great attractions I am told ; while 
 not far from it Cacouna, hardly a hundred miles from Quebec, the 
 now favorite resort of the people of Canada, with its many elegant 
 establishments and cosey summer residences graces this part of the 
 coast. Many sportsmen take pleasant fishing trips and frequent 
 baths in the chosen resorts and not cold waters about here ; while 
 the temperature, they tell me, is delightful. It is not long — not 
 twenty years ago — since this place was of comparatively little ac- 
 count and notice, with only a rocky peninsula some four hundred 
 feet high to commend it to the scientist, hardly the tourist. It is 
 now a fashionable resort. 
 
 We had hesitated for some time as to whether it were best to an- 
 chor where we at last did, or by Red Island just on our left, named 
 from the color of its soil and rocks, which is a low, flat island of 
 little importance, when we decided in favor of the former place. 
 At any rate I remembered that we were opposite the noted and far 
 famed Saguenay river, the largest on this part of the coast, rising 
 in lake St. John, about ninety miles directly west from the river's 
 mouth. A place of so much interest and importance needs better 
 words than I can give, having never even seen it, so I will copy from 
 another author, though it is not my purpose to give you other 
 peoples' adventures and descriptions as a rule. 
 
 In describing the mouth of the river, which is full of shoals and 
 reefs, he says : "Saguenay river has an entrance between Vaches 
 
 ii 
 
 w 
 
 fit: I 
 
 il t 
 
 i n 
 
 
f.-lipi 
 
 I 
 
 48 
 
 SAGUENAY RIVER. 
 
 ^I'^-ii 
 
 
 .lii! 
 
 : i.iii. 
 
 point on the northeast, and Lark point composed of low clay cliffs 
 on the southwestern side, fron. each of which dangerous reefs pro- 
 ject into the St. Lawrence. I'hese reefs leave an entrance into the 
 Saguenay only three-quarters of a mile wide, though nowhere less 
 than ten or eleven fathoms deep." 
 
 He then goes on to say, "That this extraordinary river, which was 
 imperfectly known until the late sur\'eys, is as remarkable for the great 
 volume of water which it brings down to the St. Lawrence, as for 
 the enormous depth of its bed, which is fully one hundred fathoms 
 lower than that of the St. Lawrence. It comes from the lake St. 
 John, and at Chicoutimi, a trading post of the Hudson's Bay Com- 
 pany, which is sixty-five miles above its mouth, it becomes naviga- 
 ble, and six miles above which, to the rapids, the tide ascends. To 
 point Roches, fifty-seven miles from the St. Lawrence, and eight 
 miles below Chicoutimi, it is navigable for the largest ships, and up 
 to this part there is no danger in the river ; the shores consist of 
 steep precipices, some of the headlands rising more than a thou- 
 sand feet in height. 
 
 "The current runs down with great force, the ebb tide varying from 
 three to five knots according to the breadth of the river, which is 
 from two-thirds of a mile to two miles. Tadousac, which is on 
 the eastern shore, is about one and one-half miles, within the 
 entrance of the river, and was formerly the principal post of the 
 French for trading with the Indians. It now belongs to the Hud- 
 son's Bay Company. The harbor is abreast the settlement, and is 
 well sheltered :" but for the scientific value of the facts it would 
 be needless to add that "a heavy anchor should be cast close in 
 shore on account of the eddies which sometimes set into it from 
 the river. 
 
 "Fronting the mouth of the river there is a kind of bar upon 
 which are twelve, twenty, and twenty-eight fathoms, but immedi- 
 ately within the depth increases to above one hundred, and a little 
 farther up to one hundred and fifty fathoms. The current setting 
 strongly over the bar, meeting with the spring ebbs of the St. Law- 
 rence, causes breaking and whirling eddies and ripplings ; and these 
 
THE SAGUENAY RIVER. 
 
 49 
 
 
 Streams opposed to a heavy easterly gale cause an exceedingly high, 
 cross, and breaking sea in which no boat could live. On the flood 
 at such times, there is no more sea than in other parts of the river. 
 A fixed white light is exhibited from Lark islet ; the tower is a 
 square white building ; the light is thirty-five feet above high water, 
 and visible ten miles. Two range lights (fixed white) are shown 
 on the western side of the entrance to the river ; one eighty-two 
 feet above high water, on point Noir, distant one and a half miles 
 from Lark islet lighthouse ; the other one hundred and seventeen 
 feet above the sea and distant from it about six hundred yards, and 
 are for the purpose of leading vessels clear of Prince shoal, Bar 
 reef, and Vaches patch, and visible nine miles." A curious fact, be- 
 sides all this nautical description above quoted, appears from the 
 description of another writer who says : "At spring tides a large body 
 of water passes over the Chicoutimi shoals (at a very rapid rate du- 
 ring ebb tides) , and falling suddenly into deep water seems to strike 
 downward at once, leaving but a slight current on the surface." 
 The Chicoutimi river enters the Saguenay about sixty-five miles from 
 the mouth of the l-.tter. P'urther on he adds : "Ihe strong flood 
 tides over the bar, at the entrance of Saguenay river, falling s. ddenly 
 into deep water, may also contribute to a certain extent to check 
 the strength of the surface-current of the river." A singular fact 
 appears that at Tadousac was built the first church in Canada. I 
 understand that it is still standing. 
 
 I have not forgotten that it is Monday morning, a damp, foggy 
 day — for which reason we find ourselves anchored near Green isl- 
 and, by whose reefs it is most dangerous to pass except in clear 
 weather ; thus it happens that we take the boat and are rowed to 
 the island on our first gunning expedition. We find the island low 
 and rocky, with occasional sand patches on the southern shore : on 
 the east some half a mile long, narrow rocky points, almost reefs, 
 are above high tide, while rocks and shoals are distributed quite 
 abundantly all around the island. Not three miles north and a 
 little east of Green is Apple (or Pommc as the French call it) island ; 
 between the two is a reef of slate rock, visible at low tide. The 
 
 t : 
 
 i '. t 
 i (1 
 
 n 
 
 it 
 
'^^w 
 
 50 
 
 BLOWING THE FOG HORN. 
 
 ! 
 
 
 i i 
 
 ^ ' •■ .. 
 
 I 
 
 )^, 
 
 rl' 
 
 •ii 
 
 * 
 
 passage, at high tide, between these two islands, is but very shallow 
 at best, and not fit for vessels to go between. From a square white 
 tower on the northern part of the island, a fixed light, sixty feet 
 above the sea, and visible fourteen miles, shines by night. We heard 
 the half hour gun, from this same quarter, during the fog — it is 
 also fired during dense snowstorms — all the time we were anchored ; 
 and far into the night its heavy and loud boom echoed to our ears 
 with a dull thundering roar. Our hunting, not as successful as it 
 might have been, brought several species of birds to our notice, but 
 our wet clothes called for more attention than our birds, for the 
 time being, while we dried ourselves as well as we could under 
 the circumstances. The next morning as the fog cleared away the 
 clink, clink, clinkety clink, of the anchor chain, as all hands 
 heaved at the patent windlass, sounded merrily (it was about six 
 o'clock) on the otherwise quite still air, and before long we were 
 dashing along with a breeze that had by this time nearly cleared the 
 air of fog, though it was soon on us again as thick as ever ; but we 
 were past the dangerous shoals and in free water, so we kept on 
 our course, and let the thick fog come on again. Little we cared for 
 it although we kept the old tin fog horn, with its toot-toot-toot, 
 and toctety toot, going all the morning. A vessel's fog horn is an 
 old fashioned institution, and consists of a tin horn similar to that 
 used by venders of fish, yeast, and other articles of street com- 
 merce ; or by the noisy college student in his rows between classes 
 or his midnight music, of horrid notoriety, at home. The main dif- 
 ference is that the tin is unpainted or unvarnished, and the whole 
 horn shorter and less clumsy. Any one can blow it, and on this 
 particular morning, any one did. It lay upon the deck of the cabin, 
 and while it was the special duty of the man at the wheel to "tend 
 it," as is sometimes r.aid, everybody that came around had his or 
 her turn at it. It acts as a warning to other craft that we are 
 around and coming, while all vessels use one in a dense fog. We 
 had started from Quebec with a consort, in the shape of another 
 vessel, supposed to be a slower sailer than ours, of similar size and 
 lading, bound for the same ports with us. In fact one captain 
 
OUT-SAILED BY OUR CONSORT. 
 
 51 
 
 owned both, and was on a trading expedition down along the coast. 
 In the thick fog we had lost her ; supposing her to be still behind us 
 the continued sounding of the fog horn, more often than otherwise 
 necessary, was to let her know where we were. Imagine our surprise 
 when, on the lifting of the fog for a few moments, we just saw the 
 vessel in the dim distance ahead, and not behind us ; she had got 
 ahead, and kept so, for the rest of the voyage, outsailing us ; we 
 could not and did not catch her at all until near our second stop- 
 ping place at Bonne Esperance, nearly eight hundred miles from 
 Quebec. 
 
 
 MM 
 
 \ \ 
 
 h:i 
 
.till .'III 
 
 62 
 
 WEATHER — WHITE PORPOISES. 
 
 n m y 
 
 t.f 1 1 
 
 if! 
 
 ii, 
 
 
 ' :!|it!t, 
 
 ::m 
 
 CHAPTER IV. 
 
 The weather — Beluga borenlis and other huge Animals — A Sundog — 
 Uirds — The Rusty Blackbird — Uradore hills — Varieties of rocks — 
 Coast line about .St. Augustine — Reaching Bonne Esperance — Old 
 Fort island — Eider duck — Other birds — Garden vegetables — Hay — 
 Raised beaches — Labrador dogs — Searching for drift-wood. 
 
 Wednesday, September 22. The weather to-day was cold and 
 chilly, but clear ; the wind blew strong west-southwest. I made a 
 reckoning of both thermometer and barometer, and found that 
 throughout the day they ranged as follows: at 8 a. m., then 51, 
 bar. 28.45; 'It 12 M., ther. 57, bar. 28.50; at 4 p. m., ther. 51, 
 bar. 28.57 ; and at 6 p. m., ther. 50, bar. 28.52. This is an aver- 
 age pleasant day in this locality. We are now about opposite 
 Point de Monts, and swiftly approaching the Gulf. With a fair 
 breeze the vessel flew along the water sending clouds of spray 
 way over the side of the prow. Although there was quite a breeze 
 the surface of the water a little way from the vessel was quite even. 
 We saw plenty of porpoises of both the black and white varieties. 
 The latter are very abundant in the waters of this region, and 
 about this part of the coast. From the river Saguenay to the west 
 end of Anticosti, they abound more or less frequently at all seasons 
 in open water. This is the Beluga borealis of Lesson, and accord- 
 ing to Professor Tenney, "A specimen of this animal about ten 
 feet long, and weighing about seven hundred pounds, was kept in 
 a tank in the Aquarial Gardens, Boston, for about two years. 
 He was quite docile, learned to recognize his keeper, and would 
 come and take food from his hand. He was trained to the 
 harness, and drew a young lady in a car prepared for the purpose." 
 Occasionally a large whale would be seen in the distance spouting 
 
GRAMPUS, ETC.— A SUNDOG. 
 
 68 
 
 high columns of water, but none came very near to us. The 
 captain informed me that at times they did appear close by the 
 side of the vessel. I'he Grampus is also occasional here, but they 
 are much more abundant farther north. At one time during the 
 day several horse mackerel appeared by the side of the vessel, 
 and amused us for a time as they kept up with her, swimming 
 abreast and in perfect line with each other. They would dash 
 through the spray, appearing and disappearing with surprising 
 readiness ; but we soon lost sight of them. The captain told us 
 that about this part of the coast he was almost always followed by 
 a party of four or five of these huge fish. As the strength of the 
 wind was spent, I stood looking over the prow of the ship, watch- 
 ing the most perfect rainbows formed in the water by the shining 
 of the sun through the spray as it dashed from the sides of the ves- 
 sel. Occasionally I thought that I could detect a second bow be- 
 hind the other (ahead of it I mean), but it might have been the 
 optical delusion caused by my straining my eyes by too much and 
 too close looking. Of course nearly everybody on board is to-day 
 suffering from the usual inal de mer accompanying sea voyages, 
 as too many know at their own cost. 
 
 Thursday the 23rd. The fine breeze of yesterday kept up all 
 night, and brought us a long distance on our way ; in the morning 
 the breeze calmed down, however, and left us a cool and balmy 
 though cloudy day, with only occasional glimpses of sunshine. The 
 captain foretells stormy weather from the appearance of a sundog 
 this morning. These solar appearances are supposed to be caused 
 by the presence in the air of minute six-sided ice crystals, which 
 refract and decompose the rays of light passing through them ; 
 they are seen usually at a visual angle of 22°, and are quite fre- 
 quent in the polar regions of cold c»untries generally. 
 
 Friday the 24th. This was another such a day as yesterday. 
 The morning finds us not far from the island of Great Mecattina, 
 with a fair breeze pressing us onward. In the morning one of the 
 species called pigeon hawk {Hypotrionhis columbarius) flew about 
 the vessel several times, and at last lit upon some portion of tlie 
 
 ft 
 
 
 W 
 
 \ . 
 
 i 1; 
 
 iiUBl 
 
 
 r; 
 
 i 
 
ff7'-f 
 
 54 
 
 PIGEON HAWKS — THE RUSTY BLACKBIRD. 
 
 11 
 
 I ' 
 
 It 
 
 1 1 
 ■ t 
 
 mast. These birds are quite common here all along the coast, 
 and it is not at all rare to see them accompany a vessel, now 
 lighting upon it and now again making a short flight seaward, for a 
 distance of many miles. I also saw a small owl, but could not tell 
 the species — unless perchance it was our common Sco/>s asio — 
 that acted in much the same way, alighting frequently upon the 
 masthead. About noon a rusty blackbird {Scolecophagus ferrii- 
 gineus) alighted upon the deck of the vessel and remained with us 
 some time. As we were very near the land, and somewhat of a 
 fog existed at the time, he might have mistaken us for some portion 
 of land. I found afterwards that this bird had a summer breed- 
 ing range all along the coast here as far as L'Anse au Loup, at which 
 latter place a resident, Mr. Fred. Davis, informed me that the bird 
 occasionally built its nest in his woodpile — the people there are 
 obliged to cut enough wood at one time to last the year around ; 
 thus there is alwa^'^j more or less of a pile about in the summer 
 season — and his boys called it quite common there. This, I 
 believe, is the only species of blackbird that regularly remains so 
 far north to breed ; the cow and red winged blackbird, and the 
 purple grakle, though extending quite far north and east, being 
 hardly more than of occasional occurrence. The rusty blackbird, 
 as you remember, is generally regarded as an unsocial and retiring 
 bird ; here it is the reverse, and its nest is not unlike that of a small 
 robin with many sticks outside, and its eggs about 3 or 4, bluish 
 white with spots and dashes of light brown. It feeds upon seeds 
 of various plants and a few insects. Strange to say, they are here 
 frequently kept as cage birds, and their cunning, and power of 
 mimicry of song, something rather remarkable. I have seen it in 
 confinement, and found it to keep admirably. At evening we 
 anchored in calm water at the mouth of the large rigoulette, not 
 far from St. Augustine, and had the satisfaction of passing the 
 night, at least, in calm water : there was really some satisfaction 
 in sleeping off the confusion of the four previous days of rough 
 weather. 
 
 The coast all about these partu presents the same rocky aspect. 
 
 / 
 
 / 
 
 i 
 
:ct. 
 
 NATURE OF THE COAST. 
 
 85 
 
 I have visited nearly all the important harbors from Natashquan to 
 Blanc Sablon, and find the same general appearance in the surface 
 geology, and a similar rocky contour in every place thus visited. 
 Bold masses of rock rear themselves as hills from five to seven 
 hundred feet in height ; except the three Bradore hills, which 
 are here called "the mountains," and attain the height, as marked 
 upon the charts, of 1264 feet for the highest, and 1135 for the 
 second, while the third is of nearly the same size as the second. 
 It is rare that the coast line itself presents, anywhere near the water, 
 a spot large enough upon which to build a house and have the 
 foundations rest upon an upraised seabeach, or any kind of earth. 
 Some of the houses are built on part rock and part earthy ])atches, 
 while the majority are built directly upon a rocky layer which some- 
 times appears within a few inches of the surface. 
 
 The rocks here present the same general character of coarse 
 granite or gneiss, that is formed chiefly of feldspar in great excess. 
 Occasionally, especially the farther down the coast one goes, the 
 mica is in excess, and several localities give an abundance of good 
 sheet mica that might almost be worked with profit were it not for 
 the distance that it must be transported over land, or rather over 
 rocks. The rock is for the most part syenitic gneiss ; that is, 
 hornblende takes the place of mica, and the feldspar, which is usually 
 orthoclase, as far as I was able to ascertain, is of both white and 
 flesh-colored varieties. The hornblende is of a greenish-black color, 
 and often present in large crystals, but so embedded in the mass of 
 surrounding rock, as, with the fact of its extreme brittleness, to 
 render it impossible or nearlj- so to extract them. I did not visit 
 the Bradore hills, but they are said to be of gneiss of this same 
 general character. Quartz does not seem to be abundant here. 
 I have several times had an elderly resident of the coast speak to 
 me of a "vein of marble" running through the rock not far from 
 his house, but upon examination had it proven, as I had previously 
 anticipated it to be, a vein of quite poor quartz. It is not my 
 purpose, however, to give here a general dissertation upon the 
 
 i 
 
 ,k 
 
 r 
 
 !■' 
 
 ; I 
 
 I.; \ 
 
 !ii 
 
TT- 
 
 56 
 
 ST. AUGUSTINE AND VICINITV'. 
 
 i 
 
 ill' 
 .1*11 fi 
 
 II 
 
 ! ' m 
 
 M 
 
 geology of this region, — that must be left to those better qualified 
 to do it ; a few general remarks may be of value however. 
 
 The waters about St. Augustine, as all along the coast to a 
 greater or less extent, are crowded with large and small islands, 
 that have been severed from the coast at some remote period, and 
 which now present narrow, winding, and more or less dangerous 
 passages for vessels or small boats, according to their depth. At 
 this point, and for a distance of some twenty or thirty miles up and 
 down the coast, is a most interesting and remarkable feature. It 
 seems as if the whole coast region had received a lateral and per- 
 pendicular pressure, that pressed these islands into the sea ; as if a 
 large number of small card blocks, placed side by side, had been 
 pressed from three sides so that they burst out upon the fourth. 
 From the mouth of the Kecarpwei river to that of Shecatica bay, 
 the strict coast line presents the following peculiar shape ; at point 
 I, we find the Kecarpwei river entering the sea 
 by a narrow channel between the islands ; at 2 the 
 ^ St. Augustine river enters in a similar way ; and at 
 3 the Carkevvetchepe also. At 4 is the headland 
 promontory of Shecatica bay upon its northeastern 
 side ; while from there to Bate dcs Roches (5) the coast is more or 
 less indented by small bays, and deep fiords extending in a north- 
 northeast by south-southwest direction. The almost square patch of 
 water between the St. Augustine and Carkewetchepe rivers, is filled 
 up with the small Main island. Large island, and Sandy, Cum- 
 berland, and Dukes islands in the northeast extremity, while 
 River island, at the mouth of the St. Augustine river, fills the 
 eastern end of the figure. Along the line of coast from i to 2, is 
 first Inner, and directly southeast, with only a small passage between, 
 Outer island ; while southwest of Inner is Long Island. All these 
 'slands are separated from the land and from each other by nai w 
 strips of water as I have stated, and these passages rece e 
 
 name of rigoulettes (from the French word rigole, a tr? 1 o i^ut- 
 ter), and are thus called by the inhabitants. The passage between 
 
 V 
 
BIRDS. 
 
 67 
 
 
 Inner island and tlie land is very small, and has received the name 
 of "little rigoulette ;" between Inner and Long islands is a passage 
 leading into the sea ; between Long island and the mainland is a 
 passage which really is a continuation of "little rigoulette," but there 
 is no passage for vessels at this extremity which lies oi)posite the 
 mouth of Kekarpwei river. I mention this as you will see that 
 upon our return trip we passed the opening to the sea and ascended 
 this narrowest part of the rigoulette nearly to the mouth of the 
 river, in an early foggy morning, before finding that we must retrace 
 our steps; but that, wilh a description of the scenery, which was 
 something very lovely, will occur in its place. Between Inner and 
 Outer islands, occurs "big rigoulette," the general passage for 
 coasting vessels into tht channel seaward, just ahead of the open- 
 ing of St. Augustine river and River island. As we did not remain 
 long at St. Augustine, and did not go ashore at this time, there 
 was no opportunity to examine the region carefully. 
 
 There were plenty of birds flying about in all directions over our 
 heads, and others swimming in the waters around us. The her- 
 ring gull {Larus argentatus) , and great black-backed gull (Z. 
 viariniis), were very abundant but very wild, especially the latter. 
 Ihe eider ducks {Somateria mollissima) were frequently to be 
 seen with their broods of young just ready to fly sporting in the 
 water all around us, or flying from point to point of land at our left, 
 while numerous "pigeons," or black guillemots {Uria grylk), 
 swam about us on both sides. We amused ourselves by firing 
 at these latter birds, and watching them dive almost at the flash of 
 the gun. We soon sailed by them all into the open water again, 
 where, looking over the side or prow of the vessel, the many shaped 
 and colored acalephs or jelly fishes lay swimming along with the lazy 
 flappings of their gelatinous-like disks. Although not an expert in 
 the subject, I am sure I detected several species here apparently 
 common that I have not seen in our home waters, though I may 
 mistake. What surprised me the most was finding these delicate 
 and apparently easily destroyed animals in these cold northern 
 waters at this time of the year in such abundance. 
 
 ! i 
 
 ■I 
 ■' 
 
 - 
 -•> 
 
 ■' 
 
58 
 
 OLD FORT ISLAND. 
 
 ■t- \i 
 
 We anchored here for the night, and in the morning continued 
 our journey, the destination of which was about fifty miles far- 
 ther down the coast. 
 
 Saturday the 25th. About 11 o'clock, a. m., we reached Bonne 
 Esperance safe. Mr. William Whiteley, who resides here, and is 
 the magistrate of the coast, met us and invited us to his house. 
 Here we met a charming English-American family, and were cor- 
 dially entertained by them ; but as both the people, their work, and 
 this place will be more fully described farther on, I will not now 
 linger upon its attractions. In the afternoon we took a boat to Old 
 Fort island, a small ihland, though the largest for some miles around, 
 and were soon being piloted through the narrow passages so dan- 
 gerous to those not fully acquainted with their shoals and concealed 
 rocks, to the latter named place. I will leave the general features 
 of this place, also, to be described later in a little sketch entitled : 
 A Labrador Home. Here I have given an outline of the general 
 style of living in this region, as well as a description, also, of Old 
 Fort island, the place where we found -ourselves at about five o'clock 
 the same evening, and where I remained for some time s<^udying 
 the general appearance of the region, and making frecjuent excur- 
 sions in various directions to examine the peculiarities of the local- 
 ity in which I found myself placed. It was, I assure you, quite 
 pleasant to be on shore once more. 
 
 Monday the 27th. This morning a party of us went out in a 
 boat for a short sail, taking our guns with us. The water was full 
 of birds, especially ducks and auks, as well as the other birds that 
 frequent this region in such abundance, and of which I shall speak 
 in other places ; while my attention was called particularly to the 
 "sea duck," of which we shot several from flocks that chanced to 
 fly near enough to us. As the sea or eider duck is one of the pe- 
 culiar residents of this region, a few remarks upon it, collected 
 from the experience of my year's observation may not be uninter- 
 esting, so I give them. The sea duck, as it is here called, and 
 by the word here I mean all along the coast from Mingan — if not 
 from Quebec — to Red bay, perhaps beyond the Straits of Belle Isle, 
 
THE EIDER DUCK. 
 
 69 
 
 is the eider duck of the naturahst, and the Somateria mollissima of 
 the scientist. The first specimens we obtained were shot St-pt. 27, 
 the above date, and wer ^ young birds. We saw a gieat many small 
 companies of birds scattered here and there about the harbor, but 
 they were generally, at this season, composed of old birds and their 
 broods of young ; the latter were now large enough to kill and are 
 excellent eating. In hunting these birds, especially the old ones, 
 one is obliged to proceed with the greatest caution. A good 
 sighted hunter will detect a flock or a single duck, in rough water 
 even, at a great distance — this is probably due to the fact that liv- 
 ing in a region where one must depend so much upon eyesight, 
 that sense is remarkably quickened — at the same time the duck 
 will also perceive the hunter almost as quickly as it is seen. When 
 the duck sees the suspicious object it reaches its neck to its full ex- 
 tent and takes a long, though quick sight ; if the hunter sees this 
 movement he knows that he is detected ; if he at once remains per- 
 fectly still, the duck is often outwitted, since not seeing the ol<- 
 ject move, it supposes that it is some stone or piece of wood before 
 unnoticed, and continues its feeding ; should the hunter move visi- 
 bly ever so little, the bird takes fright and is off at once. In a clear 
 day a person peering cautiously over a slight eminence can see, 
 especially if the v/ater be trancjuil, a flock of ducks often a couple 
 of miles seaward. A patient hunter will then conceal himself near 
 some chosen feeding ground, imitate the call of the male bird, and 
 decoy a flock or single bird quite close and within shooting distance. 
 The call is whistled, and sounds like the single, double, or tri])le 
 call of a snipe, repeated several times in a sort of guttural tone, if 
 such an expression may be applied to a whistle ; after every few 
 times there is an extra low and another similar high note which 
 rounds off the whistle with that i)eculiar effect so often j)ractised by 
 small boys in trying to roll the tongue, and which enters into the 
 call of so many of the water birds. At low and falling tide the 
 ducks assemble in large colonics on their feeding grounds where 
 the water is shallow and the kelj) and muscles thick, — generally 
 at evening and in the early morning ; at such times they will sit 
 ujjon the rocks and remain there until urged or driven off; their 
 
 ■ V. 
 
 if 
 
 ■1) 
 
 
 f , 
 
 ■ li 
 
 :j. 
 
 
 
n i 
 
 60 
 
 DUCKS SWIMMING UNDER WATER. 
 
 II 
 
 sight and hearing seem then to be marvellous, and the slightest 
 noise sends them off into the water. I have seen them in midday 
 thus sunning and resting themselves, but they are so watchful that 
 it is rare that you can get near enough for a shot at them. They 
 dive at the flash of the gun. I have fired at them at a rather 
 long gun-shot off, and seen them dive while the shot struck the 
 place they occupied a second previous. 
 
 An experienced hunter when on shore will get as near to a 
 flock or single bird as possible without alarming it, wait patiently 
 for it to dive, as it so often does while feeding in apparent safety, 
 when he will run ahead to some shelter nearer the object of his de- 
 sire, repeating the operation until he regards himself sufficiently 
 near, and then, remaining standing with his gun at his shoulder, 
 fire at the unconscious bird when it rises from some long dive, gen- 
 erally killing it- In the fall, when a brood of young ducks is sur- 
 prised it is (juite easy to secure a large number, though the old 
 birds generally escape by flight and swimming under the water ; 
 they accomplish this latter act with ease, and often pass long dis- 
 tances before appearing to the surface for fresh air. In the open 
 water, a flock of old birds when approached will separate and swim 
 or fly in different directions, while the young cluster, and thus ex- 
 pose themselves directly to the hunter's fire. The best way to pur- 
 sue both young and old birds is to drive them into some angular 
 indenture of the surrounding islands or land, and then wait for them 
 to appear on the surface of the water after their long dive. The 
 boat stationed too far away for them to swim clear of it, the hunter 
 has every chance for bagging his game. I have noticed that wounded 
 birds do not swim far, about eighteen inches to two feet, below the 
 water ; both bill and head are extended for\vard in a straight line. 
 The old birds will often swim over a quartei of a mile, if not 
 a full half, in extent, beneath the water without appearin.d for air. 
 As (lir as my experience goes the birds are rather tame in the win- 
 ter season, or at least in the extreme late fiiU ; they huddle together 
 in close bunches of from fifty to several hundred birds, and I am 
 informed that an old hunter once fired into a cluster thus gathered 
 and bagged fifty-nine birds with a single discharge of his gun, a 
 
am 
 ^red 
 
 BREEDING OF THE EIDER DUCK. 
 
 61 
 
 common large bore fowling piece. Occasionally old 9 birds in full 
 heat will be shot that have the back and wing coverts edged with 
 deep rusty brown, and often very deeply so ; other birds, smaller, 
 (young) at the same season of the year will have the feathers, par- 
 ticularly of the breast, edged with deep gray ; young birds gener- 
 ally have the top of the head darker and the neck much lighter. 
 In some old birds the whole plumage will be unvaried and of a dark 
 brown color. Large flocks are usually made up of a number of 
 small family broods of from five to seven birds that unite from 
 some common cause, and then pursue some common flight until 
 scattered from other causes. The usual feeding grounds of the ei- 
 der duck are shallow waters over a bed of seaweed or mud, at 
 some rods from land on its south, southwest, or west side. They 
 feed principally on moUusks, barnacles, and a variety of marine an- 
 imal life, with an occasional piece of seaweed such as may be ob- 
 tained in the shallow basins of accumulated debris, and on the 
 land- wash. In summer the ducka breed in large numbers on. the 
 islands about the harbors, and though their numbers are fast de- 
 creasing there ar'" still colonies of them, making their nests of down 
 from their own breasts, beneath some overhanging grassy clump, 
 and laying from three to five olive colored eggs. The people here 
 rob the nests several successive times during a season, while the 
 female continues to lay eggs in the hopes of securing enough to 
 hatch her brood. When setting, the eider duck remains upon her 
 nest until the very last moment, and then forced, takes a rapid flight 
 a short distance off, and does not appear again — at least I have 
 not noticed it — until the intruder has gone away. To what extent 
 the males assist the females in tiie matter of incubation I did not suc- 
 ceed in finding out positively, that they do so to a limited extent I 
 cannot doubt. They remain in the region until the last waters of 
 the bays freeze over, and are then seen no more until spring returns 
 and thaws the ice, when they appear in company with the king ei- 
 der (^S". spectabilis),, and the Pacific eider (.S". F-nix^m) — which 
 also are found in immense flocks, but distinct from, that is not min- 
 gling with, the others. 
 We here shot several other species of birds ; the titlark {Anthus 
 
 
 
 
 
 
 1 
 
 :' 
 
 . 
 
 h 
 
 ! 1 ' ■ 
 
 ':■* 
 
 ; 
 
 : 
 
 \ 
 
w^^ 
 
 62 
 
 SHOOTING BIRDS. 
 
 1/ 
 
 11 
 
 ludovicianus) , sandpipers of the species known as the white- 
 rumped {Tritiga bonapartei), as also the sanderling {Calidris 
 arenaria), all of which were more or less abundant. As the af- 
 ternoon proved rainy we remained in the house and amused our- 
 selves as best we could until bed-time. 
 
 Tuesday the 28th. To-day we were out shooting almost all day ; 
 we obtained several of the species of plover known by the many 
 different names of blackbellied, whistling field, beetlehead, ox-eye, 
 bull-head, and here by the queer title of Quebec curlew : it is in 
 fact the Squatarola helvetica of the scientist. I found this bird in 
 small flocks more or less abundant all along the coast at most of the 
 different places visited from Quebec to Blanc Sablon. It was gen- 
 erally rather wild, and, wading deep into the water, fed on the small 
 sea animals that it could there capture. Its flight was low and 
 rather short. I went out before breakfast this morning to make an 
 attempt on the small beach birds that abounded in a cove of the 
 sea not far from the house, and was successful far beyond my most 
 sanguine expectations. The great dish prepared from these small 
 but delicately flavored little birds, is to fry them in their own fat. 
 Out of several good sized flocks I managed to procure, I think, 
 eighty-seven birds, which, carried home, were soon sending up 
 their savory odor from the frying-pan. Rain in the morning obliged 
 us to stay within doors. Here it rains about one-half ^he time ; 
 it is foggy and uncomfortable weather for nearly the other half, leav- 
 ing but a short season of sunshine, and there is seldom anything 
 like hot weather. If the thermometer goes up to 70° it is hot, while 
 the usual temperature is from 45° to 60° in summer, and as low as 
 often to freeze the mercury in winter. Mercury freezes at about 
 40° Fahrenheit. 
 
 As it came off pleasant in the afternoon we had abundant op- 
 portunity for walking about the island, enjoying the prospect as 
 well as seeing the sights. In a little corner of the yard our host 
 had started a small garden, but as its care depended upon his 
 own personal labor at the height of the fishing season, when every 
 spare moment was occupied in the business of the hour, or one 
 was indulging in rest, it had fallen rather out of use. The plants 
 
 i). 
 
ts 
 
 VEGETABLES — HAY. 
 
 68 
 
 grown were of the varieties often found in similarly situated little 
 corner gardens, and consisted of turnips, cabbages, and potatoes. 
 The latter, of the species known as " Early Rose," were very small ; 
 while the turnips had turned out here as they do so often elsewhere 
 along the coast, rather small but very good. Of the cabbages, the 
 best that can be said of them is that they flavored an occasional 
 soup. Onions are rarely raised here, I saw only a few. The soil 
 is generally too sandy, being usually an elevated sea beach, to grow 
 much upon, and the vegetables suffer in consequence. The great 
 reason why farming is not made of more account here is the fact 
 that so few people have their houses where there is abundance of 
 soil proper for such work, and also the fact that they can purchase 
 quite fresh articles in the fall from the Quebec traders, at almost as 
 low prices as if they procured them in Quebec it'^elf. 
 
 Many of the people along the coast keep cattle, and the family 
 with whom we stayed possessed two heads besides numerous goats- 
 The cow gave fine milk, and a good generous amount of it, as did also 
 the goat ; the cream was thick and sweet, and the fresh butter 
 made twice or three times a week was really most excellent. The 
 cattle a^e fed upon a rather coarse, though sweet grass, which grows 
 in abundance on the level flats of the islands about. In the fall it 
 presents an appearance quite like that of a field of stunted grain, 
 since it is of a brown co^or and quite high, with light tassels hang- 
 ing drooping downward and stirred with every breath of wind. Sev- 
 eral species are very common and combined from excellent hay : 
 with Poa pratensis, we find both Hierochloa borealis^ and Elymus 
 mollis, though the first mentioned one is without all doubt the most 
 excellent grass or hay here for the cattle. The majority of the 
 low grass grows in thick matted clumps, this is undoubtedly because 
 it is not cut sufficiently often to produce a large and thinner variety ; 
 it is much like our "rowen," however, and seems to answer nicely for 
 the cows. In the summer there is plenty of first class feed, and good 
 pasturage anywhere. The cattle are let out to graze at will ; when 
 evening comes they are sought after among the dells and vales of the 
 neighboring region and driven home. It is often no slight job to 
 
 t 
 
 ,'i 
 
 I » 
 
 1 , 
 
 
WiT^ 
 
 64 
 
 RAISED BEACHES — LABRADOR DOGS. 
 
 
 1 , !: 
 
 ill 
 
 ff ' 
 
 1, 
 
 find them when they have strayed for several miles over hillocks, 
 across plains and open stretches of meadows that lead to a success- 
 ion of elevations and depressions, when perchance they have 
 rounded on their liacks and are quietly giazing not far from the 
 house in soine entirely opposite direction from that in which they 
 started out. Strange to say the grass comes up here quite early in 
 the spring, and is grown some length before even the snow has be- 
 come melted, so that there is good grazing even in early spring. 
 
 A curious fact remains to be spoken of: in digging in the gar- 
 den I was informed that at the distance of about three feet under 
 ground the sand contained abundance of shells similar to those now 
 growing alive on the beach a few rods away, and which belong, I be- 
 lieve, to the species Mytilits ediilis; a little further front, along the 
 same sand ridge, nearer the shore, — or land-wash as the expression 
 is here, — these same shells appear on the surface, or at least so very 
 near the surflice, that comparing the several lines of demarcation 
 there would seem to be a fair amount of evidence that either there 
 had been a gradual subsidence of the waters about this part of the 
 coast, or that the land had gradually risen, allowing the sea to heap 
 up a succession of beaches over which the sand had been blown 
 from above until they had been covered, as they are at present, and 
 overgrown with vegetation. 
 
 It seems not unlikely that one or the other of these actions has 
 taken place. In several different places along the island, and in sev- 
 eral of the islands, there appears to be this same peculiar sand struc- 
 ture overlying what were apparently once old beaches of the sea. 
 
 A few words only here, about the dogs of Labrador. Upon our 
 first stepping on shore we were met by a troop of about fifteen dogs. 
 These dogs are of mongrel breed between the Esquimaux, the New- 
 foundland, and the various species imported here from other regions. 
 They are used to draw the sledges in the winter, and are as valuable 
 to the inhabitants as horses are to us, yet a worse set of snarling, 
 barking, and generally fierce and also unhospitable animals, it would 
 be hard to find. They feed on blubber and any food that they can 
 obtain in the winter, which has been previously banished from the 
 
 , V v. 
 

 RLACK-TAILED GODWIT. 
 
 65 
 
 4 
 
 odds and ends of what can be made no use of in the house ; 
 while in summer they are never fed, but allowed to roam at will and 
 find what they can along shore and around the island ; they fight 
 incessantly, while, unlike our dogs, should one get undermost, the 
 rest will turn upon hmi and worry him to death. The people are 
 thus frequently obliged to arise even in the middle of the night to 
 save the lives of their animals, or their numbers would rapidly 
 decrease. They are by far too valuable animals to lose so easily. 
 
 Wednesday the 29th. Another rainy day has prevented our doing 
 much outside of the house, and we must be contented to remain 
 within doors. I have been struck with the very unusually large 
 amount of bread that is used all along the coast. It is in reality 
 "the staff of life," in these parts at least. It is raised with a yeast 
 made from the spruce. The fine tender twigs of the young plant 
 are taken, and being boiled in water made into a regular spruce 
 beer. The beer will keep fresh about a week in summer time, and 
 nearly all winter. When fresh it is an excellent drink and much 
 used here in place of water, which, poor at best, is usually taken 
 from ponds, and rarely from springs or running water ; tl.ese ponds 
 are the result of the melting of large bodies of snow that drain into 
 the low mud flats, often with rocky bottoms, which abound every- 
 where, high and low ; a fair evidence that successive elevations 
 were at one time much nearer the sea-level than they are at pres- 
 ent. When a moderate amount of this beer is mixed with a cup 
 of molasses \'c forms a yeast-like substance, or raising mixture, which 
 fully equals our ordinary raising compounds. The beer itself has 
 a peculiarly bitter taste, from lack of any article in the boiling to 
 counteract the natural taste of the spruce ; it makes a very palata- 
 ble drink, however. 
 
 Thursday the 30th. To-day I succeeded in obtaining a specimen 
 of the Limosa hudsonica, the hudsonian or black-tailed godwit, 
 also called the ring-tailed marlin. It is a rare bird, even in these 
 regions. It was the only one I obtained on the coast. It was at the 
 time flying rather high up in the air, and with the irregular flight of 
 the spotted sandpiper. Its note, uttered while in the air, sounded 
 
 ! 
 
 ■J 
 
 } 
 
 r 
 
 I \ 
 
F 
 
 
 G6 
 
 SEARCHING FOR DRIFT-WOOD. 
 
 
 I' 
 
 more like the squeak of a mouse than anything else I can name. 
 From its rarity I give the dimensions as I took them. Length from 
 end of bill to tail i6 inches ; end of bill to toes 18.50 ; extent 28 ; 
 wing 8.75; tail 3.25; bill 3.25; unfeathered tibia 1.13; tarsus 
 2.50 ; hind toe and claw .50 ; middle toe and claw 1.65. It was a 
 (J and had the gizzard full of gravel and nearly digested matter. 
 The people tell me that it is occasionally seen in fall, but that it is 
 rare. Audubon speaks of it as "rare along the Atlantic district 
 in spring and autumn. Breeds in the barren grounds of the Arctic 
 seas in great numbers." It seems to be confined to the more mid- 
 dle interior parts of the Arctic regions, and the majority of writers 
 whose works I have seen speak of it, as I have generally found it, 
 as rare along the Eastern Atlantic and Gulf coast generally. 
 
 Friday, Oct. i . To-day has been occupied in procuring firewood. 
 A number of the men who live upon the island started off with 
 their boats for a tour along the shore after drift-wood. The wood 
 collects in the coves everywhere about the islands as well as the 
 bays of the mainland, and is eagerly sought after by the people 
 who, in many cases, are obliged to go a considerable distance other- 
 wise for their supply ; it is brought in boat-loads to the shore close 
 by the house, the wet material spread out to dry in the sun, while 
 the remainder is sawed or chopped up for immediate use. In this 
 way any old wreck or pieces of vessels that have floated ashore are 
 soon spied out and secured ; limbs and branches of trees, as well 
 as boards that have fallen off from vessels and floated ashore are all 
 serviceable, for if clogged with wet they will generally soon dry 
 when exposed to the sun away from the water. Old boats are also 
 found occasionally ; all this apparent old rubbish is of value, and 
 shows that very little is lost in the economy practised on these 
 shores. 
 
 Saturday the 2nd. To-day we visited Bonne Esperance and 
 passed the day with Mr, and Mrs. Whiteley in a very pleasant 
 manner. As the locality presents some very interesting features of 
 surroundings, as well as some attractions in itself, it may be well to 
 give a general description of the place. 
 
BONNE ESPERANCE. 
 
 G7 
 
 CHAPTER V. 
 
 Bonne Esperance — Esquimaux river and island — Caribou island — F2ntering 
 Bonne Esperance harbor — Vessels in the harbor — Their nationality — 
 Activity of place — Religious character of people — Chapel and Mission 
 house — Residence of Mr. Whiteley, magistrate — Nescopies — Store 
 and shop provisions — Money — Trade — A trading story. 
 
 Bonne Esperance, or Good Hope as it would be called in Eng- 
 lish from a simple translation of the words from the French, is a 
 small island lying off the Labrador coast about ppposite the mouth 
 of Esquimaux river, and a little to the right of it, in very near lat- 
 itude 51° 24' north, and longitude 5 7° 39' west. It is quite a small 
 island, and lies within a few miles only of the mainland, with sev- 
 eral other islands all somewhat smaller than itself around it, 
 principally between it and the shore. About three miles on the 
 northwest, f^squimaux island nearly fills the mouth of the Escjuimaux 
 river, and is the largest of the small islands on this part of the 
 coast. This was undoubtedly one of the chief seats of the Indian 
 and Esquimaux, as well also as of the French and Indian wars. 
 A large number of graves were supposed formerly to exist on the 
 island, while an occasional one is still to be found there, though it 
 is of course impossible to say as to their origin ; both Indians and 
 Esquimaux were undoubtedly buried in large numbers on this very 
 island, if we can believe the records. 
 
 On the northeast of Bonne Esperance is Caribou island. This 
 is another large island, thus making Bonne Esperance the third 
 in size of the small group here situated. If we count Old Fort 
 island, which also is larger, we find Bonne Esperance, the fourth 
 and next largest of all the islands within a radius of some twelve 
 miles that contain perhaps fifty islands large and small, and as 
 
 N 
 
 
 r \ 
 
 
 
 ■ ', 
 
 
 |: : 
 
 ^ ,. 
 
 1 
 
 is 
 
 5 F 
 
 
 r 
 
 :' 
 
 
 ' : 
 
 \ ! 
 
 
 
 
 
 
 1: 
 'J 
 
 Lfl 
 
fFi; 
 
 68 
 
 now TO ENTER BONNE HARBOR. 
 
 m?' m 
 
 ill 
 
 '*;? , 
 
 I'liri 
 
 f! 
 
 many rocky knobs or crests rising aliove the water, arranged in a 
 most singular manner to form broken and winding channels which 
 are navigable with the greatest of caution, and then only by those 
 who have gradually become accustomed to their intricacies. There 
 are two ways of reaching lionne Esperance by vessel : one is to 
 take the channel inside the islands, the other to reach the harbor 
 from the outside ; the latter is easy while the former is quite 
 difficult. Very small islands lie about the harbors on this part of 
 the coast in such profusion as well as confusion, that the channels 
 formed present a most intricate network of passable and nonpassa- 
 ble openings. Right here, about Bonne Esperance, occur many 
 such places. Clustered as the islands and shoals are one can 
 hardly pass them in safety unless well acquainted with them. 
 Many of these passages, apparently safe to the sight, contain 
 sunken ledges or single rocks that render them unsafe for large ves- 
 sels ; while many places, whose narrowness and dark looking waters 
 would repel one, are the very passages chosen from their depth, 
 with fair wind, through which to glide. Strange as it may appear, a 
 wide and fair opening of water between these islets is an almost in- 
 variable sign of shallow water. Vessels can enter Bonne harbor 
 by keeping close to the mainland following carefully the evident 
 channel, keeping to the right between two small islands and fol- 
 lowing the narrow passage between them, and bearing directly to 
 the left again running straight for the stage houses then in sight ; 
 but the safer way is to enter from outside, as it is called, that is, — if 
 sailing in sight of the islands, but at a distance out to sea, — continue 
 until the white beacon of Whale island comes in sight, steer for this 
 and when it is directly northwest (all these points are magnetic 
 points, which here differ by 36° 45' west of apparent points ; that 
 is, apparent north on the (hart is 36° 45' east of true north), 
 sailing in a north or north by west direction, until the fish houses 
 are seen between the two small islands at either hand, when the pilot 
 can steer directly for them. The harbor of Bonne Esperance is 
 a neat sort of natural bay situated between the shores of the 
 opposing islands. Either shore retreats in a semicircle, so that the 
 
THE FISHING FLEET. 
 
 G9 
 
 inclosure forms an oblong semicircle of water ; the outlet to this 
 natural bay is a narrow strait in which the water is only nine feet 
 deep at high tide. Of necessity, only small crafts, such as the nu- 
 merous fishing vessels here abundant from Newfoundland, United 
 States, and other localities, can pass this opening. On the other 
 side is a similar bay, formed by the retreating sides of the other 
 part of each island, from which it is easy to gain the open water 
 again. 
 
 In this harbor it is no uncommon sight to see a hundred ves- 
 sels of all kinds and descriptions at anchor. Of ll -.e the majority 
 are from different harbors in Newfoundland, while of the others 
 many are from the Magdalen islands, and some from Nova Scotia ; 
 
 numerous French crafts and a few American are also occasionally 
 seen. 
 
 It is a pretty sight, I assure you, to see this out-of-the-world 
 region so thoroughly alive and stirring, as it is in the summer, 
 with all these small vessels from different parts of the world congre- 
 gated here, often by hundreds, filling the bay from head to entrance ; 
 while others remain outside flitting about from place to place as 
 occasion may require, all engaged in the saiie pursuit of fishing. 
 The following partial list of the localities with the number of ves- 
 sels sent out from each in 1878 will give you a glimpse of the 
 business of the season. Halifax, 8 ; Quebec, 4 ; Newl)uryport, 2 ; 
 Britain (Ang.), 2 ; Gaspe, i ; Boston, i ; Nova Scotia, 21 ; New- 
 foundland, 13 ; total, 52. These are but a small part of the num- 
 ber that were actually seen in Bonne Esperance harbor during 
 the summer of 1878 ; but these were registered as dealing more 
 or less directly in fish, and most of them remaining more or less 
 of the season in or near that locality. Bonne Esperance and 
 Salmon Bay, the latter being an establishment owned by New- 
 foundland parties, and situated about two miles away, are the 
 largest fisheries this side of Blanc Sablon on the northeast, or Nat- 
 ashquan on the west ; the former locality is only about twenty miles 
 away while the latter is some three hundred. It is a great sight for 
 the simple inhabitants dwelling near these fisheries when the vessels 
 
 I 
 
 ii 
 

 70 
 
 CHAPEL AND MISSION HOUSE. 
 
 
 come in the spring and begin their work. The long winter is past, 
 the bays at last open, and the vessels soon begin to arrive. Most of 
 the inhabitants are very quick of observation. Let them see a 
 vessel once and know her name, they will, unless great alterations 
 have been made, tell her again under almost any circumstances. 
 
 As the vessels anchor one by one in Bonne harbor, the people 
 assemble from the neighborhood around and the island becomes a 
 scene of gayety and life. Of course it is gayety in a sort of prim- 
 itive sense ; but then the people think much of it, especially as it 
 affords them so much enjoyment the coming winter in rehearsing 
 at their firesides what they have seen and heard. 
 
 I will not now mention the religious character of the people, or 
 the service as conducted in the little chapel on the top of the 
 hill ; but will only say that it is a most pleasing opportunity to visit 
 one of these native, evening church services ; to see forty or fifty of 
 the native people gathered together, and with them the transient 
 strangers from P'rance, England, the United States, and all the 
 places from which vessels sent out to the fisheries have come, from 
 the various crafts in the harbor ; to hear all join in song ; and lis- 
 ten to English, French, and American as they lead in their hearty, 
 heartfelt prayer. It is a good work that is done by the little mission 
 there established and God bless them in their work. 
 
 Below the chapel, under the hill, lies the "Mission House," as 
 it is called. Here reside several missionary ladies or gentlemen, 
 sometimes both, who have given themselves to the work of attending 
 to the spiritual wants of these poor people. It is here that the Rev. 
 Samuel R. Butler, of Northampton, Mass., and before him several 
 others, spent so much of their time while on the coast. Here, in 
 summer, the children attended regular day school, but to which, 
 owing to the distance at which everybody lives, few come ; here 
 also they come to Sunday school, which is usually well attended, 
 since the people are free to go and come as they choose from 
 their business on that day and they do so in preference to re- 
 maining at home. Not far from the Mission is the residence of 
 Mr. W. H. Whiteley, who, as I have said, is the Magistrate for 
 
 M,' 
 
 .»•... 
 
MR. WIIITELEY'S ESTABLISHMENT. 
 
 I 
 
 Canada for this part of the coast. Mr. Whiteley's house is really 
 a fine one for this section of the country. It is a good sized, two- 
 storied affair, much longer than broad, as is the custom with arctic 
 and sub-arctic houses, and very warmly built. Its white paint at- 
 tracts one from the outside, while a cosey fire and a cup of hot tea 
 are always handy within. Mrs. Whiteley is c[uite ready to welcome 
 a "new comer," and the hearty, rosy children are as delighted to 
 find a stranger to take notice of them as can be imagined. The 
 sitting and dining room are combined ; a large, square, or rather 
 oblong, stove heats the room, and there are plenty of chairs, stools, 
 benches, and an easy lounge ready for occupancy at any time. I 
 shall not soon forget the cosey times I have had in that same room 
 with the assembled family ; and the games and plays with which 
 long winter evenings have been whiled away. The many pleasant 
 faces almost recall me there as I write, and the hope of some 
 day meeting them again is present with me. 
 
 Outside of the house the contour of the island is rather peculiar ; 
 though if not strange to an inhabitant it is at least so to a visitor. 
 The whole island is one mass of granite rock overlaid here and 
 there with scanty vegetation. There are two principal crests of 
 about sixty-five, and eighty feet in height above the sea-level ; 
 their rocky summits are crowned, as is usual with nearly all such 
 elevations about the coast, with a pile of stones several feet in height 
 which serve as landmarks : they are called Nescopies : the sailors 
 call them the "American Men." 
 
 The other buildings upon the island are a small store and shop ; 
 here are kept a variety of articles of hardware, groceries, and dry 
 goods, as well as a general assortment of articles of clothing, boots 
 and shoes, hats, caps, and oilskin suits — the latter of which are 
 so much used among the men. The assortment of nicknacks 
 of one kind or another is always more or less limited, and confined 
 to a few of the many useful articles of dress and use rather than of 
 the kind to tempt the coppers of the mamma and papa, as is so cus- 
 tomary in the United States and other parts of the world. It is 
 rather a store of necessity than one of choice and amusement. 
 
 H 
 
 I • 
 
 I ' If 
 
 ! it 
 
 •|!i 
 
 '*■' 
 
 ■;t ( 
 
 I. 
 .1 
 
¥^ 
 
 72 
 
 MERCHANDISE. 
 
 
 tl 
 
 The room is a small one, and the various articles are piled in iieaps in 
 corners, and about the sides and middle of the room, for waut of 
 space. This is the regular trading jKjrt for the region around. If 
 any article of commerce is wanted it is obtained here if anywhere 
 on the coast : a barrel of flour or salt pork ; a box or bag of bis- 
 cuit as they are called, or more properly hardtack or hard crack- 
 ers, — and as a rule they merit their name since the majority of 
 them recjuire long soaking in water before they are soft enough for 
 use, to be frieil, broiled, or eaten at all, and I have often seen 
 them so tough that the rejjeated blows of a hatchet-back would 
 barely suffice to break them ; meal, of which varieties that com- 
 monly called oatmeal, is much used, while Indian meal is regarded 
 by the majority as only fit to give the dogs, to whom it is fed scalded, 
 though the poorer class are frequently obliged to mix it with their 
 flour if not use it altogether i)ure ; butter, salted down in tubs or 
 firkins, and lard, both of an inferior (]ua!ity ; sugar and molasses, 
 the latter of whicli is the chief source of the sweetenings used on 
 the coast. A strange custom prevailing here is that of sweetening, 
 with molasses the tea or coffee, though very little of the latter is 
 ever used so far northeast, tea being a i)opuIar and more healthy 
 drink, and the true, native Labrailorians invariably take the mo- 
 lasses pot, even where the sugar is eijually as easy of access, and 
 use its contents where we should use the sugar. IJesides provisions, 
 nails, hatchets, axes, and tools of various (lescriptit)ns — usually con- 
 fined to planes, saws, chisels, and screw-drivers — are always on 
 hand ; large cross-cut saws can be purchased of the traders, as can 
 other articles wanted if known and ordered beforehand. A few 
 yards of any of the cheaj) patterns of dress wear, Irinunings. gloves, 
 stockings, underwear, coarse materials for overalls, antl in tact a little 
 of anything that long experience has shown to be in constant use, 
 will usually be found somewhere in the i)romiscuous heap of ma- 
 terials stored in this room ; while it would consume a chapter in 
 itself to enumerate the variety of articles here acciunulated. It 
 is a great day when several of the inhabitants living say ten or fif- 
 teen miles up or down the coast ( onie to the store to purchase 
 
MONEY, 
 
 73 
 
 their monthly supply of necessities. It takes a long while for them 
 to settle upon what they want, and they pick out the articles they 
 wish with the eagerness and apparent pleasure of a small ■::hil(l. 
 Nearly every article has to be handed over a number of times before 
 the one desired can be settled upon, and when it is very likely it will 
 be exchanged for something more pleasing to the eye. Finally, the 
 exchange is arranged and the i)arties start for honn- with their pur- 
 chases. The general medium of exchange all through this region is 
 trade. Money is seldom used, and its value still less seldom 
 known. Both the Fuiglish and the Cann.'i; n, as well as some I'nited 
 States money fintl their way into the La!)rador markets. The Prov- 
 inces have each a money of their own, and nearly every piece has a 
 discount upon its real value. The general mode of reckoning is in 
 P^nglish pence, shillings, and jjoimds. A pound being twenty shil- 
 lings or four dollars (as it is in the majority if not all the Canadian 
 provinces), the shilling is twenty cents or twelve pence ; the Eng- 
 lish sovereign is taken for four dollars and eighty-four or eighty-six 
 cents, and consists of twenty-four shillings twopence, or threepence 
 half-peimy ; it is only in the Newfoundland province, northeast of 
 IManc Sablon, or l,abrador proper, that the English gold l)rings its 
 full value of twenty-live shillings. To compli* ate matters still more 
 both the iMiglish shilling of twenty-five cents, and the Canadian 
 twenty cent piece are freely used and e(iually abundant. New- 
 foundland silver is used as are all the other provincial pieces for 
 their full value. A»r;'ncan silver is occasional, and usually heavily 
 discounted, while .Miierican gold is, I believe, the only gold taken 
 for hill lace . ilue and without discount. In American money the 
 Canadian fifty cent piece is worth forty-eight cents, the twenty-five 
 cent piece twenty-four cents, the twenty cent nineteen cents, tlie ten 
 cent nine cents, and the five cent four cents ; on the other hand a 
 half-pence (pronoimced hajjence) is generally two-thinls of a cent 
 or one cent, a two-pence (pronounced tuppence) three cents, 
 a three-pence (thrippence) in'n cents, a (fcr-pence (fopence) 
 seven cents, six-pence ten ( ents, nine-pence iifteen cents, twelve- 
 
 ii 
 
 -j 
 
 < 
 
 
 
 1 
 
 
 ^^ 
 
 L. 
 

 74 
 
 THE MEDIUM OF EXCHANGE. 
 
 pence the same as a shilling of twenty cents, and fifteen and cight- 
 teen-pence twenty-five and thirty pence : thus our quarter is in Lab- 
 rador fifteen pence (not counting the discount which is usually 
 five cents), our twenty cent piece nine-pence (with a discount also 
 of five cents), our ten cents six-i)ence (discount of only ojie cent) ; 
 our five cent j)ieces are often taken for their full value. The three 
 cent piece is not used or recognized, and the fifty cent piece has a 
 discount of frcjm five to ten cents acconling to location. '\wz 
 Newfoundland IJank note of one pound, or four dollars, is always 
 looked at twice before it is taken, sometimes with and sometimes 
 without discount. It takes six large coppers to be worth a five 
 cent piece. 
 
 In trading, it is customary to employ the usual productions of the 
 country as a medium of exchange. Although it would be as im- 
 possible to make out a tal)le of exchange as a full table of money 
 ecpiivalents for the coast, a few of the more important may be 
 stated. Codfish at thirteen to eighteen shillings a ([uintal, herring 
 seven six-i)ence to ten shillings a barrel, cotloil two to two and six- 
 })ence a gallon ; seal skins dressed seven and six-i)ence to fifteen 
 shillings according to the species of seal, undressed five to seven 
 and six-pence, rarely ten shillings apiece ; seal oil two shillings a 
 gallon. Ik'sides this, people living near rivers sell wood for twen- 
 ty-five cents a load of ten long sticks, about forty feet long and 
 eight inches through at the butts. Old iron picked up from wrecks 
 is worth half a cent a pound, and everything usefiil has its price. 
 Nearly all the trading on the coast is done on the credit system, 
 and it is usually expected beforehand that half the people would not 
 pay their bills if they could, and many of them never intend to, 
 while still another portion are dependent entirely upon the catch of 
 the next season, having, by former delin<iuencies, overrun iheir 
 accounts, so that they are obligeil to get credit in advance of the 
 season upon that season's catch. No wonder the majority are ])oor, 
 and kejjt so not by the power of the richer class, but rather by 
 their own indolence in not i)rofiting when an opportunity oilers to 
 
A TRADING STORY. 
 
 make a little money. Most of them consider it best to let well 
 enough alone, and having earned sufficient to su])port them through 
 the winter, the rest looks after itself, while the time is then i)assed 
 in "chatting" and smoking. 
 
 A curious story is told which shows how some are really 
 incapable of keeping a perfectly understood account of their own 
 money. A certain trader offered for a i)iece of valuable fur forty- 
 five dollars in American money ; the native thought the exchange 
 over for some time and finally shook his head and said that he 
 could not sell for less than ten pounds cash. Now ten pounds you 
 rememb'^r is forty dollars, or five dollars less than the original offer. 
 The wily uader in turn shook his head as if in hesitancy, but fi- 
 nally said that he would accept the offer, while the nati\e went off 
 cluickling, if I may use the word, that he had made the trader 
 come up to his price. 
 
 :■■ > 
 
 I ill 
 
 I't 
 
 M-i 
 
 I « 
 
f:^. 
 
 76 
 
 RASPBERRIES — STORM WEATHER. 
 
 CHAPTER VI. 
 
 Raspljcrries — Weather — Iludsonian Chickadee and other birds — Black-fly 
 — Topography of country — Old Fort bay, physical features and sur- 
 roundings — Superstitions concerning the raven. 
 
 I I 
 
 Sunday, Oct. 3. To-day some of the people in the house 
 brought in a few raspl)erries picked near the house, but whetlier 
 they were of the specnes known as the dwarf raspberry {Rn/>us 
 triflonts), or the wild red raspl)erry (A*. s/n\'()sus), I could not (luite 
 tell as I did not analyze them ; the berries were quite dry and unfit 
 to eat. Neither of the sjK'cics is rare along the coast anywhere, 
 and it might liave been either or both of these. A quiet walk about 
 the island, which is about a mile and a half in diameter and four in 
 circumference, occupied us a good portion of the morning, while 
 in the afternoon the cool bree/es fountl us enjoying ourselv*es out- 
 side the house in the shade on the lawn. 
 
 Monday tlie 4th. After a stormy, windy night, the weather 
 continued rough and rainy all day. The wind increasing raised 
 great billows ujjon the surface of the water, these increasing in 
 height and fury dashed with incredible force against the rocky 
 hci^'hts so numerous in the outer waters about the island. Huge 
 waves would hurl themselves against the rocky ex]>osures, and 
 flooding them, pour down the oj>])i)>ite sides in br*)ad sheets of 
 water ; or again they would break into showers of spray thai would 
 spread themselves in crv>wnv, wreaths, and haloes of magnificence 
 that would rival the mv>st elegant artificial i)r()duction, and put to 
 shame the most wonderful of ancient or modern fountain work. 
 Looking from tlu right from otu' door, a long line of low reefy 
 
POWERFUL EYESIGHT. 
 
 77 
 
 islands stretched across our view, ending in a rising knoll at the left ; 
 over this I saw a continued display of hydraulic i)yrotechnics that 
 would have delighted the most fastidious with its magnificence of 
 display. The waves dashing ui)on the ujjright side of the cliff sent 
 spray high in the air over the hilly part of the island, while long 
 lines of billows broke into foam all along the narrow stretch <>f 
 beach. Several displa)s of lesser importance were seen in other 
 directions, but none equalled this one. All through the day the 
 sight continued, and only closed with the night. 
 
 With all this cold, wet rain, the thermometer has been averaging 
 about 50°, varying from 46° in the morning to 54° in the even- 
 ing. In spite of the weather some of the party went out and 
 shot a couple of ducks. In hunting these birds, which were eiders, 
 one has to proceed with the greatest caution. A good hunter will 
 detect a single duck or a flock of ducks even in rough water, 
 at an immense distance. I have often been surprised at being told 
 that there was a flock in a certain direction, when I could >ee 
 nothing at all in that direction. Some five or eight minutes after- 
 wards I would think that I had discovered a flock some ways yet 
 in the distance, and upon giving notice to that effect would receive 
 the reply, "why, have you just seen them? that's the flock we're 
 after, the same I pointed out to you ten minutes ago." When 
 people live so much by their eyesight, it is not hard to see how 
 that sense can become so acutely developed as it is in so many of 
 the people on the coast. I think I have mentioned l)efore tiiat 
 men will recognize a vessel that they have once seen years after, 
 and at an immense distance away. 'I'hey will tell a boat, and 
 the i)eople wh" .ire in it, before one unaccustomed can distin- 
 gui^h that there are any people at all in it. '1 liis results. I think, 
 partly from what one might almost call ''reasoning from analogy," 
 though no doubt f'jrj/j,'^///' has a great deal to do witli it. 
 
 Wednesday the 6th. This is still another foggy day, and the sun 
 made us but a short visit this morning. I shot a couple of shel- 
 drakes, shellbirds as they <all tlieni : we had them for dinner and 
 found them very lender anil delicious. They were young birds. 
 
 1 
 
 i 
 
KV " 
 
 
 78 
 
 HUDSONIAN CHICKADEE. 
 
 and of the smaller of the two species of Mergus^ of which the 
 females are so nearly alike, and the one known as serrator, the red 
 breasted Merganser, so common in these parts ; I found here the 
 savanna sparrow {Passerculus savanna) abundant and breeding 
 and also saw numbers of that familiar little hero of the north, the 
 Hudsonian chickadee {Parus hudsonicus), so very abundant here. 
 As this is a peculiarly characteristic bird of this region, a few words 
 may be said of it. I found the Hudsonian titmouse residing all 
 the y»^ar around both on the islands and on the mainland where- 
 ever I v/ent during my stay on the coast, and their cheerful ])resence 
 has dce-dee-dee'ed away the blues more than Oiice. I first met 
 them at (31d Fort island in the summer, when thoy would frequently 
 come and perch on the roof of the house, and occasionally fly in 
 at the door and pick up crumbs from the floor. They were very 
 tame, and would allow you even to catch them without much 
 opposition on their part. The people of the coast are very fond 
 of them, and will not allow a stranger to hurt one of them if they can 
 possibly help it. While flying they will often dart about from i)lace 
 to i)lace, or if it be in the air in one direction or another, with a 
 quick, whirring sort of flight which remintled one more of the 
 operation of ([uickly half opening one of our closed fons and imme- 
 diately shutting it again. They fly only a short tlistance except when 
 rather high in the air when their smallncss soon aids them in escap- 
 ing detection as they fly ; at such times their flight is swift. When 
 on the island their favorite resting place was the roof of the house, 
 anil we would often see several ]jerched near each other on the 
 ridge-pole, where they would remain a long time or until frightened 
 away. Low, stunted fir growths about the island almost always 
 contained several of these birds, who would i)lay at a veritable hide 
 and seek among the dense clumps of fir and spruce everywhere 
 abundant. If any one pretended to watch them they woukl hide, 
 and not even cliirping remain for nearly half an hour (juite stfll, 
 or if stirring at all doing so so cautiously as not to attract the least 
 attention, while often I would walk about quietly trying my best to 
 get a sight at them and yet unable to do so. At other times the 
 
WHITE-CROWNED SPARROW. 
 
