BERKELEY 
 
 LIBF ARY 
 
 UNIVEfSITY OF 
 CALIFORNIA 
 
 EARTH 
 
 SCIENCES 
 
 LIBRARY 
 
REMINISCENCES 
 AMONG THE ROCKS 
 
 IN CONNECTION WITH 
 THE GEOLOGICAL 
 SURVEY OF CANADA 
 
 BY .,.,. 
 
 THOS. CHESMER WESTC'N, F.G.S-A. 
 M 
 
 TORONTO 
 
 PRINTED FOR THE AUTHOR BY 
 
 WARWICK BRO'S & RUTTER 
 1899 
 
Wf 
 
 EARTH 
 
 SCIENCES 
 
 LIBRARY 
 
 Entered according to Act of Parliament of Canada, in the year one thou- 
 sand ei^ht, hundred and ninety-nine by Thomas Chesmer Weston in 
 fche office^ Ihe^Minister of Agriculture. 
 
PREFACE 
 
 AT the request of my children and several 
 of my friends, I have endeavoured to 
 record some of the incidents connected with 
 my explorations while in the service of the 
 Government Geological Survey of the Dominion 
 of Canada. I have also set down a few details 
 of the geological formations in'\\*hich I have 
 worked. I have prefixed to these writings some 
 autobiographical particulars which may not be 
 without interest to the reader. It is a very 
 plain tale, without the slightest striving after 
 any of the literary properties that generally 
 accompany a story. 
 
 T. C. W. 
 
 237 Daly Avenue, Ottawa. 
 May 7th, 1898 
 
 A* 
 
CONTENTS 
 
 CHAPTER I. 
 
 PAGE 
 
 A few autobiographical particulars Proposal that I 
 should come to Canada The voyage First im- 
 pressions of the land of the maple A matrimonial 
 joke My introduction to the rocks 1 
 
 CHAPTER II. 
 
 Eozoon " The Dawn of Life " A remarkable geologi- 
 cal controversy Discovery of tubuli structure 
 Unlikely that the extensive Laurentian system 
 would only contain one kind of organic remains 
 That the Huronian system has not yielded any 
 fossils . . 20 
 
 CHAPTER III. 
 
 A visit to the United States Journey to the Island of 
 Anticosti My assistant proves forgetful and un- 
 fortunate Description of the island Its geologi- 
 cal formation Some queer inhabitants I sight a 
 couple of bears Further misfortunes of Jackson 
 Departure for Quebec 28 
 
CONTENTS. 
 
 CHAPTER IV. 
 
 The South Petite Nation River The Trenton forma- 
 tion The calciferous rocks of Ontario We lose 
 our way in the bush The Chazy formation A 
 Guelph freethinker An adventure with a bull 
 Visit to Point Levis An odd mistake at our 
 boarding-house The Eozoon controversy 46 
 
 CHAPTER V. 
 
 The field season of 1869 Journey with Mr. Richard- 
 son Lost in the bush The primordial rocks in 
 New Brunswick Arisaig, Nova Scotia, and its 
 coast rocks Retirement of Sir William Logan 
 Field season of 1870 The rocks of Quebec 
 Another visit to the United States Visit to Stan- 
 fold, Que. Field work of 1871-2 The coast of 
 Labrador 59 
 
 CHAPTER VI. 
 
 Lord Dufferin visits the Survey I assist Sir William 
 Logan in the Eastern Townships Another visit 
 to Arisaig Some of the pioneer photographs of 
 the survey A storm off Cape Gaspe A visit to 
 Newfoundland Lake Temiscouata Work at 
 Hespeler, Ont. The conglomerates of Bic Death 
 of Sir William Logan The Philadelphia exhibi- 
 tion The Eastern Townships again Distribution 
 of rocks and minerals Ethnological branch of the 
 Survey 79 
 
 CHAPTER VII. 
 
 The south shore of the St. Lawrence Some good fish- 
 ing Camping by the river Ascent of the Shick- 
 shocks A poaching expedition : we catch a fine 
 salmon The lonely Mount Albert Photographs 
 and fossils Coast rocks between Digby and Yar- 
 mouth, N.S. "The Ovens" Fishing at Spindle 
 
 Cove Lost in the fog 107 
 
 vi. 
 
CONTENTS. 
 
 CHAPTER VIII. 
 
 Baie des Chaleurs and adjacent districts A fossil fish 
 and a pious critic A list of tithes Removal of 
 the headquarters of the Survey from Montreal to 
 Ottawa The Joggins Coast, N.S. Campbellton 
 and the Restigouche River A trip to Sault Ste. 
 Marie The Cypress Hill and the North- West A 
 three months geological exploring tour 127 
 
 CHAPTER IX. 
 
 To South Joggins, N.S. Palaeontology of the carbon- 
 iferous rocks Journey to Manitoba and the North - 
 West Maple Creek and the Cypress Hills 
 Vertebrate remains A herd of wild cattle Belly 
 River and the prairies Some Indians Selkirk 
 and its attractions A pelican The Red River 
 Winnipeg and its miscellaneous stores 176 
 
 CHAPTER X. 
 
 Excavations at Quebec Eozoon again A collection of 
 mammalian remains Work on the Quebec group 
 The black limestones of Quebec Supposed coal 
 deposits Evading the Scott Act With Mr. 
 Fletcher to Cape Breton Arisaig, N.S. The 
 coal mines of Sydney 208 
 
 CHAPTER XI. 
 
 Explorations on the south shore of the St. Lawrence 
 Fishing for cod A pleasant stay at Mr. Richard- 
 son's A journey to Grimsby A trip to the North- 
 West plains Maple Creek Swift Current Big 
 Jaw coulee A prize pig The bad lands Leth- 
 bridge We launch a boat on the Red Deer River 
 but afterwards give up our plan The Joggins 
 
 coast once more 226 
 
 vii. 
 
CONTENTS. 
 
 CHAPTER XII. 
 
 Supposed fossil trees A visit of investigation to 
 Kingston I start for the Red Deer River Mr. 
 McKenzie joins me at Calgary Extensive coal 
 seams Fine dinosaur remains Canada has done 
 comparatively little towards restoring these deni- 
 zens of a silent world Dangerous quicksands 
 We celebrate Dominion Day A long river trip 
 Another trip to the Cypress Hills Unpleasant 
 experiences On the borders of Lake Superior 247 
 
 CHAPTER XIII. 
 
 The "Quebec Group" controversy An interruption 
 by illness I afford amusement to the natives 
 The Montmorency rocks The gold-bearing rocks 
 of Nova Scotia Improvements in the museum of 
 the Survey A clerical geologist The pilgrims to 
 Bonne Sainte Anne The geology of the Sainte 
 Anne district Surrounded by the tide 286 
 
 CHAPTER XIV. 
 
 Further geologizing at Quebec City Examining the 
 rocks precipitated by the land-slide A victim of 
 poison -ivy A search for a meteorite on the Ot- 
 tawa River The Laurentian marble An artist's 
 sketch is useful The Joggins coast, N.S., again 
 Conclusion and thanks to my friends of the Survey. 311 
 
 Vlll. 
 
LIST OF ILLUSTRATIONS 
 
 T. C. Weston Frontispiece 
 
 Horace S. Smith facing page 4 
 
 SirW.E. Logan " " 14 
 
 T. Sterry Hunt " " 18 
 
 Eozoon Canadense Canal System " " 24 
 
 Eozoon Canadense Laminated Structure " " 24 
 
 James Richardson " " 34 
 
 E. BiUings " " 70 
 
 Alex. Murray " " 90 
 
 Shipping Buffalo Bones " " 144 
 
 Section of Miocene Tertiary Rocks " " 182 
 
 Concretions at Irwin Coulee, N. W.T " " 192 
 
 Tree-Uke Concretions at Kingston, Ont " "248 
 
 Huronian Concretions " " 292 
 
 Concretions, Cape Breton " " 292 
 
 IX. 
 
CHAPTER I. ..,..-, 
 
 A FEW AUTOBIOGRAPHICAL PARTICULARS PROPOSAL THAT 
 I SHOULD COME TO CANADA THE VOYAGE FIl'.ST 
 IMPRESSIONS OF THE LAND OF THE MAPLE A MATRI- 
 MONIAL JOKE MY INTRODUCTION TO THE ROCKS. 
 
 MY first memories are of Birmingham, Eng- 
 land, the "toy shop of Europe," the 
 birth-place of Watt's steam engine, the source 
 of " Brummagem " jewellery and Radicalism In 
 this town of various manufactures it was not 
 a city then I was born in October, 1832. My 
 father was a self-taught natural philosopher in 
 a modest way, who was enthusiastic in learning 
 about science, such as it was in those days. He 
 was particularly fond of optics, and constructed 
 both magic lanterns and microscopes for his own 
 use. His versatility and cleverness in handicraft 
 led him ultimately into the manufacture of 
 fancy ornaments such as women delight in. Part 
 of his business consisted in dealing in and 
 cutting precious stones. He also prepared micro- 
 scopic sections of stones and minerals. It was 
 this work that earned him the title of " scientific 
 lapidary," and it was in this line of duties that 
 I spent ten years of my early life. Amid the 
 various ups and downs of a busy commercial 
 life my father still pursued his scientific investi- 
 A 1 
 
REMINISCENCES 
 
 and his optical experiments. One of my 
 earliest remembrances is of finding myself in a 
 large building in which were gathered between 
 One and two thousand school children who had 
 each paid a penny to come and see one of his 
 magic lantern exhibitions a wonder in those 
 days. The place was in darkness except for the 
 light reflected from a white circle on an immense 
 hanging sheet. Then a picture of the earth 
 flashed on to this space with ships sailing around 
 it altogether disproportionate in size. My father 
 explained to his audience the rotundity of the 
 globe, and concluded his elucidation by " giving 
 out " the first verse of Isaac Watts' hymn : 
 
 O Spirit of the Lord prepare, 
 
 All the round earth her God to meet, 
 Breathe Thou abroad like morning air, 
 
 Till hearts of stone begin to beat. 
 My father's science did not deprive him of his 
 religious faith. I could linger long over his 
 efforts to combine lecturing and natural philoso- 
 phy with business, his teaching working men 
 and women and children astronomy, natural his- 
 tory and the like. My education, meanwhile, was 
 not of the most regular kind. But I had a sweet 
 and cultured mother who taught me much. I 
 learned the art of the lapidary, and the time came 
 when we were the most prominent firm in that 
 line in Birmingham. 
 
 In January, 1858, my father received from Sir 
 W. E. Logan, Director of the Geological Survey 
 2 
 
AMONG THE ROCKS. 
 
 of Canada, an offer to come to Montreal as lapi- 
 dary to the Suivey. For various reasons he 
 could not accept this, and I was recommended as 
 competent to fill the position. After some cor- 
 respondence Sir William decided to engage my 
 services. A sum of money was placed to my 
 credit in one of the Birmingham banks to defray 
 my expenses to Portland, where I was to receive 
 a pass over the Grand Trunk Railway to Mon- 
 treal. After a very stormy voyage on the Allan 
 liner Indian, I landed at Portland on February 
 15, 1859. I left England with a " stove pipe " 
 hat and a light overcoat, and almost my first 
 experience of a Canadian winter was a frozen 
 ear. 
 
 Arriving at Longueuil, opposite Montreal, the 
 scene was so different from anything I had seen 
 the crowd of sleighs, the jingling of bells, the 
 fur-clad drivers and the hundreds of people who 
 had come to see the passengers of the ship which 
 had been given up for lost, for we were many 
 days behind time attracted my attention so 
 much that I crossed the St. Lawrence without 
 knowing that I was travelling over a great body 
 of frozen water. The weather was so intensely 
 cold that when I presented a letter of introduc- 
 tion to Mr. Blackwell, director of the Grand 
 Trunk Railway, he advised me to at once get a 
 fur cap and other warm clothing, and directed me 
 to the Geological Survey. I am much indebted 
 
 3 
 
REMINISCENCES 
 
 to the late director of the Grand Trunk Railway 
 Mr. Blackwell for many personal favors. 
 
 Arriving at the Survey and Museum, St. Gab- 
 riel Street, I found Sir Wm. Logan had gone to 
 Toronto to interview the Government on survey 
 matters. I was however received kindly by Dr. 
 Sterry Hunt, chemist to the survey, and told to 
 rest myself till Sir William arrived, which he did 
 in a few days, and exercised much pains in getting 
 me comfortably settled in the building. I soon 
 arranged my instruments for work, and through 
 the kindness of the Rev. Dr. Wilkes of Zion 
 Church, to whom I brought letters of introduc- 
 tion, was soon at home in my adopted country 
 and thus commenced what lengthened out to 
 thirty-five years of active life in the office, in the 
 museum, and in the field. 
 
 On the 9th of June, 1859, (five months after my 
 arrival in Canada) I was married to Matilda 
 Allen, youngest daughter of Smith Allen, of 
 Quebec. We were married by the Rev. Mr. 
 Bancroft, at old Trinity Church. Mr. Horace S. 
 Smith, artist to the Survey, acted as my best 
 man. Horace was always fond of fun and here 
 on this (to me) serious occasion could not help 
 indulging in his " little joke." Just as we were 
 arranged at the altar and the minister was about 
 to commence the marriage service, Smith whisp- 
 ered to me " I've forgotten the ring." I made 
 up my mind at once that Smith should get out of 
 4 
 

AMONG THE ROCKS. 
 
 the matter as best he could, and the service went 
 on; but when the time came for the ring Horace 
 quietly took it from his pocket and handed it to 
 the parson. Smith remained a constant friend 
 and I attended him in his last hours of sickness 
 and saw him safely conveyed to his last resting 
 place. 
 
 Many interesting incidents occured during 
 my second year in Canada (1860) which have 
 become historical. Among these I will only 
 mention that of the Prince of Wales' visit to 
 Canada, where at Montreal he opened the 
 Victoria Bridge, that wonderful structure which 
 spans our noble St. Lawrence River, and after- 
 wards visited the Geological Survey and took 
 great interest in the rocks and minerals of 
 Canada especially in the seven pound weight of 
 native gold from the Chaudiere valley in the 
 Province of Quebec, which as on many other 
 occasions passed through my hands, from the 
 safe to the cases of the museum. 
 
 Through the kindness of Sir William I was 
 introduced to His Royal Highness and accom- 
 panied him through the Museum. 
 
 From Montreal the Prince of Wales went to 
 Ottawa where on the 1st of September accom- 
 panied by the Governor-General, Sir Edmund 
 Head, and a brilliant staff of notables, he laid 
 the foundation stone of the Parliament Buildings. 
 
 A short time after my arrival at the Survey, 
 5 
 
REMINISCENCES 
 
 I found that my duties would consist of devel- 
 oping fossils and reducing the rock in which 
 they were imbedded, which was done by a thin 
 iron disc charged with diamond dust. By this 
 means hundreds of new fossils were placed in 
 the already crowded cases. Microscopic sections 
 of various rocks and fossils were made for study 
 and I was constantly called on to assist Mr. E. 
 Billings, palaeontologist to the Survey. This 
 work occupied my whole attention till the early 
 part of 1863 when I commenced my field explor- 
 ations by accompanying Mr. J. Richardson and 
 Mr. Bell, both members of the geological staff, to 
 Drummondville. Here I was to assist in the 
 examination of the rocks, and collect, if possible, 
 fossils from certain members of what Sir W. E. 
 Logan had termed " The Quebec Group," a series 
 of strata I shall often speak of in these rambling 
 notes. 
 
 This was my first experience in field geology, 
 and also my first insight into Canadian country 
 life. 
 
 As these pages are to contain beside geological 
 notes, incidents " by the way side," I shall com- 
 mence here by relating one boarding-house 
 experience. Our home for a few days was 
 situated on the bank of the St. Francis River, 
 Province of Quebec. It is a dear old homestead 
 looking out over the falls of Drummondville, 
 and was kept by an old man and his wife, two 
 
AMONG THE ROCKS. 
 
 thoroughly Scotch bodies who had reared a large 
 family, had seen much of the hard pioneer life 
 of the back woods of Canada and had become 
 somewhat doubtful of one another's veracity as 
 the following narration shows. j 
 
 Mr. Richardson was a thorough Scotchman, 
 and like many of his class delighted in his glass 
 of toddy before retiring for the night, so before 
 commencing field operations, he provided himself 
 with a stout bottle of Scotch whiskey (his 
 favorite beverage) and having no safe place to 
 keep his bottle, decided to trust it with the old 
 man of the house. The first night, after a hard 
 day's work among the rocks of the river, he called 
 on the old man for his bottle, but on taking his 
 usual "nip" found that the strength of the 
 liquor was not what the label represented it to 
 be. The second evening it was still weaker, and 
 Mr. R. asked the old lady whether her husband 
 was to be trusted with the bottle " Ah man " 
 she replied " ne'er trust the old man wi' yer 
 whuskey for he will help himself and then fill up 
 wi' water,' ' gi' me the bottle and it will be all 
 richt." So the bottle was given to the old lady for 
 safe keeping ; but alas ! on the following night 
 the whiskey was still weaker, so the old man 
 was then interviewed as to his wife's honesty, 
 " Ah man " he replied " you should na trust the 
 auld woman wi' the whuskey for she helps her 
 sell and then waters the bottle." So concluding 
 
 7 
 
REMINISCENCES 
 
 they were two of a kind Mr. R. decided to hide 
 the bottle in our bed-room and from that time 
 the strength of the whiskey was preserved. 
 
 The rocks of the St. Francis River in the 
 vicinity of Drummondville are chiefly greenish 
 sandstones, forming in places high walls, and 
 showing beautiful examples of stratification and 
 folding ; they are underlaid by dark shales and 
 limestones holding a few graptolites and obscure 
 bivalves among which the following genera and 
 species were found : 
 
 Dendrograptus simplex, Walcott, 
 
 Dicranograptus sextans, Hall, 
 
 Ccenograptus gracUis, Hall, 
 
 Climacograptus Dicornis, Hall, 
 
 Leptobolus insignis, Hall. 
 
 These sandstones, limestones and shales are cut 
 by a variety of diorites and amygdaloids, some of 
 which are porphyritic ; they form the rocks of 
 the falls, giving a rugged aspect to this portion 
 of the river. It was here among the dark colored 
 shales and limestones of the St. Francis River, 
 I learned from my friend Mr. Richardson who 
 afterwards during our twenty years of almost 
 constant companionship taught me much of the 
 knowledge I have gained of field geology the 
 importance of fossils regarding our researches in 
 stratigraphy. 
 
 While wending our way home in the dusk of 
 the evening, after a hard day's work, we came 
 8 
 
AMONG THE ROCKS 
 
 across a man diligently engaged picking a deep 
 hole in the black shales. On enquiry as to what 
 he expected to find, he informed us that he was 
 prospecting for coal. Mr. Richardson told him 
 he would never find coal in those rocks for they 
 are thousands of feet below the coal formation. 
 But the next day when we passed the same 
 place the man was still working away and had 
 employed another man to assist him. A right 
 knowledge, as I afterwards learned, of the vari- 
 ous formations which constitute the " crust " of 
 our earth and its fossil remains, would have told 
 this man that coal could never be found in those 
 rocks. 
 
 Coal occurs in many formations. It has been 
 mined for over a hundred years in Beven, Suther- 
 landshire, in rocks of the Oolitic epoch, in other 
 countries in the Miocene-Tertiary beds, and other 
 rocks belonging to the upper crust of the earth, 
 but the Carboniferous deposits are the coal bear- 
 ing beds of Great Britain, the United States, 
 Nova Scotia, Australia, and other countries. 
 
 The miner who has studied his geological text- 
 books knows full well that it is utterly useless 
 to look for coal in rocks containing Graptolites, 
 trilobites and other organic remains which 
 existed untold ages before the coal deposits were 
 formed. I shall speak of other useless attempts 
 to find coal in Lower Silurian and Cambrian 
 rocks later on. 
 
 9 
 
REMINISCENCES 
 
 During the same year (1863) I spent several 
 weeks in the State of Vermont examining and 
 collecting from the limestones along the shore of 
 Lake Champlain, and along the road from White- 
 hall to Burlington, the whole of which distance I 
 walked, staying over night and sometimes longer 
 at the various villages or towns on the way. 
 
 Before starting on this journey several days 
 were spent at Whitehall, and here Sir William 
 Logan joined me and together we examined the 
 various rock formations of that vicinity and col- 
 lected fossils and interesting facts relating to or 
 connected with Canadian geology. 
 
 There was a good hotel at Whitehall, and here 
 Sir William and I occupied a double-bedded 
 room. On the 18th of October, Sir William 
 informed me that he would leave the following 
 morning by the early train for Albany, where he 
 was going to see Prof. James Hall. At five next 
 morning I woke to find Sir William scraping the 
 mud off his boots and doing his best not to wake 
 me, but he did, and I asked him if I could assist 
 him. " No," he said, but soon I was roused by 
 his shaking my arm and saying " Wake up, 
 Weston, and lend me some money ; I haven't 
 enough to pay for my railway ticket." I gave 
 him what he needed, but told him I had a hundred 
 miles to walk, and when I reached Burlington I 
 should want money. He promised to send a 
 cheque to the care of the postmaster. 
 10 
 
AMONG THE ROCKS. 
 
 After leaving Whitehall I spent a few pleasant 
 days at a farm house at Comstock Landing. 
 The geological structure of this vicinity is very 
 interesting, but I am not sufficiently conversant 
 with the formation here to say definitely what 
 the horizon is the fossils are obscure but look 
 like Calciferous forms. 
 
 While here I received a letter from Dr. Godfrey, 
 of Montreal, stating that my wife had presented 
 me with a son, news which caused me to hasten 
 on with my journey. The following morning 
 (Sunday) I shouldered my fishing basket (which 
 I used for carrying specimens) and started on a 
 day's tramp. It was past six in the evening 
 when I arrived tired and hungry at a small 
 village named Bread Port. Finding no hotel, 
 I enquired for a boarding house and was told 
 there was none, but Mrs. Brown down the street 
 sometimes took in travellers. I went to Mrs- 
 Brown, but she sent me to another house. They 
 too refused to take me in, and, after wandering 
 all over the village to get a vehicle, in which I 
 failed, I met a man who looked a likely person 
 to assist me. I stopped him and said : " You 
 call this place Bread Port, I believe, sir?" 
 " Yes," he said, " this is Bread Port." " Well," I 
 said, " I think it is badly named, for I can neither 
 get bread to eat nor can I get lodgings for the 
 night." Perhaps I looked like a tramp, but a few 
 words of explanation and this good Samaritan 
 11 
 
REMINISCENCES 
 
 took me to his house, and when his people 
 returned from church I was well provided for, 
 and we spent a pleasant evening singing and 
 playing. 
 
 The following morning I continued my journey 
 and in course of time arrived at my destination, 
 Burlington, where the postmaster handed me the 
 cheque I had asked Sir William to send, and also 
 gave me Sir William's letter to him, of which the 
 following is a copy. I insert it to show how 
 careful he was in money matters : 
 
 " At Prosper Hall, corner Merton St. and 
 Delaware Turnpike, 
 
 ALBANY, 19 October, 1863. 
 To the Postmaster, 
 
 Burlington, Vermont. 
 
 DEAR SIR, I send you herewith a money letter for Mr. 
 T. 0. Weston of Montreal, who will call at the post office 
 for it. I send also a letter signed by Mr. Weston that 
 you may see his signature. Will you be so good as to 
 deliever both to him when he calls. Mr. Weston has 
 been with me lately to Whitehall and to Comstock Land- 
 ing collecting fossils. He belongs to the Geological Sur- 
 vey of Canada, of which I am the director. By this 
 information you can frame questions by means of which 
 to identify him. 
 
 I am, dear Sir, 
 
 Very truly yours, 
 
 W. E. LOGAN. 
 
 I returned to Montreal a few days later and 
 spent the remainder of the time till my next 
 excursion in my usual work. 
 12 
 
AMONG THE ROCKS. 
 
 As I shall often refer to the various members 
 of the Geological staff I shall briefly mention 
 those who composed the staff when I commenced 
 my duties in the Geological Survey. 
 
 Our distinguished chief, Sir William Logan, 
 whose kind acts and fatherly counsel remained 
 forever dear to the hearts of all associated with 
 him, was a Canadian, born of Scotch parents, in 
 Montreal in 1798, educated at the High School 
 of Edinburgh. He distinguished himself as a 
 geologist in the South Wales coal fields, 1838. 
 In 1841, Sir William then Mr. Logan, visited 
 the coal fields of Pennsylvania and Nova Scotia 
 and then commenced his studies of the rocks of 
 Canada, and was shortly afterwards appointed 
 by the Government to form a geological survey 
 of Canada. During the Paris Exhibition in 1858, 
 he received the gold medal .of honor, and was 
 created a Knight of the Legion of Honor. He 
 was knighted by the Queen in 1856. Poor health 
 and a desire to spend the remaining years of his 
 working life on his beloved Eastern Township 
 rocks caused him to resign his connection with 
 the Geological Survey of Canada in 1869. He 
 was succeeded by Mr. Alfred R. C. Selwyn, who 
 was recommended to him by Sir Roderick 
 Murchison. After a wonderfully active life, dur- 
 ing which time, as Sir William often said, "I never 
 had time to get married," he died at the ripe 
 age of 77 years, at his sister's house in Wales, 
 13 
 
REMINISCENCES 
 
 22nd June, 1875. " And now he sleeps in the 
 quiet churchyard of Llechryd between his 
 brother Hart and his great friend and brother- 
 in-law, Mr. A. L. Gower. Peace to his memory. 
 Honour to his name." 
 
 Those interested in Logan's history and geo- 
 logical work I must refer to the Geological 
 Survey and to the "Life of Sir William Logan, 
 Kt., LL.D., F.R.S., F.G.S., etc., by Bernard J. 
 Harrington, for a number of years chemist to 
 the Survey, and now of McGill University, 
 Montreal. I cannot close this page without 
 recording a regret that in his memoirs of 
 Logan's life Dr. Harrington omitted to even men- 
 tion the name of the writer of these notes who 
 travelled thousands of miles with him, shared 
 some of his hard field work, and was his almost 
 constant companion in his last investigations of 
 the Eastern Townships geology. 
 
 MURRAY. Shortly after Logan's appointment, 
 by the Provincial Government of Canada in 1842, 
 to make a Geological Survey of Canada as far 
 as his means would permit he secured, through 
 the recommendation of Sir H. T. De la Beche, 
 the services of Mr. Alexander Murray, a young 
 man educated in the Royal Naval College of 
 Portsmouth, and who served in the navy. Mr. 
 Murray became a staunch friend of Sir William 
 who trusted him with many important surveys. 
 After many years faithful service through the 
 14 
 
AMONG THE ROCKS. 
 
 recommendation of Sir William, Mr. Murray be- 
 came director of the Geological Survey of New- 
 foundland, a position he held to within a short 
 time of his death, in 1884 in his 75th year. 
 
 T. STERRY HUNT. T. Sterry Hunt, afterwards 
 Dr. T. Sterry Hunt, LL.D., Officer of the French 
 Legion of Honor, etc., etc., succeeded Count De 
 Rottermund for a short time chemist to the 
 Canadian Survey in 1847. Dr. Hunt was one 
 of the most able chemists and mineralogists of 
 his day. His writings are known all over the 
 scientific world. Dr. Hunt held his position as 
 chemist and mineralogist to the Geological Survey 
 of Canada for 25 years, and resigned this posi- 
 tion in 1872, three years before Sir William's 
 death. 
 
 BILLINGS. Mr. E. Billings / a Canadian lawyer, 
 was appointed Palaeontologist to the Survey in 
 1856. Long before this Mr. Billings had been a 
 zealous worker in palaeontology, and had pub- 
 lished many contributions to that science. These 
 papers attracted Sir William's notice, hence his 
 appointment. Mr. Billings remained a faithful 
 worker in the Survey for 20 years, died in 1876, 
 and was succeeded by Mr. J. F. Whiteaves. 
 
 RICHARDSON. While on one of his geological 
 surveys in 1846 Sir William met with Mr. James 
 Richardson a Scotchman who had seen much 
 hard pioneer farm work in Canada, and had for 
 a time taught in a country school in the Eastern 
 15 
 
REMINISCENCES 
 
 Townships. Logan wanted a man to do camp 
 work and engaged Richardson. After a short 
 time, however, Mr. Richardson became so fasci- 
 nated with geology that he induced Sir William 
 to allow him to attempt a little geological work, 
 so he was sent off to collect specimens and make 
 notes of what rocks he saw. On his return to 
 camp Sir William was so pleased with this Mr. 
 Richardson's first geological work that he was 
 again sent off to make measurements by pacing, 
 which work proved so satisfactory that he was in a 
 short time appointed explorer, etc., to the Survey. 
 He became one of Sir William's most trusted field 
 men, did a great amount of good geological work, 
 as the reports of the Survey show, and was 
 trusted with the arrangement of the geological 
 specimens in the Paris, London, and other exhi- 
 bitions. He remained a faithful worker in the 
 Survey till after Sir William's death, was super- 
 annuated in 1879, and died three years later. 
 
 BARLOW. Mr. Robert Barlow, formerly in 
 charge of a corps of the Royal Engineers in a 
 topographical survey in Great Britain joined 
 the Geological Survey in 1857 as chief draughts- 
 man, which position he held until a short time 
 before the Survey was removed to Ottawa. Mr. 
 Barlow only lived two or three years after his 
 retirement from the Survey. 
 
 SCOTT BARLOW. Mr. Robert Barlow was ably 
 assisted by his son Mr. Scott Barlow, whose chief 
 16 
 
AMONG THE ROCKS. 
 
 work was to plot surveys from field note-books. 
 In later years he did some excellent field 
 geology in Nova Scotia, among the Carboniferous 
 rocks. He was appointed chief draughtsman to 
 the Survey department a short time before his 
 death, which occurred in 1894 through blood 
 poisoning. 
 
 SMITH. Mr. Horace S. Smith, artist to the 
 survey, was engaged in England, and came to 
 Canada a year or two before my appointment in 
 J 859. His duty was to draw fossils to illustrate 
 the reports of the Survey. His drawings illus- 
 trate the Geology of Canada for 1863, and other 
 Survey publications. Mr. Smith died a few years 
 before the Survey was removed to Ottawa. 
 
 BELL. Mr. Robert Bell then in 1859, a young 
 Civil Engineer although engaged at various 
 times in geological work, was not appointed to 
 the Survey till some years later, since which time 
 he has made many important surveys in various 
 parts of Canada. At this date, 1897, Dr. Bell is 
 actively engaged in geological work ; is assistant 
 director, is M.D., CM., LL.D., F.R.S., etc. Dr. 
 Bell, with the exception of myself, is the only 
 surviving member of the old regime of 1859. 
 
 I must not forget to mention here another per- 
 manent member of the staff Michael O'Farrell, 
 office keeper, whose connection with the Survey 
 dated from its commencement to 1889, when he 
 was superannuated. Michael was a thorough 
 
 B 17 
 
REMINISCENCES 
 
 Irishman, always ready to assist a friend in want, 
 but a bitter foe to those who slighted or opposed 
 him. His commanding appearance gained for 
 him from some of the members of the staff the 
 sobriquet " Figure-head of the geological ship." 
 Michael was a keen horseman, and had in his 
 younger days ridden behind the hounds with 
 some of the most popular noblemen of " dear old 
 Ireland." Michael was very much attached to, 
 and a constant attendant on his chief , Sir William 
 Logan, whose disregard for personal adornment 
 frequently raised Michael's ire. " Sir," said 
 Michael to Sir William one day : " The tailor 
 says these pants are not worth repairing." " Well," 
 said Sir William, " confound him, tell him to 
 make me another pair. " And your boots too, 
 Sir, are in a very bad state." " Well, but," Sir 
 William said, " can't they be patched ? you know 
 I can't walk in new boots." 
 
 Although Sir William did not forget Michael 
 in his will, he never seemed the same man after 
 his chief's death, and soon after the removal of 
 the Survey from Montreal to Ottawa failed in 
 health so much that it was necessary to appoint 
 another janitor to look after the offices and clean- 
 ing of the museum ; so, as I said before, Michael 
 was superannuated, and was taken care of by his 
 two daughters. Michael only survived about 
 two years after his retirement in 1889. He was 
 succeeded by Mr. Thomas Burke, who at this 
 18 
 
AMONG THE ROCKS. 
 
 date, 1897, is fulfilling his position to the satis- 
 faction of all members of the staff. 
 
 Besides the small band of permanent workers 
 spoken of, there were several other "extra hands," 
 among whom was my old friend Mr. James Low, 
 of Grenville, P. Q., who, though only a rough 
 farmer, through accompanying Sir William on 
 his explorations, became an expert in tracing 
 certain bands of Laurentian rocks. His surveys 
 were plotted by Mr Scott Barlow. 
 
 Another was Mr. Richard Oatey, a Cornish 
 miner, who did much good work in the examin- 
 ation of mineral deposits. He was associated 
 with the gold mining at the Chaudiere River, P. 
 Q., at the time when the seven pound weight of 
 gold was taken out of the beds of the rivers, and 
 fields of that district. 
 
 Many other men were occasionally employed 
 and rendered valuable service, but I can only 
 mention here one other : Principal Dawson, of 
 McGill University. Montreal, who, notwithstand- 
 ing his arduous duties at the college, found 
 time to assist Sir William in determining fossils, 
 &c., &c. Sir J. W. Dawson, C.M.G., LL.D., F.R.S., 
 F.G.S., &e., &c., retired from McGill University 
 a short time ago and is now (1897) actively 
 engaged with his books and fossils. His writings 
 are known throughout the scientific world. 
 
 19 
 
CHAPTER II. 
 
 EOZOON " THE DAWN OP LIFE " A REMARKABLE GEOLOGI- 
 CAL CONTROVERSY DISCOVERY OF TUBULI STRUCTURE 
 UNLIKELY THAT THE EXTENSIVE LAURENTIAN SYSTEM 
 WOULD ONLY CONTAIN ONE KIND OF ORGANIC REMAINS 
 THAT THE HURONIAN SYSTEM HAS NOT YIELDED 
 ANY FOSSILS. 
 
 THE great Eozoon controversy commenced 
 soon after my appointment with the 
 Survey (1859), has continued to the present 
 day, and is likely to continue, with a gradual 
 falling off of those who in the early days of its 
 discovery favored the supposition that it was a 
 fossil. 
 
 It would take many pages to record even a 
 slight summary of the literature published for 
 and against this remarkable organic looking 
 structure. Here I can only give a slight outline 
 of its discovery and a few remarks on the scien- 
 tific interest and work done to prove this concre- 
 tionary structure to be organic. 
 
 The first specimen of Eozoon known was 
 found at Burgess, Ont., by Dr. Wilson, of Perth, 
 Ont., and was sent to the Geological Survey as a 
 mineral. Years after, it became associated with the 
 supposed Laurentian fossil and was analyzed and 
 examined microscopically, but nothing was found 
 20 
 
AMONG THE ROCKS. 
 
 resembling organic matter. Later on, in 1860, 
 other coralline-looking specimens were found by 
 Mr. J. McMullen at the Grand Calumet, Ont. 
 These specimens were sliced and examined under 
 the microscope, but no fossiliferous structure 
 found a description of these can be found in 
 the Geological Survey Report for 1863. They 
 were given to Mr. Billings to study, and he finally 
 determined not to risk his palaeontological repu- 
 tation in describing such specimens to be even 
 probably of organic origin. These specimens 
 were exhibited in London in 1862 by Logan, but 
 he appears to have found few believers in these 
 supposed Laurentian fossils. 
 
 About this time Mr. Low collected some fine 
 blocks of serpentine intermixed with crystalline 
 limestone (calcite), from Cote St. Pierre, P.Q- 
 In these specimens Sir William noticed peculiar 
 concretionary matter resembling the Black River 
 fossil Stromatopora rugosa (Hall). The writer 
 was at once requested to make microscopic sec- 
 tions of this rock. Several slides were made and 
 the first looked at under the microscope revealed 
 tubuli structure, or what is called the canal 
 system, in the calcite part of the specimen, which 
 is supposed to be the test or shell of the animal. 
 Sir William examined this, the first discovery of 
 what really resembled organic matter, with great 
 interest, and, in his usual prompt way of acting, 
 had Mr. Smith, the artist of the Survey, make 
 21 
 
REMINISCENCES 
 
 drawings. These, with the micro-sections, Sir 
 William took to Principal Dawson, who was 
 delighted to find the tubuli structure in the 
 calcite as he supposed it would be, and not in the 
 serpentine, as Dr. Hunt supposed it would, if 
 found, occur. 
 
 I have been particular in recording this the 
 first specimen which appeared to have definite 
 organic structure, because in his " Dawn of Life" 
 Dr. Dawson omitted to give me credit for being 
 the first to notice this structure. The illustration 
 opposite p. 24 is from a steel engraving made from 
 Smith's drawing, magnified about twenty dia- 
 meters. 
 
 After this discovery a large portion of my time 
 was taken up in microscopical work, examining 
 and making thin slices of any Laurentian rock 
 we thought might possibly contain Eozoon, and 
 assisting Mr. Billings in palseontological work. 
 In the summer of 1873 I was sent to inspect and 
 collect specimens from our now celebrated Eozoon 
 locality at Cote St. Pierre. Many beautiful 
 specimens were found, some of which are now in 
 the cases of the Geological Survey Museum ; some 
 are beautifully weathered, causing them to resem- 
 ble the Black River fossil Stromatopora already 
 spoken of. A short time after this, 1874, while 
 collecting fossils in the rocks of the Guelph for- 
 mation at Hespeler, Ont., I was telegraphed for 
 to return to Montreal and accompany Principal 
 22 
 
AMONG THE ROCKS, 
 
 Dawson to Cote St. Pierre. We arrived at Pap- 
 ineauville in the afternoon, took the stage to St. 
 Andr^ d'- Avelin (about twelve miles) and put up 
 for the night at a small French boarding house. 
 Here we could only get one bed, which Dr. Daw- 
 son took, while I lay on a rough couch, from 
 which I was soon driven well,not by mosquitoes. 
 The following morning we continued our journey 
 and arrived at Cote St. Pierre about noon. We 
 found lodging in a small French-Canadian farm 
 house within a few paces of the original spot 
 where Mr. Low collected the fine blocks of ser- 
 pentine in whioh the first specimen of tubuli 
 structure was found. 
 
 During our journey from Papineauville we 
 passed over great ridges of gneissose rock, ser- 
 pentine, crystalline limestones, diorits, and other 
 typical exposures of the great Laurentian system 
 over hills and through valleys with patches of 
 cultivated land from which wheat and other 
 crops had just been taken. Much of the 
 land is covered with a second growth of young 
 maples and other trees. A short distance from 
 our Eozoon locality we pass over the beautiful 
 Nation River, which winds through the hills and 
 valleys of this picturesque country. 
 
 After a good meal of pork and beans, and a 
 
 walk of a few paces, we stood on the slope of 
 
 the hill in Low's excavation made in taking out 
 
 the blocks of serpentine, in which, as I said, the 
 
 23 
 
REMINISCENCES 
 
 first Eozoon tubuli structure was found. But 
 besides this excavation we discovered that much 
 work had been done by an American who was 
 in search of " Cotton rock," a fibrous serpentine, 
 or asbestos, now largely used in packing safes, etc. 
 Here there is a thickness of about a hundred feet 
 of the Eozoon-yielding rock exposed. It is asso- 
 ciated with coarse diorite, gneiss, pyroxene, 
 dolomite, tremolite, and other characteristic 
 Laurentian rocks and minerals, forming high 
 ridges and a very undulating country. The con- 
 cretionary looking (Eozoon) portion of the ser- 
 pentine rock, i.e., laminated serpentine and cal- 
 cite, has been beautifully weathered. The soft 
 calcite has been worn away and the narrower 
 bands of serpentine remain standing up above 
 the calcite, giving, as I have before remarked, a 
 close resemblance to the Protozoon fossil Stroma- 
 topora. 
 
 In a short time we made a fine collection, some 
 specimens of which are now historical, having 
 been figured to illustrate the various writings on 
 Eozoon and afterwards placed in the Geological 
 Survey, and Redpath Museums. One of these 
 specimens is reproduced in Dawsoii's " Dawn of 
 Life," Plate III, and it was from this collection 
 that the sections were made from which my 
 micro photographs were taken, some of which 
 also illustrate the " Dawn of Life." 
 
! ///>J&^* -^>. ^V -^'. 
 
 Eozoon Canadenxe : Canal System. 
 
 Eo~oon Cainideme: Laminated Structure. 
 
AMONG THE ROCKS. 
 
 I must, at present, close this brief history of 
 Eozoon the gigantic Protozoa of the Laurentian 
 seas, according to Sir J. W. Dawson; who after all 
 these years and amid all the doubt of able palaeon- 
 tologists and naturalists, is to-day the firm believer 
 in Eozoon he has ever been. For my part, and 
 I have made and examined many hundreds of 
 specimens, both as thin sections for the micro- 
 scope and decalcified specimens showing the 
 tubuli structure, I cannot say I am a believer 
 in Eozoon being organic. No one, however, 
 would be more pleased than I to see its organic 
 nature established. This I fear will never be 
 done unless something more definite than this 
 supposed Stromatopora, Stromatocerium, as it is 
 now called, is found. 
 
 The reproduction opposite p. 24 is from one of 
 the etched specimens, and with the description, 
 copied from the label on the large block of Eozoon 
 serpentine in the Geological Museum, Ottawa, will 
 give those who have not access to books and speci- 
 mens a general idea of this supposed fossil. 
 
 " A gigantic Protozoon of the order Foramini- 
 fera from the Laurentian system. 
 
 This specimen is part of an unusually large 
 mass of the fossil, contorted by the disturbance 
 of the containing beds. 
 
 The laminae of white calcite are the walls of 
 the test of the animal, and the laminae of 
 green serpentine fill the spaces originally occu- 
 25 
 
EEMINISCENCES 
 
 pied by its gelatinous animal matter. Collected 
 by Mr. James Low at Petite Nation, on the 
 Ottawa." 
 
 When one considers that in Canada and 
 British America the Laurentian system covers 
 from 2,000 to 3,000 square miles and has in 
 Canada a known thickness of 40,000 feet of 
 strata, that part of this vast body of rock was 
 originally sedimentary matter derived from older 
 rocks of which we know nothing, and that other 
 higher formations are almost wholly composed 
 of the remains of animal matter, it seems strange 
 that Eozoon should be the only thing that can in 
 any way be attributed to organic structure. 
 Whether Eozoon is a fossil or not, it will always 
 remain an interesting subject of discussion for 
 students in palaeontology and mineralogy, and 
 the specimens prepared by the writer for the 
 London, Paris, Philadelphia, Chicago and other 
 Exhibitions, together with the pile of literature 
 on the subject remain forever as a proof of the 
 deep interest taken in this supposed fossil 
 Protozoa. 
 
 Like the Laurentian system, the Huronian 
 formation, which consists of a thickness of about 
 20,000 feet of strata composed of quartzites, 
 slates, limestone, jaspers, conglomerates and 
 many other varieties of rocks, contains no defin- 
 ite fossils. Certain markings on slates and other 
 rocks have been described and figured as the 
 26 
 
AMONG THE ROCKS. 
 
 trails of mollusca, &c., but these markings are so 
 indefinite that few palaeontologists believe them 
 to be of organic origin. In one of my papers on 
 concretionary structure, published in the Tran- 
 sactions of the Nova Scotian Institute of Science, 
 Series 2, Vol. I, I mention serious mistakes made 
 in assigning to concretionary and other markings 
 organic origin. 
 
 The hundreds of miles of Huronian rocks 
 examined by the writer, both in Canada and 
 Newfoundland, have yielded nothing which can 
 definitely be regarded as organic. Billings' 
 Aspidella terranovica, (Palaeozoic fossils, Vol. II,) 
 from the Huronian slates of Newfoundland cer- 
 tainly looks somewhat like a crushed and dis- 
 torted chiton ; but this also is one of those doubt- 
 ful forms which will be looked at shyly by the 
 palaeontologist of the present day. 
 
 Thus in these two great systems Huronian 
 and Laurentian no true fossils have been found. 
 There are reasons why life should have existed 
 in both the Laurentian and Huronian forma- 
 tions ; but till definite fossils are brought to 
 light, most palaeontologists and geologists will, I 
 think, consider these two great formations 
 Archean or Azoic. 
 
 27 
 
CHAPTER III. 
 
 A VISIT TO THE UNITED STATES JOURNEY TO THE ISLAND 
 OF ANTICOSTI MY ASSISTANT PROVES FORGETFUL AND 
 UNFORTUNATE DESCRIPTION OF THE ISLAND ITS GEO- 
 LOGICAL FORMATION SOME QUEER INHABITANTS I 
 SIGHT A COUPLE OF BEARS FURTHER MISFORTUNES OF 
 JACKSON DEPARTURE FOR QUEBEC. 
 
 It/! OST of the year 1864 was spent in assisting 
 "* Mr. Billings in palasontological and 
 Museum work. Most of the month of June was 
 occupied at Troy, Bald Mountain and Glens Falls, 
 N.Y., making collections of fossils from the 
 Primordial and Trenton formations. These were 
 needed for comparison with our "Quebec Group" 
 fossils. The specimens obtained at Troy and 
 Bald Mountain proved very interesting as they 
 threw new light on the relation of certain rocks 
 of N.Y. State to those of the "Quebec Group" of 
 Canada. 
 
 At Troy I was entertained and accompanied in 
 my researches by Mr. S. W. Ford, a young palaeon- 
 tologist of bright prospects. 
 
 At Bald Mountain, to be near my work, I 
 sought and found board and lodging in a typical 
 American farm house. The great kindness 
 shown me in this hospitable home, the interesting 
 nature of my work, together with the peaceful 
 28 
 
AMONG THE ROCKS. 
 
 rest after the day's work will always cause this 
 period to be remembered as one of the most 
 pleasant of my early explorations. 
 
 On the 1st of June 1865, 1 received instructions 
 from Sir William Logan to prepare for a journey 
 to the island of Anticosti. I was to take a man 
 who could act as cook and general assistant, the 
 necessary camp equipment and provisions for 
 two months. I was to examine and collect fossils 
 from the coast rocks between English Head and 
 Becscie River, on the south side of the island. 
 
 In a few days my arrangements were complete, 
 and furnished with a Government official letter 
 to the Captain of the S. S. Napoleon III, (which 
 carried supplies to the Gulf shore lighthouses) I 
 with my assistant started on what proved to be 
 an exceedingly interesting trip, but which was 
 frequently beset with laughable and often danger- 
 ous incidents, a few of which with a slight out- 
 line of the geological structure of the island I 
 shall try and record. 
 
 I knew nothing of the man who was going to 
 be my companion on a wild sea coast, excepting 
 that my friend Mr. Root, who had recommended 
 him to me, said he was a good cook and first-rate 
 all round camp man. My first experience of any 
 short- comings in Mr. Jackson (John Jackson was 
 his name) occurred while waiting for the Mon- 
 treal boat to start for Quebec. While standing 
 near some of our baggage I saw a little woman 
 29 
 
REMINISCENCES 
 
 bustling around looking anxiously for some one. 
 Seeing me among a lot of boxes she came up and 
 said : 
 
 "Are you the man who is going to Anticosti?" 
 
 I replied that I was. 
 
 "Well sir," she said, "I hope you will take care 
 of my husband, Mr. Jackson, for he can't take 
 care of hisself, and please sir where is he now?" 
 
 " Oh, he has forgotten his overcoat," I replied, 
 " and has run home for it." 
 
 "Ah, there it is," the woman said, "he never 
 does remember anything." 
 
 Just at that moment Jackson came up and 
 "All aboard " was called out. As we hustled the 
 little woman off I gave her a few dollars on 
 account of John's wages, and her last words to 
 John were, "Don't forget the baby." 
 
 The delightful voyage down our noble St. 
 Lawrence River, the courtesy of the officers on 
 our good ship, together with the little trials and 
 troubles of my assistant, will ever remain fresh 
 in my memory. We reached English Head Light- 
 house on the 18th of June, not, however, without 
 several strong reminders of the " perils of the 
 deep." 
 
 At the time of my visit to Anticosti in 1865, 
 there were no inhabitants at the west end, 
 excepting those connected with the lighthouse, 
 but fishermen from other parts of the island 
 frequently came on business to Mr. Malouin, the 
 30 
 
AMONG THE ROCKS. 
 
 light-keeper. These men, like most fishermen, 
 are a rough, hardy set, ever ready to assist one 
 another in times of trouble, always ready to 
 indulge in what they call little jokes, but, as 
 will be seen by the following incident, very 
 much opposed to having a joke played on them. 
 For several weeks Miss Maloain (daughter of 
 the light-keeper) had been anxiously looking out 
 for a priest, who at that time visited the island 
 occasionally to administer religious functions, 
 and as there was a probability of her confessor 
 arriving that day she and I stood at the foot of 
 the flag-staff speculating on his approach. In a 
 little while we saw in the distance a boat con- 
 taining several men, and with the Union Jack 
 flying. In a moment all was ready to hoist the 
 flag in honor of the priest's arrival. When, 
 however, the boat came within a short distance 
 of the shore, to the young lady's dismay she 
 discovered that it only contained fishermen, who, 
 knowing the priest was expected, thought they 
 would play one of their little jokes by flying the 
 flag, usually displayed in honor of an ecclesiastic. 
 Seeing disappointment and anger depicted on the 
 lady's face, I pointed to a fishing basket which 
 lay near the flag stafl ; my meaning was instantly 
 interpreted. "Good," said Miss Malouin ; and 
 in a moment up went the basket to the top of 
 the pole. But, alas ! my little joke was observed 
 by the fishermen, and when they reached shore 
 31 
 
REMINISCENCES 
 
 they made straight for me and were about to do 
 me bodily harm, but the brave girl rushed in 
 between us and cried, " Strike me if you want to 
 beat any one ! " on which the men slunk off, 
 muttering threats of vengeance on my head. 
 
 At English Head, as along the greatest part 
 of the south shore, a sloping reef extends sea- 
 wards probably from half-a-mile to a mile. 
 According to Bay field, the outer edge of the reef 
 forms a precipice from twenty to a hundred feet 
 deep. Ships approach the edge of this shelving 
 reef with great caution, sounding continually, 
 and in case of fog standing off two or three miles 
 from shore. 
 
 All lighthouse supplies are carried first in the 
 ship's boats, then in a cart which is kept for that 
 purpose ; with these our supplies and camp outfit 
 were landed. Letters of introduction to the 
 light-keeper, Mr. Malouin, gained me a cordial 
 welcome, and I was hospitably housed in the 
 lighthouse till the busy time caused by the 
 arrival of the Government supply boat was over, 
 and I could get the cart to carry me to Gamache 
 Bay, a distance, I think, of eight miles, where I 
 decided to make my first camp. 
 
 During my short stay at English Head, the 
 time was occupied collecting fossils and other 
 natural history specimens. The reef, already 
 referred to, which is dry at low tide, affords an 
 excellent opportunity for collecting some of the 
 32 
 
AMONG THE ROCKS. 
 
 characteristic fossils of the island, among which 
 is the fossil to which Billings gave the generic 
 name Beatricea, and classed it in the vegetable 
 kingdom. Hyatt supposes it to belong to the 
 Cephalopoda family, placing it with Miller in 
 the animal kingdom. The latter supposes it to 
 belong to the Bryozoa family, and assorts it with 
 fossil ^sponges. After seeing many hundreds of 
 these forms in situ, I quite agree with Billings 
 in giving it a vegetable origin, and think it will 
 ultimately be described as a fucoide. I have 
 never seen a perfect specimen of Beatricea. The 
 base or root of this plant-like form is frequently 
 met with, but the upper end is always broken 
 and does not show the termination, so at present 
 we do not know what length this fossil attained. 
 The longest section seen by me measured twenty 
 feet and was about six inches in diameter, while 
 fragments of others measured a foot in diameter 
 They are cylindrical in form, slightly tapering 
 towards the upper end; the exterior is rough, 
 resembling the bark of some trees. A tube 
 crossed transversely with highly concave septa 
 runs the whole length, while between this central 
 tube and the outside, concentric layers resembling 
 those of an exogen tree and a coralline structure 
 are seen. Billings describes these species from 
 Anticosti as Beatricea undulata and B. nodulosa 
 They occur in the Lower Silurian, and are most 
 abundant in the Hudson River formation in the 
 c 33 
 
REMINISCENCES 
 
 vicinity of the west-end lighthouse. This genus 
 was found in the Hudson River rocks of Rabbit 
 Island, L.S., by Dr. Bell, of the Canadian Survey, 
 and by the writer, in 1884, at Stony Mountain, 
 Manitoba. 
 
 My main object in visiting Anticosti was to 
 collect fossils from the rocks between Macasty 
 Bay, near the north-west end of the island, and 
 Becscie River on the south side, a distance in a 
 straight line of about thirty-seven miles, but 
 very much longer taking in the numerous bends 
 of the coast. 
 
 In these memoirs I can only speak in the 
 briefest manner on the geology of that portion 
 of the island visited ; leaving those interested in 
 the subject to peruse the able report of the late 
 Mr. James Richardson, of the Canadian Geologi- 
 cal Survey. The rocks of Anticosti belong to 
 the lower and Middle Silurian. The various 
 formations recognized by stratagraphy and fossils 
 are : the Hudson River, Oneida, Medina, Clinton, 
 and Niagara. A section of the Hudson River 
 rocks at the west end of the island gives a thick- 
 ness of 959 feet of strata, composed chiefly of 
 argillaceous limestones, limestone conglomerates 
 and other similar beds. These measures contain 
 fossils characteristic of the Hudson River forma- 
 tion, a formation which occupies a great portion 
 of the island, forming in places cliffs with an 
 
 34 
 
AMONG THE ROCKS. 
 
 elevation of from 100 to 400 feet, in places almost 
 perpendicular. 
 
 Two days after our arrival at the west-end 
 lighthouse we succeeded in getting the cart to 
 carry us and part of our camp outfit to Gamache, 
 or Ellis Bay, a distance of about eight miles. I 
 sat with the driver, the light-keeper, while Jack- 
 son sat on our baggage. When within a mile or 
 so of our destination, hearing Jackson whistling 
 " Home, Sweet Home," I turned to ask him if he 
 was thinking of "the baby," and glancing 
 towards our baggage to my dismay I saw that 
 several pieces of our camp equipment were miss- 
 ing. Fortunately, however, they belonged to 
 Jackson. These things actually slipped from 
 under him as we jolted along the rocky shore. 
 
 " Good gracious, Jackson/' I exclaimed, "what's 
 gone with your dunnage-bag, long boots and 
 overcoat ?" 
 
 " Good God !" Jackson said, jumped off the 
 cart, and started back in search of his " duds." 
 
 Our camp was pitched, supper eaten and the 
 shades of night were closing in when John re- 
 turned with the lost baggage, which he had 
 found just as the tide, which had risen, was 
 about to cover them. John was in anything but 
 an amiable humor, but a nip of rye from the 
 medicine chest, and a good supper, soon restored 
 peace in our camp, and before we retired for the 
 night we bagged several large lobsters, one of 
 35 
 
REMINISCENCES 
 
 which was the largest I have ever seen in 
 Canada. 
 
 During our journey from the west- end we saw 
 many seals, Phoca Grcenlandica, (Muller), which 
 were very numerous at that time. They lay 
 basking in the sun, but on our approach wriggled 
 off into the sea. 
 
 Part of the coast between west-end lighthouse 
 and Gamache Bay is exceedingly rough High 
 escarpments of dark and light grey calcareous 
 rocks most of the beds of which are prolific in 
 fossils, especially those of Junction Cliff, which, 
 owing to the over-hanging beds, form good rest- 
 ing places for the fossils which had "weathered" 
 out, and were gathered by handfuls. 
 
