@prefix ns0: . @prefix edm: . @prefix dcterms: . @prefix dc: . @prefix skos: . ns0:identifierAIP "b5ec3cc9-d8b4-4558-b633-ea4feb15fca0"@en ; edm:dataProvider "CONTENTdm"@en ; dcterms:alternative "REPORT OF THE COMMISSIONER OF FISHERIES."@en ; dcterms:isReferencedBy "http://resolve.library.ubc.ca/cgi-bin/catsearch?bid=1198198"@en ; dcterms:isPartOf "Sessional Papers of the Province of British Columbia"@en ; dcterms:creator "British Columbia. Legislative Assembly"@en ; dcterms:issued "2016-03-21"@en, "[1919]"@en ; edm:aggregatedCHO "https://open.library.ubc.ca/collections/bcsessional/items/1.0059765/source.json"@en ; dc:format "application/pdf"@en ; skos:note """ PROVINCE OF BRITISH COLUMBIA EEPOET OP THE COMMISSIONER OF FISHERIES FOR THE YEAR ENDING DECEMBER 31st, 1918 WITH APPENDICES PRINTED BY AUTHORITY OF THE LEGISLATIVE ASSEMBLY. VICTORIA, B.C.: Printed by William FI. Cullin, Printer to the King's Most Excellent Majesty. 1919. To His Honour Sir Prank Stillman Barnard, K.C.M.G., Lieutenant-Governor of the Province of British Columbia. May it please Your Honour: I beg to submit herewith, a report reviewing the operations of the Provincial Fisheries Department for the year ending December 31st, 1918, with Appendices. WILLIAM SLOAN, Commissioner of Fisheries. Provincial Fisheries Department, Commissioner of Fisheries' Office, Victoria, British Columbia, March fyth, 1919. TABLE OF CONTENTS. FISHERIES COMMISSIONER'S REPORT FOR 1918. Paob. Standing of Provinces of the Dominion 7 Species and Value of Fish marketed 7 Number of Persons engaged and Value of Plants S The Salmon-catch of 1918 8 The Salmon-pack of Province S Summary of Reports from Spawning-grounds 10 Digest of Report of American-Canadian Fisheries Conference 10 Fishing by Indians in the Fraser River 11 Migration of Adult Sockeye Salmon in Puget Sound and Fraser River 12 Dr. Gilbert's Salmon Inspection of 1918 12 Halibut 14 Whaling 15 Pilchards 15 Deep-sea Trawling 15 Ottesen Method of freezing Fish 17 Hatching Egg Collections IS APPENDICES. The Spawning-beds of the Fraser River 19 The Spawning-beds of Rivers Inlet - 21 The Spawning-beds of Smith Inlet 24 Contribution to the Life-history of the Sockeye Salmon. (Paper No. 5.) By Dr. C. H. Gilbert 26 Hon. Wm. Sloan's Statement made at the American-Canadian Fisheries Conference .. 55 John P. Babcock's Statement made at the American-Canadian Fisheries Conference .. 56 Migration of Adult Sockeye Salmon in Puget Sound and Fraser River. By Henry O'Malley and Willis H. Rice 60 Salmon-pack of 191S in detail 92 Salmon-pack of Province, 1903 to 191S, inclusive 93 FISHERIES COMMISSIONER'S REPORT FOR 1918. Value of Canadian Fisheries and Standing op Provinces. The value of the fishery products of Canada for the year ending December 31st, 1917, totalled $52,312,044, as against $39,208,378 for the preceding year, notwithstanding that the latter was considerably greater than in any preceding year.* During the year 1917 British Columbia contributed fishery products of a total value of $21,518,595, or 41 per cent, of the total for the Dominion. As in recent years, British Columbia again leads all the Provinces of Canada in the value of her fishery products. Her output for the year exceeded in value those of Nova Scotia by $7,050,276, and exceeded that of all the other Provinces combined by $11,156,941. The following statement gives in their order of rank the value of the fishery products of the Provinces for the year ending December 31st, 1917:— British Columbia $21,518,595 00 Nova Scotia 14,468,319 00 New Brunswick 6,143,088 00 Quebec 3,414,378 00 Ontario 2,866,419 00 Prince Edward Island 1,786,310 00 Manitoba 1,543,288 00 Saskatchewan 320,238 00 Alberta 184,009 00 Yukon ' 67,400 00 Total $52,312,044 00 The Species and Value of Fish caught in British Columbia. The total value of. each species of fish taken in British Columbia for year ending December 31st, 1917, is given in the following statement:— Salmon $16,828,7S3 00 Black cod 879,404 00 Hake and husk 1,090 00 Herring 1,192,654 00 Shad 675 00 Halibut 1,721,012 00 Soles 78,649 00 Flounders 23,601 00 Skate 10,117 00 Smelts 14,270 00 Oolachans 10,991 00 Brill 51,420 00 Octopus 1,656 00 Rock-cod 8,688 OO Pilchards 11,810 00 Whiting 2,725 00 Grayfish 4,480 00 Oysters 32,202 00 Clams and quahaugs 84,000 00 Dulse, crabs, and cockles 48,424 00 Trout 10,350 00 Sturgeon 9,790 00 Perch 4,920 00 Carried forward $21,031,711 00 * The data used here is derived from the Dominion Bureau of Statistics, 1917. X 8 Report of the Commissioner of Fisheries. 1919 The Species and Value of Fish caught in British Columbia—Concluded. Brought forward $21,031,711 00 Mixed fish 13,184 00 Salmon-roe 7,820 00 Fur-seal skins • 6,540 00 Bone-meal 10,185 00 Fertilizer 70,164 00 Whale-oil 342,247 00 Fish-oil " 23,S92 00 Witches 50 00 Gill-bone p. 12,802 00 Total '. $21,518,595 00 The total value for the year shows an increase over the previous year of $6,881,249; salmon products were increased by $0,285,178, due both to an increase in size of pack and its value. The halibut-catch was less, and, notwithstanding an increase in price, shows a decrease from the previous year of $305,658. The herring-catch is valued at $1,192,654, a slight gain over the previous year, and whale products show a gain of $94,220, with a total value of $446,486. The Number of Persons engaged and Value of Fishery Plants and Apparatus in British Columbia in 191S. The following statement gives the number and the value of the plants, vessels, boats, and apparatus used In the fisheries of the Province in 1918, and the number of persons engaged:— No. Value. Vessels, tugs, and carrying smacks 514 $ 2,500,801 00 Gasolene-boats 3,172) Sail and row boats 8,548( 1,837,820 00 Value of seines, trap and smelt nets 1,829,115 00 Value of hand-lines, weirs, trawls, etc 119,3S1 00 Approximate value of salmon and other canneries, fish-houses, freezers, and fixtures 9,519,941 00 Total $15,S07,058 00 Number of Persons engaged in Fishery. Number in vessels 1,589 Number in boats 11,378 Number in canneries, fish-houses, etc 7,916 Total 20,883 The Salmon-catch of 1918. The salmon-catch of the year 191S exceeded all previous records and produced a pack of 1,616,157 cases. It exceeded the previous high record of 1917 by 5S1.672 cases. The catch of sockeye gave a pack of -276,459 cases, as against 330,209 cases in 1917; 107,354 cases of red and white springs packed, as against 76,276 cases in 1917; the pack of pink salmon totalled 527,745 cases, a gain over 1917 of 30,9S6 cases; and the pack of 497,615 cases of chum salmon exceeded that of the previous year by 21,342 cases. The 191S Salmon-pack by Districts. The Fraser River.—The total pack of all grades of salmon on the Fraser River run in British Columbia waters totalled but 210,S51 cases, as against 402,538 cases in 1917, 127,472 in 1916, 320,519 in 1915, and 349,294 cases in 1914. Over 50 per cent, of the total pack consisted of chums and pinks. The catch of sockeye was less than 10 per cent, of the total, was much less than in any former year, and produced a pack of but 19,697 cases, as against 198,183 cases in the previous fourth year, 1914. Since 1914 was the brood-year of the greater proportion of this year's sockeye run, it shows that the run was less than 10 per cent, of that of its brood-year. 9 Geo. 5 British Columbia. X 9 The season's pack of Fraser River seeking sockeye on Puget Sound totalled 50,723 cases, as against 357,374 cases four years ago. The pack of sockeye In the entire Fraser River District in 1918 totalled only 70,420 cases. Four years ago the combined catch of sockeye in that district produced a pack of 534,434 cases, or 86 per cent, more than this season. A more forceful demonstration of the depletion of the run of sockeye to that river could not be exacted. It warrants the statement that the Fraser is fished out of sockeye and the run is perilously near extinction. The conditions that have produced this result have been set forth in previous reports of the Department. As early as 1902 the Department gave warning that too many sockeye were being taken, and that, unless the fish were permitted to reach the spawning-grounds in far greater numbers, the run would be exterminated. Year after year that warning was repeated, and backed up by statements showing a steady decline in the number of sockeye that reached the spawning area. The warning fell upon ground as barren as the spawning area. The fishermen and the canners would not be denied. The amount and efficiency of the fishing-gear used to take the fish was increased without giving the sockeye any additional protection. The Canadian authorities indicated in no uncertain manner their willingness to meet the conditions by restricting fishing in their waters. The authorities of the State of Washington could not be induced to meet the situation by affording the fish that pass through their waters to reach the Fraser any adequate measure of protection. They disputed the facts submitted, and instead of restricting fishing they permitted an increase in the amount and character of gear used, thus adding to the efficiency of their methods. Even as late as the spring of 1918, at the American-Canadian Fisheries Conference, the State of Washington authorities contended that conditions on the Fraser were not of so serious a nature as to call for drastic treatment. At the Vancouver meeting of that Conference the Commissioner and his assistant both submitted written statements setting forth the views of the Provincial Government on this vital question. Those statements are reproduced in the Appendices of this report. The Salmon-catch of Northern Waters. The Skeena River.—The salmon-pack on the Skeena River in 1918 exceeded all previous records with a total of 374,216 cases, as against the previous record pack of 292,219 cases in 1917. There was a gain in all grades. Sockeye gave a pack of 123,322 cases, as against 05,760 in 1917, 60,293 cases in 1910, 116,533 in 1915, 136,166 in 1914, and 52,927 in 1913. Weather conditions throughout the sockeye season were most propitious. Fishing for spring and cohoe salmon was conducted at numerous outlying points by Hollers, many of whom had excellent season's return. Rivers Inlet.—The catch of salmon on the Rivers Inlet section shows a substantial gain over the two preceding years, with a total pack of 103,155 cases. The catch of sockeye gave a pack of 53,401 cases, as against 01,195 in 1917, 44,936 in 1916, 103,350 in 1915, 89,S90 in 1914, and 61,745 in 1913. The pack of pinks exceeded all previous years, with a total of 29,542 cases, as against the two previous high records of 8,065 in 1917 and 5,7S4 in 1914. The run of pinks to Rivers Inlet has heretofore been very small. No theory as to the increase of this season has been advanced. The Nass River.—The catch of salmon in the Nass River section is also a high-record one, totalling 143,908 cases, as against the previous high records of 119,495 in 1917 and 126,686 in 1916. The catch of sockeye again shows a decline, with a pack of 21,816 cases, as against 22.1S8 in 1917, 31,411 in 1916, 39,349 in 1915, and 31,327 cases in 1914. The_ record shows a steady decline In the run of sockeye notwithstanding the increase of gear used. The catch of pinks amp chums shows substantial increase and is attributed to the more extensive use of traps and' gear. The Vancouver Island Section. The pack of salmon taken from Vancouver Island waters, with the exception of the sockeye taken in the trap in the vicinity of Sooke, that are credited to the Fraser pack, totalled 3S9,S15 cases. Operations were conducted at all points where any salmon could be taken. Large numbers of spring and cohoe salmon were taken in the open sea, facing the west coast, by means of trolling. Some 400 small boats were so engaged. The catch of sockeye was small. All other grades show an increase. Chums, with pack of 251,266 cases, exceed all others combined. The catch of pinks gave a pack of 57,035 cases and cohoe 40,732 cases. Some 5,500,000 lb. of salmon, mostly X 10 Report of the Commissioner of Fisheries. 1919 chums, was exported in fresh state from the west coast. The catches made are regarded by the Department as ail overdraft on the runs, and if continued will exterminate them. The Catch at Nitinat Lake. The catch of fall-running salmon at Nitinat Lake, on the southern shore of Vancouver Island, in both 1917 and 1918 were of such an extraordinary and excessive nature as to call for extended comment. Nitinat Lake is but fourteen miles in length by half a mile in width. It lies at sea- level and is connected with the sea by a narrow and short passage. There is but one tributary of importance, the river at the head of the lake, which has a length of some twenty miles. Except during heavy rains, its volume is not large. Its gravelled channel offers a limited extent of spawning area. It is the only tributary frequented by salmon. A few creek sockeye are said to enter it in the late spring. Chums in large number run to that stream in the fall. Commercial fishing was first undertaken there in 1917 by the Lummi Bay Packing Company, who built a cannery there that spring. By the use of two purse-seines the company's catch in 1917 produced a pack of 51,252 of chum salmon. Not being able to handle all the fish taken, they exported to their plants in the State of Washington some millions of pounds. Their pack in 1917 was the second largest put up at any one cannery in the Province. In 191S the catch at Nitinat Lake produced a pack of 84,000 cases, 20,000 cases more than were packed at any other one cannery in the Province. The fishing-gear used by the company consisted of two purse-nets and a trap, driven at the mouth of the river at the head of the lake. In addition to the pack of 84,000 cases, some 5,000,000 lb. of salmon were exported from that district. No other water of anything like the proportions of Nitinat Lake has ever produced in two succeeding years anything like as many fish. No salmon run to any watershed of its size can stand such a drain without complete annihilation. The drain on the salmon run to all Vancouver Island waters is in the judgment of the Department excessive, and calls for the immediate and serious attention of the fishery officials of the Dominion. The fish must be given far greater measures of protection than at present afforded them, or they will cease to be of commercial importance. Reports from Salmon Spawning Areas of the Province in 1918. During the season the Department, as usual, conducted an investigation of the salmon spawning areas of the Fraser River and Rivers and Smith Inlets. Detailed reports are attached hereto. The spawning-grounds of the Fraser River basin were again inspected by John P. Babcock, Assistant to the Commissioner. He was assisted by Dr. C. H. Gilbert and Inspector of Fisheries C. P. Hickman. In the report made by Mr. Babcock it is again disclosed that the number of salmon which reached the spawning-beds was far less than in any previous year. In all the great lake sections of the Fraser above Hell's Gate Canyon there were few or no sockeye, and in the lower section, with the exception of the Lillooet Lake District, there were fewer sockeye than ever previously reported. As affecting future runs no importance need be attached to the few that did spawn in the upper section of the Fraser, and less importance than heretofore can be attached to those that spawned in the lower section. The total number of sockeye-eggs collected from the tributaries of the Fraser and placed in the hatcheries totalled 20,581,500, as against 22,484,000 four years ago. Twenty million sockeye-eggs were taken from Alaska waters by the United States Bureau of Fisheries and donated to the Dominion Government to assist in restocking the Fraser, and were shipped to the hatchery on Harrison Lake and are being incubated therein. Rivers and Smith Inlets Spawning-grounds.—Fishery Overseer A. W. Stone reports that there was a serious shortage of sockeye salmon on the spawning-beds of Owikeno Lake, at the head of Rivers Inlet. He estimates that there was 25 per cent, less sockeye on the beds than in any one of the past six years. Officer Stone, who also inspected the spawning-grounds of the Smith Inlet run of sockeye, reports that the number of fish observed there this year equalled the run of 1915, but that they were many times less than in 1914. The American-Canadian Fisheries Commission of 191S. The American-Canadian Fisheries Commission appointed by the Governments of the United States and Canada to investigate fishery questions in waters contiguous to their boundary-lines 9 Geo. 5 British Columbia. . X 11 held meetings on this Coast, in Seattle, Vancouver, Prince Rupert, and Ketchikan in the spring of 1918. The Commission consisted of the Secretary of Commerce and Labour of the United States, the Hon. William C. Redfield, E. F. Sweet, Assistant to the Secretary of Commerce, and Dr. Hugh M. Smith, United States Commissioner of Fisheries, on behalf of the United States; and the Hon. Chief Justice J. D. Hazen, of New Brunswick, G. J. Desbarats, Deputy Minister of Naval Service, and Wm. A. Found, Superintendent of Fisheries, Ottawa, on behalf of the Canadian Government. The inquiry conducted on this Coast was confined to an investigation of the salmon-fisheries of the Fraser River and the American waters of Washington Sound and the halibut-fisheries of the Pacific. At the Vancouver meeting of the International Commission, Commissioner Wm. Sloan and his Assistant, John P. Babcock, made extended statements as to conditions in the Fraser, which are reproduced in the Appendix of this report. The Conference concluded its sittings in Ottawa in September and unanimously adopted a report. At the time of writing the report has not been made public. Fishing by Indians in the Fraser River Basin above Tidal Limits. The runs of salmon to the spawning-beds of the Fraser have become so alarmingly attenuated that drastic measures will have to be taken to restore the runs. The measures to be taken must not only include the secession of all fishing in tidal limits for a period of years, but must be made to include all fishing above tidal limits by Indians for all time, notwithstanding that they have both a natural and a treaty right to take such salmon as they desire for food so long as they confine themselves to the gear originally used by them. The Indians resident in the Fraser River basin have always fished for salmon. Before the advent of the white men they depended almost wholly upon salmon for their winter food. They take them as near their homes as possible and cure them by smoking. An approximate estimate of the number of sockeye salmon taken by Indians above the commercial fishing limits on the Fraser River in the years of the big runs previous to 1913 place the number at 200,000. It probably largely exceeded that number. The number taken in former years of the big run by the Indians resident in the Chilcotin River section is placed at from 40,000 to 50,000. Those taken by Indians from the Fraser basin above the mouth of the Chilcotin is given at 20,000, and between the mouth of the Chilcotin and Bridge Rivers at 40,000-odd, while those along the banks of the Thompson and Shuswap and Adams Lakes at 30,000, in the canyon of the Fraser at Yale at 15,000, and in the Harrison-Lillooet Lakes section at 40,000. The figures here given Include only sockeye that were taken, dressed, and cured. They do not include the other species of salmon taken or the sockeye consumed in a fresh state. The catches made by Indians in the years of the big runs previous to 1913 were not a serious drain on the run of those years. The numbers taken during the lean years, though considerably less than in the big years, was a serious drain on the spawning-beds. During the last decade none of the tribes above the canyon at Yale have been able to get any considerable number in any lean year, and their catches in both 1913 and 1917 did not exceed those of early lean years. The drain on the runs, was nevertheless serious. The right of the Indians to take salmon is unquestioned, but the number of salmon they can now catch is so small as to be of little benefit to them. Owing to the fact that most of the Indians now grow the bulk of the food they use and are no longer dependent upon salmon, and that drastic measures must be taken to restore the runs of salmon to the Fraser, the Government should step in and acquire by purchase the Indians' right to take fish above the commercial boundaries. It is suggested that the Indians, if deliberately approached, would dispose of their fishing rights to the Government, and that the Government is fully warranted in entering upon negotiations to acquire those rights. The sooner the better. It is self-evident that salmon taken from the vicinity of their spawning-grounds are much more valuable individually and collectively than the fish taken within commercial fishing waters. Sockeye that were worth 75 cents each to commercial fishermen on the Fraser the past year are worth many times that amount on the spawning-grounds. The runs of the future depend upon the number of eggs deposited and hatched within the river's basin. The female sockeye of the Fraser produce averagely 2,500 eggs each. It has been suggested that the Indians be deprived of their right to fish above commercial fishing boundaries, and in lieu thereof be aided to do all their fishing within commercial limits. X 12 Report of the Commissioner of Fisheries. 1919 Possibly the Indians hi the delta of the Fraser and those resident south and west of Yale might operate to advantage and fairly economically. The suggestion is wholly impracticable so far as the great majority of the Indians are concerned. Very few of the Indians from the Chilcotin or north of Bridge River have ever engaged in fishing within commercial waters. For them to journey, with women, who do ail of the work of dressing and smoking the fish, from their homes to tidal waters and then back again with their bales of dried fish would cost the Government many times more than to purchase and transport a far greater supply of other foods for the use of the Indians at home. The Indians of the Fraser are reasonable beings. There are many clear-headed men in their councils. If this matter be taken up with them in a deliberate, clear, and painstaking manner, it is advanced that they would soon see the benefits to be derived by disposing of their fishing rights. Their right at the present time is of little value to the most of them and is steadily lessening in value. Many of them realize this to be true. The Government should act in this matter without further delay; the expense to be covered from the collection made for fishing licences. Migration of Adult Sockeye Salmon in Puget Sound and Fraser River. During the American-Canadian Fisheries Conference on the Pacific Coast in April and May. 191S, questions were raised as to the time taken by sockeye salmon to pass from the waters of Juan de Fuca Strait through Puget Sound and the Gulf of Georgia to the Fraser River. In order to settle this important economic question, experiments were made by the Fishery Departments of Canada and the United States Governments in marking with metal tags adult sockeye taken from fish-traps in the waters named. The results of this experiment are covered in a report just issued by the Bureau of Fisheries, Washington, D.C, which is reproduced in the Appendix of this report from advance sheets furnished the Department by Dr. Hugh M. Smith, United States Commissioner of Fisheries. The report is of interest, though it cannot be said that the results of the experiment have by any means determined the questions at issue. The questions involved are vital in determining measures of conservation, and must be conducted for a sufficient period of years to produce conclusions of value. Contribution to the Life-history of the Sockeye Salmon. Dr. Gilbert's fifth contribution to the reports of the Department on the life-history of the sockeye salmon, which is issued herewith, includes an analysis of the sockejTe runs of 1918 to the Fraser, Skeena, and Nass Rivers and Rivers Inlet, and throws much additional light on the mooted question of the return of mature spawning salmon to the particular part of a river's basin in which they had been batched and reared. The present paper is of great value since it conclusively demonstrates that, in the Eraser at least, the sockeye does return to spawn in the identical tributary in which it was hatched. In dealing with this important economic question Dr. Gilbert considers a number of propositions. If there'was, he submits, no tendency for a salmon to return to any definite spawning district, the sockeye entering so extensive a basin as the Fraser would be as likely to spawn above as belowT the canyon at Yale. Every spawning-bed in that event would contain individuals which had been hatched and reared in widely different portions of the basin, and the output of one centrally located hatchery would furnish stock for all sections, since each spawning area would draw indifferently on the common stock of returning migrants and the salmon frequenting it would necessarily agree in their structural characteristics with those frequenting each of the other spawning areas throughout the basin. There would be no basis for differences. Those spawning in different parts of the basin would agree in their average size as yearlings and as adults. They would agree in the relative number of "four- and five-year-old fish, in the relative numbers of those which had spent one year or two years in fresh water after hatching, or had proceeded to sea as soon as they were free-swimming. All the groups and all the variations of character which are found in the main runs off the mouth of the river would be repeated in every spawning section. Such would inevitably be the case if the salmon passed indifferently to the various spawning areas, without reference to that of their native habitat. On the other hand, he submits that If differences are found to exist in the populations of distinct spawning districts it must be considered direct evidence that segregation has occurred, and is based on some principle which has determined that certain individuals in the common 9 Geo. 5 British Columbia. X 13 run shall pass to one spawning area while others go elsewhere to spawn. The only principle adequate to explain such segregation, if it generally occurs, would be the return of fish to spawn in their native tributary. Dr. Gilbert abundantly demonstrates that the latter principle obtains, as between different river-basins, even when they are very limited in size and have their outlets to the sea in close proximity to each other. In their passage to these minor streams along the salt-water channels the schools of sockeye are often intimately mingled, and they separate only as the river-mouths are approached. Off Namu, on Fitzhugh Sound, gill-nets operated along the shore capture in about equal numbers at the same time sockeye which are bound for Bella Coola and those which are bound for Kimsquit. They have been distinguished by characteristic scale-markings. They run together along the same channels. Where they separate has not been demonstrated, but by the time Bella Coola and Kimsquit Rivers are reached complete separation has occurred. Dr. Gilbert points out that this case does not seemingly differ greatly with that of fish bound for different tributaries of the same river. Conceivably, the schools may be separated as their respective tributaries are reached. If the populations of these tributaries show distinguishing characteristics, no other satisfactory explanation presents itself. It must be considered, he states, also, that a partial segregation of spawners may occur, though not a total one. That within certain limits indifference exists as to the precise spawning-ground frequented. Thus the up-river fish may predominately run early and proceed above the canyon in the Fraser, passing without selection into any stream that becomes available; while the late-running fish may enter the Harrison, the Pitt, and Cultus Lakes again without determinate destination. That this supposition is without validity for the lower river becomes evident from the fact, emphasized in last year's report, that the fish entering Pitt Lake are so widely different from those entering- Cultus Lake, and even the different spawning-beds of the Harrison, can usually be shown to possess individual characteristics of unquestioned significance. With regard to the river above the canyon the case is not so clear, partly due, no doubt, to lack of adequate material. No material at all adequate has been obtained from any up-river tributary except the Chilcotin, because of late years the runs to the upper river have become extremely attenuated. Efforts of the last two years failed to produce the desired evidence, and it may now be too late to secure it owing to permanent depletion. Where spawning populations show distinguishing characteristics, the inference is clear and unavoidable. A physiological barrier has existed for so long a period that minor differentiations have developed and a strain or sub-race has been formed. But what is to be concluded in cases where the colonies of two distinct spawning areas present no characteristics by which we can distinguish them? Obviously, Dr. Gilbert states, we must infer either (a) that segregation in their case has not occurred, or (6) that it has occurred but hasnot been effective in producing divergence between the colonies. If the first of these were true, it would signify that the " home-stream" instinct was in most cases rigidly operative, while in one or more instances within the same river-basin it was wholly in abeyance. The second supposition would seem a priori far less improbable. Complete segregation of spawning fish mayT occur, aud each individual return to the stream in which it was hatched and reared, and yet, in certain instances, separate colonies may have failed to develop distinct methods of growth by which they can be distinguished. A failure to discover distinguishing characteristics between populations of separate spawning areas need not indicate that their progeny will fail to return at maturity to its native stream. The two colonies may have been wholly distinct and self-perpetuating for a very long period and yet no differences have been developed. If we should adopt distinctive marks for the young from each of the two streams on their downward migration to the sea, it might be shown at maturity that each stream contained spawners with a single mark only, the one that had been given to the young from that stream. Yet the adults from the two streams might be otherwise indistinguishable. Negative evidence in such a case proves nothing. But if affirmative evidence is found, it is conclusive. If differences are found to exist, there is no alternative to the acceptance of the " home-stream " theory. Reviewing the evidence collected in the Fraser basin in 1918, Dr. Gilbert says:— " In general, it can be stated that the examination of this material completely7 supported and verified the conclusions reached in 1916 with regard to the spawning-beds below the canyon; ice affirm without qualification that they are as distinctivelg populated as though located in separate streams independently entering the sea. Additional material only emphasizes the fact. X 14 Report of the Commissioner of Fisheries. 