 79 
 
 case would be exactly reversed, and I have found them in the 
 bushes on the mainland near by, and spent much time in sitting 
 down and quietly waiting while the little fellows, at first frightened 
 by my presence, afterwards gained confidence and crowded around 
 the bushes close to my feet, my hands, and my face ; they 
 did not, as a certain author once remarked of the black-capped 
 chickadee, alight on the tip of my boots and peer at me, but they 
 did peer at me from bushes within a few feet of my face, and certainly 
 within reach of my hand had I grasped for them, — though I 
 should undoubtedly have missed them had I attempted it. After 
 remaining about for some time they would fly away, only to light 
 on some bush near, or return again, to examine more fiilly the 
 mysterious object which had attracted their attention ; then they 
 would chatter away to themselves as if romj)aring notes. If finally 
 I raised my hand or my head they would dart off to some coNcrt 
 and disappear as if by magic. This chickadee is often the only bird 
 that may be seen in the woody portion of the "lake regions," where 
 in ihe heat of the day it is active and full of song, even in the 
 summer. The darker the woods, or the more tangled the under- 
 brush, the better thee birds seem to like it. It is here the counter- 
 part of our native black-cai)ped titmouse {Pants afriiiij>i//us). 
 That latter bird did not api)ear here (hiring the year to my knowl- 
 edge, although it is given by Reeks, in his list of Newfoundland 
 birds, as common and breeding. The Hudsonian titmouse breeds 
 all up and down the coast, but it usually jirefers the inte-ior md 
 tangled undeTgrowth district, which is (]uite difficult ot access. 
 The ordinary note of this titmouse adds a very ])retn' and plci-in^ 
 lisp to the I lee. (iee, dee notes which it so trf«iuentiy utii.-- 
 
 :iul 
 
 1 ) I ■! 1 ri I y ;■! -n ( 1 
 
 l,..,,-lv 
 
 which our bird usaally delivers s(^ 
 
 I alsio saw SL'venl while (TOW- , \\ . 
 and though they were not rc-iuark..: .\ i.inic 1 
 in large numl)?!^ bree'^injr in thf low everi: 
 the island. They have a prcm' song, anian 
 companions, but I will say mor" >r tiicani 
 
 Thursday the 7th. Most ot ,ie day w* spent m: 
 
Wr^ 
 
 80 
 
 SANDPIPERS — SAVANA SPARROW. 
 
 i 
 
 with my gun, and several species of birds rewarded my search which 
 will be spoken of in their place, but my attention was esi)ecially 
 drawn to the small (locks of the ruddy plover or sanderling ( Calidris 
 arena rill), which were occasionally seen alone or with large flocks 
 of otlier shore birds. I saw numbers of them during my stay on 
 the island, but seldom many at a time. They are very wild and 
 hard to approach, and kee|) tjuite close together in small flocks of 
 from ten to thirty ; their flight is wilder and their call different from 
 that of the other birds with which they associated. I found them 
 very plump and flit, and they make very nice eating. 
 
 Friday the 8th. I saw to-day several very interesting si)ecies of 
 birds, and while we are on the subject I will give you a brief 
 account of the savanna sparrow {Passerculi/s sai'anna), and the 
 white rumped sandpiper {Trinj^a honaparki). The savanna si)ar- 
 row is ])erhaps the most abundant of all the smalt land birds that 
 inhabit these regions. It is a tame and familiar little fellow, and 
 feeds without fear about the doorstei>s, and in the dooryard, 
 building its nest, laying its eggs, and rearing its young often in 
 grassy clumps not two rods from the door. They are common all 
 over the islands and on the mainland, and their song is a well 
 known attraction to a native of the place. I shot a good many and 
 found tliem to present an unusually decided shade of plumage, 
 with the dark and white colors plainly marked. There was very 
 little yellow about the head and eye, and of some twenty specimens 
 none at all on the wing shoulders. I shot, one day, four of these 
 birds, none of which had a particle of yellow upon them anywhere 
 that I could distinguish ; a small tuft of white feathers at the base 
 of the primary coverts of the shoulder gave the api)earance of a 
 white edging in the i)lace of the usual yellow. The birds were all 
 remarkably full in coloration, and decided in plumage; the white 
 very clear ; the dark inverted arrowi)oints (juite distinct, as were also 
 the grayish and buff edgings everywhere. One specimen alone 
 had the buffy suffusion covering the breast. I cannot say that the 
 rule holds good constandy, but in some thirty specimens the $ 
 had the yellow on the wing shoulder, while the 9 '^^'^ young of 
 
WIIITE-RUMPED SANDPirER. 
 
 81 
 
 the year of cither sex had white in that place. It is everywhere 
 abundant, and breeds on the ground. 
 
 Of the white-rumped sandi)iper I saw several immense flocks on 
 the flats near the house ; they were quite tame, and I shot a great 
 many of them. Some had the chestnut edgings of the wings very 
 broad and deep, while several of them had either the head or neck, 
 and one had both, quite ashy ; the greater part of them had very 
 little chestnut, that color lieing replaced by ashy ; the chestnut 
 edgings seemed to be on birds that were passing from the last 
 stage of young of the year to adult birds, but I may mistake ; 
 both varieties were in the same flock — flocks were usually of from 
 fifteen to one or two hundred in number. They would alight in 
 the mud flats, and feed, running about in the black, slimy, clay-like 
 mud or muck, running in the water nearly to the tibia and keeping 
 quite close together meanwhile ; they feed in the evening and at 
 dusk chiefly among the kelp along shore, and I rarely saw even a 
 single bird at high tide. They were very tame, and if I crouched 
 and approached them on nearly their own level, I could get very 
 close indeed. If discovered, single birds, as often small flocks, 
 would remain perfectly (|uiet for some time. It was very difficult to 
 see them as long as they remained still, since their color corres- 
 ponded so closely to that of the mud or kelp where they happened 
 to be. I have often at dusk had them fly from within a yard of my 
 feet when after careful examination I had not supposed there were 
 any near me. I saw these birds associate only with Calidris 
 arcnaria ; the birds did not mix, but each kept in its own separate 
 flock. An occasional Ercunctcs pusillus was seen, but they were 
 rare. Sometimes a single bird would be seen flying high and swiftly 
 in the air, but generally their flight was low and irregular, their 
 note uttered as two or three faint, shrill whistles, the same as when 
 running about in search of food ; they take wing on the approach 
 of danger. Large flocks are made up of the union of a large 
 number of single families. Most of the specimens obtained at 
 this season of the year had a worn and faded look, and were not 
 nearly as plump or as well plumaged as specimens that I shot later, 
 6 
 
 
WT 
 
 ii« 
 
 82 
 
 BLACK-FLY. 
 
 with bright chestnut edgings to nearly all the upper feathers. One 
 of these s[)ecinK'ns hail the head and neck nearly clear ash, of a 
 very minute pattern. I often found specimens where the tail feathers 
 were half black, and the other half white. I suspect that Trin^a 
 bainiii, that rare sandpiper, bears a stronger relation to certain 
 forms perhaps of adult worn breeding plumage of T. bonapartei 
 than is generally credited. 
 
 We were much gratified, this evening, by a fine double rainbow, 
 of most beautifiil color ; both bows were well defined agaii>st a clear 
 bank of cloud in the east. 
 
 Sunday the i oth. I saw what was either a large white hawk or owl 
 perched on a stone near the top of a small hill by the house, but he 
 soon flew away. We passed a delightfiil day out of doors. The 
 weather was fine, and the lateness of the season protected us against 
 the fury of that most pestiferous creature of these northern regions, 
 the much dreaded black- fly. This insect, the Simiilium mokstum 
 of the entomologist, is one of those vara avis that the naturalist 
 lights ujjon occasio'ally much to his own personal inconvenience. 
 It is an abundant native of all these northern regions, more especially 
 inland, during the summer months ; and of its development little 
 is known. They fill the air during the months of July, August, 
 and September, especially inland, making the country almost unin- 
 habitable. They swarm on a still day in millions, and nothing but 
 a breeze of wind will rid the air of their presence. They light on 
 any part of the face that is incautiously exposed, and bite it most 
 teiTibly tmtil the blood runs. Their favorite plu.;es of attack are be- 
 hind the ear and on the neck. This fly attacks the children even 
 more savagely than it does the grown people. I have seen a child's 
 face all bitten and swollen, while the poor thing sjjent most of 
 the time crying and rubbing the inflamed portions, which only served 
 to make the case worse. With adults the bite soon disappears. 
 The account given of this insect by Harris, in his "Insects In- 
 jurious to Vegetation " is not accurate, for this portion of the coun- 
 try at least. I have found it more or less common from the 
 beginning to the end of summer. Prof. Packard says of it, "Its 
 
REMEDIES. 
 
 83 
 
 aiitennx are eleven-jointed ; the i)alpi are four-jointed, with long, fine 
 terminal joints, and the ocelli are wanting, while the posterior tibia 
 and first joint of the hind tarsi are dilated. 'I'he body is short and 
 thick. The labrum is free, sharp as a dagger ; and the proboscis is 
 well developed and draws blood profusely." Harris says, "The 
 wings are transparent ; and their legs are short, and have a broad 
 whitish ring around them." I have verified these statements, but 
 not the following, again (juoting from Prof. Packard, who says it 
 ** lives during the larval stage in the water. The larva of a Labra- 
 dor species which we found is about a cjuarter of an inch long, and 
 with the appearance here indicated.' The pupa is also a([uatic, 
 having long respiratory filaments attached to each side of the front 
 of the thorax. According to Westwood ' the posterior part of its 
 body is enclosed in a semi-oval membranous cocoon, which is at first 
 formed by the lar\'a, the anterior part of which is eaten away before 
 changing to the pupa, so as to be open in front, 'i'he imago is pro- 
 duced beneath the surfice of the water, its fine silky covering serving 
 to repel the action of the water.' " AUhough they are undoubtedly 
 blown by the wind to great distances, the fact of their being reared in 
 fresh water — by fresh I mean not salt sea water, yet I imagine they 
 prefer stagnant pools — accounts, perhaps, for such large numbers 
 being found on the mainland, and so few on the islands outside, 
 where the continued breezes from the sea carry them away as fast as 
 they appear. If it were not for these tormenting litUe insects, explo- 
 ration into the interior would be a comparatively easy thing. As it is, 
 veils of mosquito netting do no good, for the fly is scarcely one- 
 eighth of an inch long and woukl pass through the meshes at once ; 
 handkerchiefs are used with little better success, for they will creep 
 into the smallest openings and beneath the loose folds. A mixture 
 is now made and sold, consisting largely of dilute carbolic acid 
 and sweet oil, which, when applied to the face and hands, keeps 
 these insects at a distance, for a time at least. The best and 
 only sure remedy is to keej) wJioU}' away from them. 
 
 4 
 
 • See fig. 69, Packard's " Our Common Insects," p. 73. 
 
 ■ffb 
 
 !i 
 
IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 1.0 
 
 I.I 
 
 IIIM 112.5 
 
 -^^ IM |||||22 
 
 '" m III 2.0 
 
 1.8 
 
 
 1.25 
 
 1.4 
 
 1.6 
 
 
 ^ 6" — 
 
 
 ► 
 
 V] 
 
 <^ 
 
 // 
 
 A 
 
 "W/ 
 
 
 ^^ s 
 
 m 
 
 o 
 
 7 
 
 M 
 
 Photographic 
 
 Sciences 
 
 Corporation 
 
 '^3 WEST MAIN STREET 
 
 WEBSTER, NY. M580 
 
 (;•«■ 972-4503 
 
 ^ 
 
 iV 
 
 \ 
 
 \ 
 
 % 
 
 
 ^•^ 
 
 # 
 
 
 o^ 
 
 Ji 
 
;<" c 
 
 
 I 
 
 i 
 
iv — 
 
 111 
 
 . ! 
 
 ■ii 
 
 ] 
 
 84 
 
 HORNED LARK. 
 
 Monday the i ith. This morning I shot several specimens of the 
 horned lark {Eremopliila alpcstris), and noted the extent of the 
 pinkish color on the wing coverts, rump, and neck usual in spring 
 specimens in the high eastern regions. My long stay upon the coast 
 made me quite familiar with this beautiful songster, and characteris- 
 tic bird of the region, which is abundant all along the north shores 
 of the St. Lawrence from Quebec northward. At Green Island in 
 the river St. Lawrence, I found the lark quite common ; at that 
 time it was rather tame, and could be seen on the low flats of the 
 island hopping about and feeding in close company with the sand- 
 pipers ; they were all single birds and not flocks. I saw them all 
 the fall at Old Fort island, both alone and in large flocks, and al- 
 ways more or less wild. I would often see them flying very high in 
 the air, and uttering their peculiar querulous whistling notes ; some- 
 times flying quite low and somewhat irregularly, but uttering their 
 notes at all times while on the wing. Though common everywhere, 
 they seemed to lead a sort of wild, solitary life that comported 
 well with the wild, solitary region in which they dwelt ; they pre- 
 ferred the plains, fields, and rocky knolls away from houses where 
 they would hop about in twos or threes, or small flocks, pickinf; up 
 their food ; occasionally they would perch on the tops of knolls as 
 if to reconnoitre, then retire and go to feeding again as if satisfied 
 that no enemy was near ; they are very quick and active in their 
 movements, and always wild rather than tame. I saw them often 
 on the shore, and feeding on the kelp in company with the white 
 rumped sandpiper, but never mingling with them as far as I could 
 see. As their brown color corresponds so well with the color of 
 the ground, it was often hard to detect them until a few shrill whis- 
 tles and a hurried flight announced their flushing a short distance 
 ahead. Several that I shot were remarkable for the amount of pink 
 upon them. At times large flocks fly over the island high up in the 
 air, while one of these flocks once alighted upon the island : their 
 extreme wildness was something remarkable. One can hardly say 
 enough of this most charming little fellow and beautiful songster. 
 It breeds here abundantly, and the only time I saw it tame was at, 
 
 I 'i 
 
RUBY-CROWNED WREN. 
 
 85 
 
 and immediately following, the breeding season. The nest is 
 placed on the ground, and the eggs, usually four in number, white 
 peppered with ashy in a close pattern. It enlivens the long days 
 with a fresh, and what we would call really remarkable "clean cut" 
 song that, heard at early morning or in the evening, would induce 
 one to believe the bird almost inspired. 
 
 Often on a clear, crisp morning I have seen the lark ascend 
 by a series of spirals to an immense height : then, remaining on al- 
 most stationary wing, carol forth such a thrilling warble that it 
 seemed more like the chant of a spirit than the song of a bird. 
 
 In the afternoon I shot perhaps the most cunning bird known 
 about these regions where it is probably r.ever common, the ruby 
 crowned wren {Rcgulus cakndulus). It was flitting about in a small 
 clump of bushes when I first saw it and it me. I was then obliged to 
 wait around for over half an hour before I could again catch sight of 
 and shoot it. It will cunningly crawl or flit from place to place, and 
 it is a mere question of who will continue this game of hide-and-seek 
 the longest, as to whether the bird escapes or is finally obtained. 
 
 One cannot but enjoy these rambles about a "new region ;" at 
 least a region about which so little is known as this. There is a 
 charm of novelty of situation, and to one busy hunting the delight- 
 ful natural objects with which he is surrounded, the charm is height- 
 ened. I was often reminded, especially on the water, of the enthu- 
 siasm and joy with which Audubon must have viewed these abodes 
 of the water-fowl when entering this new field of ornithology for 
 the first time. How a botanist would have revelled in the new 
 plants that would greet his eye ! and there remain many yet that 
 have not been identified ; while I often think of the field for some 
 enthusiastic collector of lichens, fungi, and mosses. The insect 
 fauna of the region likewise needs studying up, and an expedition 
 into the interior, like one through Newfoundland, would develop 
 most interesting results. 
 
 Tuesday the 1 2th. We took a trip inland this afternoon, and I 
 could not help admiring the beautiful bay which bears the name of 
 the island at its mouth, and which receives that name from being 
 
 Nh 
 
 \ 
 
 \'^ 
 
 
 1 '' i 
 
 
 
 
 
 
 ■ V 
 
 
 ■ 
 
 
 } 
 
 
 
 
 I 
 
( 'I :'l 
 
 86 
 
 OLD FORT BAY. 
 
 (ifilll 
 
 : 'i ^% 
 
 in 
 
 the supposed seat of the old French and Indian wars traditional 
 to this locality. Old Fort bay is really a most beautiful piece of 
 nature's fancy work. Just outside its mouth a large expanse of 
 water is surrounded on all sides by islands, between which narrow 
 channels, in many cases too small and shallow for the passage of 
 vessels, connect with the :;ea. Old Fort island, the largest here, 
 except Esquimaux island, as it is the largest on the coast for miles 
 around, is in a line with the mouth of the bay. The open water I 
 have mentioned is relieved by a single rock prominent at high tide 
 even ; all about is deep and safe water in which vessels may anchor. 
 
 As you enter the bay, the hilly character, at least of this part of 
 the coast, attracts your attention at once. The hills between three 
 and five hundred feet high at the left send their wooded or precip- 
 itous slopes down to the sea, while the highland extends far back 
 in series of crests. At the extreme end of this left hand promon- 
 tory, and midway between the front face of the hill, stands the 
 dark mouth of a sort of cave or natural hollow among the rocks, 
 within which tradition stores treasures inestimable, guarded by 
 apparitions morst hideous and terrible ; below it a grassy slope 
 extends slanting towards the sea ; above it on the right, the crest 
 of one of the hills forms a good outlook to the whole surrounding 
 region. 
 
 It is here that many affirm was the last standing place of a fort or 
 battery which the settlers inside the bay (and the ruins of whose 
 houses at a late date, though nothing now remains, were plainly 
 seen,) were credited as having erected for the better defence of 
 their abodes. The theory that the battery would be raised in such 
 a place is very probable, and it would be quite likely that a battery 
 would be placed outside the harbor rather than within it. I men- 
 tion this as there are some persons who believe the fort to have 
 been within the bay on the top of a long glacial ridge, which if it 
 were not known to be of nature's own formation, would strike a 
 stranger, especially one not examining the height personally, as 
 being the most likely place for such a fort. 
 
 Inside, on the right hand, the elevations are low, seldom over 
 
 I, 
 
TOPOGRAPHY OF COUNTRY. 
 
 87 
 
 one hundred feet in height. The country here extends back, like 
 that of the other side in a series of elevations. The basin of the 
 bay is like that of a fiord valley, and extends directly inward for a 
 short distance, then bends in an almost easterly direction. From 
 east to south the land is low, while from north to west and south 
 the hills are often six hundred feet high, and a more or less level 
 piece of country extends back in a west direction for a consider- 
 able distance. Towards the north the country is broken, as it is 
 all along, though not to so great an extent as in this direction, into 
 ponds and even in one instance a large lake. Labrador is known to 
 be a region of lakes in its interior. Perhaps it is more so than many 
 imagine. In some cases large lakes cover quite an area of coun- 
 try. They appear in such cases to be the resultant drainage of an 
 innumerable number of ponds, large and small. Thus these ponds 
 are ranged in areas of e(|ual height, the highest ones being the 
 smallest, while next will come an extent of country that contains 
 several larger ones, and so on, each draining into the larger below 
 and all finally into one lake. The general idea is that of a huge 
 natural amphitheatre wherein the seats are tablelands extending 
 back for several miles, so that the top layer or seat consists of several 
 isolated and high peaks whose connecting medium is a marshy bog 
 with small puddles of rain and occasional!;/ springs of water ; about 
 fifty feet lower an area of country extends inward, towards the cen- 
 tral pit of the amphitheatre, which contains several medium sized 
 ponds with a drainage from the pools above, and with innumer- 
 able tiny rills flowing into them, and receiving more or less drainage 
 from within its own districts. Below this still another series of 
 ponds, much larger, receive drainage from the waters above, and 
 empty themselves below either into another series or, uniting in 
 some large pond, receive all this drainage and flow onward and 
 forward, through valleys and rocky gorges, with precipices and 
 rocky clift*s on either side, miles into the country. After some 
 distance this again will unite with another large pond which 
 receives a similar drainage from some other direction ; thus, the 
 whole country presents an app^iarance most unusually grand and 
 
 
ffl 
 
 T 
 
 88 
 
 "HEIGHT OF LAND." 
 
 iiil 
 
 n, 
 
 ■\ Hi! 
 
 *„'!i 
 
 wild, fit region only for the Indian and trapper, where the otter, 
 beaver, and martin form the staple productions, and the flesh 
 of deer and bear the staple consumption. 
 
 About two hundred miles from the sea the "Height of Land," 
 as it is called, presents itself as a series of high hills, and really a 
 part of that great chain of mountains which one strikes directly 
 north from Quebec and which follow the country across the Labra- 
 dor peninsula to above Cape Charles on the eastern part of the 
 coast. 
 
 As this series of hills runs along the country at varying distances 
 of from two to five hundred miles from the coast, the intervening 
 region consists of a descending series of elevations, until directly 
 on the coast line itself it assumes a height of from three to seven 
 hundred feet. The "Height of Land," often contains hills from 
 looo to 20OO feet in height, and while few are over that, Mt. Bache 
 was found by the U. S. Coast Survey to be some 2 1 60 feet high. 
 According to Packard, " Its watershed " (that of Labrador) " is 
 said by Kohlmeister and Knoch, to be a chain of high mountains 
 which terminate in the lofty peaks of syenite at Aulezavic island and 
 Cape Chudleigh, which are the highest mountains in Labrador, and 
 rise probably over 3000 feet in height." 
 
 The river system of the peninsula is restricted to one or two 
 principal rivers ; while the stream system is ample, nearly every small 
 bay having for its head a body of running water — and the numerous 
 streamlets all along the coast fully support the view that there is 
 an abundance of water inland. Let us return, now, to our descrip- 
 tion of Old Fort bay, and having chosen this, of the many similar 
 places along the coast, see what beauties it possesses and what nat- 
 ural attractions it yields. 
 
 The outline of Old Fort bay, is, as I have said, a row of hills 
 three to five hundred feet in height on the left, and another and 
 smaller series of elevations scarce one hundred feet upon the right 
 of the entrance of the bay. Directly across the ridge to the right, 
 which is in fact a much twisted, contorted, and irregularly cut up pe- 
 ninsula, lies the western arm of Esquimaux river, whose mouth you 
 
 i, I, 
 
OLD FORT BAY, 
 
 89 
 
 remember is so nearly occupied at this point with Esquimaux island, 
 that the passage betvvt^en it and the mainland on the east is barely 
 large enough for small boats. The bay extends directly inward about 
 a mile, then takes a sudden turn to the right for about the same dis- 
 tance, giving to the whole the shape of a bent arm. Strange to say, 
 no stream of any size enters this large bay. A small biook only, 
 made up of the drainage of the neighboring hills, flows into it. 
 To give you an idea of the extent to which the country is cut up by 
 ponds, let me say that the distance between the end of the bay and 
 the Mission station, situated several miles up the river, near the 
 contraction in the stream just below the first rapids, is only five 
 miles ; yet there are three ways of going between the two places, 
 and each time a line of three distinct ponds is traversed, and with 
 the exception of one pond which is the first and same crossed 
 on two of the routes, no pond is traversed twice ; between all 
 of these are elevations varying often from two to three hundred feet 
 in height, though generally less than one hundred. The whole 
 country is similarly cut up into elevations and depressions whose 
 basins contain a pond, the accumulation of the drainage surround- 
 ing it. 
 
 A single stream en<"ers the bay from the centre of the left hand 
 side of the entrance ; it comes from a line of ponds extending 
 backward into the country, and I am informed that one of them is 
 an excellent trout pond. The most interesting point in all the bay, 
 the elbow as we will call it, remains to be spoken of. If we describe 
 it as having the appearance of two semicircles placed side by side 
 ^-v.^^ and with an island just off the joint, Ave shall present a very 
 
 * fair picture of the location. The larger of these curved 
 pieces is the elbow proper. The left hand (the larger) is a sandy 
 strip of beach with high cliffs about it, and a sloping ridge whose 
 stimmit is a straight line of nearly a quarter of a mile in length 
 above it. This elevation slopes gradually to the sea. Above, it 
 presents the appearance of an artificial fortification lying between 
 two high peaks, one at the left and the other at the right. The 
 top of this ridge, nearly level for some thirty or forty rods, falls 
 
 n 
 
90 
 
 BACK OF THE BAY. 
 
 II » 
 
 4 
 
 I ; 
 
 
 :'l!l 
 
 'M 
 
 ' ' .'(il. 
 
 gently to a pond beyond. The pond lies about 225 feet above 
 the sea level, while the top of the ridge is about 275 feet, thus 
 making a difference of level between the two of fifty feet. The 
 slope of the bank in front is some thirty degrees. It is evident 
 that an immense glacier, receiving arms it may be from several 
 quarters of the country in the interior, made its stand here, 
 and thence crept down into the sea polishing off this ridge, and 
 scooping out the basin of this part of the bay. It is along the 
 bank of tlie eastern part of this cove that the old colonists chose 
 their abode, and fixed their huts. At a late day foundations of 
 these were still standing, and I have been informed that there are 
 those now living who have dug up old copper coins, and relics 
 of various other kinds about this same place. No signs of either 
 relics or foundations are now to be found, although the most 
 careful search has been made all about these regions. The second 
 circular division of this beach presents a totally different aspect 
 from that just described. Its central point is a mass of rocks, and 
 rocks continue in greater or less profusion all along to the end 
 of the bay, though especially abundant right in this particular place. 
 There is a raised bank or platform, about one-quarter of a mile 
 long, and thirty or forty rods deep, through which flows a sniall 
 stream which has evidently eaten its way to and gradually sunk 
 within a few feet above high tide, while the bank itself is scarcely 
 twenty feet above the sea. Back of this are the hills, from three to 
 four hundred feet high. Close on the rocky shore the kind people, 
 who so cordially entertained us at Old Fort island, have their 
 "winter quarters," while another family occupies the other side of 
 the point. The little island opposite the house is one huge mass 
 of rock, covered with scanty vegetation. 
 
 A remarkable formation exists just east of the house. You can 
 pass by the large rock at the left of the building only at low tide, 
 when you come to a sand cove and sand bar, the latter bare at 
 high tide, that connects with another rocky mass, which, extending 
 towards the sea, corresponds in its nature to the island above 
 mentioned ; all around is rock and mud, yet this litde strip of sand 
 
ARTIFICIAL POND. 
 
 91 
 
 is prominent and peculiar in its situation. The sea washes close 
 up to it upon either side but does not cover it. A little way back 
 is seen the slope of the embankment terrace and between the two 
 the remains of a most curious structure. Here is represented the 
 last remaining traces of the old colony, in the shape probably, of a 
 concealed dock for boats and small vessels. It is an artificial pond, 
 and. covered from all observation until this part of the bay is reached, 
 its presence would never be expected unless looked for. A slight 
 sand bank in that corner of the bay had so shut off this inlet that 
 at low tide a natural basin of water was separated by this same sand 
 bank from the water of the bay. This basin shows signs of being 
 enlarged, deepened, and has an embankment around it the walls 
 of which are kept in place by huge stones ; these one can plainly see 
 have been worked and built up around it. The stones are carefully 
 arranged so that the bases are downward and the i)ointed or smallest 
 end uppermost, while the interspaces are filled up with smaller stones 
 and earth : the whole top is earthed over and covered around with 
 a layer of large flat stones. This appears to preclude the possibility 
 of the whole affair being a production of nature. Sheltered as it is, 
 it would make an excellent hiding place or repairing place for 
 small boats, which could easily make way over the bar at high water. 
 Another peculiarity, however, presents itself; there is, at low tide, a 
 drainage from the small basin into the sea through this sand bank. 
 Were there now no inlet to this small pond, as in fact there 
 is, it would be drained twice a day while the tide was low, since 
 it is very shallow at best. Advantage has been taken of this fact, 
 apparently. A short passage has been made from the northeast 
 side so that the brook, which I mentioned before as flowing in a 
 channel through the embankment beyond, is somewhat slightly 
 diverted so as to run into the basin. Now the fresh waters of this 
 brook, running in at all times with a stream of several times the 
 volume as the outflow at low tide, are continually freshening the 
 waters in the basin, until before low tide is reached you have an ar- 
 tificial pond of clear, fresh water with sandy bottom and suitable 
 for the cattle who come to its banks regularly to drink, and even 
 
 
 1 ?> 
 
 I 'i 
 
 1 
 
WTT 
 
 92 
 
 LONG -TAILED DUCK. 
 
 
 !' 
 
 ! I 
 
 I, i 
 
 - H. 
 
 the family supply is easily obtained quite pure enough for family 
 purposes. The pond would therefore supply a large number both 
 of cattle and of families. Though I have seen no evidence either 
 for or against the theory that the early colonists had cattle with them, 
 it seems to me at least very probable that such were, as I have sug- 
 gested, the intentions of the people in making this peculiar structure. 
 I will not now speak at length of the curious superstition of some 
 of the people whom I have seen, that this structure in some way con- 
 ceals large pots of gold left by the French, but will leave this and 
 other superstitions of a like nature for some other time and place, if 
 indeed they are worth mentioning at all. 
 
 Though there is litde or no beach, properly speaking, the land 
 slopes down quite close to the sea in one or two places ; these have 
 been chosen for building spots and houses erected accordingly. 
 Back of the houses all is a dense mass of low, tangled spruce and 
 fir, extending on and over the hills some distance into the interior. 
 
 I must here yield to a natural desire and describe to you two 
 new species of birds which were procured here. The first, the 
 Canada or tree sparrow {Spizc/la monticold) which is rare except 
 in its migrations, apparently ; the second, the long tailed duck 
 {Ha?rlda glacia/is). The latter bird is called the ^^coc-cait-wee" 
 from the sound made by the males which resembles the pronunci- 
 ation of these words. Another name for the birds is that of 
 " hounds." The female resembles the female of another bird called 
 the dipper or buffle head, sometimes the butter ball and spirit 
 duck, but which is here called '* sleepy diver," from the slowness of 
 its movements in the water. The marked difference between the 
 two is the absence of white on the wing of the long tailed duck ; 
 whether the young birds of the two species are distinguishable or 
 not I could not ascertain. The people here cannot tell you which 
 is the sleepy diver, as they call it, and which the female of the long 
 tailed duck, or even distinguish the young, but call them all indis- 
 criminately sleepy divers. It will never do for a stranger to tell 
 them they are wrong ; they think that you are the one wrong in all 
 cases, and you cannot possibly convince them that a stranger can 
 
BIRDS. 
 
 98 
 
 learn any more of any object connected with the region than the 
 peoi)le of that region. 
 
 Wednesday the 13th. I shot a bird to-day that has a most peculiar 
 name in the vocabulary of the natives ; it is called by them the 
 Nan-cary^ pronounced as if spelt nan-sary; it is in fact the greater 
 yellow legs {Totanus fnchinoleucus), and is more or less common 
 in the late fall either singly or in small flocks along the beach with 
 the small sandpipers and plovers. 
 
 Thursday the 14th. It seems as if I did little else but describe 
 birds ; but as there are more of these than of any other animals in 
 the place, and as I have attempted to follow the outlines of my 
 journal, I hope that in doing so I shall not render tedious a sub- 
 ject delightful in itself, but for which others may not care so much, 
 while the omission of which would disarrange what I regard as 
 the natural order of a series of articles of this kind. I will then 
 tell you of still another new and somewhat rare species of bird, 
 captured to-day, called the Lapland longspur {Pkctroplianes lap- 
 ponicus). The development of the hind claw of this bird is some- 
 thing remarkable, often reaching three-fourths of an inch or over. 
 It is found either alone or in company with the immense flocks of 
 snow buntings so common at the approach and departure of win- 
 ter. The bird was feeding at dusk near the kelp on the shore and 
 with several others probably of the same kind, — their flight and 
 notes were like those of the shore lark so common here. It was 
 rather wild, and was either a female or young bird. Though having 
 seen the bird several times to my certain knowledge, I only suc- 
 ceeded in shooting a single specimen. 
 
 The wind has blown pretty hard all day. It seems no unusual 
 thing for a wind storm to set in and last several days here, while 
 often the wind blows with such force that it is dangerous to at- 
 tempt any sort of navigation. Strange to say an Indian canoe 
 ("cranky" as it is generally regarded) will "ride " with apparent 
 safety where any other boat would be swamped in an instant. 
 
 Though the wind was not really severe, this afternoon, we had 
 hard work to manage the boat to and from Old Fort bay, where we 
 
 M 
 
 ^Ib^^ d£iLL'^':^:^l^iLj&^'.t.(U^IL^i5&. 
 

 It! li 
 
 94 
 
 RAVEN. 
 
 illMl 
 
 if if 
 
 went to tak? a load of articles to be transferred to "winter quar- 
 ters." The cold weather is rapidly coming on, and we must be 
 prepared for it by moving into our winter home before very long. 
 It is a desolate looking region, I assure you, but we will try and 
 make things as comfortable as we can for all that. We have had 
 several ravens hovering about the fish stage to-day ; the people seem 
 to regard them as birds of ill omen, and say that they are in league 
 with the devil. You can rarely get any of the natives to shoot 
 at one of these birds no matter how near they come, and they 
 seem positively afraid of the results of so doing, fearing that it will 
 bring them misfortune for the remainder of the year. The bird 
 is really a very difficult one to shoot. I have often lain in wait 
 for it with my gun, firing at it when both at rest and on the wing, 
 at a very short distance off, and had it raise its huge black wings 
 and fly slowly off with a harsh and hollow croak as much as if 
 to defy both me and my gun. I have wasted more extra large 
 ducking charges at the raven than almost any other bird, and have 
 seen the least results. The bird itself is very common everywhere, 
 summer and winter ; breeding on the high cliffs and hill tops, 
 and remaining about wherever there was any putrid flesh. It ap- 
 parently loves to walk or fly about on the tops of the hilly crests 
 on the mainland, and on the trees near the frozen bays in winter. 
 It frequents the seacoast, and is common about the inland ponds 
 and lakes. It replaces the crow, which rarely though occasion- 
 ally is found in these regions. The scientific name for the raven 
 is Corvus corax. 
 
 Friday the isth. I became quite well acquainted to-day with 
 the pigeon, as it is here called, otherwise known as the black guil- 
 lemot ( Uria gryllc) . This little bird is one of the most abundant 
 of the waterfowl next to the eider ducks and Murres, that we 
 have upon the coast. Near St. Augustine we saw this bird for the 
 first time though it is found all along the Atlantic coast, as far south 
 as New Jersey, growing more and more rare as it approaches the 
 latter place where it was found in winter. I have seen them 
 everywhere in the waters in and about the islands, though never 
 
"PIGEON" GUILLEMOT. 
 
 95 
 
 very far from land, from the opening of the bay in the spring until 
 the ice closes the last open waters early in December. I have 
 found several stages of plumage (referal)le to the ages) of this bird 
 which takes three years to mature. A very extraordinary form 
 marks the second year's growth. The whole plumage is inky 
 black both below and above, and with white blotches imperfectly 
 rounded, the size of an ordinary thimble head, scattered irregularly 
 all over it above and below ; the bill is blackish carmine, the legs 
 and feet dusky carmine. The wings with a pure white patch as 
 usual. I think the white tail coverts were present, but am not 
 sure on this point. I cannot learn that this i)lumage appears at 
 any other time than in the fall of the year ; tm ,• are rare here, and 
 apparently pass this stage in some wild place oi region where they 
 are not easily detected. The hunters abor* the coast told me that 
 they were rare. 
 
 lu uie early fall the pigeon is quite tame, but grows wilder as 
 the cold weather advances. When pursuing them from a boat 
 they arc at times easy to approach while at other'- difficult and very 
 wild without any apparent reason. It will usually dive " at tlie flash," 
 but often when feeding it allows you to come quite near. In feed- 
 ing, the bird bends its neck fonvard and dips its beak into the wa- 
 ter ; at this time, when the head is turned forward and a little away 
 from the hunter, he is generally sure of hitting his game. Often 
 the pigeon takes wing nearly as soon as it perceives a boat ap- 
 proaching, and it is then impossible to get anywhere within shoot- 
 ing distance of it ; the flight is then rapid but easy, generally 
 low and straight. \\ hen tame it usually escapes by diving " at the 
 flash " and swimming a great distance under water, easily and in 
 any direction. When wounded, they often dive, and, I think after 
 the manner of many ducks, swim to the bottom and clinging to 
 the seaweed die there. I have often watched them dive thus, but 
 have not seen them arise. They stay around near the land feed- 
 ing, on still warm days, often many together. In flying low over 
 the water, if fired at I have oft'"n seen them suddenly drop down 
 and dive, disappearing almost instantly. The flesh, especially of 
 
 
 ■i !: 
 
 S 1 
 
'*'' I 
 
 ^^•n 
 
 fT 
 
 96 
 
 MOVING TO " WINTER QUARTERS." 
 
 .■ !' 
 
 r 
 
 ill 
 
 
 I 
 
 -{f>< 
 
 
 the young birds, is * xcellent eating and they are shot in great 
 numbers ; it is called the hardest bird to kill, next to a loon, that 
 inhabits the waters of the coast. 
 
 The pigeon breeds in large numbers on several of the small isl- 
 ands along the coast. On one small island colonies of this bird 
 breed exclusively. They lay usually three eggs in some exposed 
 situation or in the cleft of some rock, making no nest, and let 
 the sun do the greatest share of the hatching ; they are oblong 
 and ovoid in shape, tapering suddenly, the ground color being 
 from greenish to pure white, and the varied streaks and blotches 
 or spots scattered more or less thickly all over their surface, es- 
 pecially so in a ring around the top of the egg, are of black, or vari- 
 ous shades of brown. Nearly all the birds of this family have what 
 are apparently purplish spots, but these are black primarily and 
 ajDpear purple only from a slight covering of the white lime of the 
 shell itself. 
 
 The weather had now become so moderated that the next few 
 days were employed in "moving in," that is, in transferring the 
 household goods and utensils together with the people and live 
 stock into their abode for the long, coming winter. The house was 
 much like the one we were about to leave, but rather more com- 
 pact and a great deal warmer. 
 
 With the exception of a species of short-eared owl, peculiar to 
 the region, no particularly new birds had been obtained. Though 
 owls are generally regarded as rare in this region, I believe them 
 more common than usually supposed, several of the brown colored 
 species having been observed. In regard to the one mentioned, 
 it was shot by one of the men who said that about dusk the bird 
 attacked him and he could not drive it away until he had put the 
 whole charge of shot through its body, and so badly blown it to 
 pieces that I was unable to do anything with it but save a few 
 feathers by which to determine the species. It appears to be an ex- 
 traordinarily dark variety of our common short-eared owl {^BracJiy- 
 otus paliistris). 
 
INDIAN TENTS. 
 
 97 
 
 CHAPTER VII. 
 
 Indian tents — New fields for research — Visit to the Indians — Seals' flesh 
 — Dogskin boots — Cattle food in hard winters — Coptis trifolia — Spruce 
 Partridge — Inland — Hypothesis of Aurora — Little Auk — Signs of 
 wreck — Ascent of the western arm of the bay — Wreck of the Edward 
 Cardwell — Picking up lumber — First snowstorm of the winter. 
 
 
 We had hardly entered the bay before we saw in the distance 
 the mishwaps of the Indians, who had chosen this for their camp- 
 ing place. The Indian mishwap, or tent, is a peculiarly arranged 
 structure, and suited only to the wants of the people who oc- 
 cupy it ; being the usual, movable dwelling place of this nomadic 
 and roving people. It is generally called by the name of wigiuam, 
 and is the same, very nearly, as the tent-like structures that go by 
 that name in western North America. It is composed of long, 
 thin, rounded sticks that have been hardened by charring in the 
 fire, and which are set about in a circle whose ground diameter 
 is some ten feet ; the tops loosely put together overlap each other 
 a foot or eighteen inches. Around this are layers of birch bark, 
 and over the bark more sticks are placed so as to fall each one be- 
 tween the other two all around the outside. All holes are then 
 patched and covered ; while the top remains loosely open to allow 
 the smoke and bad odors to escape. A small opening in front 
 permits the inhabitants to enter or leave by stooping very low in- 
 deed. The peculiarity about these abodes, and the feature that 
 characterises them even when seen at a distance, is the plume-like 
 appearand; of the top, as it is thus constituted. I know of no 
 mere piling together of sticks that will produce this peculiar ar- 
 rangement. By it, these tents can be distinguished as far as they 
 7 
 
 \ m 
 

 ! 
 
 •* w 
 
 98 
 
 FIORD VALLEYS. 
 
 can be seen. That the structure may not be visible a great dis- 
 tance, it is generally placed behind some protecting ledge or rise 
 of rocks, though not always. 
 
 From the entrance of the bay then, as I have said, we caught a 
 view of the In-'-an mishwaps, backed by the verdure of slopes, 
 hills, ravines, ridges, and the various contour of a most uneven 
 background in the profile of the evergreen spruce tops, — which low 
 shrub is everywhere abundant outside as is the large tree inland. 
 
 On either hand a succession of hilly crests marked the boundary 
 of the bay, whose inner arm or bend extended far to the right. 
 Passing beyond the little island just off the central point, we came 
 in full view of the house, with its line of rocky and sandy beach 
 running around on either side ; here we were soon seated by a 
 roaring and crackling fire of good spruce wood. 
 
 Old Fort bay has much the general shape and direction of 
 so many of the bays in this vicinity, so often termed "fiord val- 
 leys-" They are long, narrow, for the most part shallow passages, 
 between the rocks running a little north of east, and evidently of 
 glacial origin. These little bays are quite abundant all up and 
 down the coast, and undoubtedly would, at least to one well versed 
 in such readings, describe a long and ancient story of glacial phe- 
 nomena and local disturbances. The whole glacial question will 
 finally, without doubt, obtain abundance of fresh, new, and useful 
 evidence by a careful study of this region, and it is a wonder 
 that no one has examined with a greater degree of care the whole 
 Labrador peninsula. Here is a vast, and untried field for explo- 
 ration. The North Pole is undoubtedly the ultimate end of re- 
 search in this direction, but I venture to affirm that no one will be 
 successful in that end until they have made a careful land explora- 
 tion of this keystone to polar investigation, the Labrador Penin- 
 sula — from seacoast inward to the "height of land," and thence 
 followed its great streams and leading trends of northward high- 
 lands. The greatest wonder is that as yet we have no definite 
 knowledge of this new and unexplored field. But to return : — 
 
 After a good supper we went to visit the Indians and Joe Mark, 
 
 
 I, :l 
 
INDIAN DOGSKIN BOOTS. 
 
 99 
 
 the sub-chief of this division of the tribe. We found him, to 
 speak plainly, in a state of beastly intoxication. He had come out 
 to sell his furs, and having received sufficient to supply his demands 
 for food he had used the surplus for drink, and that a Uttle too 
 freely, judging from the effects wrought upon his mind and body. 
 He was lying down when we entered, but he managed to place 
 himself in a half sitting and half lying attitude and carry on a con- 
 versation with a great deal of "talk-um"on our part and consider- 
 able muttering on his. We saw how matters stood, and soon came 
 away and returned home ; not, however, until we had obtained 
 part of a young seal, that some one of the Indians had recently 
 killed, and which we broiled on a spider for our breakfast the next 
 morning. Let me say here, that the flesh of a young seal, when 
 well cooked, resembles cow's liver so nearly that one can hardly 
 tell the diflerence ; and I believe that if I could place a dish of 
 each before a stranger in Labrador who had tasted neither for a 
 long series of months he would be unable to tell accurately which 
 was which, so nearly do they resemble each other. We also ob- 
 tained some dried deer's meat, which is soaked and fried, or eaten 
 dry, and is very good. 
 
 Monday the 25th. I put on, to-day, my first pair of Indian, 
 dogskin boots. They are made like long legged moccasins, the 
 foot part being made of sealskin (the top only of dogskin), and 
 the sole being soft and pliable, and as sensitive as a glove upon the 
 hand. The foot Is therefore free to move in most any direction, 
 and thus useful especially in climbing, where one is obliged to 
 grasp and cling by the clinch of the toes. It is quite trying how- 
 ever, to walk for the first few times on stony or pebbly ground, as 
 the little corners injure the feet most terribly until they become 
 hardened and accustomed to the peculiar feeling of having the 
 bottom of the boot soft and flexible instead of hard and stiff; but 
 one soon gets used to the change and then enjoys the freedom of 
 feeling with the foot. 
 
 Wednesday the 2 7th. I attempted a sort of exploration of the 
 country back of the hills, close behind the house, but found noth* 
 
 ■ if J 
 
 « I; 
 
 1. 
 
 11 
 
wm 
 
 
 100 
 
 GOLD THREAD. 
 
 '■LtltlMfcl: 
 
 
 ing but ponds and high hills, so that with a great deal of climb- 
 ing and going around to get a very short distance only, I was 
 obliged to return at dark having accomplished very little. 
 
 I made a curious discovery this evening. On returning home 
 from one of these short daily trips I noticed a large pile of heavy 
 brush lying over against the side of the stable door (why I had not 
 seen it before I cannot tell) , and upon inquiry found that it was 
 the remains of the birch and alder tops with which the cattle had 
 been fed the previous spring. It seems that, at this time of the 
 year, when fodder is scarce, the cattle are fed with the slender, 
 tender tops of these trees, and they are eaten with avidity and 
 apparent relish. 
 
 About this time I discovered the plant Coptis trifolia, or golden 
 thread, from the little, slender golden thread of a root which it 
 possesses. It grows in abundance in these regions and farther 
 west, and is plucked and sent to market in large quantities. It is 
 a mild tonic and treated with hot water is taken with impunity. 
 
 One day, early this week, one of the men brought in a spruce 
 partridge ( Tetrao Canadensis) , which he had killed in the woods. 
 These birds are usually very tame. They fly from cover and alight 
 in some br.sh, seemingly stupefied from being flushed, while I have 
 often known the young fellows to knock them over with sticks or 
 the ends of their gun barrels, without even taking the trouble to 
 waste powder and shot upon them. One day towards the close of 
 the week we amused ourselves by digging a species of clam (ques- 
 tionably Mya arenaria), which is found in the mud at low tide just 
 here, in abundance. They were excellent eating, and made a very 
 good soup. Strange to say, though they appear to be abundant 
 here, the people seem to care very little for them, and seldom dig 
 them. 
 
 Friday, November the 2d. Although this morning I climbed the 
 ridge back of the house, and over and about the place that many 
 of the people suppose to have been the location of the old times 
 fortification and fort, but which appears upon examination to pre- 
 sent few facts to confirm such an opinion j ard though I also went 
 
 ■ ' 'i: 
 
ELEVATED SCENERY. 
 
 101 
 
 4 
 
 >.\ 
 
 over all the nearer peaks on this side of the first pond which is 
 just over the ridge, I am in as much of a maze, apparently, as I 
 was before, as to their exact position and extent. From the top 
 of the successive elevations I could see pond after pond, and 
 ridges and gullies after ridges and gullies, stretching onward and 
 outward in every direction. A fair description of the country 
 around here would be a series of ridges composed of crests of 
 unequal height div^ided both lengthwise and crosswise by gullies, 
 with basins of water filling the intervening spaces or valleys. The 
 whole region is one grand network of ponds and hills. I had no 
 instalments with me for the purpose, and if I had possessed them, 
 am no map drawer. As has been before stated, a comparison of 
 several readings of the barometer gives the height of the external 
 ridge as 275 feet above the level of the sea to which it slopes easily 
 and naturally. The pond beyond this ridge presents a lowering of 
 fifty feet, while beyond this other ponds are situated correspond- 
 ingly lower, the large pond or lake being lower yet. The hills on 
 the left are most of them on a nearly average height — the two 
 highest being each 600 feet above the sea level, and all the other 
 prominent ones nearly an even 500. 
 
 If I mistake not, evidences of glacial and water action are locally 
 very abundant, while loose bowlders and stones are scattered spar- 
 ingly all along the top crests even, and small pools of water are 
 abundant here and there on the highland levels four to five hun- 
 dred feet above the sea. Here the soil is a black muck or mud, 
 and reminds one greatly of some mucky patch of salt marsh along 
 our eastern United States, exposed a few feet only at highest tide, 
 and transported to some shallow basin between two or three sur- 
 rounding peaks of similar height. In one instance, I found a large 
 pond only one hundred feet from the top of the highest peaks 
 which surrounded it. The gulches were frequent and* full of 
 streams, in some of which, especially in deep, shady gullies, were 
 remarkably sweet and cold waters. All these places are overgrown 
 with low, dwarf spruce and fir, or birch and alder. In one place 
 a small stream started from near the top of a high peak. There 
 
 i{p 
 
im^ 
 
 li tM- 
 
 102 
 
 AURORA. 
 
 was no chance for mere surface drainage, and there must either 
 have been some secret spring or sort of artesian well pressure, 
 forcing the water thus to appear on the very top of a rocky crest. 
 The rock around seemed to be a coarse granite, with occasionally 
 a place where several feet had been apparently scraped off and 
 ground to a fine powder of sand and small pieces of rock, of 
 feldspar principally, scattered all along over the surface of the 
 underlying rock. Occasionally, veins of feldspar or quartz appear ; 
 the latter mineral is rarely found in a pure state on this coast. All 
 the exposed rock appeared thus more or less scraped about here, 
 while occasionally the patch extended down the slope of the hill 
 one-thiid or even one-half the way to the bottom. There are 
 also many small rounded cones or knolls, whose top is one mass of 
 rough sand, with the rock only a few inches beneath. Another cu- 
 rious fact is that the beaches of most of the ponds have a portion at 
 least of their extent of this same material, the rocks in place lying 
 a little way from the edge only ; the rest of the beach is often of 
 a fine pulverized quartz and feldspar sand, totally different from 
 material of the neighboring rocks. Fine beds of clear, nearly straw 
 colored sand are not uncommon. 
 
 Wednesday, the 3d. We were fortunate in observing a most 
 beautiful aurora this evening. The barometer indicated 29.00, and 
 the thermometer 30° Fahrenheit. The sky was perfectly free from 
 clouds, and the air crisp. It first appeared in the east, then in the 
 north. The primary appearance was, as is frequent in such cases, 
 a semicircular band of light. In this case small pencils of light 
 floated about in the air. Very soon a thick, heavy band of several 
 longitudinal scroll-like narrower bands appeared ; the whole, resem- 
 bling very much a long ribbon, extended itself across the horizon 
 from these points, and apparently very low down in the atmosphere. 
 The whole band was in rapid undulating motion from the north 
 towards the east, and resembled the progressive onward motion of 
 a huge serpent. The appearance was quite striking and very 
 pretty. The band was extraordinarily dense, and very bright. At 
 first its edges were very clean cut and well-defined; gradually 
 
 Ml 
 
PROCTOR ON THE AURORA. 
 
 103 
 
 the light extended itself to the northeast, while the eastern end 
 enlarging and travelling faster than the northern, the mass soon 
 presented a singular and nearly circular appearance in the north- 
 east ; ten minutes after this massing together of the light the whole 
 disappeared, and the sky was entirely overcast with a light, thin, 
 luminous curtain of this mysterious, electrical, vaporous substance. 
 I saw many very pretty auroras while here, but none surpassed the 
 one just described. I found some of the inhabitants of this region 
 singularly superstitious regarding this aurora, in that they fully 
 believed that it danced to the sound of any musical instrument. 
 It was solemnly declared to me that if I should blow the flute, or 
 play upon the violin, the cloud would descend and dance in the 
 air just above me and out of reach. On the subject of this 
 curious phenomenon, Mr. Richard A. Proctor says : — 
 
 "One of the most mysterious and beautiful of Nature's manifes- 
 tations promises soon to disclose its secret. The brilliant stream- 
 ers of colored light which wave at certain seasons over the heavens 
 have long since been recognized as among the most singular and 
 impressive of all the phenomena which the skies present to our 
 view. There is something surpassingly beautiful in the appeamnce 
 of the thin "Auroral curtain." Fringed with colored streamers it 
 waves to and fro as though shaken by some unseen hand. Then 
 from end to end there passes a succession oi undulations, the folds of 
 the curtain interwrapping and forming a series of graceful curves. 
 Suddenly, and as by magic, there succeeds a perfect stillness, as 
 though the unseen power which had been displaying the varied 
 beauties of the auroral curtain were resting for a moment. But 
 even while the motion of the curtain is stilled we see its light mys- 
 teriously waxing and waning. Then as we gaze, fresh waves of 
 disturbance traverse the magic canopy. Startling coruscations add 
 splendor to the scene, while the nobler span of the auroral arch 
 from which the waving curtain seems to depend gives a grandeur 
 to the spectacle which no words can adequately describe. Grad- 
 ually, however, the celestial fires which have illuminated the gor- 
 geous arch seem to die out. The luminous zone breaks up. The 
 
 I 
 
 I 
 
TT 
 
 104 
 
 STRANGE HYrOTHESIS. 
 
 II 
 
 
 iHi. 
 
 scene of the display becomes covered with scattered streaks and 
 patches of ashen-gray light, which hang like clouds over the north- 
 ern heavens. Then these in turn disappear, and nothing remains 
 of the brilliant spectacle but a dark smoke-like segment on the 
 horizon. 
 
 " Hitherto the nature of the aurora has been a mystery to men 
 of science. Let it be premised, then, that physicists had long 
 since recognized in the aurora a phenomenon of more than local, 
 of more even than terrestrial significance. They had learned to 
 associate it with relations which affect the whole planetary 
 scheme. 
 
 " Arago was engaged in watching from day to day, and from year 
 to year, the vibrations of the magnetic needle in the Paris Observa- 
 tory. In Jan., 1 819, he published a statement to the effect that the 
 sudden changes of the magnetic needle are often associated with 
 the occurrence of an aurora." The statements are then given in 
 his own words, and from them the following deduction is made : 
 ^^Fro7n all this it appears incontestably that there is an intimate 
 connection belioeen the causes of auroras and those of terrestrial 
 magnetism^ 
 
 This strange hypothesis, was, at first, much opposed by scientific 
 men, but gradually it was found that physicists had mistaken the 
 character of the auroral display. It appeared that the magnetic 
 needle not only swayed responsively to auroras observable in the im- 
 mediate neighborhood, but to auroras in progress hundreds, or even 
 thousands of miles away. It has been found that a much closer 
 bond of sympathy exists between the magnetized needle and the 
 auroral streamers than even Arago had supposed. It is not merely 
 the case that while an auroral display is in progress, the needle is 
 subject to unusual disturbance, but the movements of the needle 
 are actually synchronous with the waving movements of the mys- 
 terious streamers. 
 
 " I may notice in passing that two very interesting conclusions 
 follow from this peculiarity : First, every magnetic needle over the 
 whole earth must be simultaneously disturbed ; and, secondly, the 
 
 ■ . i f 
 
I '. 
 
 ZODIACAL LIGHT. 
 
 lOS 
 
 auroral streamers which wave across the skies of one country must 
 move synchronously with those which are visible in the skies of 
 another country, even though thousands of miles may separate the 
 two regions. 
 
 " Could we only associate auroras with terrestrial magnetism, we 
 should still have done much to enhance the interest which the beau- 
 tiful phenomena are calculated to excite. But when once this asso- 
 ciation has been established, others of even greater interest are 
 brought into recognition ; for " — I take the liberty of italicizing for 
 emphasis this portion which is printed in Roman in the text — " ter- 
 restrial magnetism has been clearly shown to be influenced directly 
 by the action of the sun^ 
 
 " We already begin to see, then, that auroras are associated in 
 some mysterious way with the action of the solar rays. The phe- 
 nomena which have been looked upon for so many ages as a mere 
 spectacle, caused perhaps by some process in the upper regions of 
 the air, of a simply local character, have been brought into the 
 range of planetary phenomena. 
 
 " Most of my readers have doubtless heard of the zodiacal 
 light, and many of them have perhaps seen that mysterious radi- 
 ance, pointing obliquely upward from the western horizon, soon 
 after sunset in the spring months, or in autumn shortly before sun- 
 rise, above the eastern horizon. The light, as its name indeed im- 
 plies, lies upor that region of the heavens along which the planets 
 travel. Accordingly, astronomers have associated it with the plan- 
 etary orbits, and have come to look on it as formed by the light re- 
 flected from a multitude of minute bodies travelling around the sun 
 within the orbit of our earth." After a short account of the spec- 
 troscope and its use in analyzing substances especially those reflect- 
 ing light or luminous in themselves, he says : " Recently, however, 
 zodiacal light has been analyzed by Angstrom, with a result alto- 
 gether unexpected, and at present almost unintelligible. Its spec- 
 trum exhibits a bright line, and this bright line is the same that is 
 seen in the spectrum of the aurora borealisP^ 
 
 Furthermore : " Of all the phenomena presented to the contem- 
 
 
 !:! 
 
w 
 
 106 
 
 TAILS OF COMETS. 
 
 f 
 
 It 
 
 , 
 
 ;■ ■■(■;':' 
 
 \ fill ;' 
 
 J 
 
 
 plation of astronomers, the tails of comets are undoubtedly the 
 most perplexing. Now there is one feature of comets' tails that 
 has long since attracted attention, and will remind the reader of the 
 peculiarities common to the zodiacal and the auroral light. We re- 
 fer to the sudden changes of brilliancy, and the instantaneous 
 lengthening and shortening of these appendages. And the emi- 
 inent mathematician Fouler was led by the observation of similar ap- 
 pearances to put forward the theory * T/ia^ there is a great affinity 
 between these tails, the zodiacal light, and the aurora borealis.^ . . 
 It is far from being unlikely that these long vexed questions — the 
 nature of the aurora, that of the zodiacal light, and that of comets' 
 tails — will receive their solution simultaneously;" and he further 
 adds : " I had scarcely completed the above pages when news was 
 brought from America that the spectrum of the sun's corona, as 
 seen during the recent total solar eclipse, exhibited the same bright 
 lines as the aurora. Lastly, it has been found that the peculiar 
 phosphorescent light, sometimes visible all over the sky at night, 
 gives the same spectrum (very fliint of course) as the aurora and 
 the zodiacal light. What we learn certainly, therefore, from the 
 facts above stated, is this — that substances of the same sort emit 
 the light of the aurora, of the zodiacal gleam, the tails of comets, 
 of the sun's corona, and of the phosphorescence which illuminates 
 at times the nocturnal skies. But when once we have reason — 
 as in the case of the aurora we undoubtedly have — to associate 
 electricity with any particular form of luminosity, we seem clearly 
 justified in extending the explanation to the same form of lu- 
 minosity wherever it may appear." 
 
 Although I have already taxed your patience with long quotations 
 of such a strictly scientific character, I cannot conclude them 
 without giving Prof. Proctor's own deductions from this series of 
 arranged facts. He says : — 
 
 " I believe that the key to the whole series of phenomena dealt 
 with above lies in the existence of myriads of meteoric bodies 
 travelling separately or in systems around the sun. They are 
 consumed in thousands daily by our own atmosphere. They 
 
LIT TLE AUK OR SEA DOVE. 
 
 107 
 
 probably pour in countless millions upon the solar atmosphere, and 
 from what we know of their numbers in our own neighborhood, and 
 of the probability of their being infinitely more numerous in the 
 neighborhood of the sun, we have excellent reasons for believing 
 that to them, principally, is due the appearance of the zodiacal 
 light and the solar corona." 
 
 I have occupied much more time in the discussion of this most 
 interesting and highly instructive theme than I had intended, but, 
 as you can see, the phenomena dealt with could not have been ex- 
 plained with less. We will now pass on to some of the less sci- 
 entifically important, but still, we will trust, interesting portions of 
 our diary. 
 
 Just here a short account of one of the characteristic birds of 
 this region may be of interest. I refer to the little auk, or sea 
 dove {Mergttlus allc), so common some years in the waters about 
 the islands and harbors all along the coast. P>om Oct. 15th until 
 the ice set in, I found these little fellows common everywhere in the 
 waters of the bays and harbors, and they were generally quite 
 tame. The people here regard their arrival as a sign of cold 
 weather, but it certainly did not prove to be the case this year, 
 since the birds were unusually abundant and the year an unusually 
 mild one. The popular and local name is Bonne hommc (the 
 French for good fellow) and is pronounced as if spelled Bun- 
 num. It associates with the black guillemot, and possesses with 
 it many habits in common. It dives at the flash of the gun, swims 
 long distances under the water, but is generally very tame and quite 
 easy to approach though quick in its movements. I have seen 
 them killed with an oar after a long chase in a boat. When 
 first taking flight they half fly and half push themselves along the 
 surface of the water, since their small wings make it very difficult 
 for them to fly freely. I have seen one pursued in a boat by a 
 number of men who amused themselves by throwing the oars 
 and pieces of wood, together with the ballast of the boat 
 at it, not a single missile hitting its mark, since the bird was able to 
 dodge each article thrown at it by diving and appearing in a most 
 
 I't 
 
 H: 
 
 
 
 I 
 
w 
 
 108 
 
 WRECKS, WESTERN ARM OF BAY. 
 
 i 
 I i 
 
 I 
 
 t :-i 
 
 ''■ lY.' f 
 
 contrary direction from that looked for ; to the surprise of all the 
 bird at last escaped without so much as a single wound. I have 
 noticed nearly all the changes of plumage in this bird that I have 
 seen in the pigeon during the first year, though the head, so far as 
 I have seen, is always black. It is a familiar little fellow, and sel- 
 dom killed, unless scarcity of food demands it. 
 
 Thursday the 4th. Some of the men started off early this morn- 
 ing hunting for logs or pieces of wood from what appears to be a 
 wreck. The lookout in these regions for wrecks is at all times 
 sharp and continued. It not unfrequently happens that a barcjue or 
 brig, and in one or two cases a steamer, going from Quebec or 
 Montreal to Newfoundland or Liverpool, is lost or led astray by 
 the fog and wrecked on some of the many treacherous rocks on this 
 part of the coast. Only yesterday one of the men returned with a 
 large piece of the bulwarks of some ship that had been evidently cast 
 away. It looked quite fresh, and hearing in addition that one of 
 the owners of a large establishment here had sent in a hurry for 
 several men to help him do some work the nature of which he 
 was shrewd enough to withhold, we fairly concluded that there must 
 be a wreck somewhere near. While they were away investigating 
 the matter I took my gun and started off, proposing to climb the 
 ridge on the western arm of the bay and see what results might be 
 obtained in that quarter. 
 
 I first ascended the hill at the foot of the bay. The path was an 
 old Indian foot path, and most of the way ascended almost per- 
 pendicularly. ] reached a level after much trouble, and found 
 that the needle of my barometer registered three hundred feet 
 above the sea's level. I then went carefully all over the top of the 
 height, and was much interested in finding the singular features 
 presented. The elevation seemed to be a plateau three to 
 five hundred feet high, with several crests arising from seventy-five 
 to one hundred feet higher, but with a general level at the distance 
 given above. There were several small ponds, much mucky 
 ground, and several patches of what we should call, on the sea- 
 shore, mud flats. Deep ravines were plentiful, and several times I 
 
DEER TRACKS. 
 
