Salmon Runs - Upper Columbia River, 1956-57

364;

SALMON RUNS - UPPER COLUMBIA RIVER, 1956-57

Marine Biological Laboratory

WOODS HOLE, MAtJS.

SPECIAL SCIENTIFK REPORT-FISHERIES Na 364

United States Department of the Interior, Fred A. Seaton, Secretary Fish and Wildlife Service, Arnie J. Suomela, Conunissioner Bureau of Commercial Fisheries, Donald L. McKernan, Director

SALMON RUNS - UPPER COLUMBIA RIVER. 1956-57

R. R. French and R. J. Wahle

United States Fish and Wildlife Service Special Scientific Report--Fi8herie8 No. 364

Washington, D. C. October I960

CONTENTS Page

Introduction 1 Salmon runs past Rocky Reach Dam site 2 Time and size of runs passing Rocky Reach Dam site Z Chinook salmon..... 3 Blueback sedmon 3 Steelhead trout 6 Silver salmon 7 Rate of travel--Rock Island to Zosel Dcun 7 Distribution of spawners in tributaries above Rocky Reach Dam site 8 Chinook salmon 8 Blueback salmon 14 Siunmary and conclusions 14 Literature cited 15

FIGURES

1. Columbia River watershed between Rock Island and Grand Coulee Dams iv 2. Chinook salmon p>assage at Rock Island Deim, 1954-57 (smoothed by 5) 4 3. Blueback salmon counts at Rock Island Daun, 1954-57 5 4. Steelhead trout counts at Rock Island Dam, 1954-57 (totals by weeks) 6 5. Areas of spawning in the Entiat River 9 6. Areas of spawning in the Methow River system 10 7. Areas of spawning in the Okanogan and Similkameen Rivers 11 8. Distribution of tagged chinook salmon above Rocky Reach Deun site by period of tagging at Rock Island Dam, 1954 and 1955. 12 9- Distribution of tagged chinook salmon above Rocky Reach Daxn site by period of tagging at Rock Island Dajn during 1956 13

TABLES

1. Totals of streami survey counts and counts of spring and summer chinook salmon passing Rock Island Dam - 1956 and 1957.... 2. Spawning ground surveys above Rocky Reach Dan:i site 1956-57. Peak counts (live and dead) of chinook and blueback salmon

ui Hotcher

Figure t.—Columbia River watershed between Rock Island and Grant Coulee Dams.

IV SALMON RUNS - UPPER COLUMBIA RIVER, 1956-57

R. R. French and R. J. Wahle

ABSTRACT

Important runs of salmon pass Rocky Reach Dann site on the Columbia River. The escapement of chinook salmon past Rocky Reach in 1957 was estimated to be approximately 11,000 spring chinook which spawn in the small remote tributaries and 6,000 summer chinook which spawn in the large tributaries of the Columbia River.

Blueback salmon passing Rocky Reach Dam site in 1956 and 1957 accounted for approximately 72 and 60 percent of the escapement passing Rock Island Dam, or about 67,000 fish in 1956 and 43,000 fish in 1957. They spawn in the Okanogan River in Canada. The rate of travel of blueback salmon was 13.5 miles per day.

