Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Morphological heterogeneity of kokanee, "oncorhynchus nerka kennerlyi" (Suckley), in Kootenay Lake, British… Vernon, Edwin Herman 1954

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1954_A8 V4 M7.pdf [ 5.02MB ]
Metadata
JSON: 831-1.0106827.json
JSON-LD: 831-1.0106827-ld.json
RDF/XML (Pretty): 831-1.0106827-rdf.xml
RDF/JSON: 831-1.0106827-rdf.json
Turtle: 831-1.0106827-turtle.txt
N-Triples: 831-1.0106827-rdf-ntriples.txt
Original Record: 831-1.0106827-source.json
Full Text
831-1.0106827-fulltext.txt
Citation
831-1.0106827.ris

Full Text

MORPHOLOGICAL HETEROGENEITY OF KOKANEE, Oncorhynchus nerka kennerlyi (Suckley), IN KOOTENAY LAKE, BRITISH COLUMBIA. by EDWIN HERMAN VERNON • A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS i n the Department of ZOOLOGY We accept t h i s t h e s i s as conforming to the standard required from candidates f o r the degree of MASTER OF ARTS. Members of the Department of Zoology THE UNIVERSITY OF BRITISH COLUMBIA A p r i l , 1954 ABSTRACT Samples of kokanee drawn from the spawning runs to seven streams t r i b u t a r y to Kootenay lake, B r i t i s h Columbia, were compared s t a t i s t i c a l l y f o r morphometric v a r i a t i o n between streams, between areas i n the lake and between years. Kokanee within three a r b i t r a r i l y delineated areas displayed great s i m i l a r i t y i n age, s i z e , vertebral count, scale pattern and development of secondary sexual c h a r a c t e r i s t i c s . Samples from a represent-ative stream of each area d i f f e r e d s i g n i f i c a n t l y i n age at maturity, s i z e , growth rate i n the f i r s t year, vertebral count, scale row count, g i l l raker count, head length of males, growth form of caudal peduncle, eye diameter, development of secondary sexual c h a r a c t e r i s t i c s , ova per unit t o t a l weight and commence-ment of spawning period. Some s i g n i f i c a n t v a r i a t i o n i n size and vertebral count of samples from each area occurred between years but t h i s did not obscure the significance of v a r i a t i o n between areas i n any one year. Differences i n age, s i z e , vertebral count and development of secondary sexual character-i s t i c s between areas persisted i n a l l three years examined. Kokanee straying to spawn i n areas other -than t h e i r home area were i d e n t i f i e d by age, s i z e , vertebral count, scale pattern and scale row count and these strays amounted to less than three percent of 1137 specimens examined. The strong tendency of kokanee to home t h e i r parent stream i s considered of prime importance i n the -establishment and continued existence of / three partially isolated but occasionally interbreeding popu-lations in Kootenay lake. The morphometric heterogeneity of kokanee in Kootenay lake i s most probably a result of pheno-typic variation. Some genotypic morphological variation may have occurred but the interbreeding between populations i s probably sufficient to preclude sympatric speciation. TABLE OF CONTENTS Page INTRODUCTION ... 1 ACKNOWLEDGEMENTS 2 LITERATURE ON THE.SPECIES 3 Nomenclature 3 Range 4 R e l a t i o n to Sockeye Salmon -. 6 Morphometric V a r i a t i o n 7 LOCAL GEOGRAPHY OF KOOTENAY LAKE 8 MATERIALS AND METHODS 11 F i s h C o l l e c t i o n s 11 Method of Capture 11 P r e p a r a t i o n and P r e s e r v a t i o n 12 Measurements and Counts 12 Age Determination and Scale Measurements . 14 S t a t i s t i c a l Procedures • 1$ RESULTS 20 S i m i l a r i t y of Kokanee W i t h i n Areas 20 Age composition 20 S i z e composition '. 22 V e r t e b r a l counts 2 4 Scale p a t t e r n s 25. Scale r e s o r p t i o n 25 Summary of s i m i l a r i t i e s 2 6 D i s s i m i l a r i t i e s Between Areas. 26 The samples • • 2 6 Age and s i z e at ma t u r i t y 2 7 M e r i s t i c counts 2 7 V e r t e b r a l counts 27 Scale row counts .... • 3 3 G i l l r aker counts ... . 3 3 P y l o r i c c a e c a l counts .... 3 4 R e l a t i v e body proportions 3 5 Regression of head-length on f o r k - l e n g t h ... 3 5 Regression of peduncle-length on f o r k - l e n g t h 3 7 Regression of eye-diameter on f o r k - l e n g t h . . . 4 0 Growth r a t e s 4 0 Number of ova i n r e l a t i o n t o weight 4 4 Spawning p e r i o d 4 4 Summary of d i s s i m i l a r i t i e s between areas 4 5 i i Page Comparisons Between Years 47 The samples 47 Variation i n age-class d i s t r i b u t i o n between years 4& Variation i n size between years 1+8 Variation i n vertebral count between years 50 Variation i n scale pattern and scale resorption.. 50 Summary of comparisions between years 52 Degree of Straying from Home Streams 52 DISCUSSION • 55 Recapitulation of S t a t i s t i c a l • a n d Other Evidence fo r Three Populations of Kokanee i n Kootenay Lake .... 55 The Concept of In t r a s p e c i f i c Races in_Fish.es 56 Some Effects of Environment on Anatomical 'Characters . 57 Population Characteristics of Kokanee i n Kootenay Lake i n Relation to L i f e History and Environment . . . . . "58 Factors Tending Toward I s o l a t i o n of Stocks 62 SUMMARY AND CONCLUSIONS .' 65 LITERATURE CITED i i i TABLES Table . Page I. Age composition of spawning kokanee i n Kootenay lake, 1951 w. 21 I I . Size d i s t r i b u t i o n of spawning kokanee, 1951 ..... 23 I I I . D i s t r i b u t i o n of vertebral counts of kokanee, 1951 24 IV. D i s t r i b u t i o n of fork length of kokanee, 1951 .... 28 V. D i s t r i b u t i o n of weight of Kokanee, 1951 28 VI. Difference i n mean fork-length within and between areas, 1951, with mean lengths i n , brackets 29 VII. D i s t r i b u t i o n of Enumeration Data, 1951 31 VI I I . Differences between mean vertebral count x^ithin and between areas, 1951. Mean counts i n brackets 32 IX. The r e l a t i o n of head-length to body-length, 1951 kokanee 36 X. The r e l a t i o n of peduncle-length to body length, 1951 kokanee 38 XI. The r e l a t i o n of eye-diameter to fork-length, 1951 kokanee 39 X I I . Scale measurements ( i n mm.), 1951 kokanee ....... 42 X I I I . D i s s i m i l a r c h a r a c t e r i s t i c s of kokanee from three areas, 1951 •• •••• 46 XIV. Age class d i s t r i b u t i o n between years 49 XV. Variation i n mean f ork-length between years 49 XVI. Differences i n mean vertebral count (followed by p r o b a b i l i t i e s ) within and between years 1951, 1952, 1953. Mean counts i n brackets 51 XVII* The number and o r i g i n of spawning kokanee i d e n t i f i e d as strays from t h e i r home area 53 i v FIGURES Figure Page i 1. Kootenay lake 9 2. F i e l d photograph of Kokanee 13 3. Scales of kokanee from North End of Kootenay Lake, 1951. (X50) ' 15 4. Scales of kokanee from South End of Kootenay lake, 1951. (X50) 16 5. Scales of kokanee from South End of Kootenay lake, 1951. (X50) 17 INTRODUCTION i The increasing emphasis of ecology in systematics has focussed attention on the study of intraspecific groups. Ecolo-gical studies in recent years have found evidence of par t i a l population isolation within many species and subspecies. Diver (1940), from the results of a major ecological study over a large area i n England, concluded that many plant and animal populations generally considered to be continuous are in fact composed of many.partially isolated populations. Morphology remains the chief tool of the taxonomist and the development of s t a t i s t i c a l procedures has in recent years made possible the more precise definition of taxonomic features of large groups of animals. With the a b i l i t y to analyse morphometric features has come an increased awareness of variation within species. The present paper is an analysis of morphometric variation found in an isolated population of kokanee, a lake-locked Pacific salmon (Oncorhvnchus nerka kennerlvi), from a large lake in British Columbia. S t a t i s t i c a l procedures are used to define differences between kokanee which spawn in streams tributary to various areas of the lake. While this is a f i e l d problem and no experimental work has been done, the variation found in the lake i s discussed in the light of recent researches on the effects of environment on taxonomid characters, and the effects of homing behavior in Pacific salmon on the homogeneity of spawning populations. 2 ACKNOWLEDGMENTS The w r i t e r wishes to express appreciation of the following favors: to Dr. P. A. Larkin who suggested the problem and con-t r i b u t e d much advice and support throughout the project; to Dr. C. G. Lindsey f o r advice and suggestions i n the preparation of the manuscript;; and to D. P. Scott f o r advice on s t a t i s t i c a l procedures. The wr i t e r i s also endebted to the following persons f o r help and suggestions: Dr. W. A. Clemens, S. B. Smith, and F. P. Maher. The entire project was supported by the B r i t i s h Columbia Game Commission and the following personnel gave assistance i n f i e l d c o l l e c t i o n s : Inspector C. F. Keams, Game Warden R. G. Rutherglen, Fishery O f f i c e r Rod McRae and H. Sparrow. 3 LITERATURE ON THE SPECIES Nomenclature . "The sockeye salmon (Oncorhynchus nerka, Walbaum) of the North Pacific region is typically an anadromous f i s h whose young-usually remain i n a fresh-water lake for at least a year before migrating to the ocean. In a great many lakes, however, there occur populations of 0. nerka which are not anadromous, but li v e and reproduce entirely i n fresh-water. Such f i s h have been given the varietal name 0. nerka kennerlvi. (Suckley), and are variously known as kokanee, kickaninnies, land-locked sockeye, l i t t l e redfish (when mature), and silver trout (immature)" (Ricker, 1940)* Although no broad study of the comparative morphology of kokanee and anadromous sockeye has been made, no difference has been demonstrated save the much smaller size and restricted migratory habits of the kokanee. The very close relationship between the two forms has resulted in much confusion and disa-greement as to their taxonomic status in the past and this disagreement has not yet been altogether resolved. Until the l i f e history of the sockeye was thoroughly understood, early workers often confused young, precocious males of anadromous sockeye with kokanee which they approximated in size. Thus Bendire (1882) decided that the land-locked red salmon 0. kennerlvi was the young breeding male of 0. nerka. These d i f f i c u l t i e s further delayed c l a r i f i c a t i o n of the relation-ship between the two forms. 