1919 Not only were the different colonies in 191S marked by distinctive characters, but these characters were the same for each colony as those noted in 1916." Dr. Gilbert showTs that a very conspicuous instance of racial differentiation is furnished by the Harrison watershed. It is not known how many distinct spawning areas may have existed there. Several minor colonies, like that known to have existed at Silver Creek, are now practically exterminated, and there are doubtless distinct areas in Lillooet and Tenas Lakes, but no records have been obtained. A very limited number of late spawners still frequent the slack waters of the main river above Harrison Rapids, the vanishing remnant of a once fine run visits the lake at the head of Morris Creek, and a considerable run still goes up the Birkenhead River; both of these runs pass over the spawming-beds at the Harrison Rapids, but not one of them tarries there. An examination of 200 spawners taken from the rapids in 191S by Mr. Robertson, the clever Superintendent of the hatchery on Harrison Lake, failed to yield a single fish which had been spawned elsewhere. Not one of the 200 could be confused with those constituting either the Morris Creek or the Birkenhead colony, and not one of the fish taken on the rapids was found among the 400 specimens taken at Morris Creek, and over 200 from the Skookumchuck and the Birkenhead. No Morris Creek strays were found in the Birkenhead and no Birkenhead fish were found in Morris Creek, in spite of the fact that the Birkenhead fish had all passed close to the mouth of Morris Creek on their way up. In these cases the law of segregation is most strictly in force, yet it seems impossible to imagine a locality better suited to straying of the spawning fish with resulting mixture of populations. That it does not occur here gives confidence that the same law rules everywhere, though the evidence of it may be less compelling. The most interesting colony of the Harrison, Dr. Gilbert shows, is the one which spawns late in the season at Harrison Rapids. The nature of that spawning area is in itself highly unusual. The spawning habits of the sockeye have been shown to be rigidly fixed. Unlike other species, they must resort to a lake, and fail to frequent streams save during their migrations. They spawn in lake tributaries or on the gravel shores of lakes. Their young develop in the waters of lakes and commonly reside there over a year before migrating to the sea. Of the few which descend from lakes to the sea, few ordinarily survive to return at maturity. But the Harrison Rapids spawners use gravel-bars in the main river and there is no lake between them and the sea. Their young, in consequence, have no lake to drop down to. They have, Dr. Gilbert shows, adopted the highly exceptional method of life for a sockeye, of migrating to sea as soon as they are free-swimming and before their scales have begun to grow. No data is obtainable as to the percentage of survival, but enough have survived to keep the colony flourishing. This is apparently not true of the sockeye of any other spawning area of the Fraser. It demonstrates, however, that under the spur of necessity a sockeye colony can adapt itself to the habit of sea migration in the early free-swimming fry stage, a rare habit in sockeye, though a characteristic of pink and chum salmon and frequently of the spring. The 200 specimens of Harrison Rapids spawners taken and examined in 1918 had all gone to sea in the fry stage. Not one of them exhibited in the centre of its scale any trace of growth in fresh water. The scales from all others taken from different spawning areas recorded growth in fresh water of a year or more. In 191S, as in 1916, Dr. Gilbert failed to find any individuals of sea-type spawning in any part of the Fraser basin other than the Harrison Rapids, though upwards of a thousand specimens from other sections were examined. The fact that no sockeye of sea-type was discovered elsewhere in the basin, when coupled with the further fact that every Harrison Rapids sockeye belonged to the sea-type, furnishes the strongest possible evidence of the return of spawning sockeye to their native districts. Dr. Gilbert's paper is reproduced in the Appendix of this report, together with reproductions of microscopic photographs of the centres of thirty-four scales of sockeye salmon taken from many sections of the Fraser River basin, the markings of which demonstrate that in the Fraser the sockeye returns to spawn in the identical tributary in which it was hatched. Halibut. The landing of halibut at our ports and those of Seattle and Tacoma for the year 1918 shows a decrease of 14,375,000 lb. from that of 1917. Vancouver landing for 1918 totalled but 1,902,000 lb. as against 5,162,000 lb. in 1917. Prince Rupert shows a decrease of 3,801,000 lb., with total landings of 14,777,000 lb. Such statements demonstrate the alarming decrease of this fishery and afford additional evidence of the value of the publications of this Department in 9 Geo. 5 British Columbia. X 15 1915 and 1916 on the condition affecting the fishery, and which furnished evidence to show that the banks had been seriously depleted, and that early action on the part of Canada and the United States must be taken if the fishery was not to be destroyed. Up to this time no action has been taken by either Government. Joint action alone can be effective. This subject was considered by the American-Canadian Conference and evidence taken. The conclusions reached have not been made public. The remarkable and alarming decline in the catch of 1918 is most impressive and demands early action upon the part of Canada and the United States. The correctness of W. F. Thompson's conclusions, furnished and published by this Department in 1915 and 1916, can no longer be questioned. Unless the halibut is given immediate and radical measure of protection, they will be exterminated on this Coast as they have been on the Atlantic. International action must be taken. By no other means can the halibut banks be given the necessary protection. Whaling. Whaling during 1918 was most successful. Three stations were operated on our Coast. The following statement, furnished the Department by the Consolidated Whaling Company, of Victoria, gives the number and species of whales landed at each of their stations during 1918:— Statement showing Number and Species of Whales landed at British Columbia Stations in 1918. Station. Sperm. Sulphur. Bin. Sei. Hump. Right. Bottle- nose. Total. Kyuquot Rose Harbour Naden 12 17 5 4 12 15 88 69 51 101 15 14 41 25 29 i i 246 140 114 Totals 34 31 208 130 95 1 l 500 Canned Whale-meat. A total of 29,585 cases of whale-meat was canned at the company's station of Kyuquot in 1918. The pack was made in round 1-lb. flat tins similar in all respects to salmon, each case containing forty-eight tins. The Commissioner's Assistant made a careful inspection of canning operations, and in his report said: "The canning plant at the Kyuquot whaling-station, where the meat of the whale is canned, is up to date in every respect. I followed the process through every detail. Every feature of it, from the cooling of the meat to the sealing of the can, is in every respect all the most fastidious could wish. I sampled the contents of many cans taken from the pile of those canned previous to my visit. The contents, to my. taste, equalled the best canned beef or mutton in the market. There can be no question but that canned whale-meat will command a market." A rapid expansion in output is anticipated. Pilchards. The run of pilchards to the estuaries of the west coast of Vancouver Island was greater than previously reported. Heretofore they have not been noted on the west coast in waters south of Barkley Sound. This year they were taken in numbers by the salmon-traps in Juan de Fuca Strait and in nets in Esquimalt Harbour, Cadboro Bay and Haro Strait as far north as Mayne Island. The fish were large and fat. Many were sold fresh and several salmon-canners packed them in 1-lb. and %-lb. cans, and they met with ready sale. Deep-sea Trawling. The distinguishing feature of the fishery year was the successful operating of deep-sea trawling vessels from our ports. Experiments conducted in 1917 out of Prince Rupert demonstrated that there were " banks " adjacent to that port where trawling could be conducted, and which resulted in the successful operations of 191S. The trawler " James Carruthers " made forty-nine trips out of Prince Rupert between February 28th and December 19th, occupying 151 days, which resulted in the landing of some 2,000,000 lb. of fish consisting of flounders, sole, witch, brill, cod, skate, and a limited amount of halibut. The average trip of the trawler from dock to dock was three days. On being dressed the fish were placed in cold storage and frozen, and marketed principally in the North-west, under provisions of Order No. IS of the Canada Food Board, being X 16 Report of the Commissioner of Fisheries. 1919 retailed from 10 to 11 cents per pound; the Dominion Fisheries Department assuming two-thirds of transport charges. The trawlers' operations were limited because of the lack of a market. The catch was in excess of the demand. There is no longer a question as to lack of supply or suitable trawling " banks." Given a market an abundant supply is assured. The bulk of the catch was, as stated, marketed in the North-west. Heretofore that section has been supplied only from the Coast with salmon and halibut. Other salt-water fishes were unknown. The sale of Pacific flatfish and cods was stimulated by an active publicity campaign conducted by the Canada Food Board. Sales in 1918 were sufficiently large to warrant the belief that a permanent market can be established, provided the price of that year can be maintained. There is no question as to the food values of trawl-caught fish. They are the equal of any food-fishes, with the exception of salmon. They have more flavour than halibut, and can be sold at less than half the price of either salmon or halibut. Evidence of the value in which they are estimated is demonstrated by the fact that the landing of trawl-caught fish at the chief ports of Great Britain, notwithstanding the war, between May 4th and September 14th, 191S, totalled 1,658,764,000 lb. During the last four months of 1917 an average of 1,500,000 lb. per month of trawl-caught fish were landed in San Francisco. In face of the declining catch of our estuary and halibut fisheries, it is encouraging to record the success of deep-sea trawling from our ports. A success so great as to warrant the belief that eventually a large fleet of trawlers will operate from our ports. The Department give considerable attention to the operation of the trawlers. Curtailed as operations were by a limited market and cold-storage capacity, they were extensive enough, as stated, to warrant exploitation. In advancing measures for the engagement of returned overseas service men, the Commissioner stated:— " In formulating measures for the engagement of returned overseas men, full consideration should be given to the fisheries. Deep-sea trawling conducted out of British Columbia ports in 1917 and 1918 w7ere profitable and afford evidence of the wealth of food-fishes that await exploitation. What Is required at this time to reap the wealth is more suitable vessels and more trained men to operate them. This want can be supplied ou the Pacific Coast by the establishment of a trawling-school—a school for the training of men to engage in deep-sea trawling-fishing —by providing a number of suitable vessels equipped for deep-sea fishing and officered by experienced navigators, engineers, and fishermen to act as instructors, and to which returned overseas men could be apprenticed for instruction in navigation ; the assembling and operation of steam and gas engines; the construction, care, casting, and hauling of trawl-nets; and the dressing, storing, packing, and shipment of the fish caught. " In consequence, it is suggested that the Government provide a number of suitably equipped vessels, mail them with capable instructors, and engage them in training returned overseas men in deep-sea trawling-fishing. The men during their apprenticeship to be paid a good living wage and given a bonus from the ship's earnings, after capital expenditures have been paid, that would eventually give the apprentices, who by that time aje experienced men, the ownership of the vessel. "Well managed, the operations of such vessels would be sufficiently profitable'to furnish the means for operation and maintenance and create a sinking fund that would reimburse the Government and permit the vessel being transferred to the men to be worked on their own. There is room for at least ten such training-ships on the Pacific Coast of Canada, each of which could provide tuition for twenty-five apprentices. Such a school as is here suggested could be directed and operated in connection with the Naval Training School now established at Esquimalt. " In addition to the practical trawling sea school here suggested, there should be established at Prince Rupert and Vancouver fish curing and packing schools for the practical training of overseas men in the curing, packing, and marketing of fish, and through which the catches of the training-trawlers could market their catch. " Men graduated from the land school could be sent in to the North-west and established as wholesale and retail handlers of both frozen and cured fish, first as the agents of the schools, and so bonused as to eventually acquire their plant and the business they have been enabled to establish. " The establishing and maintenance of such schools as are here suggested will not provide for the engagement of a large number of returned men, but it does provide for a number and will materially assist in building up a food-providing class of hardy men in a field where men are needed and where they may gainfully engage." 9 Geo. 5 British Columbia. X 17 The Ottesen Method of freezing Fish. A. Ottesen, an exporter of fish of Thistal, Denmark, after years of experience, has invented a system of freezing fresh fish that has widely been adopted in Europe, and which should prove of value to companies operating in the Province which are not equipped with cold-storage plants. The Fishery Director of Norway described the Ottesen method as revolutionizing the fish- refrigerator. The Inspector-General of France says: " After being thawed in water for five or six hours the fish frozen by the Ottesen method look like fresh fish and are in the same condition. The freezing does not cause the slightest injury to the flesh tissues, but, on the contrary, after the fish have been prepared for the table they keep their firmness and full flavour. Fish so frozen have been shipped long distances without change in texture or flavour." The principle of the Ottesen method is simple and easily understood. The fish are not frozen in the ordinary way in air, but in a strong solution of common salt in water, which is cooled to very low temperature and in which the fish are immersed. The freezing-point of the solution of salt depends upon the degree of saltness; the more salt, the lower the temperature. A solution with 3 per cent, of salt freezes the fish at about 27° Fahr. A solution with 17 to 20 per cent, of salt may be cooled down to from 10° to 60° Fahr. without freezing the solution. It is a solution of this kind which is used (instead of cooled air) to remove the heat from the fish and to freeze it. The apparatus consists essentially of an insulated iron vat or container of suitable dimensions in which the salt solution is placed. The cooling of the solution is accomplished by means of a central spiral, or worm, of iron tubing in connection with an ammonia refrigerating-machine, and the solution is kept in constant movement by a central revolving propeller. Large fish are hung in the solution, the smaller being placed in baskets of wire netting. The great value of the method appears in the simplicity of the apparatus, its cost, and in the rapid freezing of the fish. Herring in baskets are completely frozen in half an hour and the largest cod In from two to three hours. In ordinary methods of freezing in air the time taken would be from twenty-four to thirty hours. An ordinary Ottesen apparatus can freeze 10 tons of fish in twenty-four hours. The A. Ottesen fish-freezing process was patented in Canada several years ago and the ■ Canadian rights sold to the Det Danske Frysnings Kompagin, address 33 Ny Toldbodgade, Copenhagen. The Sockeye Salmon-pack* of the Fraser River District from 1900 to 1918, inclusive. Year. Fraser River. Puget Sound. Totals. 1900 229,800 228,704 458 504 1901 928,669 1.105.096 2,033,765 1902 293,477 339,556 633,033 372,020 1903 204.809 167,211 1904 72.6S8 123,419 196,107 1905 837.489 847,122 1,684,611 1906 183,007 182,241 365,248 1907 62,617 96,974 159,591 1908 74,574 155,218 229,792 585,435 1,005,120 1,590,555 1910 150,432 234,437 384,869 1911 62,817 126,950 189,767 1912 . 123,879 183,896 307 775 1913 736,661 1,664,827 336.251 2,401,488 534,434 155,714 1914 198,183 1915 91,130 64,584 1916 27,394 78,476 105,870 1917 148,164 411,538 559,732 1918 19.697 50,723 70,420 * Given in cases—forty-eight 1-lb cans to case. X 18 Report of the Commissioner op Fisheries. 1919 The following statement of the salmon-eggs placed in the hatcheries of the Province in 191S is furnished by Lieut.-Colonel F. H. Cunningham, Chief Inspector of Fisheries for the Dominion:— Statement showing Total Number of Eggs placed in Various Hatcheries, 1918. Name of Hatchery. Sockeye. Spring. Cohoe. Humpback. Eastern Spk. Brook Trout. Dog Salmon. Alaska Sockeye. Whitensh Anderson Lake, V.I. Babine Lake, Skeena River 2,515,500 9,100,000 3,120,000 1,873,000 2,514,875 11.960.000 3,628,500 3,024,000 465,500 3,025,000 45,000 272,000 2,172,900 1,038,000 42,000 650,000 192,000 126,000 15,363,000 30,000 V.I. Cultus Lake, Era ser Gerrard, Kootenay Harrison Lake, Fraser Kennedy Lake, V.I. New Westminster, Fraser Pemberton, Fraser 396,000 24,000 20,700,000 7,000,000 Seton Lake, Fras'r Totals 37,735,875 3,535,500 4,366,900 15,489,000 30,000 420,000 20,700,000 7,000,000 Sockeye Egg-take at Fraser River Hatcheries from 1901 to 1918. 1901 15,741,000 1902 72,034,000 1903 13,464,000 1904 9,469,000 1905 97,656,000 1906 51,121,000 1907 53,952,000 1908 46,709,000 1909 98,000,000 1910 37,343,000 1911 22,937,000 1912 38,500,000 1913 86,000,000 1914 28,589,000 1915 68,476,000 1916 '.... 40,203,000 1917 31,004,000 1918 ' 20,581,500* * In addition to the collections made from Fraser River, 20,000,000 sockeye-eggs taken in Alaska were hatched at Harrison Lake. 9 Geo. 5 Spawning-beds op Fraser River. X 19 APPENDICES. THE SPAWNING-BEDS OF THE FRASER RIVER. Hon. Wm. Sloan, Commissioner of Fisheries, Victoria, B.C. Sir,—I have the honour to submit that during the season just closed I made a thorough inspection of the salmon fishing and spawning grounds of the Fraser River. The catch of all species of salmon from the Fraser River District within the Province produced a pack of only 201,851 cases, of which more than half consisted of pinks and chums. The total pack of all grades four years ago totalled 349,294 cases, 19S,1S3 of which were sockeyes. The Catch of Sockeye. The catch of sockeye in Provincial waters of the Fraser River District this year was very much the smallest ever made. It produced a pack of only 19,697 cases, as against 148,164 cases in 1917, 32,146 in 1916, 91,130 in 1915, and 198,183 in the preceding fourth year, 1914. The catch in our waters of the district was but 10 per cent, of that made in the latter year, the brood-year of the greater proportion of this year's run. The catch of sockeye this season in the State of Washington wraters, included in the term Fraser River District, was but 14 per cent, of that made in those waters four years ago. The combined catch of sockeye in that entire district this season was the smallest ever recorded, and was 87 per cent, less than the catch in the preceding fourth year. Since the primary test of the size of a run of sockeye to the Fraser River District is the catch, and is expressed in figures by the pack, the above statements display the serious extent of the depletion in that run to that district, and fully justified the statement made by you last summer at the Vancouver meeting of the Canadian-American Fisheries Conference, " that the run of sockeye salmon to the Fraser River is perilously near to extermination." The Department's report of four years ago, in dealing with the conditions of the spawning- beds of the Fraser, stated that the beds contained less sockeye than I had previously observed there. It is therefore not surprising now to learn that the run this year was so limited. Condition on the Spawning-beds of the Fraser. During the season I personally inspected the principal sockeye-spawning area of the Fraser River basin, and was assisted by Dr. C. H! Gilbert and Inspector of Fisheries C. P. Hickman. The inspection made this year, like the sixteen former inspections made by me since 1901, was conducted during the spawning period. As a result of the last inspection, I have to submit the opinion that far less sockeye reached the spawning-beds this year than in any previous one of which there is a record. In all of the great lake sections of the Fraser above Hell's Gate Canyon there were few or no sockeye, and in the lower section of the Fraser, with the exception of Lillooet Lake, there were less sockeye than previously reported. The number of sockeye that reached Hell's Gate, in the narrow canyon of the Fraser above the village of Yale, was smaller than in any former year. The Indians who fished there, under daily supervision of Dominion fishery officials, caught less than 1,500 sockeye. They could have taken more, and would have done so, but for the fact that the run of spring salmon was above the average and satisfied their own demands, and they were not permitted to sell either fresh or'salted fish to the whites. The Indians who fished at the canyon in the Fraser above the mouth of Bridge River caught less than 400 sockeye, and those engaged on the Fraser at Chimney Creek and at Soda Creek took less than 100 at each location, while those fishing just below the mouth of Quesnel River were unable to catch any sockeye at all. No sockeye were reported from that section of the basin above the mouth of Quesnel River. X 20 Report op the Commissioner of Fisheries. 1919 The conditions at Quesnel and Chilko Lakes are best evidenced by the statements that less than fifty sockeye are recorded by the Provincial watchman as having entered Quesnel Lake, and that it took Dr. Gilbert aud myself, operating separately at the principal Indian fishing-places on the Chilcotin River, a week to obtain scales and measurements from 100 individual sockeye. Throughout the season the Chilcotin Indians at all their fishing-places caught less than 500 sockeye. Dr. Gilbert was desirous of obtaining scales and measurements of sockeye entering the Chilcotin and Quesnel Rivers. As stated, it took a week to get the data from the Chilcotin, and we could get no data from the Quesnel because so few reached that river, and they came so scatteringly that none were caught. Not to exceed 100 sockeye are known to have entered Seton-Anderson or Adams Lakes, though the creeks leading from them were under observation throughout the season, and less than 100 sockeye were found in any of the many tributaries of Shuswap Lake. Indians resident along and near the Thompson River again had no fishing-platforms on the banks of that river, and are not believed to have taken fifty sockeye or any pink salmon this year. No pink salmon were reported in the Fraser above Hell's Gate. In former reports on the spawning area of the Fraser I have dealt separately and at length with conditions at all the principal districts in the upper section of the basin. I do not do so this year because the whole story is told in the above statements. There were too few fish in those waters this season to warrant further records. No future importance need be attached to the few that did spawn there. The number of sockeye which reached the spawning area of the lower section of the Fraser basin this year were less than four years ago, and the hatcheries collected less sockeye-eggs. The run to the streams at the head of Lillooet Lake was less than four years ago. The hatchery maintained there collected 12,000.000 sockeye-eggs, as against 15,000,000 four years ago. The run was, however, greater than last year. In addition to the fish that reached the hatchery this season, it is to be noted that the Dominion Fishery and Indian Officer Grant, who was stationed at the portage between Harrison and Lillooet Lakes, records that the Indians fishing at that portage took over 6,CO0 sockeye. As in recent years, by orders of Officer Grant, the Indians retained only the male fish caught by them. The females taken were returned uninjured to the water, and were thus enabled to reach the streams at the head of lake. The number of sockeye that entered Plarrison and the lower lakes was less than four years ago, and less eggs were collected for the hatcheries. The total number of sockeye-eggs collected for the hatcheries on the Fraser from its tributaries was 12,210,000 less than four years ago. Chief Inspector of Fisheries Lieut.-Colonel F. H. Cunningham reports that some 20,000,000 sockeye-eggs taken from streams in Alaska were donated and shipped to the Harrison Lake Hatchery by the United States Bureau of Fisheries, and that they will hatch them and the young be liberated in the vicinity of Harrison Lake. This addition to the hatcheries gives a total of 32,210,000 sockeye being Incubated for liberation in the Fraser, as against 22,484,000 in 1914. Reports from the Dominion Hatcheries operating in British Columbia in 1918. 1 Sockeye. Spring. Cohoe. Chums. Pinks. Cultus Lake (tributary Westminster (tributary to Harrison) . . . to Harrison) . . . 1,873.000 3,120,000 4,000.000 3,217,000 2.500,000 2.313,000 8,000,000 2,600.000 3,025,000 40,000 417,404 42,000 1,038,000 636,000 163,000 176,000 2,019,000 396,000 23,000 Steelliead. 37,000 40,000 Pitt Lake Victoria, B.C., December 1st, 1918. I have, etc., John Pease Babcock, Assistant to the Commissioner. 9 Geo. 5 Spawning-beds of Rivers Inlet. X 21 THE SPAWNING-BEDS OF RIVERS INLET. Hon. Wm. Sloan, Commissioner of Fisheries, Victoria, B.C. Sir,—I have the honour to submit the following report in respect to the conditions existing on the spawning-beds at Rivers Inlet for the year 191S:— Leaving the cannery at Rivers Inlet on October 7th, I proceeded to the lake with the intention of going right through to the headwaters, a distance of between thirty-eight and forty miles, but the weather, which had permitted the inspection of the spawning-beds of Smith Inlet to be made under such favourable conditions, suddenly changed, and created such a sea on Lake Owikeno by the high wind that it was considered advisable to remain near the mouth until an improvement was shown. Eventually reaching Jeneesee Creek, a distance of twenty-two miles, I inspected the spawning-beds situated at this point. The creek, which in 1913 and 1914 was literally swarmed with sockeye salmon, and commented upon very favourably in my reports for those two years, was practically empty; beyond one or two small schools swimming about in the creek and a few noted breaking water outside on the gravel shores of the beach, there was nothing to indicate an exceptional run to this tributary. The hatchery officials, who usually obtain their eggs from this stream, had been unsuccessful in securing any so far. Since the experience of the past two or three years has shown that the fish have arrived late, it was deemed desirable to again visit this sectioii on my return from the headwaters of the lake, in case I was too early. The logs and jams which litter the stream all the way up to the falls form a formidable barrier to the number of sockeye which eventually reach these beds. I do not remember, in the six years of my experience of this particular stream, noting so many small grilse as on this occasion. Schools of them were observed spawning on the gravel-bars right up to the falls, having managed to wriggle their way through the fence erected by the Dominion Hatchery. The small number of sockeye salmon which had arrived at the creek were of good average size and compared favourably with the 1914 run. The Machmell River, another tributary flowing directly into the lake at this point, although one of the largest of the streams to be met with in this watershed, does not, nevertheless, appear to be favoured byr the sockeye salmon. Owing to the high stage of the water at this time the inspection was conducted under extreme difficulty. The thick muddy condition of this glacier stream did not favour accurate observation of the likely numbers spawning on the beds, but indications higher up, where the river-bed was visible, did not point to a big run; very few fish were seen. The swift-running river precluded all attempts to secure a collection of eggs or scales for the Department. It was hoped that the run of sockeye salmon up the Nookins River, a tributary to the Machmell, would make up for the indifferent number seen at the latter point, but an examination of this stream disclosed similar conditions. It was difficult to believe that these spawning-beds which in former years have been so plentifully seeded could be depleted to the extent shown this year, especially as the spawniug-beds were full of sockeye salmon in 1913 and 1914. There were no log-jams or other obstructions (with the exception of the log-jam two miles from the mouth, which does not, however, cause an obstruction to the fish) to account for the failure of the salmon to run. On arriving at the head of the lake next day, I made camp and inspected the tributaries at this section. The Indian River, which was examined under the most favourable circumstances, was absolutely empty; not one sockeye was visible as we made our way through the rapids to the falls. There were no log-jams or other obstructions to account for the failure; therefore it cannot be due to this cause. The Cheo River, one of the finest spawning-grounds for the salmon, and upon which the early-running sockeye deposit their ova, again showed up very unsatisfactorily in comparison with the large numbers noted on the occasion of my visit there in 1913 and 1914. Very few fish were spawning on the beds as we made our way up to the falls, four miles and a half distant, and only an occasional glimpse of them could be obtained in the deeper portions of the river. No collection of eggs was made possible owing to the agile salmon eluding every effort to spear them. X 22 Report of the Commissioner of Fisheries. 