Native lake trout (Salvelinus namaycush) stocks in the Canadian waters of Lake Superior prior to 1955

- 1 - NATIVE LAKE TROUT (Salvelinus namaycush) STOCKS IN THE CANADIAN WATERS OF LAKE SUPERIOR PRIOR TO 1955 John Lawrence Goodier Department of Zoology and Institute for Environmental Studies A Thesis submitted in conformity with the requirements for the Degree of Master of Science in the University of Toronto © John Lawrence Goodier ! - 2 - ABSTRACT The decline of Lake Superior lake trout stocks after the sea lamprey (Petromyzon marinus) population explosion of the 1950s was both dramatic and decisive. Few native stocks remain to permit scientific data collection. However, interviews with old-time fishermen suggest that there had existed many discrete and semi-discrete stocks within the lake. Historical records in the form of government correspondence and records, explorers' accounts, and Hudson Bay Company records yield further accounts. As background to the study, a brief history of the commercial fishery is outlined. Remote stations were scattered along the length of the Canadian border, and supported the fishing enterprises of the main towns: Sault Ste. Marie, Heron Bay, Port Coldwell, Jackfish, Rossport, Nipigon, Hurkett, and Thunder Bay. The first commercial fisheries in Canadian waters were those operated by the Hudson Bay Company. Records are available from 1800 to the 1950s and contain many valuable entries concerning species composition, seasonal movements, spawning grounds, and spawning times. The relevance of these records to fisheries investigations prompted the inclusion of not only lake trout, but all fish species, in this review. Fishermen tended to distinguish varieties of lake trout on the basis of the following criteria: 1. Spawning location and time - Trends in spawning periods may be latitudinal, runs appearing later on the grounds as one moves southward. Local variations can also relate to spawning site, i.e. whether on river, shore, or shoal beds, in windward or leeward zones, or in deep or shallow - 3 - waters. In addition, different breeding groups (runs) might sequentially utilize single grounds (up to four runs being reported at certain sites). 2. Body weight and form - Later runs tended to contain larger and heavier fish. Pulses of different size classes arrived on certain grounds, while inter-shoal variations in average size occurred. 3. Flesh colouration - Flesh colouration may, in some cases, suggest mechanisms ecologically separating lake trout groups during some portion of their lives (such as different feeding niches). Colouration differences, remaining constant through time, existed between certain spawning runs or grounds. 4. Skin colouration - Body colour may represent spawning period responses of certain lake trout groups. 5. Non-spawning movements - While discrete groups of fish were less obvious during the nonspawning seasons, there occurred vertical and/or horizontal segregation of smaller and larger lake trout. Large fish became pelagic in June or July, and some moved inshore; at a few locations a second, smaller "summer run" occurred in mid-August. In the text each of the above criteria is assessed for its validity in stock indentification. Fishermen distinguished four general variants of leans, fats or siscowets, halfbreeds, and humpers or paperbellies S. namaycush. In addition, different "breeds" or forms of lean trout were commonly recognized. Often a form was known by a name referring to its most - 4 - distinctive feature: blacks, redfins, yellowfins, grays, salmon-trout, red trout, moss trout, sand trout, racers, and deepwater leans. A form might have been quite site specific, spawning only on certain selected grounds or might, on the other hand, have composed a general spawning run of fish, appearing at many places along the shore or shoals. Each is discussed in turn, and local spawning habits, grounds, and movements are summarized by lake region. At least 250 former spawning grounds are identified, including 18 rivers which were once characterized by September runs of lake trout. Spawning locations for the deepwater, non-lean trout variants (paperbellies, halfbreeds, and siscowet) are less easily discovered. Approximately 35 fishing grounds are noted, with spawning reported at at least four of these. Consideration is given to the possible existence of phenotypically distinct stocks of siscowet. Due to the nature of the sources used in an historical approach such as this, an abundance of clues for mapping former fish stocks are generated; unfortunately, many must also remain inconclusive. ! - 5 - ACKNOWLEDGEMENTS This thesis was funded by a grant to Henry A. Regier from the Department of Fisheries and Oceans. Many thanks to Dr. Regier for his enduring help and patience. Many employees of both federal and provincial government agencies (listed in Appendix II) were most helpful. Notable among these people were W. MacCallum and M. Purvis of the Lake Superior Fisheries Assessment Unit, Ontario Ministry of Natural Resources. I would also like to acknowledge A. Lawrie for his constructive review of the thesis, and N. Martin for his advice and for placing his wonderful "lake trout library" at my disposal. Too numerous to name, the fishermen and other residents of the Lake Superior North Shore were both invaluable in their assistance and kind in their hospitality. Portions of the thesis were typed by D. Ronayne and A. Young. Regards to T. Whillans for being a source of advice, books, and pens. My special appreciation to my mother, Dorothy Goodier, for her continuing encouragement. I likewise thank the residents and friends of 87 Major for their interest and support, and for many hours of fascinating conversation concerning subjects having absolutely nothing to do with fish. ! - 6 - One fish, Two fish, Red fish, Blue fish. DR. SEUSS ! - 7 - TABLE OF CONTENTS ABSTRACT 2 ACKNOWLEDGEMENTS 5 TABLE OF CONTENTS 7 LIST OF FIGURES 9 1. INTRODUCTION 10 1.1 BACKGROUND TO LAKE SUPERIOR 12 1.2 HISTORY OF THE COMMERCIAL FISHERIES 14 1.2.1 EASTERN LAKE SUPERIOR 14 1.2.2 PORT COLDWELL 17 1.2.3 JACKFISH 19 1.2.4 ROSSPORT 20 1.2.5 THE BLACK BAY PENINSULA AND THUNDER BAY AREAS 21 2. THE FISHERIES OF THE HUDSON BAY COMPANY TRADING POSTS ON LAKE SUPERIOR 24 2.1 INTRODUCTION 24 2.2 A SEASONAL SUMMARY 25 2.2.1 PIC POST FISHERIES 29 2.2.1.1 DECEMBER TO AUGUST 29 2.2.1.2 SEPTEMBER TO NOVEMBER 27 2.2.2 FORT WILLIAM FISHERIES 36 2.2.2.1 DECEMBER TO AUGUST 36 2.2.2.2 SEPTEMBER TO NOVEMBER 40 2.2.3 MICHIPICOTEN POST FISHERIES 45 2.2.3.1 DECEMBER TO.AUGUST 45 2.2.3.2 SEPTEMBER TO NOVEMBER 48 2.2.4 RED ROCK POST FISHERIES 53 2.2.5 FORT SAULT STE. MARIE FISHERIES 54 3. METHODOLOGY FOR STOCK IDENTIFICATION 55 3.1 PERSONAL COMMUNICATION 55 3.2 THE SPAWNING GROUND MAPS 57 3.3 WRITTEN MATERIAL 61 4. NATIVE LAKE TROUT - A GENERAL DISCUSSION 63 4.1 LEANS 65 4.1.1 INTRODUCTION 65 - 8 - 4.1.2 CRITERIA FOR STOCK IDENTIFICATION 68 4.1.2.l SPAWNING LOCATION AND TIME 68 4.1.2.2 BODY WEIGHT AND FORM 71 4.1.2.3 FLESH COLOURATION 73 4.1.2.4 SKIN COLOURATION 78 4.1.2.5 NON-SPAWNING MOVEMENTS 81 4.1.3 NATIVE LEAN LAKE TROUT VARIETIES DESCRIBED BY FISHERMEN 82 4.1.3.1 BLACKS 82 4.1.3.2 REDFINS AND YELLOWFINS 84 4.1.3.3 THE LATE RUNS 86 4.1.3.4 SAND TROUT 88 4.1.3.5 DEEPWATER LEANS 87 4.1.3.6 RACER TROUT 89 4.2 FATS, HALFBREEDS, AND PAPERBELLIES 91 5. A REGIONAL DISCUSSION OF NATIVE LAKE SUPERIOR LAKE TROUT STOCKS 99 5.1 SAULT STE MARIE TO COPPERMINE POINT 99 5.2 COPPERMINE POINT TO CAPE GARGANTUA 104 5.3 CAPE GARGANTUA TO POINT ISACOR 112 5.4 POINT ISACOR TO PLAYTER HARBOUR 118 5.5 MICHIPICOTEN ISLAND TO CARIBOU ISLAND 123 5.6 PLAYTER HARBOUR TO SCHREIBER POINT 129 5.7 SCHREIBER POINT TO NIPIGON STRAIT 136 5.8 SUPERIOR SHOAL 143 5.9 NIPIGON STRAIT TO THUNDER CAPE 148 5.10 THUNDER CAPE TO PIGEON RIVER 154 6. CONCLUSION 160 7. FOOTNOTES 164 8. PERSONAL COMMUNICATION 167 9. REFERENCES 170 10. LIST OF APPENDICES 185 ! - 9 - LIST OF FIGURES Figure Page 1. THE MAJOR FISHING PORTS OF CANADIAN LAKE SUPERIOR PRIOR TO 1955 15 2. LOCATIONS OF THE FALL LAKE TROUT FISHING STATIONS ASSOCIATED WITH THE HUDSON BAY CO. TRADING POSTS OF LAKE SUPERIOR - 1800 TO 1860 28 3. MICHIPICOTEN POST'S LAKE TROUT HARVEST AT THE DOG RIVER FISHERY FISHERY IN 1840 49 4. BODY LOCATIONS OF VISIBLE FAT IN SISCOWET LAKE TROUT 93 5. SAULT STE. MARIE TO OTTER HEAD - SHOWING THE LOCATIONS OF 215 FIGURES SM1-SM17, SM24-30 (see Appendix VI) 6. OTTER HEAD TO PAYS PLAT - SHOWING THE LOCATIONS OF FIGURES 216 SM18-SM23, SM31-SM45 (see Appendix VI) 7. PAYS PLAT TO PIGEON RIVER - SHOWING THE LOCATIONS OF FIGURES SM46-SM75 (see Appendix VI) 217 SMl to SM75 LAKE CHART SECTIONAL MAPS (see Appendix VI) 218-292 ! - 16 - and Michipicoten Island (Fig. SM25; Canada. Dept. of Marine and Fisheries 1871). Both of these locations may have been operated by the Chicago-based firm known as the A. Booth Packing Co. (and later the Booth Fisheries Co.). Established in 1848 by Alfred Booth, it had acquired a lease to property on Quebec Harbour (Michipicoten Island) in 1860 (Mr. I. Purvis, personal comm. 1979). Active fishing on both Lake Winnipeg and the Great Lakes was begun about 1871 (Anon. 1955). Through a series of business acquisitions and market manipulations, the Booth Fisheries Co. grew to be the dominating force in a "fish trust" comprising about ten companies spread from Detroit to Manitoba. The trust claimed a capital of 5.5 million dollars and annual sale of 100 million lb. of fish (Anon. March 16, 1905) In the 1890s Booth assumed control of the Wiarton-based Dominion Fish Co. and extended this company's operations to the Lake Superior trade. Soon most Canadian fishermen in the east were selling to either the parent company or its subsidiary, and few licenses were independently owned. The situation proved cause for concern for one early fisheries inspector: "Mr. Duncan deplores the fact that most of the fisheries of his large district are controlled by a powerful syndicate of United States citizens, who keep the earning rates of our Canadians at a minimum... By employing foreign crews it enables the syndicate to land Canadian catch in their own ports without reporting it." (Canada. Dept. of Marine and Fisheries 1902). Such opportunistic pursuit of the industry extended fishing stations beyond Michipicoten and north along the coast line of present-day Pukaskwa Park. In addition to the major encampments at Quebec Harbour (Fig. SM25) and Gargantua Harbour (Fig. SM12), stations were situated at Caribou Island (Fig. SM30), Dog River (Fig. SM15), Richardson Harbour (Fig. SM18), and Simons Harbour (Fig. SM19) at the turn of the century (Coleman 1899; Duncan July 17, 1899). Booth Company tugs would regularly tour the various - 17 - stations and freight fish to Sault Ste. Marie. In the 1920s, the success of the Booth Fisheries Company began to wane, reportedly as a result of bad management rather than any scarcity of fish. Finally in 1937 its eastern Lake Superior interests were sold to James Purvis and Son, Ltd. of Gore Bay (Manitoulin Island). The new company began fishing with two Booth tugs. Gradually a successful enterprise was built through the exploitation of new grounds, such as Superior Shoal (Fig. SM42), and the opening of new markets, such as those for smoked trout or fat siscowet (see Section 4.2) in the "Jewish trade" of Philadelphia, Brooklyn, and Chicago (Mr. I. Purvis, personal comm. 1979). At its peak the company held four tug licenses for a total of 220,000 m. (240,000 yd.) of gill net. In the 1930s and 1940s there existed a number of fishing businesses, including those of the Mitchell Bros., Gino Nori, Carmen Talerico, the McKays, and the Lapointes. Brief historical reviews are given by Collins (n.d.a), Macdonald (1974), and Marsh (1976). In 1959 the Purvis Fisheries was sold to Ferroclad Fisheries, which still operates from Mamainse Harbour (Fig. SM5). 1.2.2 PORT COLDWELL Like so many small northern Ontario towns Port Coldwell sprang up as a station along the newly completed C.P.R. line. It was soon evident that the snug harbour of this supply depot would also prove ideal for sheltering fishing boats. In 1889 Van Every erected a large ice house and freezer and employed six boats and crews to fill it with fish (Anon. June 15, 1889). Van Every's enterprise appears to have been short-lived, but Case (1892) lists seven - 18 - fishermen as residents of Port Coldwell. By 1895 two tugs were based here, while in 1896 seven pound nets were fished east of the village (Canada. Dept. of Marine and Fisheries 1894-98). Port Coldwell fishermen probably sold their fish initially to the Booth Fisheries Company and later the Nipigon Bay Fish Co. of Rossport (see Section 1.2.4). In 1915, however, the Nicoll Brothers Fish Company was established in Port Coldwell and remained the life-giving force for the town until the 1950s when, like so many other businesses, it declined along with the lake trout populations. The Nicolls proved to be shrewd businessmen. By offering fair prices and opening up new markets other local fishermen were persuaded to sell to them. In the early 1920s the brothers directly held licenses for 47,000 m. (51,000 yd.) of gill net, increasing the number to 77,000 m. (84,000 yd.) by 1929 (Ontario Dept. of Lands and Forests 1920-1929). In the 1920s Port Coldwell fishermen were licensed strictly for grounds in the Heron Bay to Port Coldwell vicinity, but in the 1930s a "pioneering" spirit developed and tugs roamed southward along the present day Pukaskwa Park shore as far as the Otter Head (Fig. SM18; Mrs. H. Cress, personal comm. 1978). Lake trout dominated the town's fisheries and trade in fat trout grew quite lucrative (see Section 4.2). The first tugs to run to Superior Shoal after it was surveyed in 1929 were the Nicoll Bros. "Strathbelle" and Harry English's "Iris" (see Section 5.6). By the 1940s, the 88 km. (55 mi.) trips had been discontinued, however. During the peak years, the Nicoll tugs harvested an average of perhaps 63,500 kg. (140,000 lb.) of lake trout annually (Mr. F. Legault personal comm. 1979). Mrs. H. Cress (personal comm. 1978) reported that catches were still excellent until 1952, but a decline was noticeable in 1953. Today six houses are left standing in the town, but only one family resides there on a permanent - 19 - basis. For additional information, Mountain (1976) chronicles the development and operation of the town's fisheries (with some inaccuracies) 1.2.3 JACKFISH Fishermen were already established at Jackfish when Ben Almos and his two brothers arrived to fish in the early 1880s (Mr. A. Almos, personal comm. 1979). Their numbers were augmented by Peter Dahl, Sr. and Jacob Hendrickson in the early 1890s. However, it is noted by the Dominion Fisheries Commissioner, E.E. Prince (April 25, 1896): " ... fishing has never been very extensively carried on upon this part of Superior, partly because the coast is very rocky and stormy and partly because in former years very destructive wasteful fishing was pursued by U.S. poachers and the supply seriously depleted." Fishing at Jackfish was conducted mostly from rowboats and later gas boats. A single rowboat fisherman licensed for 7300 m. (8,000 yd.) of net averaged approximately 5900 kg. (13,000 lb.) of lake trout annually (Mr. A. Almos, personal comm. 1979). The Slate Islands (Fig. SM38) proved most excellent lake trout grounds. Originally privately leased to Lieutenant Governor Patterson, the earliest fishing license for the Slates was probably issued in 1884 (Kerr Feb. 13, 1886). Production levels were high for a time: "The fishing on these islands was formerly leased to Messrs. Noble who grossly abused it, and for many years it was practically worthless. The Department has applications from C.W. Gauthier and others but as there is ground for believing that these applicants would simply destroy the fisheries in the same way that the Nobles did it was decided not to grant any license in the future." (Prince Dec. 2, 1897) The Nobles were managers of the Dominion Fish Company. In the 1930s three fishermen were permanent residents of Jackfish, all sons of original inhabitants. According to Mr. A. Almos (personal comm. 1979) the most successful fishing period was from 1940 to 1948. Today only building - 20 - foundations mark the townsite. Additional historical details are to be found in Mercier et al. (1973) and Poole (1976). 1.2.4 ROSSPORT In the early years of the north shore fisheries, inhabitants maintained only tenuous contact with the outside world and must have often endured a profound sense of isolation. Although the fishery at Pays Plat Bay (Fig. SM44) was still under control of the Hudson Bay Company in the 1870s (Kerr Nov. 21, 1871), other men from the south arrived to establish scattered and meagre camps on the islands south of Nipigon Bay. William Boon began fishing on Bowman Island (Fig. SM55) in the 1870s and was followed by George Gerow in 1881. Osler (1868) encountered fishermen on St. Ignace Is. (Fig. SM54), where he also found the wife of the 1iqht-house keeper suffering in the latter stages of consumption. Years before settlers arrived at McKay's Harbour (later called Rossport) fish were cured in sugar and salt at Chummy Point (Chubby Harbour; Fig. SM45) to await the infrequent arrival of the lake steamers. Most camps were abandoned in the fall. By 1894 plans were being laid by the Booth Fisheries Company to force inroads in the Rossport industry. A letter to their Port Arthur representative expressed the desire to raise the price of fish in Rossport at the expense of the recently established Rossport Fish Company, adding that this was "liable to stir up a great deal of strife" (Turner Sept. 17, 1894). At the turn of the century, post office directories (Henderson 1900) listed 18 resident fishermen at Rossport and 9 at Nipigon. About this time J.A. Nicol took the post of Rossport station master. Struck by the possibility of shipping fresh fish on the C.P.R. line, he built a small ice - 21 - house and entered the industry as a buyer. From this modest start, his Nipigon Bay Fish Co. grew to be one of the major enterprises along the north shore, and acted as a competitor to the Booth Company. In 1940 a boxcar load of fish was being shipped each day, and fish was being brought from as far away as Black Bay (Mr. C. Macmillan personal comm. 1978). The company survived until 1953. In addition to the Nipigon Bay Fish Company, a couple of smaller independent firms flourished in Rossport. The Bowman Fish Company was started prior to 1910 and in 1918 entered the fish trade on Lake Nipigon. John Paulmart established the Independent Coop around 1915 and his company endured into the 1940s. 1.2.5 THE BLACK BAY PENINSULA AND THUNDER BAY AREAS Despite the decline of Fort William as a fur trading post, in 1866 Fountain (1904) found it to be, "... a great fish-curing depot, and this trade seems now to give it all its importance." By 1870 two .steamers, as part of their regular cargo, were carrying fish to Collingwood, and in 1875 stations were situated as far afield as Victoria Island (Fig. SM74) to the south and Point Porphory (Fig. SM63), Shaganash Island (Fig. SM61) and Roche Debout Point (Fig. SM59) on the Black Bay Peninsula (Anon. 1873; Canada. Dept. of Marine and Fisheries 1875). Near the end of the 1870s only local natives were actively fishing at Fort William, and in 1881 the Hudson Bay Company post closed completely. This decline was paralleled by the growth of the fishing industry in Prince Arthur's Landing (which became Port Arthur in 1884). In 1879 the first private company, the Lake Superior Fish Co. was established. It is not certain if it was American controlled, but much of its catch was shipped - 22 - through Minneapolis. A wharf was built, a tug ran daily to the various stations, and the business developed rapidly. By 1883 the company employed 30 men and maintained two steam tugs and 6 large fishing smacks (Anon. July, 1883). The growth of the industry during this decade, and its success in obtaining fish from waters that must have seemed to possess unlimited supplies, attracted many fishermen from other areas. The fate of the Lake Superior Fish Co. is uncertain, but it is probable that it became enfolded into the growing conglomerate of the Booth Fisheries Co. In 1884 a Booth steamer was making regular runs between Duluth and Port Arthur. It was supplied by the newly established Port Arthur Fish Co., which soon was in virtually complete control of the fisheries from Thunder Bay to Heron Bay (Anon. 1889). ln 1891 seven tugs and many sailing vessels were docked at the Lakehead (Port Arthur Board of Trade 1891): "At little Port Arthur alone the figures of the fishing industry for the market are astonishing. In 1888 the fishermen there caught 500,000 pounds of whitefish, 360,000 pounds of lake trout, 48,000 pounds of sturgeon, 90,000 pounds of pickerel, and 30,000 pounds of other fish, or more than a million pounds in all." (Ralph 1892) In the 1890s, a number of new businesses were established, including the Union Fish Company and Gagne Bros. Fishery (Henderson 1900; Anon. Jan. 23, 1892). More fishing communities and stations were also established along the Sibley and Black Bay Peninsulas. By 1883 a fishery existed at Black's Dock on Sheesheeb Point (Fig. SM59; Selwyn 1883). Fishing out of Hurkett began around the turn of the century and in 1919 A.W. Nuttall employed 16 men for the winter fisheries of Black Bay (Mrs. N. Thrower, personal comm. 1978). In addition, south of Thunder Bay fishermen were established in the Cloud Bay area in the 1920s (Mr. V. Bergman, personal comm. 1980). These fishermen sold their fish not only to the Booth Co. but also to the Fort William Fish Co., which had been established by Thomas - 23 - Craigie in 1916. In the west end, as in the east, Booth sold its Lake Superior assets about 1934, initially to the Nipigon Bay Fish Co. of Rossport. In the early 1930s, the Royal Fish Co. opened but was soon acquired by Kemp Fisheries of Duluth. This company has today become the dominant force in the western Lake Superior industry. In the 1930s fishing camps were established at Tee Harbour (Fig. SM66) on Black Bay, and in 1939 the first herring fishery was based at Squaw Bay (Fig. SM64; Mrs. E. McKay, personal comm. 1979). In the early 1940s the Tee Harbour camp was moved to Camp Bay (Fig. SM65) where today two active fishermen still remain to carry on the tradition. - 24 - 2. THE FISHERIES OF THE HUDSON BAY COMPANY TRADING POSTS ON LAKE SUPERIOR 2.1 INTRODUCTION Upon examining the written collections of the 18th and 19th century fur trading enterprises, one discovers a wealth of observations concerned with the animal resources of northern Ontario. To sift through such voluminous records requires both time and patience, but one is frequently rewarded with an intimate and detailed view of the interactions of humans and the early Canadian environment. The purposes of the trading corporations were mainly exploitive; to extract the material wealth of the land and so pursue rising profits. Yet the people they employed to man their New World outposts were likewise moulded by this land, as they sought to familiarize themselves with (and adapt themselves to) an environment which frequently grew hostile. For the purposes of this chapter, research was generally limited to the journals, correspondence, and annual reports of the Hudson's Bay Company. The original documents are now housed in the Provincial Archives of Manitoba, with microfilm copies of the written records retained in the Public Archives of Canada in Ottawa and the Public Record Office in London, England. Appendix I references the material I have examined. The authors of this material were not scientists, but many were well educated and acute observers of nature. They took a keen interest in the habits of the animals of their land, knowing that both the economic welfare of their company and the physical survival of the post inhabitants depended upon this knowledge. It was in their own interests that the directors of the Company frequently stressed the importance of maintaining detailed post journals. Directives were dispatched from London outlining the matters and affairs appropriate for journal entry; these have been reviewed by Moodie (1977). Since the journals and reports were written in the terms of the scientific layman, a certain amount of speculation and inference is required in their interpretation. Moodie and Catchpole (1975) discuss analytical problems and have successfully applied techniques of content analysis in reconstructing, from the Company journals of the Hudson Bay district, - 25 - patterns of estuary freeze-up and break-up. Although the methods of content analysis were found to be inapplicable to the present study, still it was possible to make qualitative assessments of the habits and abundances of pre-stressed fish populations in Lake Superior. 