4 The early nomenclature reflects this confusion and Sockeye have been allocated to at least three genera. Walbaum (1792) originally described the species as Salmo nerka. Among the names subsequently applied were Fario aurora (Girard, 1857) and Oncor- hvnchus paucidens (Gunther, 1866). These proved to be synonymous with Oncorhvnchus nerka. established by Bendire (187&), Bean (16*83) and Jordan and Evermann (I896). Among the specific synonyms are: Salmo paucidens (Richardson, I836); S. cooperi. S. warreni, S. richardi and S. kennerlvi (Suckley, 1862); S. aurora (Gunther, 1866). In 1862 Suckley described what was probably a kokanee from Chilliwack lake in British Columbia as Salmo kennerlvi. and Bean (1891) gave subspecific rank to the lake-locked form and applied the name Oncorhvnchus nerka kennerlvi. However, for many years this rank was not recognized, some workers giving the two forms of Oncorhvnchus specific rank (Jordan & Gilbert 1881, Green 1893, Jordan 1894, Halkett, 1913, Ward 1932), designating 0. nerka as the anadromous and 0. kennerlvi as the lake-locked forms res-pectively. On the other hand, some workers have recognized only one species, 0. nerka with an anadromous and a lake-locked form. (Jordan & Evermann 1396, Evermann 1S97, Jordan & Evermann 1922, Schultz 1929, 1935a). Of recent years Canadian workers have recognized kokanee as a subspecies of anadromous sockeye with the name 0. nerka 'kennerlvi (Suckley), (Dymond 1932, Ricker 1940, Carl & Clemens 1948, 1953). Range Oncorhvnchus nerka i s abundant in waters of the North Pacific and i t s range i s not greatly different from that of other members of the genus. In the Northeastern Pacific i t occurs from Oregon to Northwestern Alaska (Clemens & Wilby, 1946) and i s the basis of a very valuable commercial fishery.. In the Northwestern Paci-f i c i t ranges from the Anadir river i n Siberia southward to the Amur river. It i s found also on Kamchatka peninsula, Sakhal-in island and the Kurile islands (Davidson & Hutchinson, 193#). Baievsky (1926) reports that in the Russian*far east between i 1909 and 1922 the annual catch of sockeye varied between 7,000,000 and 32,000,000 pounds and constituted between 3.6 % and 10$ of the to t a l catch of Pacific salmon. In North America the range of kokanee. coincides, in general, with that of sockeye^ (Carl & Clemens, 1953). Krokhin and Krojius (1936) have reported i t from Kamchatka and although the anadromous form i s not present in Japan (Jordan & McGregor, 1925), the freshwater form i s found on Hokkaido and northern Honshu Islands. In British Columbia some lakes such as Harrison and Pitt support sockeye but not kokanee^ (Ricker, 1940). Some lakes are, or have recently been, occupied by both sockeye and kokanee such as Adams, Francois, and Chilko lakes in the Fraser drainage and Osoyoos lake i n the Columbia drainage. Many lakes support kokanee but not sockeye and among these are: Kootenay, Christina and Okanagan lakes in the Columbia drainage; Canim and La Hache in the Fraser drainage.; Cameron, Home, and Shawnigan lakes on Vancouver island. 6 Relation to Sockeye Salmon The close similarity of kokanee and sockeye indicates that the two forms have diverged in very recent geological time. Although freshwater forms of Oncorhynchus masou and 0. formasanus are common in Japan (Oshima, 1935) and small freshwater populations of 0. kisutch have been reported (Foerster and Ricker 1953, Schultz 1935a), anadromous habits are most common in the genus. The con-tinuous distribution of sockeye compared with the isolation of various kokanee stocks suggests that anadromy i s primitive and lacustrine l i f e a recent phenomenon in the species. Ricker (1935) has found that some sockeye f a i l to migrate to the sea after one or two years of l i f e and these "residuals" can be differentiated at maturity from both sockeye and kokanee. In 1940, Ricker pointed out that since both migratory sockeye and non-migratory residuals are i n the same environment the difference in habits must be due to hereditary factors. He therefore postulated two stages in the formation of kokanee from sockeye salmon: (1) production of residual offshoots of sea-going sockeye. (2) a period of genetic evolution during which better adaption to lacustrine l i f e i s attained. Foerster (1947) showed that kokanee raised in an inland hatchery from a completely lake-locked stock and released as fingerlings i n a coastal stream,, migrated to the sea and matured as typical sockeye. This suggests that the hypothesis of a fresh-water origin of the species i s not altogether untenable and would accord with the general views of Tchernavin (1939) on the freshwater ancestry of Salmonids. However Milne (194#), after an extensive study of the growth, habits and morphology of Oncorhvnchus concludes that the genus arose from freshwater ancestors but that kokanee are secondarily lacustrine. Morphometric Variation Measurements of meristic "counts and other taxonomic charac-ters of sockeye are given by Schultz, (1936) and by Clemens and Wilby (1946). The range of some of these counts has been exten-ded somewhat by Milne (1948). Although kokanee are isolated i n lakes over a very wide area and some divergence might be expected to have occurred, only very meagre descriptions are available and no evidence for divergence exists. Dymond (1936) describes two specimens from Christina lake and gives body proportions and meristic counts. Schultz (1935b) gives some body proportions of 5# kokanee from Lake Washington. 8 LOCAL GEOGRAPHY OF KOOTENAY LAKE Lying i n a deep g l a c i a l trench i n southeastern B r i t i s h Columbia, Kootenay lake i s some 65 miles long and 154 square miles i n area (Fig. 1). The main portion of the lake l i e s on a north-south axis but a v a l l e y projecting westward some 22 miles contains the West Arm. The steep sides and uniformly great depth of the basin are decisive i n placing the lake i n the oligotic!ophic c l a s s . For the purposes of t h i s paper the lake has been divided into three narrow arms rad i a t i n g from the v i c i n i t y of Balfour. ' The North End extends from Balfour northwards and' i s dominated by the large inflow of water from Lardeau r i v e r . The South End extends southward from Balfour and receives the large discharge of Kootenay r i v e r . . The West Arm has a constricted entrance arid a shallow s i l l i n the v i c i n i t y of Balfour and extends westwards to drain into Kootenay r i v e r and thence to the Columbia r i v e r . Although high and currently impassable f a l l s e x i s t imme-di a t e l y belotv Kootenay lake, the period of complete i s o l a t i o n of Kootenay lake salmon from anadromous stocks i s d i f f i c u l t to es t a b l i s h . Bean (1891) c i t e s Bendire to the effect that spring salmon (Oncorhvnchus tschawytscha) occasionally overcame i n heavy freshets the high f a l l s below the lake. However, Cottingham (1947) c i t e s early f u r traders to the effect that i n 1859 p a c i f i c salmon were unknown above the f a l l s to the resident Indians. In recent years the f a l l s have been rendered completely impassable by the construction of hydroelectric dams. The pleistocene history 9 F i g . 1. Kootenay Lake. 10 of the lake includes a period of drainage southward while the West Arm was blocked with Ice/ (Schofield, 1946). It i s not known whether the ancestors of the present kokanee population entered from the Columbia by way of this southern route or whether the present outlet was for a time passable to anadromous species. At any rate kokanee have probably been isolated in Kootenay lake for at least hundreds of generations. People l i v i n g in the vi c i n i t y of Kootenay lake have for many years noticed apparently consistant differences in size and coloration of kokanee which spawn in various areas. This paper is an evaluation of homogeneity of the lake population. It consists of a morphometric study of samples drawn,?frora kokanee spawning in seven tributary streams. 11 MATERIALS AND METHODS Fish Collections Between August 30 and September 18, 1951, mature kokanee were col l e c t e d i n the spawning migration to seven streams t r i b u -tary to Kootenay lake. Samples were taken from Cultus, Sanka and Goat creeks i n the South End; Kokanee and Redfish creeks i n the West Arm; and Lardeau r i v e r and Meadow creek i n the North Endl (Fig. 1). ' Between September 3 and September 16 i n 1952, mature kokanee were again collected from Goat creek i n the South End; Kokanee creek i n the West Arm| and Meadow creek i n the North End. In September 1953 further c o l l e c t i o n s were made from Meadow and Kokanee creeks. Method of Capture Although various methods of capture were considered and t r i e d and due thought was given to the p o s s i b i l i t y of sample bias, i t was found.that a dip net was the only practicable t o o l i n the turbulent stseams of the South End and West Arm. To assure uniformity, t h i s method was employed f o r a l l c o l l e c t i o n s . The red coloration of males rendered them more e a s i l y seen and may have resulted i n some s l i g h t bias toward t h e i r s e l e c t i o n , but t h i s was not considered to be of serious e f f e c t . , I t was not possible to c o l l e c t f i s h throughout the spawning run. To make samples representative of as large a portion of the run as possible, f i s h were taken over a long section of some ay streams, on the assumption that f i s h forthest upstream had 12 migrated e a r l i e r than those further down. In other streams, c o l l e c t i o n s were made on more r e s t r i c t e d sections but over a period of several days. In Goat and Sanka creeks, two c o l l e c -t i o n s were made, nineteen and sixteen days apart respectively. Although i n general, the f i s h c o llected are most representative of the early portions of the runs, comparison of early and l a t e c o l l e c t i o n s from Goat creek suggest that t h i s bias i s not serious Preparation and Preservation To avoid possible errors i n measurement due to d i f f e r e n t i a l shrinkage i n formalin, a l l specimens were photographed against a background g r i d of ten l i n e s to the inch before preservation i n ten percent formalin^ (Fig. 2 ) . Numbered paper tags were looped through the g i l l arches before photographing and these tags l a t e r i d e n t i f i e d each preserved f i s h with i t s respective photograph. T h i r t y - f i v e millimeter f i l m was used and the deve-loped negatives were mounted i n glass s l i d e s . A l l l i n e a l measurements were made with c a l i p e r s on the negative image projected to o r i g i n a l s i z e . Measurements and Counts Fork length - the distance i n a straight l i n e from the t i p of the snout to the end of the caudal f i n at i t s center. Head length - the distance from the t i p of the snout to the most posterior portion of the operculum. Peduncle length - the distance from the i n s e r t i o n of the anal f i n to the posterior end of the scale-bearing area. 13 Fig. 2. Field photograph of kokanee. Eye diameter - the distance from the anterior to the posterior rim of the orbit. Weight - the weight of preserved f i s h after immersion in cold, running water for at least twenty-four hours. Vertebrae - the number of centra between the cranium land.:..: uro-style, counted by exposing the l e f t side of the vertebral column. G i l l rakers - the number of g i l l rakers on the l e f t half of the f i r s t g i l l arch. Counts were made after removal of this portion of the arch and two rudimentary rakers on the anterior ventral extremity were excluded. Pyloric caeca - a l l caeca were counted by individual removal with forceps. Scale rows - the number of scale rows four to six rows above the lat e r a l l i n e . Ova - counted directly in females with ovarian membranes intact. Age Determination and Scale Measurements A l l specimens were sexually mature and resorption of scale margins was in some cases extreme, particularly in males. Scales from the dorsum immediately posterior to the dorsal f i n were found to be least resorbed and were used for age determination and scale measurements. Scales were washed i n water and mounted on glass slides with a solution of Canada balsam crystals and dioxane. Well defined annuli were found in nearly a l l scales examined and are illustrated in Figures 3, 4 and 5. Scale measurements were made with an ocular micrometer f i t t e d to a microscope. 15 Creek River Fig . 