 109 
 
 found myself suddenly and without premonition within a few feet 
 of a perpendicular wall of rock, — in fact once a regular precipice 
 reaching several hundred feet below and nearly to the sea level. 
 The whole ridge is divided by a deep gully throvigh which a large 
 stream flows from a chain of ponds, the first of which could be 
 barely seen in the distance, and which is full of trout, the inhabi- 
 tants tell me. Between the shore and the southwest sitle of this 
 stream the ridges continue but they are cut up with more gorges 
 than the northeastern portion. I .bund no less than three deep 
 clefts, starting from near a common centre and running in contrary 
 directions down to the sea. Beyond is a deep cliff, and several small 
 gullies, running into a large open place that must have reached 
 nearly tr the sea level, which i)resented an almost impassable bar- 
 rier to the coast line. After a long detour I at length reached the 
 beach, or rather the place where there might have been a beach, and 
 found nothing but rock close to the edge of the water, with no sign 
 of a footing anywhere. I clambered over the rocks for al)out three- 
 quarters of a mile and found a perpendicular face of rock that 
 must be climbed or return must be made as I had come. De- 
 termined not to return I with difficulty climbed over one hundred 
 feet of this precipice, when I am confident that a single false step 
 would have sent me to the rocks at the bottom, and thence around 
 the edge of the face of the rock I had but about two miles to 
 go to reach home, but three times I was obliged to climb steep 
 heights from near the sea level to about two hundred and fifty feet 
 above (where often the undergrowth of thick and tangled spruce 
 alone would have made the ascent quite difficult) and three times 
 descend again. It was half tide and I could not walk around as it 
 is possible to do at dead low tide. The top level of the hills above 
 was protected by overhanging ledges so that I could not ascend 
 them ; thus I was obliged to pursue a winding course over the 
 outstanding ridges of the rocks to the eastern face of the cliff. 
 It took full five hours for me to accomplish the distance home, 
 short as it was. 
 
 Friday the 5 th. This morning I saw the tracks of a small herd 
 of deer that had passed during the night. They were plainly 
 
 > ■(': 
 
 I' ''I 
 
T^ 
 
 mr. 
 
 !| it" m 
 
 M 
 
 ;l 
 
 I 
 I I 
 
 Ml 
 
 i. 
 
 110 
 
 EVIDENCES OF A WRECK. 
 
 visible for a long distance on the thick, spongy moss which every- 
 where carpets the ridge just back of the house. Towards evening 
 the men returned bringing the looked-for news — not that we wished 
 a vessel to be wrecked for our special benefit, but should there be 
 one we desired to know of it and see it if possible, — there had 
 been a large brig wrecked on her passage from Montreal to England 
 laden with lumber. The men reported the harbors and coves 
 everywhere along the shores as full of logs and deal boards. The 
 crew, it appears, were saved, but the vessel was a total wreck. As 
 a short account of this wreck may be of interest I append the 
 facts as, being present, I was able to obtain them. 
 
 It was on the evening of November 30, that the boat and 
 men returned from a short sail up the bay, where they had been 
 to visit some traps that had been set there, bringing with them 
 news that a large piece of the bulwark of a vessel was lying above 
 high tide on one of the islands, and that to all appearances there 
 had been a vest.^1 wrecked during the preceding stormy weather. 
 I think that it was later in the same day we heard that a gang of 
 men had been sent for in a great hurry by one of the chief men 
 on the coast, for the ostensible purpose of doing some work on the 
 framework of a building that he had for some time proposed con- 
 structing ; putting together these two circumstances it seemed safe 
 to conclude that something unusual had occurred, and it also 
 seemed equally safe to suppose that that something was a ship- 
 wreck. 
 
 In such a locality as this, remote from habitation, the struggle 
 for life is by no means easy ; and at such a season as this, when 
 time hangs heavy, it is not to be wondered at that the news and 
 this probable interpretation of it spread like fire, and everybody 
 was awake to be off and see what was to be seen, or find what was 
 to be found. The people about could hardly wait until the next 
 morning before starting off in their boats, and it was yet early, in 
 fact before dayligh*^. that a party from the house, with a small sack 
 of provisions, in case they should be obliged to stay away over 
 night, started in search of the supposed wreck. Those of us who re- 
 remained at home of course did our best at speculating as to the 
 
1l 
 
 SEARCHING FOR WRECK. 
 
 Ill 
 
 results of the search, but it was not until the next day that we learned 
 the truth. Meanwhile, several parties, also much excited over the 
 news, called upon us in the evening, they also eager to be off. We 
 fed and housed them for the night, and with them indulged in all 
 kinds of speculations as to the probabilities and possibilities of the 
 case. As yet none of us knew the real state of affairs outside, but 
 that there must have been a shipwreck, no one seemed to deny. 
 Then came the questions : Where was it ? How was it ? When 
 was it? 
 
 As each of these questions was discussed separately by all parties 
 present, it was some time before they were disposed of; when they 
 were, we varied the conversation with queries as to the size of the ship 
 and nature of her cargo. Of course all united in the hope that she 
 might be a large vessel, and laden with provisions, as that is the 
 thing most needed here ; but we could hardly hope that the real- 
 ity would equal our hopes and expectations, and it was finally agreed 
 that should the cargo of the supposed ship, that we felt sure had 
 been wrecked, prove to be lumber, we should be equally satisfied. 
 
 The party that had just gone from the house proposed staying over 
 night at a neighboring island, and to proceed along the coast of the 
 various islands and mainland the next morning hoping to make 
 discoveries ; the party that had just arrived, too late to accompany 
 the first, seemed unsettled as to what to do. In fact they were so 
 wrought with excitement that they seemed ready for almost any- 
 thing. Although it was already night when they arrived, some 
 proposed rowing at once to the island, where the first party had 
 gone, a distance of four miles, at least, and with them, starting 
 off the next morning. We easily dissuaded them, from such 
 an attempt, however, and talking over the outlook of the case 
 we passed away the time until slumber called us all to its em- 
 brace. The next morning we waited patiently for our party to 
 return (as our friends of last night had left early) and relieve our 
 suspense, for we too shared a feeling of anxiety as to the result ; 
 but it was afternoon before the boat returned, and not until then 
 that oui desaes were at least partially gratified. The relics brought 
 
Ti 
 
 M 
 
 112 
 
 CRUISING. 
 
 il'i 
 
 . 
 
 'I 
 
 I ■ ;TT7F.! 
 
 home in the boat gave us no longer any room to doubt that there 
 had been a tolerably good sized wreck somewhere. There were 
 iron and copper bolts, hinges, bits of rope, a bed sacking, a box of 
 books, a long pipe, a blacking brush, and any amount of small ma- 
 terial that had been picked up around the shores of the islands, but 
 what presented the most substantial evidence of the disaster was a 
 genuine barrel of flour, superfine extra. The men were full of news. 
 The islands swarmed with deal boards and logs everywhere. Oak 
 and pine lay about in confusion. Here was then at least an an- 
 swer to one of our questions. There had been a shipwreck to the 
 westward ; how far as yet we did not know, and the lumber, of which 
 the cargo must have been largely if not wholly composed, lay 
 around us, and could be had not for the asking, but for the pick- 
 ing up. 
 
 The boat soon unloaded, and the men provided with their din- 
 ners ; we then began to look about us, and prepare for another 
 cruise, to find out, if possible, where the wreck was. We were soon 
 ready, and getting into the boat started on our expedition. Know- 
 ing that the wreck was to the west, as the drift of the logs and deal 
 was from that direction, we made up our minds to stop that night (as 
 it was nearly night when we started off) at the house of a neighbor, 
 about four miles up the coast. It was now nearly dark, the wind 
 had gone down, and, much against our wills, we were compelled to 
 take down sail and row. Three stout fellows at the oars, however 
 much the swell might take the boat, were more than a match for 
 the waves, and she spun along at least seemingly fast, until we had 
 rounded the corner of the bay, and come into still rougher waters. 
 
 We passed one or two boats, and thought little of it at the time ; 
 we afterwards found that they were on the same errand as ourselves. 
 After a couple of hours of hard struggling at the oars, we came in 
 sight of a glimmer of light down the bay or deep cove past the 
 headland that had just been rounded ; soon it became brighter, and 
 we passed into the more quiet waters within the bay, and rapidly 
 approached the Ijeacon and its well known shelter. It was quite 
 dark when we moored our boat alongside of the rocks. We saw to 
 
SCENE AT DUSK. 
 
 113 
 
 our surprise, in the same shelter, and within a stone's throw from 
 us, the boats of nearly all the people on the coast, for a dozen miles 
 eastward. It seemed that other people had come in search of the 
 wreck as well as ourselves, and showed us strangers the rapidity 
 with which news travels even on this coast, where the houses are 
 four, six, or ten miles apart. Surprised as we were, we finished 
 mooring, and started towards the house. The scene that burst 
 upon us reminded one strongly of some fanciful legend of pirates 
 or sea-robbers. Here the dark outline of the house, back of which 
 tall cliffs frowned out a gloomy reception, was lighted up by a fire 
 on the rocks to the right of the doorway, around which was 
 assembled a group of men who went and came in and out of 
 the darkness beyond the flames. A crane hung over the fire and 
 kettles were suspended a little above the flames containing the 
 tea of the several parties who formed the group ; to the left of the 
 house a wide expanse of darkness wrapped the cliffs, the water, 
 and the ground in one sea of dusk. Approaching the house, the 
 scene presented a livelier aspect and we were better able to see 
 about us. The house of our friend had been literally taken 
 possession of. Later in the evening when more boats had arrived, 
 we counted twenty-four persons who had thus invaded this retreat, 
 and established themselves until morning. 
 
 You may imagine the confusion of twenty good voices in loud 
 conversation (loud talking seems to be the rule with the people 
 on the coast) over the prospects ; add to this the excitement 
 which prevailed on every hand, and the bustle over the dishes, as 
 party after party (each of whom, by the way, brought their own 
 provisions) sat down or rose up from table, and you have the 
 scene complete, — no, not complete, for the room was none too 
 warm, it being a cold night out and the atmosphere breathed and 
 felt of tobacco smoke so forcibly, that one could " cut it," as the ex- 
 pression is. 
 
 After a great deal of talking and listening at the time, with what 
 was learned afterwards, I have prepared with a great deal of care 
 the following stqry of the wreck ; — 
 
 1 !; 
 
 i i 
 
 it 
 
 !^ 
 
 I 
 
iwr 
 
 'i;i 
 
 114 
 
 WRECK OF THE EDWARD CARDWELL. 
 
 iU 
 
 
 ■ %^ 
 
 The Edward Cardwell, a full rigged barquentine, bound from 
 Quebec, P. Q., to Liverpool, England, after several days out (how 
 many we did not learn), encountered dense fogs off the banks, and 
 for about three days had been sailing in this uncertainty, feeling her 
 way slowly along, the officers not knowing where they were going, 
 but supposing themselves somewhere near the Newfoundland 
 shores. At one time the fog lifted for a few moments, and then 
 the white beacon of Whale island met their view, but the immediate 
 shutting down of a still denser fog left them again in the uncertain 
 condition in which they had been before. Steering as near as 
 possible in the direction of the beacon, it was not until rocks 
 suddenly loomed up near by that the pilot found himself at the 
 entrance of a narrow pass near ix rocky shoal with the mainland 
 some half a mile on the left. The ship was under too much 
 headway, though it was moving but slowly at best, to stop or back 
 out of her perilous position. The pilot headed her straight for the 
 opening, and called to all hands to prepare themselves for the shock. 
 One young man sprang to the cabin door and called to the captain, 
 who not havmg time even to take up his watch, which was lost, 
 rushed on deck only just in time to secure a place in the boat, 
 which the frantic men had lowered, and were about severing from 
 its fastenings to the ship. The ship struck and went to pieces in 
 a few hours afterwards ; the crew, nineteen men, were just saved, 
 but having lost all. The men rowed to the neighboring shore, and 
 finding the empty summer house of a Mr. Belvin, one of the 
 inhabitants of the coast, they broke in the door and made a fire, 
 remaining there that night. In the morning they took one of Mr. 
 Belvin's boats, as their own had been destroyed, and rowed along 
 the coast and islands until they reached the abode of Mr. W. H. 
 Whiteley, about fifteen miles from the wreck, where they tarried 
 until they were soon after carried to Greeley island lighthouse, 
 from which place they were taken to the Newfoundland coast, 
 and thus reached some port from which they took passage home. 
 Mr. Whiteley, the Magistrate of the coast, agrees substantially 
 with my statement. The ship went to pieces soon after it struck, 
 
EVENING AT THE COTTAGE 
 
 115 
 
 and the next morning (Tuesday) logs, deals, and rubbish of all 
 sorts were to be found everywhere on the mainland and islands to 
 the eastward. The vessel contained few provisions, but these 
 were mostly lost in deep water. The rocks where she struck are 
 called on the chart the forpoise rocks, and the water about the 
 shoal varies from nine to thirteen, and even thirty fathoms ; the 
 distance to the mainland is about a mile, and from the opposite 
 coast of Newfoundland some thirty miles. 
 
 Let us now return to the cottage and see what is going on there. 
 The evening wears slowly away, the men enjoy themselves and 
 pass their time in smoking and talking. Some are jovial and hearty 
 in their manner, while others, quite the reverse, are gloomy and 
 morose. It is easy to see who are friendly and who avoid each 
 other, for the men cluster together and engage in low or loud con- 
 versation as the subject of which they talk be private or public, 
 while others sit in the corners, on the floor, or in chairs resting 
 their heads upon their hands, or, leaning against the wall, are far in 
 the land of slumbers. One man stands warming himself with his 
 hands behind him, and his face away from the stove facing the 
 crowd, while another perhaps will be talking loudly and boisterously, 
 gesturing violently at the same time as if to impress the group 
 more with a sense of his own importance than to give a statement 
 of the real condition of some important issue ; perhaps this same 
 person will soon change his position to a slight bend of the head 
 and body as, with one finger held up very near to his eye, he makes 
 an outward and downward gesture, as he delivers himself of some 
 whispered secret opinion, at the same time that, having delivered 
 his opinion, he straightens himself up with the air of one who has 
 relieved himself of a, to him, tremendous thought. 
 
 In one way or another the time flies. One by one the men 
 stretch themselves on the boxes and benches in the corners ; tipped 
 up in the chairs, and on the floor besides the fire, they doze off to 
 catch a poor apology for sleep in these uncomfortable positions. 
 We can only get intervals of rest as somebody is constantly open- 
 ing and shutting the door in passing in and out ; this occasions so 
 
 M' 
 
 r 
 
 l:H 
 
 ii.r.LJu^»y:i 
 
T^ 
 
 m 
 
 :f 
 
 i|'!>'|| 
 
 ;!S 
 
 li! 
 
 : il 
 
 i-;r! 
 
 116 
 
 FINDING LOGS AND DEAL. 
 
 much noise and cold air that soon even the air from the stove 
 grows chilled and we with it. 
 
 At length, after a number of twistings and vain attempts to sleep 
 the earliest of this adventuresome party arouse the rest by their 
 preparations for early starting in the shape of making a fire, boiling 
 the kettle, and preparing and eating an early breakfast. Between 
 three and four o'clock the fi 6t party leave, and are followed by the 
 others at intervals of different length, but near enough to each 
 other to prevent our going to sleep again, until at daylight we are 
 about the last to leave the house with a good warm breakfast in 
 place of a good night's sleep. 
 
 As most of the other boats are engaged in a similar expedition 
 as ourselves, to follow us will be to follow them. The wind is against 
 us and the waves are high, but we start off without much trouble and 
 row, head to the wind and waves, out into the passage and towards 
 the nearest island. The shore is plainly visible as we row along, 
 and a shirp lookout soon discovers a log lying on the beach and a 
 deal close by ; we row to them and have soon carried the deal 
 above high water mark while we note the place for future use ; 
 beyond are several more deals, and further on others ; soon we 
 come to a cove fiill of strewn rubbish composed of bits of wood 
 and hay and straw, with many sticks and broken boards while sev- 
 eral large oak and pine logs lie, as they have been tossed by the 
 waves, wedged in between the stone and rocks. These logs are, 
 as are most of those found, from eighteen inches to two feet in 
 width and thickness, and from thirty to fifty feet in length ; while 
 the deal are about two inches thick, from ten to eighteen inches 
 wide and about twelve feet long. Each is of three qualities and 
 stamped with the Quebec market initials of "" A. F. A. K." or with 
 the word "Montmorenci." With considerable difficulty the big 
 logs are pried over and over until they reach the water when an 
 iron bolt is fastened to them and a rope attached drags them out 
 into the water where the tide, which is on the flood, rises sufficiently 
 to float them. 
 
 In this way a raft or rather tow is soon made of four fine, pine 
 
WOOD-CUTTING. 
 
 117 
 
 logs and, elated with our success as wreckers on a small scale, we 
 start off for home. In a day or two the deals were found floating 
 about by hundreds, and the work of collecting them as also of 
 marking and tying the logs to identify them, continued all the fall 
 and even into the winter. This is the first wreck that has occurred 
 on the coast, near here at least, for over thirty years, I think, if 
 my information is correct ; and though we certainly wish no harm 
 to anybody, we can but rejoice that the misfortune of the ship will 
 be so fortunate for the people of the coast. 
 
 Wednesday, Nov. loth. The men have spent a greater part of 
 the day cutting wood. Those who can obtain wood near by with- 
 out the necessity of going into the interior up the river, and rafting 
 it down, as many of them do, content themselves with a smaller ar- 
 ticle, and continue to make clearings in the low spruce and fir 
 about their own place. The majority of this wood varies from 
 four to six and even seven inches in diameter, while the trees are 
 rarely over fifteen feet in height. The tree is cut, the branches 
 trimmed, and the limbs thrown in a pile upon the ground ; the 
 trunks are then piled on the sledge and drawn by the dogs to 
 the house. In winter the men are often obliged to go chopping 
 wood after a heavy fall of snow. It is then a matter of no pleasure 
 to walk half a mile or a mile through the deep snow to some cho- 
 sen locality, and there remain cutting for many hours, while the 
 snow from the branches falls down your neck, as you stoop over 
 to chop, and the wet often finds a hole, be it ever so small, 
 in your boots ; while, to endure the cold, the thick clothing 
 one is obliged to wear renders such violent exercise anything 
 but comfortable. When the wood choppers return at night there 
 is always a hot supper waiting for them, and the roaring and 
 crackling fire sends out a genial heat that dries the wet garments 
 while it comforts the spirits of the men. 
 
 Friday, the 12th. I'he first snowstorm of winter came upon 
 us to-day. It began in the morning and snowed most furiously all 
 day ; by evening the ground was covered to the depth of nearly if 
 
 I. 
 
 11: 
 
 ! I 
 
 f* 
 
118 
 
 LABRADOR SNOW. 
 
 not quite a foot. The snow here differs from that of New Eng- 
 lan' 1 and other parts of the United States, in that it is dry and not 
 damp. It packs heavily, and when walked on generally gives out 
 that crisp sort of echo so often observed in walking anywhere over 
 lightly packed drifts. In this climate, after our first snowstomi, 
 winter is upon us, and we can safely conclude that we are shut in 
 from sunlight and society until the next summer. A thin coat- 
 ing of ice has already formed, and we can probably soon be able 
 to traverse the bays in our sledges. We have fully started upon 
 our six months of ice and snow. 
 
 Before I go any further, let me here stop for a few moments and 
 review a little. I have not fully described to you our quarters 
 either here, in our snug inland retreat, or those from which we so 
 lately removed on the outside island ; I will therefore try to give 
 you a little idea of how we live, and in what we live ; and, as most 
 of the families possess similar establishments, we will try to give 
 you the general idea of a Labrador home. 
 
DWELLING HOUSES. 
 
 119 
 
 CHAFrER VIII. 
 
 A Labrador home — Houses — Where erected — Stage — Shop — Stable — 
 The house — Papering — P'amily — Occupation of its members — Out- 
 of-door life. 
 
 In describing a Labrador home, I shall be doing justice to 
 nearly all by describing one, since all are modelled and furnished 
 on about the same plan. 
 
 There are, of course, a number of houses modelled after the 
 fashion of those "in the States," as the expression is here; but 
 they are the exception rather than the rule. The best house, per- 
 haps, is owned and occupied by Mr. W. H. Whiteley, the magis- 
 trate of this section of the coast, and situated at Bonne Esperance, 
 a little island at the mouth of the Esquimaux, or, as it is rightly 
 named, St. Paul's river. This is a comfortable mansion-like affair, 
 and is built like many of the so-called house taverns so common 
 in country places in New England and other states. Its white ex- 
 terior shows for a long distance up and down the coast, on a clear 
 day, especially if the sun be shining and serves as a beacon to the 
 inhabitant and voyager in these parts. 
 
 While Mr. Whiteley's is a palace beside the other houses, there 
 are those that are hovels beside what I am about to describe as 
 characteristic of the larger and better class of abodes. These hov- 
 els, or rather huts, for huts they are in the true sense of the word, 
 are of the rudest kind. The logs, posts, and most of the boards 
 are hewed out by hand from trees growing a little way in the in- 
 
 i ! ^. 
 
 (■■' I 
 
im 
 
 j ' 
 
 
 ■ I 
 
 
 
 f 
 
 1 ' ; ■ : ■' 
 
 1' ■ ■ 
 
 m 
 
 k 
 
 120 
 
 SITUATION CHOSEN. 
 
 terior, and brought down on rafts by the nearest river ; on the 
 border of which the hewing and trimming, which render them fit 
 for the purpose for which they are intended, are generally though 
 not always done and in the fall of the year. 
 
 The localities chosen for the erection of house or hut are gen- 
 erally two, one for a summer house in some oj)en situation, and 
 the other for what is called "winter quarters ;" and as winter em- 
 braces the greater part of the year, it is important that this latter 
 shall be, as it invariably is, in some sheltered cove on the mainland 
 where, if possible, high cliffs protect it on all sides, except from 
 the sea ; if such a place is not found, as sheltered a place as pos- 
 sible is chosen. When the house is a summer house there are usu- 
 ally cliffs on the north side, or if not cliffs high rocks, and the ex- 
 posure to the sea easterly or southerly. 
 
 A hut is of the rudest make. The sides are of logs, the bot- 
 tom and floor of single boards, the roof of rough rafters, and the 
 top of thin deal or clapboards ; but a house is of different con- 
 struction. 
 
 I shall not stop here to describe Old Fort island, as I shall 
 do so further on. I will simply say that it is the largest of the 
 neighboring islands, — except the one called Esquimaux island, 
 — and is about four miles from the mainland. The owners, or 
 rather the dwellers thereon, regard this as simply their summer 
 abode, while the winter house, is located on the mainland. The 
 name Old Fort is historical ; it being so called in memory of old 
 times in connection with the P'rench and Indian wars. 
 
 The relation between the mainland and the island, as is seen by 
 their names, is very close ; and this little family of settlers have 
 taken possession of both places, which they have held for many 
 years, and set up their abodes thereon ; the winter house being 
 at the elbow of the bend in Old Fort bay. 
 
 The house on the island is placed about forty rods from the wa- 
 ter on the east, on a small bank of rising ground about ten or 
 twelve feet above the sea level at high tide. The whole island is 
 low anywhere, yet the plain here happens to be smooth and well 
 

 DESCRIITION OF THE STAGE. 
 
 121 
 
 covered with grass and vegetation. On the southern side the land 
 slopes down regularly to the sea, anil an elliptical shaped beach 
 of rock and sand, visible only at low tide, separates the land from 
 the water, in this direction. On the west is a scries of low, 
 rocky elevations ; in the north, another series of much higher 
 knolls is terminated abruptly by the water. In the centre of this 
 little plot of ground, comprising about one one-hundredth of the 
 whole island, the house is situated. 
 
 I say the house is situated, but it would be better to say the 
 houses are situated ; for there are generally several buildings con- 
 nected with a well conducted fishing post, or summer residence. 
 I will give a brief description of them : — 
 
 The stage consists of a platform some sixty or seventy feet long 
 (according to the necessary distance), and built from the beach into 
 the sea — generally so that the farther end will always be some feet 
 above the water even at the highest tide — and about sixteen feet 
 wide. It is built on posts or poles which raise it some six or eight 
 feet above the ground, and covered with boards. The regular 
 fishing boats are generally moored only a short distance from 
 the wharf or stage in deep water, while several small boats are fas- 
 tened to the stage by means of which the men get from the wharf 
 to their boats and back. The inner half of the stage is called the 
 house, and is covered over with a sloping roof and board sides. 
 It encloses bins for salt and fish, barrels for either or both, 
 and the general necessary things contained in such places. This 
 house has usually a loft for the storage of nets, or anything not 
 needed for immediate use but which are too good to be thrown 
 away. A simple board walk leads to the front door, while the back 
 partition, generally open to the stage beyond, completes the fishing 
 house and stage, where cod are split and cured, salmon and trout 
 salted, mackerel cured, and where all kinds of fish are prepared 
 and preserved. One curious fact may be mentioned : that the 
 primitive way of fastening doors exists here almost everywhere 
 in spite of the cheapness of door fastenings as purchased of the 
 traders ; the whole contrivance is made of wood, and the door is 
 
1 
 
 
 ( ■ 
 
 %' ■ 
 
 ! i 
 1 
 
 
 
 i'', 
 
 122 
 
 INTFRTOR OF SHOP. 
 
 opened or shut from the outside by a cord which passes through it 
 by means of a small hole, the inside end being fastened to the latch, 
 and the other to the door by another hole and a knot in the string. 
 This makes a loop upon the inside which answers for the handle. 
 The doors of all the houses, barns, stages, and shops are similarly 
 fastened. 
 
 Next to "the stage" comes "the shop." This is another small 
 house with single room and a high loft, and is situated not far 
 from the house and the stage. It is built low, the foundation 
 resting either on or near the ground. Its size is between that of 
 the stage and the house pvoper. In it are kept the extra stores ; the 
 flour, potatoes, turnips, salt pork or beef, butter, tea chest, and 
 other articles not in ordinary use in the house. In one corner 
 is a tool bench, on a shelf above are numerous cans with remnants 
 of paint for painting boats and perhaps the kitchen floor ; above 
 this, on a series of nails, hang saws, shaves, planes, old iron hoops, 
 and all sorts of articles usable and unusable that can hang up, 
 while the bench beneath is cluttered up with a little of anything 
 and everything that you can imagine. In the opposite corner the 
 scythe and hoe lean against the wall, while a little way from them a 
 very small, coarse grindstone, mounted on a carriage that threatens 
 to fall to pieces every time that it is touched, leans rather than 
 stands. On the wall, over against the stone, a small window frame 
 is nailed on the inside of a square opening ; it often contains but 
 three whole panes and a broken fourth, the hole filled with an old 
 felt hat. By this, aided by the additional light of the open door, 
 barely light enough enters to enable oiif,- to see where to get or put 
 away anything, which is generally anywiere. This shop door 
 looks as if a part of the partition had been cut off", at the 
 farther end and on the same side as the window, and a couple 
 of cleats nailed crosswise, one above the other, to hold the 
 boards together ; the whole affair having very poor, jagged hinges 
 on one side, so that the door opens and shuts very hard and 
 squeaks proportionally, and a latch on the other. 
 
 On the floor, just between the door and window, lies the koma- 
 
STABLE AND HOUSE. 
 
 128 
 
 '. 
 
 tik or dog sledge ; just above hang the dog harnesses ; and next to 
 them the rackets, or snow shoes, by means of which the men, and 
 sometimes the women and children, with a little practice, walk 
 easily and (|uickly over the light and often very deep new fiillen 
 snow. In the only remaining corner stands a heap of rubbish 
 which extends underneath the bench nearly to the other side of 
 the room. It is compojed of every namable thing that you can 
 imagine : scraps of leather, old shoes and boots ; pieces of wood, 
 long and f-.liort, thick and thin, picked out for special purposes ; 
 old iron bolts, hinges, spikes, and rings ; old pans and paint dishes ; 
 pieces of rope of various sizes and thicknesses, cork bobs for fish 
 nets and wooden blocks for the same i)urpose, with an occasional 
 "snatch-block," as the sailors call it, — being an oval jjiece of wood 
 hollowed oul with a wheel inside such as is used for hoisting articles 
 by rope, — and in fact a large assortment of general rubbish beside. 
 
 At a short distance from the shop stands the stable, if the 
 family keep a cow, a goat, or any other animal reciuiring a building 
 of this kind. This is the simplest sort of a shed or barn, — with a 
 top hay loft, and a few plain partitions which serve the animals as 
 stalls, — while even a manger is wanting, and the food is given to 
 them upon the floor of the upper end of the stall. Many families 
 do not keep such animals, not being able to afford them ; then the 
 stable is of course useless and unnecessary. 
 
 Passing now directly to the house — only noticing several boats 
 lying upon the sand or on the bank above the beach, at the right of 
 the stage, and the ever present pile of wood, partly cut and partly 
 in long rugged pieces, with the fish flakes, which will be spoken of 
 in connection with the fishing business — we will try to describe it. 
 
 The house is, of course, a primitive affair, and perhaps little 
 better than the abodes which our forefathers were accustomed, 
 after a while, to erect upon "the rude and rugged shores" of some 
 seacoast town of our own New England or Atlantic states. It gener- 
 ally faces the south, — that is, the door and longest side do, and is 
 about one-third longer than wide. It is built with one full story down- 
 stairs, and an attic beneath the sloping roof; a partition running 
 
124 
 
 INTERIOR OF HOUSE. 
 
 I' ' 
 
 : ; 1 
 
 k 
 
 I ' W 
 
 fl 
 
 I 
 
 from floor to roof divides both stories into two rooms each, the 
 largest of which is about square, and the smallest one half the 
 size of the large one. Windows are few and far between; the 
 upper story usually with one at each end or with none at all as 
 the case maybe, — which case is usually governed by the money or 
 time at the disposal of the builder at the time of building — and 
 are always with the smallest kind of glasses and sashes possible to 
 admit any light ; the lower story usually has from two to four on 
 its sides, rarely any on its ends. The doors are usually three in 
 number, one upstairs and one downstairs in the partitions (these 
 are more often doorless openings), and the outside one of all ; 
 they all resemble that of the store house before mentioned, at 
 least in one sure respect if in no other, namely, that they press 
 with difficulty upon the hinges and squeak horribly when the 
 process of opening and shutting, which happens so many times 
 in the course of twenty-four hours, occurs. The upper story is 
 reached from below by a narrow pair of stairs, or a ladder, leading 
 through an opening in one corner of the room, i'. is safe to say 
 three feet square or even less, and it requires a good deal of 
 practice to perfect one's self in the art of ascending and descending 
 safely. In the construction of the house the building materials 
 used everywhere are rude or good as the tenant can afford ; the 
 outside is clapboarded or shingled, and there is no cellar, except a 
 rough hole scooped out underground and lined with hay and coarse 
 grass, and boughs perhaps ; with a cover cut from the bottom of 
 the floor. The furniture consists of a large and ample stove (two 
 storied like the house) with a baking oven running its entire length 
 above, the stovepipe ascending straight upward, forming its own 
 chimney outside of the roof; the height of the stove being some 
 four leet above its legs which raise it some eight inches from the 
 bottom of the floor. It is placed between the two partitions and 
 nearly in the centre of the room ; for its accommodation a large 
 piece of the partition is cut wholly away so tiiat the warmth will 
 heat both rooms ecjually. 
 
 Besides these the tables are all home made, and of proper 
 
INTERIOR OF HOUSE. 
 
 125 
 
 sizes to suit their different uses, being of plain deal boards varnished 
 or painted as there is no table-cloth laid at the meals on top. 
 The chairs are also home made, plain, high in the back, and with 
 seats of woven strips of deerskin or sealskin with or without a 
 cushion of patchwork or white cloth filled with ducks' feathers, — 
 withal quite comfortable. A rocking chair made after this fashion, 
 as it often is, is a comfortable affair, if not quite a luxury. In 
 the smaller of the two rooms is, as the expression goes, "a 
 poor apology" for a book-case containing a few catechisms and 
 books on the Bible, with perhaps the old times Fox's " Book of 
 Martyrs," a dilapidated Pilgrim's Progress, one or two Bibles, 
 a number of indescribable volumes many of which are in French, 
 besides a volume of "The Leisure Hour," a London magazine 
 of good reading for the household in general, and perhaps one 
 or two torn books for children. The shelves are loose or crooked, 
 and the books present the appearance of ha\ing been caught in 
 the act of tumbling. 
 
 The small room down stairs is a bedroom ; it contains a bureau, 
 and, as do the two rooms up stairs, a bed each, either boarded or 
 corded to hold it together, in which is of course a tick filled with 
 feathers from ducks and other birds, as are the pillows also ; 
 the bedding is of the coarsest kind suitable for cold winters of 
 from 20° to 40° below zero with fierce winds of unknown velocity 
 per hour. A simple mention of the angular pantry, built in the 
 corner of the house down stairs, with the lower i)art cupboarded 
 and the upper part ihelved and open, holding the plates, cups, and 
 saucers, a platter, a bowl, and one or two pitchers (of a small, 
 brown, glazed itone pattern) with the knives and forks, will do for 
 this table furniture. Placed in the opposite corner is a small table 
 with a washbowl and dish of soap, below which is a pail of water 
 with a small pail inside for a dipper, above which, a<- the side, hangs 
 a towel, and in front of which is a cjuadrangula. piece of glass, 
 .vith or without a frame, with mercury par':elled i^ehind it promis- 
 cuously, — the whole called a looking-glass; with these one or two 
 carpenters' chests for tools containing clothi' g or other articles, and 
 
 if 
 
 f 1 
 
 u 
 
 I ;■ 
 
 p 
 
 ^# 
 
 ; ■ .1 
 
 t 
 
f 
 
 ! , I 
 
 !■■■■■■■■ 
 
 l\^ 
 
 12G 
 
 UTILITY OF NEWSPAPERS. 
 
 ir 
 
 ii 
 
 a bench to sit on, and you will have before you a pretty fair picture 
 of the house and furniture. Three more subjects remain to be 
 spoken of. 
 
 First, the papering. Any one might reasonably be surprised at 
 the idea of papering such a house as above described, for two 
 reasons : that there should be the means for such a luxury here, and 
 at the way in which it is done. Roof, boards, rafters, the sides of the 
 house up and down stairs, the doors, and the cupboard are all pa- 
 pered — the articles used being anything in the shape of book, 
 pamphlet, or newspaper. I have spent more dull, gloomy mornings 
 than a few, reading the titles, looking at the pictures, or reading the 
 stories, pasted at the head and sides of my bed, and you will be 
 surprised at the following list of literature which actually occurs on 
 the walls of our palrilial mansion on this out-of-the-way Labrador 
 coast. I give you the list directly as I took it myself from the pa- 
 pers whose titles they represent : The Montreal Witness, Sunday 
 School Times, Advance, Child's World, Christian World, Child 
 at Home, Protestant, Apples of Gold, Well Spring, Herald of Mercy, 
 American Messenger, Juvenile Presbyterian, Young Reapers, Chris- 
 tian Messenger, British Messenger, Home Missionary, Christian 
 Family Almanac, Nation, Youth's Temperance Banner, A German 
 paper, Sabbath School Messenger, Boston Journal, Springfield Re- 
 publican, Christian Soldier and Christian Guest, Child's Paper, Na- 
 tional Quarterly Review, Youth's Companion, Cottager and Artisan, 
 Northern Messenger, Harper's Weekly, Every Saturday, Our Lit- 
 tle Ones, Life Boat, Foreign Missionary, Young Missionary, Sab- 
 bath School Visitor, Colporteur and Dominion Monthly, — all these 
 I saw, and there might have been others that escaped me. AltJiough 
 it may surprise you to see a list of thirty-nine papers, most of whose 
 namfca are household words at home, the way in which the people 
 get them is also curious ; it being, I am informed, a regular cus- 
 tom with the Mission to send, at Christmas time, a bundle of old 
 papers — of which they are always receiving a large number — to 
 each family living near. About two weeks after the receipt of these, 
 and usually just after the holidays which here embrace the twelve 
 
ART AND LITERATURE COMBINED. 
 
 127 
 
 days from Christmas to old Christmas day, the 6th of January, the 
 ladies of the house begin. I'hey patch up the old places, repaper 
 the dirty ones, and spend their time pasting on the new jiapers in 
 every spot that needs them until the pile is exhausted and there are 
 no more ; then the ladies scold terribly because there were not 
 half enough papers, and say that they better not have put on any, 
 unless they could have papered the whole house over anew. When 
 the papering is completed it presents a very curious appearance. As 
 I am now writing there appears on the partition beside me, a copy 
 of the Montreal Witness Extra, for May, 1870, in the centre of 
 which is a large portrait of each Mr. Moody and Mr. Sankey, and a 
 sermon in full of the former ; in another paper next to it, whose 
 title is torn off, is another portrait — of Mr. Henry Varley, and also 
 an account of his last service in the Hippodrome ; above this is the 
 fancy label of a box of Loring Brothers Malaga Raisins with a por- 
 trait of that gentleman in the centre, a vine of luscious grapes hang- 
 ing on either side of him, and a lot of vessels and water below the 
 left hand, and a steam factory on the right, the whole done in col- 
 ored ink ; a copy of the Child's Paper is above, and the Sab- 
 bath School Messenger, at the side of this, has a picture of the 
 lyre bird on one leaf, and a full sized illustration entitled *' The 
 Frozen Regions " on the other, which represents a vast number of 
 curiously formed icebergs along its sides, a sea beneath in which 
 several large blocks of ice float carrying a soldierly row of solemn 
 looking penguins and several seal, and a mass of dark clouds over- 
 ,iead. Quite near me an old Boston Journal relieves occasional 
 monotony by an editorial on Disraeli, and a sketch of the " War 
 •t: die Danube," with several other things. Near the head of my 
 bed is a picture of a lady holding a little child on her lap, a small 
 girl talking to a squirrel which is seen as a small black sj^eck away up 
 in the dense foliage of the neighboring trees, a fox-hunting party, 
 and a picture of two small boys of which the story beneath says that 
 the one because he got angry pitched the other into a pig pen, — 
 from which he was rescued by a kind old gentleman, who also 
 appears in the picture ; the paper is called " Apples of Gold." It 
 
 if 
 
 ) ;. 
 
 (I 
 
 1 1 
 
 
n I 
 
 128 
 
 THE FAMILY. 
 
 n 
 
 is supposed to be a moral lesson paper for children ; but I will not 
 stop longer to describe these things, but pass on to the family, 
 which comes next in order. 
 
 Here also we come to a difficulty : that of describing a Labrador 
 family. It would be unkind to describe the family where I stopped 
 while on the coast, also any one particular family might feel justly 
 indignant should I describe //la'r family, and yet to do justice to the 
 subject a description should be given. Still we shall not be out of 
 the way should we suppose a family composed of a middle aged man 
 and his wife ; either an old gentleman or an aged grandmother ; 
 perhaps a daughter ^v a son (from 15 to 20 years of age) or both ; 
 two or three small la. and a baby ; and, to aid in times of gen- 
 
 eral confusion, when su. times come, which happens more often 
 than the opposite extreme, several large, fierce, full-grown dogs and 
 one or two jjuppies that are always in the way, and in a continual 
 state of warfare with themselves, the people in the house, — who are 
 always scolding them — and everybody and everything in gen- 
 eral. When the houses have porches, as nearly all the winter houses 
 do, the dogs and children live in the porch together nearly all the time, 
 — in fact it is often difficult to tell which make the most noise, and in 
 the general confusion that continually prevails, to pick out or 
 distinguish the one from the other. Of course the utmost sim- 
 plicity of dress prevails among both men and women, as the richest 
 here are poor at best ; the goods worn are coarse and thick, but 
 rough as they are, they are better for the harsh treatment they re- 
 ceive than if they were of a much nicer quality. There is, of 
 course, a certain atmosphere of home even here, but it is often hard 
 to distinguish it, or tell when it is present, as the prevailing confu- 
 sion which such small quarters necessitate is rarely lulled ; and when 
 it IS, peaceful sleep usually reigns. I do not wish to convey the 
 impression that somebody is always quarrelling with something, but 
 there is a constant chatter going on most all of the time, and when 
 quarrelling is not in order, as it frec^uently is, the loud talking of 
 different parties between themselves takes its place. It seems to 
 be the usual way of putting one down, as we say, to see which can 
 
OCCUPATION OF INHABITANTS. 
 
 129 
 
 talk the loudest. If one has something to say, and anybody appears 
 to differ from the opinion expressed, the one who can talk the 
 other down always comes out ahead. 
 
 The occupation of the different members of the family can be 
 very briefly stated as that of earning a living in the summer, and 
 living in the winter. It is hardly necessary for me to say that this 
 means that what is earned during the summer in fishing and in other 
 ways is consumed, and more besides, in provisions, and clothing the 
 family for the year ; and though I make no allusions to an) spe- 
 cial family, there is hardly an exception to the rule that tht re is 
 scarcely a person on the coast but owes for one, two, or even 
 three years back provisions ; this is due to several causes, of which 
 laziness forms the chief. 
 
 In former years those persons upon the coast who kept little 
 shops of provisions and the necessities of life, as also the traders, 
 were obliged to trust out large amounts of goods on the credit 
 of the following year's fishing. This was well enough as long as the 
 fishing was tolerably good and the people were not forced by hunger 
 to give untrue reports of the work done, since a falling off of the 
 supply of fish and a little prevarication brought them provisions, 
 etc., without much overwork ; but one or two years of scarcity of fish 
 soon left the people no means of paying their old debts, and mak- 
 ing promises which they could not perform, the traders began to re- 
 fuse them credit, and now there are few families on the coast who 
 hold their own and prosper. 
 
 In the summer the men fish ; in the fall they cut wood and do 
 little odd jobs necessary to the preparation for winter ; in the win- 
 ter they keep things about their place in order and prepare for 
 spring and summer, which come almost together as the ice does 
 not go away till late in May, by mending their nets, boats, and dog 
 sledges or komatiks ; this is usually finished by the opening of the 
 season when fishing begins again. In summer two or three days 
 are taken to collect eggs of sea-birds which abound on all the 
 islands, — enough for family use only. The down of the sea-birds is 
 kept for beds and pillows and the flesh proves a pleasing variety 
 9 
 
 
 
 
 \ ■^' 
 
 ; 
 
 ^ 
 
 i 
 
 H- 
 
 ! 
 
 f 
 
 i 
 
 il 
 
 
 
"J., i j i .iyjim n 
 
 '!! 
 
 il^ 
 
 f 
 
 l( < 
 
 .'!! . 
 
 iij 
 
 i ' 
 
 180 
 
 CHILDREN— OUT-OF-DOOR LIFE. 
 
 from the usual diet of fish, othenvise so universal and abundant. 
 The women stay at home mostly and keep things in order there ; 
 cooking the meals which usually consists of the most simple fare : 
 bread with or without butter (some use lard for butter), rarely a 
 piece of pork, and tea of the usual kind called black or breakfast. 
 The main stand-by is either codfish or herring with an occasional 
 mackerel ; while the red berry and baked apple form excellent pre- 
 serves. The grandmother or the grandfather, of course, does 
 little but exist, so to speak, though the former keeps the family sup- 
 plied with good warm knit stockings, and makes and mends boots. 
 This may seem a queer statement, but one has only to consider that 
 the boots here are very different articles from what are obtained 
 in the " States," as the people here say, and that they are made 
 after the Esquimaux pattern, of sealskin, and with soft tops and 
 bottoms, to be convinced of the fact. 
 
 The children :.re, as I have said before, a constant source of dis- 
 turbance, and they with the dogs are always under foot ; if it is 
 not one, it is the other, and more often it is both. They grow up 
 together and fight together ; all at once the child becomes large 
 enough to be of use, and then he or she is up for himself or 
 herself, and is either ordered about by the women inside to attend 
 to such duties as they uie able, or by the men outside to help them. 
 The dogs are a mongrel half breed. They fight all the time, and 
 eat anything that they can get hold of, from leather, or rather seal- 
 skin, which is used in the place of leather, to meat in its most 
 putrid condition. 
 
 The out-of-door life and surroundings of the people are neither 
 varied nor pecuhar, and they live here much as such a class of people 
 do in other climates, dressing to suit the season, and paying very 
 little attention to their appearance, except during holidays or Sun- 
 days which are scrupulously regarded ; the latter, by simply keeping 
 the day — for in winter it is usually impossible to get to church, and 
 in summer it depends upon the wind, unless they live too far away, 
 as to whether they can get there at all — on holidays by dressing in 
 their best, and having a good time, a dance, or a shooting match ; 
 
HOSPITALITY TO STRANGERS. 
 
 131 
 
 the men usually, if not almost invariably, ending up with a drunk. 
 The utmost hospitality is extended to strangers, and, as the various 
 houses are often at a distance of eight or ten miles apart, it is fre- 
 quently the case that a person, travelling from one place to another, 
 is of necessity compelled to seek food and a night's lodging ; when 
 such is the case personal quarrels are invariably forgotten, as it is 
 considered the height of meanness, though I use the word only for 
 want of a better, to let personality interfere with, you might say, 
 the necessities of travel. The men are scrupulously a prayerful 
 race ; and, with the great number with whom I was often obliged 
 to "rough it," I know of hardly a case where the men did not each 
 regularly kneel in the morning and at evening before retiring, and 
 say their prayers without the least hesitation, no matter how many 
 were around. 
 
 li 
 
 y 
 
 
 *;' 
 
 1 
 
 ( 
 
 ■'V 
 
 ^i 
 
 1, 
 
 '■i 
 
 
 1 
 
! 
 
 -lUllll »WWMWli- 
 
 132 
 
 FRESH PROVISIONS. 
 
 lill! 
 
 CHAPTER IX. 
 
 Dinner off fresh meat — Credit and shiftlessness — A Labrador snowstorm — 
 Wind — Preparing for storm — Storming hard — iMrevvood — Storm in- 
 creases — Sleepless night — Another day of it — A grand sight — Vio- 
 lence of wind and wave — Destruction of stage — Calmer weather — 
 Beautiful ice scene — End of storm — Thanksgiving Day — Komatiks 
 and rackets. 
 
 Saturday, November 13. We went to Bonne Esperance partly 
 with the intention of staying and visiting, and partly to mail our 
 letters which of course we found were too late, and consequently 
 obliged to remain over until next spring. Encouraging prospect ! 
 We had for dinner a piece of fresh pork which the shipwrecked crew 
 of the Edward Cardwell had presented to Mr. Whiteley. It tasted 
 very nice, especially as we had been living on salt pork in a log 
 cabin for a month past. Pork is with me at no time a favorite dish, 
 and it is only a luxury when nothing better has been afforded for a 
 month or two previous. The following Sunday was spent in doors, 
 in reading and in pleasant conversation with this most agreeable 
 family. 
 
 Monday it came off clear and cold. I had a chance to observe 
 the island quite closely. Mr. Whiteley has greatly improved it since 
 my visit in 1875. He now has a little shop or store, and sells a 
 great many things to the people. Of course he only attempts to keep 
 in stock those things absolutely needed by the people for food and 
 clothing. One is obliged to give so much credit here that it 
 hardly pays to keep anything, much less an assortment of useless 
 articles to please the fancy of some wordly-minded maid or fellow. 
 A very few send to Quebec every year and purchase a few articles 
 of gorgeous colors, and display a dress disproportionate to their 
 
BEGINNING OF A HURRICANE. 
 
 133 
 
 manner of living ; but the majority are well contented witii more 
 simple yet durable attire, and toil on, winter and summer, with no 
 greater ambition, apparently, than to become the wife of sorne lazy 
 young man who can barely support himself, and who generally 
 succeeds finally in supporting neither himself nor her ; but the 
 people are not all of this shiftless class, though so many of them 
 incline that way. 
 
 This afternoon I gave away, with much reluctance, my thermom- 
 eter, with a promise exacted in return that an accurate series of 
 readings should be kept up four times a day throughout the winter. 
 The parties of course failed to keep their agreement. It is cold to- 
 day and freezing, with the wind northwest. 
 
 Yesterday, and to-day (Thursday) the wind has continued north- 
 west. One of the boats at work on the wreck returned this morn- 
 ing with its mizzen-mast, a huge, iron-bound affair about seventy 
 feet in length, and together the men hauled it ashore, out of reach of 
 the water. It has been cold, cloudy, and threatening weather all day. 
 
 Friday it snowed hard all day, and the wind began to blow very 
 fiercely, increasing in strength towards night, while the temperature 
 continued to fall. A snowstorm here is somewhat different from 
 one in the States. The snow begins to fall very moderate, while 
 the clouds gradually grow heavier and heavier, and the flakes fall 
 thicker and faster until the sight becomes so blinded that, dazzled 
 and bewildered, one can with difficulty distinguish objects a few 
 rods away only. It comes with a fierceness scarcely credible, and 
 a suddenness hardly less so ; while it will often clear up as quickly 
 as it came on. 
 
 Saturday was a quiet day, but cold, with the storm and wind still 
 heavy, the latter freshening towards night. 
 
 Sunday, Nov. the 21st. Last night about midnight the wind be- 
 gan to blow and the elements to war fiercely. This morning the 
 wind increased in violence, and spring tides rose to a greater 
 height than they had been seen before for a long time, and nearly 
 carried away the woodpile in front of the house : this is a large, 
 conical shaped pile of fir and larch sticks, each ten to thirty feet 
 
 i 
 
 
 1 1 
 
r r- 
 
 1 1 
 
 134 
 
 PREPARING FOR BAD WEATHER. 
 
 aiiii 
 
 
 \m 
 
 
 
 m ': ' 
 
 ■i! 
 
 «ii 
 
 high, and about eight inches in diameter at the base, stacked with 
 their small ends upward of course, thus making the whole affair 
 look like an immense Indian wigwam. It is place* m front of the 
 house and on a slightly descending slope, about ten rods from 
 the sea at low tide. The temperature throughout the day scarcely 
 varied from 18°. Frequent snow squalls, a wind direct east, and a 
 sudden and tremendous fall of the barometer, told that we might 
 expect a "spell of weather," as the people say, or even worse, with 
 the possibility of a hurricane. As a means of preparation, al- 
 though it was Sunday, the old adage of ** a work of necessity," 
 though it was hardly of "mercy," seemed to justify us in mustering 
 all hands and preparing for the worst. The worst appeared then 
 to be the hauling of the big logs, — as many of them as we were 
 able — which had been towed ashore from the wreck as far up on 
 the beach as our limited force of four men and ten children 
 would permit. This was not far, however, as solid oak logs some 
 eighteen or twenty inches square, and thirty to fifty feet in length, 
 are anything but easy handling. We then hauled all the boats, 
 those at least that seemed at all exposed to the fury of the wind 
 or water, as far up on the rocks as we were able. They were very 
 heavy, and to haul them we were obliged to use a "tackle." This 
 is an oblong block, with two wheels inside, separated by a narrow 
 partition, over which the ropes run easily, while the block is fastened 
 to the boat by a large iron hook. The end rope is fastened 
 to a ring in the top of the block, while the rope passes over 
 the wheels and through a single wheeled block some distance 
 back, this latter block being fastened to a large immovable anchor. 
 In this way a small force will literally run away with the log or boat 
 attached, which a large force may exert its utmost to move with 
 the hands. After working hard all the morning, it seemed as if 
 we really had everything prepared for the storm, should it break in 
 fury upon us. 
 
 About noon the thermometer rose to 36°, and the sky showed 
 signs of clearing ; but it was only a sign, for soon the wind started 
 up again, and blew a perfect gale from the southwest, and the 
 
ANXIETY FOR THE WOOD. 
 
 135 
 
 darkened clouds, that everywhere enveloped the sky shutting out 
 the faintest shadow of blue, began to send down rain in torrents. 
 We now began to realize that a Labrador hurricane was fairly 
 upon us. Soon the rain gave place to hail, then the hail to rain 
 again, and even fierce flurries of snow scattered huge flakes in 
 every direction. The wind blew harder than ever, and it soon be- 
 came dangerous to go outside the door. About two o'clock in the 
 afternoon the tide reached so great a height as again to threaten 
 the woodpile in front of the house, and though this was placed 
 higher up on the beach than last year even, and very near the 
 house, the waves reached the foot of the pile, and gradually began 
 to lap against its butts as if to undermine it ; there was real 
 danger that if we did not lash it with ropes the whole pile would 
 be carried away with the tide. 
 
 In this cold region, wood is by no means as abundant as it is in 
 most countries. Here the coast produces little save stunted 
 growths of fir, spruce, and birch ; the latter only is really fit for 
 firewood. Generally each man is obliged to go up some of the 
 rivers near his abode, and cut his own wood from the larger 
 growths ten, fifteen, or even twenty miles inland. There the 
 trees are from twenty to thirty or forty feet high, and six 
 to ten inches in diameter at their bases. These trees are trimmed, 
 then formed into rafts and towed down to the stage-head of the 
 dwelling for which they are intended. As all this requires labor, 
 the wood becomes correspondingly valuable. It is generally 
 kept by being piled with the butt ends resting upon the ground, 
 and the smaller ends resting upon each other in the air, like the 
 ends of the poles of an Indian mishwap. At Bonne Esperance 
 the wood is nearly all so piled ; it would thus be easilv undermined, 
 and probably all lost if carried away by the tide. TI rricanes are 
 no respecters of woodpiles, and we expected each moment that it 
 would be necessary to brave the tempest and rush out and lash the 
 pile lest the waves should carry it bodily away. 
 
 About four o'clock the tide turned, but the wind had increased 
 rather than diminished ; and blow it did, hard and strong, fierce and 
 
 11 
 
 
 
 
 
 >' 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 "■ 
 
 \ 
 
 
 
 
 ,' ll 
 
 
 > i V 
 
 
 '^m^ 
 
 
 m| 
 
 
186 
 
 STORM INCREASING. 
 
 cold ; the clouds were as thick and dark as ever. An occasional 
 flurry of snow came down, and the scattering raindrops began to 
 freeze as they fell ; it was hard to find standing places where the 
 ground was not frozen and treacherous. The wind had also drifted 
 the snow, into ridges and drifts of uncomfortable depth, though, 
 thank fortune, the cold had hardened the crust so that one could 
 readily walk upon it, could he once gain a foothold ; thus with 
 ice, glare, and drifts it was nearly impossible to face the wind that 
 whistled and blew so terribly. The temperature by this time went 
 rapidly downward, until about dusk it reached i8° above zero, — 
 not 'cry cold for this region, but just in the beginning of winter 
 cold enough to freeze everything liquid that presented its surface 
 to the air. 
 
 The wind continued to blow, fiercely and more fiercely, long 
 after the dark heavy clouds had shut out what glimpses of daylight 
 remained struggling faintly with the ajjproaching night ; and we 
 could hear the howling and roaring of the temi)est without, as the 
 mingled sissing and whistling of the wind combined with the 
 crashing and the thrashing of the waves against the shores, and the 
 thug ! thug ! of the broadsides of water that fell upon the rocky 
 beach. It was a fearful night. Any other than a strong, well built 
 house would have with difficulty stood against the tempest ; and, 
 believe me, there was very little sleeping done that night. 
 
 Monday the 2 2d. The storm is not yet over. We awoke this 
 morning and found the same war of elements to whose music we 
 listened as we sank to sleep last night. I had the pleasure of wit- 
 nessing this morning, and in fact throughout the day, a sight that I 
 shall not probably see again soon. The sea in all its rage and 
 power, lashed by the wind and augmented by the tide, vents its 
 fury upon itself. Huge ridges of wave press forward like an army ; 
 suddenly each end concentrates its waters toward the centre of the 
 ridge and with a grand onward, upward rush sends up a huge mass 
 of spray and white foam which still shoots upward in smaller angry 
 sprays and jets, while the mass of wave below sinks down into the 
 water, or boldly crashes or tumbles over itself as it splashes on 
 
SITUATION OF STAGE-HEAD. 
 
 137 
 
 the water in front of it, — an angry sea, lashing itself into foam- 
 covered patches here and there, and everywhere, as far as the 
 eye can reach. 
 
 It is next to impossible to do justice to a tempest on these Lab- 
 rador shores. While on our own coast at home 1 liavc seen 
 presented a truly magnificent sight, when a storm sent long ridges 
 of foaming and sparkling waves, with their watery foam-l)earing 
 crests towering up ten, fifteen, or even twenty feet into tlie air, 
 following each other with regular step, beating at length upon 
 the sandy and often rocky shore with cjuick splashes, that sent the 
 foaming and watery masses high up on a beach extending miles 
 in uninterrupted distance in either direction, and felt that I 
 could go home and tell of a storm at sea, and of its fierceness, 
 grandeur, and magnificence. — yet the memory of such a scene 
 is of little account in comparison to that which I shall now 
 describe, and which I not only saw uid felt, but that those 
 around me who have witnessed the terrors of a tempest upon 
 these shores not only saw but felt, and which you will soon see 
 they have cause to remember. 
 
 Bonne Esperance island, where we are at present, is a small 
 island not more than two miles in circumference. It contains two 
 small elevations, one or two small ponds of water, and is separated 
 on the north and northwest by a narrow band of water from the 
 mainland. On the western side of the island are the stage, and 
 the several buildings connected with the fishery department which 
 reach to the water's edge, extending from this backward towards the 
 path. A board walk extends up the embankment to the level above 
 where the fish flakes are spread ; back of these are sheds and 
 outhouses, while a descending slope brings you, a little to the 
 left, to the house which is nearly on a level with high tide mark in 
 front, and which looks directly out upon the waters of this little 
 bay-like inclosure, thence to the waters of the open ocean. On a 
 perfectly clear day the highlands of Newfoundland can be seen 
 fifteen miles directly east. Close in-shore the island is one mass of 
 rocks. Sometimes low, but broadly rounded tops of bowlders show 
 
 Br.Sim!air*fi»f1KST^*3SmrS:^r^r. ■■ 
 
 >af^ J,'rigB^^- ^- 
 
If I Wii > I 
 
 m 
 
 138 
 
 GRAND BUT TERRIBLE SCENE. 
 
 
 themselves, while on the southeast a low, receding cliff receives 
 the force of the waves in this direction. Just beyond the house is 
 a small s;^ndy depression called salt-water pond, since it is filled 
 with salt water at high tide only. All these peculiarities of the 
 locality are seen at a glance from the front doorstep, and the porch 
 itself is only a few rods from highest tide marks. In this out-of- 
 the-world region, this bleak, cold, desolate Labrador, over a thou- 
 sand miles from home, I am spending the winter. On this morn- 
 ing, in this terrible weather, while still uncertain as to that which 
 may follow, I stand and gaze upon a Labrador hurricane. On this 
 little island I am far enough from land to get all the benefits of the 
 raging ocean, to see its whirlpools and billows, and watch its 
 majestic, towering columns of water as they rise and break in all 
 possible shapes to the right, the left, and in front of me. It is, 
 however, sufficient ; I shall remember the sight for the rest of my 
 life. While I cannot do justice to the scene I can only add my 
 own feeble expres. ions of the terrific violence and the awful grand- 
 eur of this day's scene as it stands pictured on my mind. The whole 
 sea in commotion, actually stirred to its very depths, and break- 
 ing into foaming masses, in places where in calm the depth 
 was thirty and even forty feet ; and you could even see the water 
 black with mud from the clayey bottom, of frequent occurrence here. 
 From a short distance seaward, a long line of undulating wave 
 quickly forms, increasing till its crest is a sheet ^i foam, while 
 the wave itself breaks up into smaller masses some of which sink 
 into the waters below, and forming sort of pits or whirlpools are 
 lost in a well-like abyss, while others raise themselves again high in 
 the air and shower forth spray like the jets of a Parisian fountain. 
 Each successive wave thus turns and twists itself into the utmost 
 variety of forms and shapes, while wave succeeds wave with the 
 gi'eatest rapidity. 
 
 The narrow islets in the distance beyond are but playtliings for 
 the waves, and huge billows roll completely over the lower ones, 
 while white spray tosses itself from one side to tlie other of the higher 
 yet still low crests. It seemed as if the waves on either side of these 
 
STORM CONTINUES 
 
 139 
 
 for 
 
 heights delighted to show their mighty strength, while they vied 
 with each other in throwing their spray over them and into the 
 sea beyond. Now all the rocks and points of land, of which there 
 are many jutting out into the sea on either side, are in turn covtred 
 with a mass of foam and white-capped spray, while the water 
 rebounds high into the air fifty or a hundred feet at least ; some- 
 times it breaks over the rocks with a report like thunder, sending 
 long sheets of glistening spray vertically forwards. All these varied 
 scenes we beheld in one moment, and almost at a single glance, 
 and more beside. The wind, the rain, the blinding sleet or snow ; the 
 dark heavens, and still darker horizon ; the foaming, seething, and 
 hurrying, the turning and twisting masses of mad, white, frothy, 
 granular (for I can think of no other word to express the peculiar 
 effects of light spray upon a darker ground of water) watery crests 
 that rise and fall, or cover the whole surface of the water as for as 
 the eye can reach, are all seen in a moment. No pen can do 
 justice to the terrific grandeur of the scene ; no picture can give a 
 real impression of its awfulness. My memory can scarcely retain 
 the thoughts that the moment or moments (for the storm thus 
 continued without respite all day) inspired. And yet I have not 
 told all. 
 
 To have simply witnessed a storm of this description, and known 
 that it could do damage to any of our property, would in 
 itself nave been a sight to have recalled continually in after life. 
 Remembering the care and pains taken to save everything useful 
 in this out-of-the-world region, the expense of transportation, and 
 the real value of any little improvement when once made, you 
 can go with me to another part of the island, not far distant, and 
 look with renewed awe ui)on the destruction going on there. 
 Struggling fiercely against the wind, just before the additional dusk 
 of evening set in, we hurry down the little hill at the west to the 
 stage-head and wharf, to see if all is safe there. Three weeks ago 
 to-day the lidward Cardwell was wrecked, as I have before des- 
 cribed, and the cargo of huge oak, pine, and other varieties of 
 wood logs, beside deal boaids, scattered far and wide along the 
 
 i. 
 
 i '4 
 
T 
 
 111 
 
 in 
 
 140 
 
 FOREBODINGS. 
 
 
 ^11 
 
 1 1 
 
 i 
 
 i » ' 
 
 1 
 
 li • 
 
 1 >J 
 
 1 
 
 i; 
 
 coast. The gentleman with whom I tarried (Wm. W. H. Whiteley), 
 finding that he could dispose of them to good advantage, sent out 
 boats to pick up all that they could find, while he offered to purchase 
 for cash, or trade from the little shop of provisions and dry goods, 
 of all those who should bring him either logs or boards. In this 
 way he gathered a large lot of lumber. The huge logs had been 
 hauled above a supposable high water mark — for this heavy pine 
 and oak timber requires a dozen men to remove a single log — and 
 been allowed to accumulate on all sides about the '■•♦^age-head, 
 which was thus closely hemmed in with these solid ti -ers, many 
 of which were eighteen inches in breadth and width, and forty to 
 fifty-four feet in length. There they had lain, rising and falling 
 lazily at the highest tide only, for the past two weeks. The stage 
 head, of which I have spoken, is a sort of shed built out into 
 the water upon logs that inclose huge masses of stones to weight 
 the platform and hold it in place, while the main building is propped 
 up with upright posts, buried some distance in the earth and mud 
 below, supporting the flooring at a distance of some eight feet from 
 the bottom at low tide. 
 
 At the stage-head the boats are fastened, and here the fish are 
 cured and packed ; here are a great majority of the other stores of 
 the fishermen, — the boats are housed for the winter inside the 
 building ; while barrels of fish, flour, and hogsheads of cod oil, 
 with all sorts of implements and utensils are stored also. There is 
 often a second stage separated by a platform from the first, on 
 which platform are stored boards and plank of various kinds, 
 hogsheads empty or full of refuse cod liver, or blubber as it is 
 termed, while the stage house itself stands farther back, the platfonn 
 with its railing being nearest the sea. Vv'liilc all this hurricane is 
 raging, and while it is almost impossible to walk ahead a dozen 
 stejjs, it seems proper to "take a look at the stage" to see that all 
 is right there ; and what do we see ? In this little sort of natural 
 harbor the waves are stirred with almost the force that they are in 
 the open sea, while the water breaks to the bottom in huge billows 
 which, lifting themselves in their fury, rush forward and hurl them- 
 
DESTRUCTION. 
 
 141 
 
 selves on the jam of logs that I have mentioned as lying all 
 around the stage ; the logs are gradually breaking loose and drifting 
 about at the mercy of the billows. It is spring tide, the water is at 
 its height, and we are powerless but to stand and watch the scene, 
 while ten or fifteen huge logs are dancing about as if they were 
 playthings. Now the water breaks over both stage-heads, making 
 it impossible for one to rescue the provision and other material there 
 stored. The foam flies in blinding sheets over boards, barrels, and 
 hogsheads (or puncheons as they are here styled) alike ; it dashes 
 against the sides of the sta^e house, and freezes in heavy masses 
 wherever it touches. The wind whistles, or rather hurries past us for 
 it flies too swiftly to whistle, and still we can only watch the scene. 
 