 Cape Henry on the west side of Gamache Bay 
 has a vertical face of about 300 feet. Similar 
 cliffs occupy a large portion of the coast, while 
 intervening; spaces are low land covered with a 
 coarse grass. Many varieties of fruit-bearing 
 trees and shrubs were seen between the west-end 
 and Gamache Bay, among which were : the moun- 
 tain-ash ; gooseberry ; currant ; cranberry, &c. 
 Pine, balsam, tamarack and other trees attain a 
 good size, and much of this wood is fit for building 
 purposes. During the ten days or so we camped 
 in Gamache Bay we visited the only three people 
 living in that vicinity : Luke, who kept the 
 government store where a supply of provisions, 
 &c., was kept, to be used only by wrecked sailors 
 36 
 
AMONG THE ROCKS. 
 
 and others who had been cast ashore ; an old man 
 named Murray, whom we found in a log cabin at 
 the head of the bay, bedridden and in a terrible 
 state of dirt and wretchedness, cursing heaven and 
 earth, and all things on land and in the sea. So 
 far as I could learn, this man was the only remain- 
 ing member of a small band of wreckers, whose 
 chief was the terrible pirate-king Gamache. The 
 grave of this miscreant is situated at the end of 
 the bay. If any of the descendants of Gamache's 
 gang of pirates are living they could no doubt 
 tell some fearful stories of wrecks caused by this 
 execrable old wrecker. The only other occupant 
 of the bay was a fisherman, who was away at 
 the time, but his wife, who lived in a small log 
 cabin, furnished us with lobsters, and other 
 " fruits of the sea," and sheltered me for two or 
 three nights when I was almost driven wild by 
 mosquitoes and black flies. I was also indebted 
 to this woman for the use of a boat to make a 
 journey to River Becscie, a distance of about fif- 
 teen miles east of the bay. This journey was 
 made to get a further supply of the beautiful 
 bivalve fossil shell Pentainerus Barrandi (Bil- 
 lings), which was discovered by Mr. Richardson 
 in 1856. Knowing this was to be a dangerous 
 journey, after a good bit of coaxing and a promise 
 to pay him $8.00 per day, I got Luke, who was 
 well acquainted with the coast, to allow Jackson 
 to mind his " wrecked sailor provisions store," 
 37 
 
REMINISCENCES 
 
 and accompany me. We left the bay at 2 o'clock 
 in the morning, when the tide was well up, but 
 to avoid boulders and other obstructions we had 
 to keep well out, perhaps a mile from shore. The 
 moon was shining brightly, and as our frail boat 
 glided through the lonely sea, I too, like Jackson, 
 felt inclined to whistle " Home, Sweet Home." 
 
 At six o'clock we landed for breakfast and a 
 nap, and while Luke prepared our meal I started 
 out with my revolver to try and shoot a duck or 
 two, which in that locality, were quite numerous ; 
 but before I had gone a hundred yards I found 
 myself within fifty paces or so of two bears. 
 They were separated from me by a small stream, 
 and appeared to be feeding on dead fish. I gave 
 a shrill whistle, a signal agreed on between Luke 
 and myself, in case of danger. In a few moments he 
 was at my side, and if I was surprised at seeing 
 two bears so near me, I was more surprised to 
 hear the blasphemous curses that Luke hurled at 
 these denizens of the forest. Whether it was 
 Luke's choice language that decided the bears 
 to "get a move on" or not I don't know, but 
 after surveying us for a moment they moved off 
 towards the bush. 
 
 The next evening we reached Becscie River, 
 where, after a long search, I found and made a 
 good collection from the Pentamerous beds 
 already spoken of. Owing to stormy weather 
 we were detained here two days, and as the sea 
 
AMONG THE ROCKS. 
 
 continued rough, and our small stock of pro- 
 visions gave out, we decided to haul up our boat 
 to a safe place and walk back. It was a tedious 
 walk over those fifteen miles of rough shore, " A 
 dreary beach with green waves tumbling free," 
 but I was amply repaid by finding several new 
 and important fossils. 
 
 During our absence, while keeping a sharp 
 look-out on the Government store, Jackson had 
 busied himself in administering to the wants of 
 poor Murray, whose dirt and depravity had so 
 shocked him that he was glad to be relieved from 
 his task. 
 
 Jackson, who was an inveterate disciple of 
 Izaak Walton, had, while we were away, dis- 
 covered a fine trout stream, a short distance from 
 the head of the bay. The fish were small, but 
 very numerous, so I concluded to spend half a 
 day fishing. 
 
 Early the next morning we set off for our trout 
 stream, which I found as Jackson had represented , 
 but here in the midst of our sport, misfortune 
 again befell Jackson. While stepping on a boul- 
 der in the stream he fell fiat in the water ; his 
 ducking did not, however, prevent him from con- 
 tinuing the sport. We reached camp that even- 
 ing with sufficient fish to give all our friends in 
 the bay a good feed. 
 
 The following day, according to previous 
 arrangement, the horse and cart were brought 
 39 
 
REMINISCENCES 
 
 to convey us again to the west-end. We re- 
 mained there several days collecting fossils and 
 other objects in natural history along the shore. 
 Here, at the lighthouse work-shop poor Jackson's 
 misfortunes followed him. We had obtained some 
 leather, and John was busy mending a pair of 
 boots, while I talked to the lighthouse keeper 
 who was sitting on his work-bench. Suddenly 
 the loud report of a gun immediately behind me ; 
 the shattering of a saw which hung on the wall 
 near the light-keeper, and the exclamation " Mon 
 Dieu" caused me to turn to where the report 
 came from, and there I saw Jackson holding a 
 large old-fashioned flint gun which had been 
 tinkered into a cap -lock weapon. It appears 
 this old gun had been loaded some time ago with 
 shot and a ball to shoot a bear which had been 
 seen prowling round the light buildings. How- 
 ever the bullet and some of the shot passed so 
 close to Mr. Malouin that it grazed his coat before 
 shattering the saw. Jackson had received a cut 
 on his cheek by the nipple flying out, and his 
 wrist and face were blackened with powder- 
 marks which would remain as long as he lived. 
 John was at once ordered to leave the buildings, 
 and Malouin suggested that I put him in quar- 
 antine. 
 
 At last, after much trouble, I succeeded in 
 getting a boat and men to take us to our last 
 camping ground Macasty Bay, a distance of 
 40 
 
AMONG THE ROCKS. 
 
 about twelve miles, in a straight line from 
 the west-end lighthouse, but as in our Becscie 
 River trip we had to stand well out from 
 the shore. During the journey fog set in, and 
 that with a rolling sea, a small boat, four men, a 
 large dog, which I had borrowed from the west- 
 end lighthouse folks, eight boxes of fossils and 
 provisions, rendered the journey a heavy and 
 somewhat dangerous one, but at last we reached 
 the bay and pitched our tent in a picturesque 
 spot near a small stream of clear water and a 
 considerable bush. 
 
 After a good supper of pork and pancakes, 
 which latter Jackson was past master in making, 
 the two men left us for the west-end. Jackson 
 and I sat at our camp-fire till quite late and then 
 turned into our blankets. How long we had 
 been in the " arms of Murphy," as one of my 
 men of former travels called it, I do not know, but 
 I, always a light sleeper while at camp, was 
 awakened by hearing some beast sniffing at our 
 pork -box just outside the tent. I gave Tom, the 
 dog, who was sleeping at my feet, a push, and 
 hastily unhooked the tent door. Tom rushed 
 out and gave chase to what I supposed to be a 
 stray cow from the west-end lighthouse stables, 
 but in the morning I was surprised to find the 
 tracks of a large bear all round the tent. Tom 
 slept on the outside of our tent for the rest of 
 our camp there. 
 
 41 
 
REMINISCENCES 
 
 We remained several days in Macasty Bay, 
 during which time we made journeys in both 
 directions along the cost. 
 
 Some of the cliffs in this locality are high and 
 almost perpendicular, and a mountain rises to a 
 height of about four hundred feet. This is an 
 interesting spot for the palaeontologist, for the 
 rocks are very prolific in fossils, and besides the 
 many new species of fossils obtained by us many 
 more remain to be collected ; indeed the field 
 was so good for collecting that I had decided to 
 remain a week or two longer and run the chance 
 of another schooner other than the one which 
 was to call for us. But on the following morn- 
 ing a schooner anchored opposite the bay, and 
 the men seeing our camp came ashore. As I 
 found they were bound for Gaspe', I decided, 
 much to the delight of Jackson, who had an idea 
 that at any rate his bones would be left on " the 
 damned island/' to leave. In a short time all our 
 traps were on board and the boat was steered 
 for Gaspe , but owing to head winds the journey 
 lasted three times the duration it would have 
 done with a fair wind. It was a constant tack- 
 ing till we reached Gaspe Basin. The stench 
 was so great in the cabin, and it looked such a 
 delightful place for parasites, that I spent one 
 whole night on deck. However, we reached 
 Gaspe' safely just as the steamer had left for 
 Quebec. This obliged us to remain there a fort- 
 42 
 
AMONG THE ROCKS. 
 
 night, which time was spent in exploring some 
 of the interesting geological sections of that 
 vicinity, one of which was at Douglastown River. 
 Here considerable excitement prevailed for some 
 time owing to the discovery of Petroleum springs. 
 Oil is seen oozing from the sandstones which 
 belong to the Lower Helderberg group (Upper 
 Silurian). The few fossils found here were 
 plants which have a Devonian aspect. 
 
 Another locality visited was Silver Brook, a 
 tributary of the York River. Here petroleum is 
 seen floating on the pools of water and oozing 
 from the rocks. 
 
 At the time of my visit to this place (1865) 
 boring for oil was in operation, but so far the 
 wells had only yielded salt water and gas. 
 Although this oil excitement is constantly 
 cropping up, I am not aware, even at this date? 
 more than thirty years after, that oil in paying 
 quantities has been obtained. A few interesting 
 fossils were obtained from the Silver Brook 
 rocks, and samples of petroleum collected for the 
 museum. 
 
 And now before closing this brief account of 
 my Anticosti trip, and bidding good-bye to our 
 friend Jackson, I must relate two other misfor- 
 tunes which befell him during our stay at 
 Gasp6. 
 
 Early one morning we provided ourselves with 
 a lunch, and started for a long geological tramp. 
 43 
 
REMINISCENCES 
 
 After a time we left the main road and took a 
 bush path which followed the base of a range of 
 high hills and cliffs. We had proceeded a mile or 
 so when we came to a log cabin. Jackson, always 
 of an enquiring mind, started towards the cabin, 
 but when within a few feet of the door, a vicious 
 bull-terrier rushed out and buried his teethin John's 
 leg. At that moment a woman ran out and with 
 a stick beat the dog off. I demanded that the 
 dog should be tied up, and when this was done 
 examined the wounded leg. Fortunately the 
 flesh was not torn. I bathed the limb with water 
 from a stream close by and bound it with my 
 handkerchief, and we started back. It was a 
 painful walk for poor Jackson, and his leg was 
 very much swollen when we reached the hotel, 
 where a doctor attended to the leg. Jackson 
 stoutly declined to have me send someone to 
 shoot the dog. saying : " Perhaps it is the only 
 protection that poor woman has in that lonely 
 wood." 
 
 Like the proverbial cat with nine lives my 
 servitor was soon able to be about again. 
 
 The steamer for Quebec was due in three days> 
 so leaving John to look out for the arrival of our 
 boat, and instructing him to have all our traps on 
 the wharf in time, I drove off to inspect some 
 supposed ore deposits and an oil spring. I was 
 back half a day before the boat was due, but to 
 my surprise she was in sight and making for the 
 44 
 
AMONG THE ROCKS. 
 
 basin, but Jackson, where was he ? I hurried to 
 the hotel, paid my bill, and on enquiring for 
 Jackson was told that he was, that morning, 
 going down to the wharf with the two bottles of 
 mineral water I had previously collected, and, 
 taking a short cut down a steep embankment, 
 had fallen, cutting himself severely with the 
 broken bottles, and that he was then lying in the 
 Harbormaster's office ; there I found the poor 
 fellow on a rough lounge, and covered with the 
 Union Jack. 
 
 By this time the steamer was at the wharf. 
 Going on board and explaining matters to the 
 Captain, he kindly allowed his men to fetch our 
 luggage from the store-room, and also to have 
 Jackson carried to the ship. The pitiable sight 
 my poor man presented will probably never 
 be forgotten. We left Gaspe* on the 27th of 
 August, reached Quebec safely, and in a short 
 time were on board our train bound for Montreal, 
 which place we reached a week after leaving 
 Gaspe". There I paid Jackson off and he at once 
 started for his home, where I trust he found his 
 wife and baby. I had long ago come to the con- 
 clusion that his wife was about right when she 
 said : " He can't take care of hisself." 
 
 I never saw the poor fellow again, and think 
 he died about a year after. So ends this brief 
 account of our Anticosti trip. The remainder of 
 the year 1865 and to July the following year, was 
 occupied in labelling and arranging my Anticosti 
 and other fossils and in my usual microscopical 
 and museum work. 
 
 45 
 
\ 
 
 CHAPTER IV. 
 
 THE SOUTH PETITE NATION RIVEE THE TRENTON FORMA- 
 TION THE CALCIFEROUS ROCKS OF ONTARIO WE LOSE 
 OUR WAY IN THE BUSH THE CHAZY FORMATION A 
 GUELPH FREETHINKER AN ADVENTURE WITH A BULL 
 VISIT TO POINT LE>IS AN ODD MISTAKE AT OUR 
 BOARDING-HOUSE THE EOZOON CONTROVERSY. 
 
 ON the 27th of July 1866, 1 left for the South 
 Petite Nation River and spent about ten 
 days examining and collecting fossils from the 
 Trenton formation. This series of rocks takes 
 its name from Trenton, Oneida county, New 
 York. Although its thickness in Canada does 
 not exceed probably 1,000 feet, it is palaeonto- 
 logically one of the most interesting formations. 
 It has been studied perhaps more than any other 
 formation, and has yielded a vast number of 
 species and genera; indeed the Trenton formation 
 has been likened to a magnificent museum of 
 well preserved shells. The extensive quarries of 
 this limestone in Montreal, Ottawa and many 
 other places have not only furnished stone for 
 our principal buildings, but have supplied our 
 museums with thousands of beautiful fossils. 
 
 In the distant days of the Geological Survey of 
 Canada, of which I now write, the Government 
 grant was small and one had to tramp many use- 
 46 
 
AMONG THE ROCKS. 
 
 less (geologically) miles of country to save ex- 
 pense. It was on one of these journeys in the 
 South Petite Nation district that I took what I 
 supposed to be a short cut through the woods, 
 but long before reaching the main road leading 
 to the log cabin where I was quartered, night 
 came on, and I had frequently to look at my 
 compass by the light of a match. It was while 
 groping for my road in this way that I stumbled 
 over what I supposed to be a bear, but which I 
 found by striking another light to be a cow, and 
 to my delight I stepped into a path which led to 
 the road. In a short time I was safely housed, 
 and if it was a one-roomed log cabin, and if I did 
 have to sleep in the one bed while Mike and his 
 wife camped on the floor in close proximity, and 
 the youngsters slept in another corner of the 
 room, I was thankful for even that shelter and 
 also for the " God bless you " I received, when 
 after a hearty breakfast of milk and potatoes I 
 shouldered my large fishing basket and started 
 for fresh fields. 
 
 A short time after returning from the Petite 
 Nation River I left for Chambly (a short distance 
 from Montreal) and spent a few days on the 
 Hudson River rocks. I then visited Kempville, 
 Oxford and other localities in Ontario, and 
 worked among the Calciferous rocks, collecting 
 many interesting fossils among which were 
 several new species. 
 
 47 
 
REMINISCENCES 
 
 The Calciferous sandrock received its name 
 from the New York geologists. It is developed 
 in many parts of the United States, where, as in 
 Canada it succeeds the Potsdam formation and is 
 chiefly a magnesian limestone. Its thickness in 
 Canada is supposed to be about 300 feet. It is 
 not very prolific in fossils, nor are the species 
 numerous as in the succeeding formations. It is 
 supposed the Lamellibranchiata that numerous 
 class of bivalves which are not equilateral, or 
 divided equally on either side of the apex com- 
 menced their life in this part of the Lower 
 Silurian. 
 
 Before the season ended, I accompanied Mr. 
 Billings to Isle Bizard, where we examined the 
 Calciferous and Chazy rocks. Our chief object 
 was to collect specimens of that characteristic 
 fossil of the Chazy formation, Bolboporites. The 
 weather was lovely, and Mr. Billings, who seldom 
 left the office for field work, was like a school- 
 boy let loose for a holiday. 
 
 Here a little incident caused us a good laugh. 
 
 We had finished our work and started on our 
 return journey. Opposite the house where we 
 had boarded was a considerable bush, and to save 
 a long detour we determined to cross it. Blocks 
 of Chazy and Calciferous limestone were strewn 
 over the ground, and we kept going from one to 
 another, picking up a fossil here and there, and 
 working, as we supposed, towards our destination, 
 48 
 
AMONG THE ROCKS. 
 
 but to our surprise, instead of being a mile or so 
 away, we came plump up in front of the house 
 we had left. I laughed heartily in which Mr. 
 Billings joined at the time saying . " Take a 
 bearing with the compass, and you lead the 
 road." This time we got through the bush all 
 right and were soon back again in Montreal. 
 
 The Chazy formation takes its name from 
 Chazy, Clinton county, on the west side of Lake 
 Champlain. It has fine beds of workable lime- 
 stone, both for building and for ornamental work. 
 Some of the beds at Montreal, Caughnawaga, 
 and other localities furnish a good marble. It 
 contains many species and genera of fossils, and 
 is characterized by the small Brachiopod Rhyn- 
 chonella, of which some of the beds are almost 
 wholly composed. 
 
 Phosphatic nodules, which are supposed to be 
 coprolites, occur in abundance in some of the 
 beds ; they are composed chiefly of the remains 
 of the bivalve Lingula, which is supposed to 
 have formed the food of the animals from which 
 the coprolites were derived. It is in the Chazy 
 formation that we find many interesting Crus- 
 tacea (Trilobites), some of which are rolled up as 
 if to protect themselves from danger, as some of 
 the lower animals of the present day do. 
 
 My field work during the summer of 1867 was 
 confined to collecting fossils from the Guelph 
 formation at Guelph, Gait, Elora, Hespeler, and 
 
 D 49 
 
REMINISCENCES 
 
 other localities where the Guelph rocks are 
 exposed. This formation takes its name from 
 the town of Guelph, on the river Speed, Ontario. 
 It occupies a position next above the Niagara 
 formation, and forms the top of the Middle 
 Silurian; is about 160 feet thick and is composed 
 chiefly of yellowish white dolomite, or mag- 
 nesian limestone. Some of the beds are hard ? 
 and make a good building stone. The beds at 
 Guelph and Gait are characterized by casts of 
 the large bivalve, Megalomus Canadensis (Hall). 
 
 The Guelph formation has yielded a large 
 fauna of fossils, many of which were described 
 by the late Mr. Billings. Other new species are 
 now being worked out by his successor, Mr. 
 Whiteaves, the present palaeontologist to the 
 Geological Survey of Canada. The numerous 
 fossils many of which were new to science 
 now in the cases of the Dominion Geological 
 Museum, show some of the result of our labor 
 among these interesting rocks. 
 
 I close these brief remarks on the Guelph 
 rocks by recording the name of my old friend 
 Mr. James Harris of Hespeler at whose house I 
 boarded. Both Mr. Harris and his two daugh- 
 ters worked in the woollen factory of Hespeler. 
 They were humble people, and while with 
 them I did not " live on the fat of the land," but 
 they gave of the best they had. This was the 
 first time I had lived with a confessed infidel 
 50 
 
AMONG THE ROCKS. 
 
 for such was old man Harris and what with my 
 fossils and the learned talk of my host I found 
 plenty of food, if not for the body, for the 
 mind. There was one thing remarkable about 
 the old man's belief, and that was, when Sunday 
 came he always insisted on his children attend- 
 ing Sunday school, and learning their catechism. 
 
 Old man Harris is always associated in my 
 mind with the Guelph formation. The old man 
 must be dead long ago. I wonder whether he 
 still holds the same views of religion. As I 
 visited Hespeler and my old friend later on I 
 shall perhaps speak of him again. 
 
 The following month and until June 17th, 
 1868, I was fully occupied with museum, micro- 
 scopical and other work relating to the palaeon- 
 tological branch of the Survey. Then on the 
 last named date commenced a series of geological 
 excursions which proved highly interesting and 
 furnished much new palseontological information 
 regarding the great range of rocks on the 
 south side of the St. Lawrence between Point 
 Levis and Gaspe", and now known as a portion of 
 the " Quebec group," but which, owing to the 
 lack of fossil evidence, were in the early days of 
 the Survey assigned to various horizons. 
 
 My examination of these rocks commenced at 
 Riviere Ouelle, 80 miles below Quebec, to which 
 place I was sent by Sir William Logan to ex- 
 amine in situ, a series of black phosphatic no- 
 51 
 
REMINISCENCES 
 
 dules, which were then, as now, supposed to be 
 coprolites, and the singular, cylindrical fossil- 
 looking forms which occur with the nodules. 
 These have been described in the Geology of 
 Canada for 1863, and are spoken of by myself 
 and illustrated by micro-drawings in the Geo- 
 logical Report for 1^76-77, pp. 432-33. Part of 
 the river shore is exceedingly rough and the 
 beach is strewn, or was at that time, with the 
 bones of the white whale, or Beluga, which were 
 trapped here for their oil. Like the seal Phoca 
 Grcenlandica, these small whales, or porpoise, as 
 they are called by some, are of great antiquity, 
 their skeletons having been found in the blue 
 clay many feet below the surface, and far away 
 from river or sea. 
 
 The cases of the Geological Museum, Ottawa, 
 contain many fossil bones of the last two species, 
 a great number of fossil shells and plants from 
 the Post-Tertiary clays of Montreal, Ottawa and 
 other localities. One case contains almost an 
 entire skeleton of a harp seal Phoca Grrcenlan- 
 dica (Muller). The label attached to this valu- 
 able specimen reads : 
 
 Skeleton of Harp Seal 
 Phoca Grcenlandica (Muller), 
 
 Post-Tertiary clays, Montreal. Found at the 
 depth of 30 feet below the surface in Peel and 
 Conte's clay pit 1861. Weston. 
 52 
 
AMONG THE ROCKS. 
 
 After collecting the above specimens I spent 
 much time in freeing the bones from the clay in 
 which they were imbedded, and then in articula- 
 ting them, in which shape it remained till about 
 1887, when it was re -articulated and restored by 
 Mr. Jules Baily of Montreal, thus, in my opinion, 
 deteriorating the fossil aspect of the remains. 
 
 Besides the fossil bones and shells mentioned, 
 these Saxicava clays (which are largely used for 
 brick-making) contain, at Green's Creek, Glou- 
 cester, Ottawa, abundant nodules in which are 
 found beautiful examples of Mallotus villosus 
 (Cuvier); that delicate fish, the Capeling of the 
 lower St. Lawrence, one of the few fishes which 
 has weathered the storms and changes of cent- 
 uries. 
 
 While pacing the distance along Riviere Ouelle 
 and noting the various outcrops of rocks between 
 the railway station and the coast, I had a laugh- 
 able and what might have been a serious encoun- 
 ter with a bull. I was very intent on my task 
 as this was my first attempt to make geological 
 measurements with a view to plotting my work. 
 All went well, till, seeing a peculiar clump of 
 rocks in a field, I decided to make an offset 
 from the river road. About half the distance 
 was covered when to my surprise a wicked looking 
 bull came tearing along from the other side of 
 the field ; for a moment my chance of escape 
 seemed very slim indeed, but dropping my geo- 
 53 
 
REMINISCENCES 
 
 logical traps, I ran as one runs for his life, and 
 succeeded in gaining the clump of rocks, up which 
 I scrambled just as his bullship was about to 
 hook me in the back. A shout brought a farmer 
 from the next field, and after some little trouble 
 I regained rny instruments, but my line of meas- 
 urement was never completed. 
 
 Soon after my return from Riviere Ouelle I was 
 again summoned to Quebec to accompany and 
 assist my old friend Mr. Richardson in his inves- 
 tigation of that portion of the " Quebec Group " 
 occupying the coast between Riviere du Loup 
 and Rimouski, a distance of about 60 miles. 
 
 During the autumn of 1867, and all the sum- 
 mer of 1868, Mr. Richardson was engaged in trac- 
 ing out the structure of these Quebec rocks. Up 
 to 1867, very little was known as to the palseon- 
 tological history of the coast rocks between 
 Riviere Ouelle, already referred to, and Rimouski. 
 The discovery by Mr. Richardson this summer 
 (1868), of one or more species of Trilobites was 
 considered so important that he telegraphed the 
 fact to Sir Wm. Logan and requested him to send 
 me to assist in searching for other forms. 1 was 
 at once despatched to Point Levis, where I found 
 Mr. Richardson waiting for me at the railway 
 station. It was a dark night, and I was cold and 
 hungry. In those days we had no luxurious 
 parlor and dining cars, but in a few moments we 
 were housed in the Victoria Hotel, that comfor- 
 54 
 
AMONG THE ROCKS. 
 
 table house which has since then sheltered Logan, 
 Billings, Webster, Ells and other members of the 
 geological staff, after days of hard tramps over 
 the Levis rocks. 
 
 A glance at the fossils Mr. Richardson had col- 
 lected showed at once that they belonged to the 
 Primordial formation, a fact which threw a new 
 light on a large portion of the rocks of the St. 
 Lawrence river.* 
 
 The following morning we bade good-bye to 
 our genial hostess Mrs. Tofield, who has long 
 been numbered with the great majority, and whose 
 comfortable resting place for the traveller has 
 long been closed owing to the removal of the 
 Grand Trunk, Quebec and Levis depot to a more 
 convenient site, in close proximity to the Quebec 
 and Le" vis ferry boats and started for St. Roch, a 
 small village,close to the shore of the St. Lawrence 
 river, which place had yielded the precious fossils 
 which were to aid so materially in working out 
 the complicated structure of the Quebec rocks. 
 Our first day's researches proved highly satisfac- 
 tory, for we had bagged eight species of fossils 
 most of which were new and all of a decidedly 
 Primordial aspect. All these forms were obtained 
 from the limestone pebbles of the conglomerates, 
 a fact which left us to conclude that the matrix 
 
 *In these notes I speak of the various geological horizons 
 as they were known then, and not as they are known at the 
 present day, 1897. 
 
 55 
 
REMINISCENCES 
 
 of the conglomerates and the other bedded rocks 
 were newer and belonged to the Potsdam forma- 
 tion. 
 
 We continued our examination of these shore 
 rocks at St. Anne, St. Denis, St. Paschal, Riviere 
 du Loup, Trois Pistoles, St. Simon, Rimouski 
 and other places. The result of our work was 
 considered by Mr. Richardson very satisfactory. 
 
 Our mode of travel was chiefly by Grand 
 Trunk Railway as far as Riviere du Loup ; east 
 of that to Rimouski we travelled either by buggy 
 or by Her Majesty's mail (often being the whole 
 night on the road) which sometimes consisted of 
 a buck-board and one horse. 
 
 The construction of the Intercolonial Railway 
 between Riviere du Loup and Halifax a con- 
 tinuation of the Grand Trunk line was then in 
 progress, and afforded us both opportunities of 
 travel and an insight into rocks hitherto con- 
 cealed. During our journeys in this section of 
 the country we saw much of the poor farming 
 class of French Canadians. Few spoke even 
 a little English, and as both Richardson's and 
 my own French was very limited, we often 
 made laughable mistakes ; two examples I may 
 mention here. 
 
 Being one day in a house where none of the 
 
 inmates could speak English, I undertook while 
 
 taking my soup to ask for a table napkin, but in 
 
 some way mistook the French word " sheet " for 
 
 56 
 
AMONG THE ROCKS. 
 
 that of " napkin." Judge of my surprise when 
 the old lady brought me a large white sheet. 
 
 On another occasion our horse gave out and 
 we were obliged to put up for the night at a log 
 cabin where both food and accommodation were 
 at a low ebb. My failure regarding the table 
 napkin induced me to ask Richardson to try his 
 hand at Canadian French, so when asked by the 
 woman of the shanty " how many eggs we 
 wanted for supper, and whether boiled or fried," 
 which modes of cooking were conveyed to us by 
 signs, Richardson said in French, as he thought, 
 " four boiled eggs/' but when we sat down to 
 supper behold there were four dozen. We ate 
 four and after assuring the poor woman that we 
 would pay for the whole four dozen, we divided 
 the remaining forty -four among the members of 
 the family. They were still feasting on the 
 remains of our supper when we entered the straw 
 bed on the floor of the loft of the cabin. This 
 trip finished my field work for 1868. 
 
 Between the last date to^the following June I 
 was fully occupied in the museum and in my work- 
 room. The Eozoon controversy of which I shall 
 again speak later on was still one of the inter- 
 esting geological subjects of the day, and much 
 of my time was spent preparing microscopic sec- 
 tions of Laurentian, Huronian and other rocks. 
 Sir William Logan watched my operations 
 with keen interest, and spent hours over my 
 57 
 
REMINISCENCES 
 
 microscope examining the sections as they were 
 prepared. 
 
 About this time Sir William appointed me 
 Librarian to the Survey. I was to spend a short 
 time every morning, before going to my more 
 important duties, in recording the various publi- 
 cations purchased and presented, and attending 
 to the distribution of Survey reports, etc. I 
 retained the name of Librarian to the Survey for 
 six years, when a Mr. White was appointed to 
 relieve me from that work. The library at that 
 time consisted principally of books belonging to 
 Sir William, most of which were, after Sir 
 William's death, purchased from the Logan 
 estate. These books formed the nucleus of the 
 library of the Geological Survey of Canada 
 which to-day (1897) contains 12,000 volumes, 
 and has for Librarian the learned and venerable 
 Dr. J. Thorburn, who has two lady assistants. 
 In the last Geological Report, 1894, it is recorded 
 that during the year ending 31st December, 
 there were distributed 5,666 copies of the Survey 
 publication, This year (1897) 9,922 copies were 
 distributed. This record of the doings of the 
 library of the Geological Survey of to-day (1897) 
 marks the great progress in this department 
 since the time when I managed the affairs of the 
 library in an hour or so each day. 
 
 58 
 
CHAPTER V. 
 
 THE FIELD SEASON OF 1869 JOURNEY WITH SIB. RICH- 
 ARDSON LOST IN THE BUSH THE PRIMORDIAL ROCKS 
 IN NEW BRUNSWICK ARISAIG, NOVA SCOTIA, AND 
 ITS COAST ROCKS RETIREMENT OF SIR WILLIAM 
 LOGAN FIELD SEASON OF 1870 THE ROCKS OP 
 QUEBEC ANOTHER VISIT TO THE UNITED STATES 
 VISIT TO STANFOLD, QUE. FIELD WORK OF 1871-2 
 THE COAST OF LABRADOR. 
 
 ^T^HE geological field season of 1869,was a busy 
 and interesting one for me. Many hundreds 
 of miles were covered and much information 
 gained regarding the Primordial and other rocks 
 then under discussion. The first journey made 
 (June 12th) was to Beloeil Mountain, P. Q., 
 to examine a small outcrop of Devonian lime- 
 stone. The distance between the small village of 
 St. Hilaire and this limestone was measured by 
 pacing and with a prismatic compass ; the var- 
 ious varieties of diorite and other intrusive rocks 
 which form the mountain were noted, and a small 
 but instructive collection of fossils made from 
 the limestones. 
 
 This mountain and exposure of limestone is 
 
 always associated in my mind with my old friend 
 
 and colleague Mr. James Richardson. It was 
 
 the following summer I think when I again 
 
 59 
 
REMINISCENCES 
 
 visited this locality with him to make further 
 geological observations. 
 
 Arriving at the village already mentioned Mr. 
 Richardson decided to cross the mountain 
 through the bush while I took the cart-road to 
 our limestone locality at which place we were to 
 meet again. It was after one o'clock when I 
 reached this spot. The day was hot and sultry 
 and the mosquitoes thick and hungry. I was 
 tired, and concluded not to wait for Richardson, 
 but to take my lunch and sample the bottle of 
 claret we had provided ourselves with from the 
 village. After lunch and my after-dinner pipe I 
 began to be anxious about my companion. But 
 time wore on and the sun was near setting when 
 I reluctantly decided he was lost a strange con- 
 clusion to come to about a man who had spent a 
 large part of his life in the bush. But coming to 
 this conclusion I at once built a large smudge fire, 
 and as the smoke ascended prayed that my dear 
 old friend would see it. The shades of evening 
 were just beginning to deepen when a rustling in 
 the bush caused me to look up, and there was 
 poor Richardson with clothes torn and bathed in 
 perspiration. He sank down by the smudge fire 
 thoroughly exhausted. I at once gave him a 
 drink from our flask of claret, but for a time I 
 feared he had a sunstroke, and was doubtful 
 whether he could be got back to the village that 
 night. After bathing him with water, however, 
 60 
 
AMONG THE ROCKS. 
 
 which I carried in my hat from a spring some 
 little distance off, and drying his clothes, he 
 revived, and before dark we were on our way to 
 the village. Mr. Richardson had followed a ridge 
 of rocks for a long distance through the bush and, 
 after several hours' tramp, found he had lost the 
 bearing of the locality where he was to join me. 
 Knowing I would wait for him at the appointed 
 spot, he had struggled on, but at last saw the 
 smoke of my smudge fire and was soon with me 
 again. In after years when we sat by our camp 
 fire he would often refer to this incident and 
 say "I never was so exhausted in my life, and if 
 I had not seen the smoke from your smudge, I 
 should have lain down and perhaps never seen 
 you again." 
 
 Part of June and till the latter part of July 
 was spent in visiting various localities in New 
 Brunswick. The Primordial rocks were occupy- 
 ing Mr. Billings' attention and it was thought 
 advisable to have me examine and collect fossils 
 from various localities where this formation was 
 well developed. 
 
 Arriving at St. John I at once sought Mr. 
 George Matthew of the Customs department, who, 
 in conjunction with Prof. Baily, during their 
 vacations, have done good geological work for the 
 Survey. I wish to record here my thanks to 
 Mr. Matthew for his kindness in furnishing me 
 much information regarding the places I was 
 61 
 
REMINISCENCES 
 
 about to visit, viz Mouthpath, Ratcliffs Mill 
 Stream, Clark's Point, Oak Bay and other local- 
 ities. 
 
 The weather was beautiful, but occasionally 
 foggy, as is frequently the case in the vicinity of 
 the harbor of St. John. Prof. Baily joined me 
 for a few hours, and together we examined and 
 collected fossils from several of the exposures in 
 and near the city. Quite a number of interesting 
 forms were found, many of which are now in the 
 cases of the Survey Museum. A journey was 
 then made by buggy to RatclifFs Mill Stream, a 
 few miles from St. John, where the slates are 
 prolific in Primordial fossils. It is the home of 
 that pretty little characteristic Primordial Crus- 
 tacean, M icrodiscus which I shall mention again 
 while speaking of the Newfoundland rocks. 
 Here I obtained lodgings with some of the mem- 
 bers of the family after whom the stream was 
 named. My bed was on the floor of the dining, 
 sitting and general room. Outside a dense fog 
 prevailed and the night was chilly, but in the 
 house a wood fire blazed in a large open fire- 
 place and I expected to sleep well after a hard 
 day's work. Soon after lying down, however, it 
 seemed that there were dozens of mice playing 
 hide-and-seek, and when I did sleep I dreamed 
 that these wretched little rodents were dancing 
 a quadrille on the coverlet of my bed. In the 
 morning, while I ate uiy porridge at one end of 
 62 
 
AMONG THE ROCKS. 
 
 the table, a daughter of my host combed her 
 raven tresses at the other end. 
 
 From St. John a journey was then made to 
 Oak Bay, Clark's Point and several other places, 
 chiefly in a stage coach much resembling the 
 ancient English coaches which carried Her 
 Majesty's mail and passengers before the days of 
 railways. 
 
 Some of these excursions were exceedingly 
 pleasant. From the top of one of these coaches 
 one not only sees delightful verdant hills and 
 valleys, lakes and streams, but occasionally gets 
 a sniff of sea air, wafted across the hills and 
 valleys from the Bay of Fundy. 
 
 My stay at Oak Bay was a pleasant one. Here, 
 while collecting fossils, a man came up to me and 
 asked many questions, among which were the 
 same our Chinese friend of the yellow jacket, Li 
 Hung Chang, has lately been asking so many of 
 our country people, "Where do you come from?" 
 "How old are you?" "What salary do you get?" 
 In return for the information I gave him, he in- 
 formed me that his name was Weston, and when 
 I told him that was my name he was so tickled 
 as our American friends call it that we be- 
 came great chums, and it was through him I 
 spent some most enjoyable evenings with the 
 people of Oak Bay. 
 
 Having visited the localities mentioned and 
 completed, as far as time would allow, the work 
 63 
 
REMINISCENCES 
 
 allotted to me, I returned to St. John and was 
 soon on my way back to Montreal. 
 
 On the 7th of August, a few days after my re- 
 turn from St. John, I was directed to proceed to 
 Arisaig, Nova Scotia, and make a collection from 
 the coast rocks of that place. About three weeks 
 were spent at this work, and a large collection 
 of fossils obtained, many of which were after- 
 wards described by Mr. Billings. I shall refer 
 to this locality again when speaking of my 
 travels of 1873. The usual routine of museum, 
 microscopical and other work, fully occupied my 
 time till the following summer. 
 
 During this year (1869) two important events 
 in the history of the Survey occurred the re- 
 tirement of our dear old chief, Sir William Logan, 
 and the appointment of his successor, Mr. A. R. 
 C. Selwyn. Logan had looked forward for years 
 to the time when he would be able to lay aside 
 the cares of the Survey and devote the remainder 
 of his working days to his beloved Eastern Town- 
 ship rocks, but he continued to hold the reins of 
 the Survey till this year, 1869, when he tendered 
 his resignation to the Government, and during 
 the month of December following Mr. A. R. C. 
 Selwyn was appointed his successor. Sir William's 
 health was then failing, but before he finally gave 
 up field work, he spent many days working out 
 certain facts relating to the geology of the East- 
 ern Townships. It was my good fortune to be 
 64 
 
AMONG THE ROCKS. 
 
 chosen to accompany and assist him in the last 
 field work he did. 
 
 The following summer, 1870, was a busy one 
 in field work. I had so far been very fortunate 
 in fossil collecting; for this reason it was thought 
 advisable to have further collections made and 
 notes taken of any new cuttings or quarries in 
 localities which had already been reported on. 
 Receiving instructions from the director I started 
 June 5th for Grimsby, Ont., where I was to make 
 a collection of fossils and collect other geological 
 information regarding the Niagara rocks. I 
 was fortunate enough, at the pretty little village 
 of Grimsby, to get board and lodging with Mr. 
 Johnson Pettit, a veteran entomologist and col- 
 lector of fossils, to whom the Geological 
 Survey is indebted for many fine specimens of 
 the latter. 
 
 The Niagara formation derives its name from 
 Niagara, where, at the Falls, it is largely devel- 
 oped, consisting of shales and limestones. Many 
 of the beds are very prolific in fossils, a great 
 number of which have been figured and described 
 by Hall, Billings, and other palaeontologists. 
 This formation covers a large area of Ontario 
 (Upper Canada). After a short visit to Niagara 
 Falls, Mr. Pettit and I set diligently to work. 
 The large quarry, and constant wearing away of 
 the softer beds of the cliffs afford a good oppor- 
 tunity to the collector of fossils. It was from 
 E ' 65 
 
REMINISCENCES 
 
 the well-washed banks in the vicinity of the 
 quarries of Grimsby that we obtained the fine 
 specimens of Garyocrinus, and other forms now 
 in the cases of the Dominion Museum. 
 
 Grimsby is a delightful spot, both for the 
 palaeontologist, artist, sportsman, and agricul- 
 turist, situated almost on the banks of Lake 
 Erie, and within a few miles of Niagara thunder 
 of waters Falls, which discharge 18,000,000 
 cubic feet of water per minute over an elevation 
 of 150 feet of blueish-gray limestone. Hundreds 
 of tons of grapes, apples, pears, peaches and 
 other fruit are raised annually in this vicinity. 
 
 Johnson Petti t, of whom I have spoken, lived in 
 the old Pettit homestead, a mile or so from the 
 village of Grimsby. It was in this dear old farm 
 house I spent many pleasant evenings and rest- 
 ful nights after the day's work was finished. 
 The first night, however, was not a restful one. 
 I had been allotted the spare bed- room which 
 contained a large old-fashioned four-post bed. 
 Immediately above the head of the bed were 
 several pegs, and on one of these hung a large 
 wide-brimmed beaver hat such as our Quaker 
 great grandfathers might have worn. When I 
 extinguished the light, the moon shone through 
 the window giving the old hat and other gar- 
 ments an uncanny appearance. After a time I 
 dozed off, and in a restless sleep dreamed that a 
 skeleton had taken possession of the hat and, 
 
AMONG THE ROCKS. 
 
 with it on his skull, was walking off. I jumped 
 up and found the old hat had fallen from the 
 peg. At breakfast I related the incident to the 
 family and was informed that during the night 
 there had been a shock of earthquake, which was 
 no doubt the cause of this singular occurrence. 
 The hat and other garments were removed from 
 the room, and I slept peacefully the remaining 
 nights of my visit. 
 
 A few days after my return from Grimsby I 
 left for the island of Orleans, P.Q., to examine 
 and collect fossils at St. Laurent village and other 
 localities in that vicinity. It was at a place 
 called St. Patrick's Hole, near the village of St. 
 Laurent, that I was fortunate enough to find a 
 good specimen of Eophyton. It was the first 
 time this genus had been found in our Canadian 
 rocks, and was considered important as it helped 
 to determine the horizon of a portion of the rocks 
 of the " Quebec group." There has been much 
 discussion as to whether this fossil is a plant or 
 the track of some animal. I quote the following 
 remarks, made by Mr. Billings, from the Geolo- 
 gical Survey Report for 1870-71 : 
 
 ** On the south side of the island of Orleans, near the 
 village of St. Laurent, Eophyton was found last spring by 
 Mr. Weston. It is there associated with several species 
 of fucoids identical with those that occur on Great Bell 
 Island (Newfoundland). The rocks here referred to 
 belong to the ' Quebec group,' but as they are consider- 
 ably disturbed it is not impossible that they may be older 
 and brought up by a fault." 
 
 67 
 
REMINISCENCES 
 
 At any rate this genus was supposed to be 
 typical of the Primordial zone, or not to occur 
 higher than the Lower Potsdam formation. 
 
 Part of the month of July of this year (1870), 
 was spent on the Chaudiere and Famine Rivers, 
 in the Province of Quebec. I have already 
 spoken of the seven pound weight of gold taken 
 from the Chaudiere and its tributaries. Gold in 
 the sands of the Chaudiere has been known since 
 1835, from which time gold mining has been 
 carried on more or less annually till the present 
 time, (1897). In 1894 the gold deposits of the 
 Province of Quebec yielded 1,622 ounces of the 
 precious metal, valued at $9,196. 
 
 My work on these rocks was not connected 
 with their gold deposits ; but to try to find 
 fossils, and obtain other information which would 
 settle the disputed horizon of certain rocks of 
 this vicinity. While at work here I had the 
 pleasure of accompanying Mr. A. R. C. Selwyn, 
 our new director, who was making his first geo- 
 logical explorations in Canada. Both Mr. Selwyn 
 and I were, for a short time, the guests of Mr. 
 W. P. Lock wood, superintendent of the Canadian 
 and Northwest Land and Mining Company. To 
 this gentleman Mr. Selwyn was indebted for 
 much information about the auriferous deposits on 
 the Chaudiere and its tributaries; and to his charm- 
 ing wife, our hostess, we were both indebted for 
 many kindnesses, one of which I must relate here. 
 68 
 
AMONG THE ROCKS. 
 
 After taking breakfast one morning at Mr. 
 Lock wood's hospitable house, I started out to 
 examine the rocks of the Chaudiere. When I 
 returned in the evening I found my pocket-book 
 containing fifty or sixty dollars was missing. 
 Knowing that I must have lost it since leaving 
 the house, I started to retrace my steps along the 
 river. I had gone about a mile when I came to 
 a small pool of water over which I had jumped 
 in the morning, and there in this shallow puddle 
 was my pocket-book saturated with water. I 
 hastened back to the house, and was congratu- 
 lated on the recovery of my money, and here the 
 kindness of our hostess came in. She took the 
 bank notes and other papers from the purse, 
 spread them on plates and dried them in the 
 kitchen oven. 
 
 Leaving the Chaudiere, I next visited several 
 limestone exposures on the Famine River, and 
 obtained from known fossiliferous rocks many 
 well defined fossils of a Devonian aspect. A list 
 of some of the fossils of this locality is given 
 in the Geology of Canada for 1863. 
 
 Most of the month of August was spent on the 
 Trenton limestones of Winchester, Ont., and at 
 the village of Industry, P.Q. A large collection 
 of fossils was made and notes taken of any 
 cuttings, etc. 
 
 During the month of September I again visit- 
 ed Whitehall, State of Vermont. The work done 
 69 
 
REMINISCENCES 
 
 by Sir William Logan and myself in 1868 had 
 interested Mr. Billings so much that he decided 
 to visit the localities from which we had obtained 
 fossils. We spent several days at this work and 
 collected information which aided materially in 
 connecting the geology of this part of the States 
 with that of Canada. 
 
 We were working away at our fossil lime- 
 stones, about a mile from the hotel, when Mr. 
 Billings took a sudden desire to return to Mont- 
 real that day. Nothing I could say would in- 
 duce him to stay longer. " All right," I said, 
 " we will leave by the evening train, I have paid 
 expenses so far, give me money for our return 
 trip." " But," said Billings, " I haven't a dollar 
 in my pocket." He was not well at the time, 
 and this seemed to make him still more nervous. 
 I at once wired the office to telegraph us funds 
 to the Whitehall Canadian Express Company, 
 but strange to say we received no reply. How- 
 ever I felt bound to get iny superior officer home 
 that night. " Give me your note-book," I said, and 
 with that and my own I went to the express 
 office, saw the manager, told him the fix we were 
 in, and requested him to furnish us funds to re- 
 turn to Montreal, offering to leave our note-books 
 as security till we refunded the money either to 
 him or his company at Montreal. The old gen- 
 tleman gave a hearty laugh and at once furnished 
 the funds required, saying we could return it to 
 70 
 
\ 
 
AMONG THE ROCKS. 
 
 his company at Montreal. I hastened to the 
 hotel, paid our bill, handed Mr. Billings into a cab, 
 and in a short time we were in our train bound for 
 home. It is a rather singular coincidence that 
 when I was in the same town with Sir William, 
 in 1863, he too was without money and borrowed 
 funds from me to pay his way to Albany as I 
 have already related. 
 
 Before closing my remarks for 1870 I must 
 record the death of my friend W. B. Hartley, a 
 young engineer who joined the Survey three 
 years, I think, before his death this year. Mr. 
 Hartley was with Sir William Logan, engaged 
 investigating the structure of the Pictou Coal- 
 fields, N,S. This work Hartley continued in 
 1870, but towards the end of the season over- 
 exertion and exposure in the mines brought on a 
 sickness which soon terminated the life of this 
 bright young man. Mr. Scott Barlow, of the 
 Survey, attended him in his last sickness at the 
 hotel in New Glasgow, N.S., and it was my pain- 
 ful duty to assist his father in arranging his 
 papers and personal effects in his office at the 
 Survey. 
 
 From the winter of 1870 to July 7th, 1871, I 
 was occupied with the usual museum and other 
 work of the Survey. On the latter date I went 
 to Eaton, P. Q., to examine certain micaceous 
 schists and other rocks in which no fossils had 
 hitherto been found. Mr. Richardson, of the 
 71 
 
REMINISCENCES 
 
 Survey, who accompanied me, left Eaton to 
 attend to some private matters for Sir William 
 Logan, but before going said, " You will never 
 find fossils in these rocks." When he returned 
 in the evening, however, I showed him several 
 well defined corals, fossils of a Devonian aspect, 
 imbedded in pieces of hydro-mica schist. For a 
 moment Richardson seemed at a loss, but at last 
 said : " Well, Weston, you ought to be knighted 
 for this find." It may seem egotistical for me to 
 mention this, but I do so to show the great im- 
 portance Richardson, like Logan, attached to 
 fossils. 
 
 On August 3rd, I was again in the United 
 States, this time at Troy, N.Y., to which place I 
 had been sent to ascertain whether the conglom- 
 erates of Troy belonged to the same horizon as 
 the Bic rocks of the St. Lawrence. 
 
 I was assisted in this work by Mr. S. H. Ford 
 of Troy (already mentioned), who was then 
 studying the fossil fauna of that vicinity. We 
 were very successful in obtaining fossils iden- 
 tical with those of Bic. Poor Ford was then 
 just gaining fame as a palaeontologist, but in a 
 short time after I saw him last, his brain gave out, 
 and his relations had to place him in an asylum. 
 
 On the 22nd of August, a few days after my 
 
 return from Troy, N.Y., I left Montreal again for 
 
 Bic ; this time to accompany Mr. Billings, who 
 
 was then much interested in the fossil fauna of 
 
 72 
 
AMONG THE ROCKS, 
 
 the Lower Potsdam formation. We visited several 
 of the most interesting localities between Riviere 
 du Loup and Bic. Mr. Billings was not well at 
 the time, and as at Whitehall often became 
 nervous and would not leave the house, preferring 
 to remain indoors and study the fossils already 
 collected. After a very interesting journey we 
 returned to Quebec and spent several days in 
 exploring the rocks of the Chaudi^re Falls. 
 Billings was then studying the small bivalve 
 fossil Obolella, and it was here we obtained much 
 information relating to this genus. 
 
 I have mentioned before, that at this time we 
 travelled between Riviere du Loup and Bic chief- 
 ly at night, with the mails which were generally 
 conveyed from station to station by a buckboard 
 and one horse. 
 
 Stopping in the night to wake up a sleepy 
 post-office keeper, changing horses and waiting 
 in the cold while the mail bags were gone through 
 etc., rendered these journeys anything but pleas- 
 ant. On one occasion a fat French woman got 
 into our stage. It was midnight, and her head 
 soon began to bob from one side to the other, 
 and at last it came down with a whack on my 
 shoulder. For some time that head rested 
 there. But ah ! those days have long passed, 
 and one can now stretch in a luxurious sleeper, 
 and arrive at his destination refreshed and ready 
 for work. 
 
 73 
 
KEMINISCENCES 
 
 The next official journey was to Stanfold and 
 other localities in the Province of Quebec. Sir 
 William Logan still continued intent on his 
 Eastern Township rocks, and it was at his sug- 
 gestion that the Director of the Survey sent me 
 to these localities. The winter of this year 1871, 
 was a busy one for all hands in the Survey A 
 perusal of the report of the Geological Survey 
 for 1872-73 will show how varied my occupations 
 were when I was not engaged in field work. 
 
 My field work, in 1872, commenced May 29th, 
 when I again started out to continue geological 
 investigations in the Eastern Townships. Here- 
 ford, Farnham and other localities were visited 
 and additional information regarding the "Quebec 
 Group" obtained. 
 
 I had been working at Farnham Centre two 
 or three days, most of the time in one spot where 
 I had found a very interesting set of fossils. 
 During the day I had noticed people watching 
 me, but keeping a safe distance away. When I 
 returned to the boarding-house I heard the folks 
 talking of a crazy man who had arrived there 
 and was digging deep holes and breaking rocks 
 in a field where he expected to find gold. I took 
 no notice, but after supper was surprised to hear 
 that a deputation consisting of the chief men of 
 the village was waiting outside to interview me. 
 I went out and found several men, whose ques- 
 tions led me to think they took me to be a "little 
 74 
 
AMONG THE ROCKS. 
 
 off." But when I told them it was not gold I 
 was after, but fossils, they seemed more convinced 
 than ever that I was not "all there," and deter- 
 mined to visit me at my work the following day, 
 but bright and early next morning I was away 
 to fresh fields. Whether these men ever learned 
 what o fossil is I do not know. 
 
 The fossil fauna of the " Bic Conglomerates " 
 had proved so interesting that Mr. Billings re- 
 quested the Director of the Survey to send me to 
 the straits of Belle Isle where the Lower 
 Potsdam Group is in an undisturbed condition. 
 Sir William Logan who still took a keen interest 
 in the working of the Survey, thought valuable 
 information might be obtained there, relating to 
 the " Quebec group." So on the 29th of May, I 
 received instructions to prepare for a journey to 
 the coast of Labrador. As assistant, I decided to 
 take Mr. Willimott, a young Englishman who 
 had recently been engaged to assist me in museum 
 work. A few additions to my Anticosti camp 
 equipment were made, and two months' pro- 
 visions, packed, and we were ready to start. 
 
 We reached Quebec the following morning 
 and boarded the lighthouse supply S.S. Napo- 
 leon III., whose, captain had received government 
 instructions to land me and my assistant at 
 L'Anse au Loup, or some convenient point on 
 the coast of Labrador. 
 
 During the landing of supplies I had an oppor- 
 75 
 
EEMINISCENCES 
 
 tunity of again spending a short time on the 
 island of Anticosti near the lighthouses, and also 
 at two of the lighthouses at Newfoundland. 
 While near the straits of Belle Isle we encoun- 
 tered severe storms, and were in close proximity 
 to several icebergs ; fogs detained us for a day 
 or more. However, we landed safely and pitched 
 our tent at L'Anse au Loup, where we collected 
 a large number of beautiful specimens of Archeo- 
 cyathus, the first specimens of which were col- 
 lected by Mr. J. Richardson before 1863. All 
 Mr. Richardson's collection, except the one fig- 
 ured in the Geology of Canada for the latter 
 date, were lost at sea together with the schooner 
 on which they were shipped. The specimen 
 referred to, he fortunately had in his pocket. 
 
 While the rocks of the island of Anticosti 
 contain an almost endless variety of organic 
 remains, the rocks of the coast of Labrador 
 possess a very small fauna of fossils records 
 of past ages, when the scanty life of this planet 
 consisted of a few simple seaweeds, a few sponges, 
 a few different species of bivalves, and a few 
 varieties of trilobites animals belonging to the 
 Crustacean family. No well defined plants ex- 
 isted, and no vertebrate animal had yet appeared 
 on the face of the earth. Still the same sun 
 which shines now shone then, and rain fell in 
 those days as it does now. This is known by 
 the mud cracks we find on some of the argilla- 
 76 
 
AMONG THE ROCKS. 
 
 ceous, or clay rocks, as one sees at the present 
 day on the clay of a dried-up mud-puddle, and 
 as we often see on clay, pits made by rain drops, 
 so we find them in the rocks we are speaking of. 
 
 While the island of Anticosti abounds in trees, 
 some of a large growth, and abundant vegetation 
 of endless variety mile after mile of the Labra- 
 dor coast is covered only by lichen and scrubby 
 brush, the roots of which we used for our camp 
 fire. Returning to our camp one day we were 
 fortunate enough to find, a mile or so away, a lot 
 of small pieces of sawn pine which may have 
 drifted hundreds of miles. This we gathered 
 up carefully, carried to camp, and used for 
 kindling wood. 
 
 It would take pages to narrate all the inci- 
 dents of this trip one would like to record ; but 
 I must bring this journey to a close. Towards 
 the end of July a small steamer, "The Beaver," 
 which was built in Scotland, and was sent out 
 here for the fishery business, put into L'Anse au 
 Loup for coal. I interviewed the captain and 
 requested him to take us to Quebec, as I did not 
 wish to run the risk of remaining on that dreary 
 coast, probably till the fall. After much wrang- 
 ling about the matter he consented to take us, 
 providing I paid him 10, used our " own grub," 
 and asked no questions. This I consented to do, 
 and in a short time we were on our way up the 
 St. Lawrence River. The captain seemed to 
 77 
 
REMINISCENCES 
 
 have been drinking while on shore, and his first 
 mate acted like a typical nigger driver. Such a 
 journey as we had I never wish to experience 
 again, but after a most dangerous passage we 
 reached Quebec, and we left that ship as quickly 
 as possible, and arrived the following day at the 
 Survey with a fine, but hard-earned, collection 
 of fossils. 
 
 During the month of August I spent ten days 
 at Paquette's Rapids on the Ottawa River, 
 making a collection of Black River fossils, and 
 on the 1st of October I again went to Arisaig, 
 N.S., and returned late in the same month with 
 new information and another beautiful lot of 
 specimens. 
 
 78 
 
CHAPTER VI. 
 
 LORD DUFFERIN VISITS THE SURVEY I ASSIST SIR WILLIAM 
 LOGAN IN THE EASTERN TOWNSHIPS ANOTHER VISIT TO 
 ARISAIG SOME OF THE PIONEER PHOTOGRAPHS OF THE 
 SURVEY A STORM OFF CAPE GASPE" A VISIT TO NEW- 
 FOUNDLAND LAKE TEMISCOUATA WORK AT HESPELER, 
 ONT. THE CONGLOMERATES OF BIC DEATH OF SIR 
 WILLIAM LOGAN THE PHILADELPHIA EXHIBITION THE 
 EASTERN TOWNSHIPS AGAIN DISTRIBUTION OF ROCKS 
 AND MINERALS ETHNOLOGICAL BRANCH OF THE SUR- 
 VEY. 
 