INTRODUCTION Chelan County Public Utility Dis- trict No. lis engaged in the Upon the comipletion of Grand Cou- construction of lee Daxn in 1938, anadromous fish runs Rocky Reach Dam located approximately 21 were blocked from about 1,140 lineal miles aboveRock Island Dam. This poses miles of spawning and rearing areas dam problems for passing anadromous fish above the dam (Fish and Hanavan, runs upstream and downstream and for maintaining 1948). At that time anadromous fish adequate spawning areas if were transplanted to the tributaries major spawning between Rock Island and Grand Coulee areas are eliminated by flood waters of the forebay. Dams. These tributaries and the Co- The Fish and Wildlife Service lumbia River between Chief Joseph agreed to provide basic biological information the and Rock Island Dams, totaling about on present salmon escapements of chinook 340 lineal miles of stream, now provide salmon (Oncorhynchus tshawytscha), blueback the spawning and rearing areas for sadm- on anadromous runs passing Rock Island (Oncorhynchus nerka), silver salmon (Oncorhynchus kisutch), and the steelhead Dzim (fig. 1). trout (Salmo gairdneri). S tudi e s on the steelhead were limited to examination Note.--R, R. French and R, J. Wahle, Fishery Re- search Biologists, Bureau of Commercial Fisheries. of their abundance and times of arrival U. S. Fish and Wildlife Service, Seanle, Wasliington. at Rock Island Dam. This basic biological information will also provide in spawning streams above and below a basis for comparisons of the runs in Rocky Reach Dam site. Totals of peak future years. During 1956 and 1957 the salmon counts for the Wenatchee River

Service made studies on ( 1 ) the size and system and river systems above Rocky time of migration of various species Reach were compiled, and a ratio ob- past the dam site, (2) the rate of travel tained of the number of spawners found of migrating fish pastRocky Reach Dam in the two areas. This ratio was then site before the dam is built, and (3) applied to the escapement counted at tributary spawning areas used by Rock Island Dam to obtain an estimate various segments of the run. number of fish passing Rocky Reach Dam site. For example, if 1,000 To accomplish these objectives we spawners were counted in all streams have used the fish-count data at Rock and 700 were observed in streams above Island Dam for determining the size Rocky Reach, then 70 percent were and time of escapements atRocky Reach deemed to have passed Rocky Reach. Dam site 21 miles upstrecim. Tagging Thus, if 10,000 Chinook had been counted experinnents provided information on at Rock Island Dam, 7,000 fish would the rate of travel of blueback Sedmonj be estimated as having passed Rocky and strejmi surveys in which spawning Reach Dam site. salmon were counted in the streams were used for obtaining salmon spawn- Stream surveys were made peri- ing data. odically by foot or by boat and the number of fish of each species of salmon Financial support for this study was recorded. Peak counts, which were was provided by Public Utility District the largest number counted in any one No. 1 of Chelan County, Washington. survey, becamie the basis of comparisons. Survey counts do not represent the total number of spawners present in SALMON RUNS PAST ROCKY REACH a stream in a season. Such information DAM SITE cannot be obtained from stream surveys. However, as these survey counts Over the years salmon become were taken in the Scune manner and by specifically adapted to their environ- the same personnel, we feel they gave ment and become established in certain a usable estimate of the comparative strecims and watersheds in varying abundance of spawners in the different abundance. The purpose of the in- streams. vestigation was to determine the abundance of these upriver runs and the Time and size of runs passing Rocky arrivcil times of these runs at the new Reach Dam site Rocky Reach Dam site. Blueback salmon were found to Since it was not possible to count migrate upriver at the rate of 13.5 fish there, the arrival times of the miles per day (see later section); thus, runs at Rocky Reach Dam site was they would be expected to pass Rocky estimated from a study of the arrival Reach Dam site, 21 miles upstream times of fish at Rock Island Dam 21 from Rock Island Dam, by the second miles downstream. The daily counts day after passing Rock Island Dam. of salmon and steelhead were graphed Chinook salmon in the Snake River to show the seasonal pattern of migra- were reported to naigrate at rates of tion, and by applying known rates of 13 and 15 miles per day.* Either of travel of different species, we were these rates indicates that chinook runs able to estimate their arrival times passing Rock IslandDam would pass the at Rocky Reach Dam site. Thompson, Robert N„ James B. Haas, Lawrence M. The size of escapements passing Woodal, and Edwin K, Holmberg. 1958. Results of a tagging Rocky Reach Dam site was estimated program to enumerate the numbers and to determine the seasonal occurrence of anadromous fish in by applying to the Rock Island Dam the Snake River and its tributaries. Fish Commission of Oregon, Contract count the ratio of spawners observed Report DA 35-026-eng-20609, Portland, 202 pp. Processed, Rocky Reach site Ij or Z days later. of spring and summer chinook at Rock Therefore, the arrival times of the Island Dam are given in table 1. We blueback and chinook runs at Rock found that 77 percent of the spring Island Dam have been used as the ap- chinook counted in 1956 and 65 percent proximate times the runs would be ex- in 1957 were in areas above Rocky pected at Rocky Reach Deun site. Reach. In 1957 this meant that about 65 percent of the spring run counted over The rates of travel of individual Rock Island, or approximately 11,000 steelhead were found to vary widely in fish, were estimated as having passed the Columbia River between Bonneville Rocky Reach Dam site. Twenty-one and McNary Dams. Schoning, Johnson, percent in 1956 and 19 percent in 1957 and Merrell, in an unpublished report, of the summer chinook were counted found that these rates varied from 1.9 in areas above Rocky Reach Dam site. to 14.6 miles per day.' Thus steelhead Thus, in 1957 the population of summer could be expected to pass Rocky Reach chinook passing Rocky Reach was es- from Z to IZ days after passing Rock timated at 6,000 fish. Counts of spring Island Dam. and summer chinook runs passing Rock Island DaJTi in 1956 were not reliable