3 . Scales of kokanee from North End of Kootenay lake, 1951. ( x 50) 16 Annulus II Annulus I Goat Creek Annulus II Annulus I Cultus Creek Annulus II Annulus I Sanka Creek ^ Fig. 4. Scales of kokanee from South End of Kootenay lake, 1951. ( x 50) 17 Annulus I I Annulus I Redfish Creek Annulus I I Annulus I Kokanee Creek Fig. 5. Scales of kokanee from West Arm of Kootenay lake, 1951. ( x 50) 18 S t a t i s t i c a l Procedures Analysis of variance as outlined by Snedecor (1946) was used as a preliminary test on measurement and enumeration data. Three streams and the two sexes were set up in a 3 x 2 table and tested by the F ratio. The 0.01 probability level was termed 1 significant. If analysis of variance indicated significant variation between groups, further tests were made between means to locate the position and nature of the differences. These tests involved the null hypothesis applied with the T distribution^ (Snedecor, 1946). Analysis of covariance as outlined by Snedecor (1946) was used to establish relationships to fork length of head length, peduncle length and eye diameter. The adjusted means were tested for significant variation by comparison with the F distribution i n a manner similar to that outlined above. Further tests involving the T distribution were made between group regression parameters using methods recommended by Simpson and Roe (1939). Since the relative growth rates of different body parts with respect to time are usually not the same, then regression of one body part on another i s usually curvilinear. In such regressions, variance tends to be proportional to the mean. Analysis of covariance assumes linear regression and also assumes that variance is not proportional to the mean. The use of logarithmic values of measurement data r e c t i f i e s the tendencies toward curvilinearity and variance proportionality, and satisfies these two assumptions of regression analysis. •19 .In the present instance both logarithmic and arithmetic values of body port measurements were plotted and compared. Over the range of sizes studied a straight l i n e appeared to f i t the a r i t h -metic plot as w e l l as the logarithmic p l o t . However, the a r i t h -metic plots revealed a d e f i n i t e tendency f o r the dispersion about the mean regression to be greater with increasing size of body parts, i . e . the variance tended to be proportional to the mean. The logarithmic plots corrected t h i s tendency and therefore a l l body-part measurments were transposed to logarithms before analysis. 20 RESULTS S i m i l a r i t y of Kokanee Within Areas During c o l l e c t i o n of kokanee i n 1951 i t became evident that t h e i r size was s t r i k i n g l y uniform within various areas and also that there were equally s t r i k i n g differences between areas. Specimens collected i n streams t r i b u t a r y to the South End were uniformly small. In two' streams t r i b u t a r y to the West Arm f i s h were consistently much larger, while i n two streams of the North End they were uniformly intermediate i n s i z e . The red body and green head, t y p i c a l male spawning colouration of kokanee (Schult 1935b), appeared more highly developed i n f i s h of both West Arm streams, than i n any other area. To further investigate the apparent homogeneity of kokanee spawning i n each arm of the lake c e r t a i n characters of the samples were examined i n d e t a i l . Age composition A single age class dominated the spawning run i n each area i n 1951• The predominant age of West Arm and South End f i s h was three years although a few spawned at two years. Four year old f i s h formed the major portion of runs to North End streams with three old f i s h present i n i n s i g n i f i c a n t numbers. Apart from the North End, four year f i s h were found i n only one stream -Goat creek i n the South End where they contributed 9 % of the t o t a l c o l l e c t e d . Table I presents the number and percentage of each year class i n each stream. For a l l major calculations and comparisons i n the following pages, the dominant age group of each stream i s used exclusively. No other age groups were 21 collected in numbers sufficient to form adequate samples. TABLE I Age composition of spawning kokanee in Kootenay lake, 1951. AGE GROUPS 2 3 4 AREA STREAM No. No. No. % South Cultus Cr. 4 3.6 100 96.4 End Sanka Cr. 2 2.6 76 97.4 Goat Cr. 6 3.4 155 87.6 16 9.0 West Kokanee Cr. 4 5.4 70 94.6 Arm Redfish Cr. 2 2.1 95 97.9 , > North Lardeau R. 2 2.0 100 98.0 End Meadow Cr. 0 0 120 100.0 22 Size Composition The size of individual kokanee spawning in any one area i s strikingly uniform. Mean fork lengths of samples from three streams in the South End varied from 18.19 cm. to IB.78 cm. and the differences between means were significant at the 0.01 leve l . Mean fork lengths for two streams i n the North End were 21.40 and 21.73 (p< 0.05). The two streams sampled i n the West Arm had means of 23.85 cm. and 24»66 cm. (p<0.01). Although the differences between mean fork length for streams within the same area were significant; the differences were i n a l l cases less than one centimeter. Table II summarizes the size distributions found in 1951* With the exception of Sanka creek, a l l samples contain 26 of each sex, chosen at random from the total sample of the predo-minant age class of each stream. From the Sanka creek sample only thirteen females were available. No significant difference i n size was found between sexes in any sample, although males tended to be slightly larger. 23 TABLE II Size distribution of spawning kokanee, 1951 1 - -Fork Length South North West Interval l End End Arm B D E J H F G 16.0 - 16.9 3 2 1 17.0 - 17.9 17 3 17 18.0 - 18.9 29 23 14 19.0 - 19.9 3 21 5 1 1 20.0 - 20.9 3 2 4 6 21.0 - 21.9 27 25 2 22.0- 22.9 !7 17 2 6 23.0 - 23.9 3 3 10 17 24.0 - 24.9 25 21 25.0 - 25.9 10 5 26.0 - 26.9 3 1 27.0 - 27.9 2 Total 52 52 39 52 52 52 52 Mean 0.8.19 18.78 18.22 21.73 21.40 .24.66 23.85 B • Goat Cr. D •» Cultus Cr. .8 - Sanka Cr. J B Lardeau R. H - Me ad ow Cr. F = Kokanee Cr, G a Redfish Cr. 24 Vertebral Counts ' The distribution of vertebral counts of spawning kokanee i s shown i n Table III. The counts were made on specimens from the samples described in the previous section. No significant d i f f e -rence between sexes was found in any sample. The mean vertebral counts of f i s h from three South End streams varied from 64.Oct to 64.42 and at the OXOl level of confidence these differences were not significant. The means of f i s h from the two North End streams were;65.5.0 and 65.77 and the difference was again not significant. The means of f i s h from the West Arm streams were 64.86 and 64.90 and are not significantly different. TABLE III Distribution of vertebral counts of kokanee, 1951. (AREA »STREAM !No. of Vertebrae »No. 1 \ Mean 62 63 64 65 66 67 Cultus Cr. 1 14 17 16 4 52 64.15 S. End Goat Cr. 10 14 24 4 "52 64.42 Sanka Cr. 3 7 14 14 1, 39 64.08 Lardeau R. 7 17 23 5 52 65.50 K.End Meadow Cr. 20 24 8 52 65.77 Kokanee Cr. 2 16 20 14 52 64.90 W. Arm Redfish Cr. 2 13 27 10 52 64.86 25 Scale Patterns Scale patterns typical of kokanee from each of the three areas are illustrated in Figures 3, 4 and 5» Quantitative comparisons of scale measurements were not made between samples within each area, but marked similarities were evident during examination of scales for age determination. Scales of West Arm f i s h were characterized by a very large area within the f i r s t annulus while scales from South or North End kokanee had, comparatively a much smaller area within the f i r s t annulus. Scales of North End f i s h were further characterized by a very narrow second annulus, at times being re-duced to a single broken circulus. The uniformity of scale pattern within an area was sufficient to enable the recognition in a sample of individuals from another area, i.e. only three scale patterns were observed and a kokanee spawning outside i t s "home" area could be identified as such. Scale Resorption Progressive resorption of scales as sexual maturity is approa-ched appears to be characteristic of Pacific salmon. In kokanee the dorsal and ventral margins are m,ost extensively resorbed. Correlated with this resorption i s a thickening of the skin and an envelopment of the projecting portions of scales. These breeding modifications are most strongly developed in males. In Kootenay lake, the degree of dermal thickening and scale resorption was very similar i n breeding kokanee within each area. It was moderate in both streams of the North End and extreme i n both streams of the West Arm. On the other hand, kokanee of the three streams of the South End displayed very slight dermal thickening and scale resorp-26 tion, even in males completely spawned out. Summary of Similarities In general, kokanee which spawn within the areas designated as North End, South End and West Arm display much greater homoge-neity than do kokanee in Kootenay lake as a whole. Spawning kokanee in the North End are four years of age, display a characteristically narrow second scale annulus and moderate dermal thickening. They have a mean fork length of about 21.5 cm. and a mean vertebral count of 65.63. West Arm kokanee spawn at three years of age, display a very wide scale area within the f i r s t annulus and extreme scale resorp-tion and dermal thickening. Their mean fork length i s about 24.5 and mean vertebral count is 64»&v#. West Arm kokanee also display a uniformly high development of breeding color. Finally, kokanee from the South End form a group maturing at three years of age, having a narrow area within the f i r s t scale annulus and practically no dermal thickening or scale resorption. They have a mean fork length of about 18.5 cm. and a mean vertebral count of 64.22. Dissimilarities Between Areas The samples As shown in the previous section, examination of- kokanee samples collected from seven streams tributary to Kootenay lake indicated that considerable homogeneity resulted i f the lake were rather ar b i t r a r i l y divided into three major areas. While i t would be most desirable to examine samples from a l l streams i n further 27 detail, i t was considered most profitable to compare samples of kokanee from one stream i n each area, assuming each, provisionally, to be typical of that area. Fish from Cultus creek i n the South End, Kokanee creek i n the West Arm and Lardeau river in the North End were selected, examined and compared in more detail as repre-sentative of each area. Twenty-six of each sex were chosen at random from the predominant age group in each area and these form the samples for a l l the comparisons which follow. Age and size at maturity Whereas four year old f i s h comprised over 98 percent of the North End sample, South End and West Arm kokanee were predominantly three years old at maturity, this group contributing 96.4 percent and 94.6 percent to the respective total samples. (See Lardeau, Cultus and Kokanee in Table I ) . Kokanee from each area formed very distinct groups according to size. The distribution of fork lengths i s illustrated in Table IV. A comparison of variation i n size within and between areas is shown in Table VI. The differences between means were highly significant (p<.01). The distribution of weights i s shown in Table V. The differences between mean weights were also highly significant (p<.01). Between sexes i n each area there were no significant differences in means of weight or length. Meristic counts Vertebral count The distribution of vertebral counts i s shown in Table VII A. A preliminary analysis of variance indicated that there were significant differences between areas (p<.01), but none for sexes TABLE IV Distribution of fork length of kokanee, 1 9 5 1 Fork Length * (cm.) 1 6 . 1 1 7 . 1 1 8 . 1 1 9 . 1 2 0 . 1 2 1 . 1 2 2 . 1 2 3 . 1 2 4 . 1 2 5 . 1 2 6 . 1 27 n Area Mean 1 7 . 