 INURING the latter part of the year 1872, and 
 "-^ the early part of 1873, several distinguish- 
 ed government officials visited the Geological 
 Survey, among whom was Lord Dufferin, who 
 had recently been appointed Governor-General 
 of Canada. On his first visit to our museum 
 Sir William Logan requested me to accompany 
 His Excellency, himself and Principal Dawson, 
 through the museum, that I might answer any 
 question relating to my work. Sir William then 
 occupied a small room on the palgeontological 
 flat as sitting, and sometimes sleeping room. The 
 wall on one side of this room was covered wholly 
 with large framed slabs of Potsdam Sandstone 
 on which are well defined Crustacean (?) tracks 
 Climactichnites, so named by Logan because they 
 resemble a rope ladder. While Sir William was 
 pointing out the beauty of this specimen, Dufferin 
 79 
 
REMINISCENCES 
 
 said : ' That reminds me, when I was at 
 the skating rink the other day I saw some pecu- 
 liar tracks on the ice. I watched the skaters for 
 some time to find out what made those tracks, 
 and at last, would you believe it, I discovered 
 they were tracks made by a lady's petticoats." 
 
 By June 12th, I was again in the field assist- 
 ing Sir William Logan with his Eastern Township 
 work. 
 
 Sir William's last days of geological explora- 
 tion were fast approaching. Some of his strati- 
 graphical work in the Eastern Townships had 
 been disputed by Dr. Hunt, and feeling convinced 
 that he was correct, Sir William labored strenu- 
 ously to prove this. It was my privilege to be 
 allowed to assist him. We labored together at 
 Danville, P.Q. One day Sir William took me to 
 some plumbaginous limestones and other rocks on 
 the Nicolet River. " These," he said, " are the 
 rocks Hunt wants to make Huronian. Now if 
 we can only find fossils here, it will not only 
 be a feather in your cap, but it will save my 
 reputation as far as these rocks are concerned." 
 
 We had labored all day, going over the expos- 
 ures, sometimes on our hands and knees closely 
 scrutinizing them with our lenses hoping to find 
 some weathered -out fossil. In the evening we 
 were returning to our hotel without a fossil, 
 when I saw a thin band of limestone, less altered 
 than the rest, between the plumbaginous beds. 
 80 
 
AMONG THE ROCKS. 
 
 " Stop," I said, " let us try these rocks." I broke 
 off a piece of the limestone, looked at it carefully 
 with my lens, then looked up at Sir William and 
 said, " I will bet $10 this limestone is full of 
 fossils." Sir William took the piece of rock, 
 looked at it with his lens, shook his head and 
 said, " I don't see the ghost of a fossil." " Neither 
 do I," said I, " but we will see what my micro- 
 scope says about it." Sir William took the 
 specimen, wrapped it in his handkerchief, and 
 said : " I am going in to-morrow to Cttrtier's 
 funeral, and will put this on your table in your 
 work-room." I went back to the Survey some 
 days later, found the specimen on my table, went 
 to work at it immediately, and in half an hour 
 or so ran down to Sir William and said : " I was 
 right ; that Nicolet limestone is made up of 
 fragments of fossils, and they look so much like 
 Trenton or Chazy forms, that I can't tell the 
 difference." If ever I saw our dear old chief's 
 face radiant with joy, it was then. This settled 
 a question of grave importance to Sir William 
 Soon we were again on the Nicolet River. 
 This time I took my camera and photo chemi- 
 cals there were no dry plates then to take 
 several views of the rocks. Two of them were 
 reproduced for Dr. Ells' report " Geology of 
 Canada for 1886." 
 
 Sir William remained several days with me on 
 the Nicolet River and took a deep interest in the 
 F 81 
 
REMINISCENCES 
 
 work. Once when he was some fifty yards away 
 from me, I shouted " another fossil, Sir William," 
 To reach me quickly he waded through water 
 which ran over his boot-tops. Soon after this 
 Sir William made me a present of a microscope, 
 whether in recognition of my work on the Nico- 
 let or not I don't know. This microscope was 
 purchased from Mr. J. F. Whiteaves, of the 
 Natural History Museum, Montreal. 
 
 I have dwelt a little long on these incidents of 
 the Nicolet River because I wish to record here 
 the great importance Sir William attached to 
 fossils, and because these were almost the last 
 days of his field work. 
 
 On the 20th of July, I again left for Arisaig, 
 Nova Scotia. A nervous feeling came over me 
 when I started, which increased all the way, and 
 when I arrived at Arisaig, after a thirty mile 
 buggy drive from New Glasgow, I was in a high 
 fever. My old friends, the McDonalds, came out 
 to meet me with a welcome always extended to 
 a visitor to that rough sea coast. I told them I 
 was sick and would return when the horse was 
 rested. The old lady hustled me off to bed in 
 quick time, took my purse, and when the horse 
 and man were rested and fed, paid and started 
 them back. They ^ave me some of their simple 
 medicines, for there was no doctor within twenty 
 miles. I was delirious for a short time, but in 
 three days was up and at work again. While I 
 82 
 
AMONG THE ROCKS. 
 
 Jay in bed Mrs. McDonald came into my room and 
 smoked her black pipe, and her daughters would 
 sit around in their bare feet, " keeping me com- 
 pany " poor simple country girls. I often think 
 of those days when these kind folks walked miles 
 to get me better food than they could afford to 
 eat. The old folks are still alive, but have long 
 passed the allotted three score and ten. 
 
 I had taken my camera and chemicals with 
 me, and my first work was to take a series of 
 twenty-one photographs of the coast rocks. In 
 this work I was assisted by young Archie Mc- 
 Donald. Some of these views have been repro- 
 duced to accompany Mr. Hugh Fletcher's report 
 of the geology of this part of Nova Scotia (Geo- 
 logical Survey of Canada for 1882-1886.) A 
 section of the rocks at Joseph McDonald's cove, 
 by the writer accompanies the photo of these 
 exposures. I feel proud of these photographs, as 
 I may say they, and my Nicolet River views were 
 almost the pioneer photographs of the Geological 
 Survey of Canada. Years after when the dry 
 plate process was instituted field photographs 
 became so simple that few of the explorers of the 
 Survey went out without a camera and plates. 
 The collection of views taken in the field by the 
 various members of the Survey numbers many 
 hundreds and is of historical interest. Those 
 of Dr. George Dawson, Bell, Tyrrell, McConnell, 
 Dowling and myself give one an idea of the 
 83 
 
REMINISCENCES 
 
 wonderful rocks and charming scenery of the 
 great North-west Territories. Those of J. B. 
 Tyrrell show us the immense herds of reindeer 
 which roam over the plains of the great barren 
 lands; those of Low the dreary lands of Labrador. 
 
 A great number of the fossils collected from 
 the coast rocks of Arisaig have been described 
 and figured by the late E. Billings, "Geology of 
 Canada, Palaeozic fossils 1874." 
 
 My last official trip of this year, 1873, was to 
 Gaspe where I went to examine and collect from 
 various fossiliferous beds, known as the Gaspe* 
 limestone, or Lower Helderberg group. The 
 thickness of rocks exposed at Gaspe* is said to be 
 9000 feet, the greatest part of which are lime- 
 stones holding a large and varied fauna of fossils. 
 Most of niy stay here was spent a short distance 
 from Cape Gaspe, where I was fortunate enough 
 to get lodging in a fisherman's house close to the 
 shore, Fish formed the principal food of these 
 toilers of the sea. Fish for breakfast, fish for 
 dinner, fish for supper ; but the invigorating sea 
 breezes made me hungry enough to eat fried shark. 
 
 During my stay here a terrific storm swept 
 twenty fishing boats from their moorings. It was 
 a grand sight to see the mighty waves lashing 
 the great cliffs. I had been invited to take a day 
 off from work and accompany a picnic party to 
 Gasp^ Basin. Soon after we arrived at our camp- 
 ing grounds, the great storm came on. The only 
 84 
 
AMONG THE ROCKS. 
 
 shelter we could get was in a log cabin which 
 was soon filled with the women and girls of our 
 party. We men occupied the barn and slept in 
 the hay loft, where we spent a most unpleasant 
 night. The following afternoon the sea had 
 calmed down a little, and we started for Cape 
 Gaspe, but soon five or six of the womenfolk were 
 lying at the bottom of the boat. The sea was 
 very rough but we had a good boat and men who 
 had spent the greatest part of their lives boating. 
 At length we reached the wharf, and found all 
 the folk of that vicinity ready to render any 
 assistance in landing us, which was no easy task, 
 for as a large wave carried our boat alongside 
 the wharf only one person could be hoisted ashore. 
 I was grabbed by brawny hands and hoisted up 
 with as little ceremony as though I were a coil 
 of rope. At last we were all landed and many a 
 fervent thanksgiving was uttered for our safe 
 delivery from the perils of the deep. I returned 
 to Montreal on the 13th of September, after an . 
 instructive and profitable journey. 
 
 The winter of 1873-74, was occupied as usual 
 in museum and room work, preparing a num- 
 ber of microscopic sections of fossil woods to be 
 reported on by Dr. Dawson, developing and pre- 
 paring fossils for the museum and for study. 
 
 Sir William Logan was ever ready to assist his 
 colleague of the pioneer days of the Canadian 
 Survey, Mr. Alexander Murray, then director of 
 85 
 
REMINISCENCES 
 
 the Geological Survey of Newfoundland, and 
 when in the spring of 1874, this gentleman wrote 
 to his old chief Logan, saying he was doubtful 
 about the horizon of some of his rocks, and 
 desired him, if possible, to send Weston to try 
 and discover fossils in these rocks, which if 
 obtained would settle the, to him, important ques- 
 tion, I was requested by Sir William who had 
 recently retired from the directorship of the Sur- 
 vey, but still continued to act as adviser, to ask 
 for two months' leave of absence that I might 
 proceed to Newfoundland. The director of the 
 Canadian Survey, Mr. A. R. C. Selwyn, acting on 
 Logan's advice readily granted the leave asked 
 for, and I left by the next English mail steamer, 
 which called at St. John's. Our ship arrived 
 safely, Murray met me at the wharf, and I was 
 soon housed in the best hotel in the city. Two 
 days after, Murray came with a trap and an Indian 
 named John, who was to act as ^uide and general 
 assistant. We drove off through a country, which 
 according to Murray, " was made of the chippings 
 of the world," and at last arrived at Manuel's 
 River, twelve miles from the city. 
 
 The following extracts from Transactions of 
 the Nova Scotian Institution of Science Vol. IX., 
 Season 1895-96, will explain more fully the impor- 
 tance of fossils in geological investigations : 
 
 86 
 
AMONG THE ROCKS. 
 
 THE PRIMORDIAL SILURIAN OF NEWFOUNDLAND AND 
 CANADA. 
 
 " In spite of the oft-repeated assertion of Professor 
 Jukes and the late director of the Geological Survey of 
 Canada, ' If the fossils don't agree with the stratigraphy ; 
 so much the worse for the fossils,' my long experience as 
 a collector of fossils and close observer of the varions geo- 
 logic horizons leads me to think that if the stratigraphy 
 does not agree with the fossils ; so much the worse for 
 the stratigraphy. To illustrate the faith the late director 
 of the Newfoundland Survey had in palseontological evi- 
 dence, I will relate one incident out of many similar ones 
 known to the writer : In the summer of 1874 Murray 
 wrote to Sir W. E. Logan, then adviser to the Canadian 
 Survey, saying : * I have made my Manuel's River rocks 
 Primordial ; I am doubtful, however, whether my strati- 
 graphy is correct ; neither Howley nor I have been able 
 to find the ghost of a fossil ; could you arrange in any way 
 to send Weston down for a few weeks.' The result was 
 that I left by the next steamer which called at Newfound- 
 land, and a few days after my arrival at St. John's was 
 taken by Murray to Manuel's River where he got lodgings 
 for myself and Indian guide. The following day I com- 
 menced my search for fossils, and in a short time was 
 rewarded by finding, in the gray argilites, the well-known 
 Crustacean. Microdisciis Ikiwsoni, (Hartt) ; which occurs 
 in abundance in the Primordial slates of St. John, at. 
 Ratcliffe's Mill Stream, and other localities in New Bruns- 
 wick. This Crustacean, Microdiscns, is a puny thing, not 
 larger than the half of a small pea, but it told me a big 
 tale about the geological horizon told me that Murray's 
 stratigraphy was correct, and that I stood on Primordial 
 strata similar to those of St. John, New Brunswick." 
 
 I may mention here that the term Primordial, 
 used by Barrande and the late paleontologist of 
 the Canadian Survey, Mr. E. Billings, is seldom 
 used now ; St. John Group being thought a better 
 name for that extensive group of rocks. This 
 87 
 
REMINISCENCES 
 
 Cambrian division of the lower Silurian of New- 
 foundland, according to Murray, would, if found 
 consecutive at any one locality, represent a thick- 
 ness of 6,000 feet of black, grey and other 
 colored argillites, micaceous calcareous slates and 
 limestones, sandstones, conglomerates and other 
 rocks, some of which are prolific in fossils, 
 especially the iron-stained argillites of Manuel's 
 River and other localities in Conception Bay. 
 The fauna is similar to that of the Primordial of 
 St. John, Ratcliff's Mill Stream and other locali- 
 ties in New Brunswick. 
 
 Mr. Murray procured lodging and board for us 
 at a comfortable farm house a short distance 
 from Manuel's River, a mile or two from Con- 
 ception Bay. Before leaving us Mr. Murray said, 
 " Don't let Joe get even a smell of fire-water," so 
 when we were alone I said, " Joe, have you any 
 fire-water?" 
 
 " No, master," he replied. 
 
 I decided before starting to work next day I 
 would look round myself. Leaving Joe at the 
 house, I remained away about three hours ; when 
 I returned I found Joe so drunk that he did not 
 know who he was or where he came from. I 
 searched his bag and found one full flask of 
 whiskey and an empty bottle. These I took out- 
 side and smashed on the rocks. The old lady 
 got poor Joe to bed, and the following morning 
 my brave Indian was all right, and when I told 
 88 
 
AMONG THE ROCKS. 
 
 him what I had done with his fire-water he said, 
 " much better." 
 
 After breakfast, with some lunch in our 
 pockets, we started off for the day. Murray and 
 his assistant had searched the rocks of Manuel's 
 River for fossils but failed to discover any. I 
 was more fortunate, for the first day, not long 
 after we started down the river, I found, in an 
 exposure of clay-slate, one solitary specimen of 
 that little fossil Crustacean, Microdiscus, already 
 alluded to. Imagine my delight when I saw this 
 wee fossil which, being typical of the Primordial 
 fauna of St. John, New Brunswick, proved that 
 Murray's stratigraphy was correct, and my jour- 
 ney to Newfoundland had been successful. But 
 there was more luck in store. Some distance 
 down the river 1 saw a small island with rocks 
 in the middle of the stream. Joe found we could 
 reach this by wading and ordered me to get on 
 his back, which I did, and was soon standing on 
 an exposure of iron-stained clay-slates crowded 
 with fossils Paradoxides and other typical 
 Primordial forms. We collected specimens 
 enough to fill several cases in Mr. Murray's 
 museum. Twc days after this was Sunday, and 
 knowing how delighted Murray would be to hear 
 of my discovery, I decided to start to St. John's 
 after breakfast, but judge of my surprise when 
 the folks of the house would not lend me a horse. 
 "No," said the old lady, "Ye shall no brak the 
 89 
 
REMINISCENCES 
 
 Sabbath with one of my horses." But go to the 
 city I would. Joe was quite willing to accompany 
 with me, so, with some lunch in our pockets, we 
 started off on our twelve miles' walk. The day 
 was very hot, and when I arrived at Murray's 
 house he was astonished to find we had walked in. 
 
 Murray was a kind-hearted man, ever ready 
 to assist those in distress, but was noted for fits 
 of temper, when he would use the strongest of 
 " swear- words," and on this occasion he did use 
 swear- words, and fairly boiled over with wrath 
 against the old woman of our boarding house. 
 But in a moment his anger was over, and, with 
 a serious face, he looked up and said, " But what 
 about the rocks ? " 
 
 When I told him the Manuel's River rocks 
 were crowded with Primordial fossils he was 
 astonished and delighted, but wondered how he 
 and that Howley had missed them. 
 
 After returning from Manuel's River, Murray 
 and I made several trips to various parts of Con- 
 ception Bay, and obtained many interesting new 
 fossils. I afterwards spent a short time on Bell 
 Island, sleeping one night over the one living 
 room of the house, where fowl, and a litter of 
 pigs formed part of the family group. 
 
 Having completed the field work assigned me, 
 
 I spent the remainder of the time till my steamer 
 
 arrived in putting some of the cases in the small 
 
 museum which consisted of several rooms in 
 
 90 
 
AMONG THE ROCKS. 
 
 Mr. Murray's house in order, and also in classi- 
 fying the books in his library, during which 
 work I found four Church of England prayer 
 books. When I asked Murray how I would 
 class them, he was very angry that any one 
 should have "put their bibles in his library." 
 He flung them across the room and said some 
 of his choice swear-words. The next morning I 
 went to the cathedral church, Mr. Murray 
 was there deeply engaged in the devotions of 
 the day. 
 
 Before leaving for home Mr. Murray, after 
 paying all my expenses, insisted on presenting 
 me with 10, and on my return to the office Sir 
 William said " I have heard of your success, and 
 wish you to accept $50.00 from me." This 
 ended one of the most delightful trips I have 
 had since my connection with the Survey. 
 
 A short time after returning from Newfound- 
 land I was instructed to proceed to Lake Temis- 
 couata and if possible obtain fossils from the 
 conglomerates of that vicinity. Mr. A. H. Foord, 
 who had lately been appointed artist to the Sur- 
 vey, accompanied, me. It was his first geologi- 
 cal excursion ; the country was new to him and 
 the scenery delighted his artistic eye. We got 
 board and lodging at a French Canadian's house 
 close to the lake, but poor Foord, who had been 
 brought up in the most tender and delicate way 
 by his English parents suffered very much 
 91 
 
REMINISCENCES 
 
 with bugs. Four times during the night he 
 came to my room to report the doings of these 
 pests. In the morning I found him asleep on 
 the floor tightly coiled up in a sheet. 
 
 The rocks we wanted to see were on the other 
 side of the lake, and the only boat 1 could get 
 was a dug-out canoe, but I secured an experienced 
 half-breed to take us over. The lake was a 
 little rough, but we landed safely on the opposite 
 shore. The wind however, continued to blow, 
 and by the time the man was to return for us 
 the water was too rough for his frail canoe. 
 Darkness came on and relieved us from the ten 
 thousand mosquitoes which my companion found 
 even worse than the bugs, for this was his first 
 experience of our Canadian mosquito. As bad 
 luck would have it I had forgotten both pipe 
 and matches things Foord never carried so 
 we were unable to make a fire. I lay down 
 under a tree and got a snooze, but Foord walked 
 the beach all night, and at six o'clock the next 
 morning, the water having calmed down, our 
 man came for us. We were hungry, but none 
 the worse for our night out on the lake shore. 
 Foord refused to cross the water again so I left 
 him to make sketches while I continued to 
 explore the rocks of the lake. We returned 
 home in about ten days, and after a short time I 
 was again working at the Guelph formation at 
 Hespeler, of which I have already spoken. Mr. 
 92 
 
AMONG THE ROCKS. 
 
 Billings was then much interested in the fossil 
 bivalve Trimerella, figures of which will be 
 found in " Palaeozoic Fossil Vol. 1. 1861-65." It 
 was for the purpose of getting further informa- 
 tion of this genus, together with other new fos- 
 sils, I made these journeys to Hespeler, Gait, 
 Guelph, Elora and other places. 
 
 While at Hespeler, on the 6th of September, I 
 was called home to accompany Dr. Dawson, 
 Principal of McGill University, to the now world- 
 wide known Eozoon locality at Cote St. Pierre 
 near the Petite Nation River, province of Quebec ; 
 the home of our supposed Laurentian fossil. 
 
 In the early pages of these memoirs I have 
 spoken of this supposed organism of which page 
 after page might be written ; but I have neither 
 space nor inclination to continue this subject, 
 and shall only add here my last, published, re- 
 marks on Eozoon, " Transactions of the Nova 
 Scotian Institute of Science, Vol. IX, season 
 1895-96." 
 
 "The Laurentian rocks of Newfoundland are 
 similar to those of Canada, consisting of gneiss, 
 granite, syenite, limestone, quartzite, mica schist, 
 etc., all of which are frequently cut by granite 
 and other dykes. They form a large portion of 
 the island which, as Mr. Murray remarks, ' has 
 materially contributed to produce the remarkable 
 geological and topographical features which it 
 presents/ Probably it was the chopped-up 
 93 
 
REMINISCENCES 
 
 appearance of the Laurentian and Huronian 
 formation which caused him to remark that 
 ' Newfoundland was formed of the chippings of 
 the world.' The Laurentian of Newfoundland, 
 so far as we know, is totally destitute of the 
 remains of either vegetable or animal structure, 
 and therefore must still be considered Azoic, 
 although this term has been abandoned by some 
 geologists in the nomenclature of Canadian rocks, 
 owing to the discovery in the Upper Laurentian 
 of certain forms which resemble Stromatoceriuin 
 rugosa, one of the Protozoa of the Silurian. 
 This peculiar mineral aggregate (?), received from 
 Sir W. E. Logan and J. W. Dawson the name 
 Eozoon Canadense. Literature enough to fill a 
 cart has been published for and against this sup- 
 posed organism, among which Dawson's " Dawn 
 of Life" is the most interesting. With the excep- 
 tion of Sir J. W. Dawson, probably no one has 
 done more work at this supposed fossil than the 
 writer, who has prepared hundreds of microsco- 
 pic sections, micro -photographs, micro- drawings, 
 illustrative collections for the Paris, London, 
 Philadelphia and late Chicago Expositions, and 
 for other public and private collections ; still he 
 could never make up his mind that Eozoon C ana- 
 dense is of organic origin. Mr Billings late 
 palaeontologist to the Geological Survey of 
 Canada, pronounced strongly against the organic 
 character of Eozoon. I have frequently con- 
 94 
 
AMONG THE HOCKS 
 
 versed with Dr. Selwyn, Dr. Ami, Dr. Ells, the 
 late Mr. Vennor (who obtained the Tudor speci- 
 mens), and other members of the Canadian 
 Survey, but none of these gentlemen ever admit- 
 ted that Eozoon is a fossil. However Eozoon 
 will always remain an interesting subject for 
 students in palaeontology and mineralogy." 
 
 During part of the month of August I was 
 again among the Bic Conglomerates trying to 
 increase our already large fauna of fossils from 
 the limestones of the coast rocks. 
 
 While at St. Simon, about twenty miles above 
 Bic, I had walked many miles and worked hard 
 to collect numerous small bivalve fossils, the 
 muscular impressions of which were interesting 
 various palaeontologists. I had seen an Indian 
 watching my movements during the day, but 
 took no notice other than to wonder why he did 
 not come close to where I was working. After 
 supper my specimens, which I was very proud 
 of, were packed and given into the hands of the 
 railway station agent. While at breakfast the 
 next morning to my surprise the stationmaster 
 came in to tell me that my two boxes of speci- 
 mens had been stolen during the night. I 
 never saw my precious specimens again. May 
 that wretched Indian, for I suppose it was he 
 who stole them, be punished when he gets to his 
 happy (?) hunting grounds ! 
 
 Most of the summer of 1875, was spent col- 
 95 
 
REMINISCENCES 
 
 lectmg economic minerals for the Philadelphia 
 Exhibition of the following year. My field of 
 operation extended from Sarnia, Ontario, to 
 Beauce, Quebec. With the small sum of $400.00 
 28 boxes of specimens were obtained, consisting 
 of bricks, tiles, pottery, building and flagging 
 stones, ochres, brines, salt, marbles, &c. I may 
 mention here that through the kindness of the 
 directors of the Grand Trunk Railway I received 
 for about eight years, a pass which took me from 
 " station to station." Thus the Survey was saved 
 expending a large amount of money for travel- 
 ling expenses. 
 
 Before concluding my remarks for this year 
 1875 it is my sad duty to record the death of my 
 dear old chief Sir William Logan. Since his 
 retirement he had spent a good portion of the 
 three succeeding summers on his Eastern Town- 
 ship rocks. I have already stated that during 
 the summer of 1873 Sir William was working in 
 the Danville district and that I spent part of 
 June with him on the Nicolet River and in other 
 localities in that vicinity. This I believe was 
 the last field geology he ever did. In 1874, he 
 crossed the Atlantic to visit his friends and 
 relatives in Wales, where he gradually grew 
 weaker, till on the 22nd of June he died at Castle 
 Malgwyn Llechryd, South Wales. " And now 
 he sleeps in the quiet churchyard of Llechryd 
 between his brother Hart and his brother-in-law 
 96 
 
AMONG THE ROCKS. 
 
 Mr. A.L. Gower; "Peace to his memory ! Honour 
 to his name !" 
 
 The memory of Sir William will always be 
 dear to those who have shared in his geological 
 travels. The interesting incidents one might 
 relate of his simple mode of life and disregard 
 for anything other than the work he was at, 
 would fill many pages, but I must content my- 
 self here by relating only one. 
 
 During the summer of 1864, I think it was, I 
 spent a few days with Sir William at Point Levis 
 and St. Joseph de Levis, opposite Quebec. Sir 
 William was engaged tracing out some of the 
 bands of limestone, a map of which he had 
 already published in 1862, while I collected 
 fossils from the conglomerates. For several days 
 we had plodded over muddy roads, through bush 
 and over fields ; Sir William looked unusually 
 seedy, and his broad-brimmed weather-beaten felt 
 hat, an old coat which had seen several seasons' 
 sunshine and rain ; a pair of well worn pants, 
 tucked in a pair of muddy boots, and a week's 
 growth of beard, did not improve his appearance. 
 
 It was Saturday afternoon and Sir William 
 proposed that we should go over to the island of 
 Orleans and spend Sunday there. "All right" 
 I said, " I will go to the hotel and settle up and 
 you can meet me at St. Joseph's ferry." In a 
 short time we landed at the west-end of Orleans 
 Island, a short distance from a small hostelry 
 
 G 9A 
 
REMINISCENCES 
 
 kept by T. H. Lizotte. Sir William then pro- 
 posed that I should interview Lizotte as to dinner 
 and rooms for the night. I did so, and ordered 
 dinner for two. We remained outside looking at 
 the rocks, and in due time Lizotte came out, 
 looked for a moment at Sir William, then beck- 
 oned me aside and said : "Dinner is ready, but 
 do you wish your old man to dine with you, or 
 shall I give him his dinner in the kitchen?" It 
 is needless to say Sir William dined with me. 
 Had I been an artist like Sir William I certainly 
 should have portrayed the old gentleman as after 
 dinner he sat by the fire reading a novel, now 
 leaving off to wipe the tears from his eyes, and 
 again to stroke a cat which sat perched on his 
 shoulder. 
 
 Lizotte is still living. Last summer 1896, I 
 met him in Qnebec,and as is usual with him, when- 
 ever he greets me, he asked: "How is business?" 
 I answered " I am out of business now, and Her 
 Majesty has consented to allow me bread and 
 butter as long as I live ; but no jam." " No 
 jam," said Lizotte, " Ah that is too bad too bad." 
 
 Till the next summer my time was fully 
 occupied in arranging specimens for the forth- 
 coming exhibition. The beautiful exhibits of 
 thesupposed Laurentian fossil Eozoon Canadense, 
 which attracted so much attention from geologists 
 and palaeontologists from all parts of the world, 
 prepared in my spare hours. 
 98 
 
AMONG THE ROCKS. 
 
 My field work of 1876 commenced in June, at 
 Melbourn. The Eastern Township rocks were 
 occupying a great deal of attention and no effort 
 on my part was spared that would throw some 
 light on the geological horizon of these much 
 disputed altered rocks, a thing which seemed 
 almost impossible without the aid of fossils. I 
 have referred to the importance of fossils while 
 speaking of Logan and my work on the Nicolet 
 River. It was a continuation of this work I was 
 at when on the 13th of June, I learned by tele- 
 graph that Mr. Billings was dead. I haste^SShi 
 to Montreal to attend his funeral, ^^r long 
 previous to his death at the requfl^rof Mrs. 
 Billings I left my work in th^Jtffonships and 
 came in purposely to take Mfi^fco the country for 
 a short time, hoping th^&fhange would do him 
 good. There was ^dApmfortable summer hotel 
 near the lake a^Woeil Mountain, P. Q., which I 
 chose as a quietr retreat, but we had only been 
 there three or four days when he became nervous, 
 and so restless that I was obliged to take him 
 home again. This was the last time he left his 
 house. It was with deep sorrow that I saw the 
 last of one with whom I had been associated so 
 intimately for over fifteen years. 
 
 Mr. Billings' funeral over, I returned to Mel- 
 bourn on the St. Francis River, and as usual 
 boarded with our old friend Mr. Gee. Poor Gee, 
 I often wonder why Logan did not speak more 
 99 
 
REMINISCENCES 
 
 frequently of this singular man, at whose house 
 he preferred to stay, even if working half a 
 dozen miles away. Gee prided himself on his 
 " geological knowledge," and in his little store 
 and grog-shop, kept numerous specimens of rocks 
 and minerals, among which was his celebrated 
 " fossil duck " which was a piece of water-worn 
 limestone. 
 
 Our " grub " was none too good, but on special 
 occasions we got, by way of a treat, an English 
 plum-pudding. It was after being out all day, 
 with Logan (in 1873, I think) who never ate 
 between breakfast and his evening dinner that 
 I returned very hungry, having forgotten k) take 
 some lunch, and ate very heartily of plum-pud- 
 ding, the result of which was a severe attack of 
 night-mare from which I was wakened by Sir 
 William rushing into my room shouting " wake 
 up Weston you have been eating too much pud- 
 ding that comes of Gee feeding us too well." 
 
 Gee set apart one of the rooms of his house, 
 which he called the " Geologic room." It was in 
 this room Sir William often sat till the small 
 hours in the morning, protracting his work of 
 the day. The last night I spent with him in 
 this room, at twelve o'clock, he was dozing over 
 his papers, but at my request left them and went 
 to bed ; a few days after this he returned to 
 Montreal never to work again at his beloved 
 township rocks. 
 
 100 
 
AMONG THE ROCKS. 
 
 I had only been at Melbourn a day or two, 
 after returning from Mr. Billings' funeral, when 
 I was recalled and requested to proceed at once 
 to Philadelphia to take charge of the Canadian 
 Geological Exhibit at the Centennial Exhibition. 
 Dr. Harrington, of the survey staff, who had been 
 in charge several weeks and whom I went to 
 relieve remained a day to see me settled, and 
 then left for Montreal. 
 
 My duties were to answer any questions-relat- 
 ing to the specimens exhibited, ana -to keep a 
 general supervision over our department. Many; 
 thousands of people viewed the' -Canadian geo- 
 logical exhibit every day and many amusing 
 remarks were made about some of the specimens. 
 The large gilded pyramid representing the amount 
 of gold taken from the mines in British Colum- 
 bia during a certain number of years, attracted 
 much attention. It was pinched, scratched, 
 sounded with the knuckles, and otherwise ill- 
 treated. One old lady after rapping it with her 
 knuckles exclaimed : " Why that ain't gold, what 
 a sell." A man, after sampling it with his knife 
 said, it was " wood made to look like gold to gull 
 the public." 
 
 The fine display of amethyst quartz from 
 Thunder Bay, Lake Superior, which was not 
 under cover, gave me a lot of trouble. People 
 were constantly trying to steal some of the small 
 pieces. An elegantly dressed lady while passing 
 101 
 
REMINISCENCES 
 
 this display took a very fine specimen and hid it 
 under her cape. I saw the act, and although 
 sorry, requested her to replace it. 
 
 After eight weeks constant attention to my 
 duties, Dr. Bell, of the Geological staff, came to 
 relieve me. I remained a day to post him up, 
 and then returned to Montreal, not sorry to re- 
 sume my quieter duties of the Survey. 
 
 I have often before alluded to the supposed 
 Lanrentiaa fossil Eozoon, but I wish to record 
 here thfe interest scientific men from all parts of 
 fche world took ,in this supposed organism. Fre- 
 quently before looking at any other objects they 
 would say " I want to see the specimens of 
 Eozoon," and so it was at the late Chicago Exhi- 
 tion where the finest display yet made at any 
 exhibition was to be seen. 
 
 The winter of 1876-77 was occupied in the 
 usual museum and office work. Many micro- 
 scopic sections were made, fossils developed, etc. 
 Mr. Billings was sadly missed by those who had 
 been so closely associated with him, and it was 
 long before we could reconcile ourselves to his 
 successor, Mr. J. F. Whiteaves, late Recording 
 Secretary to the Montreal Natural History 
 Society, and Scientific Curator of the Museum. 
 
 A good portion of the summer of 1877 was 
 
 again spent in the Eastern Townships, trying to 
 
 gather material which would throw more light 
 
 on the " Quebec Group," and, at the same time, 
 
 102 
 
AMONG THE ROCKS. 
 
 to make a large collection of rocks and minerals 
 for distribution. South Stuckley, Harvey Hill 
 Copper Mines, Hancherchief, and other places 
 were visited, and on the 12th of June, I accom- 
 panied Mr. Arthur Webster, of the geological 
 staff, to Berthier, P. Q. We next went to Drum- 
 mondville, the scene of my first explorations for 
 the Survey. But the old couple who had watered 
 Mr. Richardson's whiskey in 1863 were gone, and 
 as my friend Webster said, " lay with the tips bf 
 their toes turned to the roots of the daisies." A 
 good bit of new information was gained regarding 
 the fossiliferous rocks of that locality. Kingsey 
 and Danville were then visited, after which I left 
 Mr. Webster, and on the 25th June I left Quebec to 
 join Mr. Richardson at St. Jene, island of Orleans. 
 After a short time among the rocks of that vici- 
 nity, we crossed the island to Ste. Famille, on the 
 north side, and went carefully over the rocks to 
 the west-end, where we again met our old friend 
 Lizotte, who, since entertaining Sir William and 
 myself in 186 4-, had built a much more pretentious 
 hotel. 
 
 From the island of Orleans I again went to 
 Point Levis and worked among the conglomerates. 
 It was during this visit that I collected the beau- 
 tiful group of Salteretta, at St. Joseph de Levis, 
 the first yet found in the rocks of the Quebec 
 group. Salterella was then supposed to be a 
 characteristic fossil of the Primordial or Lower 
 103 
 
REMINISCENCES 
 
 Potsdam formation. Sir William had told me 
 years ago that some day I would find Salterella 
 in the lower bands of conglomerates at Levis, and 
 I felt proud that his words ha come true, but 
 regretted he had not lived to see my beautiful 
 little group of Salterella. 
 
 Prof. Walcott, of the United States Geological 
 Survey, in one of his publications, credited Dr. 
 Selwyn, director of the Geological Survey of 
 Canada, with the discovery of this fossil ; but 
 perhaps my name got detached from the specimen. 
 
 Leaving Point Levis, I next spent several days 
 examining the conglomerates and other rocks 
 along the south shore of the St. Lawrence, op- 
 posite the east-end of the island of Orleans The 
 fossil fauna of these rocks are the same as those 
 of Levis ; but hold some forms not found at the 
 latter place. 
 
 A short time after returning to Montreal, I 
 again started off to investigate the rocks of the 
 Etchemen River. Mr. Willimott, then lately of 
 the Survey staff, accompanied me. Many new 
 facts relating to the Quebec group were obtained . 
 The first fossil sponge specula yet found in these 
 rocks were discovered in the green shales of the 
 Chaudiere Falls, together with one or two small 
 bivalve shells belonging to the Obolella family. 
 
 The geological structure of the Etchemen River 
 rocks is very interesting, and as one journeys by 
 stage or buggy up or down the river road, and 
 10* 
 
AMONG THE ROCKS 
 
 puts up for the night at the various villages, he 
 is sure to meet with incidents of Canadian life 
 which, with the stories of the " drummer," who is 
 almost sure to be met with at country villages 
 will well repay a journey through this part of 
 Canada. 
 
 Before leaving the field for this year the fol- 
 lowing localities were visited and much work 
 done with the hope of throwing more light on 
 the rocks of the " Quebec group." Sept. 7th, I 
 was at St. Garvis studying the quartzites and 
 black slates ; Sept. 13th, at Armagh ; Sept. 15th, 
 at Rivieredu Sud ; Sept. 22nd, at St. Francis ; 
 Sept. 25th, at Berthier ; Sept. 27th, at St. Thomas, 
 and other places in the Province of Quebec. 
 
 It might be supposed by one not versed in geo- 
 logical field work, indeed, it has often been said, 
 that our work is a series of pleasant trips, at the 
 expense of the Government well, let the critics 
 try a few weeks in some of the newly settled 
 districts of the Townships. 
 
 Field operations being over for this season, I 
 again settled down to my work-room and museum 
 duties. Sir William Logan was always averse 
 to giving away geological specimens ; but this 
 year, assisted by Mr. Willimott, I commenced 
 making up (on a small scale) collections for edu- 
 cational purposes ; these, when distributed to 
 various colleges, proved so important that almost 
 from this date a constant demand has been made 
 105 
 
REMINISCENCES 
 
 on the Survey for these educational collections ; 
 thus, what commenced on such a small scale 
 has become an important branch of the Survey, 
 and now, as for years past, occupies almost the 
 whole of Mr. Willimott's time. 
 
 As with the early distribution of rocks and 
 minerals, so the Ethnological branch of the Sur- 
 vey commenced on a small scale. For three or 
 four years after my appointment to the Survey, 
 half a dozen or so of Indian stone and other 
 implements stood on the mantle-piece in a small 
 room on the " fossil flat " already referred to 
 as being used by Sir William as a sitting and 
 sometimes bed room. These, together with two 
 or three casts of Indian stone pipes, and a few 
 fragments of Indian stone pottery, etc., I had 
 picked up near Niagara, with Sir William's per- 
 mission I placed in a case in the museum ; soon 
 two cases were required, and so commenced the 
 Ethnological collection which now numbers many 
 thousands of specimens, hundreds of which (and 
 some of the finest) are boxed up and stowed away 
 for the want of room to exhibit them. 
 
 106 
 
CHAPTER VII. 
 
 THE SOUTH SHORE OF THE ST. LAWRENCE SOME GOOD FISH- 
 ING CAMPING BY THE RIVER ASCENT OF THE SHICK- 
 SHOCKS A POACHING EXPEDITION : WE CATCH A FINE 
 SALMON THE LONELY MOUNT ALBERT PHOTOGRAPHS 
 AND FOSSILS COAST ROCKS BETWEEN DIGBY AND YAR- 
 MOUTH, N.S. "THE OVENS" FISHING AT SPINDLE 
 
 COVE LOST IN THE FOG. 
 
 ON the 18th of June, 1878, with camp equip- 
 ment and two months' provisions for my- 
 self and one man, I left Montreal for Gaspe. My 
 office instructions were to examine the coast rocks 
 on the south shore of the St Lawrence River 
 between Cape Rosier and Metis, especially with 
 regard to the fossil fauna. 
 
 At Gaspe I was fortunate enough to secure the 
 services of the Indian John Basque, who had in 
 1843 accompanied Sir William Logan over the 
 same ground. John (he is dead now) was a tall, 
 straight, well-built man, and handsome for an 
 Indian. Besides speaking English and French, 
 he spoke the language of several different tribes 
 of Indians, but could neither read nor write his 
 own name, things he very much wished to do 
 and which I tried hard to teach him, but it was 
 no use, and he gave it up for a bad job. 
 
 We made our first camp a short distance from 
 Cape Rosier lighthouse, and after a good supper 
 of fried ham, sea-biscuit and tea, John collected 
 107 
 
REMINISCENCES 
 
 from a neighboring bush sufficient spruce-boughs 
 to cover the floor of our tent. On this our blank- 
 ets were spread, but John seldom came inside* 
 preferring to make a wigwam for himself. 
 
 The summer nights of the Gulf shore of the 
 St. Lawrence are always more or less chilly, but 
 drift wood is plentiful in most places, and one 
 can always, without the use of the axe, keep up 
 a rousing fire. 
 
 I found John a pleasant companion, and I sat 
 at our camp fire and watched the dark clouds 
 spread over the distant mountains. 
 
 All was solitary and still excepting the ever- 
 lasting splash of the waves on the pebbly beach 
 before us. It was long after darkness had ob- 
 scured every object within a few yards of our 
 camp fire before I tumbled into my blankets. 
 When I awoke in the morning I found my man 
 busy preparing our breakfast. John, who never 
 lost an opportunity of making new friends, had 
 already been to the lighthouse and brought back 
 some fish. After breakfast we closed up our tent, 
 and with a little lunch in our collecting basket, a 
 gun, hammers and note -book, started on this the 
 first part of our geological tramp of over one 
 hundred and eighty miles along the base, and 
 frequently over the tops of the great cliffs of the 
 Gulf shore. 
 
 My examination of these exposures commenced 
 at the base of the cliffs almost under Cape Rosier 
 108 
 
AMONG THE ROCKS. 
 
 lighthouse. Here the strata consists of conglom- 
 erates, grey limestone bands, black bituminous 
 limestone, greenish and other coloured shales. 
 No fossils had hitherto been found by which the 
 geological horizon could definitely be determined, 
 but they were supposed by Logan to belong to 
 the Hudson River formation. It was my good 
 fortune, however, on this the first day's work to 
 find, almost in the shadow of Cape Rosier light- 
 house beautiful compound Graptolites which 
 evidently belong to the Levis zone which accord- 
 ing to the latest nomenclature of the Geological 
 Survey of Canada is Upper Cambrian. 
 
 The sun was setting when we returned to 
 camp. I think John thought we had done a little 
 too much work for one day, but I assured him 
 that I should not work so hard every day and 
 that he would not have to work on Sundays, as 
 with Logan, excepting to cook a thing an Indian 
 is always ready to do. 
 
 It was long after dark when I crept into my 
 blankets, but before doing so I had put some dry 
 plates in the dark boxes of my camera which I 
 had brought with me hoping to get some good 
 geological views of the coast. 
 
 I may mention here that I did get many fine 
 views, some of which have become historical, 
 having been published in the reports of the Sur- 
 vey. I may also mention here that all the nega- 
 tives taken by the Survey staff are carefully pre- 
 109 
 
REMINISCENCES 
 
 served. They have lately been re-arranged, cata- 
 logued, numbered and placed where any one 
 of them can be found at once, by Mr. Percy 
 Selwyn, who is now private Secretary to the 
 present director of the Survey, Dr. G. M. Daw- 
 son. 
 
 It would extend these notes too much to 
 recount all the adventures John and I met with 
 while journeying along this great stretch of sea 
 coast. We shifted camp frequently, generally 
 every eight or ten miles. By so doing we were 
 able to walk down the coast to where we left off 
 our examination, and up the shore to where we 
 would camp next. Our mode of shifting camp 
 equipment, etc., was by boat or cart, whichever we 
 could get most conveniently. 
 
 During our journey by boat I frequently had 
 some good fishing. On one occasion while sitting 
 at the stern of the boat, I noticed fish jumping 
 at flies, I suppose ; but I did not see any. Hav- 
 ing a trolling line with hooks and spoon, silvered 
 on one side and red on the other, I \vas not long 
 in getting it ready. The two boatmen nudged 
 each other and winked, while John gave a sar- 
 castic smile, but I let go my troll, and before the 
 men had pulled half a dozen strokes I hauled in 
 a lovely mackerel, and before we completed our 
 ten miles' journey, twelve fine mackerel and two 
 cod fish lay at the bottom of our boat. The men 
 were greatly surprised and so was I. Fishing 
 110 
 
AMONG THE ROCKS. 
 
 for mackerel and cod with a trolling spoon was 
 a new feature on the St. Lawrence. 
 
 Landing at Gritfon Cove, about ten miles above 
 Cape Rosier we pitched our tent at the base of 
 huge black cliffs composed of shale interstratified 
 with other rocks. The escarpment represented 
 here is probably 1,000 feet thick. Many of the 
 beds are highly fossiliferous, some being very 
 prolific in Graptolites, which have a decided 
 Hudson River aspect. 
 
 In the late publications of the Geological Sur- 
 vey of Canada, the formation here and for about 
 eighty miles of the coast to the Marsouin River 
 is Combeo Silurian i.e., Trenton, Utica, &c. The 
 late changes made in the geological horizon of 
 the coast rocks are chiefly due to fossil evidence 
 collected by the officers of the Survey, since the 
 time of the late Sir William Logan. All the 
 fossils collected on the journey of which I write, 
 and late journeys of Dr. Ells, Dr. Selwyn and 
 myself, have been studied by Prof. Chas. Lap- 
 worth of Birmingham, England, who I believe is 
 the best English authority on the Graptoliae 
 fauna. 
 
 The coast country between Cape Rosier and 
 Cape Chat, a distance of about 130 miles, is 
 mountainous. The hills and cliffs come close on 
 to the shore leaving for many miles no space for 
 settlement, excepting at the mouth of rivers 
 where one finds fishing stations of more or less 
 111 
 
REMINISCENCES 
 
 pretensions. We always received a warm wel- 
 come from these Gulf shore people who seldom 
 see any other than their own class. 
 
 We made many camps between Griffon Cove 
 and Cape Chatte, and gained much information 
 regarding the fossil fauna. 
 
 At Ste. Anne, ten miles below Cape Chatte I 
 found my old friend Mr. James Richardson of 
 our Survey staff camped near the foot of the 
 river. Mr. Ord also of the Survey staff, Mr. 
 John Richardson and one Indian had just returned 
 from the foot of the Shickshocks mountains, a 
 distance up the river about 30 miles not taking 
 in the many crooks of the stream. 
 
 As it was my intention to ascend one of the 
 highest peaks of the Shickshocks, and as it re- 
 quired two experienced men to pole a canoe up 
 this river I engaged the Indian Joe, who had 
 just returned with Mr. Richardson. But Joe 
 flatly refused to accompany me till he had been 
 drunk, at least one day. John, my Indian, as- 
 sured me Joe would turn up all right according 
 to promise, said he "If Joe says a day, he means 
 a day, and no longer." It was impossible to get 
 another man who knew the river like Joe, for he 
 had been tripping on that stream for years, so I 
 had to submit to Joe's terms. A certain hour 
 was fixed for his return to camp, and to my 
 surprise but not to John's, Joe turned up all 
 right, and when asked how he felt after being 
 112 
 
AMONG THE ROCKS. 
 
 drunk a day and night, said, "Bully." I preferred 
 a long narrow boat, much used on this river, to 
 the canoe Richardson had used. This I obtained 
 from the settlement at the foot of the stream, 
 and we were soon ready to start. We took our 
 blankets and three days' provisions, and of course 
 a gun and fishing tackle. We lunched off a 
 magnificient trout Joe stole from a net. It was 
 the most delicious fish I ever ate. 
 
 The Ste. Anne River is, or was, at the time I 
 ascended it, one of the finest salmon rivers in 
 Canada. In many parts the current is swift and 
 requires dexterous poleing and in places portag- 
 ing, but in places one passes over still pools where 
 the paddle cm be used. As we glided over these 
 pools and looked down into the clear water we 
 could see many salmon quietly floating or resting 
 on the bed of the river. 
 
 It was dusk when we hauled up for the night, 
 ten miles from our starting point. 
 
 After supper John and Joe collected a lot of 
 birch bark which was soon made into a flambeau. 
 Then Joe fished out from some hiding place, a 
 salmon spear which if found on him at the 
 settlement would have cost him lots of trouble. 
 When the night was as dark as it would be, we 
 stole forth and silently glided down stream. Then 
 I realized that for the first time I was on a 
 poaching expedition for the fishing rights of 
 the Ste. Anne were very strict, and I had no 
 
 H 113 
 
REMINISCENCES 
 
 permit. But almost before I had time to reflect 
 on our evil ways, a twenty pound salmon lay at 
 the bottom of our boat. I forbade the men to 
 take any more fish than we needed for a change 
 of diet. 
 
 I shall never forget this my first poaching 
 experience, the stillness of the night, the brilliant 
 light of the flambeau reflected on the eager faces 
 of the two Indians, made up a scene well worthy 
 of the artist's brush. 
 
 The following morning we resumed our journey. 
 Towards noon the sun was very hot and the 
 weather most oppressive. John had been stand- 
 ing at the bow of the boat fully an hour assist- 
 ing Joe to push through one of the most difficult 
 parts of the stream, to navigate. Suddenly he 
 let go his pole and sank to the bottom of the 
 boat. Instantly our boat swung round and shot 
 down the swift-current, but in a short time Joe 
 ran it ashore arid we hauled John out to a shady 
 spot and laid him on his back, and for the next 
 two hours I thought we were going to have a 
 dead Indian to take back. We came to the con- 
 clusion that John was suffering from sun-stroke. 
 Eemembering I had a bottle of " Pain Killer " in 
 my fishing basket I made up a strong dose, but 
 it was a pretty big job to get John to take the 
 white man's medicine. He did however, at last, 
 and in the cool of the evening we resumed our 
 journey. 
 
 114 
 
AMONG THE ROCKS. 
 
 The third evening after leaving our camp, near 
 the mouth of the Ste. Anne, we reached the foot 
 of the Shickshocks and camped for the night in 
 a trapper's deserted wigwam. At six o'clock 
 next morning we breakfasted on the remains of 
 our twenty pound salmon and then prepared for 
 the ascent of Mount Albert. John carried my 
 camera and other traps wrapped in blankets, 
 while Joe carried two days' provisions and camp 
 utensils. I carried well myself, a canvas bag 
 and small hammer, which I found quite enough- 
 I could not help wondering that while I stumbled 
 sometimes head first over fallen trees or slipped 
 down moss covered rocks, my two Indians glided 
 through all the difficulties which beset us with- 
 out a slip or stumble. 
 
 In about two hours we reached the summit of 
 Mount Albert 3,768 feet above the sea a great 
 dreary table-land with a few stunted spruce 
 trees, and strewn with weathered rein deer (?) 
 antlers. We chanced to reach the summit of this 
 mountain within a short distance of where Logan 
 and Murray in 1844, erected a flag-staff and 
 unfurled the Union Jack. The pole was still 
 lying at the foot of the mound of stones which 
 had supported it, but it was broken and weather 
 worn so John descended the mountain till he 
 found another good stick ; we then rebuilt the 
 mound, hoisted our pole and in place of a Union 
 Jack substituted a red cotton handkerchief. 
 115 
 
REMINISCENCES 
 
 The magnificent panorama presented from the 
 summit of Mount Albert is grand indeed. The 
 eye ranges over hills and valleys for a hundred 
 miles or more. The riyer we have ascended 
 looks almost like a brook winding through 
 mountains and forest till it joins the St. Lawrence, 
 the waters of which are dotted with ships and 
 fishing boats. 
 
 Although we can see some life in the distance, 
 I shall never forget the sense of loneliness I felt. 
 Even the Indians while we sat round our camp 
 fire spoke in whispers as though evil spirits 
 wandered round. 
 
 Crossing the table-land to the south side of 
 the mountain, we look down over masses of 
 serpentine and other rocks, into a great valley in 
 which a small lake nestles. I set up my camera 
 hoping to get a view of this wonderful scene; but 
 the black flies are so numerous that they partly 
 cover my lens and I see by my focussing glass 
 that a number have got inside the camera. I 
 look up to speak to Joe and find he is clasping a 
 small crucifix which hangs from his neck. He 
 is actually trembling with fright. " Joe what 
 on earth is the matter with you ?" I say ; Point- 
 ing to the lake, he says in a whisper " Look 
 see the devil is taking a swim." I look and find 
 a strong current of wind is passing through the 
 valley, which accounts for the heavy ripples on 
 the water. I asked Joe if he had any name for 
 116 
 
AMONG THE ROCKS. 
 
 that lake and he whispered yes, "that devil 
 lake." All I could say to this poor fellow would 
 not alter his opinion regarding the devil and his 
 ablutions. 
 
 Granites enter largely into the formation of the 
 Shickshocks, but hornblende, quartzites, epidotic 
 and other rocks are largely represented, together 
 with massive beds of serpentines, which are fre- 
 quently beautifully stratified. Pieces of chromic 
 iron ore, about the size of one's fist, are scattered 
 over the table-land of Mount Albert, but I could 
 not find the bed from which they came. 
 
 We only made one night camp during our 
 return journey a journey full of pleasant remem- 
 brances, and of much geological interest. 
 
 Arriving at our old camp near the mouth of 
 the river, I found Mr. Richardson waiting me. 
 Poor Mr. Richardson, the veteran explorer for the 
 Geological Survey of Canada almost from its com- 
 mencement a man who had worked his way 
 from a poor farmer to that of the most trusted 
 field geologist then on. Logan's staff, had received 
 an intimation that the present director of the 
 Survey was about to apply to the government 
 for his superannuation. This, to a man like 
 Richardson, was a blow indeed, for outside his 
 geological work he had little or no pleasure. He 
 remained with me several days, during which 
 time we visited two or three interesting fossil 
 localities on the coast, one of these was about 
 117 
 
REMINISCENCES 
 
 three miles above the Ste. Anne. Here we obtained 
 large slabs of black slate covered with the well 
 known Levis fossil Phillograptus typus (Hall), 
 and a number of other species. After a few days 
 Mr. Richardson left me and returned to Montreal. 
 John and I continued our journey along the coast, 
 camping as usual about every ten miles. Many 
 new fossil localities were discovered, and forms 
 found which were new to Canada. Duplicates 
 of all the fossils collected were sent to Prof. 
 Charles Lap worth, of Birmingham, England, for 
 identification and description. 
 
 During this long sea coast experience many 
 photographs were taken, prints of which can be 
 seen with the large collection of photographs in 
 the Geological Survey rooms. One of these views 
 shows the " Pillar Sandstones " on the coast, eight 
 and one half miles east of Ste. Anne River. I 
 focussed my view, then left John to expose the 
 plate, so that I might appear in this picture with 
 a little Indian dog who came to our camp one 
 day, whence we did not knew. He was first seen 
 sitting on his hind legs, as if asking to be taken 
 in. John at once adopted and christened him 
 Wap-e- cat, which he said was the Indian for "white 
 paw." This picture is reproduced in the Geologi- 
 cal Report for 1880-81-82, but by some mistake 
 is credited to Dr. Ellis, 1883. 
 
 I am loath to close this brief account of a jour- 
 ney during which every day brought to life some 
 118 
 
AMONG THE ROCKS. 
 
 new geological information, besides giving one 
 an insight into the life and habits of the poor 
 fishermen, many of whom labor year after year 
 and never see anything other than their immediate 
 surroundings. 
 
 We reached Little Metis Sept. 21st, where we 
 struck camp for the last time. I paid off my 
 Indian, John, saw his money safely sewn up inside 
 his vest, and started him off to his wigwam at 
 Gasp& 
 
 The winter of 1878-79 was spent in the usual 
 museum and and work-room duties and in some 
 preparations for the removal of the Survey to 
 Ottawa. 
 
 My first exploration journey of 1879 was to 
 Roxton Falls, P. Q., to examine the limestone and 
 shales associated with the copper deposits. June 
 5th, I left for Nova Scotia to examine the shore 
 rocks of St. Mary's Bay. The journey from St. 
 John, N. B., to Annapolis by steamboat 45 miles 
 across the Bay of Fundy is, in calm weather, a 
 pleasant one, but this time it was most unpleasant. 
 From the time we left St. John, till we arrived 
 at Digby Gut the entrance to Annapolis Basin 
 the waves frequently swept over our boat. All 
 passengers were below deck, and most of them 
 sick. One old body was sick unto death, and if 
 ever she reached dry land again, &c., &c. I am 
 a good sailor and did justice to my 75 cent dinner. 
 
 The occurrence of fossil remains in the gold 
 119 
 
REMINISCENCES 
 
 bearing rocks of Nova Scotia, by which the geo- 
 logical horizon could be determined, would be an 
 exceedingly important discovery. From time to 
 time forms which were thought to be of organic 
 structure have been found, as the following re- 
 marks by the writer, taken* from " Transactions 
 of the Nova Scotian Institute of Science," Series 
 2, Vol. 1, will prove. But in all cases a micro- 
 scopic examination of these forms has proved 
 them to be concretionary bodies 
 
 " Many times between the years 1860-70 the late Sir 
 Wm. E. Logan, and subsequently Dr. Selwyn, called my 
 attention to certain concretionary forms found in the gold- 
 bearing rocks of Nova Scotia. Some of these seemed to 
 be organic, and I was requested to make and examine 
 microscopic sections of them. In treating several of these 
 with acid, they proved to be composed chiefly of dolomite, 
 with a large proportion of siliceous matter, and generally 
 a little iron pyrites, which formed a nucleus. 
 