1919 The log-jam at the bend three miles up had not materially changed since my visit last year. It affords no real obstruction unless the falls are opened to permit the salmon reaching the spawning-beds above. If, as has been suggested, a fish-ladder is built through the falls, then it will become necessary to blast out the jam to provide passage to the onward movement of the fish. In making the examination of the Washwash River last year, I drew attention to the huge jam that obstructed the mouth and prevented the salmon reaching the extended spawning-beds above. I am glad to report this year that a portion of the obstruction had been washed away by an extreme freshet and permitted the salmon to reach these beds. The log-jam, however, is a source of danger and may at any time effectually block this river. Proceeding to the falls two miles and a half distant, I noted the comparatively small number of sockeye spawning on the beds, the majority having spawned out, but I was able to obtain a collection of both salmon and scales from this stream. The unsatisfactory state of the spawning-beds of the three rivers suggested that the unavoidable delay in reaching them may have been the cause for the poor results shown, but Indians who had been camping here for the past six weeks stated that even at the height of the run of sockeye, three weeks prior to my visit, very few had come into the rivers in comparison with former years, and that it was only with extreme difficulty sufficient numbers were obtained to supply their winter's need. It is evident, therefore, that the outlook does not favour a big run to this section of the lake four or five years hence. In size the run of sockeye salmon came well up to the average. Returning from the head, I inspected Sunday Creek, a small mountain stream flowing into the lake. In my report for last year I drew attention to the change in the course of the channel; the original one had been obliterated. I found again another change where an extreme freshet had burst its way through, changing the course again and obliterating the channel formed last year. It is to be feared, therefore, no reliance can be placed upon this stream to successfully fulfil its functions year by year. In the Narrows, where formerly large numbers of both sockeye and cohoe salmon were seen on the beds, was found to be lacking of both species on this occasion. The Sheemahant, the largest and most productive river of the lake, and which in past years (with the exception of 1916) had received its full complement of both sockeye and cohoe salmon, failed this year to sustain its reputation; it fell far short of the prolific numbers encountered here in 1913 and 1914. A few sockeye were observed spawning on the riffles as we proceeded up to the falls, but owing to the high stage of the water it was difficult to gauge with any degree of accuracy the real number actually on the spawning-beds; there was no room left for doubt, however, that the run this year has dropped to approximately 25 per cent, of the 1913 and 1914 runs. The few sockeye salmon observed were all fine specimens of the race and came well up to the average. No log-jams obstruct this fine river, which is open to the salmon up to the falls eighteen miles distant. In making the return visit to Jeneesee Creek, I was quite prepared to see a big change in the number of sockeye salmon spawning on the beds, but in this I was disappointed; very few had arrived. The hatcherymen* had obtained one or two boxes of eggs, but from the small number of fish seen in the creek it did not look very hopeful for the requisite number necessary to fill the hatchery being produced. In comparison with the 1913 and 1914, the run of sockeye to Jeneesee is a failure. I found great difficulty in accurately estimating the run of sockeye salmon to the Asklum River owing to the tremendous volume of water flowing down at the time of my visit; it was necessary to go right up to the falls, and then work our way down the sides of the river, before an opportunity was afforded to view the spawning-beds with any degree of accuracy. A fair number were observed spawning on the riffles, and should provide for a fair run returning here four or five years hence. In examining several of these fish I found they were not yet ready for spawning, and indicated their early arrival from the lake. They were fine specimens of the sockeye race, and here again came well up to the average of the other rivers. In estimating the run I have to record a decrease to approximately 50 per cent, of the 1913, 1914, and 1915 runs. Quap River, in comparison with former years, is a complete failure. Calling in there on my way up the lake, I noted a few sockeye spawning on the beds below the fence and a small number had reached the pen; concluding that this w-as only a forerunner of a big run which has never failed so far to eventually make its appearance, I was certainly surprised to find on my return 9 Geo. 5 Spawning-beds of Rivers Inlet. X 23 that little improvement had been shown. The hatchery officials had met with little success in their endeavour to fill the hatchery and had barely obtained 1,000,000 eggs. As the capacity of the hatchery is about 14,000,000 eggs, it is obvious that unless a big run entered very soon their efforts would be doomed to disappointment. Such a situation has occurred, so I am given to understand from information courteously supplied by the hatchery officials; only about 3,000,000 eggs were secured by them. The conditions above the fence have not changed in any way; huge trees and log-jams which obstruct the river right up to the falls are still in evidence, and will entail heavy expense in removing to permit the salmon reaching the extended spawning-beds. The Dalley River, situated directly opposite Quay, showed little improvement. Indians who had been up here three or four times during the last five weeks confirmed the lack of sockeye salmon seen on the occasion of my visit there, and reported that they did not remember having seen so few fish spawning on the beds. This has in former years been regarded as one of the most productive streams of the lake. In 1913, and especially in 1914, the sockeye salmon were seen in dense masses. It is to be feared that this river is also a failure. No log-jams interfere with the progress of the salmon up-stream. The hatchery creek appeared to be well stocked with sockeye salmon, the fine spawning- beds showing to great advantage; the bars outside were thickly covered with spawning fish and in size attained a high average. The facilities for erecting a fence to obtain eggs for the hatchery are by no means favourable owing to the raging torrent of water which comes down after a heavy rain; consequently the large numbers of spawning sockeye cannot be utilized by the hatchery for artificial propagation, the experience of past years precluding the erection of a fence. All along the shores at the mouth of the lake the sockeye salmon were observed in very large numbers; also the fine spawning-beds lying around and close to the Indian rancherie were thickly covered, and for the first time showed up in favourable contrast to the poor run of sockeye salmon to the other tributaries of Lake Owikeno. Spring salmon were breaking water all the way down through the rapids of the Owikeno River and compared very favourably with former years. I noticed the cohoe salmon repeatedly breaking water as we proceeded back from the lake; one or two had been caught in the pen at Quap River and indicated by their appearance early arrival from salt water. I have no doubt that later on the spawning-beds will receive their full complement of this species of salmon. In reviewing the results of the inspection of the Rivers Inlet watershed, I am of the opinion the serious shortage of sockeye salmon disclosed by a visit to these beds, and amounting to approximately 25 per cent, of the 1913, 1914, and 1915 runs, will have a correspondingly serious effect on the number of adult sockeye which will return from this season's spawning. It may be recorded here that an exceptional run of humpback salmon invaded the inlet during the sockeye-fishing season and permitted some of the canneries to put up large packs. The same peculiar conditions were experienced in other sections of the district, notably at Bella Coola and Kimsquit, where abnormal numbers were caught by the fishermen, who made big money in consequence, some of them making between $3,000 and $4,000. In conclusion, I wish to express my indebtedness to G. C. Johnston (manager) and J. Dawson, of Rivers Inlet Cannery; Captain Hamer (Superintendent) and J. Reid (Foreman) ; and the men at the various spawning camps. I have, etc., Arthur W. Stone, Fisheries Overseer. Rivers Inlet, B.C., November 10th, 1918. X 24 Report of the Commissioner of Fisheries. 1919 THE SPAWNING-BEDS OF SMITH INLET. Hon. Wm. Sloan, Commissioner of Fisheries, Victoria, B.C. Sir,—In obedience to instructions from the Department to inspect the spawning-grounds of Smith and Rivers Inlets respectively for the year 1918, I have the honour to submit the following reports:— Proceeding to Smith Inlet to undertake the inspection of this watershed first, I arrived there on September 22nd, and after making arrangements with Indians started for Long Lake the following day. Bearing in mind the extraordinary number of sockeye salmon seen on the spawning-beds at this point in 1914, it was certainly surprising that the run this year did not come up to expectation, and that the two canneries operating here were only able to put up a total pack of about 14,000 cases of sockeye salmon between them. Comparing this with the total pack of about 28,000 cases put up by the Smith Inlet Cannery in 1914, it will be seen there must be some explanation for the decrease shown. I am of the opinion, after closely reviewing the spawning- beds, that the poor results are due to the following reasons: A decided decrease 'In the number of sockeye salmon reaching the spawning-beds following a comparatively poor run of fish up the inlet. In comparison with the exceptionally big run which invaded the inlet in 1914, and correspondingly large numbers reaching the beds in that year, it was a distinct disappointment—i.e., if the four-year fish were the predominating factor in that year—to find the beds had not received their complement of spawning fish. Another reason may be given in the extra number of gill-net fishermen empowered to fish this inlet in comparison with former years; their nets, which were distributed over the inlet in large numbers, disorganized the schools of salmon proceeding up the inlet, and instead of schooling up at the seining-grounds went right through and on to the lake. I am given to understand from the seine-net fishermen that only one fairly big haul was made by them during the whole of the run. The inspection of the spawning-grounds was made under very favourable conditions, fine weather prevailing all the time, and the rivers which last year precluded accurate observations, due to exceptionally bad weather, were low and permitted me to observe them without interruption. The Docee River (the overflow to the lake) was well stocked with spring salmon; a very large number were seen in the clear-running water making their way up; also the fine gravel spawning-beds at the mouth of the lake contained large numbers. I am able to record a substantial increase in the number of spring salmon in comparison with former years. At no previous inspection have I noted so many cohoe salmon as on this occasion; schools were observed swimming around and intermingled with the springs, which from their healthy- looking state proclaimed their early arrival from salt water. Indications pointed to the spawning-beds receiving tlieirfull quota of eggs from this species of salmon. The run of sockeye salmon to Quay Creek, a small stream situated about seven miles from the mouth of the lake, although inspected under the most favourable conditions, showed a falling- off, and presented a great contrast to the numbers seen here in 1914. Schools of these fish were swimming around outside in the shallow waters of the lake, which no doubt will be the means of plentifully seeding the beds later on. No exceptional brood, however, may be expected from the adults which will eventually return to this stream from this season's spawning. The run may be compared to the number reaching this stream in 1915. Proceeding to the head of the lake, we made camp at the mouth of the Geluch River. In making our way. through the rapids to the falls later, it was satisfactory to note that the reputation of this fine spawning stream, although not bearing fruit to the same extent as in 1914 and 1915, nevertheless received its full quota of spawning sockeye, which I estimate came within 25 per cent, of the number seen on these beds in the years previously mentioned; they were fine healthy-looking specimens of the race and compared in size to the 1914 run. 9 Geo. 5 Spawning-beds of Smith Inlet. X 25 Each riffle appeared to be thickly covered until we reached the spawning-beds just below the falls, where the few sockeye observed in this portion of the river presented a vivid contrast to the dense masses which lined the beds in 1914. No log-jams or other obstructions impeded the progress of the salmon up-stream, A few small grilse were observed in this river for the first time in my inspection of these beds. I was very disappointed with the run of sockeye salmon to the Delabah River. In 1914, it will be recalled, large numbers were observed both in the river and in the shallow waters of the lake outside; great stress was laid upon the unsuitabillty of these beds for the propagation of salmon spawn, and suggested the benefit a hatchery would provide to counteract the all too apparent wastage which was manifest in that year. The relatively small run of sockeye which reached the.beds bears a great significance in relation to the warning given at that time. In making my way up to the falls fair numbers were noted spawning, but did not by any means reach the high standard attained in 1914; the beds should be fairly well seeded, however. In size the sockeye were fine specimens of the race and compared very favourably with the 1914 run. Stretching from one end of the bay to the other, the fine gravel spawning-beds of the lake presented a very empty appearance in comparison to the milling mass of sockeye noted here in 1914; beyond one largo school which was seen waiting at the mouth of the river, there was no indication that the remarkable run would be repeated this year. In making a comparison, I estimate the run closely resembles that of 1915, or only about 25 per cent, of the total run to this section in 1914. No log-jams or other obstructions interfere with the free movement of the salmon in this river. Returning to the mouth of the lake, the cohoe salmon were observed breaking water frequently, but none so far had reached the spawning-beds. A late run of these fish appear to have entered Smith Inlet only after the canneries had ceased operating, with the result that some fishermen who had made big hauls were unable to market their fish. One or two canneries from Rivers Inlet, however, came to their assistance later. The run of humpback salmon was very satisfactory and permitted the canneries to put up an average pack. In summarizing the results of the inspection at this watershed, I am of the opinion that the adult sockeyes which will return from this season's spawning will by no means approach the vast numbers seen on the beds in 1914, but should result in a run equal to that of 1915. In regard to the run of cohoes, humpbacks, and spring salmon, I have reason to believe that the favourable impression formed this year will result in a good run from each of this species of fish on their return as adults. In conclusion. I wish to express my indebtedness to G. F. Harris, Manager of the Wallace Fisheries at Smith Inlet. I have, etc., Arthur W. Stone, Fisheries Overseer. Rivers Inlet, B.C., November 10th, 1918. X 2G Report of the Commissioner of Fisheries. 1919 CONTRIBUTIONS TO THE LIFE-HISTORY OF THE SOCKEYE SALMON. (No. 5.) By Charles II. Gilbert, Ph.D., Professor of Zoology, Stanford University. I. SPAWNING SOCKEYE COLONIES IN THE TRIBUTARIES OF THE FRASER RIVER DURING THE SEASON OF 1918. In our report for the season of 1916 we presented for the first time evidence derived from an investigation of the salmon which frequented separate spawning areas of the Fraser River. Those which were found spawning in Morris Greek were examined separately from those that spawned at the Harrison Rapids. And in like manner independent studies wrere made of the spawners at the Harrison Hatchery, in the Birkenhead River at Pemberton, in Pitt Lake, in Cultus Lake, and in the Chilcotin River. The object of the investigation was to throw light on the much-mooted question as to the return of the mature spawning salmon to that particular part of the river-basin in which they had been hatched and reared. In the answer to this question a number of possibilities must be considered. (1.) There may conceivably be no tendency for a salmon to return to any definite spawning district, or to any geographically distinct group of spawning^ districts. In such case any sockeye entering the Fraser would be as likely to spawn above as below the canyon, and as likely to spawn in any one tributary as in any other. Every spawning-bed in that event would contain individuals which had been hatched and reared in widely different portions of the basin. Each spawning area would draw indifferently on the common stock of migrants, and the salmon frequenting it would necessarily agree in their characteristics with those frequenting each of the other spawning areas throughout the basin. There would be no basis for differences of any description. The salmon spawning in different parts of the basin would agree in their average size, as adults, and in their early history as fingerlings migrating seawards. They would agree in the relative numbers of four- and five-year fish, in the relative numbers' of those which had spent one year or two years in fresh water after hatching, or had proceeded to sea as soon as they were free-swimming. All the groups and all the variations of characters which we find in the main run examined off the mouth of the river would be repeated in approximately equal proportions on every spawning-ground. Such would inevitably be the case if the salmon passed indifferently to the various spawning areas without reference to their native habitat. On the other hand, if differences are found to exist in the populations of distinct spawning districts, this must be considered direct evidence that segregation has occurred, based on some principle which has determined that certain individuals in the common run pass to one spawning area while other individuals go elsewhere to spawn. The only principle adequate to explain such segregation, if It generally occurs, would be the return of fish to spawn in their native tributaries. This principle obtains, as we have abundantly demonstrated, as between different river- basins, even when these are of very limited size and have their outlets to the sea in close proximity to each other. In their passage to these minor streams along the salt-water channels the schools of fish are often very intimately mingled, and they separate only as the river-mouths are approached. Thus, iii the vicinity of Namu, on Fitzhugh Sound, gill-nets operated along shore capture in about equal numbers at the same time sockeyes which are bound for Bella Coola and those which are bound for Kimsquit. These can be distinguished by characteristic scale-markings. The two schools are running intimately mingled along the same channels. Where they separate has not been determined. The greater part of them may pass up Burke Channel together, or it may be the Kimsquit fish may take the Dean Channel. But by the time Bella Coola and Kimsquit are reached complete separation has occurred. The case wTould seem not greatly different from that of fish bound for different tributaries of the same river. Conceivably, the schools may be mingled along the main channels of the river and become separated as their respective tributaries are reached. If the populations of these tributaries show distinguishing characteristics, no other satisfactory explanation presents itself. 9 Geo. 5 Life-history of Sockeye Salmon. X 27 (2.) We must consider also the possibility that a partial segregation of spawners occurs, but not a total one; that within certain limits indifference exists as to the precise spawning- ground frequented. Thus the up-river fish may predominately run early and proceed above the canyon, passing without selection into any stream that becomes available; while the late-running fish may enter the Harrison, the Pitt, and Cultus Lakes again without determinate destination. „That this supposition is without validity for the lower river becomes evident from the fact previously emphasized that the fish entering Pitt Lake are so widely different from those entering Cultus Lake, and that even the different spawning-beds of the Harrison can usually be shown to possess individual characteristics of unquestioned significance. With regard to the river above the canyon, the case is not so clear, partly, no doubt, from lack of adequate material. The runs of the upper river have of late years become extremely attentuated. No material at ail adequate has been obtained from any up-river tributary except the Chilcotin, and because of permanent depletion we may now find It too late to secure any. A small run which makes it possible for natives to secure a few fish for food from the main channels of the up-river may yet disappear for all practical purposes when it becomes widely distributed among the tributaries. And unless material can be so secured after final segregation has occurred it is useless for our purposes. The differences which distinguish fish spawning in contiguous areas are usually of such small amount and compass that they become wholly masked when two or more races are mingled. This may be the case even at the mouths of the main tributaries. More than one spawning district, each with its special characteristics, may be found within a given tributary. The Harrison River, for example, has at least four spawning areas. Scales and other data collected solely at the mouth of the Harrison, or in its lower course, would be valueless. The different strains would be mingled and their characteristics obliterated. The nearer the material is obtained to the final spawning-grounds, the greater its value for investigations of racial divergence. (3.) Where spawning populations show distinguishing characteristics the inference is clear and unavoidable. A physiological barrier has existed—the racial habit of return at maturity to spawn in the native stream—for so long a period that minor differentiations have developed and a strain or sub-race has been formed. But what are we to conclude in cases where the colonies of two distinct spawning areas present no characteristics by which we can distinguish them? Obviously, we must infer either (a) that segregation in their case has not occurred, or (b) that it has occurred but has not been effective in producing divergence between the colonies. If the first of these were true, it would signify that the " home-stream " instinct was in most cases rigidly operative, while in one or more instances within the same river-basin it was wholly in abeyance. The second supposition would seem a priori far less improbable. Complete segregation of spawning fish may occur, so that each individual returns to the stream in which it was hatched and reared, and yet, in certain instances, separate colonies may have failed to develop distinct methods of growth or of habit by which they can be distinguished. Failure to discover distinguishing characteristics between populations of separate spawning areas need not indicate then that their progeny will fail to return at maturity each to its native stream. This process may have been in operation for many thousand years. The two colonies may have been wholly distinct and self-perpetuating for a very long period and yet no differences have developed. If we should adopt distinctive marks for the young from each of the two streams on their downward migration to the sea, it might be shown at maturity that each stream contained spawners with a single mark only, the one that had been applied to the young from that stream. Vet the adults from the two tributaries might be otherwise indistinguishable. Negative evidence in this case proves nothing. But if affirmative evidence is found, it is conclusive. If, in fact, differences are found to exist, there is no alternative to our acceptance of the home-stream theory. Different tributaries of the same river may be quite unlike in the sharpness of the characteristics by which their colonies may be recognized. Cultus Lake, a tributary of the Lower Fraser, possesses a strongly marked race of small salmon, pale in colour of flesh, poor in oil, scarce X 28 Report of the Commissioner of Fisheries. 1919 worthy of the name of sockeye; while Pitt Lake, also tributary of the Lower Fraser, is frequented by a large race the opposite in all respects of the Cultus Lake colony. But we have been unable to distinguish the fish of Morris Creek, tributary to the Harrison River, from those spawning in the lake-gravels in front of the Harrison Hatchery and in the small stream that supplies It. In this latter case we have obtained no direct evidence of segregation; but it is rendered highly probable by the more numerous instances in which it can elsewhere be demonstrated. The most important investigation during the season of 1918 consisted in an attempt to test the conclusions which we had based on the material of 1916 by a re-examlnation of the colonies frequenting the same spawning-grounds previously reported on. "It was hoped to secure,' also, material from additional spawning-grounds above the canyon, from which nothing heretofore had been obtained. The season of 1918 was peculiarly ill-suited to the purpose. In the recent history of the Fraser we have grown accustomed to noting that each successive year is the worst in the history of the industry, or is in any case distinctly worse than its representative in the last cycle, four years previously. But the eNtent of the disaster In 1918 was unexpected and overwhelming. A run of sockeyes to the Fraser can hardly be said to have occurred. Under such circumstances the numbers that escaped to the spawning-grounds were most limited. Along the main river- channels were a few locations, like the canyon above Yale, the Bridge River fishing-grounds above Lillooet, and those above the mouth of the Chilcotin at the Chimney Creek Bridge and at Soda Creek, where natives by persistent dipping secured a few fish. Material was obtained at all these points, but for reasons already stated it was of less value for the purposes of the present investigation. In all these river-channels it was apparent a number of local races were mingled. Because of this the range of variation in all the characters examined was excessive, far greater than when we have before us data obtained from a single spawning area. In addition to these main-river locations, we were successful in securing a limited amount of material from the Chilcotin River at Hanceville and at Fish Canyon near its mouth; from the Birkenhead at Pemberton; from the Skookumchuck above Harrison Lake; from the Harrison Hatchery, Morris Creek, and the Harrison Rapids; and from Pitt Lake. For much of this material we have to thank the very generous co-operation of Alex. Robertson, Superintendent of the Harrison Lake Hatchery, and T. W. Graham, Superintendent of the Pemberton Hatchery on the Birkenhead. To both of these gentlemen we extend our grateful acknowledgments. In general, it can be stated that the examination of this material completely supported and verified the conclusions reached in 1916. With regard to the spawning-beds below the canyon, we affirm without qualification that they are as distinctively populated as though they were located in separate streams independently entering the sea. Additional material only emphasizes this fact. Not only were the different colonies in 1918 marked by distinctive characters, but these characters were the same for each colony as those noted in 1916. We have then the picture of a number of self-perpetuating sub-races, each of which has acquired certain minor habits of growth, of migration, of age at maturity, of length of life in fresh water, of colour and quality of flesh—the same characters, in short, which in various combinations characterize the races of the separate river-basins of the simpler sort. A very conspicuous instance of racial differentiation is furnished by the Harrison watershed. We do not know how many distinct spawning areas it may have contained. Several minor colonies, like that frequenting Silver Creek, are now practically exterminated. There were spawning areas doubtless in Lillooet and in Tenas Lakes, but no records of these have reached us. A very limited number of late spawners still frequent the slack water of the main river above Harrison Rapids, the vanishing remnant of a once fine run visits the lake at the head of Morris Creek, and a considerable run still goes up the Birkenhead. Both the Birkenhead and the Morris Creek fish pass over the very spawning-beds at Harrison Rapids, but not one of them tarries there. An examination of 200 specimens, laboriously gathered at Harrison Rapids in 191S by Mr. Robertson, failed to yield a single fish which had been spawned elsewhere than in the beds of the rapids. Not one could on even a cursory examination be confused with those constituting either the Morris Creek or the Birkenhead colony. Not one of the Harrison Rapids fish wras found among over 400 specimens examined from Morris Creek, and over 200 from 9 Geo. 5 Life-history of Sockeye Salmon. X 29 Skookumchuck aud the Birkenhead. No Morris Creek strays were found in the Birkenhead, and no Birkenhead fish were found in Morris Creek, in spite of the fact that all the Birkenhead fish passed the mouth of Morris Creek on their way. In these cases, evidently, the law of segregation is most strictly in force; yet it seems impossible to imagine a locality better suited to straying of the spawning fish with resulting mixture of populations. That it does not occur here, where fish pass over and pass by one another's spawning-grounds, gives confidence that the same law rules everywhere, though the evidence of it may be less compelling. Harrison Rapids.—The most interesting colony of the Harrison is the one which spawns late in the season at Harrison Rapids. The characteristics of this colonyr were fully discussed in our report for 1916, page 50. The nature of the spawning area is in itself highly unusual. The spawning habits of sockeyes are generally regarded as rigidly fixed, more so than with any other species of salmon. Unlike other species, they- must resort to a lake, and fail to frequent any stream, otherwise suitable, which has no lake which they can succeed in entering. The lake once attained, they spawn in the gravel-beds around its margin or ascend some affluent of the lake in search of suitable spawning-beds. The young develop mainly, if not exclusively, in the waters of the lake, frequenting the depths in the daytime, but rising to feed at the surface in the evening. Under such conditions the sockeye commonly resides over a year in the lake, sometimes over two years, occasionally over three years, before seeking salt water. Of the varying number which precociously descend to salt water in their first spring, few ordinarily survive and return at maturity. But the Harrison Rapids spawners use gravel-bars in a shallow backwater region of the river, where they have no genuine lake conditions at their disposal, nor any lake into which the young can drop back after hatching. They have adopted, therefore, the highly exceptional method of life for a sockeye, of dropping down to salt water as soon as they are free-swimming, and while we have no data as to their mortality or the percentage of survival, it is clear that before man disturbed the balance of nature, enough survived to keep the colony in a flourishing condition. More than this cannot be said for any of the other colonies. Clearly, then, under the spur of necessity, a sockeye colony can adapt itself to the habits of sea migration in the early free-swimming fry stage, a habit rare in sockeyes, and so far as known never occurring among cohoes; hut almost exclusively characteristic of humpbacks and dog-salmon, and frequent among ehluooks. The 200 specimens of Harrison Rapids spawners examined in 1918 had all had the history above outlined. Not one of them exhibited in the centre of its scale any trace of growth of the young in fresh water. All had passed down to sea as young fry before any portion of the scales had made their appearance. It is very difficult to place satisfactorily on record the evidence for racial differentiation as observed on the spawning-beds. As previously pointed out, the scales have suffered extensive erosion during the later weeks which precede the spawning. Not only are the margins of the scales broken down and absorbed to such an extent that they no longer serve for determination of age, but the surface of the remaining portion of the scales has become more or less defaced. Absorption has taken place not only at the margins but also on the surfaces of the scales, and the lines of growth are blurred or obliterated. Enough may be preserved to render possible a detailed examination under the microscope, where the course of the ridges can be traced from such fragments as remain, by the aid of focusing at different levels. But such scales produce very unsatisfactory photographs. Nevertheless, it has seemed advisable to have a number of these from each of several spawning areas of the Fraser reproduced as a permanent record. Only the centres of the scales are figured, the worn and jagged edges being omitted. Figs. 1 to 5 are from the spawning race at Harrison Rapids. Each contains the extensive first year's growth which occurs wholly in salt water, together with a varying portion of the second year's growth. The boundary between the two years in these cases is not a sharp one, leading to the conclusion that fry which descend to the sea in their first year suffer less interruption to their growth during their first fall and winter in the sea than is the case with those which pass their first year or their first two years in fresh water. Rarely, however, a slackening of growth in the Harrison Rapids fish at the close of their first year is indicated by a definite line of demarcation at its outer edge. When this exists there are thirty-two to forty rings or lines of growth belonging to the first year. X 30 Report of the Commissioner of Fisheries. 