2.2. A SEASONAL SUMMARY A number of fur trading enterprises preceded the Hudson Bay Company to Lake Superior country. Posts had been maintained by the French since the early 18th century. Finally in 1804, the rival XY and North West companies merged to form the North West Company and vigourously competed with Hudson Bay Company interests in the Far West. The British firm countered by erecting inland posts between Lake Superior and Lake Athabasca in close proximity to those already established by the North West Company (Public Archives of Canada 1974). On Lake Superior the small establishment of Pointe de Meuron (later Fort William) challenged the monopoly of the French post near the mouth of the Kaministikwia River. From 1797 to 1803 and 1816 to 1821, Hudson Bay Company employees resided a short distance from the French fort at Michipicoten. The amalgamation of the two companies in 1821 brought the N.W.Co. forts under Hudson Bay Company jurisdiction. In addition to the major posts of Michipicoten House, Pic Post, Fort William, and Sault Ste. Marie, seasonal outposts were maintained at the mouth of the Agawa River1, on Batchawana Bay2, and at Red Rock (situated at the head of Nipigon Harbour). These subsidiary posts engaged in limited fisheries but were mainly supported by their parent establishments. Several authors have summarized the fishing activities of the Lake Superior fur trading establishments. Nute (1926, 1944) reviews in detail the history of the American Fur Company's involvement in the industry. Sections on the fisheries of the major Hudson Bay Company posts are presented by Marsh (1976), Weiler (1973), and Campbell (1976). The annotated bibliography of White (1977) lists sources relevant to fisheries on both sides of the lake. In - 32 - commersoni) also distributed throughout Lake Superior. It was noted on April 19, 1832, however, that "14 Red Suckers" were caught with the seine. These undoubtedly were the longnose or northern sucker which assumes a distinctive red mid-lateral colouration at spawning time. A single reference is made to a catch of "Red Carp" from a gill net (P.P.J. April 10, 1830). This may involve a confusion of the names "sucker" and "carp" but more likely refers to the redhorse (probably Moxostoma macrolepidotum): the laterally compressed body of the latter lends to it the general form of the carp. Gourley (1822) remarks that, in addition to suckers, two species of carp were identified in Lake Superior waters. The true carp (Cyprinus carpio) is an exotic to North America, and was not recognized in Lake Superior until 1915 (MacCrimmon 1968). The fish referred to as "dorees" were probably walleyes (Stizostedion vitreum). Occasional reference is made to the "pickerel", but the names are likely synonymous here. Agassiz also identified in Lake Superior the sauger (Lucioperca americana, now known as Stizostedion canadense),a fish similar enough in appearance to the walleye to be easily confused by the fur traders. Narrator J.E. Cabot recounts Agassiz's experiences at the Pic Post on July 13, 1848: "The Professor got a number of fishes, among them a brilliant green pickerel, a new species. A sturgeon was caught in the river opposite our tent, in a net belonging to one of the Indians, who dispatched him after some contest, with a fish spear. Prof. Agassiz requested me to make a sketch of this fish, which was some four or five feet long." (Agassiz 1850). What Agassiz called "pickerels" were actually Esocidae, and in Lake Superior he identified Esox boreus, or the northern pike (known now as Esox lucius). The Pic journals mention only once the netting of a pike (P.P.J. May 1, 1828) and also once a jackfish (P.P.J. May 2, 1828). Both probably refer to E. lucius. 3 From the dearth of catch we must conclude that this - 33 - species was rare in the vicinity of the Pic River (Fig. SM23). Some sturgeon (Acipenser fulvescens) were taken in nets, at an average of only 6 or 7 a year. Sturgeon nets were made of a special coarse material known as "No. 4 Twine": "Seined Morning and Evening but caught nothing. tried the Drag Seine as Sturgeon are leaping in the River but caught none, no one here, except myself know how to use it, never having fished in that way. they are ackward." (P.P.J. April 27, 1832). A few references are made to "perche" (Perea flavescens) taken during seining operations at the mouth of the Pic River and at Heron Bay. A single reference is made to a "Macqua" caught in a net July 13, 1827. This "breed" of siscowet was identified by Barnston (1874) and called Salmo ursinus (See Section 4.2). 2.2.1.2 SEPTEMBER TO NOVEMBER Fall fishing was devoted almost exclusively to the salting of lake trout. It was necessary that some stations be established at great distance from the Post for spawning runs were absent at the Pic River itself. Silt may have been to blame: "The River Peck takes its name from an Indian word, signifying mud, as it pours out an ash-coloured, and when swollen, a reddishyellow water, tinging the lake for a mile or two round its mouth, and derived from beds of yellow and white clay some d stance up the river." (Bigsby 1850). Even though the post was a small one, sufficient supplies of fish for winter consumption were far from guaranteed. Therefore, at least two, and sometimes three, stations were established. The most important were at Pays Plat, in the area of Santoy Bay and Bottle Point (called in one journal "la peche a la gros truite a l'ance aux Bouteille". P.P.J. Oct. 8, 1829), and at a site known only as "Old Sansregrette's Fishery" but probably referring to Simons Harbour.4 The distant trout fishery at Pays Plat lasted for many weeks, being - 34 - established late in August or early September and lasting into November. This suggests an extended spawning period and possibly the existence of more than one run of lake trout (see Section 5.7). Most likely, however, the majority of lake trout spawned by mid-October. An entry in the 1845 journal refers to a fishing station situated probably at Pays Plat: "In the evening the Fishermen arrived in the boat with 40 barrels salt Trout. The Trout are spawning much latter this fall than usual, which is the cause of their being so long filling the casks. On the 1st instant [October] last fall, they had secured, and brought home 70 barrels." (P.P.J. Oct. 29, 1845). The fact that a fishing station would be established a full 96 km. (60 mi.) distant from the post suggests the importance of the Pays Plat as both a breeding ground for fish and a safe harbour for fishermen, unrivalled by points farther east. Fishing at "l'Ance a la Bouteille" was conducted at Bottle Point, along the shores of Santoy Bay, and in the Steele River. In 1830 for the first time, a "weir"5 was erected in the river and tended from about September 15th to October 21st, and by October 5th had procured only half of its final yield (23 barrels). If this represents th typical peak for the Steele River spawning run, it is later than usual for rivers in the northeastern regions of Lake Superior (see Loftus 1958 and Section 5.4). However, as will be discussed below, dates for river runs could vary significantly from year to year. In certain years, in fact, lake trout would fail to enter the rivers entirely: "Cadrant and Party arrived. they only Pickled 30 Casks, 26 Less than last Autumn. the constant stormy weather was against them and at Bottle Bay River, the Trout failed, few or none entered that River to Spawn. Cadrant in a gale of wind lost a Net and 3 Cod Lines, which were carried away." (P.P.J. Nov. 2, 1833). Several journals refer to "Old Sansregrette's Fishery", where a majority of the catch was obtained prior to the start of the third week of October. An unnamed station, probably Old Sansregrette's, is mentioned in the 1828 journal: - 35 - "... Cadrant has 10 Barils of Troute Salted and ... the Troute now begins to approach the Shores of the Lake." (P.P.J. Sept. 26, 1828). Little else may be inferred, however, about the spawning period of the trout caught here. There were other more distant and less popular spawning grounds where stations were also occasionally established. In 1833 a seine fishery was maintained at the Pukaskwa River from September 4th to October 5th. Loftus (1958) notes that lake trout typically spawned in the river from September 5th to 15th (see also Section 5.4). Thompson (1822) arrived at Pic Post September 8th, only to find all hands "absent to salt trout in the Bay of Islands near the Otter Head." (This location may refer to Richardson Harbour). It is noted in 1827 that at "a bay this side of the White River", the fishermen abandoned their stat1on on October 16th as: "... they had no success at this Fishing as the Trout had done Spawning, consequently left the Shallows for the deep." (P.P.J. Oct. 16, 1827). (Between the Pic and White Rivers are a series of sheltered bays. The largest is Playter Harbour). Also in 1827, a lake trout fishery was established, for the first time, at Les Petits Ecrits. Obviously there were not as many late-spawners here as at "Old Sansregrette's Fishery", for all fishing efforts were usually shifted to this latter place after the first week in October (McIntosh Feb. 3, 1828; Swanston Sept. 27, 1828). Reference is also made to the presence of a lake trout fishery a little north of Sansregrette's, possibly at the White Spruce River (P.P.J. Sept. 22, 1827). Finally, two other "secondary" stations are mentioned: Pic Island (P.P.J. 1833) and a place about 19 km. (12 mi.) west of the post and known as "Isle Rouge" (P.P.J. 1840). Its exact location could not be determined. Whitefish formed a small and occasional part of the fall catch. It was found that whitefish spawned in Oiseau Bay (called Louison's Bay in - 36 - the journals; (Fig. SM21) in November (P.P.J. Nov. 6, 1832). The frequent failure of the stations in the Pic River region to yield whitefish in November may have been the consequence of rather late spawning periods: "La perdrix Blanc and an Indian who had remained at Ance a la Bouteille to catch White Fish arrived. he did not catch a single one. he tried several Places but no Fish Spawns or else they had done and retired to Deep Water, however on enquiry from the Indians I was told White Fish spawns late in the Season about the commencement of December." (P.P.J. Nov. 3, 1883). Such late spawning was not necessarily typical of whitefish caught at other points in the lake. At Pays Plat, for example, 300 were obtained prior to November 9th, 1835, and it is likely that these fish were approaching the shores to spawn. It will also be seen that stations operated by Fort William and Michipicoten House could depend upon whitefish catches earlier in the autumn. In the fall months, a limited amount of netting was continued at or near the Pic River. With the seines were captured "dorees" and suckers. From the nets and lines some sturgeon, whitefish, and trout were taken. Often the local fishing was discontinued entirely in order that all nets might be pressed into the service of the distant stations. 2.2.2 FORT WILLIAM FISHERIES 2.2.2.l DECEMBER TO AUGUST Winter fishing began as soon as ice had firmly set in Thunder Bay. Venturing across the ice, fishermen would set up to 20 trout lines along the west end of Pie Island or would range them northward toward the Welcome Islands. It was the practice of some Indian families to quit the fort and encamp "toward the Bottom of the Bay", possibly near Caribou Is. (Fig. SM68) where they knew the fishing usually surpassed that found nearer the establishment. Indians remaining about the post speared trout but - 37 - seldom managed to take more than a few at a time by this method. The general scarcity of lake trout in these waters suggests that their wintering grounds may have been located in the deeper waters of the western portion of the bay or to the south, outside of Thunder Bay entirely. Other native winter camps existed at Black Bay and McNab Point (Point Brule; Fig. SM71), but probably none could depend upon a steady supply of fish. To augment the winter stores, men from the post would regularly cross to the Welcome Islands and set 3 or 4 nets for suckers. Some years these fish were quite abundant. Nets might be left in the water for over a week and, with luck, yield over 100 suckers. Their proximity to the fort, and their relative importance as a source of fish in the hungry winter months, made these islands attractive to fishermen: "Visinau went to Welcome Islands in order to mark out the usual places where the company sets nets under the ice, before the freeman & Indians takes possession of the best stands." (F.W.J Dec. 26, 1831). Lines and nets were moved closer to shore as the ice grew thin in April, until finally it was too weak to walk upon. At this time fishing was suspended at the lake. As soon as the ice cleared, lines and nets were set opposite the mouth of the Current River (Fig. SM70). Here fishing for "trout" continued from the first week of May into June. These trout may have been brook trout (Salvelinus fontinalis) which would more likely be found in shallow water at this time than lake trout. Fishing within the river itself proved to be of but mediocre success, "the rocky bed of the stream not being favourable to seining operation" (quoted by Russell 1915). A nearshore gill net fishery for lake trout was established the last week in May and continued throughout June. By the first week of July, fish were becoming scarce in nets set at Pie Island, Shangoina Island (Shagoinah - 38 - Island), and the Welcome Islands, and the fisheries tended to fail by the end of the first week of July, presumably when the lake trout returned to the deeper waters of the bay. At the Shangoina Islands, an especially large summer run would appear. After the Company entered the fishery on a commercial basis, many barrels were transported from here: for example, in June of 1837 (a good season) 63 barrels of trout were salted. Swanston (June 30, 1835) remarks that the station was maintained on the west side of the island. Twice in summer whitefish would approach certain shores of Thunder Bay, where they fell victim to the post's seines and gill nets. The first run occurred in the last week of June or first week of July, while the second arrived in mid-August: "... it is an extraordinary thing that we can not catch any whitefish this late season when in former years there used to be an abundance about the middle of the month..." (F.W.J. Aug. 25, 1826). In great numbers these fish would congregate at the entrance of the Kaministikwia River. The Reverend Mountain (1844) informs us that, "Five thousand of these fish were taken in one morning before breakfast during the past summer". This pattern of two summer runs or "false spawns" of whitefish is confirmed by Purvis (1977). However, he sets a later peak for the first run; in mid-July and lasting approximately one week. Also, a longer duration for the first run is suggested by the journals. A meteorological record, included at the end of the 1839 journal supplies the following details: June 21: The small white fish come to the entrance of this river in shoals June 26: Heavy rain during the night. The whitefish have departed from the entrance of the river. June 29: Perfectly calm at sunset. The whitefish have returned. took 2506 this morning in 2 hauls with the seine... July 3: The whitefish have again disappeared from the entrance of the river. - 39 - The failure of trout-line fishing in summer was at times due to a scarcity of herring for bait. Unlike the posts at the Michipicoten and Pic Rivers whose fisheries procured herring inshore during June and part of July, herring never formed a significant part of the Fort William fisheries. No summer inshore movements are noted, and few seining attempts were ever made at this time. The failure of these species as a exploitable resource is also discussed in Section 2.2.2.2. In the first half of May, sturgeon began to appear in nets set near the mouth of the Kaministikwia River. Therefore, they were to be found running in this river at a later date than in the Pic River (see Section 2.2.1.1). In June the "Big Forks" (Mission R.; Fig. SM71) was a favourite place for seining. Flooding in the months of May and June often made seining difficult and at times impossible. During the second or third week of July sturgeon moved into the rapids to spawn. Great numbers up to large sizes were occasionally reported (F.W.J July 12, 1824): entries for 1823 record a 34-kg. (76-lb.) sturgeon (June 11), a 31-kg.(68-lb.) specimen (June 16), and finally, on July 11th, a 136-kg. (300 lb.) giant. This last enormous creature approximates the upper reported range for the species; Carlander (1969) cites a 125-kg. (275 lb.) female from Lake Winnipeg, Manitoba as the largest documented. In the early 1900s, however, a sturgeon weighing 141 kg. (310 ib.) was pulled from the waters of Batchawana Bay (Drew and Littlejohn 1975). Suckers appear to have been common in and about the Kaministikwia River from early May until mid-July: "Seine gave 300 suckers". (F.W.J. July 4, 1826). "All the Indians and Freemen about the place have been setting nets and are partly successful in catching suckers and pikes." (F.W.J. May 13, 1827). "Michel arrived from the Rapids. brought 95 suckers." (F.W.J. May 17, 1827). Suckers found in the stream after mid-May were more likely to have been - 40 - Catastomus commersoni than the longnose sucker, which spawns at an earlier date and for a shorter period. It is written that: "The principal and indeed the only river of note is Kamanaistiquia or Ft. William.... It may also be remarked that it abounds with all kinds of fish, peculiar to the country, such as sturgeon, whitefish, perch, pike, pickerel, suckers and a few catfish." (F.W.R. 1828). Catfish may refer to the brown bullhead (Ictalurus nebulosus) or the channel catfish (Ictalurus punctatus), although Scott and Crossman (1973) record the presence of these species south of Sault Ste. Marie only. Perhaps the burbot (Lota lota) is a more likely candidate. 2.2.2.2 SEPTEMBER TO NOVEMBER Early in the fall, occasional catches of suckers and pickerel were made with nets and seines employed in the vicinity of the fort. The main efforts of the fishermen, however, were devoted to the capture of lake trout and whitefish. While encampments were established at the more distant stations, nets set at Mutton Island (also called Sheep Island) were checked every two or three days (weather permitting) to supply those people remaining at the fort with fresh fish. In Thunder Bay two runs of trout were observed. Those at Mutton Island were a smaller, early-running variety and usually left the grounds in the last week of September. At this time fishing was discontinued here, and the men and nets dispatched to aid fishermen already engaged at Pie Island. Large trout were then coming into shallow water along the shores of the island. This variety did not spawn in any significant numbers at Mutton Island; nor, apparently, were they common at the Welcome Islands, which were the site of a minor station established for the exploitation of the earlier, September-running lake trout. After 1830 two or three stations were maintained each year at Pie - 41 - Island and were probably located somewhere along its north shore (F.W.J. Sept. 12, 1835). It was possible to still net trout in the second week of November when fishing efforts were usually suspended. Stations were also established at both Thunder Cape (Point Tonnère) and Hare Island (Rabbit Island). It was usual to commence setting nets at the Cape first and, as these trout began to fail, shift to the island grounds where spawning tended to continue later (about the first week of November). In 1839, the first year of the commercial fisheries, Fort William supported 17 stations. The most productive were those established on Shangoina and its neighbouring islands. While crews might arrive here early in September, the lion's share of their barrels were not filled until the large trout commenced spawning toward the end of the first week of October: "...it is astonishing what a number of Trout there are of this season amongst these Islands, many of them weigh from 15 to 20 lbs and very few less than 10 lbs... (F.W.J. Oct. 15, 1836) One may infer, therefore, that this run of large trout ended earlier than those of trout casting spawn on the shores of Pie Island or Hare Island. The Shangoina spawning terminated after October 20th when the fishermen would turn their attention to the whitefish fishery. In the 1840s there were three manned stations situated among these small islands. The yield here in 1839 was 21,500 kg. (47,400 lb. or 237 barrels) of whitefish and trout, while almost 7,700 kg. (17,000 lb.) were obtained at the Pie Island stations. Prior to 1836, a station was occasionally established at a place known as Pointe au Pere. Although less important as a trout fishery than those discussed above, late season gill netting was possible in November. In both 1826 and 1829, the Welcome Islands (also called Les Isle Travers) proved a successful station. The presence of an native fishery at Sturgeon Bay is also noted (F.W.J. Sept. 17, 1839). Closer to home, trout - 48 - smallmouth bass (Micropterus dolomieui), a species which would also have been uncommon in this region. 