UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Seasonal isolation and adaptation among chum salmon, Oncorhynchus keta (Walbaum), populations Tallman, Ross Franklin 1988

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

Item Metadata

Download

Media
831-UBC_1989_A1 T34.pdf [ 13.66MB ]
Metadata
JSON: 831-1.0098366.json
JSON-LD: 831-1.0098366-ld.json
RDF/XML (Pretty): 831-1.0098366-rdf.xml
RDF/JSON: 831-1.0098366-rdf.json
Turtle: 831-1.0098366-turtle.txt
N-Triples: 831-1.0098366-rdf-ntriples.txt
Original Record: 831-1.0098366-source.json
Full Text
831-1.0098366-fulltext.txt
Citation
831-1.0098366.ris

Full Text

SEASONAL ISOLATION AND ADAPTATION AMONG CHUM SALMON, Oncorhynchus keta (Walbaum), POPULATIONS by ROSS FRANKLIN TALLMAN B. Sc., The U n i v e r s i t y o f Manitoba, 1977 M. Sc., The U n i v e r s i t y o f Manitoba, 1980 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS OF THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Department o f Zoology, I n s t i t u t e o f Animal Resource Ecology) We accept t h i s t h e s i s as conforming t o the re q u i r e d standard THE UNIVERSITY OF BRITISH COLUMBIA October 1988 © R o s s F r a n k l i n Tallman, 1988 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Source The University of British Columbia Vancouver, Canada DE-6 (2/88) i i ABSTRACT To test the hypothesis that temporal i s o l a t i o n due to differences in i season of breeding could result i n genetic divergence among chum salmon populations, I compared the reproductive environments, phenotypes and genotypic performances of early and l a t e season breeding stocks of chum salmon, Oncorhynchus keta. These were autumn (AB) and winter (WB) spawners from one Vancouver Island stream (Bush Creek) and another winter spawning stock (W) from a nearby stream (Walker Creek). A l l stocks had s i m i l a r time of downstream migration of the f r y . No differences were found among the W and AB stocks in age at maturity, length composition of spawners, egg s i z e , vertebral count of adults and fry and time of fry migration. The WB stock d i f f e r e d from the other stocks in egg s i z e and vertebral number of adults and f r y . In 1981, WB spawners were larger than AB spawners. In 1982, WB spawners were younger than AB spawners. Analysis of ten external morphological features of the fry revealed that there was considerable overlap in body form of the stocks. To test the hypothesis that phenotypic s i m i l a r i t i e s observed among seasonal ecotypes i n the wild were due to genetic differences, sample populations were reared under c o n t r o l l e d conditions in the laboratory. When reared at 6°C, under the autumn - winter - spring progression and the winter -spring progression, the early spawning population i n Bush Creek took s i g n i f i c a n t l y more time to hatch and emerge than the s p a t i a l l y and temporally i i i i s o l a t e d population of Walker Creek, and the combined average of the l a t e spawning populations. At 10°C the populations had s i m i l a r incubation rates. Temperature regime, population and temperature regime by population i n t e r a c t i o n a l l had s i g n i f i c a n t e f f e c t s on vertebral number. AB progrency were morphologically d i s t i n c t from WB and W. These results indicate that the early spawning population d i f f e r s g e n e t i c a l l y from the l a t e spawning populations. To t e s t the hypothesis that s e l e c t i o n on the speed of development from f e r t i l i z a t i o n to hatch and emergence would reduce the additive genetic v a r i a t i o n in these t r a i t s , i n d i v i d u a l families were reared from AB, WB and W. H e r i t a b i l i t i e s were found to be between 0.27 and 0.54. The high h e r i t a b i l i t i e s suggest that some counte r v a i l i n g force i s opposing s e l e c t i o n on incubation rate. Analysis of d i v e r s i t y using biochemical genetics techniques suggest that genetic migration goes from the WB stock to the AB stock. Genetic distance measures and a Wagner-Tree analysis indicate that biochemically AB and WB are more c l o s e l y related than W. It i s postulated that the g e n e t i c a l l y determined program for incubation which results in synchronous downstream migration has s u r v i v a l value. Survival could be enhanced by predator s a t i a t i o n or synchrony with food resources in the estuary. i v TABLE OF CONTENTS ABSTRACT i i LIST OF TABLES vi LIST OF FIGURES x v i i LIST OF APPENDICES x x i i ACKNOWLEDGEMENTS xxvi INTRODUCTION 1 Selection and the Ecology of Chum Salmon 15 The D i s t r i b u t i o n and Abundance of Seasonal Races and Speculations Regarding Their Evolutionary Origin 30 Review 30 Phenotypic D i f f e r e n t a t i o n among Seasonal Ecotypes of Chum Salmon 41 Introduction 41 Materials and Methods 46 Results 6 2 Discussion 94 V Innate versus Environmental Control of Phenotypes among Seasonal Ecotypes 116 Introduction 116 M a t e r i a l s and Methods 119 Results 130 Discussion 191 Evidence for S e l e c t i o n on Incubation Rate 204 Int r o d u c t i o n 204 M a t e r i a l s and Methods 205 Results 206 Discussion . 206 An E l e c t r o p h o r e t i c A n a l y s i s of Genetic V a r i a t i o n Among Seasonally Separated Populations 210 Int r o d u c t i o n 211 M a t e r i a l s and Methods 214 Results 224 Discussion • 238 Summary and Synthesis of E m p i r i c a l F i n d i n g s 248 LITERATURE CITED 271 APPENDICES 294 v i LIST OF TABLES Table 1. Mean dates of s t a r t , peak and end of Bush and Walker Creek chum spawning runs from 1969 to 1981 49 Table 2. Comparisons between populations along s p a t i a l and temporal axes AB=Autumn Bush Population; WB=Winter Bush Population; W=Walker Population 49 Table 3. The yearly median and mode ( i n Julian day) of the temporal d i s t r i b u t i o n of adults on the Bush and Walker Creek spawning grounds during 1981 and 1982 65 Table 4. Survival time of adults in freshwater at percentiles of the d i s t r i b u t i o n of counts of l i v e and dead adults during 1981 and 1982 65 Table 5. Vertebral counts of 1982 spawners: means (X), standard deviations (SD), and sample s i z e s s t r a t i f i e d by population, sex and age 67 Table 6. Single degress of freedom comparisons of vertebral count differences among populations using the Tukey-Kramer Method. (MSD = Minimum S i g n i f i c a n t Difference) (* s i g n i f i c a n t differences P = .05) 68 vii Table 7. Age at return f o r 1981 and 1982 spawners. Means, standard d e v i a t i o n s and sample s i z e s of populations s t r a t i f i e d by sex 68 Table 8. Orbit-Hypural P l a t e Lengths of 1981 and 1982 spawners: means (50, standard d e v i a t i o n s (SD), and sample s i z e s s t r a t i f i e d by pop u l a t i o n , sex and age 70 Table 9. Single degree of freedom comparisons among spawning populations using the Tukey-Kramer Method. Minimum S i g n i f i c a n t D i f f e r e n c e - MSD. (* s i g n i f i c a n t d i f f e r e n c e , P = .05) 71 Table 10. Capture e f f i c i e n c y of i n c l i n e d plane traps below the spawning areas during 1983 73 Table 11. The ye a r l y median and mode ( i n J u l i a n day) of the temporal d i s t r i b u t i o n of f r y emigrating from Bush and Walker Creeks during 1982 and 1983 76 Table 12. V e r t e b r a l counts f o r f r y m i g r a t i n g downstream during 1982. Means, standard d e v i a t i o n s and sample s i z e s of populations s t r a t i f i e d by time. Fry samples were made at i n t e r v a l s corresponding to the approximate 17th, 33rd, 50th, 67th, and 83rd p e r c e n t i l e s of the cumulative frequency d i s t r i b u t i o n of i n d i v i d u a l s with time 76 v i i i Table 13. V e r t e b r a l counts f o r f r y migrating downstream during 1983. Means, standard d e v i a t i o n s and sample s i z e s of populations s t r a t i f i e d by time. Fry samples were made at i n t e r v a l s corresponding to the approximate 25th, 50th and 75th p e r c e n t i l e s of the cumulative frequency d i s t r i b u t i o n of i n d i v i d u a l s with time 78 Table 14. V a r i a b l e s entered i n t o the 1982 and 1983 d i s c r i m i n a n t fun c t i o n s ranked by importance with c l a s s i f i c a t i o n f u n c t i o n s for each 78 Table 15. C l a s s i f i c a t i o n matrix f o r 1982 f r y samples using the d i s c r i m i n a n t f u n c t i o n 79 Table 16. C l a s s i f i c a t i o n matrix f o r 1983 f r y samples using the d i s c r i m i n a n t f u n c t i o n 79 Table 17. Approximate transformation F s t a t i s t i c comparing group c e n t r o i d s f o r 1982 81 Table 18. Mahalanobis distance between population c e n t r o i d s f o r 1982 81 Table 19. Approximate transformation F s t a t i s t i c comparing group c e n t r o i d s among 1983 progeny 81 i x Table 20. Mahalanobis distance between population c e n t r o i d s for 1983 81 Table 21. Means and standard d e v i a t i o n s of morphological c h a r a c t e r i s t i c s of 1982 f r y samples. Lengths are i n 0.1mm. Weight i s i n 0.1gm. (S.M. Mean standardized to a common length among the samples) 82 Table 22. Means and standard d e v i a t i o n s of morphological c h a r a c t e r i s t i c s of 1983 f r y samples. (S.M. Mean standardized to a common length among the samples) 83 Table 23. C l a s s i f i c a t i o n matrix f o r 1983 f r y samples using the di s c r i m i n a n t f u n c t i o n from the 1982 data ^4 Table 24. Comparisons of accumulated degree-days (raw and adjusted data) and incubation times ( a c t u a l and predicted) f o r AB, WB, and W stocks i n 1981-82 and 1982-83. ( c a l c u l a t e d from median day of the adult and f r y runs) 87 Table 25. S i n g l e degree of freedom comparisons of egg weight d i f f e r e n c e s among populations adjusted f or c o v a r i a t e using the GT2 Method. (MSD = Minimum S i g n i f i c a n t D i f f e r e n c e ) (* s i g n i f i c a n t d i f f e r e n c e P = .05) 92 X Table 26. The temporal and s p a t i a l pattern of return of marked f i s h to Bush and Walker Creeks during 1984 and 1985 93 Table 27. Return rate of marked f i s h 93 Table 28. Mean water temperatures i n each c e l l for 1982-83 and 1983-84 experiments. (Standard Deviation i n Parenthesis) 124 Table 29. Orbit-hypural plate length of females and egg sizes used in 1982-83 and 1983-84 experiments 139 Table 30. Mean hatch t i m e for population by temperature regime treatments i n 1983-84 experiment. (Standard deviations are in parentheses) 140 Table 31. Results of comparisons and contrasts of mean time to 50 % hatch among the test populations reared within each temperature regime. Mean time to hatch used i n comparisons was adjusted for temperature v a r i a t i o n . A plus (+) indicates that the population to the l e f t of the minus sign took more time to reach 50 % hatch than the population or average of two populations to the r i g h t . (An asterisk (*) indicates that P < 0.05. Two a s t e r i s k s (**) indic a t e that xi Table 32. Mean emergence time for population by temperature regime treatments in 1982-83 experiment. (Standard deviations are in parentheses) 151 Table 33. Mean emergence time for population by temperature regime treatments in 1983-84 experiment. (Standard deviations are in parentheses) 151 Table 34. Results of comparisons and contrasts of mean time to 50 % emergence among the test populations reared within each temperature regime. Mean time to emergence used i n comparisons was adjusted for temperature v a r i a t i o n . A plus (+) indicates that the population to the l e f t of the minus sign took more time to reach 50 % emergence than the population or average of two populations to the r i g h t . (An asterisk (*) indicates that P < 0.05. Two as t e r i s k s (**) indicate that P < 0.01) 152 Table 35. Summary of the number of s i g n i f i c a n t comparisons using the SSSTP (Gabriel 1964) pooled over a l l temperature regimes 154 Table 36. Survival of chum salmon for each population, temperature regime, tank and year during d i f f e r e n t segments of embryonic development from f e r t i l i z a t i o n to emergence. (1) 156 x i i Table 37. Mean vertebral counts for population by temperature regime treatments i n 1983-84 experiment. (Standard deviations are in parentheses, N = 50) 174 Table 38. Approximate transformation F s t a t i s t i c to compare population centroids with a l l temperatures pooled 176 Table 39. Mahalanobis distance between population centroids with a l l temperatures pooled Table 40. C l a s s i f i c a t i o n matrix for 1983 fry samples using the discriminant function a l l temperatures pooled 178 Table 41. C l a s s i f i c a t i o n matrix for laboratory fry samples using the discriminant function at 6 Celsius 178 Table 42. C l a s s i f i c a t i o n matrix for laboratory fry samples using the discriminant function at 10 Celsius 179 Table 43. C l a s s i f i c a t i o n matrix for laboratory fry samples using the discriminant function at Early Regime 179 Table 44. C l a s s i f i c a t i o n matrix for laboratory fry samples using the discriminant function at Late Regime 180 xi i i Table 45. Approximate transformation F s t a t i s t i c to compare population centroids at 6 Celsius 180 Table 46. Mahalanobis distance between population centroids at 6 Celsius 180 Table 47. Approximate transformation F s t a t i s t i c to compare population centroids at 10 Celsius 180 Table 48. Mahalanobis distance between population centroids at 10 Celsius 181 Table 49. Approximate transformation F s t a t i s t i c to compare population centroids at Early Regime 181 Table 50. Mahalanobis distance between population centroids at Early Regime 181 Table 51. Approximate transformation F s t a t i s t i c to compare population centroids at Late Regime 181 Table 52. Mahalanobis distance between population centroids at Late Regime 181 xiv Table 53. Means and standard deviations of morphological c h a r a c t e r i s t i c s of emergent fry reared at 6 Celsius during 1983-84. (S.M. Mean standardized to a common length among the samples) 182 Table 54. Means and standard deviations of morphological c h a r a c t e r i s t i c s of emergent f r y reared at 10 Celsius during 1983-84 (S.M. Mean standardized to a common length among the samples) 185 Table 55. Means and standard deviations of morphological c h a r a c t e r i s t i c s of emergent fry reared under the Early Regime during 1983-84 (S.M. Mean standardized to a common length among the samples) 187 Table 56. Means and standard deviations of morphological c h a r a c t e r i s t i c s of emergent fry reared under Late Regime during 1983-84 (S.M. Mean standardized to a common length among the samples) 189 Table 57. H e r i t a b i l i t i e s and lower confidence l i m i t (LCL) (P = 0.05) for time to 50 % hatch at 8°C 207 Table 58. H e r i t a b i l i t i e s and lower confidence l i m i t (LCL) (P = 0.05) for time to 50 % emergence at 8°C 207 XV Table 59. Enzymes within tissues and buffer systems used in the electrophoretic analysis 215 Table 60. Sample s i z e s , and allozyme frequencies of polymorphic l o c i used in the analysis 225 Table 61. Heterozy gosity and test of comformance to to Hardy-Weinberg equilibrium of polymorphic l o c i within each population 231 Table 62. Summary of F - s t a t i s t i c s at a l l l o c i 233 Table 63. Contingency chi-square at a l l l o c i . Chi-square and pr o b a b i l i t y values for the hypothesis that samples were drawn from the same population 235 Table 64. Estimates of e f f e c t i v e population s i z e for each population using the algebraic mean, harmonic mean and modified harmonic mean methods described i n the text 236 Table 65. P r o b a b i l i t y that a randomly chosen a l l e l e w i l l be from a migrant i n d i v i d u a l assuming random mating, no se l e c t i o n , or mutation 237 Table 66. Genetic distance and genetic i d e n t i t y over a l l l o c i : BELOW diagonal: Rogers "D" (Wright 1978). ABOVE diagonal: Nei (1972) Genetic Identity 240 xvi Table 67. Genetic distance over a l l l o c i : BELOW diagonal: Prevosti Distance (Wright 1978). ABOVE diagonal: Nei (1972) Genetic distance 2 4 0 Table 68. Genetic distance over a l l l o c i : BELOW diagonal: C a v a l l i - S f o r z a and Edwards (1967) Chord distance. ABOVE diagonal: Nei (1978) Unbiased Minimum Distance 240 Table 69. Genetic distance over a l l l o c i : BELOW diagonal: C a v a l l i - S f o r z a and Edwards (1967) arc distance. ABOVE diagonal: Nei (1972) Minimum distance 241 Table 70. Genetic distance over a l l l o c i : BELOW diagonal: Edwards (1971, 1974) "E" distance 241 Table 71. Matrix of single-locus genetic s i m i l a r i t y or distance c o e f f i c i e n t s 242 x v i i LIST OF FIGURES Figure 1. S e r i a l and branching e v o l u t i o n 11 Figure 2. World d i s t r i b u t i o n of chum salmon 16 Figure 3. A model of chum salmon l i f e h i s t o r y showing the i n t e r a c t i o n between s e l e c t i o n (S), genotype (G) and environment (E) 27 Figure 4. Bush and Walker creeks: the study area 47 Figure 5. Distance between spawning groups of chum salmon i n Bush and Walker Creeks 50 Figure 6. Timing of spawning i n Walker Creek and the upper and lower s e c t i o n of the spawning area i n Bush Creek during 1981 and 1982. Number of l i v e spawners observed versus the J u l i a n day. Median day of each run shown by arrow 63 Figure 7. D i e l t iming of emergence of f r y from AB, WB and W 72 Figure 8. Timing of f r y downstream migrations from Walker Creek and the upper and lower s e c t i o n s of the spawning area i n Bush Creek during 1982 and 1983. Number of f r y captured i n i n c l i n e d p l a i n t r a p s versus the J u l i a n day 75 xvi i i ure 9. Temperature and flow p r o f i l e s from Bush and Walker Creeks during 1981-82 and 1982-83. Closed c i r c l e s = Bush Creek; Open c i r c l e s = Walker Creek ure 10. Egg development stage (Velsen 1980) versus log day (post-August 31, 1981) for AB and W. Horizontal bars are proportional to the number of eggs at each stage ure 11. Egg weight versus female orbit-hypural plate length of spawners from Walker Creek and the upper and lower sections of the spawning area in Bush Creek during 1981 and 1982 .... ure 12. Proposed graphical model of the i n t e r a c t i o n between vertebral count program and temperature of incubation among AB, WB, and W during 1978-79, 1979-80, 1981-82. Open c i r c l e s = AB, Closed c i r c l e s = W, Triangles = WB. Arrows indicate the c r i t i c a l temperature determining vertebral count response of progeny from "EARLY" and "LATE" spawners ure 13. Incubation temperatures (C) of the 1982-83 and 1983-84 experiments. S o l i d l i n e s represent the planned changes in temperature regime • xi x Figure 14. Time to hatch under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime for AB, WB, W, and W x WB in the 1982-83 experiment 131 Figure 15. Time to hatch under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime for AB, WB, W, and W x WB in the 1983-84 experiment 135 Figure 16. Time to emergence under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime for AB, WB, W, and W x WB in the 1982-83 experiment .142 Figure 17. Time to emergence under 6°C, 10°C, simualted autumn spawning regime, simulated winter spawning regime for AB, WB, W and W x WB in the 1983-84 experiment 146 Figure 18. Survival under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime for AB, WB, W and W x WB in the 1982-83 experiment from f e r t i l i z a t i o n to epiboly, epiboly eyed and eyed to hatch 158 Figure 19. Survival under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime for AB, WB, W, and W x WB in the 1983-84 experiment from f e r t i l i z a t i o n to epiboly, epiboly to eyed and eyed to hatch 162 XX Figure 20. Survival under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime for AB, WB, W, and W x WB in the 1982-83 experiment from f e r t i l i z a t i o n to hatch, hatch to emergence, and f e r t i l i z a t i o n to emergence 165 Figure 21. Survival under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime f o r AB, WB, W and W x WB in the 1983-84 experiment from f e r t i l i z a t i o n to hatch, hatch to emergence, and f e r t i l i z a t i o n to emergence 170 Figure 22. Schematic drawings to show the body form of progeny from AB, WB, and W rearings from 6°C treatment of the 1983-84 experiment ^83 Figure 23. Schematic drawings to show the body form of progeny from AB, WB, and W rearings from 10°C treatment of the 1983-84 experiment 186 Figure 24. Schematic drawings to show the body form of progeny from AB, WB, and W rearings from simulated autumn spawning treatment of the 1983-84 experiment 188 Figure 25. Schematic drawings to show the body form of progeny from AB, WB, and W rearings from simulated winter spawning treatment of the 1983-84 experiment 190 xxi Figure 26. Pathways for Gene flow among AB, WB and W 220 Figure 27. Genetic s i m i l a r i t y using Nei's (1978) unbiased genetic s i m i l a r i t y 239 Figure 28. Wagner Tree produced by rooting at midpoint of longest path (base measure used Prevosti Distance [Wright 1978]) 243 Figure 29. B a r r i e r s to crossmating showing the cumulative e f f e c t on genetic migration (After Mayr 1970) 258 xxi i LIST OF APPENDICES Appendix 1. Analysis of variance tables estimating the e f f e c t of population and temperature regime for mean time to hatch and emergence during the 1982-83 and 1983-84 incubation experiments 294 Appendix 2. Comparisons and contrasts among means of time to hatch and time to emergence of embryos reared in the 1982-83 and 1983-84 experiment. C a l c u l a t i o n of C r i t i c a l Value for SS-STP Test C. V. = (a - 1) x MSW x F (a-1), a(n-1), alpha Appendix 3. G-tests of independence of s u r v i v a l from f e r t i l i z a t i o n to epiboly of embryos reared during 1982-83 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg s i z e 304 Appendix 4. G-tests of independence of s u r v i v a l from epiboly to eye pigment stage of embryos reared during 1982-83 from the e f f e c t s of temperature, population, season of spawning, l o c a t i o n of spawning and egg s i z e 310 xxi i i Appendix 5. G-test of independence of s u r v i v a l from eye pigment stage to hatch of embryos reared during 1982-83 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg siz e i 316 Appendix 6. G-tests of independence of s u r v i v a l from f e r t i l i z a t i o n to hatch of embryos reared during 1982-83 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg siz e 322 Appendix 7. G-tests of independence of s u r v i v a l from hatch to emergence of embryos reared during 1982-83 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg siz e 328 Appendix 8. G-tests of independence of s u r v i v a l from f e r t i l i z a t i o n to epiboly of embryos reared during 1983-84 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg siz e 334 Appendix 9. G-tests of independence of s u r v i v a l from epiboly to eye pigment stage of embryos reared during 1983-84 from the e f f e c t s of temperature, population, season of spawning, loc a t i o n of spawning and egg s i z e 340 xxi v Appendix 10. G-tests of independence of survi v a l from eye pigment stage hatch of embryos reared during 1983-84 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg size 346 Appendix 11. G-tests of independence of s u r v i v a l from f e r t i l i z a t i o n to hatch of embryos reared during 1983-84 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg siz e 352 Appendix 12. G-tests of independence of s u r v i v a l from hatch to emergence of embryos reared during 1983-84 from the e f f e c t s of temperature, population, season of spawning, loc a t i o n of spawning and egg siz e 358 Appendix 13. Comparisons and contrasts among means of vertebral counts of progeny from the 1983-84 experiment using Scheffe's method 364 Appendix 14. Population si z e estimated by s u r v i v a l time and fish-days . 366 Appendix 15. Temperature changes (planned) in laboratory experiments .. 368 XXV Appendix 16. Egg sizes and s i z e of females used in laboratory experiments 370 Appendix 17. Water temperatures in r e l a t i o n to month and la t i t u d e i n streams u t i l i z e d by chum salmon stocks 372 xxvi ACKNOWLEDGEMENTS I wish to thank my supervisor, T.G. Northcote, for his valuable advice and support throughout the project. Bob B a l l , Ted Carter, and Clyde Murray of the P a c i f i c B i o l o g i c a l Station i n Naniaimo, B.C. helped with the c o l l e c t i o n of f i s h for the breeding experiments. T.D. Beacham, R. Withler and B. Riddell of the P a c i f i c B i o l o g i c a l Station and C. Busack of the University of M i s s i s s i p p i provided h e l p f u l advice at d i f f e r e n t times during the project. My supervisory committee of C C . Lindsey, J.D. McPhail, J. Myers and C.J. Walters of the University of B r i t i s h Columbia gave me constructive c r i t i c i s m on the research plan and the t h e s i s . In many cases they challenged me to explore some novel approaches to my ideas and r e s u l t s . I wish to e s p e c i a l l y thank C C . Wood of the P a c i f i c B i o l o g i c a l Station and T.P. Quinn of the University of Washington in Seattle who l i s t e n e d p a t i e n t l y and commented i n c i t e f u l l y on some of my more bizarre ideas. F i n a l l y , I extend my warmest appreciation to M.C. Healey of the P a c i f i c B i o l o g i c a l Station who served as my co-supervisor. Dr. Healey revolutionized my thinking by formally introducing me to the philosophy of s c i e n t i f i c i n v e s t i g a t i o n . I thank him for h i s kindness, i n t e l l i g e n c e , and honesty during my education. Project expenses were paid by the Department of F i s h e r i e s and Oceans, Canada. I was supported f i n a n c i a l l y by an N.S.E.R.C. post-graduate scholarship, and a University of B r i t i s h Columbia Research Fellowship. 