0 1 8 . 0 1 9 . 0 2 0 . 0 2 1 . 0 2 2 . 0 23.0 2 4 . 0 2 5 . 0 26.0 2 7 . 0 2 3 . 0 S. End 1 8 . 7 $ 2 2 7 17 2 N. End 2 1 . 7 3 1 8 29 1 1 3 W. Arm 2 4 . 6 6 2 1 1 2 5 9 4 1 TABLE V Distribution of weight of kokanee, 1951 Weight (grams) 46 61 76 9 1 1 0 6 1 2 1 1 3 6 1 5 1 ' 166 1 8 1 1 9 6 > 2 1 0 Area Mean < 4 6 6 o 7 5 9 0 1 0 5 1 2 0 1 3 5 1 5 0 1 6 5 1 8 0 1 9 5 2 1 0 S. End 5 7 . 9 3 32 1 6 1 N. End 1 0 5 . 4 1 2 2 3 2 1 5 W. Arm 1 5 8 . 2 2 4 1 6 13 8 5 ' 3 1 TABLE VI Differences in mean fork-length within and between^ areas, 1951 with mean lengths in brackets - South End . j North End. . . West Arm Sanka Cultus [Meadow Lardeau Redfish Kokanee (18.22) (13.78) 1(21.40) (21.73) (23.85) (24.66) South End Goat (18.19) 0.03 XX 0.59 I 3.21 3.54 5.66 6.47 South End Sanka (18.22) 0.56 | 3.18 3.51 5.63 6.44 South End Cultus (18.78) i 2.62 2.95 5.07 5.88 North End Meadow (21.40) 0.33 I 2.45 3.26 North End Lardeau (21.73) 2.12 2.93 West Arm Redfish (23.85) 0.81 • 1 ft Differences to l e f t of heavy line are within same" area, those to rigth. are between.areas.-Xft Non-significant. A l l others significant at 0.01 level. 30 or i n t e r a c t i o n . As t h i s indicated that there were no s i g n i f i c a n t differences i n vertebral counts between sexes, data f o r the two sexes were combined to form samples of f i f t y - t w o f i s h f o r each area. The differences between a l l resultant means were s i g n i f i c a n t at the 0.01 l e v e l of p r o b a b i l i t y . The North End mean was highest followed by West Arm and South End. A comparison of vertebral count means, within and between areas i a shown i n Table V I I I . Analysis of variance and T te s t s are most suited to data d i s -playing continuous and r e l a t i v e l y unlimited v a r i a t i o n . Vertebral counts of kokanee form r e s t r i c t e d d i s t r i b u t i o n s and f o r data of t h i s type a chi-square test may be used. As a check on methods, a chi-square test based on marginal p r o b a b i l i t i e s (contingency table) was applied to the vertebrae data and gave su b s t a n t i a l l y the same res u l t s as analysis of variance. The chi-square value was 36.013 with 4 d.f. (p<.01). This was regarded as reasonable assurance that analysis of variance was s u f f i c i e n t l y e f f e c t i v e and sensitive when applied to these data. TABLE VII, Distribution of Enumeration Data 1951 A Vertebrae Area 62 63 64 65 66 67 Mean D : South End 1 14 17 16 4 64.15 F : West Arm 2 16 20 14 64.90 j North End 7 17 23 5 65.50 B Lateral Line Scale Rows Area 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 Mean D F J South End 2 1 2 4 5 5 4 6 4 6 2 2 4 1 1 1 127.96 West Arm 1 2 9 5 8 5 4 5 8 2 2 1 123.06 North End 3 3 3 11 5 4 9 2 5 2 4 1 130.10 C Gil l Rakers Area 31 32 33 34 35 36 37 38 Mean D South End 2 10 12 9 9 9 1 33.8$ J North End 1 4 14 8 20 5 35.10 F West Arm 3 9 16 10 11 3 35.50 D ^Pyloric Caeca Group 50-54 55-59 60-64 65-69 70-74 75-79 80-8$ Mean S. End 1 4 7 7 6 1 65.00 S. End P 8 8 4 4 2 58.50 W; Arm 2 5 10 5 1 3 68.38 W. Arm 1 2 7 10 2 3 . 1 66*15 N. End 2 6 11 4 3 67.42 N. End p 1 8 9 5 3 67.19 3;2 TABLE VIII (Differences between mean vertebral count within and between* areas, 1951. Mean counts i n brackets , ) 1 Soutl-l End •North End iWest Arm Sanka Cultus 'Meadow 'Lardeau Re.dfish' Kokanee (64.03) (64.15) (65.77) (65.50) (64.36) (64.90) Lth End ,Goat (64.42) 0.34 0.27 1.35 XX 1.03 XX 0.44 Xft 0.43 XX Lth End , Sanka (64.03) 0.07 1.69 XX 1.42 xx 0.73 XX 0.32 AA o co iCultus (64.15) 1.62 Aft 1.35 XX 0.71 xx 0.75 XX End M^eadow (65.77) , T _ _ _ _ , 0.27 0.91 AA 0.37 XX North vLardeau (65.50) 0.64 1 XX 0.60 XX West ' Arm lRedfish 64.36 0.04 j A Mean differences to l e f t of heavy l i n e are within areas, those to the right are between areas. Aft Differences s i g n i f i c a n t at 0.01 l e v e l . 33 Scale row counts i The distribution of scale row counts is shown in Table VII B. A preliminary analysis of variance indicated that there were significant differences between streams but none between sexes. Some interaction was evident, significant at the five percent level, and reflected the reversal, in the South End sample, of a trend for males to have fewer scale rows than females. As no significant differences were found between sexes, no importance was attached to the slight interaction and sexes were combined to form samples of fifty-two f i s h for each area. North End kokanee had a mean scale row count significantly higher (p<r.01) than that of South End or West Arm f i s h . There was no significant difference between means of South End and West Arm kokanee. The high scale row count of North End kokanee may be related to a greater number of somites as suggested by the high vertebral count. However, complete correlation of scale rows with vertebrae i s not t?o be expected because a variable amount of branching of scale rows occurs above the lateral line (Neave, 1943), and scale rows were counted four to six scales above the lateral l i n e . G i l l raker counts Analysis of variance indicated significant differences in g i l l raker counts between areas but no significant differences, between sexes, and therefore sexes were again combined to form samples of fifty-two from each area. Further tests between means established that while there was no significant difference between mean g i l l raker counts of West Arm and North End kokanee (p<»5), South End kokanee had a significantly lower g i l l raker count than either of the other two groups ( p<.01) . 34 The distribution and means of g i l l raker counts are shown i n Table VII C. It can be seen that mean g i l l raker counts form a progressive (South End < North End < West Arm) which i s superficiall similar to that formed by size. (Tables IV and V). To assess any effect of size on g i l l raker count, a scatter diagram of i n d i v i -duals for each stream was constructed and this indicated no corre-lation of g i l l raker count of an individual with individual weight, i.e. within areas there was no correlation of g i l l rakers with weight. The lower g i l l raker count of South End kokanee i s thus a real difference which i s unassociated with the smaller size of these f i s h . Pyloric caecal counts A preliminary analysis of variance indicated that there were significant differences between area and sex means of pyloric cae-cal counts. As sexes could not be combined for any further tests between means, they were compared separately. No significant differences were found between males of each area or between females of the North End and West Arm; but South End females were s i g n i f i -cantly lower in caecal count than North End or West Arm females, (p <.01). The meaning of this difference i s not 4clear and may possibly be associated with a departure from normalcy in the distribution of caecal counts for South End females. Table VII D l i s t s the distributions of caecal counts for a l l areas. A scatter diagram of individuals for each stream was cons-tructed and revealed no correlation of caecal count with weight of individuals. Over the size range studied there was no tendency for larger kokanee to have more caeca. 35 Relative body proportions Regression of head-length on fork-length. Samples consisted of 26 of each sex from each of the three areas previously discussed. Preliminary analysis of covariahce indicated greater differences between sexes than between areas and therefore no samples could be combined. The analysis showed, however, that s i g n i f i c a n t differences, apart from those a t t r i b u -table to sex, existed between the adjusted head-length of various areas. From the regression equation of each sample the mean log head-length was calculated at a mean log fork-length of 1.33398. (Table. I X ) . This mean log fork-length i s the geometric mean of the t o t a l sample of 156 f i s h and i s equivalent to 21.57 cm. The head-lengths were thus adjusted f o r body-length v a r i a t i o n between samples. The corrected mean log head-length of West Arm males was s i g n i f i c a n t l y higher (p<..01) than those of North End or South End males. The means of North End and S o u t h End male samples were not s i g n i f i c a n t l y d i f f e r e n t , (p<.3). No s i g n i f i c a n t difference was found between corrected mean head-lengths of female kokanee. The shorter heads of North End females approached the l e v e l of significance (p<.05) when compared with West Arm females, but no difference was found between North and South End females (p<.3) or between West Arm and South End females (p<.2). No s i g n i f i c a n t differences were found between the rate of head growth i n r e l a t i o n to body growth (regression c o e f f i c i e n t , Table I X ) . No sample exhibited a rate s i g n i f i c a n t l y d i f f e r e n t from unity i . e . there was no r e a l deviation from isometric growth. TABLE IX The relation of head-length to body-length, 1951 kokanee. A r e a West Arm South End , North End S e x ? <** •? Number in sample 26 26 26 26 26 26 Mean log. fprk length 1.3942 1.389$ 1.27$5 1.26$1 1.3372 1.3361 Mean log. head length .7766 •7215 .6292 .5735 .6606 .6251 Regression Coefficient 0.9722 1.0$72 0.7249 1.1063 1.0934 1.0410 Prob. of Regression » 0 < .01 < .01 < .01 < .01 < .01 < .01 Mean log. head length at log. fork-fork lengthsi#3339g .7165*.0085 .660$*.0169 .6694*0103 .6349*.00$5 .6571*.0031 .6229*.005$ 37 i -In summary, West Arm spawning males have longer heads than apawning males from other areas. There is a tendency for mature females of the North End to have shorter heads than those of the West Arm. The relation.of head-length to body-length did not di f f e r significantly from isometry in any sample. The regression of peduncle-length on fork-length A preliminary analysis of variance of the regression of cau-dal peduncle-length on fork-length indicated significant differen-ces between areas but none between sexes, and therefore sexes were combined for further tests. The mean log peduncle-length at a fork-length of 21.57 cm. was calculated from the regression equa-tion of each sample. (Table X). • There were no significant diffe-rences between the corrected mean peduncle-lengths. Although the regression coefficients (Table X) show great variation (and a progression, North End < West Arm < South End), only the extremes, North End and South End, are significantly different. ( <.0l). The association of peduncle-length with fork-lenth is not very strong over the range of sizes in the data. This Is reflected in the North End sample where the probalility of regression was reduced almost to 95 percent. This lack of strong association results in large probable errors for the coefficients of regression and reduces the significance of their differences. Only North End and South End kokanee are different in the rate of increase of peduncle-length with fork length. In summary, there i s no difference in mean peduncle-length of kokanee from the three areas. In the North End, larger kokanee tended to have disproportionately shorter peduncles TABLE X The relation of peduncle-length to body-length, 1951 kokanee A r e a North End West Arm South End Number in sample 52 5 2 52 Mean log. fork length 1.3367 1.3920 1.2733 Mean log. peduncle length .4678 .4583 : * 3 7 3 6 Regression Coefficient .391 i .197 .826 - .297 1.420*.249 Prob. of Regression » 0 < .05 < .01 - < .01 Mean log. peduncle length at log. fork length of 1.33398 .4668 i .0044 .4104 - .0571 .4598 * .0047 TABLE XI The r e l a t i o n of eye-diameter to fork-length, 1951 kokanee A r e a South End North End West Arm Number i n sample 52 52 52 Mean l o g . fork-length 2.2733 2.3366 2.3920 Mean l o g . eye-diameter 0.90$1 0.93$$ 0.994$ Regression c o e f f i c i e n t .441 * .140 .