 Soon the logs are nearly all loose from their moorings ; we watch 
 the waves as they catch them and hurl them with terrific violence 
 against the slender underpinning of the stage house, — which but 
 for this gale would have been sufficiently tight to have stood all 
 ordinary weather during the winter — and crush down the foundation 
 posts as if they were small sticks. Crash follows swiftly after crash. 
 A dull //wg, and the farther part of the stage house falls, roof 
 and sides, upon the precious stores contained within, while the 
 inner edge of the platform, thus loosened of its foundation posts, 
 sinks several feet. The topmost crests of the waves sweep over 
 the whole, though not with sufficient force to carry anything away. 
 Over this slippery mass it is impossible for one to think of walking 
 for an instant. We rush for the inside of the stage, after opening 
 the large outer doors at the end, and are appalled to find half the 
 flooring already swept away and the large new boats balancing on 
 the edge of a slender support that threatens to give way and 
 engulf them instantly. There were four of these boats, each one 
 worth at least one hundred dollars, and had they been carried 
 away, it would have been a great loss and one not easily replaced. 
 As it is we are in time to save them, and do so at great risk. 
 
 While doing this another crash comes — for the logs are still being 
 hurled against the underpinning of the stage house, though their 
 number and clustered weight now hinder what otherwise one or 
 
 I 
 
!l 
 
 142 
 
 STAGE GIVES WAY — CALM. 
 
 Mi I 
 
 iP' 
 
 i '. ri-' 
 
 ti;;".: 
 
 iijiii 
 
 4 
 
 i 
 
 I ( 
 
 two logs might have easily accompUshed, in the destruction of this 
 with also perhaps that of the other stage entirely — and another 
 part of the stage and platform, thus undermined, falls. We rush 
 outside and are just in time to see the water lift it with its waves ; 
 to see the platform bend lower and lower until, with a final crack, 
 it parts, and hurls barrels and boards upon the rising crests of 
 the billows surging upward to receive them. Powerless to hinder 
 it we see several barrels of fish, and five barrels of flour of another 
 man — who could ill afford the loss — go towing about with the logs 
 on the watery billows, mingled with boards and debris of hay, chips, 
 and all sorts of material swept clean off the stage. Presently two 
 logs catch one of the barrels of flour between them ; opening for 
 a moment they come together again with a crash, the barrel is 
 burst in an instant, and a cloud of flour flies in all directions, 
 while cakes of dough go floating off on the waves. Three barrels 
 of flour and several of fish are speedily disposed of; one apparently 
 sinks bodily as it disappears and is not seen afterwards, while but 
 one is saved whole and another in part. But unless the storm 
 increases the remaining damage will be slight, and we return to the 
 house. During this gale, to-day, the barometer went down to the 
 lowest that some twenty years of readings have given it upon this part 
 of the coast, reaching 27.32 ; the thermometer 19°, and the wind 
 still southwest varying to north. At length night comes, and we go 
 to sleep again listening to the roar of the tempest, which, however, 
 shows signs of abating ; still we are anxious for the morrow. 
 
 Tuesday the 23d. Our hopes have not deceived us, and this 
 morning we awake to find the tempest moderated to such an extent 
 that we can calmly view the extent of last night's disaster. The 
 waters are comparatively calm, though they still toss about the logs, 
 boards, and miscellaneous pieces of wood that have gathered upon 
 its surface ; chips and debris are floating about on all sides matting 
 the surface of the water like a carpet which undulates with the motion 
 of the waves beneath. The further end of the stage house has fallen 
 in a mass upon the stage-head, which has lost the greater part of 
 its underpinning and part of its platform ; while the right hand 
 
AN ICE SCENE — STORM ENDS. 
 
 143 
 
 ^y\^if 
 
 floor, inside the building, has fallen out entirely, — it is a wonder 
 what has kept the whole building from going. Great masses of ice 
 cover the ruins ; the sides of the stage house that still remain 
 standing are covered with sheets of heavy frozen matter, while huge 
 blocks of icicles hang from the edge of the roof. The railings of 
 the stage beyond hang with icy masses that fall in solid sheets 
 nearly to the ground. Everything seems covered with this dull, 
 opaque, heavy saltwater ice. We see before us a sea of frozen, 
 wooden matter. We spend all the day in cleaning and clearing 
 away, and in trying to find the extent of the damage done ; in 
 relaying the broken foundations, strengthening the beams, raising 
 the broken and bending, though not yet fallen, portions of the 
 platform, and in general repairs. Night finds us more cheery, and 
 sanguine that one or two more days' work will right matters again. 
 Wednesday the 24th. To-day the storm is fully over, and the 
 men have labored hard at clearing away the rubbish, and once 
 more righting things. The wind has been mostly from the north. 
 The water has calmed down once more, its surface stirred only by 
 the low, long lines of wave that occasionally advance and break 
 upon the shoreline everywhere around the island ; the v> hole water 
 is still colored a dark clay green, an evidence that at this point the 
 sea had been stirred to its very depths. The magistrate says that 
 he has never seen it of so intense a color during the twenty years 
 he has resided on the coast. Everywhere upon the shore are clams 
 and mussels, shells of many difterent varieties, starfish, echini, and 
 holothurians of many kinds, besides innumerable other species of 
 sea dwelling animals. The beaches were everywhere literally cov- 
 ered with these treasures of the deep, while the children brought them 
 home by the armful. Beautifully colored starfish, some of them 
 immense fellows a foot or more from end to end of arms, and 
 brown, red, and gray, both light colored and dark, all told of an 
 unusual commotion in the elements of both air and sea. By 
 this time the sky had regained its usual and natural hue of hazy 
 blue, with scarce a cloud save in the far horizon where a dull, heavy 
 shadow of the hurricane hung like a ghost or ghoul that seemed 
 
 i! 
 
 ^Ui 
 
^■ 
 
 1 
 
 
 i- 
 
 f 
 
 '' 
 
 1 
 
 ! 
 
 I. 
 
 144 
 
 THANKSGIVING DAY IN EARNEST. 
 
 4 
 
 
 slowly to sink lower and lower, while it grew fainter and fainter, soon 
 disappearing entirely in the distance and leaving us once more 
 a fresh, free sort of feeling that, perfectly natural to the region, 
 seems to defy ennui. 
 
 Thursday the 25th. Alas ! that I should be obliged to spend 
 such a day in such a place. I shall hardly dare to suggest to you 
 that it is Thanksgiving Day, but here I am in an English province, 
 beyond the American frontier, where Americans seldom go, to say 
 nothing of carrying their holidays with them ; in a land, the 
 people of which are too poor to celebrate to any great extent 
 should they ever feel inclined, while living too far away to pass the 
 time in complimentary calls ; in families, whose chief diet through- 
 out the year consists of bread, without any butter, and tea, with- 
 out any sugar, — excepting what game may fortunately be cap- 
 tured ; where breakfast, dinner, and supper are the same ; in such 
 a locality, though with the better class of residents, I pass my 
 Thanksgiving. That I may not think myself worse off even than I 
 really am, I cut a slice of cold roast pork, and imagine it the best of 
 turkex' ; another slice answers for chicken ; while a third stands 
 equally well for mince pie. I then start for the woodpile where 
 I work off the effects of so hearty an exercise, as the eating of all 
 these delicacies, by sawing wood for the remainder of the after- 
 noon. Thus ends a Labrador, New England Thanksgiving. 
 
 During the remaining days of the month nothing particular has 
 happened save that the goodman of the house where I am stop- 
 ping has been in bed sick. The first of the month, however, finds 
 him much improved. 
 
 Wednesday, December i . Although it is very cold the bay shows 
 no signs of freezing yet. These winter evenings the children amuse 
 themselves with games. Many of onr New England and oth°r games 
 find favor, though dominoes seems to take precedence of all 
 others. Checkers are played quite frequently. Of course cards are 
 the prevailing game along the coast, and the old "stand-bys" seem 
 to be "high low Jack" and "forty-five," the latter being apparently 
 a characteristic Newfoundland game, as Newfoundlanders indulge 
 
THE "KOMATIK. 
 
 145 
 
 in it at all times of the night or day, and are apparently never too 
 tired to play when they can "form a hand," and 
 
 "When once agoing, 
 With pipes aglowing, 
 They sit it out till morning." 
 
 Thursday the 2d. The boys and young fellows harnessed the 
 dogs for the first time to-night, but did not use them much. As 
 I have a few hours to spare I will try to describe here the komatik 
 or dog sledge and its use, as also the racket or snow shoe of this 
 region. 
 
 In winter the greater part of the travelling is done in one of two 
 ways : either on sledges called " komatiks," drawn by dogs, or on 
 large pads that are called " racquets," and which are worn upon 
 the feet. As each of these requires a special description, I will 
 try to give one. 
 
 The komatik, as has been said, is a sort of sledge or sled, and 
 looks very much like a magnified specimen of one of those latter 
 articles. Its dimensions vary from nine to thirteen feet in length, 
 from two to three feet in width, and it stands about eight inches 
 from the ground. The wood is wholly pine, and the side bars are 
 cut out of thin deal-board, planed down to about one or rarely two 
 inches in thickness, with the front ends turned up like the front run- 
 ner of a sled ; the sides are often bevelled so that the bottom is 
 one-fourth or one-half an inch wider than the top. The upper part 
 of the sled is made of a number of thin pieces of wood (usually 
 thirty-two) of equal length and about four inches in width, with the 
 ends rounded, and then notched — for a purpose that will appear 
 hereafter. The top and bottom pieces are similar, but of double 
 the width, while the thickness of all is about the same, generally 
 one-half an inch, though the end pieces are perhaps a little the 
 thicker. Each piece has two pair of holes bored through it on 
 either end, the distance between each pair being that of the width 
 of the top of the side bar, and the distance between each hole of 
 each pair about half an inch ; between each pair it is then gouged 
 out crosswise about one-fourth of an inch deep, while the inner 
 10 
 
Ill 
 
 r 
 
 !). 
 
 II.* ■ 
 
 ' 
 
 ti 
 ): 
 
 « 
 
 
 t 
 it 
 
 ■ 
 
 tl 
 
 1' 
 
 
 146 
 
 HOW THE KOMATIK IS MADE. 
 
 pair is recrossed at right angles by another gouge, the purpose of 
 which will soon be seen. A curious fact is that all these holes are 
 bored out with a red hot iron to make them smooth and even. On 
 the side bars, at a regular and previously measured distance apart, 
 are bored holes to the exact number of the crossbars. The holes 
 are bored, one a little above, and the next a little below the pre- 
 ceding one, so that when done the whole presents two unequal 
 rows as here shown, .•.•.'.•.•.•.•. hence the liabiUty 
 of thus splitting the soft pine in the sewing process is lessened. 
 
 The next work is sewing the parts together : for this a coarse 
 salmon net twine is threaded into a needle used for the pur- 
 pose, and each crossbar is sewed into the corresponding hole 
 in the side bar, in and out of the holes on either side of the bar 
 itself, and drawn as tight as possible ; the needle then slips under 
 the twine through the groove across the inner pair of holes, and a 
 loop and a stout pull fasten it ; thus each bar is sewed on till all 
 are tight, — but we have not yet finished. The forward end of 
 each side bar must be strengthened by a long, thin iron placed 
 lengthwise along the inner side of each bar (this is the usual and 
 best way of strengthening the ends), and screwed tight to the 
 boards. Then come the shoes. 
 
 Every komatik has shoes or runners as do our sleds, but unlike 
 them they are of whalebone. Whales are so often found dead 
 on different parts of the coast and towed to some harbor where the 
 flesh is cut up for the dogs, and the bones saved for various pur- 
 poses, that the large rib-bones have become a regular article of 
 commerce among the people ; the bones are some eight or ten 
 feet long and nearly or quite a foot wide, with perhaps two inches 
 of thickness. These bones are obtained, then strips the full 
 length and an inch and a half wide are sawed off, and being 
 trimmed they answer perfectly for runners. With a gimlet, holes 
 are bored through them about six inches apart, and they are fas- 
 tened to the bar, which is also bored into, to correspond, by small 
 pointed wooden pegs driven tightly down ; the knobs of the pegs 
 are then sawed off, and the sides of the runners, if they overlap, 
 
m 
 
 HARNESSING THE DOGS. 
 
 147 
 
 are pared and the bottoms planed smooth. From the forward 
 end of the bone a small portion of each runner is trimmed down 
 thin almost to the board, and to this a piece of hoop iron is fas- 
 tened which is brought around over the point of the bar and cut off 
 close to the first crossbar; it is then pegged, through holes, or 
 screwed down. This part of the bar never touches the ground, 
 being curved slightly upward ; then with a hole through each side bar 
 at the end, like our sleds, through which to pass a rope if needed, 
 and with a plentiful supply of paint, our komatik is complete. 
 
 A komatik is drawn by harnessed dogs, and it is a strange sight 
 to see one of these hobgoblin, arctic turnouts travelling at a smait 
 pace over the ice and snow of the frozen bays from place to place, 
 and watch the eight, seven, to even three or two dogs, attached to 
 the sled by a long thin thong only, trotting or galloping along 
 many rods in front. You would hardly believe, at first sight, that 
 the two belonged together, but rather that some fairy means of con- 
 veyance had suddenly dropped from above, or appeared from an un- 
 seen comer below, and was travelling off on some mysterious Arabian 
 Night's adventure ; and truly, the snow-capped hills, the icy and snowy 
 plain of frozen sea, add to this not unpleasant romantic delusion. 
 A close inspection will show that each dog is encased with a thin, 
 narrow band harness that simply goes around the body at the belly, 
 with another piece going around the neck and around and between 
 the legs, which is fastened to the former beneath ; on the top of each 
 a long band reaching to the length of several rods follows on be- 
 hind with a loop at the very end. Each dog is thus harnessed. 
 The komatik also has a band fastened near the front crossbar by holes 
 in the runner, and buttoned together with a loop and an aiiguiar 
 button. The loop on the end of each long dog band is passed 
 over the button of this komatik band, the button fastened in the slit, 
 and the team is ready. One dog is always fastened sone distance in 
 front of the others and is styled the leader ; the others are fastened 
 at various distances apart, but always many rods in advance of the 
 sled. The team is guided usually by the voice, with or without a 
 whip of the Esquimaux make or pattern, many yards in length on 
 
r 
 
 
 (!■ 
 
 !;■ 
 
 '!|||:i 
 
 148 
 
 STRUCTURE OF THE HARNESS. 
 
 a short handle ; with a circle of thick rope to throw over the run- 
 ners and impede the progress of the sled in case of accident or sud- 
 den desire to stop, and which is called a drag ; and a long stick 
 with which to pound and sound the surface of the ice to see if it 
 be safe to cross in doubtful places, the team is ready to start. 
 
 One would think that such narrow bands as these of which the har- 
 nesses are made would be easily broken ; but though the width is sel- 
 dom three-eighths of an inch, the sealskin, of which they are made, 
 is so tough, stout, and unyielding, that they will wear rather than 
 break apart. A trace, as it is called, is often seventy or more feet 
 long ; and when you consider that a large muscular dog is either 
 dashing forward in short, quiet leaps, or straining his utmost to 
 draw a komatik on which a load of eight hundred pounds and over is 
 often fastened (three barrels of flour and other things being often 
 taken at a load), vou can judge somewhat of the strength of these 
 thongs. If the going is bad the strain is in all probability more than 
 doubled. I have said before that the thong wears rather than 
 breaks, and when it does give way it is more often by the weak- 
 ness of some closely thinned place or the rotting of some portion 
 of the trace than by any other means. Sealskin thong is used every- 
 where for rope, when it ^an he procured, for tying or binding. No 
 wonder that the old legends, which till now I had regarded as such 
 in reality, were more than true, when Indians bound their captives 
 to trees quickly with deerskin bands (for deerskin is nearly as tough 
 as sealskin, though it perhaps gives more easily), which it was 
 impossible for them to break. The quickest and easiest to make, 
 yet surest tie is called a double half-hitch ; the rope is doubled 
 over itself in two loops with both ends projecting from the inner 
 side of the loop. It is made in about three seconds, and the more 
 one tries to stretch it the tighter it becomes, especially if the cut 
 ends are fastened to some object behind or even to themselves. 
 But the dogs (these wretched half-breed of Esquimaux and wolf 
 perhaps, that spend the greater part of their time in fighting each- 
 other, to whom play is fight and fight death, and I do not exag- 
 gerate these brutal furies) have started, and the footmen follow. 
 
 Ill ' 
 
THE RACKET OR " RACQUET." 
 
 149 
 
 The foot, as I have said, is encased with a pad called a 
 " racquet." It is a bent bow of wood with two crossbars, the in- 
 tervening spaces being thickly woven of deerskin thong, except a 
 small opening where the toe goes, and which is below the middle 
 of the upper bar. There are great varieties of form, called usually 
 from some fancied resemblance to the tail of an animal. The 
 beaver, the otter, the porcupine, and the bobtailed rackets are used 
 perhaps more frequently than any others about this coast, though 
 the long racket is used throughout many of the Canadian prov- 
 inces. 
 
 In walking one takes the usual step, though perhaps a longer 
 one than otherwise, having the feet much farther apart, and in the 
 walking position the convex portion of the top part of the racket 
 of the hind foot fits almost exactly into the concave portion of the 
 bottom part of the racket of the front foot. In stepping, the 
 wide part of each pad passes over that of the other ; you will 
 thus easily see that in hurried walking, if each step does not at first 
 clear the racket of the other foot, it is impossible to take a step 
 with that foot which thus sustains the whole weight of the body, 
 and the impetus given to the body invariably plunges one down 
 into the snow. With the swinging motion necessitated by the 
 unusual space between the feet, a little practice soon accustoms 
 one to walking easily over the lightest and deepest snow if it be 
 dry, as it usually is here, and not wet and sloppy. 
 
 A short time since the dogs were ready to start off on a journey 
 or trip with their master or driver ; and now several strong men 
 without stooping, and with a motion peculiar to habit from long 
 practice, drop their rackets upon the ground and adjust their feet 
 in them, — now both start together for the trip. The going is not 
 very good for the sledge, and the nmners cut deep ruts in the snow, 
 already upon the ground to the depth of over a foot on a level, 
 as it is slowly passed over. The dogs travel with difficulty, and 
 need constant urging from their driver who keeps up a continual 
 shouting and crying out at them. 
 
 A string of a dozen or more rapidly repeated words which sound 
 
 
 m 
 % 
 
u 
 
 1 
 
 1 
 
 
 
 
 1 
 
 1 ' ! 
 
 (r'-^ 
 1 
 
 
 il 
 
 ■i'i 
 
 '''M 
 
 
 
 : 
 
 ! 
 
 j^^H| 
 
 
 
 
 t 
 
 l^ffl'^ 
 
 
 160 
 
 HOW TO DRIVE THE DOGS. 
 
 riMl'j! 
 
 more like a series of hi, hi, hi, hi, etc., or ki, i, i, i, i, etc., sends the 
 troop off at a quick pace, while the leader is easily turned to the 
 left by a series of sounds, of which either rudder, rudder, rudder, 
 or da, da, da, da, da, or udder, udder, udder, udder, seems to be 
 the only interpretation (I have found no one yet that can tell me 
 the exact word used) ; and to the right, by another sound like 
 ouk, ouk, ouk, or owk, owk, owk, repeated in a hollow, guttural 
 tone. These are the regular sounds for guiding the leader, while 
 a continual shouting or clapping of hands and a variety of any 
 small words are poured forth till one would easily imagine the 
 driver to have seriously shouted himself hoarse. If any object 
 appears in the distance, this too is a theme for the urging forward 
 of the dogs, who seem to go well enough without all this noise. 
 A crow flying close along the ground, and a komatik in the distance 
 share alike the o-look', o-look (which is grunted rather than spoken 
 out) , crow, crow, or caw, caw, caw, or komatik, komatik, komatik, 
 to which is often appended a single shrill hi. Whether the dogs 
 need this constant urging, or whether it be a custom that seems 
 to have been handed down or not, I cannot conceive. Mean- 
 while the men who are walking have nearly or quite kept up with 
 the sled. The long strides enable one easily to double the ordinary 
 walk of a person, so those behind have kept quite near. On they 
 both go — the dogs at a smart trot which occasionally breaks into 
 a gallop, and the men who present a truly comical appearance as 
 they press forward swinging their legs with their huge appendages 
 and their arms, occasionally their whole bodies, with the violence of 
 the exercise. The wind, which blows the light snow on the top 
 of the thick crust of the late storm in whirls and clouds all about 
 them, sings around their ears and faces, making the one tingle, 
 though encased in a warm covering, and the other ruddy Mi jc 
 glow of health. It is foolishness for one to venture uch 
 
 weather, as is here of every-day occurrence, without c iticient 
 amount of good thick and warm clothing ; and even the do^ sef a 
 to be provided by nature with an unusually thick coat of long flow- 
 
A "TRIP." 
 
 151 
 
 ing hair as their best protection from the severe cold liable to occur 
 here at almost any time with a few hours' notice. 
 
 A joui ney of six to ten miles brings them to their rlestination, 
 and then a halt is made. The trip may be simply to a neigh- 
 bor for some visit, or to borrow or return some loaned article, 
 or it may be to the provision depot for a supply of goods or 
 articles for the use of the family. The molasses keg needs re- 
 filling or the tea chest is empty, perhaps the butter is all gone, or 
 soap, matches, or broken dishes need replacing with new supplies. 
 Often the men are out of tobacco which, to them at least, is a neces- 
 sary article, when the greater portion of the time is occupied in 
 filling the room with volumes of deep blue smoke while assembled 
 around a (almost literally) blazing hot stove telling stories or 
 silently enjoying the smoke and passing time away. It may be 
 that the housewife needs a new dress, or the children extra cloth- 
 ing, while often warm shirts and mittens of swanskin are to be 
 procured for both. At any rate the articles are obtained, often, I 
 am sorry (for the families' sake) to say, on credit of next year's 
 " ketch" of fish, and the little party turns toward home again, which 
 is soon reached, as are also the warm fire, the hot supper, and bed. 
 
 ;of 
 
153 
 
 A TRAMP. 
 
 ) 
 
 CHAPTER X. 
 
 It 
 
 ■ ; I 
 
 1^ 
 
 ■II: 
 
 !li 
 
 Trip up the River to the Mission — Ice pictures — Bad walking — On the 
 Old Fort — New scenes and bad walking — Pleasant Sunday — The 
 return — Journal — A komatik ride — Christmas gathering — Wood cut- 
 ting — Work for the evenings — Making sealskin boots, mittens, and 
 other needful and fancy work. 
 
 Saturday the 4th. With one of the men I started to-day on a 
 tramp up the river to the Mission station, about seven miles dis- 
 tant. It seemed good to be on the move once again, although we 
 got much more of it than we had expected before the day was 
 endtu. The river was frozen but a part of the distance, so we 
 were obliged to go by boat to the land at its mouth. It did not 
 take iong, however, to get the boat ready, and we were soon 
 on our way. At first we rowed through the soft masses of ice that 
 coated the water everywhere around the north side of the island, 
 but finding this passage closed with ice too thick for our boat to 
 break, we retraced our way, and, fortunately, found the passage on 
 the other side clear all the distance to the land on the opposite 
 side of the bay. The ice that seriously impedes passage at this 
 season of the year is called here s/ol>. It is a thick, consistent 
 mass of frozen salt-water that lies in huge patches all over the sur- 
 face of the water from land to land. While in this soft condition 
 it is easily rowed through ; but the danger is, that, as the cold 
 strengthens at any time of the day or night, it is liable to congeal 
 suddenly, even at a few moments' warning, when it grows harder 
 and thicker every hour. If a boat is within it, the congealed 
 mass has at times been known to surround it so as to prevent es- 
 cape. The mass is thus too hard to allow the men to rescue their 
 
MAGNIFICENT SCENERY, 
 
 153 
 
 -' 
 
 
 heavy boat, and too soft to walk upon. The whole sheet, mean- 
 while, cracked fr :ni the edge on either side, slowly but surely 
 moves with the current until it is gradually taken out to sea to 
 be lost in the gulf or ocean. Judge then the fate of those unfor- 
 tunate beings who have remained in the boat, thus inclosed, mov- 
 ing only when it moves, halting only when it halts. 
 
 The passage on the west side of the island, as I have said, was 
 fortunately clear, so that we gained the opposite side of the bay 
 without accident or delay. This left us half-way to our place of 
 destination, the remainder to be tramped over the hills, with snow 
 knee-deep to opposite the houses, to which we must get by shouting 
 for a boat to come and take us over should the water between 
 be open, as we had reason to fear. Having called at a house 
 of one of the inhabitants near by to deliver some messages, we 
 returned and began our walk, or rather tramp, over the partially 
 frozen ice of the river at this point and towards the hills beyond, 
 over which we were obliged to travel. 
 
 I have read, in books of arctic travel, of snow-capped hills and 
 of snow-filled valleys ; of rivers and bays frozen over with dark, 
 semi-transparent salt-water ice, but never before had I experienced 
 the pleasure of beholding its reality. Now I saw them in all the 
 half-frozen splendor of a semi-arctic latitude. I could also tramp 
 over their treacherous surfaces with a feeling of perfect safety 
 knowing that the distance to the houses up the river was not far. 
 
 The walking was by no means easy. We first crossed the river 
 on the ice which at this point was sufficiently strong to bear us. 
 The ice was not very solid, and big holes anil huge cracks were 
 to be seen everywhere about us, so that it required a great deal of 
 manoeuvring to find secure footing and a safe passage between 
 them. We would walk along safely enough for some distance, 
 and then the dark colored patches of water, seen on the surface 
 of the snow, v/ould show us where the dangerous places were ; 
 then again the ice contained large cracks, and we were ol)liged to 
 prove its solidity with a stick before trusting our feet to step upon 
 it. Sometimes the ice would be as large floating cakes when 
 
hr 
 
 '» *' 
 
 154 
 
 DIFFICULT WALKING. 
 
 we were compelled to jump swiftly to get safe from one to an- 
 other, each step sending the cake deep into the water beneath, so 
 that the person behind would be obliged to wait until it had re- 
 gained its position once again before attempting a similar passage. 
 
 Along the edges of the river the ice had formed and cracked 
 again, leaving the walking very dangerous as well as difficult ; this 
 often necessitated the climbing of long, steep, icy slopes before 
 gaining a secure footing. Then again the ice had formed over 
 beds of sharp rocks, often at high tide, and the water retreating 
 had undermined the support, when suddenly, while walking over 
 these, they would give way and we would fall often several feet 
 on the rocks below, making most treacherous as also dangerous 
 travelling. Again long open stretches of water would compel 
 us to go by the land, either around some inland pond or lac sale 
 (salt lake, as the people call these inlets of the sea, and of 
 which there are so many all along the coastline), or over some 
 low and narrow, or high strip of land to the river again. Some- 
 times we would crawl along the edge of some high and sloping, 
 or steep and rocky crest, often coming suddenly to the brink of a 
 precipitous height of twenty to thirty feet, where a single false 
 step would have sent us to the rocks or sharp ice below; or 
 perhaps we would suddenly sink to our armpits into some con- 
 cealed, snow-covered spruce thicket, from which we would ex- 
 tricate ourselves only with great difficulty. 
 
 From some high hilltop we could see lofty crests sloping to some 
 narrow or often perpendicular cut, through which the river, now 
 frozen and now with open glistening waters, ran towards the sea, a 
 distant maze on our south. A mile or so ahead we could see the 
 buildings, and the white church with its steeple, of the little Mission 
 station. Here are about a dozen buildings where many of the 
 families that live upon the islands outside in the summer, together 
 with the foreigners who conduct the school and Sunday church 
 services, pass their winter quite cosily and more or less merrily 
 in teaching and being taught. 
 
 It was a difficult tramp, but a delightful one. At length we 
 
RESTING: THEN ON AGAIN. 
 
 155 
 
 we 
 
 reached a position on the hills directly opposite the houses on the 
 other bank. After a great deal of hallooing and shooting blank 
 discharges of powder from our guns, a boat started out from the 
 other side, and slowly approached us ; soon we were on the oppo- 
 site side and close beside a nice warm fire, drying our wet clothes 
 and partaking of the hospitality of a kind-hearted inhabitant of this 
 little settlement. 
 
 Here I found two young men who were destined for the same 
 point that I hoped to reach that night, so I decided to avail 
 myself of the opportunity and accompany them. 
 
 We first called on some of the inhabitants of the place, and 
 found them cordial and genial people, well meaning and hospitable 
 to strangers. I found myself invited to partake of bread and tea at 
 each place, and at last was obliged to refuse absolutely even the tea 
 which was thus generously pressed upon me, and which is the 
 beverage so abundantly partaken of all along this coast. At last 
 we had finished all our calls, for the present at least, and so we all 
 started for Old Fort Bay, seven miles distant across the river, the hills, 
 and several ponds. Since our journey presents a continuation of 
 the one I had just taken, except inland instead of along the di- 
 rect coast line, I will try to give you some idea of it. 
 
 The scenery here is very similar to that we had just witnessed, 
 with the same features of frozen river and lakes, and journeyings 
 over hills and snowy slopes ; but there was enough variety to 
 make the trip new and pleasant, rather than full of tedious mo- 
 notony. 
 
 We started on our journey by rounding the bend of the river 
 which had this year been frozen over earlier than usual, when we 
 came at once upon a long stretch of open ice. In this part of the 
 river, as in fact in any of these salt-water lake basins, the water usu- 
 ally freezes several times before it finally becomes caught on either 
 side. Each time it freezes, the soft ice loosens and goes floating down 
 the river ; thus several days of alternate freezing and open water 
 occur usually before the ire ,i .ally catches for the winter. Strange 
 to say, this year the ice c- ^nt and held the first time, and so, as 
 
F^^^iwjw: r:»» 
 
 ^^mmmmmi^^mmmmm 
 
 
 l! I 
 
 r,i 
 
 156 
 
 BETWEEN THE MISSION AND OLD FORT. 
 
 we were yet quite early in the season, we walked over a freshly 
 frozen yet perfectly solid surface in the direction in which we were 
 going. 
 
 We had hardly rounded the turn before one of the fellows, 
 whose eyes had in this case been sharpened by expectancy, espied 
 the fresh tracts of the ptarmigan in the snow of the left bank of 
 the river ; but of course the birds had long since gone. Un- 
 doubtedly they were not far off, though we had no time to hr.nt 
 them up. We continued our walk, therefore, enjoying the clear, 
 cool, and healthy air while drinking in the unusual sights, at least 
 to me, of hummocks and hillocks rising in the distance, one above 
 another, and stretching one beyond another, with varying gorges, 
 and again solid walls of granite, into the distance beyond. After 
 travelling an hour with such views constantly presented to us on 
 either hand, we passed a little ridge of ground and came to another 
 so-called /ac sale, a bend in the river almost converted into a lake 
 by the stricture at its mouth. Passing over the frozen surface of 
 the lake we came to another ridge which, passed, brought us again 
 to water — this time a pond. A succession of two or three of 
 these ponds crossed by low, narrow ridges of rising ground brought 
 us to the head of the eastern arm of Old Fort Bay, which we 
 found barely frozen, and for the most part a terrible mixture of 
 broken ice, rocks, and water. We had great difficulty in getting 
 from the ice to the "hore, as the cracks were large, and very wide ; 
 but we succeeded in doing so at last, when we found ourselves at 
 the foot of a large, heavy ridge of hills which extended for the 
 remainder of the distance between us and our destination, varying 
 from four to five hundred feet in height, and by no means easy to 
 ascend. Had it been high tide our way would have certainly led 
 over these hills, as their perpendicular faces towards the v/ater 
 would have rendered walking in that direction entirely out of the 
 question ; while the way would have been picked out only with the 
 greatest of care in the dusk of evening, in which we would walk 
 among spruce thickets, over stones, and down steep declivities, 
 quite dangerous indeed in the daytime. 
 
 r 
 
 II 
 
TREACHEROUS WALKING. 
 
 157 
 
 
 We had no boat by which we could row the short distance now 
 separating us from the house to which we were going. The tide was 
 dead low. On we went, the stones and corners of the rocks hurting 
 our feet — for we wore the dogskin boots of the Esquimaux, — 
 while the huge cakes of broken ice often caused us to stumble in 
 quite a hopeless manner. Again, in spite of the low tide, the 
 cliffs came almost down to the water, and we were frequently 
 obliged to wade in the water up to our knees and much above the 
 top of our boots, or climb over most dangerous places at the risk 
 of slipping down into the water beneath. To add to the dis- 
 agreeableness of our situation the darkness now became so dense 
 that we could scarcely see a couple of rods ahead ; and this was 
 what we termed luck being on our side n gainst walking over the 
 tops of the high hills before mentioned ; but we stood it man- 
 fully, and at last reached the little basin just before the bend that 
 would bring us to the shore. Here the ice was quite fresh, being 
 formed from a little brook that runs into this natural or perhaps 
 artificial hollow. It was here so slippery that we could hardly 
 stand upon it, while the broken cakes made the walking almost 
 impossible. We stumbled and fell, then tried to climb up the 
 slippery inclines only to fall quite back again. I think that that 
 little distance of scarce a mile and a half, from the head of the 
 bay to the house to which we were going, cost me more severe 
 travelling and labor to attain, than any other similar trip that 
 I ever remember of taking in my life ; but we reached it at last. 
 
 It seenis to me almost as if I could recollect each individual 
 step of the way, and sitting in a cosey chair by a large snapping 
 fire, I wondered how I had the strength to go through with it 
 after the twelve steady miles that I had already trami)ed since 
 morning, — and of such tramping ! To give an idea of how lit- 
 tle these strong, robust fellows think of such travelling, one of the 
 two men who had just come with us insisted upon returning after 
 tea with a young fellow whom we met here and who was going 
 that way. Seven o'clock at night and so dark that one could 
 hardly see a rod ahead 1 over slippery ice on which one could 
 
^11 
 
 ^i;: 
 
 .)' 
 
 ; 1 
 > 
 
 1 
 1 
 
 
 ' 
 
 :(« ' 
 
 \ 
 
 
 
 n 
 
 'i 
 
 11 
 1 
 
 i 
 
 
 168 
 
 SUNDAY AT OLD FORT — THE RETURN. 
 
 scarcely stand much less walk, through water up to the knees, 
 crawling over ridges of rocks to fall from which would have been 
 indeed dangerous, then five miles more of steady tramping be- 
 fore reaching home. One may draw his own conclusions as to 
 the endurance of these men, as this is a fair sample of what they 
 will do every day for weeks at a time, while hunting, without any 
 apparent ill effects. 
 
 One may well imagine that I did full justice to the nice, hot 
 supper that was laid out before me, and was soon after snugly 
 tucked beneath a double covering of heavy blankets to seek the 
 rest so much needed. 
 
 Sunday, the next day, was most charming. The sun came out 
 bright and warm and we all enjoyed the beautiful weather. We 
 had some birds for dinner, the first we had had for several weeks. 
 After the meal, although it was Sunday, since the men must get 
 back to their work the next day, we all started for the river again. 
 The water had become so open that we accomplished the worst 
 part of the journey in a boat, and walked the remainder in com- 
 parative ease and with very litde trouble, reaching the Mission 
 station by evening. 
 
 I will not say much regarding the Mission here, leaving that for 
 another place, as it is worthy of special consideration. I simply 
 called upon the genial people there, and enjoyed a short but very 
 pleasant interview. 
 
 I left a small stock of medicines at the Mission house, having 
 brought some with me from home, and soon started off. I saw 
 some poultry in one of the houses. They were barred up and 
 carefully fed upon scraps from the table. The hens furnished 
 a few eggs, and were apparently about the only specimens of 
 this species of fowl on this part of the coast. Strange as it may 
 be, on the lower part of the coast, some sixty miles below this 
 place, I found hens so abundant — having been brought over from 
 Newfoundland — that I purchased several dozen of eggs for the 
 reasonable sum of a shilling, or twenty cents, a dozen. 
 
 At the house of one of the inhabitants where we stopped to 
 
JOURNAL — A KOMATIK RIDE. 
 
 159 
 
 f 
 
 I 
 
 visit, we found four Indians who had come from several hundred 
 miles in the interior, with their fall catch of fur to trade for provisions. 
 As they were journeying in the same direction with us, we pro- 
 ceeded on together. We started about ten o'clock in the forenoon 
 and, after a tedious tramp over the hills, reached the bend of the 
 river, found the boat, and were soon rowing the remaining distance 
 to Bonne Esperance where we arrived in time for supper. 
 
 Tuesday the 7th. The Indians, iiaving finished their trading, 
 returned home this afternoon. 
 
 Saturday the nth. A small party went out in the boat to-day 
 and shot four ducks and a pigeon, which gave us a taste of fresh 
 meat once more. 
 
 Monday the 13th. The men spent a greater part of the day in 
 mending their fish-nets. 
 
 Tuesday the 14th. A flock of ducks appeared just off the 
 island, and we got a shot at them as they clustered, killing seven. 
 In the evening the young folks amused themselves making molasses 
 candy, while the elder people joined heartily v/ith them in the 
 disposal of it. 
 
 Friday the 1 7th. The cow was killed to-day, and we had the first 
 taste of beef we had had for months. 
 
 Friday ti\e 24th. Yesterday I again went "into the river," as it is 
 here cabled ; that is, went to the settlement up the river, and to-day, 
 the ice being regarded as safe, I had my first real ride on the Labra- 
 dor dog sledge, or komatik, with a native driver as guide. Our team 
 was a small one, only three dogs, but they drew the sled so fast 
 that it might have been twice the number and I not have known 
 it. Along the ice of the river and bays we glided, over low hills 
 and across snow patches, and over grass and moss laid bare by the 
 wind ; we went, literally, "over hill and dell," while we often passed 
 rapidly places flooded with water, where the delay of an instant 
 would have sunk the sledge. Now up some steep hill we helped 
 to pull the sled, while down on the opposite side we went 
 faster than the dogs could go at full gallop. On across the 
 country we went, until, landing upon the very verge of a high 
 
 I 
 
m I 
 
 1 ?, 
 
 r 
 
 ■ ■■ 
 
 
 ■ ' ! ■■ 
 \ 1 1 
 
 
 
 
 A 
 
 160 
 
 CHRISTMAS EVENING. 
 
 precipice, we were obliged to come to a halt. Here we tied the 
 dogs, turned the sled bottom upwards, and descended the slope 
 by a narrow, circuitous path, which brought us to the house of 
 one of the residents, where we dined. We found here a very pleasant 
 family, two members of which had spent some time "in the States" 
 in good society; after enjoying with them a very pleasant chat, 
 we returned to the sled and continued our journey. I was amused 
 to see the eagerness of the dogs to be " on the go " once more. 
 They strainod at their harnesses, and whined and barked while 
 jumping, with the evident intention of either starling the sled or 
 breaking their traces. 
 
 Our way to Cld Fort Bay lay in much the same general direction 
 as that in which we had walked a few days before ; but the hills are 
 so cut up with gorges that come nearly to their base, that nature 
 has formed foui different routes between the Mission and Old 
 Fort Bay. The scenery was quite similar, but riding we had more 
 of an opportunity to take in the arctic-like views everywhere pre- 
 sented. The day was fine, and the outline of the hills beyond 
 was very decided upon a clear blue sky. We appeared to be rush- 
 ing along through a narrow, winding valley road with receding 
 heights on either hand. On a level with us, yet on either side 
 near the base of the hills, the falling tide had caused the ice 
 to break in cones, and blocks of all sorts of forms and sizes. 
 Still on we went, the ice often bending beneath us, while the dogs 
 reminded one of childhood's fairy tales. The sight was grand, 
 while the weather was fine, cold, and with little wind. The inner bay 
 was frozen over, and we rode to the doorway of the dwelling for 
 which we had started. 
 
 Christmas evening there was a social gathering at one of the 
 houses up the river. About fifty people were present and passed 
 a very pleasant evening. In this case the old adage that "distance 
 lends enchantment to the view" was well illustrated. Most of the 
 inhabitants live at distances of five to ten miles from each other, 
 which makes calling upon one another a circumstance of some 
 importance. When once people do get out on such an occasion as 
 
wooD-curriNG in winter. 
 
 IGl 
 
 an annual Christmas dance, they stay long enough to enjoy the 
 visit thoroughly, — they did so on this occasion. 
 
 Wednesday the 29th. Pearly in the morning we started off with 
 our dog-team for the bottom of the bay. Here we tied the dogs, 
 and with our axes started up the snow-covered slope of the hills 
 to where the spruces and firs were abundant. None of the wood was 
 large. The snow was up to our knees ; it covered the evergreen 
 thickets with dense patches, so that when we started to cut a tree 
 the shaking would dislodge the snow in large patches that covered 
 us completely, while it slowly crept down our backs and up our 
 sleeves. Still on we labored until we had loaded the sledge, when 
 all started for home with appetites sharpened by the keen, sub- 
 arctic air and the healthy exercise. It was quite fortunate that we 
 procured our wood when we did, for Thursday was as warm as it 
 had been the night before, and the ice, so much melted, began to 
 break to such an extent as to render the travelling extremely 
 hazardous, though it soon froze up again. 
 
 During the long winter months the women of the house spend 
 their evenings, and for the most part their days also, in making 
 boots, shoes, and nicknacks of various kinds. Of course they 
 can do this, since their time is almost entirely at their own disposal, 
 and, after the regular work of the family is done, the remainder of 
 the day and evening falls heavily upon anyone of a naturally 
 nervous or industrious disposition. The location of the dwelling 
 forms another inducement to industry, since the long winter eve- 
 nings must be spent entirely within doors, and work of some 
 sort must be constantly provided. 
 
 It must be remembered that except in a few places, such as 
 the Mission and several small collections of houses used as winter 
 quarters and which have been built from a quarter to half a mile 
 distant from each other, the houses are, for the most part, miles 
 apart, and visiting in the evening is quite out of the question. 
 
 It is comparatively easy to point out the industrious portion 
 of the female population of the coast ; everyone knows them. 
 To substitute fictitious names for real, Mrs. Goodey will make 
 II 
 
 I 1 1^ 
 
 ' tl '■ 
 
 |i 
 
—IP 
 
 r ; 
 
 w- 
 
 162 
 
 MAKING SEALSKIN BOOTS. 
 
 fine Esquimaux sealskin boots, and will do other similar work, 
 though attention to her large family makes it impossible for her 
 to spare much of her time for outside work. Grannie Roberts is 
 also noted for her nice, careful sewing of boots, pouches, or 
 sealskin bags to wear on the back, in which to carry provisions, 
 game, or other articles that the hunters may require, while she takes 
 in such work as filling pillow cases and bed ticks, making hunting and 
 warm working jackets for winter, or mending socks, mittens, and 
 other articles for general use or wear. She is a good old lady, 
 and thought well of everywhere, while her work is always well 
 and cheaply done, and her nearly fourscore years combined with 
 her remarkable cheerfulness give her a good word from every- 
 body on the coast. Further to the westward work from Aunt 
 Jane's is known in all directions. 
 
 If I have described elsewhere a pair of native shoes, as they 
 are often called, a brief description of them here again may not 
 be out of place, since it is necessary to have a clear under- 
 standing of the process of making them. A good deal more 
 than at first appears to a purchaser depends upon every little point 
 in the operation. To begin, — suppose Grannie Roberts is to make 
 a pair of sealskin boots for some buyer. From a lot of sealskins 
 one is selected either from a harbor seal with the hair on, or a 
 large harp from which the hair has all been scraped off; in either 
 case the skin, to be the most serviceable, must be well scraped of 
 fat on the inside and dried for two or three months on some 
 frame on which it has been stretched to its fullest extent in the 
 sun, exposed on the woodpile or roof of the house (after the hair 
 has been taken off if a harp, and with the hair on if a harbor 
 seal). These dry skins will not shrink, and for every purpose of 
 wear are infinitely better than the shoes, sold in large numbers, 
 made of quickly dried skins, sewed upon wooden forms, which 
 shrink and tear, while they soon wear useless. Out of them 
 the bootleg is cut, from a paper pattern of any kind the wearer 
 may choose. All, or nearly all, bottoms are cut from like patterns 
 to fit a foot of any shape, but invariably from the dried skin of 
 
"NICE POINTS" OF THE WORK. 
 
 163 
 
 :h 
 m 
 er 
 ns 
 of 
 
 the harp seal, the dryer and older the better, since they stand more 
 wear the older they are. The pattern of the sole is an oblong 
 oval, while the tongue or top piece is more or less lance-shaped. 
 After soaking over night in water to soften it, the sole is taken 
 and the whole edge for about an inch and a half is bent inward, 
 then the toe is puckered in creases, as is also the heel, while the 
 tongue fits the space left after the bootleg is temporarily fastened on, 
 all the pieces overlapping enough to allow for sewing. These puck- 
 erings are made by simple creases of the needle at the time of 
 sewing. All seams are made — if the sewing is done in a scientific 
 manner, and not simply to " sell the boot," as the expression goes — 
 by the simple overlapping of the two pieces and sewing each edge 
 tightly to the part beneath, while the ritlge thus made by the seam, 
 if rubbed with a piece of wood shoemaker fashion, will be hard and 
 shiny as well as very tight. In all sewing the skin is so thick that 
 the needle can be run through it and out the same side without 
 perforating the skin ; thus a seam admits no water through the 
 sewing, if the thread and overlapping pieces are drawn tight. 
 
 The upper edge of the bootleg has a doubled piece of cloth sewn 
 around its edge, though sometimes sealskin replaces it, through 
 which a piece of tape or braid of any color to suit the wearer, 
 about a yard and a half long, is threaded, and the skin, being ipiite 
 flexible when on the foot, is drawn tightly about the kg, the 
 braid wound about twice and tied with the string end hanging out- 
 ward ; this secures the boot firmly and yet not painfiiUy to the foot 
 by the leg and, though the string often gets loose and the bootleg 
 often slips down, it seldom gives much trouble to the wearer. 
 
 A curious operation that might escape one's attention, as well as 
 a curious fact in connection with this oi)eration, is that the put ker- 
 ings of the heel are held together by running two, three, or four 
 small threads at about ecjual distances from each other — the stitches 
 being taken through the bend in the creases on the inside of the 
 boot, from side to side, around the heel where they are drawn 
 tight and fastened to the seam above ; another fact is that the 
 creases of the toe are not thus fastened. Why the former should be 
 
Hlf^ 
 
 i 
 1 
 
 i- 
 
 y 
 
 
 li 
 
 
 1 
 
 
 164 
 
 SELECTING SKINS. — PATCHING. 
 
 done and not the latter I cannot ascertain ; it would be a curious fact 
 to study into if one could spare the time. A proper sealskin for 
 bootlegs will cut from two to five pairs, according to the size of the 
 skin, — as the pattern for all adults is usually the same, — while a 
 proper skin for bottoms will fit six to eight pairs of boots. The 
 bottoms are not scraped, but the legs are scraped (juite clear of the 
 vellum from the inside of the skin. A skin that is dried in the house 
 has a yellowish look while one dried out of doors in the sun is white 
 as parchment on the inside. Should it happen that a person's 
 feet are in the habit of sweating much, the whole inside of the 
 boot is rubbed with a tan made of birch rind, — but I do not 
 understand the exact philosophy of it. When one first purchases 
 a pair of boots they are generally quite dry ; they are then oiled 
 carefully with the hand, with seal oil, until every part is fully 
 lubricated, the inside is then rubi °d on a stick with a polished 
 and nearly sharp edge, as is often done by shoemakers, in a certain 
 stage of their own bootmaking, to accomplish the same result, — 
 the operation is nearly similar to the manner in which pegs are 
 rubbed out of boots. After this operation they are hung in a warm 
 place near the fire until the oil has soaked into the skin when they 
 are ready to be worn, and if properly made will, with the roughest 
 wear over stones and ice, unless cut or otherwise unusually injured, 
 remain, with occasional reoilings, water-tight for at least two months. 
 At the end of about that time, the bottom or the heel is worn 
 through and the sole must be tapped, — this is done by simply cutting 
 out a piece of skin, round and the size of the whole heel if for 
 the heel, oblong and the size of the sole if for the sole, which 
 is then sewn on with a tightly drawn single thread from the 
 outside. Such a patched bottom will last a month or six weeks 
 longer and then the whole bottom gives way and is usually replaced 
 by a whole new piece from the ankle down. 
 
 All shoes are made substantially upon the same pattern ; while 
 for house wear the leg part is dispensed with and the bottom 
 extended upwards far enough to bind as a slipper or tie around the 
 ankle as a shoe. Such is Labrador and Esquimaux foot-gear as 
 
MOCCASINS. 
 
 165 
 
 worn on the whole or nearly the whole coast. Their price varies 
 from poorly made boots at $2.00 to the best at $2.50 to $3.00 ; 
 while the scarcity of seals will sometimes render them even more 
 expensive. A dried sealskin for making boot bottoms averages 
 in price at $1.50. The bootleg is often made of dogskin similarly 
 dressed to the sealskin, but the same general character prevails in 
 all, however made. In winter the men usually wear moccasins 
 on their feet. These are generally made of dressed deerskin or 
 mooseskin, and worn by those on the coast who are fortunate enough 
 to have procured the skins either by purchase or trade, by their 
 own success in hunting, or, as is generally the case, directly from 
 the Indians. The process by which the skins are tanned in the best 
 manner seems to be kept a secret. A moccasin is generally made 
 in much the same manner as a boot bottom, and in place of the 
 leg is a simple binding of colored cloth doubled to allow a piece 
 of braid to be inserted which, tied, holds the moccasin to the 
 foot, while many are simply bound without strings. In some cases 
 a wide piece is sewed on to the top that may reach aroiuid the 
 ankle, while a loop on each side holds a strip of deerskin that 
 ties around the ankle holding the moccasin fast to the leg. 
 
 Moccasins are of endless patterns and varieties : souiC are like 
 slippers and very plainly made ; others are more carefully pre- 
 pared, by ornamenting the tongue with beads, colored cotton, or 
 porcupine quills wrought into figures, flowers, and forms of many 
 varieties. Very often the shoe is tanned and rendered quite brown 
 by this process. Sometimes the toes are pointed, the side fringed, 
 and many are the devices for varying the make and ))attern so 
 that the taste of the purchaser may be gratified in his selection 
 of a pair. Deerskin moccasins are only worn on snow and in 
 snowy weather ; for being of a soft skin the slightest wet will at 
 once shrink and spoil them. The snow here often falls in large, 
 deep masses of a very dry nature, and only at this time these 
 shoes are worn, — but they are never worn alone ; the universal 
 accompaniment is a pair of leggings made of thick swanskin, 
 which is a sort of very thick, woolly, cotton-like cloth in common 
 
 I 
 
 
166 
 
 LEGGINGS. — VAMPS. 
 
 use here for all sorts of articles of warm wear. The legging answers 
 the purpose of a bootleg, and since it is only worn with the 
 moccasins in the snow serves remarkably well for that purpose. 
 Leggings, made either with a sock or without one, pull on the 
 legs over the ])antaloons ; a loop often passing from side to side, 
 under the foot as in riding pants and patent hunting oilskin or 
 rubber breeches. The moccasin is then put on over the bottom 
 I)art of the legging, and the top secured by some bright braid as 
 before mentioned in the case of the boot. A little matter of pride 
 comes into notice here, — that is, the tasty bordering of the seam of 
 the legging, on either side, since the seam is always worn outward 
 and the braid tied with colored worsteds so that the ends hang 
 jauntily outward also. The whole outfit forms a very pretty fancy 
 piece, and reflects the good taste of the wearer. 
 
 Neither shoes, lioots, nor moccasins, of a soft bottom, are ever 
 worn except with several pair of thick stockings and one or two pair 
 of swanskin vamps, as they are called ; these are simply cloth-like 
 slippers, and much resemble a stocking cut off just above the instep 
 with the edges bound or sewed over and over with worsted, and a 
 central flap an inch or two long from the middle of the front edge 
 in which is made a loop and by which the pair are looped and 
 fastened, the one to the other, when they are hung up to dry, as 
 tiiey usually are every night. One can easily see that, compelled 
 as the people are to take long walks, the foot-gear must be, as it is, 
 very warm and protecting to the feet ; and yet, though at first sight 
 those skin coverings would seem cold and productive of cold feet 
 the reverse is so extraordinarily the case that one can stand in 
 water all day and not wet the feet, unless the boot is poorly made, or 
 stand on the ice in the coldest weather and seldom suffer from the 
 cold penetrating to the soles of the feet. The main difficulty is 
 the getting used to such "strange feeling shoes," and in walking over 
 small, loose, angular pebbles which are everywhere abundant on 
 the shore, large rocks giving little or no trouble at all. In summer 
 generally the smallest amount possible is worn on the foot, 
 ytrange to say, sometimes sealskin bottoms arc put on leather- 
 
 \y 
 
SEALSKIN MITTENS. — HATS. 
 
 167 
 
 w: 
 
 \>' 
 
 topped boots. Often American and English boots and shoes, or 
 rubbers, are used, but these seldou'. in winter except on Sundays 
 or extraordinary occasions. In the house sealskin shoes, much 
 like a moccasin, are worn by the members of the family large and 
 small, almost without exception, while in warm weather the chil- 
 dren go barefooted. 
 
 The next most important articles of wear that the women make 
 are sealskin mittens. The sealskin mittens of the Ksciuimaux, 
 and those which are worn all aUuig the coast, are articles probably 
 peculiar to this, as to other ..r ic regions. The sealskin is the 
 same as that used for most articles oi wear here, and must be remem- 
 bered as the harbor, not the Russian, Northwest, Alaskan hair seal. 
 The skin is dressed as usual, anil the mittfiih made with the fur 
 outward, 'i'liey are odd looking articles outside, wliile the inside 
 is at first (]uite tough and rougli with the natiu'al wrinkles of the 
 skin, caused by heat or dryness. These, like the boots, are W(jrth 
 more if made of well rather than imperfectly cured skins. The 
 whole length of the mitten is from twelve to fourteen inches. The 
 top is quite wide while the rest of the mit narrows to the hand which 
 is made of a short, wide, rounded j)iecc of skin sewed to the main 
 part of the mitten. It is a simple affair, and yet, like the boots, 
 when worn with ever so thin an extra covering inside, it is nearly 
 impervious to the cold. As a further protection there is a border 
 at the top consisting of a strip of mu^krat, otter, or beaver skin, 
 with the fui c 'itward, and from one to twt) inches wide. Though 
 somewhat clumsy, it is a rather pretty affair to look at, especially if 
 the pcvttern of the sealskiij, as it often does, varies from silky ash to 
 almost biack with njundish spots of ash black centred ; sometimes 
 a skin is nearly pure black (the hairs, silky at the tijjs only, reflect- 
 ing a delicate velvet). The < olor of the skin and the curing setra 
 to sell the mits better than anything else, and a, prettily spotted pair 
 is almost always chosen m preference to one [)lainly colored. 
 
 We have still to describe two article jf wear before our hunting 
 apparel is comi)leted. First, the sealskin hat or cap : this is made 
 like an ordinary cap with a rather pointed crown (made of cornered 
 
168 
 
 A "NUNNY BAG." — COSSACKS. 
 
 
 liifji i"'- 
 
 
 |I|P^-' 
 
 1 
 
 J; 
 
 pieces of skin running from top to bottom), with ear flaps and 
 strings to tie beneath tiie chin. Secondly, a hunting or *' nunny 
 ba^,^^ to sHng over the shoulders and carry the provisions or any 
 articles of use or luggage : this, too, is entirely of sealskin make. 
 It is unnecessary to enter closely into the details of the work on 
 these articles as the sewing is usually of the same type in each 
 article as that on the seams of the bootlegs, while the st^^e may 
 vary according to the taste of the person for whom they are made. 
 The nunny Ijag is a spacious bag, wider than high, with an immense 
 lapel. It is carried on the back between the shoulders, and se- 
 cured by bands which slip over the arms and rest on the shoulders. 
 
 As we have gone so far in the hunting or tramping dress of the 
 Labradorians, a short account of their other princii)al articles of 
 outside wear may not here be out of i)lace. Of course the general 
 clothing in the winter months, whatever it may be, is of the warmest 
 possible kind ; the heaviest flannels are always used, and sometimes 
 several pair, while the cloth of which the coat and pantaloons are 
 made is as stout and thick as can be obtained. 'l"he men wear 
 cotton or perhaps a sort of duck overalls, and cossacks ; the former 
 need very little description being simply what T have named them, 
 and, worn over the pantaloons, are with them, tucked into the boots. 
 A cossack is a loose short jacket. It is made of swanskin, and 
 the long sleeves reach to the hand while the robe or hood for the 
 head is cornucopia-shap*,'d, and fastened to the collar behind. 
 The binding is of calico. It is secured around the waist by a 
 scarf the ends of which hang to the left ; or the belt holding 
 the hunting knife (an article always worn, concealeil or open) 
 is simply strapped around the body. 
 
 A person attired in such an outfit as I have mentioned is warm 
 and well provided for almost any sort of weather that »nay come. 
 He is ready to meet the therinometer at 30° or even 40° below 
 zero, while l)y leaving off some of the flannel tmderwear he can 
 readily adapt himself to a warm day. I-ong tramps are thought 
 nothing of; hospitality, as I have said before, is extended to him 
 from every house upon the road ; hardships seem nothing ; while 
 
VISITING THE TRAPS. — HUNTER'S CABIN. 
 
 169 
 
 he who endures the most, and speaks the least about it, soon 
 aaiuires the reputation which he has worked to earn, of a fellow 
 who will stand almost anything. The great occupation in winter 
 is visiting the traps and hunting. I might have jilaced these 
 occupations in exactly the reverse order but that more is secured 
 by trapping, hence the more important, than by hunting, which 
 usually furnishes only a temi)orary supply of game for the table, such 
 as partridges, rabbits, and in some cases deer. The partridges are 
 the ptarmigan, the spruce partridge, and rarely the sharp-tailed 
 grouse ; the two former being very abundant, while the latter is rare. 
 The various animals, e.xcept the jjorcupine, which is clubbed with 
 a stick, since its gait is so slow that it cannot run away from 
 one, though sometimes shot are more often trapped, and these 
 will be si)oken of in another i)lace. The deer-hunting forms 
 an occupation in itself. There are several cabins built near the 
 well-known deer hunting grounds in the interior of the country 
 some ten or twenty miles away, where j)arties, wishing to enjt)y this 
 sport, and secure some fi«.'sji reat for the table, make their head- 
 (juarters ; from here they take 1. ig tramps, and often shoot several 
 deer to a man in the course of . season. The cabins are very small 
 within, and some ten or twenty nen will huddle together in this 
 confined place in the evening (.the deer be unusually abundant 
 during the day) and spend their ime smoking and playing cards. 
 A dingy little cal)in scarcely uger than an ordinary room 
 sixteen by twenty feet in width, wi'i a long bin-shaped bunk or 
 berth on each side, capable of ho ling six or eiyht men each, 
 as a sleeping apartment ; a stove in the middle of the room ; a 
 room full of tobacco smoke ; and the shouts and confused voices 
 of so many men, are anything but pK tsant to one whose nerves 
 are at all delicately strung ; yet, as I i ive before said, these men 
 think nothing of iiardshijjs, being bren to them from childhood, 
 and though they often go forty miles a day looking for game, 
 over hills and down gorges of the nn st difficult walking, and 
 have started off early in the morning S( metimes before light, go- 
 ing all day with only a taste of fooil, retu ning late in the cve.iing, 
 
170 
 
 " HOSSACS. 
 
 !| 
 
 they will tfien spend the greater part of the remainder of the 
 night in revelling, with only a few hours of sleep between this 
 and another day of the same sort. Nor do I exaggerate a particle, 
 when I say that this life is continued sometimes a month at a time, 
 through all sorts of weather. Such is the life of these hardy people 
 in winter when they go out for a "good time ;" and in summer, 
 during the fishing season, it is fully as bad ; but the people are, 
 for the most part, pleasant and friendly, and a stranger has no 
 cause to complain for want of being favored. 
 
 The fancy work in which the ladies indulge to a greater or less 
 extent, especially in the winter evenings, may be known as pouches, 
 pockets, " /wsscics," watch cases, and sometimes cushions and a 
 variety of beadwork of all sorts and patterns, — the latter being 
 mostly worked by Indian s(iuaws, who, with the men, often live in 
 tents near the coast in the spring and summer time when they sell 
 all sorts of articles of their own manuAicture and handiwork. While 
 watches arc rare on the coast, the various patterns for watch cases, 
 of beadwork or cloth or on sealskin, are various and very pretty. 
 As I am writing I have one now hung on the wall at my right. It 
 is about six inches long, and three broad at the widest part ; the 
 shape is rather sharp ovoid. The main part is of harbor sealskin 
 plainly ir ^ very neatly worked with colored beads on the upper 
 part, while the outside of the pocket has three little knots of the 
 same on either side of the round opening for the face of the watch, 
 which opening, with the upper edge of the pocket, and the whole 
 outer edge of the case has an edging of purple ribbon, on which 
 a string of white beads is sewed in smnll scallops entirely around 
 it; the inside pocket lining, as shown through the opening 
 in front, is of red silk, and the l)ack of the whole affair is of 
 brown cotton cloth. It i^ really cry ornamental, and 'luite useful ; 
 various patterns are made btMiles this ; but, as a rule, the plainer 
 and simpler the more auractive they are. Hossacks are long strips 
 of cloth on which a series of three, four, or even more pockets, of 
 a variety of depth and width to suit the maker, are sewed, the 
 lop of each j^Kket just reaching to the bottom of the other, and the 
 
 'P^'l'^ 
 
DUCKS' HEADS AS ORNAMENTS. 
 
 171 
 
 whole making a strip some twelve or fourteen inches long by four 
 or six wide. When the body of the piece and the lining of the 
 pockets are of good cloth, the trimmings an edging of colored 
 ribbons with bows at each corner of the top of each pocket and 
 at the bottom of the last one, and the face of each pocket filled 
 with sealskin, beadwork on sealskin, or colored cloth groui. Iwork, 
 or, ns is more often the case, of the heads of ducks and other 
 birds, the effect is most striking, and the ornament one which would 
 grace the boudoir of the wealthy as well as the library of the student. 
 The head of a merganser or sheldrake, with its purple reflections, 
 and long, linear, glossy, and most delicately cirved feathers ; that 
 of the loon with its metallic gloss; of the kmg eider, or kingbird 
 as it is called, with its delicate green on a white backgrouml, — 
 are the most frecjuently used pieces ; while the teal, widgeon, and 
 various other birds are sometimes substituted and, placed in the most 
 conspicuous places, add greatly to the attractiveness of the article. 
 Pouches and pockets, especially for carrying tobacco, are usually 
 made and sold in large numbers. They are manutacturetl of itiarbor 
 sealskin — lined or not — bordered, fringed, worked with beads some- 
 times over nearly their entire surface, or plain according as the 
 maker may see fit to work them. The rounded lapel is termin- 
 ated by a long, thin strip of sealskin, to which is attached a piece 
 of wood, bone, or tlie tooth of some animal, which ties and fastens 
 t'le article when rolled \\\). If it is made to roll up tightly it is 
 c \Ueil a pouch, while if it retain its shape it is called a pocket. 'I'he 
 market for all these articles is chiefly near home and their making 
 a matter of jjleasure rather than work. 
 
 ii 
 
 if; 
 
 
 
^H! 
 
 r 
 
 ■ 
 
 li) vt 
 
 ;,' I;.;* 
 
 U|('.- j 
 
 :i||vi 1 
 
 in;u|-> < 
 
 !||v 
 
 
 il' ' 
 
 <; 
 
 ^'t ■ 
 
 172 
 
 NEW YEAR'S DAY. — RACKET WALKING. 
 
 CHAFFER XL 
 
 New Year's Day — How to walk on Rackets — "Fish, dogs and seal," the 
 general topics of conversation — Obtaining skeletons — Larch poultices — 
 "Small Talk — Low temperature — Deer stories — Trapping — Indians 
 — Up the river — At the Mission — Harnessing the puppies — A racket 
 walk. 
 
 Saturday, January i, 1881. New Year's clay passed much like 
 any other day here, and differed little from either the Christmas or 
 Thanksgiving that had preceded it. The "old wife's" saying, that 
 the twelve days after Christmas determine the weather for the twelve 
 months in the year, is here strongly and strangely believeil, espec- 
 ially by the elderly people. I am of the opinion that, generally, 
 the people are rather inclined to superstitions. Sayings like the 
 above, together with such as the month *' comes in like a lion 
 and goes out like a lamb," or the reverse, are frecpiently quoted ; and 
 though half in jest I fiincy them to be more than half believed. 
 One will find a very fliir practical example of the use or belief in 
 the value of Herschel's weather tables by the moon, here in the old 
 wife's almanac. Such elderly ladies almost invariably inquire when 
 the moon ajjpeared before telling what weather may be exi)ecte(i. 
 Strange as it may be, they seldom assert that the weather will be 
 " so and so," but say** I have always noticetl ;" or " the last time" 
 it was so and so, so and so happened — thus they predict the fu- 
 ture from the past. 
 