Chinook salmon . --The season of because of the overlap of the runs that passage and the number of chinook season. The reason for this overlap is salmon passing Rock Island Dam are discussed in a following section. in Z. Chinook runs in the given figure Fall chinook salmon spawn in the River are composed of Columbia Columbia River in the limited area be- spring, fall segments ac- summer, and tween Rock Island Dami and Rocky Reach cording to the time they enter the river. Dam site, and on extensive spawning Each season, the chinook salmon escap- riffles above Rocky Reach. Determina- ing the fishery appear during the same tion of their abundance in this stretch time period at Rock Island Dam; spring of the main Colunibia River was as- chinook pass in May and June, summer signed to another fishery agency. chinook pass in July and August, and fall chinook make their appearance Blueback s almon.--Bluebac k after August. The spring and summer salmon, first appearing at Rock Island chinook runs are fairly well defined by Dam in early July, show a peak of their period of passage at Rock Island abundance around July 15 to Z5. (see Dam. The fall run, however, is not figure 3.) To determine the size of the clearly marked and appears as a minor blueback escapement passing Rocky part of the escapement late in the Reach Dam site we have used the sea- season. Fish and Hanavan (1948) re- sonal counts of blueback at Tumwater ported on the bimodality of the chinook Dam (fig. 1) as the number of blueback runs at Rock Island Dam and reported entering the Wenatchee River system. that the spring run apparently termi- The remaining part of the run, or Rock nates during early July. They did not Island Dam counts minus the Tumwater report on the existence of a fcill run, Dam counts, has been used as the es- but evidence of a fall run atRock Island timated number passing Rocky Reach Deim is presented in a later section of Dam site. This procedure was in pref- this report. Considerable overlap of the erence to usirig stream surveys for three segments occurs. All three obtaining ratios of blueback salmon in groups should be expected at Rocky the two areas. In recent years the Reach Dam site during the same time Okanogan run has been the larger of period as at Rock Island Dam. the two runs. Counts of the Wenatchee River run at Tumwater Dam for 1954- Spawning ground counts above and 57 have ranged from 28 to40 percentof below Rocky Reach Dajn site and counts the total escapement counted at Rock Island Dam. The remaining 60 to 72 ^Schoning,. Robert W., Donald R.Johnson, and Theodore percent of the blueback runs for these R. Merrell. 1950. The expected appearance of certain runs years has been attributed to the of Columbia River salmon at McNary Dam. Fish Commis- sion of Oregon. February, unpublished report., IS pp. Okanogan run. The estimate of the blue- Portland. back run passing Rocky Reach Dam site lO C\l lO to 00 — OD — cj_ r--_ q. r>-_ n Hi in O fO CM M "O