 " In 1890 a number of similar forms were found by Mr. 
 Willis, in the rocks of the Northup Gold mines, Rawdon, 
 Nova Scotia. They were handed to Professor Hind, who 
 supposed them to be fossils, and assigned them to Lower 
 Silurian age. Wishing the "judgment of a specialist," 
 he gave them to Professor Kennedy, of King's College, 
 who confirmed Professor Hind's opinion and pronounced 
 the " fossils " to be Stromatopora. Mr. Fletcher and Mr. 
 Faribault, of the Geological Survey, while in the vicinity 
 where this discovery was made, visited the mines, and 
 brought away a number of these so-called fossils. They 
 were given to me by Dr. Selwyn, director of the Dominion 
 Geological Survey, for microscopic examination ; and I 
 regret to say, the result is precisely the same as for those 
 examined thirty years ago. 
 
 120 
 
AMONG THE ROCKS. 
 
 " They appear to be composed of dolomite, and, when 
 dissolved in hydrochloric acid, leave a good percentage of 
 insoluble matter, probably felspar and silica. It is likely 
 that they were spheroidal or ovoidal in form before being 
 flattened by the pressure of overlying beds. One of the 
 specimens before me is a piece of greenish-grey laminated 
 mica-schist five inches long and one inch thick. Inclosed 
 in this are four of these concretionary forms broken 
 through the centre, each measuring one inch in length and 
 half an inch in breadth. Two of these are connected with 
 each other by a thin strip of the material of which they 
 are composed. 
 
 "In broken sections some of these bodies show slight 
 concentric layers which in microscopic sections are not 
 seen. Not a trace of organic structure was found. 
 
 ' ' I quite agree with Professors Hind and Kennedy as to 
 the importance of finding fossils in the auriferous rocks of 
 Nova Scotia, and trust they may be more fortunate than I. 
 
 " It is well known that concretions occur in all rock for- 
 mations. One or two instances will be worth recording to 
 show how careful one should be in referring any forms of 
 a concretionary nature to organic structure (Read Nov. 
 9th, 1891.)" 
 
 It was chiefly to try and discover fossils in 
 these auriferous rocks that the present journey 
 was made. 
 
 Gold-bearing deposits occupy a great portion 
 of the Atlantic coast of Nova Scotia. The rocks 
 consist of greenish and grey slates, quartzites and 
 other rocks, which, if as stated in the Geological 
 Survey Reports, belong to the Primordial Silurian 
 and Cambrian formations should contain fossils, 
 but at present nothing with decided organic struc- 
 121 
 
REMINISCENCES 
 
 ture has been found. The so-called Eophyfon 
 from the " Ovens " gold-bearing slates is in my 
 opinion not of organic origin. 
 
 Many miles of coast rocks between Digby and 
 Yarmouth were examined, especially those of 
 Wey mouth, Church Point, Montegan, Salmon 
 River, Bear Cove, Cape St. Mary, Cranberry 
 Head, &c. 
 
 Having a good camera and dry plates with me 
 I was able to obtain excellent photographs show- 
 ing the stratification, folding, &c. of these rocks. 
 The Nictaux iron ore deposits were also visited. 
 Here the Hematite ore-beds contain well de- 
 fined fossils which belong to the Devonian for- 
 mation. 
 
 I travelled along the coast road by stage or 
 buggy. The journey through the Annapolis val- 
 ley is a most pleasant one. It is the great apple 
 raising country of Nova Scotia. Continuing 
 along the coast and shore road I reached Yar- 
 mouth, where after a few days' work among the 
 rocks of that vicinity I left for Lunenburg the 
 oldest British town in the Province of Nova 
 Scotia and on the 9th of July, crossed the harbor 
 to the u Ovens," already mentioned with regard 
 to the so called Eophyton, (Torell), eos, dawn ; 
 phyton a plant. From the harbor of Lune-nberg 
 are seen on the west side a range of cliffs con- 
 taining a number of even cavities, the regularity 
 of which whether made by human agencies or 
 122 
 
AMONG THE ROCKS. 
 
 by the action of the sea no doubt suggested the 
 name " Ovens." 
 
 These rocks were known to contain gold many 
 years ago, and considerable excitment has at 
 various times prevailed, when a specimen of 
 quartz or slate had been found which held a 
 good sample of the precious metal. 
 
 The only mining going on at the time of my 
 visit (1879), was done by a few of the fishermen 
 of Spindle Cove, on the " Ovens " shore, who with 
 a rude cradle, shovel, or other simple implements 
 gathered from the sea sand of the shore, gold 
 dust to the amount of from fifty cents to a dol- 
 lar a day. But my visit to the Ovens was not 
 for gold ; but for fossils, which after a most dili- 
 gent search I failed to find, nor could I find 
 another specimen ot the so-called Eophyton 
 marks, which could readily be made by a bit of 
 rock passing over a soft substance such as these 
 slates were once. 
 
 At the time of my visit to the " Ovens " there 
 were only five or six families living near thie 
 shore ; they were fishermen. 1 found board and 
 lodging with a German family at Spindle Cove. 
 Among my accounts of this journey I find the 
 item : Four days' board and lodging at " Ovens " 
 $2.00, but besides this I paid $2.50 for 
 boating and assisting with my camera. This 
 small sum of $4.50 to these poor people meant 
 much. When shown to my room the first night 
 123 
 
REMINISCENCES 
 
 of my stay with these fisherfolk I found there 
 were two feather beds only. Thinking they had 
 forgotten the sheets and blankets, I found my 
 way down stairs where already they were at 
 their family devotions. Waiting outside the 
 room till prayer was over, I then requested the 
 man to go up to my room where I explained that 
 there was no covering to the bed. He looked 
 puzzled and called his wife up, when I made 
 known what I wanted she looked surprised, but 
 a happy thought seemed to strike her, and she 
 went away with a smile on her face, soon she 
 returned with a quilt, but at the same time 
 thought I would find it more comfortable to 
 sleep between the feather beds. 
 
 I have spoken of catching mackerel with a 
 hook and spoon, but that fishing was tame to 
 the sport I had at Spindle Cove. Shoals of her- 
 ring the well known Yarmouth bloater came 
 into Lunenburg harbor, followed by the horse 
 mackerel even into the fish traps, or seines. It 
 is a powerful fish and much dreaded by the fish- 
 ermen who have herring nets or seines set, and 
 who lose no time in getting them out of the way 
 by harpooning. 
 
 During my stay at Spindle Cove I assisted in 
 harpooning two of these creatures which were at 
 last killed by severing the caudal vertebrae with 
 an axe. One of these measured five feet long 
 and was over two feet in diameter. I had it 
 124 
 
AMONG THE ROCKS. 
 
 towed to shore and then photographed it. In 
 appearance these huge fish are just the same 
 shape and color as our mackerel ; but the tlesh 
 is coarse ]ike that of the seal. 
 
 On the 12th of July, I crossed the harbour 
 again (four miles) with the people I had been 
 staying with. They, at least the man and his 
 two daughters, were taking fish to the Lunenburg 
 market. There was a light fog when we started 
 which soon increased till we could not see a 
 yard before us. At last the man and his daugh- 
 ters ceased rowing, confessed they were lost 
 and did not know where they were. Fortunately 
 the man knew the compass bearing from his 
 shore. I looked at my compass and found we 
 were headed almost in the opposite direction. 
 This however, these folks could not believe, but 
 at last decided to be guided. After pulling an 
 hour or so through the dense fog we ran right 
 against the wharf at Lunenburg. I wished my 
 companions of the fog good-bye, took steamer 
 and was soon at Halifax, were I spent a short 
 time with my friend Kev. Dr. Honeyman, cur- 
 ator of the Provincial Museum, who had been in 
 1876 my companion for eight weeks at the 
 Centennial Exhibition, Philadelphia. 
 
 While on this journey I had received official 
 instructions to visit on my way back, the Jog- 
 gins shore and get photographs of various cliffs, 
 to illustrate a paper Sir William Dawson was 
 125 
 
REMINISCENCES 
 
 writing for the Royal Society. I was fortunate 
 in getting what Sir William wanted, and these 
 were published in the Royal Society Reports. 
 I reached Montreal on the 20th of July. 
 
 On the llth of August I started out again to 
 the Eastern Townships, visiting West Farnham, 
 Bedford, St. Dominique, Richmond, St. Pie and 
 again to West Farnham. October 1 9th, I accom- 
 panied Mr. Richardson to St. Armond and 
 Grand Ling. This was the last official journey 
 Mr. Richardson made. He was superannuated 
 greatly against his will shortly afterwards, 
 This completed my field work for 1879. 
 
 The winter of 1879-80 was occupied in my 
 usual microscopic and museum work. A great 
 number of microscopic sections of rocks and fos- 
 sils and preliminary examinations were made. 
 In the museum many new fossils were labelled 
 and arranged in the cases. In some of the work 
 I was assisted by Mr. Willimot. 
 
 126 
 
CHAPTER VIII. 
 
 BAIE DE CHALEURS AND ADJACENT DISTRICTS A FOSSIL PISH 
 AND A PIOUS CRITIC A LIST OF TITHES -REMOVAL OF 
 THK HEADQUARTERS OF THE SURVEY FROM MONTREAL 
 TO OTTAWA THE JOGGINS COAST, N. S. CAMPBELLTON 
 AND THE RESTIGOUCHE RIVER A TRIP TO SAULT STE 
 MARIE THE CYPRESS HILL AND THE NORTH-WEST A 
 THREE MONTHS GEOLOGICAL EXPLORING TOUR. 
 
 of the summer of 1880, was spent in 
 the examination of the shore, rocks of Baie 
 des Chaleurs, Cascapedia and Matapedia Rivers, 
 Campbell ton and other localities in that district. 
 
 The shore rocks of the Restigouche River near 
 the back of Campbellton Railway Station and 
 those on the opposite shore are highly interest- 
 ing, containing as they do, a remarkable fauna 
 and flora of fish and plant remains of Devonian 
 age. Many fine fossils were collected from this 
 locality by myself and later by Mr. Foord. 
 
 On the llth of June, I crossed from Dalhousie 
 to the north shore of Scaumenac Bay where I got 
 board and lodging at a farm house, the occupants 
 of which were an old couple and a grown up son 
 and daughter. 
 
 My field of research here was along the shore 
 
 towards Pt. Maquasha, Baie des Chaleurs, but 
 
 chiefly on the shore of Scaumenac Bay. Here 
 
 high cliffs of lightyellowish grey sandstone occupy 
 
 127 
 
REMINISCENCES 
 
 the shore. The rocks are of Devonian age and 
 
 o 
 
 underlie the lower Carboniferous conglomerates 
 of the north shore of the Restigouche River. It 
 was in these Devonian cliffs of Scaumenac Bay 
 that Mr. R. W. Ells of the Geological Survey in 
 1879, discovered that curious crustacean Pterich- 
 ihys ; or fish belonging to the Ganoids which- 
 ever it is. This discovery led to my present 
 researches and during the next summer to those 
 of Mr. Foord, also of the Geological Survey staff. 
 
 Some of the beds of these great sandstone cliffs 
 contain nodules, or concretionary forms. The 
 following description from my notes on concre- 
 tionary forms published in Transactions of the 
 Nova Scotian Institution of Science will suffice. 
 
 'The Upper Devonian fish and plant-bearing beds of 
 Scaumenac Bay, New Brunswick, are prolific in fossilifer- 
 ous concretions, which are composed of calcareo -arenac- 
 eous rock, and take various forms according to the shape 
 of the nucleus, which, when a fish, is often so well pre- 
 served that every bone can be seen. One of these con- 
 cretions obtained by A. H. Foord measures over twenty- 
 one inches in length, and contains the skeleton of a fish 
 almost as long. It is Chirolepis Canadensis (Whiteaves). 
 In other concretions from this locality the writer and Mr. 
 A. H. Foord found : Glyptolepis microlepidotus (Agassiz), 
 Phaiwroplenrvn curium, Pterichthys Canadensis (Whit- 
 eaves), Eusthenopteron Foordi, etc." 
 
 These concretions are scattered along the shore, 
 
 being washed up by the waves of the bay. It 
 
 was in one of these that I discovered the first 
 
 fossil fish known to the Survey, from these 
 
 128 
 
AMONG THE ROCKS. 
 
 rocks This find is always associated with the 
 old couple with whom I boarded. When in the 
 barn carefully chiselling off a portion of rock 
 which concealed part of my fossil fish, the old 
 lady came in, and after watching me for a short 
 time said : " Well now, it do look like a mackerel, 
 but if God made stone fishes it was for some wise 
 purpose that we poor mortals can't understand 
 and ought not to meddle with, and Sir, it would 
 be much better for you to leave them where the 
 Lord placed them." I tried to explain that a few 
 million of years ago this fish swam in the sea, 
 then died and was buried in the sand and mud 
 at the bottom, then the sand and mud became 
 stone. But the few millions of years seemed to 
 frighten the old lady and she left me. Then the 
 old man came to interview me on the subject, and 
 chided me for even thinking of a few millions of 
 years ago, said he " Do you dispute the words of 
 the Holy Bible, etc., etc." That evening I was 
 requested to attend service before retiring for 
 the night. The old man read a chapter from the 
 Bible explaining the same as he went on, then 
 the family sang a long hymn, drawling out each 
 word till I dozed off to sleep. But at last sing- 
 ing was over, and then the old man delivered 
 what some church people would call a most pow- 
 erful prayer, full of beautiful thoughts, which 
 ended with " Oh Lord bless the stranger within 
 our gates, and keep him from vain babbling ; and 
 J 129 
 
REMINISCENCES 
 
 turn his mind from things of the past, to his 
 future salvation, Amen." No doubt the few mil- 
 lions of years was in the old man's thoughts 
 when he spoke of vain babbling. 
 
 The following day I collected from these same 
 fossil fish-beds, fossil plants, among which were 
 specimens of an old fashioned fern which Sir 
 William Dawson has since named Archaeopteris 
 Jacksoni. This specimen puzzled the old man of 
 my boarding house, but no further allusion was 
 made to vain babbling in the old man's prayer 
 that evening. 
 
 Leaving Scaumenac Bay July 1st, I proceeded 
 to New Richmond, where a few days were spent 
 on the rocks of that vicinity, a journey was then 
 made to Causapscal railway station, on the Inter- 
 colonial Railway, and on the Campbellton road, 
 near the Matapedia River. Here I obtained 
 lodging at the trackman's cottage, where I fed on 
 fat pork, brown bread and potatoes, while H. R. 
 H. Princess Louise two or three hundred paces 
 away lived on the best of the land, and while 
 I fished with my hammer among the rocks for 
 fossils, she fished in the river close by, for sal- 
 mon ; and got them, too. Several members of 
 the Royal family have been the guests of Sir 
 Donald Smith, who has a summer house or fish- 
 ing station on the banks of the Patapedia River, 
 a lovely spot close to the Causapscal Ry. station, 
 which, since being patronized by royalty, has 
 130 
 
AMONG THE ROCKS. 
 
 become quite a noted spot on our long railway. 
 The platform of this station is the lounging place 
 for Indians while waiting to be engaged by sports- 
 men who go up the Matapedia River, either to 
 fish or shoot. 
 
 Having completed my examination of the rocks 
 at the Devil's Elbow, a turn on the river two 
 miles or so below the station, with a canoe, two 
 Indians and a few day's provisions, we left for a 
 journey up a portion of the Matapedia River. 
 The rocks of this river belong to the Gasps' series, 
 and are almost destitute of fossils. One who has 
 camped on this beautiful river will never forget 
 the charming scenery, and if he is fortunate 
 enough to " hook " a salmon or one of the large 
 trout found in the rivers of this vicinity, and to 
 have his Indian canoe-men cook it in their 
 fashion, he will long remember his camp on the 
 Matapedia River. 
 
 Arriving at Campbellton on Saturday evening, 
 I concluded to stay there nntil Monday. Judge 
 
 who was on his circuit, and staying at 
 
 the same house, invited me on Sunday morning 
 to accompany him to service at a sraall Catholic 
 church on the north side of the river. We en- 
 gaged an Indian to paddle us over in his canoe. 
 The little church was already crowded with 
 people of various nationalities, with a good 
 sprinkling of Indians. 
 
 After the sermon in which the priest exhorted 
 131 
 
REMINISCENCES 
 
 his flock to attend to their religious duties more 
 diligently, on pain of excommunication he pro- 
 duced a roll of paper in which were the names of 
 those who had and had not paid their tithes : 
 John Brown, four cords of wood; Patrick O'Farity, 
 ten pecks of potatoes ; Peter Basque, fifteen pecks 
 of beans ; John Gabriel, one side of pork ; Francis 
 Cye, one ton of hay ; Narcisse Cromk, NOTHING, 
 and the priest looked round with fire in his eye, 
 but he could not spot poor Narcisse, and so the 
 list went on. 
 
 My wanderings during the remainder of the 
 field season of 1880 covered many hundreds of 
 miles, and new geological facts were obtained 
 from the rocks at Father Point, Rimouski, 
 Negette, Bic, and many other localities on the St. 
 Lawrence shore and in the Eastern Townships. 
 
 The most important event connected with the 
 Survey this year, 1880, was the removal to 
 Ottawa. 
 
 The director's summary report for 1881 states 
 that : "The total number and weight of packages 
 forwarded from Montreal between the month of 
 November, 1880, and May, 1881, was 1,729 boxes; 
 101 barrels ; 162 miscellaneous packages gross 
 weight, 282,585 Ibs. 
 
 The work of packing all the type fossils from 
 
 the cases, specimens from the drawers, wall 
 
 cases and the large wall specimens, was done by 
 
 myself and those under my direction. In this 
 
 132 
 
AMONG THE ROCKS. 
 
 work I was assisted by Mr. Broadbent and Mr. 
 Holmes. It gives me great pleasure to record 
 here the valuable services rendered in this work 
 by Mr. Broadbent, a young Englishman, who had 
 just come to this country to seek his fortune, and 
 who, with the Director's permission, I engaged to 
 assist in the removal of the fossil department of 
 the Survey. Mr. Broadbent being a good writer, 
 I appointed him the task of cataloguing the species 
 and formation of every type fossil taken from 
 the cases a work he accomplished most satisfac- 
 torily, indeed, his diligence in the duties as- 
 signed him decided me with the Director's per- 
 mission, to take him to Ottawa to assist in re- 
 organizing the collection. After assisting me 
 for several months, Mr. Broadbent was removed 
 to the mineral department, and was soon after 
 appointed a member of the staff. He now holds 
 the position of museum assistant in the mineral 
 department. 
 
 The work of replacing all the type specimens 
 in the cases, and the thousands of specimens kept 
 in drawers under the cases required much time 
 and patience. Mr. J. B. Tyrrell, so well known 
 now as field geologist and explorer in the distant 
 lands of the North-west Territory, assisted me 
 in part of this work. 
 
 After spending most of the summer months 
 since 1863, in travelling and field work, it was 
 not pleasant to spend the whole of the summer, 
 133 
 
REMINISCENCES 
 
 excepting five days in museum work. The five 
 off days were, at the Director's request, spent in 
 accompanying Mr. Walter Billings, Mr. Ami and 
 Mr. Souter to Paquett's Rapids, on the Ottawa 
 river. It was dark when we arrived, with our 
 camp outfit and a few days' provisions at our 
 destination on the 5th of September. My three 
 companions were students in Palaeontology, and 
 so eager to obtain some of the fine Black River 
 fossils which occur in thePaquette's Rapids rocks, 
 that instead of pitching tent and getting off to 
 bed, they started off with a supply of matches to 
 hunt fossils. To an old collector like myself this 
 was a strange procedure. When they returned I 
 was rolled in my blanket and sleeping. Before 
 my companions were up the following morning I 
 had co lected one of the finest specimens of Stro- 
 matocerium rugosum now in the geological 
 museum. 
 
 Mr. Ami, now Dr. Ami, M.A., etc., etc., is assist- 
 ant palaeontologist to the Survey, and is one of 
 the bright scientists of the Survey staff. I had 
 a jolly time, and returned with renewed energy 
 to deal with the Climactichnites tracks of which 
 I spoke in connection with Lord Dufferin's visit 
 to the Survey museum. 
 
 Unfortunately the director who superintended 
 
 the taking down of this large specimen had not 
 
 calculated its weight, and in lowering it from the 
 
 wall, when at an angle of about 45 it fell with a 
 
 134 
 
AMONG THE ROCKS. 
 
 crash, shattering the plaster casts round the speci- 
 men into fragments and breaking some of the 
 sandstone. The director got a black eye which 
 he afterwards told his friends was not a disreput- 
 able one, and the two men assisting got hurt. It 
 is a wonder it did not go through the floor to the 
 next flat. I was then requested to take it in 
 hand. The fragments were boxed up and shipped 
 to Ottawa, and after three weeks' work, assisted 
 by Mr. Broadbent, a carpenter and a man who 
 made new casts to fill up the frame, we succeeded 
 in getting it placed against the wall once more. 
 We made the best it was possible to make of 
 it, but the specimen, which was one of Sir Wil- 
 liam's pet fossil slabs, is far from looking as well 
 as when in his bed -room in the Montreal museum. 
 
 Before and while the removal of the Survey 
 was in progress, the families of the members of 
 the staff were removed to Ottawa at the Govern- 
 ment's expense. 
 
 After seeing my family settled in the capital, 
 I returned to Montreal, and boarded, strange to 
 say, in the "Logan House," which had been rented 
 by the administrator of the Logan estate, Mr. 
 Grant, to my friends the Austins. The room as- 
 signed me was the one formerly used by Sir Wil- 
 liam as library and sitting room, a delightful 
 room opening by French windows on to the 
 spacious garden. I spent many happy evenings 
 in this old homestead where there were still 
 135 
 
REMINISCENCES 
 
 things left to recall the memory of our dear old 
 chief, and the time when he was ever ready to 
 welcome the humblest members of his staff, and 
 to listen patiently to their troubles. I recall the 
 time, one Sunday in 1865, when oppressed and 
 despondent, I went to tell him I thought of re- 
 turning to England. " Fiddle-de-dee " said Sir 
 William. He then like a father reasoned with 
 me and pointed to a brighter side of my troubles. 
 When I left his house it was with renewed energy 
 to continue my duties at the Survey. Such was 
 the kind influence ever extended to those with 
 whom Sir William had to deal. 
 
 Many other incidents worth record ing occurred 
 during the year of our removal to the capital, 
 but I must hasten on with my travels. 
 
 By the spring of 1882 the geological museum 
 was in fair order and my field work was resumed 
 early in June, by a journey to the Joggins coast, 
 N.S., with a view to increase, by fossils, our 
 knowledge of the Carboniferous formation. 
 
 To the student who wishes to study the geology 
 and palaeontology of the Carboniferous forma- 
 tion there is probably no better opportunity af- 
 forded in the world than the Joggins coast. 
 Here he sees a magnificent range of cliffs extend- 
 ing for miles along a shore washed by the waves 
 of the Bay of Fundy. 
 
 Occasionally in the great red sandstorfe cliff's 
 one sees portions of erect trees, or rather the 
 136 
 
AMONG THE ROCKS. 
 
 casts of portions of large exogen trees, the woody 
 structure of which decayed and disappeared leav- 
 ing the bark standing, to be filled with the sand 
 of the sea or lakes near which they grew. While 
 in a decayed form the base of some of these 
 trees were penetrated by small reptiles who no 
 doubt took shelter there for safety. These little 
 reptiles represent the first vertebrate animals of 
 this world. Some of the sandstones and shales 
 contain the remains of beautiful ferns and trop- 
 ical plants, on the leaves of which we find small 
 shells, representatives of the Mollusk family. 
 Other beds are made up almost entirely of small 
 bivalve shells, which Sir William Dawson called 
 Naiadites carbonaria. 
 
 In the geological museum can be found a fair 
 representation of the Carboniferous fossil fauna 
 and flora of Canada ; but as my last discovery 
 in the Joggins rocks proves there is still much 
 information to be gained with regard to the life 
 of the Carboniferous period. 
 
 The splendid geological work accomplished in 
 the Carboniferous formation of Novia Scotia by 
 Logan and Dawson in the early days of the Sur- 
 vey, and later by other members of the staff, is a 
 most important feature in the geology of Canada, 
 for on the correct interpretation of the Carboni- 
 ferous strata often depend^ vast sums of money. 
 
 Having secured fine sections of the fossil trees, 
 Siyillaria, so named from the seal-like scales on 
 137 
 
REMINISCENCES 
 
 the bark, and specimens among which were 
 scales of fishes, ferns, and other fossil plants 
 work in which I was assisted by Mr. James 
 Devine, a coal miner, I spent a tew hours visit- 
 ing the mines, which are situated half a mile or 
 so from the coast. 
 
 In the superintendent's house I was surprised 
 to see several mounted moose heads (Cerius 
 alres) which my friend the superintendent said 
 he shot at his hunting grounds a few miles from 
 Joggins. 
 
 At this time there was only one main road in 
 the village and another, the coast road, which ran 
 near the brow of the cliff. On these roads are 
 dotted the homes of the miners. The coal is 
 conveyed from the mines by tram-ways and 
 dumped down a chute to schooners and other 
 vessels 
 
 Fifty years ago few people excepting those 
 engaged in mining, or in the manufacture of 
 grindstones were to be seen in the village, and 
 along the Joggins shore. All communication 
 being by stage, buggy, or boat. 
 
 Having packed and shipped my specimens I 
 again mounted the rambling old stage and after a 
 pleasant ride through a verdant country over 
 which refreshing sea breezes are wafted from the 
 Bay of Fundy, I arrived at Maccan Station, and 
 on June 26th, again landed at Campbellton 
 where a few days were spent collecting from the 
 138 
 
AMONG THE ROCKS. 
 
 Devonian rocks on the banks of the Restigouche 
 River. 
 
 Among the interesting specimens secured this 
 time were good examples of the fossil-plant 
 Psilophyton princeps, (Dawson) .Miller says it is 
 the oldest known plant in America, and is sup- 
 posed to have grown in a marsh (psilon, smooth ; 
 phyton, stem.) 
 
 The agglomerates of this vicinity are rich in 
 fish remains, and the conglomerates with their 
 various bright colored pebbles of jasper, sand- 
 stone, quartz, agates, and other rocks, enhance the 
 beauty of this lovely spot where not so very 
 many years ago the red man alone pitched his 
 wigwam, built his birch bark canoe, and speared 
 salmon as he glided down the often swift waters 
 of the Restigouche to one of his wigwams, where 
 now stands the little church of which we have 
 read in previous pages of these notes. Now the 
 whistle of the locomotive echoes among the hills 
 and through the valleys, the white man plies his . 
 axe, Royalty dips its fishing-line in the waters, 
 and the geologist pokes his nose I mean his 
 hammer into the rocks. 
 
 A few days after my return to Ottawa I 
 received instructions to proceed to Sault Ste. 
 Marie, to examine and collect typical specimens 
 of the rocks of that locality, and along the coast 
 of Georgian Bay and part of the north coast of 
 Lake Huron. 
 
 139 
 
REMINISCENCES 
 
 It was then, and is now, an important question 
 as to the exact geological horizon of the red sand- 
 stones, so largely displayed in the construction of 
 the Soo Locks. It is supposed that they belong 
 to the Potsdam formation, but the want of fossils 
 to confirm this question still remains, as my 
 researches revealecfnbthing which could definitely 
 be pronounced organic. It is recorded that one 
 fossil of a Potsdam type was found in some part 
 of these sandstones ; but as no one knows when, 
 where, or who collected it, I think we may con- 
 sider it doubtful. Still I think with Logan that 
 these sandstones are Potsdam, and that fossils 
 will be found to confirm this fact. 
 
 My short stay at the Soo was full of pleas- 
 ant incidents > most of which were due to Mr. 
 Cousins, a prominent Civil Engineer of that place. 
 But our pleasant moonlight boating parties 
 like all other good things, came too soon to an 
 end. 
 
 Leaving Sault Ste. Marie July 22nd, with two 
 half-breed Indians and a boat, a journey was 
 made down that beautiful stream called Garden 
 River the highway for ships passing through 
 the Soo Locks to Lake Superior. Returning to 
 Sault Ste Marie a few more days were spent in 
 the examination of the rocks of that vicinity, 
 after which, providing myself with a horse and 
 cart and an Indian guide, camp outfit and pro- 
 visions, a journey was then made by road to 
 140 
 
AMONG THE ROCKS. 
 
 Echo Lake. All rock exposures were noted and 
 typical specimens collected. 
 
 Arriving at Echo Lake, I was cordially received 
 by those in charge of the copper mines property, 
 which is situated on the north side of the lake, 
 two or three miles from Lake George. The rocks 
 here are chloretic slates, quartzites, conglomerates, 
 &c. They belong to the Huronian series and 
 contain veins of yellow copper ore. 
 
 Although extensive preparations have been 
 made for mining and shipping the ore, I do not 
 think up to the present time any profit has been 
 derived from these mines. 
 
 A short distance from the mines at Limestone 
 Point, a stratified dolometic limestone occurs in 
 large quantities as on the banks of Garden River. 
 This limestone is well adapted for ornamental 
 purposes. The scenery in this vicinity is wild 
 and rugged ; it is a delightful spot for the sports- 
 man. A pike weighing fifteen pounds was taken 
 from the lake by my Indian guide. Returning 
 to Sault Ste. Marie, I engaged two half-breed 
 Indians and their boat and we started on a coast- 
 ing journey down St. Mary's River, part of Lake 
 George, along the north side of St. Joseph's Island, 
 where at Gravel Point I secured a large number 
 of fossils belonging to the Black River formation, 
 and finally arrived at the Bruce Mines on the 
 north side of Lake Huron, where at fifty fathoms 
 from the surface, in 1847-9, 400 tons of copper 
 141 
 
REMINISCENCES 
 
 ore was raised. But the working of these mines 
 was discontinued many years ago. 
 
 At the time of my visit one oi the pioneers 
 if not the discoverer of these mines Col. Rankin 
 was engaged loading a barque with the quartz 
 debris from the stamp mills. He hoped to dis- 
 pose of this broken rock to a Boston firm who 
 were to use it for covering the walks of parks 
 and private residences. There were then many 
 thousands of tons of this material, but whether 
 Col. Rankin made a paying speculation of it or 
 not I do not know. Associated with these cop- 
 per-bearing rocks are great bands of jasper con- 
 glomerates, which form quite a feature in the 
 Huronian district. It is a beautiful rock, remark- 
 able for its bright red jasper pebbles. Masses of 
 detached pieces, some almost as large as a small 
 log cabin were seen in or near the village of the 
 Bruce mines. I shipped a large block of this con- 
 glomerate to the Survey, together with many 
 other specimens characteristic of the Huronian 
 formation, and on the 19th of August, returned 
 to Sault Ste Marie where I again engaged two 
 half breeds, and with a good boat, provisions, 
 &c. we started on a ten days' journey along the 
 North shore of Lake Huron. 
 
 It would be tedious to record all the incidents 
 
 and adventures we met with in this long journey, 
 
 most of which was made on foot with one man, 
 
 while the other followed slowly with our boat, 
 
 142 
 
AMONG THE HOCKS 
 
 landing whenever signals were made for him to 
 do so. Some of our night camps were made on 
 the small islands near the shore. They are com- 
 posed of similar rocks to those of the coast, but 
 in many instances are smoothed by the waters of 
 the lake, affording a good opportunity to study 
 the stratification of the strata. On one of these 
 islands, composed of chloretic slate^ we found 
 although very much water worn, cavities made in 
 taking the rock out with rude instruments, for 
 (as my Indian guide said) making stone pipes. 
 Each of my men carried away pieces for this pur- 
 pose. They said this spot was known to their 
 forefathers, who journeyed many miles to obtain 
 this favorite " pipe stone." 
 
 Arriving at Thessalon Point we ascended the 
 river of that name for twenty miles or so, pass- 
 ing over many of the typical rocks of the 
 Huronian series. Returning we continued along 
 the coast arriving at Algoma Mills Sept. 2nd, 
 where my men were paid off. I saw them and 
 their boat safely on board the steamer bound for 
 Sault Ste. Marie, their home, and after spending 
 a day or two on the large exposures of rocks in 
 that vicinity, boarded the returning steamer for 
 Owen Sound, and thence by rail returned to 
 Ottawa which place I reached Sept. 6th, af cer a 
 journey not free from dangers, and geologically, 
 of less interest than any of the previous journeys 
 I had made. A few days after my return from 
 143 
 
REMINISCENCES 
 
 the above lakes I was again in New Brunswick 
 working up palseontological evidence regarding 
 certain formations then under discussion. 
 
 Returning to Point Levis a few days were spent 
 among the Graptoletic rocks of the " Quebec 
 group," and on the 5th of October, I once more 
 settled down to my winter's work connected 
 with the Survey, which consisted, as usual, in 
 musuem work, mending and restoring a number 
 of fossil bones collected by Dr. G. M. Dawson 
 and Mr. R. G. McConnell. My microscopical 
 work consisted in preparing 300 sections of rocks 
 and fossils. A large number of specimens were 
 developed and prepared for study and for the 
 museum, and the remainder of my time was 
 spent in arranging for my next summer's field 
 work. 
 
 The duty assigned me for the summer of 1883, 
 was to make a geological examination of that 
 portion of the North-west lying between the 
 Cypress Hills, Manitoba, and the Rocky Mount- 
 ains in the vicinity of Kootenay, or Water ton 
 Lake, a distance of about 500 miles. 
 
 As Dr. G. M. Dawson now director of the 
 Survey accompanied by Mr. J. B. Tyrrell, also 
 of the Survey was about to make a Survey of a 
 portion of the North-west Territory this summer, 
 it was thought advisable that I should accom- 
 pany him as far as Fort McLeod, Alberta, near 
 the line of the fifth principal meridian. 
 144 
 
AMONG THE ROCKS. 
 
 I left Ottawa, May 22nd, picked up my son G. 
 H. Weston, then a student at the Agricultural 
 College Guelph, Ont., who was to accompany me 
 as assistant and proceeded to Toronto, there to 
 wait the arrival of Dr. Dawson, who joined us a 
 few days later. We left Toronto, May 27th, and 
 started by rail via Hamilton, Chicago, St. Paul 
 &c., arriving at Winnipeg on the 30th, and Bran- 
 don the same evening, where we purchased part 
 of our travelling outfit and chartered a freight 
 car which for the remainder of our rail journey 
 became a home not only for our horses, waggons, 
 hay, &c., but for Dr. Dawson, Mr. Tyrrell, myself 
 and our assistants. 
 
 The construction of the Canadian Pacific 
 Railway was then in progress, and in the vicinity 
 where we were, the one train running on this 
 section consisted of construction cars and one 
 construction passenger car. We got our car 
 attached to this local train and started towards 
 Swift Current. The nrst town we struck, after 
 leaving Brandon was called Red Jacket. It con- 
 sisted of two large canvas tents, one a general 
 store, and a sign on the other stated that it was 
 " Red Jacket Hotel." 
 
 We reached Maple Creek June 9th, where we 
 found Mr. R. G. McConnell of the Survey, 
 camped a short distance from the Mounted 
 Police Station. This was the end of our railway 
 journey, and here we pitched our tents and com- 
 
 K 145 
 
REMINISCENCES 
 
 pleted our arrangements for the journey across 
 the plains. 
 
 Long before the great Canada Pacific Railway 
 passed through that portion of the North-west 
 Territory known as the Assiniboine district, 
 Maple Creek./ was one of the principal Indian 
 posts. Here Crees, Assiniboines, Sarcees, Bloods, 
 Blackfoots and other tribes brought their furs 
 and other produce and traded with the white 
 men. Although most of the land of this district 
 is now owned by the white settlers it is still the 
 favorite camping ground of the aborigines. 
 
 A mile or so from our camp were congregated 
 a large number of Indians belonging to various 
 tribes. In their tepees, we were told, they were 
 preparing for their annual " Sun-dance." 
 
 Thinking t-his a favorable opportunity to get 
 a few good photographs we caused one of onr 
 vehicles to be hitched up, then Dr. Dawson and 
 I drove across the prairie towards the wigwams. 
 When within half a mile of this camping 
 ground we noticed that close to the tepees were 
 probably a hundred Indian ponies already brid- 
 led. A moment later, as many bucks in all their 
 war paint and feathers issued from their 
 tepees, hastily mounted their horses, formed a 
 line and started at full gallop in a direct line for 
 us. On they came till within a hundred paces 
 or so of us. For a moment I sat paralyzed, 
 expecting to be trampled under their horses' 
 146 
 
AMONG THE ROCKS. 
 
 feet, but when within thirty or forty feet of us 
 they opened out in the centre allowing room for 
 us to pass, and with a rush and the Indian salu- 
 tation " How," they were soon far away. It 
 was a beautiful sight, one not soon forgotten. 
 Arriving at the Indian camp, Dr. Dawson 
 attempted to get a photograph of some of the 
 squaws and young Indians, but the moment we 
 got the camera up, away they scooted into their 
 tepees. However I seized the hand of an almost 
 nude youngster, and he and I form part of a 
 pretty picture (?) 
 
 Both inside and outside the general stores of 
 Maple Creek at this time, 1883, were favorite 
 lounging places for the Indians. Many of the 
 aborigines were quite picturesque looking, adher- 
 ing strictly to the Indian costumes of their 
 fathers, while others adopted any of the white 
 man's cast off clothing they could find. The 
 respect accorded an Indian appeared to depend 
 on how many ponies he possessed and his ability 
 to support himself without the aid of the white 
 man. In this district there appeared to be a 
 great many Indians who absolutely refused to 
 remain on the Government reserves, therefore 
 obliging themselves to pick up a precarious liv- 
 ing, one method being the collecting of buffalo 
 bones for which so much per hundred pounds 
 weight was given by men commissioned to receive 
 and ship them to various towns to be used in 
 147 
 
REMINISCENCES 
 
 refining sugar, fertilizing, etc. It no doubt seems 
 hard to those who a few years ago hunted the 
 buffalo in all its wild grandeur to be gathering 
 and selling for the small sum of money or provi- 
 sions the bones which in their happy hunting 
 days they left to bleach on the plains. 
 
 By various signs I talked to an old Indian 
 whose photograph now adorns a wall in the 
 ethnological room of the geological museum of 
 the glorious days of the buffalo hunt could one 
 have photographed the various expressions of 
 this aged buck's face and the beautiful way he 
 illustrated by signs and gestures the buffalo 
 chase, and later the total extinction of this noble 
 monarch of the plains it would have been a 
 picture indeed. 
 
 We had completed our arrangements and pro- 
 posed leaving the following morning, June 6th. 
 Our horses, eight in number, two heavy waggons, 
 one buggy, and one buck-board were all arranged 
 round our tents. It was late when we left our 
 camp fires and tumbled in for the night. At 6 
 o'clock next morning Thompson one of Daw- 
 son's men reported that three of our horses 
 were missing. We had been warned by the 
 mounted police that Indians were stealing horses 
 in all directions. This made us anxious and 
 men were at once sent in various directions to 
 look them up. It was dusk in the evening when 
 my man Haultain returned with our horses, 
 148 
 
AMONG THE EOCKS. 
 
 which he said he found among a number of 
 Indian ponies miles away. Dr. Dawson had 
 brought with him a branding iron, a broad arrow, 
 the government brand mark, and each horse was 
 at once branded. A sharp look out was kept 
 during the night and the following morning 
 after an early breakfast and a good-by to our 
 friends of Maple Creek we started on our journey 
 towards the Cypress Hills. Our mode of pro- 
 cedure was : Dr. Dawson and his assistant, Tyr- 
 rell, in a buggy, Johnson on horse back, who 
 chose the best road for our waggons, then followed 
 their supply waggons. Next came my equip- 
 ment waggon, and my son with a buck-board, 
 and lastly myself on horseback. 
 
 In five days we had reached and crossed a 
 portion of the Cypress Hill and were camped in 
 a valley about 150 miles from Fort McLeod. 
 The discovery this year, 1883, by Mr. R. G. Mc- 
 Connell of the Geological Survey, and later by 
 myself, of mammalian and other fossil remains, 
 had rendered the Cypress Hills' district a most 
 interesting localities for the geologist and osteo- 
 logist. But I shall speak of the rocks and fossils 
 of the hills and valleys of the Cypress Hills in 
 my notes of next summer's travels. 
 
 The 12th of June found us in the vicinity of 
 
 Pagon Creek on the Fort Walsh trail. Here we 
 
 gathered as many mushrooms as we wanted. The 
 
 scenery here is very fine. By compass we had 
 
 149 
 
REMINISCENCES 
 
 travelled over hills and through valleys where 
 we had found both wood and water, but now we 
 are about to journey for a few days through a 
 district where wood is more scarce than the 
 "Prairie dog," Spermopoilus Ludovicianus, and 
 our men are filling every available space in the 
 waggons with food for our camp stoves. 
 
 As we travel these plains I can't help thinking 
 of the time not many years ago when vast 
 herds of buffalo followed each other, in single 
 file, along the narrow trails we so often cross, 
 which if we follow, we will surely come to 
 water. Then the Indian was perfectly inde- 
 pendent of the white man, for the buffalo fur- 
 nished him his clothing, his tents to live in, and 
 meat to eat. But all is changed now, bark and 
 dirty rags form their wigwams, and those who 
 will not remain on the reserves, furnished them 
 by government, must eke out a living as best 
 they can. The small cash treaty annuities (which 
 I think average from $3 to $25 according to age 
 and rank)is frequently spent the day it is received, 
 generally on a few luxuries for the buck, not for 
 his wife and children. 
 
 While passing through the beautiful valleys of 
 the Cypress Hills, and over the Sweet Grass Hills, 
 we followed the advice of the Mounted Police 
 and kept a sharp look out for Indian or other 
 horse thieves, by having our men take turns in 
 mounting guard during the night. The horses 
 150 
 
AMONG THE ROCKS. 
 
 were picketed near our camps and each man, well 
 armed, took a two hours' watch. 
 
 Many mishaps incidental to travelling the 
 prairies were met with a break down while 
 crossing a soft -bottom stream, an axle-tree broken, 
 through a wheel dumping in a badger hole and 
 so on. But the beautiful scenery and the impor- 
 tant geological facts gleaned as we went along amply 
 repaid us for the cares of the day, and at night 
 when safely housed in our tents with the spoils 
 of our journey consisting sometimes of Indian 
 skulls, fossils, &c., new food for the osteologist 
 and palaeontologist, we felt that we had done 
 some little for our " Queen and country ". The 
 chief incidents of interest during the next few 
 days were the discoveries of small seams of coal 
 (lignite) ; rocks with beds of fossil oyster shells, 
 small herds of antelope ; water which when sep- 
 arated from thousands of small red insects by 
 straining through a handkerchief, was not so bad, 
 no wood and a scarcity of buffalo chips (dung) 
 for our camp stoves, a constant look out for red- 
 skin horse thieves, and many other things which 
 tended to impress this section of the Territories 
 on one's memory. 
 
 Passing through Verdigris Coulee and along 
 the margin of Suds Lake we arrived at Milk 
 River Ridge in the vicinity of which a remark- 
 able fossil fauna was found. It is probable that 
 fifty different species were obtained from a patch 
 151 
 
KEMINISCENCES 
 
 of rocks a couple of yards square. We named 
 this spot Fossil Coulee. It can be found in Daw- 
 son's map of this district. This unique collection 
 together with all specimens obtained during our 
 long journey from Maple Creek, now lies at the 
 bottom of Lake Superior owing to the sinking of 
 the ship Glenfinlas. It is fourteen years since 
 this occurred, but I still mourn the loss of these 
 hard-earned specimens which were shipped by 
 bull-team from Fort McLeod to Maple Creek. 
 
 From the top of Milk Ridge we got our first 
 view of the Rocky Mountains. On the 25th, of 
 June we struck the Benton trail and saw one of 
 the interesting sights of prairie travel, a train of 
 twelve great schooner waggons divided into four 
 sections, each section having six teams of mules 
 It is worth a long ride to see one of these cara- 
 vans cross a soft-bottom stream, especially if any 
 of the harness breaks and the mules get mixed 
 up then one hears language which well I cer- 
 tainly should not like to have a member of the 
 fair sex with me at that time. The mule driver 
 is swift in overcoming difficulties, but though his 
 poor dumb companions may strain every nerve 
 to straighten things out again, they never go 
 unpunished for any mishap that may occur, and 
 at such a time it is well for the tenderfoot not to 
 interfere. 
 
 Passing Ed. Mahan's coulee we camped for the 
 night at tifteen mile lake, within a short distance 
 152 
 
AMONG THE ROCKS. 
 
 of a small encampment of Blood Indians who 
 were not long in paying us a friendly visit, 
 shaking hands all round and saying " How " 
 We presented each with a slice of pork and while 
 Dr. Dawson was fixing his camera to get one of 
 his historical pictures I presented each one with 
 half a plug of black tobacco, but at the words " all 
 right, steady," from Dawson, off they went mak- 
 ing a bee-line for their wigwams. 
 
 June 27th, we reached Belly Valley through 
 which the swift and often turbid Belly River 
 courses. To our right are drift deposits forming 
 high steep bluffs under which are extensive beds 
 of coal which were being worked. The opposite 
 side of the river is Coal Banks, so named from 
 the coal deposits. The river here is said to be 
 437 feet wide ; the prairie level 300 feet above 
 the stream, and the scarped banks which pre- 
 sent beautiful sections of these stratified rocks 
 200 feet high. The rocks of the Coal Bank dis- 
 trict belong to that portion of the Cretaceous 
 formation known as Pierre shales, Belly River 
 series, &c., some portions of which are almost 
 destitute of fossils, while other beds are exceed- 
 ingly prolific in organic remains. 
 
 After days of travel over woodless plains, 
 it is exceedingly pleasant to strike a verdant 
 spot like Belly Valley where wood is plenti- 
 ful and the blossoms of the thousands of wild 
 roses perfume the air but, tread carefully, 
 153 
 
REMINISCENCES 
 
 for the deadly rattlesnake lurks among the 
 plants. 
 
 One would have liked to linger many days in 
 this charming spot, but having secured some 
 baking powder, a few other luxuries, and two 
 or three good photographs, we crossed the stream 
 on the Coal Banks ferry for which service we 
 paid $4.75. 
 
 The following morning we left Coal Banks, 
 climbed the steep hill and were again on the 
 prairie level ; continuing our journey we passed 
 Rye Grass flat, of which I shall speak later on, 
 and camped for the night on the banks of the Old 
 Man River where a large number of Indians had 
 pitched their tepees, and appeared to be holding a 
 pow-wow. 
 
 The following morning, after a journey of 
 eight miles over an excellent trail, we arrived at 
 Fort McLeod, one of the important trading posts 
 of the Alberta district. Here all our precious 
 rocks and fossils which I have already stated 
 now lie at the bottom of Lake Superior were 
 repacked and left with J. G. Baker & Co. for 
 shipment by the first bull or mule train which 
 left for Maple Creek. 
 
 Lieut. Gov. Dewdney arrived at this time and 
 was given a good reception by the few white 
 men of the village and a dozen or so of Indian 
 bucks who came from well I don't know where, 
 but I do know that among them seated astride 
 154 
 
AMONG THE ROCKS, 
 
 on an Indian pony was a fair(?) daughter of the 
 plains who cast sheep's eyes at our worthy Lieut. 
 Governor. " Nothing venture, nothing have " 
 was quoted by one of our men. 
 
 How shall I describe this place where dirt pre- 
 vaileth and righteousness enter eth not within 
 her gates well perhaps it is better to leave that 
 to a more prolific pen than mine. 
 
 Sunday, July 1st, and the two following days 
 we had a busy time getting horses shod, waggons 
 mended, laying in provisions, &e., preparatory to 
 our journey to Kootenay Lake, about 150 miles 
 west. 
 
 So far Dr. Dawson and his assistant Mr. J. B. 
 Tyrrell, had been our guides and companions, 
 but on July 3rd, they left us and started across 
 the plains in an opposite direction to the one we 
 had to take. 
 
 Any one who has travelled with Dr. G. M. 
 Dawson will I feel sure look back with pleasure 
 to their journey with that most courteous gentle- 
 man whose faculty for overcoming difficulties, 
 and great knowledge, has won him the high 
 position of Director of the Geological Survey of 
 Canada. 
 
 We watched the doctor's outfit till it was lost 
 in the distance and a few hours afterwards the 
 " Weston outfit " was heading for the Rocky 
 Mountains. It consisted of one heavy waggon 
 drawn by two fine strong chestnut horses 
 155 
 
REMINISCENCES 
 
 " Baby, and Dick ;" a buck-board with a yellow 
 horse, " Buck," and my saddle-horse " Pink eye." 
 Haultain our driver and my son G. H. West/on, 
 camp equipment &c., for a journey of six or eight 
 hundred miles over the plains. As on these 
 journeys one's life often depends on one's horses, 
 it is not surprising that they are well looked 
 after and that their various peculiarities are 
 noted. Continuing our journey, sometimes by 
 trail and sometimes by compass, we reached 
 Pincher Creek, an important fossil locality, 
 July 9th. 
 
 After leaving Fort McLeod, we journeyed over 
 hills and through valleys, forded rivers and 
 creeks. The high cut banks of the rivers in this 
 locality often render it necessary to travel long 
 distances before a crossing can be found, and 
 then to cross these swift streams required a steady 
 head. On one occasion Haultain had got his 
 waggon safely over a broad swift river, and my 
 assistant George on his buck -board was anxiously 
 watching Buck's legs to see if they moved, but 
 turning for a second to see how I was getting 
 along, and seeing that my horse was turning 
 round, shouted, <l for God's sake let go your reins ;" 
 I did so, got a tight grip on the saddle, closed my 
 eyes, and in a few moments felt my horse 
 scramble out on the bank of the river ; the rush- 
 ing water had made me giddy and I was keep- 
 ing a tight rein on my horse, but the moment 
 156 
 
AMONG THE ROCKS. 
 
 Pink Eye felt his head free he made for his com- 
 panions on the shore. Since this incident I have 
 often trusted to my horse, feeling sure his brute 
 instinct would overcome what to me were grave 
 difficulties. Halting for our noon meal about 
 fifty miles from Port McLeod we discovered that 
 one of our dunnage bags was missing. Haultain 
 was just about to return on our trail to look for 
 it when in the distance we saw four wild looking 
 Indians mounted on bare-backed ponies. One 
 horse carried our dunnage bag tightly lashed on 
 his back. The day was very hot, and for the 
 journey the Indians had divested themselves of 
 almost all their clothing displaying a good por- 
 tion of their forms. By a diagram drawn in the 
 sand they reported that the bag had been found 
 in one corner of the Indian reserve, and by signs 
 demanded $2 each. I showed them the govern- 
 ment brand on our horses ; we made them under- 
 stand that the bag was the property of the gov- 
 ernment and that fifty cents each was the price, 
 of the work they had done ; after a little parley 
 with each other they decided to accept our terms, 
 shook hands, said " How," mounted their horses 
 and at full gallop started back for the Blood 
 Indian Reserve. 
 
 Reaching Pincher Creek, July 9th, we pitched 
 
 our tents in a picturesque locality on the bank of 
 
 the creek, a mile west of the North West Mounted 
 
 Police farm. Near us were camped two families 
 
 157 
 
REMINISCENCES 
 
 of West Percy Indians. The bucks were fine 
 strong dark skinned fellows who during our two 
 or three days' stay in this locality became very 
 friendly, especially a stout well formed youngster 
 who made his initial visit to our camp in a per- 
 fect state of nudity. 
 
 At Col. McLeod's house, a mile or so from our 
 camp, we found our mail from Ottawa. It had 
 been twenty-two days on the road, but was very 
 acceptable. 
 
 The geology of Pincher Creek is very interest- 
 ing, winding its way, as it does, through the Foot 
 Hills of the Rocky Mountains, the cut banks of 
 the creek afford an excellent opportunity to study 
 the various beds of rock which constitute the 
 Laramie or Eocene of Canada either the base of 
 the Tertiary or the upper beds of the Cretaceous. 
 Some of the rocks here are very prolific in fossils. 
 One cliff perhaps 200 feet high holds a thick bed 
 of limestone partly made up of well preserved 
 gasteropods and other shells. The beautiful 
 spiral shell Physa Copei described by Mr. Whit- 
 eaves in " Contributions to Canadian Palaeonto- 
 logy/' Vol. 1, Part 1, is very abundant. Some 
 of the shaly beds hold fossil plants. Red, 
 purple and grey clays overlie the fossiliferous 
 rocks. 
 
 We reached Mill Creek another important 
 fossil locality July 13th, while a big hail storm 
 prevailed. As at Pincher Creek, we succeeded in 
 158 
 
AMONG THE ROCKS. 
 
 getting several good geological photographs of 
 these localities. 
 
 The rocks, which are well displayed here, form 
 a portion of the Upper Cretaceous formation. 
 They have been divided into Mill Creek and 
 Niobrara beds. The latter beds hold the well 
 known bivalve Inoceramus problematicus, and 
 the former beds, many beautiful shells, some of 
 which have been described by Mr. Whiteaves. 
 
 Among the fossil flora of this locality are sev- 
 eral species of ferns, and leaves of exogen plants. 
 Some of these have been described and figured 
 by Sir J. W. Dawson in Transactions of the Royal 
 Society of Canada. 
 
 The Mill Creek sandstones of this vicinity are 
 represented by high clitfs. In these we saw casts 
 of large trees relics of that epoch of this world's 
 history when great carnivorous reptiles peopled 
 a large portion of America a time when croco- 
 diles tilled the waters and swarmed over the 
 banks of rivers which now flow past the city of 
 Philadelphia. 
 
 Our three days' sojourn in Mill Creek val- 
 ley was a most pleasant one surrounded by 
 green pastures on which the finest cattle of the 
 North-west roam. We had good trout fishing at 
 the South Fork of the Old Man River ; a grand 
 view of the snow-capped Rocky Mountains ; civ- 
 ilization in the way of a blacksmith's forge, car- 
 penter shop, Gladstone's House, a boarding place, 
 159 
 
REMIJNISCENCES 
 
 mill, white men, half breeds, Indians, and abund- 
 ant food for the mind of the geologist and 
 palaeontologist. 
 
 It was not without regret that on the 16th of 
 July, we pulled up stakes and continued our 
 journey towards Kootenay Lake. 
 
 Following the trail Canada's Governor General, 
 Lord Lome, travelled over in J882, and the 
 Kootenay River, after many difficulties fording 
 streams and creeks, in which a broken axle of our 
 waggon figured conspicuously, as it obliged two 
 of us to return to Mill Creek ; thence on horse- 
 back over rivers and hills to Garnet's farm where 
 Dr. Dawson had left a makeshift axle which his 
 men made from a stick of hard wood then carried 
 swung under his waggon. This my son George 
 brought down the hills, and over rivers and 
 creeks, on horseback no small task ; which it 
 gives me pleasure to record. 
 
 After a good night's rest at the creek, we 
 returned with our precious stick of maple and 
 some telegraph wire, to obtain which George and 
 I had driven eighty miles, and with a broken 
 shafted buckboard had twice forded the swift 
 waters of Kootenay River. But at last we were 
 again at the scene of our wrecked waggon, where 
 Haultain had been guarding our provisions from 
 prairie wolves and prowling Indians. 
 
 At daylight the following morning the wheels 
 of our waggon were firmly lashed on our ama- 
 160 
 
AMONG THE ROCKS. 
 
 teur axle, a new shaft for our buck-board was 
 roughly made from a stick of cotton-wood cut in 
 a small bush miles away, and other repairs were 
 attended to. Our present difficulties being over, 
 we resumed our journey and in two days camped 
 in the vicinity of Kootenay Brown's shanty at 
 the south end of Waterton (Kootenay) Lake, a 
 magnificent sheet of water nestling at the foot of 
 grand, rugged, picturesque mountains, some of 
 which have an altitude of 4000 feet above the 
 lake. 
 
 I did not see any fish from the lake, but from 
 the home-made fishing tackle seen in Brown's 
 house I should fancy some of the largest fish 
 found in our Canadian Lakes inhabit these waters. 
 Brown appeared to be an educated American, 
 living with an Indian woman. Why he chose to 
 become a squatter in this lonely spot, he refused 
 to reveal to us. The heads and skins of the 
 grizzly bear, mountain sheep and other wild ani- 
 mals which adorned his humble dwelling proved 
 that he was no mean hunter. 
 