$2$ * .117 .511 * .16$ Prob. of Regression = 0 < .01 < .01 < .01 Mean l o g . eye-diam. at log. fork length B 1.3339$ .934$ ± .00$$ .9366 ± .0019 .9652 * .00$1 40 (negative allometry). In the South End larger kokanee tended to have disproportionately longer rpeducles (positive alleraetry). In the West Arm, kokanee tended to have a peduncle-length proportionate to size (isometry). Regression of eye-diameter on fork-length Preliminary analysis of variance indicated significant d i f f e -rences between areas but none between sexes and therefore sexes were combined again to form samples of 52 kokanee from each area (Table XI). Mean eye-diameter was adjusted in each sample to a fork length of 21.57 cm. The adjusted maan eye-diameter of West Arm kokanee was significantly greater than that .of North End kokanee (p«c.01) and that of South End kokanee (p<.02). Fish from the latter two areas had adjusted mean eye-diameters which were not significantly different. Detailed comparisons of regression coefficients indicated no significant difference between f i s h of any area in growth of eyes relative to size. A total mean regression coefficient of 0.709 was calculated for the three samples combined. In summary, West Arm kokanee had larger eyes than North End or South End kokanee. There was no difference between areas in the rate of growth of eye-diameter with fork-length. Growth rates Adequate samples of kokanee from each area were available only for mature f i s h and therefore growth rates could not be calcula-ted directly from lengths. However, Mottley (1941) has shown that for Kamloops trout in Paul lake, British Columbia, scale size can 41 be regarded as an unbiased estimate of the length of f i s h , and can be used directly as an index of growth. Milne (1948) has, i n effect, demonstrated the same situation in sockeye salmon. Because scales of spawning kokanee have suffered marginal resorption to some degree, and this feature precludes the rigorous mathematical treatment necessary to establish the relationship of scale size to f i s h length, i t has been assumed that scale size gives an unbiased estimate of fi s h lenth. In support of the assumption that scale size i s an estimate of f i s h length, a very high correlation was found between scale radius to the second annulus and fork length of spawning kokanee from the West Arm and South End. The correlation coefficient was 0.830 (p<.001). Table XII C illustrates the close correlation by comparing differences of f i s h from the two areas expressed as percentages. Mean fork length of-West Arm kokanee i s 31*31$ greater than that of South End kokanee. Mean scale radius to annulus II in the West Arm f i s h i s 31.69$ greater than that of South End f i s h . The correspondence i s very close. Examination of scales from kokanee angled in Kootenay lake between May and September suggested that annuli are formed late in June. As can be seen in Figures 2, 4 and 5, scales of mature kokanee taken in September show very l i t t l e marginal growth after the formation of the last annulus. This probably reflects slacke-ning of body growth correlated with rapid growth of sexual pro-ducts, as well as some resorption of scale margins. The following scale measurements were nade on scales of 26 kokanee of each sex from each area: TABLE XII A. Scale measurements (in m.m.)., 1951 kokanee West Arm North End South End a 1 Mean anterior radius to annulus I 1.5721 .3631 .8081 a b Mean anterior radius to annulus II 3.0412 2.7756 2.3092 b c Mean ant. radius from annulus I to annulus II 1.4765 1.5210 1.5007 c B. Comparisons of scale measurments; probability of drawing any two samples from the same population, a, b and c refer to respective means of Table XITA. West Arm North End North End a. b c a b c a v> .01 b >.01 c South End a >.0l ^.05 b •>.01 * .1 c *j±i -C. Correlation of scale size and fork length Mean anterior radius to annulus II Mean fork length at maturity 3.0412 24.66 Cultus Creek 2.3092 18i 78 Difference .0732 5.88 Percent difference 31.69 % 31.31 % 43 (a) Anterior radius to annulus I (b) Anterior radius to annulus I I (c) Anterior radius from annulus I to annulus I I . A preliminary analysis of variance established no s i g n i f i c a n t v a r i a t i o n between areas, and therefore the sexe.s were combined f o r further tests between area means. The mean scale measurements are l i s t e d i n Table X I I A and the p r o b a b i l i t i e s of drawing any two of these means from a single population are l i s t e d i n Table X I I B. The outstanding feature of t h i s analysis i s the comparati-vely large size of scales from West Arm kokanee at the f i r s t annulus. They were 82 $ larger than North End scales and 88 % larger than South End scales; these differences were highly s i g n i f i c a n t . Scale radius to the f i r s t annulus was greater i n North End f i s h than i n South End f i s h but the p r o b a b i l i t y of t h i s being due to chance was approximately .03. Scale size of West Arm kokanee at the second annulus was correspondingly larger than those of North End or South End f i s h . Mean scale r a d i i from the f i r s t to the second annulus showed no s i g n i f i c a n t differences between means of any groups. In summary, assuming scale measurements to be estimates of r e l a t i v e f i s h lengths, the following facts emerge: (a) The large size of kokanee i n the West Arm i s the r e s u l t of a much greater rate of growth i n the f i r s t year. Growth rate i n the second year i s not greater than that of other areas. (b) Growth rate i n the second year appears reasonably constant f o r f i s h of a l l areas i n Kootenay lake, and t h i s suggests 44 that environment is f a i r l y constant in i t s effects on yearling kokanee. (c) Mature kokanee from the North End are larger than those of the South End as a result of slightly better growth in the f i r s t year but chiefly because of an extra years growth. Number of ova in relation to weight Ten females from the West Arm, 26 from the South End and 23 from the North End had ovarian membranes intact. The eggs of each female were counted and the total divided by the individual total body weight. The mean number of eggs per gram total body weight was calculated for each area. South End females averaged 3.44 eggs per gram and this was significantly greater (p<.01) than either North End females at 2.33 eggs per gram or West Arm females at 2.06 eggs per gram. The latter two means did not differ signi-ficantly. Thus South End females produced a relatively large number of eggs for their size. Spawning period Observations have not been sufficiently extensive to ascer- -tain the relative times of peak spawning or termination of spaw-ning in the various areas of Kootenay lake. However, observations by the writer, supported by those of local inhabitants, indicate that kokanee in the South End commence spawning 10 to 15 days earlier than those of the North End. Commencement of spawning in West Arm kokanee appears to be intermediate in time. In general, the spawning period i s September and October, and after the i n i t i a l two weeks, kokanee are spawning in nearly a l l areas. 4 5 Summary of dissimilarities between areas Mature kokanee from the West Arm differed from those of other areas in the following features: large size, intermediate number of vertebrae, large heads and eyes of males, high rate of growth in f i r s t year and extreme resorption of scales and dermal thicke- -ning. Kokanee from the North End were characterized by: maturing at four rather than three years, intermediate size, high number of vertebrae, high scale-row count, small heads of females, caudal peduncle growth displaying negative allometry, intermediate growth in the f i r s t year, and late commencement of spawning. South End kokanee were characterized by: small size at maturity, low vertebral count, low number of g i l l rakers, caudal peduncle growth displaying positive allometry, very slight scale resorption and dermal thickening, large number of ova in relation to total body weight, small growth in the f i r s t year and early commencement of spawning, A summary, in general terms, of variation between areas i s presented in Table XIII. Kokanee from the West Arm of Kootenay lake display , "... breeding modifications such as scale resorption, dermal thickening and heightened coloration that are more highly developed than in kokanee from other areas in the lake. The larger heads and eyes of males in West Arm kokanee probably result from a greater deve-lopment of secondary sexual characters. Caution is necessary in interpreting morphometric variation in mature Pacific salmon. Milne ( 1 9 4 8 ) found that a sharp change TABLE XIII •Dissimilar characteristics of kokanee from three areas, 195l« West Arm Worth End South End Predominant age at maturity 3 ,4 3 Size at maturity large . intermediate small Growth rate in f i r s t year high low very low Vertebral count intermediate high low Scale row count - low high low G i l l raker count high high low Head length (males) long short short Growth form of caudal peduncle isometric neg.allometric pos. allometric Eye diameter large small small Scale resorption and dermal thickening extreme moderate slight Spawning period (commencement) intermediate late early Ova per unit total weight of fis h low low high 47 in the relation of body parts (particularly head parts) to total length took place at the time of sexual maturation. Schultz (1935) has shown that breeding modifications of male kokanee include lengthening of the snout and jaws with a resultant increase i n head-length. Davidson (1935) found that in pink salmon (Oncorhvnchus  gorbuscha). the changes in the head region associated with develop-ment of secondary sex characters included elongation and displace-ment of the pterortic, frontal, maxilla and premaxilla bones. Comparisons Between Years The Samples Kokanee were collected from three areas in 1952 and two areas i n 1953* These samples were not examined in respect to a l l fea-tures studied in the 1951 samples but ages of a l l specimens were determined, scale pattersn observed, sizes recorded and vertebrae counted. Samples were not taken from the same streams of each area in a l l cases. Thus in 1952 the South End sample was taken for convenience from Goat creek rather than Cultus. In 1951, when both streams were sampled the mean fork-length of Goat creek kokanee was 0.59 cm. less than that of Cultus. This was a signi-ficant difference (p<.01) but was much less in magnitude than the differences between areas (See Table VI). However, to make com-parisons more legitimate the Goat creek sample i s made representa-tive of the South End in 1951. In the North End, Meadow creek was sampled in 1952 and 1953 and therefore the Meadow creek samples were taken as representative of the North End for comparisons between years. Comparisons of "Meadow creek and Lardeau river sizes and vertebral counts are 48 shown in Tables VI and VIII. Variation in age-class distribution between years Table XIV shows that age-class distribution of spawning kokanee remains reasonably constant between years. The runs i n each area are dominated by a single age-class each year. In the West Arm three year old f i s h contributed 87 to 96 percent of the spawning population. Kokanee in the North End are almost 100 percent four year f i s h at maturity. South End spawning kokanee consist of from 87 to 97 percent three year old f i s h . Only in the South End (Goat creek) were kokanee found which were older than the predominant year class. Most of the young fish in the samples were precocious males or "jacks". It w i l l be shown below that some of these deviates from the predominant year class are strays from other areas. Variation in size between years Table XV illustrates the variation in mean fork-length between years. As no significant difference was found between sexes, they were combined to form larger samples in a l l cases. Although the size of kokanee varies from year to year the progression West Arm> North End > South End i s stable. A significant reduction in size occurred in a l l areas in 1952. It was particularly noticeable i n the South End where fork-length dropped 14.1 percent from the previous year. The reductions in the North End and West Arm were 6.8 percent and 8.3 percent respectively. Examination of South End scales in 1952 disclosed a broad second annulus sugges-ting that these f i s h had met adverse conditions i n their second winter. 