1 INTRODUCTION The c l a s s i c a l and apparently most widespread mechanism of interpopulation d i f f e r e n t i a t i o n i s through the reduction of gene flow between two populations by virt u e of t h e i r geographic i s o l a t i o n (Templeton 1981). Selection acts d i f f e r e n t i a l l y on each i s o l a t e d gene pool r e s u l t i n g in population s p e c i f i c adaptations to the l o c a l environmental conditions (Mayr 1963). This i s known as the a l l o p a t r i c adaptation or speciation model. Another possible mode of adaptive genetic divergence i s through i s o l a t i o n of populations i n time rather than i n space (Carson and Templeton 1984). An example of t h i s i s when a species has separate populations breeding i n d i f f e r e n t seasons or years. For example, Alexander and Bigelow (1960) proposed that seasonal environmental constraints on the l i f e h istory of f i e l d c r i c k e t species in temperate zones would re s u l t in the development of temporally i s o l a t e d populations adapted to a p a r t i c u l a r seasonal cycle of reproduction. They coined the term " a l l o c h r o n i c " adaptation or speciation for t h i s s i t u a t i o n . Seasonally separated reproductive populations have been reported i n diverse taxonomic groups such as flowering plants ( C h i a r e l l o and Roughgarden 1984), insects (Alexander and Bigelow 1960, Naraoka 1987, Tauber et a l . 1977, Tauber and Tauber 1976, 1981), amphibians ( B l a i r 1941), r e p t i l e s (Mendonca 1987) and fishes (Berg 1959). An example in the flowering plants i s the seasonal ecotypes of Hemizonia l u z u l i f o l i a , an annual plant of the C a l i f o r n i a grasslands. One form (subspecies l u z u l i f o l i a ) germinates i n winter, flowers in A p r i l , and l i v e s u n t i l e arly summer, whereas the other (subspecies rudis) 2 germinates i n winter but flowers in midsummer, and often l i v e s u n t i l early autumn ( C h i a r e l l o and Roughgarden 1984). In the Insecta there are two groups that have been studied extensively: the lacewings and the f i e l d c r i c k e t s . In both there are species complexes which at some la t i t u d e s act as seasonally i s o l a t e d species while at other l a t i t u d e s a broad hybrid zone of overlap e x i s t s (Harrison 1985). Temporal i s o l a t i o n may occur when development rates to maturity d i f f e r as i s the case between Gryllus firmus and Gryllus pennsylvanicus. The timing of the onset and termination of diapause i s thought to control seasonal cycles in Chrysopa (Tauber and Tauber 1982). In southern l a t i t u d e s diapause i s short in duration and breeding seasons long. Overlapping breeding seasons r e s u l t and seasonal i s o l a t i o n breaks down. In the northern l a t i t u d e s the seasonal environmental constraints are more severe, diapause i s lengthened in each group and breeding seasons do not overlap. Mating i n the "Chrysopa downesi type" occurs during A p r i l whereas mating i n the "Chrysopa cornuta type" occurs during June to August and i n early March (Tauber and Tauber 1982). Tauber et a l . (1977) proposed that seasonal i s o l a t i o n of these types occurred as a res u l t of changes at only two l o c i . Interpopulation v a r i a t i o n i n season of reproduction occurs i n several families of temperate zone fishes (Berg 1959). There are seasonally separated reproductive populations in the Petromyzontidae, Clupeidae, Salmonidae, Cyprinidae, Belonidae and Percidae (Neave 1949, Berg 1959, Goldberg and Pizzorno 1985). 3 Several researchers have suggested that salmonid populations which d i f f e r with respect to t h e i r seasonal timing of reproduction function as separate gene pools (Smith 1969, Berg 1934, 1959, Okazaki 1978, 1981). Studies of polymorphic enzymes among seasonally d i s t i n c t populations support the claim of limited gene flow among seasonally separated stocks (Altukov and Salmenkova 1981, Altukov 1981, Okazaki 1978). For example, Okazaki (1978) compared the electrophoretic patterns at several enzyme l o c i of two seasonally separated chum salmon spawning populations of the Tokachi River, Japan. He found that the l a t e r spawning run was more si m i l a r in enzyme frequencies to a nearby la t e spawning population than the early spawning population in the same stream. Okazaki (1978) postulated that the l a t e spawning population had arisen from transplants from the other stream and that the two runs within the stream were reproductively i s o l a t e d . Smith (1969) recorded differences in ph y s i o l o g i c a l and s t r u c t u r a l characters such as fat storage and g i l l raker number among winter and summer migrants of steelhead trout (Salmo gairdneri) i n B r i t i s h Columbia streams. Differences in these characters were maintained among the two groups when reared under controlled conditions i n d i c a t i n g that the differences were i n h e r i t e d . Leider et a l . (1984) found that p a r t i a l reproductive i s o l a t i o n existed between summer and winter steelhead t r o u t . Except for the work of Smith (1969) on winter and summer races of steelhead trout there i s l i t t l e information regarding genetic divergence in polygenic t r a i t s among salmonid populations reproducing during d i f f e r e n t seasons. The comparative study of seasonally separated spawning populations of fishes has focussed on the adult phase of the l i f e c y c l e . However, the 4 e c o l o g i c a l consequences of allochrony may be greatest during the early l i f e history stages. Vladimirov (1975) proposed that the s t a r t of exogenous feeding i s a c r i t i c a l period of development in f i s h larvae. Fish may perish at t h i s time due to defects in t h e i r organ systems connected with the search, capture and a s s i m i l a t i o n of food organisms. In Salmonidae, t h i s period occurs at the time of the f i n a l resorption of the yolk and the s t a r t of active feeding. Concurrently, s e l e c t i o n pressure by predation can cause convergent avoidence t a c t i c s i n prey belonging to widely divergent taxa (Kneib 1987, Main 1987). P e r f e c t l y formed fry may s t i l l perish i f they complete embryonic development when there i s a lack of s u f f i c i e n t exogenous food or there are predators present (MacNamara and Houston 1987). Recently, Rutberg (1987) proposed that b i r t h was synchronized in ruminant species in order to reduce predation as well as to coincide with optimal environmental conditions. Consequentially, some researchers have suggested that the timing of post l a r v a l migration and early l i f e h i s t o r y morphs of juvenile salmonids are under seasonal s e l e c t i v e constraints (Godin 1982, M i l l e r and Brannon 1982). Fry must appear on the nursery grounds at a p a r t i c u l a r time of the year with a p a r t i c u l a r form to be successful (Godin 1982, Fresh et a l . 1982, Healey 1979). For example, young salmon released at d i f f e r e n t times of the year showed variable s u r v i v a l to maturity ( B i l t o n 1980, Taylor 1980). B i l t o n (1980) proposed that there was an optimal yearly time of downstream migration. Healey (1979) and Sibert (1979) found that the annual timing of chum salmon fry migrations i n the Nanimo River, B.C. corresponded to the peak densities of t h e i r preferred food type in the estuary. 5 Allochrony complicates the development of a cohesive phenotype whose form and behaviour are "tuned" to the environmental c o n d i t i o n s . Consider two populations of ectothermic organisms: one reproducing during the s p r i n g , the other reproducing during the summer months. The o f f s p r i n g of the f i r s t population must then develop while temperatures are r i s i n g from c o o l to warm. The o f f s p r i n g of the second population s t a r t development at a plateau of warm temperatures. The rates of developmental processes, dependent on temperature of in c u b a t i o n , w i l l vary between populations as a r e s u l t . I t i s not d i f f i c u l t to e n v i s i o n that the progeny of the e a r l y group w i l l be somewhat more developed than those of the l a t e group at any p a r t i c u l a r time. However, i f there i s al s o a seasonal c o n s t r a i n t on the l i f e h i s t o r y , f o r example, an optimal time to begin winter h i b e r n a t i o n , both groups may be se l e c t e d to converge i n p h y s i o l o g i c a l s t a t e and stage of development. I f the genes c o n t r o l l i n g the t r a i t act on the expression of the t r a i t i n an a d d i t i v e f a s h i o n , the e f f e c t on the genetic composition of t h i s c u t - o f f point w i l l , be s i m i l a r to a t r u n c a t i o n s e l e c t i o n experiment (Falconer 1981) where each population i s s e l e c t e d i n opposite d i r e c t i o n s . However, the phenotypic composition, s e l e c t e d to conform to a s i n g l e optimum, w i l l be e f f e c t i v e l y under s t a b i l i z i n g s e l e c t i o n . P a c i f i c salmon have adapted to a large v a r i e t y of spawning environments that d i f f e r i n many ways, e.g. annual patterns of water flow, temperature, i n s o l a t i o n , g r a v e l s i z e s , d i s t a n c e from the ocean, e t c . The adaptation of a stock to an environment i s probably a f f e c t e d by the s e l e c t i o n of a l l e l e s at l o c i which c o n t r o l c e r t a i n key f i t n e s s t r a i t s of the stock such as annual 6 timing of spawning, the temperature dependent development rate of embryos and timing of emergence of larvae, the si z e of eggs, or the choosing of appropriate nest s i t e s . I f the v a r i a b i l i t y of these f i t n e s s t r a i t s in the stock i s at least partly due to additive genetic v a r i a b i l i t y at the c o n t r o l l i n g l o c i , then these t r a i t s can re a d i l y change in response to natural selection, allowing the stock to colonize new environments. For example, sockeye salmon spawn i n t r i b u t a r i e s of Mendenhall Lake near Juneau, Alaska; yet 60 years ago Mendenhall Glacier covered those spawning grounds. Although, i t i s l i k e l y that man-made introductions are responsible (According to J.H. Helle National Marine Fi s h e r i e s Service, Auke Bay, Alaska, pers. comm., sockeye fry were planted i n the Mendenhall River i n 1922 and 1951) adaptation must have been rapid. The v a r i a b i l i t y can also allow the stock to adapt to changes i n e x i s t i n g spawning habitats. For example, i n 1964 an earthquake l i f t e d land masses around Prince Williams Sound, Alaska, s u b s t a n t i a l l y a l t e r i n g the spawning habitat of many chum and pink salmon (fj. gorbuscha) stocks (Noerenberg, 1971) but most of these stocks have endured and are very productive. According to J.H. Helle (pers. comm.) at least two major land elevation changes have occurred with earthquakes i n the l a s t 20 years. These apparently al t e r e d the i n t e r t i d a l areas and forced adaptation to a new landscape. Pink and chum salmon have both developed i n t e r t i d a l spawning in these areas. It seems l o g i c a l that genetic adaptations should evolve in temperate ectotherms, such as salmonids, that reproduce during d i f f e r e n t seasons as the progeny s t a r t development at d i f f e r e n t times of the year, and develop i n d i f f e r e n t thermal conditions (that i s during d i f f e r e n t seasonal 7 c y c l e s ) . Many t r a i t s are s e n s i t i v e to the temperature of incubation and there may be s e l e c t i v e constraints on the expression of these t r a i t s . The p o s s i b i l i t y that season of reproduction i s important i n genetic divergence among populations not only has implications for the study of evolution but also for the more p r a c t i c a l e f f o r t s made in the science of f i s h e r i e s management. In p a r t i c u l a r , i t may broaden the working d e f i n i t i o n of "stock". This i s a t h e o r e t i c a l construct used in a l l f i s h e r i e s management e f f o r t s . Larkin (1972) introduced the "stock concept" as a valuable t h e o r e t i c a l t o o l for f i s h e r i e s management e f f o r t s . A stock may be defined (sensu Helle, 1981) as a group of fishes that form a unit in time and space. Subsequent authors have emphasized the u t i l i t y of the stock concept towards conservation of the ec o l o g i c a l and genetic c h a r a c t e r i s t i c s of commercially exploited species (Harlan 1981, Helle 1981, Kapuscinski and Lannan 1986). A grand e f f o r t has been made to define, in p r a c t i c a l terms, what represents a d i s c r e t e stock or management unit (Ihssen 1976, Ihssen et a l . 1981). Stocks have been described on the basis of e c o l o g i c a l differences such as the distance from the sea of the spawning grounds. As well, morphological, p h y s i o l o g i c a l and behavioural c h a r a c t e r i s t i c s may d i f f e r between stocks (Taylor and McPhail 1985, Riddell and Leggett 1981). In most cases, stocks have been bounded by t h e i r location of reproduction. For example, a l l fishes of the same species that spawn in the same stream or a l l fishes of the same 8 species that spawn in streams of a p a r t i c u l a r area may be considered as disc r e t e stocks. Beacham et a l . (1985) proposed that a l l chum salmon, Oncorhynchus keta, reproducing in streams of the Fraser River should be considered as one stock. S i m i l a r l y , Taylor and McPhail (1985) proposed that coastal spawning coho salmon, Oncorhynchus kisutch, were discr e t e from i n t e r i o r spawning coho. Migratory timing has been used in f i s h e r i e s management to segregate stocks in many systems including the Fraser River ( K i l l i c k 1955), Yukon River (Buklis 1982), Copper River (Merritt and Roberson 1986), and the Great Lakes (Biette et a l . 1981). In general, phenotypic and genetic v a r i a t i o n associated with seasonal timing of reproduction has not been investigated as thoroughly as that associated with the l o c a t i o n of spawning. For reviews of the stock concept in r e l a t i o n to f i s h e r i e s management see Larkin (1981) and Loftus (1981). Stock d e f i n i t i o n r e l i e s on e f f e c t i v e methods of stock i d e n t i f i c a t i o n . Information c o l l e c t e d to define wild stocks f a l l s into two broad categories: 1) t r a i t s that are known or presumed to be unaffected by the environment of development; 2) t r a i t s that are affected by the developmental environment. The former includes karyotypic, electrophoretic and immunological analyses as well as analyses of mitochondrial DNA. The l a t t e r includes analyses of morphological, p h y s i o l o g i c a l , and behavioural v a r i a t i o n among proposed stocks. At f i r s t glance, the f i r s t methods seem to be the most sensible to use because they are not confounded by the e f f e c t s of the environment. Electrophoretic analysis has been the most popular for f i s h e r i e s work because large samples can be handled r a p i d l y . This lends i t s e l f well to "In Season" 9 management operations. However, i t i s often d i f f i c u l t to r e l a t e the pattern of electrophoretic v a r i a t i o n to the e c o l o g i c a l v a r i a b i l i t y i n the species. It may be more important for future maintenance and production to preserve the stock s p e c i f i c adaptations in morphology, physiology or behaviour than conservation of a pattern of polymorphic enzyme var i a t i o n that has occurred due to genetic d r i f t . Relatively intensive studies of the phenotypic v a r i a t i o n and i t s underlying genetic v a r i a b i l i t y in key morphological, p h y s i o l o g i c a l and behavioural c h a r a c t e r i s t i c s of proposed stocks of a species can also be used to make management decisions. C l e a r l y , there i s a concordance between the information needed to define "stocks" and the information needed to understand evolutionary processes. For both the f i s h e r i e s manager and the evolutionary b i o l o g i s t the f i r s t question i s : Does a p a r t i c u l a r observable d i s c o n t i n u i t y in the ecology of a species ( i . e . a migration timing difference) indicate that evolution has occurred? Evolution i s defined as genetic change as a r e s u l t of adaptation to an e c o l o g i c a l s i t u a t i o n (Hartl 1981). Secondarily, what information i s required to determine i f evolution has occurred? This depends upon the approach one wishes to take. Studies must always be comparative. The words "to change" implies t h i s . In e f f e c t , we must always compare two populations in one way or another (Endler 1986). 10 Evolution can proceed in two ways: 1) A population can change g e n e t i c a l l y and, perhaps, phenotypically in response to a change in the environment. This i s the s e r i a l progression approach (Figure 1). One compares what a sing l e population i s now as opposed to what i t used to be. Examples of progressions are found in the f o s s i l record: i . e . Homo sapiens, horses Ephippus etc. (Stanley 1979) and in extant populations i . e . change of wing c o l o r a t i o n i n moths (Biston spp) with i n d u s t r i a l i z a t i o n i n Great B r i t a i n (Kettlewell 1956). 2) Populations can branch away from each other. They can change g e n e t i c a l l y and often phenotypically to occupy d i f f e r e n t niches. This i s thebranching evolution approach since one compares two populations that are assumed to have a common d e r i v a t i o n . This l a t t e r mode of evolution i s i n t e r e s t i n g because i t i s the process by which speciation ultimately occurs. An understanding of t h i s process allows one to explain why there are so many kinds of organisms (Hutchison 1959). Since I am interested i n speciation I have chosen t h i s approach. 11 I SERIAL BRANCHING Figure 1. S e r i a l and branching e v o l u t i o n 12 So again ... What information i s required to determine i f evolution, as defined above, has occurred? If an ec o l o g i c a l s i t u a t i o n does result i n s p l i t t i n g , one might predict the following: 1) Phenotypic differences among the populations. 2) An adaptive basis to phenotypic v a r i a t i o n between the groups ( s e l e c t i v i t y d i f f e r e n c e s ) . 3) The differences i n phenotype have a genetic basis. 4) Aspects of the species ecology would be conducive to genetic change ( i . e . "homing" to place of b i r t h , small e f f e c t i v e population s i z e s , strong s e l e c t i o n at points in the l i f e h i s t o r y ) . Restated as questions to be answered: 1) Are there phenotypic differences among the populations? 2) Do these features have an adaptive basis? 3) Are the differences i n phenotype g e n e t i c a l l y determined? Are the stocks g e n e t i c a l l y d i f f e r e n t ? 4) How did the stocks become ge n e t i c a l l y d i s t i n c t ? What were the casual factors in genetic d i f f e r e n t i a t i o n ? 5) Are there a l t e r n a t i v e explanations for the observations? The purpose of t h i s work i s to determine i f populations that spawn during d i f f e r e n t seasons: 1) are temporally i s o l a t e d even when i n the same loca l e ; 2) show c h a r a c t e r i s t i c phenotypic differences i n the adults or f r y ; 3) show 13 evidence of adaptations for s u r v i v a l in t h e i r respective seasonal environments; 4) are phenotypically and/or g e n e t i c a l l y d i s t i n c t stocks. In the f i r s t section I review the l i t e r a t u r e on chum salmon ecology to describe the factors that a f f e c t the d i s t r i b u t i o n and abundance during the adult, embryonic, and juvenile phases of the l i f e c y c l e . Three main points are i l l u s t r a t e d in t h i s section: 1) the population structure of t h i s species has both a geographic and seasonal component; 2) migrations form a major component of the l i f e c y c l e ; 3) mortality and hence s e l e c t i o n i s greatest in the early stages of l i f e . The next section b r i e f l y reviews explanations of the evolutionary o r i g i n s of 'seasonal races' in salmonids. The next section presents empirical information on chum salmon populations returning to the same loca l e but reported to spawn during d i f f e r e n t seasons to describe the degree of temporal i s o l a t i o n through measures of the seasonal overlap in time of spawning, and the rate of straying of spawners in time and space. Evidence for phenotypic s t r u c t u r i n g among seasonally and s p a t i a l l y separated groups i s investigated by comparisons of age at maturity, length at maturity, vertebral counts, freshwater s u r v i v a l time of adults and the time to hatch, timing of downstream migration, vertebral counts and morphological features of the progeny of these populations. S t a b i l i z i n g s e l e c t i o n i s suggested as a possible cause of the r e l a t i v e uniformity of phenotype among the populations. Adaptation of the incubation rate program to the season of reproduction i s suggested by comparisons of the time to emerge and incubation environments. The next two sections confirm the genetic basis of stock s p e c i f i c adaptations to season of reproduction in the incubation rate, vertebral number and 14 external morphology of the fry based on the re s u l t s of laboratory rearings of the progeny of the stocks under co n t r o l l e d conditions. These two sections are followed by an analysis of genetic divergence among the populations using polymorphic enzymes. This i s an e n t i r e l y d i f f e r e n t approach to the question of genetic r e l a t i o n s h i p s among populations than analysis using .quantitative t r a i t s . It i s also a standard p r a c t i s e used in determining stock differences for f i s h e r i e s management. The r e s u l t s of th i s section provide an i n t e r e s t i n g contrast to those of the previous sections. Some conclusions drawn from the information contradict those drawn from quantitative t r a i t s and therefore force a change i n perspective regarding the r e l a t i o n s h i p s among the populations. In the f i n a l section I summarize the important conclusions of t h i s work and synthesize to provide resolution to the apparent contradictory information presented. 15 SELECTION AND THE ECOLOGY OF CHUM SALMON Chum salmon, Oncorhynchus ket a, i s a highly variable anadromous species of f i s h of the North P a c i f i c Ocean. It has the widest endemic d i s t r i b u t i o n of the P a c i f i c salmons (Oncorhynchus spp.) (Figure 2) (Bakkala 1970). The standing crop of chum salmon in the north P a c i f i c Ocean considerably exceeds the other salmon species (Bakkala 1970). Chum salmon undergo several transformations during t h e i r l i f e . As a result the l i f e h i s t o r y i s most conveniently divided into stages: egg to f r y ; downstream migrant; coastal j u v e n i l e ; pelagic j u v e n i l e ; coastal adult; reproductive adult. These may be grouped into freshwater stages (egg to f r y , reproductive adult) concerned mainly with reproductive biology, and marine stages (coastal j u v e n i l e , pelagic juvenile) concerned mainly with growth. Between these are important t r a n s i t i o n stages (coastal adult; downstream migrant) that involve anadromous and catadromous migrations. Each stage has associated causes of mortality and mortality rates. Chum salmon spawn i n freshwater streams throughout the North P a c i f i c Rim. In North America the southernmost record of occurrence of chum salmon i s the San Lorenzo River in Monterey, C a l i f o r n i a (long. 122°W., lat.°37 30°N.) (Atkinson et a l . 1967). Spawning streams exist from t h i s point northward as far as the Mackenzie River (long. 135°W., l a t . 69°N.) on the A r t i e coast of the Northwest T e r r i t o r i e s (Aro and Shepard 1967). In Asia the southern l i m i t Figure 2. World d i s t r i b u t i o n of chum salmon. 17 i s in the Saga Prefecture (long. 130°E., l a t . 33°N.) of Kyushu Island, in the Sea of Japan (Kimura 1981) while spawning occurs as far north as the Lena River (long. 125°E., l a t . 73°N) on the A r t i e coast of the U.S.S.R. (Sano 1966) (See Figure 2). In Asia there are two forms of chum salmon recognized by t h e i r season of reproduction: the summer chum salmon native to Kamchatka, the Okhotsk coast, the Amur River and the east coast of Sakhalin Island and the autumn chum salmon native to Japan, Sakhalin Island, the southern Kurile Islands and the Amur River (Sano 1966). Summer chum salmon migrate to spawn in June, July and August. The autumn chum salmon spawn mostly from September onward. In the Soviet d i s t r i c t s of O l y u t o r s k i i , Anadyr, Okhotsk and the west coast of Kamchatka mature adults ascend to spawn from June to September with peak runs in July or August (Sano 1966). In the Amur River there are two separate spawning runs of chum salmon, one during July and August and another from August through to early October (Sano 1966). In the r i v e r s of the Sakhalin and Kurile Islands spawning occurs in September and October. In Japan, the peak spawning runs occur in September and October on Hokkaido Island and October and November on Honshu and Kyushu Islands. Seasonal forms or "races" are less accepted in North America although there i s evidence that they e x i s t (Bakkala 1970). A general trend of spawning l a t e r i n the year with decreasing l a t i t u d e occurs throughout the range. 