 With him as our guide I visited the " Cascade," 
 a beautiful waterfall on the western shore of the 
 lake. Many of the characteristic rocks of the 
 mountains are well displayed here, and I hoped 
 to obtain fossils from some of the beds, but my 
 researches were fruitless. 
 
 The following day Brown was engaged to 
 accompany my son and me through a portion of 
 
 L 161 
 
REMINISCENCES 
 
 the South Kootenay Pass a lovely valley through 
 which a small stream of water courses. It was 
 here while my horse was picking his way over a 
 rough trail at the foot of high cliffs that Brown 
 shouted " look out for the grizzly ". Looking 
 across the valley we saw a full grown grizzly 
 slowly piloting her cub down the side of a pine 
 clad mountain. Being unwilling to take any 
 risks with her ladyship I moved off as quickly as 
 possible, but Brown who was fully prepared for 
 any emergency begged me to allow him to follow 
 up his big game. When George and I were half 
 a mile or so away, we heard the report of his 
 rifle, but later on when he overtook us he repor- 
 ted that just as he fired his horse baulked, and as 
 the bear charged at him, and his horse not being 
 accustomed to that work, he thought it best to 
 retreat. He said however, that he should renew 
 the chase next day as he had an order for a 
 grizzly cub, and if he could despatch the mother 
 he could readily bag the young one. 
 
 The sun had set, when after a horse-back ride 
 of forty miles or so we returned to camp. We 
 had followed the narrow trail which for hundreds 
 of years had been traversed by the Kootenay and 
 other tribes of Indians when going to their buffalo 
 hunts. 
 
 The n agnificent ranges of snow capped moun- 
 tains the home of the mountain sheep and goat 
 which tower thousands of feet above the thickly 
 162 
 
AMONG THE ROCKS. 
 
 forested sides of portions of the valley, the home 
 of the grizzly bear, render the Kootenay Pass a 
 most interesting spot for both the artist and the 
 sportsman. 
 
 In wishing us good -night, Brown said " Keep 
 up a good fire, boys, for there are more than two 
 grizzlies round these diggings." 
 
 The geological horizon of the Waterton (Koote- 
 nay) Lake rocks, owing to the absence of fossils, 
 has not been determined definitely, but in Dr. 
 Dawsoii's survey of this district, he divided them 
 into several zones Cambrian, Carboniferous, 
 Devonian, and probably Triassic. 
 
 Among the great variety of rocks which form 
 the magnificent elevations of the Waterton Lake 
 and South Kootenay Pass district one of which 
 reaches a height of 10,535 feet above the sea 
 are magnesian limestones, slates and shales, 
 quartzites, amygdaloidal traps, etc., rocks of 
 every tint ; bat fawn, bright red, yellowish and 
 brown colors predominate. Sheep Mountain, one 
 of the Kootenay Lake ranges which was measured 
 by Mr. R. G. McConnell, of the Survey, who has 
 done so much valuable exploration work in the 
 Northwest Territories, has an elevation of 1,955 
 feet above the lake. It is composed of sandstones 
 of various tints, red and green shales, jasper con- 
 glomerates, limestones, etc., some of the sand- 
 stones are ripple-marked. The south Kootenay 
 rocks represent a thickness of about 4,100 feet, 
 163 
 
REMINISCENCES 
 
 but in all this series of strata no fossils have yet 
 been discovered. 
 
 Our geological work being finished, as far as 
 time would allow, in this district, we struck 
 camp July 25, and the following day forded the 
 Kootenay River. Heavy rain and the melting 
 of snow on the mountains had swollen the river 
 so much that crossing became difficult my horse 
 lost his footing and swam for it, landing half a 
 mile down the river. George thought Buck 
 towed his light vehicle over, Haultain's waggon 
 being heavy, did not float. 
 
 By July 27th, we were again at Fort McLeod, 
 where we got a new axle for our waggon and a 
 shaft for our buck -board. 
 
 While camped near the village waiting for re- 
 pairs to our outfit, etc., we saw a good bit of the 
 Indian life of this district. The Blood Indian 
 reserve being only a short distance away, Indians 
 were constantly travelling between the reserve 
 and the village. Although associating with the 
 white man, a great number of the Indians of this 
 district still adhere to the paint and feathers of 
 their forefathers. Many of the young Indian 
 women of this place appear to be as vain as 
 some of their white sisters in their adornments. 
 But this is a mounted police settlement! 
 
 From sunset till the small hours in the morn- 
 ing we heard the monotonous turn, turn, turn, on 
 the Indian drum, which seemed to accompany all 
 164 
 
AMONG THE ROCKS. 
 
 gambling, the nightly pastime of the red man of 
 this village. 
 
 In the corner on the mud floor of one of the 
 Indian shacks lay a mother and her babe sleep- 
 ing, while close by squatted two " bucks " deep 
 in an Indian game of chance, One of the men 
 was stripped almost to his breech-cloth. He had 
 staked his blanket and lost; staked the few mis- 
 erable household possessions, and lost ; and still 
 the play went on, and the half clad Indian boy 
 sat in the corner and beat his turn, turn. 
 
 July 30th, we left Fort McLeod, and camped 
 for the night at Willow Creek, and the following 
 day reached Rye Grass Flat, Old Man River, a 
 locality of much geological interest. Here thick 
 beds of yellowish white limestone are crowded 
 with well preserved fossil bivalve shells, which 
 belong to the Lamellibranchiata family. Some 
 of these forms have been described and figured 
 by Mr. Whiteaves ; Contributions to Canadian 
 Palaeontology, Vol. 1, Part 1. These fossils which 
 are well represented in the cases of the Geologi- 
 cal Museum belong to the Laramie, which forms 
 the base of our Canadian Tertiary formation. 
 
 Leaving Rye Grass Flat so called from grass 
 of that name which is characteristic of this val- 
 ley, Aug. 3rd, we proceeded by compass across 
 the prairie level to Scabby Butte, where 'we 
 expected to camp for several days, but on our 
 165 
 
REMINISCENCES 
 
 arrival we found the small lake of that locality 
 dried up, and after vain attempts to get water, 
 even enough to make a cup of tea, we decided to 
 hasten back to our old camping ground, the only 
 place for miles where we could get water. 
 
 A few days later we were again at Coal Banks, 
 following the banks of the Belly River. High 
 scarped yellowish sandstones form an interesting 
 feature of this locality. Besides holding many 
 important fossil shells, we found several bones of 
 a great Dinosaurian reptile, probably the Lee- 
 laps incrassalus (Cope). The prairie level here 
 is probably 200 feet above the river. In one or 
 more instances we found much difficulty in get- 
 ting our horses down to the water, and several 
 times had to carry water for camp use half a 
 mile or more up these steep banks. 
 
 Aug. 12th, we were following the banks on the 
 north side of the St. Mary River, which enters 
 the Belly River a few miles above Coal Banks. 
 It has its source in the mountains south of the 
 49th parallel, is exceedingly rapid, and passes 
 through a beautiful verdant valley, in places 200 
 or 300 feet below the prairie level. 
 
 The rocks which belong to the Upper Cretac- 
 eous and form a portion of the Pierre and Fox 
 Hill formations, contain many interesting fossils, 
 but as yet, considering the extent of these forma- 
 tions, very little has been done regarding the 
 palaeontology of these great exposures. 
 166 
 
AMONG THE ROCKS. 
 
 I must pass over our wanderings for the next 
 few days and take the reader of these notes once 
 more to our old camping ground at Rye Grass 
 Flat, in the valley of Old Man River. We were 
 returning DO Fort McLeod and remained here to 
 pay another visit to Scabby Butte. It was the 
 15th of August. The water in all the rivers was 
 low and the ponds and swamps dried up. Fear- 
 ing we would again find the Scabby Butte district 
 destitute of water, I decided, with George, to 
 make the journey with one horse and a buck- 
 board. So with a large tin pail of water, for our 
 horse Buck, and a small keg of water for our- 
 selves slung under the buck-board, we started at 
 7.30 am., and following our old track reached 
 the Buttes in about three hours. Scabby Butte 
 is a remarkable place, reminding one of the exca- 
 vations of an ancient city. The sands, clays, 
 lignite shales, lignite coal, sandstone and other 
 rocks, have been scooped out, as it were, from 
 the prairie land, to a depth of fifty or sixty feet, 
 leaving portions of stratified rocks standing up 
 like the walls of a burned city. There is evi- 
 dence that at one time this was a great rendez- 
 vous and hiding place for Indians during their 
 buffalo hunts and probably during their time of 
 war. The geological horizon is Laramie, sup- 
 posed Lower Tertiary, or Eocene. Some of the 
 same fossil shells found at Rye Grass Flat, and 
 other localities mentioned, occur here, but the 
 167 
 
REMINISCENCES 
 
 chief interest of Scabby Butte lies in its fossil 
 bones. It was the home of the great Dinosaurian 
 reptiles. 
 
 . . . Monsters of the prime, 
 Who tare each other in their slime. 
 
 Mr. R G. McConnell, of whom I have several 
 times spoken, was the first to discover, in 1881, 
 fossil bones at this locality. In the year men- 
 tioned he dug from a soft sandstone cliff, a 
 femur or thigh bone of one of these extinct 
 kangaroo-like reptiles. This bone is about the 
 size of the thigh bone of a mastodon. It fell 
 to my lot to restore it, for it was in many frag- 
 ments. 
 
 Another interesting feature of Scabby Butte 
 is the fossil wood found in the sandstones and 
 other rocks. Many feet below the prairie level 
 we saw the stumps of exogenous trees trees 
 which grew in the silent ages of this world 
 silent but for the struggles of fierce reptiles, 
 whose sharp cutting teeth show that they were 
 flesh eating animals ; and for the sound of rest- 
 less waters which have left us as souvenirs ex- 
 amples of ripplemar kings. Here, too, these large 
 trees spread their branches over delicate ferns 
 and grasses which have left their impressions in 
 the sandstones and shales. The fossil wood of 
 this, and many other places in the North-west, 
 is of much palyeontological value when it is 
 found in place, as it grew. Being silicified these 
 168 
 
AMONG THE ROCKS. 
 
 woods stone woods they are frequently called 
 retain their cellular structure. This enables the 
 palaeontologist to determine their species. 
 
 Sir J. W. Dawson has described in Trans, of 
 Royal Soc. of Canada, from sections made by the 
 writer, a number of these woods. They show 
 that the flora of those days contained, poplars, 
 pines, oaks, and other trees which grow at the 
 present time. 
 
 One of the specimens of fossil wood we 
 brought from Scabby Butte is about five feet 
 long and a foot wide. It shows it has been very 
 much flattened by the pressure of overlying beds, 
 rocks of great thickness which must have over- 
 lain the present beds and been denuded before 
 the time of the present prairie level. 
 
 The sun was getting low when Buck who 
 had refused to drink the water we had taken so 
 much trouble to carry for him was released 
 from his picket rope, and in a short time we 
 were retracing our trail over this trackless por- 
 tion of the prairies. The only animal life we 
 saw while on our Scabby Butte journey was a 
 small herd (six or eight) of antelope and some 
 blackbirds, one of which for a more elevated 
 position perched on Buck's head, and on this 
 perch Mr. blackbird got a free ride for several 
 miles flying off at times, then returning. But 
 this is not an unusual sight on the prairies. We 
 reached camp in the gloaming, where we found 
 169 
 
REMINISCENCES 
 
 Haultain prepared to give us a good supper of 
 prairie chicken and pork. 
 
 Continuing our journey from Rye Grass Flat, 
 past Coal Banks and through a part of Chin 
 Coulee over a portion of the country where 
 long distances have to be made between drink- 
 able water we reached the South Saskatchewan 
 River, a little below the confluence of the Bow 
 and Belly Rivers, August 23rd. The valley here, 
 which is 250 feet deep, exposes a fine section of 
 Cretaceous rocks. Many of the beds are prolific 
 in fossils especially the genus Ostra. Some of 
 these oyster beds are very interesting as they 
 contain fragments of bones which probably be- 
 long to the Dinosaurian reptile, Lcelaps. 
 
 The current of the river here is swift, and the 
 water, which is beautifully clear, of an aqua- 
 marine color. 
 
 We pitched our camp on a flat a little below 
 the Forks of the Bow and Belly rivers. The 
 vegetation was chiefly wild sage, grass, and sun- 
 flower plants, which latter at this time where in 
 full bloom, looking from a distance, like a cloth 
 of gold. These plants formed an excellent covert 
 for prairie chickens which were quite numerous. 
 
 Before leaving this interesting but lonely spot, 
 George and I visited an Indian cairn a few 
 miles below the Forks of the Bow and Belly 
 rivers and while returning to camp in the even- 
 ing saw, on the opposite side of the banks 
 170 
 
AMONG THE ROCKS. 
 
 of the river, a man on horseback, making signs 
 which indicated that he was lost and hungry. 
 We signalled him to follow the banks of the 
 river towards our camp, but he evidently did not 
 mean to lose sight of us for he dismounted and 
 at once commenced to lead his horse down the 
 250 feet of shelving rocks a feat one might 
 suppose only a goat could accomplish. We 
 watched them with almost abated breath expect- 
 ing every moment to see both horse and man roll 
 to the foot of the cliff. But no, they landed safe 
 at the water's edge where on a small grassy flat 
 the man picketed his horse and then followed 
 the stream keeping a sharp eye on us as we 
 followed the banks two or three hundred feet 
 above. 
 
 In due time we reached camp, but how to get 
 our man over a river too deep to wade and too 
 swift to swim we did not know, till Haultain said 
 he had seen up the river a roughly made weather- 
 beaten boat left there probably by surveyors, 
 years ago. This we decided to try and use, and 
 as there was no stick of wood we could employ 
 as a paddle down came our tent, and with the 
 poles for paddles, and partly stripped for any 
 emergency, George and Haultain started across 
 the stream, but before they were half over the 
 current almost swamped their leaky boat. How- 
 ever they landed safely, and with the lost and 
 hungry man, and a short slab of wood they had 
 171 
 
REMINISCENCES 
 
 found where they landed, to assist, they crossed 
 the river again, landing a mile below our camp. 
 
 We found that our visitor was a butcher from 
 Medicine Hat, who had been hunting stray cat- 
 tle, and had been without food for three days. 
 Our supper, which was already waiting us, con- 
 sisted of boiled prairie chickens, pork, beans and 
 hard-tack biscuit, of which Mr. Butcher ate till 
 we were obliged to stop the supply as we feared 
 he might " overdo the thing " as Haultain said. 
 
 In a short time the stranger was fast asleep 
 and next morning after a good breakfast, and 
 with three days' provisions and a sketch of the 
 river he was to follow, he crossed the stream 
 again. The winding of the shore prevented us 
 from seeing where he landed or how his horse 
 and he succeeded in reaching the prairie level 
 again. 
 
 This man whose name was George Spindle 
 stated that within five miles of our camp Indians 
 were hunting buffalo and two days ago he had 
 seen more than one, and would have chased them 
 but for the want of cartridges for his rifle 
 which we supplied him with before he left us. 
 I am doubtful however about his having seen 
 buffalo, as we had seen no fresh tracks of them 
 on all our journey. 
 
 Leaving the Forks of the Bow and Belly 
 rivers we continued our journey along near the 
 top of the scarped banks of the South Saskatch- 
 172 
 
AMONG THE ROCKS. 
 
 ewan which in places here is 100 feet wide and 
 runs through a valley 250 or 300 feet deep. 
 
 A prairie fire had lately swept over this 
 portion of the country leaving 15 or 20 miles of 
 land destitute of grass. We descended the banks 
 of the river when an opportunity afforded, and 
 while our horses fed, made our geological exami- 
 nations; always finding interesting palaeozoic 
 objects now the scale of a fish, which must 
 have been of great size ; now fragments of bone 
 belonging probably to our great Dinosaurian 
 reptile, now fossil shells belonging to the oyster 
 family, and in the shaly beds, fragments of 
 grasses and leaves relics of the Cretaceous 
 epoch of our world's history. 
 
 We reached Medicine Hat, Aug. 28th. " The 
 Hat " is one of the live prairie towns of the 
 Northwest an important station on the Cana- 
 dian Pacific Railway which crosses the South 
 Saskatchewan River a little below. At this 
 time (1883) most of the shops and dwellings 
 were of wood. Besides twenty or more merchants 
 there were two doctors, two barristers and 
 solicitors, four hotels, and several boarding 
 houses. 
 
 A large flat in front of the " town " afforded 
 an excellent camping ground for Indians whose 
 tepees numbered a hundred or so. While some of 
 the N itches as they are called here pick up a 
 precarious living, selling to the white man any 
 173 
 
REMINISCENCES 
 
 fancy ornaments they may be able to make, a 
 few of the " gentle sex," with painted faces and 
 gaudy Indian attire, solicit favors from those 
 men who have no regard for either body or soul. 
 And a Salvation Army man stood at the cor- 
 ner of the street and beat his big drum. 
 
 Rattlesnakes must not be. uncommon in this 
 section of the country, for several of these deadly 
 reptiles were exhibited in the shop windows of 
 the town ; one was in a case outside the C.P.R- 
 ticket office. On a subsequent visit to this 
 vicinity I had the pleasure (?) of almost treading 
 on one of these creatures, and should have done 
 so, had his snakeship not saunded an alarm with 
 his rattles. 
 
 Many interesting memories are associated 
 with our short stay at Medicine Hat. Here we 
 packed and shipped several boxes of specimens 
 for study and for the museum. One of the 
 boxes contained a fine bleached skull and other 
 buffalo bones from one of the many " buffalo 
 graves " we had seen along our line of travel. 
 We had now completed our geological work, and 
 at 3 p.m. Aug. 29th, started on our last prairie 
 journey. We made 17 miles that evening before 
 camping for the night. Our horses seemed to 
 know that they were nearing their home, for 
 they were very restless all night. We were 
 greatly annoyed too by the howling or barking 
 of prairie wolves. With great care we kept our 
 174 
 
AMONG THE ROCKS. 
 
 horses till early morning, when after our last 
 breakfast on the plains, we left for and arrived 
 at Maple Creek, which place we had left on the 
 8th of June. 
 
 During our almost three months of prairie camp 
 life I had ridden horseback many hundreds of miles 
 over the barren and fertile fields of our Canadian 
 North-west, chased the antelope and coyote, seen 
 that important institution of prairie ranches, 
 the cow-boy, in his dangerous and exciting work, 
 " rounding up," visited the Indians in their wig- 
 wam, etc., and what was more interesting to me, 
 secured many valuable specimens which have 
 found a resting place in the Geological Survey 
 museum of the Dominion. After storing part 
 of our camp ouifit and other government pro- 
 perty we left for Ottawa which place we reached 
 Sept. 10th. 
 
 175 
 
CHAPTER IX. 
 
 TO SOUTH JOGGINS, N. S. PALAEONTOLOGY OF THE CAR- 
 BONIFEROUS ROCKS JOURNEY TO MANITOBA AND THE 
 NORTH WEST MAPLE CREEK AND THE CYPRESS HILLS 
 VERTEBRATE REMAIN'S A HERD OFWILDCATTLE BELLY 
 RIVER AND THE PRAIRIES SOME INDIANS SELKIRK AND 
 ITS ATTRACTIONS A PELICAN THE RED RIVER WIN- 
 NIPEG AND ITS MISCELLANEOUS STORES. 
 
 days after my long North-West trip, I 
 was on the road to the South Joggins, N. S., 
 where I again worked along the coast gathering 
 new information regarding the palaeontology of 
 the Carboniferous rocks. I returned to Ottawa, 
 Oct. 5th, with many fine specimens. My explora- 
 tion expenses for this year, 1883, were $1321.46. 
 
 The winter of 1883-84 was as usual fully 
 occupied in office and museum work. 
 
 The preparation of " Contributions to Micro- 
 Palseontology " by A. H. Foord, artist and assist- 
 ant Palaeontologist to the Survey, required a great 
 number of micro-sections of fossil Polyozoa, or 
 Bryozoa. This interesting work took up a good 
 portion of my time, and the remainder was spent 
 in preparing fossils for study and for the cases of 
 the museum, writing labels, superintending the 
 mounting of specimens, &c. &c. 
 
 To defray my expenses during the summer 
 1884, I received from the director of the Geolog- 
 ical Survey of Canada a letter of credit for 
 $1,700.00 with instructions to proceed to Man- 
 176 
 
AMONG THE ROCKS. 
 
 itoba and the North- West Territories to collect 
 fossils and obtain all the geological information 
 I could from the rocks of the Cypress Hills, and 
 from the deposits on a portion of the south shore 
 of Lake Winnipeg. * 
 
 As in other cases already mentioned I was to 
 give another young member of the staff his 
 initial exploration journey. This young student 
 in geology was James A. Macoun, son of the 
 veteran botanist of the Survey, Prof. Macoun, 
 M.A., F.L.S., F.R.S.C. 
 
 The little I knew of James Macoun led me to 
 believe he would make a most agreeable com- 
 panion. " What about Macoun?" said the chief 
 on our return to Ottawa ? '"Give me James 
 every time," was my^reply " he knows no fatigue, 
 and fears no danger." 
 
 With camp equipment which included a good 
 Winchester rifle, a Remington gun, fishing tackle 
 &c., we left Ottawa May 21st, and proceeded by 
 rail to Toronto and Owen Sound, thence by S. S. 
 Athabasca through a portion of Georgian Bay, 
 Lake Huron, and Lake Superior to Port Arthur. 
 The voyage over part of these two magnificent 
 lakes, and along a large portion of Georgian Bay 
 which together have an estimated length of 
 635 miles, and a breadth of 350 miles on one of 
 
 * Dr. G. M. Dawson the present Director of the Geological 
 Survey of Canada has kindly allowed me to make use of my field 
 note-books, for dates during the remainder of these Reminis- 
 cepces. 
 
 M 177 
 
REMINISCENCES 
 
 the powerful and well equipped steamers of the 
 Canadian Pacific Railway, from Owen Sound to 
 Port Arthur, occupied from 8.30 p. m. May 22nd, 
 to 12 p. m. May 25th. While in Georgian Bay 
 we were detained some time by fog, and soon 
 after passing through Sault Ste. Marie locks into 
 Lake Superior, we passed through a large body 
 of field ice. The nights were cold, and while on 
 deck an overcoat felt comfortable. We had sev- 
 eral opportunities of seeing some of the extensive 
 fisheries of these great lakes, and of tasting 
 white fish and salmon trout just after being 
 taken from the water. The scenery en route is 
 in many places especially the range of various 
 colored cliffs along the south shore and Thunder 
 Cape very grand. The gold, copper and other 
 minerals taken from the rocks in this section of 
 Canada represent many millions of dollars, and 
 yet it is thought that vast mineral wealth remains 
 to be developed. 
 
 A motley group of people met our steamer at 
 the Port Arthur landing ; rough miners, Indians, 
 wrapped in their bright colored blankets, beaded 
 leggings and moccasins, the cow-boy with his 
 broad brimmed hat, leather trousers and brace 
 of revolvers stuck in his cartridge belt ; the 
 tourist with his photographing camera, and the 
 man shark, " seeking whom he may devour." 
 But we had little time to study this picturesque 
 group, fur ai'ter a hasty meal at the hotel we 
 178 
 
AMONG THE ROCKS. 
 
 were soon in one of the handsome Pullman cars 
 bowling along towards Maple Creek, a distance 
 of about a thousand miles from Port Arthur 
 from which place, it will be ; remembered, we 
 started on our prairie journey in 1883. 
 
 Arriving at Maple Creek, May 29th, we found 
 the " town," which consisted of three stores, two 
 hotels and a few scattered dwellings in a state 
 of excitement over a white man who had been 
 killed by an Indian. This trouble, with the com- 
 forting remark that Indian horse thieves were 
 numerous over that part of the country we were 
 going to, made us feel a little anxious. To make 
 the journey at this time the people of the village 
 thought dangerous, and the sergeant in charge at 
 the mounted police fort kindly offered to give us 
 an escort, but we decided to risk the matter, and 
 at once set about getting our outfit in order. As 
 usual the J. G. Baker Company, who have stores 
 in all the principal North-west towns came to 
 our assistance, and in a short time we were pro- 
 vided with a waggon, buck board, four horses, two 
 half-breed Indians (cook and driver) and a month's 
 provisions. All being ready we left Maple Creek 
 at 12 noon, May 31st, and struck off for the 
 Cypress Hills, and our interesting fossil bone 
 locality, which I have already stated was dis- 
 covered by Mr. R G. McConnell of the Canadian 
 Geological Survey. We made 18 miles that day 
 and camped near a creek in close proximity to a 
 179 
 
REMINISCENCES 
 
 large encampment of Blackfoot Indians who 
 were engaged in collecting buffalo bones to be 
 used for refining sugar and for fertilizing pur- 
 poses. Continuing our journey the following 
 morning, partly by trail and partly by compass, 
 we reached our field of research 48 miles from 
 Maple Creek, Sunday, June 2nd a great verdant 
 valley extending many miles in a north and south 
 direction. Here we pitched our camp within a 
 short distance of where the water flowed south 
 into the White Mud River and north into Swift 
 Current. From this main valley coulees branch 
 off at right angles and extend from a short dis- 
 tance to several miles till they are lost in the 
 prairies. In these coulees are found the best 
 escarpments of Miocene tertiary rocks; which 
 although several hundred feet in thickness are 
 seldom seen in vertical sections of more than fifty 
 feet. The sides of the coulees slope at various 
 angles up towards the table land, and are partly 
 covered with scrubby brush, grass and wild sage 
 (Artemwiacaiid), while in the bottom a stream 
 of clear water flows, and willow, pine and other 
 trees form a shelter for the antelope and other 
 smaller game. 
 
 The Miocene rocks of the Cypress Hills plateau 
 consist largely of gravel in which pebbles of 
 quartzite from half an inch to a foot in diameter 
 predominate. In all the overlying strata frag- 
 ments of fossil bones, either fish, reptile or mam- 
 180 
 
AMONG THE ROCKS. 
 
 mal have been found, but it is in the agglomerate 
 bands that most of the vertebrate remains, now 
 in the Dominion Museum, were obtained. This 
 consists of (in places) a four- foot bed of yellowish 
 sandy limestone which when treated with hydro- 
 chloric acid leaves a residue of grains of quartz 
 and fragments of a variety of other rocks, angu- 
 lar and partly rounded pieces of rock similar to 
 the matrix. Occasionally pebbles of red, black 
 and other colored jaspers, banded quartzites, 
 chert and porphyry are found ; all derived from 
 the Lauren tian or Huronian mountains east of 
 these deposits, and transported during the glacial 
 epochs. In these rocks are entombed the remains 
 of an extinct Rhinoceros which Prof. Cope 
 named Menodus angustigenis. It is the largest 
 hoofed animal yet found in Canadian rocks. It 
 was our good fortune to discover, besides many 
 vertebrae, limb and other smaller bones, the 
 largest part of a cranium of this ponderous beast. 
 Some of the teeth are almost perfect great 
 sharp-edged cutting molars. The nasal bone is 
 ornamented with two horn cores. 
 
 SECTION OF MIOCENE TERTIARY ROCKS NEAB THE HEAD 
 WATERS OF SWIFT CURRENT. 
 
 a. Superficial and other deposits in which, in a bed of 
 yellowish- white silt 20 feet above the agglomerate beds 
 (6 of section), the teeth of an extinct deer, Leptomeryx 
 mammifer Cope, were found. 30 feet. 
 
 b. Agglomerate beds, containing rounded pebbles of 
 quartzite, jasper and chert. These are the beds from 
 which Menodus angustiqenis, M. Proutii, M. Americanus, 
 
 181 
 
REMINISCENCES 
 
 Hemipsalodon grandis, and other genera and species were 
 found. 4 to 20 feet. 
 
 c. Yellowish sand with a few pebbles of quartzite. 
 
 r 2". 
 
 d . Fine conglomerates in which a few fish remains were 
 found. 0' r. 
 
 e. Sand and thin beds of conglomerate. 5 feet. 
 
 /. Conglomerates in which a large bone was seen. 2 
 feet. 
 
 g. Sand and fine gravel. 2 feet. 
 
 h. Gravel, loose pebbles varying in size from half an 
 inch to a foot in diameter. 15 feet. 
 
 so' - ol' 
 
 '.r^^tC^M^^-.-a* & 
 
 - - "~^ ;. -r^m- t. C? 
 
 FROM TRAN. OF THE NOVA SCOTIA INST. OF SCIENCE, 1892-93. 
 
 182 
 
AMONG THE ROCKS. 
 
 Another interesting specimen obtained is one 
 of the ranii of a large flesh-eating animal allied 
 to the hysena a portion of one of the canine 
 teeth of this powerful and ferocious animal 
 measures one and a half inch in diameter. 
 Then we found an almost perfect lower jaw of 
 an extinct wild boar, Mothxriwn arctatum. Cope 
 named this. Then we found a portion of the 
 jaw of an extinct horse, which could not have 
 been as large as a Newfoundland dog Ancliithe- 
 rium Westonii. Prof. Cope named this after 
 the discoverer and writer of these notes. As the 
 remains of our little horse were found in the 
 same beds with the remains of the great flesh- 
 eating animal, it makes one think he must have 
 had a tough time keeping clear of this savage 
 beast. 
 
 Among other specimens from this interesting 
 locality are the remains of extinct fishes, one 
 bone of which Cope named Amia Whiteavesian, 
 after the distinguished palaeontologist of the 
 Canadian Geological Survey. Then we saw arid 
 collected portions of large trunks of silicified 
 trees, the lines of growth of which showed that 
 they were exogenous and of great age. It is 
 singular that in all these fossil bone beds, not a 
 single specimen of a leaf or blade of grass was 
 found. There must have been leaves to these 
 trees, and if there were no grass or other vegeta- 
 tion, what did our little horse live on ? Another 
 183 
 
REMINISCENCES 
 
 singular thing is, that in this vast body of Miocene 
 rocks only one cast of a shell was found, while 
 the formations above and below are crowded with 
 fossil shells. 
 
 By the numerous stone chippings and an 
 occasional perfect arrow-head, we found that the 
 pebbles of these Miocene rocks had been used by 
 the Indians in forming their implements and 
 weapons of war. A fine quartz arrow-head was 
 found in one of the cut banks, two feet below 
 the prairie level. How long did these two feet 
 of earth take to accumulate over this relic of 
 the red man ? 
 
 In rambling through the great valleys and 
 over the buttes of the Cypress Hills, one meets 
 with incidents not soon forgotten. Out of many 
 I could relate I must only mention one, for there 
 is still much to record before we get to the end of 
 our '84 travels. 
 
 Joe, my half-breed cook, and I were making a 
 journey in our buck-board along the great main 
 valley already mentioned, when on rounding a 
 butte which had obstructed our view, we came 
 face to face with a great herd of almost wild 
 cattle. They were directly in our path, filling 
 the space between the hills to our right, and the 
 stream to our left. Had we been on foot we 
 certainly would have been trampled to death. 
 For a moment only, Joe hesitated, then grabbing 
 a picket-line that lay at our feet, with tremen- 
 184 
 
AMONG THE ROCKS. 
 
 dous shouts swung it around his head, while I at 
 the same time urged the horse on It was a 
 critical time, and to this day I can't think how 
 we got through this herd of cattle who had yet 
 to be subjugated by man. 
 
 Continuing south, we at last came to a place 
 in English " The old man lying on his back." 
 It was here during a war between two tribes of 
 Indians a red man finding one of his opponents 
 sleeping on his back, drove his tomahawk through 
 his skull. Near this locality we came to what 
 Joe was anxious to show me, viz : A butte, com- 
 posed but I will quote from my notes on con- 
 cretions, Transactions of the Nova Scotian 
 Institute of Science, Vol. IX., Session 1894-95. 
 " Another interesting concretion locality lies half 
 a mile west of White Mud River, near the Fort 
 Walsh trail in the Assiniboine district, N.W.T., 
 the rocks belong to the Laramie formation. Here 
 a small butte was pointed out to me by my half- 
 breed Indian guide, who called the place gun- 
 shot butte, and said, a few years ago when he, 
 with others, hunted buffalo in that locality, they 
 sometimes when ammunition was scarce used 
 these balls in their guns and rifles. I found the 
 hill or butte to consist largely of calcareous sand, 
 which contained enormous quantities of spheroidal 
 concretions, varying in size from that of buck- 
 shot to an inch in diameter, the ordinary size 
 being that of rifle balls, etc." These curious forms 
 185 
 
REMINISCENCES 
 
 can be seen in the cases of the Dominion Museum. 
 Near this spot we saw the remains of shacks 
 which had been used during the buffalo hunt. 
 " Ah ! " said Joe, " they was the good times, good 
 lot to eat and good lot of fun. But we must 
 get back to camp before the sun he go down." 
 So our horse's head, much to his satisfaction, was 
 turned towards Skull-bone coulee, so named be- 
 cause of the fossil skull we had found. 
 
 It is known to all who have travelled the 
 prairies with a small band of horses, how very 
 " chummy " they become ; so much so that if one 
 is taken away, it requires a strong picket to keep 
 the others from following ; and how sensible they 
 are to the approach of a companion ! One day I 
 sent Joe on horseback to bring me a specimen of 
 rock from a far distant cliff. The sun was set 
 and I was becoming anxious about their return, 
 when after a time one of our waggon horses 
 began whinnying. I looked in all directions, 
 but could see neither man nor horse, and it was 
 two minutes or more before Joe and his horse 
 appeared, making their way over the top of a 
 butte half a mile distant. How the horse recog- 
 nized the approach of his companion half a mile 
 away, and before he saw him, we shall never 
 know. 
 
 During our journey back to camp we saw 
 several Indian graves. Joe told me that some 
 years ago many half-breeds and Indians had 
 186 
 
AMONG THE ROCKS 
 
 died in this valley of smallpox and were hastily 
 buried. The graves were covered with stones to 
 prevent the prairie wolf or coyote from getting 
 at the bodies, but one grave had been dug up 
 and the bones scattered around. I felt loath to 
 leave these great valleys and coulees of the 
 Cypress Hills where every day brought to light 
 some relic of days when the savage hyaena, the 
 bulky rhinoceros, our little horse and many 
 other long extinct creatures lived and were never 
 hunted by man. But we had spent all the time 
 we could afford, and with our precious specimens 
 and several excellent photographs of the rock 
 exposures, etc., we started on our return journey 
 to Maple Creek, which place we reached June 
 19th, and were soon busy repacking specimens 
 for shipment, settling accounts, photographing 
 typical Indians for the ethnological department 
 of the Survey Museum, preparing for our next 
 field of research, etc. 
 
 Maple Creek was a busy place just then 
 ranchers were in for supplies; cow-boys flirted 
 with dark-skinned Indian girls, and the red- 
 coated mounted police officers and men who had 
 driven in from " The Fort " for the mail, added 
 a warmth to the picture. But it was sad to see 
 the poor Indians toiling in with their rude carts 
 laden with buffalo bones. Thousands of tons of 
 these bones have been gathered by Indians from 
 the Cypress Hills district. 
 187 
 
REMINISCENCES 
 
 And now, before starting on our journey down 
 the Red River, I must say a few words about 
 another very interesting fossil locality about fifty 
 miles east of Maple Creek, and a few miles from 
 Medicine Hat. 
 
 At 2.30 o'clock June 23rd, Mr. Macoun, Joe 
 and myself boarded a freight train on the C.P.R. 
 and started for Irvin Coule'e. 
 
 At a lonely part of the prairie called Walsh, 
 our train was delayed eight hours waiting for 
 right of way. The only building there, was a 
 log cabin put up for telegraph purposes. The 
 operator kindly gave us some supper, and then 
 we spread our blankets and tried to get to sleep, 
 but the constant ticking of the telegraph machine 
 and the knowledge that in a box a few feet from 
 me was a live rattlesnake, prevented me from 
 getting even the proverbial forty winks. This 
 was a lonely place for a man to live, but the 
 operator delighted in nature and the wild prairies 
 and spent much of his time with his rifle and dog. 
 The prairie wolf or coyote now in one of the 
 cases of the Geological Museum was shot by him, 
 and presented at my request to the Survey. 
 
 At 2.30 we boarded our train again and at 6 
 o'clock the same morning reached Irvin. The 
 only house here was the railway section- house, a 
 rough place with rough line-men, but we were 
 glad to get breakfast here, then to spread our 
 blankets on the floor and take a few hours sleep. 
 188 
 
AMONG THE ROCKS. 
 
 Fortunately for us, a man with a yoke of oxen 
 was here from " The Hat " and we got him to 
 take our camp outfit two or three miles down 
 Irvin coulee, our next place of research. 
 
 We pitched our camp on the banks of Ross 
 Creek at the head of a verdant coulee near the 
 line of the C. P. Ry. A great portion of this dis- 
 trict is a sandy desert, with masses of soft sand- 
 stone interstratified with harder beds and in 
 places thin beds of lignite shales. By the action 
 of water these rocks have been worn into all sorts 
 of grotesque forms. See photo, p. 46, C. Geologi- 
 cal Survey Report for 1884. These rocks which 
 belong to the Belly River series already spoken 
 of contain fish and reptile remains. Mingled 
 with the bones of extinct turtle, on the water 
 washed sun-baked sands, we found many bony 
 enamelled plates, or scales of the Ganoid, and 
 teeth which may belong to the same fish. The 
 most important geological specimen found in 
 Irvin or Ross Coulee was the scapula of an ex- 
 tinct animal, probably a Dinosaurian. It was a 
 bone three feet long, eight or ten inches wide and 
 half an inch to one or two inches thick. I spent 
 six or eight hours uncovering this bone, and Mr. 
 Macoun walked sixteen miles to Medicine Hat 
 and back to get glue or some other material 
 with a view to preserving this specimen. It was 
 like most of these fossil bones, cracked in all 
 directions. Well, after all our trouble, while lift- 
 189 
 
REMINISCENCES 
 
 ing it from its sandy bed, it fell into a thousand 
 fragments, and now lies at the bottom of one of 
 the great excavations in these soft sandstones. 
 Still I mourn the loss of this bone. 
 
 The fact that the coulee in which we were 
 camped was a favorite resort for prairie wolves, 
 rattlesnakes, and other animals, gave rise, as we 
 sat round our camp-fire to many wolf and snake 
 stories. One told by Joe, our cook, was that a 
 rattlesnake will never touch a pregnant woman. 
 Both half-breeds and Indians believe this. We 
 saw several wolves and one antelope in this cou- 
 le'e. In shooting at one of these, the end of my 
 rifle burst, the break only affected an inch at the 
 muzzle, and Joe was soon busy sawing this off 
 with the bread-knife. It took three hours to do 
 it, but it was accomplished as well as though we 
 had proper tools for the job. This rifle, which 
 hangs close to where I now write, reminds me of 
 an incident I may relate. While camped at Maple 
 Creek this summer a very fancy dressed Blood 
 Indian came along, a thick shock of hair stood 
 straight up from his forehead, the lower part was 
 colored green and the upper part red ; his dark 
 brown face was striped with red and yellow 
 paint; his bright colored blanket half covered 
 his almost nude body. Joe, at my request, inter- 
 viewed this dude, and found that in a few hours 
 he was to be married. I at once set up my 
 camera to get a picture for our ethnological room 
 190 
 
AMONG THE ROCKS. 
 
 in the Survey Museum, but the moment Mr. 
 Indian saw the camera away he went. Joe 
 fetched him back, showed him a plug of tobacco 
 and gave him the rifle to shoot me if the " picture 
 maker " hurt him in any way. By these means 
 I succeeded in getting a good photo, of this bride- 
 groom elect, and also of Joe (who was almost as 
 dark-skinned as the Indian). 
 
 While Macoun and Joe had gone in search of a 
 vehicle no easy task in this locality I wandered 
 alone among the sandy buttes and clays of the 
 " bad lands " of this district. I had found, and 
 was bagging some small fossil bones, when a slight 
 noise caused me to look up, and there, fifty paces 
 away, was a large hungry looking wolf. It did 
 not appear to be the night-prowling coyote, but a 
 larger and darker colored species probably the 
 dusky wolf. For a moment I did not know what 
 to do, bub remembering what Joe had said, I gave 
 the biggest war-whoop this valley had perhaps 
 ever heard, and to my delight Mr. Wolf turned 
 and walked off. Towards evening my assistants 
 returned with an ox cart and driver. We camped 
 that night near the railway track close to Irvin 
 section- house. Here I paid off Joe and started 
 him on his road back to Maple Creek. He had 
 been a good man and I tried hard to persuade 
 him to accompany us oil the remainder of our 
 journey, but he would not go so far from 
 home. 
 
 191 
 
REMINISCENCES 
 
 The following morning Mr. Macoun and I visi- 
 ted a very interesting locality, first noted by Mr. 
 McConnell of the Survey in 1883. In my notes 
 on Concretions found in Canadian Rocks Trans 
 of the Nova Scotian Inst. of Science, Vol. IX., 
 1894-5 I remark as follows : 
 
 44 The largest concretions seen by the writer were found 
 in the Fox Hill and Pierre shales and clays of the North - 
 
 CONCRETIONS AT IRVIN COULEE, N.W.T. 
 West ; a very interesting exhibition of these giant forms 
 may be seen three miles north of Irvin station-house on 
 the Canadian Pacific Railway. Here huge boulder-like 
 spheroidal and ovoid concretions once held in the rocks, 
 but removed by denudation (probably in the Glacial epoch, 
 for Glacial striae are seen on some of the flat beds), stand 
 out in bold relief, resting on the flat and up-turned edges 
 of shale and sandstone ; and on the top of one of them, 
 about twenty feet high, an eagle had built its nest of 
 buffalo bones and the roots of the wild sage, for want of a 
 192 
 
AMONG THE ROCKS. 
 
 more elevated situation, which does not occur in this 
 locality." 
 
 We found many interesting fossils here, among 
 which was Linyula, nitidu ; Protocarilia sub- 
 quadrata ; Liopistha, and other species new to 
 these rocks. Crystals of selenite (gypsum) are 
 scattered over the sands and clays of these bad 
 lands lands which till only a year ago, when our 
 great Pacific Railway wound its way across almost 
 trackless prairies were seldom traversed, and 
 then only by the red man and wild beasts, and 
 now in hundreds of places, where previous to 
 1883, the Indian pitched his wigwam and the 
 prairie wolf roamed, one sees fields of grain and 
 probably runs against a barbed wire fence. 
 
 We left Irvin July 1st, and the following 
 day drove from Medicine Hat to the Saskatch- 
 ewan coal mines (distant about eight miles). Mr. 
 Lawson the manager received us kindly, and we 
 remained at his house two days. This is a 
 remarkable place, with great coule'es and buttes, 
 which contain thick beds of lignite coal. 
 
 From some of the sandstone and iron -clay 
 stone we obtained many good fossils of both 
 plants and shells. One of the most interesting 
 fossil plants found here at a spot where Mr. 
 Lawson the day before killed a rattlesnake 
 has been described and figured by Sir J. W. 
 Dawson in Trans. Royal Soc. Canada 1885. Sir 
 William calls this pretty fossil leaf Brasenia 
 
 N 193 
 
REMINISCENCES 
 
 antiqua, but by mistake gives Mr. Lawson credit 
 for its discovery. The geological formation here, 
 is Upper Cretaceous, part of the Belly River 
 series. The photographs we obtained of these 
 Coal mines, give a good idea of the country and 
 also of the Lignite beds. We returned to Medi- 
 cine Hat, July 4th, and spent several days 
 examining the rocks of that vicinity. We 
 reached Winnipeg July 14th, and the same day 
 we landed at East Selkirk on the banks of the 
 Red River 30 miles from Winnipeg. Here we 
 found lodging at a small but comfortable hotel. 
 The Village of Selkirk is about three miles dis- 
 tant from the bank of the Red River. The chief 
 attractions of the place then (1884) were the 
 quarries, where great loose pieces of limestone 
 are dug out of the clay and shale and sent to 
 -Winnipeg to be used in the construction of some 
 of the fine buildings of that city ; the arrival and 
 departure of C. P. Ry. trains, and the lime kilns. 
 The limestones of this locality and a good por- 
 tion of the Red River Valley belong to the 
 (Galena) Trenton and Black River formation. 
 They are whitish yellow in color and contain a 
 large fauna of fossils, many of which have been 
 figured and described by Mr. Whiteaves of the 
 Canadian Survey. The fossils obtained by the 
 writer and Mr. Macoun, and later by Mr. 
 Lawrence M. Lambe artist to the Geological 
 Survey whose beautiful drawings adorn the 
 194 
 
AMONG THE ROCKS. 
 
 Reports form one of the important fossil 
 exhibits of the Dominion Museum. 
 
 It was several days before we could obtain a 
 boat and men, but at last we secured a good 
 stout sail-boat, one elderly Cree Indian who 
 had spent most of his life at the Hudson Bay 
 Post one half-breed, who owned the boat, and 
 one month's provisions. At 9.30 a.m. July 19th, 
 we left Selkirk Village and drifted down the Red 
 River a muddy stream with a strong current, 
 which in a few hours after passing through the 
 large and handsome city of Winnipeg, empties 
 its sediment-laden waters into Lake Winnipeg. 
 
 For some miles after leaving East Selkirk, the 
 banks of the river on each side are dotted with 
 small thatched cabins or cottages, and a few 
 larger and more pretentious houses. Most of 
 these dwellings belong to half-breed Indians, 
 some of whom have good farms and fine cattle. 
 After this, for miles before reaching the lake, 
 the land on either side of the river is low and 
 marshy, with a thick growth of rank grass, and 
 rushes growing to a height of five, ten, and even 
 eighteen feet. This swampy land appears to be 
 the home of wild birds, especially the bittern 
 and other members of the heron family. Camp- 
 ing ground is very scarce, and when found is 
 well like most Indian camping grounds. 
 
 We pitched our first camp on the shore of 
 Lake Winnipeg, July 21st. It would occupy 
 195 
 
REMINISCENCES 
 
 too much space in these notes and perhaps 
 weary the reader to follow us from island to 
 island, and over long stretches of water and the 
 shores of this beautiful lake, so I will only relate 
 a few of the many incidents still fresh in my 
 memory. 
 
 Our second camp was made on Elk Island, 
 near the north shore and thirty or forty miles 
 from the mouth of the Red River. It is a pretty 
 island, three miles or so in length, well wooded 
 with birch, poplar, pine, &c. My men said the 
 lake was well stocked with fish, among which are 
 pike, cat-fish, sturgeon, whitefish, etc. I can 
 vouch for the cat-fish, for Macoun, by wading in 
 up to his waist and slinging a line and hook 
 baited with pork, brought out some very fine 
 specimens. I was surprised to find how different 
 fish, just taken from the lake, tasted from those 
 of our canals and lakes near cities. 
 
 Boulders of every description of Laurentian 
 rocks line part of the west shore. The bedded 
 rocks and the fossils they contain resemble 
 those of the Hudson River formation of Anti- 
 costi. 
 
 After spending a short time at Observation 
 Point, we sailed to Big Island, where the rocks 
 are pretty much the same as at Elk Island. 
 Among the few fossils obtained here was an 
 interesting specimen of Bryozoa, which has been 
 figured and described by E. 0. Ulrich, of the 
 196 
 
AMONG THE ROCKS. 
 
 Geological Survey of Illinois, Contributions to 
 micro-palaeontology, Part II, 1889, as Diplotrypa 
 Westoni. 
 
 Our 5th canip was made on Deer Island Here 
 sandstone, green shale and limestone form a 
 ridge of cliffs twenty to sixty feet high some of 
 the shaly beds are characterized by fucoidal 
 markings which have been, perhaps, twenty feet 
 in length ; among other fossils observed here are 
 Orthoceras, of large dimensions ; Receptaculites, 
 which resemble the seed portion of large sun- 
 flowers one or two species of coral, and a few 
 other obscure forms. 
 
 Punk Island was the next locality visited. A 
 strong north-west wind drove us on this island, 
 very nearly wrecking our boat. The shore at 
 the east end is white sand, which would make 'a 
 good glass. This sand is characteristic of several 
 parts of the shores of this lake. There was little 
 to interest us here, and, while waiting for the 
 weather to clear, we amused ourselves with fish- 
 ing, several " gold eyes " and two fine cat-fish 
 were caught. 
 
 From Punk Island we crossed the lake to Big 
 Grindstone Point, on the south shore. Between 
 this and Washow Bay the shore is rugged with 
 shelving cliffs of sandstone and limestone rocks 
 similar to those of the islands already spoken of. 
 
 Our lake Winnipeg photographs which may 
 be seen at the Geological Survey Department, 
 197 
 
REMINISCENCES 
 
 give an excellent idea of a portion of this beau- 
 tiful lake. 
 
 July 31st we crossed Washow Bay to Bull's 
 Head. While our two men went slowly on with 
 boat and equipment, Macoun and I walked many 
 miles along the shore, noting the various ex- 
 posures of rock, collecting fossils and other objects 
 in natural history. 
 
 Low ridges of broken cliffs occupy several miles 
 of the shore of this locality. Many interesting 
 fossils were found here in the vicinity of Dog's 
 Head, where our next camp was made. Here, 
 one evening while wandering alone along the shore 
 I captured with my hands a fine pelican. It ap- 
 peared to be resting after a long flight, and sleep- 
 ing ; when grabbed by the neck it fought bravely. 
 I shouted, and Macoun ran to my assistance. We 
 wanted this specimen for the museum, or we 
 should have given it its liberty. But how differ- 
 ent this fish-eating bird of the wilderness looks 
 in the glass case ! The following day (Sunday) 
 Macoun skinned and prepared the pelican while 
 I photographed the Hudson's Bay Post, of Dog's 
 Head a log cabin occupied by a half-breed 
 Indian, his wife and children. A portion of the 
 building appeared to be a stable for a number of 
 esquimaux and other dogs which were then lying 
 around the buildings. This H. B. Post was the 
 only building seen along the coast. 
 
 Near any H. B. Post one is sure to meet with 
 198 
 
AMONG THE ROCKS. 
 
 Indians; such was the case here, two families, 
 representatives of the Bungo and Swampy Cree 
 tribes. Twelve members of" one family were 
 crowded under a poorly constructed bark wigwam. 
 It cost half a pound of tea to get a photo, of this 
 outfit. The other family said they were also 
 Bungo Indians, who I believe are an offshoot of 
 Swampy Crees. These descendants of once brave 
 races of red-skins appear to have discarded, or 
 perhaps for a few of the white man's luxuries 
 parted with all the gaudy trappings so dear to 
 their forefathers. With very few exceptions all 
 the Indians seen on the shores of Lake Winnipeg 
 were clothed in the cast-off garments of the white 
 man. One weather-beaten buck considered him- 
 self in full dress when he wore a battered stove- 
 pipe hat, which was probably made before the pre- 
 sent owner was born. Through my boatmen, 
 who spoke this old buck's language, I found he 
 was chief of one of the Swampy Cree tribes, and 
 had lately been married to a young squaw. One 
 of my most valued Indian stone pipes* was ob- 
 tained from this man for a plug of tobacco. 
 
 Many of these old chiefs although very poor 
 are proud and courteous in manner and always 
 find a way of returning, in some small way, any 
 gift or compliment paid them. After trading for 
 the stone pipe and a fine pike just fresh from the 
 lake, I invited the old man to tea with us. We 
 
 * See Toronto Hail and Empire, Feb. 2nd, 1898 Fig. 1. 
 199 
 
REMINISCENCES 
 
 had canned beef, pancakes, &c., &c., for supper. 
 It was perhaps the feast of the old man's life ; he 
 went away with his hand on his stomach and 
 tears of gratitude in his eyes. Just before turn- 
 ing in for the night, we were surprised to see the 
 old man appear at our camp fire with a young 
 squaw his new wife. Speaking in his own 
 language, and with great courtesy, he present- 
 ed his bride to me ; I called our Indian cook 
 Basque to interpret arid found that the old chief 
 presented his wife to the white chief for the 
 remainder of the time he pitched his wigwam at 
 Dog's Head ! Through Basque I presented my 
 sincere thanks for the noble chief's kind offer, 
 which I could not accept as I had a white squaw 
 in the big city, and she would not like me to 
 take another into my wigwam. The old man 
 looked hurt and surprised, but a little tea to the 
 fair lady, and a slice of pork to the old man, 
 fixed matters up nicely, and they went away 
 into the darkness of the night. 
 
 Continuing our journey along the coast, we 
 reached Kinwow Bay, Aug. 8th. From this 
 beautiful spot we sailed to a small island a few 
 miles distant. This we found to be the most 
 interesting yet struck, so far as fossils are con- 
 cerned. Thousands of slabs of yellowish white 
 dolomitic limestone have been washed up from 
 the lake, many are much water-worn and bleach- 
 ed by the rain and sun till they look like big pan- 
 200 
 
AMONG THE ROCKS. 
 
 cakes. These rocks, which are seen in place at 
 low water, dip at a slight angle towards the 
 north. During the few hours we were able to 
 remain on this little island, we collected many 
 fossils new to the Galena limestones of that 
 district ; in fact the rocks were so rich in organic 
 remains that, on being requested by the men to 
 name the island, I called it Bonus Island. From 
 this prolific spot of land and rock we crossed to 
 Cat Head. Here a large range of cliffs 60 to 70 
 feet high occupy the coast. Geologically they 
 are interesting rocks as they contain fossils not 
 found in this formation elsewhere. These cliffs 
 hold a great number of kidney shaped " chalk 
 balls" which give the face of the cliff the appear- 
 ance of having been bombarded. Some of these 
 forms have a nucleus of flint. 
 
 This is the end of our journey towards the 
 head of the lake. Since leaving the mouth of 
 the Red River we have covered about half the 
 length of this magnificent lake, which is about 
 270 miles long. While returning we encountered 
 several heavy storms and often had to run into 
 some sheltered spot. 
 
 Aug. 12th, we were again at Dog's head this 
 time to find a large number of Bungo or Swampy 
 Cree Indians assembling from all parts. They 
 were congregating to hold their " Dog Feast." 
 This accounted for the constant turn, turn of 
 the Indian drum, which we had heard, wafted 
 201 
 
REMIiNISCENCES 
 
 over the waters of the lake from distant 
 shores. 
 
 The feast was to take place the following day, 
 on the opposite shore of the lake, which here is 
 perhaps only half a mile across so we de- 
 cided to stay and witness this feast of the white 
 dog. 
 
 After a great deal of formality, in which two 
 pounds of tobacco and a few other traps figured, 
 Chief Sac-a-chew-we-ass allowed the servant of 
 the great white chiefs at Ottawa to point the 
 nose of his camera over the top of the blanket 
 door of their sacred wigwam, which consisted of 
 a long shed-like tent, built of branches of trees, 
 and decorated with any scraps of bright colored 
 ribbons, rags, tin cans, etc., they could beg, bor- 
 row or steal. In the centre of this feast-wigwam 
 was a large iron pot, in which the white dog was 
 to be boiled. The occasion of this dog-feast 
 gathering was to initiate young Indian braves, 
 lads who were old enough to be instructed in the 
 art of Indian warfare. When part of the object 
 of our visit was accomplished, through an inter- 
 preter, we were politely requested to move off, 
 that the great Dog Feast might proceed. As we 
 pushed our boat from shore, the blanket door of 
 the wigwam was raised to admit twenty or thirty 
 Indian braves (?) who, with rattles made of 
 tomato and other fruit and meat cans containing 
 a few pebbles and fastened to the end of sticks, 
 202 
 
AMONG THE ROCKS. 
 
 had been driving away any evil spirits lurking 
 around the ceremonial lodge. 
 
 August 17th. We had a splendid run from 
 Goose Island, where we had been weather-bound, 
 to Grassy Narrows. Here we shot several ducks, 
 and Macoun clubbed a pike three feet three 
 inches long. We had little or no provisions left, 
 so these edibles were very acceptable. 
 
 We pitched our twenty-first lake camp near 
 Willow Point, and the following day ran our 
 boat into the Red River and made our night 
 camp on a marshy spot of land, the right of 
 which we had to dispute with hundreds of frogs. 
 
 At 5.30 the following morning we found a 
 stiff head wind blowing down stream. Rowing 
 against wind and current was out of the ques- 
 tion, so the remainder of our boat journey was 
 made by tracking, and at 2 p.m., August 21st, we 
 reached the point at Selkirk we had left on the 
 19th of July. 
 
 Our grub had been poor and scanty for several 
 days, so on reaching the hotel all hands sat down 
 to a jolly dinner, after which Basque and our 
 skipper were paid off; the former proceeded at 
 once to get drunk, while the man of our little 
 craft who had brought us safely through storms 
 and sunshine slipped quietly down stream to- 
 wards his humble home on the north shore of 
 the great lake over which we had safely glided. 
 