49 TABLE XIV. Age class distribution between years Year Age West Arm North End South End No. < No. fo No. 1951 2 4 5,40 6 3.39 3 70 94.60 155 87.57 4 120 100 16 9.04 2 2 2.32 1952 3 49 100 84 97.68 4 73 100 2 8 12.3.1 • ... 1953 3 57 87.69 1 1.05 4 95 98.95 TABLE XV Variation i n mean fork-length between years 1951 1952 1953 West Arm 24.66 22.62 23.28 North End 21.40 19.95 20.49 South End 18.19 15.62 I 50 In 1953, North End f i s h remained significantly smaller than 1951 f i s h and were not significantly different from 1952. West Arm kokanee in 1953 were significantly larger than in 1952 but remained significantly smaller than in 1951. Variation in vertebral count between years In Table XVI are li s t e d marginally, the mean vertebral count of samples from each area in 1951, 1952 and 1953 (a sample was not taken from the South End in 1953)• Also l i s t e d , i n the body of the table, are the differences between means and the probabi-l i t i e s of these differences occurring by chance from a single population. It has been shown previously that .vertebral count does not vary significantly between streams i n the same area. From Table XVI i t can be seen that vertebral count can vary significantly in the same area between years as this occurred each year in the North End. Although variation i n vertebral count between years i s not always in the same direction i n each area, this variation does not obscure the significance of variation between areas in any one year. Variation in scale pattern and scale resorption Scale characteristics were not compared quantitatively between years but scales were examined for age determination each year. The comparatively large radius to the f i r s t annulus of West Arm kokanee was maintained as a striking characteristic of this area in 1952 and 1953. Scale patterns of North End and South End kokanee also appeared to remain stable except for a previously mentioned widening of the second annulus of South End kokanee in 1952. •TABLE XVI .Differences i n mean vertebral count (followed by p r o b a b i l i t i e s ) within and .between* years 1951, 1952, 1953. Mean counts i n brackets. 51 North End 52 (66.19) N. End- 53| (65.25) ] ¥.- Arm' 51 (64.90) W.- Arm 52 (64.66) ¥.- Arm 53 (64.67) S. End 51 (64.42) S. Eiid 52 (64.09) N. End 51 (65.77) 0.42 p^.01 0.52 p ^.01 XX 0.37 p < .01 1.11 p«c.01 • 1.10 p <• .01 XX 1.35 p-^.Ol 1.68 p-^.01 tf. End 52 (66.19) 0.94 p<c.01 J 1.29 p ^ .01 xx ' 1.53 p -c.Ol 1.52 p-c .01 1.77 p<. .01 XX 2.10 p-i .01 N. End 53 (65.25) " 0.35 .02*p<.05 0.59 P <..01 AA 0.58 p^.01 0.83 . P^-01 1.-16 p-s. .01 W. Arm 51 (64.90) 0.24' ,2-c p *L .3 0.23 XX 0.48 p-£ .01 o.si P< .01 W. Arm 52 (64.66) ©.01 p-c.5 0.24" ,2<^p <_.3 AA 0.57 p.*: .01 W. Arm 53 (64.67) 0.25 ,Z*c p -c.3 0.58 p^.01 S. End 51 (64.42) t- •  • I 0.33 J .02*p^0^ A Differences to l e f t of heavy l i n e are between years i n same area; those to right; are between areas. AA Differences between areas i n same year. 52 ; West Arm kokanee continued to display extreme scale resorption and dermal' thickening, as well as a higher development of breeding coloration in comparison to f i s h from other areas. Summary of comparisons between years Examination of age, size, vertebral count, scale pattern and scale resorption in samples from three areas for three years indi -cates that significant variation occurred between years in some cases, but the variation did not obscure the significance of variation between areas in any one year. Differences between areas were essentially the same each year. Degree of Straying from Home Streams By examination of scale pattern, size, scale resorption and in one case, vertebrae and scale rows, i t was possible to identify kokanee spawning in an area other than their native or home area. The number and proportion of these strays are shown in Table XVII. Of the 1131 kokanee sampled over a three year period a total of 32 or 2.83 % were identified as strays^ (21 males, 11 females). Of this number, 15 were South End fis h which were spawning in the West Arm, one was a West Arm f i s h spawning in the South End, while no strays from other areas were found among the 391 kokanee exa-mined from the North End. The largest group of strays was found in Goat creek in 1951 (South End). This group of 16 f i s h constituted 9 % of the total collection and was sufficiently large to from a sample for statis-t i c a l comparisons. Mean fork-length, vertebral count, scale row count and scale r a d i i did not differ significantly from North End —kokanee and except for scale r a d i i , did di f f e r significantly from 53 TABLE XVII •The number and origin of spawning kokanee identified as strays ^from their home area. A r e a Stream , Total No. Strays Origin of of sample No. Strays 1951 West Arm Kokanee Cr. 74 3 4.0 S. End Redfish Cr. 97 8 8.2 S. End North End Lardeau Ri 102 0 0 Meadow Cr. 120 0 0 South End Cultus Cr. 114 0 0 Sanka Cr. 78 1 1.3 W. Arm Goat Cr. 177 16 9.0 N. End 1952 West Arm Kokanee Cr. 49 2 4.1 S. End North End Meadow Cr. 73 0 0 South End Goat Cr. 86 0 0 1953 West Arm Kokanee .Cr. 65 2 3.1 S. End North End Meadow Cr. 96 0 0 Totals 1131 32 2.33 . 1 54 South End kokanee. Thus these f i s h could be assigned to the North End with some confidence. Because morphological variation between streams was slight, and significant only in the case of fork-length and weight, i t was impossible to identify strays from within one area. It i s possible that considerable straying occurs between adjacent streams but i t has been impossible to assess i t s degree. 55 DISCUSSION Recapitulation of Statistical and Other Evidence for Three Populations of Kokanee in Kootenay Lake. The samples taken from seven streams tributary to Kootenay lake in 1951 demonstrate that much homogeneity becomes apparent i f the lake i s considered as three large, interconnected areas. This was concluded after an examination of age and size compo-sition, vertebral count, scale pattern and degree of breeding modification. These characteristics tended to have much less variation within areas than between areas. Investigation of further characteristics of kokanee from a representative stream of each area disclosed further di f f e -rences between areas. These differences were in scale-row count, gill-raker count, head-length, growth of caudal-peduncle, eye-diameter, growth rates, number of eggs in relation to weigth and spawning times. Examination of age and size composition, vertebral-count, scale pattern and breeding modifications disclosed considerable stability of these characteristics over a period of three years. Variation between years was less than that between areas. Individual kokanee spawning in areas other than those in which they originated were identifiable. The degree of stray-ing from one area to spawn in another was slight, with strays amounting to less than 3 percent of 1131 specimens examined. Straying within areas could not be evaluated. A l l evidence indicates that kokanee i n Kootenay lake are 56 grouped into at least three p a r t i a l l y i s o l a t e d , but occasionally interbreeding populations. The Concept of In t r a s p e c i f i c Races i n Fishes Investigation of morphometric v a r i a t i o n , p a r t i c u l a r l y of meristic series, i n wide-ranging species of fishes has i n many cases disclosed a lack of homogeneity with respect.to c e r t a i n anatomical characters (Heincke 1898, Schmidt 1918, Hubbs, 1934, Lissner 1934, Tester 1937). By a r b i t r a r i l y d i v i d i n g the t o t a l range of a species into several areas homogeneity could be demonstrated within the resultant i n t r a s p e c i f i c groups. These homogeneous groups, usually with rather i l l - d e f i n e d ranges, have most commonly been c a l l e d races and have been considered to be p a r t i a l l y i s o l a t e d interbreeding populations. Evidence of discontinuity, derived from morphometric analyses, have f o r many years been u t i l i z e d by fishery b i o l o g i s t s as a basis for conservation measures designed to protect econo-mically important f i s h . Heincke (1898)described differences d i s -tinguishing B a l t i c from North Sea races of herring and i n sub-sequent years much data was accumulated on i n t r a s p e c i f i c v a r i a -t i o n of ground-fish as well as herring. Meristic characters, p a r t i c u l a r l y vertebrae, were u t i l i z e d to demonstrate s t a t i s t i c a l differences between various spawning populations. The factors c o n t r o l l i n g the observed v a r i a t i o n were l i t t l e understood and were not of great importance f o r the purposes of establishing the degree of independence of stocks and the formulation of regulatory programs. In 1929, the International Council f o r 57 the Study of the Sea held a symposium at which papers were pre-sented dealing with herring races from Spain to Scandinavia. Most participants appeared to take for granted that the morpho-metric variation they were discussing was a genotypic phenome-non, although Bjerkan Ts paper cast serious doubts on the v a l i -dity of this concept. Some Effects of Environment on Anatomical Characters While fishery biologists had largely confined themselves to measuring and analysing s t a t i s t i c a l l y the morphometric variation within species, others had studied the effects of environment on morphology. Schmidt (1918) showed that tempe-rature and salinity could affect the number of meristic ports. Hubbs (1926) discussed the effects of development rate on meris-t i c ports. D'Ancona (1927) summarized observations of several authors on both animals and plants, showing that temperature can affect the number of various counts used in taxonomic distinc-tions. Rquiisefell and Dahlgren (1932) found that differences in vertebral counts between year classes of Pacific herring in Alaska were probably due entirely to environmental temperature during early development. Runnstr^m (1933) reported a similar analysis for Atlantic herring in Norwegian waters, while Tester (1938) found a similar situation in British Columbia waters. Mottley (1934) was able experimentally to decrease the number of lateral line scales of Kamloops trout by increasing the temperature during early development. Taning (1944, 1952) showed by a series of carefully controlled experiments on brown trout that some meristic counts can be raised by either high or 58 low temperature during development within the egg. Oxygen and carbon dioxide tensions were also shown to have an effect on meristic counts. The effects of environment on growth rate are well known. Martin (1949) has shown that rate of early development and