1919 As was observed in 1916 and during previous years, sockeyes belonging to the type which alone frequents the Harrison Rapids spawning-beds—the sea-type—make their appearance during the latter part of the run only, at the mouth of the river and in the approaches to it. In 1916 the firs* to appear were taken on July 17th, and after that date they occurred constantly to the close of the season. In 1918 the first to make their appearance at Esquimalt, from traps along the Vancouver Island shore, were found on July 14th. Not a single specimen had been found among the 250 which had been previously examined, distributed over June 17th, 20th, and 28th, July 2nd and 10th. They were present on July 14th, 18th, 21st, 25th, 29th, August 2nd, 4th, 8th, 11th, and 26th. The only dates on which examinations were made at Esquimalt after-July 14th that yielded no individuals of sea-type were August 13th, 16th, and 23rd. The numbers of this type were evidently decreasing during the very last of the run. The defective condition of the scales made it impossible to determine directly the age of the Harrison Rapids fish, but the lengths were secured, and a valuable clue can be obtained from the range in size found in three-year fish and in four-year fish of this type before they enter the river. The complete range may not be available because of the small number of individuals of sea-type which were obtained at Esquimalt during 1918. Only twenty-four in all were observed. Five were three-year males; one a three-year female; five were four-year males; thirteen were four-year females. The three-year males ranged from 20% to 24 inches long; the four-year males from 26 to 28 inches. Comparing these figures with the Harrison Rapids column in Table I., it is seen that the majority of the Harrison Rapids males are from 25 to 29 inches long, and that the lengths of these, when tabulated, form a fairly regular curve, with the mode at 27. Lying below 25 inches, extending from 17% to 24%, are sixteen thinly scattered records, which form no part of the curve referred to. As these two elements in the table agree closely with the observed lengths of three- and of four-year males obtained in salt water where age can be determined, we are justified in concluding that approximately all Harrison Rapids males between 17% and 24% inches long are in their third year, while those between 25 and 29 inches long are in their fourth year. The single three-year female of this type secured from salt water in 191S is 20% inches long, while thirteen four-year females range from 24 to 26% inches. Table II. gives the length distribution of Harrison Rapids females, all but three of which fall within the scope from 23 to 26 inches and form again a regular curve, while three individuals are 20, 22, and 22% inches long respectively. It is highly probable that the three last mentioned are in their third year, while the remainder are in their fourth year. The earlier dates, on which sea-type individuals appeared in the run, produced only three-year males and four-year females, the other groups appearing subsequently. In 1918, as in 1916, we failed to find any individuals of sea-type spawning in any part of the Fraser River basin other than Harrison Rapids. Upwards of a thousand specimens were examined from other spawning districts, in which it might be thought an occasional sea-type individual might be found, if only as a stray. The fact that not one individual was discovered elsewhere, when coupled with the further fact that every Harrison Rapids sockeye belonged to this group, furnishes the strongest possible evidence of the return of spawning fish to their native districts. Morris Creek.—This stream offered in the early days one of the most valuable spawning districts in the Harrison watershed. In his report for 1902, pages 24 and 25, J. P. Babcock states: " Morris Creek and Lake are insignificant bodies of water, but as a spawning-ground for the late run of sockeye, and from the standpoint of artificial propagation, they constitute one of the most important and valuable points on the Fraser. The Dominion Government has operated a spawning-station here since 1885, and with the exception of the year 1900 has never failed to take eggs. All the salmon-eggs taken by the Dominion in this Province up to 1901 were secured at this station." Much later than 1901 Morris Creek still served as the principal source of eggs, first for the Bon Accord Hatchery near New Westminster, and later for the Harrison Lake Hatchery. None of the fry in the early days were returned to Morris Creek, as it was not believed that such procedure was necessary to maintain the spawning run. Even 9 Geo. 5 Life-history of Sockeye Salmon. X 31 in 1902 (I.e.) Mr. Babcock wrote: "It is generally believed that fish bred in a given watershed, such as the Fraser, return to it upon reaching maturity, and there is considerable evidence to warrant it. It has not, however, been settled, and probably never will be, that the fish bred in a given tributary of a large river seek only that tributary to spawn." Believing, then, that any increase in the run to a. large river would equally benefit all the tributaries, there seemed no reason for laboriously returning fry to Morris Lake in order that they should return there and help maintain the spawning run. Knowing, as we now do, that salmon will in general return to the district in which they are liberated, the fate of the Morris Creek run seems to have been inevitable. It has steadily dwindled with the years until it can no longer be depended on for any considerable take of eggs. Natural propagation was reduced to a minimum in order to obtain eggs for the hatcheries, while the hatchery-reared fry did not at maturity return to Morris Creek. The same has been the history of Silver Creek, where " the Dominion Government placed a weir in August, 1902, and took nearly two and a half million eggs, which were transferred to the hatehery near New Westminster." The run in Silver Creek is now practically extinct, and we cannot doubt that this process has been hastened by failure to replenish its run through the planting of fry. No better examples than these can be found of the necessity of working out completely the entire life-history of our commercial fishes before it is possible to propagate them with success or to legislate wisely for their protection. Failure to follow this principle has discredited hatchery-work from the beginning, and has led to the well-founded suspicion that in many instances they have been more of a detriment than an advantage to the runs. Morris Creek would seem to have been an ideal hatching and rearing ground for the sockeye. The enormous run entering this very insignificant stream in early days furnishes evidence of extraordinarily favourable conditions. Examination of the scales of the fish now running shows a striking uniformity in their development. They form an impressively homogeneous lot. All have large sharply defined nuclear regions, testifying to a vigorous growth during their first year, which is uniformly passed in the lake. The number of nuclear rings varies from fifteen to twenty-eight, with the mode at twenty. The frequency curve is given in Table III. In all probability the fingerlings on migrating from the lake in the spring of their second year were 4 to 6 inches in length, and better fitted to cope with their enemies than the smaller weaker yearlings from less favourable localities. Figs. 6 to 8 give typical centres of Morris Creek scales. No individuals from this district had lived two years in the lake before migrating, and none had proceeded to sea in their first year. The length distribution, given in Tables I. and II. in the columns for Morris Creek, seems to indicate that the majority of the fish were in their fifth year. The modal length for the males at 27 inches and for the females at 25 or 26 inches closely agrees with that usually characteristic of five-year.Fraser River fish. Birkenhead River.—This affluent of the Harrison has now the most reliable sockeye run, and is the only spawning district of value remaining in this watershed. Its present importance is due in part doubtless to the fact that it was not drawn on for hatchery purposes until a much later date than Morris Creek and the Harrison Lake region, and in perhaps larger part to the significant fact that the output of the hatchery is constantly planted in the Birkenhead and helps maintain the run. The characteristics of the Birkenhead race in 1918 are in general the same as those described for the 1916 run. The growth of the fingerlings in fresh water must be much less than in the case of the Morris Creek and Harrison Lake fingerlings. The nuclear area of the scale averages small, with densely crowded rings. These were somewhat more numerous in 1918 than in either of the two years preceding and the extremes range far more widely. The lot in 1918 was far less homogeneous than in 1916 and 1917, even those taken at the Pemberton Hatchery presenting an unexpected amount of variation. While the nuclear regions average small and the number of rings in the great majority of individuals is less than sixteen (the mode lying somewhere between eight and eleven), occasional specimens have nuclear rings running as high as nineteen, twenty, and twenty-one. Furthermore, the frequency curve for nuclear rings in 1918 has none of the regularity so well defined in 1916 and 1917. The irregularities in 1918 are equally marked and X 32 Report of the Commissioner of Fisheries. 1919 are similar, whether the specimens were procured at Skookumchuck or at the Pemberton Hatchery on the Birkenhead. As seen in Table III., there is at both localities a mode at eight rings, one at eleven, and a less regularly defined one at fifteen. Yet, so far as known, all come from the same spawning-grounds. It is not clear what significance, if indeed any, attaches to this circumstance. In spite of the variation in nuclear rings, racial peculiarities are apparent. Especially marked is the small size of the first year's growth in the sea, a character that was also conspicuous in the two years preceding. Occasionally this is carried so far that the first year's growth in the sea resembles a second year in a lake, but only one undoubted two-years-in-lake individual was observed. It was found possible to determine age from the scales in a larger percentage of individuals than was possible in other spawning localities. Both at Skookumchuck in the Lillooet River and at Pemberton on the Birkenhead between 40 and 50 per cent, of all specimens could be grouped by age. This does not, however, give a reliable estimate of the relative numbers of four- and of five-year fish present. With imperfect scales, the age of five-year fish is more readily determined than of four-year fish. Four-year scales cannot be determined unless some portion of the original margin of the scale is preserved, thus demonstrating that a fifth year's growth had not been present when the scale was intact. But five-year scales are unmistakably such in eases where not only the margin but the entire fifth summer's growth has been lost. If the last winter's band of crowded rings is present beyond the growth of the fourth summer, or any part of such band, the fish was evidently in its fifth year. The practical absence of six-year fish of the one-year-in-lake type in the Fraser River facilitates this determination. In 108 specimens from Skookumchuck, the age was ascertained in fifty-four, or exactly one-half. Thirty-four of these were in their fourth year (twelve males and twenty-two females) and twenty were in their fifth year (twelve males and eight females). In 180 specimens from the Birkenhead, age could be determined in seventy-eight, or 43 per cent. Of these, forty-one were in their fourth year (three males and thirty-eight females) and thirty-seven in their fifth year (twenty-one males and sixteen females). If these proportions were reliable, we should have 37 per cent, of five-year fish at Skookumchuck and 47 per cent, at Pemberton, the percentage for both localities being 43. As stated above, this percentage is too high. The minimum percentage of five-year fish can be obtained by assuming that all the undetermined individuals were four-year fish, which reduces the proportion to 19 per cent. Obviously, this extreme assumption cannot be justified. If one-fifth of the undetermined Individuals were five-year fish, there would be slightly over 30 per cent, of five-year fish in the Birkenhead-Lillooet section, and this must be considered a fairly satisfactory estimate. The percentage of five-year fish of the one-year-in-lake "type in the total run, as determined by samples takeii frequently throughout the season at Esquimalt, was 23, and these ran largely in the early part of the season. The percentage during June and the first half of July was higher than that indicated. Pitt Lake.—Over 200 specimens were examined from Pitt Lake, and of these only twenty-five could have their age determined. These proved to be in their fifth year in every instance except one, the latter having spent two years in the lake as a fingerling and having returned from the sea in its sixth year. Reference to Table I. indicates that the sizes of the Pitt Lake spawners were similar to those from Morris Creek and Birkenhead, the majority ranging from 25% to 29% inches, with an average of 27.7 in the males, and from 23 to 27, with average at 25.4 in the females. These averages are in the case of the males much higher than for five-jTear males taken from the main run in salt water (26.3) ; while the average for females agrees exactly with that obtained from five-year females taken in the Vancouver Island traps. It is our belief that practically all of the Pitt Lake run was made up of fish in their fifth year. The lot proved most homogeneous. The nuclear regions represented a fresh-water growth obviously under highly favourable conditions, though as fingerlings they had failed to attain as large a size as did the fingerlings in Morris Creek. The nuclear rings are bold, firm, well spaced, closely parallel, and regular. The number of rings, as seen in Table II., ranges from thirteen to twenty-one, with mode at seventeen, much lower than Morris Creek. One individual has as few as ten rings; another as many as twentjT-three. It is a well-marked race. Figs. 22, 23, and 24 illustrate the nuclear regions of characteristic scales from this region. Like the Morris Creek race, the first year's growth in the sea is large and vigorous. 9 Geo. 5 Life-history of Sockeye Salmon. X 33 Table I.—Fraser River Male Sockeyes, examined on their Spaivning-beds, 1918, distributed by Lengths and by Locality. Inches. 17% 18 . 18% 19 . 19% 20 . 20% 21 . 21% 22 . 22% 23 . 23% 24 . 24% 25 . 25% 26 . 26% 27 . 27% 28 . 28% 29 . 29% 30 . 30% 31 . 3 4 5 14 16 20 22 18 19 10 8 5 5 1 2 3 3 2 2 4 8 24 24 34 11 8 1 1 CC! 033 -J-S 1 1 1 i l 3 7 7 17 13 30 8 16 6 7 3 1 3 3 21 5 9 7 12 4 6 4 1 1 3 7 6 15 7 8 3 2 5 1 Table II.—Fraser River Female Sockeyes examined on their Spawnhig-beds, 1918, by Lengths and by Locality. distributed Inches. 4) c 'Jl U o \\W o it II U OJ a o . 013 013 B2 fi a Em A3 £ Oi 03, Hrl CJ OD CC! CJ fl a o r*i r. s OJ Br. CjrH S cc B o AS O S3 VI — '?o cj cj CJm •jo a . 5" CC O ■S3 20 20% 21 21% 22 22% 23 23% 24 24% 1 2 5 9 10 23 29 40 41 33 35 12 10 3 o i i i 3 5 1 5 1 1 i 1 i l 8 8 20 15 6 4 1 l 12 9 24 17 26 17 6 i l 2 4 12 6 20 14 14 7 9 2 2 1 1 i i 4 2 6 1 12 3 4 1 5 1 2 1 2 i l l 3 2 1 1 1 i l 4 3 5 9. 25 25% 26 26% 27 27% 28 28% 29 X 34 Report of the Commissioner of Fisheries. 1919 Table III.—Fraser River Sockeyes examined on their Spawning-beds, 1918, distributed by Number of Nuclear Rings and by Locality. No. of Rings. 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 11 20 21 27 22 14 10 9 6 3 2 2 CJ Cr, CC! OJ 2 3 6 S 12 11 6 5 3 1 6 13 15 10 19 14 8 7 10 3 as 3 o 9 20 11 9 11 12 6 1 £G The Chilcotin River.—It was with difficulty that any material was obtained from this district. Mr. Babcock and the writer spent August 21st to 23rd at Hanceville and Alexis Creek. One specimen only was seen at Alexis Creek, while at Hanceville, although dipping was constantly iu progress, only eighteen individuals were taken. At Fish Canyon, near the mouth of the Chilcotin, on August 24th and 25th, thirty freshly caught sockeyes were examined, and portions of the skin containing scales were dissected off from sixty-two fish found drying on the racks. In the case of the dried specimens we could obtain no reliable measurements and no record of sex. But the history of the individuals is, of course, found equally recorded on the scales. Such material, in default of better, may be availed of for the present purpose. A cursory examination of the Chilcotin material would be sufficient to demonstrate that it is totally distinct from any of the spawning runs investigated in the part of the Fraser River basin which lies below the canyon. No single specimen from the Chilcotin agrees with the Harrison Rapids colony in having migrated to salt water in its first year. All had spent at least one year in the fresh-water lakes before descending to the sea. Furthermore, a considerable proportion of the Chilcotin fish (from 35 to 40 per cent, of the whole) had spent two full years as fingerlings in their lake. This is a very significant fact, in view of the almost total absence of representatives of this group from the Harrison watershed and the Pitt during the season of 1918. Obviously there is wide variation from year to year in the extent to which fingerlings in the same spawning area will tarry in the lake for a second year. Thus, as has been shown in our report for 1916 (page 51), the two-years-in-lake type was abundantly represented in the Birkenhead, where they constituted 10 per cent, of the run. But only one specimen was found in 1918 among 288 examples from Skookumchuck and Pemberton. Only 6 per cent, two-years-in-lake fish were found in the Chilcotin in 1916. It seems evident that the occurrence of this type in some abundance is more probable within certain spawning areas than in others; but extensive fluctuations occur in successive years. From such evidence as is before us, it would seem that in 1918 the members of this group were exclusively bound for the Chilcotin, and perhaps other spawning-grounds above the Yale Canyon. Their distribution 9 Geo. 5 Life-history of Sockeye Salmon. X 35 in the main run becomes then a matter of special interest. Referring to our Esquimalt records, we find this class sparingly represented during the early part of June, but the number increasing during the month, until on June 2Sth it constituted 14 per cent, of the whole. Up to this point the history of 1918 exactly parallels that of 1916. On July 2nd they constituted 12 per cent.; July 10th, 3.5 per cent.; July 14th, 9 per cent.; July 18th, 9 per cent.; July 21st, 11 per cent.; July 25th, 11 per cent.; July 29th, 13 per cent.; August 2nd, 5.5 per cent.; August 4th, 5.5 per cent.: August 8th, 5.5 per cent.; August 11th, 2 per cent.; and subsequent dates occasionally with one specimen, but usually without any. The striking difference between 1918 and 1916, as regards the run of the two-years-in-lake type, lies in the heavier early run in 1918 and the almost total disappearance of these forms in the latter part of August. Oil August 10th, 1916, 10 per cent, and on August 27th 10 per cent, of the run belonged to this class. It seems highly probable that the numerous late-running individuals in 1916 were bound to streams below the canyon. The material obtained at Fish Canyon did not differ in any respect from that obtained higher up-stream at Hanceville. The nuclear areas of the scales are characteristically marked with slender rings, not greatly crowded, nor widely spaced. The rings are fairly regular and in general parallel, but the individual rings are frequently broken and interrupted. Figs. 14 fo 20 present characteristic centres of Chilcotin scales, the majority of these from the two-years-in- lake form. The first year's growth in the sea usually averages small, with the scale-rings often crowded and irregularly spaced, indicating partial checks in the middle of the growing season. Variations are found in this regard, as is always the case, but the general picture is of a single spawning type, displaying not more than the usual amount of variation. Two forms can be roughly distinguished, one with large, the other with small nuclear regions, and these two exist among those individuals that spent two years in the lake as well as among those that migrated seawards in their second spring. No other differences have been detected between the two groups. Those with the larger type nucleus may have been the first to hatch. Intermediate individuals are found. It was impossible to determine the age in the majority of the specimens. In thirty-three out of a total of 108 the scale was complete at some point on the margin, and made it possible to ascertain in each case that the fish when captured was in its third year since entering the sea. Those of the one-year-in-lake type were all in their fourth year; those of the two-years- in-lake type were in their fifth year. Both types had spent an equal period on the sea-feeding grounds and had reached essentially the same size. A comparison of length-frequencies of Chilcotin fish (Table I.) with those of Pitt Lake, Morris Creek, and the Birkenhead will show the much smaller range in size of the Chilcotin material. We failed to find evidence of five-year fish (one-year-in-lake type) in the Chilcotin, either in 1916 or in 1918; or of six-year fish (.two- years-in-lake type), although numerous examples of these six-year fish are known to have entered the mouth of the river in June and July. The Quesnel River.—No material was obtained from the Quesnel River. Mr. Babcock and the writer visited the dam and flshway at the outlet of Quesnel Lake on September 3rd and 4th. No salmon had been observed to enter the fishway, and only two or three which we did not secure were seen in the pool at the foot of the race. No run was reported at any later date. Specimens were secured, mainly from drying-racks, at Chimney Creek Bridge and at Soda Creek, points on the main river between the mouths of the Chilcotin and the Quesnel. It might be assumed with some justice that the majority, if not all, of these fish were headed for the Quesnel. Along the river between Quesnel and Fort George no success appeared to attend fishing at any point. Here aud there we could learn of the capture of a few strays, but that was all. Yet it Is hazardous to adopt conclusions based on evidence of this nature. Some proportion of the Soda Creek and Chimney Creek fish were undoubtedly bound up-river beyond the Quesnel. What the relative numbers were it is impossible to estimate. The same kind of evidence that would convince us no fish were running in the upper river can also be adduced to demonstrate that none entered the Quesnel. The material secured at Chimney Creek Bridge and at Soda Creek was clearly different from that obtained in the Chilcotin. For one point of distinction, there were few individuals of the two-years-in-lake type. Out of a total of seventy-one specimens, only five, or 7 per cent., were of this class, which constituted 35 per cent, of the Chilcotin fish. This abrupt change from X 36 Report of the Commissioner op Fisheries. 1919 the mouth of the Chilcotin in the few miles to Soda Creek could have only one meaning. A distinct body of fish, developed from fry which had been hatched in the Chilcotin, had entered that river. Other minor differences marked the nuclear portions of the scales from fish taken above the mouth of the Chilcotin—differences which hardly lend themselves to description, as they appear in the ever-varying material, but nevertheless in a composite image are found to have a certain distinctness and an unquestioned significance. The nuclear rings are on the average more closely crowded, while the number of rings remains about the same, ranging widely from seven to twenty-one, with the mode at ten. The first year's growth in the sea is small, and the individuals are all of those sizes which characterize Fraser River sockeyes that have spent three years at sea. The condition of the scales permitted us to ascertain the age of but one specimen, a four-year fish, but we entertain no doubt that all the individuals observed had spent three years at sea, and were either in their fourth or fifth year, depending on whether as fingerlings they had remained one year or two years in fresh water. Fig. 21 presents for comparison the centre of the scale of a Soda Creek specimen. Fraser River Canyon above Yale.—Fifty specimens were obtained from natives dipping in the canyon and above, as far as Lytton. Inasmuch as it is supposed no sockeyes entered the Thompson in 1918, and the Seton-Anderson section was wholly bare of spawning-fish, it would be natural to assume that sockeyes captured in the canyon above Yale were all bound for the Chilcotin and the Quesnel, and would wholly resemble the fish captured at Fish Canyon and Chimney Creek. Nothing could be farther from the truth. The Yale Canyon fish averaged much larger than the up-river fish, and contained a considerable proportion of five-year individuals, of which not one was detected above. Apparently some 30 per cent, of the fish dipped in the canyon were five years old, a much larger percentage than was at any part of the run entering the mouth of the river. The destination of these five-year sockeyes is a question of much importance. The only direct evidence we have is derived from two specimens, the only material obtained from the Thompson River—one taken two miles east of Lytton on the Thompson; the other at Thompson Siding, nine miles east of Lytton. Both of these are five years old, one 24 the other 26 inches long, and both females. Only four of the fifty are of the two-years-in-lake type, so we could not be dealing to any considerable extent with Chilcotin material. Furthermore, many of the nuclear regions of the scales are of the robust type, with strong, bold, parallel ridges, quite different from the delicate slender-ringed nuclei of the upper river. Figs. 25 to 28 illustrate certain of the types of scales characterizing the sockeyes taken in Yale Canyon and above, while Fig. 29 is from the Thompson. We. can apparently not escape the conclusion that a large proportion of these canyon fish, taken during the month of August, were bound up the Thompson River. Certainly they were not bound for the Chilcotin or the upper district. Although few in number, they were concentrated in the canyon, but when they reached the quiet waters above the canyon, and afterwards scattered over the immense Shuswap-Adams district, they were wholly lost to sight. If this inference is correct, it gives us the first glimpse of the Thompson race or races. The suggestion that these fish, like those spawning below the canyon, may in considerable numbers attain five years of age is an interesting one, and must be taken into account in all discussions of the four-year cycle and the striking variation in the proportion of the age-groups in the Fraser River. Bridge River.—No material was obtained from the stretch of the Fraser between the mouth of the Thompson at Lytton and the outlet to Seton Lake at Lillooet. Above Lillooet, at Bridge River, fifty specimens were obtained from August 21st to September Sth, and proved on examination to contain a mixture of the type found in the Chilcotin at Fish Canyon, together with that secured above the Chilcotin in the main Fraser at Chimney Creek Bridge and at Soda Creek. The Chilcotin type predominated. Twenty-six per cent, were constituted of two-years-in-lake fish. The sizes ran low and the entire assemblage was markedly different from that taken at the same period and earlier in the month of August in the Yale Canyon. One specimen, a male 27 inches long, was in its fifth year, but all the others were of smaller size and apparently four-year fish. The nuclear rings of those of the one-year-in-lake form ranged from twelve to twenty-one, the smaller sizes predominating, and largely concentrated between thirteen and seventeen; the mode at sixteen. Those of the two-years-in-lake type have the rings of the first year ranging from six to eleven; those of the second year from twelve to twenty. 9 Geo. 5 Life-history of Sockeye Salmon. X 37 II. THE FRASER RUN OF 1918. In our report for 1916 we remarked: " The run of sockeye salmon to the Fraser River in 1916 was the poorest known." In 1917 we were obliged to record the disappearance of the big year of the cycle from Fraser River runs, with no prospect of its recovery. And in 1918 the run reaches still lower dimensions than in 1916, the total catch on Puget Sound being 50,617 cases and in Canadian waters 20,955 cases, a total of 71,572 cases. This is but six-tenths' the catch of 1916, the poorest previous year. The brood-years for 1918 were 1913 for the five-year fish and 1914 for those in their fourth year, the four-year fish largely predominating, as is customary in the Fraser River. 1.913 was the big year of that cycle, but big years are now known to owe their magnitude to four-year fish hatched in the upper waters of the Fraser, in tributaries which produce few or no five-year fish. The tributaries of the Eraser which are responsible for five-year fish lie below the Yale Canyon, with the possible exception of the Thompson River, and have never experienced large increases in the size of the run during the big years of the cycle. W-e would have no reason to expert, therefore, any unusual influx of five-year fish in 191S because of its relation to 1913. On the other hand, the 1913 run to the Lillooet Lake section, which supports the Pemberton Hatchery on the Birkenhead, the most important spawning stream of the Harrlson-Lillooet section of the river, was distinctly poorer than in any big year since the hatchery had been established there. The history of 1913 gave no grounds for hope of any large number of five-year fish in 1918. The four-year fish were hatched from the 1914 brood stock, which would be considered of more than average size for an off-year if we relied on the figures for the commercial returns (555,557 cases). On inspecting the reports from the spawning-grounds presented by Mr. Babcock for 1914, we find, however, that serious obstructions in the canyon above Yale prevented the ascent of the salmon during the greater part of August, and hampered them greatly at other portions of the run. The result was what might have been anticipated. The numbers that reached the up-river spawning-grounds were everywhere greatly reduced. Fewer sockeye reached the Chilcotin than during any other recorded year. The run at Quesnel Lake was one of the poorest since the fishway was built in 1903, the total number that entered the lake being too small to make any noticeable showing there. Only a few hundred reached the Seton-Anderson Lake section, and the run to the Shuswap-Adams District was distinctly light. There was a fair run to Lillooet Lake and a rather small run to the Harrison. This review of the situation is given to show how hazardous it is to predict future runs solely on the basis of the number of cases packed during the brood-years. This cannot be depended on to give reliable data on the number of fish that will succeed in spawning. A careful study of the conditions described on the spawning-beds of the Fraser in 1913 and 1914 by Mr. Babcock would have prepared us for a decidedly limited run in 1918, though perhaps it would not have foreshadowed the magnitude of the disaster. Age-groups.—In spite of the phenomenally small size of the run, the number of five-year fish shows no marked increase above the normal, although in other years of reduced run it has frequently shown such increase. Taking into consideration those sockeyes only which remained one year in the lake before passing to sea, we find the average percentage of five-year fish in the run to have been 23 per cent, varying on different days of the run from 5 to 59 per cent. No orderly sequence was observed, although in general the heavier percentages were found in June and July. Striking changes in the relative proportions of age-groups in successive years accompanies violent alternations of good and bad years. In such cases we find the poor years in such a series are signalized on the Fraser by a high percentage of five-year fish. But when the runs have declined almost uniformly to a low level, as seems now to be the case, the age- groups regain their normal relations and remain about the same from year to year. Succession of Types in the Run.—As in previous years, it was obvious in examining the takes of different dates in orderly sequence that changes were constantly occurring in the constitution of the run. On July 2nd and 10th, for example, two distinct types could be recognized, one with a small nuclear region, containing most frequently eight to fourteen rings, usually with a band of intermediate rings outside it, and a small first year in the sea. The nuclear rings were fine, crowded, and more or less broken, and numerous five-year fish were included in this series. The characteristics here given are so closely those of the Lillooet-Birkenhead race as to raise a presumption that the fish of this type were bound to that portion of the Fraser watershed. X 38 Report of the Commissioner of Fisheries. 1919 The second type was marked by the large size of the nucleus, containing sixteen to twenty- two rings, and without distinct band of Intermediates surrounding it. The nuclear rings are boldly drawn and strong, well spaced, and not interrupted. The first year's growth in the sea is not reduced, with more or less crowded, irregularly spaced rings, as in the preceding type, and very few are in their fifth year. This second type apparently disappears abruptly about the middle of July and is replaced by several partially recognizable strains, which, together with type number one, form a very mixed assemblage. Figs. 30 to 34 Illustrate some of the different forms which appear in this run. Tables IV. to XL give the succession of sizes appearing on a series of dates, and indicate that the average size increases during the advance of the season, largely by the disappearance of the smaller individuals included in the early part of the run. Taking into consideration only the predominating class in the Fraser, which after hatching spends one full year (fifteen months) in the lake, the average sizes for the 1918 run are as follows:—■ Length, Weight, Inches. Lb. Four-year males 24.9 6.5 Four-year females 23.8 5.7 Five-year males 26.3 7.5 Five-year females 25.4 6.7 From the above it appears the conditions on the feeding-grounds at sea were normal and an average growth was effected. Compare with Table XL, page 35 of Paper No. 3 (1915), of this series. Table IV.