18 There are two temporally d i s t i n c t spawning migrations in the Yukon River, Alaska (Buklis and Barton 1984). The summer chum salmon enter the Yukon River in early May and continue spawning u n t i l the middle of July. The autumn spawning run occurs in June and July. In the rest of the state spawning periods range from July and August in northern Alaskan r i v e r s to August and September in southeastern Alaska (Salo 1986). Further south, on Prince of Wales Island, chum salmon spawn mainly in September and October. There are over 880 chum salmon spawning streams in B r i t i s h Columbia. Spawning takes place from October to January in the south of B r i t i s h Columbia. For example, peak spawning occurs i n October in the Chehalis River as compared to November to January in the mainstem of the Fraser River, Chilliwack-Veddar and Harrison Rivers. In the northern part of B r i t i s h Columbia spawning runs occur e a r l i e r in the year. Spawning occurs in August and September on the Queen Charlotte Islands and during July and August on the northern B r i t i s h Columbia mainland (Aro and Shepard 1967). In Washington spawning occurs mainly in December and January (Atkinson et a l . 1967) although there are spawning runs in October (Koski 1975). Atkinson et a l . (1967) note that spawning occurs mainly in l a t e November and early December in Oregon. 19 LIFE HISTORY IN FRESHWATER Adult Phase Coastal Adult Like a l l P a c i f i c salmon, chum salmon return to t h e i r stream of b i r t h to spawn and then d i e . Once mature the chum salmon begins a migration towards freshwater. Parker (1962) estimated the mean monthly instantaneous mortality rate to be 0.035 at t h i s stage. A t o t a l mortality of 0.070 occurred over two months. Taguchi (1961) gives a mean monthly instantaneous mortality rate of 0.381 for the l a s t 100 days of ocean existence of the chum salmon. However, Ricker (1976) regards t h i s estimate as'much too high. Spalding (1964) reported that harbor seals and sea l i o n s are the main predators during the return journey. Spawning Adult Chum salmon generally spawn less than 200 km from the sea i n smaller streams (Bakkala 1970). In these s i t u a t i o n s some spawning may occur in the t i d a l zone. Migrating f i s h are notably reluctant to surmount b a r r i e r s in the stream. However, some populations undertake long migrations to spawning grounds up to 2500 km upstream in large r i v e r s such as the Yukon and the Amur River (Sano 1966, Neave 1966). The l a t e r spawning populations migrate further upstream. The increased water flow at t h i s time may allow them to surmount b a r r i e r s that impede the progress of e a r l i e r a r r i v i n g f i s h . 20 Sex r a t i o s have been shown to change during the spawning migration but for the e n t i r e period of migration they approach 1:1. The number of males to females declines during the course of the spawning run (Salo 1986). Age composition changes as the spawning season progresses. Generally, e a r l i e r migrants are older than l a t e migrants. For example, Salo (1986) noted that the mean age of spawning chum salmon in the Fraser River declined from 2.98 i n October to 2.78 years in December during the 1970's. In contrast, Mattson and Roland (1963) found that age 3 f i s h dominated early migrants and age 4 f i s h dominated late migrants at T r a i t o r s Cove, Alaska in one year. Although a l l spawners die once reproduction i s completed t h e i r r e l a t i v e longevity in freshwater i s important to mating success and egg to fry s u r v i v a l . The longer a male chum salmon can remain active the greater number of p o t e n t i a l mates he can encounter. Greater longevity in the female means a lower p r o b a b i l i t y of redd superimposition by another female. Freshwater l i f e varies greatly from a few days to several months (Trasky et a l . 1974, Koski 1975), Spawners are subject to a variety of predators while they are attempting to migrate and spawn. Bears, g u l l s and bald eagles take advantage of the presence of the f i s h i n a r e l a t i v e l y confined area during t h i s phase. 21 Egg to Fry Phase Both a b i o t i c and b i o t i c factors influence the s u r v i v a l of embryos and the c h a r a c t e r i s t i c s of the emerging fry (Koski 1975). Stream flow, water temperature, dissolved oxygen, gravel composition, spawning time, density of spawners, and stock c h a r a c t e r i s t i c s may influence the well-being of the embryonic chum salmon (Koski 1975). Parker (1962) estimated that 92.2 per cent of chum salmon in Hook Nose Creek perished during the seven months of the egg to fry phase. The mean monthly instantaneous rate of mortality during t h i s period was 0.364. Drought may cause egg mortality d i r e c t l y by leaving the redds dry or i n d i r e c t l y by allowing other mortality causing factors to operate (Neave 1953). McNeil (1966) and Wickett (1958) found low oxygen and high mortality when stream discharge was low during and af t e r spawning. Poor s u r v i v a l in dry years may be the result of a greater incidence of freezing of redds (McNeil 1966). The duration of the incubation period and the s i z e of the fry may be affected by the v e l o c i t y of water at the nest s i t e . Developmental rate increases and larger fry are produced at higher stream flows than at lower flows (Bams 1982). Summer chum salmon spawn i n deeper waters and at higher water v e l o c i t i e s than do the autumn chum salmon (Salo 1986). The progeny of early stocks may therefore migrate e a r l i e r and at a larger s i z e than the progeny of stocks that spawn l a t e . 22 Water temperatures near or at freezing during the incubation can account for s i g n i f i c a n t m o r t a l i t i e s of salmonid eggs and alevins (McNeil 1966). According to Schroder (1973) a drop i n water temperature i n h i b i t s spawning. Chum salmon eggs incubated below 1.5°C during early development had s i g n i f i c a n t l y higher m o r t a l i t i e s than controls (Schroder 1974). Late spawning stocks often select areas with upwelling groundwater that remains above 4°C. During severe winters the autumn chum salmon have greater egg to fry s u r v i v a l than summer chum salmon ( N i k o l s k i i 1952). Late spawning stocks may also require fewer temperature units to emergence than early spawning stocks (Koski 1975). Survival of chum salmon eggs and alevins i s d i r e c t l y related to dissolved oxygen content (Wickett 1954). Wickett (1954) calculated the l e t h a l l e v e l to be 1.67 mg/liter. Koski (1975) found that s u r v i v a l was reduced at oxygen l e v e l s below 2 mg/liter. Lethal l e v e l s r i s e from f e r t i l i z a t i o n to hatching (Alderdice et a l . 1958, Fast and Stober 1984). Low oxygen during early development can cause a reduction in the incubation rate of the embryo (Alderdice et a l . 1958). Survival rates can vary greatly depending on the composition of materials i n the gravel. For example, Koski (1975), by systematically a l t e r i n g the composition of the gravel, found that s u r v i v a l s ranged from 7.2 per cent to 88.4 per cent depending on the mix. When fine materials such as sand and s i l t are deposited on the streambed, permeability i s reduced. Gravel composition 23 a f f e c t s the s u r v i v a l of salmonid alevins i n three ways: 1) di r e c t suffocation of eggs and alevins; 2) reduced intragravel water flow and dissolved oxygen content; and 3) a physical b a r r i e r to emergence (Iwamoto et a l . 1978). The morphology and behaviour of chum salmon alevins can influence t h e i r s u r v i v a l . Chum salmon alevins are r e l a t i v e l y slender compared to other P a c i f i c salmon embryos (Fast and Stober 1984). This allows them to move through smaller i n t e r s t i t i a l spaces. P o t e n t i a l l y , t h i s could make them less vulnerable to predators although no information e x i s t s to confirm t h i s . Chum salmon alevins exhibit photonegative behaviour sh o r t l y a f t e r hatching. This i s thought to be an adaptation for predator avoidance by keeping the alevins s a f e l y i n the gravel u n t i l they a t t a i n the morphology necessary for e f f e c t i v e predator avoidance (Fast 1985). MARINE LIFE HISTORY Downstream Migrant Phase After completing t h e i r embryonic development the fry emerge from the gravel and migrate downstream to the sea in the spring. Downstream migration generally occurs during one to two hours following n i g h t f a l l (Kobayashi and Ishikawa 1964, Godin 1982). 24 In coastal streams the migration of fry to the estuary requires only one night (Hoar 1958). However, i n some systems fry w i l l migrate during the day also (Mason 1974). Migration of Fraser River chum salmon fry occurred during daylight hours (Todd 1966). In longer r i v e r s , such as the Amur or the Yukon, fry may require two to three months to complete t h e i r downstream migration (Smirnov 1975). Downstream migrants may feed in freshwater p a r t i c u l a r l y i f the migration requires more than one night (Smirnov 1975). However, peak feeding has been recorded a f t e r sunset rather than during the daylight hours (Kobayashi 1960). Chironomidae, Trichoptera, Emphemeroptera, and Plecoptera larvae are important freshwater foods (Kostarev 1970; Frolenko 1970; Kobayashi and Ishikawa 1964). Rosly (1972) demonstrated that the condition of f r y was greater in years where there were lower flows. Rosly (1972) speculated that the better condition of the fry was due to improved feeding opportunities during the low flow years. Godin (1982) suggested that for each population there was an optimal annual timing of entry into seawater. The annual mean entry date increases with l a t i t u d e (Godin 1982). Walters et a l . (1978) using a computer simulation model combining timing of annual production of zooplankton, ration l e v e l , growth rate of young chum salmon and the e f f e c t of si z e s e l e c t i v e predation for the Fraser River chum salmon concluded that fry migrating before or a f t e r the peak would suffe r greater r e l a t i v e mortality compared to those fry migrating during the peak of the run. 25 Coastal Juvenile Bailey et a l . (1975) found that the chum salmon juv e n i l e s fed heavily on zooplankton in the estuary of T r a i t o r s Cove, Alaska. They noted that the su r v i v a l of fry depends on t h e i r growth rate and their a b i l i t y to escape from predators. The concluded that competative interactions for food could result in m o r t a l i t i e s . Parker (1962) estimated mortality at 94.6 per cent over f i v e months. The mean monthly instantaneous rate of mortality was 0.582 during t h i s phase. Pelagic Juvenile Chum salmon w i l l remain dispersed over the North P a c i f i c Ocean for several years. Asian chum salmon extend eastward as far as 140°W compared to a westward l i m i t of 175°E for North American chum salmon. L i t t l e i s known of the causes of mortality during t h i s stage. Parker (1962) estimated mortality at 43.4 per cent over 34 months with a mean monthly instantaneous mortality rate of 0.017. Low water temperatures and low s a l i n i t y during early ocean residence may a f f e c t s u r v i v a l adversely (Wickett 1958; Birman 1959). Predators of pelagic juveniles include the hagfish (Polistotrema s t o u t i i ) , lamprey (Entosphenus tr i d e n t a t u s ) , mackerel shark (Lamna d i t r o p i s ) , fur seal (Callorhinus ursinus), sea l i o n (Eumetopis jubata), harbor seal (Phoca v i t u l i n a ) , f i n whale (Balaenoptera physalus), humpback 26 whale (Megaptera nodosa), k i l l e r whale (Orcinus orca) and beluga (Delphinapterus leucas) (Clemens and Wilby 1946). Most of the mortality in chum salmon occurs in the egg to fry and coastal juvenile phases. Survival in the egg to fry phase depends on the conditions in incubation environment. These are determined by the spawning adult's choice of redd s i t e as well as the annual timing of spawning. Survival of the coastal marine juvenile may also depend on the incubation environment. Differences in development rate w i l l a f f e c t the timing of the migration of f r y into the estuary (See Figure 3). Subspecific l e v e l s of organization Several l e v e l s of genetic organization below the l e v e l of the species appear to ex i s t i n chum salmon. The sexes, populations and 'races' exhibit differences in the length, weight, age at maturity and, in the l a t t e r two categories, fecundity. The sexes show differences in length at age, weight at age and age at maturity. Male chum salmon have a r e l a t i v e l y more rapid growth rate i n t h e i r f i n a l year at sea and hence return to spawn at a greater length and weight than females of the same age (Salo 1986, Bakkala 1970). In general, more males than females return to spawn at age 0.2 while more females than males return to spawn at age 0.4 (Thorsteinson et a l . 1963). Egg to Fry TEMPERATURE ol INCUBATION Downstream migrant Fry Emergence size date t ime of day number Coastal Juvenile E Eggs sizes hatch dates numbers Spawners ages sizes dates fecundit ies £ '^W sexes GENOTYPIC VALUES spawning date egg size development rate* Fry- -Subadult growth maturation threshold . . . . . . . . . . . . ages sizes dates sexes numbers NERITIC ZONE Pelagic Juvenile remain al sea Return Migration t iming .4 duration — PELAGIC ZONE Coastal Adult \ return to s p a w n Final Sea Year growth survival Size maturation threshold Reproductive Adult Figure 3 . A model of chum salmon l i f e history showing the interaction between selection (S), genotype (C) and environment ( E ) . 28 Among chum salmon populations there are differences in the average length of spawners, weight at age, age at maturity, fecundity and external morphology. According to Sano (1966) spawners in more northernly populations tend to mature at older ages than those to the south. In Japan spawners from the southern r i v e r s are larger and heavier than those of the same age spawning i n the northern r i v e r s (Sano 1966). The absolute fecundity also d i f f e r s among populations along l a t i t u d i n a l gradients. More southerly populations had higher absolute fecundity than northerly populations (Sano 1966). However, Beacham (1982) demonstrated that northern populations had higher fecundity at a given s i z e than southern populations. Birman (1956) found s i g n i f i c a n t differences in body depth of chum salmon from the B i r a and Issuri Rivers as compared to chum salmon from the Amgun River and the Amur River estuary. Svetovidova (1961) found external morphological differences in trunk length, head length, diameter of the eye, and body depth between populations spawning in d i f f e r e n t t r i b u t a r i e s of the Amur River. F i n a l l y , a number of stocks may more or l e s s function as a g e n e t i c a l l y cohesive u n i t . Such a group may be characterized as having a c e r t a i n geographic d i s t r i b u t i o n or return to spawn at a p a r t i c u l a r time of the year. For example, Okazaki (1981) on the basis of an electrophoretic analysis of polymorphic enzymes characterized North American chum salmon as being composed of several 'regional populations'. These consisted of a number of populations i n a broad geographic area. Another example of t h i s i s the phenomenon of seasonal races. These d i f f e r in l o c a t i o n of spawning, time of spawning migration,, time of spawning, s i z e at maturity, morphological features, 29 distance migrated from the sea, average weight of spawners, age at maturity, and absolute fecundity. In Asia, the summer chum salmon spawns during August and September in more northern locations such as Kamchatka, the northern coast of the Okhotsk Sea and the Amur River area. The autumn chum salmon spawns in more southerly locations including the Amur River, Sakhalin, and northern Japan during September to the end of November. Autumn chum salmon usually weigh above 3.5 kg as compared to less than 2.5 kg for summer chum salmon. Autumn chum salmon carry 500 to 1000 more eggs on average than the summer chum salmon. In the Amur River the autumn chum salmon migrates to t r i b u t a r i e s 1000-2000 km from the sea whereas the summer chum salmon migrate to t r i b u t a r i e s l e s s than 100 km from the sea (Sano 1966). Grigo (1953) compared external morphological features of summer and autumn chum salmon. Summer chum salmon had r e l a t i v e l y deeper bodies, shorter pectoral f i n s and deeper heads compared to autumn chum salmon. 30 THE DISTRIBUTION AND ABUNDANCE OF SEASONAL RACES ANC SPECULATIONS REGARDING THEIR EVOLUTIONARY ORIGIN REVIEW I s h a l l present a s e l e c t i v e review to show that although h i s t o r i c a l methods of c l a s s i f i c a t i o n must be revised, seasonal ecotypes are a widespread and common feature among anadromous fishes and chum salmon in p a r t i c u l a r . I s h a l l also review hypotheses regarding the evolutionary o r i g i n of seasonal races. The habitat s h i f t hypothesis proposed by White (1978) for seasonally separated insect populations i s presented as a convincing theory for the o r i g i n of seasonal races. The e a r l i e s t attempt at coalescing the d i f f u s e l i t e r a t u r e on i n t r a - s p e c i f i c seasonal groupings in fishes was made in 1934 by the Soviet icht h y o l o g i s t and geographer, L.S. Berg. Berg (1934), using cereal grains as an analogy, c l a s s i f i e d seasonal races into two groups: "vernal" and "heimal" races. According to Berg (1934) i n d i v i d u a l s of the heimal race are larger, more fecund, spawn further upstream and enter freshwater up to a year p r i o r to spawning. Individuals of the vernal race were smaller, le s s fecund, spawned shortly a f t e r entering freshwater and close r to the mouth of the r i v e r than t h e i r heimal counterparts. Subsequent observations indicate that f i s h exhibit almost every conceivable combination of season of a r r i v a l , season of spawning, fecundity adaptation and length of stay in freshwater before spawning (Ricker 1959). A r i g i d c l a s s i f i c a t i o n of two opposing phenotypes as proposed by Berg 31 (1934) i s not adequate to encompass the variety seen in nature. Ricker (1959) recommends that no attempt be made to force a l l known stocks of anadromous fishes into the vernal or heimal category. H i s t o r i c a l l y , i n t r a s p e c i f i c seasonal migratory groupings of fishes have been c a l l e d 'races'. This term has problems associated with i t . F i r s t l y , i t implies reproductive i s o l a t i o n of one large grouping (many l o c a l populations) within a species from another such grouping. However, such reproductively is o l a t e d groupings might as well be c a l l e d s i b l i n g species. In t h i s study I am interested in whether seasonal groupings in a l o c a l i z e d area represent reproductively i s o l a t e d populations or demes within a species and whether such i s o l a t i o n r e s u l t s i n a genetic change. I believe the most s a t i s f a c t o r y approach to c l a s s i f i c a t i o n of seasonal groups i s to consider whether the v a r i a b i l i t y i n spawning migration times among populations also r e f l e c t s temporal separation at the time of reproduction. This c l a s s i f i c a t i o n method i s useful from an evolutionary ecology perspective because i t separates seasonal groupings according to the p o t e n t i a l for reproduction i s o l a t i o n and hence genetic divergence among populations. Thus, seasonally d i s t i n c t migrating groups f a l l into two categories: a) populations that migrate at d i f f e r e n t times but do not spawn during d i f f e r e n t seasons (e.g. chinook salmon); b) populations that migrate at d i f f e r e n t times and are also temporally separated during reproduction (e.g. chum salmon). 32 Many species of f i s h have two or more forms that d i f f e r in t h e i r timing of migration. For example, Packer (1972) notes that in the Columbia River C h i n o o k salmon, Oncorhynchus tshawytscha, there are four temporally d i s t i n c t stocks: early spring c h i n o o k s ; late spring c h i n o o k s ; summer c h i n o o k s ; and f a l l chinooks. Other f i s h species where time of migration d i f f e r s but time o f spawning i s not d i s t i n c t are brown trout (Salmo t r u t t a ) , A t l a n t i c salmon (Salmo s a l a r ) (Berg 1934), sockeye salmon (Oncorhynchus nerka) (Altukov 1981), Ar t i e char (Salvelinus alpinus), beluga sturgeon (Huso huso), sharpnose sturgeon (Acipenser n u d i v e n t r i s ) , s t e r l e t (Acipenser ruthenus), cut-tooth minnow (Rutilus f r i s i i ) , Aral barbel (Barbus brachycephalus), Shemaia (Chalcalbumus chalcoides) and chekon (Pelecus cultranus) (Berg 1959). Populations within species that spawn during d i f f e r e n t seasons have been recorded for a wide variety of fishes such as chum salmon (Smirnov 1975), pink salmon (Oncorhynchus gorbuscha) (Ivankov 1967), sockeye salmon (Brannon 1984), chinook salmon (Burger et a l . 1985), A t l a n t i c salmon (Neave 1949) Windermere charr (Frost 1966), s t e l l a t e sturgeon (Acipenser guldenstadii), r i v e r lamprey (Lampetra a y r e s i ) , Bream (Abramis brama), carp (Cyprinus c a r p i o ) , osman (Diptychus dybowskii), sander (Lucioperca lucioperca) (Berg 1959), volba (Rutilus r u t i l u s ) , A t l a n t i c herring (Clupea harenqus) (Norman 1975), and redbelly dace, (Chrosomus eos) (Tyler 1966). This l i s t by no means encompasses a l l species that have seasonally a l l o c h r o n i c populations but does show that t h i s phenomenon e x i s t s i n widely divergent taxa such as the Petromyzontidae, Clupeidae, Salmonidae, Cyprinidae and Percidae. 33 Seasonally i s o l a t e d populations may occur throughout the species range. Chum salmon spawns in streams around the north P a c i f i c Rim. There are seasonally d i s t i n c t spawning populations in Japan (Ricker 1972), U.S.S.R. (Berg 1959, Smirnov 1975, Altukov 1981), Alaska (Buklis and Barton 1984), B r i t i s h Columbia (Neave 1966), and in Puget Sound, Washington (Koski 1975). There are no reports of seasonal demes in Oregon. Berg (1934) noted that there were two types of chum salmon, summer and autumn, in Siberian and Kamchatkan r i v e r s . The summer form, which Berg (1934) designated infraspecies fJ. keta keta, spawned in the Amur and Anadyr Rivers mainly during mid-July to early September whereas the autumn form, infraspecies f l . keta autumnalis, spawned during early September to early November. A s i m i l a r timing difference i s observed among stocks u t i l i z i n g the Yukon River in Alaska (Buklis 1982; Buklis and Barton 1984). According to Koski (1975) many of the r i v e r s of northern B r i t i s h Columbia and southeastern Alaska have stocks that spawn i n July and August but the majority of stocks spawn from September to November (Neave 1966, Atkinson et a l . 1967). Two runs of chum salmon, 'early' and ' l a t e ' , spawn in the streams within Puget Sound, Washington (Koski 1975). The early run, as t y p i f i e d by Koski's data from Big Beef Creek, Washington, spawned from the f i r s t week of September to the middle of October. The la t e run spawned from mid to la t e October to the beginning of January. The difference in spawning time among a l l o c h r o n i c chum salmon populations appears to be around seven to eight weeks which i s comparable to the recorded differences among early and la t e stocks i n the Vancouver Island -southern B r i t i s h Columbia mainland area and in Hokkaido (Ricker 1972). Early 34 and late runs in the high l a t i t u d e stocks in the Yukon R. and the U.S.S.R. are, in months, associated with summer and autumn whereas the southern stocks spawn in autumn and winter months. I believe t h i s difference to be an i l l u s i o n because 'summer and autumn month' stream temperatures in high l a t i t u d e s are comparable to 'autumn to winter month' stream temperatures in the mid-latitudes (Appendix 17). Explanations of the evolutionary o r i g i n s of seasonal demes in fishes have focussed on e c o l o g i c a l or p h y s i o l o g i c a l factors a f f e c t i n g the return migration of the adults. Although Berg (1934) believed i t u n l i k e l y that a summer chum salmon could produce an autumn chum salmon and vice versa, subsequent authors ignored genetics by considering seasonal races to be a consequence of the energetics of adult migration or feeding. They assume that the v a r i a t i o n i n migration timing i s caused by environmental factors rather than genetic divergence. Several hypotheses have been proposed regarding the evolutionary o r i g i n of seasonal demes. Schmidt (1947) hypothesized that seasonal demes in fishes resulted from the successive lengthening and shortening of r i v e r courses at the end of the l a s t ice age. According to Schmidt (1947) fishes o r i g i n a l l y developed t h e i r anadromy and reproductive strategy of homing in r i v e r s a short distance from the sea. With the growth of the g l a c i e r s these r i v e r s were p e r i o d i c a l l y incorporated into much longer systems such as the Paleo-Yukon and the Paleo-Amur. Schmidt (1947) proposed that two s t r a t e g i e s evolved within many species: (a) some f i s h retained t h e i r s i t e f i d e l i t y and migrated the 35 length of the longer Paleozoic r i v e r systems to spawn in t h e i r t r a d i t i o n a l gravel beds; (b) some f i s h spawned in the lower reaches of the lengthened r i v e r s . Each group developed adaptations i n accordance with t h e i r mode of reproduction. Thus, Schmidt (1947) argues that since the upper r i v e r spawners could not complete the journey i n one season before the r i v e r became impassible they evolved to enter freshwater up to a year p r i o r to spawning. To o f f s e t the energetic costs of the journey and the increased r i s k s of adult freshwater mortality, these f i s h became larger and more fecund than the fishes spawning i n the lower reaches. As the g l a c i e r s receded and the r i v e r s shrank, the two groups came to spawn more clo s e l y together. Each group retained i t s season of reproduction and c h a r a c t e r i s t i c spawner siz e and fecundity. A t l a n t i c salmon populations i n the Soviet Union have provided a good model for t h i s hypothesis (Berg 1959). Unfortunately, f i s h e s with seasonal races do not always f i t the phenotypic requirements of t h i s model. In P a c i f i c salmon there i s no record of l a t e season spawners remaining in the r i v e r u n t i l the next year to spawn. Late season chum salmon spawners generally are larger and more fecund than f i s h of the early spawning stock i n a system (Koski 1975, Berg 1934, Ricker 1972, Krogius et a l . 