* lO lO m o> a> o> ff>

so o

c o E

HSU dO bl3SNnN Table 1. —Stream survey and Eock Island Dam counts of spring and summer Chinook, 1956-57 in 1956 was 67,000 fish, or 72 percent October and forms the larger group of the Rock Island Deun counts. In of the two, figure 4. As counting at 1957 the estimate was 43,000 blueback Rock Island Dam is not continuous dur- or 60 percent of the Rock Island Dam ing winter months but consists of counts. periodic counts at the three ladders, the migration during this period is not Steelhead trout . --Two groups of fully known. These periodic counts at steelhead runs pass Rock Island Dam, the various ladders at Rock Island have the spring run, which pass in mature revealed some movement of steelhead condition and are destined to spawn that throughout the winter months. Pre- same season, and the fall run, which sumably, spawning of steelhead occurs pass Rock Island from July through throughout the accessible river systems 700

1954 N = 4578 1955 N = 3200 1956 N = 1531 600 1957 N =3927

( Counts are for periods of confinuous counting at the 3 ladders.)

500

X 400-

300 ir oui 2

200-

100

-•^..^p,,,, •••-—^-~

7 14 21 28 4 II 18 25 2 9 16 23 30 6 13 20 27 3 10 17 24 I 8 15 22 MAY JUNE JULY AUGUST SEPTEMBER OCTOBER

Figure 4.--Steelhead trout counts at Rock Island Dam, 1954-57. (totals bf weeks).

6 above Rock Island Dcun, although no pre-existing conditions; and if delays spawning ground counts were obtained. are caused, what might be the cumula- Sizes of steelhead rxins passing Rocky tive effect of many dams. Reach were not estimated. Knowing Rocky Reach Dam was to be constructed we decided to determine Silver salmon . --Silver salmon are the rate of travel of salmon through this few in this area at the present time. stretch of river before construction. Total yearly counts of silver Scilmon Then in the future it miay be possible at Rock Island Daun for 1954-58 have to determine if Rocky Reach Dam de- ranged from 40 to 94 fish. These fish lays passage of migrating fish. p>ass Rock Island Dam during Septem- ber, October, and November. Formerly In 1957, to determine the rate of silver salmon were abundant. Records travel through this stretch of river, show that approximately 6 to 7 million blueback salmon were tagged with var- silver salmon eggs were taken annually ious colored tags, were released in from the Wenatchee River alone during the forebay at Rock Island Dsuna, and the years 1899 to 1902. From 1910 to were observed at Zosel Dam approxi- 1920 the nvimbers of silver salmon eggs mately 145 miles upstream on the taken annually at the Methow River Okanogan River. The use of combina- hatcheries ranged from a few thousand tions of colored tags enabled us to to nearly 3 million (Washington State identify tagging dates as tagged fish Department of Fisheries, 1899-1920). passed an observation point. Zosel Dam, These silver salmon runs, though of the observation point, is a small wooden little value now, could p>ossibly be im- structure designed to forni a pool for proved. logging operations at a nearby mill. Observers recorded the daily passage Rate of travel—Rock Island to Zosel of tagged and untagged fish through the Dam fishways at this dam; however, fish were also able to pass the dam through The rate of travel for blueback cracks in the wooden weirs. As few Scilmon between Rock Island Dam and recoveries were made from the 1957 Zosel Dam (figure 1) was determined experiments, the data from tagging prior to the building of Rocky Reach experiments of 1954 and 1955 were used Dam in order to ascertaun if delays to obtain rates of travel between Rock in migration result from the new dam Island and Zosel Dams. These experi- now under construction. The question ments in 1954 and 1955 were similar of delay and its effect upon migrating to the 1957 experiments and were for salnrion has long been a matter of con- the purpose of studying the efficiency jecture among fishery biologists. The of the Rock Island Dcim fishways. actual effect of a short-term delay or the accumulated effect of several short- term delays to migrating salmon is In addition to tagging experiments, not known. The most notable example of peak counts at Rock Island and Zosel the effect of a delay in migration has Dams were used to determine the rate been the blockade at Hell's Gate on of travel by computing the difference the Fraser River. Here, fish that were in time between the peaks in abundance delayed longer than 14 days did not at the two places. reach the spawning ground. During these experiments fish captured, tagged, and The data obtained from the blue- liberated were often recaptured below back tagging experiments at Rock the gate and the interval between cap- Island Dam in 1954-55 are presented tures was termed delay (Talbot, 1950). below. "Days out" represents the time In reference to the Columbia River and interval between tagging at Rock Island many dams now built or contemplated and recovery at Zosel Dam. The day along salmon migration routes, one of of tagging is treated as zero day. The the problems is whether dams cause data from the two seasons were com- delays over and above the effects of bined because the means and variances of the days -out period of the two re- spawners from year to year and indif- covery groups were not statistically ferent streams, providing the surveys different. are made in the Seime manner. Aerial surveys, of the number of salmon Number of tag recoveries 30 nests were used as an add in locating the area of spawning in the Okanogan Average number of days out 10.7 River from Oroville to Brewster.