subsequent rate of growth may both influence body JTorm through their effect on body size at inflections between growth stanzas, the direction of inflection determining whether the subsequent relative sizes of body parts are large or small. Population Characteristics of Kokanee in Kootenay Lake in Relation to Life History and Environment. It i s evident from the foregoing that environmental fac-tors could have been effective in producing most of the observed variation of morphometric features of kokanee in Kootenay lake. Kokanee spawn in September and October in tributary streams and development within the egg is subject to the environmental influence of stream conditions during the winter-months. Taning (1952) findsthat in brown trout the f i n a l number.of vertebrae i s fixed by the time the embryonic eye becomes pigmen-ted. This i s commonly known as the "eyed-egg" stage; A similar situation might be expected in kokanee and therefore stream conditions, particularly temperature, have probably been deci-sive in producing the observed heterogeneity of vertebral counts. Mottley (1934) found that the number of lateral line scales was fixed sometime after the "eyed egg" stage in Kamloops 59 trout. Neave (1936) found rudimentary lateral line scale papilae i n brown trout several weeks after hatching. As kokanee migrate to the lake almost immediately after hatching i t i s possible that both stream and lake environment may influence the number of scale rows • The number of g i l l rakers, i f subject to phenotypic variation, would probably be influenced chiefly by the lake environment of kokanee. Svardsen (1949) finds that the number of g i l l rakers of Coregonids i s not determined u n t i l after the f i r s t summer. Svardson also finds that the number of g i l l rakers in coregonids is under gentetic control and i s remarkably stable under various environ-mental conditions. The response of growth rate in fishes to environmental condi-tions, particularly food, is well known. Active ingestion of food by kokanee is confined to the portion of their lifecycle spent in the lake and therefore lake environment is probably decisive in modifying growth rates. Stream conditions could modify growth rates by affecting the length of the developmental period before hatching, and in turn, the time at which active feeding begins. The length of the incubation period in fishes i s directly related to temperature, and i t i s possible that temperature is not uniform in a l l areas in which kokanee spawn. No temperature data are available but i t is possible, for example, that the high rate of f i r s t year growth in West Arm kokanee i s at least partially caused by higher incubation temperatures and a resultant advance-ment of the feeding period. Thus growth rates are dependent on both stream and lake environment. 60 Martin (1949) has shown that growth rates can af f e c t body-form. Svardson (1950) suggests that growth rates of fishes are r e l a t e d to age at sexual maturity. A r i c h food supply may cause accelerated physiological aging and an early sexual maturity. On the other hand, very poor food supply also may produce forms which show early sexual maturity i n spite of retarded physiological aging. Selection under these extreme conditions may favor animals possessing hormonal mechanisms permitting early sexual maturity. Thus there may be selective pressure with the effect that sexual maturity i s quickened i f n u t r i t i o n a l standards become too poor. While there i s no experimental evidence f o r the operation of these mechanisms i n Kootenay lake kokanee, i t i s i n t e r e s t i n g to note that the two populations having the smallest and largest mean size mature a year e a r l i e r than the population having an i n t e r -mediate mean s i z e . The influence of environment on the development of secondary sex characters i s unknown, but appears to be associated with size at maturity i n kokanee of Kootenay lake. Packer (1938) found a difference i n degree of scale resorption i n two populations of none-migratory sockeye i n Cultus lake. Although Packer (1940) ascribed a genetic basis f o r the difference i n scale resorption i t was i n t h i s case also associated with size at maturity. From the foregoing i t might be expected that differences i n growth rates between the three areas i n Kootenay lake would contribute l a r g e l y to the observed d i s s i m i l a r i t y of other.charac-t e r i s t i c s . It has been shown that growth rates of kokanee a f t e r the f i r s t year are s i m i l a r i n a l l areas, and that size at maturity 61 i s dependent (aside from age) on growth in the f i r s t year. Assuming a gradual and random dispersal of kokanee after entering the lake from the spawning areas, they would be most subject to the modifying influence of local conditions during their f i r s t year. With increasing size greater mobility is achie-ved and temporary penetration of other areas would tend to reduce, in older age classes, differences in mean growth rates between fi s h originating in different areas. It i s also possible that the quality of food i s such that i t i s not equally available i n a l l areas to underyearling f i s h , but i s equally available>in a l l areas to older f i s h . The North End of Kootenay lake i s dominated by the large inflow of Lardeau river. During the summer this river i s colder than surface water of the lake and, on entering, goes.beneath the surface to a depth of equal density (temperature) and flows south-ward along the westward lake margin for a considerable distance. In the process of mixing with Kootenay lake water, heat is added to the deeper lake strata. The South End i s dominated by the inflow of Kootenay river which, being warmer than Kootenay lake, spreads out over the surface. The mineral content of Kootenay river is much higher than that of Lardeau river (Larkin, personal communication) and has a f e r t i l i z i n g effect on the South End which i s reflected i n the higher production of bottom organisms i n this area (Larkin, 1950). The West Arm i s comparatively narrow and, as i t drains the entire inflow to the lake, i t has a perceptible current throughout 62 i t s length. Because of a shallow s i l l at the entrance to the arm, water i s drawn from surface strata of the main body of the lake. During the growing season, these surface strata are at a compara-tively high temperature and should supply relatively warm water to a l l levels of the West Arm. Plankton produced in the main body of the lake w i l l be drawn off by the flow of surface water into the West Arm. These features probably have a profound effect on the productivity of this area and could be expected to result in an abundance of food for kokanee. A detailed limnological comparison of the three areas i s not possible with the data available. Physical features indicate that environmental conditions di f f e r greatly between areas but are f a i r l y uniform within areas. Factors Tending Toward Isolation of Stocks The pronounced tendency of Pacific salmon to return to and spawn in the stream of their birth i s well known. Walford (1939) after an extensive survey of literature, concluded that on the basis of scale marks, stocking experiments, c y c l i c a l abundance, marking and tagging experiments i t has emerged that each river and even each tributary may have i t s own local self-perpetuating stock of Pacific salmon. McGregor (1924) counted extensive series of ova, caeca and g i l l rakers in King salmon (Oncorhvnchus tschawytscha) from Klamath and Sacramento rivers in California. He was able to demonstrate distinct differences in these characters between the two river rac&s. Hubbs (1934) recognized that the homing habits and consequent stock isolation in Pacific salmon i s correlated 63 with marked structural differentiation. He also pointed out that r a c i a l segregation tends to be most intense during reproduction in many animals. The existence of racial stocks and homing habits i s extensively used as a basis for conservation measures for sockeye (Royal, 1953). In Kootenay lake, the homogeneity of kokanee on their spawning grounds i s considered to be strong evidence of a pronounced tenden-cy to "home" to the parent stream. The degree of straying between streams i n the same area i s unknown and may be extensive but the amount between areas has been shown to be less than three percent. The homing behaviour of kokanee is considered to be of prime / importance in the isolation of populations i n Kootenay lake. Kootenay lake i s a large body of water and the three arbitra-r i l y chosen sub-areas are also large. The lake l i f e of South End and West Arm kokanee is less than two and one-half years; that of North End kokanee less than three and one-half years. In the f i r s t year of l i f e particularly, the f i s h are very small in rela-tion to the available range. They are plankton feeders and thus tend to spread out over a wide area of lake surface ra%her than over an extensive section of shore-line. Their movements i n pursuit of food can be assumed to be nearly random and spacial conditions alone would result in a tendency-for kokanee to inhabit areas adjacent to their home stream, especially during -early l i f e . Their short lives may preclude their random distribution throu-ghout the entire lake. No sharp geographic barriers exist within Kootenay lake -although the constricted entrance to the West Arm may be a partial 64 b a r r i e r , but distance alone may prevent, random interbreeding. Thompson (1931) showed a c o r r e l a t i o n between water distance and morphological divergence within a species of Darter i n I l l i n o i s r i v e r s . He found that the number of spines and f i n r a y s varied roughly as a function of distance by water. The types did not form a series i n a gradient and therefore did not from a c l i n e . Schmidt (1918) found that v a r i a t i o n i n Zoarces viviparus was a function of distance i n the open sea but a c l i n e i n f i o r d s . Although there i s no evidence f o r a geocline i n Kootenay lake kokanee i n the present data, i t i s possible that the samples are of the extremes of v a r i a t i o n , and that i f intermediate areas were sampled, intermediate types might be found. This i s not consi-dered very probable because most intermediate areas are not known to have any extensive spawning runs. Svardson (1950) believes that, i n sympatric species of Coregonids, schooling behaviour i s important i n mainteining the i d e n t i t y of each species and t h i s may also aid i n maintaining separation of diverse groups of kokanee i n Kootenay lake. Although much of the observed morphological heterogeneity of kokanee i n Kootenay lake i s undoubtedly phenotypic the stocks are probably s u f f i c i e n t l y i s o l a t e d to allow some genotypic mor-phological d i f f e r e n t i a t i o n . However the degree of interbreeding between populations probably precludes the p o s s i b i l i t y of such d i f f e r e n t i a t i o n continuing toward sympatric speciation. An analysis of morphometric v a r i a t i o n cannot separate geno-ty p i c from phenotypic d i f f e r e n t i a t i o n . . Experimental work i s es s e n t i a l to separate effects of environment from effects of heredity. The present paper i s severely l i m i t e d i n t h i s respect 65 but i t has served to delineate to some extent the sub-groups involved and may serve as a basis f o r experimental work. SUMMARY AND-CONCLUSIONS 1. Kokanee (Oncorhynchus nerka kennerlvi) were collected from spawning runs to seven streams t r i b u t a r y to Kootenay lake, B r i t i s h Columbia. 