1934, Kubo 1950) but often they spawn in the same locale rather than farther upstream. Ivankov (1967) notes that in the Sakhalin River and Iturup Island the summer pink salmon spawns upstream from the autumn pink salmon. A reversal of the migration pattern among seasonally separated spawning stocks also occurs in the sockeye salmon of the Kamchatka River (Birman 1981) and the chinook salmon (Healey 1983). It appears that Schmidt 36 (1947) was determined to f i t an explanation of the o r i g i n of seasonal races into Berg's (1934) model. By r e s t r i c t i n g himself he f a i l e d to explain the d i v e r s i t y observed in nature. Abakumov (1961) proposed that seasonal demes resulted from repeated intrusions and withdrawals of the sea at the end of the l a s t g l a c i a l period. This hypothesis i s s i m i l a r to that of Schmidt (1947) since the intrusions and regressions of the sea led the shortening and lengthened of r i v e r drainages. However, i t d i f f e r s i n that the r i v e r s were lengthening or shortened by changes in sea l e v e l . The process would have been more gradual than that described by Schmidt. I think i t less probable that two groups could form by t h i s method since a gradual change in spawning grounds could be "tracked" by the f i s h . Schmidt's hypothesis, for a l l i t s f a u l t s , suggests a rapid s e l e c t i v e event had occurred. This i s consistant with ideas proposed in other studies of .salmonid evolution such as those of Grotnes (1980) and Quinn (1985). As well, current models such as the "Punctuated Equilibrium" model of evolution (Stanley 1979, Gould 1982) propose that s p l i t t i n g of species occurs over r e l a t i v e l y short periods of time. Birman (1981) related the formation of seasonal demes in pink and chum salmon in Asia to the conditions of t h e i r marine period of l i f e and t h e i r oceanic migrations. The summer form u t i l i z e s ocean feeding grounds, such as the Sea of Japan, that are protected from cold ocean currents. Birman (1981) proposed that the spawning timing difference occurred because ocean temperatures on the feeding grounds exceeded the optimum for e f f e c i e n t growth 37 thereby forcing the summer form to migrate e a r l i e r . Temperatures outside t h i s area allow the other stock to remain at sea longer. Presumably during e a r l i e r epochs ocean temperatures experienced by the two forms were more s i m i l a r than today and there were no differences in spawning time. There are problems with t h i s hypothesis. Birman's (1981) hypothesis f a i l s to explain why there are seasonal races of chum salmon in the North America. There i s no thermal discontinuity l i k e the Sea of Japan i n the ocean range of the U.S.A. and Canadian chum salmon. Birman (1981) ac t u a l l y states that there are no seasonally d i s t i n c t runs of chum salmon in the Yukon River. Secondly, since the summer and autumn keta are i s o l a t e d by season of reproduction I think i t l i k e l y that some degree of adaptation to the temperature in the Sea of Japan would occur so that the f i s h could stay longer at sea. F i n a l l y , Birman's (1981) hypothesis assumes the two groups are g e n e t i c a l l y i d e n t i c a l . F i n a l l y , there i s the case of the Windermere charr, Salvelinus w i l l u g h b i i (Frost 1966). In t h i s case there are two seasonally i s o l a t e d populations of charr that spawn in spring and autumn that d i f f e r in t h e i r pattern of scale growth and number of g i l l rakers (Frost 1966). Frost (1966) reared both forms under the same conditions and determined that season of spawning was under environmental and not genetic c o n t r o l . She concluded, therefore, that these were merely two forms of the same animal and differences occurred due to differences in developmental environment. However, Bush (1975) does not appear to agree as he l i s t s the Windermere charr as a case of sympatric 38 speciation. More recently, C h i l d (1980) found that Windermere spring and autumn charr could be separated by esterase gene frequencies. By comparing the gene frequencies to other populations C h i l d (1980) concluded that the two forms had evolved t h e i r breeding seasons in i s o l a t i o n from each other and then were introduced into Lake Windermere. Frost (1966) proposed t h i s scenario as a p o s s i b i l i t y also, but did not rule out that the divergence could have occurred within the lake basin although s t i l l as a result of a geographic b a r r i e r . Blaxter (1958) advanced a s i m i l a r idea of geographic separation for the o r i g i n of spring and autumn spawning populations of herring. He proposed that spawning season s h i f t e d from spring to autumn in one or more populations during a period of geographic i s o l a t i o n . Frost (1966) advanced two other p o s s i b i l i t i e s , both of which suppose sympatric divergence of the stocks. She supposed that breeding time of the charr was continuous from la t e autumn to early spring and at any depth. "Then some advantage favored s u r v i v a l of those charr which spawned early in the season in shallow water and another advantage favored the s u r v i v a l of those charr which spawned l a t e s t i n the season and in deep water. The two sets of s e l e c t i o n pressures, one working towards autumn and shallow-water spawning the other towards spring and deep-water spawning, would mean s e l e c t i o n against those charr which spawned intermediately in time and place." Thus, Frost (1966) proposes separation by disruptive s e l e c t i o n . The other hypothesis i s that spawning i s triggerred when daylength i s near 8j h. Frost (1966) speculates that i f some autumn spawning charr matured 39 la t e they might miss t h i s supposed spawning cue and be forced to wait u n t i l a s i m i l a r 8^ h daylight regime would occur again in the spring. Of a l l the hypotheses put forward those of Frost (1966) concerning the Windermere charr are the most palatable. However, as Frost (1966) points out the o r i g i n of seasonal races in fishes must be a matter for speculation since one was not there to record the event f i r s t hand. An actual event of seasonal divergence was noted by White (1978) i n monophagous insects. According to White (1978) seasonal 'races' of insects have come about due to habitat s h i f t s and subsequent adaptation of the seasonal reproductive cycle to the new habitat. For example, the species Rhagoletis pomonella, a monophagous insect, infested only Hawthorn f r u i t s u n t i l 1864 when i t appeared on apples. The emergence period of the apple 'race' i s from June 15 to the end of August with an emergence period about a month before the maturation of apples in the area. The Hawthorn 'race' emerges between August 5 and October 15 approximately a month before maturation of the Hawthorn f r u i t s . Presumably then the apple 'race' arose from the Hawthorn when some insects oviposited on the apple and subsequently survived to reproduce. In general, the hypotheses of the o r i g i n of the seasonal races have focussed on v i s i b l e phenotypic v a r i a b i l i t y among the two groups. The problem with t h i s e c o l o g i c a l approach i s that one cannot e f f e c t i v e l y test the assumption that phenotypic v a r i a t i o n i s either g e n e t i c a l l y or environmentally generated. The most important aspect of seasonal races i s the p o t e n t i a l for 40 genetic i s o l a t i o n of populations coupled with the fact that the o f f s p r i n g are incubated i n dramatically d i f f e r e n t thermal environments. This i s l i k e l y to drive either the phenotypes or genotypes of each seasonally i s o l a t e d pair of stocks in d i f f e r e n t d i r e c t i o n s . The question of genetic divergence among chum salmon seasonal races has been l a r g e l y overlooked. 41 PHENOTYPIC DIFFERENTIATION IN SEASONAL ECOTYPES OF CHUM SALMON (Oncorhynchus keta) INTRODUCTION According to Birman (1981) a l l species of the genus Oncorhynchus have seasonal races. This phenomenon reaches i t s highest development within the genus in the chum salmon, 0. keta (Berg 1934). Phenotypic v a r i a t i o n among chum salmon populations occurs along s p a t i a l and temporal axes. Geographically separated stocks vary in many c h a r a c t e r i s t i c s such as polymorphic enzymes (Okazaki 1981), vertebral number (Kubo 1956) and embryonic development (Smoker 1982). Temporally d i s t i n c t stocks vary in many c h a r a c t e r i s t i c s such as time of a r r i v a l , sexual development, spawning time, distance of migration, length, weight, fecundity, shape, polymorphic enzymes and karyotype (Altukov 1981, Altukov and Salmenkova 1981, Berg 1959, Bakkala 1970, Okazaki 1978, Smirnov 1975, Kulikova 1970). Given the strong c o r r e l a t i o n s observed between phenotypic characters and season of reproduction, one might conclude that these are linked. The l o g i c a l point to follow i s that the phenotype has evolved in response to the needs of the animals spawning at d i f f e r e n t times of the year. Thus, the phenotype should r e l a t e to conditions i n the spawning environment. (Bakkala 1970, Koski 1975, Berg 1934, Ricker 1959, 1972). As well, as a consequence of development i n d i f f e r e n t environments, the phenotype w i l l be altered in a way so that 42 morphologically d i s t i n c t races w i l l e x i s t throughout the l i f e h i s t o r y (Bakkala 1970, Birman 1981). Thus a hypothesis e x i s t s which states that as a consequence of separation i n time of reproduction, the phenotype has evolved in d i f f e r e n t d i r e c t i o n s and so "seasonal races" occupy d i f f e r e n t niches (Saviattova 1983, Packer 1972, Bakkala 1970, Altukov 1981). Since consequence i s d i f f i c u l t to discern using c o r r e l a t i o n s of variables, t h i s hypothesis has also been stated in the reverse fashion. Seasonal differences in spawning time have evolved due to changes i n phenotypic characters and in the niche i n other periods of the l i f e h i s t o r y (Birman 1981, White 1978). The c e n t r a l testable point i s that the differences i n the environment associated with a seasonal s h i f t in the ecology of the animal are linked c l o s e l y to differences in phenotype. Thus one can d i s t i n c t l y i d e n t i f y a summer race from an autumn race by phenotype. Incubation temperature can p o t e n t i a l l y a f f e c t a l l developing c h a r a c t e r i s t i c s . Populations reproducing during d i f f e r e n t seasons w i l l experience quite d i f f e r e n t thermal environments in incubation. However, s t a b i l i z i n g s e l e c t i o n , or the pressure in the environment for the conservation of form and function of a t r a i t , w i l l encourage genotypes that can compensate for incubation environment variance to produce the optimum phenotype. Compensatory adaptive responses among allochronic populations w i l l be most l i k e l y i n t r a i t s where phenotypic response varies with temperature. Incubation rate to hatch or emergence (Godin 1982), vertebral number (Seymour 1959, Taning 1950), and external morphology (Barlow 1961, Fowler 1970) are a l l 43 known to be altered by temperature of incubation. These t r a i t s are appropriate variates as a l l have been correlated to i n d i v i d u a l f i t n e s s in fishes (Godin 1982, Swain and Lindsey 1984, R i d d e l l and Leggett 1981). On the other hand, the c o r r e l a t i o n s observed between spawning season and phenotype have been confounded by other factors such as geographic separation. For example, not only do the summer and autumn chum salmon of the Amur River c l e a r l y d i f f e r .in body s i z e , egg s i z e , fecundity, and age at return, but autumn chum salmon also spawn 100 miles further upstream. Thus, geographic i s o l a t i o n may be the explanation for the differences in phenotype. S i m i l a r l y , data used by Brannon (1981, 1982) and M i l l e r and Brannon (1983) to investigate run timing and r e s u l t i n g phenotypic differences among salmonid populations suff e r from a confounding geographic component. In other cases the proposed l i n k between environment, season of spawning and phenotype has not been c l e a r l y tested. Authors such as Bakkala (1970), Birman (1981) and Berg (1934) have used environmental data from one system and phenotypic data from another to attempt to form a casual l i n k . In studies such as Beacham (1984, 1987), Beacham and Murray (1986a, 1986b, 1987a, 1987b) and Beacham et a l . (1987) with the p o t e n t i a l to investigate the s i t u a t i o n on a fine scale without confounding due to geographic factors, the environmental data was either not c o l l e c t e d , c o l l e c t e d in such a s u p e r f i c i a l manner as to be meaningless, or assumed to follow a pattern observed in neighboring systems. Further, in these studies no e f f o r t was made to investigate i n d e t a i l the temporal and s p a t i a l d i s t r i b u t i o n of the spawners to see i f populations were 44 t r u l y temporally i s o l a t e d ( i . e . were separate populations) or i f there were geographic b a r r i e r s between groups that they sampled. To quote Murray and Beacham (1987): " D i f f e r e n t trends i n embryo and a l e v i n developmental characters among f a m i l i e s w i t h i n a species were assumed to r e f l e c t adaptations to v a r i a b l e n a t u r a l incubation c o n d i t i o n s " . In general l i t t l e e f f o r t has been a l l o t t e d to i n v e s t i g a t i o n s of seasonal groupings to l i n k , on a f i n e s c a l e , the environments, timing of reproduction and phenotypic t r a i t s . An important exception i s the work of Koski (1975) at Big Beef Creek, Washington. Koski (1975) was able to compare a wide v a r i e t y of phenotypic t r a i t s of two temporally d i s t i n c t "runs" of chum salmon that enter B i g Beef Creek each year (There are three runs i n t o t a l but as one of these was small Koski combined i t with another f or a n a l y t i c a l purposes). He presented d e t a i l e d information on the spawning environments, phenotypic c h a r a c t e r i s t i c s of the spawners and f r y . While some c h a r a c t e r i s t i c s of the spawners d i f f e r r e d between the groups, there appeared a s u r p r i s i n g degree of convergence among some key c h a r a c t e r i s t i c s of the o f f s p r i n g . Most i n t e r e s t i n g was the tendency of the f r y spawned l a t e r i n the season to emerge a f t e r fewer accumulated temperature u n i t s than those spawned e a r l i e r i n the year. The s l i g h t convergence i n downstream migration t i m i n g (of the o f f s p r i n g ) seemed par a d o x i c a l given the d i f f e r e n c e i n ti m i n g of spawning and temperature o f inc u b a t i o n . Koski (1975) suggested that s e l e c t i v e facors could o v e r r i d e the obvious d i s c o n t i n u i t i e s i n the environment during i n c u b a t i o n . Unfortunately, Koski's work was not published widely, and, perhaps has not received the r e c o g n i t i o n that i t deserves. 45 One caveate to h i s work provided by Koski (1975) i s that the d i f f e r e n c e s i n spawning timing at Big Beef Creek may be more a r e s u l t of recent d i s r u p t i v e s e l e c t i o n by an intense Puget Sound f i s h e r y . Two or three separate runs appear i n Big Beef Creek because the f i s h e r y removes the f i s h at the centre run. The purpose of t h i s s e c t i o n i s to t e s t the idea that phenotype and season of spawning are c o r r e l a t e d and that "seasonal races" occupy d i f f e r e n t niches. Ricker (1972) i n d i c a t e s that g e n e t i c a l l y d i s t i n c t geographic races of P a c i f i c Salmon are a widely accepted phenomenon. In c o n t r a s t , a genetic b a s i s for "seasonal races" i s mainly s p e c u l a t i v e i n nature. I propose the hypothesis that the reduction of gene flow caused by temporal i s o l a t i o n w i l l allow genetic divergence to occur among populations. Divergence w i l l occur because s t a b i l i z i n g s e l e c t i o n f o r optimum phenotype w i l l favor genotypes which can compensate f o r the temperature d i f f e r e n c e s experienced by the a l l o c h r o n i c populations. In t h i s chapter I compare the reproductive environments and phenotypic c h a r a c t e r i s t i c s of s p a t i a l l y and temporally i s o l a t e d p o pulations. MATERIALS AND METHODS 46 Study area I confined my investigations to three chum salmon populations spawning in two adjacent creeks on Vancouver Island. I assume that the chum salmon populations in these creeks are representative of other chum salmon populations where there are run timing differences. As well, I propose that the general p r i n c i p l e being investigated applies to a l l a l l o c h r o n i c ectotherm populations. Bush and Walker creeks are small coastal streams that enter Ladysmith Harbor (B.C.) less than a km apart at roughly 49°N and 123°W (Figure 4). The surrounding vegetation i s t y p i c a l protected temperate coast forest with the dominant species being Douglas F i r , Pseudotsuga m e z i e s i i , Western Red Cedar, Thuja p l i c a t a , Broadleaf Maple, Acer macrophyllum, Hemlock, Tsuga  heterophylla, and Alder, Alnus rubra (Hitchcock and Cronquist 1973). The t o t a l stream bed a v a i l a b l e to Bush Creek spawners i s approximately 8000 m2 and to Walker Creek spawners i s 3000 m2. Their chum salmon populations are accessible, spawn a l i m i t e d distance from the mouth of the stream (1.5 - 2.0 km) and have an approximate two month separation in the s t a r t , peak and end of reproduction according to t h i r t y years of Canadian Dept. of F i s h e r i e s and Oceans, (DF0) escapement records (1950 - 1980) (Appendix 18). Records from 1969 to 1981 were s u f f i c i e n t l y d e t a i l e d to calculate means and variances for spawning timing. The year to year v a r i a b i l i t y in run timing was between (+) or (-) 2.5 and 4.0 days (Table 1). 4. Bush and Walker creeks: the study area 48 Adult c h a r a c t e r i s t i c s To determine temporal and s p a t i a l d i s t r i b u t i o n of spawners, observers walked the creeks three to f i v e days per week from September to February in 1981-82, 1982-83, and 1983-84. Two temporally d i s t i n c t spawning populations were observed in Bush Creek (Designated "Autumn Bush" or "AB" and "Winter Bush" or "WB" hereafter) while a late run was found i n Walker Creek (Designated "Walker" or "W"). These circumstances allowed comparisons of phentic c h a r a c t e r i s t i c s among the three populations along s p a t i a l and temporal axes (see Table 2). Chum salmon spawners have a s p a t i a l l y contiguous d i s t r i b u t i o n i n Bush Creek. Late a r r i v i n g spawners tend to migrate further upstream than those a r r i v i n g e a r l i e r but there can be a large amount of s p a t i a l overlap. For example in 1983-84 there was a great amount of s p a t i a l overlap among the Bush Creek populations. However, in 1981-82 and 1982-83 the late spawning population spawned in the upstream section of the spawning area while the early f i s h spawned in the sections below. This was convenient because I could sample the migrating f r y from the l a t e population without contamination from the progeny of the early population. However, the s p a t i a l distance between the two reproductive populations was less than 25 meters, a distance comparable to the distance separating c l u s t e r s of spawners within e i t h e r population (Figure 5). Chum salmon are thought to die a short time after entering freshwater in small coastal streams (Koski 1975, Smirnov 1975 and Salo 1986). 49 Table 1. Mean dates o f s t a r t , peak and end o f Bush and Walker creek chum spawning runs from 1969 to 1981. START PEAK END BUSH CREEK OCT 1+/-3.5 OCT 28+/-4.5 NOV 20+/-3.0 WALKER CREEK NOV 16+/-3.5 DEC 10+/-2.5 JAN 1+/-4.0 Table 2. Comparisons between pop u l a t i o n s along s p a t i a l and temporal axes. AB=Autumn Bush P o p u l a t i o n ; WB=Winter Bush P o p u l a t i o n ; W=Walker Po p u l a t i o n . SPATIAL SEPARATION YES NO YES AB-W AB-WB TEMPORAL SEPARATION NO WB-W WITHIN DIFFERENCES 50 FREQUENCY 4 0 - i 12 2 4 3 6 4 8 6 0 7 2 8 4 D I S T A N C E BETWEEN GROUPS ( m ) Figure 5. Distance between spawning groups of chum salmon in Bush and Walker Creeks 51 To check my assumption that spawning time could be reasonably estimated by the temporal d i s t r i b u t i o n of adult abundance, I calculated freshwater s u r v i v a l time of adults. The dead were counted and pitched from the stream. Since a l l reproducing animals die the temporal pattern of l i v e adult abundance w i l l be followed by a s i m i l a r temporal pattern of pitched dead adult abundance. The phase lag between the two curves w i l l represent the duration of freshwater l i f e of spawners. The cumulative frequencies by sampling day were then computed for both l i v e and dead counts. As a much smaller proportion of the dead were located, conversion to pe r c e n t i l e values was necessary to provide equal s c a l i n g for the l i v e and dead counts. Probit analysis was used to determine the appropriate l i n e a r regression for each stock. The mean difference between the transformed l i v e and dead count l i n e s estimates freshwater residence time. The age at maturity varies from 0.1 to 0.6 (Bakkala 1970). Three systems have been used to record the ages of P a c i f i c salmon: G i l b e r t and Rich (1927); Chugunova (1959); and Koo (1962). The Gilbe r t and Rich method has been most widely used i n North America. This method records age from the time of egg deposition. The other methods record age from time of hatching. In the Gil b e r t and Rich method a large arabic numeral represents the t o t a l age of the f i s h while a subscript records the years spent i n freshwater. Therefore, a 52 f i s h has spent two years in freshwater including incubation time and three years i n saltwater. The European system formalized by Koo (1962) uses two numerals separated by a decimal point recording the number of freshwater and saltwater annuli, r e s p e c t i v e l y . Since each annulus i s a c l o s e l y separated set 52 of growth rings on the scale during the slow growth of winter, the f i r s t number represents the number of winters i n freshwater whereas the second number represents the number of winters spent i n saltwater. Thus, a 52 f i s h using the Gilbe r t and Rich method would be a 1.3 f i s h using the European system. The Soviets add a + to the age of mature salmon to indicate that the f i s h has undergone an addit i o n a l summer of growth beyond the l a s t annulus. In t h i s report I have used the European method throughout. Chum salmon never have a freshwater annulus and so I have dropped the zero representing freshwater age when reporting mean age at maturity. There i s a l a t i t u d i n a l c l i n e in age at maturity such that more northerly populations consist of older spawners (Bakkala 1970). For example, average age of spawners ranges from 3.23 years in the Yukon River to 3.17 years in B r i t i s h Columbia to 2.48 i n Bellingham, Washington (G i l b e r t 1922, Pritchard 1943). Northern stocks grow more slowly than southern stocks (Sano 1966). It i s uncertain whether the trend in age at maturity r e f l e c t s environmentally induced slower growth to threshold s i z e for reproduction in the more northerly stocks or a genetic response of the more southerly stocks to reduce age at maturity to o f f s e t higher mortality. P a c i f i c salmon undergo a dramatic metamorphosis during t h e i r reproductive phase. Secondary sexual c h a r a c t e r i s t i c s are most obviously expressed in s t r i k i n g changes i n the external morphology. These changes are more a res u l t of sexual s e l e c t i v e processes within stocks rather than a s t a b i l i z i n g e f f e c t from a common s e l e c t i v e environment. The rapid flux in the morphological 53 features as the animal i s being transformed made comparisons