2.2.3.2 SEPTEMBER TO NOVEMBER Of great importance as a fall fishery was the Dog River, where, prior to 1821, Hudson Bay Company fishermen would ply their nets in close proximity to those of the rival North West Company. Until the 1820s the Dog was known to all simply as "Trout River". In early years riverspawning lake trout were exploited solely with seines (of varying dimensions). Then, in 1829, there was introduced to Lake Superior the fishing weir, an innovation modelled upon those employed by both natives and white men west of the Rocky Mountains (M.P.J. Sept. 5, 1829). Proving a great success, it yielded at the Dog River 36 barrels of lake trout that year: "not only with respect to the productiveness (equivalent to the usual result of at least three fisheries (27 Nets) and the labour of 6 men during a period of 40 days) of the experiment (which occupied a period of 22 days and the labour of one man with his family) but also productive of immense saving of labour and considerable expense of fishing Tackle." (M.P.J. Sept. 26, 1829). The advance guard of the fall runs began to gather about the river mouths, preparatory to moving upstream (where they were followed by the seiners). Journal entries indicate that the spawning period at the Dog River could vary greatly from year to year. Fig. 3 depicts catch data recorded for the year 1840 (M.P.J. Sept. 21, 1840). We see that the actual period of spawning lasted less than 2 weeks, and that few trout remained after September 20th. Similarly, in the years 1829, 1830, and 1839, the conclusion of spawning, and the general retreat of the stock, occurred by September 23rd (earlier than dates recorded by recent authorities see Section 5.3). On the other hand, certain years were characterized by a delayed spawning period: ! - 49 - ! - 50 - "... the Trout are not begun to gather in the River..." (M.P.J. Sept. 13, 1799). "the fish were not plentiful till the last four days." ( M.·P.J. Sept. 25, 1799). "It was near a fortnight [after September 9th] ere the fish were plentiful enough to admit of [the men] curing any during which time they could not get sufficient for their own subsistence. tis but these few days past that they became numerous." (M.P.J. Oct. 3, 1801). In 1840 the average individual weight of Dog River lake trout was 1.2 kg. (2.7 lb. -- i.e. 19 barrels were found to hold a total of 1412 fish; M.P.J. Sept. 21, 1840). Weir fisheries were also established at the Makua River8, and the lake trout run here proved almost as productive as that of the Dog River. We also learn of an unsuccessful attempt in 1828: "... to make a fall fishery at Montreal River (which got a great name last year for being abundantly stocked with excellent Trout) which entirely failed and where too much time was lost..." (M.P.J. Oct. 22, 1828). No further reference is made to establishing a fishing station at Montreal River, although it was frequented by native fishermen. Stream-running stocks of lake trout may have been a common phenomenon in this region, frequenting not only the larger streams: "Sent Mr. Robertson to order McKay to establish a Barriere at a small rivulet (a little beyond his station [at the Dog River]) where I hear Trout cast their spawn." (M.P.J. Sept. 10, 1830). Two fishing stations were located about Cape Gargantua; one at a place known as the "Stoney Islands" (which may allude to the area about Devil's Warehouse; Fig SM12; M.P.J. Sept. 19, 1827). Spawning grounds also existed about Cap Chaillon, where a station was periodically established at a place known as "Mousseau's ascent to his Sugar Bush" (M.P.J. Oct. 6, 1828). Occasional mention is made of a native fishery at the "Ecores", possibly referring to Point Isacor (M.P.J. Sept. 2, 1839). In addition, it is noted September 9th, 1858 (M.P.J.), that a fishing party was dispatched to Michipicoten Island for lake trout, equipped with "a lake boat with jib - 51 - & main sail & spars". The inhabitants of Michipicoten House, like those of Fort William (see Section 2.2.2.2), identified two varieties of lake trout, distinguishing them according to size and spawning sequence. The first run of small trout included both river and early shore-spawners, while large trout would often run through October and past the first week of November. At Cap Chaillon, for example: "they have collected a very superior lot of rich Trout and Whitefish.... [Boucher] reports that some kind of Trout have not concluded casting their spawn but his salt and Barrels were spent." (M.P.J. Oct. 16, 1839). Trout and whitefish could run concurrently, and were not necessarily isolated by their spawning periods as was the case at some other locations in the lake (as at Pic River or Shangoina Island, both previously discussed). It is also interesting to note that lake trout and whitefish south of Michipicoten Bay would spawn to a later date. The following information is extracted from the Thermometrical Journal for 1839: Sept. 4-6 Small trout collecting at Small Rivers to Cast their spawn Sept. 13 Small trout begun spawning Sept. 23-24 Small trout in this vicinity done spawning Oct. 18-19 Spawning of small trout about ended Nov. 18-20 Large trout & Tittamingue cease spawning in this vicinity altho not elsewhere In 1832 lake trout disappeared from the area about Perkwakwia Point around October 8th (M.P.J.). Sometimes November whitefish fisheries were established at the Point, but these were quite minor in scale. The earliest fall whitefish were those approaching the shores ·of Michipicoten Bay prior to spawning in the river. These tended to be smaller than whitefish taken elsewhere, but at an average weight of 0.7 kg. (1.5 lb.) each, they were heavier than the fish of the Kaministikwia River stock (Section 2.2.2.2).9 The earliest recorded date on which whitefish began entering the river was September 10th in 1859. Typically, however, upstream movements - 52 - commenced between September 23rd and October 3rd. Seining operations were occasionally extended a distance of 16 km. (10 mi.) upstream to Scot's Falls (but it is likely that only a small percentage of the stock would migrate this far): "The men took the seine up the River almost to the Portage and descending thru it out at 10 different places without catching a fish -- In the evening at the mouth of the River 26 W. Fish were caught." (M.P.J. Oct. 11, 1827). "The seine was worked today in different places of the River down from the Grand Peche [exact location not known] and only procured one Tittamingue and two pickerel fishes. Hence we presume the fish have got up amongst the Rapids and dispersed here and there." (M.P.J. Oct. 9, 1829). Other favourite locations for plying the seines included the mouth of the Magpie River, the "first point above the House", and Point Parisienne (location not known -- M.P.J. Oct. 8, 1829). As in the Kaministikwia River, rapid waters were preferential spawning sites. In certain years the river would attract whitefish for over 3 weeks, and in 1830 supported a seine and weir fishery from September 23rd to November 10th. Sensitive to turbid waters, however, whitefish would avoid the river in times of storm and resulting siltation. Natural modification of the river bottom (including, no doubt spawning beds) was a common phenomenon in the Michipicoten River and other rivers of the North Shore: "... [the Tinsmith] is besides very expert at working the Seine, by which some years ago a great number of white fish used to be caught during the spawning season in this River -- an alteration since in the bed of the River occasioned by a flood or flush of water seems to have given occasion to the fish to resort elsewhere." (M.P.J. Sept. 25, 1828). Storm damage could be quite dramatic: "In the evening the gale increased to a perfect Hurricane, rolling the Water of the Lake into the River at a most rapid and alarming rate flooding and scouping off its banks everywhere - so much so that a strong quay at the end of our Mess House was demolished and the Hause itself partly undermined. Indeed the violence of the gale and consequent influx of water in the River surpassed anything of the kind remembered." (M.P.J. Oct. 25, 1829). - 53 - The Dog River was likewise susceptible to the effects of wind and sea, with their power to interrupt normal spawning patterns of the lake trout and whitefish stocks (the latter arriving to cast spawn a short time after the trout had departed -- M.P.J. Sept. 27, 1831): "J. Clousten and assistants arrived from the fishery at Dog River - which owing to the frequent furious gales and heavy rains, is quite over - The bed and channel of the river being quite cut up and deranged." (M.P.J. Oct. 29, 1831). In October incidental catches of lake herring were periodically made in whitefish seines plied in the Michipicoten River. No spawning run is described, however, and the fish were at no time remotely as abundant as in July. C. F. George Keith noted the presence of a variety of fish he identified as a "Ciskeche": " a small species about the same size and bearing considerable resemblance to the Herring." (M.P.J. Oct. 17, 1831). Keith also refers once to the capture of four small "Toulibee" in the river (M.P.J. Sept. 23, 1840). Certainly the names cisco, herring and tullibee have all been used to describe Coregonus artedii. Koelz (1929) saw fit to recognize two subspecies in Lake Superior, C. artedii artedii and C. artedi albus, the former distinguished by its body of slenderer proportions. Jordan and Evermann (1908) added another, C. artedii arcturus to the list. While it is unlikely that any of these forms should be granted subspecific status, it is possible that they represent different stocks of ciscoes. These were perhaps familiar to the fishermen of Michipicoten Post. 2.2.4 RED ROCK POST FISHERIES The Hudson Bay Company post on the northwest shore of Lake Nipigon maintained an outpost at the mouth of the Nipigon River. It was a small, relatively minor establishment, and Stevenson (1865) describes the site in its later years: "The station consisting primarily of two large log buildings, viz a - 54 - one-roomed dwelling house and a store - also a potato patch, has an unimportant neglected look. There are besides two small log shanties ... Outside the enclosure there are a few wigwams, but as we had already heard - all the Indians had gone up the river to hunt and collect birch bark " Whitefish tended to dominate the fall catch. In 1837 the total harvest was 5,500 whitefish and 2 kegs of Trout (either lake trout or brook trout). It may be noted, in passing, that the fishing stations situated on Lake Nipigon are identified as English Bay, Sandy River, Peche au Large, Grande Peche, Roche qui Frappe, and Isle aux Brochel. (The locations of some of these are not clear.) 2.2.5 FORT SAULT STE. MARIE FISHERIES Scores of historical accounts survive describing the native whitefish fisheries at the St. Mary's River rapids (Macdonald 1979). Men of the Hudson Bay Company Post also exploited the rapids in order to obtain fish both for personal consumption and the foreign market. The major part of the fishing was in October: "There are but few fish yet in the Rapid according to report and the few that are caught some petty traders from the American, side immediately trade being encamped on the Island for that purpose." (S.S.M.J. Sept. 21, 1824). "The whitefish have not been plentiful these few nights back in the Rapid. Oct. 24, 1824). I got upwards of 100 today." (S.S.M.J. "La Verdue came here from the Rapid. He had with him but one fish. The fishing is at an end for this season." (S.S.M.J. Nov. 27, 1824). ! - 55 - 3. METHOOOLOGY FOR STOCK IDENTIFICATION 3.1 PERSONAL COMMUNICATION Many of the pioneer families which played a vital role in developing the Lake Superior country have not yet been lost from this land. Second and third generation descendants remain to continue the family name and recall earlier days of settlement and industry. Their towns have modernized, but civilization still hugs the broad sweep of shoreline from Sault Ste. Marie to Thunder Bay. This close proximity to the lake insured that inhabitants knew intimately its ways and resources. During the summers of 1978 to 1980, I interviewed individuals personally involved with the pre-1960 Lake Superior fisheries. Some of these people had been contacted by letter during the spring of 1978. The most obvious sources, of course, were the commercial fishermen. Certain of these people had entered the profession prior to 1920, learning their trade under the tutelage of their fathers and pursuing the lake trout fishery until its collapse in the 1950s. The oldest interviewee was 93 years of age. Many commercial fishermen had retained grounds licensed by their fathers, fishing small blocks of water near shore and close to their home towns. Occasionally a ground might be fished out for a time; then it became necessary to move to more distant sites. Some fishermen employed tugs rather than rowboats or gas boats and could range far afield. The western end of the lake has traditionally been more accessible to the small-scale operator with its greater number of sheltered bay and island areas, nearshore grounds, and less intense prevailing winds during the fishing season. Harsher conditions in eastern waters demanded larger and more capital intensive enterprises. - 56 - I considered it also desirable to seek out participants in the recreational fisheries and charter boat operators. The sports people employed different equipment, different techniques, and frequently fished at different times of the year (see Loftus 1979). Their experiences and impressions of the fish therefore offered many insights not available from commercial fishermen. In total 80 people directly involved with the former lake trout fisheries were interviewed; a number of those most informative were interviewed more than once. In addition, helpful suggestions and valuable information was kindly supplied by employees of a number of government agencies (See Appendix II). A few interviews were recorded on tape. However, it was generally found that simple note-taking was more conducive to relaxed conversation and detailed explanations. Questionnaires were prepared as an interview guide but seldom rigidly adhered to -- an informal atmosphere was encouraged. Given sufficient time, the most valuable route to obtaining oral historical information is not necessarily the most direct. During the interviews emphasis was placed upon: - background history of the lake trout fishery; - lake trout appearances, habitats, and habits, including spawning ground locations and characteristics, movements; - fishing techniques, including equipment, depths fished, frequency of fishing; and - stock transformations. Interviewees were requested to mark on Canadian Hydrographic Service (or National Oceanic and Atmospheric Administration) Great Lakes navigational charts the native lake trout spawning grounds with which they were personally familiar. - 57 - 3.2 THE SPAWNING GROUND MAPS This report is but one in a continuing series of efforts at delineating the natural habitats of Great Lakes' fish species. A pioneer attempt was made by Reighard (1910), who produced whole lake maps, large in scale but lacking in detail. Careful attention to detail, on the other hand, is shown by Smith (1968) in his study of former lake trout spawning grounds in the Canadian waters of Lake Huron and Georgian Bay. More ambitious still is the unpublished work of Organ et al. (1978) which maps spawning grounds of all species in the coastal waters of Michigan state. Horrall (1976), Peck (1978), and Coberly (personal comm. 1979) are also pursuing relevant studies. It is hoped that the volume currently being prepared by Goodyear et al. (1981) will draw together these many threads to form a definitive summary of spawning grounds in the Great Lakes. For the present study, the reference publication of Lawrie (1967), "An Atlas of Canadian Lake Superior Fishing Grounds", has proved invaluable. (It does not, however, deal with spawning grounds specifically). When the name of a particular ground was in doubt this work was considered to be the final authority. Usually names shown on the hydrographic charts were retained, although there are many instances where these derive from the inventive imaginations of the original surveying parties and ignore the historical precedents. The complaints of local residents that these names are neither the first nor the best are being investigated by the Geographical Names Board (Queen's Park, Toronto) -- hopefully changes will be seen in future map publications. Page size sections of the lake charts were photocopied to produce the maps of Fig. SMl to Fig. SM75. Their collective lake locations are outlined in Figures 5, 6, and 7 of Appendix VI. In mapping the spawning grounds for lean lake trout, an attempt was made to assess these as major - 64 - 1. spawning location, 2. spawning time, 3. gross physical features such as flesh colour, skin colour, body size, body form, and weight, 4. spatial-seasonal distribution, and 5. feeding habits. Each form was usually known by a name referring to its most obvious unique feature, such as redfin, black, sand trout, moss trout, and so on. It was during the spawning season that fishermen became most aware of the existence of these types. A particular type might be quite site-specific, spawning only on certain selected grounds. On the other hand, a type could refer to a general spawning run of fish, appearing at many places along the shore or shoals. The problem of detailing intraspecific variants as perceived by fishermen is occasionally complicated by the fact that the same name may be employed by people from different localities to describe groups of fish which bore little resemblance to each other. In addition, that which fishermen presumed to be a unique form constant through time might only have represented the responses of individual fish to local environmental conditions, responses as transient as the conditions themselves. In many instances, however, the criteria listed above can imply some degree of isolation between groups of lake trout, and it is possible to infer from these the former existence of discrete or semidiscrete stocks and subpopulations within Lake Superior. Historical documents reveal that people have recognized stocks for many years, and although not now supported in their judgements of taxonomic status, their observations do have biological significance. Section 4 reviews the evidence for the existence of lake trout stocks (or subpopulations) and discusses the general characteristics - 65 - of the intraspecific groups known to fishermen of the Great Lakes in general, and Lake Superior in particular. Section 5 continues with a detailed regional account of the Canadian waters of Lake Superior. In the absence of quantitative data (from tagging experiments, meristic or morphological studies or electrophoretic analyses), conclusions concerning the presence of former stocks must remain, unfortunately, quite speculative in many instances. Despite the loss of most of these native stocks prior to 1960, the survival of local remnants holds promise for further fruitful study. 4.1 LEANS 4.1.1 INTRODUCTION Early explorers to the Upper Great Lakes marvelled at the great sizes of lake trout they encountered there. Three forms were identified by the Jesuits at Michilimackinac, Lake Huron: "... one, the common kind; the second, larger than three feet in length and one in width; and third, monstrous, for no other expresses it -- being moreover so fat that the Savages, who delight in grease, have difficulty in eating it." (Jesuit Relations 1670 - 71).10 Wintering here in 1680, the Recollet Father Hennepin (1698) describes techniques of fishing through holes cut in the ice with nets set in 46 m. (150 ft.) of water. He coins the term "salmon-trout" for the fish he found here. Trout fishing at Michilimackinac also employed hooks and lines, sunk in some places to depths as great as 90 m. (300 ft. -- Henry 1809). The explorer Charlevoix comments upon the piscatorial situation of Lake Huron in 1721: "There are three sorts of [trout] taken, among which is one of monstrous size and in so great quantities that the Indian with his spear will strike to the number of fifty in the space of three hours." (Charlevoix 1744). As Alexander Mackenzie (1801) paddled westward through Lake Superior in - 66 - 1789, he also found three different forms of charr but offers no description. Gourley (1822), however, supplies some details. One form was the brook trout; the other two lake trout, the largest reaching: "... 