Range of days out 7 to 16 Tagging experiments atRock Island Dam gave additional information about Since the distance between Rock Island spawning areas of the different seg- Dam and Zosel Dam is about 145 males, ments of the runs which pass Rocky the average rate of travel is 13.5 miles Reach Dam site. Although these ex- per day. periments were designed for evaluating the fishways, they also yield data which The results of a comparison of the are applicable to this study. Chinook time of peak counts of blueback salmon and blueback salmon were tagged at Rock Island and Zosel Dams were throughout most of their period of mi- found to agree with results of the tagging gration at Rock Island Dam during the studies. In 1953 the mode at Rock 1954, 1955, and 1966 experiments. Island Dam occurred during the period Many tagged salmon were recovered Jxily 15-22, with a maximum count on on the spawning grounds, and from July 18. At Zosel Dam the mode oc- their distribution we learned which curred during the period July 23-30, spawning areas were used by various with the maximum noted on July 28. segments of the run. Thus, the difference in time between maximum counts (10 days) compares Chinook salmon closely with the 10.7 days travel time established by tagging studies. Other Peak counts of salmon and the time seasons could not be compared as no period encompassing the peak of spawn- definite peaks in abundance were noted ing are listed in table 2. These counts at Zosel Dam. serve to exemplify the relative im- portance of each tributary stream- Note the increased covmts in each DISTRIBUTION OF SPAWNERS IN stream in 1957 over those of 1956. TRIBUTARIES ABOVE ROCKY Total counts of chinook salmon are REACH DAM SITE about 3 times the counts obtained in 1956. Salmon are known to be attracted The distribution of fish within the to certain spawning areas in a river streams is illustrated in figures 5, 6, if the flow, temperature, system, and 7. The stream lengths have been gravel, and other requirements are drawn to sccile; the widths, however, there is very little fulfilled. Normally, have been exaggerated in order to show these spawning areas straying from areas of spawning. The symbols rep- the from year to year. The purpose of resent the comparative abundance of investigation was to discover the spawn- individual species within each stream. ing areas now used by salmon in the As observed, spring chinooks spawn tributaries above Rocky Reach Dam in the smaller tributaries and more site. remote areas of the tributaries, and summer chinooks in the larger main to Stream surveys were made tributaries. determine the distribution and comparative abvmdance of spawners in the The distribution of spawning Chi- tributary streeims. While the survey nook salmon according to the time of counts do not represent the total num- season they passed Rock Island Dam is bers of spawners present, the counts graphically illustrated by the recovery are of value in deternaining the dis- of tagged fish on the spawning grounds. tribution and coinparative abundance of The history of chinook salmon escape- Table 2. —Spawning ground surveys above Bocky Reach Dam site 1956-57. Peak counts (live and dead) of chinook and blueback salmon