among the stocks using a wide range of morphological c h a r a c t e r i s t i c s u n r e l i a b l e . However, orbit-hypural length remains r e l a t i v e l y constant even af t e r transformation to the reproductive morph. Size was p o s i t i v e l y correlated with f i t n e s s i n Oncorhynchus kisutch by Van den Berghe and Gross (1984). E n e r g e t i c a l l y , larger s i z e may be adaptive in the l a t e r season when water flows are high. Contrastingly, when flows are low as in summer or early autumn larger f i s h may be at a disadvantage because i t i s more d i f f i c u l t for them to navigate the shallow creek. Such a r e l a t i o n s h i p appears to be consistent with a l l published reports of length differences among seasonally d i s t i n c t stocks of chum salmon (Koski 1975, Smirnov 1975, Berg 1934, Ricker 1972, Kubo 1956). Lengths were measured on carcasses recovered in the stream at regular i n t e r v a l s throughout the duration of the runs. The number of adults sampled at each i n t e r v a l was proportional to the abundance of spawners present. Vertebral counts also remain constant throughout the animal's l i f e . Kubo (1956) found that early and l a t e spawning stocks d i f f e r r e d in vertebral count. To compare the mean and variance in vertebral number among the populations, counts were made from adult carcasses recovered in 1982. To make counts from the adults, the fle s h was stripped away from the backbone with a k n i f e . Samples were s t r a t i f i e d r e l a t i v e to the duration of the runs. Where 54 possible, samples of the adults were made at time i n t e r v a l s corresponding to the f i r s t , second and t h i r d q u a r t i l e s of the run. Comparisons of interpopulation v a r i a b i l i t y in vertebral counts are subject to noise from many other factors. For example, according to Lindsey (1975), vertebral number i s p o s i t i v e l y correlated with larger si z e within and among fishes. Thus, I s t r a t i f i e d my sampling of vertebral counts by the orbit-hypural length of the f i s h . Chum spawners return mainly as 0.2 and 0.3 year-olds with a small proportion returning as 0.1 and 0.4 year-olds. This means that spawning populations are composed of f i s h that were born during several d i f f e r e n t years. Year to year c l i m a t i c f l uctuations would re s u l t in a d i f f e r e n t thermal incubation regime for each year c l a s s . Even i f cohorts are g e n e t i c a l l y i d e n t i c a l with respect to a vertebral number program, vertebral count v a r i a t i o n could occur among age groups. Thus, vertebral number sampling was also s t r a t i f i e d by age of the spawner. Length of adults i s related to sex and age in most f i s h species and so I s t r a t i f i e d my population sampling of length by these v a r i a b l e s . Counting the number of annuli on scales i s the least time consuming and simplest method of ageing fishes (Chilton and Beamish 1982). Scale ageing in P a c i f i c salmon may be biased due to resorption of the outer scale edges during spawning. Helle (1979) found that scale resorption did not preclude use of scales for age determination among chum salmon spawning in Olsen Creek, Alaska. To reassure myself that Helle's (1979) r e s u l t s are u n i v e r s a l l y true 55 for spawning chum salmon, I checked 70% of scale sample ages with age estimations from o t o l i t h c o l l e c t i o n s . O t o l i t h s are more d i f f i c u l t and time consuming to sample but are generally believed to be unaffected by the reproductive metabolic transformation which occurs in salmon (Calaprice 1969). Agreement between scales and o t o l i t h s was 84%. To compare vari a t i o n s i n sex, age and length of spawners entering Bush and Walker creeks, sex, s c a l e / o t o l i t h age and orbit-hypural plate length were recorded on recovered carcasses in 1981 and 1982. Progeny c h a r a c t e r i s t i c s To determine the timing of the downstream migration of f r y , i n c l i n e d p l a i n traps were set overnight below the spawning areas one to seven times per week, from February to July in 1982 and 1983. Once each week in 1983 the length of time for f r y to t r a v e l from the spawning area to the trap was estimated by releasing marked fry at the head of the spawning area. Diurnal samples were also occasionally taken to estimate the proportion of daytime migrants. Differences in the mean and variance of vertebral counts of progeny were compared from x-rays of fry samples captured by i n c l i n e d p l a i n traps during the 1982 and 1983 fry migrations. In 1982 fry were sampled at i n t e r v a l s corresponding to the approximate 17th, 33rd, 50th, 67th, and 83rd p e r c e n t i l e s of the cumulative frequency d i s t r i b u t i o n of i n d i v i d u a l s with time. In 1983 56 fry were sampled at i n t e r v a l s corresponding to the approximate 25th, 50th and 75th pe r c e n t i l e s of the cumulative frequency d i s t r i b u t i o n . To compare morphology of the o f f s p r i n g , the downstream migrating progeny were sampled using i n c l i n e d plane traps during February to July in 1982 and 1983. Total length, standard length, head length, eye diameter, snout length, pectoral f i n length, body depth, head depth (to the nearest 0.1 mm), weight (to the nearest 0.1 g), and the number of parr marks were measured for comparison of morphology. Incubation rate Average population incubation rate may be estimated in the f i e l d i f one knows: (a) when the eggs are spawned; (b) when the f r y have completed development. If one knows the temperature during incubation then i t i s possible to predict the time to emergence and compare i t to the actual time to emergence. Chum salmon are p a r t i c u l a r l y s u i t a b l e for estimation of these va r i a b l e s . The adults spawn shor t l y a f t e r entering freshwater in coastal streams (Neave 1953). According to Schroder (1981) spawning occurs within 30 hours af t e r t e r r i t o r y formation. Chum salmon in Bush and Walker Creeks form t e r r i t o r i e s a f t e r one to two days of freshwater residence. Thus, escapement timing may be roughly considered as spawning timing. The f r y migrate to sea aft e r completing embryonic development so timing of downstream migration i s a marker for the end of embryonic development. 57 Although chum salmon fry generally migrate downstream immediately af t e r completing development, t h i s does not occur in a l l systems (Mason 1 9 7 4 ) . Another way to compare incubation rate would be to take samples of the embryos of each population at several times during the run and compare t h e i r stage of development. Providing only a small portion of the t o t a l embryo population was sampled, I should have two independent measures of incubation rate differences among the populations. To compare the pre-hatch incubation rates, samples of eggs were c o l l e c t e d from several redds at i n t e r v a l s during the 1981-82 incubation period. Sampling af t e r hatch was largely unsuccessful probably due to the a b i l i t y of the alevins to avoid disturbances ( D i l l and Northcote 1969). Sampling was unsuccessful i n WB due to sparcely situated redds and high water l e v e l s . Also, a considerable number of coho salmon spawned i n upper reaches and i t was d i f f i c u l t to be c e r t a i n i f eggs sampled were coho or chum salmon. Such sampling proved c o s t l y in terms of the number of embryos destroyed i n c i d e n t a l l y and so was not continued during the 1982-83 season. Phenotype can be influenced by many environmental factors but the dominant variable influencing incubation time i s temperature (Murray 1980). The importance of temperature i s enhanced among populations reproducing during d i f f e r e n t seasons. The next most important factor a f f e c t i n g speed of incubation i s dissolved oxygen which i s a function of flow in streams (D. Alderdice pers. comm.). Therefore to avoid unnecessary complexity I limited my comparison of incubation environments to measurements of column 58 temperature, intragravel temperature, flow and dissolved O 2 . The flow and temperature were compared using spot checks every one to two days in 1981. In 1982 and 1983 temperature was measured using continuous temperature recorders. Intragravel temperature was measured continuously near the recorder and by spot check within redds. Redd depth may range from 21.5 cm (Helle 1979) to 41 cm (Neave 1966). Intragravel measurements were taken between 20 and 40 cm into the gravel at known redd s i t e s . Flow measurements were made using the wier method described by Hynes (1976). Dissolved Oxygen was measured by spot checks in the redds. According to Murray (1980) the r e l a t i o n s h i p between time to emergence and temperature i s negative exponential rather than hyperbolic. Temperature of incubation may d i f f e r greatly among temporally separated populations. Therefore I used Murray's (1980) model of the r e l a t i o n s h i p between temperature and incubation rate to standardize comparisons among the populations: (1 ) In E - 5.603 - 0.097 T Where: E i s the time to 50 per cent emergence T i s the temperature (constant) of incubation ( 2) E = e 5 - 6 0 3 " Q - 0 9 7 T Murray (1980) developed the parameter estimates using o f f s p r i n g from a sing l e pair mating of Weaver Creek chum salmon reared in the laboratory at d i f f e r e n t temperatures. Murray's (1980) model was chosen because i t i s conservative regarding the speed of incubation at low temperatures compared 59 with other incubation models (Jensen, In Press). I considered t h i s important because much of the embryonic development of progeny from l a t e spawning adults w i l l occur at low temperatures. To demonstrate the genetic divergence among the early and lat e spawning populations, I estimate the time to emergence in each using the model and the mean temperature of incubation. However, such an estimate does not account for the seasonally varying temperatures experienced by embryos in nature. To standardize comparisons so that both the thermal regime curve and the curved response of incubation rate to temperature were accounted f o r , I simulated the ef f e c t of the incubation rate - temperature r e l a t i o n s h i p on the d a i l y temperature unit accumulations using 6°C as a baseline. The adjusted d a i l y thermal units were summed to give an adjusted t o t a l accumulated thermal units to emergence for each population. The computations used are i l l u s t r a t e d below for a day i n which the incubation temperature was 10°C: E10 (3) 6 E6 Where: NTj i s the thermal units accumulated during day j . Ei i s the emergence time at temperature i . 60 (4) NTj = X 6 6 x e5.603 - 0.097 X 10 NT i _ e5.603 - 0.097 X 6 the adjusted t o t a l thermal units (T.U.s) to emergence. Maternal e f f e c t s represent a source of phenotypic v a r i a b i l i t y that can be both genetic and environmental i n o r i g i n (Van Vleck 1973). Among salmonids maternal e f f e c t s include behavioural differences among females that a f f e c t s u r v i v a l of the eggs such as depth of redd, choice of redd s i t e and longevity and f i d e l i t y in guarding eggs af t e r egglaying (Van den Berghe and Gross 1984) plus energy supplied to the embryo in the form of yolk. Egg s i z e i s p o s i t i v e l y correlated with incubation time (Smoker 1982) and perhaps vertebral number (D. Swain, pers. comm.). Smoker (1982) determined that egg size was an important factor in the generation of d i v e r s i t y i n quantitative t r a i t s of chum salmon. Thus, I measured damp dry weight of 100 water hardened eggs from females from each population in 1982 and 1983. sum of NT, 61 Homing and straying As an index of p o t e n t i a l gene flow among the populations I estimated the amount of straying among the three populations. Fry migrating to the estuary were marked using p e l v i c f i n c l i p s during the spring of 1982. I marked 12,432 fry of the AB populations, 8053 fry of the W population and 2111 fry of the WB population. Marked adults were recovered from the spawning grounds during the f a l l and winter of 1984 and 1985. S t a t i s t i c a l Methods I was uncertain as to the normality of the d i s t r i b u t i o n of spawners and outmigrating progeny with time and therefore chose to use non-parametric measures of l o c a t i o n to describe the timing of the adult and f r y runs. I used Peterson mark-recapture method to estimate the proportion of f r y captured by my traps. Comparisons of phenotypic c h a r a c t e r i s t i c s among the populations were generally made using a Model 1 or mixed model two or three way analysis of variance with i n t e r a c t i o n (Sokal and Rohlf 1981). I considered population to be a fixed e f f e c t because I was most interested in the autumn spawning versus winter spawning comparison. I compared means using the Tukey-Kramer procedure to minimize the p r o b a b i l i t y of making Type 1 Error with unequal sample s i z e s . Results using t h i s method are clo s e r to the intended s i g n i f i c a n c e l e v e l than 62 those of other non-orthogonal tests (Dunnett 1980). Other methods such as the GT2 method are too conservative (Dunnett 1980). Analysis to test the equality of egg weight-female length regressions among populations was performed using BMDP s t a t i s t i c a l software program 1R. Population mean egg weights, adjusted for the length covariate, were compared using the GT2 method (Hochberg 1976). Although I planned the t e s t s beforehand, t h i s method i s appropriate because i t compensates for the pr o b a b i l i t y of Type 1 Error when a l l possible comparisons are made. According to Sokal and Rohlf (1981) the GT2 method i s preferred with unequal sample s i z e s . I chose t h i s method over the Tukey-Kramer method for t h i s case because I was comparing adjusted treatment means and wished to be as conservative as possible. RESULTS Spawner c h a r a c t e r i s t i c s The temporal d i s t r i b u t i o n of spawner abundance in AB, WB and W was sim i l a r in 1981 and 1982 (Figure 6). T y p i c a l l y , spawners arrived i n AB during the l a s t week in September, the median occurred in the t h i r d week of October and spawning ended by mid-November. WB and W spawners arrived i n mid-November. The median of WB spawner abundance peaked at the end of November or early December with most a c t i v i t y f i n i s h e d by mid-December. 63 1000-i 1981 AB 1 875- T 0 750-625- 0 500- 0 0 375-250-CT 125-0 ( oo o 0 LU 1 4 co 1000-875-750-625-500 375-250-125-0 WB W • WALKER A WINTER BUSH o AUTUMN BUSH 1982 A A A A. ID AB OCD TO i i i r * O o CD 0"" O O 00 255 267 279 291 303 315 327 339 351 363 10 22 l | * A ^ f ^ J M J U L I A N D A Y Figure 6. Timing of spawning i n Walker Creek and the upper and lower s e c t i o n of the spawning area i n Bush Creek during 1981 and 1982. Number of l i v e spawners observed versus the J u l i a n day. Median day of each run shown by arrow 64 W spawner abundance occurred in early to mid-December with spawning being completed by lat e December or early January. Variation in the median day of spawner abundance was much greater among populations than among years (Table 3). The maximum year to year v a r i a t i o n of seven days occurred in W. Year to year v a r i a t i o n i n AB was one day while there was no year to year v a r i a t i o n i n WB. Year to year v a r i a t i o n in the modal time of spawner abundance was even less with a maximum two days difference occurring between the 1981 and 1982 WB runs. In contrast the median of the WB run was 25-26 days l a t e r than the peak of the AB run i n 1981 and 1982. Walker Creek adult numbers peaked 37-45 days post AB run (Figure 6). The differences among the early stock (AB) and the lat e stocks (WB and W) were even greater using modal values. The 95 % confidence l i m i t s of s u r v i v a l time i n freshwater for Bush Creek adults were 0.51 to 5.95 days (mean = 3.26) and 1.59 to 4.33 days (mean = 2.96) i n 1981 and 1982, respectively (Table 4). The 95 % confidence l i m i t s of su r v i v a l time for Walker Creek adults were 7.16 to 8.92 days (mean = 8.04) i n 1981 and 5.35 to 6.53 days (mean = 5.94) i n 1982 (Table 4). Temporal overlap among the early and late stocks was s l i g h t with over 95 per cent of the adults of the AB run having spawned and died before f i v e per cent of the W and WB spawners a r r i v e d . I believe that the actual temporal separation may be greater than i t appears because WB pre-spawners migrating to the upstream areas or post-spawning WB f i s h d r i f t i n g back p r i o r to death would be observed in the downstream area and i n c o r r e c t l y c l a s s i f i e d as late spawning AB f i s h . 65 Table 3. The y e a r l y median and mode ( i n J u l i a n day) of the temporal d i s t r i b u t i o n o f a d u l t s on the Bush and Walker creek spawning grounds during 1981 and 1982. POPULATIONS METHOD OF ESTIMATION YEAR AUTUMN BUSH WINTER BUSH WALKER MEDIAN 1981 301 ' 326 338 1982 300 326 345 MODE 1981 298 331 344 1982 299 329 345 Table 4. S u r v i v a l time o f (days) a d u l t s i n freshwater at p e r c e n t i l e s o f the d i s t r i b u t i o n o f counts o f l i v e and dead a d u l t s d u r ing 1981 and 1982. Estimated Freshwater Survival Time Stream Percentile of Count Bush Creek Walker Creek lit h Time 1981 1982 1981 1982 5 8.38 0.36 9.65 4.86 15 7.09 1.01 9.25 5.13 25 5.81 1.66 8.92 5.39 35 4.53 2.31 8.46 5.67 45 3.25 2.96 8.07 5.94 55 1.99 3.61 7.67 6.21 65 0.70 4.26 7.28 6.48 75 -0.59 4.91 6.88 6.75 85 -1.81 5.56 6.19 7.02 66 Three-way analysis of variance (population, sex, age) with i n t e r a c t i o n was performed to compare vertebral count differences among the 1982 spawning populations. S i g n i f i c a n t difference (p < 0.05) among populations occurred (F = 10.95, d.f. - 2, P r .05). Comparisons among means (Tukey-Kramer Method) revealed s i g n i f i c a n t differences among s p a t i a l l y and temporally separated stocks. WB spawners (5? = 63.0) had fewer vertebrae than s p a t i a l l y separated W (X = 65.1) and the temporally separated AB (5< - 64.9) spawners (Table 5). No s i g n i f i c a n t d ifferences were observed among AB and W adults where there was s p a t i a l plus temporal separation among stocks (Table 6). Samples were not s u f f i c i e n t to compare a l l populations including the e f f e c t s of sex, age, and length but comparisons between WB and W samples revealed s i g n i f i c a n t differences i n vertebral counts duetto population e f f e c t s (F = 16.20, d.f. = 1, 113, P = .05). Two-way analysis of variance with i n t e r a c t i o n (population, sex, population by sex interaction) revealed s i g n i f i c a n t differences i n age at return among populations for 1982 (F = 7.00, d.f. = 2, 193, N = 199, P = .05, Mean of AB = 2.83, Mean of WB = 2.39, Mean of W = 2.59). No s i g n i f i c a n t differences i n treatment or i n t e r a c t i o n e f f e c t s were discovered i n the 1981 data (N = 431). Comparisons among means (Tukey-Kramer Method) showed that during 1982 the l a t e stocks combined, WB and W, were comprised of s i g n i f i c a n t l y younger spawners (ft - 2.52 seawater years, N = 138) than the early stock, AB (ft = 2.83 years, N = 59) (Tables 7, 9). No s i g n i f i c a n t differences in the age of return of spawners was found in comparisons where there was a s p a t i a l or s p a t i a l plus temporal separation among stocks. 6 7 Table 5. V e r t e b r a l counts o f 1 9 8 2 spawners : means ( X ) , standard d e v i a t i o n s ( S O ) , and sample s i z e s s t r a t i f i e d by p o p u l a t i o n , sex and age. POPULATION AB WB W SEX AGE X SD n X* SD n X SD n 2 6 5 . 8 3 1 . 1 6 9 6 6 3 . 3 8 2 . 2 7 3 2 4 6 5 . 3 5 1 . 5 9 6 1 4 FEMALE 3 6 3 . 3 3 1 . 5 2 8 3 6 1 . 8 0 2 . 7 6 1 1 5 6 4 . 5 0 1 . 9 4 7 2 3 2 6 5 . 4 2 1 . 4 3 2 6 6 3 . 8 9 3 . 8 7 8 9 6 5 . 1 2 1 . 7 5 9 1 7 MALE 3 6 5 . 0 0 1 . 4 1 0 2 6 2 . 8 5 2 . 7 2 1 8 6 4 . 7 5 1 . 3 9 0 2 0 TOTAL 6 5 . 1 5 1 . 5 1 8 1 7 6 2 . 9 6 2 . 7 8 1 5 6 6 4 . 8 7 1 . 7 0 4 7 4 Table 6. S i n g l e degree o f freedom comparisons o f v e r t e b r a l count d i f f e r e n c e s among populations by the Tukey-Kramer Method. (MSD ^Minimum S i g n i f i c a n t D i f f e r e n c e ) (* s i g n i f i c a n t d i f f e r e n c e P = .05) Difference Among Means Di fference Comparison 1 9 8 2 1 9 8 2 1 9 8 3 Investigated FRY ADULTS FRY * * * Temporal XWB- X"AB 0 . 5 8 ( 0 . 3 7 2 ) - 2 . 1 9 ( 1 . 5 9 1 ) - 2 . 3 9 ( 0 . 2 9 ) Temporal/ X A B- X W 0 . 1 4 ( 0 . 3 7 9 ) 0 . 2 8 ( 1 . 5 4 5 ) * 0 . 6 3 ( 0 . 2 9 ) S p a t i a l * * * S p a t i a l x W B - x"w 0 . 7 2 ( 0 . 3 8 4 ) - 1 . 9 1 ( 1 . 0 1 8 ) - 1 . 7 6 ( 0 . 2 9 ) 68 Table 7. Seawater age at r e t u r n f o r 1981 and .1982 spawners. Means, standard d e v i a t i o n s and sample s i z e s o f populations s t r a t i f i e d by sex. POPULATION AB WB W YEAR SEX X" SD n X" SD n X" SD n FEMALE 2.88 0.426 130 2.87 0.344 23 2.72 0.484 65 1981 MALE 2.79 0.490 135 2.88 0.353 8 2.78 0.508 70 FEMALE 2.96 0.615 30 2.36 0.487 44 2.65 0.538 37 1982 MALE 2.69 0.541 29 2.45 0.510 20 2.54 0.505 37 1981 TOTAL 2.84 0.461 265 2.87 0.341 31 2.76 0.496 .135 1982 TOTAL 2.83 0.592 59 2.39 0.492 64 2.59 0.521 74 69 Three-way analysis of variance (population, sex, age) with i n t e r a c t i o n of length differences revealed s i g n i f i c a n t differences among populations and age groups in 1981. [F(pops) = 4.19, d.f. = 2, 418, P = .05; F(ages) = 19.86, d.f. = 1, 418, P = .05], No s i g n i f i c a n t differences were found due to treatment or i n t e r a c t i o n i n the 1982 comparisons. Comparisons among means (Tukey-Kramer Method) showed that 1981 spawners comprising the la t e stock, WB was s i g n i f i c a n t l y larger (X* = 58.52 cm, N = 31) than the early stock, AB (X - 55.75 cm, N = 265) (Tables 8, 9). No s i g n i f i c a n t differences among stocks were found in comparisons with s p a t i a l or s p a t i a l temporal separation among stocks. Progeny c h a r a c t e r i s t i c s Sampling both streams over a 24 hour period showed over 95 per cent of the fry migrated at night (Figure 7). Ninety-eight per cent of marked fry re-released upstream from the i n c l i n e d p l a i n traps were recaptured the same night suggesting that fry complete the journey to the estuary the same night as they emerge from the gravel. Approximately 33 per cent of the AB fry migrating downstream were captured when traps were set (Table 10). Fourty-eight per cent of the W fry and 57 per cent of the WB fry were captured by the traps (Table 10). Year-to-year v a r i a t i o n i n fry run timing was s i m i l a r to the average between population v a r i a b i l i t y . The average timing difference between years was 7 days whereas the average timing difference among populations was 70 Table 8. Orbit-Hypural P l a t e Lengths o f 1981 and 1982 spawners : means ( X ) , standard d e v i a t i o n s (SO), and sample s i z e s s t r a t i f i e d by p o p u l a t i o n , sex and age. POPULATION AB WB W YEAR SEX AGE X SD n X SD n X SD n 1981 FEMALE 0.2 51.10 2.639 20 52.00 1.732 3 53.16 2.873 19 0.3 55.83 6.276 110 58.75 3.998 20 57.37 2.916 46 MALE 0.2 52.19 2.851 31 54.00 0.000 1 55.44 3.110 1.8 0.3 57.61 4.029 104 61.29 3.729 7 57.81 8.655 52 TOTAL 55.75 5.322 265 58.52 4.456 31 56.68 6.029 135 1982 FEMALE 0.2 54.30 2.042 5 53.52 4.143 28 53.56 3.021 14 0.3 56.03 3.248 25 53.84 5.094 16 54.41 4.497 23 MALE 0.2 54.31 2.964 10 53.84 4.132 11 53.16 40.06 17 0.3 55.34 3.249 19 55.31 6.198 9 53.85 5.280 20 TOTAL 55.37 3.100 59 53.90 4.634 64 53.81 4.325 74 71 Table 9. S i n g l e degree o f freedom comparisons among spawning populations using the Tukey-Kramer Method. Minimum S i g n i f i c a n t D i f f e r e n c e = MSD. (* s i g n i f i c a n t d i f f e r e n c e , P = .05) t - s t a t i s t i c Difference Comparison 1982 1981 Investigated Equation Age Length Temporal X"WB- X A B -0.44 (0.251) 2.77 (2.318) Temporal/ X A B- X w 0.24 (0.251) -0.93 (1.292) Spa t i a l S p a t i a l X W B- X w -0.20 (0.230) 1.84 (2.432) 02:00 04:00 06:00 08:00 10:00 12:00 14:00 1&00 18:00 2 0 0 0 22:00 24:00 T I M E OF DAY Figure 7. D i e l timing of emergence of fry from AB, WB and W 73 Table 10. Capture e f f i c i e n c y o f i n c l i n e d plane t r a p s below the spawning areas during 1983. Population (1) Date AB WB W N.M. N.R. T.C. N.M. N.R. T.C. N.M. N.R. T.C. A p r i l 5 25 11 174 0 0 0 50 26 322 A p r i l 12 50 21 201 0 0 0 25 10 67 A p r i l 19 50 15 1507 25 18 99 50 30 373 Apr 1 26 100 36 1588 100 62 797 50 24 236 May 3 100 30 1013 100 51 697 100 53 462 May 14 25 3 322 25 11 248 25 2 107 TOTAL 350 116 4805 250 142 1841 300 145 1567 (1) N.M. = Number Marked N.R. - Number Recovered T.C. = Total Catch 74 9 days. The temporal pattern of fry run timing was reversed among stocks compared to that of the adult populations such that the l a t e s t spawned progeny were the e a r l i e s t to migrate. The median of the WB fry run was 1-2 days p r i o r to the AB run in 1981 and 1982. The median of the W fry run was 5 to 22 days pr i o r to the AB run during those years (Figure 8). Differences in timing of downstream fry runs among the populations were greater in comparisons among s p a t i a l l y separated stocks versus comparisons among temporally separated stocks (Table 11). The median of the WB run was only 1-2 days p r i o r to the AB run compared to 4-20 days af t e r the W run. Analysis of variance (with run time as a block e f f e c t ) of vertebral counts in the 1982 emerging juveniles revealed s i g n i