—Fraser River Sockeyes, Four-year Males, Esquimau, 1918, distributed by Lengths and Dates of Capture. Inches. rH OJ a 3 l-o 0 CN OJ H 0 ri CO ID oi I"! © rs re*' t-H 3 He CO rH 3 >, 3 1-5 to CM k. 3 He. ci CM >c 3 l-J ci ci 3 ti 3 < cd si 3 ti 3 CO rH 3 06 ri ti) 3 < CO CM ti 3 < CD CN ti 3 < 1 1 1 1 1 1 1 1 2 3 23 4 2 5 1 i 1 3 2 1 1 2 2 2 2 5 1 1 3 1 4 7 6 1 2 26 2 5 3' 1 2 1 16 2 4 4 4 3 3 1 3 2 3 1 1 '4' 4 4 1 4 1 3 2 1 3 1 1 23% 1 1 1 4 1 2 1 1 6 5 6 4 1 3 2 10 5 1 2 3 6 .3 7 3 1 ,1 94 1 2 ?, 24% 4 3 5 s 1 12 3 2 1 1 1 25 . 3 25% 2 26 . ?6Vo 27 . . 1 19 17 2 16 2 19 Totals — 5 18 7 12 10 23 22 25 12 27 3 7 9 Geo. 5 Life-history of Sockeye Salmon. X 39 Table V.—Fraser River Sockeyes, Four-year Females, Esquimalt, 1918, distributed by Lengths and Dates of Capture. Inches. ri CJ a 3 1-5 ©' CM CJ 3 3 1-5 cd CM OJ a 3 ri ci t» 3 1-5 d rH r-. 3 1-5 rj! rH J-? 3 1-5 CO rH 3 1-5 T-i CI 3 1-5 IO CN 3 1-5 d CM >/ 3 ro cm' to 3 <1 -, ro od rH >> 3 1-5 CI r> 10 CI >> 3 r-3, d CI >> 3 1-5 ci to 3 < -*> to 3 CO to 3 < ti 3 CO rH to d rH to CO CI to 3 <5 d CI M 3 22 22% v 23 23% 1 24 1 1 1 1 3 1 3 1 1 1 3 1 2 1 1 25 1 2 1 4 3 i' 3 2 2 2 1 1 1 2 2 1 T 1' 2 1 1 25% 1 1 3 1 1 26 1 1 26% 1 1 1 1 1 27 1 1 a 2 1 27% 3 28 1 28% 1 1 11 29 Totals 4 — 11 13 8 1 4 9 4 6 1 7 4 2 1 2 2 X 10 Report of the Commissioner of Fisheries. 1919 Table VII.—Fraser River Sockeyes, Five-year Females, Esquimalt, 1918, distributed by Lengths and Dates of Capture. Inches. t- rH CJ a a 1-5 d CM 0 a hs cd Cl a a I? ci rt d 003 >> r-> CO He! CI a 1-5 IO CI 3 He d CI >c 3 H5 ci . to 3 to 4 od to < rH to 3 •< CO rH CO 3 < d to < CO CI 60 3 d CI to 3 < 22 1 22% 23 23% 1 1 2 24 . 1 1 2 3 2 2 2 3 1 1 1 1 1 1 24% •. 1 2 25 1 2 i 1 1 3 2 2 1 2 2 1 25% 2 1 1 2 1 1 1 26 1 1 2 26% 1 1 1 27 2 271/, . - . 1 6 Totals 2 5 S 5 7 1 4 2 2 4 3 3 3 1 1 3 1 Table VIII.—Fraser River Sockeyes, Four-year Males, Esquimau, 1918, distributed by Weights and Dates of Capture. Pounds. c 3 3 He d CI CO a He od CI CJ a 3 He ci t» 3 He d rH a He r-i >, 3 Hj od 3 He CI >> 3 He id Cl ci CI _C-c 3 He ci ti a r*' to < GO to CO ri ti 3 ■"1 d rH ti a < CO CI to 3 CD CI ti 3 «! 3 1 1 1 1 1 4 8 1 1 1 3% 4 2 2 1 1 1 5 3 2 2 1 ixe, 1 1 1 5 6 6 1 4 1 2 4 3 1 2 2 1 4 4 5 4 1 1 2 1 5 1 1 1 5 2 1 2 1 8 4 5 2 i 1 9 1 8 2 1 2 4» 4 5 6 3 5% ' 5 i l 4 1 4 4 2 2 2 5 5 7 1 3 3 3 3 1 1 1 2 9 4 10 2 1 1 1 i 1 6 1 6% 0 1 t-U, '>. 8 8% 9 9% 1 26 1 19 Totals 5 18 7 19 16 IT 16 12 10 23 22 25 12 27 3 7 9 Geo. 5 Life-history" of Sockeye Salmon. X 41 Table IX.—Fraser River Sockeyes, Four-year Females, Esquimalt, 191.8, distributed by Weights and Dates of Capture. Pounds. rH CJ a a He d CM OJ a 3 He od CN CJ a 3 He ci 3 He d ri >3 3 He -*' rH 3 He od rH 3 HS CI 3 He LO Ct >, 3 ro d CI >> 3 He CM* to < ti <1 od ti 3 < H ti 3 < CO rH ti 3 O d to 3 CO CI ti 3 < d CI ti 3 < CO a CJ 02 1 1 ■ii/„ 1 6 4 7 1 2 4 1 6 i 1 1 1 3 1 3 2 4. 5 2 1 1 2 1 7 1 3 i 3 4 2 5 2 2 5 2 4 6 2 5 3 8 2 1 3 6 3 5 3 4 5 7 7 2 1 1 1 1 3 i 1 io' 6 3 1 2 1 4 4 7 2 1 8 2 4 1 1 514 1 2 4 2 7 5 5 6 i 2 1 1 6% 1 7% 8 1 8% 9 l 9% i Totals 10 21 12 12 16 19 15 21 17 28 9 21 20 17 9 18 i 3 6 Table X.—Fraser River Sockeyes, Five-year Males, Esquimau, 1918, distributed by Weights and Dates of Capture. Pounds. rH CJ a a He d CI OJ a 3 He od CI CJ fl 3 He ci >> 3 Hj d rH H5 >> 3 He od >> 3 He ci 6c. Irs LO CI >c 3 He d CI 3 He ci to 3 to < 00 to -5J rH to 3 -<1 cd tH 60 a < d rH 60 3 CO CI 60 < oi CM to 3 < 4V„ 1 1 2 Ky„ 1 6 2 1 2 3 3 2 Y 1 2 1 1 1 2 1 1 2 RU, 1 6 1 4 1 3 2 1 2 2 1 1 1 1 1 1 2 1 1 1 1 i 1 1 7 VV, 1 1 1 1 1 1 8 1 8% 1 9 1 ' 2 3 9% 1 2 1 10 10% 1 11 11% 1 11 Totals 4 11 13 S 4 9 4 6 1 7 4 2 1 2 •>, X 42 Report of the Commissioner of Fisheries. 1919 Table XI.—Fraser River Sockeyes, Five-year Females, Esquimau, 1918, distributed by Weights and Dates of Capture. Pounds. 7-i OJ a a He d CN CJ a 3 He od CI cc a 3 He ci 3 He d rH >c 3 1-5 rH >c 3 H5 cd rH >, 3 He CI >> 3 H5 fl t-s d Cl >c 3 He ci to 3 <1 ^' to 5, od to 3 0 to 3 <1 CO rH to 3 Percentage. Four and Five Years old. 5 yrs. 79% 4 yrs. 21% 5 yrs. 4 yrs. 5 yrs. 4 yrs. 5 yrs. 4 yrs. 5 yrs. 4 yrs. 5 yrs. 4 yrs. 5 yrs. 4 yrs. 20% 80% 65% 35% 87% 13% 76% 24% 67% 33% 57% 43% Brood-year from which derived. 1907 (87,874 cases). 1908 (64,652 cases). 1909 (89,027 cases). 1910 (126,921 cases). - 1911 (88,763 cases). [ 1912 (112,884 cases). I 1913 (61,745 cases). 1914 (89,890 cases). Table XIII.—Percentages of Five-year Rivers Inlet Sockeyes occurring at Different Dates in the 1918 Run. Dates, 191$. Percentages of Five-year Fish. Numher of Specimens examined. June 27 7 19 25 32 37 64 59 69 44 59 54 75 ., 29 75 July 2 75 4 75 „ 10 75 „ 12 75 „ 16 75 19 75 „ 23 70 „ 26 75 „ 30 70 Table XIV.—Percentages of Males and Females in Rivers Inlet Sockeyes occurring on Different Dates, Season of 1918. d ci CN TH rH rH >, r-. >> r-l 3 3 3 3 ro He He He 91 98 81 59 9 2 19 41 63 92 43 50 37 8 57 50 Four-year males Four-year females Five-year males Five-year females 60 40 85 15 57 43 77 23 39 61 78 22 48 52 51 49 58 42 51 49 34 66 48 42 43 67 X 44 Report of the Commissioner of Fisheries. 1919 Average percentages— Four-year males 71 Four-year females 26 Five-year males 53 Five-year females 47 Average total males throughout season 66 per cent. Average total females throughout season 34 „ Table XV.—Rivers Inlet Sockeye, Run of 1918, grouped by Length, Age, and Sex. 19 .. 19% 20 .. 20% 21 .. 21% 22 .. 22% 23 .. 23% 24 .. 24% 25 .. 25% 26 .. 26% 27 .. 27% 28 .. Length in Inches. Totals Average length in inches Number of Individuals. Four Years old. Males. 1 1 6 10 50 55 69 49 35 26 31 15 7 355 Females. Males. Five Years old. 1 4 11 23 25 9 11 8 1 2 96 22.5 1 4 4 S 23 20 14 '21 17 28 15 11 4 1 171 24.9 Females. 1 3 17 13 39 35 25 20 16 4 2 175 24.5 Table XVI. -Rivers Inlet Soclceye, Run of 1918, grouped by Weight, Age, an d Sex. 3 .. 3% 4 .. 4% 5 .. 5% 6% 7 .. 7% 8% 9 .. 9% 10 .. Weight in Pounds. Totals Average weight in pounds Number of Individuals. Four Y'ears old. Males.. 7 42 117 77 51 30 20 355 4.9 Females. 3 24 34 20 9 Five Years old. 96 5.1 2 11 26 24 22 29 21 15 12 5 3 1 171 6.7 2 20 22 46 29 26 20 6 2 1 1 175 6.3 9 Geo. 5 Life-history of Sockeye Salmon. X 45 IV. SKEENA RIVER SOCKEYE RUN OF 1918. (1.) General Characteristics and the Age-groups. The Skeena River sockeye-pack for 1918 consisted of 123,322 cases, and ranks among the best half-dozen years in the history of the industry. Coming as it does after two phenomenally poor years, it gives grounds for hope that the Skeena may not as yet be suffering the results of overfishing. Evidences of the decline of a run, with standardized fishing methods, are, however, generally of slow and irregular approach. The returns from this river should be most carefully scrutinized. The necessity for this is all the more evident for the reason that some important factor which must be largely concerned in determining the success or failure of the runs remains unknown to us. Attention has been repeatedly drawn to the total lack of relation between the run of any year and the apparent size of the runs four and five years before. The year 1918 presents a striking example of this lack of correlation. Its brood-years were 1913 and 1914, and as its five-year fish constituted 59 per cent, of the total run of 1918, the year 1913 was relatively the most important. But 1913 was apparently the very poorest year the Skeena has ever experienced. The four-year fish derived from the 1913 run constituted only 38 per cent, of the run of 1917, which was but little better than 1913. There seemed no reason to anticipate that 1913 would deliver five-year fish in 1918 far in excess of its entire yield of both four- and five-year fish. The nature of the unknown factor must be at present wholly a matter for conjecture. We have called attention previously to the effect of boisterous weather on the commercial success of a season. A stormy summer is popularly supposed largely to diminish the proportion of salmon captured. The number of cases packed during such a season might indicate a poor run, whereas the run may have been above the average, and the escape to the spawning-beds unusually large. An attempt to correlate size of pack with weather conditions has not been made. It is possible that weather may exert a directly favourable or unfavourable influence on the size of the run in any year, in addition to its effect on fishing operations. This also is a matter inviting careful investigation. During previous years we have observed an extensive variation in the relative abundance of the two-years-in-lake group. It has varied from 7 per cent, in 1914 to 27 per cent, in 1916, but in two of the four years in which this proportion has been determined it has been 14 and 15 per cent. We have observed in other rivers a wide disparity in this regard between successive years. Evidently, at times, a much larger percentage of the yearlings fail to migrate in their second spring than is usually the case. We have no clue to the significance of this change of habit. It is perhaps to be sought in fluctuating conditions which in certain seasons are less favourable to vigorous growth of the fingerlings than in other seasons. We have previously called attention to the fact that the larger sizes of yearlings migrate at the end of their first year, while the smaller sizes more frequently remain behind for additional growth. If a racial habit has been formed in a given river favouring migration seawards when a certain average stature has been attained, a much larger proportion of individuals may fail to reach it in one year than in another. In 1918 the Skeena River run contained what appears to be a fair average number of adults that had when fingerlings spent two years in fresh water. Fifteen per cent, of the run belonged to this class. As shown in Table XVIII., 60 per cent, of these returned at the age of five and 40 per cent, at the age of six. A tendency appears for the members of this group to run into the river in larger numbers during the early part of the season, although this tendency is not so strongly marked as in the case of the five-year fish that as fingerlings spent only one year in the lake. As shown in the table, the individuals belonging to the latter group become proportionally less abundant during the latter half of July, while the four-year fish that have spent only one year in fresh water become more abundant. (2.) Relative Numbers op Males and Females. A well-marked tendency for the males to precede the females was shown in 1918 in the Skeena, and is evidenced by Table XX., giving the proportions of the sexes in both four-year and five-year fish, taken at intervals of three or four days throughout the run. During the dates in June males were in excess of females generally in both groups, largely in excess in X 46 Report of the Commissioner of Fisheries. 1919 the younger group, less so in the five-year contingent. In both groups, as the season progresses, there is an obvious increase in the number of females at the expense of the males. During the latter half of July the five-year females are greatly predominating, and the four-year females are nearly as abundant as the males of their class. Taking into consideration all the groups present in the run, the one-year-in-lake and the two-years-in-lake groups, of all ages, we find the males notably exceeding the females and constituting 57 per cent, of the entire run. Considering the run as a unit and ignoring the fluctuations in its constitution from its origin to its close, we find the proportions of the sexes in the different groups, brought into comparison with previous years, presented in Table XXI. The close correspondence in a series of years is most striking, as is also the fact that the proportions of the sexes in 1918 represents in each of the groups the average for all the years of which we have, a record. AVe note again that males predominate in both groups of the two-years-in-lake series. The five-year members have a slightly higher percentage of males than the six-year class, but in both classes the males are constantly in excess. It is not clear why the two years spent in fresh water in their early history should have influenced them in this manner. We should have expected males to predominate in the five-year class, while females would be equally in excess with the six-year-olds. (3.) Lengths and Weights. The following tables indicate that the size of Skeena River sockeyes in 1918 was fully up to the average observed over a series of years. The conditions at sea then had been favourable, and no adverse weather conditions had stunted growth during the late winter and early spring months of the spawning year. We again call attention to the comparatively unreliable averages for fish of the two-years-in-lake group, due to the small size of these groups. The weights for the year agree as closely as do the lengths. Comparing 1918 with 1915, we find the greatest difference in the average weights for the various classes is two-tenths of a pound; while the greatest difference in the average lengths equals two-tenths of an inch. Table XVII.—Percentages of Four- and Five-year Skeena River Sockeyes that spent One Year in Lake, in Runs of Successive Years. Run of the Year. 1912 (92,498 cases) . 1913 (52,927 cases) . 1914 (130,166 cases) 1915 (116,553 cases) 1916 (60,923 cases) . 1917 (65,760 cases) , 1918 (123,322 cases) Percentage. Four and Five Years old. 5 yrs. 4 yrs. 5 yrs. 4 yrs. 5 yrs. 4 yrs. 5 yrs. 4 yrs. yrs. yrs. yrs. yrs. yrs. yrs. 43% 57% 50% 50% 75% 25% 64% 36% 60% 40% 62% 38% 59% 41% Brood-year from which derived. 1907 (108,413 cases). 1908 (139,846 cases). 1909 (87,901 cases). 1910 (187,246 cases). 1911 (131,066 cases). 1912 (92,49S cases). \\ 1913 (52,927 cases). j 1914 (130,166 cases). 9 Geo. 5 Life-histori' of Sockeye Salmon. X 47 Table XVIII.—Percentages of Different Age-groups, Skeena River Sockeyes, found to constitute the Run on a Succession of Dates, Season of 1918. Dates, 1918. One Year in Lake. Four Years old. Five Years old Two Y'ears in Lake. Five Y'ears old. Six Years old June 22 „ 25 „ 28 July 1 „ 4 „ 8 „ 12 „ 15 „ 19 „ 22 „ 27 „ 30 •£i 33 30 27 59 46 42 44 67 76 83 82 33 55 53 48 23 48 38 39 26 13 13 16 13 6 11 11 16 3 12 14 5 11 4 27 6 6 14 2 3 8 3 2 Table XIX.—Relative Numbers in One-year-in-lake and Tico-years-in-laJce Groups, Skeena River Sockeyes, 1918, on a Succession of Dates. Dates. June 22 „ 25 „ 28 July 1 „ 4 „ 8 „ 12 „ 15 „ 19 „ 22 „ 27 „ 30 One Year in Lake. Two Years in Lake. 60 40 88 12 S3 17 75 25 82 18 94 6 80 20 83 17 93 7 89 11 96 4 98 2 Table XX.—Percentages of Males and Females, One-year-in-lake Group, Skeena River Sockeyes, 1918, on a Succession of Dales. Four Years. Five Years. Males. Females. Males. Females. June 22 „ 25 „ 28 July 1 4 8 61 74 94 82 59 85 67 62 63 56 44 52 39 26 6 18 41 15 33 38 37 44 56 48 61 59 46 60 53 46 . 46 30 12 17 33 33 39 41 54 40 47 54 „ 12 „ 15 „ 19 „ 22 „ 27 „ 30 54 70 88 83 67 67 X 48 Report of the Commissioner of Fisheries. 1919 Table XXI.—Relative Numbers of Males and Females in Different Year-groups, Skeena River Sockeyes, in a Series of Years. One Year in Lake. Two Years in Lake. Years. Four Years old. Five Years old. Five Years old. Six Years old. Males. Females. Males. Females. Males. Females. Males. Females. 1912 1913 1914 1915 1916 1917 1918 54 69 60 55 70 65 63 46 31 40 45 30 35 37 42 47 47 45 43 4S 46 58 53 53 55 57 52 54 56 65 61 44 35 39 54 58 56 46 42 44 Table XXII.—Skeena River Sockeyes, Run of 1918, One Year and Sex. Lake, grouped by Length, Age, Length in Inches. 21 21% 22 • 22% 23 23% 24 24% 25 25% 26 26% 27 27% 28 Totals Average length in inches Number of Individuals. Four Years old. Males. Females. 8 27 35 43 39 41 9 12 1 226 24.1 1 3 7 21 39 23 21 9 6 2 1 133 23.3 Five Years old. Males. Females. 25.8 1 4 1 3 6 19 2 22 13 41 20 20 34 16 17 6 14 2 5 1 114 133 24.9 a fa *^et ." CD h -a C3 rj W-5 ■■.■^^ co £ fe S M S 03 co 5 S V CO a .2 fl toO as s ■ . 3^ a» % a fe S - u.S OJ CD >-. fa Fig. 9. Fraser River sockeye from the Birkenhead River, taken September 22nd, 1918. Female, 21 inches long. -.■-;. •.v,i'....-.--..•C.«\\»'- ••-■■ - , \\ ■ • s*XvV< ■ ->$&r- ir•".:' Fig. 10. Fraser River sockeye from the Birkenhead River taken September 7th, 1918. Female, 24 inches long. Fig. 11. Fraser River sockeye from the Birkenhead River, taken September 10th, 1918. Female, 25 inches long. ///: Fig. 12. Fraser River sockeye from the Birkenhead River, taken September 7th, 1918. Female, 24 inches long. 3 6 at 3 CD ^0 5 3 g fl .£ .M Ol «5H . CD 00 CO O rH rH 9 CD »-l g « O +J i-i «J a. <] SS;$?Ss^—--"■ M < Ol cD a 5 r* ® . 2 E5 Fig. 16. Fraser River sockeye from Fish Canyon, Chilcotin River, taken August 24th, 1918. Female, 28% inches long. vmfg Fig 17. Fraser River sockeye from Fish Canyon, Chilcotin River, taken August 25th, 1918. Male, 23*4 inches long. Fig. 18. Fraser River sockeye from Fish Canyon, Chilcotin River, taken August 24th, 1918. 3 co ":-3-u"-» • "*■ **X» ':'.="■ Xsc^XCi;:*;,',!- '• •. Ji&-.'c-I-:'-;" ■■■ >SiV>xxx.x - t'X.X • " . X ^ -X '- M 3 < rH r3 rH a v.'•" • .-"••-'X"'" * ■■•,.-Vt V**»£-St ■-'■ '■':■' -...'."-.Otc S S8 O ri ■Z co- E X^ iililMi H .0 Ph M 3 be rH a. o, fe « - to ti 3.3 g j3 3 3. .3 a . CD CD CO Jd cs ca fe " . bo . ca b u Ol -O ri -O CD biS ^ fe bS O «w 80 G r'.^2 9 4 1 1 1 1 4 6 6 3 9 9 4 0 5 2 i 4 5 3 2 1 3 4 2 1 1 1 1 5 1 1 1 1 2 i 2 1 2 3 2 3 1 Total No... 75 70 33 49 91 79 27 13 4 11 Ave. length 24.9 24.1 25.6 24.7 25.7 24.8 27.1 25.0 27.3 25.4 Ave. weight 6.2 5.7 7.1 6.2 7.1 6.2 8.2 6.6 8.1 6.8 Table XXIX.—Nass River Sockeyes which ran in August, 1918, grouped by Age, Sex, Length, and by their Early History. Number of Individuals that spent Length in Inches. One Year in Lake. Two Years in Lake. Three Years in Lake. Four Years old. Five Years old. Five Years old. Six Years old. Six Years old. Males. Females. Males. Females. Males. Females. Males. Females. Males. Females. 92 22% 23 23% 24 24% 25 25% 26 'i 1 1 i i i "2 5 6 5 - 2 1 i '5 13 6 6 i i 4 2 i 26% 27 27% 28 28% 1 1 1 1 2 4 2 i 4 1 Total No... 3 2 22 32 6 1 11 8 Ave. length 25.5 25 26.4 1 25.2 27.6 26.5 26.5 26.1 Ave. weight 7 6 • • 7.2 6.5 8 .5 7.5 7 .8 6.8 X 52 Keport of the Commissioner of Fisheries. 1919 Table XXX.—Percentages in each Class of Nass River Sockeyes running at Different Dates in 1918. 1918. One Year in Lake. Two Y'ears in Lake. Three Y'ears in -Lake. Four Years. Five Years. Five Years. Six Years. Six Years. Seven Years. June 26 July 2 35 27 57 43 40 27 22 22 10 5 10 2 25 14 13 37 32 22 13 19 2 9 40 59 23 12 22 37 45 49 63 61 60 68 7 6 6 14 16 8 6 18 7 9 2 4 2 19 7 23 21 8 „ 11 „ 16 „ 19 „ 23 „ 25 „ 30 Aug. 2 „ 5 9 Geo. 5 Statement by Hon. Wm. Sloan. X 53 STATEMENT SUBMITTED TO THE AMERICAN-CANADIAN FISHERIES COMMISSION BY THE HON. WILLIAM SLOAN, COMMISSIONER OF FISHERIES, AT THE VANCOUVER MEETING, 1918. Gentlemen of the Commission,—As the Commissioner of Fisheries for the Province of British Columbia, I desire to comment briefly on the present precarious condition of the sockeye- salmon fishery of the Eraser River. The watershed of that river and its channels lie wholly within the Province. From its watershed have come all but a fraction of the sockeye that have been taken in the waters contiguous to the International Boundary-line which separates the Province of British Columbia and the State of Washington. From that watershed must continue to come the seaward migrants of sockeye which produce the commercial runs of sockeye to those waters, because there is no other watershed tributary to those waters which affords sufficient spawning and rearing waters for sockeye salmon. The vast runs of former big years demonstrates the extent and the value of the runs that that watershed can produce. It is the greatest sockeye-producing watershed known. No other watershed has produced such vast numbers of sockeye. That watershed is to-day as capable as ever of producing the vast run of the past. It has not been contaminated. Settlement, power and irrigation have not injured it in any way. It needs only to be protected to produce the great runs of the past. Its produce has been lessened—alm5st destroyed—because a sufficient number of spawning sockeye has not been permitted to reach it. The runs in the three last years have steadily decreased because too few of the adult salmon have escaped capture in Dominion and State waters; because of excessive fishing-too many have been captured. The run of the big year was further destroyed by a rock- slide blocking the river-channel at Hell's Gate in 3913. That the runs in the three last years have been almost wiped out by commercial fishing and that the run in the big year has been alarmingly decreased has been ably demonstrated by statements already submitted to you. It is unnecessary here to more than call your attention to the evidence of depletion already in your hands. I do, however, accentuate the fact that the evidence submitted to you by our Department is founded upon scientifically ascertained facts. The watershed of the Fraser and the fishing areas supplied by It have been under close scrutiny since 1901. No other has been so carefully observed. The history of the race of sockeye that frequent the Fraser is better known than that of any other district. The facts are no longer questioned. The run of sockeye to the Fraser is periously near to extermination. They will be exterminated if conditions remain as they are, and in so short a period as to wipe out all interests of both fishermen and Banners. In view of the evidence there is, in my judgment, but one thing to do. Adopt measures that will ensure to the watershed all the sockeye that still survive. To that end I would suggest that the total prohibition of sockeye-fishing in the waters frequented by those produced in the Fraser River until such time as they have recovered from their depleted condition. I suggest this though it does involve compensation to resident fishermen and canners who can establish that they are entitled to compensation by their respective Governments. It is fruitless to rely upon concurrent regulations in British Columbia and the State of Washington waters. That has been tried and failed. Such efforts will continue to fail. Present commercial and monetary considerations must be eliminated. It must be in an international way, because it is an international question. It is one of the greatest fishery questions in which Canada and the United States are now concerned. The only adequate, the only permanent solution of this question, I submit, is-the acquisition by Canada and the United States of all the rights in this fishery of which they may not now be in possession. That being established, the waters should be closed to sockeye-fishing for such a period of time as is necessary to restore the runs to the abundance of former big years. AVhen that has been accomplished, let fishing be resumed under supervision and for the benefit of the two nations until such time as they have been recouped for their expenditures, and thereafter in such manner and to the end that the supply may not again become depleted. I submit, gentlemen, that the Governments of Canada and the United States should recognize that the conditions confronting them demand such treatment. There is no gainsaying the evidence. The watershed of the Fraser River will, when adequately protected, produce more X 54 Report of the Commissioner of Fisheries. 1919 sockeye salmon than any known watershed. It produced in 1913 2,300,000 cases. In the three following lean years it produced an average of but 267,000 cases per year. Being in possession of a watershed capable of producing 2,300,000 cases a year, can any Government be content with conditions by which but 267,000 cases are produced, and the continuance of which will entirely destroy any production whatever? The evidence in the case is conclusive—it is undisputed; the fishery is in a precarious condition. There is, however, a difference of opinion as to the remedies to be applied. I submit that those best qualified to speak have made it plain that no temporary measures will produce desired results. The races of sockeye that frequent the Fraser cannot be restored by any half-way measure. To allow the destruction of the sockeye-flsheries of the Fraser River would be an unnatural, immoral, and unpatriotic policy. The questions here involved are similar to those in the fur-seal case. They are international in character, and not Provincial or State questions, and must be dealt with upon broad national lines and in the interests of the people of Canada and the United States. Faithfully yours, Wm. Sloan, Commissioner of Fisheries for British Columbia. STATEMENT MADE BY JOHN P. BABCOCK, ASSISTANT TO THE COMMISSIONER OF FISHERIES FOR BRITISH COLUMBIA, AT VANCOUVER MEETING OF THE AMERICAN-CANADIAN FISHERY CONFERENCE, 1918. Mb. Chairman and Gentlemen of the International Fisheries Commission,—The outstanding features of the salinoii-fishery of the Fraser River District, in my judgment, are the depletion of the runs and the potentialities of the watershed. Because depletion has been shown, and is universally admitted, I shall confine attention to the latter. The Watershed,—The watershed of the Fraser River contains a greater area of tributary fresh-water lakes than are found in any other on the Coast. The Fraser drains the major portion of the south-eastern section of the Province of British Columbia. Three of the largest lakes on the Pacific slope and five others of large area contribute their waters to the Fraser and afford spawning areas and rearing waters for a countless number of sockeye salmon. No other known watershed affords such an extended spawning area. No other watershed produced in a single year such vast numbers of sockeye. The great runs of 1901, 1905, 1909, and 1913 demonstrate the harvest that watershed will afford when abundantly seeded. The great catches of those years—ranging from 1,572,000 to 2,401,000 cases—demonstrates the number of fish that may safely be taken without, injury to the runs of the future, because, notwithstanding such great catches, every section of the spawning area of the watershed was shown to have been abundantly seeded in 1901, 1905, and 1909, and there is evidence to show that its spawning area would have been as abundantly seeded in 1913 but for an accident. Since 1901 I have made a study of conditions on the fishing and spawning grounds of the Fraser River. I first inspected its spawning area in 1901. As the agent of the Provincial Government I have inspected that watershed during the spawning period every year since, with the exception of the years 1910 and 1911. The annual publication of the Provincial Government contains my yearly reports. As the result of my inspections, I feel fully justified in submitting that the major portion of the great runs of 1905, 1909, and 1913 were the product of the sockeye that spawned in that section of the watershed of the Fraser that lies north of the great canyon in the Coast ranges, commonly termed " the Fraser River Canyon." And that the major portion of the runs in the alternate years—the lean years—were the product of the sockeye that spawned in that section of the watershed that lies to the south and west of the Fraser River Canyon. In the discussion of this question the former is here termed the upper section of the Fraser River watershed, and the latter the lower section of the Fraser watershed. In my judgment, the great runs of 9 Geo.- 5 Statement by John P. Babcock. X 55 the big years have very largely consisted of fish propagated in the upper section, plus the normal yearly product of the lower section. The runs in the alternate, or lean, years have been the normal yearly product of the lower section plus the small numbers produced in the lean years from the beds of the upper section. In the big years 1901, 1905, and 1909 every spawning-bed in the upper section was crowded with sockeye. They were found there in incredible numbers. The beds of that section in 1913 showed an alarming decrease. While over 4,000,000 sockeye were recorded as entering Quesnel Lake in 1909, but 550,000 were shown to have entered that lake in 1913, and less than 2S,000 in 1917. Furthermore, similar conditions were shown to have existed in all the lake districts of the upper section in 1913 and 1917. The number of sockeye that reached the beds of the upper section in 1917 were shown to have been very much less than in 1913, and little, if any, more numerous than in some reeent lean years. The records of the upper section in 1913, and again in 1917, demonstrate that the conditions which produced the big runs in 1905, 1909, and 1913 no longer exist, that the big year run has been destroyed, and that hereafter the runs of those years must be classed with the runs in the lean years. In the alternate—the lean—years the spawning- beds of the upper section were but sparingly seeded up to 1906 and have not been as well seeded since. Gradually, with one or two exceptional years, the number of sockeye which reached the upper section in the lean years has notably declined. Every district in the upper section shows a decline. Hatcheries located at Shuswap and Seton Lakes, the only hatcheries in the upper section, have been closed because a sufficient number of sockeye have not reached those lakes in recent years to afford a supply of eggs. No eggs were or could have been collected at either of those lakes in the last .three years. Passing to the spawning area of the lower section of the Fraser, the record discloses that from 1901 to 1917 there was no pronounced increase in the- run in the big years over the run of the lean years. That there has been as many fish on those beds in the lean years as in the big years. This is especially true of the runs to Lillooet and Harrison Lakes, the two great lakes of the lower section. The runs to this section have shown a steady decline. There were less sockeye in this section in 1917 than in any former year, big or lean, on record, and less eggs were secured for the hatcheries. Because the bulk of the run of the big years issues from the upper section of the watershed, and because there has been no noticeable increase in the number of sockeye on the beds of the lower section in those years, it appears that the condition which originally brought about the phenomenon of the big run and the three following small runs of sockeye to the Fraser was of such a character as to have affected only the run to the upper section and yet did not affect the run to the lower section. Because the run to every lake district of the upper section was equally affected we are warranted in assuming that the point of obstruction was located below the junction of the Thompson and the Fraser Rivers. Because the run was affected for three years only, it is assumed that the barrier, or blockade, was of such a character as to have affected the run in those years only and did not affect the run in the fourth year; that in the fourth year it had worn away or been removed by high water to an extent that permitted the run of that year to reach the spawning-beds of the upper section. The channel of the Fraser through the canyon extending from Yale to Cisco is exceedingly narrow at many points. Towering cliffs of rock line its banks. A rock-slide such as the one that occurred in that canyon, at Hell's Gate, in 1913 could easily have produced a similar result at an earlier period of time, and just as effectively cut off the run for a number of years as the slide of 1913 would have done had it not been removed by the Dominion Government in 1914. Great as was the slide of 1913, it did not cut off all of the early run of sockeye of that year. Owing to extreme high water in July numbers of sockeye were enabled to pass through, as was demonstrated by the fish reaching Quesnel and Chilko Lakes. Assuming, then, that we have here a reasonable theory of the origin and the nature of the barrier that cut off the sockeye from the spawning area of the upper section of the Fraser which resulted in the phenomenon of the one big year and three lean year runs of sockeye, let us speculate as to the extent of its effect upon the runs of the three lean years. The barrier may have been sufficient to have cut off a portion of the run only, or it may have cut off the entire run. If a portion of the run at extreme high or low water was enabled to pass, that portion would have furnished the nucleus—the seed—for a run four years later. X 56 Report of the Commissioner of Fisheries. 