80 and 90 pounds weight.. [and] at Michilimackinac also are said to weigh 50 and some over 70 pounds.11 They resemble the salmon of the sea, and may be of the same family, with the difference occasioned by their different water and food. The meat is similar in colour but not so highly flavoured. I think the proper name of this fish is the "lake salmon". The next species is more frequent, but of whiter meat, and smaller in size, weighing from 10 to 20 pounds and taken in greatest abundance in the fall. This fish is like the trout of the brooks, except being larger and without shining red spots on the sides. For the sake of distinction he may with' propriety be named the "salmon-trout". The habit of larger lake trout of moving into shallow water for a short period in spring or early summer is discussed in detail in Section 4.1.2.5. It is natural that this was the time of year when they were most evident to early travellers, for these trout generally spawned late in the fall season -- at a time when few people forced to move about in fragile canoes would be fishing). The large trout which Gourley calls "lake salmon" were also known as Mackinaw trout(or Mackinaw salmon), after the Mackinac Straits where they were first noted. Mitchill (1818) described them as Salmo amethystus, selecting this name because of a purplish tint, resembling that of amethystus quartz, which he found at the base of the teeth. (No fisherman interviewed by myself remembered seeing such dental colouration on Lake Superior lake trout). DeKay (1842), in his Natural History of New York, distinguished the Mackinaw salmon from the "Lake Trout", which he identified as Salmo confinis. He described the colour of the former as dark or dusky gray dorsally, with many light gray spots on the back or sides, and somewhat yellowish ventral, anal and pectoral fins. The flesh was reddish in colour. The latter fish he found to be a smaller "species" of more robust - 67 - shape, with a much blacker back, fins tinged slightly with red, and paler "coarser" flesh. The British naturalist Adams (1873) notes that the Mackinaw salmon (or Namaycush) differed from the Lake Trout (also referred to as the Tague) and siscowet in having a more deeply forked caudal fin, ventral fins placed farther back on the body, and a characteristically extended labial crest. Mackinaw trout were believed to be limited to the Upper Lakes for the most part, although DeKay (1842) notes their presence along the southern shores of Lake Erie. On the average, however, the Lower Great Lakes and inland waters of Ontario have always held smaller lake trout than those native to the Upper Great Lakes: " examples (of Lake Ontario trout) weighing 24 lbs. are sometimes taken, but the average weight is much less than that. The fish caught in seines, on lines, etc. do not average more than 2 pounds, but in the large meshed gill nets, set especially for trout, the average is probably 8 pounds." (United States Commission of Fish and Fisheries 1890). A number of sportsman's accounts published in the 19th century (e.g. Roosevelt 1865; King 1866) identified both the above "species" in Lake Superior. One cannot, of course, be certain of the investigative care with which these writers would study their fish specimens. Insufficient details are supplied concerning habitat and habit to propose that these men were in fact observing discrete subpopulations. Nevertheless, Lake Superior fishermen, past and present, have found visibly distinct forms of lake trout to exist not only in different parts of the lake but also to appear concurrently or consecutively upon the same fishing grounds. As examples, historical references to this great array of lake trout variants are reproduced in Appendix III: one pertaining to the Upper Great Lakes in general (Goode 1884), and one specifically to the St. Ignace Island area of Canadian Lake Superior (Thomson 1883, and elaborated by Kerr Jan. 3, 1885). The bewilderment of early taxonomists is quite understandable. - 68 - 4.1.2 CRITERIA FOR STOCK IDENTIFICATION 4.1.2.1 SPAWNING LOCATION AND TIME Throughout the early years of Canadian fisheries management, much heated debate centred on whether closed season regulations were representative of the true spawning periods of the fish species they were designed to protect. It was obvious that the gravidity of a particular species could vary widely under the influence of genetic makeup, age, and local conditions such as food supply, current, and temperature. Many decried the blanket application of province-wide close seasons, advocating instead zoned fishing regulations. Fielding (1915), for instance, wrote: "Let me give as examples only two instances of these varying periods in the province of Ontario. On the west shore of the Saugeen or Bruce Peninsula the lake trout (Cristivomer) come on the "redds" a fortnight earlier than it does on the southeast shores of Georgian Bay. Again, in lake Nipigon, the speckled trout becomes gravid nearly a month later than it does in the river flowing out of it." In bodies of water as vast as the Great Lakes, it was clear that one could not even consider the whole lake as a single spawning unit. Along such great stretches of shoreline, fishermen have long recognized serial trends in the annual lake trout spawning periods. For example, information collected by Smith (1968) during interviews with local Canadian fishermen, indicates that spawning in Lake Huron commenced latest in the vicinity of the Straits of Mackinac (Nov. 1 - 15) and became progressively earlier as one moved along the shore of Manitoulin Island, southward toward Kettle Point (Oct. 20 - Nov. 10). In Georgian Bay, with the exception of the Collingwood area, a general counter-clockwise trend, from latest to earliest, was apparent. It will be seen below that comparable large-scale trends were also known within Lake Superior waters (see Sections 5.3 and 5.9). These may, in part, be related to differential cooling of near surface waters in the lake. Falling water temperatures can trigger - 69 - spawning activity. Evidence collected by Van Oosten (1927) along the American shores of Lake Huron fails to reveal obvious north-south trends in the spawning periods. Local conditions, affecting only local stocks, may obscure such general trends. Different spawning groups of lake trout may be observed in Lake Superior, and these well show temporal, as well as habitat segregation according to whether they spawn in: 1. river, shore, or bank areas, 2. windward or leeward zones, 3. bay or inshore areas, and 4. deep or shallow waters. River-running subpopulations, for example, often spawn somewhat earlier than those along the lakeshore. Stocks frequenting the warmer waters of a shallow bay may ripen more slowly, reproductively isolating them from those spending most of the year in the waters of the main lake. In addition, historical documents dealing with the Upper Great Lake fisheries often noted two major runs of lake trout -- one spawning inshore, and another spawning predominantly on the offshore reefs and shoals. In Lake Huron, for example, a "shallow-water trout" or "shoal trout" came ashore near Alpena, Port Austin, and Southhampton between September 15th and 25th and remained there for approximately 2 weeks (On certain grounds these might be taken in 1 m. (3 ft.) of water). A later, heavier run of "deepwater trout" (apparently also a variety of lean) did not approach the shore and was most common upon the reefs between October 10th and the latter part of November -- the season being later in the northern part of the lake than at Alpena and Southampton (United States House of Representatives 1897). In the same document, however, it is stated than Georgian Bay fishermen knew their offshore trout to spawn earlier than the larger inshore trout. In Lake Superior the first run at Marquette, Michigan, on - 70 - the other hand, was the "shoal trout", a smaller fish (Cook 1929). On the basis of the interviews conducted along the Canadian shore, it is evident that large native trout could be taken at spawning time on either the offshore banks or shore shoals depending upon location in the lake. A majority of fishermen reported inshore spawning to commence prior to offshore spawning, but this too could vary with location. The spawning scenario in Lake Superior was considerably complicated by the fact that more than one seemingly distinct spawning group (or run) could appear on the same grounds. This phenomenon seems to have been more common in Lake Superior than in any other of the Great Lakes. In contrast, a different situation prevailed along the Wisconsin shore of Lake Michigan, where fishermen found no more than one run of lake trout utilizing a single ground (Ms. C. Coberly, personal comm. 1979). This multiple usage of Lake Superior breeding grounds resulted in a rather extended spawning season in certain parts of the lake. The duration of the spawning period for each group could vary. Successive runs showed definite peak concentrations of individuals, and these might be separated in time by as much as 2 weeks or more. If spawning is a response to declining lake temperatures, intraspecific variation is evident in the temperatures preferred by each spawning group. Despite intermingling of individuals at the end of one run and the beginning of the next these differential preferences seem to have remained fairly consistent through many years. In addition, the presence of breeding grounds in close proximity (i.e within the normal range of lake trout movements; see Section 4.1.3.6), characterized by different spawning dates, suggests the sympatric association of different groups of fish varying in their responses to spawning stimuli. According to Fabricius (1950), "the spawning activities in fish are released by an inherited neural releasing mechanism, in which various internal and external stimuli cooperate by - 71 - heterogeneous summation". The individuals of a particular spawning run might vary not only in their spawning periods but also in their spawning depths, as well as in certain gross physical features (discussed in detail below). 4.1.2.2 BODY WEIGHT AND FORM As a general rule, seasonally later runs were composed of larger and heavier fish. Certain grounds were famous for the large sizes of their spawning trout. Although the weight ranges characteristic of successive runs could show much overlap as these arrived on the same grounds, consistencies were found in the modal weights reported by fishermen for each run. On the basis of the interviews, it seems that there was not merely an intergradation of sizes arriving on the grounds, but actual pulses of differentsize classes. Several factors, inherent to the trout populations, might account for these inter-run variations: 1. Differences in age composition; i.e. different age classes may show temporal separation at spawning time, with younger fish spawning at earlier dates. 2. Differences in growth rates; i.e. earlier runs may include slower-growing individuals. 3. Differences in "age at maturity"; i.e. precocious maturity may characterize earlier runs which have a greater percentage of younger individuals. 4. Sex difference, which might give the appearance of different breeding runs. Such confusion was not evident in the interviews. In addition, most evidence indicates no sexual divergence in the growth rate of lake trout (Martin and 0lver 1980). It is obvious that these factors are, to a large extent, interrelated. Alm - 72 - (1959) presents an experimental basis for stating: "In a population with good growth rate maturity appears at a lower age and usually also at a smaller size than in a population with poor growth rate, where maturity is reached only at a higher age and in most cases at a bigger size. In populations with very poor growth rate and high age for maturity the average length at maturity can again be lower, and approach that of the earliest mature and fastgrowing populations." Furthermore, his experiments with various species at Kälarne Fishery Research Station in Sweden revealed that "age at maturity": "... can be genetically determined in different species and also in different forms of the same species. It is then usually higher in large-sized and often fast-growing forms than in small-sized and often slow-growing ones. This is exemplified by big and small forms of trout, and perhaps also by similar forms of char and. whitefish." Genetic bases for differential growth and maturation rates are always difficult to infer, and many authors stress the importance of environmental influences such as water temperature, diet, and fishing pressures. The densest stock, for example, will tend to have the smallest fish as a result of competition for food (see review in Ricker 1959) -- extreme examples of this phenomenon are dwarf forms developed by certain species (see Nilsson 1955). In some cases, a fishing-up phenomenon might occur, whereby large individuals are removed from the more heavily exploited stocks. Nevertheless, the fact that specific size classes appeared annually on certain Lake Superior grounds, and at relatively the same time each year suggests a strong stability in composition at these spawning locations. It is possible that this stability is linked to local racial differences. It is useful to cite related studies. Partly on the basis of variations in growth rates, Tomkins (1951) postulates the existence of two distinct lake trout populations within Georgian Bay. Employing length criteria, Cope (1953, 1957) outlines various races of cutthroat trout (Salmo clarki) within Yellowstone Lake. Analysis of scale features can also prove useful in relating size to age composition. Scale studies have been - 73 - utilized in the identification of salmon stocks, for example (Major et al. 1970; Henry 1961), but unfortunately no efforts have been made at applying these techniques to lake trout at the level of their individual spawning runs. Differences in body form between the lake trout of different grounds or different runs on the same grounds have also been reported by Lake Superior fishermen. Variations in body form between segments of a population may suggest different rates of growth. Martin (1949) has demonstrated that fish pass through a number of relative growth stanzas during their lifetimes. Logarithmic plots of body part length versus total body length show that the transition from one stanza to the next is marked by a sharp inflection of the slope. Body size at inflection is the most important factor in creating intraspecific variation in the relative sizes of body parts (not changes in the slopes of the growth lines). According to Martin (1949), stanza changes occur during the eyed egg stage, hatching, ossification, and the onset of sexual maturity -- early stages are the most significant in determining body form. To the extent that growth rate and body form are under the influence of environmental factors, intraspecific variations may reflect different habitat conditions at the critical stages of development. Inter-lake comparisons of fish populations reveal effects of temperature: northern, colder water races have been found to often possess smaller heads, eyes, maxillaries, and fins than faster-growing, southern races (Martin 1939). It is possible that physical differences such as these may find expression within a single lake between stocks frequenting different depth zones (see also Section 4.2). 4.1.2.3 FLESH COLOURATIOH Questions concerning the factors controlling Salmoninae flesh shades - 80 - (Mr. N. V. Martin, personal comm. 1979). Bean (1902) notes that Alaskan lake trout specimens are usually very dark. Sage et al. (1904) state that black skin colouration was common among the confined trout of fish hatcheries. In Lake Nipigon, Dymond (1926) found a very dark-skinned variety, known locally as "black trout", to exist there. These were of medium size, seldom exceeding 1.8 kg. (4 lb.) or 2.3 kg. (5 lb.), and ascended rivers to spawn. As none were to be found in summer, Dymond suggests that the colouration may have been stimulated by the darkness of the river waters. As noted by Peterson et al. (1966): "It is well known that the bright colours of trout disappear rather rapidly after exposure to light." During interviews many Lake Superior fishermen remarked that dark skin colours would lighten after the fish had been for some time out of the water. It was; on the other hand, the experience of a couple of interviewees that black trout left in a fish box would only fade at spots where their bodies had been touching, while several fishermen maintained that blacks out of water usually did not lose their colour. In general, however, dark colouration probably represented a natural spawning period response of certain groups of Lake Superior lake trout and was used to identify them at this time. At certain locations black trout would appear in deep-set nets several weeks before the actual spawning period, but their presence was generally considered rare in spring and summer (although a few fishermen noted that they often captured very dark specimens indeed when trolling in certain areas: the Port Coldwell area (Fig. SM33) and near Montreal Island (Fig. SM9) were specifically mentioned). Precedents exist for the use of skin colouration as a criterion for salmonid stock identification. Bulkley (1963) found significant differences in spotting counts and body colouration between six spawning runs of cutthroat - 81 - trout in Yellowstone Lake. Nilsson and Filipsson (1971) noted significant differences in both skin and flesh colouration between two discrete forms of Arctic char (Salvelinus alpinus) in a lake of the Swedish Lapland, despite similar preferences for crustacean species and identical points of capture. Genetic determinants are proposed by Ricker (1959) to explain differences in colour between three stocks of Kootenay Lake (British Columbia) kokanee (Onchorhynchus nerka), as well as between introduced kokanee and residual progeny of anadromous sockeye salmon in Cultus Lake. 4.1.2.5 NON-SPAWNING MOVEMENTS The regional studies of Section 5 concentrate upon the behavior and appearance of lean lake trout at their spawning times. It was at these times that fishermen were most aware of their movements and habits. Visually distinct group of lake trout were less obvious at other seasons. In this report some attention is also paid to the non-spawning movements of Lake Superior lake trout. Both vertical and horizontal separation of larger (and older) trout from smaller trout was frequently noted. Trout caught in bottom-set nets in early spring were usually of smaller size. As the weeks progressed many trout moved into shallower zones along the off-shore shoals or shore banks. Shoals in the same general area might be characterized by fish of different size classes differences sometimes not attributable simply to the effects of differential fishing pressure. Still later, in June or July, large lake trout tended to become pelagic, at which time many fishermen knew them to feed on insects at the surface (see also Lawrie and Rahrer 1973). Some movements have also been associated with the pursuit of lake herring stocks. At certain sites nearshore, movements of large lake trout were known and might be referred to as a summer "run". Some years these fish would be available for several weeks. In the shallow waters, gill nets were fished "corks out", set right at the surface. In more open waters, far greater numbers of large trout were taken on troll lines while swimming above the - 82 - bottom-set nets. In the 1950s some fishermen began the practice of "canning" (or floating) their nets at the surface during the early summer. The technique, although successful in catching more fish, was not widely employed on the Canadian side of Lake Superior. Preparation of the nets was time-consuming and there was a greater chance of net damage. At scattered locations along the Canadian shore, large lake trout again moved into shallower waters in the second part of August. Usually the fish of this movement were not as abundant as the first. These fish were not yet preparing to spawn and would shortly depart again for greater depths. It was not unusual for this run to fail entirely, perhaps due to atypical winds or temperatures. 4.1.3 NATIVE LEAN LAKE TROUT VARIETIES DESCRIBED BY FISHERMEN 4.1.3.1 BLACKS The smallest mature lake trout tended to spawn inshore during the earliest part of the fall season in September or early October. On certain grounds, these were typically gray or silver-gray in colour with lighter gray spots and light ventral surface and were described by various fishermen as "ordinary trout" or "little grays".However,at many locations both in the eastern and the western waters of the lake, the earliest run trout were very dark. Their sides were slightly lighter than their backs, which were noted as having been brown-green, dark brown, or black in colour. The ventral region was darker than that of other lake trout, and the variety was generally known as "black trout". Average weights cited for trout in this run ranged from 1 kg. (2.2 lb.) to 2.7 kg. (6 lb.). At this size, the body was found to be somewhat robust and stocky -- the form may relate to that which was identified as Salmo confinis in the 19th century (and perhaps also the Truite de Greve of Agassiz 1850). Many fishermen found the flesh colour of the blacks to - 83 - have been reddish, but certain areas were visited by those with pale flesh tones (see Sections 5.6 and 5.7). In addition to this distinctive inshore run, other trout of larger size might also be quite dark in colour. The Caribou Island trout for example, were undoubtedly a discrete subpopulation almost exclusively black and somewhat heavier than those described above (see Section 5.5; Fig. SM30). In the island environs south of Nipigon Bay and the Black Bay Peninsula, fishermen sought not only an early run of small blacks but also the spawning "big blacks", which appeared later in the season. These were not, however, regarded as a distinct spawning run. (Thomson (1883) refers to these as Rock trout, a name also employed by some Lake Huron fishermen when referring to the native blacks of that lake; see Appendix III).River-spawners composed the most obvious stocks of dark coloured trout. No physical variations from lake-spawning blacks were reported by fishermen. Native spawning stocks utilized many of the Canadian rivers of Lake Superior. These stocks were most prevalent south of the Pic River on the eastern shore, while none were known to have existed west of Nipigon Bay or along the American shore. Although certain streams entertained a wide range of fish sizes, there is no evidence to suggest that any one stream played host to more than one stock. Certain streams might contain a number of separate spawning sites -- extensive movements of lake trout could occur within the Michipicoten River, for example (Section 5.3). It is not known, however, if groups of fish would selectively home to specific grounds within a single stream. Authors in the past have pointed out that, with the exception of - 84 - Lake Superior, stream-spawning habits were rare to lake trout (although Martin and Olver (1980) suggest a greater prevalence among Arctic populations). Appendix IV, however, cites reported instances of lake trout entering rivers other than those of Lake Superior. It is clear that this phenomena while atypical of the species, should not be regarded as having been rare. It is undoubtedly characteristic of the populations of certain inland lakes and was probably, at one time, more common in the Great Lakes than in recent decades. Stream populations, of course, are more susceptible than those of the lake to natural hazards such as water fluctuations or inhospitable weather conditions, as well as to the effects of exploitive fisheries. Many former populations have been lost. During the Lake Superior spawning season, lake trout tended to enter the rivers after dusk and retreat to the lake before dawn. The peak of the daily run was usually between 11 PM and 1 AM. Some trout would not leave the river however (especially when the spawning sites were a long distance upstream), and would rest by day in spots of quieter flow (see also Loftus 1958). 