 I made my first trial to-day of walking on rackets. Racket- 
 walking is a feat very difficult for a stranger to acquire readily. 
 The motion is different from anything I know of; the peculiar swing 
 of the btuly nuich like that of a sailor walking the slippery or un- 
 
 ! I 
 
" FISH, DOGS, AND SEAL." 
 
 173 
 
 
 balanced deck of a vessel. The impression upon the snow is about 
 
 ) as shown in the figure. Place a number of these figures in a 
 ( line, one before the other, and you have the impression ; im- 
 
 ) agine the foot as it swings the circle of the loop and you have 
 ( the step. I walked with one of the neighbors into the river, 
 
 ) as it is called, that is, to the Mission settlement, seven miles 
 away ; my feet were sore for days afterwards, but I soon became 
 expert in the use of the rackets, and before many weeks was 
 using them as freely as if I had lived in Labrador all my life. 
 
 Wednesday evening a party of neighbors called from about fif- 
 teen miles to the westward, and amused us with the news, of which 
 there was very little, and the gossip of which there was a great deal. 
 They were very pleasant, and concerning their hobbies — fish, dogs, 
 and seals — quite intelligent. Hunting is generally the chief topic of 
 conversation in the winter, and though game appears scarce this 
 year, I have the word of so many that I cannot doubt the asser- 
 tion that last year ('79-'8o), two young fellows shot and snared 
 eight hvuidred rabbits and about three hundred ptarmigans ; they 
 sold part of them farther to the north for a shilling (twenty cents) 
 a pair, while those remaining not eaten were salted down for future 
 use. 
 
 Tuesday the iitli. We were nearly turned out of the house 
 for the day and the rest of the week, by the owner using the only 
 available roojii for a carpenter's shop, while he made him a new 
 komatik ; for some trifling reason, however, the sled was mot a suc- 
 cess, and Wife used but very little, the old one being preferred in 
 its place- 
 
 I was \-ery- fortunate to-day in getting a large nunilKT 
 ot oaanraakia, tnym. a ^^entieina? now ui 
 unuseil hahitatituTT' of the Li .soaa's } 
 A -^ ->;!tne. — I iTcan Tinuseti this year 
 \\. ."cral stt-wtonv ')f the be;i^ - *V 
 c<L_e-.-, v-'r Caunada lynx (Zi//; _ ^_ 
 each ot' the common red. tiio itj-w 
 
 with one t.b. tlie white or arct:L fujk. (^^M^ex 
 

 
 w 
 
 !.!?: 
 
 V 
 
 174 
 
 PURCHASING SKELETONS. 
 
 several of the weasel and mink, one martin (A/us/c/a Americana), 
 and a variety of the common red squirrel {Sciurus Hudsoniiis) . 
 It often amuses me to hear the reports, so frecjuently current, of 
 the fabulous " wealth of the American gentleman who has lately 
 come on the coast, buying everything that is of no use to anybody." 
 
 One fellow brought me a common porcupine and said that he 
 heard that I was giving five dollars for i)orcupines, but wanted to 
 know if I could not give him six ! I gave him a dollar, and 
 considered that he had made a pretty good bargain at that. He 
 was contented, since he had been induced to ask the price by his 
 father and others who undoubtedly thought that they could easily 
 " make a good thing" out of the American who might not know 
 the difference. With reference to one other animal whose skele- 
 ton I procured, I would say that the wolverine, or glutton ( Gulo 
 iuscus), — called " Carcajou " by the Indians — is becoming either 
 more and more rare about this locality, in fact all along the coast, 
 or has retired far inland. It is seldom caught ; its fur rarely 
 appears in collections — that is in comparison to what it did ten 
 years ago even — and is so poor, and the price so low, that it 
 would hardly pay one to hunt it. The Indians have ceased 
 to capture it, and out of a thousand dollar batch of fur you will 
 not find r,c';e than ten dollars wortii of its skins. The animal has 
 undoubtedly gone farther into the interior, where it hunts the 
 deer by pouncing upon it from some concealed spot in the 
 trees above, when the animal passes, and tearing its throat sucks 
 its blood. 
 
 Monday the 1 7th. While up the bay this morning there came on 
 a tremendous snowstorm. The flakes were larger I think than any 
 that I ever saw before or have seen since ; it often appeared as if I 
 saw them fiiUy half as large as my hand. 
 
 For the last three or four days one of the men has been suffering 
 intensely with a felon on his thumb ; all manner of poultices and 
 a])pli( ations have been tried to cure it, but none have been so 
 effectual as the soft, outer rind of the common larch {Larix Amer- 
 icana), boiled in hot water and then kneaded into a poultice. It 
 
"SMALL TALK." 
 
 175 
 
 jring 
 and 
 
 healed the sore with wonderful rapidity. It reminds one how rare 
 serious cases are in this region, where physicians are so " few and 
 far between." The i)eoi)le are seldom sick, and when they are 
 nature is its own nurse assisted by the exjjerience of the " old wo- 
 men " of the coast. 
 
 Saturday all hands amused themselves making molasses candy, 
 a great treat in this region ; it tasted very good and homelike I 
 assure you. A little luxury of this kind is highly ai)preciated in 
 such a region, and especially since so little sweetening of any kind 
 can be ])rocured. This year provisions have been unusually high 
 everywhere along the coast. Flour, which usually sells for from seven 
 to eight dollars a barrel, has already reached twelve dollars, and is 
 still going up ; it is hard to be obtained even at this price. Many 
 of the poorer families are feeding on corn meal, which is generally 
 refused here until the last extremity, since the majority of the peo- 
 ple regard it as hardly fit to feed the dogs with. I sui)i)ose, in this 
 case, the climate has a great deal to do with the prejudice. C'orn 
 meal is evidently and eminently a southern dish, and suitable to 
 the warm climates. Besides what I have mentioned, the saltfish 
 has already given out in the majority of the families along the 
 coast, and the preserved salmon, trout, herring, and mackerel, 
 are being used up fast. Last year the game was so abundant that 
 not half of it could be eaten ; this year there are no indications 
 of any, and the Indians, from two hundred miles in the interior, 
 are telling the same story. Some of the old inhabitants say that 
 they never knew such a dearth of game during thirty or forty years' 
 residence on the coast. 
 
 Several cases have come to my knowledge where whole families 
 have been obliged to separate and hire around in turn to their 
 more i'ortunate neighbors for the winter for their board alone, to 
 escape the poverty and destitution which would surely have over- 
 taken some of them had they continued living together for any 
 length of time during this excessively cold weather. Adil to this 
 the wrangling of families and often membtrs of the same family 
 (happening in several cases, over the cargo of the wrecked lOdward 
 
I 
 
 I- 
 
 I 
 
 1 f 
 
 ^r 
 
 Wi: 
 
 176 
 
 DOGS DRAWING HEAVY LOADS. 
 
 Carclwc'U), and it is not strange that this should he regarded as an 
 unusual year on the coast of Labrador. Still another drawback 
 exists, and one that would work in the exact opposite way in any 
 other country : the weather has been mild instead of cold, — the 
 bays hardly frozen over. Now if the bays do not freeze fast 
 sledding between places becomes almost impossible, since the dogs 
 must draw the sleds over steej) and high hills, and can carry only 
 a small load ; while if the weather be cold, the bays freeze cjuickly 
 and sleds draw immense loads with the greatest of ease. 
 
 It is an old saying that "you cannot load a well-made komatik." 
 The sled balances so nicely and draws so evenly on the ice that you 
 may i)ile on all that you can, yet once start the load and the dogs 
 will draw it fast enough. This does not necessarily prove that the 
 dogs are not powerful creatures, or that they do not exert great 
 power in drawing the loads imposed upon them. I have seen three 
 stout dogs draw over a thousand jjounds of old iron (upon one of 
 these sledges) over seven miles in about an hour and twenty-five 
 minutes ; while again I have seen these same dogs draw a load 
 of twelve twelve-foot deal board plank, weighing not one-half as 
 much, and when the sled became stuck in the slush, all their 
 force could not move it. Though this would seem to prove that 
 the sled is easily drawn after being once started, such is not always 
 the case. It must be remembered that dogs must exert a tremen- 
 dous effort to draw even the small weight of an emi)ty komatik over 
 the ice that is smooth as glass and presents no unevenness ujwn 
 which they may rest the foot to keep ii from slipping. I have 
 seen several of these creatures harnessed in such a position where it 
 was impossible to assist in the carrying, and even very little in the 
 starting of the load ; for instance, there were five large, smart ilogs 
 harnessed to a huge oak log, eighteen inches square and fifty feet 
 long, embedded in the snow antl ice. A few pries with the i)ick 
 loosened but did not start the log. The dogs strained, often 
 running backward, then rushing forward with a great jump that 
 brought them to the end of the line, with a thug like the sound 
 of a drum, while at other times they strained every nerve, before 
 the log started ; but, once started, was drawn without difficulty. 
 
LOW TEMPERATURE — DEER STORIES. 
 
 177 
 
 uiid 
 fore 
 
 The lowest temperature that I observed occurred this week when, 
 about seven o'clock one morning, the glass was at 27° below zero ; 
 it must at the very least have been 30° below during the night. 
 This is generally the average winter temperature in this region, on 
 cold winter nights and mornings, while it often goes down to 40° 
 below and sometimes freezes the mercury. This winter the majority 
 of readings for the thermometer have been hardly to zero, or barely 
 below it — consequently the bays have not frozen over, travelling has 
 been most difficult, and the people have suffered accordingly. 
 
 In the poorer communities I have been struck with the greed with 
 which everything is saved or hoarded up against a time when it 
 may be needed. Nails are often of as much value as if they were 
 made of nickel ; pieces of wood of peculiar shape are saved ; the 
 hoop, stave, or head of a barrel, if strong, is carefully laid away ; 
 while each little thing is saved or hoarded with the care bestowed 
 upon some sacred Penates, that is kept to be looked upon rather 
 than for use. Everything has its value while there is little that is 
 apparently useless. 
 
 I suppose that to-day, Thursday, Feb. 10, one of the most 
 exciting stories (since verified) came to us that we have heard 
 for a long time. I have reference to a party of deer hunters who 
 have just returned from the interior with report of a large herd of 
 deer having appeared there, about a day's journey eastward from 
 where we now live. Their success has been unusual. Large herds 
 have been seen, and a great many deer killed by different parties 
 along the coast. One man living in the little settlement up the 
 river had the rare fortune to kill four in one day. It appears that 
 the hunter alluded to started out alone from the cabin, one morning 
 before breakfast, and, coming up with a herd, was fortunate enough 
 to kill two deer with the same ball — both fidling together. He re- 
 turned to the cabin to breakfast, and said that he should want some 
 one to go with him with a sledge and dogs to procure the deer and 
 bring them to the cabin for him. While waiting for the men to har- 
 ness the dogs, and get in readiness, he said to them that he would 
 just take his rifle and have a turn over the hill beyond to see if 
 19 
 
 I! 
 
IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 V 
 
 A 
 
 {/ 
 
 :/ 
 
 ,<" Cx 
 
 >■' WJ- 
 
 <f^4 
 
 4 
 
 1.0 
 
 I.I 
 
 m 1^ 
 
 - ilM 
 
 1^ 
 
 122 
 
 t m liiio 
 
 1.8 
 
 
 1.25 1.4 
 
 1.6 
 
 
 -• 6" — 
 
 
 ► 
 
 p% 
 
 <? 
 
 Si'" 
 
 >7 
 
 
 c"? 
 
 o 
 
 7 
 
 /A 
 
 Photographic 
 
 Sciences 
 Corporation 
 
 23 WEST MAIN STREET 
 
 WEBSTEK N Y. 14580 
 
 (716) 8/i-4.S03 
 
 €S 
 
 ^^' 
 
 i-V 
 
 # 
 
 '^ 
 
 O 
 
 XS" 
 
 
 ^,.^>^ 
 
 % 
 
 v^ 
 
 -<j,^ 
 
 <n? 
 
^j^ ,„„ 
 
 L<? 
 
 i/.A 
 
 > 
 
178 
 
 HOW INDIANS HUNT. 
 
 i'i' 
 
 II f 
 
 there was anything in the shape of game to be seen, hardly ex- 
 pecting to find any so near the cabin. He returned in a short 
 time saying that he had shot two more deer, as before, both with 
 the same ball, taking two of them as they happened to be running 
 abreast. Of course, those in the cabin were wild with excitement. 
 During the hunting season several parties shot deer and some of 
 them more than one, but no one approached the above record. 
 
 We now had fresh meat, and made good use of it, I assure you. 
 It was quite palatable after a uniform diet of salt fish and pork for 
 four months previous. Later on we obtained deer's meat of the 
 Indians, smoked by them, and prepared as our dried beef; 
 they cut it up into small pieces, and eat it from their tin dish 
 much as we would bread and milk, without the milk, however. 
 This dried deer's meat is, in fact, the Indians' bread for a large part 
 of the year, while they are out on their hunting expeditions. The 
 Indians conduct their hunts in a much more systematic manner 
 than would at first appear. They start from the coast and travel 
 about a hundred miles directly inland before pitching their tents, 
 carrying everything with them, of course, by the water of the 
 numerous inland ponds and lakes, or on their back over portages, — 
 a portage being generally spoken of as a narrow passage over some 
 embankment which separates the waters of one lake or pond from 
 those of some other farther on — when they camp, and the men 
 immediately start out on the look, first, before any hunting or trap- 
 ping of other game is attended to, for deer. They travel by 
 concentric circles gradually widening outward, and return in much 
 the same way home. Thus it is impossible for them to miss the 
 tracks, on the snow, of these animals, should they cross them in any 
 direction. The tracks first found, the impressions are followed with 
 accuracy and swiftness, day and night, until the herd is found, when, 
 I believe, generally, part remain to hunt while the others return and 
 fetch the tent and utensils to within a few miles of this new locality. 
 As long as there are deer within thirty miles of the tent the 
 Indians remain in a given place and proceed with their hunting. 
 Several deer are killed, their flesh hung in strips over the fire and 
 
 iii 
 
 'I H 
 
INDIANS IN WINTER. 
 
 179 
 
 smoked — to be afterwards laid away for extreme use only ; the bones 
 are either made into a soup — as they are full of oil and marrow, 
 or roasted, then cracked and eaten ; while the hide is soaked in 
 brine and the hair taken off by a scraper, much like our choj)ping 
 knife — if indeed there be any difference — but by them called *^ood- 
 loo^^ or, as pronounced, " hoodloo ;" it is then washed, cleaned, 
 dried, and white-tanned and preserved (dressed) with the brain of 
 the animal, in which the skin is rubbed and kneaded with both 
 hands and feet until rendered quite soft and flexible. 
 
 When the supply of food has been assured the hunters next turn 
 their attention to the traps. They set deadfalls, principally, for all 
 the variety of animals captured, and are usually more successful 
 in capturing martins, than any other game, tl e number of these 
 animals annually captured being simply enormous. The next most a- 
 bundant species is the beaver ; then come the otter and Canada lynx j 
 perhaps the red fox should come in after the martin, at least after 
 the beaver, in point of numbers at any rate. In this manner the 
 encampment, consisting perhaps of four men, three women, and five 
 or six children, will continue to work their way five or six hun- 
 dred miles inland in winter, eventually coming out in the spring by 
 the same road — though not always — that they went in on the pre- 
 ceding fall, with a load of valuable furs, and in a half famished 
 condition. They then sell their furs, pay part of their debts — for 
 every one gets large credit from the traders who often charge enough 
 to make up for it — and live in plenty for a time following. It looks 
 strange to see them, as I have, furnish their wigwam, or mishwap ; 
 they borrow a stove, let the chimney run out at the top of the tent, 
 and try to semi-civilize themselves, while they buy large quantities 
 of everything that is eatable from the particular trader of whom 
 they procure their goods — generally each family getting their pur- 
 chases of a different party — and do little for the next month but 
 eat and sleep. They are, generally speaking, a peaceable set, and 
 only savage when their apparent — to them at least — rights are in- 
 fringed on. The whole number on the coast is about 1400 ; they do 
 
 ■-■■ ■*! 
 
 
H 
 
 )';■ 
 
 180 
 
 A TRIP UP THE RIVER. 
 
 not exceed a well ordered regiment, and doubtless one-half of these 
 only could be relied on to fight, even should any real difficulty 
 arise, which is not likely to happen at any time in a country at once 
 so cold and so remote from human habitation generally. They 
 have enough sense to see that it is for their own interest to keep on 
 good terms with their white neighbors, — for who would purchase 
 their furs if they did not do so' Strangely enough, an Indian 
 will purchase anything that, setting his eyes upon, he desires, pro- 
 vided it come within the reach of his means, or that his credit allows ; 
 it may be a shotgun worth a hundred dollars, a boat worth twice 
 that amount, or a solid and expensive gold watch, — and cases of the 
 latter kind have occurred frequently. I have in mind one where 
 an Indian ordered and obtained a valuable gold watch and chain 
 worth about $150.00 ; he used them for awhile, and then exchanged 
 it about a month after for a $3.00 silver one that, at the time, equally 
 pleased his fancy. Nor is this either an extravagant or an ex- 
 ceptional instance on record of like stupidity. 
 
 In the afternoon of the same day, Thursday the loth, I took a 
 guide and went several miles farther up the river. It was a most 
 pleasant and delightful walk. We followed a winding path, or 
 portage, over a series of hills and vales ; the deep snow on which 
 we walked nearly covering a small growth of firs and spruces, pro- 
 bably six to eight feet high — the path lying over these. On the 
 distant left rocky knobs and crests, with rounded tops, were 
 everywhere apparent, while, on the right, high cliffs bordered 
 the river. 
 
 We were now in the valley of the narrow river. High hills were 
 on either side ; beyond and ahead, uprearing tier upon tier, dark 
 blue and long, uneven ridges of crests had been upheaved by nature 
 towards the light blue sky above. The river, bending to the left, 
 seemed to lose itself in the distance among the bases of one or two 
 bold cliffs which appeared to bar its passage. It was a grand scene : 
 the passive river, bound in ice and crowned with snow. At this 
 point the stream is nearly a mile and a half wide. About the same 
 
TRAINING THE DOGS. 
 
 181 
 
 , 
 
 distance can be seen ahead before the cliffs shut the view in that 
 direction. The scene is one of a vast inland lake, enclosed on all 
 sides by cliffs and hills. 
 
 Returning home I had a full view of the ridge across which the 
 portage here runs. Like many, which I saw in various other 
 places about the region, it is without doubt of glacial origin, and 
 was perhaps formerly of equal height with the surrounding elevations 
 on either side, but was worn down by the abrasion of some local 
 arms of the glacier entering the river — perhaps the course of the 
 former main glacial stream — at this point. 
 
 In the evening I visited the Mission, and found a young folks 
 party in progress. It pleased me to see the effectual way that 
 the teachers had taken to reach the juvenile hearts, and promote, 
 for a couple of weeks to come, scholarly attainments in the reci- 
 tation room. 
 
 Sunday the j:3th. It snowed hard all day. Monday was no 
 better. It was St. Valentine's day, which is here rigidly kept like 
 the majority of other Saints' days and holidays, as a " fete day ;" 
 no one did any work, and while the women " slicked up," the men 
 got "set up " — if they could find anything to get set up with — and 
 spent the most of their time in smoking and sleeping. It snowed 
 steadily until Wednesday morning, when it stopped, for a short time 
 only, as if to ascertain the results, to begin ag.iin with renewed force 
 and violence. Sunday was about one of the most uncomfortable 
 days I experienced during the winter. The wind was cold and 
 piercing, strong and penetrating. The ground was covered deep 
 with snow, and the mist in the air — if I remember right — heavy 
 and wet. Monday the little puppies, now nearly two-thirds grown, 
 were broken into the komatik ; it was a most curious sight. The 
 process, though it appears brutal, is the only one that seems effect- 
 ual. The young dogs are taken and tied tightly into a harness, 
 an extra thong is passed into the mouth and then tied above and 
 around the whole muzzle as tightly as it can be drawn ; this is 
 done principally to prevent the animal from biting the thongs and 
 freeing itself. The other end of the harness is then fastened to 
 
182 
 
 WASHINGTON'S BIRTHDAY. 
 
 that of one of the large and fierce dogs already secured to the 
 sled. Of course, then, as the large dogs go, the smaller ones are 
 obliged to follow. If they trip and fall, as they frequently do, the 
 big dog drags them along until they are jounced on to their feet 
 again. Thus on they go, rolling over and over, bounding from side 
 to side, and all the time uttering most dismal and horrible groans 
 and cries. Gradually they become actually hardened and seem 
 to enjoy being thus pulled about, and soon they are loosened and 
 harnessed, several of them separately, under the lead of some old 
 head dog. After this the best and most obedient of them is fastened 
 to the head dog, in advance of the others, and taught to be urged 
 forward, or to be turned to the left or right at the shout of the 
 proper turning words from the master ; they are then harnessed 
 all alone with this trained head puppy as leader, and if they go all 
 right, exercise and the whip alone are necessary to keep them in 
 training and practice. 
 
 Tuesday, February 22. Washington's birthday! How happy 
 I am that the dear old gentleman was not born here in this remote 
 frozen region ; he verily would have died of the blues without 
 having handed down to America an event to celebrate. I fired 
 several rounds in honor of the occasion, — much to the disgust of 
 the people with whom I was staying, who, as Englishmen, of course 
 cared little about celebrating any save their own fete-days — and 
 concluded the day with a walk, upon those curious articles the 
 rackets, which I will try to describe. 
 
 In describing a walk in rackets I hardly know how, at first, to 
 begin. I might follow on foot any one of the many excursionists 
 who are so constantly and continually going and coming to and 
 from the interior, or the dwellings along the coast, but that, being un- 
 accustomed to long tramps, I could not give a true account of such 
 experience ; or I might give fanciful accounts of dangers and 
 occurrences that possibly might have happened to me or anyone 
 else unacquainted with the ground to be gone over had such a 
 journey been attempted. I do not like to mix the fanciful and the 
 real unless circumstances so happen that there seems to be no 
 
\f 
 
 ENDURANCE OF NATIVES. 
 
 183 
 
 escape, and yet I fear that the faint description of a really short, 
 but to me long and tiresome walk, or rather tramp of this kind 
 which shall be told, will be far from giving you a true idea of what 
 a Labradorian day's march, over hill and bay, may be. Though the 
 hills are high, their caps snowy, and their sides slippery ; though 
 strangers would soon be tired out by excessive fatigue in climbing 
 and slipping over them ; the people here think no more of it than 
 if the tramp were one over a level plain that we would walk 
 with ease and comfort. The whole coast for many miles inland is 
 one vast extent of hills, from three to five hundred feet above 
 the sea level ; and yet the men go over them from place to place, 
 visiting house after house, stopping simply for a few moments' rest 
 or chat with some neighbor, or to eat a frugal meal at another's 
 hospitality, — for hospitality is offered here even among those who are 
 enemies in every other way, and it is counted a sin to refuse such 
 when houses are often a day's march apart — while a day's march of 
 twenty, thirty, and in rare instances forty miles, of such travelling 
 is far from being unusual. One may wonder how such journeys 
 are made, but it is impossible to account for them other than by the 
 natural hardihood of the people here, who live and thrive on the 
 coarsest of food. Not long since a man walked a distance of 
 about seventy miles in two days ; quite recently, parties have gone 
 deer hunting, twenty miles into the interior, stalking deer all 
 day, returning to thv.ir simple cabin at night, having travelled forty 
 miles. Nor are such events uncommon; the people here think 
 nothing of it, and, in fact, tell of the many miles that they have 
 walked with considerable pleasure. It is of course pleasant to 
 ramble over these snow-capped hills ; to see the beauties of nature, 
 expressed in the indescribable language of nature, from elevations 
 and depressions, from ponds and gorges, and snow alternating with 
 rock or discovered lichens, or chilled vegetation ; but a stroll 
 over a bay of frozen ice, with hills about and around one, and 
 islands here and there, while you walk comfortably along and fill 
 yourself with the scene, is, perhaps, still more pleasant. 
 
 My walk from Old Fort Bay to Bonne Esperance was undertaken 
 
 It 
 
1 
 
 ~ 
 
 
 184 
 
 ENGLISH VERSUS AMERICAN. 
 
 m 
 
 I! '1^' 
 
 IIM 
 
 in honor of the day. Here I was, in a lonely out-of-the-way 
 region of the globe, midway between the temperate and the arc- 
 tic zones, and snowed and iced up, so to speak, from the rest 
 of the world ; to this, add the fact that I was upon English soil, 
 and you understand my feelings. I shall never forget in this 
 connection (though I mean no offence) a passage in one of Jules 
 Verne's works, in which he recounts a dispute upon the naming of 
 a point hitherto unreached in the unknown lands in the polar regions. 
 The parties were an Englishman and an American ; both had come 
 together unexpectedly in search, the one of the northwest passage, 
 the other, of the north pole ; the name of Cape Washington 
 was given by the American. The Englishman angrily asserted, 
 " you might choose a name less offensive to an Englishman ;" to 
 which the other replied, proudly, ** but not one which sounds so 
 sweet to an American." I am afraid that my English friends 
 shared a touch of this sentiment, at least the former part of it, as 
 I maintained the national honor of the occasion and fired a salute of 
 four guns to the memory of the day ; for, on coming to breakfast, 
 I was greeted with a sharp, "what's that for?" and an explanation 
 only elicited a shrug of the shoulders accompanied by an humph/ 
 as if of no consequence to them and entirely unnecessary on my 
 part. 
 
 After breakfast I prepared to carry out my scheme. It was 
 a fine morning, the air was cold and clear, scarcely any wind, 
 the roads in tolerably good condition as I supposed, and all 
 nature invited me forth. In return for the reception my patriot- 
 ism of the morning had received, I determined to pay my friends 
 by not telling them where I was going — having but one object 
 in view all the time ; many were their solicitations, but I was 
 firm. The dogs seemed unusually happy as they frisked about and 
 around me, and it seemed good to get out in the air; so 
 putting on my rackets, with my gun in hand, I started off. At first 
 I had to walk on the level bay about a mile and a half to the head- 
 land on the left. The walking would have been called good by a 
 native Labradorian ; to me it was terrible, and I stumbled along 
 
 (m; 
 
* 
 
 RACKET WALK ACROSS THE BAY, 
 
 185 
 
 over ridges of drifted snow and patches of smooth ice, on which 
 as with the schoolboy it seemed as if for ever)- step forAvard I 
 took two backward, and that it would have been much better to 
 have turned, as he did, and walked backwards ; but at last the diffi- 
 culty was overcome and the snow reached again when the travelling 
 was smooth for some distance. 
 
 The wind had as yet hardly sprung up, yet the clear bul rather 
 cold air was unusually invigorating. The hills on either side 
 looked quite fresh and clear in outline against the sky, and 
 the day gave promise of being unusually fine. Ever)'where, snow 1 
 snow ! snow ! It seemed as if the elements combined to make bad 
 travelling for the sledges, for it is hard work for the dogs to haul 
 the komatiks through the snow on the bay, especially when the salt 
 water penetrates the ice, or rather the partially thawed top of the 
 salt water ice, which, it must be remembered, requires a much lower 
 temperature to freeze than fresh water, alternate with the deep 
 ridges of soft snow ; but with rackets the case is different and these 
 snow heaps were just the kind of material for good walking. I soon 
 struck for the highland on the left and was just approaching a lit- 
 tle opening between the mainland and an island — such openings 
 are called here by the name of "tickle" — when a gust of wind 
 from the northeast, and in the exact line through which I was to pass, 
 struck me fiercely. The snow was piled in deep, uneven ridges ; 
 cakes of ice had been thrown upon each other by a recent partial 
 thaw blocking the way before me ; and the wind, added to these 
 hindrances, almost forced me to abandon my design. The gust 
 swept fiercely through the narrow opening, and I was glad when 
 safe on the other side and once more set out across the long 
 stretch of bay between me and the island half-way to my stopping 
 place. 
 
 The hills on the coast receded on the left in an outline of un- 
 dulating crests far eastward ; on the far west was the ridge of high 
 hills that formed the entrance of Old Fort Bay ; beyond lay in- 
 numerable islands, often so close together that they seemed to form 
 
 m 
 
 ! I 
 
 I 
 
 1 i ' 
 
^li 
 
 4' 
 
 ■'I 
 
 |!l; 
 
 A' ■ 
 
 186 
 
 AN ICE SCENE. 
 
 a band of mainland in the distance of several miles that separated 
 them from me. On ! on ! The scene was one of strange lieauty, 
 while words are utterly unable to portray its grandeur. 
 
 The ice and snow of a winter on the hills and plains of north- 
 ern latitudes cannot be described to a dweller in city, country, or 
 town. The few houses of the region, miles apart, hidden by sur- 
 rounding cliffs ; undulating hill tops and deep gorges ; isolated 
 knobs, now high, now low ; near ridges looking far, and far ones 
 looking near, the effects of refraction in a clear, northern air, — 
 all unite with an irregular plain of level bay in being covered 
 everywhere with snow, snow, snow ! A dull glare of ice, and 
 occasional bare places or peaks on the rocky masses of hill — for 
 I can call the whole coast by no other name — relieve occasionally 
 a monotony of snow ; while the clouds above add white masses 
 of stratus to the scene. Tell me, now, how can pen describe what 
 sight and sense almost fail to appreciate ? Arctic travellers (not 
 that I dare assume a place among them) tell of the sights that 
 everywhere greet them and which are indescribable ; yet few believe 
 them until they likewise catch a glimpse of a similar display of 
 nature's arctic grandeur. In the narrative of explorers, we read of 
 one by one who return to give evidence to their truth and attempt 
 to describe similar experiences. 
 
 The process of walking on rackets is one easier to describe than 
 to attain to perfection. The large size of the rackets occasions an 
 unusual difficulty in walking. The feet must be kept quite wide 
 apart, and usually the proper step requires the leg to be swung in 
 a semicircular direction around the racket of the other foot. ThI 
 effect to the eye is very peculiar and ludicrous to one observing 
 it for the first time ; and, in fact, to me it always appears ludicrous 
 to see a man thus easily, though apparently awkwardly and with 
 difficulty, laboring his way along, claperty-clap, as the big pads 
 fall in their proper step places on the snow, and the person wear- 
 ing them advances rapidly along. 
 
 By this time the novelty of such a scene as that I was gazing 
 
DIFFICULT WALKING. 
 
 187 
 
 upon had liegun to wear away and given place to uneasiness regard- 
 ing the distance ahead rather tlian of that already travelled. 'I'he 
 walking, now that the open bay was reached, began to be anything 
 but good ; for while the lighter and more crispy — for I can think 
 of no other word to express the condition — the snow, the l)etter 
 the walking, the case is quite the contrary when a partial thaw sets 
 in and renders the walking watery and sloshy. At such a time 
 every hole in the sieve-like interlacing of deerskin, which fills the 
 interior of the racket, is filled with the sticky snow, and the bars 
 covered also ; soon, by frequent pressure, the upper side is cov- 
 ered with the same, and so with the weight above, and ihe sticking 
 of the snow beneath, the foot must fairly drag such a mass of cloggy 
 matter along with it and the racket, and soon renders one un- 
 used to this kind of walking quite weary ; but there is no help, and 
 on, on, through the slosh and snow a fatiguing tramp is continued. 
 It is almost useless to stop and rest, as, strange to say, the best rest 
 is to keep on walking. I do not know just how to explain this ap- 
 parent contradiction ; the truth, however, is revealed by a practical 
 test. At length the walking is so bad that I take off my rackets 
 and try my dogskin boots alone, though the condition of things 
 is hardly improved until the walking is bettered by an approach 
 to the island before mentioned, where a layer of snow materially 
 helps the ftiatter and forms a pleasant change from the sticky 
 material that I leave behind. 
 
 After pressing on with renewed courage I soon reached the 
 other side of the island, and couid see Bonne before me about 
 three miles ahead with a small point of mainland between us. 
 The walking was now good, but the weather had changed. The 
 wind blew hard from the northeast, and the cold was biting. Soon 
 the clouds began to gather, and, almost before I knew it, the at 
 first gently falling flakes of snow had agreed with the wind and 
 came drifting down in perfect sheets ; harder and harder came 
 the blinding storm, fiercer and fiercer blew the wind, heavier and 
 heavier became the rackets upon my feet, and then the feet them- 
 

 
 (, n 
 
 188 
 
 A SNOW FLURRY. 
 
 selves ; but there was nothing to be done but keep up the spirits 
 and march along. I have heard of fierce storms quickly aris- 
 ing in these arctic regions, that frequently bewildered the best of 
 even native travellers and turned him astray from an almost straight 
 road. I have, no doubt, frequently seen such from some house or 
 safe harbor without fully appreciating its intensity ; but I then saw 
 it and felt its full meaning, though I was too near a shelter to be 
 lost, unless overcome by fatigue or bewildered with fright, neither 
 of which seemed probable though of course possible. The wind 
 blew the snow in fine dust all over my clothes, and the cold pene- 
 trated even the thick cloth of which they were made. It was a 
 terrible walk to the mainland, but I soon gained it, and sheltered 
 myself in the fish stage of one of the establishments on the east 
 side. Here I rested while watching the storm. There was a strug- 
 gle in my mind as to whether to proceed at once across the snmll 
 bay to my friend's house a distance of about half a mile, or wait 
 until the storm had passed ; a determination, howevc-, to brave the 
 worst ; and, perhaps, a sort of feeling that it was better to con- 
 quer the elements, than be concjuered by them, took possession of 
 me, so I braced on the rackets again, buttoned my coat tightly 
 about me, and with an extra pull to my cap I started off. How the 
 wind did blow ! how the blinding snow beat o\ er and around me ! 
 that half mile seemed longer than all the other seven that I had 
 gone over combined. At last the other side was reached, the well 
 known place gained, and in a few moments I was in the presence 
 of the nicest and most pleasant family on the coast of Labrador, 
 without exception, while the weariness vanished as I was greeted 
 by friends and sat down with them to a capital vtx^ison dinner. 
 
 Any one who has been half starved for the space of several 
 weeks (almost months) on bread without butter, and tea and coffee 
 without sugar or milk, with any kind of meat save salt pork out of 
 the question, will appreciate a meal of good food when they get 
 it. The case was mine exactly ; and the savory smell of a good 
 venison stew with some large mealy potatoes as an addition, as it 
 
BONNE ESPERANCE IN WINTER. 
 
 189 
 
 was put upon my plate, clouded my vision to all other surround- 
 ings, so that I had eyes for this and nothing else, while I lost all 
 sense of the outer world in contemplation of that delicious plate 
 of food that was being passed to me. The tea with both milk and 
 sugar — for Mr. Whitely's is one of three families on the coast that 
 keep a cow — proved an excellent beverage ; while a real currant 
 pudding presented an additional feature that was wholly irresistible. 
 My friends must have thought me a barbarian and half starved at 
 that ; the latter I was without doubt. 
 
 Few changes had occurred at Bonne since my visit there in the 
 fall. The ice had frozen in the bay in front of the house, and snow 
 still topped the summits of the island ; big drifts lay about here 
 and there, showing the general direction of the wind in heaping 
 them up, but no one had attempted to dig through .hem, and it 
 would indeed have been labor lost since the wind would surely 
 have filled the paths in a single night to their f mer height. The 
 m.i^e portion of the family occupied their time chieflv in most in- 
 teresting conversation about deer and deer hu.iling, and some 
 eijj.aeen or twenty men had already gone off into the country on a 
 hunt ; the unusual abundance of these animals had fully rai-^ed 
 a deer craze that attracted nearly all the eligible young men of the 
 coast. Already some thirty deer had been killed, and others were 
 reported every day. I saw a beautiful head and horns of a young 
 buck shot by my friend, and it truly was a beauty. One man 
 twice killed two deer with a single ball ; this is remarkably fine 
 shooting and shots of this kind are only made by a good hunter 
 and one of steady nerves and eye. The deer, it is said, are very 
 tame ; but a man must stard a great deal and be able to walk many 
 miles a day if he would become a deer hunter in this country ; and 
 though I do not suppose I shall attempt a trip of this kind, I can 
 yet hope that chance may throw me in the way of shooting one 
 of these fine animals. This has been the only fresh meat that the 
 people about the river have had this winter, excepting the small 
 piece each received of the cow that was killed late in the fall. 
 
 '/ 
 
W) 
 
 
 
 I Si 
 
 I •: 
 
 It 
 
 
 190 
 
 TO THE VICTOR THE SPOILS. 
 
 All game has fled, and the only fear is that the cupidity — I can 
 call it nothing else — of the successful hunters will overcome their 
 natural good feeling and prevent them from making that distri- 
 bution of their spoils to the unsuccessful ones that they otherwise 
 would. There is a strong spirit of selfishness among some of the 
 inhabitants of the coast who have risen to comfort solely by their 
 own hard work, for many that are now poor might also have become 
 comfortably off by their own efforts, had they possessed courage to 
 press forward to the work. 
 
HOT WEATHER — TRIP TO AMOUR. 
 
 191 
 
 I can 
 their 
 distri- 
 jrwise 
 5f the 
 ' their 
 :come 
 ige to 
 
 CHAPTER XII. 
 
 80° in winter — Trip eastward — Starting — Esquimaux river and island — 
 Salmon bay — Bradore bay — Caribou island — Five leagues — Middle 
 bay — Belles Amour — Over Bradore hills — L'Anse and dunes — Blanc 
 Sablon — L'Anse Coteau — L'Anse Clair — Forteau — Amour. 
 
 
 
 Thursday, March 3. This was a remarkable day — though such 
 a day as does at times occur in this region. The thermometer 
 went steadily up, higher and higher, until about two o'clock it 
 reached 80°, and hot enough to make us all wish for summer clothing. 
 Soon it began to descend as rapidly as it had risen, until in the 
 night it must have reached zero, or very near it, again. This, I 
 believe, was the hottest day that I experienced on the coast ; 
 summer or winter. 
 
 On Friday morning, the 4th of March, I started on my long 
 expected trip from old Fort Bay to L'Anse Amour, a distance in 
 a straight line of only about thirty-five miles, while, by the way we 
 were obliged to go, along the bays and points of the coast, 
 it was nearly, if not quite, fifty miles. It took us two days to go, 
 and as many to return, while a stay of the same time at difierent 
 places on the way consumed the greater part of the week. It 
 was a pleasant trip, and the time seemed only too short, while it 
 was quite impossible to see all the numerous sights and go to all the 
 attractive places awaiting us on every hand. The weather was 
 most beautiful, and the air in a continual state of exhilarating 
 freshness. Only one or two small flurries of snow passed over 
 us, while, for the most part, the sky was of a cloudless blue. 
 Occasionally a light haziness concealed the sun, and once a dark 
 
I'.v 
 
 J' t • 1 
 
 192 
 
 PREPARING FOR A TRIP. 
 
 V,v 
 
 r-:' 
 
 gathering in the southwest reminded one forcibly of the coming on 
 of a thunderstorm as we frequently see it at home. Once, also, a 
 light fog arose. I mention it simply because our guide said that 
 it was very rare to see a fog in the month of March. The cool 
 breezes were mostly from points from north to south by east. The 
 cold was not intense, ^nd at no time did I find it necessary to wear 
 an overcoat over the heavy clothes that I had on though it proved a 
 very convenient cushion for the komatik. I think there was but 
 one day when the sun thawed the snow ; but the roads were good 
 all the time. Over hill and on bay, wherever we went the travelling 
 was good, for dogs properly shod, and there was no reason to 
 complain of it. We did not need rackets, and the snow was good 
 and firm, while a beaten path most of the way united with everything 
 else to make my journey a most enjoyable affair. 
 
 In starting upon a journey, in these parts, little or no provision 
 is made for food upon the way, unless the distance to be passed 
 over contains no houses ; for, as I have frequently observed, to 
 refuse hospitality even to an enemy is considered the worst crime 
 of which a man here can be guilty. That important matter settled, 
 the articles to be carried are carefully selected. As it requires no 
 great time to select the articles for such a trip, one is, so to speak, 
 ready at any time for a journey. 
 
 Friday we were awake and about bright and early. The day was 
 fair, and it was not long before a hearty breakfast was awaiting us. 
 After our simple meal we hastily got together our effects and 
 lashed them to the komatik. They presented the appearance of 
 of a rather high cushion, and could be easily taken for a seat 
 prepared for the occasion. Large sacks of rough bagging for carry- 
 ing articles which might be picked up on the way, such as food or 
 meat for the dogs, etc., we added two such bags to our list, and 
 with our overcoats, a pair of rackets in case such should be needed 
 on account of the roads, and my gun, the equipment was as com- 
 plete as it was simple. 
 
 The harnessing of the dogs did not take long. The early hour, 
 and the unusual stir seemed to show them that something out of 
 
PREPARING FOR OUR JOURNEY. 
 
 193 
 
 the ordinary course of proceedings was in order, and they too 
 showed their deHght at the prospect by their gambols and frolics. 
 It was really somewhat difficult work to harness them. They would 
 allow themselves to be caught only with great difficulty, and then 
 would soon manage one way or another to break loose just when 
 we thought we had secured them, and were off dgain barking at 
 us from behind the woodpile some yards distant. It is a matter of 
 no small difficulty to "rig" a komatik for a trip, getting all the lashings 
 and the ropes in proper order, especially when seven dogs are to be 
 caught and harnessed separately, their lines fastened to the petook 
 (if this is the proper spelling, which I could not ascertain cer- 
 tainly), being what we call the rope to a sled. The coldness 
 of the air chilled our fingers ; the dogs provoked us with their 
 uncalled-for gambols and frolics, refusing to be caught, though 
 they seemed to regard it as great sport. At last the three large 
 dogs and the three puppies, which latter had lately been broken 
 in were all harnessed ; the effects mentioned were tightly strapped 
 on the sled, and all was ready. It is a curious sight to see the 
 dogs thus harnessed striving to start the sled, each on his own 
 account. When the last one is harnessed, if the team is a good 
 one, each dog starts off with a run in a separate direction ; it needs 
 a good driver to control them or hold them in, or I should say 
 hold the sled in, until he is ready to start. The effect presented 
 reminds one much of the directions in which the five fingers point 
 when the hand is spread out to its fullest extent. Each dog starts off 
 with a rush, brings up at the end of his harness with a thug, as the 
 narrow but short sealskin thong tightens, and the dog is elevated 
 several feet in the air. The driver stands holding the komatik back 
 with both hands by main force, while he shouts at the top of his 
 voice to the dogs, with very little effect, to stop them or hold them 
 in until he is ready. It was a cold morning, but we had not long 
 to wait. Hurrying out of doors, and waving good-bye, we were off 
 the moment the hi, hi, hi, told that all was ready. 
 When a team of good dogs first starts off it is almost, and one might 
 
 !■ .; 
 
 Hill. 
 
 18 
 
mm 
 
 I 
 
 V„ '1: 
 
 -I .'■. 
 
 ■I- . ; 
 
 M 
 
 194 
 
 FROM OLD FORT BAY TO THE RIVER. 
 
 well say quite, impossible to guide them. They will start off in any 
 direction of their own accord, perhaps towards the very opposite 
 side of the bay, and it is not until the first rush is over, and they 
 have somewhat abated their eagerness, that the driver's guiding 
 words are heeded at all. At last the head dog minds the voice 
 and gradually turns into the right track, when the others follow 
 a little behind, and the trip is really begun. It was thus with us, 
 when at five minutes past eight o'clock we started off in the face 
 of a good cool breeze from the north, in a clear air, with a bright 
 sun shining on us from an almost perfectly cloudless sky, for our 
 trip to the eastward. 
 
 The road from Old Fort Bay to the river, the same which I had 
 gone over in komatik and on foot so many times, was not changed 
 much this morning ; of course one would hardly expect it to change, 
 except by accumulation or disappearance of snow and the thawing 
 and freezing anew of the ice of the bay. The same level platform 
 of thick ice (broken only on the sides of the bay near the foot of 
 the hills, while its surface by a slight recent thaw was covered with 
 innumerable sharp needles of ice pointing upward occasioned by 
 a smart frost in the midst of the thaw, rendering the going anything 
 but good for the dogs' feet) spread before us for about a mile and a 
 half to the firot portage, or path over the low hills, to the nearest 
 pond over which one is obliged to travel. As we rode through this 
 long, narrow cleft of the hills, I could not but think of the mighty 
 force that had here been at work for centuries to render the place 
 what it is ; as also of the process by which indentations were made 
 that left a succession of peaks separated by partial valleys on either 
 side, from the mouth of the bay to the bottom, and extending for- 
 ward and inward in successive chains far into the interior ; and of 
 the gorges and individual shape of each crest, bending gently 
 down to the water from the upper end of the bay, while ending 
 generally in rather abrupt, low, and yet often nearly perpendicular 
 cliffs or cliff-like rocks on each side. Then the probable agency 
 in polishing the rocks, often quite smooth, and in rounding off the 
 
 h 
 
NATURE OF THE COUNTRY. 
 
 195 
 
 I 
 
 peaks — for it is seldom that you anywhere see a pointed hill top — 
 and apparently shaping the whole general appearance of hills, 
 valleys, and surrounding elevations and depressions. 
 
 Thus thinking, we rode swiftly and pleasantly along the bay ; we 
 crossed the first portage, over a low bank, with hills on either 
 side, at a little to the right of the bottom of the bay, where 
 a rather high hill slopes to the water and forms its boundary line, 
 and descended to the first pond. This pond is on the same level 
 as the bay, and separated from it by only a low ridge of land from 
 which the rock of the region crops out on all hands. It is nestled 
 in a hollow with hills all around it while the outline, more or 
 less circular in form, encloses an area of perhaps a half a mile. 
 Across this we go, and then slowly glide along the second portage 
 which, like the ridge containing it, is nearly on a level with the 
 pond. On the right, the low, rather swampy looking level is covered 
 with the gnarled trunks and stems of the dwarf birch so common in 
 this locality, the Betula populifolia of the botanists, and the best 
 fire-wood that the region affords, while on the left we pass close 
 to a long, perpendicular wall of rock that, rising some six feet and 
 extending some fifty in an even line, abruptly ends the ridge in 
 this place ; its surface is as smooth as if polished by some mighty 
 force, and yet I saw no scratches, or even signs of cleavage any- 
 where upon its face. The extent of this portage is about the 
 same as that of the one we had recently passed. 
 
 We now descend a little to a second pond ; a very small stream of 
 good drinking water (to which a large hole has been made through 
 the snow) runs along this path. This pond was soon passed, and 
 after ascending a low bank similar to that we crossed in going 
 to the first pond, and like it a low ridge between hills, we de- 
 scended to the river. Esquimaux or St. Paul's river, as it is called 
 — the latter being the proper name, though it figures upon the 
 charts as the former, — is peculiar in many respects. Its mouth 
 is quite large, and with several flexures or bends ; while Esquimaux 
 Island lies directly on the right hand, leaving only a small, narrow, 
 f 
 
 
'''hlk 
 
 I'-i 
 
 r :i 
 
 <; 
 
 
 ■■J ■ ,' i 1 
 
 i 
 
 »iHl 
 
 ill '111 I ! 
 
 196 
 
 SCENERY AROUND ESQUIMAUX ISLAND. 
 
 and almost straight passage to the open sea beyond. About seven 
 miles inland the river narrows, passes some rapids, then enlarges 
 again to the next series of rapids, beyond which I have not been. 
 The waters are covered with ice the greater part of the time from 
 November to June. The broad bay-like portion, on which we now 
 entered, is everywhere surrounded with high hills which resemble 
 those inland. Several deep indentations, formerly the sites of one 
 or more houses, but which have been abandoned from unknown 
 causes, are to be seen on the left, as we enter the western side and 
 observe Esquimaux Island in the near distance in front of us ; 
 while the water reaches the sea on the right, and the tops only of 
 several little islands appear to relieve in summer the otherwise open 
 sea, and in winter the level ice. It is a wild scene, — nothing but 
 snow-capped hills, in the distance, above and around, while huge 
 heads of rocks jut out in long, stubbed, and rounded points of land, 
 dividing the bay-like river bend into several smaller pond-like bends 
 or bays. We skim along over the ice, in our hobgoblin team of 
 dogs with sled, enjoying the beauty of nature thus opened to us in 
 her icy realm, and drink in the fresh, yet not too cold, air that 
 stings our ears and gives us ruddy faces. 
 
 The usual path to the mission is soon traversed, and we pass on 
 our right the long, even, low sloping ridge on the north side of 
 Esquimaux Island with rather higher crests on either side, extend- 
 ing still further north, the whole looking as if it might once have 
 been the terminus of some glacial mass that swept over it, as well 
 probably as of most of these regions, and on our left the group of 
 some e'ght or ten houses which, with the church, constitute the 
 winter quarters of the inhabitants of the neighborhood. Now we 
 enter a small bend in that side of the river directly ahead of us, 
 which leads to a gorge-like opening between either two very high 
 hills (high for this region) or a single hill cleft in twain by some 
 mighty force, the latter of which is more probable, and begin to 
 climb the steep bank above covered with short spruce and fir trees, 
 by a narrow and winding portage. Up, up the bank we climb, 
 
LAC SALE. 
 
 197 
 
 that 
 
 at least the dogs climb while we get off the sled and walk, at 
 the top of which we still see the hills apparently four times the 
 distance we have reached above us ; a walk of a few hundred rods 
 brings us to the opposite slope. On our left the huge mass of rock 
 that forms the greater part of the hill has been broken off or 
 rent asunder with the greatest nicety, presenting huge columns 
 of granite, more or less square, on a nearly perpendicular face of 
 rock. If you take a piece of paper (stiff writing paper is the best) 
 fold it backwards and forwards in creases about three-eighths to 
 five-eighths of an inch thick alternately, it will give you a very 
 fair representation of the cliff as I saw it : nor did I see a cor- 
 responding rent anywhere in the rock on the opposite side. We 
 now descended the slope and entered the pond-hke arm of the 
 river, or lac sale, as it is here called, which signifies a salt lake, or 
 a salt water lake-like bay of the sea or river, which latter it is in this 
 case. We had a nice path across lac sale, as, in fact, we did on 
 nearly or quite all the ponds between there and our journey's end, 
 and on this litrie arm of the sea, not a mile in uneven extent of 
 coast line, we saw hills, cliffs, and wooded slopes on either hand 
 around, nor was the passage through which we were to pass visible 
 anywhere before us. It was a complete picture, and worthy the 
 pencil of an artist. 
 
 We crossed a small pond directly after passing the salt lake just 
 mentioned, and enclosed, like it, with high hills on either side. 
 It was about on the same level with the water which we passed. Be- 
 tween lac sale and this pond, we found a long, narrow pass, where 
 the rocks came close to the passage with their abruptly broken 
 sides and edges gently sloping upward to low, rounded crests on 
 either hand. Through the pass thus formed flowed a small stream 
 of water running over rocks, the bed of a brook, and evidently 
 broken pieces from the neighboring hills. It formed a pretty sight 
 to see this little stream in such a natural and yet rather unlooked-for 
 place, and the broken ice above showed the quite clear water as it 
 flowed gently beneath. 
 
 1 1 
 
■•^r 
 
 
 it 
 
 'ii 
 
 
 ;i 
 
 
 1' ■ 
 
 ■• I >-i 
 
 I 
 
 
 » 
 
 
 
 
 1' 
 
 j ■■ 
 
 ■1 . 
 
 
 : 
 
 m 
 
 I ! 
 
 l!n 
 
 198 
 
 COUNTRY ABOUT SALMON BAY. 
 
 Soon these scenes of picturesque beauty began to be familiar 
 to us though we never tired of them ; while the new scenes or new 
 variations of old scenes repeated the picture in a more beautiful 
 manner and added the charm of freshness and expected change 
 to those which were to come. The long, narrow portages soon 
 gave way to wider and broader tracts of country through which the 
 way wound as before. The hills were reduced to low crests, and 
 open field-like slopes on ridge and high level plains appeared in 
 the distance. We soon crossed Salmon Bay, coming out of a long 
 arm-like expansion of the neighboring lands, and which formed the 
 centre of three, all similar expansions, the right and left of which 
 appeared more like sort of bays or rounded arms, as one looked 
 back upon the view from the opposite side of the bay. One or 
 two houses were visible, the first seen since leaving the river, a 
 distance of some five or six miles behind us. And here we came 
 to another of these strangely formed places or ridge-like openings 
 in the valley, between hills so often seen in different places all along 
 the coast, and which remind one of artificial, though they are in 
 truth quite natural, terraces. The height of this formation is about 
 thirty feet, its top is nearly level. On the left, looking towards the 
 sea, are the high hills of Bradore — though I do not mean the Bra- 
 dore hills so called, which are over eleven hundred feet in height 
 and situated some way back in the country, while those of 
 which I speak form the boundary of Bradore Bay on its extreme 
 western side and are only about four hundred feet high — with a 
 rather abrupt slope to this ridge. On the right a rather low crest 
 separated it from the sea. The ridge itself slopes to the water 
 on each side, at an angle of about 30° on the Salmon Bay, and by 
 a gentle rather than abrupt slope on the seaward side. The ridge 
 points in a northeastward and southwestward direction as do most 
 other similar ridges where I have seen them, and the direction of 
 the slopes, and general appearance of the top of the plateau 
 suggest to one's mind the final resting place of some arm of the 
 mighty Labrador glacier before or rather during the final plunge of 
 
 I 
 
CARIBOU ISLAND AND VICINITY. 
 
 199 
 
 its huge mass of ice into the sea. Surely these well-defined and 
 peculiar shaped plateaus are of glacial origin, and being in all 
 cases, so far as I have seen them, of pretty much the same form 
 and shape, the suggestion is of a common glacial one. Before we 
 leave the subject a few words about Salmon Bay may be of interest. 
 Salmon Bay, like the majority of the long, fiord-like bays so numerous 
 on the coast east of it, is a long, rather narrow inlet, about four 
 miles in length, extending into the land in a northwesterly di- 
 rection. At its mouth Caribou Island, a little island a mile wide 
 and a mile and a half in extent, occupies the greater part of the 
 western side of the entrance ; beyond a small pond-like piece of 
 water, about three-quarters of a mile in either direction, is nearly 
 enclosed by the small points of land that reach towards each other 
 from the opposite side of the basin which admits one to the bay 
 proper. In summer this is the seat of one of the largest fishing 
 posts on this portion of the coast. Some eight or ten establish- 
 ments here employ about one hundred and fifty men, and seven 
 cod seines valued at twenty-five hundred dollars are used in the 
 fisheries. 
 
 I must not forget that the country to the east of Salmon Bay, 
 and before reaching the high cliffs and bluffs just next the river, is 
 rather low, and made up Df rounded crests of hills covered here and 
 there with a scanty vegetation, presenting something of the appear- 
 ance of a hilly New England farm lot. The tof« of the hills were 
 strewed with a few loc se stones either broken from the rocks them- 
 selves or left there by some agency of ice or water. The same 
 general features existed in most parts of the bay on both sides, 
 and as I looked upon the scene I could almost imagine the mighty 
 force that, slowly pushing itself and creeping onward by its own 
 plastic mobility down the channels from some interior region, 
 gradually ground off" these hill tops, and scattered these stones, while 
 it levelled the feeble barriers that attempted to confine it to the 
 land, and buried itself in the sea. 
 
 Frdm Salmon Bay, across a narrdw plateau-like ridge, we de- 
 scended to the sea again— frozen in a mass of icy needles that flashed 
 
 1 ! 
 
 H 
 
 Ij 
 
 !i 
 
i f '^-i' 
 
 'I' ! 
 
 'Am 
 
 200 
 
 FIVE LEAGUES AND VICINITY. 
 
 and glittered in the sun, most treacherous and cutting to the dogs' 
 feet, the precursors and frequent occasion of that malady, so terrible 
 in these regions, snow blindness — and continued our course — with 
 a glimpse of the far-off shores of Newfoundland, towards which 
 spread an icy plain with a distant channel of open water, at our right, 
 and the shore, skirted with the broken and fantastically shaped 
 blocks of ice that lay piled, in all sorts of positions, on or against 
 each other often several feet high, like tide marks on the left, — 
 across the three full miles of Five Leagues Harbor to Five Leagues 
 Point, the residence of a well known and thrifty fisherman and 
 seal-catcher. Taking a cup of tea here, we continued our way to 
 Middle Bay, on the east side of which, with another friendly ac- 
 quaintance, we took our dinner and nooning before continuing 
 farther. As we had made full sixteen miles since morning, both 
 the dogs and ourselves needed and enjoyed this rest. After our din- 
 ner, which consisted of bread and tea, the latter a concoction of the 
 twigs or rather boughs of the spruce, which are often used as a 
 substitute for the real article and so carefully prepared that it is 
 sometimes hard to distinguish from an inferior quality of it, we har- 
 nessed our dogs, and were once more on our journey. 
 
 I cannot quite recollect the exact position of the house of this 
 hospitable fisherman, or of a large, old-fashioned, red-painted, 
 country looking house that we passed somewhere here, which looked 
 more like a comfortable establishment than any I had seen for a 
 long while, and which, though I did not stop there, no doubt 
 would have fully equalled the expectations excited from the outside 
 appearance. Behind this was another of those peculiar ridges 
 such as have been mentioned as occurring in other localities. As 
 the direction of Middle Bay is about north northeast, this narrow 
 ridge barely separates the eastern baylet — if I may use the word — 
 of the bottom of the bay, from the western baylet of the opposite 
 or Belles Amour Bay. The intervening neck (the ridge I have 
 mentioned) extends outward on either side and forms a square 
 block of land nearly two miles each way ; on this are one or two 
 
 I 
 
 i 
 
 W\' 
 
 k' 
 
 ii 
 
 IK 
 
 i vi 
 
 '1 j!i 
 
BRADORE AND VICINITY. 
 
 201 
 
 I 
 
 fishing establishments that yield or ought to yield a good profit to 
 their owners, but I sadly fear that industry is greatly crippled here 
 as in other places about the Labrador coast, l>y a feeling that says 
 let well enough (which is sometimes poor at best) alone, and hin- 
 ders any real progress either in wealth or desire for cultivating the 
 intellectual qualities. 
 
 I have been struck with the really musical and pretty names which 
 appear all along the coast, and bear the marks of that society lan- 
 guage — the French. Belles Amours signifies something like loves 
 or passions of beautiful women, and undoubtedly took its name from 
 some conquest of love by or over some fair inhabitant of the place. 
 
 We now descended to the famous Bradore bay, the " Brest" of 
 the early inhabitants when Cartier discovered the St. Lawrence, 
 when Cabot discovered Newfoundland, and when Corterel is said 
 to have discovered Labrador. The shape of this bay is very nearly 
 that of an acute angled triangle — the upper side, extending due 
 east by north, in an almost straight line for nearly seven miles ; the 
 other side (which I shall call the outer side) due north by slightly 
 west (the centre part bulging out giving to the whole side the appear- 
 ance of a bent bow with the end slightly recurved) for six miles ; 
 while the base, or a straight line from the extremity of Stony Point 
 its western, to the extremity of Grand Point its eastern, boundary, 
 the distance is eight miles. The only island of any account in 
 this large bay is called the Ledges Island and is about a mile wide 
 by a mile and a half long only. It is about a mile from shore near 
 the centre of the outer side, and, with a few scattering rocks, forms 
 a mass of dangerous shoals in this northeastern end of the bay. 
 
 The little island near the land on the extremity of the outer side, 
 and quite near Grand Point, is Paroquet (also spelled Peroquet) 
 Island. It is a very small island, a mere rock rather than island 
 scarcely half a mile in either direction, but small as it is it is as noted 
 perhaps in its way as any of the larger islands on the coast. As its 
 name would indicate, it is the abode of the paroquet or puffin, the 
 sea parrot as it is called, or the Fratercula arcticus of the or- 
 nithologists. On this small isknd thousands of these plump little 
 
 Ml 
 
1! ■ 
 
 ^i 
 
 I 
 
 I 
 
 
 
 1: 
 
 m 
 
 ■ 
 
 ■A] 
 
 I'i^l' 
 
 
 1 ■ i. i 
 ! :! 
 
 -■ ■( ■'- ■( 
 
 i 
 
 
 jj 
 
 
 
 ■|< 
 
 1 
 
 1 ' - '■ 
 
 1 
 
 1 
 
 
 \ • 
 
 202 
 
 SCENE ON BRADORE HEIGHTS. 
 
 birds, with their huge scissors-shaped, horny bills, collect in the sum- 
 mer and breed in large companies. Their nests are made in holes 
 in the ground, which is tunnelled with their burrows in every 
 direction. A great trick of the inhabitants with strangers who visit 
 this island is to get some person — the greener the better — to 
 put his hand into the hole when the bird is within sitting upon its 
 eggs. If you could but see the bill, you would readily understand 
 why the attempt is never made a second time. Imagine a huge pair 
 of horny scissors, two inches long, two inches high at the base, and 
 half an inch thick yielded with force, and no wonder one does not 
 care to repeat the experiment. 
 
 As the ice in Bradore bay was not all quite fast, that is, as open 
 water appeared on it in many places, we were obliged to take to 
 the hills and cross by the land. The hills and high cliffs which 
 everywhere skirt the north shore, sometimes almost perpendicular to 
 the water at their base, sustain a plateau extending far back into the 
 country. A mass of hillocks crowns its summit, varying from 
 two hundred and seventy-five to three hundred and twenty-five 
 feet above the sea-level. The whole plateau, if I may so call it, 
 spreads out for a long distance inland, and forms nearly or quite all 
 the country upon the upper side of the bay line. During our whole 
 ride over these hills the constant sight of Something new, the 
 unlooked-for contrasts presented here and there, rendered the trip 
 across them a source of pleasure and delight — one well worthy 
 the express journey of some geologist or glacial specialist, who 
 should read to us the sermons these rocks contain. Now a long and 
 wide (almost level) plain with its rather uneven surface would end in 
 a row of low crests of unequal height, beyond which the slope would 
 carry us some fifty or seventy-five feet to a pond ; beyond a varied 
 plain and hilly patch of ground would often give place to another 
 pond, almost at a level with the surrounding hillocks. Snow covered 
 almost everything, and only the rounded or broken hill-tops, and 
 the rough pieces scattered from their place in the rock thus crushed, 
 or scattered over the slopes by some mighty agency, here and there, 
 
 iteaiiiiiiiiiHiiiHli 
 
LRADORF TO BLANC SABLON. 
 
 203 
 
 rather evenly though sparingly deposited, alone offered a glimpse of 
 the appearance of the surface beneath the snow. These blocks 
 were for the most part very angular, often nearly scjuare ; occasion- 
 ally a roimded one appeared, und I saw one huge, almost round 
 stone, nearly two feet through each way, balanced nicely 'etween two 
 upright rocks, while on the opposite side an almost perfect and 
 solid block cube of stone was nicely balanced upon the fine point 
 of a small upright chip of rock, from which but a slight push would 
 have dislodged it, — as also the rounded stone opposite — showing 
 that the force or agency that left it there c luld hardly have been a 
 convulsive, but rather a slow-moving and methodical one. At the 
 northeast corner of the bay we took to the ice again, and finding it 
 firm and hard we skirted along the edge of the outer side, and 
 crossing Grand Point — where the formation begins to differ from 
 what I have ju t described, and to take on the aspect of the whole 
 plateau this side of Bradore and extending to Forteau bay, a distance 
 of some eleven miles across, being a sandstone deposited in regular 
 layers of which I shall speak soon — we landed at Blanc Sablon, 
 the eastern termination or boundary line of the Province of Quebec. 
 
 Beyond, the whole broad plateau or table-land of Labrador 
 proper extends. It exists as a dependent of the provincial gov- 
 ernment of British Newfoundland, to which Labrador pays duty 
 and tribute ir. the shape of revenue on all goods received within 
 its boundaries. 
 