f i c a n t differences among the populations (F = 9.94, d.f. = 2, 342, P = .05, N - 358). Comparisons among means (Tukey-Kramer Method) revealed s i g n i f i c a n t differences in mean vertebral count among s p a t i a l l y and temporally separated stocks (Table 6). WB fry had more vertebrae (X = 67.0, N r 120) than s p a t i a l l y separated W (X = 66.3, 6.3, N = 111) and temporally separated AB (X = 66.4, N = 126) fry (Table 12). No s i g n i f i c a n t differences were observed where there was temporal plus s p a t i a l separation among populations (Table 6). Analysis of variance (with time as a block e f f e c t ) of vertebral counts in the 1983 emerging juv e n i l e s revealed s i g n i f i c a n t differences among the populations (F = 197.84, d.f. = 2, 445, P = .05, N = 450). Comparisons among means (Tukey-Kramer Method) revealed s i g n i f i c a n t differences i n mean vertebral 75 h-X O < O cc L L O 7 2 0 0 • 6300 • 5 4 0 0 • 4500 3600 2700 1800 900 0 1600 1400 1200 1000 800 600 4 0 0 200 0 • WALKER A WINTER BUSH o AUTUMN BUSH 1982 o 0 o o u o A 1983 oo oo A A i • A • A A CD n O i " i 36 48 60 72 84 96 J U L I A N D A Y 1—r ' v — i — r 108 120 132 144 156 168 Figure 8. Timing of fry downstream migrations from Walker Creek and the upper and lower sections of the spawning area in Bush Creek during 1982 and 1983. Number of fry captured i n in c l i n e d p l a i n traps versus the J u l i a n day 76 Table 11. The y e a r l y median and mode ( i n J u l i a n day) o f the temporal d i s t r i b u t i o n o f f r y emigrating from Bush and Walker creeks during 1982 and 1983. METHOD OP CALCULATION YEAR DOWNSTREAM ' POPULATIONS BUSH UPSTREAM BUSH WALKER MEDIAN 1982 128 126 106 1983 115 114 110 MODE 1982 130 122 101 1983 115 116 90 Table 12. V e r t e b r a l counts f o r f r y m i g r a t i n g downstream during 1982. Means, standard d e v i a t i o n s and sample s i z e s o f populations s t r a t i f i e d by time. Fry samples were made at i n t e r v a l s corresponding t o the t o approximate 17th, 33rd, 50th, 67th, and 83rd p e r c e n t i l e s o f the cumulative frequency d i s t r i b u t i o n o f i n d i v i d u a l s w i t h time. POPULATION AB WB W TIME X (PERCENTILE] i X SD n X SD n SD n 17 66.71 1.517 24 67.56 1.044 25 66.00 0.666 10 33 66.84 1.666 26 67.05 1.463 22 65.45 1.061 8 50 65.85 1.047 26 66.57 0.945 23 66.19 1.387 26 67 65.58 1.176 24 67.19 0.939 26 66.45 1.214 42 83 67.08 2.412 26 66.63 1.245 24 66.57 0.866 25 TOTAL 66.42 1.482 126 67.00 1.177 120 66.28 1.146 111 77 count among s p a t i a l l y and temporally separated stocks (Table 6). WB fry had fewer vertebrae (X* - 64.9, N = 150) than s p a t i a l l y separated W (X = 66.6, N - 150) and temporally separated AB (X = 67.3, N = 150) f r y (Table 13). As well, AB fry had more vertebrae than the s p a t i a l l y and temporally separated fry of the stock, W, but the magnitude of the difference was not as great as in the other comparisons. Year-to-year v a r i a t i o n i n vertebral count of the juveniles was not s i g n i f i c a n t in W and AB. However, WB juveniles showed substantial year-to-year v a r i a t i o n in vertebral count. As a consequence year-to-year v a r i a t i o n i n interpopulation divergence of vertebral counts of juveniles was large. The s h i f t i n mean vertebral counts between the years suggests that year-to-year environmental differences may influence the expression of vertebral counts. The reversal i n the d i r e c t i o n of vertebral count differences among the populations suggests that genotype-environment i n t e r a c t i o n plays a role in the expression of the t r a i t . The pooled error variance of the adult vertebral counts was s i g n i f i c a n t l y greater than the f r y vertebral counts [VAR(adult) = 5.87; VAR(juvenile, 1982) r 1.53; VAR(juvenile, 1983) = 1.17, B a r t l e t t ' s t e s t , P < 0.01]. Discriminant analysis of morphometric measurements from emerging fry was moderately successful in discriminating among the AB, WB, and W populations when comparing samples within each year (Tables 15, 16) but d i f f e r e n t characters were important discriminators between years (Table 14). The 78 Table 13. V e r t e b r a l counts f o r f r y m i g r a t i n g downstream during 1983. Means, standard d e v i a t i o n s and sample s i z e s o f populations s t r a t i f i e d by time. Fry samples were made at i n t e r v a l s corresponding t o the t o approximate 25th, 50th and 75th p e r c e n t i l e s o f the cumulative frequency d i s t r i b u t i o n of i n d i v i d u a l s w i t h time. POPULATION AB WB W TIME (PERCENTILE) X SD n X SD n X SD n 25 67. 50 0.647 50 64.78 1. 298 50 66.540 0.762 50 50 67.64 0.693 50 64.88 1. 350 50 66.66 1.042 50 75 66. 68 1.096 50 64.98 1. 407 50 66.72 0.927 50 TOTAL 67. 27 0.933 150 64.88 1. 346 ' 150 66.64 1.346 150 Table 14. V a r i a b l e s entered i n t o the 1982 and ranked by importance w i t h c l a s s i f i c a t i o n f u n c t i o n s 1983 d i s c r i m i n a n t f u n c t i o n s f o r each. Variable Coefficient for Discriminant Function Ranking 1 2 3 82 83 1982 1983 1982 1983 1982 1983 Eye Diameter 1 - 9.29 9.31 8.81 Snout Length 2 4 0.20 -1.05 0.63 •0.83 1.31 -1.12 Weight 3 - ' -0.55 -0.57 -0.58 Parr Marks 4 1 0.71 5.61 0.79 5.42 0.02 5.02 Head Length 5 2 1.12 3.75 1.20 3.94 2.09 3.95 Head Depth 6 5 3.33 1.41 3.27 1.66 3.06 1.44 Total Length 7 7 2.04 -0.71 2.05 •0.85 2.09 -0.72 Body Depth - 3 2.84 2.66 2.99 Standard Length - • 6 2.04 2.18 2.01 79 Table 15. C l a s s i f i c a t i o n matrix f o r 1982 f r y samples us i n g the d i s c r i m i n a n t f u n c t i o n . GROUP PERCENT CORRECT CASES AB CLASSIFIED INTO WB GROUP W AB 64.0 96 46 8 WB 58.0 28 87 35 W 81.3 3 25 122 TOTAL 67.8 127 158 165 Table 16. f u n c t i o n . C l a s s i f i c a t i o n matrix f o r 1983 f r y samples using the d i s c r i m i n a n t GROUP PERCENT CORRECT CASES AB CLASSIFIED INTO WB GROUP W AB 52.6 80 30 42 WB 55.3 40 84 28 W 61.22 3 36 92 TOTAL 56.4 142 150 162 80 greatest difference in morphology occurred between the W and AB fry i n both years (Tables 17-20). In 1982, W fry had s i g n i f i c a n t l y smaller eyes than WB and AB fry (P < 0.05) (Table 14). AB f r y had s i g n i f i c a n t l y shorter snouts than WB and W f r y (P < 0.05). WB fry had s i g n i f i c a n t l y shorter snouts than W fry (P < 0.05). W fry were s i g n i f i c a n t l y l i g h t e r than WB and AB fry (P < 0.05). WB fry were s i g n i f i c a n t l y l i g h t e r than AB fry (P < 0.05). W fry had s i g n i f i c a n t l y fewer parr marks than WB and AB fry (P < 0.05). AB fry had s i g n i f i c a n t l y deeper heads than WB and W fry (P < 0.05). WB fry had s i g n i f i c a n t l y deeper heads than W f r y (P < 0.05). In 1983, W fry had s i g n i f i c a n t l y fewer parr marks than WB and AB fry (P < 0.05) (Table 22). AB fry had s i g n i f i c a n t l y shorter heads than WB and W fry (P < 0.05). WB f r y had s i g n i f i c a n t l y longer snouts than W fry (P < 0.05). AB fry had s i g n i f i c a n t l y shallower heads than WB f r y (P < 0.05). When c l a s s i f i c a t i o n of the 1983 samples were attempted using 1982 discriminant function the r e s u l t s were poor (Table 23). Important to note i s the high misclassi f i c a t i o n of the WB samples. In both years well over 40% of the samples were m i s c l a s s i f i e d . Incubation rates Walker Creek was 0.58°C warmer than Bush Creek during the 1981-82 incubation period (P < 0.001). Walker Creek was 0.59°C warmer than Bush Creek during the 1982-83 incubation period (P < 0.001). Water column temperatures in Bush Creek were on average 0.91°C warmer during the 1982-83 incubation period compared to the 1981-82 incubation period (P = 0.021). Water column temperatures i n Walker Creek were on average 0.90°C warmer during the 1982-83 incubation than i n the 1981-82 period (P = 0.002) (Figure 9). In 1981-82 81 Table 17. fo r 1982. Approximate tran s f o r m a t i o n F s t a t i s t i c comparing group c e n t r o i d s AB WB WB 13.04 W 62.42 29.28 Table 18. Mahalanobis d i s t a n c e between p o p u l a t i o n c e n t r o i d s f o r 1982. AB WB WB 1.19 W 5.72 2.68 Table among 19. 1983 Approximate tran s f o r m a t i o n F s t a t i s t i c comparing group c e n t r o i d s progeny. AB WB WB 6.12 W 9.70 7.32 Table 20. Mahalanobis d i s t a n c e between pop u l a t i o n c e n t r o i d s f o r 1983. AB WB WB 0.554 W 0.911 0.688 82 Table 21. Means and standard d e v i a t i o n s o f morphological c h a r a c t e r i s t i c s o f 1982 f r y samples. Lengths are i n 0.1mm. Weight i s i n 0.1gm.. (S.M. = Mean standardized t o a common length among the samples). POPULATION AB WB W CHARACTERISTIC MEAN S.D. MEAN S.D. MEAN S.D. S.M. S.M. S.M. Head Length 80.91 5.58 80.93 4.11 81.00 2.75 80.55 81.14 81.08 Snout Length 12.20 0.72 12.79 1.35 13.83 1.75 12.17 12.86 • 13.84 Pectoral F i n 39.82 0.93 40.23 2.28 40.45 2.20 Length 39.77 40.29 40.52 Eye Diameter 29.73 1.67 29.27 1.74 27.69 1.28 29.62 29.29 27.70 Head Depth 49.08 3.26 47.89 2.63 46.61 2.33 48.85 48.06 46.69 Body Depth 51.23 3.53 49.73 2.53 49.92 2.64 50.98 49.82 50.05 Weight (gms) 413.97 74.19 367.50 46.52 341.73 37.53 406.36 370.52 344.10 Parr Marks 9.35 1.20 9.53 1.19 8.53 1.16 9.310 9.54 8.54 Trunk Length 286.26 13.73 282.31 14.08 282.33 12.96 284.03 283.49 283.50 Caudal F i n 14.97 9.47 12.67 4.16 12.40 4.89 Length 15.47 12.79 12.39 WT/TL 1.08 0.16 0.98 0.09 0.91 0, .08 1.06 0.98 0.91 KD (Bams 1976) 1.94 0.54 1.91 0.45 1.86 0.43 83 Table 22. Means and standard d e v i a t i o n s o f morphological c h a r a c t e r i s t i c s o f 1983 f r y samples. Lengths are i n 0.1mm. Weight i n 0.1gm.> (S.M. = Mean standardized t o a common length among the samples). POPULATION AB WB W CHARACTERISTIC MEAN S.D. MEAN S.D. MEAN S.D. S.M. S.M. S.M. Head Length 81.07 81.11 3.07 83.08 82.77 3.58 82.57 82.82 3.54 Snout Length 14.91 14.92 2.35 15.75 15.62 2.28 14.75 14.78 1.88 Pectoral F i n Length 36.71 36.74 3.77 37.92 37.69 3.79 36.76 37.00 4.07 Eye Diameter 29.47 29.48 1.22 29.87 29.81 0.85 29.81 29.85 0.93 Head Depth 44.34 44.35 3.33 45.93 45.74 2.74 45.30 45.40 2.81 Body Depth 48.75 48.79 2.84 48.43 48.27 2.47 49.57 49.65 2.55 Weight (gms) 364.96 366.07 47.69 369.34 364.59 45.60 376.07 380.05 44.64 Parr Marks 9.50 9.50 1.37 9.25 9.21 1.13 8.61 8.60 1.09 Trunk Length 236.22 243.32 10.51 236.26 234.85 10.53 233.64 234.80 10.42 Caudal Fin Length 53.75 53.79 4.83 53.74 53.42 4.68 54.61 54.81 5.12 WT/TL 0.98 0.98 0.101 0.99 0.98 0.095 1.01 1.02 0.091 KD 1.92 0.47 1.92 0.46 1.94 0.43 84 Table 23. C l a s s i f i c a t i o n matrix f o r 1983 f r y samples using the d i s c r i m i n a n t f u n c t i o n from the 1982 data. GROUP PERCENT CASES CLASSIFIED INTO GROUP CORRECT AB WB W AB 3.8 6 54 92 WB 19.0 6 30 116 W 66.7 6 44 100 TOTAL 30.0 18 128 308 85 2 300 250 200 150 100 50 0 450 4. 360 270 H 180 90 H • OCU> O i ? ^ i ^ i n i ™ i 1 1 1 r T i 1 1982-83 * V i — f — i 1 1 ^ I T ^ t t T T p i ^ j r n r y ^ y - r y — y - J 1 "1 i 1 T 15 15 14 • 14 13 12 14 13 13 12 12 AUG SEPT OCT NOV DEC JAN FEB MAR APR MAY JUN JUL 31 30 30 29 29 28 27 29 28 28 27 27 Fiaure 9. Temperature and flow p r o f i l e s from Bush and Walker Creeks during 1981-82 and 1982-83. Closed c i r c l e s = Bush Creek; Open c i r c l e s = Walker Creek. 86 gravel temperatures taken in the redds were on average 0.132°C warmer than the column temperatures in Bush Creek (t - 3.60, P r .05). No s i g n i f i c a n t difference occurred between the column and gravel temperatures of Walker Creek in 1981-82. The temporal pattern of spawning i n the creeks r e s u l t s in a marked va r i a t i o n in the temperature regimes experienced by the embryos of the populations (Figure 9). AB embryos experience a temperature incubation curve i n which temperatures s t a r t out high in October, then decline to minimums in January before climbing again. WB and W embryos experience temperature regimes i n which temperatures s t a r t out close to the minimum then climb to spring values. These differences are important because the embryos of early and l a t e spawning stocks experience d i f f e r e n t temperatures during p a r t i c u l a r developmental stages. Progeny of the AB run accumulated more thermal units before emerging than did WB or W progeny (Table 24). A l l populations apparently required fewer degree-days to reach emergence in 1982-83 than in 1981-82 (Table 24). Predicted (assuming no genetic v a r i a t i o n among stocks) incubation times using the mean temperature of incubation and Murray's (1980) regression equation are compared to the actual incubation times in Table 24. In both years AB actual incubation time was longer than the predicted while WB and W actual incubation times were a l l shorter than predicted. 87 Table 24. Comparisons o f accumulated degree-days (raw and adjusted data) and inc u b a t i o n times ( a c t u a l and pr e d i c t e d ) f o r AB, WB, and W stocks i n 1981-82 and 1982-83. ( c a l c u l a t e d from median day of the adult and f r y r u n s ) . POPULATION YEAR AB WB W MEAN TEMP. OF INCUBATION 1981-82 4.82 3.66 4.36 1982-83 5.68 5.48 6.45 TOTAL DAYS TO EMERGENCE 1981-82 192 165 133 1982-83 180 153 130 PREDICTED DAYS TO EMERGENCE 1981-82 182 204 191 1982-83 168 172 156 DIFFERENCE (ACTUAL-PREDICTED) 1981-82 +10 -39 -58 1982-83 +12 -19 -26 ACCUMULATED DEGREE-DAYS 1981-82 925.25 603.35 579.65 1982-83 1022.20 839.20 838.25 ADJUSTED DEGREE-DAYS 1981-82 1039 845 873 1982-83 1078 853 834 88 When degree days are adjusted to account for the variable incubation rates at d i f f e r e n t temperatures, the year-to-year v a r i a t i o n disappears while the v a r i a t i o n among stocks remains constant (Table 24). Comparisons of samples of eggs taken from AB and W in 1981-82 incubation period show that W lags behind AB i n i t i a l l y but gradually catches up in stage of development (Figure 10). Regression equations describing the r e l a t i o n s h i p between egg weight and female orbit-hypural plate length are as follows: AB Egg weight = -0.23808 + 0.00980 X Length (R-squared = 48.5 P < 0.001) WB Egg weight = -0.19542 + 0.00862 X Length (R-squared = 67.5 P < 0.001) W Egg weight = -0.20122 + 0.00926 X Length (R-squared = 56.4 0/ P < 0.001) Analysis of variance of regression c o e f f i c i e n t s over populations revealed that the l i n e s were not equivalent (F = 26.00, d.f. = 4,404, P < 0.00001) (see 89 c c BUSH WALKER 1.8 2.0 LOG DAY 2.2 Figure 10. Egg development stage (Velsen 1980) versus log day (post-August 31, 1981) for AB and W. Horizontal bars are proportional to the number of eggs at each stage 90 Figure 11). Analysis of variance of regression c o e f f i c i e n t s of pai r s of populations revealed that WB was d i f f e r e n t from AB (F = 22.07, d.f. - 2,286, P < 0.00001 ) and W (F r 62.81, d.f. = 2,266, P < 0.00001 ). W and AB regression l i n e s were equivalent (F =• 2.61, d.f. = 2,266, P < 0.07568). The slopes of the WB and AB regressions of egg weight versus length were s i g n i f i c a n t l y d i f f e r e n t (P = 0.029). The intercepts of the WB and AB regressions were also s i g n i f i c a n t l y d i f f e r e n t (P < .001). The slopes of the WB and W l i n e s were equivalent (P = 0.203) but the intercepts were d i f f e r e n t (P < 0.001). Analysis of covariance revealed that the adjusted mean egg weight d i f f e r r e d among WB and W (F = 25.25, d.f. r 1,197, P < 0.0001). Comparisons among adjusted means using the GT2 method revealed that WB egg weights (X = 0.26903) were s i g n i f i c a n t l y lower than W (X = 0.29793) (Table 25). Homing and straying The returns of marked f i s h i ndicate that straying among the three populations was substantial (Tables 26, 27). The exchange was greatest between AB and WB with 9.03 % of the returns born i n AB returning to WB and 40.30 % of the returns born i n WB returning to AB. A l l strays from W (50.0 % of the returning marked fish) were found i n the WB area. A small percentage ( 1 . 8 1 %) of the returns i n AB returned to W. In 1984, the composition of 0.3 year olds i n the AB spawning population was 88.16 % f i s h that originated from AB parents, 11.84 % that originated from F i g u r e 1 1 . Egg w e i g h t v e r s u s f e m a l e o r b i t - h y p u r a l p l a t e l e n g t h o f spawners from W a l k e r C r e e k and t h e u p p e r and l o w e r s e c t i o n s o f t he s p a w n i n g a r e a i n Bush C r e e k d u r i n g 1981 and 1982. 92 Table 25. S i n g l e degree o f freedom comparisons of egg weight d i f f e r e n c e s among populations adjusted f o r c o v a r i a t e using the GT2 Method. (MSD -Minimum S i g n i f i c a n t D i f f e r e n c e ) (* s i g n i f i c a n t d i f f e r e n c e P = .05). Difference Among Means Difference Comparison Investigated Difference (MSD) Temporal XWB" XAB -0.0202 (0.0111) Temporal/ Sp a t i a l X A B- X w -0.0087 (0.0109) S p a t i a l XAB" XW -0.0289 (0.0101) 93 Table 26. The temporal and spatial pattern of return of marked fish to Bush and Walker Creeks during 1984 and 1985. Recovery Location Estimated Total Marks Rate of AB W3 W AB WB W Straying Population of Origin 1984,1985 1984,1985 1984,1985 1984,1985 1984,1985 1984,1985 1984,1985 AB 15, 3 3, 0 1, 0 201,194 40, 0 8, 0 1 9.3 55,0 58 WB 2, 0 3, 0 0, 0 27, 0 40, 0 0, 0 40.3 % — W 0, 0 4, 0 7, 0 0, 0 54, 0 54, 0 50.0 % -Table 27. Return rate of marked fi s h . Population Nurter Marked Nutter Recovered 1984, 1985 Ex ani noted for Marks 1984, 1985 Population Size 1984, 1985 Rate of Return AB 12,432 19, 3 682, 75 9161, 4842 3.611 % WB 2,111 5, 0 100, 20 1339, 708 3.171 % W 8,053 11, 0 150, 20 1150, 200 1.047 % 94 WB parents and 0 % that originated from W parents. In 1985, only f i s h marked as AB progeny were recovered. These were a l l observed to home to AB. In 1984, composition of 0.3 year olds i n the WB spawning population was 29.85 % f i s h that originated from AB parents, 29.85 % that originated from WB parents, and 40.30 % f i s h that originated from W parents. The composition of 0.3 year olds in the W population was 12.90 % f i s h o r i g i n a t i n g from AB parents, 0 % f i s h o r i g i n a t i n g from WB parents, and 87.10 % o r i g i n a t i n g from W parents. DISCUSSION Spawning and subsequent development of the embryos in W and WB occur under s i m i l a r conditions. Spawning occurs during d i f f e r e n t seasons i n WB and AB. As a consequence the incubation environment d i f f e r s greatly between the two groups. Spawning also occurs during d i f f e r e n t seasons i n W and AB. In contrast to the separation i n time of the spawning populations, the fry migrate i n r e l a t i v e synchrony. Thus, during the egg to fry phase, the of f s p r i n g of ea r l y spawning and l a t e spawning stocks develop under sharply d i f f e r i n g conditions, but by vi r t u e of t h e i r timing of downstream migration, occupy a common environment during the coastal juvenile phase. The synchronization of migration among the populations appears to be the res u l t of s t r i k i n g i n t r i n s i c incubation rate differences between e a r l y and lat e stocks rather than due to the e f f e c t s of temperature or other environmental e f f e c t s . S i g n i f i c a n t d ifferences occurred in adult and fry vertebral number in comparisons between s p a t i a l l y and temporally separated populations but not between populations separated by both space and time. No differences occurred among populations 95 in length and age at return of spawners except between the lengths of WB and AB spawners in 1981 and age at return of WB and AB spawners in 1982. The greatest difference in morphology occurred between AB and W progeny. Straying was found to be considerable with substantial straying into WB by both W and AB. M i l l e r and Brannon (1982) and Brannon (1984) suggested two models to account for the evolution of v a r i a t i o n i n spawning timing among sockeye salmon stocks. Spawning timing could be adjusted to o f f s e t differences in mean temperatures among the streams. Brannon (1984) presents evidence of a c o r r e l a t i o n between the date of spawning and the mean temperature of the stream during the incubation period. Populations that reproduce in colder streams spawn e a r l i e r than those reproducing in r e l a t i v e l y warmer streams. The adjustment of spawning timing ensured that fry migrated to the nursery lake at a time that minimized thermal s t r e s s , predation, food supply l i m i t a t i o n s and space and shelter c o n s t r a i n t s . Thus, incubation rate program i s s i m i l a r among the stocks. However, M i l l e r and Brannon (1982) observed differences in spawning timing at Chilko Lake without differences in the temperature of incubation. The r e l a t i v e l y synchronous migration of fry out of the gravel suggests that the stocks were innately d i f f e r e n t in the incubation rate program. My r e s u l t s present a t h i r d case. Walker Creek i s warmer than Bush Creek and has the l a t e s t spawning population of the three. However, synchrony of fry migrations i s achieved due to differences i n the rate of development among the early and l a t e stocks. 96 ADULT CHARACTERISTICS Chebanov (1986) proposed that several factors, such as spawner density, sex r a t i o , duration of the spawning period, body length of spawners, time of a r r i v a l and duration of stay on the spawning grounds influenced reproductive success i n Oncorhychus. The consistency of the timing of return to the spawning grounds from year to year suggests that annual timing of spawning i s somewhat independent of environmental influences. Bams (1976) found timing of return of pink salmon, Oncorhynchus gorbuscha, to have a genetic component. Recently, G a l l et a l . (1988) also suggested that season of spawning was h e r i t a b l e . However, the returns of marked f i s h to Bush and Walker creeks suggest that straying between streams and seasons of spawning was a common occurrence i n these streams. The computed adult freshwater residence time i s approximately one week or less so I int e r p r e t the escapement timing as an approximation of the time of spawning. Freshwater residence time of the Walker Creek population i s s i g n i f i c a n t l y longer than that of f i s h i n Bush Creek. Three factors might be hypothesized to be responsible for longer persistence i n Walker Creek: (a) lower temperatures l a t e r i n the spawning season may reduce energy demands upon the f i s h ; (b) weaker flows i n Walker Creek may s i m i l a r l y reduce energy demands; (c) genetic differences between stocks. If (a) were true then one would expect that freshwater residence time within runs would increase with time as temperatures became colder. Conversely, hypothesis (b) would predict 97 a decrease in freshwater residence time within runs with time. In 1981 residence time decreased within both runs with time while in 1982 residence time increased (Table 4). These data suggest that another factor, such as genetic di f f e r e n c e s , i s important. Stream s u r v i v a l time, as determined by tag recoveries, has been reported in a number of other chum populations. Elson (1975) examined f a l l chum salmon stream s u r v i v a l time i n the Fishing Branch River. Average stream l i f e was 30.5 days in 1972 and 21 days i n 1973. Trasky (1974) determined that stream l i f e in the Delta- River averaged 20.4 days in 1973 and 18 days i n 1974. According to Bakkala (1970) freshwater l i f e in southeast Alaskan coastal streams varied from 11.4 days to 18.3 days. Helle (1979) recorded freshwater stream l i f e of chum i n Olsen Creek, Alaska to be between 7.5 and 10.8 days. Schroder (1973) determined that stream l i f e varied between 13.1 days and 15.9 days at Big Beef Creek, Washington. Buklis and Barton (1984) hypothesize that r i v e r s i z e and length of freshwater migration are p o s i t i v e l y correlated with the length of freshwater l i f e while stream temperature i s negatively c o r r e l a t e d . This would explain the d i s p a r i t y between my r e s u l t s and those of Helle (1979), Trasky (1974), Bakalla (1970) and Elson (1975). The discrepancy between my r e s u l t s and Schroder's (1973) may be due to a tagging induced delay in reproduction at Big Beef Creek or an undetermined systematic bias i n my sampling methods. A l t e r n a t i v e l y , both r e s u l t s could be accurate and simply r e f l e c t d i f f e r e n c e s i n freshwater s u r v i v a l due to the differences in conditions between an a r t i f i c i a l spawning channel and natural streams. 98 Both genetic and environmental factors can influence v a r i a t i o n in meristic characters among populations of f i s h (Lindsey 1988). Laboratory studies have shown that vertebral counts w i l l vary with temperature of incubation (Lindsey 1975). Beacham (1984) found that early spawning chum salmon stocks averaged fewer g i l l rakers than l a t e spawning stocks. He suggested that t h i s occurred because early stocks probably spawned at warmer temperatures than l a t e spawning stocks. Kubo (1950) incubated eggs from a single chum salmon cross under d i f f e r e n t temperatures and found young derived from eggs incubated at lower temperatures had more vertebrae than those derived from eggs incubated at higher temperatures. However, the genetic component has also been shown to be important. For example, i n salmonidae the genetic basis of such v a r i a t i o n has been confirmed among s p a t i a l l y i s o l a t e d populations of chinook salmon, Oncorhynchus tshawytcha (Seymour 1959), and among summer and winter run steelhead trout, Salmo gairdneri (Smith 1969). In chum salmon g e n e t i c a l l y based vertebral count differences have been confirmed among s p a t i a l l y separated populations and among temporally separated populations (Kubo 1950 ; Kubo 1956 ; Ricker 1972). According to Lindsey (1988) the operational factor in l a t i t u d i n a l v a r i a t i o n of vertebral counts i s temperature. Vertebral counts are more highly correlated with temperature at spawning time than other factors, such as l a t i t u d e (Lindsey 1988). 