4.1.3.2 REDFINS AND YELLOWFINS In general, any lake trout possessing brightly coloured fins was liable to be called "redfin". It was possible for dark trout to display red-coloured fins, a condition especially obvious at the Slate Islands (Fig. SM38; Section 5.6) and among certain river-spawning stocks. The name, however was also employed by many Lake Superior fishermen to describe a specific run of lake trout which they understood to follow closely upon that of the earliest spawners (i.e. the small black trout or "regular" grayish trout) but to represent a different "breed". It is - 85 - possible that fin colouration could be triggered in certain fish by falling water temperatures, thus creating the illusion of a new run. Blacks and redfins might spawn on the same grounds. At certain locations, the black trout would retreat from these grounds before redfins had even appeared. Although this was not always the case, generally the two forms were not found spawning at the same time. Redfins were the heavier variety, averaging 3.2 kg. (7 lb.) to 3.6 kg. (8 lb.), and were frequently over 6.8 kg. (15 lb.). Some fishermen described these fish as lighter in colour and more elongated (proportionately less deep of body) than the blacks. Flesh colour was pink, orange, or red, whereas black trout might show pale shades. Redfin grounds were situated at many locations on the Canadian side of the lake. In addition, one finds this classification employed by fishermen of Isle Royale (Rakestraw 1968), the Keweenaw Peninsula (Organ et al. 1978), and the Green Bay area of Lake Michigan (C. Coberly, personal comm. 1979). Fishermen of Georgian Bay also identified redfins as a second run of lake trout (Kerr Nov. 19, 1886; Loftus 1980). Former fishermen of Jackfish and Port Coldwell areas tended to refer to the fish of their second run as yellowfins rather than redfins. (a couple of local residents pointed out that, in many instances, yellowfins should more properly be called "orangefins" (their fins being an oranqered colour). Fishermen of the Black Bay Peninsula - Thunder Bay area also tended not to distinguish redfins and yellowfins as different spawning classes. Certain grounds, however, particularly those along windward island shores, were frequented predominantly by lake trout with yellow fins (Section 5.9). In addition, certain fishermen of the Rossport area believed yellowfins to represent a unique run at a few locations - 86 - (discussed 1n Section 5.7). It has been suggested that lake trout which were larger and spawned at later dates were more likely to exhibit yellow colouration. In Lake Superior it was rare to capture trout with coloured fins anytime except during the late summer or fall (with the exception of those at Superior Shoal; Fig. SM42; see Section 5.8), suggesting that fin colouration was a phenomenon initiated by the approach of the spawning season. (This response was also observed by Royce 1943.) Exceptions were encountered, however, and a couple of fishermen believed fins to be coloured all year but darker when spawning time drew nigh. This may also be typical of certain populations of small inland lakes. A stock of lake trout native to Killala Lake (at the headwaters of the Little Pic River; Fig. SM33), for example, is extremely dark in colour, of average weight 1.1 kg. (2.5 lb., and is known to retain its red-coloured fins throughout the year (Mr. R. Ryder, personal comm. 1978). It is probable that large fish such as yellowfins, or those of the late runs discussed below, corresponded to the Mackinaw trout (Salmo amethystus) described in the 19th century (see Section 4.1.1). 4.1.3.3 THE LATE RUNS Passing along the northern shores of the lake toward Thunder Bay, fishermen identified very late runs of large trout at certain locations. Residents of Port Coldwell and Jackfish referred to the fish of the third spawning runs, in general, as "salmon-trout". These fish began spawning in late October and continued through November: "... The salmon-trout were streamlined and had a small head like a West Coast salmon. The tail was not too big, and all possessed red flesh. They did not have coloured fins, and they had a silvery body, not as dark as the blacks and yellowfins. Salmon-trout were taken in a little deeper water at 5 or 6 fathoms." (See Section - 87 - 5.6.; Mr. A. Almos, personal comm. 1979). Fishermen whose grounds were situated along the Black Bay Peninsula knew large red fleshed trout as "red trout" (see Section 5.9). However, those men who plied the waters among the islands south of Nipigon Bay applied the name red trout to a separate spawning run. Here red trout were easily distinguished from the black trout which they followed and differed mainly in size from later-running trout. This form is discussed in more detail in Section 5.7. Kerr (Jan. 3, 1885) records the existence of red trout in the Upper Lakes, noting it to be "the finest of all, best for procuring eggs from." Some fishermen may also refer to groups of late-spawners as "gray trout". Thomson (1883) notes the large gray, or "shovel-nose trout", a variety existing about St. Ignace Island (Fig. SM54) and weighing up to 32 kg. (70 lb.; Appendix III). In this general region, gray trout were understood to be the last run of the season, having uncoloured fins, a body short but deep in appearance, and flesh white or yellow-white in colour (Section 5.7). Approaching the shore banks of these islands, grays showed a predilection to spawn at relatively greater depths than most other varieties of lean trout; depths quoted varied from 9 m. (30 ft.) to 20 m. (65 ft.). Here (as well as in certain areas farther west) large grays (and less frequently yellowfins) were found spawning over shore beds of filamentous green algae, designated "moss" by local fishermen (and marked as such on the appropriate spawning ground maps in Appendix V). In the Rock, Wright's, and Washington Harbour areas of Isle Royale (Fig. 7), "channel salmon" lake trout selected similar substrates and so earned the additional nickname "moss trout". Mr. S. Sivertson (personal comm. 1979) explains that moss trout eggs were unusually small, even though these fish were a big "breed", ranging from 6.8 kg. (15 lb.) to 16 kg. (35 lb.). Organ et. al. (1978) note the presence of moss trout more often in the - 88 - Michigan waters of Lake Michigan than Lake Superior. In a biological survey of Lake George, New York, Needham et al. (1922) identified the "moss" described by local fishermen as Dichotomosiphon tuberosus, finding "its best development at depths of forty or fifty feet, or just below the nitella zone". 4.1.3.4 SAND TROUT In the southeastern portion of the lake, fishermen recognized not only two major runs of fall spawners but also another type they knew as "sand trout". These fish were considered distinct, having very red meat and light silvery colouration, with darker blotches (not spots) on the back. Also called "silver trout", individuals were described as slender and small, a 1.8 kg. (4 lb.) specimen being considered a large one. In Lake Superior they were usually taken in summer. Some fishermen described sand trout as a shoal fish, mostly taken at depths of approximately 55 m. (180 ft.) over sandy bottom. Oth2r people knew this form to move to inshore flats in June and July. It is possible that the Lake Superior sand trout corresponded to the form known to Lake Huron fishermen as "summer trout", which Kennedy (1941) describes as "... lighter in colour than the Winter Trout (fall spawners). Fishermen say they have taken them with spawn running in the summer. These fish "disappear" in the autumn or winter. According to one report, summer trout were immature fish (Canada. Dept. of Marine and Fisheries. 1904). The sand trout may have similarly represented young age classes. In the Apostle Island region of Lake Superior, Dryer (1966) found the greatest abundance of small lake trout between 49 m.(160 ft.) and 53 m. (174 ft.) in summer, and 55 m. (180 ft.) and 71 m. (234 ft.) in spring. On the other hand, sand trout may represent a seasonal phenotypic variation in certain groups of smaller earlyspawning black or "regular" trout. Few spawning grounds for sand trout were discovered, although several fishermen knew trout of this description to spawn at a few sites along the eastern shore. It was remarked by Mr. G. Jones (personal comm. 1978) that the last time he saw sand trout was in the late 1960s in summer on the Pukaskwa Flats (see Section 5.4; Fig. SM17). Here, according to - 89 - Loftus (1952) sand trout would spawn between October 20th and November 30th, a late run for that area. 4.1.3.5 DEEPWATER LEANS On a couple of the deep offshore shoals in the Caribou Island to Michipicoten Island area (Section 5.5), as well as at Superior Shoal (Fig. SM42), certain lake trout were identified as "deepwater leans". These appeared rounder in shape than the "ordinary leans" and possessed a thin "filament" of fat along the abdominal wall, a consequence perhaps of feeding habits (although the fat was never as thick as that found in siscowets and halfbreeds). Flesh colours were pinkish and never as red as that shown by certain shallower leans in the same areas, also suggesting partia1 differences in diet (see Section 4.1.2.3). Caught at a depth of about 55 m. (180 ft.), deepwater leans may have spawned here in the fall. Feeding may have occurred at even greater depths. Lake Nipigon fishermen likewise identified a form of deepwater lake trout, unknown above depths of 37 m.(120 ft.), 'and spawning at depths of 37 m. (120 ft.) to 55 m. (180 ft.; Dymond 1926). These were fatter, with small heads and white flesh and were seldom, if ever, found in the same nets with the other lake trout variants (Mr. K. McLeod, pers. comm. 1979). It is uncertain to what degree the deepwater leans of a particular shoal in Lake Superior were isolated from other leans above and the fat trout below, but reproductive isolation from the shallow-spawning stocks seems indicated 4.1.3.6 RACER TROUT Tagging returns have shown that occasionally individual lake trout may travel great distances. In a study of Apostle Island trout, for example, Eschmeyer et. al. (1953) found that most remained within 80 km. (50 mi.) of - 96 - other waters by the American Fish Commissioners, including Mystic Pond on Cape Cod (Massachusetts Commission on Inland Fisheries 1872, 1884). Siscowet were rare in the southwestern part of Lake Superior, only beginning to become more plentiful in the American waters about Whitefish Point (Fig. 5). Historical references mention particularly high abundances about Isle Royale and the Keeweenaw Peninsula. However, according to Sweeny (1880), Chairman of the Minnesota Fish Commission, siscowet were nowhere as abundant as lean trout: one barrel of siscowet to 15 of "Namaycush" was considered a plentiful catch. Relative abundances in Canadian waters in this early period are not known. Siscowet would outnumber lean trout in the spring catches at the Pic Post (Section 2.2.1.1). Agassiz (1850) found siscowet to occur "everywhere along the northern shores", although this is undoubtedly exaggerated. His specimens were obtained at Michipicoten (Bay?), but he notes also the plenitude about Isle Royale. (A century later Hubbs and Lagler (1949) cite accounts of catches at Good Little Boat Harbour, Chippewa Harbour, Amygdaloid Channel, and elsewhere along the north shore of the island -- Fig.7). Gill netting for fat trout was generally conducted at depths ranging from 73 m. (240 ft.) to over 183 m. (600 ft. -- which caused a high attrition rate of net floats prior to the days of reinforced aluminum. Even the early cedar floats were flattened to the form of a shingle). At certain locations in Lake Superior, siscowet would move into near-surface waters for a short period in June or July. Milner (1874) reports, and Eschmeyer (1955) supports, the apparent absence of any migratory instinct at spawning time. Interviews with Lake Superior fishermen suggest, however, that at least some stocks would move to shallower depths on a bank to spawn, although definite annual patterns are difficult to discern. - 97 - In light of the apparent variability of depth distribution, one imagines a situation in which fat trout are distributed unevenly along the sides of the deep offshore shoals, but relatively free to move between different depths, probably in opportunistic pursuit of food. Siscowet have been found in spawning condition during most months of the fishing season (Eschmeyer 1955), and it is possible that they do not spawn every year. Sweeny (1890) recounts that: "In the neighbourhood of Isle Royale ... fish were found in various stages of development; in some the eggs are firm and hard and undeveloped, in others fully developed, soft, and ripe, ready for spawning. The males are also in the same stages of readiness; from some the milt flows freely and in great abundance; others seem spent, while others, again, seem immature and the milt undeveloped." Occasionally Canadian fishermen lifted siscowet with spawn running in nets set over bottoms of clay. Perhaps fat trout are less selective than leans in their spawning sites. When fat trout were brought up from great depths their eggs would burst; consequently they were not preferred by spawntakers (Mr. W. Sanders, personal comm. 1978). It is thought that halfbreeds are immature fat trout (Khan and Qadri 1970), although Mr. L. Morden (pers. comm. 1979) found the largest trout he identified as halfbreeds to have been 3.4 kg. (7.5 lb.). Also some have been reported in spawning condition (although these may have been fat trout). Most halfbreeds were 1.4 kg. (3 lb.) and under (Mr. I. Purvis, personal comm. 1979). Halfbreeds were found at shallower depths than fats, from 36 m. (120 ft.) to 128 m. (420 ft.), and there were certain popular grounds where siscowets were netted in greatest abundance. Halfbreeds were usually preferred over siscowets for smoking. Electrophoretic analysis of the serum enzyme lactic dehydrogenase (LDH) suggested three strains of lake trout: lean, fat, and paperbelly. Patterns for fish identified as halfbreeds by fishermen showed the same - 98 - patterns as either paperbellies or, more frequently, fats according to studies conducted by Laarman and Ulrickson (1970). Paperbellies, known to American fishermen as humpers, have been described by Eschmeyer and Phillips (1965), Rahrer (1965), and Lawrie and Rahrer (1973). This variant is unique in having a very white belly, sometimes distended, with abnormally thin skin. Paperbellies are a lighter gray colour than the deepwater fats. It has been suggested by Lawrie and Rahrer (1973), that paperbellies derive from wandering lean lake trout which colonized the offshore banks and underwent subsequent polymorphic development. Opinions differ widely on the fat content of paperbellies. Some fishermen found them oily and of inferior quality to even the fat trout. On the other hand, it was often possible to include paperbellies with lean trout in market shipments. As discovered by Eschmeyer and Phillips (1965), the rate of increase of paperbelly fat content with length was nearly as rapid as that of siscowets and more rapid than that of lean lake trout. Occasionally fishermen will use the term "banker" to variously describe small siscowets, halfbreeds, or paperbellies. It was not always evident from the interviews which variant was intended. In these instances the text simply refers to bankers 5. A REGIONAL DISCUSSION OF NATIVE LAKE SUPERIOR LAKE TROUT STOCKS For discussion purposes, Lake Superior has been divided into ten regions. Criteria of geography, commercial fishing patterns (see Section 1.2), and/or stock discreteness determined their selection. The summary of Appendix V arranges grounds from east to west by region. - 99 - 5.1 SAULT STE. MARIE TO COPPERMINE POINT In the southeastern waters, commercial lake trout fishing traditionally located north of the twin promontories Agassiz (1850) called the "Pillars of Hercules of Lake Superior" -- the Canadian Gros Cap and American Iroquois Point (Fig. SMl). Beyond this gateway the land opens to form Whitefish Bay, narrowing again at the larger northern gateway, flanked by Whitefish and Coppermine Points (Fig. SM4). This marks the entrance to the lake proper. Southernmost Whitefish Bay (Fig. SMl) possessed excellent herring and whitefish grounds, but lake trout spawning grounds were scarce. There are indications, however, that trout movements into these waters (even to the St. Mary's River) were more common prior to 50 years ago. On the Michigan state side, Organ et al. (1978) indicate a substantial spawning ground for "native" trout which was centred between Point Iroquois and Nadoway Point and extended westward to the Canadian Gros Cap Reefs (through water as deep as 17 m. (55 ft.); Fig. SMl). They describe a single spawning run commencing about November 1st. Fishermen familiar with the Canadian inshore grounds found the first run to begin earlier in the season. The bottom between Gros Cap and Kelly Creek (Fig. SMl and Fig. SM2) is rocky and gravelly, and both "regular" and black trout would usually move close to shore to spawn during the first week of October. (It is known that the late Mr. Alec Daigle of Gros Cap witnessed each year trout cleaning debris from the rocks about North Gros Cap; 1947 was the last year this phenomenon was seen; Mr. M. Purvis, personal comm. 1978). Roosevelt (1865) believed Goulais Bay (Fig. SM2) to have been one of the "favourite haunts" of the Mackinaw Salmon: - 100 - "Late in the fall they retire to the sombre depths, and are only taken by still fishing with a long line and live bait, and at such times the deep water abreast of Gros Cap is one of their favourite localities, and they are there frequently caught by the Indians from fifty to seventy-five pounds." (Specimens of these weights were relatively rare after 1918). In 1766 Indians informed Alexander Henry (1809) of excellent autumn fishing at "Oak-bay" (Goulais Bay), which was known to the French as "Anse a la Peche" or "Fishing Cove". It is not known if lake trout or whitefish were being caught here, but both have been praised for their abundances in this area. Fishermen still living, however, remember grounds centred about Goulais Point rather than inside the bay itself where the bottom is predominantly sand and mud and therefore discouraging to spawning activity. The shoals about Maple Island (Fig. SM3) were considered to have been indisputably one of the finest trouting grounds in this part of the lake. The fishermen of the .Goulais Bay Indian Reserve always made this a popular spot in the fall when both black trout and redfins would congregate on the rubble bottom between the island and the shore. Between October 5th and 10th black trout began arriving on the ground, the run lasting a week to 10 days. Redfins appeared shortly before the end of this run and usually departed by the end of October. These dates correspond to those reported by American fishermen for southern Whitefish Bay (Iroquois Island, Salt Point, and Tahqueamenon Island) as recorded by Van Oosten (1927). Much of the shoreline of Batchawana Bay, including Harmony and Havilland Bays, drops off too rapidly to serve as proper grounds, but Fig. SM3 indicates important spawning areas in the southern portion of the bay and about Batchawana Island. Provincial Land Surveyor Wilson (1865) heard Indian reports of grounds off the east and northeast shores of the island. In his notes we discover one possible reason for their excellence: - 101 - "A strong current from the west sweeps around to the north and northeast of the island. the water is of considerable depth, and the current so strong that a considerable portion of it is open during the winter." An area of suitable bottom is made more advantageous as a spawning ground if washed by currents. Thus will it be kept clear of debris and constantly replenished by fresh water. Lake trout seem especially attracted to exposed points or open shorelines and a moderate wind will bring many trout inshore to spawn (see Martin and Olver 1980). It has been the experience of some Lake Superior fishermen that spawning is heaviest on the windward side of islands. This is not, however, always borne out in reality, and excessively high winds can drive trout away from the shallow grounds and back to the depths. The importance of southern Batchawana Bay as a spawning area prompted A.G. Duncan, Inspector of Fisheries, to recommend in 1911 that the waters "east of a line drawn from Rudderhead Point along the west side of the Batchewana Island to Sand Point" be set aside as a preserve (Canada. Dept. Marine and Fisheries. 1911). It is possible that his recommendation arose from an impression that the bay was being overfished. The eastern half of Batchawana Bay reaches a maximum depth of 45 m. (149 ft.). This area of deep water is effectively separated from the main lake by Batchawana Island and its shallow southern shoals. Trout, spending most of the year in these deep inner waters, would move onto the shallows at spawning time. Along the western side of Havilland Bay and northward toward Batchawana Is., they would spawn in 1 m. (3 ft.) of water (Mr. G.A. Jones, personal comm. 1978). The bay trout were a smaller black, most weighing 0.9 kg. (2 lb.) or 1. 4 kg. (3 lb.), with 4.5 kg. (10 lb.) representing a large one. It is probable that they formed a stock discrete from those of the lake. Following completion of bay spawning, fishermen would move outside of Batchawana Island where lake trout ran to large sizes and were characterized by a somewhat later - 102 - spawning period (commencing about October 15th). Recent experimental netting by the Ontario Ministry of Natural Resources suggests that planted lake trout may be successfully recolonizing the deeper waters of Batchawana Bay (Wohlgemuth 1974). It is interesting to note that the whitefish of Batchawana Bay may also have been, at one time, a discrete stock. In the Report of the Ontario Game and Fish Commission (1892) reference is made to an assertion of local fishermen that there existed whitefish peculiar to the bay. Barnston (1874), once factor of the Hudson Bay Company Post at Michipicoten, wrote: " in spring we sometimes had sent to us from a small outpost at Bachewaino Bay a fish or two, longer than our own and much thicker and heavier.... I never had an opportunity of submitting these white-fish to a close comparison with the large specimens taken at the Sault Ste. Marie, below the rapids, but I conjecture they might be of the same species." Within Pancake Bay (Fig. SM4) there was little spawning activity, possibly due to a sandy condition of the bottom. The variety known locally as sand trout (see Section 4.1.3.4) resorted to the flats in this area (and it was trout of this same appearance and size class which moved into the bay in early summer to feed). North from Pancake Point to Coppermine Point both black and "regular" trout ran close to shore. Fishermen described the flesh of the black trout as predominantly reddish in colour. Generally, prior to 1945, one found the average weight of the trout in the first run to be 1.1 kg. (2.5 lb.) to 2.7 kg. (6 lb.), with an average weight of almost 1.8 kg. (4 lb.). Pancake Shoal (Fig. SM4) was once a popular fishing ground and served as a source of lake trout spawn for the Sault Ste. Marie Hatchery (Parker 1931). Trout probably spawned all around the Sandy Islands (Fig. SM4), but a couple of fishermen remember hauling better catches along the shores of South Sandy Island. On Parisienne Shoal (Fig. SM4), it was possible for a single boat to take many tons during the few weeks that the - 103 - blacks and redfins were present there. It is the opinion of some that the shoal runs occurred somewhat later than the shore runs in this region. The Indian fishery of Goulais Bay, for example, was shifted to the shoals as the season progressed, Parisienne Shoal being the latest site of spawning (Mr. G. Agawa, personal comm. 1978). Blacks typically began to cast their spawn here between October 15th to 20th. After one week, redfins arrived according to their time. The shores of Parisienne Island (Fig. SM2) were noted for whitefish breeding, and in the 1890s fears of depletion caused these grounds to be set aside as a preserve for a number of years (Canada. Dept. of Marine and Fisheries. 