 I will now speak of a little cove on the southern end of this 
 eastern side of Bradore bay. I am almost certain that it is that 
 portion marked in the charts where the water encroaches upon the 
 land in a semicircular-like baylet almost directly opposite Paroquet 
 Island. I discovered here a beach of pure sand, extending backward 
 into a series of regular hillocks of sand or sand dunes, as they are 
 called, which rise above the level of the surrounding country back of 
 them. That they appear in this place, marking as they seem to do 
 the separation between the Bradore granite and Blanc Sablon sand- 
 stone, and running away back in gentle slope to what seems to have 
 been once a bay of the sea with its nearly level extension between a 
 
 T 
 
 i\ 
 
^i; 
 
 I 
 
 li 
 
 
 204 
 
 L'ANSE AUX DUNES. 
 
 W: 
 
 high plateau on either side, is perhaps rather remarkable, and of 
 geological interest. The peculiar name given to this place of 
 L'Anse aux Dunes, would seem appropriate were the word {funes in 
 French translated dunes in English — the word anse signifying a 
 small bay or cove. Here five or six men set some nine nets for 
 seal in the spring, and also catch a few salmon and cod in the 
 summer, but it is only a small station at best. Near the water the 
 beach was strewn with shells, especially of the razor-billed clam as 
 it is often called, and a variety of sea animals, while huge drifts of sea- 
 weed, like that so common in many places on the coast here and 
 on the Atlantic generally, were entangled with blocks of ice and lay 
 around from nearly one end of the beach to the other. That night 
 we spent at Blanc Sablon, having travelled a distance of twenty-five 
 miles, nearer thirty by the route we took, since leaving home in the 
 morning ; accomplishing the whole, including stops and all, in a 
 little less than ten hours. The distance would have been gone over 
 in much less time had the dogs been properly shod ; but, as 
 it was, the ice cut their feet so badly that blood flowed in nearly 
 each print in the snow or on the ice, while the poor animals were 
 quite stiff" and lame the next morning. 
 
 Of Blanc Sablon I shall say very little, as I had so little time here 
 to visit the place. On all sides of the harbor the country was one 
 mass of stratified deposit of sandstone, as is the whole plateau, as 
 I have before called it, between this place and Forteau. On the 
 western side of the harbor the deposit is much broken, numerous 
 hill-like elevations and broken ridges cover the surface everywhere 
 on the tops of which a scattered vegetation flourishes ; on the east- 
 ern side the deposit forms a ridge that extends for some distance in 
 a north and northeasterly direction. For a mile or so in a north- 
 erly direction the water, retreating, has left the hollow occupied 
 formerly by the sea so perfect that but little imagination is required 
 to trace the former outline of the harbor. With the place itself 
 I was much pleased. It is a busy looking station, and from its 
 geographical position, large size, and importance as a fishing locality, 
 it is often called the Boston of Labrador. It contains perhaps a 
 
BLANC SABLON AND VICINITY. 
 
 205 
 
 hundred buildings on both sides the harbor and the adjoining 
 islands, of which I shall speak presently, and some three hundred 
 inhabitants, increased to twice that number in the summer and fish- 
 ing season. The "large room," as it is called, contains an abundance 
 of articles for sale or trade in exchange for furs, fish, and other 
 articles used as a fair medium of exchange, at their current price 
 either in trade or cash. The stock is owned by merchants from the 
 isle of Jersey in the English channel — whence a great number of 
 people come who live on this coast — and now in charge of a smart, 
 gentlemanly clerk already some eight years in the companies' em- 
 ployment though still in his teens. The fishery connected with 
 this same firm, employing some ten men in winter, and eighty in 
 summer, is perhaps the largest in the place, and no wonder this 
 young fellow clerk who oversees the whole is called the young 
 master, while he still holds his position and retains the confidence 
 of his employers as well as his employes. The two important islands, 
 just outside the harbor, are Wood {I'tle au Bois as it is usually 
 called) and Greenley Islands. The former contains several fishing 
 establishments, and another shop where various articles are bought 
 and sold, and is the larger of the two being about two miles long, of 
 triangular shape, with the apex pointing northwest, and a base about 
 a mile wide facing southeast ; the latter the light island, where the 
 lighthouse is situated, one of the only two upon the coast from 
 Point des Monts in the river St. Lawrence, eastward. It lies west 
 and a very little south of Wood island, and is scarcely three-quarters 
 of a mile either long or wide. To give a little, perhaps rather dry 
 detail, I find the light described in the report of the Canadian 
 government, somewhat as follows : "light upon the S. W. part of the 
 island, latitude, 51" 22' 35", longitude (west of Greenwich), 57" 
 10' 50". Single light, revolving white for \, then red ^, then white 
 again ^, b'ank i^ minutes (three minutes for full revolution). The 
 tower octagonal, wooden with dwelling attached, 100 feet high, the 
 light visible fifteen miles. It is an established light of the second 
 order, and a fog gun is fired every half hour." 
 
 Taking a good rest at the house of the friendly people with whom 
 
 ■■ \ 
 

 ;li' -■ 
 
 \ 
 
 
 ■■ -ii 
 
 f 
 
 
 ! ': il'i;; 
 
 I', ^ 
 
 ,.!' 
 
 ,'kr 
 
 206 
 
 HOW THE INHABITANTS LIVE. 
 
 we had taken up our quarters for the night, we gave ourselves to 
 the enjoyment of the clear, cool, evening air, and soon after supper 
 retired to sleep. The people in many places along the coast where 
 I have visited, at least the men, have a habit of closing their day's 
 work in winter with their supper. At that time all the work of the 
 day is supposed to be over, and after the meal the principal occu- 
 pation is visiting one's neighbors and either there or at home spend- 
 ing the evening in chatting, and smoking very poor and bad 
 smelling tobacco, which, however, seems to be the best the coast 
 affords. What would at home be thrown away as worthless is here 
 sold in large quantities for a high price. I mention this fact to 
 show the imposition often practised upon these people who are un- 
 able to help themselves, and who must either put up with it or go 
 without ; strange to say, most prefer to put up with the bad article, 
 and hundreds of hundred-weight boxes of bad tobacco are used 
 each season. Nearly all of the men on the coast smoke, and it is 
 very rare to find a man, old or young, to whom the pipe is a stranger, 
 or who does not use tobacco in one form or another. Even the 
 young boys smoke as soon as they can secure the necessary articles. 
 It seems to be a matter of pride with them here as elsewhere, and 
 the fellow who can begin the earliest and smoke the most tobacco 
 is the best fellow, and takes the lead among his comrades and 
 companions. 
 
 The houses here are much like the houses elsewhere, a huge 
 kitchen and sitting-room with a small bedroom on the ground story, 
 and a loft with two rooms reached by a ladder leading to a small 
 aperture about two feet wide in the floor above. The bed or bunk 
 is roughly made, and more often the mattress simply lies on the floor. 
 The people with hardly an exception are fishermen, and the best 
 off are poor compared to what they might be should a spirit of 
 economy and industry take possession of them and drive away the 
 spirit of procrastination. The occupation during the day is chiefly 
 that of mending and netting nets and seines, or laying plans for 
 and chatting with neighbors on the next season's work among the 
 fisheries. The fishing season is usually from the first of May, the 
 
 m 
 
 !,.!: 
 
 lit !>; 
 
EASTWARD OF BLANC SABLON. 
 
 207 
 
 opening of spring, until the first of September, the beginning of 
 cold weather, but during these four months all is stir and excitement 
 both day and night. Men in boats are out fishing all the time, and 
 every device thought of is put into operation for the securing of a 
 good " catch," as the work of the season is called. Labor is cheap, 
 and none but lazy persons fail to enter into the full excitement of 
 the occasion ; these well deserve the poverty brought upon them 
 by their own idleness and want of exertion. I have known whole 
 families to exist on the charity of their neighbors with perhaps only 
 a single or at most two barrels, one of flour and the other of meal, 
 during the long six months of winter ; but the idle always have 
 their own reward, at least on this coast where no one is rich enough 
 to supply such persons at their own expense, even if they were so 
 disposed. 
 
 Early the next morning we tackled our still lame and tired dogs 
 and proceeded on our journey. The dogs went badly at first but 
 they bravely overcame the tendency to weakness, caused by their 
 lame feet, and soon trotted along at so brisk a pace that, but for 
 the occasional drops of blood left in their tracks from wounds 
 ttidde by the sharp ice cutting their feet, one would scarcely notice 
 their tendency to fatigue. We soon climbed the steep, high hill 
 directly back of the houses and the ridge or rather long, low gran- 
 ite mound — if one may so call it — and found ourselves on an 
 uneven plateau of sandstone, the upper surface of which looked 
 much like that of the Bradore granite ; but while the vegetation 
 on the Bradore Hills was scant and poor, here dwarf spruce and 
 fir trees showed occasional green tops and branches from beneath 
 the snow ; while broken and cut sticks, appearing in all directions, 
 showed where the ground, once completely covered with these trees, 
 had been cleared for fuel, and left to again grow over, probably 
 for the same ultimate use. I could not but reflect that I was riding 
 over a mass of stratified sandstone, extending miles in three di- 
 rections with the sea at the south, and once a part of the ocean, 
 which receded either by elevation of the land, by sinking of the sea 
 basin, or a combination of both. With about the same slope that 
 
 I 
 
■ im' 
 
 208 
 
 L'ANSE AU CLAIRE. 
 
 
 r t 
 
 \ f! 
 
 we ascend from Blanc Sablon to this plateau, we descended to the 
 bay on the opposite side from an elevation of about four hundred 
 and thirty feet, and, as I later observed, the slope was nearly the 
 same on the southern side also, though the distance passed over 
 here was only about three miles. 
 
 The ride was so full of interest that I scarcely realized when a 
 view of the little inlet of St. Claire or L'Anse au Claire Bay came 
 upon us quite unexpectedly, and the ride down the hill and along 
 the western bank of the bay afforded a fine view of the surrounding 
 beauties of nature. Here we took dinner with a most excellent 
 family, and the pleasant conversation of both host and hostess, who 
 were great friends with my guide, proved most interesting. The 
 gentleman who had for a long time resided here as a fisherman — 
 and certainly a very intelligent one — gave me much useful inform- 
 ation about the country, and stated the fact that he had, from 
 personal examination, found that this same sandstone formation 
 continued beneath the sea, as he found from soundings while fishing 
 for cod, and appeared again on the Newfoundland shore, which is 
 here about fourteen miles across its nearest point. I saw here, also, 
 that this little bay partook of the same general character of so many 
 other bays along the coast here, and distincdy showed the former 
 extension of the sea in its basin beyond, while the limestone cliffs 
 and hillocks appeared in all directions, to the right, to the left, and 
 directly ahead After our dinner and a little time allowed for rest- 
 ing ourselves we continued our journey in a cool, bracing air, across 
 the bay. We ascended the opposite ridge, a continuation of what 
 we had just passed over, and with a similar height above the sea, 
 only perhaps with a rather more even surface, and crossing several 
 large ponds, continued our way over these hills in a triangular 
 direction to the height above Forieau Bay, a distance of five miles, 
 passing through a slight snow fog, a peculiarity of this district per- 
 haps (though I am assured of rare occurrence in the month of 
 March), which, barely lifting, gave us a hazy view of what was 
 around. We now descended, in a tremendous hurry, the famous 
 steep Forteau hill, that is situated here, to the bay ; the dogs could 
 
FORTEAU BAY. 
 
 209 
 
 over 
 
 not keep up to the sled in spite of double drags, or huge rings of 
 thick rope, thrown over the runners, fall between them and the 
 snow and form a powerful resistance to the forward motion of the 
 sled, — yet we reached the bottom in safety. 
 
 As this is another of the important indentations of the coast, it 
 may be well to give simply an outline of the principal points of inter- 
 est connected with it. If then we measure the distance between Point 
 Forteau on the western and L'Anse Amour on the eastern extremity 
 of the bay, we shall find a straight line between these two places to 
 run in a northeast by easterly direction for a distance of nearly four 
 miles. The outline of the bay itself is triangular ; the apex from 
 the centre of the mouih of the bay lies inland in an almost ex- 
 act northwesterly direction. The western side of the bay is nearly 
 straight, while the eastern represents the figure 5 with the dash at 
 the top off, and the curve at the bottom of the bulge not quite 
 reaching a point exactly below the straight line above as it neces- 
 sarily does in a well made figure of that character. Of course 
 there are many little irregularities of coast-line, but the general 
 form is quite as I have described it. At the extremity of Amour 
 Point stands Amour lighthouse, the other of the two lights on 
 the coast, and which will soon be described. Strange as it may 
 be, Forteau Bay basin shows still another of those plainly marked, 
 former extensions of the sea to some inland point ; and, as this 
 peculiar sandstone formation ends at the beginning of the bulge, 
 on the eastern side, the whole forms an even plateau from Blanc 
 Sablon to this point. A formation of more or less stratified lime- 
 stone then begins, whose eastern terminus I have not yet ascertained, 
 though it is said to extend under the sea and appear again like 
 the sandstone at L'Anse Claire, on the opposite Newfoundland shore. 
 The whole thus presents an interesting field for future investigators, 
 as it forms an additional link in the chain of points of interest, 
 whose combination forms a culmination of great geological import- 
 ance in reading the former history of the region. It was quite 
 fortunate that at about five o'clock in the evening we were so 
 near our journey's end, for it would have been an easy matter 
 u 
 
 m 
 
 .y 
 
nm 
 
 - ''■ ■ 
 
 W ' 
 
 !l! ii 
 
 lli !l!i|!;: 
 
 I ? 
 
 210 
 
 L'ANSE AMOUR. 
 
 to have tracked us across the bay ( Forteau Bay I mean ) on 
 the ice by the blood alone from the dogs' feet. We nad six dogs in 
 our team, three full grown and three young ones. As there were not 
 enough shoes for all of them even wlien we started, and as we had 
 not been able to procure a sealskin or any canvas to make others 
 to replace those already worn out, by this time most of the dogs 
 had one or two shoes only, while one had none. We had driven 
 him without any all this distance, pearly twenty miles, because hav- 
 ing a full set of four nice shoes, he had turned to and pulled them 
 off with his teeth ai:d eaten them up at our last stopping place, and 
 his punishment was tying his mouth up tightly and driving him as I 
 have said. This fellow bled from all his feet, and nearly all the others 
 showed drops of blood in each footprint ; but there was no help for 
 it, and so we kept on as well as possible until our destination was 
 reached. 
 
 Forteau Bay is the first bay of any importance on the Labrador 
 peninsula, and though the principal port on this part of the v-oast 
 is Red Bay, twenty-two or twenty-three miles farther east, the 
 former is not without its objects of general interest. Each side 
 of the bay is lined with buildings, — houses, shops, and fishing 
 store-houses — to the number of some twenty or thirty. On the 
 eastern side of the bay a neat little church, well built and painted 
 white, rears a small but homely looking, short, square steeple with a 
 small spire upon it against the dark background of the ridge 
 beyond ; by its side stands a tastily built parsonage, silently awaiting 
 occupancy by some quiet and practical Christian people whose 
 "charity shall begin at" this their village "home," and end there also, 
 which, wisely administered, "shall cover" for the people "a multitude 
 of sins." And truly the people here need such a family, for they 
 are sadly deficient in intellectual ideas, neighborly charity in little 
 things, and peisonal virtues, though there are many exceptionally 
 fine families among these plain, rough, good-hearted people. 
 
 It was just about dusk when we reached our destination on the 
 extremity of L'Anse Amour, and being Saturday night we looked 
 ior a quiet rest on the following Sunday, both for ourselves and 
 
A SUNDAY OF REST. 
 
 211 
 
 our sore-footed clogs. It was a nice old-fashioned couple that met 
 us here, and, sending the boys to take care of our team and the 
 articles upon it, they invited us into the house, where, after cordial 
 greetings ( for the guide was well acquainted here ) they placed 
 before us, for refreshment, cups of hot tea and buttered graham 
 bread — the first I had seen on the coast. The large house, with its 
 low studded ceiling and ample apartments, presented a cosey and 
 homelike appearance. The kitchen served as a dining-room and 
 place of general assembly, and there we all met together, — those 
 who smoked did so, and those of us who did not made ourselves 
 comfortable and looked on, while we all chatted pleasantly together 
 until supper time ; this over, we spent the evening in chatting 
 pleasantly and familiarly with our host and hostess. 
 
 Sunday, March 6. Two pleasures fill me to-day, the oue that 
 I am here in a new place with the beauty and freshness of new 
 scenes before me, the open bay in front, the sea with a distant 
 view of the Newfoundland coast at the left, and a huge bank of 
 limestone containing unknown treasures in the shape of new and 
 rare fossils behind the house and not a hundred rods away : the 
 other the prospect of a good, quiet, homelike Sunday with 
 pleasant people who greet me most cordially, upon my appear- 
 ing at the breakfast table. The simple but relishable meal 
 over, the morning was spent in talking, walking, and in reading. 
 The dinner disposed of — an equally satisfactory repast — all hands 
 were invited to take a walk to the limestone cliff and see its attrac- 
 tions. This we did, and were soon examining this mound ridge 
 of sea-deposit which appears to stretch a long distance in a north- 
 westerly direction, though just how far I could not tell. The edges 
 of the cliffs, here some seventy- five feet high, were exposed, and 
 presented a most uneven, and irregular jagged surface, facing the 
 sea ; below, the broken pieces that had tumbled down by weather, 
 wind, frosts, rain, and their own weight, strewed the ground every- 
 where in a sloping pile of fine chips — like slate rock — and stones 
 of various sizes with occasional huge bowlder-like lumps of solid 
 rock. I picked up several well formed fossils, and accepted our 
 
212 
 
 POINT AMOUR. 
 
 i. 
 
 '' 'I 
 'J I, 
 
 
 
 ll 
 
 llltl 
 
 host's kind invitation to spend a week or more with him in the 
 spring, when the snow and ice should have disappeared, in examin- 
 ing the place with reference to its geology and for specimens. 
 
 A rough climb brought us to the top of the ridge, and gave us a 
 fine view of the surrounding sea and country. We had been 
 followed by eight or ten of our dogs, and it was an amusing sight 
 to watch their efforts to follow us in our climb, and their repeated 
 tumbles and consequent slides down the icy slope some fifty feet to 
 the bottom of the descent. At length we are all safely landed, and 
 a brisk walk in the strong wind and rather cold, bracing air brought 
 us to the road. On our right was the lighthouse, but we did not 
 visit it. It is situated on the extremity of the point, and we learn 
 from the reports that it is in latitude 51° 27' 35", longitude 56° 
 50' 55". It is a circular lighthouse, height 155 feet, built in 
 1855. The light is of the second order, and visible eighteen miles ; 
 it is fixed and whJte. A whistle sounds every ten seconds in 
 foggy weather. This and the light on Greenley Island are the only 
 two lights on the coast ; no doubt another will soon be built, perhaps 
 near the rocks upon which the Edward Cardwell was so lately 
 wrecked. There at least ought to be one near that point of the 
 coast, though Beacon or Old Fort Island would seem to be the 
 best place for it. Our walk home was soon accomplished ; the after- 
 noon and evening passed away quietly, — a sermon read by one of 
 our number occupied the latter part of the evening, and we all re- 
 tired to rest early. 
 
 Monday morning we left these kind-hearted people, with many 
 expressions of cordiality on both sides, and were soon travelling 
 towards home again, as this was the end of our journey. It was a 
 bright and clear, cool morning, and the dogs spun merrily along, well 
 shod and apparently quite over their lameness and sore feet. We re- 
 turned by the same road as that by which we had come, as far as 
 L'Anse Claire, and then, veering to the left, followed the coast to 
 the strangely located abode of another hospitable and well known 
 family, at the foot of the central point of the sandstone cliff here 
 situated midway between L'Anse Claire and Blanc Sablon j this is 
 
If 
 
 RETURNING HOxME. 
 
 213 
 
 called L'Anse Coteau, and occupies a little niche in the coast just 
 east of the extreme eastern end of Wood Island, which is seen just 
 beyond the point. The ride around this cliff presented somewhat 
 the appearance of our walk around the limestone of L'Anse Amour. 
 The sandstone, everywhere broken, showed us plainly the layers or 
 Virata, and its various colors of brown, red, and gray. We stayed 
 here over night'and continued our journey the next day. Returning 
 by Bradore bay, which was now quite frozen over and safe to travel 
 upon, I could now see the appearance of the granite at this point. 
 Several huge and irregular dikes, apparently of trap, marked the 
 surface of some of the rocks. Most striking irregularities of break- 
 age, in some cases amounting almost to cleavage, appeared in the 
 clift's and in the pieces broken off and tumbled to its foot. Some 
 of the bowlders were cleft by frost and weather, while ever^'^vhere 
 evidence of most interesting structure prevailed ; but I had too 
 little time to spend here, so continued my homeward journey. The 
 rest of the return trip was pleasant and delightful ; the air cool yet 
 comfortable, with a clear sky. The dogs, though wear)', seemed to 
 know that they were going home and trotted along quite rapidly. 
 Soon darkness began to close in upon us, and in a short time 
 hearty home welcomes and a nice hot supper were the entertain- 
 ments of the evening. " Early to bed and early to rise," says the 
 old proverb ; we tried the former part of the saying and soon 
 drowned our cares in peaceful slumber. 
 
 I'll 
 
 ji/VVOi 
 
ih! 
 
 H 
 
 
 
 214 
 
 CANADIAN PORCUPINE. 
 
 CHAFfER XIII. 
 
 Canadian porcupine — Picking fall berries in spring — Carrying wood to 
 summer quarters — Anticipating Fourth of July — Summer quarters in 
 winter — Capsized — Fox hunt on rackets — A mile of soft snow without 
 rackets. 
 
 Thursday, March lo, I spent the day in skinning and stuffing 
 a porcupine, that was brought in to me by one of th ; neighbors, 
 and a most difficult task I found it. The Canadian porcupine 
 i^Ercthizon dorsatuni) is quite common all along the coast, and 
 in the interior of Labrador. It is an animal more or less peculiar 
 to this region, and is generally found in winter by following, on the 
 newly fallen snow, the impressions of its feet which look much like 
 the imprints of a small child's foot. It is found more frequently 
 in winter than in summer, when it conies out from its hiding places, 
 which are in caves, under rocks, and often in the hollows of trees, 
 and travels about for food. It lives on berries and the bark of 
 trees, gnawing them in such a manner that the hunter knows at 
 once when one of these animals is about, after having found its 
 cuttings. The animal returns to the same place each day to feed 
 until its supply ceases, when it seeks a new place. Though naturally 
 a sluggish animal, the porcupine, when pursued, takes at once to 
 the tree-tops and is then very spry and agile. When caught upon 
 the ground it immediately rolls itself together into a sort of ball, 
 and with spines erected meets the intruder. Only the belly, and 
 extremities of the animal are destitute of these spines so that it can 
 be safely attacked only on these parts. The Indians kill it with 
 a blow from some stick which they carry for the purpose, or, if 
 
TRIP TO OUR SUMMER HOUSE. 
 
 91S 
 
 or. 
 
 if 
 
 it ascends a tree, they cut the tree down and then soon dispatch 
 the animal. It is a forniidaljle object, when thus on the defensive, 
 for other animals to attack, as to touch it detaches many of the 
 spines which work their way into the flesh causing terrible wounds. 
 It occasionally utters a sort of plaintive cry, and this, with many other 
 circumstances, has won for it, by the whites along the coast, the 
 name of " Indian papoose. " The Indians regard its flesh very 
 highly. One can easily see that it is no small job to skin one of 
 these animals, and that the quills, liable to be lost, afterwards 
 cause serious trouble should they get into the foot or hand. The 
 Indians when they wish to use the quills for fancy work, as they 
 often do, carefully pick them out and arrange them to form 
 mats, or various ornaments of one kind or another; when the 
 flesh is to be eaten they are generally hung up over a flaming fire 
 and singed to the skin, thus burning off all the chaff of the (juills. 
 The young are covered with long, black silky hair with whitish ends. 
 
 Friday the nth. This afternoon the snow, melted by the hot 
 sun, having i)artially disappeared from some of tb.c elevations, 
 we picked a dish of nice, fresh berries for supper. Here the snow 
 covers the red berries and keeps them in a perfectly fresh condi- 
 tion until the ensuing spring when they are picked and eaten in 
 large numbers. They taste very fresh and nice. 
 
 Saturday, March 26. This day, looking to be a fine one, with wind 
 south and a mild thaw, seemed to invite a tramp ; so accordingly, 
 as the men were to carry out a load of wood to the summer house 
 on the island — only about four miles distant — I determined to take 
 my gun and follow them ; the report that a large, white owl had 
 been seen there, helping materially to decide me in the matter. 
 The men started off quite early, directly after breakfast, for the 
 woods, with their dogs and komatiks, to get their load. If the going 
 is bad, or the dogs are obliged to wear their sealskin or canvas, pocket- 
 like shoes, only a light load is put upon the komatik, but if the ice 
 is smooth, and the going good, it is an old saying that you can- 
 not "load a komatik ;" that is, the sled slips along so easily when once 
 it is started that you can pile on all the weight that it will hold, and 
 

 216 
 
 BAD TRAVKLLING. 
 
 w 
 
 ¥ 
 
 
 '\ 
 
 the (logs will draw it with case. 1 have known a team of four good 
 dogs to draw a load of nearly a thousand pounds of iron for eight 
 miles, and the roads not in the best of conditions. 
 
 While the men were piling the sleds with wood — we had two 
 teams — to carry across to the summer house, I put on my warm 
 clothing and prepared to accompany them on foot. It was a clear, 
 beautiful morning, and after walking some little distance the bracing 
 air seemed to give me renewed spirits ; but soon a misfortune quite 
 dampened them again, and rendered me unfit both in body and 
 mind to enjoy fully the sport of the day, or rather the sport that 
 might have been during the day. This was the loosening of the 
 stopper of my powder tlask, s j that before I could stop it, the horn 
 had swung behind me, mouth downward, and the powder, to the 
 amount of about half a pound, was instantly a long, black, snake- 
 like train on the snow beneath. Strange to say, my first impulse 
 was (thinking that the snow would soon wet it, and that it was im- 
 possible to collect any again) to utilize the waste ; so imagining that 
 it was Fourth of July, or Washington's birthday, I cannot remember 
 which, I hastily drew a match and touched the train. With a whiz 
 it exploded and a long burnt track alone remained to tell the story. 
 I had then time to think that but a single load remained in my gun ; 
 and that the sleds were fast overtaking me. 
 
 It was very nearly as far home as forward to the island, and the 
 thought that a single load will often do the work of a dozen 
 impelled me fonvard rather than backward. By this time the sleds 
 had arrived, and together we went forward at a medium pace, a 
 fast walk, over the not very good ice toward the island. When I 
 say not very good ice, I mean bad for the dogs' feet and the easy 
 running of the komatik, yet not thin. We did not come to thin ice 
 until the point of the island opposite us was reached, and here we 
 came upon a narrow platform of thin ice. with the water close to 
 us on each side, over which we must pass. The water rippled up 
 to its edge in a treacherous manner, and the surface looked dark, 
 thin and wet. Thinking that we could clear it, with a shout the 
 driver urged forward his dogs, and with a rush we made for the 
 
 niiii^ 
 
^ 
 
 SUMMKR (XJAUTKUS IN WINTKR. 
 
 L'17 
 
 opposite side, but crack, crack, crack ; the dogs stood still and 
 looked to see what was the matter, only for a moment, but that 
 moment was enough ; the ice gave way and the koinatik and 
 its load of wood, on top of which were several jackets, and my 
 powder horn and gun, slowly sank into the water. To spring from 
 cake to cake to the firm ice. and to reach down and seize the above 
 mentioned articles as they were slowly disapi^earing, and before 
 they had reached the water, was the work of a moment ; but the 
 precious articles were saved. 
 
 Then came the work of picking out the sticks and reloading the 
 sled. Fortunately wood floats, and the sled had not been l)roken or 
 otherwise damaged (they are made very stout), and soon we were 
 again on the move. Luckily, nothing was lost, while both myself and 
 iriend, the driver, escaped most miraculously, not having even wet 
 our feet. Arriving at our stopping i)lace, we left the wood at its 
 destination, and went up to the house. 
 
 How changed the i)lace looked from what I had seen it in the sum- 
 mer ! It was then fertile and flowering ; it was now snow, snow, snow 
 everywhere. I took my gun, with its only charge, put in on leav- 
 ing home very luckily, and started for the hillock on which the white 
 owl had last been seen ; but, alas ! he was not there. I climbed to the 
 top of the crest, but could see nothing ; on the crest beyond, still 
 nothing of the owl ; but on the third crest I espied a fox trotting 
 off at a slow gait up the ridge before me, not two hundred yards 
 away. With my telescope, I could see that it was a red fox, and 
 that he held up his left front foot, using only three feet to walk upon, 
 as if lame from being caught in some trap or otherwise, so I deter- 
 mined to chase him ; no quicker said than done. 'I'he fox was about 
 three hundred yards away from me when the chase began. I saw 
 him through the glass for a moment as he trotted along the 
 snow ahead of me, and then taking up my gun, started after him ; 
 he was just behind the ridge, and I had hoped to gain upon him, 
 but a fox is a cute animal if he has but three legs to run on, and 
 by the time I had gained the ridge, he had heard me coming and 
 was now far to the right and still out of gunshot. On I pressed, 
 14* 
 
 

 a, 
 
 it 
 111 
 1« 
 
 
 i I 
 
 Mil ' 
 
 II , i 
 
 il' 
 
 
 1 
 
 1- 
 
 - 
 
 ll 
 
 1 
 
 i 
 
 
 I 
 
 il' 
 
 '■'^5 
 
 218 
 
 A TRICKY FOX. 
 
 occasionally losing sight of the game altogether and now having 
 the stimulus of seeing him before me, either tracking the snow with 
 blood from the wounded foot, or slowly climbing the hillock in 
 front. Sometimes he would make a turn to the right, expecting to 
 throw me off the scent, and often to the left for the same purpose ; 
 but the island was small, and the ridges so far distant, compar- 
 atively, that it was impossible, now at least, for him to gain his 
 object. On we went over hill and dell, sometimes one gaining and 
 sometimes the other. At last the end of the island was reached, 
 and the fox was running across a narrow pass, covered with ice, 
 between this and another island ; I had gained on him and he was 
 now less than two hundred yards away, so I determined to fire. 
 Hastily kneeling behind the crest of a hillock, and resting on a 
 stone there, I fired ; the ball reached the fox but did not hit him ; 
 and, though I had already run several miics across the island, and 
 had shot away my only charge, I started in pursuit again, across 
 the pass and away over the next small island. Here comes the 
 strange part of the chase ; the footing led plainly across the snow 
 for a short distance, and then, mingling with another recent one, 
 diverged ; but which was which I could not tell. 
 
 The island was hilly though small ; upon gaining the summit of the 
 crest the fox had disappeared. Following the slope downward to- 
 wards a small point of land to the westward, I soon came across 
 the bloody track of the fox again ; it disappeared when I reached 
 the front, then appeared again along the edge of the rocks quite 
 near the ice of the bay and below the rocks. This was a cute 
 trick to escape detection, but the next step was more so. The 
 tracks disappeared all of a sudden in the centre of a patch of snow 
 with moss around it, and not a trace of them could be found 
 anywhere, I carefully searched the narrow point over, and looked 
 behind each rock, ascended each crest, examined each old fox 
 footing, partially filled with drifting snow, so that I could not have 
 mistaken it, and not a sign appeared. The fine fellow had played 
 me another of the many tricks for which Sir Reynard is noted, and 
 left me, four miles from home, heated with running, cross and 
 
FOX HUNTING ON RACKETS. 
 
 219 
 
 hungry. Vowing vengeance, home I started, awaiting the next 
 pleasant day for anothei hunt. 
 
 The next day, though more or less snowy and otherwise uncom- 
 fortable, we went out to the island with another load of wood. I 
 took my gun, hoping to see some signs of game, and started a little 
 ahead of the loaded sleds which soon caught up to me, when we 
 proceeded along together. It seemed as if we never went so slowly, 
 and though the loads were not large, the previous day's thaw ren- 
 dered it treacherous travelling. We plodded on cautiously, beguiling 
 the way with conversation until within a short distance of the island, 
 when sharp eyes detected, on the slope of a low hillock ( of which 
 there were several), among the stones and rocks, a small moving 
 object. A few steps nearer decided the question of what it was, and 
 no doubt remained in my mind but that my friend the fox was mak- 
 ing another tour of the island, perhaps searching for me for the fun 
 of another chase. Several crows, or rather ravens, hovered occasion- 
 ally over him, and vented their spite in repeated dashes and dabs at 
 his head, wishing, no doubt, that they could enjoy a good meal from 
 his carcass, or refresh themselves with his eyes ; but no such fortune 
 for them occurred, and we saw him slowly disappear over the ridge, 
 trotting in the direction of the komatik and ourselves, whom he 
 evidently had not yet seen. Seizing my gun, which was lying on 
 top of one of the loads, I started on the run to head him off and 
 try and get a shot. 
 
 It was hard work running on such snow, and I sank to the knees 
 at nearly every step ; but passing over the ridge I saw the fox 
 climbing the opposite sloj^e ahead of me. He had turned, appar- 
 ently, and started off in a new direction ; so I started off after him. 
 Soon he disappeared over this second ridge, and when half-way up 
 the bank, chancing to turn around, I saw a fox just crossing the 
 path by which I had come. P'or a moment, and only a moment, I 
 hesitated which one to follow — then, turning, I sta'-'^ea for the 
 ne" it and one last seen. The fox used only three feet (the other 
 bleeding from some trap wound) an^I did not riin fast ; yet I was 
 obliged to do my best to keep up with and not lose sight oi him. 
 At last I gained perceptibly. I could see h- j I Jack lail and stern — 
 
mm 
 
 
 f ' 
 
 IP I 
 
 i .1 • 
 
 ! 
 
 r 
 
 
 220 
 
 A WINTER TRAMP WITHOUT RACKETS. 
 
 it was probably a cross or patch fox — and soon, making for a huge 
 rock in the centre of an ahnost level plain, I had a chance, long 
 shot. The ball sped swiftly along and missed ; but it struck so 
 near as to bewilder the animal, who, for a moment, seemed unde- 
 cided which way to turn, but quickly disappeared over the ridge. 
 I followed him until too tired to run farther, and was obliged finally 
 to give up the chase and returned to my companions quite crest- 
 fallen, yet ready for another trial. 
 
 Sunday, the 2 7th. We are alone, the family having gone to visit a 
 neighbor about eight miles away, and it does seem a relief to be 
 quiet — to hear no noisy children, and no rough, coarse scoldings 
 or threats of violence to them in case of disobedience. The day 
 was pleasant, and we all enjoyed it ; it aeemed more like Sunday 
 than any similar days we had passed, to me at least, while on this 
 part of the coast. The evening came before we really knew it. 
 
 Monday the 2 8th was a damp, snowy,"dismal day : we all remained 
 in the house and amused ourselves as l)est we could in reading, 
 and writing, or netting nets and winding twine. Towards evening 
 several of the neighbors, whom we knew well, came in to chat with 
 us, and thus this day passed like many others. 
 
 Wednesday the 30th, To-day I took a tramp with two or three 
 others, inland over the ponds : but we carried no rackets and 
 found the walking terrible. In many places we sank to our arm- 
 pits, and we were obliged to progress Indian fashion, that is, crawl 
 on the hands and knees, using a long, round stick, with which to press 
 upon the snow while extricating ourselves from the drifts. It was a 
 most tedious and difficult work. We would walk on the treacher- 
 ous crust for a few steps, then sink in with one foot up to the knee, 
 recovering our position only to fall in with the other foot and again 
 to sink nearly to the armpits. The stick then kept us from sinking 
 farther — the snow often being fifteen feet deep beneath us, — and 
 we would slowly and with difficulty crawl out and onward. Thus 
 we soon were obliged to turn back, and though having gone scarcely 
 a mile we were completely tired out when we reached home. I 
 shot several species of birds, however, which i)aid well for all my 
 trouble. 
 
!li 
 
 WORK AT SUMMER QUARTERS. 
 
 221 
 
 CHAPTER XIV. 
 
 Preparing the summer house to live in — Moving out — A spring rescue — 
 Seals on ice — Larks — A home scene — Spring duck shooting — Re- 
 pairing the boats — Visit to th.^ Indians — Indian canoes — Netting nets 
 — Labrador mail — Natural scenery of Labrador — Repairing canoes — 
 Visit to Eskimo graves — Ornithological notes. 
 
 
 During the early part of April the weather still continued cold and 
 disagreeable, while it snowed more or less the whole time. About 
 the middle of the month we could spend the greater part of the 
 day at the island in refurnishing our cabin, and soon had the roof 
 finished, the inside ceiled, and the partitions put in their proper 
 place, while new flooring was laid down up stairs. While this was 
 in progress, the boys cleared up around the house outside as much 
 as possible considering the amount of work in progress, and a 
 couple of good, stout men wielded a large saw which converted 
 several huge logs, from a large pile of similar pieces a litde to one 
 ;jide of the house, into boards for the use of the carpenters inside. 
 A!i hands worked hard, and one could see the improvement from 
 day to day, as affairs progressed and drew to completion. Towards 
 tiie t nd of the month the sun got high enough to allow us to move 
 out r.:gain, and the almost new house furnished very comfortable 
 nu ter.. There is one peculiarity of the Labrador climate that I 
 must mention. There is, strictly speakmg, no spring in Lal)rador, 
 and though the i-^e does not begin to (lisap})ear much before the 
 end of May — while large masses often remain until July, and ice- 
 bergs float through the straits even in August — it goes very sud- 
 denly when once it really begins. In si)ite of all these drawbacks 
 there is a certain mildness in the atmosphere that takes the place, 
 
(I 
 
 222 
 
 MOVING OUT. 
 
 apparently, of spring proper, and accordingly it thus becomes pos- 
 sible to move out of " winter quarters," and into the summer house 
 in its more exposed situation, from the middle to the last of April ; 
 strange as it may appear, the time set generally falls on the week 
 after Easter. 
 
 The process of moving out is much like that of moving in, 
 though in the latter case you go by boat, and in the former by sled ; 
 as the baggage consists principally of p'-ovisions and boxes there is 
 not much to move. Of course in this region trunks are seldom 
 used, being replaced by the sailor's chest or carpenter's box which 
 holds each individual's goods. It was a glad day when we turned 
 our backs upon the he sv. we had occupied throughout the long, 
 tedious winter, and fina! ; jd, with our "goods and chattels," 
 
 at our summer house. It v, a beautiful morning. The air was 
 keen and sharp with the freshness of the still lingering winter, 
 though the absence of that intense cold, so usual at this time, was 
 marked as quite a pleasant change. The sky was clear, and towards 
 noon the sun came out warm and nice. The hills were still more 
 or less covered with snow, yet here and there a suggestion of a 
 much greener foliage than could have been seen a month earlier 
 even had made its appearance. The greatest change, however, 
 occurred in the inroads which the waters of the sea, owing to the 
 previous mild weather, had made upon the ice of the bay. From 
 a point of the mainland we could see that half the pass between 
 the two islands was already open, and far onward and outward we 
 could view an open sea. I say open sea ; yet though the ice had 
 broken up, leaving the water visible, there were still large masses 
 of drift, or as it is called shee-shec ice floating up or down with the 
 current or drifting about at the mercy of wind and tide, and no 
 boat could yet be safe far outside. As we stood viewing the waters, 
 with their distant long lines of drift ice, which resembled the 
 pictures often seen of arctic winter, we would occasionally see 
 short, round bodies upon it, that upon inspection proved to be 
 seals. We could with our glass often see these animals sporting in 
 die open waters of some lagoon-shaped mass of ice, and occasion- 
 
"SKY" LARKS. 
 
 22d 
 
 ally notice both old and young seal together on the blocks. We 
 could only look at them, however, since the ice prevented us from 
 launching the boats, and they were too far off to be reached by 
 either rifle or shotgun. The distance appeared to be about two 
 miles and the inhabitants considered it as about such. When 
 tired of watching the seals, we turned our glass to the loom of 
 the Newfoundland coast that appeared in the distance like the 
 rising cloud of a heavy bank of autumnal Newfoundland fog. 
 Then again, nearer home, the open water in the bay reminded us 
 that soon the ice would disappear and the boats could be launched. 
 I was much impressed with the beautiful song of the larks 
 (^EremopJiila cornuta) , which everywhere greeted us. Sometimes 
 they would send forth distant trills as they flew past the island high 
 over our heads, while at others they would sing the most beautiful 
 carols as they rose in spirals high into the air ; at still other times 
 they would hop about the dooryardfree from distrust while they fed 
 upon the scraps scattered for them, for which they had chirped a few 
 notes of thankfulness before finally taking wing. About this time 
 flocks of snow buntings filled the air, and lit upon the snowy slopes 
 of the crests and knolls about the island, hotly chased by the boys 
 who kill them with stones, or catch them in snares. These birds 
 are generally time and unusually fat, which renders them delicious 
 eating. The yard was now full of busy men ; the doorstep of happy 
 children ; and the house itself with the industrious housewives — all 
 intent upon putting things in order for present comfort and future 
 summer use. Meanwhile the dogs and little puppies enjoyed them- 
 selves romping over the island, while I remembered the two pussies 
 as they stood in the doorway, or rubbed about the doorsash, or 
 stood their ground against some venturesome dog who mildly poked 
 his nose into and inquired who was within the door. The whole 
 scene is painted on my memory as one of delight, as the ap- 
 proaching season wafted its breezes to a house and yard full of 
 delighted beings who had been cooped up for six long months in 
 the inmost recesses of a most piratical looking harbor where no 
 one would have suspected that such a thing as a house ever could 
 
KiT 
 
 224 
 
 SPRING DUCK SHOOTIW;. 
 
 M 
 
 W 
 
 fM 
 
 III illlila; 
 
 Hii 
 
 exist. Soon the ducks began to fly, and then such sport as we had. 
 The king eider came first, then the common eider ; the former is 
 called the passing, the latter the laying, duck. The birds at first 
 fly in large flocks, often thousands in a flock, and generally the dif- 
 ferent species do not mingle. They have a certain course which 
 they pursue, and the shoals over which they fly are called the 
 J "gunning points." Here the men and boys congregate, and, lying 
 low, behind some rock or cake of ice, await the flight. Some days 
 the birds fly thickly, others rarely any pass. The people see them 
 at a great distance, and often hear the beating of their wings before 
 they see them. The birds fly over or along the side of the station, 
 and the minute the head of the flock has passed the first or head 
 . gunner, he rises or turns and fires when all hands do likewise, and 
 the slaughter begins. Often twenty or thirty birds are thus knocked 
 down by a party of two or three persons with double barrel guns. 
 
 My first spring ducking was in the afternoon of Tuesday, the 1 2th 
 of April, when several of u j drew one of the small, flat boats over 
 the ice to the clear water beyond, and, launching it, started for the 
 gunning point. We brought home a good bagful of birds that 
 night, and you may be sure that they were well served, and well 
 disposed of the next day. 
 
 After a few days' work the house was prepared both outside and 
 inside for our comfortable abode. A coat of blubber — or the 
 remains of the livers of the codfish, after the oil has been boiled 
 and tried out of them — as a final touch was put on to the roof 
 with an old broom, and the whole declared to be water-tight, as a 
 hard rain soon proved it to be. We were now comfortably settled, 
 and viewed with satisfaction the progress made each day in the de- 
 struction of the bay ice, and the lessening of the snow on the hills 
 beyond and the ground about the island, by the rays of the sun 
 which gained strength each day as the season advanced. 
 
 The men now turned their attention to the boats, and began 
 repairing them for immediate use when the ice should break up 
 enough to enable them to be launched. 
 
 Monday the i8th. I visited the Indians this afternoon, and 
 
 ii 'i 
 
 1 ( 
 
BIRCH-BARK CANOES. 
 
 225 
 
 completed a bargain for tlie purchase of an Indian canoe. These 
 Indian or birch-bark canoes are getting to be more and more vaki- 
 able all along the coast. Formerly the bark grew in abundance in 
 certain localities on the Labrador shores, but the call for it lately 
 has been so great, since both boats and huts are made of it, that 
 the home supply has become exhausted, and the Indians now send 
 for it to Anticosti, and even Gaspe on the south shore of the St. 
 Lawrence, whence the greater part is now obtained. The Betula 
 papyracea, the tree from which this valuable product is obtained 
 has, strange to say, in these regions at least, become nearly extinct. 
 Of course, the stripping of a tree of all its bark or outer covering 
 at once kills it. The great danger is, that soon the probably already 
 limited quantity south of the St. Lawrence will also become ex- 
 hausted, when, unless considerable growths are found upon the 
 island of Anticosti, where the Indians were recently engaged in 
 searching for it, the supply will become entirely exhausted. At 
 present the Indians, and those whites who are so fortunate as to 
 have bark on hand, are very sparing of it, and will sell only for 
 cash to those parties with whom they regularly trade. 
 
 An Indian canoe is apparently (to use an American slang 
 phrase) a most "cranky" affair. It is light, weighing according to 
 its size from seventy-five to two hundred pounds. The ordinary 
 canoe is about twelve to fifteen feet long, and two feet and six inches 
 wide in the centre, its widest part, while the depth is about three 
 fifths of its width. From the middle both ends taper, cigar-shaped 
 to the bow and stern. Each end is slightly elevated and pointed ; 
 but it is needless to describe further the shape of an Indian canoe. 
 Within this apparently frail craft the natives go from place to place 
 — of course seldom venturing far out to sea — with the utmost 
 freedom. I have seen them rocked about near shore in the surf 
 when it seemed as if the waves would overpower them at any 
 moment ; and then again I have seen a canoe with a single individ- 
 ual paddling as easily and regularly through a narrow pass against 
 high waves that had appearcMl too tlaiigerous {.,r i,i,.,i\ of \\\ <.!.i( : 
 boatsmen with their wooden boats. In th.Coc small canoes, that 
 15 
 
1 i 
 
 226 
 
 SEAL NETS. 
 
 i 
 
 
 appear to toss and rock about so frightfully to one not acquainted 
 with their use, whole families of Indians, often comprising three men, 
 two women, and several small children, will go from island to island 
 or harbor to harbor miles apart, in most squally weather ; when 
 thus moving about they carry their luggage with them. What, then, 
 would be the surprise to see, as is jften the case, a canoe literally 
 and absolutely loaded to within three to five inches of the top with 
 a family moving, in fine weather, to some location miles farther up 
 or down the coast. In a number of years there has come but a 
 single instance, that I can recall, when a load of this kind has been 
 swamped and any of the family have perished. 
 
 It, of course, takes some time for any one to become an adept 
 in the use of these canoes, and though I finally purchased a good 
 second-hand one, I made little use of it save to paddle about 
 around the islands where we were staying, and over to the nearest 
 island beyond, where neighbors resided, a distance of perhaps a 
 quarter of a mile from point to point ; but it was most excellent 
 exercise, and I enjoyed it heartily. The price here of a new canoe 
 that will hold say four people and their baggage, and eighty odd 
 pounds, is somewhere in the region of from ;^7 to ^8 of New- 
 foundland money, or, being four dollars to a pound, some twenty- 
 eight dollars, varying two or three dollars either way. There are 
 generally but one or two Indians in a tribe that can make a first- 
 class canoe, and these have their hands full the greater part of the 
 time. 
 
 In the evening we lashed the canoe that I had purchased to our 
 komatik and started for home. 
 
 The next two or three days were spent in netting a net with 
 which to capture seals. A seal net is an immense affair made of 
 stout salmon twine, and netted in meshes usually about six inches 
 from knot to knot, the best size being apparently twenty-seven 
 meshes wide, with a length of about forty-five fathoms, or about 
 two hundred and sixty feet. The process of netting is one that 
 cannot well be described. The needle that holds the twine is of 
 peculiar make. The meshes are made by doubling the thread over 
 
 ;i' 
 
ARRIVAL OF THE MAILMAN. 
 
 227 
 
 a small card of the required size, while the knot is tied and the 
 whole is drawn tight. When made the knot comes on the ridge of 
 the card and cannot be slipped ; this forms a string of loops into 
 each one of which the needle 's passed while a new series of loops 
 are knotted on the card just below these. In this way the netter 
 proceeds with twine, or needle or card, of any size, and fashions a 
 tightly knotted network of any size he may desire. In this way all 
 the cod seines, nets or traps, are made, while the six inch seal mesh, 
 and the quarter or eighth inch bottom to the dip net or lance seine, 
 undergo the same process. The work, however, is quite easy and 
 most fascinating. 
 
 Wednesday the 20th. Great has been the joy of all hands to-day. 
 The mailman has at last arrived with news from home. Our last 
 letter bore the postmark of Sept. 23, and here it is over seven 
 months since knowing whether those we left at home are dead or 
 alive, or our country prosperous. How eagerly those letters were 
 read, and the papers studied, I will leave you to imagine. It was 
 indeed a joyful day. 
 
 The winter mail for the coast of Labrador arrives, usually, some 
 time the latter end of March, by special carrier. This man starts 
 from Bersimis, which is a small village about one hundred and forty 
 or fifty miles from Quebec, and for which place the mail starts from 
 Quebec on or about the first of February, about the fifteenth of 
 that month. He travels on foot with snowshoes — sometimes by 
 komatik — with the mail upon his back in a stout leathern bag, and 
 makes daily marches of from fifteen to thirty miles, according to 
 the weather and the travelling. Though the distance between 
 Bersimis and Esquimaux Point, the end of the route generally for 
 the first carrier, is only about two hundred and six direct 
 miles, it becomes at least a quarter to a third more by the route 
 taken, since there are bays to go round and rivers to cross ; so 
 that often instead of ten it takes twenty days to accomplish the 
 distance. It is forwarded to Bersimis from Quebec, I believe, by 
 coach, this being the stage limit ; should it be late, however, the 
 time is extended so much. As there are houses every few miles 
 
 

 I 
 
 I 
 
 * 
 
 Vi 
 
 I ii 
 
 228 
 
 LABRADOR MAILS. 
 
 apart all up and down the coast it is an easy matter to drop into 
 one of them for the night, and along the route the mailman is 
 always welcome. Strange to say, there are branch post-offices at 
 stated places along the line, and at these places only is the bag 
 opened, while the carrier takes the letters to the van us people to 
 whom they are directed on his homeward trip ; and even this is 
 done by courtesy, since it is required by law, I believe, that the 
 letters shall only be delivered to those to whom they are addressed 
 by the lawful mailman ; the carrier, being only a paid messenger, 
 has no responsibility but to deliver the bag containing the letters 
 and papers in safe condition to the postmaster at the end of the 
 route, who takes upon himself the responsibility of sending them 
 along the line by the returning carrier — thus making himself liable, 
 while trusting only in the good faith of the carrier, to damage for 
 any losses sustained. At Esquimaux Point, the mail is taken by 
 another man who has travelled up the coast while the first man 
 ' was travelling down. The carrier from Bersimis returns to that 
 place with the return mail, while the one from Bonne B^sperance, 
 who has thus disposed of the up mail on its way to Quebec, returns 
 to the former place with the down mail. This is taken in the usual 
 way, but more frequently by komatik, since the bays are generally 
 frozen over by this time, and travelling upon them is infinitely 
 better than over the deep snow-clad hills which here begin to line 
 the coast. I should have said that the mail first stops at Mingan, a 
 post of the Hudson's Bay company, where one of that company's 
 agents is usually an authorized mail agent also. From Esquimaux 
 Island the mail goes to Natashquan, the next regular office, a dis- 
 tance of about one hundred miles, and from that place, through a 
 tract of country the most difficult yet travelled, especially in bad or 
 mild winter weather. The carrier is often obliged to go over high hills, 
 away inland, to cross creeks or bays that are not yet frozen over, 
 and which are scarcely a mile or two across, and at St. Augustine, 
 much out of the regular route, to cross the deep and irregular cut 
 of Shecatica Bay to Bonne Esperance. The whole distance, thus 
 reckoned, is about two hundred and fifty miles. Thus from 
 
 
 i:' 
 
'; n 
 
 HOW THF.Y ARE CARRIED. 
 
 229 
 
 Bersimis to Bonne Esperance, a distance of about seven hundred 
 miles, the mail is ta.ken by the carrier who tramps the distance gen- 
 erally with snowshoes or rackets at the rate of from fifteen to thirty 
 miles a day, as I have said ; hence, though the mail generally starts 
 from Quebec on the first or second of February, it seldom reaches 
 its terminus before the last of March or the first of April. A long 
 time to wait for one's letters, you will say ; and indeed it is, but you 
 hardly mind it when your news from home does come, especially 
 if it be good news. 
 
 After reaching Bonne Esperance, a special carrier is sent on to 
 Blanc Sablon, eighteen miles farther down the coast, with the 
 remainder of the mail. The peninsula of Labrador proper beyond 
 receives no mail until the lighthouse steamer, a small vessel which 
 is sent yearly by the Canadian government at Quebec to supply the 
 lighthouses along the coast with provisions, oil for the lights, and 
 coal, brings a second mail, in the early spring, as soon generally as 
 open water will allow passage. In the summer, the mail comes and 
 goes, several times in the course of the season, by the traders who 
 bring it from and return it to Quebec ; while another also arrives 
 from Natashquan, via packet from Gaspe on the south shore, twice 
 a month which is taken up by some authorized trading vessel — 
 most every trader stops at this port — and carried onward; and 
 still another mail comes from the States via Newfoundland, by a 
 mail steamer, employed at one of the fishing stations, that makes 
 bi-weekly trips from St. John's to one or two localities on this part 
 of the coast. In this way the people on the coast of Labrador re- 
 ceive and send their yearly mails. A cable runs from Anticosti to 
 Quebec, by way of the south shore and it is soon hoped that one 
 will be laid from that island to Mingan, at least, the post of the 
 Hudson's Bay Company before alluded to, if not farther down the 
 coast, — but there is yet much to hope for on this score. 
 
 The weather now began to be fine, and the warm sun i- ; lelt 
 away the snow, and turn to green the brown winter-killed leaves of 
 the plants already brought to light, in the damp mosses on the hill 
 tops and occasionally on the plains. The birds were becoming 
 
 MM> 
 
^p 
 
 S80 
 
 ROBINS — SPRING SCENES. 
 
 I 
 
 'V 
 
 1 H 
 
 abundant once more, and apparently sighing for their mates, while 
 all nature began to rejoice. 
 
 I cannot describe the characteristic aspect of a Labrr^dor scene. 
 It is one peculiar to the region itself and must be seen and felt to 
 be appreciated. 
 
 The extensively broken coast line affords you a varied view of 
 now sea and now land, in the shape of some island or promontory 
 of the mainland — the former being more usually the case. High 
 and low hill tops crown the immediate coast line, in the majority 
 of places, while often their somewhat distant outline reminds you 
 much — if you close the view in other directions — of our own 
 eastern United States scenes of hilly outline in the distant horizon. 
 A spring month, to one situated on one of these little islands, is 
 always most delightful. The novelty of busy preparation for the 
 summer season combined with the animation which all nature pre- 
 sents as well as the balmy air and the peculiarities of situation and 
 surrounding objects, cause one to feels a freedom scarcely to be 
 i magined, since care is thrown to the winds and you immediately 
 find something that occupies you pleasantly, almost without know- 
 ing it ; while each day, as it passes, adds to the pleasure felt. 
 
 My joy can easily be imagined when, one day, towards the latter 
 end of the month, one of the men, knowing how I sought for birds 
 and other natural objects, brought me a robin which he had shot 
 a short time previous outside the yard, and not far from the house. 
 It was the first real home bird that had greeted me, and was all the 
 more appreciated for that. The robin does not appear abun- 
 dant in Labrador ; though a few breed here, inland and along the 
 rivers. 
 
 This week I rosined the bottom and sides of my canoe. In this 
 operation, as generally performed by the Indians, the seams of 
 the canoes are covered tightly with spruce gum and rosin, while 
 should any leaks appear, they are covered over with a different 
 preparation usually made by the Indians, thus : a quality of rosin is put 
 into a tin can, and a small portion of oil — seal or whale, usually 
 the former — is added to it, with a very little paint, — white lead — the 
 
 li, ' 
 
"BANQUESE" ICE. 
 
 S81 
 
 whole is then put over the fire and heated. As it melts, oil is con- 
 stantly added, a little at a time, and the mixture stirred with a stick. 
 As soon as the substance will drop from the stick, instead of run- 
 ning freely from the end, the pot is taken off the fire immediately 
 and the leaks in the canoe are thickly coated with it while yet hot. 
 An Indian finds a leak in his canoe most ingeniously. After rosin- 
 ing all the apparent cracks, he goes all over the boat again apply- 
 ing his lips to every suspicious looking place, and, exerting both 
 suction and pressure, soon discovers the places where air is still ad- 
 mitted ; these he rosins, and then his canoe is launched and tested 
 by paddling a short distance or rolling and rocking it violently 
 from side to side. If, after a sufficient time, water appears, the bot- 
 tom is again examined, and redressed with rosin until it is tight. 
 When an Indian has finished with his canoe for the day, he takes it 
 out of the water and turns it bottom side up, and, if possible, throws 
 spruce boughs upon it to keep the sun from melting the seams ; if no 
 covering can be obtained he leaves it bottom down upon the softest 
 place he can find, that the bottom maybe prot cted from scratches. 
 A canoe is always lifted, as so much dead weight both from and 
 into the water. 
 
 One of the peculiarities of a Labrador spring scene is that of 
 drifting or "banquese" ice. This is ice that breaks up in some 
 northern locality, and flows, in greater or smaller masses, through 
 the Straits towards the Gulf and open sea, generally melting by the 
 time it reaches the Magdalene Islands. The "banquese" ice often 
 lies a mile or so off shore, and, to the leaked eye, appears to be cov- 
 ered with logs and sticks, apparently from some wreck. A good 
 glass will show the mistake at once, and you find, with surpirise, 
 that they are so many seals, both old and young. If they are 
 abundant, boats are put off and the seal-hunting begins. 
 
 Tuesday, May 3d. I spent most of the day in examining several 
 Esquimaux graves, but found only a few old bones, and a substance 
 resembling seals' hair in each case within eighteen inches to two feet 
 of the top of the mound. The graves were five in number, but 
 
 'i IR 
 
 , ^.j-.i 
 
«»"■*■*■ 
 
 
 232 
 
 BIRDS. 
 
 
 . 
 
 IV- 
 
 m 
 
 eveiything had disappeared so that nothing of any worth rewarded 
 my work. 
 
 Wednesday the i ith. I started off on a trip down the coast, but 
 not before having secured a specimen of the pintail duck (^Dafila 
 aaita) that one of the men shot while it, with its mate, was feed- 
 ing in a shallow bay of the sea, in front of the house. It is very 
 rare in these parts, and seldom seen here, though not uncommon, I 
 believe, in parts of Newfoundland. 
 
 Thursday the 1 2th. Passed into Blanc Sablon to-day ; the bay, 
 at its entrance, was full of birds of all kinds. I noted several spe- 
 cies of gulls and terns, including several jaegers, or what are often 
 called hagdowns or shearwaters, while ducks were plentiful, and the 
 puffins or paroquets abounded in tens of thousands. Greenley 
 Island, at the entrance of the harbor, has always been a noted place 
 for these birds. They breed here in vast numbers, covering the 
 ground everywhere with burrows. There are now only two or three 
 places on this part of the coast where the puffin breeds in any num- 
 bers, and Greenlc) Island is one of them ; another place is the Para- 
 keet Islands, in Bradore harbor. The bird seems now more 
 abundant at the former place. Though 1 visited the island, and 
 made personal observations there, I find that the ground was so 
 well and thoroughly gone over by Dr. Elliott Coues, some years 
 previous, that I give his description of this species, as covering the 
 ground in so complete and natural a manner that I am sure it can- 
 not fail to interest. Though I have seen and identified nearly all 
 of Dr. Coues' points here given, a few of them are new however. 
 He says : — " The habit of collecting in immense numbers at par- 
 ticular localities during the breeding season, so characteristic of the 
 whole family of Alcidte, is a trait exhibited in the highest degree by 
 the species now under consideration. With scarcely the exception 
 of the common murre, no bird of the family shows so preeminently 
 gregarious a disposition as does the Arctic puffin. Collecting, as it 
 does, in thousands, on particular islands of small extent, it becomes 
 a matter of astonishment that food can be procured in sufficient 
 
PARRAKEET ISLANDS. 
 
 288 
 
 irded 
 
 
 quantity to sustain them, so that each pair can find a place to de- 
 posit its eggs. The pertinacity, too, with which they cling to the 
 immediate vicinity of their breeding place is remarkable. But a 
 very short distance from an island where there are thousands, it is 
 a comparatively uncommon thing to see a puffin. The most ex- 
 tensive of these breeding places appears to be an island near the 
 harbor of Bradore, visited by Audubon in 1833, of which he has 
 written so graphic and instructive an account. The one, however, 
 that I had an opportunity of visiting cannot be much behind it in 
 point of the numbers of the birds breeding on it ; and during a stay 
 of three days I had ample opportunity of examining the island and 
 noting the manners of its curious poi)ulation. My visit was on the 
 25th, 26th, and 27th of July. Let a short extract from my journal 
 descriloe our approach to the island. 
 
 " We were now within less than a mile from the island, towards 
 which all eyes were anxiously turned, and still not a bird met our 
 gaze. But a few minutes more, however, and they commenced to 
 appear, flying around the boat or resting on the water ; all were 
 'parrakeets,' and 'tinkers,' except now and then a solitary ' turre.' 
 They were tamer than I had ever seen birds before, almost flying 
 into our little whale boat ; it was hard to restrain from firing. 
 As we rounded the island close to the shore, they came tumbling 
 out of their holes by hundreds, and, with the thousands we disturbed 
 from the surface of the water, soon made a perfect cloud above and 
 around us, no longer flying in flocks, but forming one dense, con- 
 tinuous mass, and yet not a gun had been fired. 
 
 "The Parrakeet Islands are three in number, lying along the 
 western shore of Blanc Sablon Bay, just at its mouth. The one I 
 visited is the innermost as well as the largest, though the others are 
 equally crammed with the birds. It is about a miio in circumfer- 
 ence ; in shape almost a perfect semicircle, with two points stretch- 
 ing out and inclosing a snug cove, where only can a landing be 
 effected with safety. It is abrupt and precipitous on three sides, the 
 fourth sloping gradually down to the cove. The top is nearly flat, 
 and covered with a rather luxuriant growth of grass, the soil l>eing 
 15* 
 
234 
 
 HOW PARRAKEETS BREED. 
 
 enriched by the innumerable droppings of the birds. The three 
 sides in which the holes are dug are so steep and precipitous that it re- 
 quires considerable agility to scramble along them, the danger of fall- 
 ing into the water being increased by the slipperiness of the soil, 
 worn smooth by innumerable feet, and continually moistened with 
 ordure. The sides are composed of soft, loamy earth, with rocks 
 of every size and shape jutting out in all directions, and afford the 
 most favorable possible conditions for the excavation of the bur- 
 rows. The fourth side between the two points is composed mostly 
 of masses of rock, in the crevices of which the auks chiefly r" ■' osit 
 their eggs ; they very often appropriate the deserted holes the 
 puffins. 
 
 " The holes in the ground in which the puffins deposit their eggs 
 (a habit, as far as I am aware, entirely peculiar to the genus in this 
 family of birds) are excavated by the birds themselves, an operation 
 for which their powerful beaks, and long, strong, and sharp claws 
 admirably adapt them. They extend nearly or quite in an horizon- 
 tal direction, and are semicircular in shape, with the diameter scarcely 
 largr r than is necessary for the free passage of a single bird. They 
 vary much in length, but the majority are not so deep but that the 
 egg may be reached by thrusting in the arm to its full extent. 
 Their course is seldom in a straight direction ; they curve and wind 
 in a most tortuous manner, many burrows being connected to- 
 gether by winding passages. The entrances to the holes are worn flat 
 and smooth by continual paddling from the feet of the birds, and, 
 as well as the whole sides of the island, are moist and slippery with 
 the ordure. The sides of the island, just above high water mark to 
 the very top, are perforated with innumerable holes, but on the top 
 itself not a single burrow is to be seen. 
 
 *' At the further extremity of the hole, which is usually a little en- 
 larged, the single egg is deposited, always a slight bed of dried 
 grasses being first arranged to keep it from the moist earth. I 
 have indeed found eggs lying on the bare ground near the entrance 
 of the burrows, whither they had apparently been dragged by the 
 bird as it hurriedly made its exit ; but in no instance did I find one 
 
A BIRD WITH A BILL. 
 