1919 If we assume that the entire run was cut off from the upper section for one or all of the three years of the cycle, we must furnish a plausible, a workable, theory to account for the restoration of the runs in the three lean years. That is not difficult. Dr. Gilbert, by his study of the growth and structure of the scales of the sockeye, has demonstrated that, while the race of sockeye that frequent the Fraser are predominately four-year old fish, there is in the run of every year three-year-old, four-year-old, and five-year-old fish. He demonstrated "clearly that in the Fraser basin a high percentage of four-year fish accompanied a large pack, or, stated conversely, a small pack is occasioned mainly by a deficiency in four-year-old fish."* In four runs he found an average of 82 per cent, of four-year-old fish. In the catch of 1912 he found 21.5 per cent, of three-year-old fish and 10 per cent, of five-year-old fish. The presence of from 54 to 99 per cent, of four-year-old fish in five consecutive runs, and plus per cent, to 46 per cent, of five-year-old fish in the same runs, establishes the fact that from each year's spawning a proportion of five-year-old fish are produced. We have here evidence of the existence in the run of the year that was not obstructed the nucleus for a future run. The period taken to build up a run from such a scant seeding must have been an extended one. Had this constituted the only seed that reached the beds in the three lean years, the run following the big year should have been more readily built up than the run in the two following years. In fact, the record of the pack demonstrates that the catch in the year following the big run has always been greater than in the two following years. We are not, however, confined to the conclusion that if the run for three years was entirely destroyed the run in the lean years was built up entirely from the spawning of five-year-old fish. Granting the premises of the home-stream theorists that salmon bred in a stream return to that stream to spawn because the weight of evidence favours their contention, there is abundant evidence on this Coast and in New Zealand, where the Pacific salmon have been established, to show that some of the salmon bred in one stream have on reaching maturity entered a different stream to spawn. Further, it has not been shown or claimed that all the fish bred in a watershed return to the identical tributary of that watershed in which they were propagated. There is sufficient evidence to warrant the conclusion that some salmon propagated in the lower section of the Fraser have entered the upper section and spawned there. We have therefore three strings to our bow to account for the existence of the runs of the lean years: (1) The escapes at high or low stages of water during the original blockade; (2) the overlapping five-year-old fish bred in the upper section by the run that did survive; and (3) the fish bred in the lower section that passed to the upper section and spawned there. The building-up of a run to the upper section of the Fraser by any or all of the measures here indicated would of necessity have takeii a long period of time. We cannot estimate it, but we are not compelled to place it in the dark ages, or even two or three centuries ago. We can, however, postulate that during that period the Indians of the upper section and in the Fraser Canyon were catching such fish as they could secure up to the limit of their demands, and were in consequence interfering with a more rapid development of a run. The records of Simon Fraser, the discoverer and original navigator of the river that bears his name, written more than a century ago—1806-11—establishes the fact that there was in those years a big run every fourth year and a light run in each of the three succeeding years in the Upper Fraser. He also furnishes evidence to show that in some of the lean years the Indians at Stuart and Fraser Lakes could not supply his post with all the salmon he required, and that an additional supply was obtained from the post at Kamloops, where the run appears to have been larger or the demand less; indicating at that time the Indians could not obtain from Stuart and Fraser Lakes all the fish they desired. This matter has been treated at length that it may be appreciated that the period of time necessary to restore a depleted run to a run of commercial importance is under natural conditions a long one, and because it has been stated that the failure of nature to have equalized the runs of sockeye in the Fraser demonstrated that the runs in the lean years could not be developed to the proportions of a big-year run. I sumbit that there is no force in that contention. The runs in the lean years will equal the runs of former big years four years following as abundant seeding as the beds received in either 1901, 1905, and 1909. It is only a matter of seeding the spawning area abundantly. * B.C. Fishermen Report, 1915, pages 29' and 30. 9 Geo. 5 Statement by John P. Babcock. X 57 Mr. Secretary Redfield requested that in advocating restrictive measures I should consider the present demand for foods, and that he would like me to express my views on the subject. In reply, let me submit: The average pack of Fraser River sockeye in Washington aud British Columbia in the last three lean years totalled but 267,000 cases. The reports from the spawning- grounds of the Fraser in these years give no promise that the runs in the following three lean years will equal that average. If that be true, the best we can expect is 267,000 cases per year. Is the food shortage so serious as to demand that the few remaining sockeye of the Fraser run be drawn upon to supply that number of cases? Permit me, gentlemen, to draw my remarks to a conclusion by quoting from the very able statement furnished you by my mentor, Dr. Gilbert:— " The Fraser River presents unexampled opportunities for productiveness and wealth. The people need the enormous supplies of highly valuable food which the river is able to produce annually. It should not be permitted to remain at its present low rate of production. . . . The people need the food. They will come to need it in future years even more sorely than they do at the present. No private interests should be permitted to stand in the way of restoring this producer of food to the public. " If the Fraser River were a private monopoly, to be henceforth wisely handled, there can be no doubt it would now be promptly closed to commercial fishing for a term of years, and the entire run—now so sadly dwindled—dedicated to purposes of propagation. This should be done without further delay for at least one cycle of four years, and the results carefully noted by a continued study of the spawning-beds. ... If the river were closed to fishing for one cycle of four years, we could know fairly well in advance what the result was to be, and could then either open the river and sound to restricted fishing if conditions should warrant, or, if necessary, close it for a future period of four years. This is the only method to restore the sockeye run with any promptness and with any certainty of success. " So great has been the reduction of the runs, we cannot predict with any optimism what would be the result of less drastic measures. If the amount of fishing-gear in use be limited and the weekly closed season be extended, undoubtedly a somewhat larger proportion of fish would reach the beds. But it must be borne in mind that it is not the proportion of a given run which spells success, but the actual numbers of spawners. The whole of a sadly depleted run may be all too few to produce the desired results. It is greatly to be feared that any restriction hi the present case which would be so moderate as still to leave it profitable for canners to operate in the face of such reduced runs can accomplish little or nothing towards restocking of the river. The only wise course—the only adequate remedy—is to close the river for a term of years by concurrent action of the two Governments." Yours respectfully, John Pease Babcock, Assistant Commissioner of Fisheries. X 58 Report of the Commissioner of Fisheries. 1919 MIGRATION OF ADULT SOCKEYE SALMON IN PUGET SOUND AND FRASER RIVER.* By Henry O'Malley, Field Assistant, U.S. Bureau op Fisheries, in Charge op Operations on the Pacific Coast, and Willis H. Rich, Field Assistant, U.S. Bureau op Fisheries. INTRODUCTION. During the American-Canadian Fisheries Conference on the Pacific Coast in April and May, 1918, the question was constantly raised as to the time required for the sockeye salmon of Fraser River to pass through Puget Sound. In order to determine this important question, Dr. Hugh M. Smith, United States Commissioner of Fisheries, and W. A. Found, Superintendent of the Canadian Fisheries, decided to carry out an extensive marking experiment. Early in July the authors were detailed to take immediate charge for the United States Bureau of Fisheries, and to confer with Lieut.-Colonel F. H. Cunningham, acting for the Canadian Government. The final arrangements were that the Canadian fisheries authorities would handle the marking operations in Canadian waters and the United States Bureau of Fisheries the marking operations in Puget Sound, and that each Government would attend to the collecting of data in its respective waters. The authors are especially indebted to H. J. Todd & Sons; E. B. Deming, Pacific American Fisheries; W. A. Lowman, Coast Fish Co.; Frank Wright, Carlisle Packing Co.; and J. W. Elliott, Alaska Packers' Association. Without the co-operation of these men and the members of their respective organizations it would have been impossible to have successfully conducted this experiment. L. H. Darwin, Washington State Fish Commissioner, and his deputies also rendered valuable assistance. Many helpful suggestions were received from Dr. Charles H. Gilbert, of Stanford University. PROCEDURE. Tags and their Attachment. The method adopted for procuring the desired data was to mark with serially numbered tags enough adult sockeye salmon so that a sufficient number of returns could reasonably be expected. After consideration of several types of tags the one finally accepted was of the " bachelor-button " type, similar to those in use for marking cattle and other live stock. This button comes in two separate halves, each with a hollow central extension which forms the shaft of the complete 'button. The shaft of one half of the button is small enough to pass through the shaft and long enough to extend slightly beyond the.face of the other half. When placed together in the proper position the two halves are fastened by means of a special crimping-tool. This is built on the general plan of a pair of pliers; but the jaws meet only toward the tips, at which points are conical elevations. By fitting these elevations into the open ends of the shaft and closing the tool firmly, the end of the smaller inner shaft is spread sufficiently to prevent its being pulled back through the outer shaft. The buttons were made of either silver or aluminium. Some fear was entertained that the aluminium might corrode in the salt water sufficiently to make it diflicult or impossible to read the numbers, but in this experiment there was absolutely no difficulty on this account. No corrosion is noticeable on any of the buttons returned, although some of them must have been three or four weeks in practically pure salt water. This is contrary to the experience of Greene,t who found that salt water had a strongly corrosive action on aluminium buttons of similar style. This may be due to some slight difference in the alloy of which the buttons were made. * Reproduced from advance sheets furnished the Department by Dr. Hugh M. Smith, United States Commissioner of Fisheries. t Greene, Charles W.: The Migration of Salmon in the Columbia River. Bulletin, U.S. Bureau of Fisheries for 1909, Vol. XXIX., pages 129-148. Washington, 1911. 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 59 The buttons were attached to the upper lobe of the caudal (tail) fin. It was first necessary to cut a hole of the proper size to admit the shaft of the button. This was done by means of an ordinary leather-punch. Figs. 1 to 4 show, in situ, both sides of two buttons which were returned.* The operation of attaching the tags required, usually, less than one minute and no particularly unfavourable results of tagging were noticed. At some of the stations where the marking was done the fish were held in crates until after the marking for the day was complete, and it was noticed that they soon recovered from the effects of being out of water while the tag was being attached and were apparently in perfect condition when liberated. Reports that the tags were " worrying " the fish and causing them to wear their tails in an effort to remove the tags were investigated and found to be quite unfouuded. Tails of several fish have been preserved and show no indication of such wearing as had been asserted. Figs. 1 and 2 are from such a specimen. The fish to be marked were taken from the traps at the time of lifting and were held either in crates or in one of the pockets of the spiller until they could be marked and released. Marking-stations. Marking-stations were established at five points in Puget Sound, as follows: Near Sooke, Vancouver Island, British Columbia; Salmon Banks, just south of San Juan Archipelago ; near Point Partridge, Whidbey Island; near Village Point, Luinmi Island; and at the last trap in American waters on Point Roberts. For convenience these have been designated in the following pages as Stations A, B, C, D, and E, respectively. The work at these stations was in charge of the following men:— A. Sooke, B.C.—W. H. Rich, Field Assistant, United States Bureau of Fisheries (July 14th to 23rd) ; Dr. C. McLean Fraser, Director of the Biological Station, Nanaimo, British Columbia (July 23rd to the end of the season) ; Alex. Robertson, Superintendent, Harrison Lake Hatchery, Dominion Fisheries Department. B. Salmon Banks—Walter C. Buckmaster, Apprentice Fish-culturist, United States Bureau of Fisheries. C. Point Partridge—Clive L. Henry, Apprentice Fish-culturist, United States Bureau of Fisheries. D. Lmnmi Island—Don E. Courser, Apprentice Fish-culturist, United States Bureau of Fisheries. E. Point Roberts—Joseph Kemmerich, Foreman, United States Bureau of Fisheries. The last four of these stations were under the general supervision of Dennis Winn, Field Superintendent, United States Bureau of Fisheries. Collection and Organization of Data. In the collection of data reliance was necessarily placed in fishermen, trap-tenders, and cannerymen. A reward of 25 cents was offered for the return of each button accompanied by information giving the date and place of capture. Specimens taken in American waters were taken care of by Dennis Winn, at the Seattle office of the United States Bureau of Fisheries, and those taken in Canadian waters by Colonel Cunningham, at the office of the Dominion Fisheries Department at New Westminster, British Columbia. One of the tags reported from the upper, regions of the Fraser River was secured by J. P. Babcock, Assistant to the Commissioner of Fisheries for the Province of British Columbia. While the majority of these records are considered approximately correct, data secured in this manner are necessarily subject to some inaccuracies, both as to time and place of capture. In the authors' opinion, however, these inaccuracies will be balanced, so that with reasonably large series the averages should be reliable. In cases where the data were obviously wrong the records have been omitted. This has been done in several instances where the date given for the capture was earlier than that on which the fish was recorded as having been marked. A number of tags were returned with incomplete data, either the date or the place of capture, or both, wanting. All such cases have been omitted entirely from consideration. For the convenient organization of the data the American waters .of the sound from which returns were reported have been divided into nine regions. Several factors guided in establishing * Not shown in this reprint. X 60 Report of the Commissioner of Fisheries. 1919 O J^-'J U a N D E F u 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 61 the boundaries of these regions : (1) The general geography of the district; (2) the grouping of traps; and (3) the nature of the descriptions giving the locality where the tags were recovered. These last frequently indicated merely the general region in which the tag was taken; i.e., Rosario Strait. The boundaries of these regions, as well as the location of the marking-stations, are showii on the accompanying map. All of the records reported from the Canadian waters which are open to commercial fishing have been treated together. For this reason that part of the Strait of Georgia just outside the mouths of the Fraser River and the river itself, from the mouth to Mission Bridge, together constitute the tenth region. The reasons for so treating the Canadian returns will be given later. STATISTICAL STUDY OF DATA. General Features. In the following study the chief concern is with what is believed to be a strict and reasonably complete presentation of the facts. There has been no attempt to draw conclusions regarding the bearing of these facts upon the particular problems connected with the conservation of the sockeyes of the Fraser River. Some of the tables present similar data as seen from different points of view. This has, perhaps, been carried to aii extreme in order that all obtainable facts might be available. Other tables are presented rather as matters of record than as having any especial bearing on- the main problems in hand. For the most part such tables are presented without detailed comment. Raw Data. Table 1 gives,for each marking-station the numbers of the tags attached each day. As noted in the table, silver tags were used only at Sooke (Station A). At all other stations aluminium tags were used, and the first few tags used at Sooke were also of aluminium, Nos. 5 to 29, inclusive. Table 1.—List of Tag Numbers attached at each Marking-station. Station A, Sooke, British Columbia.* Numbers. Date. Total. Numbers. Date. Total. From To From To 5 8 29 1350 1362 1382 1390 1447 1545 1617 1-70.7 1918. July 14 „ 21 4 21 25 10 20 8 56 98 72 99 1718 1823 1901 2039 20S8 2OS0 2101 2125 1822 1900 2038 2078 2100 2087 2123 2127 1918. July 31 105 9 Aug. 1 2 77 138 1326 „ 21 1353 .. 22 ., 23 „ 24 „ 4 41 1363 1383 4 „ 6 23 8 1391 „ 6 23 144S 27 1546 1.618 „ 28 „ 30 Tot al 831 Station B , Salmon Banks, Wash. 3001 3025 3065 3160 3216 3266 3377 3454 3394 3400 July 21 25 40 95 56 50 111 54 17 6 3455 3497 3601 3665 3759 3S01 349fl 3600 3664 3758 3800 3S36 42 3026 22 104 3066 ,. 23 24 . : „ 2 64 „ 4 94 25 ., 5 42 „ 26 3401 3378 3395 ,,. 28 „ 30 „ 31 To 836 Station C, Point Partridge, WASH.f 1001 1043 July 17 1044 1093 ,, 19 1094 1154 22 1155 1230 ,, 24 1231 1382 ., 20 1383 1546 Aug. 6 43 50 61 76 152 164 1547 1665 1770 1925 1664 1769 1924 2000 Aug. 13 14 Total 118 105 155 1,000 * Tag Nos. 5 to 29, attached at Station A, and the tags used at all other stations were aluminium. With the exception of those noted, all tags attached at Station A were of silver. t This is the official list as given by the assistant having charge of the marking at this station. The records for August 6th and subsequent dates are subject to suspicion on account of an admitted lack of veracity in the accounts. Previous to this date Dennis Winn was present at the markings. X 62 Report of the Commissioner of Fisheries. 1919 Table 1.—List of Tag Numbers attached at each Marking-station—Concluded. Station D, Lummi Island, Wash. Station E, Point Roberts, Wash. Numbers. Date. Total. Numbers. Date. Total. From To From To 2001 2085 2177 2251 2314 2360 2416 2477 July Aug. 19 23 24 28 30 2 4 1918. 85 92 74 63 46 56 61 247S 2556 2650 2691 2555 2649 2690 2737 Aug. 5 6 ., 7 9 1918. 7S 20S6 2178 94 41 2252 2315 2361 47 737 2417 4001 4096 4166 4259 4373 4495 4534 4631 4744 4819 4835 30 86 98 121 152 Total Table 2 gives in serial order a complete list of the tags returned, showing the place where and the date on which the tags were attached and the date and region of capture. Tables 1 and 2 contain the raw data from which all of the subsequent tables were constructed. Table 2.—List of Tags returned. Tags attached at Station A, Sooke, British Columbia. ] Date Date Region Days Date Date Region Days. Tag No. fish Fish where en Tag No. Fish Fish where en m irked. recaught. taken. Route. marked. recaught. taken. Route. 3 1918. 1918. 1918. 1918. ly 14 July 19 2 5 1455 July 27 July 36 1 3 12 , 21 .„ 23 9 9 1463 „ 27 Aug. 2 6 6 14 , 21 Aug. 1 8 11 1467 ,, 27 July 31 9 4 16 , 21 „ 20 * 3© 1468 27 „ 30 6 3 19 , 21 1 t 11 1473 „ 27 Aug. 3 7 7 21 . , 21 July 24 7 3 1477 „ 27 ,. 0 7' 9 25 , 21 „ 25 9 4 1481 27 July 29 6 2 26 , 21 „ 23 2 2 I486 „ 27 ,. 31 6 4 9~ , 21 „ 25 1 4 1505' „ 27 Aug. 1 t o 28 , 21 ,. 24 2 3 1512 „ 27 6 6 10 29 , 2.1 „ 31 6 10 1513 ., 27 July 28 ' § 1 1328 , 21 „ 25 2 4 1515 ,. 27 Aug. 4 1 8 1331 , 21 „ 26 2, o 1519 „ 27 Sept. 14 1! 49 1332 , 21 „ 23 1 2 1529 „ 27 July 30 6 3 1335 .... , ■ 21 „ 24 9 r> 1538 „ 27 Aug. 5 7 9 1339 , 2-1 ,, 24 2 o 1543 „ 28 July 30 7 2 1342 , 21 „ 28 9 7 - 1546 „ 2.8 „ 29 6 1 1353 22 „ 28 9 6 1549 „ 28 29 1 1 135? 22 „ 29 10 4 1531 „ 28 Aug. 1 f 4 1339 , 2'2 „ 80 10 8 1565 „ 28 „ 4 7 7 1362 92 „ 23 1 1 1573 ,, 28 „ 23 9 26 1368 , 23 Aug. 30 10 7 1588 „ 28 9 fi 4 1371 , 23 July 31 10 8 15«1 „ 28 9 6 12 1390 24 „ 29 6 0 1502s „ 28 ,, o 10 8 1404 , 25 „ 29 9 4 1593 „ 28 4 8 7 140io , 25 „ 28 1 3 1507 „ 28 9 6 12 1407 , 25 Aug. 4 1 10 1602 „ 28 4 7 7 1423 , 25 July 30 8 0 1607 „ 28 4 8 7 1449 27 Aug. 1 10 o 1610 „ 28 July 31 6 3 * Hell's Gate. SO miles above Mission, Fraser River. t Yale. 60 miles above Mission, Fraser River. I No Point trap, Vancouver Island. § Beachy trap. Vancouver Island. || Samaquam, Lillooet River, 25 miles above Harrison Lake, 90 miles above Mission. 9 Geo. 5 Migration of Cockeye in Puget Sound and Fraser River. X 63 Table 2.—List of Tags returned—Continued. Tags attached at Station A, Sooke, British Columbia—Concluded. Date Date Region Days ] Date Date Region Days Tag No. Fish Fish where en Tag No. , o 6 6 1903 2 ., ? 9 o 1682 „ 30 1 1 1 1904 2 6 8 4 1686 „ 30 a 2 7 1905 2 ,, 7 8 o 1697 „ 30 July 31 1 1 1907 9 ,, o ? 3 1699 „ 30 Aug. 8 9 9 1920 2 „ 13 9 11 1712 „ 30 „ o 9 6 1929 9 „ 13 9 2 1714 „ 30 1 9 2 1935 2 8 7 6 1721 „ 31 ,, a 8 5 19157 2 4 1 2 1727 „ 31 3 7 3 1958 2 6 7 4 1735 „ 3(1 „ 18 10 IS 1964 2 ,, o * 2 1738 „ 31 i 2, 0 1986 9 i, o 2 3 1739 „ 31 8 7 8 1992 2 9 7 1 1753 „ 31 „ 14 8 14 1996 2 4 7 2 1756 „ 31 ,, o 8 5 2004 2 8 8 6 1761 „ 31 6 7 6 2015 9 F, 8 6 4 1764 „ 31 4 7 4 2018 9 „ 12 10 10 176? „ 31 „ 11 7 11 2024 o 6 6 4 1769 „ 31 9 10 9 2029 9 ,, 7 7 o 1778 „ 31 „ 11 7 11 2032 2 9 8 7 1780 „ 31 „ 16 8 16 2050 4 8 8 4 1783 „ 31 ,, 0 9 0 2053 4 ;6 10. 33 1786 „ 34 „ 18 10 18 2071 4 7 7 3 1796 „ 31 „ 12 7 12 2073 4 „ ii 9 7 179? „ 31 ,, ? 8 7 2074 4 ,, 7 1 3 1801 „ 31 „ 18 8 18 2093 4 ,, ? 2 3 1806 „ 31 „ 13 8 13 2097 4 8 8 4 1814 „ 31 G 6 6 2100 4 9 8 0 1813 „ -31 ,, o 7 5 2103 6 9 8 3 1816 „ 31 ,, 0 8 a 2113 6 9 7 3 1818 „ 31 0 8 6 2116 6 „ 11 7 0 1821 „ 31 », a 6 o Tags attached at Station B, Salmon Banks, Wash. 3007 July 21 July 26 10 5 3118 Ju ly 23 July 24 1 1 3009 ,, 21 „ 29 1 8 3121 , 23 26 1 3 3014 „ 21 „ 99 1 8 3125 , 23 25 2 2 3015 „ 21 >, 24 9 o 3128 , 23 26 10' 3 3024 ,, 21 „ 25 10 4 3129 9|.3 24 1 1 3028' „ 22 „ 24 2 2 3133 , 23 28 10 0 3036 22 Aug. 6 '7 15 3134 , 23 26 4 2 3037 „ 23 July 24 2 2 3143 , 23 29 10 6 3039 22 „ 24 1 2 3146 , 23 28 10 0 3040 „ 22 „ 25 2 3 • 3147 , 23 29 .9 0 3042 22 „ 29 8 7 3148 , 23 29 0 6 3050 2-2 ,, 24 9 2 3149 , 23 • 23 10 1 3052 22 „ 23 2 1 3130 , 23 26 8 3 3060 2.2 „ 24 2 2 3153 , , 23 31 2 8 30-62 „ 22 ,, 25 8 3 3158 , 23 Aug. 1 2 9 3070 „ 23 „ 2-5 1 2 3160 , 23 July 29 9 6 3075 „ 23 „ 24 1 2 3168 , 24 26 9 2 3079 „ 23 „ 25 1 2 3169 , 24 24 1 1 3086 „ 23 „ 27 10 4 3173 , 24 ,, 28 9 4 3087 „ 23 „ 26 fi 3 3174 , 24 29 10 0 3088 „ 23 „ 25 9 2 3175 , 24 25 - 1 1 3089 „ 23 „ 26 2 3' 3176 , 24 28 2 4 3093 „ 23 „ 26 9 o 31SO , 24 Aug. 2 2 9 3103 „ 23 ., 25 9 2 3184 , 24 July 'Mi 9 2 ■3104 „ 23 „ 28 10' 0 3185 , 24 Aug. 2 10 9 3111 „ 23 „ 29 10 6 * Beachy trap, Vancouver Island. X 64 Report of the Commissioner of Fisheries. 1919 Table 2.—List of Tags returned—Continued. Tags attached at Station B, Salmon Banks, Wash.—Continued. Date Date Region Davs Date Date Region Days Tag No. Fish Fish where en Tag No. Fish Fish where en marked. recaught. taken. Route. marked. recaught. taken. Route. 1918. 1918. 1918. 1918. 31S8 July 24 July 2S 10 4 3399 July 31 Aug. 2 2 2 3189 ,. 24 „ 30 6 6 ' 340-2 „ 28 July 30 9 2 3194 „ 24 „ 26 6 2 3403 „ 28 Aug. 1 ?, 4 3198 „ 24 Sept. 15 * 53 3404 „ 28 July 28 1 1 3100 „ 24 July 2,9 10 0 3405 „ 28 „ 28 1 1 3-900 „ 24 ., 25 4 1 3409 „ 28 ,, 28 1 1 3203 „ 24 Aug. 1 7 8- 3410 „ 28 Aug. 2 10 0 3210 „ 24 July 25 6 1 3412 28 9 9 12 3216 u 24 „ 25 1 1 3414 „ 28 July 30 7 2 3222 „ 25 Aug. 2 8 8 3415 „ 2S Aug. 4 7 3223 „ 25 July 30 10 0 3418 „ 28 July 30 9 3226 „ 25 „ 26 6 1 3423 ,, 28 „ 30 9 3232 ;, 9-i „ 30 10 0 3425 „ 28 „ 29 1 1 3244 „ 25 Aug. 7 10 13 3428 „ 28 ., 31 8 3 3245 „ 25 „ 27 33 3432 „ 28 Aug. 8 8 11 3246 „ 9,5 ,. 21 10' 27 3433 „ 28 July 31 6 3 3251 „ 93 July 28 10 3 3434 „ 28 „ 30 10 2 3233 „ "5 .. 28 6 1 3437 28 Aug. 5 1 8 ,, 25 Aug. 19 10 23 3441 „ 28 ,, 9 10 6 3257 „ °5 July 29 10' 4 3442 „ 28 July 30 9 3259 „ 25 Aug. 8 2 14 3443 „ 28 Aug. 13 9 16 S'efi ,, 25 1 S 7 3444 „ 28 July 29 1 1 3268 „ 26 July 31 1 0 3449 „ 28 ,, 29 1 1 39fi9 „ "6 „ 99 7 3 3446 ,, 28 „ 30 2 3270 „ 26 „ 28 1 2 3448 „ 28 ,. 31 8 3 3272 „ 26 ,, 29 8 3 3450 „ 28 Aug. 1 9 4 3273 „ °6 „ 28 10 2 3451 „ 28 July 31 10 3 3974 „ 26 „ 28 1 2 3432 „ 28 „ 30 2 3273 ,, 2fl „ 28 1 2 :343? „ 31 Aug. -6 7 6 3278 „ 26 „ 30 10 4 3460 „ 31 4 8 4 3281 „ 26 „ 31 t 0 3461 „ 31 3 8 3 3286 „ 26 „ 29 7 3 34'8'2 ...... „ 31 6 8 fi 3288 „ 26 , ,28 1 2 3463 „ 31 „ 30 II 30 3290 „ 26 „ 29 1 3 3-469 „ 31 1 1 1 3294 „ 26 90 9 3 3471 „ 31 1 '7 1 3995 „ 26 „ 28 1 9 3474 „ 31 9 6 2 3297 „ 26 „ 31 1 0 3475' ,. 31 2 1 3298 „ 26 „ 26 10 1 3478 „ 31 1 1 1 3299 „ 26 Aug. 18 8 23 3481 „ 31 1 8 1 3301 „ 26 July 30 10 4 3484 „ 31 ,> 8 2 0 3309 „ 26 ,. 29 1 3 3486 „ 31 1 1 3312 „ 26 Aug. 1 9 0 3487 „ 31 2 8 2 3317 „ 26 ,, 0 7 10 3490 „ 31 4 7 4 3321 „ 26 Sept. 15 § 51 3495 „ 31 2 8 2 3322 „ 20 July 30 10 4 348*0 „ '31 6 1 6 3326 ,, 26 ,, 29 1 3 3501 Aug. 1 9 1 1 3328 „ 26 „ 29 9 3 3502 1 >, 0 8 4 3399 „ 26 „ 31 1 0 3508 1 9 6 1 3330 „ 26 Aug. 22 9 27 3310 1 „ H12 10 11 3335 „ 26 July 29 10 3 3513 1 4 8 3 3337 ,, 26 „ 30 10 4 3514 1 1 1 3339 „ 26 Aug. 5 9 10 3518 1 „ 15 10 14 3340 „ 26 July 30 9 4 3522 1 2 6 1 3350 „ 26 „ 31 6 0 3523 1 2 1 1 3351 „ 26 „ 29 9 3 3524 1 ,, 0 8 4 „ 26 „ 30 9 4 3526 1 2 9 1 3335 „ 26 „ 29 8 3 3527 1 4 8 3 3336 „ 26 „ 31 10 0 3528 1 6 8 5 3357 „ 26 „ 28 9 2 3529 1 4 10 3 3358 „ 26 „ 29 10 3 3335 1 ,, 0 7 4 3361 „ 26 „ 29 6 3 3537 1 2 6 1 3366 „ 26 Aug. 8 9 13 3541 1 ,, 0 2 4 3367 „ 26 July 28 1 2 3544 1 ,, 0 9 4 3368 „ 26 Aug. 1 8 6 3545 1 2 6 1 3372 „ 26 July 30 10 4 3547 1 ,, 0 4 3373 „ 26 „ 31 10 0 3548 . , 1 3 8 2 3377 „ 26 „ 31 ■ 8 5 3330 1 4 2 3 3378 „ 30 Aug. 4 7 0 35-51 1 2 6 1 3382 „ 30 July 31 6 1 3552 1 4 10 3 3394 „ 30 Aug. 5 i 1 3553 1 ,, 7 8 6 3397 „ 31 1 7 1 1 4 7 3398 „ 31 " ! 7 4 1 4 8 3 * Four miles above Pitt Lake, 30 miles from Fraser River, 50 miles trom ocean, t Soda Creek, 280 miles above Mission. I Devil's Run, 10 miles above Mission. § Four-mile Creek, Pitt River, 50 miles above Fraser River, 50 miles from ocean. II Hagenson Slough, 30 miles above Fraser, 50 miles from ocean. 11 The record is September 12th, but this is possibly a mistake, and the more probable date is August 12th, as tabulated. ** Fraser River, 5 miles above Yale, 60 miles above Mission. 9 Geo. 5 Migration of Cockeye in Puget Sound and Fraser River. X 65 Table 2.—List of Tags returned—Continued. Tags attached at Station B, Salmon Banks, Wash.—Concluded. Date Date Region Davs Date Date Region Days Tag No. Fish Fish where en Tag No. Fish Fish where en marked. recaught. taken. Route. marked. recaught. taken. Route. 19-18. 1918. 1918. 1918. 3556 Aug. 1 Aug. 5 8 4 3699 Aug. 4 Aug. 8 9 4 1 .. 11 8 10 3700 4 6 8 2 3562 1 2 6 1 3702 4 , 0 6 1 3567 1 1 4 1 3703 4 9 10 0' 3574 1 4 8 3 3704 4 6 7 2 3577 1 4 7 4 3705 4 , 1 10 3 3378 , X 2 6 1 3711 4 6 2 2 3579 1 4 ? 3 3713 4 6 7 9 3582 1 6 4 o 3714 4 9 9 a 3583 1 6 8 D 3715 4 6 9 2 3584 1 , 12 7 11 3718 4 , 19 10 15 1 , 3 0 4 3721 4 , 7 9 3 3387 1 2 1 1 3722 4 9 10 0 3588 .- 1 j o 1 4 3727 4 , 1 8 3 3589 1 7 8 6 3798 4 7 9 3 3590 1 , o 8 4 3730 4 6 9 2 3591 1 i o 10 4 4 6 7 2 3394 1 6 6 0 3737 4 , 0 7 1 3598 1 2 6 1 3730 4 , o! 1 1 3605 2 , 3 8 3 ■ 3741 4 6 ? 2 3606 2 , o 9 3 3742 4 8 -8 4 3610 2 , 14 8 12 3743 4 7 6 3 3613 2 , o 8 3 3744 4 Sept. 16 -!- 43 3617 9 4 1 9 3746 4 Aug. 8 10 4 3619 9 , 5 7 3 3747 4 , 11 7 7 3'6->4 9 , 0 1 3 37-52 4 , 23 9 19 3625 9 7 8 o 3754 4 , 1 1 3 3630 2 4 1 2 3756 4 6 7 2 3632 2 4 1 2 3757 4 , 0 1 1 3633 9 , 7 8 5 3760 5 9 9 4 3634 9 , 7 9 0 3762 , o , 7 2 2 3635 9 4 1 2 3764 , o 9 8 4 3637 9 , 14 2 12 3773 , 0 7 8 2 3639 2 , *17 10 15 , o , 13 7 8 3642 9 , o 9 3 3777 , 5 9 7 4 3643 9 , o 9 3 3778 , o 12 9 7 3644 9 i o 7 3 3779 > o Sept. 20 § 46 3646 9 , 0 9 3 3780 , o Aug. 7 6 2 3648 9 , 0 8 3 3783 , 5 , 10 8 0 3653 9 , o i 3 3789 0 8 8 3 3658 2 6 10 4 3790 , o 8 9 3 3662 , 2 6 10 4 3791 , o , 7 7 2 3663 2 7 8 0 3792 , o 6 7 1 3667 4 , o 1 1 3794 ..... , o 8 2 3 3670 4 , 10 8 6 379-5 o 9 9 4 3672 4 6 6 2 3797 , 0 , 12 10 7 3673 , 4 6 2 2 3800 i o , 18 9 13 3677 4 6 9 2 3801 7 , 12 9 0 3678 , 4 9 8 0 3803 7 , 10 9 3 3680 4 , 20 10 16 3806 , 7 , 10 8 3 3682 4 , 7 8 3 3807 , 7 9 7 2 3685 4 7 8 3 3812 , - 7 , 11 8 4 3687 4 7 8 ■ 3 3818 7 Sept. 3 II 27 3688 4 , IS 9 14 3819 , 7 Aug. 14 9 7 3693 4 Sept. 17 t 44 3820 7 „ 12 10 0 3694 4 Aug. 14 2 10 3826 7 „ 13 10 6 3695 4 6 8 2 3829 . 7, „ 1118 10 11 3696 4 6 7 2 383:5 , 7 „ 10 8 3 Tags attached at Station C, Point Paktkidge, Wash. 1001 July 17 July 19 7 2 1023 July 17 July 20 4 3 1003 , 17 , 20 4 3 1030 , 17 99 1 0 1010 , 17 , 22 4 0 1031 , 1? „ 21 10 4 1011 , 17 , 23 9 6 1036 , 1? „ 20 1 3 1017 , 17 , 20 8 3 1037 , 17 „ 25 1 8 1018 , 17 23 9 6 103S , 17 Aug. 17 ** 31 1019 , 17 22 1 0 1041 17 July 24 ' 7 * The record is September 17th, but this is possibly a mistake, and the more probable date i > August 17th, as tabulated. t Birkenhead River, 25 miles above Lillooet Lake, 130 miles above Mission, Fraser River. X Fraser River, 65 miles above Mission. § Birkenhead River, 25 miles above Lillooet Lake, 130 miles above Mission, Fraser River. ]| Four-mile Creek, Pitt River, 30 miles above Fraser River, 50 miles from ocean. If The record is September ISth, but this is possibly a mistake, and the more probable date i 5 August 18th, as tabulated. *• Devil's Run, 10 miles above Mission, Fraser River. 5 X 66 Report of the Commissioner of Fisheries. 1919 Table 2.—List of Tags returned—Continued. Tags attached at Station C, Point Pabteidge, Wash.—Continued. Tag No. 1044 1050 1052 1055 1057 1038 10 50 1061 1063 1067 1069 1075 1078 1079 1080 1085 1088 1090 1092 1094 1098 1104 1106 1107 1109 1111 1116 1124 1128 1129 1130 1132 1133 1138 1151 1156 1157 1163 1165 1167 1169 1173 1175 1176 1177 1185 1192 1193 1194 1196 1197 1199 1205 1207 1212 1213 1221 1223 1228 1229 1231 1235 1238 1239 1241 1242 1244 1249 1251 1264 1269 1278 1285 1289 1200 1293 1296 Date Fish marked. 191S. July 22 22 92 22 22 22 9 2 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 24 26 26 26 26 26 26 26 26 26 26 Date Fish recaught. Region where taken. 1918. July 24 22 24 92 22 9 9 23 21 92 24 Aug. 9 July 22 99 24 22 22 21 24 30 26 23 23 24 25 Aug. 12 Julv 24 25 24 30 30 28 29 29 24 26 31 Aug. 8 10 1 July 29 30 Aug. o 6 July 29- 31 25 31 30 Aug. 10 July 29 29 31 30 Aug. 1 July 29 Aug. 7 .lulv 30 31 31 29 Aug. 0 Julv 30 29 Aug. 18 July 29 Aug. 1 .lulv 2.9 Aug. 6 Julv 29 30 31 Aug. 1 Julv 29 Aug. 18 July 29 Aug. 0 July 30 1 1 2 10 1 8 9 10 1 10 2 10 2 1 10 9 * 1 9 6 2: 9 10 10 8 9 1 2 10 7 10 Davs Date Date Region Davs en Tag No. Fish Fish where en Route. marked. recaught. taken. Route. 1918. 1918. 5 1307 July 26 Aug. 1 8 6 3 1308 „ 26 ,, o 10 10 0 1309 „ 26 2 2 7 3 1310 „ 26 July 30 9 4 3 1313 „ 26 „ 28 8 9 3 1317 „ 26 Aug. 2 9 7 4 1320 „ 26 „ 11 8 16 2 1322 „ 26 ,, 0 10 10 3 1328 „ 26 July 30 6 4 5; 1330 „ 26 „ 30 1 4 14 1334 „ 2:6 „ 30 10 4 3 1337 „ 26 „ 30 9 4 3 1344 „ 26 „ 29 1 3 0' 1346 „ 26 „ 29 1 3 3 1347 „ 26 „ 29 1 3 3 1351 ,, 26 Aug. 4 7 9 1353 „ 26 1 8 6 o 1354 „ 26 1 8 6 11 1355 „ 26 1 10 6 4 1357 „ 26 9 6 14 1 1360 „ 26 1 2 6 1 1363 „ 26 „ 12 7 17 3 1368 „ 26 4 7 9 3 1368 .. 