1893). But the sandy bottom that made these grounds so attractive to whitefish discouraged spawning of the lake trout, and only a few spots proved acceptable. Redfins were known to occasionally spawn at the southern end and on the hump just north of the island. East of Parisienne Island the deep bottom of Whitefish Bay rises to form a ridge known as Halfway Bank (Fig. SM2). The waters are as little as 17 m. (55 ft.) to 24 m. (80 ft.) deep, and in summer nets were anchored to catch trout and whitefish as they moved near the top of the bank. Some trout may have crossed the 4 km. (2.5 mi.) to 7 km. (4.3 mi.)-wide ditch of deep water (of maximum depth 79.2 m. or 260 ft.) to spawn on the main shore in the fall. Fishermen believe, however, that it is more likely that these stocks followed the sides of Halfway Bank northward to arrive in October at Parisienne Shoal and the Sandy Islands. North of Pancake Shoal, Cubin Bank (Fig. SM5) was a popular ground in June and July for lean trout, although none chose to spawn here. This region was not a favoured one for siscowet, halfbreed, or paperbelly fishing. Agassiz's (1850) expedition caught siscowet somewhere in the vicinity of South Sandy Island or Parisienne Shoal. Summaries of the 1948 CF-1 commercial catch records show this to be the only site where fat - 104 - trout were reported -- a total of 117 kg. (259 lb.) netted in May (Ontario Ministry of Natural Resources. 1980). In contrast to areas farther north, this region supported only limited river-spawning ,stocks. It is likely that few or no lake trout ever moved into the Pancake (Fig. SM4), Batchawana (Fig. SM3), or Goulais (Fig. SM2) Rivers. A minor run frequented the Harmony River (Fig. SM3). An early photograph shows a catch of lake trout and whitefish from the St. Mary's River (possibly taken in summer; Young 1900). Lake trout continue to be found near the the rapids in late spring and summer (Mr. G. Primeau, personal comm. 1979). 5.2 COPPERMINE POINT TO CAPE GARGANTUA If a line were drawn between Coppermine Point and Cape Gargantua contained therein would be a sort of great bay, whose bottom rolls and swells to form many islands and shallows. Here, in fact, is the last extensive area of near-shore shoal waters until Schreiber is reached, far to the northwest. The gill net boats found much profit here, as also did the trollers, though at times much equipment could be lost. The rugged bottom made it seem, according to the fishermen, as if one were "fishing off the sides of mountains". Along the shore in fall the greatest concentrations of trout tended to be found near the mouths of the rivers. Spawning would commence here a short time, perhaps several days, after the river-spawners began to run. Much of the shore is rubble and gravel and, potentially at least, offers suitable spawning substrates. Important grounds were located about Coppermine Point (Fig. SMS) and in the vicinity of the Devil's Warehouse, at Gargantua (Fig. SM12). Along many other stretches though, the bottom slopes rapidly away from shore at angles greater than 40° and so - 105 - discourages spawners. Taken as a whole, the shore was less productive in the fall than the offshore shoals. Six fishermen, all veterans of these waters, were interviewed. None spoke of an early run of black trout except in the vicinity of Coppermine Point. Instead the first spawners were referred to as "regular" or "ordinary", trout. This is apparently not a confusion of terminology, and the reasons for the paucity of the smaller dark trout remain obscure. Also of note, is the fact that redfins, while abundant on the shoals, spawned at only a couple of isolated grounds along the shore. One gathering place for redfins was a short stretch of beach just north of Mica Bay and around the small nearshore Crow's Nest Island (Fig. SM6). Another was near Robertson Cove (Fig. SMlO). Spawning was reported all around Montreal Island (Fig. SM9) where the "regular" grayish trout could be taken in as little as 1.5 m. (5 ft.) of water. The best grounds were on the windward, southwestern shores. Eight km. (5 mi.) to the south, Montreal Shoal (Fig. SM7) rises to within 3 m. (10 ft.) of the surface. Not only were the regular trout here, but large redfins would come in great abundance. Neither at Montreal Island, nor along the adjacent main shore, were these redfins noted as having been prevalent. In certain years at Montreal Shoal, and possibly on some of the other offshore grounds as well, spawning may have continued much later into the season than was typical in eastern Lake Superior. Occasionally when fishing for whitefish in early December, Mr. W. Renner (personal comm. 1979) would also take a few large trout with ripe spawn. Mr. G. Primeau (personal comm. 1979) reports an incident he describes as unusual. One November day in the late 1950s, he set in the shallow waters of Montreal Shoal and procured 500 kg. (1100 lb.) of large - 112 - hug a good percentage of the shore and offer some resort to spawning fish. But beyond, until the sand beaches of Michipicoten Harbour are reached, one can pull a boat tightly against the mainland rock and still drop anchor into many metres of water without touching bottom. Only Old Woman Bay and Brule Harbour (Fig. SM13) present shoals of limited extent. The area about Indian Harbour (Fig. SM12) possessed lake trout spawning grounds of special note. Centuries before any commercial fishermen discovered their value, Ojibway natives would drift among the rocks and tiny islands to set their nets in fall. North of Ryan Point and around Cap Chaillon and the rubble strewn ledges of Grindstone Point, both redfins and the smaller regular trout would spawn. Along the Low Lands north of Grindstone Point trout could be netted shortly before their spawning time (Fig. SM13). Here was little opportunity for spawning, and it is believed that most of these fish were moving southward toward more suitable grounds. Prior to the general spawning movement onto the beach shoals, trout would also congregate upon the shelf extending outward from Grindstone Point (approximately 30 m. (100 ft.) deep). During the first week of October trout would approach the shore bank about Red Rock River (Fig. SM12) but avoided running into the river itself. When the run was on, it is said, fishermen took some of the region's largest trout from here. The average weight of each fish was almost 7 kg. (15 lb.). According to Mr. M. Krezak (pers. comm. 1979) many of these were known locally as "half-dollar fish", a variety unique to this bank (although similar in appearance to the Gull Island trout described in Section 5.2; Fig.SMll). A deep red coloured about two-thirds of the pectoral and dorsal fins (but not the tail), and the flesh was pink in colour. Their silvergray bodies were highlighted by large "coin-sized" spots or b.lotches of a darker shade. Fish marked in this manner could be - 113 - found neither along adjacent stretches of shore, nor at nearby MacKinnon Bank. No "halfdollar trout" were ever evident except at spawning time. These trout may have composed a stock distinguishable by the unique colouration they assumed in the fall, although, as stressed previously.in Section 4.1.2.4, a variety of factors of habit and habitat can influence skin colouration. The opinion was expressed by at least one fisherman that local water conditions affected by the sedimentary geology of the area, and perhaps associated with some properties of the Red Rock River outfall, acted to "create" the half-dollar trout. Upon the ledge encompassed by Old Woman Bay (Fig. SM13), lake trout would gather prior to spawning in late September at the northern point and, to a lesser degree, upon the shoals about the river's mouth. Farther to the north, spawners came just inside the entrance to Brule Harbour and along the shore of Entrance Island (Fig. SM13). Some. of the larger trout would spawn at depths as great as 18 m. (60 ft.; Mr. 0. Bjornaa, personal comm. 1979), and it is reported that the onset of spawning was several days later than at Old Woman Bay. During stormy weather it was the habit of many trout to move onto the shoal just inside the mouth of the eastern bay. Lake trout also spawned in nearby False Harbour in water too shallow for gill nets. To the north in Michipicoten Bay (Fig. SM14), spawning began at an earlier date, a fact already revealed by the Hudson Bay Company journals (see section 2.2.3.2). An entry for September 26th in the diary of William H. Keating (geologist with the 1823 American surveying expedition under Major Stephen Long) describes fish from what was probably the earliest run of lake trout sought each year by fishermen of Michipicoten House: "The trouts which we ate there appeared to us distinct from any other fish, and we regret that they had all been cut up before Mr. Say [biologist] was enabled to obtain specimens for study and preparation. The trout is a - 114 - dusky colour, with light spots irregularly scattered upon its surface; it is a richer and more substantial food than the white-fish, but not so delicate; its flesh is of a reddish tinge which approaches that of European salmon. The season for catching the trout had nearly expired..." (Keating 1825). In addition to these, redfins were once common about Michipicoten Harbour. West of Michipicoten Bay the densest spawning occurred in the close vicinity of the rivers: the Dore, the Makua (or Bear), the Eagle, and especially the Dog, now known as the University River. Holly (1953) notes that "from Dog Harbour to Ghost River, and especially off Eagle River, the whole coastal area is excellent lake trout ground any time of the year". The Dog Banks (Fig. SM15) were fished both spring and summer and it is probable that many of the shore and river-spawners remained along its sides year round. Tagging studies made by Loftus (1958) at Dog River revealed two separate movements away from the spawning grounds. A majority of recaptures by commercial fishermen occurred at the Dog Banks, although a scattering of trout westward along the shore of the lake was apparent. The arrival of trout in shallow waters in summer tended to occur earlier in Michipicoten Bay than in the vicinity of the Montreal River (Fig. SM7). Fishermen were in general agreement with the dates noted in the Hudson Bay Company journals. Trout could be seen feeding near the surface in late June, frequently in the company of ciscoes. The largest specimens were taken by trolling in July, and were usually scarce by the end of that month. A similar pattern was manifest around Grindstone Point. Trout remained on or near the top of Mackinnon Bank (approximately 27 m. (90 ft.) deep) later into the summer season (Fig. SM12). When the Purvis tugs fished here they might drop gangs of nets from Cap Chaillon around the outside of the bank toward Cape Gargantua (Mr. L. Morden, personal comm. 1979). Holly (1953) includes a report: "These form an excellent trolling ground and are a favourite spot for - 115 - 1 "hookers" or poachers to drop their illegal hook lines. The trout are large and plentiful in late July and. August and easily accessible harbours are near at hand...." South of Point Isacor (Fig. SM15) are a number of deepwater banks which yielded a good showing of trout. In May and early June, nets would be set for lean trout on the southern portion of the shallower Isacor Bank (the minimum depth of which is 36 m. -- 120 ft.) in about 82 m.(270 ft.) of water (Fig. SM16)12. Later in June, and into the summer, the nets were moved closer to shore between the Eagle and Ghost Rivers. As early as the first week of August, the Purvis tugs would occasionally set for bankers (halfbreeds?) on Isacor Bank. By mid-October siscowet ranging up to large sizes dominated the catch and continued to do so until the conclusion of fishing, usually around November 20th (James Purvis and Son, Ltd. 1934-55). Similarly, in October siscowet moved from deeper waters outside of East Bank (Fig. SM15) onto the bank itself (the minimum depth of which is 73 m. -- 240 ft.). The dates for actual fat trout spawning in this region are uncertain. Additional fat trout grounds existed closer to Michipicoten Bay, although the exact locations of some are not known. The Rev. James Evans, dispatched by the Methodist Church as minister to the Ojibway Indians, wintered at the Hudson's Bay Company Post at the mouth of what he called the "Meshebegwasdoong River" (Michipicoten River). Here in 1838 he found fat trout available within easy rowing distance: " the trout are not infrequently an inch thick with fine white fat resembling the leaf of a hog, and if hung in the sun will almost melt away leaving nothing save the skin and bones. They fry without anything to grease the pan and leave therein an abundance of fine pure sweet oil which is even good to shorten cake without leaving any disagreeably fish taste. Now laugh all."14 (Landon 1973). During the first week of July, 1974, native fat trout were taken in less than 9 m. (30 ft.) of water near Michipicoten River (Mr. H. Orr, personal comm. 1978). Fishermen confirm that in former years such an - 116 - occurrence was not unusual, although fat trout would never appear in as great numbers as at the P.Lc River and certain other localities (see Section 2.2.1.1), and there seems to have been no regular annual movement of Michipicoten fats into shallow water. A lake trout run into the Michipicoten River is known to have existed at least as long ago as the 1920s, and is described by fishermen as having been of major importance (although not as important as that of the Dog River). Often trout would work their way upstream unusually great distances -- sometimes as far as Scot's Fall 16 km. (10 mi.) from the lake. Here they would cast their spawn in the pool below the falls. About September 17th the first fish would enter the river, and while most trout spawned within ten days some could still be found 2 to 3 weeks later. Their average weight is not known, although specimens over 13.5 kg. (30 lb.) have been reported. A great number of the trout in the river were redfinned. A spawning run as large as the one described by the old-time fishermen would most likely have been mentioned in the Hudson Bay Company journals of Michipicoten House. Since it is not, presumably none existed in the 1850s, and it would seem that the Michipicoten River spawners had more recent origins in lake-spawning stocks. Trout began to appear in the Dog River (Fig. SM15) each year at about the same time as in the Michipicoten River. A spawning period from September 17th to 30th is noted by Loftus (1958), and according to Devereaux (1963), September 24th was the approximate peak of the spawning period, compared to October 21st for trout from shore sites in this area. This last date, however, refers to the later runs of lake trout -- in the vicinity of the, river mouths trout began to spawn only a short time after those actually in the rivers. It should be noted that these dates for the Dog River spawning run are later than those found to be typical during the - 117 - operation of the Hudson Bay Company post, a discrepancy for which there is no obvious reason. However, as mentioned in Section 2.2.3.2, the spawning period could show much yearly variation. Loftus (1958) found Dog River trout to be "three to four inches shorter than Montreal River trout at maturity" (measuring fork length).This represents an average weight difference of 0.7 kg. (1.5 lb.; Eschmeyer and Phillips 1965). Interviewees agreed that Dog River trout tended to be smaller than Montreal River trout. Average weights reported ranged from 1.8 kg. (4 lb.) to 3.2 kg. (7lb.), and fish over 6.6 kg. (14 lb.) were considered to have been rare. One experienced old-timer found the trout to have been "at one time so thick you could almost walk across the river on their backs". The Dog River, like the Montreal; was a favourite with sportsmen and poachers, although being less accessible, fishing pressure was reduced somewhat. Nevertheless, it is apparent that by the second half of the 1940s the stock had significantly declined over earlier years. Opinions regarding dates for the Eagle River run varied widely (Fig. SM15). It was the practice of many fishermen to commence their operations in the Pukaskwa River area (Fig. SM17) in order to catch the earliest run of trout there, and to then move consecutively to the Eagle, the Dog (and sometimes the Bear), and lastly the Montreal River. As a consequence of this practice fishermen may have arrived during a run, and so be uncertain as to the actual dates it normally began. Loftus (1958) supplies dates from September 12th to 20th for the Eagle River, and these may be the best approximations. Some years, however, trout might begin up to a week earlier, and might remain in the river three weeks. A pool inside the mouth where fish congregated made this a popular river for illegal scoopnetting. At the Bear River (Fig. SM14) a minor run moved as far as 360 m. (1120 - 118 - ft.) upstream. Similarly, trout moved into the Dore River (Fig. SM14) as far as the falls 182 m. (600 ft.) from the lake. Some trout spawned in the Old Woman River, but most spawning was on the lakeshore near the mouth (Fig. SM13). 5.4 POINT ISACOR TO PLAYTER HARBOUR The broad sweep of shore from Point Isacor northward toward the Pic River is as exposed to the moods of Superior as any place on the lake. It receives the full brunt of the howling autumn winds, and strong currents push close to its craggy banks. Places such as La Canadienne Point, Point Isacor, and outside Otter Island proved to be, at times, especially difficult to fish. Yet for those fishermen with sufficient nets and horsepower, it usually paid them well to come, set their nets, and lay overnight in the little harbours that pockmark the shore. From the impressions of several fishermen, it seems that lake trout spawned along much of the shore stretching from Ganley Harbour to the Otter Head (Fig. SM17 and Fig. SM18). Some of the heaviest spawning was at La Canadienne Point. A little to the south is the long stretch of shoals called the Pukaskwa Flats (Fig. SM17). Here tugs would commence setting nets in August at about 64 m. (210 ft.) and gradually move them up onto the Flats and the shore as the trout came on to spawn. Redfins arrived in great numbers, as did whitefish later in the season. The earliest run was predominantly silver gray bodied, regular trout, but mixed among them were easily distinguishable black trout. Most of these dark trout would probably move northward to the Pukaskwa River to spawn. In this region, as in the previous (Section 5.3), trout were particularly abundant about the mouths of the rivers early in the fall season. Here were the major grounds for the blacks, although these fish were reported to have also spawned at Simon's Harbour, Spruce Harbour - 119 - (Fig. SM.19), and the Otter Head (Fig. SM18). Mr. F. Legault (personal comm. 1978) remembers the 1930s when, at the height of the run in the vicinity of Spruce Harbour, 1200 m. (3940 ft.) of nets could be set in the evening and pulled at 11 P.M. containing 181 kg. (400 lb.), then again at 12 P.M., and again in the morning, each time with similar weight. The log of the Pijitic, a pleasure vessel operated by Marathon Paper Mills for the entertainment of its employees and clients, contains many accounts of the Pukaskwa area shore (Stevens 1950-51). An entry for September 9th, 1950 notes that lake trout were being caught along the beach and the old dock foundations at Richardson Harbour (Fig. SM18). These trout were probably just beginning to spawn. The notation for September 12th made in the vicinity of the White Gravel River (Fig. SM21), sheds more light on spawning times: "...We then trolled the beach for lake trout near nets that had been placed on the beach by the trappers and Talricco. We picked up one 8-pound laker. The brook trout had some lake trout eggs in them showing they had been working the beach where the lake trout were spawning." A spawning season commencing in early September was fairly typical of most of the grounds north of Ganley Harbour (Fig. SM17). In general, spawning in the vicinity of a river would begin at the same time, or slightly after, riverrunning began. It is less certain how late into the season spawning would continue as few boats would be fishing the grounds at this time. According to Mr. I. Johnson (personal comm. 1979), a redfin run would begin near the White Gravel River in as little as 0.6 m. (2 ft.) of water about the 1st of November (or "at the first of the dark moon"). This may not have been typical of the whole shore however. Farther to the north near Morrison Harbour (Fig. SM22), McNab {Sept. 7, 1922) suggests that the second run (which he refers to as yellowfins) would begin at an earlier date, i.e. about October 25th. Few fishermen familiar with the shore north of the White Gravel River, a part of what is now Pukaskwa National Park, are still living. Descriptions of former fish - 120 - spawning habits here must, for the most part, remain unwritten. Examination of the Purvis records for the 193Os reveals that lean trout could still be caught in the vicinity of Isacor Point early in November, long after the Eagle River (Fig. SM15) spawning run had ceased. Many No. 1 lake trout (under 5 lbs -- 2.3 kg.) were evident throughout October and may represent, in part, individuals that had moved away from the Eagle or Dog Rivers. The majority of the total daily poundage taken was made up of trout classed as medium (from 2.3 to 3.6 kg.; 5 to 8 lb.), Similarly, in the vicinity of the Pukaskwa River (Fig. SM17) large numbers of leans were still being caught when the tugs quit the waters in November. Again the major percentage of the catch was medium trout, although, after the second week in October, many large trout classed as "headless" appeared (the typical total daily weight ratio of headless to medium trout was between 1:2 and 1:3). Along the main shore at the Otter Head (Fig. SM18), pound nets were established by the Purvis operation and would still capture lean trout after midOctober (James Purvis and Son, Ltd. 1934-1955). Holly (1953) noted