 Late that night poor Basque was found under 
 203 
 
REMINISCENCES 
 
 a hedge sleeping off the effects of too much fire- 
 water. The money I had paid him money 
 which would have furnished his family, who 
 dwelt in a wretched shack down the river, with 
 pork and flour enough for the winter was gone. 
 We spent two or three days collecting specimens 
 and visiting several historical spots on the banks 
 of the Red River. One of these was Fort Garry 
 renowned for the many attacks its walls had 
 withstood in the fierce warfare with savage 
 Indians. Another interesting spot was the Gun 
 homestead, where we found lodging for a day or 
 two. The Gun family date back to the early 
 settlement of half-breed Indians on the banks of 
 the Red River. Stone implements of war and 
 the chase are often plowed up on the banks of 
 this muddy stream. It was our good fortune to 
 find two fine stone mauls or hammers. Now 
 where the Indian prepared his flint arrow-heads 
 and weapons of war a church stands, and a bell 
 calls both Indian and white man to prayer. 
 
 August 30th, we were again in Winnipeg 
 (" Win " muddy and " Nipi " water) the largest 
 prairie city of Manitoba, over 760 feet above the 
 sea ; in 1871 a village with a population of 241 
 souls, which in thirteen years increased to 20,000. 
 Where thirty years ago the red man pitched 
 his bark wigwam, now stone churches, banks, 
 schools, factories, handsome dwellings and stores 
 stand. 
 
 204 
 
AMONG THE ROCKS 
 
 At this time, 1884, cheap, hastily erected frame 
 buildings occupied a large portion of the town, 
 prominent among these were many " second- 
 hand " shops, where for a small portion of their 
 original value might be purchased guns, rifles, 
 revolvers, dagger-knives, fishing tackle, cartridge 
 and money belts, together with hundreds of other 
 articles which were to figure in the new life 
 their former owners were to pursue 
 
 But we must hasten to the locality where our 
 last geological examinations are to be made 
 Stony Mountain, which place we reached August 
 30th, Stony Mountain, that dreaded name to 
 all evildoers in Manitoba, for it is here, a few 
 miles from Winnipeg, on elevated ground, sur- 
 rounded by fertile prairie land, that the hand- 
 some Penitentiary is situated. 
 
 In the spacious vestibule of this building were 
 mounted specimens of buffalo and other denizens 
 of the surrounding country ; the walls were hung 
 with Indian trophies of war and the chase. 
 
 Mr. Bedson, warden of the institution, received 
 us kindly, and after dinner detailed one of his 
 men to show us his fine herd of buffalo, which, 
 with the exception of having only a limited 
 range of their native prairie to roam, are in a 
 state of nature. Two prairie wolves, a badger, 
 a bald headed eagle, and other live stock served 
 to amuse those prisoners who were allowed to 
 leave the jail walls. 
 
 205 
 
REMINISCENCES 
 
 The fossils found in the white limestones of 
 Stony Mountain indicate that the geological 
 horizon here is higher than that of East Selkirk 
 and Lake Winnipeg. Our fortunate discovery 
 of Beatricea ungulata, a fossil (?) which is char- 
 acteristic of the Hudson River formation of An- 
 ticosti arid Rabbit Island L. S., seems to leave no 
 doubt about the formation. 
 
 In digging a well to supply the Penitentiary 
 with water a thick bed of purplish colored argil- 
 laceous limestone was cut through. This bed is 
 crowded with organic remains corals, bryozoa, 
 brachiopoda, gasteropoda and other genera are 
 abundantly represented in these rocks, which, but 
 for the excavation made for water, might have 
 remained unknown. Apart from my study of 
 the rocks, bugology occupied a good portion of 
 two nights. Caesar's ghost ! when I think of 
 those two nights, and yet our hostess expressed 
 no surprise when shown how many boarders 
 occupied one bed. 
 
 And now our geological work in Manitoba is 
 over for this season and we are homeward bound. 
 
 During our explorations my companion, James 
 Macoun has, whenever an opportunity presented, 
 pursued his favorite study botany, and returns to 
 Ottawa with an important addition to his father's 
 already large collection of Manitoba plants. 
 
 At Barkley a small way station between 
 Winnipeg and Port Arthur owing to a big 
 206 
 
AMONG THE ROCKS. 
 
 wash-out, our train was delayed thirty-six hours. 
 Alex. Mackenzie, M.R, ex-Premier of Canada, 
 and wife were passengers on this train. The old 
 gentleman asked, and I answered, many ques- 
 tions regarding the working of the Geological 
 Survey. 
 
 At Port Arthur we found a number of mem- 
 bers of the British Association, who, with the 
 Director of the Canadian Geological Survey, were 
 taking a run through a portion of our great 
 North-west. 
 
 Another voyage over the waters of Lake 
 Superior, Lake Huron and Georgian Bay to 
 Owen Sound, a railway journey of two hundred 
 miles or so and we were again, on the 6th of 
 September in Ottawa. 
 
 207 
 
CHAPTER X. 
 
 EXCAVATIONS AT QUEBEC EOZOON AGAIN A COLLECTION 
 OF MAMMALIAN REMAINS WORK ON THE QUEBEC 
 GROUP THE BLACK LIMESTONES OP QUEBEC SUP- 
 POSED COAL DEPOSITS EVADING THE SCOTT ACT 
 WITH MR. FLETCHER TO CAPE BRETON ARISAIG, 
 N.S. THE COAL MINES OF SYDNEY. 
 
 ^EPTEMBER 17th. Having heard that some 
 *** obscure fossils had been found in the rocks 
 then being taken out of the excavations of the 
 Drill Hall on the Cove Fields, Quebec, I lost no 
 time in reaching that locality, and was rewarded 
 by adding to our exceedingly scanty knowledge 
 of the fauna of the Quebec City rocks. Among 
 the interesting collection of Graptolites made 
 were several forms new to this portion of the 
 Quebec group I shall take this opportunity of 
 recording that in all the researches made in these 
 city rocks, not one typical Lie" vis fossil has been 
 found, such as Phyllograptus typus, Hall. Leav- 
 ing Quebec City rocks, a short time was again 
 spent among the graptolitic shales of Point 
 LeVis, and again, as is always the case in these 
 remarkably prolific rocks, many interesting ad- 
 ditions were made to our already large collection 
 of graptolites. Some of these forms show fea- 
 tures not known to Prof. Hall when he wrote 
 his important Monograph, Decade II, 1865, 
 208 
 
AMONG THE ROCKS. 
 
 " Graptolites of the Quebec Group," written 
 for the Geological Survey of Canada. 
 
 This was the year of the discovery of that 
 interesting graptolite Tetragraptus approxi- 
 matus. 
 
 It seems strange that this species, which is 
 common in some beds of the Levis shales, should 
 not have been known at the time Hall wrote the 
 above report. My last official journey of 1884 
 was made Oct. 2nd, to Cote St. Pierre to obtain 
 a new supply of specimens of the supposed fossil 
 Eozoon of which I have already spoken. Another 
 fine collection was made from the locality where 
 Sir J. W. Dawson and I collected in 1874. 
 
 Mr. Larique, a farmer, occupant of the small 
 house figured by Dawson in his sketch of the 
 Cote St. Pierre Eozoon locality " Dawn of Life," 
 p. 22 drove me to Papineauville. Here at the 
 railway station I met two English geologists who 
 had been in search of Eozoon and " the man with 
 the hammer." These gentlemen Dr. James 
 Murray and G. H. Parke had been told in 
 Ottawa that I was in the Eozoon locality, so they 
 hunted both Eozoon and the man with the ham- 
 mer. The} 7 discovered the latter just as he was 
 about to start for home, but their precious collec- 
 tion of " Eozoon " did not happen to have a single 
 representative of this distinguished "Laurentian 
 fossil." I gave them several specimens from my 
 collection, and they went away happy. 
 
 o 209 
 
REMINISCENCES 
 
 Although so few geologists now believe in 
 Eozoon being organic, Sir William Dawson, who, 
 by the way, a few weeks ago, celebrated his golden 
 wedding is still as firm a believer in Eozoon as 
 he was when he wrote his " Dawn of Life." At 
 this time (April '97) I believe his last efforts in 
 behalf of Eozoon are going through the press. 
 
 During the winter of 1884-85 many duties 
 connected with the museum and the restoration 
 of specimens for study and exhibition fell to my 
 lot. The first of these was the removal and re- 
 arrangement of the ethnological collection from 
 the upper to the middle flat. This being ac- 
 complished, the somewhat tedious task of restor- 
 ing, labelling, etc., the large collection of mam- 
 malian fossil bones, collected by Dr. Dawson, 
 Mr. McConnell and myself from the Miocene 
 rocks of the Cypress Hills, N.W.T., was under- 
 taken. There being no osteologist connected 
 with the Canadian Geological Survey, it was de- 
 cided that this, or a portion of this collection 
 should be placed in the hands of Professor E. D. 
 Cope, of Philadelphia, the renowned professor of 
 comparative anatomy, etc., etc., etc. Having 
 arranged this large collection of Miocene fossils 
 for inspection, the Professor was invited to come 
 to Ottawa and select such specimens as he 
 thought worthy of description and illustration. 
 Prof. Cope came, made his selection, and they 
 were shipped to his rooms in Philadelphia. The 
 210 
 
AMONG THE ROCKS. 
 
 result of Professor Cope's studies of this collec- 
 tion can be seen in " Contributions to Canadian 
 Palaeontology/' Vdi. Ill (Quarto), 1891. Later 
 on I shall speak of Prof. Cope again, but there is 
 an incident I must record here. 
 
 Yesterday, April 15th, 1897, while writing the 
 above lines, and while my thoughts were wan- 
 dering to Prof. Cope's work-rooms, 2102 Pine 
 Street, Philadelphia, where another large collec- 
 tion of fossil bones belonging to the Geological 
 Survey of Canada chiefly of my collecting 
 were being studied, I received the Philadelphia 
 "Evening Bulletin" for April 12th, 1897, in which 
 the death of Prof. Cope was announced. I need 
 not say what a shock this was to me, After two 
 weeks' sickness, this great scientist was cut down 
 in the midst of his geological labors. And now 
 shall I have to go to the Quaker City and see to 
 the returning of our specimens ? 
 
 The remainder of the winter and till June 2nd 
 was spent in over-hauling all the Carboniferous 
 fossils in the museum. Many of these were 
 developed, relabelled, etc., and several weeks at 
 microscopic work completed the indoor duties 
 for 1884. 
 
 The short time allotted me for field work 
 during the summer of 1885 was spent on the 
 rocks of the " Quebec Group." The want of 
 fossils to determine the geological horizon of the 
 Quebec City, and a large portion of the Eastern 
 211 
 
REMINISCENCES 
 
 Township rocks, was the cause of my researches 
 this summer. As the following notes will show, 
 Dr. Selwyn, director of the Geological Survey, 
 was much interested in this "Quebec Group" 
 work, and joined me several times for the pur- 
 pose of studying the stratigraphical structure of 
 this formation. Arriving at Quebec June 23rd, 
 I found our chief awaiting me at the St. Louis 
 Hotel. Together we carefully went over the 
 rocks of the " Cove Fields " (already mentioned). 
 I was then left alone to examine the rocks of the 
 city, and here I must record one of the most 
 interesting fossil discoveries made in the shales 
 and limestones of this place. 
 
 It seems but why, no one knows till this 
 time to have been taken for granted that with 
 the exception of two or three species of grapto- 
 lites found last year in the " Cove Field " beds, 
 a few fragments of crinoiclal columns and plates, 
 and two species of Bryozoa, discovered by the 
 writer during Logan's connection with the Survey, 
 that these city rocks were destitute of fossil 
 remains. The result of these discoveries and 
 those of the last three or four subsequent years 
 has proved that the limestones, shales and con- 
 glomerates of Quebec City hold many interest- 
 ing organic remains ; in fact, a large fauna of 
 both graptolites and other genera, among which 
 those found this year 1885 deserve special 
 mention. I refer chiefly to the discovery in the 
 
AMONG THE ROCKS. 
 
 rocks near the back of St. John's St., or Mont- 
 calm Market. Here I found the black shales and 
 limestones to contain numerous small bivalve 
 shells belonging to the Lingula family, also a 
 great number of minute Crustaceans belonging to 
 the Trilobite family. Specimens of the latter 
 fossils were forwarded te Mr. Ford, of Troy, 
 N.Y., who was then engaged in the study of the 
 trilobites of the primordial formation. Ford's 
 interesting report on this collection can be found 
 in " Transactions N. Y. Academy of Science," 
 vol. 7, 1897. 
 
 On June 25th, Dr. Selwyn again joined me 
 in the study of the black limestones of the city, 
 more particularly those of the quarry near to 
 St. Patrick's Street, where this rock has been 
 largely quarried for making hydraulic cement. 
 They are the characteristic limestones of the city, 
 and a portion of the north-west end of the island 
 of Orleans, and contain one or two species of 
 graptolites which occur abundantlyin the Hudson 
 River formation. On June 27th, the well-known 
 surgeon, Dr. Ahem, of Quebec, accompanied me 
 to Point Levis, where the previous year he had 
 discovered, near Fort No. 2, fine specimens of 
 the typical bivalve shell, Lingula Quebecensis. 
 A few good specimens were obtained, but I had 
 to leave further researches for some other time, 
 which at present has not come. Dr. Ahern has 
 frequently done the Geological Survey good 
 213 
 
REMINISCENCES 
 
 service, and to this day remains one of my most 
 esteemed friends. 
 
 On July 1st, Dr. Selwyn again joined me, and 
 together we examined some miles of the rocks on 
 the south side of the island of Orleans. Two or 
 three days were then spent between the east and 
 west end of the island, on the north shore. 
 
 Our present visit afforded us a good oppor- 
 tunity of visiting the so-called coal deposits, 
 which were then being so much talked of. We 
 found the locality, but as usual these " coal beds " 
 turned out to be a deposit of black shale, owing 
 its blackness, in part, to being very much slicken- 
 sided caused by two beds of rock rubbing to- 
 gether and to thin veins of Anthraxolite, a meta- 
 morphosed or hardened bitumen, a mineral found 
 in thin seams in the various formations of the 
 " Quebec Group." I think I have mentioned 
 before that the rocks of the island of Orleans are 
 many thousands of feet below the coal beds of 
 Nova Scotia, and were deposited many thousands 
 of years before any coal deposits we know of. 
 
 The following incident relates to the first out- 
 cry I heard of coal on the island of Orleans. 
 
 Somewhere near the year 1868, one or more 
 gentlemen waited on Sir William Logan and 
 informed him that a deposit of coal had been 
 found in the rocks of the island of Orleans, and 
 requested that he would accompany them and 
 make a report on this discovery. " Fiddle-de- 
 214 
 
AMONG THE ROCKS. 
 
 dee !" said Sir William ; " the rocks of the island 
 of Orleans are thousands of feet below the car- 
 boniferous rocks, and we have no coal in Canada." 
 "But, Sir," replied one of these men, taking some 
 lumps of coal from his pocket, "here are specimens 
 from our coal beds." " Yes," replied Sir William, 
 " these are good samples of Newcastle coal." The 
 men were persistent, however, that they had a 
 valuable coal deposit, and at last Sir William 
 consented to visit these mines. A careful investi- 
 gation of the locality resulted in the discovery 
 that many years ago a blacksmith's forge had 
 stood on this spot, and later on a rock slide had 
 buried the blacksmith's forge and the coal which 
 constituted this valuable coal mine. 
 
 In 1846, Sir William was called on to inspect 
 some " coal deposits " in the vicinity of Baie St. 
 Paul, below Quebec. Here with the exception 
 of a patch of Trenton rocks the formation is 
 Laurentian. The upshot of Sir William's re- 
 searches here resulted in the discovery that im- 
 ported coal had been carefully " packed " to 
 deceive the inhabitants. 
 
 Other similar incidents might be related with 
 regard to the discovery of coal in the Province of 
 Quebec, but the above will serve to show that at 
 present no true coal has been found in beds 
 lower than the Carboniferous formation, which 
 formation is not represented there so far as we 
 know. Our nearest coal deposits are in Nova 
 215 
 
REMINISCENCES 
 
 Scotia. Continuing our journey eastwards from 
 the " coal mine " partly by buggy and partly on 
 foot along the margin of the island, we were soon 
 opposite the magnificent Falls of Montmorency, 
 which are said to be nearly a hundred feet 
 higher than those of Niagara. Here the foamy 
 waters leap 250 feet over the face of rugged 
 rocks. A view of the falls from Orleans island 
 is worth going a long distance to see ; but the 
 grandeur of this mighty body of water rushing 
 from its summit to the valley below can be seen 
 to its best advantage only at the foot ot the fall, 
 where if one is so inclined he can in a few 
 moments get drenched to the skin in the icy 
 spray which rises like a cloud from where the 
 fleecy waters bury themselves beneath the waters 
 of the river. Some years after the time of which 
 I am writing, while making a typical collection 
 of the rocks of the falls, I experienced what it 
 was to get a bath from this icy spray. Perhaps 
 it was the exertion of climbing the 400 steps to 
 the top of the cliff that counteracted the chill 
 received. 
 
 Taking to our buggy again, we continued east 
 along a road dotted with farms owned by French 
 habitants, descendants of the early French sett- 
 lers of the island. The numerous wild plum 
 trees and vines of this island perhaps account 
 for Jacques Cartier calling it the " Isle of 
 Bacchus." 
 
 216 
 
AMONG THE ROCKS. 
 
 Arriving at St. Famille a thoroughly French 
 village " as old as the hills " on Saturday after- 
 noon, the doctor decided we had better remain 
 there till Sunday, so our driver, a French Cana- 
 dian, interviewed the only people who kept 
 boarders, and after a considerable amount of talk 
 we were at last quartered till the following morn- 
 ing. After a sumptuous dinner of pork and 
 beans my good chief who never lost an hour 
 working time unless his strength gave out pro- 
 posed that we should descend the banks to the 
 shore and proceed towards the east end of the 
 island. The shades of night were falling when 
 after a four hours* geological tramp we returned 
 to our French Canadian boarding-house, where 
 no one seemed to speak English. We were both 
 a little played out and I proposed taking a little 
 stimulant, if it could be got in a village where 
 the " Scott Act " was in full force. Hunting up 
 our landlord, and taking him to one side, I placed 
 my hand on my stomach and said : " Monsieur, 
 mon ami et moi bien fatigue's, nous-avons mal-dau 
 coeur, avez-nous du quoi, du whiskey ou du 
 brandy." The old man gazed in a frightened way 
 around the room, then with one finger up to his 
 nose to enjoin silence locked the front door. Dr. 
 S. and I were then taken into a back room, the 
 door of which was also locked ; then the old man 
 unlocked a quaint old-fashioned corner cupboard 
 and produced a bottle of gin ; well, we had our 
 217 
 
REMINISCENCES 
 
 drink, but while taking it I felt like a thief and 
 was glad when the doors were unlocked and open 
 once more. 
 
 At 8 o'clock the following morning (Sunday), 
 we were in our buggy again and crossing to the 
 south side of the island. The geological section 
 presented along the south shore of this pretty 
 island, affords an excellent opportunity of study- 
 ing a number of the characteristic rocks of the 
 " Quebec group." 
 
 Towards evening we reached the west end of 
 the island and again took up our quarters for 
 the night in our friend Lizotte's comfortable 
 hotel long since destroyed by fire, but restored 
 again on a much more extensive plan by the 
 enterprising owner who, like our friend Gee, of 
 whom I shall speak later on, has sheltered and 
 fed many members of the geological staff, and to 
 whom so many of the improvements of the west 
 end of this pretty island were due. 
 
 July 6th. We crossed over to St. Joseph de 
 Levis where the government graving dock was in 
 course of construction. It is a structure 484 feet 
 long and 100 feet wide. This great work has 
 long been completed, and I have there often 
 walked under the keels of our large ocean steam- 
 ships. In excavating for this great dry -dock 
 many interesting beds of the LeVis rocks were 
 exposed. One large block of conglomerate taken 
 out we thought worth having cut into slabs for 
 218 
 
AMONG THE ROCKS. 
 
 the museum, and it was shipped for that pur- 
 pose. 
 
 From St Joseph we took the I.C.R. track to 
 Point L6 vis, stopping every little way to examine 
 rock -cuttings, some of which yielded many fossils 
 new to the " Quebec Group." We returned to 
 Ottawa that evening, July 6th. 
 
 Part of the months of July and August were 
 spent in the Eastern Townships examining the 
 rock exposures near Richmond on the St. Francis 
 River and Hereford, where I crossed the boundary 
 line into the State of New Hampshire. 
 
 The first indication I got that I was in Uncle 
 Sam's domains was a cemetery with most of the 
 graves decorated with small American flags, 
 flowers and children's toys. It looked strange 
 to us to see all these little Yankee flags stuck 
 into the graves, but since then I have seen many 
 American burying-grounds, and find the people 
 of the United States pay much attention to the 
 decoration of graves. 
 
 The absence of fossils in the great stretch of 
 rocks examined rendered this my last official 
 journey of 1885 monotonous; still it was pleasant 
 to travel along the banks of the picturesque St. 
 Francis River, almost every foot of which was 
 known to my driver, who was our old friend 
 Gee, whose house at Melbourne had for months, 
 sheltered our greatly respected late chief Sir 
 William Logan, Richardson, Webster, myself and 
 219 
 
REMINISCENCES 
 
 other members of the Geological Survey staff. 
 Poor Gee! Incidents relating to his " historical 
 house " crowd thickly on my memory. Many 
 pages might be written about the " Gee home- 
 stead," where Logan spent his last working 
 days. Gee was never more happy than when 
 driving his renowned guest, Sir William, to and 
 from his field work in distant localities. All 
 of the old members of the staff of those days 
 excepting Dr. Bell, who is still actively en- 
 gaged in field work for the Survey, and my- 
 self (at my own request I was superannuated 
 in 1894) are gone to rest. Logan, Murray, 
 Billings, Richardson, Barlow, Hunt, and many 
 others, who, though not members of the staff, 
 took an active part in the Survey work. Last 
 of these (to the present time, 1897) I must men- 
 tion our old friend Gee. of whom I have been 
 speaking. Poor Gee, who fed us on the fat of 
 the land (?) and delighted in piloting us to some 
 new discovery he had made of serpentine, slate 
 or other minerals. I believe this geological ex- 
 cursion of 1885 was the last Gee ever assisted in. 
 But even though all these late members of the 
 staff' have passed away, the geological work as 
 the reports of the Survey will show still goes 
 on, being diligently pursued by Dr. R. W. Ells 
 and other members of the staff. 
 
 As usual, the winter months of 1885-86 were 
 occupied in office and museum work. During 
 220 
 
AMONG THE ROCKS, 
 
 this season in conjunction with Mr. (now Dr.) Ami, 
 many improvements were made in classifying, 
 arranging and labelling specimens, preparing 
 samples of gold and Eozoon for the Colonial and 
 Indian Exhibition. 
 
 My geological excursions of 1886 commenced 
 early in July, when I joined Mr. Hugh Fletcher, 
 B.A., one of the Survey's most energetic field 
 geologists. The geological work accomplished 
 by Mr. Fletcher in Nova Scotia and Cape Breton 
 will remain forever a monument to his energy 
 and perseverance. It was to assist this gentle- 
 man in researches for fossils to determine by 
 palaeontological evidence the geological horizon 
 of certain rocks, that I joined him. To be with 
 Mr. Fletcher while he is pursuing his field work 
 is a geological feast. This, with the beautiful 
 country through which we travelled, was refresh- 
 ing indeed after spending weeks on the non-fos- 
 siliferous rocks of Lake Huron and the Georgian 
 Bay. With Mr. Fletcher I drove and walked 
 through a good portion of Antigonish, gleaning 
 from the rocks much evidence of their past his- 
 tory. About ten days were spent examining the 
 various rock exposures in the vicinity of beauti- 
 ful Lochaber Lake, near the shore of which we 
 boarded with a charming family, who, after the 
 day's duties were over, gathered around the piano 
 or organ and sang the good old melodies of long 
 ago. A game of euchre with the old folks, or a 
 221 
 
REMINISCENCES 
 
 sail on the lake usually ended the days days all 
 too short for both work and pleasure. 
 
 From Lochaber Mr. Fletcher and I journeyed 
 to Arisaig, N. S., which place we reached July 
 20th. Since my first visit to this locality in 1869 
 many changes had taken place. The tedious 
 buggy drive of 30 miles from New Glasgow is 
 now accomplished in a short time by rail. 
 Changes, too, had occurred in my old friend 
 Joseph McDonald's homestead; marriages, deaths, 
 departures to fields more promising than this 
 rough sea coast and the scanty livelihood to be 
 got from these rocky lands. But " The sea rolled 
 on as it rolled before," and here, as we travel 
 under the sea-splashed cliffs, we see no change 
 and decay. I have spoken of the geology of the 
 Arisaig rocks before, and shall only mention here 
 that many new fossil forms were added to our 
 already large collection. 
 
 Aug. 8th Mr. J A. Robert one of Mr Fletcher's 
 assistants joined me, and together we proceeded 
 to Cape Breton to examine some of the Cambro 
 Silurian rocks of that region. Railway, stage 
 and steamboat carried us to North Sydney, where 
 I had an interesting chat with Mr. Richard 
 Brown, manager of the Sydney coal mines. To 
 this gentleman the Survey is indebted for many 
 beautiful Carboniferous coal plants. 
 
 Leaving Sydney we drove to George River, 
 Bras d'or Bay, where, in the vicinity of Mr. 
 222 
 
AMONG THE ROCKS. 
 
 McFee's farm, we discovered beds of limestone 
 holding numerous fossils belonging to the genera 
 Obolella, Lingula, Hyolithes, &c., forms, some of 
 which have been described by Mr. Matthew, of 
 St. John, N. B. 
 
 After wandering along the shores of Bras d'or 
 Bay which were strewn with thousands of jelly- 
 fish cast up by the waves over verdure -clad 
 mountains, along the shores of Escasoni Lake, 
 where we fought the rocks, and caused them to 
 yield up many precious geological relics, and 
 where we boarded at some of the picturesque 
 farm houses in the vicinity of Christmas Island 
 settlement, where some of the people end most of 
 their sentences with " whatever," we wended our 
 way to Riversdale station, on the Sydney branch 
 line of the Intercolonial Ry., a short distance from 
 Truro. Here we found lodgings at the Post 
 Office, kept at one of the few houses in the im- 
 mediate locality of the station, where the arrival 
 and departure of trains formed the chief excite- 
 ment of the place. 
 
 We spent five quiet, restful days with the 
 homely old lady and her daughter, who carried 
 the mail- bag between the station and the house. 
 
 In this quiet old-fashioned home, where the 
 table in the best room was strewn with goody- 
 goody books, to my surprise I found a charming 
 novel, " Broken Toys," by I forget the author 
 written many years ago, evidently by one who 
 223 
 
KEMINISUEISCES 
 
 understood the ways of children well, for it gives 
 a beautiful description of a little girl who tried 
 to bring up a large family of dolls in the way 
 they should go, but with all her love and care 
 some of them went sadly wrong ; one lost an eye, 
 another a leg, then Mary Jane became so 
 thoroughly demoralized that, as an example to 
 her other sisters, she had to be buried and gotten 
 out of the way for ever and ever. 
 
 The rocks in the vicinity of Riversdale Ry. 
 station represent a portion of the Carboniferous 
 or Devonian system. The railway, a little east 
 of the depot, cuts through a thickness of about 
 115 feet of argillites of various colors, beds of 
 sandstone and layers partly made up of concre- 
 tions. Some of these beds are prolific in the 
 bivalve shell Naiadites (Dawson), while other 
 strata hold numerous fragments of fossil plants, 
 chiefly Cordaites leaves. Catamites, Neuropteris 
 and many other fossils of the Carboniferous 
 or Devonian period occur in these interesting 
 rock sections. This is a locality where many 
 days might be spent profitably collecting. 
 
 August 31st, I passed through Point LeVis and 
 spent a few hours on the graptolite beds, adding 
 one new species to our list. 
 
 Oct. 7th, I was again at our famous Eozoon 
 
 locality at C6te St. Pierre, where, thirteen years 
 
 ago with Sir William (then Principal) Dawson, we 
 
 studied in place the supposed Laurentian fossil, 
 
 224 
 
AMONG THE ROCKS. 
 
 Again this year, 1886, with Glen Gordon, of 
 Ottawa, I was entertained in the little farm 
 house made historical in that interesting book 
 " The Dawn of Life." But I have spoken several 
 times about this supposed Lauren tian fossil, and 
 shall only say here that, although cart loads of 
 pieces of this supposed ancient coral reef have 
 been carried away, hundreds of tons weight still 
 remain for those interested in Eozoon Canadense. 
 
 While writing the above I have received a let- 
 ter, and several photographs, from Mr. Fletcher, 
 of the Survey, dated May 20th, 1897, in which 
 he says : " A crowd of us, including Dr. Selwyn 
 and Adams, went to Cote St. Pierre last Saturday 
 to find Eozoon" which, he adds, some think " will 
 die a natural death." 
 
 During the winter months of 1886-87 I was 
 occupied as usual with microscopic and other 
 work connected with the ethnological and 
 palaeontological branches of the Survey and 
 museum. It gives me pleasure to record here 
 the advice and kindly aid I always received from 
 those officers of the Survey who were more 
 learned than myself in some matters I had to 
 deal with. My thanks are specially due to Mr. 
 Whiteaves, Mr. Fletcher and Dr. Ami, for kind 
 assistance in anything pertaining to literary 
 work. 
 
 22* 
 
CHAPTER XT. 
 
 EXPLORATIONS ON THE SOUTH SHORE OF THE ST. LAWRENCE 
 FISHING FOR COD A PLEASANT STAY AT MR. RICH- 
 ARDSON'S A JOURNEY TO GRIMSBY A TRIP TO THE 
 NORTH-WEST PLAINS MAPLE CREEK SWIFT CURRENT 
 BIG JAW COULEE A PRIZE PIG THE BAD LANDS 
 LETHBRIDGE WE LAUNCH A BOAT ON THE RED DEER 
 RIVER BUT AFTERWARDS GIVE UP OUR PLAN THE JOG- 
 GINS COAST ONCE MORE. 
 
 DURING the winter of 1886, doubts regarding 
 certain portions of the geological structure 
 on the south shore of the St. Lawrence, below 
 Levis, existed in the mind of the Director of the 
 Survey and he decided to visit this section of the 
 " Quebec Group " the following summer, and that 
 I should accompany him. 
 
 Accordingly on the 14th of June, 1887, 1 joined 
 Dr. Selwyn in Quebec and together we proceeded 
 by caleche to Beaumont, the small village next 
 below St. Joseph de Levis. Here we bagged some 
 lunch and started down the shore in search of 
 various outcrops recorded by Mr. Richardson. 
 We had a most interesting tramp, and many facts 
 regarding the relation of the south shore rocks to 
 those opposite on Orleans Island were obtained. 
 The following day we examined the exposures in 
 the vicinity of St. Michel de Bellechasse. Along 
 this section of the coast, many fine examples of 
 anticlinal and synclinal foldings are seen. Miles 
 226 
 
AMONG THE ROCKS. 
 
 of red, green and gray argillites, quartzites, lime- 
 stone, conglomerates, and other rocks tilted at 
 various angles, form a rugged coast. There is 
 considerable evidence that the rocks in this local- 
 ity are lower than the Lie" vis proper. Since the 
 date of which I write, the geology of this shore 
 has been worked out and mapped by Dr. Ells of 
 the Survey. 
 
 Continuing along the shore towards Levis we 
 came on what Dr. Selwyn supposed to be the 
 fault between the Lower and Upper Levis. The 
 former horizon as regarded by Selwyn extended 
 past the snow shed on the Intercolonial Railway. 
 But here in the shed, fossils came to our aid, and 
 for the first time since the " Quebec Group " con- 
 troversy typical LeVis fossils were found, among 
 which was the bivalve shell Lingula Quebecensis, 
 thus proving that we had passed the Lower Levis 
 beds, and were then on the Upper LeVis formation. 
 
 On our return to Point LeVis Dr. Selwyn left 
 me, and I spent several days alone collecting fos- 
 sils from the cliffs at Le'vis and from the shore 
 rocks in the vicinity of Cap Rouge on the Que- 
 bec side of the river. 
 
 On the 7th of July, I again left Ottawa for 
 Quebec to join Dr. Selwyn, who was to accom- 
 pany and spend a short time with me and Mr. 
 Lambe the artist of the Survey, who was to be 
 my assistant on the rocks in the vicinity of 
 Gaspe and Cape Rosier. 
 
 227 
 
REMINISCENCES 
 
 In a former visit to Cape Rosier I had discov- 
 ered compound graptolites which indicated that 
 the escarpments in the vicinity of the light- 
 house were older than they had hitherto been sup- 
 posed to be, and it was chiefly to see these grap- 
 tolitic beds that the chief of the Survey accom- 
 panied me. Going by rail to Dalhousie, and by 
 steamer Admiral to Gaspe was a delightful trip. 
 The Government S. S. La Canadenne which was 
 about to leave Gaspe Basin with lighthouse sup- 
 plies, carried us to Grand Greve where through 
 the kindness of Dr. Wakeham, commander, we 
 were landed by one of the steamer's boats. A cart 
 carried us from Grand Greve to Cape Rosier 
 lighthouse where we were kindly received and 
 accommodated with board and lodging. Although 
 it was the llth of July, the weather was cold, 
 and we were glad to sit by a wood fire and listen 
 to the waves as they dashed against the rugged 
 cliffs a short distance below our bedroom window. 
 The following morning our worthy chief, anxious 
 to see the rocks in which I had a few years before 
 discovered the compound graptolites, hurried us 
 off to the shore below the lighthouse. For a 
 short time I failed to find my graptolite beds, and 
 think the Doctor doubted my veracity. How- 
 ever in clue time I pointed to some thin bands 
 of blackish gray limestone, perhaps thirty feet 
 up the cliff, and said, <: you are a good climber, 
 Sir will you see what those black beds con- 
 228 
 
AMONG THE ROCKS. 
 
 tain ?" Up went the doctor, and in a few mo- 
 ments, holding a piece of limestone in his hand, 
 and with a beaming face called out : " You are 
 right, Weston, here are your compound forms." 
 While the doctor threw down pieces of the rock 
 Mr. Lambe and I made a selection of the best 
 fossils, some of which are now in the hands of 
 Prof. Lapworth, of Birmingham, England, for 
 identification. The remainder of the day was 
 spent along the coast east of the lighthouse. Two 
 days later the chief left us and returned to Gaspe* 
 to join his daughter, who had accompanied us 
 from Levis to that place. 
 
 During the third night of our stay at Cape 
 Rosier lighthouse, while the wind roared and the 
 angry waves lashed the shore, a wee wail from a 
 new-born babe was blended with the roar of the 
 sea, and we were called on to drink the health of 
 the little stranger who had come to help swell 
 the French- Canadian population. 
 
 Mr. Lambe and I left our snug retreat in the 
 lighthouse July 19th, and proceeded by cart 
 along the coast road, making our geological 
 observations and collecting fossils wherever we 
 could find them. 
 
 We reached L'Anse au Griffon July 20th, 
 where bold cliffs of aroillite and other rocks 
 
 O 
 
 occupy the coast. The rocks of North Bluff at 
 
 Griffon Cove are prolific in graptolites, many of 
 
 which are characteristic Hudson River forms. 
 
 229 
 
REMINISCENCES 
 
 In a hay cart we pursued our way along the 
 shore road, which, as my companion Lam be 
 observed, " was just rough enough." At Cape 
 Magdalen we spent a very pleasant day or two, 
 boarding as usual in fishermen's homes where 
 each member of the family vied with the others 
 in trying to make us comfortable. 
 
 Black slates containing graptolites of the Hud- 
 son River formation are well displayed in the 
 vicinity of the lighthouse, and the sand stones of 
 this vicinity hold a few Brachiopods. From Cape 
 Magdalen lighthouse we examined the coast rocks 
 to Mont Louis and Gros Male. Graptolites of 
 the Hudson River series occur in abundance in 
 some of the beds of black shale. 
 
 The scenery of this section of the river is very 
 fine; in fact all along the shore of the grand 
 stream one finds an endless variety of scenery, 
 and a constant change in the arrangement of the 
 rock formations. The pleasure, however, of a 
 journey such as the one I am speaking of, is much 
 greater when with camp and provisions as in 
 my first journey up this coast as then one is 
 independent of the (very often) poor, but cheer- 
 fully given accommodation in fishermen's homes. 
 
 We reached the Marsouin River July 29th. 
 This is another very interesting geological 
 locality. Its black shales are abundantly stocked 
 with graptolites, presenting a rich feast for the 
 palaeontologist. 
 
 230 
 
AMONG THE ROCKS. 
 
 A considerable amount of fishing is done at 
 the various places near where we found lodging 
 for the night. If at some of these stations one 
 has time to spend half a day cod fishing with one 
 or two of the sturdy fishermen of this coast, he 
 will find food for the mind as well as food for 
 the body. Generally I am not a successful 
 angler, but standing one evening on a block of 
 stone a few paces from the shore, I cast my fishing 
 line, with a hook baited with pork, into the dark 
 waters. In a few moments I felt a vigorous tug 
 at my line and hauling in I found a good sized cod 
 on my hook. In a shallow stream which empties 
 into the river, small trout were abundant, but as 
 they feed on the decaying refuse of thecod fish, 
 prepared here for market, I did not disturb them. 
 
 On the first of August we reached Ste. Anne 
 des Monts, where the St. Anne River one of the 
 finest salmon streams in Canada empties into 
 the St. Lawrence River. I have already spoken 
 of this river and my journey up it to the Shick- 
 shock Mountains, in 1878. 
 
 The coast here presents many interesting 
 features to the geologist and palaeontologist. 
 Here an extensive series of dark colored slates 
 and shales occur, in which we find three miles 
 or so above the mouth of the St. Anne River 
 typical graptolites of the " Quebec Group," the 
 most prolific forms are the well known Phyllo- 
 graptus typus and P. angustifolius. 
 231 
 
REMINISCENCES 
 
 Above Ste. Anne, black, green, red and other 
 colored slates and shales with massive beds of 
 conglomerate are seen for miles up the coast. 
 The fossils found in the limestone conglomerates 
 leave no doubt as to their geological horizon, viz., 
 " Bic conglomerates " of the " Quebec Group." 
 
 Following this varied strata we reached 
 Ruisseau A'sem, Aug. 4th, where we were hospit- 
 ably received by Mr. John Richardson, son of 
 James Richardson of whom I have frequently 
 spoken in these memoirs. We found Mr. Rich- 
 ardson busily engaged in his saw-mill, where 
 thousands of cords of birch wood had that 
 summer been stowed, and were then being cut 
 into long strips, tied up in bundles and shipped 
 to Scotland and other places for making spools. 
 At this time a large barque was being loaded 
 with bundles of this product birch cut from the 
 extensive forests near the shores of the St. 
 Lawrence River some of which would no doubt 
 in course of time be returned to Canada in the 
 form of cotton spools. 
 
 The weather was delightfully fine during our 
 stay at Ruisseau and we made long tramps along 
 the rugged shore, returning at sunset to find the 
 busy operations of the day suspended, and the 
 mmates of Mr. Richardson's refined seaside home 
 gathered around a cheerful wood fire. After a 
 pleasant evening, in which music and cards fig- 
 ured, and while the mighty waves dashed against 
 232 
 
AMONG THE ROCKS. 
 
 the great cliffs just below, each member of the 
 household read a verse from a chapter in the 
 Bible, after which a prayer followed in which 
 the " strangers within our gates " were not for- 
 gotten, and the good Lord was asked to guide 
 the footsteps of those who sought sermons in 
 stones. 
 
 We left this delightful retreat Aug. 5th, and 
 followed the coast to Grande Matane where we 
 were kindly received and entertained at the 
 house of Mr. and Mrs. James Russell son-in-law 
 and daughter of the late Mr. James Richardson. 
 
 Following the coast to Little Matane we passed 
 over large exposures of black slates, some of 
 which are rich in graptolites, especially in two 
 or more species of Dictyonema fossil Polyzoa, 
 or Hydrozoa, which resemble on a small scale 
 the coral sea fan Gorgonial. 
 
 Having finished as far as time would permit, 
 our coast examinations, Mr. Lambe and I pro- 
 ceeded by rail to Point LeVis, where after a few 
 hours on the ever interesting rocks of that local- 
 ity, we returned to Ottawa. 
 
 The excellent water color sketches of some of 
 the coast rocks made by Mr Lambe, on this his 
 first geological journey, together with the series 
 of photographs taken by myself in 1878, serve 
 to remind us of pleasant days days spent in 
 trying to unravel the complicated structure of the 
 Quebec rocks, pleasant days for those who, as 
 233 
 
REMINISCENCES 
 
 John Richardson said in his prayers, " sought 
 sermons in stones." 
 
 On the 1 4th of September I was again in 
 Grimsby, Ont., that pretty little village near the 
 shores of Lake Erie, and a few miles from Niag- 
 ara Falls. Here again I met my old friend Mr. 
 Johnson Pettite who was ever ready to lay aside 
 his books and " bugs " and accompany me in a 
 fossil hunt. But I have spoken of Grimsby and 
 our friend Pettite before, so will only mention 
 here that again we had a profitable time among 
 the (in places) highly fossiliferous rocks of the 
 Medina and Niagara formations. Besides secur- 
 ing many good specimens belonging to the order 
 Echinoidea, or sea urchins, we collected many 
 other fossils of interest, and the large slab of 
 Arthrophycus Harlandi which adorns the wall of 
 the geological museum. It is supposed to be a 
 fucoid or sea plant arthron a joint, pkykos, a 
 sea plant. 
 
 September 17th I again settled down to my 
 usual winter duties in the Survey building. 
 Many new specimens were added to the museum 
 and much good work done by each member of 
 the staff work done not simply for our daily 
 bread and butter, but for the love and interest 
 each member of the Survey took in the welfare 
 of the institution. 
 
 During the early summer of 1888, I was again 
 preparing for another long journey across the 
 234 
 
AMONG THE ROCKS. 
 
 North-west plains. Our former collection of 
 mammalian and reptilian fossil bones had proved 
 so interesting that it was thought advisable to 
 visit the Cypress Hills district again, and then 
 proceed to Red Deer River, Alberta, where in 
 1887 Mr. J. B. Tyrrell of the Survey had ob- 
 tained a large portion of a Dinosaurian skull and 
 other reptilian fossil bones. 
 
 On June 28th, provided with camp equipment 
 I left Ottawa and proceeded by Canadian Pacific 
 Railway, about 1600 miles, to Maple Creek which 
 place I reached July 2nd. Here I was again in- 
 debted to Horace Greely, of Powers and Bros, 
 general store, for fitting me out with waggon, 
 buck- board, horses, two half-breed Indians and 
 provisions. 
 
 We left Maple Creek July 4th, and struck out 
 for the head waters of Swift Current. On reach- 
 ing a prairie farm a few miles distant one of the 
 men repented his contract and refused to go 
 further with us. I was just about to return to 
 the Creek to try and get another teamster when 
 a typically dressed cowboy stepped up and said : 
 " What's the matter with me going with you 
 uncle ; I don't like boastin', but there aint many 
 places around these parts Joe Lowry and Billy 
 don't know, and for that matter Billy and me kin 
 take you any wheres between here and Cali- 
 fornia." 
 
 I found Billy was a French Canadian pony 
 235 
 
REMINISCENCES 
 
 that had been the constant companion of his 
 master Joe in all his wanderings from the Pacific 
 coast to the Atlantic, and back to our present 
 locality. I liked Lowry and we soon proceeded 
 on our journey Jim our teamster with the wag- 
 gon, Lowry with the buck-board, I on horseback, 
 and Billy following as close to his master as he 
 could get. We reached the head waters of Swift 
 Current July 10th, and camped in the main val- 
 ley called Frenchman from which many cou- 
 le"es branch. It was in one of these coulees, on 
 our present explorations, that the largest por- 
 tion of the skull of an extinct Rhinoceros was 
 found. 
 
 Big Jaw cou!4e, through which a beautiful 
 stream of clear cold water flows and in which 
 bush and green pastures form a delightful retreat 
 for the antelope and a few other wild animals 
 who still roam the Cypress Hills and valleys, 
 besides yielding to us a large portion of the extinct 
 Rhinoceros skull, has been the home of ferocious 
 flesh-eating animals animals nothing but the 
 bones of which have ever been seen by the eyes 
 of man, who appeared on this planet ages after 
 these creatures had ceased to exist. 
 
 Lowr\' became greatly interested in my 
 researches and determined that he too would add 
 something of value to our collection. I was 
 making the geological section which appears on 
 a previous page, when Joe, who was far up the 
 236 
 
AMONG THE ROCKS. 
 
 other side of the couie'e shouted " Uncle " the 
 only name Lowry ever gave me "I've found the 
 big bone you wanted." I found the specimen to 
 be a well preserved femur or thigh bone of our 
 extinct Rhinoceros, or as Prof. Cope calls it 
 Menodus angustigenis. 
 
 It would require too much space here to record 
 all the interesting incidents of our journey over 
 the plains, and through the valleys of the Cypress 
 Hill district. 
 
 A few more words however which by permis- 
 sion I copy from my field-note book and for 
 this year we must leave our Miocene rocks with 
 their wonderful records of a silent world silent 
 except for the roar of savage beasts as they 
 sought their prey among the peaceful animals of 
 
 those past ages. 
 
 Big Jaw Coulee N. W. T. 
 
 Sunday July 15th, 1888. 
 
 "8 a.m. Bright clear morning Bar. 26.1 Ther. 54. strong 
 S. E. wind. 9 p.m. Scout (Lowry) and I have ridden about 
 ten miles over this locality and have visited one of my bone 
 coulees I think the one we named Cold Water coulee, as 
 besides having a stream of water through it, it had several 
 springs near the mouth. This is the head waters of Swift 
 Current, about five miles north of our camp or where the Pie 
 Pod trail enters the main coulee (Frenchman). The day is 
 closing in very cold I have to keep on my over-coat. My 
 two cow-boys are merry to-night, owing to the prospect of 
 starting for civilization to-morrow." 
 
 Two lunch and one night camp which was 
 made at Bear Creek on the East Post trail, and 
 we had again crossed the Cypress Hills and are 
 237 * 
 
REMINISCENCES 
 
 at Maple Creek busy shipping specimens, settling 
 accounts, &c. 
 
 Barring the little accidents almost always met 
 with in crossing the North-west plains the jour- 
 ney, to one who loves nature, is full of pleasure, 
 especially if one has men like those I have had 
 on this trip who when the day's work was done 
 sat around the camp fire and told of their wand- 
 erings in distant lands ; of their cow-punching 
 incidents, feats in lassoing wild steers, &c. I 
 was inclined to doubt some of the tales Lowry 
 told of his lassoing exploits, but while at " The 
 Creek " I had an opportunity of seeing that he 
 was no boaster in the art of lassoing. While we 
 were waiting for our train we joined a prairie 
 picnic which was going on near the " town." 
 Among a number of prizes offered for various 
 feats, was a pig which was to be given to the 
 first man, Indian or white, who could lasso Mr. 
 Pig at a distance of fifty yards. The pig was 
 dumped out of a box on the prairie flat ; a sharp 
 prod in the hind quarters started him off at full 
 gallop, but in a moment six lassos were thrown, 
 in another moment a wild shout from hundreds 
 of Indians, cow-boys and the motley group which 
 always composes a prairie picnic for Lowry had 
 his lasso round the pig's neck, thus winning the 
 prize, which I believe he sold for $5.00. 
 
 On July 22nd, I was again at Lethbridge, and 
 the following day (Sunday) with a team of horses, 
 238 
 
AMONG THE ROCKS. 
 
 waggon, saddle-horse, two men one of whom 
 was a typical North-west mule driver and pro- 
 visions for a few days, we started for Milk River 
 Ridge. 
 
 I have stated before that the collection of fos- 
 sils Dr. Dawson, Mr. Tyrrell, my son George and 
 I made in 1883 were all lost by the sinking of 
 the ship on which they were freighted. It was 
 with the hope of replacing some of these speci- 
 mens that the present journey was made. 
 
 Soon after leaving the settled portions in the 
 vicinity of Leth bridge bad luck seemed to follow 
 me. Travelling the bleached clays and sands of 
 the " bad lands " of the North- West Territories 
 under favorable circumstances has not many 
 redeeming features, but in a blazing sun with 
 one's horses almost mad with the stings of that 
 great pest of the North-West, the horse fly, and 
 one's self swarmed with mosquitoes, to have to 
 listen constantly to the blasphemous language 
 of two of the lowest types of prairie mule-drivers, 
 takes away much of the pleasure of a geological 
 journey. 
 
 Many miles away from civilization we woke 
 at sunrise to find our horses gone. Jim stood 
 for a few moments, using language not fit for 
 these or any other pages, then without breakfast 
 started to track our horses The sun had set 
 when he returned with our three horses and two 
 extra ones he had picked up on the plains. It 
 239 
 
REMINISCENCES 
 
 was during Jim's absence that the fine large 
 bivalve fossil shells which are to be seen in a 
 wall case in the Survey Museum, were discov- 
 ered ; other fossils of much interest were found 
 in the natural excavations of these barren lands. 
 Here was the tooth or rib of some extinct rep- 
 tile, and there the cast of a delicate shell whose 
 pearly beauty had long perished, and here again 
 the leaf of a plant which received its name many 
 thousands of years after its verdant life had 
 decayed. These and many other relics of past 
 ages were gathered from the rocks of the Milk 
 River Ridge. 
 
 I was not sorry to hurry away from these 
 dreary, though highly interesting bad lands of 
 the Milk River district, and rid myself of two 
 men who were pessimists of the first water. We 
 reached Lethbridge July 26th, where men and 
 travelling equipment were at once returned, and 
 the hearts of an Indian and his bride, who 
 arrived at that moment both seated astride an 
 Indian pony, were made happy with the remains 
 of our provision larder. 
 
 During the winter of 1887, 1 had, through cor- 
 respondence with the Rev. Leo. GSfetz of the Red 
 Deer Crossing (about 100 miles from Calgary), 
 arranged to have a large flat bottomed boat built 
 in which to descend to and below the confluence 
 of the Red Deer and South Saskatchewan rivers. 
 I left Lethbridge July 27th, and arrived at Cal- 
 240 
 
AMONG THE ROCKS. 
 
 gary two days later, where a short time was 
 spent on the rocks of the Bow River, where the 
 Canadian Pacific cuts through high banks of 
 gravel, sandstone, and shales. The two latter 
 series of strata hold many fragments of fossil 
 plants, but the most prolific rocks in fossil flora 
 are those of the " Hog's Back," one mile down 
 stream from the railway depot. Here in a fine 
 display of the Belly River sandstones, capped 
 with drift deposits which together form a bluff 
 of perhaps 200 feet in height, large fossil leaves 
 of exogeii plants, associated with smaller leaves 
 and flowering plants were found. Some of these 
 interesting fossils have been described by Sir 
 J. W. Dawson in Proceedings of the Royal 
 Society of Canada. Those figured, are from 
 drawings made by my friend L. M. Lambe, the 
 artist of the Geological Survey. These are relics 
 of an epoch ages prior to the time when man 
 appeared and shaped his weapons of war and the 
 chase out of some of the stone found here, and 
 when no human eye had seen the mighty sea, 
 which has left tracings of its winds and waves in 
 these stony records. Early on Thursday morn- 
 ing, August 2nd, perched on the top of a large 
 lumbering stage coach I started for the Red 
 Deer crossing, which, as I said before, is about 
 100 miles north of Calgary. The journey co one 
 not acquainted with prairie life is one of much 
 interest. Our first night was spent at a prairie 
 Q 241 
 
REMINISCENCES 
 
 farm where, after a good supper of pork aud 
 potatoes, with six other men, rolled in buffalo 
 robes, we slept on the floor. 
 
 At five the following morning the horn sound- 
 ed for breakfast, and at 7 o'clock we were again 
 on our road. The heavy rains of the two pre- 
 vious weeks had made much of the trail almost 
 impassable, and many times the horses were knee 
 deep in black peaty muck. Our stage broke 
 down several times. Only an experienced 
 driver, which we had, could have got through 
 these sloughs. One of the interesting features 
 of a journey to Edmonton from Calgary is the 
 meeting of hundreds of " freighters," all half- 
 breed Indians, men who from infancy have toiled 
 along the prairie trails with their heavily laden 
 carts. 
 
 We reached the small village of the Red Deer 
 Crossing Aug. 5th. Here I found the boat I had 
 expected to be ready, only just commenced, so I 
 pitched my tent and waited. The Red Deer 
 Crossing village consisted of two general stores ; 
 one run by the son of the Rev. Leo Geatz, who 
 has a residence and farm about a mile distant, 
 and one, during the absence of the owner, kept 
 by Mr. W. B. Vennor, brother of the late H. G. 
 Vennor, late of the Geological Survey of Can- 
 ada. The latter gentleman will be remembered 
 by many as at one time holding the important (?) 
 position of Canadian weather prophet. These 
 242 
 
AMONG THE ROCKS. 
 
 two stores and a few scattered houses form the 
 welcome resting places of many weary freighters 
 before and after crossing one of the most rapid 
 streams of Alberta. 
 
 I failed to find a man who had been more than 
 ten miles down the Red Deer River from the 
 Crossing, in the summer, and was glad to get two 
 half breeds who were willing to leave civiliza- 
 tion for an uncertain length of time, and ven- 
 ture down a part of a river unknown to them or 
 any of their associates. But at last these two 
 men were engaged, and while waiting the com- 
 pletion of our boat, which was being built by 
 half-breed Indians from pine cut on the bank of 
 the river a few miles below the crossing, and 
 sawn for that purpose, I had several rambles 
 with my friend Vennor who years ago had left 
 his position in the Montreal Bank to wander in his 
 "adopted country" who pointed out many 
 places of geological interest and caused me to 
 assist him in making a collection of the wild 
 flowers of that district which appeared to be, in 
 variety and abundance, more beautiful than in 
 any other portion of the North-West I had been 
 over. 
 
 Monday, Aug. 13th, our boat was launched, 
 and after taking dinner in the parsonage and 
 farm-house of the Rev. Mr. Geatz and his charm- 
 ing family we three men, with camp equipment 
 and three months' provisions, embarked in our 
 243 
 
REMINISCENCES 
 
 rudely constructed boat, and drifted down the 
 swift current, watched by all the members of the 
 Geatz homestead, who made their adieus and 
 wished us God speed. I think from their anxious 
 faces they had doubts as to whether our boat 
 would carry us far, before some calamity over- 
 took us. If they did think that, they were right, 
 for in a short time I found that we were incap- 
 able of guiding our heavy-laden boat through the 
 swift waters, in fact my men seemed to know 
 nothing about navigating rapid streams. Part 
 of our first night was spent in trying to patch up 
 our boat, and the following morning we succeed- 
 ed in reaching Mr. McKenzie's farm, eight miles 
 below the crossing, and the last settlement on the 
 banks of the Red Deer River, before entering 
 an unsettled and little known portion of that 
 land. 
 
 Mr. McKenzie, a noble specimen of the half- 
 breeds of that district, was greatly surprised to 
 find us attempting to descend the river equipped 
 as we were, and strongly advised me to abandon 
 the journey, promising if I did, to have another 
 boat built by the following spring and to 
 accompany me himself. This I decided to do, 
 and in a short time boat and part of our provis- 
 ions, camp equipment, &c. was stored, my men 
 settled with and soon I was on my way back by 
 stage to Calgary. It seemed hard after so much 
 trouble to abandon my long looked for trip, 
 244 
 
AMONG THE ROCKS. 
 
 but as McKenzie said we could not afford to 
 lose our lives for a little geological work. 
 
 I reached Calgary Aug. 18th, and Irvin, or 
 Ross Coule'e, the following day I have already 
 spoken of this interesting locality and after 
 spending a few days gleaning new facts regarding 
 the geology of this district, returned to Ottawa, 
 which place I reached August 30th . 
 