99 A simple r e l a t i o n s h i p between temperature of incubation and vertebral count does not explain the pattern of v a r i a t i o n that I observed. WB and W have the greatest vertebral count difference among the comparisons but t h e i r embryos experience the most subtle difference in temperature regime during development. The progeny of temporally separated populations experience s u b s t a n t i a l l y greater temperature differences in t h e i r embryonic incubation environment. Yet, the magnitude of vertebral count differences i s less in comparisons among temporally separated stocks and i n s i g n i f i c a n t where there i s both temporal and s p a t i a l separation. Age at maturity i s d i r e c t l y related to the p o t e n t i a l reproductive e f f o r t . According to Smirnov (1975) the age at return may vary among chum salmon stocks entering d i f f e r e n t r i v e r s but generally does not vary among seasonally separated stocks. However, Beacham (1984) and Beacham and Murray (1987a) recorded that age composition changed during the course of the spawning migration into the Fraser River. The e a r l i e r a r r i v i n g f i s h were older than the l a t e a r r i v i n g f i s h . Birman (1977) also found that mean age at return was greater in early spawning stocks than in l a t e r ones. Similar observations were reported by Sano (1966) and Helle (1979). This pattern of changing age composition with time occurred in Bush Creek in 1982. Mean age at return of a cohort has been reported to increase with cohort abundance (Helle 1979, Beacham and Starr 1982). The number of spawners returning in 1982 exceeded that of 1981 in a l l populations but the mean age of 100 1982 spawners was less than 1981. This suggests that density e f f e c t s may be less l i m i t i n g on age of return than environmental factors influencing growth. The importance of age at maturity in terms of reproductive e f f o r t i s re f l e c t e d i n the generation time (Cole 1954) and in correlated responses of fecundity and egg siz e (Gall 1975). E a r l i e r age at maturity w i l l be favored in an organism whose reproductive p o t e n t i a l increases slowly with s i z e (Gadgil and Bossert 1970). An e a r l i e r reproducer can minimize the generation time and avoid further r i s k of mortality in the ocean. However, Gall (1974) found that older rainbow trout (Salmo gairdneri) had a higher fecundity and produced larger eggs than younger f i s h . The f r y of the older f i s h had better growth rates. Thus, older f i s h may produce greater numbers of more f i t fry than the younger spawners (Shine 1988). Recently, Gall et a l . (1988) demonstrated that the age at maturity i n rainbow trout i s highly h e r i t a b l e (h2 r 0.38). In a given environment these factors probably balance each other so that reproduction w i l l take place at the age that r e s u l t s in the greatest number of off s p r i n g produced that reach reproductive age. The l i m i t e d information available suggests that older spawners are larger than younger spawners. For example, Bakkala (1970) reports fork lengths of female chum salmon as 58.7 cm at age 0.2 and 61.4 cm at age 0.3. Male chum salmon had fork lengths of 58.5 cm at age 0.2 and 59.9 cm at age 0.3. Larger females are generally more fecund than smaller females. Larger s i z e may carry an advantage in courtship. Schroder (1981) observed that, in chum salmon, larger males had greater mating success compared to smaller males. During 101 1981 older spawners in Bush and Walker Creeks were larger than younger spawners. However, 1982 female spawners d i f f e r r e d only s l i g h t l y in the length at age (AB - 54.3 at age 0.2 and 56.0 at age 0.3, WB - 53.42 at age 0.2 and 53.84 at age 0.3, W - 53.56 at age 0.2 and 54.41 at age 0.3). S i m i l a r l y , length at age of male spawners d i f f e r r e d l i t t l e (AB - 54.31 at age 0.2 and 55.34 at age 0.3, WB - 53.84 at age 0.2 and 55.31 at age 0.3, W - 53.16 at age 0.2 and 53.85 at age 0.3). Individuals of early run stocks are often smaller than l a t e r run f i s h (Berg 1934, Ricker 1972,). Smaller s i z e may be an advantage early i n the year when flows are lower. Smaller f i s h may be able to more e f f e c t i v e l y maneuver to avoid predators, such as bears or g u l l s , in shallow water. Larger si z e may be an advantage in migrating and maintaining p o s i t i o n against stronger currents that occur l a t e r in the year. However, Leider et a l . (1986) found that summer steelhead trout were larger at a given age than winter steelhead. My 1981 r e s u l t s concur with data gathered by Berg (1934) and Ricker (1972) but in 1982 the pattern reverses and i s s i m i l a r to that found by Leider et a l . (1986). Fry C h a r a c t e r i s t i c s Although the adult runs from AB are allochronous r e l a t i v e to the W or WB runs the fry runs i n Bush and Walker Creek are r e l a t i v e l y synchronous. Since the fry reach the estuary in one night I conclude that there i s a synchrony both in the timing of emergence and the subsequent migration into the 102 estuary. According to Buklis and Barton (1984) synchrony of fry runs occurs among the summer and autumn chum stocks in the Yukon River. In contrast, at Big Beef Creek the early stock progeny emerged 25 to 30 days p r i o r to the l a t e stock o f f s p r i n g (Koski 1975). However, from spawning to emergence, progeny of the winter stock in Big Beef Creek gained 12 to 14 days on the autumn spawned f i sh. Godin (1982) argued that the degree of synchrony i n emergence timing within a population could have important e f f e c t s on the predation rate on migrant f r y during t h e i r d i s p e r s a l movements from the nest s i t e . The greater the degree of synchrony i n fry emergence, the lower the r e l a t i v e mortality rate of these f r y because predators w i l l be swamped or s a t i a t e d (Peterman and Gatto 1978). Hence, s e l e c t i v e pressure w i l l be greater on asynchronous in d i v i d u a l s and population synchrony w i l l emerge. This argument i s e a s i l y extended to provide a mechanism for fry migration synchrony among populations which migrate through a common estuarine environment as must the progeny of most allochronic chum salmon stocks. Another suggestion i s that synchrony of fry migrations should evolve in response to the short term abundance of food in the estuary. Bams (1969), Northcote (1978) and M i l l e r and Brannon (1982) have suggested that seasonally predictable periods of maximum secondary production in coastal marine areas of temperate regions promote genetic d i f f e r e n t i a t i o n for predictable optimal times of fry migration into the nursery habitat which maximize feeding opportunities, growth and s u r v i v a l . Healey (1979) and Sibert (1979) performed 103 cooperative studies which showed that the seasonal pattern of chum fry migration on the Nanaimo River corresponded c l o s e l y with seasonal patterns of density of the f i s h ' s perferred epibenthic prey on estuarine mud f l a t s . . Walters et a l . (1978) combined the predation and feeding hypotheses using computer simulations. An optimal mean time of downstream migration and entry into the estuary maximizing early marine s u r v i v a l was suggested for Fraser River chum salmon f r y . , Predicted optimal mean date coincided c l o s e l y with the known peak time of abundance of chum salmon fry in the Fraser River estuary. Walters et a l . ' s (1978) model suggests that the lower s u r v i v a l of fry entering the estuary on either side of the optimal date i s due to the lower surface zooplankton d e n s i t i e s during these non-optimal periods. Individuals on either side of the optimum grow more slowly becuase of the lower r a t i o n and thus spend a longer period at r i s k from s i z e - s e l e c t i v e mortality r e l a t i v e to fry migrating at the optimum date. This model also predicts that the greater the variance about the mean time of entry the lower the subsequent aggregate s u r v i v a l . Thus, a r e l a t i v e l y narrow time window i s available for downstream migration i f the f r y wish to procure adequate food and reduce the p r o b a b i l i t y of being eaten. Although, B i l t o n (1980) showed that the timing of release of hatchery coho salmon smolts could greatly a f f e c t s u r v i v a l , the evidence for natural s e l e c t i o n against asynchronous wild f r y i s scant. Taylor (1980) found that s u r v i v a l of pink salmon fry migrating at the peak time was 8 times higher than fry released 33 days e a r l i e r . Unfortunately, the f r y with lower s u r v i v a l were 104 also considerably smaller than the others at the time of release. Fresh et a l . (1982) found that i n d i v i d u a l s u r v i v a l increased with increasing numbers of chum salmon fry released. Interpopulation synchrony of downstream migration i s not that s u r p r i s i n g i f the eggs are l a i d at the same time. Synchrony of fry runs produced from allochronic populations suggests that s t a b i l i z i n g s e l e c t i o n operates or has operated on the timing of fry runs. Two adaptive mechanisms are hypothesized for how fry run synchrony i s achieved among seasonal races of chum salmon. Buklis and Barton (1984), based on the observation that Yukon and Amur River autumn chum salmon spawn i n spring-warmed locations, proposed that synchrony i s achieved through behavioural adaptations i n redd s i t e choice by the adult. Late spawning adults choose spring-fed redd s i t e s so that development i s accelerated. A l t e r n a t i v e l y , the stocks could adjust t h e i r genetic incubation rate program to compensate for the e f f e c t of occupying greatly d i f f e r e n t thermal environments as embryos. These hypotheses are not mutually exclusive and both types of adaptation may function i n concert to create the desired phenotype. Buklis and Barton's (1984) hypothesis recieves some support from my data because Walker Creek i s warmer than Bush Creek during the embryonic incubation period. On the other hand I found no evidence from my measurements of intragravel temperatures that the l a t e stocks were u t i l i z i n g s i t e s with warmer temperatures than the values recorded in the water column. It i s c l e a r that the l a t e stocks, WB and W, require s u b s t a n t i a l l y fewer temperature units to 105 reach emergence than the early stock, AB. These differences are probably genetic because when the thermal unit accumulations are adjusted to account for the non-linear response of the genotype to temperature a l l of the year to year v a r i a t i o n disappears. The s l i g h t l y warmer regime in Walker Creek may have been a factor i n allowing i n i t i a l c o l o n i z a t i o n by l a t e spawners but t h i s alone cannot account for the pattern of fry run timing. Independent information provided by Smoker (1982) also suggests that chum stocks have the capacity to make genetic adjustments in incubation rate. Smoker (1982) performed laboratory experiments comparing the time to emergence among several geographically separated chum salmon stocks reared at the same temperature. Emergence time was dependent upon stock and had an estimated h e r i t a b i l i t y of 0.900 (Table 3 in Smoker 1982). Koski (1975) recorded the day-degrees accumulated by the progeny of the early and late spawning chum salmon populations at Big Beef Creek. Within years the la t e stock always required s u b s t a n t i a l l y fewer accumulated day-degrees to emerge. In 1968 and 1969, respectively, the early stock required 106 and 195 more day-degrees (C) to emerge than did the l a t e stock. The average compensation by the AB stock was 208 adjusted thermal units. The agreement between my r e s u l t s and those of Koski (1975) suggests that genetic adjustment of incubation rate i s a common adaptive response among seasonal races to the s t a b i l i z i n g s e l e c t i o n pressure for synchronous fry emergence and downstream migration. 106 Smoker (1982) suggested that incubation rate was negatively correlated with egg s i z e . If so, then egg size differences among early and late populations could account for the differences in incubation rate. However, my results show no correspondence between incubation rate differences and egg size differences among the populations. W has a rapid rate of incubation compared to AB but does not show a corresponding smaller egg s i z e . Relative egg s i z e i s greater in AB compared to WB. Since egg s i z e increases with increasing female length and in 1981 length of spawners was s i g n i f i c a n t l y greater in WB than AB I speculate that the smaller egg s i z e of the WB when adjusted for the length covariate represents a compensatory response by the population to produce an optimum average egg s i z e . The possible genetic adjustment of egg s i z e among stocks towards an optimum i s supported by the c o r r e l a t i o n between the degree of reproductive i s o l a t i o n among stocks and the degree of s i m i l a r i t y i n egg s i z e . While the pattern of differences among the populations remained the same the 1983 fry had more vertebrae than t h e i r parents. This probably r e f l e c t s a difference in the conditions during embryonic development of the adults compared to the f r y . The d i r e c t i o n of the vertebral count difference between WB and AB or WB and W d i f f e r r e d between the fry of 1982 and the adults of 1982 and t h e i r progeny. Two circumstances could produce t h i s r e s u l t . F i r s t , s e l e c t i o n in the estuary or the ocean could truncate the vertebral count d i s t r i b u t i o n . This does not appear l i k e l y as the adult vertebral count variance estimate i s 107 higher than the f r y v e r t e b r a l count variance estimate. A l t e r n a t i v e l y , the observed phenotypes could be produced by the i n t e r p l a y between the genotype of the stocks and the i n f l u e n c e of year to year v a r i a t i o n s of temperature during the incubation p e r i o d . T y p i c a l l y , the v e r t e b r a l number versus temperature of inc u b a t i o n curve i n t e l e o s t s i s "V" shaped so that higher v e r t e b r a l counts are recorded at high and low temperatures of incubation with the i n f l e c t i o n point of low counts o c c u r r i n g at some intermediate temperature (Seymour 1959). Genetic compensatory measures t o incubation temperature v a r i a t i o n among stocks probably take the form of phase s h i f t s i n the "V" or p o s s i b l y d i f f e r e n c e s i n the tangent to each curve (Figure 12). Genetic adjustment occurs i n response to the average temperature d i f f e r e n c e s experienced among stocks over many generations. Superimposed on the v a r i a t i o n w i t h i n the system i s the year to year c l i m a t i c v a r i a b i l i t y . Hence, during years when the temperature of incubation i s higher than the point of i n t e r s e c t i o n between the curves the r e l a t i o n s h i p between the stocks should be reversed (Figure 12). However, support for t h i s hypothesis i s l a c k i n g i n the a v a i l a b l e climate data. Naniamo a i r p o r t records show that temperatures during the incubation period were lower i n 78-79 and higher during 79-80 compared to the temperatures during 81-82. To corroborate the hypothesis temperatures experienced by the 78-79 and 79-80 embryos should be higher than the 81-82 temperatures. However, the greater variance i n the adult v e r t e b r a l counts may r e f l e c t the temperature f l u c t u a t i o n s betweeen the 78-79 and 79-80 seasons. 108 6 7 • A O O 6 6 - • A O Z) o So -< CD 6 4 -LxJ h-UJ > ez4 • A • A t A L A T E 81-82 t L A T E 7 8 - 7 9 7 9 - 80 E A R L Y 81-82 INCUBATION T E M P E R A T U R E O O E A R L Y 7 8 - 7 9 7 9 - 80 Figure 12. Proposed graphical model of the i n t e r a c t i o n between vertebral count program and temperature of incubation among AB, WB, and W during 1978-79, 1979-80, 1981-82. Open c i r c l e s = AB, Closed c i r c l e s = W, Triangles = WB. Arrows indicate the c r i t i c a l temperature determining vertebral count response of progeny from "EARLY" and "LATE" spawners . 109 The lack of f i t in the 1982 discriminant function to the 1983 fry morphology probably r e f l e c t s the e f f e c t of year to year v a r i a t i o n i n incubation temperature on body proportions (Barlow 1961). According to Barlow (1961) the r e l a t i v e proportions w i l l change depending on the temperature of incubation. Since 1981-82 was s i g n i f i c a n t l y colder than 1982-83 incubation season I a t t r i b u t e the s h i f t i n f r y morphology to t h i s . The percentage of m i s c l a s s i f i c a t i o n s i s an estimate of the phenotypic d i s t i n c t n e s s among the stocks i n fry morphology. The fewest m i s - c l a s s i f i c a t i o n s occurred among W and AB (9.35 %) with quite s i m i l a r rates of m i s - c l a s s i f i c a t i o h occurring among W and WB (24 %) and among AB and WB (20.7 % ) . WB f r y are morphologically intermediate between W and AB. Given the p o t e n t i a l rate of genetic migration among the stocks t h i s result could occur i n two ways: (a) in the absence of se l e c t i o n on these t r a i t s ; (b) i f disruptive s e l e c t i o n operates on the fry morphology. In section 4 I w i l l show that there i s r e l a t i v e uniformity i n morphology among the wild fry samples from the three populations when compared to sample populations reared under con t r o l l e d conditions. Bengtson et. a l . (1987) recorded that the time of spawning season influenced the subsequent s i z e of progeny of the A t l a n t i c s i v e r s i d e , Menidia  menidia. However, Fresh and Schroder (1987) found that predators did not choose emergent chum salmon based on s i z e but that predation r i s k was i n v e r s l y related to abundance of f r y . 110 Studies such as those of R i d d e l l and Leggett (1981) and R i d d e l l et a l . (1981) l i n k morphological v a r i a t i o n to s e l e c t i v e regimes. However, a narrow optimum phenotype may not be selected for during the estuarine phase of the l i f e h i s t o r y . Variations in body morphology among the stocks could reduce competition for resources, such as food, in the estuary. A l t e r n a t i v e l y , disruptive s e l e c t i o n on body morphology may occur a f t e r the animals leave the estuary. Possibly the stocks occupy somewhat d i f f e r e n t oceanic environments. Birman (1981) suggested that early spawning Asian chum salmon populations returned from the Sea of Japan while l a t e spawning populations occupied a d i f f e r e n t part of the ocean. Quinn (1984) hypothesized that straying could be an a l t e r n a t i v e l i f e h istory t a c t i c to homing. Quinn (1984) and Quinn and Tallman (1987) predicted that straying should be r e l a t i v e l y common in populations spawning in unstable streams and in species and populations spawning in geographically simple streams with s i m i l a r nearby streams. Bush and Walker Creeks are f i r s t order streams that would on occasion experience winter freshets. The high l e v e l s of straying observed between the population in Bush and Walker Creeks may have evolved i n response to i n s t a b i l i t y of the spawning environment. There are many problems i n i n t e r p r e t i n g i n d i r e c t measures of gene flow from the recovery of marked i n d i v i d u a l s . Marked i n d i v i d u a l s may behave d i f f e r e n t l y than unmarked i n d i v i d u a l s . I f marked in d i v i d u a l s stray more gene flow w i l l be overestimated. If marked i n d i v i d u a l s stray less gene flow w i l l be underestimated. The recovery of marked in d i v i d u a l s does not necessarily 111 mean that these i n d i v i d u a l s had been successful in finding a mate. Although, only a limited number of marks were recovered many of the recovered i n d i v i d u a l s were spent suggesting that the f i s h had spawned with other f i s h in the population. Straying rates among other populations of chum salmon vary. F r e i t a g (1980 c i t e d i n L i s t e r et a l . 1981) recorded no strays from releases of chum salmon f r y at Disappearance Creek, southeast Alaska. Fourteen of 159 returns strayed from the Beaver F a l l s Hatchery release s i t e i n Alaska (note t h i s stock was a transplant). Chum salmon returning to Inches-Barnes Creek i n B r i t i s h Columbia strayed to nearby streams at a rate of 1% ( L i s t e r et a l . 1981). Year to year v a r i a t i o n i n straying rates from Blaney Creek, B r i t i s h Columbia into the North and South Alouette Rivers were highly variable ranging from 46?o in one year to 9% in another ( L i s t e r et a l . 1981). For a de t a i l e d discussion of straying rates r e l a t i v e to season of reproduction and environmental p r e d i c t a b i l i t y r efer to Quinn and Tallman (1987). A high l e v e l of straying does not necessarily mean that there i s gene flow. Dowling and Moore (1985) demonstrated that hybrids may be strongly selected against. My r e s u l t s indicate that vertebral count phenotype d i f f e r s among s p a t i a l l y and temporally separated populations but not among populations that are separated by both space and time. I believe t h i s pattern represents the outcome of genetic adjustment among the populations to a t t a i n an optimal 112 phenotype rather than s o l e l y the e f f e c t of incubation environment on s i m i l a r genotypes. S i g n i f i c a n t differences were found in single degree of freedom tests for the 1982 age differences and the 1981 length differences i n comparisons among stocks with temporal di f f e r e n c e s , only. The most i n t e r e s t i n g comparison i s between the two stocks that are most l i k e l y to have the opportunity to diverge g e n e t i c a l l y , W and AB. None of these comparisons were s i g n i f i c a n t l y d i f f e r e n t . I believe that these s i n g l e comparison r e s u l t s are best explained by proposing that s t a b i l i z i n g s e l e c t i o n encourages a p a r t i c u l a r phenotype to f l o u r i s h . The genotypes which f a i l to compensate for environment differences experienced by the stocks are selected against. This system works best when i s o l a t i n g factors among stocks are e f f e c t i v e as among the AB and W populations. In the absence of s e l e c t i o n successful t r a i t d i f f e r e n t i a t i o n among populations requires that less than one i n d i v i d u a l per generation migrate among the populations (Spieth 1974). The s p a t i a l and temporal pattern of return of marked f i s h suggests that gene flow among these populations i s su b s t a n t i a l . In p a r t i c u l a r , the WB population receives migrants from both the AB and W populations. While AB and W have the most opportunity to diverge phenotypically they c l e a r l y have not i n many t r a i t s . The WB population appears to be the most unusual of the group in terms of phenotype. The pro b a b i l i t y of s u r v i v a l of an i n d i v i d u a l depends upon the a b i l i t y of h i s genetic program to compensate for the e f f e c t of the incubation environment to 113 produce the optimal phenotype. In complete i s o l a t i o n s e l e c t i o n w i l l remove un f i t genotypes so that the mean genotype of a population w i l l approach an adaptive peak. However, immigration of genes from other populations where the genetic program has adapted to compensate for a d i f f e r e n t incubation environment w i l l r e s u l t in movement away from the optimum. WB has the greatest p o t e n t i a l in-migration of less f i t genes from other populations r e l a t i v e to i t s population s i z e . The r e l a t i v e intermediacy of fry in morphology represents contrary evidence to the general model suggested for the system. I acknowledge that t h i s result i s puzzling but i t must also be considered r e l a t i v e to my hierarchy of importance of variables measured. I consider that the strongest predictions may be made regarding t r a i t s where the r e l a t i o n s h i p between temperature and development of that t r a i t i s simple and well understood. Thus, I order the r e s u l t s with respect to my a b i l i t y to generate a cle a r expectation to test as follows: (1) incubation rate and fry migration timing; (2) vertebral number comparisons; (3) morphological comparisons, age at return and egg size comparisons. The r e l a t i o n s h i p between rate of incubation and temperature i s simple and well understood. The r e l a t i o n s h i p between vertebral number and temperature of incubation i s complex but reasonable p r e d i c t i v e models e x i s t . In contrast, some r e l a t i o n s h i p between body morphology and temperature e x i s t s but p r e d i c t i v e models have not been developed. I believe that the propensity of ecologists to look for phenotypic differences in nature has prevented i n v e s t i g a t i o n of a s i z a b l e portion of the 114 genetic v a r i a t i o n among