that excellent summer grounds for both trout and whitefish were to be found on the Pukaskwa Flats. Trolling about "the large rocks off Otter Island" (Fig. SM18) usually also met with success. He writes: "Leaving Otter Head in 1948, we delayed long enough to troll around the bare rock at the harbour entrance and at each swing around we caught one or two four to eight lb. lake trout. We ended up with about 150 lbs." In addition to these grounds, interviewees reported that lean trout could be found in shallow water in July at Simon's Harbour (Fig. SM19) and along the shore from Point Isacor to Pilot Harbour (Fig. SM16). Evidence is not sufficient to infer possible differences between these summer movements and those of the other regions. Grounds famous for siscowet and halfbreeds were the Spruce Banks (Fig. - 121 - SM20), and throughout the history of the commercial fishery they were frequented by many of the major eastern operations. A Purvis tug would start fishing for trout in the spring and continue to periodically visit the banks until November. When the fishing was concentrated more inshore for leans perhaps one or two gangs of nets at (11,890 yd. to 10,872 m. per gang) would be dropped in deep water (Mr. L. Morden, personal comm. 1978). Fishermen responded to movements of the fish up and down the bank in relation to either their feeding or spawning migrations. In summer the best sets for fats were made on the outside banks. By October effort was concentrated more inshore closer to the White Spruce River (Fig. SM19). It is not known when these trout actually spawned on the banks. Off the Oiseau Banks (Fig. SM21; exact location unknown) in 64 m. (210 ft.) of water, Mr. F. Legault (personal comm. 1979) would catch siscowet that were beginning to spawn about the 15th or 20th of October. Other siscowet and halfbreed grounds included those near the Otter Head and south of the Pukaskwa Flats, although these were not as important as the Spruce Banks. Approximately 40 km. (25 mi.) west of Simons Harbour is a small shoal (of 62 m.; 204 ft.) where fats were scarce, but paperbellies were often plentiful (Fig. 6 ).The black trout which spawned in the pools below the first rapids of the Pukaskwa River where also known to local residents as "yellowfins", many possessing this fin colouration. Colin Macmillan (1975), a Pukaskwa area resident and guide for many years in the 1940s and 1950s, describes their habits in his notes (based upon information given him by Gideon Nicoll of Port Coldwell): "These Black Trout spend most of the year in very deep water, below 400', where the commercial fishermen cannot net them. Usually about the first week or so of August a few male Black Trout would be caught in nets set in 400' of water. These fish were the vanguard of the spawning run. The females usually followed the males by two weeks. During the remainder of August and into September these fish moved in ever increasing number up from the deep water into the shallows, then into the rivers to spawn. Gib Nicol said the dates when these fish entered the rivers could vary as much as a month, he thought water temperature was the controlling factor." Usually blacks entered the Pukaskwa River between September 1st and 5th. The - 128 - and East Banks, The Hummock, and 16 km. (10 mi.) east of The Hummock (Fig. SM27). Mr. w. Saunders (personal comm. 1979) referred to the Chummy Bank trout as "bankers" which he distinguished from paperbellies. Fats were often taken at 128 m. (420 ft.) to 183 m. (600 ft.) between certain of the banks and in water off the shore bank of Michipicoten Island. What may be recognized roughly as a "ditch" surrounds the island, and in this fat trout would congregate. Little is known of the spawning habits of the non-lean varieties. Purvis (1977) records the opinion of some fishermen that siscowet south of Michipicoten Island would move to the surface in August to spawn over deep waters. Most certainly, as he states, "further investigation is likely warranted in this area." West of Michipicoten Island, the Big Flats (Fig. 6) were a source of fats, halfbreeds, and paperbellies both in summer and in fall. Here the records of James Purvis and Son, Ltd. (1934-1955) show, the best summer catches were made during the· first three weeks of July (although the best were still much less than those lifted after Mid-September). Average catch per unit effort for 5 lifts in October and November of 1936 was 240 kg./1000 m. (480 lb./1000 yd.) of gill net, which far exceeds that of Superior Shoal (Section 5.8). From September to November large fats (over 4.5 kg.; 10 lb.) dominated the catch in an average ratio of 3:1 with bankers, and medium and small fats (ranging from 2.0:l to 4.5:1, determined from 10 different lifts). Leans were only taken in small quantities as incidental catch. 5.6 PLAYTER HARBOUR TO SCHREIBER POINT The waters west of Playter Harbour (Fig. SM23) contained grounds exploited by fishermen from the three major fishing centres of Rossport, Port Coldwell (Fig. SM33), and Jackfish (Fig. SM39). Places such as Almos Shoal (Fig. SM39), - 129 - Foster Island, McKay Rocks, Hawkins Island (Fig. SM33), and Nicoll Shoal (Fig. SM35) still bear their names and remember them as early pioneers to this part of Ontario. Most of the old-time resident commercial fishermen of Port Coldwell, Peninsula (or Marathon, Fig. SM31), and Heron Bay (Fig. SM23) are gone, although a few younger people remain who carry memories back to the 1940s. Still, lack of informants has prevented some of the spawning ground maps east of Jackfish from being as comprehensive as might be desired. Jackfish fishermen knew three runs of trout. Along these shores, the earliest that trout would approach the spawning banks was August 15th to 20th. The 1911 Report of the Ontario Department of Game and Fisheries notes an unusually early spawning run at Jackfish that year: "...the run of gray and black trout was on about the 15th of August and right up to the 15th of September. These fish were then filled with ripe spawn." Spawning of the first-run trout usually began between September 20th and 25th, continuing for 10 days to 2 weeks -- dates fairly typical of the shores from Port Coldwell to Rossport. It may have been a frequent occurrence to find trout commencing to spawn in Jackfish Bay (Fig. SM39) a few days earlier than along the main-shore. Two extracts from the Port Arthur Fish Hatchery correspondence, referring to Rossport and Port Coldwell respectively, further testify to the influences of temperature on trout behaviour: "...[it] is due to the high temperature. The black trout that should spawn between September 22nd and October 5th are out in deep water and wandering all over the lake." (McNab Oct. 15, 1922). "I find that the extreme hot weather has delaid the spawning season of the black trout and when entering the shallow water, the storm of last Wednesday drove them out in deep water and I am doubtful if they will ever return (McNab Oct. 2, 1920). Both blacks and "regular" silver-gray trout formed the first run. From Cape Victoria to Worthington Bay (Fig. SM39 and Fig. SM41) the grounds were almost continuous, and the old Hudson Bay fishing stations at L'Ance a la Bouteille (Fig. SM37) and Les Petits Ecrits (Fig. SM41) were still showing especially heavy runs in the late 1940s. On the Jackfish grounds of the lakeshore, blacks - 130 - averaged 2.7 kg. (6 lb.) with few taken over 3.6 kg. (8 lb.) -- i.e. larger fish than the blacks spawning south of Coppermine Point (see Section 5.1). Usually their flesh was white. In veritable droves the trout would arrive on the shores of the Slate Islands (Fig. SM38) in early fall. The spawning grounds between Mortimer and Patterson Islands were well known, especially by the rowboat fishermen of Jackfish who tended to avoid the exposed eastern and south-eastern shores, seeking instead protection inside the islands. It was possible, however, to fish with comparative safety in the protection of the southern bays of Patterson Island, and Mr. A. King (personal comm. 1978) remembers that some of the heaviest fishing he has ever seen was at the east point of Horace Cove. Black trout were the most common variety at the islands. Spawning commenced at the same time as at Jackfish, although it was reported that they might be taken on the sheltered grounds of the islands 1.1r1.til later in the season. Smaller, on the average, than the shore blacks, most were 1.4 kg. (3 lb.) in size with few over 2.0 kg. (4.5 lb.). Red-coloured fins marked many, although these were not regarded as a discrete run or stock. As early as mid-August they were taken by fishermen trolling near the lighthouse south of Sunday Harbour. Larger trout more typically spawned at the Leadman Islands (formerly called the Little Slates). Here in the course of the fall season, blacks, yellowfins, and salmontrout (see below) came successively to cast their spawn on the rubble shoals. Mr. C. Cress (personal comm. 1978) found areas of honeycomb rock (uncommon to most of Lake Superior) to exist around Leadman Island. South of the Pic River (Fig. SM23) the first run came on the grounds at an earlier date than west of Neys Provincial Park (Fig. SM33). At Randle Point, Happy Harbour and Playter Harbour (Fig. SM23) spawn would be seen running from black trout just after the first week in September. The general westward trend of earlier spawning periods, noted in Section 5.3, was not apparent at the Pic River. South of the Pic River, this trend may have paralleled the annual September cooling rate - 131 - for near surface waters (Environment Canada 1970, 1973). Martin and Olver (1980) summarize observations concerning the importance of temperature in determining the initiation and duration of the spawning periods for lake trout populations. By August 20th it was usually possible to net large trout within Jackfish Bay at a depth of 36 m. (120 ft.) Shortly before the end of the black (or "regular") trout run, the yellow-finned trout would appear on certain of the same shoals. Mr. A. Almos (personal comm. 1979), son of one of the earliest fishermen at Jackfish, noted the presence of a "yellowfin" racer form in that area: "Yellowfins had big heads, big teeth, and some had small bodies -- they were very ugly." These tended to range from 4.5 kg. (12 lb.) to 11.3 kg. (25 lb.). (See Section 4.1.3.2). Mr. Almos reported that very occasionally a yellowfinned trout would be taken in summer. In this region the "salmon-trout of the third run"(see Section 4.1.3.3) might continue spawning throughout the month of November and perhaps into December. Pic Island (Fig. SM33) was not an important spawning ground as much of the shore is surrounded by unsuitable sandy bottom. In August, however, large lake trout would congregate in the 36 m. (120 ft.) waters of Thompson Channel where they might be captured by trolling, or in nets of up to 20 cm. (8 in.) mesh. In July and early August, large trout could also be taken by trolling close to shore. Interviewees spoke of catches similar to those made by Stevenson (1865): "August 1 ... The sun shone out warm a little before midday, and as we neared the island, passing not far from some rocks just showing above the surface with the heavy rolling swells breaking over them, I trolled for half an hour and caught three lake trout, one about 30 lbs., another over 20, and the smallest about 10 lbs." It would seem that throughout the summer these large trout would remain in the environs of the island. On "moss-covered" bottom in Thompson Channel, Mr. F. Legault (personal comm. 1980) located spawning gray trout - 132 - (salmon-trout?). According to this fisherman, the large trout common in the Thompson Channel-Detention Island area were absent east of Port Coldwell, while 11 km. (7 mi.) west of Port Coldwell the fish which he knew as gray trout were absent. Salmon-trout spawned along certain small stretches of shore in the area of Jackfish Bay (Fig. SM39). Above the town in the east arm of the bay, Mr. A. Almos (personal comm. 1979) and his father would take an average of 2 or 3 salmon-trout at spawning time. (The largest lake trout he ever saw were two individuals taken from here: one 14.1 kg. (31 lb.) and one 15 kg. (33 lb.)). Prior to the construction of the pulp mill in 1948, yellowfins would spawn inside Moberly Bay and on the shoal about Cody Island (although salmon-trout were absent from these particular grounds), as well as at Lawson Island, south of Santoy Bay (Fig. SM37). According to Mr. P. Dahl (personal comm. 1979), large lake trout with "a hooked jaw" were taken near Jackfish in December. Unfortunately, no detailed description is available (perhaps these correspond to the "shovel-nosed grays" mentioned by Thomson (1883); Appendix III). In spring, lake trout would begin to move into shallow water. In May Jackfish fishermen would start to sink their nets along the main shore between 33.0 m. (108 ft.) and 43.0 m. (144 ft.) -- shallower depths than those at which fishermen in the southeastern part of the lake would commence setting the1r nets (see Section 5.2). Most trout captured were 1.1 kg. (2.5 lb.) to 1.4 kg. (3.0 lb,), for, according to the fishermen, larger fish lay in deeper realms. Until the latter part of June, trout were seldom fished at depths shallower than 33 m. (108 ft.). At the end of June, lake trout up to large sizes would approach the shallows where they could be taken on the troll -- this movement to shallow water occurred earlier than in the southeastern part of the lake. The general retreat to deeper waters came by the end of July. (Along this portion of the northern shore, the period of these movements reported by interviewees was later than that noted in the journals of the Hudson Bay - 133 - Company post at the Pic River; see Section 2.2.1.1.) Close to the Steel River (Fig. SM37) proved an excellent site for trolling. Here boats would move in as close as 3.6 m. (12 ft.) from shore. On the bank, about the river's mouth, trout up to 13.6 kg. (30 lb.) were found, although most were from 2.3 kg. (5 lb.) to 4.5 kg. (10 lb.). Prior to the installation of the Marathon mill, lean lake trout would move close to shore in Peninsula Harbour (Fig. SM31). According to Mr. K. T. McQuaig (personal comm. 1978), it was possible to catch 90 kg. (200 lb.) to 136 kg. (300 lb.) of trout per day by trolling. The annual appearance of these summer runs was noted in the Pic Post journals (Section 2.2.1.1). In 1974 planted trout again resumed this habit by running into less than 2 m. (6 ft.) of water near the town of Marathon (where trout had not come for many years); arriving in July, they remained on into August (Mr. B. Honan, personal comm. 1978). Farther west, Mr. O. LaBarr (personal comm. 1978) of Port Coldwell engaged in recreational trolling in the vicinity of the townsite as early as the 1930s. On his hooks in summer he took black as well as gray-coloured trout. In both summer and fall, the shoals at the Black River (now dammed to create the present Aquasoban River; Fig. SM39) were once famous as trouting grounds. Over 100 years ago, a provincial geologist praised the site: "The mouth of the Black River affords one of the most esteemed fisheries upon the lake, and has long been the resort of neighbouring Indians during the proper season. Our own little net, scarcely twenty yards in length, supplied us abundantly, and a little surplus. The fish here taken are the speckled and salmon trout, white fish, siscowit, pike and herring. The labour of one or two men, during the first run, would supply a large settlement with most excellent fish...." (Anon 1872). There are indications that these stocks of lean lake trout and siscowet suffered depletion prior to the ravages of the sea lamprey. A series of deep offshore banks extending far to the southwest of Peninsula Harbour formed one of the most extensive areas of fat trout - 134 - ground in the entire lake. It is little wonder that the Nicoll Brothers of Port Coldwell, despite lower market values, found fishing for these trout quite profitable. Large siscowet of 4.5 kg. (10 lb.) to 9.0 kg. (20 lb.), and occasionally up to 11.3 kg. (25 lb.), reportedly ranged along the length of the Pic Bank (Fig. SM34 and Fig. $M36). Fishing might commence in August around the shallow hump 16 km. (10 mi.) south of Pic Island (Fig. SM32). In search of siscowet, one might also follow along the crest of the Copper Island Bank (Fig. SM40). Spawn was seen to run from these trout on September 1st, but the peak of their spawning period was probably mid-October to early November (Mr. F. Legault, personal comm. 1980). The late Mr. F. Gerow of Rossport would fish for fats along the bank after December 10th and on toward Christmas (Mr. W. Schelling, personal comm. 1979). Individuals up to 11 kg. (25 lb.) were known; large siscowet might be taken all summer but were not as abundant as those of the fall season. (It should be noted that many siscowet were also taken off the extensive Jackfish Bank (Fig. SM36), although exact locations were not determined). Closer to Port Coldwell, Uncle Bobs Bank (Fig. SM31) and the banks in the vicinity of Fred Shoal (Fig. SM33) were considered among the best grounds in the region. Perhaps the siscowet that the Hudson Bay Company fishermen captured in June near the Pic River moved to these banks in fall to spawn. (This inshore movement of fats had never been observed by several fishermen questioned at Heron Bay and Marathon, however).Another instance of siscowet moving into shallow water came from the Slate Islands. A siscowet was caught south of Dupuis Island (Fig. SM38) in approximately 3 m. (10 ft.) of water in July by Mr. P. Dahl (personal comm. 1978). It is said that this was occasionally the experience of other fishermen as well. Fats were fished in deep water until August when nets were moved onto the edges of the various banks. Purvis (1977) was informed by local fishermen that fats came onto the banks east of Pic Island (Fig. SM33) - 135 - to spawn at this time. No fishermen interviewed by myself, however, remembered having actually seen spawn running from these siscowet at this time. While certainly many were present on the banks in August, most may have actually spawned much later in the season. Nicoll Brothers Fishery used not to be able to obtain spawn from fat trout for the Port Arthur Hatchery until November and December (Nicoll Sept. 27, 1920; Nicoll Oct. 22, 1917). Purvis (1977) notes the presence of paperbellies on banks west of the Pic River (Fig. SM23). About June 1st, before any lean trout had appeared, these fish moved into shallow waters in the vicinity of Port Coldwell harbour and southward along the shore (Fig. SM33). Reportedly, the paperbellies (or bankers as the local fishermen will call them) would not bite on trolling hooks. Those captured in nets averaged 1.8kg. (4 lb.) to 2.7 kg. (6 lb.). There is no suggestion that spawning ever occurred in inshore waters. River-spawning is a much less common phenomenon along the shore of western Lake Superior than along the eastern shores. At the Steele River (Fig. SM37) the fishing success enjoyed by the Hudson Bay Company Pic Post endured into the 20th century (see Section 2.2.1.2). Local fishermen set pound nets about the mouth, and residents of Jackfish scooped upstream for black trout to supply winter provisions. These fish might run 6 km. (3.7 mi.) upstream in September. By the 1930s, however, the run had been drastically reduced. It is possible that it fell victim, in part, to overfishing, but one might also consider the possible effects of logging drives made down the Steele River. The Pic River (Fig. SM31), both in the days of the Hudson Bay Company and in more recent decades, received no river-run lake trout. It is said, however, that trout would run up the Prairie River (Fig. SM35) in autumns when storms had not forced obstructing debris into the mouth, but no further details were obtained. In the early 1900s spawning trout would also move the short distance into the Black (Aquasoban) River - 136 - as far as the falls (where poachers were known to fish; Mr. P. Dahl, personal comm. 1980; Fig. SM39). Water fluctuations generated by the construction of a dam in 1948 and the creation of artificial Hays Lake conceivably had negative effects upon this stock. One fisherman knew lake trout to be present in the Little Pic River (Fig. SM33) in the 1930s, although the run must be considered to have been minor in scale. Reasons for its disappearance are not known, but it is reported that planted lake trout again moved into the Little Pic River during the spawning season of 1978. 