 On Sept. 9th, I was again on the Joggins 
 Coast, Nova Scotia, and with my old friend Mr. 
 James Devine,a coal miner who to assist me for a 
 short time, had left his pick and shovel far down 
 in the bowels of the earth was again searching 
 the shore rocks for new geological information. 
 It must indeed be a poor student in geology and 
 palaeontology who cannot find abundant material 
 for study from the magnificent cliffs of the 
 Joggins Coast, which are constantly crumbling 
 away and exposing to view erect sections of 
 fossil trees, trees in the base of which small rep- 
 tiles some of the first representatives of verte- 
 brate animals of this world crept for safety, 
 trees whose branches overhung shallow waters 
 in which some of the first fishes known by these 
 fossil remains, sported ; waters on the shores of 
 which grew delicate and noble ferns, grasses and 
 other plants ; representatives of our early endogen 
 and exogen flora. 
 
 The view of the Carboniferous rooks as repre- 
 sented on the Joggins coast is one of great inter- 
 245 
 
REMINISCENCES 
 
 est ; its massive beds of sandstones intersected 
 by bands of coal, beds of bituminous shale hold- 
 ing millions of bivalve-looking fossils, which 
 some writers think belong to the Crustacea 
 family; beds of limestone in which scales of 
 extinct fishes are mingled with delicate jaws of 
 minute reptiles. But to unravel and read from 
 these stony records an epoch when the great coal 
 deposits of Nova Scotia were formed has taken 
 years of patient work, to which the writer of 
 these notes is proud to have been privileged to 
 add his mite. 
 
 A stage journey of about three hours brought 
 me again to Macan where, after dinner and the 
 shipping of my specimens one of which was the 
 large section of a fossil tree, now exhibited in 
 one of the upright cases of the geological musem 
 I boarded the west bound train on the Inter- 
 colonial Railway, and two days later was again 
 in Ottawa. 
 
 246 
 
CHAPTER XII. 
 
 SUPPOSED FOSSIL TREES A VISIT OF INVESTIGATION TC 
 KINGSTON I START FOR THE RED DEER RIVER 
 MR. MCKENZIE JOINS ME AT CALGARY EXTENSIVE COAL 
 SEAMS FINE DINOSAUR REMAINS CANADA HA.S PONE 
 COMPARATIVELY LITTLE TOWARDS RESTORING THESE 
 DENIZENS OF A SILENT WORLD DANGEROUS QUICK- 
 SANDS WE CELEBRATE DOMINION DAY A LONG RIVER 
 TRIP ANOTHER TRIP TO THE CYPRESS HILLS UN- 
 PLEASANT EXPERIENCES ON THE BORDERS OF LAKE 
 SUPERIOR. 
 
 DURING this year, 1888, much discussion 
 prevailed concerning the discovery of sup- 
 posed fossil trees, many examples of which are 
 seen in the Cambrian (Potsdam) formation on 
 the banks of the Rideau Canal, Kingston, Ont. 
 The attention of the director of the Canadian 
 Geological Survey was drawn to these singular 
 forms ; he visited the locality and caused a sec- 
 tion of one measuring four feet in diameter to 
 be sent to the museum in Ottawa. (Notes on 
 concretionary structure, by the writer ; Transac- 
 tions of the Nova Scotian Institution of Science, 
 Series 2, Vol. 1.) 
 
 On the director's return I was requested to 
 
 visit the locality, and on the 16th of October, 
 
 accompanied by Mr. Topley, the Government 
 
 photographer, we started, and on arriving at 
 
 247 
 
REMINISCENCES 
 
 Kingston found several people much interested 
 in the discovery of these so called " fossil trees." 
 In a shop window we saw a section of one labelled 
 " Section of a stone tree." Hiring a team we 
 drove through the historical city of Kingston. 
 Our twelve mile drive to the celebrated quarry 
 where these so-called fossil trees occur was a 
 pleasant one. 
 
 We found quite a number of these cylindrical 
 forms, some examples would weigh many tons ; 
 some of the weathered sections show irregular 
 concentric rings which resemble somewhat the 
 lines of growth in exogen plants. No fossil trees 
 have yet been discovered in the Cambrian form- 
 ation to which these deposits belong, and the 
 conclusion arrived at was that these tree- like 
 bodies are of concretionary structure, formed 
 probably in geyser cavities. The "potato-stones" 
 found in these sandstones are nodules, formed of 
 the same material which composes these rocks. 
 
 The photographs taken at this interesting 
 locality are the property of the Geological Survey 
 and can be seen at that institution. This com- 
 pleted my geological journeys for the summer. 
 The winter of 1888, was fully occupied in office 
 and museum work ; work to me always interest- 
 ing and instructive, and of tener than not a labor 
 of love. 
 
 On Saturday, June 1st, I again started for the 
 Red Deer River. Dr. G. M. Dawson. now Di- 
 248 
 
TREE-LIKE (_'<.< KETIONS AT KINOST^A*. O.ST 
 
AMONG THE ROCKS. 
 
 rector of the Geological Survey, left at the same 
 time for British Columbia, and we travelled to- 
 gether as far as Calgary, which place we reached 
 June 6th. 
 
 Comfortably seated in one of the luxurious 
 Pullman cars of the Canadian Pacific Railway I 
 often thought of our first journey over a portion 
 of this railway, when we shared part of a freight 
 car with three horses, dismembered waggons, 
 bales of hay, etc., and took our meals in tents 
 and rude buildings by the track-side. 
 
 At this time, during the construction of the 
 C.P.R. through the districts of Saskatchewan, 
 Assiniboine, and Alberta, the red man of the 
 plains, together with their dark-skinned squaws 
 and children, all attired in bright colored blank- 
 ets, beaded leggings and moccasins, feathers and 
 paint, assembled at various points of the track 
 to watch the doings of the white man. As the 
 status of the Indian is estimated by the number 
 of ponies he has, these " cayuses " always accom- 
 pany their owners, who flocked to watch the 
 iron horse plough its way through their late 
 hunting grounds, and to pick up a few pence 
 from the travelling tenderfoot in exchange for 
 polished buffalo horns and other trinkets made 
 in their " spare time." 
 
 Although only a dozen years or so have passed 
 since our first journey over this country, it is 
 remarkable what a change has taken place in the 
 249 
 
REMINISCENCES 
 
 appearance of many of these children of the 
 plains. The blanket, by most, has been discarded 
 for the cast-off clothing of the white man, and 
 many bucks and squaws out of sheer necessity 
 are working alongside the white man. Still 
 there are many elderly aborigines who scorn to 
 labour, and will, so long as life and memory last 
 adhere to the pursuits of their forefathers. 
 
 June 6th. Mr. McKenzie a fine example of 
 the half-breed Indians of the Canadian North- 
 West Territory joined me at Calgary. He had 
 made the journey of a hundred miles or so with 
 waggon and horses from his farm on the banks 
 of the Red Deer River, to meet and convey me 
 to my starting point down this so far as verte- 
 brate palaeontology goes one of the most inter- 
 esting rivers of the North- West. 
 
 After another short examination of the fossili- 
 ferous rocks of Calgary, from which several fine 
 specimens of fossil leaves one 12 inches in length 
 were collected, and a supply of provisions laid 
 in for our journey, we started for the McKenzie 
 farm. I have spoken of my previous journey by 
 stage, from Calgary to the Red Deer Village. 
 Our present trip by waggon was much less inter- 
 esting, and occupied a day longer. Part of it 
 was made on the Edmonton stage road, and part 
 over the plains, till we reached the banks of the 
 Red Deer River, opposite the McKenzie farm 
 eignt miles or so below the Red Deer Village 
 250 
 
AMONG THE ROCKS. 
 
 which place we reached at 10 p.m. July 13th. 
 Our horses had long scented their stables and 
 were very impatient to cross the river, then a 
 very rapid stream which had risen a foot since 
 " Mac." left, and it was a question whether we 
 would not have to go to the ferry, eight miles up 
 stream. Mac. gave me the reins while he exam- 
 ined his sounding marks on the banks of the 
 river. The night was cold, and in the distance 
 we heard the barking of numerous prairie wolves 
 or Coyotes (Canis latrans) ; I had anything but a 
 soft job to hold in our horses till Mac. returned 
 and said we would cross on his lower ford. Our 
 camp equipment was arranged to keep it from 
 the water, which Mac. expected would cover the 
 floor of the waggon, then he took the reins and 
 with the cheering words, " keep your eyes closed 
 and don't move/' allowed the horses to plunge 
 into the stream. For a moment I felt the wag- 
 gon sway with the current and the water covered 
 its floor, but in another minute or two our horses 
 gained a footing on the opposite shore, and with 
 a last brave pull ascended the steep bank, and in 
 a few moments I was comfortably seated before 
 a large fire in the hospitable home of one of the 
 early pioneers of the Red Deer River district. 
 
 June 15th was spent at the mouth of the Blind 
 
 Man River, eight and a half miles up stream, 
 
 where a fine geological section of the Edmonton 
 
 series is seen. One of the beds of rock here holds 
 
 251 
 
REMINISCENCES 
 
 beautifully preserved fossil plants ; leaves which 
 have been pressed in nature's stony-book 
 pressed so closely that every vein is plainly seen. 
 Our small but valuable collection of fossil plants 
 from this locality, together with photographs 
 showing the confluence of the Blind Man and 
 Red Deer rivers, form an interesting souvenir of 
 this picturesque spot where we saw a graceful 
 deer come to the water's edge, drink and then 
 glide away into the bush. 
 
 My friend McRenzie, who was versed in several 
 Indian languages, and had spent many years 
 among the aborigines of the Canadian North- 
 West, gave me many interesting accounts of the 
 early settlers and explorers of the Alberta and 
 other districts ; which reminded me of having 
 read that among the earliest explorations of 
 which any record remains, we find the name of 
 M. Bigot, the French Intendant, who after his 
 succession to the intendancy in 1748, sent out 
 explorers to spy the land in various parts of 
 unsettled Canada, hoping, it is presumed, to add 
 to his already luxurious and libidinous ways of 
 living. 
 
 In Alberta, as in almost all sections of the 
 North- West Territories, many of the beautiful 
 names given by the Cree and other Indians to 
 lakes, rivers and mountains have been changed 
 by the white man, but between themselves I 
 found the Indians used the old Indian appella- 
 252 
 
AMONG THE ROCKS. 
 
 tions, and even the half-breed seemed to resent 
 the re-naming of rivers, localities, etc., a thing 
 too much done by young explorers of the present 
 day. 
 
 The following morning, June 17th, after a 
 successful day's work at the confluence of the 
 Blind Man and Red Deer rivers, the McKenzie 
 family, typical representatives of the half-breed 
 Indians of the Canadian North -West were 
 assembled and photographed, then our two boats 
 were launched and provisioned, and at 2 p.m. we 
 waved adieu to our friends of the last habitation 
 we should see for many days, and drifted swiftly 
 down a portion of a river unknown in the 
 summer to any of the settlers I had met with 
 in Red Deer village. This is probably due to 
 the swiftness of the current, which in places is 
 so rapid that it is almost impossible to track a 
 boat up stream again. The waters of the Red 
 Deer flow from one of the snow-clad interior 
 ranges of the Rocky Mountains, vhere, after 
 coursing through the foot-hills it is a smooth 
 stream of clear blue water two hundred feet 
 wide. 
 
 Soon after leaving the McKenzie farm we 
 reached the " canon " in which are high, steep 
 and scarped banks. Here we found the water 
 rushing between large Laurentian boulders, ren- 
 dering navigation dangerous ; to me the scene 
 was very fascinating. Once the bow of my boat 
 253 
 
REMINISCENCES 
 
 was driven on a projecting boulder, and in an 
 instant the stern swung around and our flat 
 bottomed craft was heading in that fashion down 
 stream. I heaved a sigh of relief when a few 
 moments later we swept into smooth water. In 
 a short time we reached the great, yet (1889) 
 undisturbed coal seams of this district. Here an 
 almost vertical section of seams of coal rises fifty 
 feet above the water's edge and as far as can be 
 seen below the surface of the river. In 1858, 
 when Dr. Hector navigated this stream, large 
 exposures of this coal were seen to be burning, 
 and now the large outcrops of red and yellow 
 burnt coaly matter lend a picturesque aspect to 
 this vicinity. 
 
 Bold escarpments several hundred feet high 
 occupy portions of the river-sides for many 
 miles, and from the table- land, sloping valleys, 
 thickly wooded in places, form lovely retreats 
 for the numerous wild animals that roam the 
 sheltered recesses. Along the shore of the river 
 and muddy flats we noticed the tracks of grizzly 
 and smaller bears, coyotes or prairie wolves, 
 woodchucks or ground-hogs, the prairie dog, 
 and other rodents. I must not forget to men- 
 tion the beaver, individuals of which tribe were 
 seen in several places along the margin of the 
 river. One fine fellow was busily engaged in 
 the construction of a dam, but disappeared the 
 moment he caught sight of us. Outside a deserted 
 254 
 
AMONG THE ROCKS. 
 
 hunter's shack near the confluence of the Red 
 Deer and South Saskatchewan rivers, I saw 
 several skeletons of these animals. It is a pity 
 that this valuable rodent, which together with 
 the maple leaf, forms our Canadian emblem, 
 should be becoming extinct on many of our rivers 
 and streams, where a few years ago they were 
 numerous. The rancid spelling oil secreted near 
 the root of the tail of the beaver furnished the 
 castor oil of the early medical man. I remember 
 reading, I think in some book of fables, of these 
 animals, when being pursued, biting out these 
 oil-glands and casting them before the hunters. 
 We all know that the castor oil now used is 
 made from a bean. 
 
 As our boats glide silently past numerous small 
 thickly wooded islands and occasionally verdant 
 valleys, we constantly hear the twittering and 
 sweet song of birds, among which is the soft 
 cooing of the wood-dove. The high sandy buttes, 
 where little vegetation, excepting wild sage 
 grows, is the home of the bald-headed and other 
 eagles. On the ragged projecting portions of 
 these escarpments, out of reach of the prowling 
 coyotes, these noble birds build their nests of 
 twigs and buffalo bones; I found in one nest a tin 
 tea spoon. 
 
 In the twilight of the evening the hoarse cry 
 of an owl is frequently heard, and the king-fisher 
 is seen to dart from his river-side perch and 
 255 
 
REMINISCENCES 
 
 swoop down on an unlucky " gold- eye " who has 
 ventured too near the surface of the water. 
 
 The banks of the Red Deer River are a favorite 
 resort for wild geese ; the proverbial stupidity of 
 these birds is very noticeable. Many times after 
 our night camp, we saw a few hundred yards 
 away, a flock of geese feeding on short mossy grass 
 which grows on the muddy shore of the river. 
 They remained feeding till our boat came within 
 gun shot, then there was a hasty, partly running 
 and partly flying, not into the bush or fields, but 
 straight ahead of our boats, stopping to feed when 
 we camped for lunch, or landed to examine some 
 favorable spot for fossils, and then on again till 
 night. Sunday June 23rd, we were fairly in the 
 bad-land district of the Red Deer River. The 
 great sandy buttes and escarpments of this part 
 of Alberta extend for many miles, sometimes 
 coming close to the river side, and in other places 
 lying back, leaving room for grassy and sandy 
 flats. These great exposures belong to the Lara- 
 mie or upper Cretaceous formation. In this geo- 
 logical horizon, far below the Miocene Tertiary 
 in which the remains of an extinct rhinoceros, 
 already referred to, were found is a series of 
 rocks of special interest to the osteologist as they 
 contain the remains of great Dinosaurs, those 
 terrible flesh-eating reptiles which inhabited cer- 
 tain portions of our North-west, at a time when 
 what is now prairie land and rivers was occupied 
 256 
 
AMONG THE ROCKS. 
 
 by great lagoons and jungles. On one of the 
 great sand and shelving sandstone buttes, a hun- 
 dred or more feet above the river, and in close 
 proximity to several eagles' nests, we came on the 
 remains of one of these huge Dinosaurs (deinos, 
 terrible ; saura, lizard) the skull and other bones 
 of which now occupy a prominent position in our 
 Geological Survey Museum. In the sandstone 
 beds that contain these remains I first saw the 
 two lower jaws ; the right ramus covers a por- 
 tion of the left, concealing the teeth, which are 
 seen in the other jaw. A close examination 
 showed that the rami lay upon the roof or palate 
 of the cranium. Only the roots of the teeth of 
 the upper jaws remained, but the teeth of one of 
 the lower maxillaries is almost perfect large, 
 curved, and beautifully serrated at the edges. 
 There were also portions of limb- bones and claws 
 dreadful claws the sight of which carried one 
 back into past ages. 
 
 Besides these remains we found slabs of sand- 
 stone covered with ripple marks and rain-prints 
 showing us that in those remote ages before any 
 human eye had gazed on the wonderful works of 
 our Creator the wind blew and the rain fell as 
 at the present time. Overlying this Dinosaurian 
 grave the bones in which have been since deter- 
 mined by Prof. Cope to be Lcelaps incrassatus 
 were slabs of rock holding beautifully preserved 
 leaves of exogen plants, and grasses, 
 K 257 
 
REMINISCENCES 
 
 In the geological section of this earth's crust, 
 these fierce Dinosaurians occupied a period in 
 the animal kingdom millions of years after the 
 Devonian and Carboniferous age, when the first 
 fish and smaller reptiles appeared on life's stage. 
 
 When one contemplates the vast ages which 
 have rolled away since the days when these great 
 kangaroo-like saurians disappeared, never again- - 
 we may presume to be repeated on this planet, 
 and then think that the Laramie rocks of which 
 we are speaking, only occupy one leaf of nature's 
 great stony records, our minds are bewildered 
 among the immensities of the past. Scientists 
 tell us that while it takes or has taken 730 years 
 to deposit one foot of some of our sedimentary 
 rocks, other formations have taken 6,800 years to 
 deposit one foot. Thus, remarks one writer, the 
 period of time required to build up 100,000 feet 
 of sedimentary rock has varied according to 
 locality, from 37,000,000 to 680,000,000 of years. 
 
 Leaving the high buttes and desert-like sandy 
 flats which have yielded so many important geo- 
 logical specimens food for the mind and con- 
 tinuing down the beautiful but sometimes treach - 
 erous Red Deer River, we pass through verdant 
 valleys, in places 150 feet below the prairie level. 
 A flock of wild geese keeps ahead of our boat 
 and a beaver slides down the bank into its shelt- 
 ered dam, a little further on as our boat glides 
 swiftly over a long stretch of rapids we pass a 
 258 
 
AMONG THE ROCKS. 
 
 hungry -looking coyote, who is evidently looking 
 for a breakfast of goose, and towards evening 
 find ourselves among a series of similar rocks to 
 the Laramie, but lower in our geological section ; 
 they belong to that part of the Canadian Cretace- 
 ous known as the Belly River Series. As in the 
 Laramie formation, we find these sandy buttes 
 and escarpments strewn with fossil bones, chiefly 
 fragmentary here several vertebrae twice the 
 size of those of the buffalo ; here the distal end 
 of a femur or thigh bone, which must have be- 
 longed to a larger animal than that great ele- 
 phant " Jumbo "whose skeleton now adorns one 
 of the American museums. These are the bones 
 of Dinosaur reptiles who lived prior to those al- 
 ready spoken of. But besides these great carni- 
 vorous beasts there lived in those remote ages, 
 huge herbivorous Dinosaurs, one of which has 
 been named by Prof. March Stegosaurus ungul- 
 atus. It is remarkable for a series of armour 
 plates on its back and for its small head, com- 
 pared with the size of its thigh bone. Scatt- 
 ered on these sandy flats are the remains of 
 turtles, scales of Ganoid and other fishes creat- 
 ures who existed ages before the Mastodon and 
 Mammoth appeared, and at the time when the 
 lignite coal of our North-west Territories was 
 being deposited. The restorations and writings 
 of Cope and March furnish us with a grenf 
 amount of information about these extinct ani- 
 259 
 
REMINISCENCES 
 
 mals, especially about those remains found in 
 the United States, where a large sum of money 
 has been spent in pursuing this most interesting 
 branch of fossil osteology. Compared with the 
 researches of the United States, however, com- 
 paratively little has been done in Canada towards 
 the collecting and restoring of these wonderful 
 denizens of a silent world silent except for the 
 roar of these wild beasts and the occasional rag- 
 ing of the tempest. While wandering alone over 
 one of these desolate bone-strewn spots, and 
 wondering whether any human foot had ever 
 trodden the sandy ground, I stumbled over a 
 human skull. It was bleached and weathered 
 and is probably the skull of an Indian who lost 
 his life in one of the fierce battles waged between 
 the Bloods and Crees. This skull is remarkable 
 inasmuch as the atlas or first cervical vertebra 
 is firmly ankylosed to the occipital or base of the 
 skull, which peculiar defect must have deprived 
 Mr. Indian of the pleasure of turning his head 
 either to the right or left. This skull is now in 
 the ethnological room of the Geological Survey 
 Museum, Ottawa. 
 
 The patches of quicksand met with in parts of 
 the river bottom and on the flats near the stream, 
 often caused us much trouble. In some parts of 
 the river where the water spreads out consider- 
 ably our boat, though not drawing more than 
 six inches of water, frequently stuck fast in the 
 260 
 
AMONG THE ROCKS. 
 
 sands which are continually drifting with the 
 current, and will in a short time completely sur- 
 round one's boat unless it is immediately got out 
 into deeper water, not always an easy task to 
 accomplish as our boat-pole sunk its full length 
 without finding bottom. On one occasion I was 
 much struck with the danger of quicksands. I 
 had been rambling alone over some of the high 
 sandy buttes in Range 21, Tp. 33, and when re- 
 turning to our tents stepped on what I thought 
 hard sand, in a dried up brook-course, only two 
 yards or so wide, and in a moment down I went, 
 up to my waistcoat pockets. Almost before I had 
 time to think I had stretched out my arms and 
 grabbed a small bush which grew on the opposite 
 bank, and with some difficulty pulled myself out, 
 to find I was covered up to my waist with ice- 
 cold slimy sandy mud. I scraped part of this mud 
 oft with my sheath-knife, and then hastened to my 
 tent which was hidden behind a clump of cotton- 
 wood trees. On turning this obstruction to a 
 straight path I almost stumbled over two coyotes 
 who had evidently been trying to get into our 
 provision boxes. Although these sneaky wolves 
 are very cowardly, it is not pleasant to be in 
 close proximity to them, for when hungry they 
 will sometimes attack anything. 
 
 When my men returned to camp I got Mc- 
 Kenzie to sound the place which had almost en- 
 gulfed me, and when no bottom was found with 
 261 
 
REMINISCENCES 
 
 our fifteen foot boat-pole, Mac looked at me and 
 said : " Well we should have knowed where you 
 had went to, for your hat would have been left." 
 
 The old saying that birds of a feather flock 
 together, holds good regarding the distribution 
 of fossil bones over the bad lands of the Red 
 Deer. After leaving the interesting locality 
 where the Dinosaurian jaws were found, I 
 walked over many sandy buttes and flats with- 
 out finding any tiling of interest. The portion of 
 river we had been travelling for the past few 
 days, passed through valleys 600 or more feet 
 deep. Small but well wooded islands formed 
 sheltered nooks for our night camps. Often the 
 surface of the water fairly bubbled as hundreds 
 of " gold eyes " snapped at flies. 
 
 In this vicinity, while one's boat glides swiftly 
 with the current and one is congratulating him- 
 self that he is making probably ten miles an 
 hour, he finds his boat stuck fast in quicksand. 
 Then it is get out if you can, with your boat pole, 
 but if it is not quicksand one can drag the boat 
 into deeper water. 
 
 Sunday, June 30th, was spent in exploring 
 some of the buttes in the vicinity of Range 21, 
 Tp. 30. Here in places the buttes and escarp- 
 ments came close up to the river. In one of the 
 ranges seams of coal are seen ; further on a burnt 
 butte exhibits in a marked manner the stratifica- 
 tion of various coaly layers or beds, some of 
 262 
 
AMONG THE ROCKS. 
 
 which are of a bright Indian red while others 
 are of various yellow tints. The river scenes in 
 this vicinity are charming, a flock of wild geese 
 keeps ahead of our boats, a large eagle hovers 
 above us, and the constant twitter of birds adds 
 to the pleasure of our journey. The following 
 morning was bright and beautiful. My men had 
 been looking forward to this day for some time, 
 for it was Dominion Day, and we were to cele- 
 brate it by tapping our only bottle of brandy 
 which I had guarded diligently in case of getting 
 a rattlesnake bite. Hitherto our " nips " had 
 been made of a decoction of pain-killer, sugar 
 and hot water, a splendid drink when one is 
 " chilled to the bone." We were early in our 
 boats, and while drifting with the current, passed 
 several beaver dams, shot a wild goose, and had 
 climbed a high butte by 11.30 Here we hoisted 
 a long pole we had brought up, which with a 
 large red pocket-handkerchief served as a flag 
 staff. Having christened the elevation '*' Domin- 
 ion Butte," fired a salute, sung " God Save the 
 Queen," drunk to all absent friends and relations, 
 we descended to our boats which we reached just 
 in time to find that they were being, inspected by 
 several Indians. These were the first people of 
 any sort we had seen since leaving the McKenzie 
 farm. We found that these red skins were on 
 their way to a far distant lake, where, through 
 some of their tribe, they had heard that there 
 263 
 
REMINISCENCES 
 
 were buffalo. Mac got uneasy and motioned me 
 to get in the boat, which I did at once, and as we 
 shot down a stretch of rapids he said : " It's a 
 
 d good job, while we were away, they did 
 
 not take boats, provisions and everything they 
 could lay their hands on." 
 
 For two or three days after leaving Dominion 
 Butte we struggled with sand bars (for the river 
 was very low), mosquitoes, head winds, etc. But 
 as an offset to these little troubles we had rousing 
 camp-fires and jolly suppers of fish (gold eyes), 
 which were so our cook expressed it " too 
 quick to take the bait, and so spoilt the sport." 
 In this section of the country, one must be a 
 good sleeper, indeed, not to be continually dis- 
 turbed with prowling rodents, the barking of 
 coyotes and the croupy cry of the big grey 
 owl. 
 
 But morning comes, and with a fair wind and 
 rudely constructed sail at the stern of our boat, 
 which seems to dance over the rippled waters, 
 we forget the troubles of yesterday and those 
 ahead of us. 
 
 July 5th, we had reached Dead Lodge Canon, 
 Range 13, Tp. 24. In this vicinity the river cuts 
 through a fertile valley from 400 to 600 feet 
 deep. Here nature had used her scooping shovel 
 to an enormous extent, for between the prairie 
 level and the river sandy buttes interstratified 
 with bands of sandstone form pyramid -like 
 264- 
 
AMONG THE ROOKS. 
 
 structures. Here to my delight I found on the 
 bleached sands numerous fossil bones, and further 
 search revealed the fact that this at one time 
 had been the home or rendezvous of both flesh- 
 eating and herbivorous extinct animals, among 
 which figured our old friend, Lcelaps incrusatus, 
 Cope, for here in the cut side of a high butte sticks 
 out the proxhemal end of a femur which we know 
 to be that of a Dinosaur. Three hours were spent 
 digging out this bone, and three pairs of hands 
 carefully lifted our precious specimen to put it 
 in the rude box we had made from part of the 
 upper floor of our boat, when to our surprise the 
 thing crumbled into a thousand fragments. Many 
 of the bones found in these formations are solid 
 stone, the bony pores having been filled with 
 silicious matter. In this case, however, only a 
 thin outside crust remained, while the interior 
 was a decayed mass which crumbled into dust 
 when exposed to the atmosphere. 
 
 In these great sand banks must be buried 
 thousands of tons of fossil bones, for as the 
 weathering away goes on, these relics of distant 
 ages weather out and remain on the sandy flats 
 or roll to the foot of the banks and cliffs. These 
 bones belonged to animals that existed in that 
 wonderful epoch called the Reptilian age, turtles 
 six feet long held their own with the great kanga- 
 roo-shaped carnivorous reptiles who with their 
 curved knife-like teeth would have been formid- 
 265 
 
REMINISCENCES 
 
 able foes even to the great Mammoth and 
 Mastodon who did not appear till ages after the 
 great Dinosaurs " had laid their bones down," as 
 one of the men expressed it. It was with great 
 reluctance I left this interesting bone locality, 
 but provisions were getting scarce, and we still 
 had to explore many miles of a country unknown 
 excepting for the valuable map of this river, 
 compiled by Messrs. McConnell and Tyrrell of the 
 Canadian Geological Survey. 
 
 Sunday July 7th, we waved adieu to what is 
 probably the most important field in Canada, so 
 far as bones of extinct animals is concerned. It 
 would take years to glean from these great sandy 
 buttes, flats and cliffs even a part of the inform- 
 ation they hold relating to that period when the 
 Cretaceous rocks of Canada were laid down a 
 time contemporaneous to that when crocodilian 
 reptiles swarmed the banks of the river, now 
 part of Philadelphia, and the lagoons of 
 Penny si vania. Continuing our journey we still 
 pass elevations varying in height to perhaps 600 
 feet, but the Unio and other bivalve shells they 
 contain tell us that they are a little higher in the 
 geological horizon. 
 
 The river is still low, and though our boats 
 only draw seven or eight inches of water we 
 have continually to cross and re-cross, keep- 
 ing them in the small channels made by the 
 current. 
 
 266 
 
AMONG THE ROCKS. 
 
 Laurentian boulders of almost every variety 
 are scattered over the table-land, the buttes, and 
 the bed of the river. 
 
 Sunday 14th, we reached the confluence of the 
 Red Deer and South Saskatchewan rivers. In this 
 vicinity there are some fine flats well timbered 
 with cotton-wood, black birch, elder, willow and 
 other woods. During the past few days we have 
 seen many flocks of wild geese, a few ducks, 
 several coyotes, and Mac had no trouble in 
 supplying our camp with fish, but with the excep 
 tion of an occasional black bass it was always 
 " gold eyes." 
 
 Near the junction of the two rivers we came 
 on two " shacks." These were the first signs of 
 civilization we had seen since seeing the small 
 band of Indians at the foot of Dominion Butte, so 
 we went ashore (as Mac said) to see how the cat 
 jumped. Several sharp raps on the first door 
 brought no response, and as the door was not 
 locked we entered and found probably all the 
 worldly possessions of a North- West trapper. 
 A candle stuck in a black bottle adorned a rough 
 home-made table, several guns and a rifle hung 
 on the wall, there was a cupboard in one corner 
 of the room, but " when we went to the cup- 
 board we found it was bare." At the back of 
 the shanty we found the festering remains of 
 coyotes, beavers, skunks and other animals. 
 With a blazing sun and the thermometer register- 
 267 
 
REMINISCENCES 
 
 ing 90 in the shade the stench of this decayed 
 matter was so great that we only remained to 
 tack our cards on the door. 
 
 Here at the junction of the two rivers the 
 Saskatchewan takes a northerly course and at a 
 distance of about 70 miles, in a straight line 
 but probably about double that distance by 
 water passes through Medicine Hat (of which I 
 have already spoken) where one of the fine iron 
 bridges of the Canadian Pacific Railway spans the 
 stream. But from the mouth of the Red Deer 
 we continue westward. 
 
 Monday July 15th, at 8a.m. the thermometer 
 registered 80 in the shade. For the past few 
 days we had been drifting with the current down 
 the Saskatchewan where in many places the 
 water is both swift and deep quite a relief after 
 the shallows of the Red Deer. Mac tells me that 
 these rivers, after heavy rains, rise six inches in one 
 night, so the next persons to follow our tracks in 
 search of geological treasures, may be more 
 fortunate, and not even know half the difficulties 
 one experiences when the water is unusually 
 low. 
 
 The distance from the mouth of tke Red Deer 
 River to the Battleford and Swift Current cross- 
 ing is not counting the many windings of the 
 river about 100 miles. A great part of this 
 distance is uninteresting to the geologist; as the 
 river is flanked on both sides with mud banks 
 268 
 
AMONG THE ROCKS, 
 
 and the low valleys are covered with vegetation ; 
 the sportsman, however, can find plenty of work 
 for his gun and dogs. 
 
 Leaving our Belly River rocks so named 
 because they are largely developed on a river of 
 that name which have yielded us so many 
 interesting fossil bones, we 20 miles or so be- 
 fore reaching the Saskatchewan Landing come 
 to a series of strata made up of sand, clays, muds, 
 sandstone and limestones. They represent the 
 upper part of the Canadian Cretaceous known as 
 the Pierre and Fox Hill Group. We hastily pass 
 over these few remaining miles, and at noon, 
 July 19th, reach the Battleford and Swift 
 Current crossing. Here we find a small settle- 
 ment of half-breed Indians ; many of these hardy 
 sons and daughters of toil hae been occupied 
 from almost childhood in freighting over the 
 great plains of our North- West. Some of these 
 freighters can tell wonderful tales of the fierce 
 struggles between various tribes of Indians who 
 waged war with one another in this vicinity 
 where the old worn-out freighter now cultivates 
 his little garden, while the stronger members 
 of his family continue their freighting occupa- 
 tion. 
 
 We pitched our tent for the night on the shore 
 
 near a " shack " or log cabin which was occupied 
 
 by one of the North-West mounted police officers. 
 
 Here, in exchange for one of our boats, we 
 
 269 
 
REMINISCENCES 
 
 obtained sugar and pork which we were much 
 in need of. 
 
 The following morning having arranged with 
 a freighter to meet us with a cart, in four days, 
 we drifted down stream about ten miles and 
 pitched our tent on a verdant plateau opposite 
 Swift Current coule'e. 
 
 I have already mentioned that the geological 
 formation of this section of the country is the 
 upper part of the Cretaceous. This formation 
 here, is characterized by high cut banks, buttes 
 and flats. These deposits of various colored 
 clays, sands, etc., in places, hold concretions or 
 nodules of iron-stone, in which are found beauti- 
 fully preserved fossil bivalve and other shells 
 ranging in size from a quarter of an inch to a 
 foot or more in diameter, the latter being the 
 large bivalve Inoceramus, which must have 
 held meat enough for a hungry man's dinner 
 had he existed in those days. 
 
 Among the great variety of fossils found in 
 the upper Cretaceous which represents the 
 upper part of the Mesozoic rocks of Canada so 
 wonderfully developed in this locality, we find 
 several species of Ammonites, representatives 
 of the Cephalopoda family so called because 
 they have some resemblance to the horns of the 
 Egyptian God, Jupiter Ammon. In the rocks 
 and clays of this locality we frequently find 
 portions and sometimes whole specimens of these 
 270 
 
AMONG THE ROCKS. 
 
 forms, some of which measure two feet in 
 diameter, and on the outer shell show all the 
 beautiful opalescent colors as when in ages ago 
 these " shells of the ocean " floated on silent seas 
 seas no human eyes had ever gazed upon. 
 Over these clays are also scattered the fossil 
 remains of animals whose form and character we 
 shall probably never know, for the osteologist 
 who can build representations of some of the 
 extinct vertebrate animals out of a limb bone 
 fails to recognize some of the specimens from 
 this locality. 
 
 Now I have come to our last night's camp on 
 the shores of the Saskatchewan River. Long 
 after darkness had closed in, my two faithful 
 half-breeds and I sat beside a roaring camp-fire 
 and talked of days when the aborigines of the 
 North-West pitched their wigwams on these 
 shores and engaged in all the cruelties of 
 Indian warfare. But hark ! the crackling of 
 our camp fire does not prevent the half Indian 
 instinct of Joe hearing an unusual sound, and 
 in a moment he seizes my rifle and creeps 
 through the bush followed by Mac with a gun. 
 For a time all is quiet, then a rifle shot is 
 heard, and in a short time the two men re- 
 turn to camp, but without any trophy of the 
 chase. 
 
 Presently I turn in and sleep soundly till 
 towards morning, when I awake to find a brilliant 
 271 
 
REMINISCENCES 
 
 reflection of our camp fire on the walls of my 
 tent, and to hear Mac singing : 
 
 ' Prowling beasts about Thy way, 
 Stones Thy pillow, earth Thy bed.' 
 
 In his young days, Mac had attended the 
 Mission School on the banks of the Red River 
 near Winnipeg. 
 
 We found that Joe's rifle shot wounded some- 
 thing, for there were blood stains on the grass. 
 While at breakfast the sight of an Indian 
 freighter wending his way over the rough undu- 
 lating country on the other side of the river, was 
 a signal for us to strike camp and cross the river 
 which just then, owing to the recent rains we 
 had had, was exceedingly swift and, as after all 
 heavy rains, very muddy. In a short time we 
 had crossed the stream and had drap*p;ed our boat 
 
 OO 
 
 above high- water mark, there to leave our good 
 but roughly constructed and now much battered 
 craft, which had carried us safely over hundreds 
 of miles of the two winding rivers we had hastily 
 examined for the remains of a world uninhabited 
 by man. 
 
 Our traps were soon packed in the freighter's 
 cart, and after a journey of ten miles over hills 
 and through valleys we again reached the ferry 
 crossing. 
 
 I spent the night in the police shack while 
 the men camped outside. Early next morning, 
 with our dozen or more boxes of fossils weighing 
 272 
 
AMONG THE ROCKS. 
 
 about a thousand pounds snugly packed in 
 freighter's carts, we left the banks of the Saskat- 
 chewan River, and following the Battleford trail 
 arrived at Swift Current at sunset. The dis- 
 tance, although only thirty miles, presents many 
 interesting features incident to prairie travel, 
 especially when one makes the journey, as we 
 did, with an old-time freighter (and his numerous 
 family) who had spent most of his life on this 
 trail and could point out spots where, as our old 
 guide said, brave Indians buried their tomahawks 
 in each other's skulls. But now here all is peace, 
 and the " iron horse " of that wonderful Canadian 
 Pacific Railway snorts its way through the quiet 
 village of Swift Current, where a few Indians 
 still remain, loath to quit their buffalo hunting 
 grounds of a few years ago, loath also to assist 
 the white man, yet forced to make a scanty 
 living from him. The following morning I bade 
 good-bye to my two genial companions of many 
 days and nights spent " near but yet so far " 
 from the haunts of the white man. 
 
 July 26th. I was again at Maple Creek pre- 
 paring for another trip to Cypress Hills. I have 
 already spoken of the verdant hills and valleys 
 of this bright spot in our North-west Territories, 
 and recorded a few of the fossil remains found 
 in the Miocene Tertiary rocks exposed in the 
 coulees and hills of this locality, so shall hastily 
 pass over the present journey to these interesting 
 
 s 273 " 
 
REMINISCENCES 
 
 rocks rocks which have thrown much light on 
 the fossil fauna of the North-west. 
 
 Having hired a wagon, three horses, one a 
 saddle horse, and two men, and with a month's 
 provisions, we left Maple Creek July 27th, and 
 followed the trail James Macoun and I took in 
 1884. 
 
 Between Maple Creek and our present desti- 
 nation I notice many changes have taken 
 place during the past five years ; land then 
 covered with prairie grass is now under cultiva- 
 tion. In other parts ranches appear, and the 
 typical cow-boy herds his cattle ; further on, 
 where the trail descends into a deep valley, we 
 pass the bleaching bones of a horse " rode to 
 death while tracking an Indian horse-thief " 
 says one of my men. Still further on in a 
 valley we come to four Indian graves; stones 
 that cover the recently turned earth have not pre- 
 vented the hungry coyote from digging up por- 
 tions of the buiied bodies, for fragments of limb 
 bones and a piece of a skull lie near by. Some 
 distance from this locality we came on another 
 Indian grave, but this is in a tree a rudely con- 
 structed wooden box rests on two limbs of the 
 tree, out of the reach of prowling beasts. We 
 looked into this box and found the remains of an 
 Indian woman, who in life was no doubt loved 
 and considered beautiful, for beside her had been 
 placed two skilfully worked bead bags, each con- 
 274 
 
AMONG THE ROCKS. 
 
 taining pigments for painting the face ; a string 
 of glass beads and a few brass ornaments com- 
 pleted the toilet articles which were to adorn this 
 dusky lady of the forest and plain when she 
 entered the happy hunting grounds of her fore- 
 fathers. 
 
 The great numbers of buffalo bones that lay 
 bleaching on the plains and in the valleys are now 
 all gone gone to refine sugar and fertilize other 
 lands gathered up by the same Indians who a 
 few years ago only gathered the hides and a 
 little flesh of the now almost extinct noble bison 
 which roamed these prairie lands in vast herds. 
 Only, in many miles, one skull was seen, and 
 this served as a finger-post, for on it was roughly 
 painted " water 100 yards north." 
 
 We pitched our tents near the head waters of 
 Swift Current where a large coulee entered the 
 main valley of this district. 
 
 About a week had been spent collecting fos- 
 sils and gaining new geological information 
 relating to our Miocene deposits, when to my 
 surprise I found a deadly enmity had sprung up 
 between my two men, one of whom I had days 
 before discovered was a deserter from the Ameri- 
 can Army a man of the very lowest type of 
 the scam of the earth. For several days the 
 sound of this man's blasphemous tongue had 
 grated on my nerves so much that I was almost 
 unfit for work. This together with the loss 
 275 
 
REMINISCENCES 
 
 of one week's provisions, " stolen by our 
 Yankee soldier " said my other man, who was 
 not a bad fellow, decided me to return to Maple 
 Creek ap soon as possible. Quietly loading my 
 rifle while talking to these men I ordered them 
 to load up at once as I intended to sleep at the 
 police camp that night. It was thirty miles 
 away and we had nine hours to do it before 
 sunset. With rifle in hand I mounted my horse 
 and waited patiently till all was ready for the 
 start. Our Yankee was particularly sullen and 
 I feared he would not go ; but he knew full well 
 that if he deserted me he could not appear in 
 Maple Creek again, so at last we started. Fol- 
 lowing the waggon enabled me to keep a sharp 
 eye on Sam, for that was our Yankee's first 
 name. Our road lay through valleys and over 
 high buttes and was an exceedingly tough jour- 
 ney. 
 
 Some one has said that the vilest of men have 
 some redeeming traits, so I found it with this 
 man. Large herds of half wild cattle unmol- 
 ested excepting during the branding seasons 
 wander up and down the valleys of the Cypress 
 Hills district. In one of these valleys we found 
 a young steer mired in a creek which crossed the 
 valley. Without uttering a word our Yankee 
 friend (?) jumped out of the waggon and motioned 
 the driver to stop, then taking two picket-ropes 
 from the waggon beckoned him to assist. In a 
 276 
 
AMONG THE ROCKS. 
 
 few minutes the ropes were fastened around the 
 horns and body of the poor beast, then the horses 
 were taken from the waggon and the ropes 
 attached to the whipple-pole, and the almost 
 exhausted steer dragged to dry land, probably to 
 die, for it was too feeble to stand when we moved 
 on. During our long journey, this was the one 
 redeeming act of our Yankee outlaw. 
 
 Half an hour or so later in a narrow part of 
 the valley we came on a large herd of cattle, 
 creatures who for a few years of their lives 
 range as much at will as the buffalo who once 
 darkened these valleys with their dusky and 
 bulky forms. We were close on these cattle, 
 and for a moment I wondered what these cow- 
 boys would do to get a way through this moving 
 mass of animals. But when we were within a 
 few yards of the head of the column, which 
 appeared to be headed by several very fierce 
 looking bulls, one of the men turned towards me 
 and motioned me to put spurs to my horse and 
 ascend the banks of the valley then each man 
 seized a picket rope, and with great shouts swung 
 them around their heads, at the same time urging 
 the horses on. Out of danger I watched the slow 
 progress of our waggon through this great drove 
 of cattle, which probably numbered two thousand. 
 I breathed a sigh of relief when at last the team 
 passed safely with the exception of a nasty 
 wound one of the horses got. 
 277 
 
REMINISCENCES 
 
 Our way now lay over a very rough portion of the 
 country, so much so that in descending some of 
 the hills all four wheels of the waggon had to be 
 firmly lashed together. But we reached the 
 police station (shacks, or log cabins) safely, where 
 I was hospitably received and sheltered for the 
 night, while the men camped outside. To the 
 sergeant in charge I related the trouble I had 
 had with my men and that serious trouble might 
 happen before I got them back. The following 
 morning, Sunday, I found a splendidly-mounted 
 police officer had been detailed to escort me 
 " into town," he had received instructions not to 
 leave my side till we were in Maple Creek. We 
 rode ahead of the waggon, but twice the police 
 officer had to return, and presenting his revolver 
 order the driver to "get a gait on." We 
 reached our lunch camping-ground, where the 
 buffalo's skull was, with the notice ; " water 100 
 yards north " ; then after giving our horses an 
 hour's rest continued and reached Pie-pod's Creek 
 and at 6 o'clock our camping ground for the 
 night. Here we found ^ large encampment of 
 Indians, aborigines who prefer a roaming life 
 and often hungry stomachs rather than remain 
 on the Government Reserves with daily rations 
 of food. This (Bear Creek) is a favourite camp- 
 ing ground, and just then there appeared to 
 be an unusual gathering of red- skins, for some 
 special purpose. We would have preferred 
 278 
 
AMONG THE ROCKS. 
 
 spending the night elsewhere than in such close 
 proximity to these Indian camps, but as this was 
 the only place we could get water, for some 
 miles, we had no option. 
 
 After supper our horses were safely picketed a 
 few yards from our tents, and at 11 o'clock our 
 police escort took off his boots and coat, spread 
 his blanket and was soon asleep by my side. 
 Although tired after being ten hours in the sad- 
 dle and much worried, as several annoying inci- 
 dents (due to my men) had occurred during the 
 day, I did not doze off till near midnight. Then, 
 with the exception of a faint light in two or 
 three of the numerous tepees, and a few prowling 
 dogs, all was quiet, but it seemed I had scarcely 
 closed my eyes when a tremendous discharge of 
 fire-arms, accompanied with savage yelping, 
 startled me, and at the same moment the police 
 officer sprung to his feet, and with revolver in 
 hand rushed from the tent ; returning in a few 
 moments, he said " without my boots, my horse 
 
 takes me for one of those b y red skins and 
 
 will not let me go near him. I'm not afraid that 
 they will do us any harm, but I am afraid they 
 might play some trick on my horse." This man's 
 magnificent horse although the night was 
 unusually dark recognized his master at once, 
 with his boots on, and allowed him to lead him 
 to our tent and picket him close to us. 
 
 The Indians continued firing and shouting 
 279 
 
REMINISCENCES 
 
 their war-whoops till early morning, when all at 
 once they quieted down to gambling, a fact indi- 
 cated by the constant beating of several " tom- 
 toms." But soon after they were up and prepar- 
 ing for other celebrations. 
 
 The two men I have so frequently mentioned 
 during this description of my third trip to Cypress 
 Hills, for some cause not known to me, appeared 
 very anxious to be away from this Indian camp, 
 and, as I shared in this anxiety, we were in our 
 saddles by six o'clock and on the last stage of 
 our journey. While the police officer and I rode 
 ahead, our team and men followed. The trail 
 was good and, although my Yankee attendant 
 made a bold attempt to run me down with his 
 horses and waggon which almost cost him a 
 bullet in his unworthy body we reached Maple 
 Creek safely, where, nothing loath, I gave up my 
 outfit and prepared to reach the last locality I 
 had been instructed to examine before returning 
 to Ottawa. 
 
 Although I had been unfortunate in this, my 
 last journey to the Cypress Hills, we secured 
 several specimens new to our Canadian Miocene 
 rocks, among which may be mentioned a left 
 ramus of mandibule of an extinct wild boar. I 
 think I have already mentioned that Professor 
 Cope named this fossil jaw Elotherium arctatum. 
 It is described and figured in " Contributions to 
 Canadian Paleontology Vol. III. It is probable 
 .280 
 
AMONG THE ROCKS. 
 
 further researches in the Miocene rocks of the 
 Cypress Hills may reveal many other remains of 
 extinct animals, other than those now exhibited 
 in the cases of the Dominion Geological Museum. 
 
 Taking the C. P. Ry., Aug. 6th, I arrived two 
 days later at Port Arthur, and at once com- 
 menced my researches for fossils in that portion 
 of the Lower Cambrian rocks known as the Ani- 
 mikie series, which has an area of over 1,000 
 miles, extending from Port Arthur along the 
 west shore of Thunder Bay, Lake Superior, and 
 other localities in that region. 
 
 The rocks of the Animikie formation consist 
 chiefly of even bedded grey and blackish argillites, 
 diabasis and other traps, cherts and jaspers. In 
 places, the argillites (clay slates) are very 
 micaceous and stained with oxide of iron, some are 
 soft am! stain the fingers with a carbonaceous 
 matter, others are hard and dolomitic. They are 
 supposed to have a thickness of 12,000 feet and 
 to rest upon the Archian formation. They form 
 the silver-bearing rocks of Lake Superior. Among 
 the group of mines now being worked in the 
 Animikie formation are the well known Rabbit 
 Mountain and Beaver mines, which are situated 
 a few miles from Port Arthur. 
 
 Although the stratigraphical relations of the 
 
 Animikie series seem to prove that they are not 
 
 far removed from the Laurentian formation, 
 
 they are thought by some writers to belong to a 
 
 281 
 
REMINISCENCES 
 
 higher series of strata a series in which one 
 might expect to find fossils of a Primordial type, 
 but, although the enormous exposures of these 
 slates at the silver mines, Rossport, and other 
 localities afford the collector such a splendid 
 chance, no fossils have yet been found in the so- 
 called Animikie series. For various reasons it is 
 exceedingly important that fossils should be 
 found in these rocks, and the geological or other 
 investigator who does find organic remains, whe- 
 ther belonging to the animal or vegetable king- 
 dom, will solve a most important geologic question. 
 
 Mountains, hills, and valleys part of which 
 are heavily timbered characterize the Thunder 
 Bay, Lake Superior silver mining district. Bold 
 escarpments of trap and other eruptive rocks 
 give a rugged aspect to portions of the country. 
 
 A three or four hours' stage ride carried me 
 from Port Arthur to the noted Beaver and Rab- 
 bit Mountain silver mines. At the Beaver mines 
 I found my old friend Capt. Williams, with whom 
 J had spent many pleasant hours at the Echo 
 Lake Copper mines. Capt. Williams, who was 
 then in charge of the Beaver mines, invited me 
 to accompany him through the workings. I at 
 once embraced the opportunity, and after being 
 enveloped in rubber boots, coat and cap, with a 
 tallow candle in my hand, we were lowered into 
 the bowels of the earth. Down, down we went 
 till my head became dizzy, but I closed my eyes, 
 282 
 
AMONG THE ROCKS. 
 
 and held on to the side of the cage till a series of 
 sharp bumps under our feet notified us that we 
 had reached the bottom. Just at that moment, 
 far away in one of the tunnels, an explosion took 
 place shaking the foundation on which we stood, 
 but my guide said it was all right his men 
 were blasting. 
 
 One who is not nervous finds much to interest 
 him in these silver mines. Here and there in the 
 great walls of black slate were partly imbedded 
 cannon-shot-like concretions ; the miners call 
 them "bombs." They vary in size from half an 
 inch to two feet or so in diameter. In the dense 
 darkness of this " silver pit " our tallow candles 
 showed but a faint light, sufficient, however, to 
 reveal the beauty of some of the trap and other 
 rocks whose sides fairly glistened with iron 
 pyrites, calc-spar and other minerals. But so 
 little silver ore is seen that one wonders where 
 the $5,000 worth of that metal, taken out last 
 week, came from. 
 
 The vast accumulation of dump matter taken 
 from these mines and deposited near the mouth 
 of the pits affords the searcher for fossils an excel- 
 lent opportunity, but in the many thousand tons 
 of rock examined not the least trace of organic 
 structure was seen. 
 
 Leaving the Beaver mines Aug. 14th, I again 
 took the C. P. Ry. to Rossport, a small fishing 
 station on the south shore of Lake Superior, 
 283 
 
REMINISCENCES 
 
 about eighty miles east of Port Arthur. On my 
 way I met Dr. Lawson, one of the field geologists 
 of the Survey, who was then engaged tracing out 
 the boundaries of the Animikie formation. Hav- 
 ing secured lodgings at the only boarding house 
 in that vicinity and hired a small fishing smack, 
 together we visited several small islands within 
 easy reach of the shore. On one of these, Quarry 
 Island, sandstones of the Nipigon series are 
 largely displayed and are seen to rest on the 
 black slates of the Animikie formation. They 
 resemble some of the Potsdam rocks of the St. 
 Lawrence River, but, so far as we know, contain 
 no fossils ; they have been largely used in the 
 construction of bridges, etc. 
 
 Another island is partly composed of Amygda- 
 loidal traps, others display fine sections of the 
 Animikie carbonaceous slates ; one is not likely to 
 soon forget a sail around these beautiful islands, 
 nor the delicious supper of fish one can always 
 get here fish just fresh from the deep cold 
 waters of an almost matchless lake. 
 
 Rossport if one can do without the luxuries 
 to be had in the smallest town is a charming 
 place to rusticate in for a short time during the 
 hot summer weather. Boating, fishing, gunning, 
 and a hundred and one shady nooks to explore, 
 nooks which remind one of the song in the Opera 
 " Wang," 
 
 "A shady nook, a babbling brook." 
 
 284 
 
AMONG THE ROCKS. 
 
 August 21st, I was again in Ottawa unpacking 
 my gleanings, from a few of the rocky leaves con- 
 taining the relics of that wonderful epoch the 
 Age of Reptiles. The great extent of Miocene 
 Tertiary and adjacent Belly River series, dis- 
 played along and near the banks of the Red 
 Deer River, have as yet only yielded up to us, so 
 far, a few of the many palaeontological relics they 
 must contain. Let us hope that others interested 
 in the past history of the great Dinosaurian age, 
 will continue researches in these rocksonly a 
 few drops in the bucket, as it were, of which have 
 yet been partially examined, and yet have added 
 so much more valuable knowledge to the story of 
 our Canadian fossil vertebrates. 
 
 285 
 
CHAPTER XIII. 
 
 THE " QUEBEC GROUP " CONTROVERSY AN INTERRUPTION 
 BY ILLNESS I AFFORD AMUSEMENT TO THE NATIVES 
 THE MONTMORENCY ROCKS THE GOLD-BEARING ROCKS 
 OF NOVA SCOTIA IMPROVEMENTS IN THE MUSEUM OF 
 THE SURVEY A CLERICAL GEOLOGIST THE PILGRIMS 
 TO BONNE SAINTE ANNE THE GEOLOGY OF THE SAINT 
 ANNE DISTRICT SURROUNDED BY THE TIDE. 
 
 pvURING the year 1889, the " Quebec Group " 
 *"^ controversy, was still a bone of contention, 
 and again I received instructions to examine 
 especially for fossils certain rocks in various 
 parts of the Eastern Townships beginning at St. 
 Julie, a small village, a short stage ride from the 
 Becancour Station on the G. T. Railway, from 
 which place I followed the rock exposures to 
 Inverness, Lloyd's Mills, through the village of 
 Ste. Agathe and St. Sylvester to the Chaudiere 
 River, thence to Point LeVis (South Quebec.) 
 
 More than one person had said to me : " you 
 geological fellows have a grand picnic all the 
 summer, and spend the winter preparing for the 
 next summer's sport " ; well, if these grumblers 
 had followed my footsteps even over this short 
 journey they would have found there was not 
 much sport in it, and " those geological fellows " 
 well earn all the money they get for " picnicing." 
 286 
 
AMONG THE ROCKS. 
 
 It would take too long to record all the inter- 
 esting incidents of a geological trip through the 
 Eastern Townships, so I shall only mention one 
 which may be of use to those following up this 
 work' 
 
 At Lloyd's mills, in a deep gorge, through 
 which the waters of the River Thames rush 
 over rugged escarpments of slate and limestone 
 conglomerates, I found in the latter rocks, several 
 examples of the pretty little fossil Salterella 
 rugosa, (Billings,) a very small conical shell, 
 belonging to the Sub-Kingdom Articulata. 1 
 found that these conglomerates had a dolomite 
 matrix and resembled very much the dolomitic 
 conglomerates of the west-end of the island of 
 Orleans (over fifty miles away.) Now, I thought, 
 when I reach Quebec, I will at once cross over to 
 Orleans Island, and if I can find Salterella in the 
 dolomitic conglomerates there, it will be an 
 interesting link in the palseontological chain of 
 the " Quebec Group." I reached Quebec Sept. 
 12th, crossed over to my conglomerates, and 
 there a few yards distant from where Sir 
 William Logan and I had sat years ago I found 
 a beautiful little group of Salterella. I thought 
 how Logan's eyes would have glistened had we 
 found these fossils then ! My little prize now 
 occupies a place in the cases of the Survey 
 museum. 
 
 A few days were spent on the rocks of Point 
 287 
 
REMINISCENCES 
 
 LeVis and Quebec; and then I returned to 
 Ottawa, having finished my field work for 1889. 
 The summary Report of the Geological Survey 
 for 1889, after giving a brief account of my 
 explorations for this year, says : 
 
 " The remainder of the year has been occupied by Mr. 
 Weston in preparing, labelling and arranging for exhibi- 
 tion in the Museum the collections received during the sum- 
 mer." 
 