26 July 29 ■7 3 21 1399 ,, 26 „ 30 9 4 2 1372 „ 26 Aug. 16 9 27 3 1373 ,, 26 „ »8 10 2-3 1374 „ 26 July 30 8 4 8 1375 „ 26 „ 29 1 3 8 1377 „ 26 Aug. 6 2 11 6 1380 . „ 26 1 9 6 7 1424 Aug. 6 7 2 1 7 1438 ■6 „ 11 7 0 1548 7 9 1 2 4 1559 7 ,, 11 8 7 7 1567 7 >> 7 2 1 15 1569 7 „ 11 8 4 17 7 ,, 11 8 4 8 1577 7 „ 14 10 7 o 1579 ,, 7 9 2 2 6 1591 ,, 7 ,, 13 8 '5 12 1598 7 „ 13 4 6 13 1604 ,, ? „ 19 6 9 1609 7 „ 13 8 6 1 7 6 17 1617 7 9 7 2 1628 7 „ 11 8 4 1639 „ 7 „ 13 9 0 1673 9 ,, 16 8 7 1677 9 „ 13' 9 4 1681 9 „ 18 8 9 7 6 8 5 14 6 7 7 1687 9 „ 14 6 j> 1689 9 „ 15 8 6 1691 9 „ 14 9 ■0 1695 9 „ 13 8 4 1704 9 „ 12 6 3 1705 9 „ 13 8 4 1717 9 „ 13 7 4 1719 ,, 9 „ 13 9 4 1721 9 ;, 14 8 0 10 4 3 1?9'3 9 „ 11 1 2 1724 9 „ 12 6 3 1732 9 „ 12 7 3 1735 9 „ 22 9 13 3 6 3 1736 9 „ IS 10 9 1741 9 9 2 1 1743 9 „ 13 8 4 11 1746 9 „ 18 8 9 3 1749 9 „ 12 8 3 4 1762 9 „ 23 • 6 14 5 1763 9 „ 13 2' 4 6 1768 9 „ 14 9 5 3 1769 9 „ 13 8 4 23 1773 „ 13 „ 15 8 2 3 1787 „ 13 „ 21 9 8 10 1793 „ 13 „ 18 9 0 4 1796 „ 13 „ 13 2 1 * Ebey's Landing, below Point Partridge. t Otter Point, Vancouver Island. X The record is September 18th, but this is possibly a mistake, and the more probable date is August 18th, as tabulated. 9 Geo. 5 Migration of S-ockeye in Puget Sound and Fraser River. X 67 Table 2.—List of Tags returned—Continued. Tags attached at Station C, Point Partridge, Wash.—Concluded. Date Date Region Days Date Date Region Days Tag No. Fish Fish where en Tag No. Fish Fish where en marked. recaught. taken. Route. marked. recaught. taken. Route. 1918. 1918. 1918. 1918. 1807 Aug. 13 Aug. 17 9 4 1907 Aug. 13 Aug. 15 1 2 1812 „ 13 „ 14 2 1 „ 13 ,, 16 9 3 1820 „ 13 , 19 8 3 1931 ,. 14 „ 19 7 0 1835 „ 13 , 14 9 1 1932 „ 14 „ 19 7 5 1846 „ 13 , 16 4 3 1933 ,, 14 ■ „ 23 10 9 1831 „ 13 , 16 8 3 1935 „ 14 „ 16 6 2 1853 „ 13 , IS 8 0 1939 „ 14 „ 16 2 1854 „ 13 , 16 8 3 1945 ,, 14 Sept. 3 10 19 1856 „ 13 , 14 1 1 1932 „ 14 Aug. 19 ? 0 186-5 „ 13 , 15 1 '-> 1957 „ 14 „ 17 1 3 1867 „ 13 , 19 7 6 1969 „ 14 „ 16 2 2 1878 „ 13 , 16 1 3 1970 „ 14 „ 16 2 2 1-887 „ 13 92 9 9 1976 „ 14 „ 19 2 0' 1889 „ 13 , 15 9 2 1982 „ 14 „ 15 2 1 1891 „ 13 , 16 8 3 1986 „ 14 „ 20 8 6 1897 „ 13 , 15 8 2 1993 „ 14 ,, 15 2 1 1899 „ 13 , 19 7 6 1999 „ 14 „ 25 10 11 1900 „ 13 , 15 2 2 „ 14 „ 16 2 2 Tags attached at Station D, Lummi Island, Wash. 2002 July 19 July 30 9 11 2177 July 23 July 24 8 1 2003 „ 19 99 8 3 2181 „ 24 Aug. 2 10 9 2013 „ 19 92 8 3 2183 ,, 24 July 25 9 1 2015 „ 19 22 8 3 2185 „ 24 ,, 29 9 0 2020 „ 19 22 10 3 2186 „ 24 „ 25 8 1 2031 „ 19 22 8 3 2189 „ 24 „ 28 9 4 2033 „ 19 ,. 23 9 4 2190 „ 24 „ 26 9 2 "037 „ 19 Aug. 6 8 18 2201 ' „ 24 Aug. 6 9- 13 2038 „ 19 July 21 8 2 2210 „ 24 ,. 12 10 19 2039 „ 19 22 8 3 2214 „ 24 July 26 9 2 2040 „ 19 „ 25 9 6- 2216 „ 24 „ 25 9 1 2044 „ 19 „ 25 9 3 2221 „ 24 „ 29 10 0 2049 „ 19 „ 25 8 O 2226 „ 24 „ 26 9 2 9051 „ 19 Aug. 9, 8 21 2231 „ 24 1, 29 9 0 2052' „ 19 July 22 8 3 9->32 „ 24 „ 27 9 3 20pi5 „ 19 92 7 3 2234 „ 24 „ 28 9 4 2057 „ 19 „ 23 a 4 2237 „ 24 „ 29 10 0 2061 „ 19 „ 22 8 3 2238 „ 24 „ 26 9 2 2063 „ 19 „ 24 8 1 2240 „ 24 „ 29 9 0 2065 ,, 19 „ 22 8 3 2241 „ 24 „ 26 9 2 2069 „ 19 „ 22 8 3 9942: „ 24 „ 30 9 6 2072 „ 19 „ 21 9 2 2246 ,. 2-4 „ 25 8 1 2073 „ 10 „ 24 8 0 2250 „ 24 „ 28 8 4 9075 ,. 19 ., °4 7 0 2251 „ 94 Aug. 5 10 12 2078 19 ,, 22 8 3 2252 „ 28 „ 30 10 33 2079 „ 19 99 9 3 2253 ,, 28 „ 7 9 10 2082 „ 19 „ 23 9 4 2255 „ 28 July 29 9 1 2084 „ 19 Aug. 16 * 28 2257 „ 28 ,, 30 8 2 2088 „ 23 July 26 8 3 2259 „ 28 Aug. 1 9 4 2090 „ 23 „ 24 8 1 2261 „ 28 July 30 8 2 2092 „ 23 „ 24 8 1 2263 „ 28 „ 29 8 1 2094 93 ,, 24 8 1 2268 „ 28 Aug. 26 t 29 '095 „ 23 „ 24 8 1 2269 „ 28 July 31 8 3 2096 „ 23 „ 26 8 3 22 iO „ 28 ., 30 9 2098 „ 23 „ 24 8 1 2271 „ 28 Aug. 1 10 4 9100 „ 23 Aug. 2 t 10 2280 „ 28 July 30 8 9 2103 „ 23 July 26 8 3 2282 „ 28 „ 30 9 2 2108 „ 23 „ 25 9 2 2285 „ 28 AUg. 11 10 15 2112 „ 23 „ 24 9 1 2286 „ 28 July 29 1 1 2122 „ 23 „ 24 8 1 2287 „ 28 „ 30 9 2 2123 „ 23 „ 96 8 3 2288 „ 28 „ 30 9 2 2137 „ 23 „ 24 8 1 2289 „ 28 Aug. 8 10 11 2147 „ 23 95 8 2 2290 „ 28 July 30 8 2 9148 „ 23 „ 30 10 7 2291 „ - 28 Aug. 1 9 3 2158 „ 23 „ 24 8 1 2293 „ 28 July 30 9 2 2160 „ 23 ., 24 9 1 230O „ 28 „ 29 8 1 2161 „ 23 Aug. 13 10 21 2302 „ 28 „ 30 9 2. 2166 „ 23 July 26 10 3 2.304 „ 28 „ 29 8 1 2171 „ 23 „ 25 9 2 2310 „ 28 „ 29 1 1 2173 „ 23 „ 24 8 1 2312 „ 28 „ 30 9 2 2174 „ 23 „ 24 8 1 2313 „ 28 „ 29 8 1 2175 „ 23 „ 31 8 8 2314 „ 28 „ 30 9 2 * Soda Creek Canyon, 280 miles above Mission, Fraser River. t Yale, 60 miles above Mission, Fraser River. X Hell's Gate, 80 miles above Mission, Fraser River. X 68 Report of the Commissioner of Fisheries. 1919 Table 2.—List of Tags returned—Continued. Tags attached at Station D, Lummi Island, Wash.—Continued. Date D-te Region Davs Date Date Region Davs Tag No. Fish Fish where en Tag No. Fish Fish where en marked. recaught. taken. Route. marked. recaught. taken. Route. . 1918. 1918. 1918. 101S. 2315 July 30 Aug. 30 * 31 2485 Aug. 5 Aug. 7 9 2 2319 „ 30 1 9 2 "487 5 8 9 2323 „ 30 2 8 3 2488 ,, o „ ? 9 9, 2324 „ 30 1 9 2 •'492 „ 5 ,, ? 9 2 2328 ,, 30 1 9 2 2493 >, 5 6 8 1 2330 ,, 30 July 31 10 1 249-5 „ o 6 8 1 2331 „ 30 Aug. 2 10 3 9407 „ i> 6 8 1 2335 „ 30 2 8 3 2501 j, o 6 9 1 2338 „ 30 July 31 10 1 2S013 ...:. ,, o t23 10 18 2344 „ 30 „ 30 8 1 •'507 ,, o „ 20 § 46 9346 „ 30 Aug. 2 2 3 2510 ,, 0 ,. 7 8 2 2347 „ 30 1 8 2, 2512 ,, o 6 8 1 2349 „ 30 July 31 9 1 2515 ,, 5 9 9 4 2358 „ 30 Aug. 1 8 2 "516 ,, o 6 9 1 2359 „ 30 9 10 3 2517 „ 5 ,, 7 9 2 2365 Aug. . 2 4 8 2 2518 „ o 6 8 1 2367 2 „ zy 7 3 2522 i> o 7 8 2 2368 2 „ o 10 3 2523 ,, o 6 8 1 2370 9 4 8 2 2529 j j o 7 8 2 2373 9 4 8 2 2530 ,, o 6 8 1 2375 2 4 8 2 2531 » o ,, ? 8 2 2376 9 „ o 9 3 2333 ,, 3 6 9 1 2381 9 ,, o 9 3 2539 ,, o 7 8 2 2384 9 ,, 0' 8 3 2541 ,, o 6 8 1 2389 9 9 9 i 2542 ,, o 8 8 3 2391 9 6 8 4 2544 „ o .; 7 8. 2 2394 2 „ 25 10 23 2546 ,, 5 „ 7 8 2 2401 9, 4 9 2 2547 ,, 5 6 8 1 2403 9 >, o 9 3 2548 „ o 8 ID 3 2404 9 „ 13 10 11 2549 ,, o 8 S' 3 2405 2 ii o 9 3 2531 ,, E» 6 8 1 .2410 2 9 9 1 2532 ,, o 8 10 3 2413 9 9 9 7 2354 ,, 5 a ? 8 2 2414 2 ,, o 9 3 2556 6 „ ? 9 1 2416 2 S 9 6 "557 6 8 8 2 2420 4 „ ? 8 O 2558 6 9 10 3 2421 4 ,, o 8 1 2559 6 6 9 1 2424 4 6 8 2 2560 6 ,, ? 9 1 2496 4 4 7 1 2566 6 9 10 3 9495 4 „ o 7 1 2569 6 9 9 3 2429 4 „ tl9 10 15 2570 6 6 9 1 2431 4 ,, o 7 1 2574 6 9 9 3 2432 4 „ o 7 1 2575 6 '9 10 3 2434 4 ,, 7 8 o 257© 6 „ 16 II 10 2435 4 6 8 9 2578 6 „ 11 7 o 2436 4 6 9 2 9579 6 9 8 3 2437 4 ,, o 8 1 2580 6 9 8- 9438 4 „ o 8 1 2583 6 7 8 1 2439 4 4 7 1 2384 6 ,, 10 8 4 2440 4 ,, o 9 3 2585 6 „ 10 8 4 2441 4 „ o 8 1 2587 6 „ 11 8 '0 2442 4 „ 16 8 2 2588 6 8 8 2 2443 4 6 9 2 9589 6 „ ? 9 1 2444 4 ,, o 9 1 2591 6 ,, 7 9 1 2443 4 „ o 8 1 2592 6 7 9 1 2449 4 „ o 8 1 2593 6 6 9 1 2453 4 ,, o 8 1 2595 6 9 8 3 2455 4 » o 8 1 2398 6 „ 12 8 6 2456 4 6 8 9 2002 6 9 9 3 2457 4 6 8 2. 2603 6 9 10 3 2459 4 7 9 3 2604 6 8 8 9 2464 4 ,, 5 9 1 9605 6 9 0 3 2465 4 Sept. 3 10 30 2610 6 9 9 3 2466 4 Aug. 6 9 2 2611 6 ,, ? 8 1 2469 4 >, o 8 1 2612 6 „ 12 8 6 2471 4 6 8 2 9614 6 8 9 9 2472 4 ,, -o 9 1 9616 6 7 9 1 2474 4 6 9 2 9618 6 8 10 2 2475 4 6 8 9 2621 6 9 8 3 2476 4 6 8 2 2622 6 „ 10 9 4 2478 „ 5 6 8 1 2693 6 „ 11 9 0 2482 ,, o 7 8 2 2625 6 8 9 2 « Lytton, 110 miles above Mission, Fraser River. t The record is September 19th, but this is possibly a mistake, and the more probable date i s August 19th, as tabulated. X The record is September 23rd, but this is possibly a mistake, and the more probable date I s August 23rd, as tabulated. § Birkenhead River, 25 miles above Lillooet Lake, 130 miles above Mission, Fraser River. || Strawb erry Island , 50' miles above Mi ssion, Fr tser River. 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 69 Table 2.—List of Tags returned—Continued. Tags attached at Station D, Lum mi Island, Wash.—Concluded. Date Date Region Days Date Date Region Days Tag No. Fish Fish where en Tag No. Fish Fish where en marked. recaught. taken. Route. marked. recaught. taken. Route. 1918. 1918. 1918. 191S. 2627 Aug. 6 Aug. 7 8 1 2678 Aug. 7 Aug. 10 8 3 2628 0 ,. 7 9 1 2682 ,, 7 „ 12 10 0 2629 6 „ 11 10 o 2683 ,, 7 9 8 2 2630 6 8 9 2 2684 ,, 7 9 9 9 2631 6 „ 11 7 o 2685 ,, 7 9 9 2 2633 6 9 9 3 2686 7 „ 12 9 0 2634 6 „ 15 9 9 2687 ,, 7 8 8 1 2633 6 ,, 7 9 1 2688 ,, ? 8 8 1 2636 6 - „ 8 8 2 2689 » 7 8 9 1 2642 6 8 9 9 2692 9 „ 11 8 9 2645 6 ,, ? 8 1 2699 9 „ 11 8 2 2646 6 8 8 9 2707 9 „ 13 9 4 2648 6 Sept. 19 * 44 2710 9 Sept. 4 X 26 2651 7 Aug. 12 8 0 2713 9 Aug. 16 8 7 2635 7 9 8 9 2717 9 „ 12 9 3 2657 ,, 7 8 8 1 2718 9 „ 14 10 0 2659 7 8 9 1 2719 9 „ 12 10 3 2681 ,' 7 8 8 1 2720 9 „ 15 8 6 2662 ,, i 9 9 2 2722 9 „ 13 9 4 2666 7 „ 10 8 3 2726 9 ,, 14 9 0 2669 ,, 7 „ 10 9 3 2728 9 „ 12 10 3 2670 7 9 9 2 2733 9 „ 11 8 9 2672 ,, 7 9 9 9 2734 9 ,,11-13 9 o 2675 ,, 7 9 8 9, 2736 9 ,, 14 9 0 2676 ,, 7 „ 16 1 40 2737 9 „ 19 ■ 7 10 Tags attached at Station E, Point Roberts, Wash. 4001 4007 4008 4012 4025 4047 4065 4071 4103 4104 -1107 4109 4117 4121 4122 4129 4138 4142 4130 4166 4174 4183 4187 4196 4198 4203 4207 4209 4211 4216 4219 4220 4921 4246 July 25 23 25 25 25 25 25 25 26 26 26 26 26 26 26 26 26 26 26 29 29 29 29 29 29 29 29 29 29 29 29 29 29 29 July 29 Aug. 4 .lulv 26 Aug. o July 30 29 28 29 Aug. 3 July 29 30 29 31 30 28 29 29 28 Aug. o 9 16 2 4 12 Sept 12 Aug. 9 July 30 Aug. 0 July 30 Aug. 2 July 31 31 30 Aug. 1 10 7 10 § 10 II 10 10 10 10 10 10 10 IO 10 10 10 10 II 10 10 10 tt 8 10 10 10 9 10 10 10 9 4 4249 11 4252 1 4258 11 4264 o 4267 4 4268 3 4270 4 4271 8 4274 ...... 3 4277 4 4279 3 4283 o 4*84 4 4286 2 4288 3 4289 3 4291 2 42, o- „ 20 9 15 4407 „ 31 „ 15 10 15 4633 ,, o 9 6 4 4415 „ 31 9 10 2 4639 9 0 „ 20 10 15 4417 „ 31 ,, o 10 6 4646 „ o 9 10 4 4423 „ 31 1 10 1 4653 ,, »' Nov. 16 §§ 103 4432 „ 31 1 9 1 4654 ,, 5 Aug. 11 10 6 4434 „ 31 July 31 10 1 4657 ,, & „ 13 10 8 4436 „ 31 Aug. 30 10 30 4660 5 7 10 2 4438 „ 31 „ 11 7 11 4663 » o 9 10 4 1446 „ 31 2 10 2 4664 ,, 5 „ 11 9 6 4447 „ 31 „ 21 10 21 4669 ;, 3 6 10 1 4454 .. 31 6 8 6 4670 9 O „ 15 10 10 4461 „ 31 „ i 10 7 4677 ,i 0 7 10 2 4464 „ 31 ,, 7 10 7 4686 ,, a „ 18 10 13 4467 „ 31 ,, o 8 5 4687 3 8 10 3 4468 „ 31 „ 10 8 16 4691 9 0 „ 11 7 6 4474 „ 31 „ 20 10 20 4692 . ,, 5 ., 16 9 11 4476 „ 31 ,, o 9 0' 4694 ,, 3 Sept. 6 10 32 4478 „ 31 9 10 9 4708 ,, o Aug. 12 9 i 4480 „ 31 ,, o 10 0 4l'09: ,, 0 ,. 12 10 7 4481 „ 31 7 10 7 4713 ,i o Sept. 5 10 31 44S2 „ 31 „ 28 § 2S 4714 ,, o Aug. 12 10 7 4483 ,, 31 9 10 2 4732 ,, o ,, 20 10 13 4483 „ 31 ,, o 9 0 4733 ,, o1 6 10 1 4486 ,1 31 2 10 9 4735 ,> a 9 9 4 4487 „ 31 ,, 5 8 O 473S ,, o „ 26 10 21 4488 „ 31 1 10 1 4740 9 0 9 10 4 4489 „ 31 July 31 10 1 4751 6 9 10 3 4491 „ 31 Aug. 1 9 1 4753 6 „ 12 9 G 4492 „ 31 6 -10 6 4 15 4 6 6 10 1 4499 Aug. 1 4 10 8 4756 6 18 10: 12 450O 1 „ lis 10 7 4739 fi „ 30 10 24 4502 1 „ 12 9 11 4760 6 „ 16 10 10 4306 1 „ 19 10 18 4765 6 8 10 2 430-8 1 „ 13 10 12 4773 6 „ 12 10 6 4509 .1 4 10 3 4776 6 „ 20 10 14 4511 1 Sept. 2 10 32 4777 6 ,, 7 10 1 4315 1 4 10 3 4780 6 „ 11 9 -o 4516 1 „ 12 10 11 4784 6 9 9 3 4521 1 6 10 5 4790 6 „ 18 10 12 4524 1 „ 21 10 20 '4791 6 7 10 1 1 9 10 1 4<96 6 ,, 19 10 13 4526 1 „ 6 10 o 4799 6 „ 26 10 20 4533 1 2 10 1 4800 9 6 „ 13 10 7. 4534 9 >, o 10 3 4801 6 6 10 10 4540 2 >> 5 10 3 4803 6 ,, 7 10 1 4544 2 „ 11 10 9 4808 6 8 10 2 2 ,, 7 1 5 480-9 6 „ 30 10 24 4547 2 ,, 5 ** 3 4812 6 „ 12 10 6 2 4 10 9 4813 6 8 10 2 4556 2 ,, o 10 3 4S14 6 8 10 2 4558 9 4 10 2 4818 6 ,,111122 10 19 4560 9 „ 12 8 10 4823 ,, 7 „ 13 10 6 4562 2 8 10 6 4825 7 9 9 2 4569 2 4 10 2 4828 ,, ? „ ? 10 1 * Skookumchuck, on Lillooet River, 12 miles above Harrison Lake, 73 miles above Mission. t Birkenhead River, 25 miles above Lillooet River, 130 miles above Mission, Fraser River. X Portage Creek, foot of Anderson, 170 miles above Mission. § Hell's Gate, 80 miles above Mission. || The record is September 8th, but this is possibly a mistake, and the more probable date is August Sth, as tabulated. \\\\ Devil's Run, 10 miles above Mission, Fraser River. ** Three miles above Mission. Fraser River. ft Trafalgar Flat. 50 miles above Mission. Fraser River. XX Samaquam. Lillooet River, 25 miles above Harrison Lake, 90 miles above Mission. §§ Mouth of Seymour Creek, Burrard Inlet, at North Vancouver. ill The record is September 22nd, but this is.possibly a mistake, and the more probable date is August 22nd, as tabulated. 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 71 Table 2.—List of Tags returned—Concluded. Tags attached at Station E, Point Robekts, Wash.—Concluded. Date Dr.te Region Days Date Date Region Days Tag No. Fish Fish where en Tag No. Fish Fish where en marked. recaught. taken. Route. marked. recaught. taken. Route. 1918. 1918. 1918. 1918. 4830 Aug. 7 Aug. 16 * 9 74 Aug. 15 Sept. 24 J 40 4831 ,, 7 ,, ? 10 1 „ 15 „ 21 t 37 4833 7 „ 10 10 3 83 „ 15 „ 15 % 31 4836 8 „ 18 10 10 -84 „ 15 „ 10 10 26 4847 8 „ 12 10 4 90 „ 16 9 § 17 4861 8 8 10 1 100 „ 19 Aug. 19 10 1 4869 8 „ 10 10 2 113 „ 19 „ 22 10 3 4870 8 „ 19 t 11 126 „ 20 „ 21 10 1 4879 8 „ 18 8 10 127 „ 20 22 10 2 4883 8 „ 12 9 4 130 „ 20 Sept. 5 10 16 4886 8 „ 10 10 2 137 „ 20 Aug. 28 10 8 4896 » 8 12 10 4 140 „ 20 22 10 2 4913 8 „ 12 10 4 149 „ 20 22 10 2 4914 8 9 10 1 151 „ 20 22 10 4921 8 „ 11 10 3 136 ,. 21 „ 21 10 1 36 „ 15 „ 18 10 3 164 „ 21 Oct. 7 t 47 38 „ 15 „ 16 10 1 165 „ 21 Aug. 21 10 1 72 „ 15 Sept. 20 X 36 167 .... „ 21 „ 1127 10 6 * American Bar, 50 miles above Mission, Fraser River. t Trafalgar Flat, 50 miles above Mission, Fraser River. X Birkenhead River, 25 miles above Lillooet Lake, 130 miles above Mission, Fraser River. § Samaquam, Lillooet River, 25 miles above Harrison,Lake, 90 miles above Mission. || The record is September 27th, but this is possibly a mistake, and the more probable date is August 27th, as tabulated. , Peecentage of Returns. Talile 3 shows the percentage of returns from each day's marking at each station. The data from which the percentages were obtained are also given. In general the returns indicate that the marking was quite uniformly successful. This is especially true with that done at Stations B and D. The marking done at Station A suffered somewhat in efficiency for a few days at the time the change was made in the personnel, hut otherwise is satisfactory. The records from Station E show a sudden diminution in the percentage of returns on August Sth and for the following three days on which tish were marked. No explanation can be given for this. As noted (note to Table 1, Station C), the accuracy of the August records for Station C is subject to considerable question, and the operator at this point has admitted reporting incorrectly the number of fish marked on August 6th. But 2 out of 164 reported marked on this date were recovered, a much lower percentage than was obtained from any other day's marking. These facts have thrown the record of marking done at this station during August so much under suspicion that it was considered necessary, in certain phases of the study, to disregard entirely the returns. It is important to call attention to the fact that the figures given In Table 3 cannot be accepted as giving any adequate idea of the percentage of fish entering from the ocean which are caught while passing through the waters where commercial fishing is permitted. Several indeterminate factors must modify the percentage of returns to such an extent that, while they are reasonably comparable inter se, the actual figures give a much-distorted idea of the toll taken from the run as it is passing through the sound and river. Two of these factors are especially obvious, i.e.: (1) The figures as given here do not include all of the actual returns since some were omitted on account of faulty or incomplete data; the error from this factor, however, is not great, approximately 5 per cent.; (2) a much more important source of error is due to lost tags. There are no means of knowing just how many were taken and not turned in. Nor can anything more be learned as to the number of tags lost from the fish between the time they were attached aiid the time the fish were captured. Numerous reliable reports came to us of fish which showed a split in the tail-fin terminating in a hole similar to the one made with the leather-punch. Apparently the tags had in some manner become caught and pulled out. Another possibility is that some of the fish were unable to stand the operation necessary to the attachment of the tag, more particularly the attendant handling and removal from the water. All of these factors would tend to increase the percentage of fish captured, but obviously there are no means for determining to what extent. X 72 Report of the Commissioner of Fisheries. 1919 Table 3.—Number of Fish marked each Day, Total Number of Returns from each Day's Marking, ■ and Percentages of Returns for each Station, 1918. Station- A. Station B. Station C. Date marked. Percent Percent Percent Marked. Returned. age returned. Marked. Returned. age returned. Marked. Returned age returned. July 14 4 1 25 43 ii 30 „ 19 50 19 38 „ 21 46 16 35 25 o 20 „ 22 10 4 40 40 10 25 61 16 26 „ 23 20 2 10 95 27 28 „ 2-4 8 1 12 56 18 32 76 25 30 „ 25 56 4 7 50 13 26 „ 26 111 45 40 152 4» 33 „ 27 9S 16 16 „ 28 72 15 21 54 27 50 „ 30 99 20 20 117 3 IS „ 31 105 24 23 4S 20 42 i i 12 16 104 46 45 2 138 21 15 64 24 38 4 64 8 12 94 . 44 48 42 18 43 6 34 3 9 164 2 1 36 11 31 118 14 12 9 105 25 24 „ 13 155 24 15 „ 14 76 16 21 Totals 831 147 17.7 836 311 -37.5 1,000 203 20.2 Date marked. Station D. Station E. Marked. Returned. Percentage returned. Marked. Returned. Percentage returned. .Tulv 19 .". 85 92 74 '63 'ie '06 61 78 94 41 '47 28 25 23 'as io '26 35 35 52 21 'io 33 27 31 'ie '33 '36 57 45 55 51 '34 95 70 '93 114 122 39 97 113 75 16 118 36 12 23 31 16 ' 's 11 'is 33 41 14 24 '27 25 6 12 7 1 2 7 4 23 24 ' 25 8 16 ' 28 29 19 ' 30 29 1 31 34 All 36 2 25 4 5 24 a 33 37 8 10 9 15 12 16 8 19 9 20 23 21 25 ' Totals ■ 737 298 40.7 1,0:90 240 99 Total number marked 4,494 Total number returned 1,199 Percentage returned 26.8 Among; the specimens listed in Table 2 the following were recorded as having been taken outside the limits of the regions indicated on the map :— Days out. 1 specimen, Burrard Inlet, marked August Sth, Station E 103 2 specimens, Ebey's Lauding, marked July 24th,. Station C 17 1 specimen, near Sooke, British Columbia, marked July 26th, Station C .. 11 2 specimens, near Sooke, British Columbia, marked July 27th, Station A . 1-5 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 73 1 specimen, near Sooke, British Columbia, marked August 28th, Station A 4 4 specimens, near Sooke, British Columbia, marked July 20th, Station A . 2 1 specimen, near Sooke, British Columbia, marked August 2nd, Station A . . 2 These specimens from without the limits were so few that they have been omitted from further consideration. It seems probable that, with the exception of the specimens marked at Station A and taken near Sooke, these represent runs of sockeyes which were bound to some stream other than the Fraser River. In the following tables separate consideration has been given to those fish which came from the tributaries of the Eraser and from the main river above Mission Bridge. The problems connected with the fish from these upper waters are sufficiently distinct from those connected with the fish taken by commercial fishermen to demand such separate treatment. Table 4 gives the number of specimens returned (1) from the main commercial fishing- grounds, (2) from the Upper Fraser River, and (3) from outside the limits which we have set. Table 4.—Number of Marked Fish taken in- Numbered Districts of-Puget Sound and Fraser River, Upper Fraser River, and outside these Boundaries. Total returned. Number from Marking-station. Main Districts. Upper Fraser River. Outside Limits. 147 311 203 298 240 136 301 190 280 212 3 10 1 9 27 8 3 1,199 1,137 50 1'2 Detailed Study or Returns from the Commercial Fishing Districts. Two tables have been prepared for each of the marking-stations, presenting in detail the data obtained from the marking: First, a table showing for each day's marking the number and percentage of specimens recovered from each region and the average number of days required for the journey; second, a table showing the distribution of returns according to the region from which the return was reported and the number of days en route. This last is given as a matter of record and in order to show something of the range of variation underlying the average rates of progress as given in various other tables. It is upon the data given in these tables that the general discussions of the routes and rates of travel which immediately follow are based. Table 5.—Number and Percentage of Specimens from each Day's Marking at Station A taken in each Region and Average Number of Days en Route, 1918* Region. Date marked. 1. 2 4. 6. 7. Fish. Days. Fish. Days. Fish. Days. Fish. Days. Fish. Days. ' 9 i 2 i 1 2 'i 1 1 3.0 1.0 6.5 3.0 1.0 1.0 .io 2.0 3.0 1 6 1 9: i 9. 5 i 5.0 3.3 6.0 7.0 5.0 4.0 2.5 3.0 'i 2.'0 'i 'i 'e 5 2 3 2 3 16.0 5.0 3.7 6.4 9.0 5.3 4.5 4.0 'i 'i 3 3 8 4 6 1 2 „ 21 99 3.0 8.5 5.3 „ 30 .. 31 Aug. 1 9 8.3 7.5 4.2 3.5 3.0 4.0 Total 12 16 1 23 32 Averagest .. • 2.8 4.1 2.0 5.7 5 8 * No specimens marked at Station A were recovered in Regions 3 and t Weighted mean. X 74 Report of the Commissioner of Fisheries. 1919 Table 5.—Number and Percentage of Specimens from each Day's Marking at Station A taken in each Region and Average Number of Days en Route, 1918—Concluded. Date marked. Region. Fish. Days. Fish. Days. 10. Fish. Days. Total marked. Total returned. Percentage returned. July 14 „ 21 99 „ 2§X '.'.'.WW '.'. „ 24 „ 25 „ 27 „ 28' „ 30 „ 31 Aug. 1 2 „ ■ i.'.'.,'.'.'.'.'.'.'.'. 6 Total Averages* 'i li.'o 3 'i 5.0 'i l 2 7.0 l 2 10.0 4 9 9.9 2 5.0 4 o.o 2 3 4.3 1 1 3.0 25 13 " 4.3 4.0 4.0 26.0 6.0 8.0 7.0 6.0 5.0 6.5 15.0 8.0 8.0 33.0 10.5 4 46 10 20 105 77 138 64 34 831 1 14 4 2 1 4 13 14 16 24 12 20 136 30 40 10 12 7 13 19 16 23 16 14 12 9 16.3 * Weighted mean. Table 6.—Distribution of Returns from Marking at Station A according lo Region in which captured and Number of Days en Route.* Days en Route. Region. 1. 2, 4. 6. 7. S. 9. 10. 1 9 4 9 4 1 'i 2 6 1 4 1 2 i 1 1 3 6 5 2 '2 3 1 9 6 4 3 3 4 3 9 1 2 i 1 4 6 3 4 1 'i 1 1 1 1 1 2 3 1 1 3 'i -'i 'i 3 4 'i 9 6 .. . . 1 7 8 9 1 4 1 10 11 1 12 .. . 13 14 16 18 .. 2 26 33 'i Totals 12 16 1 23 32 25 1 13 14 * No specimen's marked at Station A were recovered in Regions 3 and 5. 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 75 Table 7.—Number and Percentage of Specimens from each Day's Marking at Station B taken in each Region and Average Number of Days en Route, 1918.* Region. Date marked. 1. 2. 4. 6. 7. Fish. Days. Fish. Days. Fish. Days. Fish. Days. Fish. Days. July 21 9 9 2 1 6 3 12 7 'i 5 5 4 S.O 2.0 1.8 1.0 s.'o 2.0 2.5 1.6 2.2 1.2 5 4 2 1 '3 '3 2 i 6 2 2.0 5.5 6.5 14.0 3.7 2.7 3.5 12.0 3.3 2.3 'i 1 '2 2.0 1.0 3.0 '2 3 2 2 1 1 1 9 '3 1 4.5 3.0 1.0 4.0 3.0 1.0 2.0 1.4 2.0 2.0 'i 'i '3 2 2: 5 5 3 8 4 1 15.0 „ 23 ,, 24 9g 8.0 26 5.3 28 3.0 „ 30 „ 31 9; 5.5 3.2 5.0 3.0 4 2.5 3.7 2.0 Totals 49 29 4 25 35 Averages! . . . 2.4 4.2 2.2 2.3 4.4 Region. Total marked. Total returned. Date marked. 8 9. 10. Percentage returned. Fish. Days. Fish. Days. Fish. Days. July 21 2' 1 2 5 3 'ti 14 7 9 4 3 5.0 3.0 7.5 8.0 5.7 3.0 3.0 5.1 3.4 3.3 3.3 1 1 51 3 io 7 '3 5 6 5 3 ■ 3.0 2.0 3.6 2.7 7.3 5.7 s.'o 3.4 8.0 6.2 5.0 2 's 4 7 11 4 3 3 6 1 3 4.5 4.4 5.7 11.7 3.5 4.0 7.0 8.0 8.7 7.0 7.3 25 40 95 56 50 111 54 1? 48 104 64 94 42 36 5 10 27 17 12 43 27 3 19 43 24 42 17 10 20 29 25 23 28 „ 24 25 30' 24 26 39' „ 28 30' 50 18 „ 31 Aug. 1 9 40 44 38 4 45 41 28 Totals 56 49 54 836 301 36 Averagest . . • 4.7 5.4 6.3 * No specimens marked at Station B were recovered in Regions 3 and t Weighted mean. X 76 Report of the Commissioner of Fisheries. 1919 Table 8.—Distribution of Returns from Marking at Station B according to Region in which captured and Number of Days en Route* Davs en Route. Region. 1. 2, 4. 6. 7. 8. 9. 10. 1 9 20 15 6 1 3 1 9, 13 4 2 i 'i l 2 1 i i 2 1 , 13 4 4 2 4 9 7 6 1 9 1 2 'i l 'i 1 6 21 8 7 5 2 1 'i 1 1 1 'i 1 10 13 9 4 2 2 'i 'i 2 1 'i i i 2 2 3 4 . i 8 14 14 6 1 8 9 10 3 i 11 12 l 13 i 14 15 '9 16 19 23 25 1 'i 27 1 Totals * 49 20 4 25 35 36 49 54 * No specimens marked at Station B were recovered in Regions 3 and 5. Table 9.—Number and Percentage of Specimens from each Day's Marking at Station C taken in each Region and Average Number of Days en Route, 1.918.* Region. Date marked. 1. 2. 3. 4. 0. 7. Fish. Days. Fish. Days. Fish. Days. Fish. Days. Fish. Days. Fish. Days. July 17 4 'i 3 5 'i 1 4 1 5.2 3.6 4.3 3.2 2.6 2.0 2.0 3.0 5 9 4 6 1 9 5 7 2.8 2.5 10.0 5.5 1.0 1.-5 2.5 1.4 2.1 'i 'i 3.6 7.6 'i i 4.3 6.6 3.6 5 1 4 O 'i 4 'i 6.2 2.0 4.2 7.0 9.6 6.2 2.6 2 i '6 1 1 2 2 3 4 5 „ 19 99 3.5 24 „ 26- Aug. 6 9.0 5.0 2.0 9 13 . . 3.5 6.0 14 5.0 Totals 213 34 2 5 19 20 Averages! . . . 3.0 3.8 5.0 4.4 " 5.7 5.6 Region. Total marked. Total returned. Date marked. 8. 9. 10. Percentage returned. Fish. Days. Fish. Days. Fish. Days. 1 5 3 2 13 'e 10 7 1 3.0 5.4 5.0 7.0 5.6 5.5 3.0 6.0 2 1 O 4 7 "i 5 5 6.0 • 3.0 6.3 6.5 8.1 5.6 6.2 5.8 1 9 5 7 'i 1 '3 4.0 13.5 8.7 10.9 7.6 9.0 13.6 43 50 61 76 152 164 118 105 135 76 13 19 16 23 47 2 14. 25 24 16 30.0 38,0 29 96.0 30.0 33.0 1.0 12.0 9 „ 13 „ 14 24.0 15.5 21.0 Totals 48 28 20 1,000 199 19.0 Averagesf.. ■ ly.Q 6.3 9.8 * No specimens marked at Station C were taken in Region f Weighted mean. 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 77 Table 10.—Distribution of Returns from Marking at Station C according to Region in which captured and Number of Days en Route* Days en Route. Region. 1. 2. 3. 4. 6. 7. 8. 9. 10. 1 2 3 4 . . 3 8 6 9 7 7 1 1 3 1 3 2 1 1 1 ! i i ! 'i 1 1 2 i 2 2 2 1 8 1 'i : '} 2 1 2 9 4 3 1 *2 1 'i 'i 12 11 4 8 1 9, 9 'i 'i -2 7 ■5 3 3 4 1 1 i i 's 2 6 2 3 8 9 10 11 2 1 12 1 13 14 1 15 16 17 "1 9Q 1 4 Totals >3 34 2 5 19 20 4S 28 20 * No specimens marked at Station C were recovered in Region 5. Table 11.—Number and Percentage of Specimens from each Day's Marking at Station D taken in each Region and Average Number of Dags en Route, 1918* Region. 1 2 - S . 10. JO a X3 a . a rjrj rd m ^ w A of. rd oil A oti A oc 55 T3 •A% i cci 0 P E 03] D r33i cc3 0 E cd Q s CC! Q o ^3 H9 It Sg July 19 .. 2 4.0 16 5.1 8 4.6 1 3.0 85 27 32 „ 23 17 1.9 4 1.5 o 10.3 92 24 26 ., 24 3 2.0 15 3.8 5 10.2 74 23 31 „ 28 9 1.0 9 1.4 12 2.8 4 15.7 63 27 43 „ 30 1 3.0 0 9 2 4 1.7 4 2.0 46 14 30 1 3.0 6 2.5 10 2.8 o 12.3 56 20 36 .. s rt 1.0 19 1.6 9 1.6 2 22.5 61 35" 57 23 1.6 8 1.9 3 8.0 78 34 44 6 2 5.0 18 2.8 24 9 3 6 3.2 94 50 53 i 10 1.8 9 2.2 1 5.0- 41 20 49 9 i 10.0 ° 3.8 6 4.0 3 3.7 47 15 32 Totals 9 1 11 .. | 131 109 ■• 33 737 289 39 Averagest- . 1.0 .. 3.0 3.3 .. 2.5 2.7 8.5 .. * No specimens marked at Station D were recovered in Regions 3', 4, 5, and 6. t Weighted mean. Table 12.—Distribution of Returns from Marking at Station D according to Region in which captured and Number of Days en Route, 1918* Davs en Route. Region. Davs en Route. Region. 1. 2. 7. 1 8. 9. 10. 1. j 2. 7. 8. 9. 10. 1 9 'i 5 ' 2 47' 39 28 6 4 3 1 1 28 38 19 10 7 2 1 'i 1 1 2 i 12 2 5 'i 'i 2 12 ' i "1 ' i 1 9 13 15 : . 3 4 2 18 1 20 "1 23 1 6 1 1 8 9 30 1 1 .. | X Totals 2 1 1 1 11 131 109 11 35 * No specimens marked at Station D recovered in Regions 3, 4, 5, and X 78 Report of the Commissioner of Fisheries. 1919 Table 13.—Number and Percentage of Specimens from each Day's Marking at Station E taken in each Region and Average Number of Dags en Route, 1918* Region. •a CD J* Cr Cd a "cd +j 0 H 00 a c "cd 0 H Date marked. 6. 7. 8. 9. 10. A 013 S cd a 3d fe tfl A V3 s OC cs fi A 013 >, CC! fi Hi 5 m ri cd fi bD . §S S2 July 25 i.'o 1 '2 'i 11.0 7.5 8.0 '3 3 5 '9 'i 3.7 4.0 7.8 8.5 too '2 5 5' 1 4 5 3 1 1 3.5 7.4 3.4 11.0 5.2 8.6 4.7 2.0 4.0 5 i S4 95 70 93 114 122 39 97 113 75 16 118 56 12 23 31 16 6 10 14 27 39 14 20 26 23 5' 11 3 '2 7 3 6 26 29 30 31 2 10 9 19 27 13 14 19 22 4 9 3 ' '9 7 3 3.7 5.0 5.5 8.3 9.3 4.9 9.8 7i.9 2.7 3.4 10.0 2.0 4.7 2,7 14 15 23 32 36 21 23 33 31 6 8 10 5 '8 23 19 16 19 20 21 Totals l • • ! 4 14 27 166 1,000 212 19.5 Averagesf .... 4.0 ! .. 8.0 6.4 5.8 6.7 * No specimens marked at Station E were recovered in Regions 1, t Weighted mean. 4, aud 5. Table 14.—Distribution of Returns from Marking at Station E according to Region captured and Number of Days en Route* which Days en Route. Region. . Days en Route. Region. 6. 7. 8. 9. 10. 6. 7. 8. £ 10. 1 'i 'i 'i '2 '2 3 3 1 2 '2 3 1 3 4 6 2 2 '2 '2 34 25 24 10 7 12 10 4 3 3 3 3 2 2 15 2 4 2 3 1? :;.•.:::::>: 1 2 1 18 2 5 6 7 8 20 21 24 26 27 3 3 2 1 1 10 11 30 31 32 Totals .... 2 1 12 13 14 2 1 4 14 5 7 1 166 * No specimens marked at Station E were recovered in Regions 1, 2, 3, 4, and 5. It was hoped that much more detailed information as to the migration up the Fraser Kiver might be presented, and to that end the river and the waters of the Strait of Georgia around the mouth of the river had been divided into twenty regions, each comprising about five miles of the river or off shore about the mouths. The returns from Canada were at first studied on this basis, but it finally became clear that the subdivisions were much too small to give results of any significance. These twenty regions were then divided into seven, as follows:— Region 10. Banks off the mouth of the South Arm and Canoe Pass. Region 11. South Arm and Canoe Pass up as far as Deas Island. Region 12. South Arm from Deas Island to New Westminster. Region 13. Banks off the mouth of the North Arm. Region 14. North Arm. Region 15. Fraser River from New Westminster to the mouth of Pitt River. Region 16. Fraser River from the mouth of Pitt River to Mission Bridge. 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 79 Table 15 gives the number of specimens which were marked at each of the marking-stations recovered from each of these regions, and the average number of days en route. It is apparent from this that even with this increase in the size of the regions no significant and consistent difference in the length of time required to reach the different regions appears, and for the purposes of statistical analysis it was necessary to consider all of the returns from the main Canadian waters as constituting a single group, and to combine Regions 10 to 16 into one region, 10. It is much to be regretted that the data are not such as to make possible a reliable estimate of the rate of travel in the river itself. Whether this is due to the manner of collecting and recording the data cannot be stated. Although these records from Canadian waters do not admit of detailed analysis, yvhen combined as a single group they agree well with the results obtained from American waters. Table 15.—Number of Specimens marked at each Station recovered in each Region in Canadian Waters, including the Fraser River to Mission Bridge only, and Average Number of Days en Route. Station. Total Fish. Average Number A. B. C. D. E. Days en Route.* Region 10 : 3 7.0 8 . 13.5 2 6.5 1 5.0 15 6.3 27 6.0 5 5.6 1 5.0 2 0.0 4 7.8 7 10.3 8 6.9 2 13.5 9 14.5 1 10.0 11 8.0 15 9.1 3 12.3 3 2.7 1 1.0 2 3.0 35 6.8 77 6.1 11 5.2 11 9.9 1:8 6.8 4 9.7 10 6.0 66 134 23 14 19 9 16 Region 11 : 7.2 Region 12 : Fish Region 13 : Fish 6.9 7.6 Region 14 : Fish 8.4 Region 15 : Fish Region 16 : Fish 6.7 8.1 6.6 Total fish 14 54 20 35 166 2.S9 Average number days en route* .... 10.5 6.1 9.8 8.3 6.6 7.1 * Weighted mean. In Table 16 is shown the total number of individuals marked at each station which were taken in each region, together with the average time en route. This combines the totals and averages developed in Tables 5 to 14. Table 16.—Number of Specimens marked at each Station taken in each Region and Average _..... _ . Number of Days en Route* Region. 1. 2. 3. 4. b Fish. Days. Fish. Days. Fish. | Days. Fish. Days. Fish. Days. 12 49 23 2 2.8 2.4 3.0 1.0 16 29- 34 1 4.1 4.2 3.8 3.0 '•2 s.'o 1 4 5 2.0 2.2 4.4 23 25 19 'i 5.7 B 2.3 5.7 D F, 4.0 Totals 86 80 2. 10 68 No specimens were taken in Region 5. X 80 Report of the Commissioner of Fisheries. 1919 Table 16.—Number of Specimens marked at each Station taken, in each Region and Average Number of Days en Route—Concluded. Region. 7. 5 9. 10. Total. Fish. Days. Fish. Days. Fish. Days. Fish. Days. A B 32 35 20 11 4 5.S 4.4 5.6 3.3 8.0 25 58 48 131 14 7.5 4.7 5.6 2.5 6.4 13 40 28 109 27 7 2 5.4 6.5 2.7 5.8 14 54 20 35 166 10.5 6.3 9.8 8.5 6.7 136 301 C 199' D E 2S9 212 Totals 102 274 226 289 1,137 Routes taken by Fraser River Sockeyes through Puget Sound.—Tables 17 and 18 show in complementary ways the distribution of returned fish according to the station at which they were marked and the region in which they were taken. Table 17 gives the percentages of the total returns from each station which were taken in each region and Table 18 the percentages of the total returns from each region which were marked at each station. The returns from Regions 3, 4, and 5 are practically negligible. It is quite obvious that the great majority of the fish, on entering the Sound through the Strait of Juan de Fuca, pass across to Washington Sound, and especially the southern shores of the southern' islands of the San Juan Archipelago (Region 1) and the western shore of Whidbey Island (Region 2) ; 8.8 per cent, of the returned fish which were marked at Sooke were taken in Region 1 and 11.8 per cent, in Region 2. Table 17.- -Percentages of Recovered Fish from each Marking-station ivhich were taken in each Region* Region. 1. 2. 3. 4. 0. 7. 8. 9. 10. A S.S 16.2 11.4 0.7 11.8 9.6 16.8 0.4 'i 0.7 1.3 2.5 17.0 8.3 9.9 6.5 23.5 11.6 9.9, 3.8 1.9 18.4 18.9 24.3 45.3 6.6 9.5 16.2 13.8 37.7 12.7 10.3 B 17.9 C 10.4 D . ... 