5.8 SCHREIBER POINT TO NIPIGON STRAIT Those fishing memories of the big haul or the big fish, the kind that the old-timers repeat for generations, have always been both common and amazing in the Pays Plat-Rossport area of Lake Superior. In the shelter of the islands, on the rubble of the many shoals, trout and whitefish spawned late into the season in numbers that were a fisherman's dream. In July of 1823, Major Delafield was impressed by the great fishing potential: "In this group of islands there are strong currents, forming eddies by their points, where the Indians usually fish for whitefish. The whitefish of Lake Superior are said not to be so good as those in Mackinac. The salmon trout are very fine. Sturgeon are much eaten but the fat is too rank, otherwise the sturgeon would be delicate and good. A place called Gun-Point [Rossport Point; Drew and Littlejohn 1975], which is formed by a long and flat neck of land projecting into the lake amongst islands affords a good fishery. This is the first extensive flat seen on the lake.'' (Delafield 1943). Along the shores about the town of Rossport (Fig. SM44), those fish called "red trout" congregated at spawning time in numbers greater than anywhere else. The rather ambiguous nature of the term has been previously discussed in Section 4.1.3.3. The sportsman Thomson (1883), however, gave credit to the form's distinctiveness: " the Red Trout is far superior to any of its confreres, and is called by the Indians - Pugwashooaneg, that is, Pays-Platt-District fish, as it is taken only in this locality, and only in the fall of the year as a rule. The Indians come from Nepigon expressly to fish - 137 - for it, and care for no other trout in comparison. It is much more highly esteemed than the brook trout.... I found the flesh of a bright pink, and the flavour exquisite." A few experienced fishermen from Rossport defined red trout as a separate spawning run. Around August 10th, red trout first began appearing in the nets. They thus preceded the blacks into shallower water, but actually spawned at a later date. Throughout September they would remain on the edges of the bank in 18 m. (60 ft.) to 36 m. (120 ft.) of water, while blacks might be spawning above them nearer to shore. Finally in the first week of October, they too would move close to shore for about two weeks in order to deposit their roe. Red trout could be seen on some of the same grounds as the blacks, but their sites, as indicated by fishermen, tended to be smaller in area and less numerous. (Examples include Bread Rock, Armour Island, and Fraser Point, Fig. SM41, Fig. SM54, and Fig. SM55 respectively). Red trout were noted as having paler skin than blacks, red flesh, and weights up to 6.8 kg. (15 lb.). The early-spawning black trout were less than 4.5 kg. (10 lb.) and typically averaged between 1.4 kg. (3 lb.) and 2.3 kg. (5 lb.). Like the black trout of the Jackfish grounds (Section 5.6), flesh colour was predominantly off-white in colour. Differences in the flesh colouration of red trout and black trout could suggest a segregation in diet of the two forms (see Section 4.1.2.3); unfortunately, fishermen very seldom note the stomach contents of their catch. Along the lakeside shores of the islands the small-class blacks would spawn almost everywhere (and at about the same time each place); separate stocks are therefore difficult to discern. Certain bay areas of clay and mud were avoided by the trout, of course, but on the whole, the blacks were pervasive, and in good years their runs were heavier than of most other places in the lakes. Although spawning grounds are not shown on the outside of Puff Island and other islands to the south (Fig. SM57), the fish probably came here as well. Fishermen were wary of certain of these wave-swept shores and would set only in the lee waters, especially - 144 - nicest lean trout for market -- few deformed fish or paperbellies were taken here. It was on the East Shoal that trout of racer appearance were most common. Upon Centre Shoal a number of different varieties were reported to have existed. Some fishermen found deformed leans in great numbers, but others took large lifts of fine trout here (Mr. R. Hamilton, personal comm.1979). Nets were set in deep water between the banks. According to Mr. I. Johnson (personal comm. 1979), a fisherman for the Nicoll Bros. of Port Coldwell, Superior Shoal siscowet were fatter than normal and less desired by markets. In general, the lean trout of Superior Shoal tended to be dark in colour (although not black like the Caribou Island trout; Section 5.5), having a dark brown back and mottled appearance. Typically the trout were slow-growing and long-lived. Prior to 1956, when the shoal was closed to commercial fishing, they tended to be, on the average, smaller than inshore trout. Interviewees reported the average round weight to have been 0.9 kg. (2 lb.) to 1.4 kg. (3 lb.): only rarely were any over 6.8 kg. (15 lb.) or9.1 kg. (20 lb.). In 1964 Superior Shoal was again opened to fishermen. Catch samples in this year found the fish to average 0.97 kg. (2.13 lb.) dressed weight. Trout from elsewhere in the lake averaged 1.12 kg. (2.48 lb.). At this time about 75 percent of the inshore samples was composed of planted trout, of which virtually none were taken on the offshore grounds (Anon. 1964). Therefore, the disparity in weight between Superior Shoal lake trout and inshore lake trout would have been even more pronounced in earlier years when native fish (generally heavier than planted trout) dominated the inshore catch. One fisherman reported having seen fish approximately 0.35 kg. (0.75 lb.) in weight containing spawn on Superior Shoal. (According to the weight-length relationship of Eschmeyer and Phillips (1965), this represents a total length of 0.35 m. -- 1.2 ft.). Samples taken from Superior Shoal in the early 1960s included trout ranging up to 16 years of age (Lawrie 1964). A spawning run of trout in the Marquette-Munising area was found by Rahrer (1967) to be comprised of - 145 - individuals of only 7 through 12 years of age, with ten-year olds being most abundant. (The smallest mature male was 0.56 m. or 1.9 ft.). At Superior Shoal lean trout have been found in spawning condition throughout the fishing season from June to October (see also Loftus 1958). Isolated fish populations, less exposed to inshore climatic controls such as temperature and wind and not limited by unavailability of suitable substrate, could potentially evolve a wide range of spawning times. It is also quite possible that the lake trout of the Shoal did not spawn every year. This situation is common in more northern lakes. In Keller Lake (Northwest Territories) Johnson (1973) found that each fish spawned only three times during its lifetime and tended to be in very poor condition. Martin and Olver (1980) have summarized the literature pertaining to the factors that affect the growth rate of lake trout. They stress the importance of food supply, population density, and temperature. Nilsson and Svardson (1968), for example, have shown the effect that a good forage base can have on stimulating growth of trout in Swedish lakes. It is probable that all three of the above factors were acting in combination to determine fish growth on Superior Shoal. Here fishermen found a great scarcity of forage fish. Gut analyses of adult trout confirmed this, revealing Mysis relicta to have been the major food item (Budd 1955). This also helps to explain, at least to some degree, the fact that most of the lean trout had very red flesh (see Section 4.1.2.3). According to Mr. J. Macdonald (personal comm. 1978), owner of Ferroclad Fisheries of Mamainse, "even the bladders are very pink, the gut barrels would run red from the bladders." The leans, and to a lesser extent the siscowets and paperbellies, showed a high percentage of deformities on the Shoal. Certainly not all were of poor condition, but many appeared large-headed with emaciated bodies of the racer form and were easily recognized by inspectors and fishermen as coming from Superior Shoal (Section 4.1.3.6). When discussing these fish with oldtime fishermen the term "freaks" will often be heard - 146 - (or less frequently the designation "steamboat-chasers"; Mr. P. Hamilton, personal comm. 1980). Mr. L. Morden (1978) comments: "... some were humpbacked, some had too big a head, some had huge fins, which were sometimes orange or red. There were so many different kinds of deformities. The fish were very thick and there may have been interbreeding." This situation still prevailed when Adolph King started to journey out to the Shoal in 1951. He referred to these fish as "reptiles", with prominent teeth and eyes sunken in their sockets. "They were vicious, ripping much net twine," he stated, adding humourously, "We were afraid they might even snap at our legs" (personal comm. 1978; also in Drew and Littlejohn (1975) but misquoted). The Nipigon Bay Fish Company of Rossport was forced to ship the "grotesque" Superior Shoal leans headless if they were to sell. They brought two or three cents less per pound. Their saleability, however, was fortunately enhanced by the redness of their flesh. Lake trout, while perhaps of poorer market quality, have always been relatively abundant on the shoals, and this has been a successful fishing ground for those fishermen owning tugs and willing to make the long trip here. Numbers of fish were so great that the pressures of population density may have put food resources at a premium. In addition to physical deformities, the existence of cannibalism was frequently discovered (also see Lawrie and Rahrer 1973). A noteworthy example was a 5.4 kg. (12 lb.) trout found by one fisherman to have a 0.9 kg. (2 lb.) trout in its stomach. Martin and Olver (1980) have reviewed observations of cannibalism: "Cannibalism is apparently of minor significance in most lake charr populations. It appears to be most common in arctic lakes where there is not a diversity of forage species and the lake charr itself may be one of the few species present." From the records of the Purvis Company it is possible to obtain some idea of C.P.E. for lake trout (fats, bankers, and leans combined) in the 1930s. The greatest catch from a single lift was obtained August 24th, - 147 - 1938 from 2½ gangs of net (with 3658 m.- 12000 yd.- to a gang) which yielded 3360 kg. (7400 lb.) of trout. Catch per 914 m. (1000 yd.) of gill net was 92 kg. (203 lb.) in 1938 and 93 kg. (206 lb.) in 1939 (James Purvis and Son, Ltd. 1934-1955). By the time he quit fishing Superior Shoal in 1956, Mr. A. King (personal comm. 1978) had perceived a decrease in the percentage of trout that showed cannibalism or deformity. During the six years that Ferroclad Fisheries fished here after the Shoal was reopened in 1964, the thinbodied racers were a rarity in the catch (Mr. J. Macdonald, personal comm. 1978). The reason for this is not clear. It is possible that the populations were sufficiently thinned by fishing during the 1950s to remove pressure on limited resources, permitting improved growth of the surviving fish. However, the 1967/1970 average C.P.E. from the Shoal was much higher than the Purvis estimate -- 213 kg. (470 lb.) per 914 m. (1000 yd.) of comparable net (assuming an average weight of 0.97 kg., or 2.13 lb., per fish; Anon 1964, Lawrie 1978) No apparent change in the food supply had occurred. Smelt had not colonized the shoal in the 1960s. Neither had possibly more robust hatchery trout colonized the Shoal, nor was sea lamprey predation a significant problem. A similar phenomenon involving the revitalization of trout populations of poor condition was also witnessed in Lake Nipigon. In 1916 the lake was opened to commercial operators as a source of fish during World War I, and the Fort William Fish Co. maintained the steam tug Pewabic here. Mr. A. King (personal comm. 1978), engineer, found conditions of cannibalism and deformity similar to those he later found on Superior Shoal. Despite this, Mr. King explains, a "fishing bonanza" occurred in which many outfits with tugs and gas boats arrived in Macdiarmid to take advantage of the large tonnages which were to be had. He writes: " Therefore these conditions for three seasons as stated reduced the population of fish in Lake Nipigon to such a point that "cannibalism" almost was a thing of the past and Lake Trout, - 148 - especially now, were able to find enough feed to become a very much improved and healthier specimen which kept improving up to the present day. Lake Nipigon trout has been for many years since that era a very fine species... The writer ... had seen in 1917 [fish], especially trout, netted in Lake Nipigon with their head near the weight of the rest of the body." (King n.d.) 5.9 NIPIGON STRAIT TO THUNDER CAPE The great archipelago of islands that commences near Schreiber arcs southward as far as Pigeon River and the American border. The coastline of the Black Bay Peninsula is shattered into a series of deeply indented bays and small nearshore islands. Gollat (1976) notes that the larger bays (Otter Cove, Sheesheeb Bay, and Sturgeon Bay; Fig. SM59 and Fig. SM60) were very popular with lake trout anglers. Inside the majority of the bays, however, sand and mud deposits are laid down, and they proved generally unsuitable sites for spawning stocks. About the multitude of islands, the shoals are predominantly covered by more acceptable rubble and coarse gravel. The initiation of spawning for the different runs of trout followed a definite trend in time as one moved westward toward Thunder Bay. Along the shore, fishermen would maintain contact with one another, and when someone to the east announced that a run was on his friends farther west would prepare nets for their own fall fishing. At Black's Wharf near Sheesheeb Bay (Fig. SM59), blacks were on their spawning grounds around September 17th, while about three days would pass until the general commencement of spawning between Shaganash Island (Fig. SM61) and Black Bay (Fig. SM63). According to Mr. M. Gerow (personal comm. 1980) fishing at Point Magnet (Fig. SM61) began somewhat earlier than in the vicinity of Rossport. Spawning dates at Silver Islet (Fig. SM65) differed from those noted at Number 10 Light (Shaganash Island, Mr. G. McKay, personal comm. 1979) and, in turn, spawning about Thunder Cape (Fig. SM66) was later than at Silver Islet. Along the Black Bay Peninsula, the early-spawning blacks were - 149 - described· as usually weighing 0.9 kg. (2 lb.) to 2.3 kg. (5 lb.), with some reaching 3.2 kg. (7 lb.); i.e. similar in weight to those described in Sections 5.1 and 5.7 but somewhat lighter (on the average) than those taken on the mainshore Jackfish grounds (Section 5.6). Nets of 12.7 cm. (5 in.) or 14.0 cm. (5.5 in.) mesh were used in their capture. Here Mr. R. McKay (personal comm. 1979), whose grounds were in the Shaganash Island area, found some blacks to have coloured fins. Most of these possessed fins edged with white, although some had reddish and others slightly bluish-coloured fins. Certain years the blacks would be available on these grounds until almost the middle of October. Their flesh was usually pink in colour. Especially abundant were the blacks spawning on the inside of Edward and Porphory Islands (Fig. SM63) and Point Magnet (Fig. SM61). From pound nets set at Point Magnet, the Port Arthur Hatchery could usually be confident of a good supply of black trout eggs (McNab Oct. 19, 1923). On the whole, black trout tended to spawn in greatest numbers on the leeward side of islands, while larger red trout (see below) were most prevalent on the windward lakeside shores. Later runs of trout followed a spawning sequence along the Peninsula toward Thunder Bay similar to that of the black trout. The second included many red-finned individuals. Near Jean Pierre Bay (Fig. SM60) these would range from 2.3 kg. (5 lb.) to 6.8 kg. (15 lb.). The smaller individuals began spawning near the end of the first week in October and the peak of the run occurred between October 15th and 20th. Larger size classes might continue into November. Yellowfins reportedly spawned at Stanton Island (Fig. SM60) and on a couple of shoals west of Shaganash Island (although not many spawned around the island itself; Fig. SM61). At some locations one would also discover gray trout and large blacks present as late spawners. In general, however, fishermen in this region, unlike those from the Rossport area, do not remember the different runs as having been obviously discrete in time. Some lumped grays, yellowfins, and redfins - 150 - under a general class "red trout", perceiving them to have arrived more or less continuously on certain grounds. On the shoals about Shaganash Island, blacks were still on the grounds when the reds appeared, and there was a mingling of both until the larger reds became dominant in late October and November. In this region, as compared to the Rossport area, stocks of large lake trout seem to have been less significant to the fall fishery, although less effort was probably devoted to their capture. Smaller mesh sizes were usually employed and fishermen tended not to set trout nets as late into the season, either retiring them for the winter or switching their attention to the herring fishery. It is interesting to note that spawning period trends analogous to those of the lake trout also existed for lake herring. At Camp Bay on the Sibley Peninsula (Fig. SM65) one could catch herring as much as two to three weeks before they were spawning in Thunder Bay. Certain autumns the fishing here was finished before it had even begun in Thunder Bay. The initiation of herring spawning (like lake trout spawning) is, at least in part, under the control of declining water temperatures: the shallowness of Black Bay causes its waters to cool more rapidly during the course of the fall than those of Thunder Bay. It is also the opinion of local fishermen that a number of different and discrete herring stocks frequent the waters of these two bays at spawning time. Very little lake trout spawning occurred within Black Bay itself except near the entrance; south from George Point on the western side and south from Foster Point on the eastern (Fig. SM63 and Fig. SM64). Much of the bottom is composed of clay with scattered patches of sand, weed, and rock, and there are far better grounds here for walleye, whitefish, and herring than for lake trout (although, in former years, Mr. A. Nuttall (personal comm. 1980) would take small numbers of fall trout at Bent Island, inset Fig. SM64). During the month of September, its water can be several Co warmer than that of the main lake. When the currents of Black Bay push southward along the Sibley Peninsula (Fig. SM63) they may exert - 151 - strong influences on the fishing there, especially at those times of the year when the fish frequent the shallows. As also happens in Nipigon Strait (see Section 5.7), nets set in spring near the edge of the ice in Black Bay would yield good trout catches. When the ice lingered fishing proved quite successful, but when strong winds broke up the ice and sent it out into the main lake the lake trout would leave for deep water (Mr. 0. Kukko, personal comm. 1978). Later, in early summer, wind from the southeast brought trout close to shore along Middlebrun Channel and at Middlebrun Point (Fig. SM65). But those days when the wind changed to the northeast and pushed the warm, brownish water of Black Bay southward, the trout again retreated to open water. Similar effects could be seen in the fall when the bay waters could act to delay spawning. Early in spring, trout were taken outside the islands along the Black Bay Peninsula where the banks slip away into very deep water and also in certain of the deeper inside channels (such as that west of Lasher and Spain Islands; Fig. SM60). In May, movements to shallower zones occurred as trout pushed into Sheesheeb Bay (Fig. SM59) and also congregated in the 27 m. (90ft.) shelf west of Gourdeau, Swede, and Macoun islands (Fig. SM60). Early in June, they would move to shallower depths, and large trout were available to trollers near Jean Pierre and Sturgeon Bays in July (Fig. SM60). Here they might be seen surface feeding on insects. According to a fisherman interviewed by Gollat (1976), an excellent trolling area for large trout in the 1940s ran from Swede to Brodeur Islands and across to Otter Island and Roche Debout Point (Fig. SM59 and Fig. SM60). In the l930s it was not uncommon to take trout weighing from 18 kg. (40 lb.) to 29 kg. (65 lb.) on the shoals outside Black's Dock (Fig. SM59; Mrs. E. McKay, personal comm.1978). Spring and summer movements of lake trout seem to have occurred at approximately the same time in this region and in the Rossport region. (No westward time trends are apparent, and perhaps these seasonal movements are less under the influence of water temperature and current than those - 152 - occurring at spawning time). Along Middlebrun Channel (such as at Finlay Bay; Fig. SM65) pound nets were set about the 1st of June. Trout first appeared in shallow water around June 10th. (and some would occasionally push northward almost to the bottom of Black Bay). The large "pound-net trout" were most evident in 3.7 m. (12 ft..) of water and less at the end of June and early in July (Mr. M. Gerow, personal comm. 1980). This behaviour paralleled that of trout among the Rossport Islands, where they were available at shallow depths until after mid-July (see Section 5.7). Along the Black Bay and Sibley Peninsula, the general retreat would occur about this time, although in some years large stragglers would drift in and out of the near-surface waters until August. It should be noted that the trout found in Sheesheeb Bay in June were of a small class, averaging 0.9 kg. (2 lb.) to 1.4 kg. (3 lb.). Large trout were uncommon here. Stockier than trout taken at other nearby locations, these small trout were leant distinction by their somewhat "humped" appearance (although they were not paperbellies and are described by Mr. O. Nordlander (personal comm. 1979) as having been a "nice-eating, beautiful lean trout"). Small spots dotted the skin and flesh colour ranged from off-white to red. According to Mr. Nordlander, only two or three net lifts were possible before these fish disappeared until the following summer -- in this area they came only as a feeding group. When Sheesheeb Bay fish were sold to the Nipigon Bay Fish Company of Rossport, employees in the packing plant found them to be easily recognizable (Mr. Baker, personal comm. 1978). The Bateau Rock area (Fig. SM62) consists of a series of jagged peaks and shoals, which rise to the water's surface at one point. The shallowest peaks are contained within a 2 km.2 (1.2 mi.2) area lying . 24 km. (15 mi.) due south of Flour Island (Fig. SM57) and 11 km. (7 mi.) east of the Gull Island shoals, which straddle the international boundary. Like Superior Shoal, the deep lake locations of these two fishing grounds could act as a genetic isolating mechanism for the fish stocks located there. - 153 - Two fishermen, experienced with trout fishing on both Superior Shoal and the Bateau Rock shoals, stressed the similarity in appearance of the leans at the two locations: flesh colour was deep red, skin colour was dark (and quite mottled), heads appeared to be large, and body shapes were comparable. (The racer form seems to have been much less evident at Bateau Rock, however). Crawford (1966) describes the lean trout at the latter location as "dark green laterally, with light red markings on the fins." Lean trout netted at Bateau Rock were small, ranging from 1.1 kg. (2.5 lb.) to 3.2 kg. (7 lb.). Net sets made by the Fisheries Research Board of Canada revealed that the population contained unusually large numbers of older lake trout up to ten years of age. (The age distribution was younger than that of Superior Shoal leans, however; Lawrie 1964). Also found along the sides of the Bateau Rock shoals were fats, halfbreeds, and paperbellies. It has been the experience of some fishermen that many of the lake trout of the greater depths possessed abnormally large and protruding eyes, a trait which earned them the name "popeyes". (One is reminded of the deep-water form of chub, Coregonus kiyi, endowed with eyes comparatively large for its family). This unusual trait may have arisen as an isolated adaptation to life in deep water. It has also been noted among Arctic char, whitefish and salmon that some deep-water forms possess larger eyes than shallower zone forms (Nikol'skiy 1969). Mr. B. Humby (personal comm. 1979) also reports that Bateau Rock fats were abnormally mottled in appearance. To the west, at the Gull Island shoals, American fishermen found humpers (or paperbellies) spawning at a depth of about 75 m. (245 ft.). These fish were known as "mooneyes", but, unfortunately, no description is available (Rakestraw 1968, and Mr. S. Sivertson, personal comm. 1979). Elsewhere, nets could be set to capture siscowets and halfbreeds at various locations along the outside dropoff of the Black Bay Peninsula islands. South of Puff Island offered deepwater fishing during the summer months (Fig. SM57 and Fig. SM58). Another important summer location east