Salmon counts Species and Period of peak of spawning name of stream 1956 1957

CHINOOK SALMON

Entiat River 149 273 Aug. 20-30 Twisp River 191 702 Aug. 10-20 Chewack River 404 1,065 Aug. 15-25 Nfethow River Above Winthrop 249 909 Aug. 20-30

Winthrop to Twisp 141 148 Sept . 5-15 Twisp to Pateros 375 1,398 Oct. 5-15 Simllkameen River 101 189 Oct. 10-20 Okanogan River Oroville to Brewster "" 99 Oct. 15-25 BLUEBACK SAIM)N Okanogan River Above Lake Osoyoos 16,067 8,533 Oct. 1-10

During aerial surveys of this section following the peak of spawning, 37 salmon redds were counted in 1956 and 53 redds in 1957. ^Survey too early for spawning peak.

ment at Rock Island Dam shows the

3 mites separation of spring and summer Chi- (widths eiagQeroled ) nook runs to occur sometime between o Spnnq Chinook • Summer Chinook June 25 and July 9. Accordingly, tag recoveries were plotted by these periods in which spring and summer Chinook are known to have passed the dajxi, figure 8. The period between June 25 and July 9 was chosen as a period of overlap of the two runs. It is observed that for the combined tagging data of 1954 and 1955 no tagged spring Chinook were recovered on the summer Chinook spawning areas; only one tagged sumimer chinook was noted in spring Chinook spawning areas. Fish tagged during the period of overlapwere found in both areas.

The 1956 tagging experiments at Rock Island Dcm:i indicate that spring chinook may not always pass Rock Island Dami and Rocky Reach Dcm^i site in the periods indicated. The distribution of tagged fish by season of passage at Rock Island for 1956 is showninfig- Figure 5. —Areas of spavming in the Entiat River, ure 9. Many fish tagged after Jvily 9, 1956 Rottli

3niiles

(widths exaggerated )

o Spftng Chinook • Summer Chinook X Blueback Salmon

Figure 6.—Areas of spawning in the Metiiow River system.

10 \Loke Voseoui P

I Oliver Oiverson

Tuc ul nuit Lohe J^ r ^\\ ( Inkoneep Cfeeh

BRITISH COLUM Bl A WASHINGTON

3 miles

( l»l(ll^s eioggeroted )

Summer Chinook redds Bluebock Salmon

Figure 7.- -Areas of spawning in the Okanogan and Similkameea Rivers.

11 Figure 8.--I}istrlbutioa of tagged chioook salmon above Rocky Readi Dam site by period of tagging at Rock Island Dam, 1954 and 1955.

12 Legend

D Tagged prior to June 25 O Togged June 25- July 9 A Tagged ofter July 9

BRITISH COLUMBIA WASHINGTON

Leovenworfh Hotchery

Figure 9.—DistributioD of tagged chinook salmon above Rocky Readi Dam site by iieriod of tagging at Rock Island Dam during 1956.