12.1 E 78.3 * No specimens were recovered in Region 5. Table 18.—Percentage of Recovered Fish from each Region ichich were marked at each Station* Region. 1. 2. 3. 4. 6. 7. 8. 9. 10. A B C 14.0 57.0 26.7 2.3 20.0 36.2 42.5 1.3 100 10 40 50 33.3 36.3 29.0 1.4 31.4 34.4 19.6 10.8 3.8 9.1 20.6 17.7 47.5 5.1 5.8 21.6 12.4 48.3 11.9 4.S 18.6 7.2 D 12.1 B 57.3 ' * No specimens were recovered in Region 5. There is apparently considerable interchange of fish between Regions 1 and 2, as is evidenced by the fact that 9.6 per cent, of the returned fish marked at Station B, located in Region 1, were taken in Region 2, and that 11.4 per cent, of the fish marked at Station C, in Region 2, were taken in Region 1. From the region of Washington Sound a very few fish pass northward through Haro Strait (Region 4), but the proportion must be small compared with those passing through Rosario Strait and on up through the Strait of Georgia. The apparent proportions may, however, be modified by the fact that there are many more traps in Rosario Strait than in Haro 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 81 Strait. The few fish taken in Haro Strait came from Stations A, B, aud C, as would be expected. Only two specimens were taken in Deception Pass, both marked at Station C. No fish are recorded as having been taken in the central channels of the San Juan Archipelago. From the southern part of the Strait of Georgia (locally designated the Gulf of Georgia) the fish pass Point Roberts and enter Canadian territorial waters. Very soon thereafter they must enter the river, especially through Canoe Pass and the main mouth of the South Arm, and begin their journey up the river to the spawning-grounds. In summarizing, the route followed by the very great majority of the salmon entering the Strait of Juan de Fuca and bound for the Fraser River may be stated as follows: Across Washington Sound to the " Banks " south of the San Juan Islands and to the western shore of Whidbey Island; from here northward through Rosario Strait to the Strait of Georgia, past Point Roberts to the mouths of the Fraser River. There is no evidence to indicate that this route is varied in different parts of the season. As might be expected, this route is well indicated by the location of the various trap-sites. The approximate number of traps in the various regions of Puget Sound for which licences were issued in 1918 is as follows:*— Region 1 14. 20 13 11 4 26 Region 7 , 11 „ 8 40 „ 9 39 Total 178 In addition to the above, there are six traps located on the Canadian side of the Straits of Juan de Fuca, on Vancouver Island, eleven scattering traps along the southern shore of these straits, and twenty-nine in Admiralty Straits, chiefly between Admiralty Bay and Double Bluff. There are also a few in Hood Canal and that part of the sound extending south past Seattle, Saratoga Passage, and Padiila Bay. The ones listed above are those of chief concern here. There is apparently only a very low degree of correlation between the number of traps and the number of marked fish taken in any particular region. In Table 19 the six regions from which most of the returns came have been arranged in the order of the number of traps located in each. Then is given, for each region, the percentages of fish marked at Stations A and B which were recaptured in the respective regions. No correlation is apparent between the number of traps and the returns from Station A. There seems to be a distinct correlation in the case of the returns from Station B (especially if Region 6 is omitted), but the correlation is not well proportioned. Table 19.—Correlation between Number of Traps in Various Regions and Number of Marked Fish taken in those Regions. Region. Number of Traps. Percentage of Returns from Stations. A. B. 11 14 20 26 39 40 4.12 1.44 1.95 2.87 1.82 3.35 4.98 5.88 6.23 3.28 8.31 8.62 In a few instances a retrograde migration has apparently taken place and the fish have travelled away from rather than toward the mouth of the Fraser River. It is possible that faulty data may account for this, especially in such extreme cases as those fish marked at Station D and reported taken in Regions 1 and 2; or it may be that these are not Fraser River fish, but are sockeyes bound for some other stream. * Based upon licences issued by the Fish and Game Commission of the State of Washington. X 82 Report of the Commissioner of Fisheries. 1919 Rate of Migration.—A number of the preceding tables give, variously grouped, the average number of days required to pass from each marking-station to each region. Tables 5, 7, 9, 11, and 13 show the total range of variation in this regard for each station, and the general averages for each station are shown in Table 16. From the last-mentioned table it is apparent that, as would be expected, the time en route usually increases as the distance between the station and the regions where the fish were recovered increases. Those fish which were captured in the same region in which they were marked—such, for instance, as were marked at Station B and recovered from Region 1—have evidently been slow to resume the migration after the marking. Forty-nine specimens marked at Station B were taken in Region 1 after being out an average of 2.4 days. From Station C thirty-four specimens were taken in Region 2 after an interval of 3.8 days. Eleven specimens from Station D were taken in Region 7 after an average of 3.3 days, and twenty-seven specimens from Station R were taken in Region 9 after 5.8 days. It seems quite clear that, in the case of fish which have been recaptured comparatively close to the point at which they were liberated, the rate of progress is slower than that of fish taken a greater distance from the marking-station. Also, as may be seen from Table 16, in cases where a retrograde migration has taken place the rate of progress has usually been slow. These occurrences may possibly be due to some slight injury, or other unfavourable result, due to the handling attendant upon tagging. It may also be due to inaccurate data. In either case it would seem justifiable, when estimating the normal rate of progress through the sound, to consider only those cases of forward migration in which the fish have travelled a reasonably long distance from the station at which they were marked, say not less than twenty miles. In the case of the Canadian records it will be noted that the correlation between the distance travelled and the time en route is very low. The average time required to pass from Station E, at Point Roberts, into the Fraser River is 6.6 days. This would indicate a much slower rate of migration here than in other parts of the sound. In order to throw additional light upon this point, the difference between the time required for fish marked at Stations A, B, C, and D to reach the vicinity of Point Roberts (Region 9) and that which elapsed before the fish from these same stations were taken in Canadian waters has been calculated as follows :— Days. Station A 3.3 „ B 0.9 C 3.3 D ! 5.6 Average (simple mean) 3.27 This indicates a rate of progress between Point Roberts and the Fraser River over twice as rapid as that indicated solely by the results of the marking at Point Roberts, and one agreeing much better with the results obtained from the American records. The previous tables have shown the rate of travel as the number of days required to pass from the marking-station to the various regions of the sound. In order to make these figures comparable, the rate in miles per day has been calculated for each Instance. This has been done by simply dividing the distance" travelled by the time required to make the journey. Table 20 gives the distances from each marking-station to each of the established regions of the sound. These distances were measured, by means of aii ordinary map measure, from the marking-station to approximately the centre of the region in question. The most direct route possible was chosen. 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 83 Table 20.—Distances on Puget Sound from Marking-stations to Regions where Fish were taken* Station. Region. Miles. Station. Region. Miles. Station. Region. Miles. r l 35.0 ' 1 15.5 r i 62.0 2 47.5 2 2 59.5 3 57.0 3 15.5 3 62.5 4 45 4 35.0 4 35 A 5 6 52.5 60 C J 5 6 23.5 20 E 5 6 35 7 71 7 33 7 35 8 S3 8 45 8 20.5 9 92.5 9 59.5 9 . 10 122.5 L io 89.5 ,, io 30 " 1 l 33.5 2 15 2 28 3 21.5 3 29.5 4 17.5 4 51.5 B 5 10 D 0 41.5 6 21.5 6 13 7 33.5 7 8 47.5 8 11.5 9 62 9 27.5 10 92 10 57.5 * As it is obvious that the majority of the fish pass through Rosario Strait, the distances, wherever applicable, have been measured over this route. Table 21.—Number of Specimens marked at each Station taken in each Region and Rate of Progress in Miles per Day* A. B. C. D E. A. B. C. D. E. Region 1 : Region 7 : Fish 12 49 23 2 Fish 32 35 20 11 4 Rate 12.5 5.2 33.5 Rate 12.2 7.6 5.9 4.4 Region 2 : Region S : Fish 16 29' 34 1 Fish ^5 56 48. 131 14 Rate 11.5 3.6 9.3 Rate 11.1 10.1 8.0 4.6 3.2 Region 3 : Region 9 : Fish 2 Fish 13 49 28 109 27 Rate 3.1 Rate 12.8 11.5 9.1 10.2 Region 4 : Region 10 : Fish 1 4 0 Fish 14 54 20 35 166 22.5 8.0 8.0 Rate 11.7 14.G 9.1 6.8 4.5 Region 6 : Fish 2.3 25 19 1 Rate 10.5 9.3 3.5 11.2 * No specimens were taken in Region 5. Table 21 gives the results of these calculations of the rate in miles per day, together with the number of individuals on which the calculations are based. From this has been calculated the mean rate of travel.* In doing this all cases have been omitted which were based on (1) less than ten individuals, (2) a distance travelled of less than twenty miles, and (3) a retrograde migration relative to the mouth of the Fraser River. This has been done to exclude, as nearly as possible, all abnormal figures. With these exclusions there remain nineteen categories on which to base the final conclusions as to the rate of migration. The data for these nineteen categories are given separately in the following table:— * This and other similar means have been computed by use of the " harmonic mean " recommended by Yule (An Introduction to the Theory of Statistics), Rugg (Statistical Methods Applied to Education), and others, for the calculation of mean rates based on units of work. The harmonic mean is denned by Rugg as " the reciprocal of the arithmetic mean of the reciprocals of the individual measures of the series." X 84 Report of the Commissioner of Fisheries. 1919 Table 22.—Rate of Progress in the Nineteen Categories containing the Most Reliable Data. Station marked. Region where captured. Distance travelled. Specimens. Rate per Day. A 1 r i 2 6 7 ■ 8 9 L io- f 7 8 Miles. 35.0 47.5 60 71 83 92.5 122.5 21.5 33.5 47.5 62 92 33 45 59.5 89.5 27.5 57.5 30 Number. 12 16 23 32 25 13 14 25 35 56 49 54 20 4'8 28 20 109 35 166 Miles. 12.5 11.5 10.5 12.2 B 11.1 12.8 11.7 9.3 7.6 10.1 C 9 I 10 \\ l I io9 \\ 10 in 11.5 14.6 5.9 8 D 9.1 9.1 10.2 E 6.8 4 5 Calculated from these figures the mean rate of migration is 7.8 miles per day. If the Canadian records are omitted, the rate is 9.7 miles per day. Greene,* in his study of the migration of salmon (Chinook, silver salmon, and steelhead) in the Columbia River, obtained results quite different from these. He estimates that from thirty to forty days are usually required for the process of acclimatization to fresh water, during which time the fish work back and forth with the tides. After entering water which is wholly fresh he estimates the rate of travel of silver salmon and steelhead at 6.36 to 7.50 miles per day, although he concedes that this is only about one-third of the rate of travel as. estimated by men engaged in the fishing industry and seems to imply that their figures are probably more reliable than his own. Greene's figures should be accepted with considerable caution on account of the comparatively few individuals marked and the fact that three species were represented and all combined to give his final results. The sockeye-tagging experiment has produced no evidence indicating that the migration is especially retarded during the passage from salt to fresh water. As may he seen from Table 15, there is no great difference in the time required to pass from American waters to the Fraser River, just within the mouths, and that required to pass above New Westminster, where the water is entirely fresh. Furthermore, as will be seen later, in the case of fish taken on or near the spawning-grounds in the tributaries, the time elapsed since marking averages between thirty and forty days, the time given by Greene as required for the process of acclimatization. It may be concluded either that Greene's results are wholly unreliable or that the migration of sockeye is quite different from that of the species studied by him, which is, of course, entirely possible. The rate of migration of the quinnat (Chinook) salmon of the Sacramento River as given by Rutterf is much more nearly in accord with the results of this sockeye study. He estimates the rate of progress through brackish water to average seven or eight miles per day and that through fresh water to be between fifteen and twenty miles per day. These figures apply only to the spring run on the Sacramento River. The migration of the fall run is considerably slower, averaging but four to five miles per day through the fresh water. Greene's observations were, of course, made upon fall-run fish. A further analysis of data has shown some interesting variations in the rate of migration. A calculation of the mean rate of migration from each marking-station indicates that the rate of travel decreases as the fish approach nearer the mouth of the Fraser as follows: Station B, 10.5; Station C, 7.9; Station D, 9.1; Station E, 4.5. * Greene. Charles W.: The Migration of Salmon in the Columbia River. Bulletin U.S. Bureau of Fisheries for 1909, Vol. XXIX., pages 129-14S. Washington, 1911. t Rutter, Cloudsley: Natural History of the Quinnat Salmon. Bulletin U.S. Fish Commission for 1902, Vol. XXII., pages 65-142. Washington, 1904. 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 85 Finally the rates of migration for different parts of the season have been calculated. The data given in Tables 1 and 2 were first condensed into quartiles and then into halves, and the rate of travel in miles per day was calculated for each of these. The positions of the quarter points determining the quartiles were based on the number of fish marked, not on the number returned. In adopting a central point on which to separate the halves it was rather arbitrarily determined to consider all fish marked during July as belonging to the first half and all those marked during August as belonging to the second half. It is more convenient to think of the halves as separated in this manner, and the central point separating the fish marked into halves comes sufficiently close to August 1st to warrant the division on this basis. These calculations have not been made for Stations C and E. The inaccuracies in the marking record at Station C would have made such treatment valueless. In the case of Station E, over 78 per cent, of the fish recovered were taken in Canadian waters. Inasmuch as the authors were unable to make a detailed analysis of the Canadian records, it has seemed best not to attempt too detailed an analysis of the records obtained from the marking at Station E. 'Table 23.—Rate in Miles per Day in Different Quartiles—Returns from Station A. Region. I. 2. 7. S 3. 10. Quartiles. fl on fa CD ba cd* a o ■ fl +-: fl CD ■51 CD A CD fl CD fl CD 03 oc+J oi CC] CS O rr M fa « fa K S rr fa H fa M fa B hi ■4U First 6 9.2 8 12.2 6 13.9 3 11.3 2 10.4 5 22.0 5 19.1 35 13.0 3 35.0 2 6.8 9 8.6 8 9.9 4 9.8 0 9.2 2 18.8 33 10.0 Third 1 11.6 4 12.8 6 12.5 13 11.4 13 10.0 2 11.6 0 10.4 44 11.1 Fourth 2 14.0 2 15.8 2 15.0 8 19.7 0 17.3 1 13.2 2 5.7 23 14.5 * Harmonic mean. Table 21,.—Rate in Miles per Day in Different Halves—Returns from Station A. Region. 1. 2. rt 8 9. 10. Halves. fl fa CD ci cd# fl CD fl -t-> fl CD fl 4J ■s fl +j fl C3 cj cci oi OJ ci !> =3 fc- S fa « fa M fa 85 fa « fa « fa « EH Xi 4 6 11 4 5.1 9.3 14.3 10.8 2 5 18 15 2.9 7.6 10.8 9.6 3 9 25 19 11.0 6.2 10.3 13.2 10 14 '7 18 20.O 10.3 12.6 10.3 14 21 6 13 18.4 14.4 14.6 13.3 33 55 68 69 11.0 9,9 Third 11.5 11.3 * Harmonic mean. X 86 Report of the Commissioner of Fisheries. 1919 Table 26.—Rate in Miles per Day in Different Halves—Returns from Station B. Region. fl 013 fa rt O B Halves. 6. 7. 8. 9. 10. fl cc CD ci M 013 fa CD +c ci « fl oil fa 03 ta fl 013 ri CD ci 'A fl 013 fa oj +-> ci CD4-J t» rt 12 13 7.7 12.6 14 21 6.0 8.6 19 37 S.S 10.8 27 22 12.1 11.3 36 1 16.1 18 13.4 10S 112 10.5 10.9 Harmonic mean. Table 27.—Rate in Miles per Day in Different Quartiles—Returns from Station D. Region. Total Fish. Average Rate.* Quartiles. 8. 9. 10. Average Rate (for Regions 9 and 10 only). fl on fa CD 4-c cj K fl on CD ci fl Of! fa CD oi Third 33 IS 46 31 3.7 6.4 6.S 4.1 13 30 27 39 8.1 8.6 13.1 11.0 5 13 7 10 6.7 6.5 3.9 16.4 51 61 80 83 4.6 7.3 7.5 7.0 7.6 7.S 8.8 11.8 * Harmonic mean. Table 28.—Rate in Miles per Day in Different Halves—Returns from Station D. Region. Total Fish.. Halves. 8. 9. 10. Average Fish. Rate. Fish. Rate. Fish. Rate. Rate.* First 50 81 4.3 5.5 43 - 66 S.S 17 6.2 7.4 110 165 5 6 11.5 IS 7.2 * Harmonic mean. In the case of the quartiles, the indications are that the rate Is more rapid in the first than in the second quartile, and that the rates in the third and fourth quartiles are successively more rapid than in the second. This does not appear clearly in the results from Station D, especially when the returns from Region S are included. Omitting the returns from Region 8 (which is deemed legitimate, considering the comparatively short distance travelled, just twenty miles from Station D to the centre of Region S) brings the results more nearly in accord with those obtained from Stations A and B, the only disagreement being in the rate during the first quartile. This is, in the case of Station D, slightly less, rather than greater, than the rate in the second quartile. It seems not unlikely that the difference in rates of travel shown in the different quartiles may be explained as indicating racial differences existing in the fish composing different parts of the run. Gilbert* has demonstrated that the Fraser River run is composed of a number of distinguishable races, each bound to a different spawning region. The assumption that these races may have different rates of migration while passing through the waters of the sound does not seem to be a difficult one to accept. * Gilbert, Charles H. : Contributions to the Life History of the Sockeye Salmon. Report. Commissioner of Fisheries for the Province of British Columbia for 1017, Paper No. 4, pages Q33-S0. Victoria, 191S. 9 Geo. 5 Migration of Sockeye in Puget Sound and Fraser River. X 87 In the case of the halves, the results from all three stations are in complete agreement, indicating that the rate during the second half is distinctly more rapid than during the first half. In connection with the above determination of the rate of progress through Puget Sound, it must be borne in mind that the rates are undoubtedly lower than they should be. This necessarily follows when it is considered that seldom less than one day, and frequently three or four days, elapses between the times when the traps are lifted and the fish reported. Again, the fish may spend some time in the hearts and pot of the trap before entering the spiller, from which they may be taken at the next lift. A fish entering the trap immediately after it has been lifted would not be reported before the next day at the earliest, thus adding at least one day to the actual time required to make the journey from the marking-station. If traps were uniformly lifted at twenty-four-hour intervals, twelve hours could with safety be subtracted from the average number of days en route and the rate calculated accordingly. The authors' knowledge on this point, however, is not sufficient to warrant such treatment. It would undoubtedly be a step in the right direction, but would tend to give an opinion that the rates have been fully corrected, which would be quite unwarranted by the facts. It is possible that the error involved is within the limits of error dependent upon other factors. Returns from Tributary Streams and the Eraser River above Mission Bridge. Returns were obtained from various sections of the Fraser River watershed which have not been included in the regions indicated on the map. These sections may be indicated as follows :— - • Specimens. Main river, 3 to 27 miles above Mission 7 Main river, near Hope, Yale, Hell's Gate, and Spuzzum 16 Main river at Lytton and Lillooet 2 Main river at Soda Creek 2 Pitt River 6 Harrison Lake system, between Harrison and Lillooet Lakes 5 Harrison Lake system, Birkenhead River 9 Seton Lake system, - Portage Creek 2 Chilcotin River 1 The complete data concerning these returns are given in Table 29. The mean dates on which the fish captured in each of the above districts have been calculated and appear in the summary. This was done in order to see, if possible, when the fish bound to the different tributaries were to be found in Puget Sound. We have suggested above the possibility of a segregation in time of the races demonstrated by Dr. Gilbert. 'X 88 Report of the Commissioner of Fisheries. 1919 •8 cq ft. o r- r- r* o e C5 ft, I e t. fl V CO W © nc lO © HH © O III'8! © ... Ol iH i-t O CO H iH co in - rO CO CO CO CO HH HH HH . 1- © O CM CO CM © © © Oi Is CM CO CO CM r-i CO CO - .^ S3 >-. >> >- >v be bi bi f*a >i s Sft "3 3 33 » 5 5 3 9: cd W "-^ ""3 >"3 i-S < <3 <5 ^ 0 ■ f» tD Ol CM CD iO © CM rH © ,3 S i-t m , , _ A ^ ^~~""' ""» ( "^' "' -, < m ^ ffiCsMlQ'JHOKIClCOHffiHH HCOOf-OHSKONit-HiQ tH rH -H rH CM rH HM CMCMHi-htjicOCOCOHHHHH'HHH' • ° PS ■* Mfl NO O IT, N C M M O C C ONrHM^irjOHHU-rlNOCO H -2 d CM iH CM rH r-l rH CO bi be bi. b/, -*- bi ii, bi bi: bi bi br. bi bi 3 3 3 33 333333333 CMCMiH rHiHCMCMCM i—i iH rH ri hV*^ 4^4J4ij^4-=4J .434343 H-rojojcucvoa.oi'aja.aj < "1 (B cc x I- x x k _ a; 'X M cr © cm =-i ir. i- -j-- ~ © i-h cm i-coo O CT ^1 ^ —. .Q Id lO un rH © rH © r,,"9 CMCM CM CM CM CM CO CO CO COCO rH CO rH rH rH rH CM CM CO CO >*> >> bi bi ^^ ^^j >> bi bi bij^ "33^^"33S33r3!=^=:3 J$}Z' b'r; It. '"■ bi bt bi bi bi X '>• 33P3-;33333r:jr-;=s ; c: bb^^-i^^-ITn^^^^ ■ -shs<. bi bi ■ jfe >» E?,E? &J 3 p p :3,333 ^ bi bi : S3 : s l-S <■*$ 'l-3H-i-Sl-Si >i fil P u )w OJ ■ ■ rt • cs : (S ^5 1 to ; : "S « jzj u Sc ^ '■ ^ S » is ° t-" 'E q3 •1 g Sc ? -O 3 rid c3 p *c 1 *| 5 - c 3 P bu J illooet kes, Bi es, Por 0 o CD si on ■puzz Brid ver . bove -3 3 -a m I I I O —! b- 27 miles above ear Hope, Ya main river ytton and LilL ada Creek, ma itt River, 5 mi ii Harris n and Li nd Ande in River Total.. etwee: arriso ston a hilcot: co S hH 05 -H 'A W cc O 9 Geo. 5 Migration of Cockeye in Puget Sound and Fraser River. X 89 The data here presented are quite too fragmentary to afford a basis for more than the most tentative conclusions ; but the following points are suggested:— (1.) The fish which go farthest up the main river before turning into the tributary in which they are to spawn run earlier in the season. (2.) The main bulk of the fish which pass up the main river past Hope are to be found in Puget Sound at the height of the season, the latter part of July and early in August. (3.) The fish entering the lower tributaries, Pitt River and the Harrison Lake system, come largely from those fish constituting the last half of the run. In the event that it were deemed desirable to stop commercial fishing for part of the season, it would seem that the first part should be selected, as the indications are that there is a larger proportion of the up-river fish in the first half of the run, and these are the fish which have suffered more severely as a result of the disaster of 1913. SUMMARY. (1.) The experiment of tagging adult sockeye salmon in Puget Sound was initiated in an effort to determine the routes and rates of migration of Fraser River sockeye in passing through the waters yvhere commercial fishing is permitted. (2.) During July and August, 1918, numbered silver or aluminium buttons were attached to 4,494 adult sockeyes. Of these, 1,199 were later recovered and data as to time and.places of capture secured. (3.) The route most commonly followed passes from the Strait of Juan de Fuca across Washington Sound to the Salmon Banks and Whidbey Island, then through Rosario Strait and the southern part of the Strait of Georgia, past Point Roberts to the mouths of the Fraser River. (4.) The rate of migration as determined by the data for American waters is approximately ten miles per day. (5.) The migration is more rapid during the last half than during the first half of the season. X 90 Report of the Commissioner of Fisheries. 1919 PACK OF BRITISH COLUMBIA SALMON, SEASON, 1918. Compiled from Data furnished the Department ry the B.C, Salmon Canners' Association. Fraser River District— B.C. Packers' Association J. H. Todd & Sons J. H. Todd & Sons (Esquimalt) Glen Rose Canning Co., Ltd Great West Packing Co., Ltd M. DesBrisay & Co Gosse-Millerd Packing Co., Ltd Defiance Packing Co., Ltd C.L. Packing Co., Ltd St. Mungo Canning Co., Ltd Eagle Harbour Packing Co., Ltd .. Liverpool Canning Co., Ltd Canadian Fishing Co., Ltd Totals Skeena River District— B.C. Packers' Association Anglo B.C. Packing Co., Ltd J. H. Todd & Sons Kildala Packing Co., Ltd B.C. Canning Co., Ltd Skeena River Cora. Co., Ltd Cassiar Packing Co., Ltd. Wallace Fisheries, Ltd Gosse-Millerd Packing Co., Ltd. — Canadian Fish & Cold Stor. Co., Ltd Port Edward Fisheries, Ltd Totals Rivers Inlet District— B.C. Packers' Association Anglo B.C. Packing Co., Ltd J. H. Todd & Sons B.C. Canning Co., Ltd Kildala Packing Co., Ltd Wallace Fisheries, Ltd Provincial Canning Co., Ltd McTavish Fisheries, Ltd Totals Nass River District— B.C. Packers' Association Anglo B.C. Packing Co., Ltd Kiiicolith Fisheries, Ltd M. DesBrisay & Co Western Salmon Packers, Ltd Portland Fisheries, Ltd Totals Vancouver Island District— Anglo B.C. Packing Co J. H. Todd &Sons Wallace Fisheries, Ltd Quathiaski Canning Co., Ltd....... Western Packers, Ltd C. L. Packing Co., Ltd X Preston Packing Co., Ltd Clayquot Sound Canning Co., Ltd.. Nanaimo Canning & Packing, Ltd.. Redonda Canning & Cold Storage Co, Gulf Islands Fishing & Can. Co., Ltd. Sidney Canning Co., Ltd Lumrai Bay Packing Co., Ltd Nootka Packing Co., Ltd Defiance Packing Co., Ltd Puntledge Canning Co., Ltd Totals Outlying Districts— B.C. Packers' Association Gosse-Millerd Packing Co., Ltd Kildala Packing Co., Ltd Wallace Fisheries, Ltd Draney Fisheries, Ltd Kimsquit Fisheries, Ltd Western Packers, Ltd Maritime Fisheries, Ltd Tallheo Fisheries, Ltd Western Salmon Packers, Ltd Lockport Canning Co Totals Grand totals Sockeyes. Red Springs. White Springs. Chums. Pinks. Cohoes. Blue backs and Steelheads (IRAND Totals (Cases). 8,462 4,472 11,471 10,696 779 14,177 599 50,656 611 615 884 2,605 33 3,816 416 8,980 2,848 2,848 883 267 862 2,079 3,586 416 7,677 792 667 998 1,995 461 4,717 677 10,307 874 727 1,646 819 4 1,329 5,399 1,500 1,162 2,978 7,238 154 2,839 36 15,907 294 5,154 429 18,355 2,241 1,625 28,098 918 291 1,447 5,455 19 29 9,068 1,034 494 1,188 5,857 5,142 909 13,715 212 379 428 3,735 2,836 1,211 1,760 10,551 725 342 1,476 10,866 185 932 8 14,534 544 19,697 622 15,192 1,046 16,515 11,676 1S,3S8 708 31,111 24,853 2,916 86,215 40,111 4,395 208,851 24,526 3,050 2,743 34,907 12,113 1,709 ■ 81,964 19,917 2,262 1,173 2,469 26,181 4,258 667 56,927 8,016 3,838 456 487 12,861 9,319 34,977 9,438 746 330 118 13,355 631 358 24,976 12,257 957 124 194 9,070 2,139 362 25,103 5,729 826 426 679 11,543 1,475 367 21,045 10,118 695 314 124 8,256 799 347 20,653 14,384 1,402 419 15,506 2,377 526 34,614 7,823 995 152 552 10,640 1,879 366 22,407 6,944 897 115 2,810 8,730 1,513 6 21,015 4,170 345 16,013 367 822 11,978 10,678 2,256 38,759 4,671 286 4,994 30,535 123,322 14,626 ' 6,828 22,573 161,727 374,216 152 990 10,771 31,577 5.282 121 23 2,018 56 7,500 4,939 144 3,962 2,688' 11,733 7,295 235 48 2,332 21 9,931 5,951 131 40 1,345 3,492 2,919 13,878 6,153 93 39 4,181 2,129 833 13,428 3,800 18 27 3,906 210 7,961 5,355 53,401 4,776 13l 957 894 34 19 932 676 392 7,147 452 6,729 29,542 12,074 103,155 659 3,503 3,454 2,039 15,717 7,208 511 549 11,262 18,569 2,580 425 41,104 4,427 634 563 11,354 6,014 4,718 405 28,115 2,608 191 8,071 15,888 6,048 13 32,819 2,142 97 49 2,315 8,366 1,278 70 14,317 655 5 2,332 3,863 40,368 6,915 398 1,305 11,836 21,816 1,820 59,206 17,061 143,908 285 375 12,164 10,479 4,813 28,116 4,300 1,016 348 302 5,546 11,512 1,258 6,758 373 21,904 2,049 8,433 40,775 606 215 310 5,038 18,055 3.806 28,030 179 481 9,103 3,375 23 13,161 13,719 2,430 258 231 16,638 1,285 570 1,694 10,856 2,117 16,522 2,607 1,465 41 1,425 368 6,906 39 30 25,303 657 838 122 26,989 189 337 9,877 2,001 1,126 13,530 121 90 46 8,719 724 2,073 1,489 13,262 1,174 1,394 671 13,147 320 5,769 22,475 13 1,326 290 81,890 59 1,224 84,802 8,399 805 20,206 24,753 2,796 32,206 24,753 6,243 17,084 25,460 2,171 4,864 2,642 10,598 540 11,138 251,266 57,035 40,732 4,215 389,815 7,601 35,100 17,793 98 82,489 780 25 24,485 9,090 3,167 4 37,551 5,501 166 77 2,326 10.071 1,816 357 20,313 8,613 343 1,511 43,812 463 51,742 6,424 19 10 13,708 23,930 8.373 71 52,635 4,204 68 18 2,271 12,872 1,637 368 21,438 7,907 96 2,690 10,503 2,304 23,508 2,508 182 9,024 31,033 368 78 . 43,193 1,467 186 73 2,107 13,762 6,871 31 23,497 11,732 3,174 540 15,446 13,000 8,500 21,500 51,980 276,459 6,581 3,002 90,464 201,847 42,331 1,007 396,212 65,535 41,819 497,615 527,745 191,068 15,916 1,616,157 9 Cxeo. 5 Statement showing Salmon-pack op the Province. X 91 rr I—I O Ph 02 fi rr < m H O i—i rr EH m PQ rr> r-i H o |2S W1 W CO H rH °o -* co eo NO© m oi cn CC —i ->] — ■ 7 1 -?. *OatHO N-OHNhO CC ~ I O MO* ■*qi- cn i-ii> OTNcDt-'d'or ci co co r- -* o l- iQ Cl Cl I-- Cl ■00 00 GO Ol IC r-H CD CC- ci O CO TO CI-hknnh t- W3 © :c ~ ~- cc ci -n co r r ci i- ■—. i-H cn co cc cj m 0»01CK-H X rH CO rH rH OC4J>r*lQ IO CN X CO i-H CO O CO Ol t-h co"i-h>o"cm"i> >o r- i-h io *# Cl rH O CO CN I-H X I-H O rH r-i" CD' CO Of rH Ol Ol CN -* COOlHiN -^^-t~ iO o C0~rH"cD"ai"tN ;?ci rtf.g ^Scd -3 5 o ".3 O So co rr rH rr Oo'rT rf ■X • X Ol Sf . CO . O Cl i-H IO iO O r^ CO I.C 01 CO -rH X Cl CO X CJ rH t* Cl Cl J>- CN Ol iO CC CO ■* WNrH-mcn I> CO m CO 00 co co ■* r- cO O Oi cn ■* ■* cd'co"co"cn""(m" >* i-H t- Ol CM NulCOCQl'- iO H O O H CC'co'cMO'co" Ol rH CO Ol iH CO MNOMHH Cl" co" i-T ■* "O CO HCQH 31,411 3,845 11,200 69,693 19,139 1,498 CD 00 CD CO 22,188 4,496 24,938 44,568 22,180 1,125 m cc rjl cc" CO C- X CD r HU5SOC- X i-H CO CN C in a> dagra X 92 Report of the Commissioner of Fisheries. 1919 Ol Ol rM CO CO CO © X lO CO co »o co i-h r- SO tH © XO CM tH CM © CO "O i-H CO -N © I~ ■ COCC c SO CO CM ■ © i—l - Cl CO . r-f . IO tH CO • CO CM - © lO rH rH -tHCS • UO CD Ol I-- • © Tf< CN CM Oi -* . CM CM Ol n, © X o Ol o Os CO l> CN rH . s X © rH CO Tt* ■ CO H K CO ■* CM CM tKOONCMCO ■ t- CM CO CM © . © © X © CO • "" i-H r-H t~- Cl CM • - CO © CO Tfl X © ■* I> Ol Ol 1-H i-H 1> Cl CO X © IO Tf X CO © *-, Oi -* o S H Ti H &M. •o cu ga-c.£o3 a a. Ur> <- Ui f CO X co fi" ©" io" ucTio" Ol 1 SO © © t~ CM o Ol © o CM © Ol O r-H © CO © CO Ol CM IO O © Cl X Cl X CM © © O X Tf X CM I-~ CO Ti"di0OiO © IO tK CN i-h iO © CO rH CM O X CO" Cl" TjT r-T © io © CO lOt-OOD 0.1X50 ■*HC ©_ cn'oTio"© rH © I— tH i-h CM © © Tfl O © CO rH Cl Tf- r~ o © 035 398 720 Tf IO © o CN r~ im os cs Tf Ol Ol CO CM CM rH CO CM © -f C CD -t Tf rH CO X CO Tf © CO HTflr-HO © IO © © CO CM i-h © iO CO t> © iO CD CO Clt- © Tf< Tf CO o Cl CO CM i-h O © CD tH Cl N ■* O CO CM i-H rH ■x ~ ci io co r» CO r- O © Cl © i-C Cl CO ^ N IO cm" cm" io" ci" io" th" C- © © rH CM © Th Ol rH CO CM i—I © IO- X icC cc IO rH IO O i— © X Ol rH © X I- C Tt .;,- ^^-i rH I"- © Tf X ©■ Ol CO rH rH CO Tfl S3C'S§I I CM X x as © Tfl X rH TH Tf CO rn rH Ol © rH Ol t- X CM NNOCO WONH jrf I * t- O © © X X X i-H -H IO -f © X X CO TfTio'rH t-* Mm oi X © Tf< IO cor- X X iO Ol X CM CO O IO X CM © O CO I— X © o go CD >. cn — no S o a- s O -i H -C coOPMO"""@en ; edm:hasType "Legislative proceedings"@en ; dcterms:identifier "J110.L5 S7"@en, "1919_V02_23_X1_X92"@en ; edm:isShownAt "10.14288/1.0059765"@en ; dcterms:language "English"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "Victoria, BC : Government Printer"@en ; dcterms:rights "Images provided for research and reference use only. For permission to publish, copy or otherwise distribute these images please contact the Legislative Library of British Columbia"@en ; dcterms:source "Original Format: Legislative Assembly of British Columbia. Library. Sessional Papers of the Province of British Columbia"@en ; dcterms:title "PROVINCE OF BRITISH COLUMBIA REPORT OF THE COMMISSIONER OF FISHERIES FOR THE YEAR ENDING DECEMBER 31ST, 1918 WITH APPENDICES"@en ; dcterms:type "Text"@en ; dcterms:description ""@en .