 235 
 
 in its usual position at the farther extremity, that was not upon a 
 layer of grass. I noticed this fact the more particularly since Au- 
 dubon especially states that no nest whatever is formed for the re- 
 ception of the egg. Without for a moment doubting the accuracy 
 of that great naturalist's observations, the present case is only ad- 
 ditional proof of the extent to which the birds' habits are influenced 
 by circumstances ; the position of nests, the number of eggs, etc., 
 varying much, and the food changing in a measure with every 
 change of locality. The eggs measure two and a half inches in 
 length, by one and three-fourths in greatest diameter, varying very 
 little from this standard ; in shape, which is rather rounded ovate, 
 they differ in being more or less obtuse at the smaller end. The 
 greatest diameter is nearly opposite the middle. The shell is usu- 
 ally more or less granulated, but differs much in the extent of the gi-an- 
 ulation. The color is white or whitish, varying from nearly pure to 
 a brownish hue, the latter color being in the shell, and not caused 
 by soiling or discoloration. They are marked with obsolete, almost 
 imperceptible dots, spots, and lines of light purplish, mostly con- 
 centrated into a ring around the large end. There are sometimes 
 a few irregular splashes of very light yellowish brown. Audubon is 
 clearly in error when he states that they are simply ' pure white.' 
 At that date (July 25) they all, with few exceptions contained 
 young about to be hatched." 
 
 A great trick of the Labradorians is to get a greenhorn to stick 
 his hand into one of the burrows of this bird when the bird is sup- 
 posed to be within. If you carefully examine the bill — of horn, 
 nearly two inches in length and about the same in height — one will 
 see that a most alarming species of forceps may be thus put in mo- 
 tion, and, as the bird is one of the fiercest of its kind, can readily 
 imagine why the victim never repeats the experiment. 
 
 The number of birds that I saw on Greenley Island was simply 
 immense, and could easily have been a multiple, and not a small 
 one, of ten thousand. I have often seen the water covered with a 
 clustered flock, all engaged in making a hoarse, rasping sound, not 
 unlike the filing of a saw ; this is also done both by the " murre," 
 and the " turre," and at such times, which ever species is present. 
 
 / 
 
^mX ';I|M 
 
 « 
 
 m 
 
 
 r 
 
 
 
 
 fil- '■' 
 
 
 i', 
 
 I:;.- > 
 
 236 
 
 A DAY'S " SHOOT.' 
 
 they receive from the sailors the name of " guds," from a fancied 
 resemblance to that sound. When on the wing I seldom if ever 
 saw them mix with other birds. Though they appear in large num- 
 bers, at stated times, they disapperr, or rather disperse after breed- 
 ing, almost as suddenly as they came ; yet stragglers do not leave 
 until the harbors are blocked up with ice. At Greenley Island al- 
 though there is a large fish-curing establishment, houses, and a 
 light-house on the northeast end of the island, the birds occupy 
 the other side unmolested, and are seldom interfered with by gun- 
 ners ; 3'et tlie island is scarcely three-quarters of a mile long in its 
 longest direction, and even less than half a mile wide. Though I 
 have used up much space for this bird, I can but finish Dr. Coues' 
 most interesting description of this strange species of the feathered 
 tribe, especially as it accords so nearly with our own experience. 
 He says ; — 
 
 " Hardly had our boat touched the shore than we leaped out, 
 guns in hand, and at once scattered over the isl^d. As we wan- 
 dered along the sides, the affrighted birds darted past us like arrows, 
 issuing from their burrows beneath our feet and around us, and all 
 making directly for the water. Those already disturbed flew in 
 every direction above us, while thousands rested on the water in a 
 dense mass at a little distance. I took my stand on a flat rock, and 
 in less than an hour a pile of puffins, niore than I could carry, lay 
 at my feet. Shortly after I commenced firing the birds formed 
 themselves into an immense circle, of a diameter of perhaps a third 
 of a mile, one point of which just grazed the island. It was aston- 
 ishing to see with what precision this circle was preserved, each 
 bird flying directly in the wake of the one that preceded. I had 
 merely to stand facing the advancing birds, and no better oppor- 
 tunity for continual slaughter could be desired. I now realized 
 what I had been told, but had found hard to believe, that a wagon 
 might be filled wiih the birds by a tolerably expert workman, shoot- 
 ing them at just such a moment that they should fall into it. The 
 poor things seemed not at all aware of the nature of the danger that 
 threatened them ; flying so close past me that I could almost strike 
 them with my gun. During the continual firing the birds would 
 
 *Wl 
 
nr 
 
 SOME CHARACTERISTICS OF THE PUFFIN. 
 
 287 
 
 y 
 
 emerge from their holes every minute or two ; and after shooting for 
 half an hour on one spot I was not a little surprised to see two or 
 three start out almost from between my feet, and in a great fright 
 make the best of their way down to the water. On emerging from 
 their holes the birds generally looked around for a moment to see 
 what was the matter, and then, in great haste, fluttered and tumbled 
 down to the water below, in which they immediately dove, and, swim- 
 ming swiftly under water, reappeared at some distance. From the 
 countless thousands around me I did not hear the slightest note of 
 any kind ; they flew in perfect silence. 
 
 " The flight of the puflin, when once on the wing, is firm, well sus- 
 tained, very swift, and performed with short, quick, vigorous beats. 
 When it takes wing from a rock, whence it can project itself into 
 the air, it at once supports itself without difficulty ; but when on 
 the water, it is obliged to flap over the surface for several yards be- 
 fore it can rise on wing. When getting untler way, the feet are 
 extended backwards and outwards on each side of the tail, which 
 is spread, but they are soon drawn up, and tlv^* tail closed. AVhen 
 shot at and not touched, like the auks, tliey swerve from their 
 course, open and aliut the tail, and extend the feet. When stand- 
 ing on a rock at the entrance of their burrows, where they alight 
 without the slightest difficulty, they present a peculiar grotesque 
 appearance, such as is afforded by no other bird. 
 
 " When taken in the hand the puffin utters a loud, hoarse, croak- 
 ing scream, at the same time fighting furiously. They are capable 
 of inflicting a very severe wound >vith their powerful bills, easily draw- 
 ing blood. Their long and strong inner claw is also an effective 
 weapon, so that by dint of scratching, biting, and struggling, they 
 proved difficult customers to manage." With regard to this same 
 claw he says : — "I could not but admire the beautiful provision of 
 nature with regard to furnishing this bird witii the means of excava- 
 ting its burrows with facility. The inner claw of each foot is very 
 long, mucii curved, and excessively sharp. To preserve it so, when 
 not in use, it always lies perfectly flat, so that the point does not 
 rest on the ground. In digging and fighting, however, it is held 
 upright, and then becomes a very effective weapon." 
 
mim 
 
 t if. 
 
 I 
 
 k' 
 
 
 Sk 
 
 :1 
 
 i! 
 
 I 
 
 238 
 
 "TURRES," AND "MURRES." 
 
 With regard to the razor-billed auk, the "tinker," or " turre," 
 while on the subject of auks, I will say that I have noticed them 
 breeding on a small island in almost as large colonies as the parra- 
 keets. At the Fox Islands, off Kecarpwei River, they are very abun- 
 dant ; several of us landed, and had about the same sport Dr. Coues 
 mentions as having had with the puffins ; I noticed them in thous- 
 ands about several other small islands also, and am informed by 
 the inhabitants that this species was always very abundant about 
 this locality while much rarer and replaced by the " murre " farther 
 northward. They breed in the crevices of the rocks, long, deep, 
 and narrow ones being preferred. I did not find but a single egg, 
 but was repeatedly told by the people, that if I took the eggs, the 
 bird would lay again the next day. The inhabitants systematically 
 take all the eggs they can find on a given island regularly twice a 
 week throughout the breeding season, and find the birds so wonder- 
 fully accommodating that the last batch taken is nearly as numerous 
 as the first. The " turres " associate both with the " murres " and 
 the black guillemot. The latter bird lays its eggs, smaller, and 
 otherwise distinct, in similar situations and often the two are found 
 breeding side by side, but seldom in any quantities. 
 
 With regard to the foolish guillemot, or " murres," I found them 
 breeding in similar situations and together with the razor-billed 
 auks. The egg is noted for its variable size and the nature of its 
 marking. I have taken them all the way from pt/re white ^ through 
 an endless series of blotches, and waved lines of black, purple, and 
 brown, to an almost pure green, and blue, and even a delicate pink 
 barely spotted or marked at the larger end. The people on the 
 coast cannot tell whether either the turre or murre lays more 
 than a single egg, or whether they sit upon their eggs or allow the 
 sun to hatch them. I have been told, on apparently good authority, 
 that they do sit upon their eggs, and consequently are furnished 
 with a large, bare place upon the lower belly where they have 
 picked the feathers from themselves in order to make the proper 
 hollow in their downy covering for the egg to rest in — but I failed 
 to notice the spot upon any of the birds shot. I could not ascer- 
 tain, either, the period of incubation. 
 
 W 
 
BLANC SABLON IN SPRING. 
 
 239 
 
 CHAPTER XV. 
 
 Blanc Sablon again — Northern limits of the bittern — Return along the coast 
 of Natashquan — Spring scene in Red Bay — Other places — St. Mary 
 islands — Cormorants — At Natashquan — Ramble about the place — 
 Appearance of the birds — The Dark Day — Arrival at Mingan. 
 
 At Blanc Sablon we anchored for the rest of the day and night. 
 I landed, and walked along the beach to examine the general struc- 
 ture of the place, and found the prospect much more pleasant than 
 it had been in the winter. I caught sight, for a moment only, of a 
 small sparrow, which, I distinctly saw had a black cheek and face. 
 The bird was about the size of a chipping sparrow, and yet I dis- 
 tinctly noticed a large black patch either on the cheek or chin, 
 which I could not tell certainly, as he hopped past me into the 
 rushes beyond the sandy beach, and thence flew quickly to the 
 other side of the creek. Here, also, I found a person who in- 
 formed me that, a short time previous, one of the men had shot " in 
 the marshes a bird with a long neck, long bill, and long legs." 
 The American bittern {Botaiirus minor) is probably not rare here, 
 being in fact the only bird of the family which we know as venturing 
 so far north as this region, and undoubtedly this was the bird re- 
 ferred to. 
 
 In the afternoon I took a short walk around the shore. I crossed 
 a shallow streamlet that, flowing through a rather wide bed of sand 
 and small pebbles alternating with patches of muck, of which ma- 
 terial neighboring lands are largely composed, empties into the salt 
 water at the head of the bay. The beach here I found to consist of 
 yellowish white sand. 
 
 Leaving this sandy beach we next came to a beach of rocks and 
 debris from a sandstone ridge close by. After passing several 
 

 
 1'^ 
 
 
 II 
 
 240 
 
 GLACIAL "EVIDENCES." 
 
 ridges, all broken and crumbling on their face towards the sea, we 
 came to a long stretch of mucky meadow, nearly if not quite level 
 with the sea at high tide. Soon these marshy lands gave place to 
 long ridges of glacial-worn, rounded, rocky elevations, which run 
 into the water in the direction of Greenley Island. This singular 
 formation of rock reminds one much of an immense table, of lon- 
 gitudinal and transverse elevations and depressions of several acres 
 in extent, which slowly fall to the sea easterly on both sides. 
 
 Several houses and sheds were perched here and there with 
 apparently little beneath to support them, while the rock itself fur- 
 nished a natural " fish flakes," upon which fish were spread to dry 
 in the sun. The direction of these ridges, as they fell off into the 
 sea, was almost exactly north by east to south by west, and, as they 
 entered the .vater, they were pointing directly for Greenley Jsland, 
 where a grand confusion of broken rocky debris would undoubtedly 
 tell the geologist its history. Everywhere were present rounded 
 tops, and cleft sides of rocks while I was confident I found upon 
 their surface several glacial scratches lying in about the same general 
 direction. The whole mass reminds one of an immense checker- 
 board, where the boundary lines between the squares were deep 
 clefts and the very top of the squares rounded hummocks, covered 
 with glacial marks. There is hardly another such location "on the 
 Labrador," as it is here called, that presents so unusual and decid- 
 edly remarkable an appearance. 
 
 Friday, the 13th. The wind has played us one of its tricks, to- 
 day, and left for parts unknown. We tried all day to get into 
 Forteau but were unable. We consoled ourselves however, as well 
 as possible, with roast duck, and waited for the wind. What little 
 breeze we did have was dead ahead. Finally we returned to Blanc 
 Sablon, and prepared to send letters home by a small schooner 
 that we found leaving that afternoon for Gaspe. Saturday it was 
 foggy, and we were unable to proceed but a short distance from 
 the harbor. The gun at Greenley Island lighthouse, and the steam 
 whistle at L'Anse Loup, kept up a continual hooting, while, near 
 to, we could hear and almost see — several times we actually did 
 
WHALES, "BLOWING"— RED BAY. 
 
 241 
 
 near 
 
 see them — the whales as they sported and spouted close to our 
 vessel, in spite of the foggy weather. We afterwards found 
 them very abundant in this vicinity, and often saw half a dozen 
 of them within a short distance of the vessel in different directions, 
 all spouting at the same time. They were probably of the species 
 called finbacks, since the " spout " corresponded to this class of 
 whales more closely, being a single stream sent upwards ten or 
 twelve feet in an almost perpendicular line, falling back again as 
 spray in nearly the same place. The sulphur bottom, the Green- 
 land, and the hump-backed whales, it is said, all blow a double col- 
 umn of water rather obliquely backwards; the finbacks, a high 
 straight stream some twelve feet directly upwards to fall back again 
 to the same place as spray ; while the sperm whale shoots its single 
 stream somewhat forward, the spray falling forward and beyond the 
 head. Whether this distinction holds good in all cases I do not 
 know, but the seamen insist that it is invariably the case. 
 
 During the night the wind sprang up and we drifted nearly to 
 Newfoundland. In the morning we found ourselves in sight of 
 houses two miles off the land ; in a short time we would have been 
 on shore. About daylight the wind started up. The outline of 
 the Labrador coast, the fog having cleared away by this time, showed 
 plain and clear. The various buildings along the coast, and the 
 highlands of Bradore, L'Anse Dune, Blanc Sablon, L'Anse Loup, 
 Forteau, and L'Anse Diable, appeared in rapid succession. We 
 noticed particularly the difference between Black or Pirouette 
 River, with its receding slopes, on the east, while the ridges of 
 the west coast extended away back into the country in the shape 
 of huge rows of hills, that, though well wooded, appeared barren 
 and desolate enough. 
 
 On we sailed : from the correspondence of the coast lino 
 to the position of Red Bay on our charts, I felt we must be near 
 that place ; and soon Saddle Island appeared ahead of us, and, with a 
 glass, we could plainly see the church on the hilltop, and the houses 
 on the slopes below. The passage to the harbor of Red Bay is 
 between Saddle Island and the mainland. It is very small and 
 16 
 
ri 
 
 iV 
 
 H 
 I* 
 
 r^ 
 
 • 
 
 
 242 
 
 SADDLE ISLAND. 
 
 narrow, and scarcely perceived until you are close to the west of 
 it. The island appears like a part of the mainland. It is probably 
 one of the most secluded harbors of the coast. Our surprise can be 
 imagined. We had supposed ourselves at least twenty if not thirty 
 miles farther up the coast, when we had taken our first reckoning. 
 When close upon Saddle Island the opening begins first to appear. 
 The water is deep in the passage — as everywhere in the harbor — 
 close to the very edges of the land on either side, and though the 
 passage was small we beat in without the slightest difficulty, run- 
 ning so near the rocks that our boom nearly touched the rocky 
 ledge ashore as we tacked ship. Here we entered a new region : 
 I wish I could picture it to you. The air was clear, fresh, and 
 crisp. The sky was almost cloudless. The hills presented most 
 charming natural pictures with their coverings of already green foli- 
 age. Close to the edge of the water nestled the houses with the 
 stages and workshops of some twenty-five or thirty families. 
 
 When once in the harbor we were shut in completely. The high 
 top of the crests on Saddle Island prevented even our mast from being 
 seen, while we were easily led to imagine ourselves in one of those 
 strongholds of ancient piratical romance, — always so well chosen 
 and impregnable. In the evening we, went ashore and heard a 
 good, earnest Christian, of the Wesleyan faith, lead a good old New 
 England sen'ice of evenirig prayer. We afterwards occupied some 
 time in calling upon various people dwelling here, and found them 
 very hospitable. 
 
 Red Bay is the Newfoundland headquarters, for this part of the 
 coast, of trade and of the traders. It is a very old settlement, and 
 contains several very aged native inhabitants. Though there are 
 one or two large fisheries here in summer, there are twenty-one 
 families only that winter and live here the year around. A. M. 
 Pike, one of the oldest inhabitants, told me that he had lived on 
 the coast for thirty-two years, and that the place was never larger — 
 except during a little while in the summer — than at present. Of 
 the twenty-one families, seven are named Pike, three Pennie, and 
 three Ash. One old gentleman showed me a sample of lead ore 
 
RED BAY HARBOR. 
 
 243 
 
 which he said had been taken from the neighboring hills, and also 
 informed me that mica, in large sheets, was abundant not far off; 
 but no one could inform me as to Labradorite, which, however, 
 occurs as a bright green and as a black, glittering, scaly variety in 
 several places along the coast. 
 
 As a short description of the harbor may be of interest I will 
 try to give it. From Twin Island, the eastern boundary of the 
 eastern mouth, which is shaped much like a carpenter's square, we 
 approach directly to Saddle Island, which stands in the very centre 
 of the mouth of the harbor, so that, in approaching, neither the 
 eastern or western entrance is perceived until one is directly upon 
 it. This island is about twice as long as it is broad, and lies about 
 southeast by northwest. There is a hill on either end one of 
 which only contains a beacon. The island is almost wholly rock. 
 The depression between these two crests reaches nearly to the level 
 of the water and yet it contains a pond of tolerably good water 
 where the people obtain their usual supply for daily use in the sum- 
 mer time. In what is known as the outer harbor, the water is deep 
 on both sides close to shore, while the eastern entrance is more 
 shallow than the western ; so that while men-of-war vessels often 
 pass in at the western, and anchor within the second or inner harbor, 
 only small boats can pass through the eastern entrance. The west- 
 ern entrance enlarges into what is called West Bay. The shores 
 of the outer harbor gradually approach each other to within almost 
 a stone's throw, and yet there is ample room for the largest man-of- 
 war vessel to pass safely and anchor, as the revenue and other ves- 
 sels that often enter here do, within the inner harbor. 
 
 Close to shore the inner harbor is shoal, yet in its centre, it is 
 very deep. In the farther end of this peculiar hour-glass are situ- 
 ated the winter houses of the inhabitants, :it the foot of high, reced- 
 ing, unevenly sloped and gorged hillocks that look like a vast 
 amphitheatre. Near by, on the east side, is a little brook ; here 
 plenty of fresh water is always to be obtained, while the slopes of 
 the hills furnish abundance of firewood, if one will only cut it. Taken 
 all in all, this is a most beautiful place, and well calculated to ex- 
 
nn* 
 
 
 N 
 
 L 
 
 I;'. 
 
 ■J; 
 
 I 
 
 "4- 
 
 tin 
 
 '^^ 
 
 
 lu 
 
 I: .;' 
 
 11 
 
 244 
 
 CHARACTER OF SADDLE ISLAND. 
 
 cite the enthusiasm as, hkewise, the envy of all lovers of beautiful 
 scenery. While in general outline the eastern side of the harbor is 
 low and marshy, the western is composed of granite hills from four 
 hundred to nearly five hundred feet in height, coming down to 
 within a few fathoms of the sea. Again on the east the little eleva- 
 tion, on which stands the church, is only seventy-eight feet above 
 the sea, the hill immediately behind it one hundred and seventy, 
 while the highest and next in order is only two hundred and ten 
 feet. From this elevation I could see beyond low earthy and gran- 
 ite hillocks. Immense ridges and bowlder-like rocks of granite were 
 lying about us in every direction ; beneath were bays, ponds, and low 
 marshy spots reaching in succession nearly to the sea level. The 
 formation is so much different on the opposite sides of this harbor 
 that it is quite remarkable. On the west red feldspar predomi- 
 nates, in large cliffs, whence the name Red Bay, while on the east 
 the rock is almost entirely granite or gneiss. Saddle Island, so 
 named from its resemblance to a saddle, contains a most peculiar 
 phenomenon. In the low marshy depression between the higher 
 extremities of the island is situated a small pond. Though this 
 pond is within a few feet of the level of the salt water, it is so sweet 
 and fresh that the people use it for drinking water. The pond is 
 very deep, yet with no apparent inlet any^vhere. The outlet runs 
 continually, yet the supply does not appear to diminish. 
 
 I shall never forget the clear, beautiful, varying shades of green on 
 the slopes, and the dark outlines of the houses, as the s\m sank behind 
 the western hills, overshadowing them for an instant, the first night 
 of our entrance into this charming little harbor. We could see the 
 people all along the shore, wending their way to church ; while in 
 place of the well-known music of the church bells, the robins, here 
 equally abundant as at home, and the "russingels," or fox sparrows, 
 sent forth a perfect medley of harmony that accorded well with *h\s 
 scene. 
 
 Give me the rustic harmony of a woodland scene like thi id 
 I will defy the best laid argument of philosophy that would ati^i. :it 
 to prove that such people, if sincere in their worship, have not 
 
A RUSTIC SCENE — RETURN. 
 
 245 
 
 ♦ 1 
 
 tis 
 
 reached the outermost circle of heaven's horizon. I care not for 
 pompous argument, and flowery speech, that would attempt to con- 
 vince one by its mighty utterances, when the glimpse of such a 
 scene as the one I represent is before one's very eyes. Surely 
 such a scene as this produces a quietness, peace, and serenity of 
 mind that no argument can prove false, and no philosophy can 
 shake. It is in itself at once the reality of both argument and 
 philosophy, and presents to us an end, which, having reached, calms 
 our fears and bids us drink and live : drink in the beauties already 
 fading with a falling sun, and live in newness of revived hope to 
 return home with a satisfaction such as no mere sermon could give. 
 
 From Red Bay we retraced our steps along the coast, calling at 
 various places on the shore. I will try to give you an idea of the 
 coast about here. From Saddle Island (Red Bay) to Carrol's Cove 
 is five miles. Carrol's Cove is a sort of carpenter's square attach- 
 ment of the main coast, which here consists of high cliffs and 
 hilltops. It is low and scarcely above the sea level at high tide. 
 The extremity points in an easterly direction ; the shores everywhere 
 are narrow strips of pebbly beach ; while the point itself consists 
 almost entirely of rocks. The bend in the land forms a sort of 
 harbor that, in mild weather, is not an altogether bad one. Here 
 some eight families live in about as many houses. Many of them 
 are quite intelligent, and some of them have sons or daughters that 
 are now residing in different portions of the states, and I was 
 often accosted with the question of, " Do you know my son living 
 in such and such a place ? " 
 
 We soon passed from Carrol's Cove to Eastern Modest a 
 distance of four miles where eight families reside ; thence to 
 Pirouette River, three miles and a half, where there are five fam- 
 ilies (this place is also called Black Bay and the river at the head 
 of the bay Black River) ; thence to Western Modest, nearly three 
 miles farther, where seventeen families live ; still farther up the 
 coast to Cape Diable, about opposite which is Capsan Island, one 
 and one-half miles, where there are five families ; into Diable bay, 
 DT L'Anse Diable which is, perhaps, some two and a third miles 
 

 246 
 
 CENSUS — FOWLS. 
 
 farther, where ther<; reside seven families. We continued our 
 journey three and one-half miles to L'Anse Loup where we found 
 eight families residing ; from the bottom of L'Anse Loup Bay to 
 Schooner Cove, the property of a merchant from Newfoundland a 
 Mr. Watson by name, the distance is a mile and three quarters ; 
 and from here to the lighthouse, at Point Amour, two and one- 
 lialf miles farther. We did not anchor at either Capsan Island or 
 L'Anse Diable, since there is no good ground to anchor there, and 
 vessels generally remain either farther up or down the coast, while 
 the travel is in small boats between these two places. Thus from 
 Red Bay to Point Amour, a distance of about twenty-six miles, 
 following a straight line from settlement to settlement, there live 
 about eight families and four hundred people. 
 
 The census from Pt. Amour to Blanc Sablon could be estimated 
 in about the same proportion, while the latter place is perhaps the 
 largest, with its surroundings of Wood and Greenley Island, east 
 of Esquimaux Point and Natashquan. Strange as it may appear, 
 nearly every family of this little colony keeps fowls and domestic ani- 
 mals, and it is no unusual sight to see thirty or forty of the former 
 running about the settlements, the hens picking for food here and 
 there, and the cocks crowing continually. We were able to pur- 
 chase about eight or ten dozen fresh eggs for the nominal price of 
 a shilling (twenty cents) a dozen. We would willingly have paid 
 twice that amount. 
 
 I will not stop to describe all the scenes along the coast here, 
 but will simply say that some of the places were gems of loveliness. 
 Some were backed by tablelands of sandstone, others were pecu- 
 liar formations of granite. Most of these places we had only time 
 to see from our vessel as we passed by them, yet they presented a 
 panorama of loveliness. When opposite Pirouette I could plainly 
 hear the pipings of the marsh frogs in the swamps of the river 
 beyond, and have no doubt but that, as one of the inhabitants 
 told me, they are very common there. Several of these places ap- 
 pear to have a formatioti much different from that of the country 
 around them, and I have no doubt but that most interesting results 
 
 n 
 
THE BATTERY AND RED CLIFFS: THEIR DANGERS. 247 
 
 would be obtained by careful study both of the fauna and flora, 
 as well as of the geology of the region. 
 
 Thursday night we reached L'Anse Loup, passing the Battery as 
 it is called and Red Cliffs. The Battery consists of a plateau with 
 perpendicular, stratified cliffs in front, about three hundred feet high, 
 though I did not go ashore to measure it exactly. It forms a most 
 striking, and grandly beautiful feature of the surrounding scenery. 
 ! am informed that in winter it hangs with masses of icicles, that, 
 from the strata of colored rock and soil back of it, forms most beau- 
 tiful and almost dazzling reflections of colors of all sorts and plays. 
 It must truly be a grand sight to see this huge mass of rock show- 
 ing forth crystals of water in such mingled brilliancy. 
 
 A path runs below the overhanging cliffs, narrow, but distinct 
 even at high tide. In winter, when blocked with snow, the people 
 are forced to scale the height and travel with great difficulty over 
 instead of beneath this platform of rock. Though travelling below 
 is much more dangerous, the danger is generally laughed at, and 
 the space crossed as quickly as possible. Several very narrow es- 
 capes from serious accident have been recorded here. At one 
 time a huge rock, of several tons weight, fell directly between the 
 dogs and the komatik, as the sledge was hurrying along at its ut- 
 most speed ; no one was injured, though all were very badly fright- 
 ened. At another time a sledful of several young ladies had just 
 passed one of the most dangerous parts of the road when a shower 
 of rocks fell behind and between them and the next komatik which 
 was but a short distance behind, yet nobody was injured. Soon 
 we had passed this place and were snugly anchored in L'Anse Loup 
 Bay for the night. The morning gave us a good view o/ the bay, 
 which is a simple indenture of the coast, about as deep as it is 
 wide, and nearly a mile and a half in either direction. The eastern 
 and western sides are ridges of hills, while their slopes are rather 
 pretty and picturesque. A small stream enters the foot of the bay, 
 wide but shallow ; while west of this is a huge, low ridge of rocks, 
 and farther west still a beach of pure, clear sand. The extension 
 of country beyond appeared as if it might once have been a part of 
 this same e/tcnsion of bay, as it no doubt was. 
 
'!IB.: 
 
 M 
 
 248 
 
 FORTEAU BAY. 
 
 
 ■: md 
 
 Forteau Bay is very similar to Loup bay. It is about twice as 
 deep, and more triangular in shape, while the eastern extremity 
 bends inward much like an inverted comma, giving the appearance 
 almost of a second bay infringing upon the first. The eastern side 
 seems to have been formed by the washing away of the limestone 
 ridges which are here several hundred feet high, and extending far 
 back into the country. The western side is principally of limestone 
 which recedes to the top of the cliffs beyond, at a height of about 
 four hundred feet. The bottom of the bay is mostly a sandy beach, 
 except where occasional outcrops of rock appear. There is a 
 small stream here also, as in nearly every bay along the coast, that 
 seems to drain the remains of some former extension of this same 
 bay, from the low and extended valley beyond. I noticed particu- 
 larly the cliffs on the eastern side, which reached almost to the sea, 
 and were nearly perpendicular ; whose stratifications were apparent 
 even from our vessel, and whose uneven fractures presented the 
 appearance of ruined towers or castles and temples. I plucked 
 several sweet-scented spring flowers ; and saw with pleasure signs 
 of the springing mto new life of evergreen as well as alders, shrubs 
 and closely clustered vegetation everywhere abundant. 
 
 Our stay here was short, however, and a fair breeze soon brought 
 us again to Greenley Island. I will describe it. 
 
 Greenley Island is a low island, scarcely seventy feet, at most, 
 above the sea level. The northern part is short and wide, the 
 southern point long and blunt. Several rising hillocks mark the 
 centre of the island ; from these the land slopes on one side close 
 to the sea, on the other to the western portions of the island which 
 are covered with rocks and bowlders as if once the terminal mo- 
 raines of a glacier, yet much waterworn ; while the water beyond is 
 clear and sandy. On the east, and I think also the north side of the 
 island is sand ; on the southwest the rocks extend but a little distance 
 into the water, while on the east point is an immense flat table of 
 rock, seamed, scarred, and rounded by the water, which covers it 
 at low tide. This is a curiously formed ledge of granite, and does 
 not appear to have been broken excepting by the scars of time and 
 water, yet it appears to form the underwork of this part of the island. 
 
 »■: 
 
GREENLEY ISLAND. 
 
 249 
 
 t twice as 
 extremity 
 pearance 
 item side 
 imestone 
 nding far 
 imestone 
 of about 
 iy beach, 
 lere is a 
 3ast, that 
 his same 
 particu- 
 » the sea, 
 apparent 
 rated the 
 plucked 
 ire signs 
 3, shrubs 
 
 brought 
 
 at most, 
 ide, the 
 nark the 
 de close 
 id which 
 nal mo- 
 jyond is 
 e of the 
 iistance 
 table of 
 overs it 
 id does 
 me and 
 '■ island. 
 
 Upon the island is an occasional outcropping, apparently of this 
 same rock, though for the most part the ground is soft grassy and 
 mossy muck, with occasional fresh water ponds. Here and there 
 are strewn angular blocks of coarse, poor, grayish-yellow granite, 
 quite different from the predominant rocks on the island. These 
 are more or less square in outline and varying in size from two to 
 five feet. The tops are flat, and the edges all clean cut, and hardly 
 if at all water-worn, being distinct from the worn edges of the bowl- 
 ders, which were common near to the water's edge. These blocks I 
 have mentioned are scattered at irregular distances all over the 
 southwest portion of the island, being more abundant near the table 
 rock before mentioned, and west of the lighthouse. All the beach 
 on that side of the island which faces Wood Island is sand, while 
 the high point here seaward is mostly of the same origin so far as I 
 could ascertain. 
 
 The lighthouse is situated on the southern extremity of the island. 
 It is an octagonal wooden structure, marked like stone or brick, and 
 painted white upon the outside. The apparatus for illumination 
 consists of an iron framework fastened to an upright revolving rod, 
 that moves by a simple arrangement of wheels ; one of these is 
 heavily weighted while a handle is furnished for winding the weight 
 and keeping the apparatus in motion. Another small upright rod 
 with an upright screw attached keeps in motion a patent governor, 
 for regulating the time of each revolution. This governor is a 
 crossbar, on each end of which is a sort of fan, of brass, which 
 turns in a perpendicular or horizontal direction to indicate slow or 
 fast. Ordinarily the governor turns one hundred and twenty times 
 in a minute, and the lamps are four minutes in making a complete 
 revolution. The speed is regulated by government authority. The 
 top of the tower has twelve large, double plate glass windows, each 
 a quarter of an inch thicic. The lights are twelve, arranged to shine 
 as four, three being white and one red, which flash as they revolve 
 at given intervals far out to sea, warning the sailors that the coast 
 is nigh. 
 
 We were very cordially received and shown around by the light- 
 
'.rJt 
 
 
 I 
 
 
 1 , 
 
 250 
 
 OFF ST. MARY ISLAND — CORMORANTS. 
 
 keeper, who, being a French gentleman, however, understood very 
 little English. There is here also a fog gun, fired every half hour 
 in foggy weather. That, with the five minute steam whistle at 
 Point Amour, near by, renders it quite lively here at times, and 
 furnishes much amusement to the natives, especially those, who, 
 coming from a distance, hear these things for the first time. 
 
 Monday the 23rd. At last we have a breeze, and hoisting full sail 
 we are soon sliding through the water at the rate of about six miles 
 an hour. By night we are far on our way to the westward. 
 
 Tuesday the 24th. Though the wind has slackened a little it still 
 holds good. The air about us is perfection itself, so clear and 
 bracing is it. At eight o'clock we were just off the St. Mary Islands, 
 having gone about eighty miles in twelve hours, and, counting the 
 curvature of the coast, a full hundred and sixty in the last twenty- 
 four; and yet on we go, — dashing through the water. We pass 
 Shag rocks, a long row of bare rocks, without vegetation of any 
 kind, where the cormorants, or shags as they are here called, breed 
 in large numbers upon the ledges of bare rock ; they use their 
 own guano deposits for a nest. There are two species of cormo- 
 rants here, the common cormorant {Graculus carbo), and the 
 double crested cormorant ( G. dilophtis) ; both arc called shags, 
 but the latter are generally designated by the Indian name which 
 is, if I am informed correctly, Wapitougan. Both species breed 
 equally abundant apparently. I have seen thousands at a time 
 lining the rocks. They sit upright, in rows, upon the edges of the 
 rocks, and seldom one sits behind another, so that, to accommodate 
 them, every edge of each crag presents a trimming of cormorants ; 
 a lively looking trimming just as some shot is fired that sends all 
 into the air. The eggs are two to three, and, though really bluish 
 white in color, are invariably covered with a calcareous deposit 
 that renders them exceedingly chalk-like in appearance. 
 
 At a distance these rocks present the appearance of being cov- 
 ered with snow, but a nearer approach shows that this is a covering 
 of guano from the continual droppings of the birds ; while the tops 
 of the rocks are thickly embedded with an accumulation of guano 
 
 ii'' 
 
 ■I'll 
 
If 
 
 NATASHQUAN — SPRING BIRDS. 
 
 251 
 
 
 who, 
 
 from the same cause, firmly stamped down with the continual pat- 
 tering of numberless feet. 
 
 Tuesday the 25th. We anchored off Natashquan. Here we ob- 
 tained a bundle of newspapers, mostly Harpers' Weekly, the first 
 we had seen for nearly seven months. Natashquan point is a httle 
 better than a sand bank, with an overgrowth of low spruce and fir 
 trees, and but a poor attempt at vegetation. At its extremity are 
 placed the houses of one of the Hudson's Bay Company's posts, 
 and one or two native huts, while on a small island opposite is a 
 very neat house belonging to a trader, who passes the greater part 
 of his time between this place and the neighboring shores and 
 Quebec trading along shore. The water all about the point is very 
 shoal with dangerous sandbanks in every direction. I saw the hulk 
 of a vessel of about eighty tons, that several days before had run 
 aground here ; it had become stuck fast in the sand and was now 
 a total wreck. A few miles almost direct east takes us to the mouth 
 of the river where the settlement is. The harbor, even here, is full 
 of sunken ledges, most of which appear only at low tide. Vessels 
 cannot approach near the shore even here, since the sandbanks 
 again interfere, and the water is quite shallow even at high tide. A 
 little distance above this is another river, and the intervening space 
 is a picturesque little peninsula, of coarse and fine granite of feld- 
 spar, over which a scanty vegetation forms a groundwork with 
 dwarf spruce and fir trees extending far back into the country. 
 
 In the bushes, just back of the few houses that line the stream, I 
 saw large numbers of birds. The white throated sparrows sang 
 their well known pea-body, pea-body, pea-body, prefaced by their 
 usual whistle, from nearly every prominent tree-top, while I amused 
 myself for nearly an hour watching the robins as they flew or 
 hopped about on the lawn, or in and out the wood-pile and other 
 debris in front of the houses, enjoying the bright sun which 
 shone down, warm and gladdening, upon the ground just springing 
 into greenness again, wet with the moisture of melting snow. I saw 
 quite a number of other birds, and recognized the white crowned 
 sparrow, and also a shy Maryland yellow-throat ( Gcothlypis tri- 
 
 

 IV ■ 
 
 
 252 
 
 GLACIAL MARKS — THE "RUSSINGEL." 
 
 cAas) , who appeared several times, for my express benefit, I sup- 
 pose, in plain sight, before finally disappearing into the underbrush 
 at the right. The remainder of the forenoon was spent in examining 
 the glacial rounded rocks, on which I found several well defined 
 scratches, and in following the stream for a short distance to the 
 meadows, or low marshy districts just beyond the houses ; walking 
 all the way on shelving rocks that, nearly level with both meadow 
 and stream, sloped off in large platforms into the water. This river, 
 I am told, is navigable only for small boats, and for only two miles 
 from its mouth, though I believe that the Indians travel somewhat 
 farther in their canoes. As the people living here were mostly 
 French I could glean but little from them. The harbor, however, 
 is a mass of shoals even to the mouth of the river, whose eastern 
 bank is sand, while the bed of tlje stream partakes somewhat of 
 the character of the harbor, as far as I saw it, at least. I here heard 
 the Canadian " russingel, " full of most tuneful melody for which it 
 is so noted throughout Canada. On the right bank of the river, 
 and bordering the beach, are quite a number of houses, while a 
 small island near by contains a cluster of as many more ; altogether 
 quite a settlement for this region. One large fishing establishment 
 has about forty boats and two hundred hands engaged in the fish- 
 eries, during their season ; while there is a postoffice which receives 
 mails to and from Quebec, Bonne Esperance, and the South Shore 
 by packet, via Gaspe, touching at Anticosti, between which latter 
 places a submarine cable has been recently laid, to give warning 
 of the shipwrecks which are so constantly occurring on this island. 
 Saturday the 28th. It was to-day, if I remember right, that we 
 experienced the first of the two noted dark days of 1 881. We were 
 about half-way between Natashquan, the point we had just left, and 
 Mingan, a post of the Hudson's Bay Company, to which we were 
 bound. Early in the morning the wind left us, and soon the atmos- 
 phere was clouded over with a dull murky and yellowish smoke- 
 colored light, that almost hid the sun. We could only see a faint 
 light spot where that luminary was protesting unsuccessfully against 
 such an infringement of its illuminating powers. By eleven o'clock 
 
DARK DAY — WHALES. 
 
 S58 
 
 % 
 
 t, I sup- 
 derbrush 
 icamining 
 
 defined 
 e to the 
 
 walking 
 meadow 
 his river, 
 vo miles 
 )mewhat 
 J mostly 
 lowever, 
 
 eastern 
 ;what of 
 re heard 
 which it 
 le river, 
 while a 
 together 
 lishment 
 the fish- 
 receives 
 h Shore 
 h latter 
 ivaming 
 I island, 
 that we 
 l^e were 
 jft, and 
 i^e were 
 atmos- 
 smoke- 
 a faint 
 against 
 3'clock 
 
 the sky was like the lurid smoke of a furnace, and about as yellow, 
 the air was so heavy as scarcely to be breathed, while the sailors all 
 turned at once to Mother Shipton's prophecy, and believed, for sure, 
 that the end of the world was about to approach. I afterwards 
 learned that as far down the Gulf as Bonne Esperance, and even 
 at L'Anse Loup, the lamps were lighted at various times from eleven 
 in the morning until four in the afternoon, while nearly all along 
 the coast, some part of the day presented for a short time, an ap- 
 pearance nearly as black as night. I will mention another circum- 
 stance and its interpretation by the captain and crew that appeared 
 most striking. As we drifted along we passed successively, long 
 waves of a fine, yellowish dust powder that lay upon the surface of 
 the water often so thickly as to cover acres of water at a time. It 
 looked precisely like a deposit of sulphur. Part of those spoken to 
 on the subject declared that it was a deposit from the smoke of 
 several forest fires, which we afterwards know to have raged on 
 Anticosti, as well as several places on the mainland. I believe they 
 said that they had seen something like it before — at least it did not 
 particularly surprise them ; while another class declared that it was 
 the dust or rather pollen from the innumerable alders that line the 
 banks of the rivers all about the coast. The extent of these patches 
 was remarkable, whichever explanation was true. The patches 
 were often ovei- a roile in length, while we passed over many acres 
 of it before leaving it. 
 
 Several whales had played near us during the night, and when 
 first seen, the ridges were declared oil from the whale blowings, — 
 which, often indeed, cover the water for miles ; but during the 
 night all passed off, and in the morning we had passed Esquimaux 
 point, a Roman Catholic sealing settlement of about one hundred 
 and sixty houses, and soon reached Mingan, a most beautiful little 
 harbor of clustered islands, where the varied figures of the tide 
 ripple in the shallow water are one of the most remarkable displays 
 of the kind along the coast. 
 
 m 
 
v> 
 
 254 
 
 IN SIGHT OF MINGAN". 
 
 CHAPTER XVI. 
 
 ■^ . 
 
 Mingan and surroundings — Hudson's Bay Companies' buildings — Mingan 
 River — Indians of this region; their habits, religion, etc. — "Montag- 
 nais" and " Nascopies " — The Indian trade at the various places along 
 the north shore — Romaine or Olomanosheebo — Natashquan again — 
 French steamer and salmon freezing — Jewelry peddler — Agwanus, Na- 
 bisippi — Terns and gulls — Codfish " schooling " — Esquimaux Point — 
 Indian names, etc. : St. Genevieve Island, Watcheeshoo, Manicouagan, 
 Saddle Hill, Mt, St. John, Washatnagunashka Bay, Mushkoniatawee, 
 Pashasheeboo, Peashtebai — Shooting at the Fox Islands — Mutton Bay, 
 Great Mecattina Islands — Old Fort Island again at last. 
 
 I can well remember the morning when our little schooner headed 
 for the well sheltered harbor of Mingan. We had started from Es- 
 quimaux Point, off which place the vessel had been anchored during 
 the night, about nine o'clock, and, passing the point with its low 
 extension of sand beach and its one hundred and sixty houses, — 
 some of which were of quite unique pattern and pleasing outward 
 appearance, — had fallen in with a light breeze which took us along 
 right merrily through the tide ripples and shallow waters between 
 the long chain of islands seaward, and the mainland on our right. 
 Early in the afternoon we sighted a long, low point of land in the 
 distant horizon, where a small white dot was plainly visible, repre- 
 senting the buildings of the establishment toward which we were 
 fast approaching, and which was our destination. There are times 
 when the motion of a contrary tide against the sides of a vessel ap- 
 pears to show that one is fast gaining, when in reality losing ground ; 
 there are also times when such a tide is in the vessel's favor, and a 
 swift, yet almost imperceptible motion, carries her on at a rate of 
 five, six, or even more miles an hour. At times our vessel had each 
 of these motions, now going swiftly, now slowly, as the case might 
 
"TIDE RIP"— MINGAN HARBOR. 
 
 25S 
 
 be, while the low and sandy, or the high and cliff-like sides of the 
 neighboring islands grew nearer and nearer, to fall slowly behind us 
 and grow dim in the distance, as the point toward which we were 
 aiming grew each moment plainer and plainer. 
 
 We were between the mainland and a few low islands on our left, 
 when suddenly we struck a most remarkable " tide rip," as it is 
 called. The rippling of the water, with low, even lappings, sending 
 forth music like the laugh of " Laughing Water," extended all around 
 and about us, a mile at least in each direction. Cross currents 
 made long furrows here, while there, broad sheets presented the 
 same even ripple as far as the eye could reach- On through this 
 charming and fascinating little place we slowly glided, as we listened 
 to the merry waters and feasted our eyes on the playful ripples, 
 watching our sure approach to the harbor now so near us. Flocks 
 of ducks, hurrying to and fro, passed us on our right and left, while 
 jruUs and other birds sported around and above us, or floated on 
 the surface of the water, often a few rods only from the bow of our 
 vessel as we glided along. At last we reached harbor, and were 
 soon safely anchored. 
 
 Mingan harbor is a narrow but deep stretch of water open to the 
 sea on the east and west, with the mainland on the north, and Min- 
 gan Island on the south. Of Mingan Island I can say little ; it is a 
 low piece of land, long and narrow with a foundation of sand which 
 is visible here and there along its surface. The main body of the 
 island seems to be composed of a series of rocks deposited in layers 
 one above the other, and forming low, irregularly faced cliffs, with 
 slate-like cleavage and fracture, which are nearly or quite perpen- 
 dicular in outline. The upper surface of this mass bends in various 
 directions, while its height is twenty to forty feet above the sea. 
 Low, stunted firs abound everywhere, so that the island forms a not 
 unattractive sight in this really romantic little spot. 
 
 On the mainland, or Mingan proper, contrary to what might be 
 expected from the appearance of the island opposite, an entirely 
 different formation exists. Nowhere along the coast, for a consid- 
 erable distance at least, does a rock of any size appear, either in 
 
I' 
 
 256 
 
 THE NEIGHBORING COAST LINE. 
 
 place or loose as bowlder, stone, or pebble. Strange to say, as will 
 be shown further on, the rocky precipices, or rather steps of the 
 rapids in Mingan river, some three miles from its mouth, seem to 
 be the first indications of rock formation in this locality, while these 
 are simply the eastern and southeastern boundary of a tremendous 
 mass of high rocky ground that extends inland for miles, perhaps 
 thousands of miles. 
 
 The coast and its beach, as the whole country to the rocks inland, 
 is everywhere low and sandy. On the beach itself the sand is dense 
 and very fine. Farther in shore there is a very scant, occasional 
 streak of low vegetation where are pastured a few heads of cattle 
 and goats that graze on the lawns, here and there, where they can 
 find food. A few acres of good grass are fenced in, and this sup- 
 plies an excellent feed for the animals during the winter, which 
 here is neither so long nor so severe as is usually the case farther 
 north, at Bonne Esperance even. Froni Mingan west to Long 
 Point, a distance of about six miles, this low sand beach extends 
 almost without a single rock, I believe, while the east beach is en- 
 tirely of sand. The river itself passes through a ridge of this same 
 material which forms a high bank on the left and a low one on the 
 right, as one passes inland, while the whole land rises directly from 
 the sea then falls in a northeasterly direction, and the trend of 
 greatest height, here, as nearly everywhere along this part of the 
 coast, is in a northwesterly direction. In the background, the dis- 
 tant hills rise to a height of at least a thousand feet, while dim out- 
 lines of others, of perhaps greater height, appear in the horizon. 
 This is the picture whose charming outline at once attracts and 
 captivates one upon entering the harbor of this sequestered little 
 spot. Exhilarated by the sharp, fresh air we land, and soon count 
 our trout from the waters where few but the Indians have preceded 
 us. 
 
 Mingan has been for many years a post of the Hudson's Bay 
 Company. Thither the Indians from the interior resort to sell their 
 furs, for which they receive in exchange provision, clothing, ammu- 
 nition, and those useful articles of which they may be in need. 
 
 II 
 
MINGAN : ITS ATTRACTIONS. 
 
 257 
 
 Mingan is situated about three hundred and sixty miles from Que- 
 bec, in a straight Une, while it is only twenty-five miles from the 
 opposite sliore of Anticosti and is a little east of a point opposite 
 the extreme southwestern portion of that island. The attractions 
 of the place are only its fishing and the houses of the company's 
 post with that of the Guardikphche. PVom your position on board 
 your little vessel in the harbor, you can see them all ; the long 
 plank walk with the net shop at its eastern extremity has near it 
 the officers' house with a cosy little office close by ; farther down 
 are the storehouses, wherein are the provisions, cloKang and such 
 like stores with which the general trading of the establishment is 
 done, while bundles of fine fur hang suspended from cross bars and 
 nails in the lofts above. Just around the bend, the walk continues 
 to an old unused wharf where it terminates ; being a distance of 
 one-tenth of a mile. I say one-tenth of a mile, — many is the time 
 that this plank walk has been paced, while sauntering for pleasure 
 or to pass the time away, undoubtedly by each individual dweller 
 of this establishment. A few barns or outbuildings, of one sort or 
 another, placed here or there as the case may be, complete that 
 portion of the post which has been built up to the j -resent time. 
 Most of these buildings have been tastily painted, thus presenting 
 a neat and attractive picture to one viewing it from the vessel in 
 the harbor, or the deck of the little steamer, that, plying between 
 here and Quebec, touches all the important points along the coast 
 for a distance of some few hundred miles. If you visit this charm- 
 ing little spot, you are sure of a cordial reception from the gentle- 
 manly officers of the post, who are only too pleased to welcome 
 strangers amongst them. 
 
 If there is plenty of time at your disposal, hire, for a small amount, 
 a guide and canoe and make a day's trip up the river. The Min- 
 gan river flows in a west varying to southeasterly direction, and 
 enters the Gulf not far from the Romaine river which flows from 
 nearly a contrary direction. At the mouth of both rivers are shal- 
 lows and accumulations filling the water with ridges that control 
 strongly the current at this point. These sand bars are constantly 
 17 
 
.:)■■' 
 
 
 258 
 
 TRIP UP MINGAN RIVER. 
 
 
 shifting, while in places they have overrun each other and piled up 
 small islands of sand which becoming overgrown with grass or scant 
 vegetation have become the nesting places of gulls and ducks, thus 
 supplying the people with birds and eggs in large numbers when- 
 ever they are desired. Following up the river you will find sand 
 and sand banks on either hand, and extending, with scant vegeta- 
 tion, far inland. Soon we approach the rapids, where the river 
 widens into a baylike expansion of water bounded everywhere by 
 granite cliffs, which, rising to a height of three hundred feet extend 
 in undulating billows of uneven height, far towards the dim out- 
 lines of the hills beyond, while tangled vegetation, and bowlders 
 and loose rock lie mingled in confusion on all sides. 
 
 In front of the rapids, and just before reaching them, we came 
 to a partial clearing, on our right, in the dense mass of fir and 
 spruce trees here grown so closely together, where a party of pleas- 
 ure seekers from " the States " had built a " cabin " which they 
 used as a rendezvous during the summer sporting season. The 
 rapids, directly opposite this charming spot, consist of a body of 
 water pouring over a series of rocky ridges running from shore to 
 shore, and which form a regular pair of steps of fifteen to twenty 
 feet in height. The opening of the river is between perpendicular 
 walls of granite which have apparently been worn down and crum- 
 bled away by time and the water so constantly dashing and splash- 
 ing over it in its path to the ocean. Above the rapids the stream 
 is narrow but deep. 
 
 To the left of these rapids formerly existed a narrow pathway 
 which, ascending to the height above, led to a position whence the 
 descent to the river again was comparatively easy. This path was 
 once much used by the Indians who sought the interior of the 
 country by this route, but it has long since been discontinued. I 
 ascended this path with the guide for nearly two miles, but the 
 walking was so difficult, the pathway often so obscure, and the whole 
 surface of the plateau exhibiting such a sameness of general 
 feature that we soon gave it up as a bad job and returned with 
 difficulty the way we had come. I found in the rocks several small 
 
 liii' 
 
INDIANS OF THIS REGION. 
 
 259 
 
 veins of magnetic iron, and also of lead, while a few sheets of 
 mica several inches in either direction were also picked up, but the 
 nature of the formation precludes the probability of finding much 
 of value in the mineral line either here or anywhere else upon the 
 coast. 
 
 The Mingan river, like many if not most of the other rivers along 
 the coast, extends inland about twenty miles when it reaches a 
 pond ; it is then connected by a series of ponds to some lake 
 whose altitude usually exceeds that of the surrounding country, and 
 whose waters, descending in a contrary direction, form gradually a 
 second river which flows to the sea by a similarly circuitous route to 
 that of its congener. 
 
 We descended the river much more rapidly than we had ascended 
 it, the cunent being very swift in places, and the wind also being in 
 our favor. From the shore we could see the summit of Mt. St. 
 John's, lying some fifteen miles inland in a northwesterly direction, 
 which mountain is said to be a little over fourteen hundred feet in 
 height. Directly inland the country is said to rise in successsive 
 steppes — if one might use the word in this connection, — to what 
 is termed the " height of land, " some five hundred miles inland, 
 where a chain of mountains, peculiar to the whole lower St. Law- 
 rence region, and northern Quebec, with peaks varying from one 
 to three thousand feet in height, continues in an eastern trend to- 
 wards the sea, which it reaches at the extremity of the Labrador 
 peninsula, near Ivucktoke, or Hamilton Inlet. 
 
 I had intended saying something further here upon the subject 
 of the Indians themselves, of this locality, but they do not differ 
 greatly from those of the whole coast, and all agree in the same 
 general characteristics. Mingan is the camping ground, so to speak, 
 for all of this class of people for several hundred miles of coast and 
 as many inland. Their chapel or church is also situated here, and 
 weekly worship is conducted by their chief, except at such times 
 as their priest makes them a special visit. 
 
 The religion of the Indians seems to be of a sort of Roman 
 Catholic order. I am told that it is quite similar to that of the 
 
n 
 
 'f 1 1 I 
 
 2C0 
 
 RELIGION OF THE INDIANS. 
 
 I't ; 
 
 French Canadian Roman Catholics, though having a distinct In- 
 dian characteristic which marks it at once as pecuhar to that class 
 of people. Their church, at Mingan, is a low, wooden affair, very 
 plain, and with only the necessary paraphernalia connecteil with 
 their worship within it. Outside and near to is the burying ground, 
 and above each tomb, — at least the majority of them — a simple 
 cross of stained wood marked the head, the size of the cross being 
 the sign of the importance of the individual in his tribe and village. 
 Hack from the burying ground, and some distance in the neighbor- 
 ing woods, was a large cross, and a bower of fir boughs a little tiis- 
 tance from it, representing some further ceremonies in their mystic 
 religion. Mere, I am informed, the i)eople go to bow and rever- 
 ence the cross, and to dance or weep within the bower, as the 
 occasion may require. With regard to the Indian religion, Mr. 
 I'utler says : — 
 
 "The worship of ihe Indians at Mingan is in accordance with 
 the teachings of the Romish church, I imagine. I have never heard 
 of any separate forn\ of their own. The cross and bower you speak 
 of are, I believe, a sort of memorial or votive shrine. There is a story 
 connected with it but I have forgotten what it is. " Mr. iJutler 
 has thus touched upon a genuine relic which, could it be recalled, 
 would ])robably interest every intelligent reader in the United States 
 and abroad, lie further says : " I supi)ose their religious ideas, 
 ai)art from the Roman Catholic Church, are very vague. " Much 
 lo my annoyance, the attendance of strangers was forbidden rather 
 than bidden to these their mystic rites, and I was unable to observe 
 them at their worship, although their priest visited them and per- 
 formed services while I was there. 
 
 'I'he respect shown by the Indians to their ilead results in a 
 species of religious superstition, peculiar, perhaps, to all communi- 
 ties of their race and color, that the bodies of their dead friends 
 must, at all hazards, be |)rotected from anything that would defile, 
 or in any way injure them, while any relative of the deceased re- 
 mains alive. They regaril their burying ground thcnr fmal home, 
 and even from far distant caiviping grounds they arc said to send 
 
HABITS OK LABRADOR INDIANS. 
 
 ?G1 
 
 delegations on long trips, ^^oth in summer cind winter, for the 
 purpose of making sure that intruders have not disturbed tlie re- 
 mains of their friends. 
 
 Although for the most part, these Indians, both as a tribe and as 
 individuals composing it, are quiet and peaceably disposed 
 creatures, on no occasion, prol)ably, would one of them kill a 
 white man, or one of another tribe, more quickly than if in any 
 way interfering with the bodies or burial places of any of their 
 tril)e. Socially, the Indians are (juiet ;;,id peaceable, if treated 
 with respect and kindness, though they are quick to take offence 
 at one who attempts to " bully " them or infringe upon their rights. 
 
 Of the Indians of Labrador I will now say a few words. There 
 are two principal tribes of Indians, not counting the Ks(iuimaux 
 of the extreme northern portion of the plateau, who inhabit this 
 region : the Montagnais who inhabit the coast, especially of the 
 river and G\\\( of St. Lawrence, and the Nascopies who dwell 
 principally inland and whose visits to the seashore are peri(jdii:al 
 and chiefly to secure su))plies by selling their furs and obtaining 
 credit on their probable winter's catch. Of the origin of these 
 tribes it is not necessary to enter here into any discussion. The 
 Montagnais are probably a pari of the .\lgon<iuin, anil the Nasco- 
 pies are undoubtedly a sub-branch of the same stock. 
 
 The " Montagnards," or " Montagnets," or " Montagnais " are 
 found in abund;<'v<' hR-tly along the shores of the lower St. Law- 
 rence, and thi.^' e- teiul more or less abundantly from (J^uebec, 
 even, as lar^ii Mingan, while stragglers occur even farther down 
 the coast. Tadousac was I'ormerly a great trading post of theirs 
 in the St. Lawrence, and even as early as the middle of the six- 
 teenth century when Jac(]ues C'artier visited this region these Indi- 
 ans flourished and were friendly, as ihey always have been, to tiie 
 French colonists. The Montagnais were always active in war, and 
 not so retired as their neighbors of the inlerio' From their resi- 
 dence (ju the coast we hear more of tlieu» an . their exploits as we 
 also know them belter. 'l"he .Montagnais were one of the first 
 Indian tribes to subscribe to Chri>Uanity, tliougli nothing has 
 

 I 
 
 fHT 
 
 NASCOPIES AND OTHER TRIBES. 
 
 ' 
 
 I 
 
 induced them to give up their nomadic existence and settle perma- 
 nently on reservations which have been offered by government. 
 They showed aversion to agriculture, and preferred to live in idleness 
 rather than to cultivate the soil, though many of their neighbors 
 of other tribes reared five plantations of maize. They were early 
 beset with Jesuit missionaries who labored in vain to civilize them. 
 Perhaps no one thing tended to demoralize the Indians then, as 
 now, more than the trafific in spirituous litpiors, which was every- 
 where encouraged rather than discouraged. 
 
 The " Ounadcapis," " Ounascapis," " Naskapis," " Naspapees," 
 "Nascopi," •* Naskupi," or " Nasr[uapee," as they have been 
 variously calle<i, formed at that period a distinct people inhabiting 
 the territory lying north of Lake St. Johns and extending towards 
 Hudson's Bay. In a communication from Mr. J. H. Trumluill, of 
 Hartford, Ct., regarding this tribe he says : "They speak a dialect 
 of the Cree language, nearly like, but not identical with, that of 
 the Montagnais Indians, of the same stock. The word " Nascopi " 
 is i)roi)erly the name for an " Indian man" (/. <?., 7'ir) in their 
 dialect, and that, in all Algontpiin languages, the name for " man " 
 and the verb " to stanil erect " are nearly related." In about the 
 year 1674 the Nascopies came in great numbers to Tadousac and 
 intermarried with the Montagnais, though since then the tribes in 
 general have been hardly distinct, yet a !l'w of each race still 
 retain the peculiarities of their tribe, so that there are still direct 
 descendants from the pure sto< k. In relation to the name Nascojji, 
 I found it in common use all along the coast for a pile or heap 
 of stones thrown up into some form several feet in height and 
 \isually placed on top of an island or neighboring height to mark 
 some ])osition, or imi)«)rtant spot or event. These heaps occur 
 everywhere, and are known, as I have said, by the name " Nascopi," 
 by the natives, also " American Man " by the sailors. 'I'hey are 
 common everywhere. The name does not occur in any of our local 
 dictionaries or encyclopedias that I can ascertain. 
 
 The Indians traded with the French as early as 1504, both 
 15as(|ues and Normans fre(|uenting their chief post at railousa<- for 
 
MONTAGNAIS — INDIAN STATIONS. 
 
 2G3 
 
 this puq^ose. In 1871 a census of the Montagnais Indians along 
 the north shore of the St. Lawrence rcsiUted in i ,685 ; there being 
 190 at Seven Islands, 552 at Rersamis, and 560 at Mingan alone. 
 For the Labrador division exclusively there were 1,309. In 1877 
 the Nascopies of the lower St, Lawrence numbered 2,860, though 
 doubtless through error, 1,860 would have probably reached nearer 
 the truth if indeed there were so many as that. In that same year 
 (1871) there were estimated in Labrador and (Quebec to be 2,500 
 Nascopies, and 1,745 Montagnais, Since that lime the number of 
 families have constantly diminished. Some have renounced their 
 trading auc. trapping voyages and settled in comfortable cabins and 
 are turning their attention to farming and the raising of such crops 
 as potatoes, turnips, oats, etc., and secure about enough hay to 
 feed a few head of cattle during the winter, but the traders are 
 fast undoing what has been done, and sell them goods as well as 
 liquor largely on the credit principle so that, so to speak, they are 
 demoralized almost as fast as, :♦" not Tister than, they are moralized, 
 (iodbout river was formerly a grca , Milan station and gathering 
 place ; not more than a dozen familie, ow reside there, and they 
 only in the summer time. Farther down the river, at Seven Islands, 
 about five times that number gather in tl :; summer, descending by 
 the Moisie river, and strive to recupcrai ' from the half famished 
 condition in which scarcity of provision: in the winter time has 
 left them. Moisie itself, at one time (juit a rendezvous for these 
 families, is now almost deserted for this »lace (Seven Islands). 
 Mingan, now the favorite resort of this tribe is a good location for 
 their wants. It has been previously describe 1. The Indians meet 
 here in the summer, and have a general re ting time. Many of 
 them, if the season's " catch " has been good buy them boats and 
 barges for $80 to $150 and make hunting ex» irsions both up and 
 down the coast, or to the island of Anticosti thirty miles distant 
 across the channel. In the year 1878 there we e about 80 families 
 and 375 people, young and old, stationed hei ) for the summer. 
 The year I was at Mingan there were about he same number, 
 Natashipum is another station of these Indians ( uring the summer 
 
ill 
 
 
 I ' n 
 
 264 
 
 INDIAN SETTLEMENTS. 
 
 here bein[; a post of the Hudson's bay company here also, 
 though at the extreme point of the mainland on the east side of the 
 channel. Their settlement here is entirely by mishwaps arranged 
 for temporary accommodations only during the summer months, 
 and rarely more than fifty families and two hundred people, old and 
 young, assemble here from their inland excursions to sell their furs 
 and to recuperate. Musijuarro, though ftjrmerly a great Indian 
 rendezvous, has now become deserted for Romaine or Olomano- 
 sheebo, which is really a most delightful and picturesque place, — 
 at least so it seemed to me as we entered its snug and (juiet harbor 
 one beautiful evening, and viewed it with the lights and shadows of 
 the sun's last rays upon it. Romaine river is now a great highway 
 for the Indians who visit the interior or those who descend to the 
 seacoast. 
 
 St. Augustine, about thirty miles to the westward of Ilonne Esper- 
 ancc, was formerly the great resort of the Nascopie Indians. Here 
 for a long time the Hudson's IJay (.'ompany kept a nourishing post, 
 which was afterwards deserted by them, and a generous, honest 
 dweller of that region was allowed to take possession of it, 
 who now supi)lies the Indians who come from the interior, to this, 
 the only post of the region, for a distance of many miles in either 
 direction. Many of the Indians wintering here came directly 
 across the country even from Ungava and the shores of Hudson's 
 Bay itself. About 1871 there were as many as one hundred and 
 twenty families of Nascopies at this i)lace, with about a third as 
 many Montagnais ; the two elements an; so intermingled now that 
 the numbers are ])retty e\t nly divided between the two tribes. It 
 has been said of the Nascopies in general, that "aUhough intelligent 
 they are yet very superstitious, believe in dreams, in their 'jong- 
 leurs,' or medicine-men," etc. ( )flat' years the Indians have en- 
 camped largely in the islands and mainland near Bonne Espcrance 
 where Mr. W'hiteley and t)thers furnish them with supplies to con- 
 tinue their himting and trapping in the interior the following winter. 
 There are thus about three hundred and seventy-five families, and 
 1,700 people leading a nomadic existence, and dwelling about and 
 
INDIAN TRADE. 
 
 2G5 
 
 around the various resorts and islands from Quebec to Belle Isle. 
 'J'hese Indians are seldom if ever worth over a few hundred dollars, 
 their furs are noi abundant, they themselves being thus obliged to 
 eke out a miserable existence, starving to death in the winter to 
 procure food for their families, and secure enough furs to pay for 
 their summer's provisions, at which season they are usually obliged 
 to glut themselves to gain sufficient strength to pursue their hunt- 
 ing the next winter, elsf they would surely starve and die in the 
 midst of plenty, from their previous winter's want. 
 