 During my connection with the Geological 
 Survey, to the year 1890, I had seldom been 
 away from official duties, more than a day or 
 two at a time, but this year was an unfortunate 
 one for me. A severe attack of la grippe brought 
 on another complication which taxed the medical 
 skill of my friend, Dr. H. P. Wright, to whose 
 constant care I probably owe my life. 
 
 The following item is from the Summary 
 Report of the Geological Survey for 1890 : 
 
 " From the 6th of January to the 26th of June, Mr. 
 T. C. Weston was confined to his house by severe illness. 
 He subsequently obtained leave of absence until the tirst of 
 August, and during that time and up to the 8th of Sept. 
 he was occupied in the examination of the rocks in the 
 neighborhood of Quebec City, from which he was success- 
 ful in obtaining a remarkable and in some respects a 
 unique set of fossils, which throw considerable light on 
 the age of these rocks, which is still under discussion. 
 From the 10th of September until the close of the 
 year, Mr. Weston's time has been spent in museum work 
 in the palaeontological and archaeological sections, in 
 arranging several hundred new specimens in the cases, in 
 preparing descriptive labels for them and the like." 
 288 
 
AMONG THE ROCKS. 
 
 The fossils from the Quebec City rocks, men- 
 tioned in the above quotation, were found in the 
 grey slates near the back of St. John's or Mont- 
 calm Market. 
 
 It is a singular fact, that in all the geological 
 work done by Logan, Billings, Richardson and 
 others, up to this date (1890,) and two or 
 three years previous, when I discovered minute 
 trilobites, small brachiopods belonging to the 
 Lingulidse family, and other forms in this same 
 series of slates, no fossils had been found, so far 
 as I know, in the city rocks. These trilobite 
 and Lingulae beds have long been under the road- 
 bed of the improved portion of St. Patrick's 
 Street, and now while I write (1897) the rocks in 
 which the remarkable series of graptolites, 
 bivalve and other fossils were found in 1890, are 
 being covered up in making local improvements ; 
 we regret that we did not do more work at these 
 rocks now " Lost to sight, but still to memory 
 dear.'' Many of the fossils just referred to were 
 obtained under rather trying circumstances. 
 The " St. John Market rocks" now covered 
 are in a thickly populated part of the city, where 
 the youthful French element is almost unlimited. 
 To these juvenile members of (the "Quebec 
 Group " I was going to say,) the ancient city, I 
 owe more than one blow which might have been 
 serious. To see a man standing on a ledge of 
 rocks, hour after hour in the blazing sun pound- 
 
 T 289 
 
REMINISCENCES 
 
 ing these rocks, and every now and then care- 
 fully wrapping up and bagging the pieces, was 
 so said one boy in French " as good as a 
 circus." One day, a crowd of rough boys return- 
 ing from school came down the bank in a body 
 to interview me. I tried to explain what I was 
 doing, but my little lecture on palaeontology 
 evidently did not please them, for they left in a 
 body, ascended the bank, and opened fire on me 
 with corn-cobs and other vegetable matter of 
 which there was plenty at hand. I was getting 
 the worst of it when fortunately an old priest 
 came along, and seeing the trouble, came to my 
 assistance and rescued me from the hands of the 
 Philistines. 
 
 Collecting "diamonds" to sell to the many 
 tourists who visit the "Ancient City " is a favorite 
 pastime for many Quebec children, and as my 
 " market rocks " yielded many specimens which 
 could be sold for ten cents or so a dozen they 
 were a perfect mine of wealth to the little ones 
 who on half day holidays gathered here armed 
 with any iron implements, from a railway spike 
 to an old horse shoe, indeed one little girl con- 
 fessed that she had stolen her father's shoe- 
 making hammer for this work. Seeing me, one 
 day, working in a certain spot for a long time, a 
 number of these children, concluding I had made 
 a good strike, crowded me so much that I was 
 at>out to give up work when the thought 
 290 
 
AMONG THE ROCKS. 
 
 occurred : Suppose I offer these youngsters five 
 cents a dozen for all the diamonds they can get 
 anywhere ten yards from where I am working. 
 The offer was made and gladly accepted, and for 
 two hours I worked in peace ; then at five o'clock 
 I called time, and the youngsters gathered around 
 me with their *' diamonds," some had two or 
 three, while others had many. I found it no 
 easy task to settle up accounts with these young 
 folks, but at last matters were squared satisfac- 
 torily, and I went away with about five dozen 
 " diamonds," which after all cost me fifty cents. 
 Now, when I look at the fossils collected that 
 day, I seem to see those twenty or thirty child- 
 ren pounding away at the rocks some singing 
 French songs, while another small group sang 
 
 " Digging dusky diamonds all the season round, 
 " Down in a coal mine underneath the ground." 
 
 Passing two young women in Quebec the other 
 day, one who carried a baby nudged the other 
 and said: "That's the fellow that bought our 
 diamonds, when we were youngsters, wasn't them 
 happy days ? " 
 
 This year (1890), owing to differences of opinion 
 regarding the rocks of Montmorency Falls, it was 
 decided to have each variety of these rocks 
 examined microscopically. This duty was 
 assigned to the lithologist of the Survey, Mr. 
 Ferrier, and I was requested by the director to 
 make a collection of all the typical rocks of the 
 291 
 
REMINISCENCES 
 
 falls. I was still very shaky, after my iate ill- 
 ness, and this task, although accomplished in one 
 day, was a severe one. However, a very inter- 
 esting stratigraphical collection was made, which 
 I regret to say has not yet been reported on. 
 
 While trimming "hand specimens," a short 
 distance up the river from the falls, a lady came 
 along with her " kodak," and after taking several 
 snap-shots at the beautiful scenery of this delight- 
 ful spot walked away in the direction of the 
 hotel. In a short time I too made my way to 
 the hotel, where I learned that this lady was the 
 Countess of Aberdeen, who with her noble hus- 
 band the Earl of Aberdeen was, I believe, making 
 a flying visit to Quebec. A few years later 
 (1895) Lord Aberdeen became Governor-General 
 of Canada. To their Excellencies the people of 
 Ottawa will always be indebted for their good 
 works, delightful winter sports, and other social 
 gatherings at Eideau Hall. 
 
 During the winter of this year (1890), the 
 geological horizon of the gold-bearing rocks of 
 Nova Scotia afforded considerable discussion. 
 Certain nodular forms, which when weathered 
 seemed to show coralline structure, had been 
 found by Mr. Willis in the rocks of the Northrup 
 gold mines of Nova Scotia. These forms were 
 handed to Professor H. Y. Hind, who supposed 
 them to be fossils of Lower Silurian age. They 
 were then handed to Professor G. T. Kennedy, of 
 292 
 
HURON IAN CONCRETIONS: Once supposed to be the fossil 
 Oli.lhatma so dint n. From the Huronian argillites, New- 
 foundland. 
 
 CoNCRETictys : Micro-drawing from a thin slice of Cape 
 Breton oolitic limestone, magnified about twenty times. 
 
AMONG THE ROCKS. 
 
 King's College, who confirmed Professor Hind's 
 opinion, and pronounced these " fossils " to be 
 Stromatopora. Had this been a fact, it would 
 have altered our present idea of the stratigraphy 
 of the Nova Scotia gold-bearing rocks, consider- 
 ably. Mr. Fletcher and Mr. Faribault of the 
 Geological Survey having collected numerous 
 examples of these supposed fossils, I devoted 
 considerable time preparing microscopical sec- 
 tions, etc. of these forms ; the result of which was 
 that not a particle of organic structure was 
 found, and like the wonderful discovery of 
 Oldhamia in the Huronian slates of Newfound- 
 land mentioned elsewhere were found to be 
 only concretions.* 
 
 A considerable time was spent this winter in 
 hunting up the dates when certain old museum 
 specimens were collected. Although the locali- 
 ties were on these specimens, for some reason the 
 date of collection was omitted. It is due to Dr. 
 Selwyn, then director, that now almost every 
 specimen in the museum bears the date of collec- 
 tion. This and other museum work together 
 with a large share of time spent in the restora- 
 tion of fossil bones, the developing of other speci- 
 mens, etc. kept me fully occupied till the spring. 
 Before closing my remarks for this year I take 
 much pleasure in recording my sincere thanks to 
 
 (* See notes on concretionary structure by T. C. VVestou 
 Trans ol the Nova Scotiau lust, of Science Ser. 2, Vol. 1.) 
 
 293 
 
REMINISCENCES 
 
 Dr. H. M. Ami, M. A., F. G. S., of the Survey, who 
 during my sickness besides his own work 
 devoted much time attending to many of my 
 duties, both in the palseontological and ethno- 
 logical branches of the Survey, and later in assis- 
 ting me to select typical graptolites of the Point 
 LeVis and Quebec City rocks, to send to Prof. 
 Walcott, director of the Geological Survey of the 
 United States, and Prof. Lapworth, LLD., F.G.S. 
 of Birmingham, England, who had kindly con- 
 sented to identify these fossils. Prof. Lapworth's 
 interesting report on these Quebec graptolites 
 will be found in Trans. Royal Society of Canada, 
 Section IV. 1886. 
 
 As will be seen by the reports of the Geologi- 
 cal Survey, the year 1891 was an interesting 
 epoch in the history of Canadian geology and 
 also in the history of our National Museum. 
 
 It was my privilege to be allowed to make many 
 improvements in the arrangement of specimens, 
 both in the palaeontological and ethnological 
 branches, to add many new specimens to our 
 already splendid geological exhibit, and to restore 
 many of the fine fossil bones which now form 
 one of the interesting features of the museum. 
 In this work the winter months passed pleasantly. 
 The " Quebec Group " controversy continued, and 
 more than one unbeliever in our director's strati- 
 graphy of the LeVis formation got an occasional 
 roasting ! 
 
 294 
 
AMONG THE ROCKS. 
 
 To try and prove certain theories by fossils I 
 was again preparing to leave for further work at 
 the Quebec City, and island of Orleans rocks, 
 when Ottawa in fact the whole of Canada 
 was cast into mourning, by the death of our Pre- 
 mier and Minister of Railways The Right Hon. Sir 
 John A. Macdonald, which took place on the 6th of 
 June. By the death of this great man Canada 
 lost its greatest statesman, and the Geological 
 Survey one of its best friends. 
 
 On the 30th of June, I was again in the City 
 of Quebec, spend ing a few days on the St. John's 
 St. " Market rocks" which now, as I have 
 already stated are covered up. It was during 
 my work this summer, that a very interesting 
 series of fossil Graptolitidce, belonging to the 
 genera Dictyonema, were found in one of the 
 layers of slate composing part of these " Market 
 rocks." Now that these rocks are for ever hid- 
 den from us, the fossils obtained from them with 
 so much trouble and perseverance should be con- 
 sidered among the choicest specimens of the geo- 
 logical Survey. 
 
 During my visit to Quebec, I have often spent 
 interesting half hours with Abbe Laflamme of 
 Laval University, who, as the reports of the Sur- 
 vey show, has, during his vacation from the 
 University, done much important geological work 
 for the Survey. Like all true geologists, the pro- 
 fessor had a sharp eye for fossils, but in some 
 295 
 
REMINISCENCES 
 
 way, he like others seems to have taken it for 
 granted that the Quebec City rocks contained no 
 fossils, an idea which might be due to the fact 
 that thousands of tons of the black limestone of 
 the city may be broken without finding the trace 
 of a fossil. Having on the 6th of July, mentioned 
 to Prof. Laflamme the discovery of Dictyonema 
 and other graptolites, in the " market rocks," he 
 at once locked his study in the University, and 
 accompanied me to these rocks whose fossil evi- 
 dence had so long escaped the notice of geologi- 
 cal workers in the " Quebec group " formation. 
 
 To see a black robed priest, with hammer in 
 hand, pounding rocks in the heart of the city 
 must have seemed strange to many passers-by, 
 but even the curiosity of the small boy was sus- 
 pended during the priest's stay with me. Had I 
 been alone no doubt there would have been a few 
 corn-cobs flying around as on former occasions. 
 
 During my few days sojourn in Quebec this 
 summer, I had the pleasure of taking my friend 
 Mr. R. G. McConnell, of the Canadian Geological 
 Survey who, during a vacation, was making a 
 hasty trip to England over a portion of the 
 Quebec City 'and Levis rocks, McConnell, who 
 is one of the "long distance" Northwest explorers 
 of the Survey, had never been in Quebec before, 
 or seen in situ any of the Quebec or LeVis rocks, 
 so often discussed (sometimes hotly) in his pres- 
 ence in the work-rooms of the Survey. To a 
 296 
 
AMONG THE ROCKS. 
 
 veteran geologist a different series of rocks from 
 those he has been studying for years always 
 proves interesting ; so it was with McConnell's 
 hasty glance over a portion of our Quebec group 
 rocks. 
 
 Accompanied by Mr. Bedard a Laval Univer- 
 sity student as assistant, we boarded one of the 
 market boats which runs from Quebec to several 
 small villages along the north side of the island 
 of Orleans. In a few hours we landed at St. 
 Famille the small French village spoken of 
 while recording a previous journey with Dr. 
 Selwyn some years ago. From this village we 
 went carefully over the shore rocks, both east 
 and west, hoping to gather new facts regarding 
 the relation of these rocks to those of Quebec 
 city. After much hard work and weary tramps 
 nothing was found to disagree with Logan and 
 Richardson's interpretations of these rocks. The 
 only new feature recognized this summer was 
 that near the west point of the island of Orleans, 
 nearly opposite Beauport church, there is a set of 
 black limestones containing fossils identical with 
 some of those found in the St. John's market 
 rocks. This proves, without doubt, that these 
 Orleans Island limestones are a portion of the 
 black rocks seen in most parts of Quebec City. 
 
 To try and confirm, by fossil evidence, some of 
 the statements made regarding the geological 
 horizon of the rocks on the north shore oppo- 
 297 
 
REMINISCENCES 
 
 site the island of Orleans I proceeded alone by 
 boat, to Ste. Anne de Beaupre' a village situated 
 on the banks of the St. Lawrence River 21 miles 
 from Quebec. But who has not heard of that 
 wonderful spot where thousands yes, hundreds 
 of thousands of pilgrims assemble yearly to do 
 homage at the shrine of St. Anne, to hear mass 
 in the magnificent church, and drink water at 
 the " miraculous " fountain. Rich and poor, old 
 and young, kneel side by side at the beautiful 
 altar, and are blessed by one of the numerous 
 priests who are ever ready to administer spiritual 
 consolation to the afflicted, many of whom make 
 heroic sacrifices to reach this wonderful place of 
 miracles : 
 
 "The waters of the grand Saint Lawrence glide 
 In calm majestic motion, on their way 
 
 Past Bonne Saint Anne, as if the gentle tide, 
 Its silent humble homage thus would pay 
 
 Before the ancient shrine, as on its breast 
 It bears the pilgrims to this place of rest. 
 
 At this time, 1891, there were no large hotels, 
 as now, at Ste. Anne, and I found shelter in one 
 of the many cheap boarding houses of this strange 
 place of religious associations. 
 
 A great many people who visit Bonne Sainte 
 Anne for religious comfort are poor indeed ! a 
 fact readily seen when one looks at the hundreds 
 of rudely made crutches, trusses, and other instru- 
 ments, constructed to support the crippled and 
 298 
 
AMONG THE ROCKS. 
 
 otherwise afflicted. Year by year these instru- 
 ments of suffering humanity have accumulated. 
 Day by day these relics of pain and decay are 
 laid at the shrine of Saint Anne. Tobacco and 
 snuff boxes, pipes, spectacles, jewellery, and many 
 other things have been laid with grateful hearts, 
 for health restored and hopes renewed, at the 
 feet of Saint Anne. Hundreds of these offerings 
 adorn the walls of the church, while other large 
 instruments made for the afflicted cover one of 
 the great pillars of this noble building. 
 
 But to return to our boarding house, where, to 
 reach the small bed -room assigned me, one has to 
 pass through a large sitting room which, at this 
 time, was used by " women only," pilgrims who 
 come and go every hour of the day. Having 
 wended my way past twenty or more chattering 
 French Canadians and gained my " chamber," I 
 prepared at once for a good geological tramp, 
 then sought the dining room where, seated on 
 long benches before a large, roughly made deal 
 table, were a motley group of pilgrims. Each of 
 these visitors to Ste. Anne appeared to be eating 
 provisions they had brought some brown bread 
 and a little pork, while others seemed to fare 
 luxuriously on sweet corn, bread, butter, and 
 cakes, purchased in the village. Some of the 
 more fortunate ones indulged in a plate of soup, 
 or a cup of tea, furnished by the house for five 
 cents each. But poor and scanty as the food of 
 
REMINISCENCES 
 
 many of these people seemed, they were all 
 happy, and why not ? They liad made their 
 oblations to the good Saint Anne, drunk of the 
 water of the " miraculous " fountain, been blessed 
 by the good priest, and were now waiting for the 
 steamboat to carry them back to their homes. 
 
 After dinner which was taken at the same 
 table with the pilgrims, I hurried off to my work, 
 returning late in the evening to find the house 
 crowded with a new lot of pilgrims, who for 
 some reason had been detained on their journey 
 and were now obliged to remain over at Ste. 
 Anne for the night, an unusual thing for pilgrims 
 to this place. 
 
 The night was beautiful, and after supper I 
 wended my way down to the foot of the long 
 wharf, over which, during the last year (1890), 
 105,672 pilgrims had walked on their road to do 
 penance and receive Holy Communion at the 
 shrine of Saint Anne, to do which many had 
 made brave sacrifices. 
 
 Alone I sat in the gloaming on the foot- worn 
 steps of the pier and watched the thousand 
 twinkling lights of the Quebec and Point Levis 
 shores, and their reflection in the dark waters of 
 the St. Lawrence. 
 
 From this distant point one sees a great por- 
 tion of the district surroundings of the ancient 
 capital of Quebec, every yard of which is freighted 
 with historic interest, the cliffs of the Citadel, 
 300 
 
AMONG THE ROCKS. 
 
 the magnificent Dufferin Terrace 1,400 feet 
 long and 200 feet above the level of the river, 
 the towering cliffs of Levis which have yielded 
 us such an abundant fossil fauna, and, after 
 researches made almost every summer for the 
 past fifty years, still give us an occasional new 
 form, but 
 
 " The day is done, and the darkness 
 
 Falls from the wing of night, 
 
 As a feather is wafted downwards 
 
 From an eagle in his flight." 
 
 And again I wended iny way to the pilgrims' 
 lodging house, passing the wharf toll-gate through 
 which 2,000 seekers after religious consolation 
 had passed that day, each paying a small toll, 
 which, to save time, is arranged for by the good 
 fathers who organize the excursion. 
 
 A short walk and I was again in my boarding- 
 house. All was quiet, and unobserved I reached 
 the large room that led to my bed-room. On 
 opening the door, to my astonishment I saw the 
 floor covered with cheap mattresses or straw 
 beds on each of which was stretched, one, two or 
 three women. I backed out of the room, closed 
 the door and went in search of the good woman 
 of the house, and having found and explained 
 matters to her, she smiled and said " Monsieur, 
 I will escort you to your chamber." Following 
 my hostess, we picked our way carefully between 
 these sleeping pilgrims and gained my chamber, 
 301 
 
REMINISCENCES 
 
 I was tired and slept well till late the following 
 morning, when on opening my door I found all 
 quiet again, and the people of the house busy 
 preparing for another batch of visitors to this 
 village of miracles. And so these pilgrims come 
 and go every day during the summer months, 
 each leaving their small oblations towards the 
 support of the church, and each carrying away 
 some small souvenir of their visit to the shrine 
 of Saint Anne. 
 
 It is a beautiful sight, while leaning on the 
 rails of Dufferin Terrace at sunset, to watch one 
 of our large St. Lawrence steamers heavily laden 
 with homeward bound pilgrims glide gracefully 
 past the walls of the Citadel and through the 
 shadows of the towering cliffs of Cape Diamond; 
 to watch the brilliant reflections of the evening 
 sun on the waters, the cliffs and the numerous 
 windows in the Catholic buildings of Levis. But 
 hark ! a thousand voices on the throbbing steamer 
 are chanting their vesper hymns. Happy souls, 
 many of whom endured much to make this pilgri- 
 mage to their Bonne Sainte Anne, to 
 
 '* Confess their sins, receive the Bread of Life." 
 
 But to our geology again. Returning to Ste. 
 Anne by the Beauport and Montmorenci road we 
 pass numerous outcrops of Trenton limestone and 
 many small lime kilns in which the French Can- 
 adian farmers burn lime for agricultural purposes 
 and for whiting the walls of the houses. 
 302 
 
AMONG THE ROCKS. 
 
 At the summit of Montmorenci Falls of which 
 I have already spoken the Trenton limestone is 
 beautifully displayed in the " natural steps," of 
 the Montmorenci River. Here the formation has 
 an estimated thickness of fifty feet of dark grey 
 limestone (resting on gneiss) holding most of the 
 typical fossils, among which are found good 
 examples of our Lower Silurian Bryozoa, Brachi- 
 opoda, Gasteropoda, Cephalopoda, Crustacea and 
 other divisions of the fossil fauna of the Trenton 
 formation. 
 
 An interesting geological feature of this delight- 
 ful locality in the shadows of pine woods, and 
 the roar of waters as they hasten on to hurl them- 
 selves over the great cliffs and mingle with the 
 waters of the St. Lawrence is the occurrence of 
 a rich deposit of ochre, covering an area of several 
 acres, and having a depth of one to seventeen 
 feet. The colors of these pigments vary from a 
 light yellow to a dark brown. In 1875 while 
 making a collection of these ochres for the Phila- 
 delphia Centennial Exhibition I found a number 
 of flint chippings and bits of Indian pottery. 
 Some of these relics were a foot or more below 
 the surface. It is probable this was a favorite 
 Indian camping ground during their hunting 
 expeditions hundreds of years before the first 
 page of Canadian history was written. 
 
 Continuing our journey by road, from the 
 Falls of Montmorenci ; towards the Ste. Anne 
 303 
 
REMINISCENCES 
 
 Biver and at the lower falls, almost all the 
 typical rocks of the Trenton, Utica and Hudson 
 River formations. These have characteristic 
 fossils and a great variety of strata of bluish - 
 grey limestone, dark bituminous shales, grey 
 and other colored sandstones, etc. Together with 
 the Laurentian gneiss on which one of these 
 formations (Trenton) is seen to lean or rest, they 
 form a thickness of about 1200 feet. 
 
 Arriving again at Ste. Anne I take another 
 meal at our pilgrim boarding house, then pro- 
 ceed by buggy to St. Joachim another small 
 French village on the banks of the St. Lawrence 
 River, a few miles below Ste. Anne. 
 
 It is a lovely drive, the river tq the right and 
 verdant sloping banks or terraces to the left. 
 Outcrops of sandstone and other rocks belonging 
 to the Hudson River formation are numerous. 
 Most of these I examined carefully for fossils. 
 In one ledge of sandstone good examples of the 
 Graptolite Diplograptus pristis were found. 
 These were important specimens, and I was in- 
 tent on my find when a typical lower-class 
 French Canadian of this district who was pass- 
 ,ng stopped to see what I was doing. Seeing 
 me carefully wrap up and bag a piece of this 
 fossiliferous rock, he turned to my carter and 
 said : 
 
 "Sacre" crapaud! pourquoi emportes tu dans 
 ton sac, ce petit morceau de pierre ? Mon Dieu! 
 304 
 
AMONG THE ROCKS. 
 
 il peut bien prendre, toutes les pierres qu'il 
 trouvera sur ma ferme ! 
 
 Continuing our journey in a short time we 
 came to the Grand River, a small but rapid 
 stream hurrying on to plunge and mingle its 
 waters with those of the St. Lawrence. 
 
 Here where the road crosses the stream, we 
 find a beautiful example of rock crumpling or 
 folding, forming a number of small anticlinals 
 and synclinals But to which formation do 
 these strata belong ? They look like some of our 
 Levis rocks ; and they ought to belong to the 
 Hudson River formation what will the Index 
 to our stony book say? With hammer in hand 
 T wade along the margin of the stream to where 
 it cuts through a series of black shales and thin 
 bedded sandstones A few minutes work and I 
 have in my bag several typical fossils of the 
 Hudson River formation. The stratigraphy of 
 our rocks across the river on the north shore of 
 the island of Orleans told what we might ex- 
 pect here, and the fossils in my bag confirm the 
 supposition. 
 
 Quite satisfied, and the best of friends, with 
 this fine example of corrugated strata and the 
 fossils obtained from some of the beds, my carter 
 and I sat down in a shady nook on the river bank 
 to enjoy a lunch the good lady of ourpilgrim board- 
 ing house had provided us with a bottle of milk 
 four hard boiled eggs, bread, butter and cheese, 
 
 u 305 
 
REMINISCENCES 
 
 we were both hungry and it was a delicious 
 meal so my carter thought and when I gave 
 him a cigar, he murmured words of thanks in 
 which le Bon Dieu figured ! Man and beast 
 being fed and rested, we resumed our journey, 
 and when within a mile of St. Joachim Church 
 came to an interesting outcrop of dark grey lime- 
 stone mentioned in the Canadian Geological 
 Report for 1863. Here we do not have to 
 hammer for hours as in some cases for a fossil, 
 for the index of this chapter of our stony book is 
 opened out before us, and we see at a glance 
 that the fossils exposed on the weathered surface 
 of this sedimentary rock are characteristic forms 
 of the Trenton formation. 
 
 A few minutes suffice to give us all the typical 
 forms we need to identify this strata, some of 
 which belong to the following genera : Stenopora, 
 Leptcena, Strophomena, Orthis, Catymene, etc. 
 
 Between this Lower Silurian strata and St. 
 Joachim village only Laurentian rocks are 
 seen, but the limestone is met with again at Bay 
 St. Paul, fifty miles or so lower down a loca- 
 lity I shall mention in my next year's (1892) 
 rambles 
 
 St. Joachim village, like St. Anne, and I think 
 all the villages along the lower St. Lawrence 
 coast, are essentially French Canadian settle- 
 ments. In every village a large Catholic church 
 occupies a prominent position. They are all 
 300 
 
AMOJSG THE ROCKS. 
 
 dedicated to some saint. " In the gospel accord- 
 ing to St. Luke, Joachim is mentioned under the 
 abbreviated name of Eli, as father-in-law to St. 
 Joseph. The only but glorious offspring of this 
 marriage was the Blessed Virgin Mary." How- 
 ever large these churches are, they are seldom 
 large enough to accommodate the great number 
 of people who assemble for Sunday morning 
 mass. Sunday is a red letter day for the toiling 
 Canadian farmer and his family. To some who 
 have to make long distances to their church, it 
 means most of the day at mass and on the road ; 
 but after their devotions are over, the remainder 
 of the day is given up to pleasant intercourse 
 with- one another, and for a short time the trials 
 and cares of life are forgotten. After seeing 
 much of the poor class of French Canadians one 
 comes to the conclusion that they are indeed a 
 happy race of people. 
 
 Leaving the village of St. Joachim with its 
 grassy flats where the " sport " finds good work 
 for his dogs to do, we return to Ste. Anne, where 
 again a few more geological tramps were made 
 along the shore and shore road. Although the 
 rocks had yielded many interesting fossils, I 
 found nothing to refute the work done by Logan 
 and his colleagues, during the early history of 
 the Survey. 
 
 July 20th, my friend Mr. Bedard joined me in 
 Quebec and together we spent a day on the 
 307 
 
REMINISCENCES 
 
 Utica shales between the mouth of the St. 
 Charles River and Beauport. 
 
 The Utica formation, which takes its name 
 from Utica, New York, is represented in Canada 
 by about 300 feet of black bituminous shales, 
 interbedded with harder rocks, sandstones, lime- 
 stones, etc. 
 
 At low tide near the mouth of the St. Charles 
 River a fine section of this strata is seen, its dip 
 is about S.E <40. To the palaeontologist who 
 is interested in Utica and Hudson River fossils 
 the latter formation in Canada has a thick- 
 ness of 2,000 feet and is intimately associ- 
 ated with the Utica deposits this locality pre- 
 sents a fine field for collecting most of the typical 
 Utica and Hudson River graptolites. It is 
 singular though that while much of the strata 
 here is very prolific in graptolites, I failed to find 
 one of the trilobite or fossil bivalve shells, which 
 are so numerous in the Utica of other localities. 
 
 The tide rises rapidly over the Utica rocks 
 and extensive marshy flats near the mouth of 
 the St. Charles River. This I found to my cost 
 one day, for while intently examining some fine 
 specimens of the graptolite Climacograptus 
 bicornis, I felt & splash of water over my boots, 
 and looking up, I found myself surrounded with 
 water ; the twenty or more cows who were near 
 me a short time ago feeding on the marsh grass, 
 had evidently paid more attention to the tide 
 308 
 
AMONG THE ROCKS. 
 
 than I, for they were on the shore half a mile 
 away. Slinging my collecting bag, which con- 
 tained my lunch and perhaps fifty pounds weight 
 of rocks, over my shoulder I made for the shore, 
 which I reached just after the water had found 
 its way into my trousers pockets. The day was 
 very hot, and while my clothes dried in the sun 
 I Watched the burning of a schooner which had 
 been beached here for that purpose, and thought 
 of those startling military times of 1759, when 
 the siege of Quebec took place and all this shore 
 between the mouth of the St. Charles River and 
 Beauport was lined with batteries, arid the sur- 
 rounding country for miles was covered with 
 French encampments. 
 
 A spot a little to the right of where I sat enjoy- 
 ing a bottle of ginger ale and some sandwiches, 
 provided me by my good sister-in-law, Mrs. M. G. 
 Mountain, an old resident of Quebec is still 
 pointed out as the place where in 1535 Jacques 
 Cartier and his companions passed the winter. 
 It was here also that the French constructed 
 floating batteries and fire-ships which floated 
 with the current down the St. Lawrence and did 
 much damage to the English fleet. 
 
 July 22nd, I was again on the island of 
 Orleans trying to connect by fossil evidence the 
 black shales and limestones of the " cement 
 quarry " with the bituminous limestones and 
 slates of Quebec City. The fossils obtained 
 309 
 
REMINISCENCES 
 
 threw much light on this complicated strata, but 
 much still remains to be done. I returned to 
 Ottawa July 27th, and owing to delicate health 
 did not return to field work again this year. 
 
 During the remaining part of the summer and 
 winter of 1891-92 my time was occupied again 
 in palaeontological and archaeological work con- 
 nected with the museum, varied with microsco- 
 pical and official work. 
 
 During the past winter my health had been 
 poor and my request to be allowed to make 
 short geological excursions during the summer of 
 1892, was readily granted. 
 
 Knowing that extensive excavations for the 
 foundation of the Chateau Frontenac were in 
 operation, and hoping to obtain more fossil 
 evidence regarding the Quebec City rocks, I 
 again on the 18th of July returned to the ancient 
 capital. 
 
 310 
 
CHAPTER XIV. 
 
 FURTHER GEOLOGISING AT QUEBEC CITY EXAMINING THE 
 ROCKS PRECIPITATED BY THE LAND-SLIDE A VICTIM 
 OF POISON-IVY A SEARCH FOR A METEORITE ON THE 
 OTTAWA RIVER THE LAURENTIAN MARBLE AN 
 
 ARTIST'S SKETCH is USEFUL THE JOGGINS COAST, N.S. 
 
 AGAIN CONCLUSION AND THANKS TO MY FRIENDS OF 
 THE SURVEY. 
 
 thousands of tons of black limestone 
 excavated for the new Frontenac hotel, 
 Quebec, afforded an excellent opportunity for 
 fossil researches, and I worked hard to get new 
 facts, but regret to say that with the exception 
 of one species of graptolite Dyplograptus 
 pristinijormis these rocks were destitute of 
 organic remains. 
 
 Returning to the St. John St. market rocks I 
 found my Dictyonema beds gone ; gone to pave 
 the streets of the city. I have already stated 
 that the road-bed of St. Patrick's street covers 
 my trilobite beds, now the Dictyonema beds 
 only a few yards distant, are gone. It is a singu- 
 lar fact that the fossil Dictyonema has not so far 
 been found in any other than the St. John Street 
 market rocks, in all the city. The loss of these 
 rocks which have been made historical by the 
 writings of Professor Lapworth of Birmingham, 
 311 
 
REMINISCENCES 
 
 England, and several members of the Geological 
 staff will always be regretted. 
 
 While still at work at these ever interesting 
 Quebec and Levis rocks I received official 
 instructions to proceed to Bay St. Paul, there to 
 make a large collection of the garnetiferous 
 gneiss which forms one of the interesting geo- 
 logical features of the nor tli- east side of the bay. 
 
 Bay St. Paul which is one of the lesser salt 
 water resorts of the river St. Lawrence, is about 
 60 miles below Quebec, 30 miles above the popu- 
 lar summer resort, Murray Bay, and 90 miles 
 below Quebec. Between the latter Bay and Bay 
 St. Paul, the coast is very rugged, and affords an 
 excellent opportunity for the student in geology 
 to study Canadian Laurentian rocks, and also 
 some of the Lower Silurian strata ; as at Bay St. 
 Paul, where the Trenton limestones are largely 
 developed and contain the characteristic fossils 
 of that formation. 
 
 A little saline spring issues from the Silurian 
 rocks of this bay, and a short distance away, in 
 the parish of St. Urban large deposits of iron 
 ore (limonite) occur. 
 
 The journey from Quebec to Bay St. Paul in 
 one of the fine Saguenay steamers occupies 
 about five hours. Before reaching Riviere 
 Ouelle, about 80 miles below Quebec we enter 
 salt water and see, perhaps, several white por- 
 poises, that peculiar Cetacea, white whale or 
 312 
 
AMONG THE ROCKS. 
 
 Beluga Delphinapterus catodon, which as a 
 skeleton in the Geological museum, Ottawa 
 shows a length of 18 Dr 20 feet, and fossil bones, 
 also seen in the same institution, from the 
 Post Tertiary clays of Montreal, show that this 
 aquatic mammal existed prior to the formation 
 of the Leda clay deposits in which such a variety 
 of " chalk shells " are found. 
 
 A Quebec paper states that during this spring 
 (1898, large droves of porpoises have been seen 
 in the St. Lawrence following the small fish on 
 which they feed. Nearly one hundred have been 
 captured at the River Ouelle fishery. This catch 
 represents a value of over three thousand dollars. 
 One of these porpoise, or white whales, was sold 
 to an American, representing the Barnum Museum 
 for $80. 
 
 August 9th, I was again in Quebec accomp- 
 anying Dr. Selwyn, director of the Geological 
 Survey of Canada, over some of the Quebec City 
 rocks he had not yet studied. We first examined 
 the great pile of rock debris which fell from 
 almost under the foundation of part of the Cita- 
 del walls in 1889, crushing through several 
 houses and hurling fifty-one souls into eternity. 
 
 The falling of these rocks brought to light sev- 
 eral beds of strata exhibited in no other part of 
 the city, thus affording an excellent opportunity 
 for fossil researches ; but although we climbed 
 over thousands of tons weight of this land-slide 
 313 
 
REMINISCENCES 
 
 we failed to find more than one species of fossil, 
 that being a graptolite common to the Hudson 
 River and Utica formations. 
 
 From these fallen rocks we ascended the steep 
 partially verdant cliffs till we were immediately 
 under the walls of the Citadel. It was an exceed- 
 ingly hot day, and the journey up about 180 
 feet of slippery shingle was not easy. When 
 about half way up we came to a large bed of 
 poison ivy; turning to me the doctor said, 
 "Weston is this poison ivy?" "Yes" I said, 
 " put on your other glove and if possible don't 
 touch it. For my part I have no gloves nor 
 umbrella, and must hold on to this villainous 
 plant or perhaps roll to the bottom of the 
 cliff." I had suffered from this poisonous plant 
 before, and knowing that I was then in a fit 
 state to receive the poison felt sure I should 
 suffer again, but there was no help for it, so I 
 used the stems of the plant as a ladder. Reach- 
 ing the top of the cliff, without thinking where 
 my hands had been, I brushed the perspiration 
 from my forehead. The following day I took 
 my chief to see the Utica slates, near the mouth 
 of the St. Charles River (already spoken of). 
 Before leaving Dr. Selwyn that evening, I held 
 out my wrists and said, " look doctor, the poison 
 is beginning to do its work." The next and fol- 
 lowing seven days I was in bed and suffered well 
 I shudder now when I think of the awful distress 
 314 
 
AMONG THE ROCKS. 
 
 that infernal plant can cause large water-blisters 
 over one's face, one's hands tied up in linen sat- 
 urated with lotions, and all the rest of it. It is 
 a singular fact, however, that this poisonous 
 plant, which covers a large portion of the Citadel 
 rocks, only affects some people, and that only at 
 certain times of the year, and when the skin is 
 in a fit state to receive the poison. We are told 
 in classical mythology that ivy was sacred to 
 Bacchus, whose head was surrounded by a wreath 
 of it. I wonder what the effect would have been 
 had they crowned him with the same species of 
 ivy that caused me so much trouble I The species 
 of ivy spoken of above is Rhus toxicodendron. 
 French Canadians call it Herbe a la puce. It is 
 found in abundance under the rampart and Cita- 
 del wall, along the shores and on the islands of 
 the St. Lawrence river. Although in handling 
 this plant it may only affect one person in fifty, 
 it is better to avoid touching it in the hot months 
 of summer. 
 
 During my stay in Quebec this summer (1892) 
 I found an opportunity of again examining the 
 Mountain Hill Cliff. Geologically this is a very 
 interesting portion of the Quebec City rocks ; 
 but here too the moment one touches the 
 rocks with his hammer, he is surrounded by peo- 
 ple who want to know why you break the 
 rocks while there is so much already broken, etc., 
 etc. 
 
 315 
 
REMINISCENCES 
 
 I was delighted one day when Mr. G. A. 
 Laf ranee, whose book-binding establishment comes 
 close up to the cliff, allowed me to get out of a 
 back door on to the escarpment. I worked 
 without interruption, and found such important 
 information regarding the Quebec city rock de- 
 posits, that some days after when the director of 
 the Survey returned to the city, he gladly 
 accompanied me, and together we crept through 
 the little back-door, or trap, in Mr. Laf ranee's 
 factory. We left this locality with quite a num- 
 ber of fossils which prove that a large part of 
 these Mountain Hill rocks belong to the Trenton 
 and Hudson River formations. 
 
 The following extract, from a paper contributed 
 by me to the Ottawa Naturalist, will serve to 
 record in these memoirs, several interesting facts 
 regarding this portion of our Quebec City rocks: 
 
 " Out of the 12,000 feet or more of strata which form 
 the much discussed "Quebec Group," there are several 
 interesting escarpments and sections which have hitherto 
 not received the attention they deserve. One of these 
 escarpments is the Mountain Hill cliff, which forms a 
 portion of the height over which the ramparts of the city 
 of Quebec are built. 
 
 ** A close examination of the cliff immediately facing 
 Mountain Hill House, on the lower part of the hill, shows 
 it to be composed of a coarse grey nodular limestone ; in 
 places, bedded structure may be seen, while the principal 
 portion (which is the matrix of the conglomerate), is com- 
 pact, and sometimes flinty, with seams of carbonaceous or 
 bituminous matter. 
 
 316 
 
AMONG THE ROCKS. 
 
 14 This portion of the cliff is prolific in fossils, but they 
 are chiefly fragmentary, and might readily be overlooked. 
 This is probably the reason why, in the early study of the 
 geologic structure of the city portion of the ' Quebec 
 Group, ' these were included in the Levis division of the 
 same. 
 
 ''No fossil remains had been found or observed in the 
 Mountain Hill cliff until the summer of 1877, at which 
 time the writer discovered a D umber of interesting species. 
 In 1892, another opportunity was afforded me to examine 
 that portion of the exposure immediately back of the 
 Express office, and adjoining the book-binding establish- 
 ment. On that occasion there were found some remark- 
 ably well-preserved fossils, some of which were immediately 
 recognized as being characteristic Trenton forms. Dr. 
 Selwyn arrived in the city at the time, accompanied me 
 to this locality, and several new species were added to 
 our former collection." 
 
 My last official journey of this season (1892) 
 was in search of a meteoric stone, or large piece 
 of meteoric iron which celestial bodies, we 
 know, occasionally fall from the heavens. 
 
 On the 3rd of September, Mr. Robert Abbott, of 
 Ottawa, who has spent many years on the Ottawa 
 River, mentioned to the director of the Geological 
 Survey that some years ago he had seen a "chunk 
 of iron," which he supposed to be a meteorite, on 
 the banks of the Ottawa River near the village 
 of Portage du Fort. 
 
 That same evening Mr. Abbott and I left 
 
 Ottawa in search of this specimen. After five 
 
 hours' railway journey we arrived at Portage du 
 
 Fort, where, on seeing the rock deposits of that 
 
 317 
 
REMINISCENCES 
 
 locality, I said "if anyone wishes to build a 
 marble palace on the banks of this beautiful 
 river, there is plenty of material here." 
 
 Although the Laurentian crystalline limestone 
 of Portage du Fort, and other localities on the 
 Ottawa, is not fit for statuary, it is well fitted 
 for any ornamental work, and has been largely 
 used for decoration purposes in the Parliament 
 buildings of Ottawa. The following morning 
 (Sunday), Mr. Abbott and I started after the 
 " thunder-bolt," arriving at the spot where he 
 had seen the " chunk of iron " some years ago. 
 We found we had as Mr. Abbott expressed 
 it had a wild goose chase. We searched dili- 
 gently among the Laurentian boulders which 
 cover a large portion of the bank of the river 
 here, but failed to find the "Thunder-bolt." 
 Returning to the village, we made enquiries from 
 some of the old habitants of the place. One 
 said that some boys, having heard that the "iron 
 stone " was valuable, rolled it into the river till 
 such time as they could find a way of disposing 
 of it. Another said that priests from Montreal 
 had attempted to cart it away, but finding 
 they could not do so with the means they then 
 had, buried it till they could find a way to trans- 
 port it to the railway. Mr. Abbott affirms the 
 specimen was there, and that he had seen it many 
 times, each time thinking to report the matter to 
 some public institution. It is a great pity he did 
 318 
 
AMONG THE ROCKS. 
 
 not while it was there. However, he still hopes 
 to find this valuable specimen 
 
 The large and only meteorite in the Ottawa 
 geological museum now, is said to be of extra- 
 terrestrial origin and an aerolite. It was found 
 in 1N54 upon the surface of a field at Madoc, and 
 weighs 370 pounds so says the Geological Sur- 
 vey Report for 1863. It is probable that this 
 specimen was first found on the surface of a field; 
 but Sir William Logan told me that he found it 
 propping up the corner of a barn, and at once 
 sought the owner of the barn and offered to put 
 a good square stone in its place; the offer was 
 accepted and Sir William immediately had this 
 valuable specimen removed and placed in the 
 museum. 
 
 My official duties during the winter months of 
 1892 consisted as usual, in museum work con- 
 nected with the palseontological and ethnological 
 branches of the Survey. During the first three 
 months of 1893, some of fche members of the geo- 
 logical staff were busily engaged preparing maps, 
 rocks fossils, etc., for the forthcoming World's 
 Fair at Chicago I was requested, as usual, to 
 prepare a series of specimens, illustrating, as far 
 as possible, the supposed Laurentian fossil Eozoon 
 Canaden.se already spoken of in these pages. 
 The preparation of these specimens occupied 
 about two months. They consisted of decalcified 
 and other specimens showing the tubuli structure, 
 319 
 
REMINISCENCES 
 
 microscopic sections, micro-photographs, micro- 
 drawings, etc., altogether forming one of the 
 most interesting collections of Eozoon yet exhi- 
 bited. 
 
 My next duty was to prepare a stratigraphical 
 collection of Canadian fossilif erous rocks. These 
 were included in the large collection of hand 
 specimens which represented as far as possible 
 examples of all our Canadian rocks. 
 
 My next work, also for the coming exhibition, 
 was the re-labelling and re-arranging a large 
 number of specimens of native gold, models of 
 gold from the Chaudiere River, specimens of 
 platinum, etc., and lastly in preparing a series 
 of labels, tablets, etc., to illustrate the me- 
 thods employed by the Survey in labelling 
 and exhibiting geological specimens in the muse- 
 um. These specimens filled three of our muse- 
 um flat show cases, which when in place in 
 the Chicago Exhibition formed an interesting 
 and instructive feature of our geological ex- 
 hibit. 
 
 June 17th, 1893, through indisposition I ob- 
 tained a short leave of absence and again return- 
 ed to Quebec. After a few days' rest, although 
 the flesh was weak, the rocks on the other side 
 of the river had such magnetic attraction, that 
 again, with a boy to hammer, and carry my col- 
 lecting bag, another important collection of fos- 
 sils was made. 
 
 320 
 
AMONG THE ROCKS. 
 
 I think I have already mentioned that fossils 
 are not evenly distributed through the LeVis 
 rocks, but occur generally in colonies ; so one 
 making his fossil debut to our Quebec city and 
 Levis rocks must not be like the foreigner who 
 said to me : " I have tapped the Levis cliffs, and 
 find them to be Laurentian, for there is not even 
 the ghost of a fossil in them." 
 
 Expecting to visit Murray Bay for a few days, 
 I had, before leaving Ottawa, requested my 
 friend, Mr. Lawrence M. Lambe, artist to the 
 Survey, who was well acquainted with that 
 locality, to recommend me to a good hotel. He 
 at once made a pencil sketch of a little, stout, 
 dumpy man, and said : " That is something like 
 my friend, Mr. C., who keeps the best hotel at 
 Murray Bay." 
 
 Monday morning, June 27th, I boarded the 
 fine S. S. Saguenay, and after a delightful trip 
 arrived at this charming salt-water summer re- 
 sort at 4 p. m. With Mr. Lambe's sketch in my 
 hand I looked over the gay crowd of summer 
 visitors, who always flock to the wharf to see 
 new arrivals, and sure enough among this 
 group was our little dumpy man. The sketch 
 was so like him that when I presented it to 
 Mr. C. and said, that is the man I am looking 
 for, he fairly shook with laughter, said, " that's 
 me," and requested to be allowed to keep the 
 sketch. 
 
 v 321 
 
REMINISCENCES 
 
 The days were beautiful, and in the evenings 
 when music and laughter prevailed, one thought 
 of Longfellow's words : 
 
 " And the nights shall be filled with music, 
 And the cares that infest the day, 
 Shall fold their tents like the Arabs, 
 And as silently steal away. " 
 
 Here, at Murray Bay, as at most of the salt 
 water summer resorts below Quebec, the man 
 with the hammer finds the rocks more interest- 
 ing than the festivities of a gay hotel, and 
 wanders forth into the valley of Murray Bay 
 River to search for new fossil evidence, which 
 there is always a chance of obtaining in any of 
 nature's rock storehouses. 
 
 But to one interested in the Laurentian form- 
 ation of the Lower St. Lawrence River, a tramp 
 over the sea-splashed rocks between Murray Bay 
 and Bay St. Paul, gives much pleasure, for here, 
 if one is fitting up a cabinet of Laurentian rocks, 
 he can find a charming variety. But while at 
 this work, he must not forget like the writer of 
 these notes did and get cut off by the tide, and 
 have to remain in an awkward position till ebb 
 tide. 
 
 Before leaving Quebec for Ottawa, I was re- 
 quested to again visit officially the Joggins 
 Coast, N. S. In previous pages of these memoirs 
 I have spoken of the magnificent section of car- 
 boniferous rocks which occupy the south coast of 
 322 
 
AMONG THE ROCKS. 
 
 Cumberland Basin, or the head of the Bay of 
 Fundy. 
 
 The writings of Sir William Logan, Sir Wil- 
 liam Dawson. and other celebrated geologists, 
 have told us of the remarkable fauna and flora 
 of these coast rocks, and given us sections and 
 descriptions of the Joggins Coast deposits till one 
 would almost fancy the geological structure, and 
 palseontological evidence was exhausted, but that 
 is npt so, for during this my last visit to this 
 ever-interesting locality it was my good fortune 
 to discover two very large unio-like shells. They 
 are fresh water bivalves, one measures a little 
 less than eight inches in length, and the other 
 a little more than eight inches. They were 
 found in the rocks belonging to the productive 
 coal measures, with SigUlaria, LepidodeTidron 
 and other Carboniferous fossil plants, and an 
 almost perfect lower jaw of a reptile whose 
 largest tooth was half an inch long. It was pro- 
 bably the maxillary bone Baphetes plciniceps. 
 Owen, (Acadian Geology, Dawson, p. 360). Like 
 many fossil bones of the carboniferous formation 
 this was so fractured that it could not be pre- 
 served with the means then at hand. 
 
 One of these remarkable fossil bivalves has 
 been described and figured by Mr. J. F. Whit- 
 eaves, F. R. S. C., F. G. S., palaeontologist to the 
 Geological Survey of Canada, in Transactions of 
 the Royal Society of Canada. It received the 
 323 
 
REMINISCENCES 
 
 name Asthedonta Westoni, being dedicated to 
 the discoverer. 
 
 My friend, Mr. G. F. Matthew of St. John, 
 N. B., who has done much good geological work 
 for the Survey, having heard of this Carbonifer- 
 ous fossil bivalve, writes : " Let me congratulate 
 you on the giant clam you have found; quite a 
 respectable companion to him of the Pacific ocean 
 and a head and shoulders taller than any modern 
 fresh water beast of our country." 
 
 Another important fossil find was made this 
 season in the productive coal measures of the 
 Joggins Coast. Sunday July 15th while stroll- 
 ing along the shore, in the vicinity of Hard- 
 scrabble Wiiarf, with my coal miner friend, Mr. 
 James Devine who has a sharp eye for fossils 
 we discovered on the surface of a large block of 
 sandstone which had fallen from one of the 
 upper beds of the cliff, the tracks of a small 
 reptile. It is probable that these are . the foot- 
 prints of the small saurian reptile, or lizard 
 Hylonomus Lyelli, Dawson, (Acadian Geology, 
 Dawson 1863 p. 373.) It is probable that the 
 rami of the small reptile which made these 
 tracks did not exceed half an inch in length. 
 These, almost the first vertebrate animals which 
 appeared on this earth, were the small reptiles 
 whose bones Sir William Dawson has found in 
 the base of erect fossil trees belonging to the 
 genus Sigillaria. These delicate little lizards, as 
 324 
 
AMONG THE ROCKS. 
 
 the name Hylonomus indicates, were forest 
 dwellers, and lived chiefly in the base of the 
 coneiferous trees which formed a large portion of 
 the forests of coal period. The slab of sandstone 
 just spoken of now in the Geological Survey 
 museum containing the footprints, was obtain- 
 ed with much difficulty, apart from our having 
 to carry it for more than a mile over a rough 
 sea coast, but one is always ready to risk a little 
 to obtain a good geological specimen. 
 
 The study of the Carboniferous system in Cum- 
 berland county, Nova Scotia, is very fascinating ; 
 the great variety of rocks including sandstones 
 shales, conglomerates, limestones, clays, gravels, 
 etc., with their vastly important bituminous coal 
 beds, and great cliffs of sandstone peculiarly 
 adapted for grindstones, thousands of which are 
 yearly manufactured close to the sea shore, a few 
 miles distant from the Joggins coal mines. These 
 rocks which in Canada are spread over a good 
 portion of New Brunswick and Nova Scotia 
 have an estimated thickness of about 11,000 
 feet, not including the Permian beds. They were 
 formed in the ancient carboniferous seas, and 
 contain, besides a very large number of species of 
 fossil plants, the bones of extinct fish and rep- 
 tiles ; the foot prints of great Saurians and, as 
 we have just seen, the tracks of small lizards 
 whose whole body would not measure more than 
 the length of a claw of the great Saurian who 
 325 
 
EEMINISCENCES 
 
 has, so far as we know, only left his footprints 
 behind. 
 
 From the e ver-in teresting shores of the Joggins 
 coast, where the sea lashes the great sandstone 
 cliffs, which as I have already said occasionally 
 exhibit sections of erect fossil trees, I paid my 
 initial visit to Spring Hill. Arriving there July 
 24th, I at once visited the coal mines and through 
 the courtesy of Mr. Alex. Mclnnes got a slight 
 idea of the extensive mining then going on. 
 
 The village or town of Spring Hill is situated 
 inland a few miles from the Joggins coast, reach- 
 ed now by railway. Geologically, above ground, 
 this is not an interesting locality ; I saw no rock 
 exposures near the village, and the only chance 
 of examining the geological strata is in the mines, 
 or on the dumping ground where one can get 
 some idea of the underlying deposits, and 
 occasional^ pick up a good fossil plant or perhaps 
 a section of a " petrified " trunk of a tree such as 
 those found in the cliffs of the Joggins coast. 
 
 I had only been at Spring Hill a few hours, and 
 was getting much interested in the deposits on 
 the dumping ground where several good fossil 
 ferns were found, when suddenly a severe sick- 
 ness overtook me, resulting probably from some 
 fish I had eaten for dinner. Attended by Dr. 
 Cove and the cook of the hotel I remained in 
 bed five da}'s the longest five days I ever spent. 
 But there was a silver lining to my dark cloud, 
 326 
 
AMONG THE ROCKS, 
 
 which was represented by wonderful extempore 
 music created by the host of the hotel, a young 
 man who had recently become proprietor of this 
 famous old hostelry. 
 
 To Mr. -Alex. Mclnnes, manager of the Spring 
 Hill coal mines, and Dr. Cove, M.A., my thanks 
 were specially due for many kindesses, espec- 
 ially for collecting a number of fossil plants for 
 me during my short but almost fatal sickness 
 which lasted some days after I again returned 
 to Ottawa. It was this sickness and general debi- 
 lity that caused me to apply for superannuation, 
 which after a further leave of absence for four 
 months, during which my time, whenever able, 
 was spent on the Quebec rocks in the interest of 
 the Survey was granted on the 1st of July 1894. 
 
 A few more words before closing these 
 rambling remarks. During my 35 years' con- 
 nection with the Geological Survey of Canada 
 changes have constantly taken place, and con- 
 stant improvements in this Institution have been 
 made, especially since the Dominion Govern- 
 ment decided to remove the Survey to Ottawa. 
 The palaeontological branch has (like all the other 
 branches) prospe.ed under the able chief of that 
 department Mr.J.F. Whiteaves and his assistants, 
 one of whom it has been my privilege to be since 
 the death of the former palaeontologist, Mr. E. 
 Billings. It has not been as some members of the 
 Survey prophesied "the beginning of the end," 
 327 
 
REMINISCENCES 
 
 but rather the beginning of a new epoch in the 
 history of this important institution, which I feel 
 sure will continue to prosper under the able 
 management of its present director, Dr. G. M. 
 Dawson. I may mention here that while the 
 Survey was in Montreal the largest number of 
 visitors, to the museum, in any year, was 1652. 
 In 1897 (this year) the number has increased to 
 32,357. Year by year these numbers have risen 
 and will probably be doubled when a proper 
 building is provided for displaying the thousands 
 of specimens now stored away. 
 
 Passing through the various departments of 
 the Survey one is constantly reminded of absent 
 friends who have long passed away, but whose 
 faithful works in the interests of geology 
 will always remain as monuments of their in- 
 dustry and love of their chosen pursuits. 
 
 In previous pages of these notes I have spoken 
 of some of these diligent workers in our Canadian 
 Geological field, and now to the present members 
 of the Geological staff: Dr. G. M. Dawson, Dr. Bell, 
 Mr. J. F. Whiteaves, Dr. Ami, Mr. Fletcher, Mr. 
 Lambe, Mr. Dowling, Prof. Macoun, and others of 
 the Survey, who are diligently continuing the 
 good work of their predecessors, I tender my 
 sincere thanks for friendship and valued counsel. 
 
 328 
 
RETURN EARTH SCIENCES LIBRARY 
 
 TO -* 230 McCone Hall 642-2997 
 
 LOAN PERIOD 1 
 1 MONTH 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 ALL BOOKS MAY BE RECALLED AFTER 7 DAYS 
 
 Books needed for class reserve are subject to immediate recall 
 
 DUE AS STAMPED BELOW 
 
 FORM NO. DD8 
 
 UNIVERSITY OF CALIFORNIA, BERKELEY 
 BERKELEY, CA 94720 
 
267 
 
 U.C.BERKELEY LIBRARIES 
 
 fill 
 ill 
 
 CD3MblS017