2. Preliminary examination of morphometric features of these kokanee indicated that i n Kootenay lake they were separable into three groups: (a) South End f i s h , averaging approximately 18.5 cm. and 55 gm. and spawning at 3 years. (b) North End f i s h , averaging approximately 21.5 cm. and 105 gm. and spawning at 4 years. (c) West Arm f i s h averaging approximately 24.5 cm. and 155 gm. and spawning at 3 years. 3. Samples from streams within each of these areas were examined i n more det a i l . . Within each area kokanee were similar i n growth rate, vertebral count, development of secondary sexual character-i s t i c s and s i z e . 4. Samples from a representative stream of each area were found to d i f f e r i n predominant age at maturity, s i z e , growth rate i n the f i r s t year, vertebral count, scale row count, g i l l raker count, head length of males, growth form of caudal peduncle, eye-66 diameter, development of secondary sexual c h a r a c t e r i s t i c s , ova per u n i t t o t a l weight and commencement of spawning p e r i o d . 5. Examination o f age, s i z e , v e r t e b r a l count, scale p a t t e r n and s c a l e r e s o r p t i o n i n samples from t h r e e areas over a per i o d of three years i n d i c a t e d t h a t w h i l e s i g n i f i c a n t v a r i a t i o n occurred between years i n some cases, the v a r i a t i o n d i d not obscure the s i g n i f i c a n c e of v a r i a t i o n between areas i n any one year. D i f f e r e n c e s between areas were e s s e n t i a l l y the same each year. 6 . By examination of scale p a t t e r n , s i z e , s c a l e r e s o r p t i o n , v e r t e b r a l and scale rows, i t was p o s s i b l e t o i d e n t i f y kokanee spawning i n an area othe r than t h e i r n a t i v e or home area. Less than three percent of specimens examined had.strayed from t h e i r home area. 7. The e f f e c t s of environment on morphometric f e a t u r e s i s discussed and i t i s considered probable t h a t most o f the heterogeneity of kokanee i n Kootenay.lake i s due to phenotypic v a r i a t i o n . 8. I t i s concluded t h a t the strong tendency o f kokanee. to home t o t h e i r parent stream i s t h e p r i n c i p a l reason f o r the" existence of at l e a s t three e s s e n t i a l l y d i s c r e t e kokanee populations i n Kootenay l a k e . 9. Although some genotypic morphological d i f f e r e n t i a t i o n may have occurred, the i n t e r b r e e d i n g between populations i s probably s u f f i c i e n t t o preclude sympatric s p e c i a t i o n . 10. Further experimental work i s suggested to separate geno-t y p i c from phenotypic v a r i a t i o n . 67 LITERATURE CITED Baievsky, Boris. 1926. Fisheries of Siberia. Rep^t. U. S. Comm. Fish. 1926. Append. II : 37-64-. Bean, T. H. 1883. Notes on fishes collected by Capt. Chas. Bendire, U. S. A. i n Washington Territory and Oregon, May to October, 1881. Proc. U. S. Nat. Mus. 6 : $9-93. 1891. Kennerly's Salmon. Forest and Stream. XL : 498-499. Bendire, C. E. 1878. Red trout, or redfish of Oregon and Idaho. Forest and Stream. 10 : 156-157. 1$$2. Notes on Salmonidae of the Upper Columbia. Proc. U. S. Nat. Mus. ^ : 81-87. Bjerkan, Paul 1929. On the validity of "Race Characters" of food fishes. Conseil per. inter. Expl. Mer. Rapp. et Proc.-Verb, des Reunions. 5J_ : 73-7$. Carl, G. Clifford & W. A. Clemens. 194$ & 1953. The freshwater fishes of British Columbia. Handbook # 5, B. C. Prov. Museum. Clemens, W. A. and G. V. Wilby. Fishes of the Pacific coast of Canada. Bull. Fish. Res. Bd. Can. # 6$ : 1-36$. Cottingham, M. E. 1947. History of the West Kootenay d i s t r i c t in B. C. M. A. Thesis. Univ. B. C. D'Ancona. 1927. Sul valore sistematice delle unita morphologiche di numero variabile ed in particular sulla sistematical della Alose mediterranee. Mon. Zool. I t a l . 3_$ ($): 1$7-191. 6 8 Davidson, F. A. 1935. The development of the secondary sexual ' characters in the pink salmon (Qncorhvnchus gorbuscha). 1 J. Morph. 52 169-183. Davidson, F. A. and S. J. Hutchinson. 1938. The geographic distribution and environmental limitations of thewPacific salmon. Bull. U. S. Bur. Fish, 43 (26) : 667-692. Diver, C. 1940. The problem of closely related species l i v i n g i n the same area. The New Systematics, ed. J. Hurley. Oxford. Dymond, J. R. 1932. The trout and other game fishes of B. C. Dept. Fish. Ottawa. 55 pp. 1936. Some fresh-water fishes of British Columbia. Rept. B. C. Comm. Fish. L 60 - L 73. Evermann, B. ¥. 1897. A report on the salmon investigations in the headwaters of the Columbia river i n the State of Idaho in 1895. Bull. U.'S. Fish. Comm. 16 : 151-202. Forester, R. E. 1947. Experiment to develop sea-run from land-locked sockeye salmon (Qncorhvnchus nerka kennerlvi). J. Fish. Res. Bd, Can. 2 (2) : 88-91. & W. E. %cker, 1953. The coho salmon of Cultus lake and Sweltzer creek. J. Fish. Res. Bd. Can. 10 : 293-319. Girard, C. F* 1857. Notice upon the species of the genus Salmo of authors, observed chiefly in Oregon and California. Proc. Acad. Nat. Sci.Philad. ,8 : 217-220. Green, A. H. 1893. Notes on the occurrence of new and rare f i s h in British Columbia. Bull. Nat. Hist. Soc. B. C. pp. 9-10. 69 •7 Gunther, A. 1866. Description of f i s h from Vancouver (In Lord, J. K. The naturalist of Vancouver Island and British Columbia. Vol. 2 London, 2 vols.) Halkett, A. 1913. Check l i s t of the fishes of the Dominion of Canada and Newfoundland. Dept. of Marine and Fisheries, Canada, pp. 1-138, 14 p i . Heincke, F. 1898. Naturgeschichte des Herings. I. Die Lokal-formen und die Wanderungen des Herings in den europ&ischen Meeren. Abh. deut. Seefischerei Ver. 2 (1) : 1-128. Hubbs, C. L. 1926. The structural consequences of modifications of the development rate in fishes. Amer. Nat. 60 : 57-81. 1934* Racial and individual variation -in animals, especially fishes. Amer. Nat. 68 : 115-128. Jordan, D. S. 1894. Description of a new species of cyprinoid f i s h from the headwaters of Fraser river in British Columbia. Proc. U. S. Nat. Mus. 16 : 313-3L4. & C. H. Gilbert. 1881. List of the fishes of the Pacific coast of the U. S., with a table showing the distribution of the species. Proc. U. S. Nat. Mus. £ : 452-453. & B. ¥. Everasinn. 1396. The fishes of North and Middle • America. Bull. U. S. Nat. Mus. 4_Z (1) : 1-1240. & 1922. American food and game fishes. 572 pp., 110 p i . Doubleday & Page. N. York. ; & E. H. McGregor. 1925. Record of fishes obtained by David Star Jordan i n Japan. Mem. Carnegie Mus. 10 : 122-146. 70 Krokhin, E. M. & F. V. Krojius. 1936. The lake form of Qncorhvn-chus nerka from Lake Kromotsk, Kamchatka. C. R. Acad. Sci. Moscow N. S. ^  : 89-92. . Larkin, P.. A. 1950. Summary of investigations conducted by the Fisheries Research Group of the B. C. Game Dept. Unpub-lished manuscript on f i l e Univ. B. C. Lissner, H. 1934. On races of herrings. J. Conseil 2 (3):346-364. Martin, W. R. 1949. The mechanics of environmental control of body form i n fishes. Pub. Ont. Fish. Res. Lab. No. 70: 1-91. McGregor, E. A. 1924. A possible separation of the river races of King salmon in ocean-caught f i s h by means of anatomical characters. Galif. Fish & Game £. : 139-150. Milne, D. J. 1948. The growth, morphology and relationship of the species of Pacific salmon and the steelhead trout. Doctoral .thesis, McGill Univ. Mottley, C. McC, 1934. The. effect of temperature during develop-ment on the number of, scales in Kamloops trout. Salmo  kamloops, Jordan., Contr. Can. Biol. 8 : 253-263. "1941. The use of scales of rainbow trout (Salmo gairdnerii) to., make dirct comparisons of growth. Trans. Amer. Fish. Soc. 71 : 74-7-9. • . Neave, F. 1936. The development of the scales of. Salmo. Trans. Roy. Soc. Can. 3.0 (5): 55r-72. 1943. Scale pattern and scale counting methods in relation to certain trout and, other, Salmonids. Trans. Roy. Soc. Can. 22 (5) : 79-91. Oshima, M. 1935. Life history and distribution of the fresh-water 71 salmons found i n the waters of Japan. Proc. 5th Pac. S c i . Congress. 1933. (5) : 3751-3773. Richardson, John. 1$36. Fauna Boreali-Americana. 4 v o l s . London & Norwich, 1831-37. Ricker, W. E. 1938. Residual and kokanee salmon i n Cultus lake B r i t i s h Columbia. J . B i o l . Bd. Can. 4_ (3) : 192-218. 1940. On the o r i g i n of kokanee, a fresh-water type of sockeye salmon. Trans. Roy. Soc. Can. 3J_ (5) : 121-135. Rounsefell, G. A. & E. H. Dahlgren. 1932. Fluctuations i n the supply of herring. Clupea p a l l a s i i . i n Prince William Sound Alaska. B u l l . U. S. Bur. F i s h . 4J No. 9 : 263-291. Royal, L. A. 1953. The e f f e c t s of regulatory s e l e c t i v i t y on the productivity of Fraser r i v e r sockeye. Can. F i s h . Cult. No. 14 : 1-12. Runnstr^m, Sven. 1933. Ober die RassenverhaTtnisse bei dem norwegische Frtthjahrshering mit besonderer Bertlcksichtigung der Konstang der Rassenmerkmale. J. Conseil 8_ (2): 235-249. Schmidt, J . 1918. Racial studies i n f i s h e s . I. S t a t i s t i c a l investigations with Zoarces viviparus L. J . Genet. 7 (2) : 107-118. Schofield, S. J . 1946. The o r i g i n of Kootenay lake. Proc. Roy. Soc. Can. l± : 93-98. Schultz, L. P. 1929. Check l i s t of fresh-water f i s h e s of Oregon and Washington. Univ. Wash. Pub. F i s h . 2 (4). : 43-50. 1935(a). Species of salmon and trout i n the Northwestern United States. Proc. 5th Pac. S c i . Cong. 1933 (5): 3777-3782 . & Students. 1935(b). The breeding a c t i v i t i e s of the l i t t l e r e d f i s h , a land-locked form of the sockeye salmon, Oncorhvna': 72 chus nerka. Jour. Pan. Pac. Res. Inst. 10 (1) : 67-77. Schultz, L, P. 1936* Keys to the fishes of Washington, Oregon and closely adjoining regions. Univ. Wash. Pub. Biol. 2 : 103-228. Simpson, G. G. & A. Roe. 1939. Quantitative Zoology. 414 pp. McGraw-Hill. New York and London. Snedecor, G. W. 1946. S t a t i s t i c a l methods. 485 pp. 4th Ed. Iowa State College Press. Ames, Iowa. Suckley, G. 1862. Notices of certain new species .of North American Salmonidae, chifly in the collection of the North-west Boundary Commission. Amer. Ann. Lyc. Nat. His. U. Y. 2 : 306-313. SvSrdson, G. 1949. The Coregonid problem. II. Morphology of two Coregonid species in different environments. Inst. Freshwater Res., Drottningholm. Rept. No. 31 *• 151-162. • 1950. The Coregonid problem III. Whitefish from the Baltic successfully introduced into fresh waters in the north of Sweden. Inst. Freshwater Res., Drottningholm. Rept. No. 32 : 79-125. Tlning, A. V. 1944. Experiments on meristic and other charac-ters in fishes. Meddeleser f r a Kommissioner for Danmarks Fiskeriog Havundersogelser. Series Fiskeri. Bind 11. m 1952. Experimental study of meristic characters of fishes. Biol. Rev. g2 : 169-193. Tchernavin, V. 1939. The origin of Salmon. Salmon and Trout Mag. 25. s 1-21. 73 Tester, A. L. 1937. Populations of herring (Clupea p a l l a s i i ) in coastal waters of British Columbia. J. Biol. Bd. Can. I (2) : 103-144. 1938. Variation i n the mean vertebral count of herring (Clupea pa l l a s i i ) with water temperature. J. Conseil I I (1) : 71-75. Thompson, D. H. 1931. Variation i n fishes as a function of distance. Trans. I l l i n o i s State Acad. Sci. 23_ : 276-281. Walbaum, J. J. 1792. Petri Artedi Sueci Genera^i Piscium. Grypeswaldiae. 723 pp. 3 p i . Walford, L, A. 1939. The migration and conservation of salmon. Pub. Amer. Assoc. Adv* Sci. No. 3, 106 pp. Ward, H. B. 1932. . The origin of the landlocked habit in salmon. Proc. U. S. Nat. Acad. Sci. 13 (9) : 569-530. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0106827/manifest

Comment

Related Items