13 appeared in spring chinook spawning bound for the spawning areas in the areas. These were spring chinook and Okanogan River in Canada as evi- not summer chinook as evidenced denced by the strezim surveys (see by their characteristic small size, fig. 7). Very few blueback salmon greenish-gray appearance, and early enter the Entiat, Methow, and Similka- spawning. Summer chinook average meen Rivers. Blueback seilmon tagged larger, take on a reddish color in spawn- during all parts of the migration period ing streams, and spawn later in the at Rock Island Dam were recovered on season. Apparently a delay along the the Okanogan River spawning grounds. migration route was responsible for Spawning commences in the Okanogan the unusually late passage of mainy River during the lastweek in September spring chinook at Rock Island Daan, and continues throughout the month of Davidson' noted that the upstream October. salmon migrationin the ColxunbiaRiver is retarded by floods, and at stages of extreme high water may be inhibited. SUMMARY AND CONCLUSIONS In 1956 a minor flood occurred in the Coliunbia River and the unusually late Investigations of fish runs passing passage of many spring chinook could Rocky Reach Dam site have revealed be attributed to this occurrence. It that important runs of chinook and blue- should be noted that none of the fish back salmon pass the site bound for tsigged prior to July 9 was recovered in S|>awiiing areas in all mciin tributaries summer chinook spawning areas that between Rocky Reach and Chief Joseph season. Dam. The runs passing Rock Island Dam would be expected at Rocky Reach Another tagging experiment con- Dam site within Ij or Z days since the ducted during 1956 confirmed our be- dams are only Zl miles apart. lief that fall chinook spawn in the main Columbia River. Sixty- two chinook from the September migration at Rock For 1957 the spring and stimmer Island Dam were tagged and released chinook escapement pass Rocky Reach over a 3-day period. No tagged fish Dam site was approximately 11,000 were recovered fronn this lot during and 6,000 fish respectively. The es- the streaun surveys made on all tribu- capement of blueback salmon passing tary streauns. For similar experiments Rocky Reach Dam site was approxi- in August, however, in which the num- mately 67,000 fish in 1956 and 43,000 ber tagged ranged from 24 to 70 chinook, fish in 1957. some tags were recovered. Total recoveries from these summer experi- Steelhead trout and silver salmon ments averaged 10 percent. If the fall are assumed to pass Rocky Reach; escapement of chinook salmon spawned however, their numbers were not deter- in the tributaries, surely some tags mined. Although numerous in the area would have been recovered; six fish up to the year 1920, very few silver would have been recovered if the 10 salmon have been counted at Rock percent average were obtained. It Island Dam since counting commenced seems certain that these fish sp>awn in 1933. only in the main Columbia River, but because of the difficulty of finding The rate of travel of blueback tagged fish in so large a river, no re- salmon between Rock Island Dam and covery was attempted there. Zosel Dam established from tagging experiments was found to be 10.7 days Blueback salmon for the 145 miles distance or 13.5 Majority of the blueback scdmon miles per day. The difference in time passing Rocky Reach Dam site are between peaks of abundance at Rock Island and Zosel Dam also indicated ' Davidson, F. A. 1957. The effect of floods on the up- that about 10 days were required stream migratioii of the salmon in the Columbia River. Riblic Utility District No. 2 of Grant County (Ephrata, for fish to travel between the two Washington). 20 pp. Processed. points.

14 In general, spring chinook spawn 1939-1947. U. S. Fish and Wild- in the smaller, more remote tribu- life Service, Special Scientific taries, and summer chinook spawn in Report No. 55, November, the lower main tributaries. Fall chinook 63 pp., Washington, D. C. spawn in the main Columbia River. Blueback salmon spawn in that part of TALBOT, G. B. the Okanogan located in Canada, from 1950. A biological study of the ef- Liake Vaseaux to Lake Osoyoos. fectiveness of the Hell's Gate Scatterings of blueback were also found fishways. International. Pacific in the £ntiat, Methow, andSimilkameen Salmon Fisheries Commission, Rivers. Bulletin III, Part 1, pp. 3-80. New Westminster, B. C, Can- ada. LITERATURE CITED WASHINGTON STATE DEPARTMENT FISH, FREDERIC F., AND MITCHELL OF FISHERIES. G. HANAVAN. 1899- 19Z0. Annual Report of State 1948. A report upon the Grand Cou- Fish Conunissioner. Olympia, lee fish-maintenance project Washington. Various paging.

GPO (0ZS34

MBL WHO! Library Serials

I I ! \ ( lllll" I'll' 'I'ltllliVft ;

'•• "111(1111:11 I II I. Mill 5 WHSE 01492