 Another great source of misery to tho Indians is credit. This 
 effectually bars them from enterprise, and prevents their advance- 
 ment in every possible respect. Tiiey cannot consider themselves 
 free when once they have become involved to any trader or dealer. 
 The uncertainty of their business makes it all the worse. If a year 
 comes when game is scarce, and little fur can be obtained, they 
 come from the interior disheartened, for who will give them \ixo- 
 visions if they have no furs? At last some one furnishes tiiem with 
 enough provisions, to keep them from starving, on credit of the next 
 winter's ''catch." These are soon eaten up. and the same trader 
 finds himself ol)liged, to save himself, to fit them out for the coming 
 winter, with the proviso that he shall be paid from liie results of 
 that catch. The Indians are obliged to promise or starve. The 
 next year is a poor year also. The Indians take little fur, and, 
 knowing that if they go to the trader who fitted them out they must 
 give him all they have with no pros[)ect of any return, they come 
 to some other pari of the coast, and try the same experiment with 
 another party ; thus they go from place to place, leaving debts 
 wherever they go, with no prospect of ever paying iheni, and the 
 fear that soon every avenue will l)c blocked to their approach. The 
 trailers, eiiually in desi)air, are but just opening their eyes to the 
 situation, and finding the only solution of the difficulty in a com- 
 promise in which each side shares. 
 
 When the ice breaks up in the s])ring, usually the last of Afav, 
 the Indians seek the coast with their Airs to trade and to re( uperate. 
 Many of the traders practise all sorts of arts whereby to deceive 
 
266 
 
 "BARTER TRADE." 
 
 the Indians both as to the quality of tiieir furs, and that of the ar- 
 ticles which they offer in exchange for them ; the majority, however, 
 ask only a fair price in exchange for cash, and the charges seem ex- 
 orbitant only as the proj^osed credit is large or doubtful. Flour 
 is generally $7.00 to $9.00 a barrel ; lard, 20 cents, butter, 30 to 40 
 cents, and ship biscuits, 10 cents per pound ; pork a barrel, $25.00 
 to $28.00 ; cotton, 25 to 40 cents per yard ; molasses, 60 to 80 
 cents a gallon. In exchange, beaver skin is worth $1.50 to $2.50 
 a pound, martins $1.50 to $3.00 a piece, lynx, $1.50 to $2.00, 
 bear, $5.00 to $7.50, fox, $1.50, silver fox, $15.00 to 30.00, cross 
 fox, $7.00 to $9.00, and mink about $1.00. This will show the 
 genera,! average of the trade, varying slightly according to *.he 
 season all along tiie St. Lawrence and Labrador peninsula. 
 
 The summer visits of these Indians are dreaded by all fishermen. 
 They loam at will, fishing or shooting, visiting salmon and seal nets 
 at night, and stealing the " catch" of their more fortunate white 
 neighbors, while they scare away all species of game and render 
 others as miserable as they themselves are, till fall again disperses 
 them towards their winter hunting grounds in the interior of this 
 wild, bleak, desolate country. In 1881 the value of their whole 
 hunt was only about $20,000. The same year their numbers all told, 
 from Quebec to Iklle Isle, were about 11,000 young and old; this 
 was probably too large an estimate by several thousands. One can 
 thus see that the Indian population of this region is anything but 
 an easy one to manage. They come down from the interior in 
 small bands to the various places named above, in a half starved 
 condition, and attemi)t to make up for their loss in hunting and 
 trapping, by begging from any antl all that are more fortunate than 
 themselves, and by obtaining large credit from the traders. The 
 government can do little to help them. If one year it sends sup- 
 plies, the following year the Indians come down en masse to their 
 rendezvous and loaf around doing nothing to earn their own living, 
 while waitnig for a similar shipment from the government. If this 
 comes it dcmorali/.es ihcm more than ever, and if it does not rowe 
 they have then lost all the benefit of the fishery, where they nii.ulit 
 
MODE OF LIFE AND DRESS. 
 
 267 
 
 have reaped a good harvest. A few of them are furnished with 
 nets, hooks, and hnes, and work industriously during the summer 
 months. The soil, at the places where they remain during that 
 season, is unfit for cultivation, and as none of the" whites attempt 
 cultivation, the Indians would be hardly more fortimate. One 
 year the government sent them a number of bushels of potatoes to 
 plant, as an experiment. The Indians cut them up at once, and 
 thanked the agent for being so good as to send them supplies, hop- 
 ing that they would send more the next year. Hunger has re- 
 duced them as a tribe, though a few are haughty and troublesome 
 at times. It is impossible to civilize them to any great extent, 
 and there are but few schools, and these only in the lower prov- 
 inces where there are also a few reservations. 
 
 There has recently been appointed, by the government, an In- 
 dian agent who is now hard at work trying to better the condition 
 of these poor, unfortunate people, and make it possible for each 
 family to become sober and industrious and earn its own living in 
 a thrifty way; but their nature rebels against most all treatment, 
 and they prefer the nomadic existence of their fathers, or the uncer- 
 tainties of the chase, and are delighted to roam at will and fish the 
 stream, thus gaining an uncertain sort of existence from year to 
 year until at last they disappear and are heard from no more. 
 Disease very seldom attacks an Indian. Such diseases as small- 
 pox and measles arc the most prevalent, — for the former they are 
 vjKTcinated freely l)v the government agent while he is on the coast. 
 The aged and inftrni are disposed of as speedily as p«>ssible, being 
 lieft to starve n .;ic or turned off upon anybody who will uxkjbt- 
 take to tare Air them. CWten in the summer time large t , 
 
 occirr, but these are riivlv caused by he Inri !i- i- 
 to think. \ti iTidlian is very circtul v.i this \> jnci; i: 
 
 l>e a great >lTenrc to tort- guiltA' of wilful dcstru .luu at tie "^l 
 ing grounds." 
 
 I tiaw noc say moch concerning the dhB»«fff*feF Irrrfewv; T^»««r 
 wear anythsiK: ami everything. OKI ganneat; -in« • '^ ^ "-' 
 ch.. *. ami h^.u\. .roH-n, white, or black; kflteroraiktai i >«>(.. 
 
Pf^ 
 
 a 
 
 
 . 
 
 ■ 
 
 It 
 
 ■ li 
 
 1 
 
 h 
 
 i' 
 
 4 
 
 2G8 
 
 LEAVING MINGAN. 
 
 rubber boots or moccasins, according to the weather, state of their 
 purse or their credit. The squaws wear caHco, or woollen dresses 
 of any shade, quality or pattern, and a peculiar, long, peaked cap 
 usually of red flannel cloth. Dressed -^leer-skin waistcoats are 
 common with the men and are very warm ; moccasins of all sorts 
 and shapes are made, and used or sold. But this extraordinary 
 race demands far greater attention than has been heretofore given 
 them, or that can possibly be contained in these fragments of their 
 history, and it is to be hoped that the government which has so 
 far acted judiciously in the matter will be able to do some perma- 
 nent good to this unfortunate and destitute people. 
 
 Wednesday, the 8th, We left Mingan, on our return trip, yes- 
 terday, with a good wind and fair sky, and to-day we made Romaine 
 river or Olomanosheebo as the Indians call it, another now de- 
 serted i)ost of the Hudson's Bay Company. This location, (Irand 
 Romaine, as it is called, has recently become a great rendezvous 
 of the Indians. The place was crowtled. There must have been 
 thirty or forty families camping there, just on the eve of their de- 
 parture on a hunting excursion, so that we were too late to visit 
 them. They were in the midst of a celebration of some grand fete 
 day, and were saluting their priest, who had just arrived to conduct 
 the ceremonies of the occasion. They continued an almost imin- 
 terrupted firing of guns for about two hours. In the evening all 
 attended service ; ten minutes after every Indian had left the place, 
 and it was as still and deserted as if no one had ever existed there. 
 Romaine is as pretty a i)lace as its name is musical. It has a se- 
 cluded yet snug ai)i)earance, while the hills form a sort of barrier 
 within which the natural beauties of the place unfold themselves. 
 The rocky islets on the outside enclose a safe harbor. 
 
 Tuesday, the 14th. We reached Natasluiuan to-day, and there 
 found letters from home awaiting us. Here we found, surroimtled 
 by petty traders, a large French steamer that was coasting the 
 shore for fresh salmon, which, first uiulergcjing a freezing process, 
 were packed, to be transjjorted to the I'Yench coast where they 
 are retailed at a large i)rofit. The captain luul hail no cause to 
 
NATASIIQUAN AGAIN. 
 
 269 
 
 complain of the " catch," and all the men were hard at work cu- 
 ring and packing a lot of splendid large fish recently brought in by 
 one of the fishermen from a short distance down the shore. The 
 captain was an intelligent and excellent apfHraring fellow, and 
 extended to us the usual courtesies, for which the French generally 
 are so well noted, when he found we were Americans and strangers ; 
 and I should do him injustice did I not make i)articular mention of 
 the excellence of his table, to which he cordially invited us, and 
 which invitation we as cordially accepted. After dinner we were 
 shown all over the steamer, and she was indeed a lx;auty. She 
 was almost new, and her engines in perfect trim, while all her ap- 
 purtenances showed the signs of newness and solidity. Her tonnage 
 was a little over one thousand. Though the captain showed us 
 everything freely, he would not describe to us the new freezing 
 process to which he subjected the fish he presen-ed, claiming it to 
 be a new invention and not that of vaporization of ether or any 
 kindred process in use at the present time. One of the steamer's 
 tug boats was put at our disposal for an excursion the next day, 
 but the rain deprived us of this very great anticipated pleasure. 
 
 Saturday the i8th. Quite an excitement was caused this even- 
 ing by the arrival of a jewelry peddler. He had sold a large <iuun- 
 tily of this artificial and counterfeit material, and now was eager to 
 get away from the village by the (luickest possible conveyance, and 
 I cannot wonder if a sample of his goods tliat came under my eye 
 is a fair sample of all of which he disi)osed to this honest-hearted 
 community. It was that of a very poorly plated ring worth about 
 a shilling that was sold for ten sliillings. The young fellow who 
 purchased it intended it for his young lady, but, finding that it 
 tarnished the first night his rage may be easily inuigined. 
 
 At Natashquan we obtained a fresh supjily of provisions, and a 
 few pounds of most excellent fresh maple sugar ; it was put 
 up in S(iuare cakes of five pounds each and sold at the surprisingly 
 cheap price often cents a pound. 
 
 Our next stopping place was at Agwanus river, twelve miles 
 above Natashquan and but a few miles below Nabisippi, which 
 
w 
 
 270 
 
 AGWANUS — NARISIPPI. 
 
 r 
 
 latter place we could plainly see with the naked eye. Agwanus is 
 a very pretty place. A beach of (juite pure sand is crowned by a 
 number of grass-plats upon which several houses have been built 
 while the wiiole is backed by the cliffs and crags of the neighboring 
 hills. We sailed by a number of small, low islands or rather rocks, 
 for there was scarcely any vegetation upon them, the breeding 
 place of innumerable terns (probably S/crna macroura), which 
 fairly swarmed everywhere we went. They took good pains, how- 
 ever, to keep just out of gim-shot, so that we did not procure any 
 of them. 
 
 It was dusk when we passed through this region, and as wc 
 slowly glided along the channel among the rocks, we watched the 
 beauties of the scene. Low patches of green topped the brown 
 rocks, and were set off by the disi)lay of the darker green of the 
 spruces farther inland and the shadows on the still darker green of 
 the cloud-shadowed rocks still farther away. Soon we came ujjon 
 a small cabin, snugly tucked away in a sheltered place among the 
 rocks near the shore. Close by here we anchored, while at our 
 right another small island swarmed with gulls and terns at which 
 we practised shooting until dusk. This place we learned was 
 Washtawooka bay. The harbor protected us perfectly on all sides, 
 and here we anchored for the night. 
 
 Thursday, the 21st. Although the longest day of the year, we 
 to-day made our shortest run, being but six miles, only to return 
 to Nabisippi, another now deserted post of the Hudson's Hay 
 Company, for the night. We spent part of the time lying to about 
 a mile off shore and shooting at the gulls, of which large numbers 
 surrounded us. It was the species known as Bonaparte's gull 
 {^Chra'Cticephalus pJiihiiic/phia) which abounds about the shoal 
 waters and fishing grounds everywhere along this part of the coast. 
 We saw here a most beautiful but rare sight, — it consisted of 
 large body of codfish "schooling," as it is called, in reality playing, 
 upon the surf:ice of the water. The water would be covered for 
 acres with the heads or bodies of these fish as they dashed madly 
 about or scooted forward on the surface of the sea, often in per- 
 
RETURN TRIP — INDIA?' VILLAGES. 
 
 271 
 
 feet line and in perfect time. Again they would disappear as sud- 
 denly as they appeared. I have seen them at times leap several 
 inches clear of the water. Again they would form in long ranks and 
 rush through the water with such velocity as to make it actually hiss 
 as it closed i)ehind them. Sometimes they would suddenly dis- 
 appear and again as suddenly appear headed in some other direction 
 and proceed with their " playing " as before. They often sport in 
 this way for hours, api)earing in thousands and tens of thousands, 
 and then disappear as suddenly as they came not to be seen again 
 during the season. At this, the "old salts" say " they have taken 
 to deep water again." 
 
 Wednesday, the 23d. We started off early this morning with a 
 fine breeze, on our return trip. (We had visited, but did not stay 
 long, at the Roman Catholic settlement of al)out one hundred and 
 sixty houses, at Ksfjuimaux Point, about twenty miles east of Min- 
 gan. I simply mention the location as it is an important one in 
 sealing industries, though otherwise it does not, I believe, attract 
 particular attention.) Farther to the eastward the small island of 
 St. Oenevieve (pronounced here ^^cnyunr) forms a prominent 
 landmark, as does also the locality of Watcheeshoo (pronounced 
 7c>atch-a-s/ioe) which is near Manicouagan l)ay. Close by is Sad- 
 dle hill, the highest part of this neighboring coast, being about 
 three hundred and seventy-five feet above the sea, and about sixty- 
 five miles to the north and west of which lies Mt. St. John, the 
 highest peak on the North Shore at this point, being 1614 feet 
 high. A few other Indian villages were passed, such as Washatna- 
 gunashka bay, or Mushkonialawee, Pashasheebo, as also Peash- 
 tebai, and Appeeletat, all of which names, strange to say, really 
 sound (piite musical when pronounced properly by one accpiainted 
 with their Indian sound. 
 
 Friday, the 24th. We lay to, awhile this morning, off the Fox 
 islands, and filled several pails and buckets with murrcs' eggs 
 {Lomviu troiie), v.'hile with our guns we shot about a hundred of 
 these birds in less than an hour, and yet we left them flying as 
 thickly over and by the island as when we had first landed. \Ve 
 
IMAGE EVALUATION 
 TEST TARGET (MT-3) 
 
 1.0 
 
 I.I 
 
 m 111^ 
 
 ^ m 
 
 M 
 
 2.2 
 
 36 
 
 i: 
 
 1^0 IJI 2.0 
 
 1.8 
 
 
 1.25 
 
 1.4 
 
 1.6 
 
 
 M 
 
 6" — 
 
 
 ► 
 
 v: 
 
 ^ 
 
 //, 
 
 m 
 
 e. 
 
 ^ 
 
 w 
 
 A, 
 
 *a^ oj^f' 
 
 >' 
 
 O 
 
 /a 
 
 ^» y 
 
 7 
 
 /A 
 
 Photograpnic 
 
 Sciences 
 Corporation 
 
 i\ 
 
 # 
 
 «■ 
 
 -u 
 
 S5 
 
 :\ 
 
 
 <^ 
 
 \ 
 
 <^" 
 
 
 o^ 
 
 23 WEST MAIN STREET 
 
 WEBSTER, N.Y. 14580 
 
 "16) 872-4503 
 
S" c^ 
 
 * 
 
 
 Us 
 
272 
 
 GUNNING — FORT ISLAND AGAIN. 
 
 boiled the eggs and found them excellent eating. They are not 
 quite as rich in flavor as hens' eggs, but certainly fully equal them 
 in quality. With this addition to our stock of provisions we had 
 several new dishes, among them a most delicious rice pudding. 
 In the evening we passed the highlands of Meccatina, nearly six 
 hundred feet above the sea, and also those of Mutton bay, about 
 seven hundred feet high, a small settlemeat where dwells a mission- 
 ary who ministers to the people on this part of the coast ; and 
 anchored in the snug harbor of Great Meccatina, where we lay all 
 night. Here we again saw the magnificent comet, we had occa- 
 sionally seen before, in a northeast by north direction, and appar- 
 ently about 30° above the horizon. It was about 1 1 o'clock p. m. 
 that I made this observation. 
 
 In the morning we proceeded eastward with a fine breeze. As 
 we passed, all the harbors were full of Newfoundland fishing vessels, 
 many of them from Harbor Grace. Bale des Roches was filled 
 so full that we could scarce obtain anchorage. I found on many 
 of the rocks distinct and well defined glacial scratches, and noted 
 other peculiarities of the location. We remained here all day on ac- 
 count of the intense fog everywhere. The next day we -lanaged, 
 after beating about for some time, to reach Old Fort island again 
 whence we had started, after one of the most enjoyable trips I ever 
 experienced. 
 
 ir 
 
AT OLD FORT ONCE MORE. 
 
 273 
 
 CHAPTER XVII. 
 
 Affairs at Old Fort Island — The fishing season — Thunder storm — Arrival 
 of vessel which is to take us home — Our trip in her to L'Anse Loup and 
 scenes at intervening places — "Off for home" — Double reefed fore 
 and main sails — Island of Anticosti — A hurricane — Quebec and 
 home. 
 
 Affairs at the Island had not changed much since I had left it, 
 and everything moved on in the same quiet, well ordered style. 
 There had been an arrival which pleased the children greatly, in the 
 shape of a young calf, in the bam, and most of the day was occu- 
 pied by them in watching its movements, which they did with the 
 greatest of interest, reporting hourly at the house. There were 
 three children in the family, but one would easily have imagined 
 that there were at least thirty, by the succession of continual ap- 
 pearances and as sudden disappearances at the door, as each came 
 to relate the dispatches from the barn. It, however, afforded them 
 occupation, which was the main point. 
 
 The animals about the place, not counting in the dogs, con- 
 sisted of a bull, a cow with its calf, and two goats — one of which 
 was a billy. The goat furnished milk as well as the cow. Many 
 people along the coast keep goats, and prefer the milk for the chil- 
 dren to that of the cow. Both the billy and the bull were turned 
 loose on the ground with simply a board before their eyes, and one 
 of their fore feet tied to a rope passing around the horns. In this 
 manner all roam together : dogs, children, animals and all. 
 
 It really seemed quite refreshing to be fed once more on good 
 home-made bread and butter, to drink fresh, rich, creamy milk, and 
 to have the luxury of real cream in one's tea and coffee. The gun 
 18 
 
sm 
 
 ■Hi 
 
 
 274 
 
 AFFAIRS AT THE ISLAND. 
 
 If 
 
 Still afforded the roast duck of the noon or evening meal ; and 
 many a morning have I been out before breakfast and brought 
 home a bag literally full of plover and beach birds for the meal. 
 Good cheer, and plenty of it, was never wanting now. Soon the 
 men began to trim their boats for the fishing season, now nearly 
 upon them, and all around us was activity and life. A curious ex- 
 ample of how everything is turned to use occurred recently. There 
 had been a wreck somewhere off the coast, and several dead sheep 
 had been picked up just off shore. Of these the skin was carefully 
 saved, the tallow melted down and turned into homemade candles, 
 while the carcasses were salted down to serve as food for the dogs 
 the ensuing winter. The candles were, to be sure, a rather poor 
 apology for such articles, and their wicking simply pieces of cotton 
 cloth, but they answered every purpose. 
 
 The fishing season was now well advanced, and a trip to Bonne 
 Esperance assured me that the people generally along the coast 
 were in "great humor" over the unusually large "catch" offish 
 which was everywhere reported ; but our chief delight centred 
 in the letters and papers from home which were found await- 
 ing our arrival, and glad indeed we were to get them. In the 
 evening we were treated to what at home we should call a mild 
 thunderstorm, and were informed that it was " the most terrible one 
 that had ever visited the coast. " Such storms are, indeed, quite 
 rare here, and there is seldom even what we should call a " hard 
 rain. " The next day the sky cleared most beautifully, and we 
 had one of the finest days I have ever seen on the coast. 
 In the evening one of the fishermen brought in a seventy-five 
 pound cod, and it was an enormous fellow, the largest I had ever 
 seen. The day following, as it was warm and pleasant, I took my 
 first and last plunge bath and swim, off the dock, in Labrador water. 
 It is safe to say, that in those five minutes there must have been a 
 change of temperature in my body of at the very least one hundred 
 degrees, in the shade at that. I never attempted it again. 
 
 Thus we pass away the time. Day follows day in quick succes- 
 sion, while all is pleasure both within and without ; but all pleasures 
 
 i 
 
TRIP DOWN THE COAST. 
 
 875 
 
 have an end, and too soon we find the vessel that is to convey us 
 home already in the harbor awaiting us. The Captain is to make 
 a two weeks' trip " down along " visiting, for purposes of trade, the 
 various places over which we had been before, yet we preferred 
 to join him, and so by evening we were packed and " all aboard " 
 for our trip. We passed Salmon Bay, everywhere abounding in 
 glacial evidences, rounded its trough-like hollow, which appeared 
 as if gauged out from the land toward the sea, as it undoubtedly 
 was, and noted everywhere its peculiarities and attractions. We 
 visited Blanc Sablon in full summer activity, and saw, even now, 
 early August, snow-clad hills in the distance beyond ; while nearer 
 to green slopes and verdure were everywhere scattered by Nature's 
 profuse hand. 
 
 I took my gun and started toward the distant hills. I climbed 
 the crests close to the beach, and at the head of the bay. They 
 were huge sand heaps sparingly grassed over, and reaching to two 
 hundred or more feet above the sea level. Again huge depressions, 
 nearly as deep as the hills were high, lay before me. These ele- 
 vated tablelands, and smooth even valleys everywhere intervening 
 stretched forward and inland in all directions. From the highest 
 of these tabled hills, what a view burst upon my sight ! On one 
 side a long extent of country (apparently rich pasture land for herds 
 of cattle, though we searched for them in vain, and listened in vain 
 for the tinkling of the cowbell) receded to the hills on the right, 
 while all was enveloped in dark shadows upon the green ground- 
 work from the hills and the clouds. The whole outline of the pic- 
 ture was that of a large funnel ; the tube extending in the distance, 
 and the mouth occupied, in greater part, by a magnificent large 
 pond, the outlet of which, draining through the centre of the tun- 
 nel, followed the tube, and, lost in the distance, eventually found 
 its way, dashir.g down the rocky Bradore heights, into the sea. 
 
 On the other side Blanc Sablon harbor and the sea beyond came 
 in view. To the west were Long Point and L'Anse Couteau, while 
 Wood and Greenley islands occupied the centre of the picture, and 
 the fishing boats (I counted two hundred or more) everywhere dot- 
 
 f .• 
 
i 
 
 li 
 
 h 
 
 t 
 
 •■■■ 
 
 1 
 
 I , 
 
 276 
 
 FORTEAU — INDIAN IMPLEMENTS. 
 
 ted the scene, like flies upon the wall. My game bag was empty 
 upon my return to the ship, notwithstanding several large flocks of 
 curlew at which I might have shot, but my mind was filled with this 
 glimpse of one of the most beautiful of Nature's pictures, so that I 
 minded not the revilings of our captain at sight of the empty bag. 
 
 The next day we reached Forteau. A walk around the base of 
 the hills and cliffs on the northern side of the harbor, through 
 growth of tangled spruce and fir where an occasional bluebell cr 
 bed of Cornell peered at us from some sheltered retreat, brought us 
 to the house of a pleasant and homely family who cordially wel- 
 comed us. The following day I took a short walk inland, to what 
 are here called "the deserts, " which are elevations and depressions 
 of low sand dunes, whose summits are more or less grassy. They are 
 from ten to fifty feet high, with corresponding depressions of dry 
 sand. We found in many of these hollows several well formed 
 spear and arrow heads, and could have picked up a bushel basket 
 full of "chips," which were everywhere abundant. The story is 
 told, that when the Esquimaux were driven from this part of the 
 coast by the mountain Indians, who at the time nearly extermina- 
 ted that curious race, all this region was the scene of the Indian 
 descent to the plains from the highlands, hence the chief place of 
 the engagement between the two parties ; here the Indians made 
 and used their arrows and other implements. The chippings that 
 we found were mostly of a white, chalcedony-like flint rock, not at 
 all like anything that we could find along the coast, but said to be 
 " common inland. " A few quartz and ordinary flint pieces were 
 found, but they were comparatively rare. But we were soon forced 
 to leave this interesting locality. Returning to the house we spent 
 the evening, while yet light, in watching the feats and gambols of 
 one of the young dogs as it leaped into the water from a pier over 
 fifteen feet in height, and even dove for stones thrown at it to a 
 depth of nearly four feet, bringing the stones to the surface between 
 and over its fore paws ; then, as the evening had advanced, we re- 
 turned to the house and retired to rest. 
 
 The next morning several of the boys joined us in a walk across 
 
INLAND POND — OFF FOR HOME. 
 
 277 
 
 the limestone ledge above the house to Schooner Cove, a small 
 western branch of L'Anse Loup. Here one of the Newfoundland 
 merchants has a well-to-do fishing establishment, as also a small 
 storehouse for the sale of supplies for this part of the coast ; here 
 we were to meet our vessel which had preceded us by sea. The 
 walk was over a fertile limestone ridge about two hundred feet high, 
 where still further evidences of glaciers could be seen about us. I 
 found the dry beds of several ponds as I walked, and in one place 
 one small pond still existed. It was situated in a hollow formed in 
 gradually sloping hillocks. Considerable vegetation grew around it, 
 and it was altogether quite a picturesque spot. Our guide then 
 told us its history. He said that the water of the pond was fresh, 
 but yet so deep that after repeated trials no one had ever been 
 able to reach the bottom with the sounding lead. It was supposed 
 to have an underground connection with the sea ; but, he further 
 explained, that the water itself was nol salt since the fresh water, 
 the results of springs and drainage was lighter than the salt water, 
 hence would and did not mix with it. At times, he said, the sur- 
 face was agitated with heavy waves, and covered with a frothy sub- 
 stance like that which the sea waves cast upon the beach. The 
 pond appeared to be about seventy-five feet below the height of 
 the surrounding plateau which averaged two hundred and forty 
 feet above the ,iea level. 
 
 Saturday, August 13th. We are fairly off ibr home, I mean our 
 United States home, to-day, having left our farthest point north- 
 east, and started towards Quebec. Oh, what beautiful evenings ! 
 What superb weather ! How long will it last ? Shall we ever 
 visit here again ? These were the exclamations and questions that 
 escaped us as we ploughed on, — on through the waters towards 
 home. 
 
 Sunday we reached Old Fort Island, received on board the re- 
 mainder of our luggage, and in a few hours were off again. To- 
 wards night it came in foggy, and in the morning it rained quite 
 hard. In this state of the atmosphere we missed the proper entrance 
 to the sea that we should have taken, and passed on through what is 
 
i-'if ' 
 
 i 
 
 if 
 
 278 
 
 INCIDENTS OF THE TRIP HOME. 
 
 called the little rigoulette, at the bottom of which enters Kekarp- 
 wei River, and where there is no exit. Never shall I forget the 
 clearing up of the fog and the disclosing to us of one of those 
 charming Labrador scenes, so characteristic of the locality ; low 
 shores, sandy beaches, grassy slopes and tops, and a forest of low 
 and tall spruce and fir, intermingled with cliffs and rocks every- 
 where. The passage was so narrow we could with difficulty haul 
 the ship around without her touching the boom on one shore and 
 the stern on the other. Narrow and shallow as the passage was, 
 we succeeded at last, and returned to the opening that we should 
 have taken at first into the sea. I do not believe that there was a 
 person, however, save the captain and the crew, who regretted the 
 extra time spent in this charming region. At length we were once 
 more out lo sea, and right merrily we sped onward, with " de//e 
 brise, " as the captain called it, towards Quebec. 
 
 In passing Whale Head, one of the numerous small fishing sta- 
 tions on this part of the coast, we found a small sail-boat of Indians 
 following in our wake, a short distance off. The captain, wishing 
 to send a letter ashore, produced a large bottle, placed the letter 
 within, carefully corked it again, and, with gestures calling the at- 
 tention of the Indians to himself, threw it into the sea. We 
 watched eagerly, and soon had the satisfaction of seeing the Indians' 
 boat secure the bottle. The captain assured us that the letter would 
 reach its destination, and that messages were often transported in 
 a similar jnanner from place to place, when a favorable wind ren- 
 dered it impossible to stop the ship, and some sail-boat was near 
 to pick up the bottle. 
 
 Night soon set in, and with it the breeze freshened so that soon the 
 order was given to reef sails ; this was succeeded by an order 
 to double-reef both fore and main sails; and on, on we sped, — 
 through the treacherous Gulf of St. Lawrence towards home, the 
 water fairly sissing as we cleaved it, and the wind fairly hissing 
 as it slid from our sails. I think our Captain told us that we had 
 made better time, in a given time, than he had ever made before 
 on any trip ; I fully believe that it might have been true. Wednes- 
 
 ^ 
 
 \ 1 
 
 t' 1 
 
 •V ^ 
 
OFF ANTICOSTI — QUEBEC. 
 
 279 
 
 day morning the 24th, we shook the reefs from our sails, and in 
 the evening were anchored off Natashquan. A boat carried off 
 the mail and returned with the up mail, and we were again mov- 
 ing. By evening of the next day we were off Anticosti Island, this 
 scene of so many mournful wrecks and disasters. 
 
 The sailors about this part of the coast say that the Gulf of St. 
 Lawrence ends and the mouth of the river begins just off the west- 
 ern end of this Island. Most writers make the dividing part at the 
 bend farther up the river ; which is right is not a point for us to 
 settle. Off Anticosti the nearest land is Mingan, which has been 
 so fully described. On the south shore its nearest point is Cape 
 Gaspe. The island itself is about one hundred and twenty miles 
 long by thirty wide. It is well wooded in portions of its surface, 
 though it is generally barren and uninhabited save by a few bears 
 and other wild animals. Very little and very poor water can be 
 found upon it, while the two government lighthouses upon either 
 extremity alone represent the life of the island. Indians occasionally 
 hunt for bears and procure birch bark for their canoes along its 
 shores and in its interior. Upon its southern side are dangerous 
 reefs. Heavy gales occur frequently near by, and many are the 
 tales of horror told of wreck and disaster upon its coast, but I 
 hasten on. On Wednesday evening, the 31st, a gale struck us. 
 Fierce and heavy it blew. Sharp thunder and lightning were followed 
 by alternate puffs of wind, cold and hot as if from the mouth of a 
 furnace. The squall fairly turned into a hurricane. Before we 
 had gone to rest the gale subsided, the dark clouds gave way, the 
 rain ceased, and the clear evening predicted a beautiful day for the 
 morrow. The morrow arrived : it found us once more at Quebec. 
 
 ij^Jj' 
 
w 
 
 I 
 
 a 
 
 it- 
 
 280 
 
 THIRD TRIP TO LABRADOR. 
 
 CHAPTER XVIII. 
 
 Third voyage; summer of '82 — Puffin-shooting — Dredging — Bad weather 
 — Main boom breaks — Chateau and Temple bays — Places of interest — 
 Mines and minerals — Aurora and phosphorescence — Icebergs — Fox 
 harbor — Battle island — Indians and Esquimaux — Indian vocabulary — 
 Square island — Dead island — A water garden — Triangle harbor — 
 Homeward bound — Notes on Dutch and Esquimaux settlements. 
 
 I HAVE nowneared the end of my explorations. Yet, strange to say, 
 I find myself once more embarked upon the perilous enterprise 
 of again stemming the sea and stormy Gulf of St. Lawrence for a 
 further extension of my researches upon the Labrador coast. This 
 time I start from Boston, July 13, 1882, in a small schooner of 
 about one hundred tons, the " Polar Star " by name, with a party 
 of some dozen gentlemen, young and old, who, with me, are also 
 bound for this bleak and rocky, yet picturesque coast. The party 
 consists of ornithologists and mineralogists, conchologists and 
 ichthyologists, photographists and pleasurists. All in an amateur 
 sense however, except possibly the pleasurists, and they might 
 almost be said to outweigh everything else and form the main 
 feature of the party. In fact it was a pleasure party and nothing 
 else, of which guns, ammunition, and fishing tackle formed the chief 
 topics of conversation from the time we left until we returned. 
 
 I will not stop to describe our journey to Halifax where we pro- 
 cured a relay of supplies for our voyage ; or of our stop on the 
 Cape Breton shores, at several points of interest ; or our journey 
 through Canso to the stormy Gulf of St. Lawrence, where, after 
 passing the famous Bird Rocks near the Magdalene Islands, just 
 
 
A GALE — MORAVIAN SETTLEMENTS. 
 
 281 
 
 east of Anticosti, we encountered one of the shortest but severest 
 gales that the captain had ever met, that nearly plunged our good 
 ship with its cargo of precious souls into the bosom of the raging 
 deep ; but we will pass at once to the sights and scenes which seem 
 of sufficient importance to induce us to add a single chapter to our 
 already quite full account of the "men and things " seen and done 
 upon " the Labrador. '* 
 
 And now I find that we go over a great deal of the ground al- 
 ready gone over so fully and often at great length in our previous 
 pages. Our dozen pleasure seekers, however, feather a great many 
 new facts, yet we find that the daily life of the native inhabitants 
 does not differ essentially, as we progress east and north along the 
 coast, from such as we have already described it to be. The houses 
 grow smaller and smaller the farther north we go, yet the same 
 general character of life everywhere prevails. 
 
 The time at our command for the voyage is limited, yet we reach 
 to within a few miles of Rigoulette, the chief station on the east 
 coast, if we except some of the Dutch settlements much farther 
 north, and the mouth of the only river of importance in all Labra- 
 dor. Beyond are the " Dutch settlements, " the headquarters of 
 the Moravian missionaries, who are supported in their work chiefly 
 by the Moravians in this country at Bethlehem, Pennsylvania. 
 Their headquarters on the coast are at Hopedale, Nain, Okkak, 
 Port Manvers, Hebron, and possibly even at Ungava bay itself, 
 though at the last named place I am uncertain whether there be an 
 established station or not. Situated as we are, then, on a flying 
 visit much farther north than we have ever been before in this most 
 interesting country, with little time at our command for investiga- 
 tion, and with so many avenues open through which to procure 
 facts concerning the places visited, I fiiid myself compelled to as- 
 sume the diary form of writing ; and, instead of a systematic ac- 
 count of each place as we visited it, and an arranging of facts 
 always in their proper connection, I give them as they were pro- 
 cured, being sure that, in this case at least, it is by far the better 
 way, and will give more satisfaction in the end. 
 
r 
 
 -. II 
 
 ( , 
 
 4 
 
 282 
 
 PUFFIN-SHOOTING — GALE OFF CHATEAU. 
 
 Our party first visited Bonne Esperance, from there L'Anse 
 Loup and Forteau ; at each of these places we spent some time 
 dredging in the harbor, and filling our bottles with specimens and 
 our cans of alcohol with fishes and other marine treasures. When 
 off Blanc Sablon, we lowered a boat and all hands went for a hunt. 
 Some went to Pigeon Island and some the next morning to the west 
 end of Greenley Island : the puffins were as abundant as when des- 
 cribed by Dr. Coues earlier in our pages. We had no difficulty in 
 killing three hundred in a single day, and could have procured twice 
 that amount without the least trouble. We found the island on this 
 end literally tunnelled with the holes of this bird, and the appear- 
 ance was much as if thousands of woodchucks had been at work 
 burrowing the ground ; there was scarcely a square yard of earth that 
 did not have at least one burrow in it while more often there were 
 twenty. 
 
 We lay in Forteau and L'Anse Loup harbors, becalmed, for 
 nearly two weeks, during which time we used the dredge with great 
 success and brought up from depths varying from six to twenty 
 fathoms a large assortment and variety of marine invertebrates, 
 which were carefully labelled and packed away to be sent to the 
 Smithsonian Institution at Washington, The waters seemed alive 
 with new and curious forms, however, and every haul of the dredge 
 procured something that we had not before obtained. 
 
 I shall not soon forget our run to Chateau. We started with a 
 fair breeze but it soon became so foggy that we could not see where 
 we were going. We reefed sail and continued to beat about look- 
 ing for some signs of the harbor. Sudder' ,/e passed some fishing 
 boats but as suddenly found huge rocks looming up directly in 
 front of us. A shout from our captain, while he put the helm hard 
 to, brought the vessel around, but so narrowly did we escape the 
 rocks that as we jibbed we could have touched them with an oar. 
 We put the ship about, and scud for the open sea. The waves were 
 lashed to fury by the wind which blew abaft our starboard side ; yet, 
 blowing as it was in this howling tempest, we succeeded in signal- 
 ling one of the small boats for a pilot, who boarded us with a great 
 
SOME DAMAGE — CHATEAU AND TEMPLE BAY. 
 
 283 
 
 deal of difficulty, and again putting about we steered for the harbor 
 of Chateau or rather Temple Bay, which is the true harbor at this 
 point, but we did not part with the gale so easily. 
 
 As we entered the narrow passage to the bay our captain shouted 
 to take down sail, but the pilot countermanded the order ; at least, 
 between the two the order was not obeyed, and, as we were run- 
 ning with the wind now almost dead ahead, though a littie on our 
 port, suddenly rounding the point of land on our left an unusually 
 terrific squall struck us, and with a sound like thunder our main 
 boom snapped like a pipe stem. Had the sailors not rushed to the 
 ropes and instantly taken down the sail it would have been blown 
 to threads in a moment. Our jib alone v.i'' now sufiicient to take 
 us to the safest position in the harbor, where <; soon let down both 
 anchors, paying out nearly all the chain on e.ich. Here we stayed 
 in imminent peril all night of being hur'^ I upon the rrcKS the other 
 side of the harbor, not forty rods away, at ctny moment. Thank 
 Providence our good ship weathered the Masc. The next day the 
 sailors spliced the boom with spruce side pieces and about twenty 
 fathom of a nice three-quarter inch manilla dredge rope belongmg 
 to our largest dredge ; as we stayed at Temple Bay and ajjout Cha- 
 teau for two or three days, owing to bad weather, I will try to des- 
 cribe the place. 
 
 Chateau Bay is a small bay and comprises really the outer en- 
 trance only of a much larger inside bay. The larger is cal'ed Tem- 
 ple Bay, though it is often mistaken for Chateau proper. Henley 
 Harbor is a small harbor outside the main bay and a little to the 
 eastward of it. '^fhe point of land separating Temple from Pitt's 
 Bay, east of it, is called Pitt's point, just outside of this is Whale 
 island which fills the cove so closely that but a very narrow entrance 
 admits vessels to the harbor within, yet the water is deep enough 
 here for even large vessels. 
 
 Whale island is thus named from a most extraordinary resem- 
 blance to the shape of a whale's back. It can only be seen from a 
 single position (the foot of a large hill said to be seven hundred 
 feet high) in Temple bay anu in a northwesterly direction from the 
 
w 
 
 
 .1 ifi' 
 
 I: 
 
 ■(.iS i 
 
 
 
 if 
 
 Li 
 
 284 
 
 WHALE ISLAND — FRENCH FORTS. 
 
 island. The hight st point on the hump of the back of this marine 
 granite monster was carefully measured and found to be about three 
 hundred and forty-five feet above the sea level. It would be a 
 curious fact if the supposed basaltic prisms of Castle island should 
 prove to be only a peculiar formation of rocky cliffs similar in origin 
 with the other rock formations about this part of the coast, but 
 close investigation seems to show this to be the case. The crouch- 
 ing figures represented in profile near this place are truly remarka- 
 ble ; they are figures of men, and could hardly bear a closer 
 resemblance to the real objects if they were indeed genuine. One 
 of them, especially, so nearly resembles a person in a crouching 
 attitude as to deceive the keenest eye if watched in the most favor- 
 able light, and well do these stony sentinels maintain their ceaseless 
 vigil, year by year, century by century. 
 
 The forts, located here, — remnants of the old wars which exter- 
 minated the Indians and Esquimaux residing on the coast, with the 
 white settlers then temporarily residing here also, — are objects of 
 great interest. At one time their outlines were quite distinctly 
 traceable, and even their internal structure well planed out, but 
 time has effaced nearly everything but the merest outlines of their 
 positions, which are now barely sufficient to enable one to deter- 
 mine their site. The fort at Henley seems to be the larger, more 
 important, and more distinctly outlined of the two; the one at 
 Temple bay the more accessible. At the latter place we dug up 
 canister and grape shot in abundance and could undoubtedly have 
 procured a large variety of articles had we spent some time there. 
 The situation of this fort was the summit of a huge hill the basal 
 point of which was the separating line between Temple and Cha- 
 teau bays. On the southeastern side of the slope we found and 
 examined three caves, or rather clefts in the rocks, of which local 
 tradition told startling stories of robbers, fierce animals and money, 
 which were said to exist there. We found no difficulty in " find- 
 ing bottom, " and saw no cause for alarm, whatever other? might 
 have feared. Superstitions are abundant everywhere on the coast and 
 there were undoubtedly those who believed the various stories of 
 
POPULAR SUPERSTITIONS — HOUSES. 
 
 285 
 
 
 dragons, blackmen, and lobsters with terrible claws, who were said 
 to guard treasures of inestimable value in these and other abodes 
 of foxes and weasels. People told us of a large stone monument or 
 house about three miles inland said to have been built by parties 
 from a man-of-war who entered the harbor secretly, some years 
 since, by night, and proceeded inland for some mysterious pur- 
 pose. One man, I am told, declared solemnly, that they came for 
 " buried treasures, " and built this " stone house " to guard the 
 remainder until they should come for it. Undoubtedly it was 
 done by the sailors to excite this very superstition. They succeeded 
 admirably. 
 
 Between the hill containing the fort and the opposite height west- 
 ward, is a most beautifully formed glacial ridge and slide. Seaward 
 it is more or less abrupt and precipitous though I should judge not 
 over forty feet high. Towards the bay it slopes evenly and smoothly 
 to the water. On this slope the houses are built. They are like 
 other houses on the coast, with a single peculiarity in the shape of 
 the fireplace. This was a hearth of some iron piece, usually the 
 sides of an old iron stove, a back fireboard of a similar character, 
 while the remainder of the chimney was of wood and formed part 
 of the house proper. Why the blazing, roaring, crackling hearth fire 
 that I often saw did not burn the house down I cannot well see. 
 An iron frame, made like an old-fashioned steelyards, hung over the 
 fire, the hooks of which could be raised or lowered at will, with 
 two hooks (if I remember correctly) upon which were hung the 
 kettles — here all the cooking of the family was done. One house- 
 wife showed me with great exultation, how nicely the bread kettle 
 hung over the fire. I thought that if a modern mother was obliged 
 to bake the quantity of bread weekly that this mother did and in 
 such a primitive manner, she would show the arrangement with 
 anything but satisfaction. 
 
 At one pcint in Temple bay I visited a mine of white mica that 
 was just opened. It was romantically situated about 150 feet abov^e 
 the sea and by the side of a winding brook, that tumbled down 
 among the confusion of loose rocks below and abounded in trout. 
 
■ t 
 
 tl'- 
 
 ■ hi ■ 
 
 m 
 
 wm 
 
 ■V 
 
 
 286 
 
 MINERALS. 
 
 The mica, though very pure, was only in very small pieces. The 
 rock in which it occurred seemed to be a gigantic bowlder rather 
 than a rock in place. 
 
 While on the subject of minerals I will say that at Dead Island, 
 some miles farther north, a party of miners boarded us, but did not 
 remain long. On our whole trip along the coast we found quite a 
 number of minerals, but none in quantities sufficient to pay for be- 
 ing worked. We found a mica mine at Dead Island, also, but it 
 did not appear to be so good even, as that at Temple bay. At 
 another locality one of the party found a quarry of very poor Labra- 
 dorite of the black variety ; we obtained a few pieces only. Every- 
 where granite rock predominated. In one locality veins of quartz 
 and mica alternating in thin layers, one above the other, extended 
 for some distance in several directions, bounded, I believe, by gran- 
 ite also. White and black mica were abundant everywhere, but in 
 small pieces. Tourmaline of the black variety was not rare and 
 several remarkably finely terminated crystals were procured. Some 
 fine rubellite was also found. Copper pyrite was common, as was 
 also iron pyrite ; sulphide of iron was very abundant : several pieces 
 of apatite were found ; a large amount of galena and in one locality 
 molybdenite, the latter probably in no very great amount. Poor 
 labradorite in various forms and color ; quite large but very brittle 
 garnets ; argentiferous lead in small quantities ; quartz very clear 
 and glassy ; feldspar of various colors ; and occasionally greenstone 
 like Pike's Peak greenstone, wrongly called labradorite, was 
 found. With this brief account of the mineralogy of the region let 
 us pass to other and more important subjects. 
 
 At Temple bay we spent considerable time dredging, and brought 
 up many curious and rare specimens from the bottom of the bay. 
 A large, bright red holothurian or sea-cucumber, seemed particularly 
 common to this bay, while the smaller varieties also abounded. 
 Shells of many kinds were dredged without number and at every 
 haul of the dredge the shrimps and crustaceans seemed equally 
 abundant. I shall not soon forget a dredging party that took place 
 one evening in this same bay. We started out from the vessel di- 
 
 ll,, 1^' 
 
DREDGING — AURORA — PHOSPHORESCENCE. 
 
 287 
 
 rectly after supper, and rowed across the bay nearly to the opposite 
 side, then, letting down our dredge, we began our work. That 
 night we secured a most strange collection of objects. Our buck- 
 ets were quickly filled with the profusion of material that was pro- 
 cured, and we were soon obliged to row back to the ship and get 
 more pails and buckets. Occasionally, when too near shore, we 
 would bring up a dredge, full to the top, of spiny sea urchins or 
 echini among which an occasional star-fish or holothurian would 
 be found ; in such cases we were obliged to empty the net over- 
 board and make a new haul. 
 
 In the midst of our work, when we had nearly filled our buckets 
 and pails with rich material, to be looked over in the morning, sud- 
 denly a most brilliant Aurora gathered in the heavens. It gathered 
 itself as if it were an immense snake, and with many undulations 
 seemed to coil and recoil itself to make room for the enormous 
 length with which it spanned the heavens and reached almost di- 
 rectly over our heads, from horizon to horizon. It was broad and 
 of a most intense white. It undulated like a ribbon, and changed 
 its form continually, sometimes concentrating much like an immense 
 drop, then as suddenly lengthening again. Then its direction would 
 change from northeast and northwest to nearly east and west. 
 
 While these contortions were attracting our attention in the heav- 
 ens, another peculiar phenomenon was beginning to appear in the 
 water, which suddenly became magnetized as it were to an unusual 
 degree, and that most remarkable occurrence of phosphorescence 
 began to display itself in a most intense degree. I have never seen 
 it so beautiful and so luminous. We had now reached the side of 
 the ship, and every dash of the oar sent large, whirling eddies of fire 
 off at our right and left. The boat left a long, luminous wake, like 
 the reflection in the water of the auroral ribbon above. A dash of 
 the oar would cover the surface with bubbles of fire, while, occasion- 
 ally, large disks of the same would sail by apparently ten or fifteen feet 
 below the surface, we could not touch them with our oar, though that 
 was twelve feet long. We all sat up some time watching these cu- 
 rious appearances, and each decided that they were the most beauti- 
 
u 
 
 (■'■; 
 
 U:^ 
 
 l>4 
 
 t, ! 
 
 it i 
 
 • ■■ Mi 
 
 288 
 
 ICEBERGS —FOX HARBOR. 
 
 ful that they had ever witnessed. We caught a huge ball of fire on 
 our dredge rope, and hauled it in. It was placed in a small can of 
 water and found to be one of the jelly fishes, with which, in the 
 daytime, the water abounded. It was nearly an inch long. Is it 
 possible that these animals, large and small, are luminous to such 
 an extreme degree ? 
 
 After remaining for several days in Chateau we left it for our 
 next port north. We had no more than cleared the land and taken 
 our course than the fog shut down upon us again, thick and heavy. 
 We kept steadily onward, however, but soon found we were 
 rushing into a tremendous ice mass. The vessel's course was changed 
 just as an immense ice tower, between two and three hundred feet in 
 height, surged by us with a velocity that would have crushed us in- 
 stantly had we struck it ; another and another followed, and we be- 
 gan to fear for our lives, but soon, by good fortune, the fog lifted 
 and we sighted Belle Isle, turning our prow, then, to a similar 
 course northward, to that which we had been steering eastward, we 
 were, by evening of the same day, quietly anchored at Fox harbor, 
 where we passed a most comfortable night. 
 
 Fox Harbor is one of the principal harbors of St. Lewis Sound, 
 and jujt north of Cape Charles. It is a large indentation, and con- 
 tains several very important harbors, the most so being Battle Island, 
 often called the Boston of Labrador. St. Lewis inlet runs a long 
 way into the interior of the country and is navigable nearly its whole 
 length of some twenty to thirty miles, I believe. We did consider- 
 able dredging here, and produced capital results. 
 
 I think that it was on Saturday, Aug. 1 2, that we arrived at Fox 
 Harbor, and glad enough were we to be in safe anchorage, once 
 more, while the fog continued to settle upon us. The next three 
 days were quite pleasant however, and we made the most of them. 
 Everybody on board started in the boats for a tour of investiga- 
 tion on shore. Here, at length, we had struck a real semi-arctic 
 habitation, inhabited by Indians, Esquimaux, and several half-breed 
 families. The houses were similar, but poorer, than those we had 
 seen all along the coast. The children were everywhere followed 
 
 \'m 
 
DOGS — IMPLEMENTS — DREDGING. 
 
 289 
 
 with troups of dogs, but they were not savage, being mainly pure 
 Newfoundland and a race of large Indian dogs. Several of these 
 dogs were brought home by various parties on board, and have 
 since thrived well, excepting the two Indian hunting dogs, pure 
 breed, which, I understand, have since died. A peculiarity of these 
 dogs is said to be the fact, that they will gorge themselves, and 
 carry food for a longtime in their stomachs untouched by the gastric 
 juice, disgorging it from time to time for their young. One man 
 affirmed that he had known them to keep food in this way for two 
 days, throwing it up in the apparently perfect condition of fresh 
 meat, upon which young puffins fed greedily. 
 
 Our men returned to the vessel loaded with spears, bows and 
 arrows, komatik whips, sealskin boots and mittens, and several 
 finely spotted skins. One of the party procured the tusks of a young 
 walrus, two of these animals having been killed by the natives the 
 previous winter. They told me here that this animal though occa- 
 sionally seen about this part of the coast was rarely captured. 
 Upon inquiring I found that no white bears had been seen here for 
 several years. 
 
 Across the harbor lies Battle Island, on the eastern side of which 
 is Battle Harbor. It is a village of about fifty houses, and a place of 
 much importance upon the coast. A mail steamer calls every fort- 
 night and returns directly to St. Johns. It is a fishing community, 
 and does not differ much from the settlements at Red Bay or Blanc 
 Sablon. 
 
 We did some good work at Fox Harbor dredging, and this 
 was the only place where we found squids, although they doubtless 
 occur more or less abundantly at nearly or quite every harbor. 
 Another peculiarity of our finds here was the immense number 
 of small Crustacea, sandfleas and worms, that were everywhere 
 abundant under rocks and in pools of water wherever we 
 searched for them. Here, too, as at other places, we dredged sev- 
 eral Terebralulay but found them generally rare. One of our 
 party secured a most beautiful and magnificent large salmon trout 
 nearly two feet long. 
 19 
 
,1 V 
 
 t 
 
 if J| 'I 
 ^ '■ ' 'i 
 
 Ill 
 
 ri 
 
 
 I i 
 
 'h 
 
 290 
 
 SQUARE ISLAND AND VICINITY. 
 
 Leaving Fox harbor we passed Mecklenburg harbor but did 
 not enter it, as it was not of sufficient consequence ; we also 
 passed St. Frances harbor, the mouth of the Alexis River, naviga- 
 ble for about twenty miles inland, and said to present as beautiful 
 scenery as any harbor along the coast, while the intricacies of its 
 mouth being as complex as any river south of Rigoulette, and 
 steered for St. Michaels and Square Island, the next place of in- 
 terest, where we stopped over night. We anchored at the farther 
 end of the harbor, and a most beautiful and picturesque little spot 
 it was. This place is named Square Island from the large almost 
 square island which nearly blocks the entrance to the harbor. 
 We entered through a very shallow and narrow passage, crossed 
 the harbor, and were soon anchored in a sequestered little spot safe 
 from everything save those intolerable torments the blackflies and 
 mosquitoes. The houses here looked more like the nests of an 
 army of cliff swallows than anything else that I can imagine ; they 
 were perched everywhere on the high rocks close to the cliffs, and 
 looked as if glued to them, so closely did they stand ; and so near the 
 color of the rocks were the weathered boards and boughs of which 
 they were composed, while so snug was the harbor, that one might 
 have hunted for weeks for the location, did he not know it from 
 previous visitation, and then have passed it without discovering the 
 entrance passage, while high cliffs everywhere surrounded it. A 
 party went over to Nolan's harbor, a few miles distant, and met 
 Capt. Fitzgerald, of Harbor Grace, Newfoundland, who oversees 
 quite a fishery at this point ; but one place so closely resembles 
 another on this part of the coast, that the intricacies of bays, coves, 
 islands, and narrow passages of water present every possible shape, 
 size, and form of harbor. It is, in fact, like the " Bower of Fau: 
 Rosamond," in which one would easily lose one's self without the 
 help of a most skilful pilot. 
 
 On Thursday we left Square Island and sailed to Dead Island, a 
 few miles north only of our former position, and found quite a com- 
 munity of fishermen living here. The inhabitants were chiefly sum- 
 
 IJb 
 
WHAT WE DREDGED. 
 
 291 
 
 mer visitors from Newfoundland, engaged in the herring fisheries. 
 They had quite a good catch, and were about preparing to return 
 home with their cargoes. Tlie majority of these fishermen, we 
 found, were from Harbor Grace, Newfoundland, and we everywhere 
 found them a rough, tough, but not ill-natured race, yet one with 
 whom we did not care to deal any more than it was found abso- 
 lutely necessary. 
 
 At Dead Island we produced some of our best results at dredg- 
 ing. The harbor was not deep, but the seaweed, with which the 
 bottom was covered, was everywhere full of life, and covered with 
 shells and minute crustaceans. In one small pass to the north 
 of which we were anchored, I found one of the most remarkable 
 nests of marine animals that I had ever come across, either on La- 
 brador or in any of the harbors of Massachusetts. The water was 
 from one to five fathoms deep, and clear as crystal. The bottom 
 was one large, magnificent flower bed of anemones growing on a 
 ground of red nullipore that covered everything. The extent of 
 this growth must have been acres, and as we sailed along in our 
 boat we could see the magnificent animals as plainly as if they had 
 been before us in the bottom of the boat, and we could pick 
 them with our hands. 
 
 The unevenness of the bottom furnished a pleasing variation of 
 elevations and depressions. Choice nooks and hiding places, or 
 plains were everywhere interspersed. Natural grottoes, and varie- 
 ties of rock-work, all were there, and all covered with " red rock," 
 or "live rock " as the people call this peculiar growth, probably the 
 red nullipore {Lithothamnion polymorphtini) , and some so heavily 
 incrusted as to represent miniature shrubs and trees even by 
 their own incrusted additions. Over this growth the anemone 
 {Metridiuni) grew as luxuriant as flowers reared in plant house 
 or hot-bed, and a most gorgeous hot-bed, even of beautiful 
 and rare tropical plants could not have excited more admiration. 
 I hauled up, with a huge scoop of the dredge, specimens whose 
 base measured ten and even more inches across, and whose ex- 
 
293 
 
 SEA ANEMONES. 
 
 panded disk was nearly as wide. Fringed tentacle several inches 
 in length would surround a disk whose neck was more often over 
 than under six inches in length. When touched they would 
 coil into a mass of leathery pulp as large as a medium-sized musk 
 melon. 
 
 This fine floral carpet was everywhere interspersed with green 
 echini, some large as a good-sized sunflower ; myriads of starfish, 
 with yellow, red, or brown backs were basking in some clear space 
 or curled up to fit the surfaces of the rocks upon which they 
 lay. Suddenly we came upon a single bed, it must have been 
 nearly half an acre in extent, in the middle of the passage, where an 
 almost perfectly level spot appeared so covered with these anemones 
 that their waving fringes intermingled to hide the bottom and pre- 
 sent a bed truly of the most exquisite and fairylike texture. Truly 
 were I to be buried in the ocean, I could not nor would I ask 
 for a more enchanting spot upon which to rest. I am sure 
 that no Arabian Nights' tale ever pictured a more imaginative 
 scene of splendor, than this simple bed of sea flowers — this " wind 
 flower" of the water formed in reality. 
 
 We left Dead Island on Wednesday the 23rd, and sailing through 
 the same narrow pass, which opened into a broad bay-like harbor 
 beyond and a series of intricate channels, coves and islands, we at 
 length reached Triangle harbor, another of those pirate-like coves 
 in which the coast so profusely abounds. The harbor is well shel- 
 tered from the seas, but hard to enter, being narrow and shallow. 
 High hills and cliffs are all about it, the highest point, on the 
 right, being a hill with lofty sides extending perpendicularly 
 nearly to its very crest, which is about three hundred and eighty 
 feet above the sea. We threw stones from this crest into the sea 
 and found that the top receded, from a point that overhung the 
 water about midway of this height, so much that the stones we 
 threw with all our force fell into the water behind the cliff" and con- 
 sequently out of our sight. Back from the hill grew a luxuriant 
 growth of vegetation, while a small but deep pond, partially sur- 
 
END OF VOYAGE. 
 
 298 
 
 / 
 
 rounded with high cliffs, afforded us a most excellent opportunity for 
 a fine bath in not too cold firesh water. The pond was one of a series, 
 doubtless extending way into the interior and forming the head 
 waters of some stream that flowed to the sea in one of the neigh- 
 boring fiord valleys, abundant here also. At this, our final post 
 north, we secured capital results with the dredge, though we found 
 a clayey and muddy bottom, which, I recollect, gave us great trouble 
 to strain in our sieves. Black bears are abundant in these parts, 
 I was told, and the very morning after our arrival one was seen 
 perched on the summit of a huge cliff above us. It was pursued 
 by the natives but not captured. 
 
 At length, on Friday the 25 th, we started for home, as our time 
 of absence had already nearly expired. We stopped at Fox har- 
 bor on our way back, and the next day had a most excellent run 
 of about one hundred and ten miles to Bonne Esperance where 
 we remained over Sunday and Monday. 
 
 At Bonne we found a new machine in operation for the conversion 
 of the refuse of cod and other fish into a sort of fish guano. Sev- 
 eral barrels had been sent to the United States to be analyzed and 
 tried for mixture with other materials for a soluble guano for the 
 land and crops, but I am not informed as to whether it was a suc- 
 cess or not. 
 
 Monday night, for our benefit, the natives performed a Labrador, 
 or rather Newfoundland dance, at one of the native cabins near 
 by. A crowd of about thirty assembled and danced till nearly 
 morning. Their main object seemed to be to " start the sweat, 
 and see who could make the most noise. " It seemed as if the 
 very house would come down over our heads as they hammered 
 on the floor with their top-legged boots pounding with the full force 
 of their powers : this serenade to the departing guests closed our 
 sight-seeing upon the Labrador Coast, while the hearty good-bye 
 of Mr. Whiteley, the magistrate, compensated for all bad weather 
 and mishaps that we had previously experienced as we slowly 
 spread sail and started for home. 
 
*'i! 
 
 
 294 
 
 NATIVE INDIAN DIALECT. 
 
 The following brief Indian vocabulary will give a slight clew to 
 that most obscure yet interesting language of the Labrador Indians, 
 as spoken all along the coast. 
 
 
 AMONG THE 
 
 NUMERALS. 
 
 
 1 pay yuck. 
 
 
 7 nish wash. 
 
 
 2 nesh. 
 
 
 8 nursh wash 
 
 • 
 
 8 nurst. 
 
 
 9 pay-er coush. 
 
 4 nao. 
 
 
 10 pay-er cannou. 
 
 6 pertater. 
 
 
 20 nesh ennou 
 
 • 
 
 6 goot wash. 
 
 
 30 nurst ennou. 
 
 . 
 
 OTHER 
 
 WORDS. 
 
 
 and, 
 
 ash-shoo. 
 
 no, 
 
 mar watch. 
 
 I, me, 
 
 nin. 
 
 yes, 
 
 topway. 
 
 you, 
 
 tin. 
 
 what, 
 
 chaquin. 
 
 he, 
 
 win. 
 
 water, 
 
 nee-pe. 
 
 that, 
 
 ne-ya. 
 
 spirits, liquor, 
 
 skutee-wabee. ^ 
 
 all of us (we), 
 
 casslno. 
 
 an axe, 
 
 eustaclie. 
 
 money, 
 
 soumentish. 
 
 salmon, 
 
 oush-a-muck. 
 
 quick, 
 
 sellorpe. 
 
 trout, 
 
 meta-muck. 
 
 girl. 
 
 squish. 
 
 cod. 
 
 ohm-zhee. 
 
 boy. 
 
 nowpee. 
 
 seal, 
 
 ar-chook. 
 
 mother. 
 
 uaga. 
 
 deer, 
 
 atchick. 
 
 give, 
 
 pcrta. 
 
 martin. 
 
 wablstan. ^y 
 
 I go with you 
 
 , mu pou shoo. 
 
 meat, 
 
 mce-asli. 
 
 how much, 
 
 ten ash push. 
 
 butter. 
 
 tootoosh pimme. 
 
 ship, 
 
 jonne push. 
 
 millc. 
 
 tootoosh ackee. 
 
 come on. 
 
 stammetay. 
 
 pork. 
 
 coocoosh. / 
 pimme. 
 
 goon, 
 
 mate. 
 
 lard, 
 
 canoe, 
 
 oushe. 
 
 
 
 V 
 
 Mi 
 
 Note. The other important localities on the Labrador Peninsula are as 
 follows : — 
 
 Hamilton Inlet (^Ivucktoke) which extends about ninety navigable miles in 
 a westerly direction. Its entrance is a wide bay which is succeeded by lake- 
 like expansions of this same inlet connected by narrow passes or channels. In 
 
 vs ■ 
 
NOTE. 
 
 295 
 
 the first channel, or " the narrows," as it is called, are Indian Village and Rig- 
 oulette, — the latter the most important station hereabouts; and formerly, I 
 believe, if not now, connected with the Hudson's Bay Company. The river, 
 which flows into this inlet and is called by the natives Esquimaux river, is 
 the largest and most important in all Labrador. It is the terminus of the 
 French and English settlements on the coast : below it the Dutch and Esquimaux 
 are the chief and more often the only inhabitants, to Ungava Bay itself which 
 borders the entrance to Hudson's Bay. 
 
 These almost mongrel inhabitants are known chiefly by the names of their 
 villages, as the people of Umiakkoviktanuk or Cape Strawberry, which is some 
 1 200 feet above the sea, in the background of which is Altagaiyaivik or Monkey 
 Hill, said to be over 2000 feet high. Next comes a post of the Hudson's Bay Com- 
 pany at Aillik, and one on the opposite side of the bay at Kaipokok. Towards 
 the sea are the Gull rocks, called by the musical ( ?) name of Nanyaktikiluk, 
 and farther along Gull island or Nanyakaigaluk. On the mainland nearly 
 opposite Gull island comes the important Mission station of Hopedale. Be- 
 low this is Nain, the next important Moravian mission. East of Nain is 
 Tunnuluaoak or Pownal or Paul Island, the place so often quoted as abound- 
 ing in Labradorite. Okkuk, followed by Port Manvers and Hebron, other 
 missionary stations follow while Cape Chudleigh (or Chidley as some give it) 
 forms the northern terminus, from which this promontory descends to Un- 
 gava Bay the northern boundary of the Labrador Peninsula. Of the interior of 
 this vast plateau little or nothing is known. The blackflies and mosquitoes form 
 an almost impassable barrier to investigation; try it once and you will thor- 
 oughly believe it.