populations. Breeding groups i s o l a t e themselves from each other by u t i l i z i n g p a r t i c u l a r l o c a l s i t e s i n time and space for reproduction. Their progeny, having gone through embryonic development in d i f f e r e n t l o c a l environments, often mingle in a common s e l e c t i v e environment during the non-reproductive phases of t h e i r l i v e s . S t a b i l i z i n g s e l e c t i o n w i l l eliminate those genotypes which cannot compensate for the differences among the populations i n the embryonic incubation environment. Thus, the pressure for phenotypic uniformity w i l l r e s u l t i n genetic change among populations. Ecologists formulate t h e i r view of the world by measurement of phenotypes, often without consideration of the underlying processes which have generated those phenotypes (Stearns 1977). Yet, genetic differences among populations may be manifest by the absence of phenotypic difference rather than i t s presence. Populations of ectotherms spawning during d i f f e r e n t seasons d i f f e r greatly in the thermal incubation environments experienced by t h e i r progeny. Temperature of incubation i s a powerful environmental influence on many phenotypic characters such as time to emergence, vertebral number and external morphology. Many characters w i l l be constrained by s t a b i l i z i n g s e l e c t i o n so that i n d i v i d u a l s who can g e n e t i c a l l y compensate for the e f f e c t of temperature of incubation w i l l p e r s i s t . Seasonal races of chum salmon appear to diverge g e n e t i c a l l y i n incubation rate in response to constraints s t a b i l i z i n g the phenotype for downstream fry migration timing. I speculate that among seasonal races vertebral count and egg s i z e i s also phenotypically s t a b i l i z e d 115 by an underlying genetic compensation. These re s u l t s suggest that seasonal races may have a genetic basis. SECTION A 116 INNATE VERSUS ENVIRONMENTAL CONTROL OF PHENOTYPE AMONG SEASONAL ECOTYPES INTRODUCTION Reproduction i n temperate ectotherms occurs within r e s t r i c t e d time periods and locations r e l a t i v e to other a c t i v i t i e s . A s p e c i f i c place and season for reproduction characterizes each population. Deviations among populations in the season or location of reproduction w i l l reduce or prevent cross-mating and thus act as a b a r r i e r to gene flow (Mayr 1963). Although r e s t r i c t e d gene flow i s not a necessity for interpopulation divergence i t i s generally accepted that the reduction or lack of genetic migration among populations w i l l hasten l o c a l adaptation (Endler 1977). Depending on the degree and permanence of i s o l a t i o n among populations and the nature of s e l e c t i v e pressures acting upon them they may become separate species, semi-species or races (Saviaatova 1983). While the existence and veracity of geographic races or subspecies i s well documented (Mayr 1963) studies regarding genetic differences among "seasonal races" i n f i s h e r i e s are scant. This i s s u r p r i s i n g considering the vast l i t e r a t u r e on the ecology and genetics of salmonids and the following: 117 (a) embryos produced by all o c h r o n i c populations are l i k e l y to experience d i f f e r e n t temperature regimes;(b) -many c h a r a c t e r i s t i c s of poikilotherms are fixed early in development through combined temperature and genotypic e f f e c t s ; (c) s t a b i l i z i n g s e l e c t i o n in the environment w i l l encourage phenotypic uniformity. Genotypes which can compensate for the temperature differences to produce the optimum phenotype w i l l f l o u r i s h . Hence genetic differences should evolve among seasonal ecotypes. The evolutionary implications of allochrony are sweeping considering that many processes of the developing poikilothermic embryo from mitosis to behaviour are a function of temperature. Conservation of function, e n a n t i o s t a t i s , must be maintained even though temperature differences experienced by embryos i n i t i a t e d in d i f f e r e n t seasons may be comparable to thermal v a r i a t i o n throughout the l a t i t u d i n a l range of the species. Local adaptation to a p a r t i c u l a r thermal regime i s l i k e l y , given the s e l e c t i o n pressure for a harmonius, e f f i c i e n t development to a juvenile stage. The average incubation period to emergence of the WB and W populations was much shorter than that of AB even though AB embryos incubated at a much warmer average temperature than the embryos of the l a t e r spawning stocks. Phenotypic s i m i l a r i t y between AB and W in several other characters, such as vertebral count, egg s i z e , length at maturity and age at maturity, was observed even though the breeding environment d i f f e r e d . These observations suggest that genetic divergence i n incubation rate and perhaps vertebral count has evolved among seasonally d i s t i n c t spawning populations of chum salmon. On 118 the other hand, high straying rates among the populations could homogenize the gene pools. The d i f f e r e n t temperatures i n early development among the early and l a t e spawning stocks could switch development onto d i f f e r e n t pathways on the same genome. To d i s t i n g u i s h between these p o s s i b i l i t i e s one must compare the performance of the populations when reared under s i m i l a r environmental conditions. Within a standard environment there are three classes of factors that may af f e c t stock performance : 1) genotypic differences among stocks; 2) maternal e f f e c t s which may be genotypic or environmentally induced; and 3) genotype -environment i n t e r a c t i o n s . Okazaki (1981) suggested that s p a t i a l l y separated chum salmon stocks are g e n e t i c a l l y d i f f e r e n t . Genotypic differences have also been suggested for temporally separated stocks (Okazaki 1978, Kulikova 1971). The incubation rate of chum salmon has been suggested to be under maternal control through egg size differences among populations (Beacham and Murray 1985, Smoker 1982). In many cases the performance of populations have been compared using a singl e temperature regime ( i . e Smoker 1982). This can be misleading i f the r e l a t i v e performances of the populations depends on the temperature regime. Such in t e r a c t i o n may obscure r e l a t i o n s h i p s among the populations. For example, Beacham (1987) found s i g n i f i c a n t genotype-environment i n t e r a c t i o n i n growth of chum salmon. Levins (1963) suggested that under c e r t a i n conditions environmentally induced switches i n development should be favored over s t r i c t genetic control of morphology. Recent experimental studies indicate that such 119 developmental switches may be widespread ( L i v e l y 1986). Occasionally, population performance i s compared in a range of constant temperature regimes ( i . e. Beacham and Murray 1986). However, constant thermal regimes are rare in the natural environment. It i s important to simulate the natural condition which i s a seasonally varying regime. This i s e s p e c i a l l y important when comparing the performance of stocks spawning in d i f f e r e n t seasons. The purpose of t h i s chapter i s to determine i f the pattern of phenotypic v a r i a t i o n observed in the wild i s due to differences in the genomes of the populations. To achieve t h i s I reared progeny in the laboratory to determine the r e l a t i v e importance of genotypic, maternal and i n t e r a c t i v e (genotype-environment) factors i n the expression of phenotype. MATERIALS AND METHODS Experiment 1 - 1982-83 Eggs and milt were c o l l e c t e d from spawners near the peak of the spawning run in each population. Gametes were c o l l e c t e d on Oct. 20, 1982 from the AB stock, on Nov. 24, 1982 from the WB stock and on Dec. 15, 1982 from the W stock. As well, eggs were c o l l e c t e d from females of the WB stock in December to cross with W males. Five males and 5 females within each stock were mated 120 in a l l possible combinations to produce 25 f a m i l i e s . Crosses were made separately to ensure the widest v a r i a b i l i t y with the number of spawners av a i l a b l e . Pooling milt from many males can cause a lessening of genetic v a r i a t i o n due to sperm competition (Withler 1988). The spawn was transported to the laboratory on i c e . Eggs were f e r t i l i z e d at 8 C. The zygotes were water hardened for 2 hours, then the fa m i l i e s were pooled to form test populations. As well, 25 families generated from 5 WB females and 5 W males were combined to form a fourth t e s t population. Each group was reared under four temperature regimes: constant 6 c e l s i u s , constant 10 c e l s i u s , a simulated autumn - winter - spring temperature progression and a simulated winter - spring temperature progression. Each population temperature treatment was r e p l i c a t e d . A sample of 10 water-hardened eggs per female was preserved in Stockard's s o l u t i o n (Rugh 1952). The blot-dry weight of these eggs ( to the nearest 1.0 mg) was measured using a top-loading electrobalance. Embryos were reared on screens in c i r c u l a r 45 1 f i b e r g l a s s tanks. The embryos were exposed to the surrounding l i g h t conditions. Dechlorinated Naniamo tapwater was sprayed onto the surface of each tank at a rate of 1 to 1.5 L/min. Oxygen concentrations were above 90 % a i r sat u r a t i o n . Loading den s i t i e s averaged about 1000 eggs/tank. Water temperatures were recorded twice weekly. Temperatures were reset ( i f necessary) each week. Variation from the planned temperature regimes i s shown i n Figure 13. Mean temperatures for each c e l l are given in Table 28. 121 Dead eggs were removed from each tank and stored in Stockard's s o l u t i o n . Dead eggs were inspected to determine the stage at development at which they died according to Velsen's (1980) c l a s s i f i c a t i o n . With the onset of hatching I recorded the cumulative number of hatched alevins each day. Once hatch was completed the alevins were provided with a gravel substrate to a depth of about four cm. Fry that remained at or near the surface with the water o f f for 15 minutes were considered to have emerged. The cumulative numbers of these emerged fry were recorded d a i l y u n t i l there was no increase. Survival from hatch to emergence was calculated by subtracting the t o t a l emerged fry from the number in the tray at hatching. When 50 per cent of the hatchlings had emerged I selected 50 fry at random from the tank for l a t e r measurement and preserved them in 5 % buffered formalin. Experiment 2 - 1983-84 Experiment 2 was b a s i c a l l y a r e p l i c a t i o n of experiment 1 . However, the protocol was al t e r e d i n number of important ways: 1) Eggs and milt were c o l l e c t e d from spawners on October 19, 1983 from the AB stock, on December 4, 1983 from the WB stock and on December 9, 1983 from the W stock. 2) After f e r t i l i z a t i o n eggs were moved d i r e c t l y into t h e i r incubation tank rather than water hardening for 2 hours. This was done i n order to reduce mortality from v i b r a t i o n a l shock that the eggs might have experienced the 122 previous year. According to Jensen and Alderdice (1983) s e n s i t i v i t y of eggs to disturbance increases rapidly and s t e a d i l y after f e r t i l i z a t i o n . The eggs were then d i s i n f e c t e d using Erythromycin (Evelyn et a l . 1986). 3) The tanks were kept dark throughout the incubation period. 4) Pressure reducing valves were i n s t a l l e d so that incubation temperatures would not fluctuate as widely as they had in 1982-83. In 1982-83, temperature adjustments by other experimenters sharing the water l i n e caused pressure changes in the incoming l i n e s to my tanks. This would change the mix between warm and c h i l l e d water supplies and hence a l t e r the temperature at the tank. Pressure reducing valves squeeze the pressure to a set minimum so that pressure changes up the l i n e do not a f f e c t the pressure in the l i n e s coming into the lab. Hence, temperatures remained r e l a t i v e l y stable in the tanks during 1983-84 incubations (Figure 13, Table 28). 5) Temperatures were checked and reset d a i l y rather than weekly to ensure that the planned regimes were more c l o s e l y met. 6) After the embryos had hatched the incubation baskets were removed and inte r n a l stand-pipes were set in the tanks. Alevins were dropped into a 4 cm layer of gravel at the bottom of the tank. C o l l e c t i n g bottles were placed at the end of the tank drain pipe. Emerging fry had to move from the tank bottom to the top of the stand-pipe and then down the drain into the c o l l e c t i n g b o t t l e . Thus, alevins had to migrate much more v e r t i c a l l y and t r a v e l considerably further to be c l a s s i f i e d as emergent compared with 1982-83. A d a i l y count of fry appearing i n the c o l l e c t i n g b o t t l e provided a much less subjective measure of f r y emergence than in Experiment 1. 1 9 8 2 - 8 3 10 CELSIUS 1 9 8 3 - 8 4 10 CELSIUS L A B T E M P 16 15 14 13 12 11 10 9 • • • A B A A A W B o o o W B W * * W A 9 • O r!ri iT UJ 0 2 0 4 0 6 0 8 0 0 5 0 100 150 D A Y S F R O M FERTIL IZATION Figure 13. Incubation temperatures (C) of the 1982-83 and 1983-84 experiments. Solid lines represent the planned changes in temperature regime. 124 Table 28. experiments. Mean water temperatures i n each c e l l f o r (Standard Deviation i n Parentheses). 1982-83 and 1983-84 Population- Temperature Temperature Treatment 1982- 33 1983--84 AB - 6 6.1 (0.73) 6.0 (0.20) AB - 10 10.8 (1.79) 10.0 (0.14) AB - EARLY 4.8 (1.78) 4.8 (1.59) AB - LATE 6.2 (2.24) 6.2 (2.32) WB - 6 6.7 (1.34) 5.9 (0.14) WB - 10 10.9 (2.14) 9.9 (0.17) WB - EARLY 4.5 (1.73) 4.8 (1.62) WB - LATE 6.1 (2.21) 6.0 (2.26) WBxW - 6 6.0 (0.41) 6.1 (0.20) WBxW - 10 10.4 (1.58) 10.0 (0.21) WBxW - EARLY 4.6 (1.66) 4.8 (1.63) WBxW - LATE 6.2 (2.26) 5.8 (2.19) W - 6 6.0 (0.19) 6.0 (0.19) W - 10 10.0 (3.27) 10.0 (0.14) W - EARLY 4.8 (1.76) 4.8 (1.60) W - LATE 5.9 (2.14) 5.6 (2.16) 125 7) General water system shutdowns did not occur during 1983-84. There were three in 1982-83 r e s u l t i n g in sharp temperature spikes i n the incubation tank temperatures. S t a t i s t i c a l Analysis The i n d i v i d u a l hatch and emergence time of embryos in the same tank may not be independent. Thus, to avoid pseudoreplication, (Hurlbert 1984) I w i l l use the mean hatch and emergence times in each tank as the measure of in t e r e s t . Differences observed between years could be due to b i o l o g i c a l factors or more l i k e l y due to technical differences in the experiments. Therefore I chose to analyze them separately using a two way f a c t o r i a l analysis of variance. The model used i s as follows: Y i j k = u + Ai + Bj + ABij + Ei jk where Y - mean days to hatch or emergence u = the parametric mean of the mean days hatch or emergence A = e f f e c t of population, i = 1, 4 B = e f f e c t of temperature regime j = 1, 4 AB r population by temperature regime i n t e r a c t i o n E = error term for the k'th observation in tank i j k 126 I used the Sum of Squares Simultaneous Test Procedure (SS-STP) (Gabriel 1964) to compare populations within temperature regimes and temperature regimes within population. This procedure i s conservative with respect to the pr o b a b i l i t y of Type I error (Sokal and Rohlf 1981). A l l possible comparisons and contrasts between group means may be tested. The 'a p r i o r i 1 comparisons of in t e r e s t were between AB and WB representing a comparison of temporally separated populations within the same stream; AB and W representing a comparison between two populations with both temporal and s p a t i a l i s o l a t i o n ; WB and W representing a comparison between temporally s i m i l a r , s p a t i a l l y separated populations; AB versus the average of WB and W combined, a contrast between the early spawning population and the lat e spawning populations; W versus the average of AB and WB, a contrast of the population in one creek against the populations i n the other creek, WB versus the average of AB and W, a contrast between the population with small eggs and those with large eggs; and comparisons of the cross, WBxW, with each of the donors stocks WB and W as well as t h e i r combined average. Fluc t a t i o n s i n temperature could bias comparisons between populations within each temperature regime. Mean days to hatch and emergence were adjusted to account for temperature differences by multiplying the recorded number of days by the r a t i o of the expected mean temperature time the observed mean temperature. The adjusted mean days to hatch and emergence were used in the comparisons. 127 Survival For the purpose of s u r v i v a l estimates the dead embryos were grouped into three major i n t e r v a l s of development from f e r t i l i z a t i o n to hatching: f e r t i l i z a t i o n to epiboly; epiboly to eye pigment stage; eye pigment stage to hatch (Velsen 1980). Survival from hatching to emergence was also estimated. I used the G-test (Sokal and Rohlf 1981)to determine i f s u r v i v a l from f e r t i l i z a t i o n to hatch and hatch to emergence was dependent on temperature regime, population, season of reproduction, location of spawning, or egg s i z e . Vertebral Counts Vertebral counts were read from X-rays of the preserved fry from the 1983-84 experiment. I decided i t was unwise to compare vertebral counts of the fry from the 1982-83 experiment because temperature control was poor. Also, high m o r t a l i t i e s i n some c e l l s increased the p o s s i b i l i t y that a non-representative sample would be obtained. The r e l i a b i l i t y of the vertebral counts was checked against X-ray vertebral counts of fry reared for several weeks (size range 50-70 cm) and against cleared and stained specimens. Preliminary t e s t s indicated that tank e f f e c t s were minimal. Therefore, I combined samples from r e p l i c a t e s . The e f f e c t of population, temperature regime and t h e i r i n t e r a c t i o n on the vertebral counts of emergent fry was 128 estimated using a two way f a c t o r i a l analysis of variance. The model used i s as follows: Yijt< = u + Ai + Bj + " ABij + Ei jk where Y = vertebral count of f i s h k reared under temperature regime j from population i u = the parametric mean vertebral count A = e f f e c t of population, i = 1, 4 B = e f f e c t of temperature regime j = 1, 4 AB = population by temperature regime i n t e r a c t i o n E = error term for the k'th observation on f i s h i j k I used the Scheffe's Method to compare populations within temperature regimes. The 'a p r i o r i 1 comparisons of inte r e s t were between AB and WB representing a comparison of temporally separated populations within the same stream; AB and W representing a comparison between two populations with both temporal and s p a t i a l i s o l a t i o n ; WB and W representing a comparison between temporally s i m i l a r , s p a t i a l l y separated populations; AB versus the average of WB and W combined, a contrast between the early spawning population and the late spawning populations; W versus the average of AB and WB, a contrast of the population in one creek against the populations i n the other creek, WB versus the average of AB and W, a contrast between the population with small eggs and those with large eggs. 129 External Morphology For the purpose of comparing external morphology of emergent fry among the populations the following procedure was used: 25 f i s h were chosen alternately from preserved samples of each r e p l i c a t e tank. I measured t o t a l length (TL), standard length (STL), head length (HL), snout length (SNL), pectoral f i n length (PFL), eye diameter (ED), head depth (HD), body depth (BD), and wet weight (WT) as described by Hubbs and Lagler (1958) plus the number of parr marks per f i s h . The length measurements were to the nearest 0.1 mm. WT was measured to the nearest 0.1 gram. A parr mark was defined a r b i t r a r i l y as any disc r e t e v e r t i c a l bar of dark pigment exceeding 40 per cent of the distance from the ventral to dorsal surface of the f i s h . I used t h i s c r iterium to d i s t i n g u i s h parr marks from spots. As with the vertebral counts I do not report external morphology from the 1982-83 experiment because temperature control was unsatisfactory. Stepwise discriminant analysis was used to determine the most important t r a i t s for separating the populations as well as estimating the degree of separation. The e f f e c t of incubation environment on the separation among the populations was compared by discriminant analysis of the sample populations s t r a t i f i e d by temperature regime. Variables were added into the discriminant function u n t i l the 'multiple c o r r e l a t i o n coefficient,R2> of each of the remaining variables with those already entered was greater than 0.40. 130 RESULTS Time to Hatch The frequency histograms of time to hatch among the populations within each temperature regime for the 1983-84 experiments are shown i n Figures 14 and 15 respectively. In the 1982-83 experiment both population ( P < 0.0001 ) and temperature regime ( P < 0.0001) had s i g n i f i c a n t e f f e c t on time to hatch. However, population by temperature regime i n t e r a c t i o n occurred ( P < 0.0001 ) and therefore i n t e r p r e t a t i o n of each main e f f e c t i s conditional on the state of the other factor (Table 28). S i m i l a r l y , in the 1983-84 experiment population ( P < 0.0001 ), temperature regime ( P < 0.0001) and population by temperature regime in t e r a c t i o n (P < 0.0001) had s i g n i f i c a n t e f f e c t s on time to 50 % hatch. Thus, here also, i n t e r p r e t a t i o n of each main e f f e c t i s conditional on the state of the other factor (Table 30). Within each temperature treatment the mean number of days required for the early spawning stock, AB, and the l a t e spawning stock of the same creek, WB, to reach 50 % hatch did not d i f f e r s i g n i f i c a n t l y except at 6 C and under the winter-spring regime in the 1982-83 experiment (Tables 29, 30, and 31). However, i n both of the s i g n i f i c a n t r e s u l t s the AB stock required more days to 131 T I M E TO H A T C H AT 6 C E L S I U S FREQUENCE . 300 -, 700 -600 -500 -T I M E TO H A T C H A T 10 C E L S I U S FREQUENCY 800 -, , 700 -600 -500 -1982-83, BENCH SIDE Figure 14. Time to hatch under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime for AB, WB, W, and W x WB in the 1982-83 experiment. . 132 T I M E T O H A T C H A T S I M U L A T E D W I N T E R S P A W N E D R E G I M E FREQUENCY 300 -, • l l . _96. 100_..J04 108 112 116 120 124 128 132 136 T I M E TO H A T C H A T S I M U L A T E D A U T U M N S P A W N E D R E G I M E FREQUENCY 800 700 -M l 70 75 80 85 90 95 100 105 110 115 120 125 DAY M I D P O I N T P 0 P • • • ' 2 [ X X X I 3 ( 7 7 7 ] 4 1982-83, BENCH SIDE gure 14 continued 133 T I M E T O H A T C H A T 6 C E L S I U S FREQUENCY 800 68 72 76 80 84 92 96 100 104 108 T I M E T O H A T C H A T 10 C E L S I U S FREQUENCY 800 POP DAY MIDPOINT 2 CXXXl 3 1982-83, WALL SIDE 17771 4 Figure 14 continued 1 3 4 TIME TO HATCH AT SIMULATED WINTER SPAWNED REGIME FREQUENCY 800 -i 700 600 J TIME TO HATCH AT SIMULATED AUTUMN SPAWNED REGIME FREQUENCY 800 -I • 700 -600 -500 A 5 0 5 0 5 0 5 6 5 6 5 ° 6 DAY MIDPOINT POP WmfM , 2 3 C Z Z 2 4 1982-83, WALL SIDE Figure 14 continued 135 T I M E TO H A T C H AT 6 C E L S I U S FREQUENCY 800 700 600 -500 -400 V I * 78 80 82 84 86 90 92 94 96 98 100 T I M E TO H A T C H AT 10 C E L S I U S FREQUENCY 800 700 -POP 5 0 5 0 DAY MIDPOINT ' O Q 2 E X S 2 1983-84, BENCH SIDE 6 7 7 7 7 0 2 5 f 7 7 7 ] 4 Figure 15. Time to hatch under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime for AB, WB, W, and W x WB in the 1983-84 experiment . 136 TIME TO HATCH AT SIMULATED WINTER SPAWNED REGIME FREQUENCY 800 -, 700 -T I M E T O H A T C H A T SIMULATED AUTUMN SPAWNED REGIME FREQUENCY 800 -i 700 -68 72 76 80 84 88 92 96 100 104 108 112 DAY MIDPOINT POP' • 1 CZXZ2 2 SSS 3 r7771 4 1983-84, BENCH SIDE Figure 15 continued 137 TIME TO H A T C H AT 6 C E L S I U S FREQUENCY 800 -, 700 -600 -500 -TIME TO H A T C H AT 10 C E L S I U S FREQUENCY 800 -, 48 51 54 57 60 63 66 69 72 75 78 DAY MIDPOINT p mm 1 c m 2 E S S 3 LZZZI 4 1983-84, WALL SIDE gure 15 continued 138 TIME TO HATCH AT SIMULATED WINTER SPAWNED REGIME FREQUENCY 300 -i TIME TO HATCH AT SIMULATED AUTUMN SPAWNED REGIME FREQUENCY 800 - i 700 -600 -7 2 75 78 81 84 87 90 93 96 99 102 105 DAY MIDPOINT P 0 P mm 1 2 reXXI 3 r7771 4 1983-84. WALL SIDE Figure 15 continued 139 Table 29. Mean hatch time f o r pop u l a t i o n by temperature regime treatments i n 1982-83 experiment. (Standard d e v i a t i o n s are i n parentheses). Temperature Regime Population 6 10 EARLY LATE AB 101.7 (0.01) 56.0 (0.95). 84.4 (0.72) 128.1 (0.22) WB 78.4 (1.31) 57.3 (1.74) 86.9 (2.15) 114.3 (1.40) WBxW 88.7 (3.50) 59.4 (4.00) 95.9 (5.27) 113.2 (2.00) W 84.1 (0.72) 53.7 (0.74) 84.2 (1.46) 104.5 (1.34) Table 30. Mean hatch time f o r p o p u l a t i o n by temperature regime treatments i n 1983-84 experiment. (Standard d e v i a t i o n s are i n parentheses). Temperature Regime Population 6 10 EARLY LATE AB 96.2 (3.07) 54.4 (0.52) 94.4 (1.23) 130.7 (0.69) WB 87.8 (1.72) 64.4 (0.66) 86.0 (1.00) 125.2 (1.07) WBxW 89.1 (6.47) 74.0 (0.83) 100.5 (5.90) 113.5 (0.09) W 83.7 (0.84) 58.7 (0.93) 80.4 (0.06) 107.6 (0.37) 140 Table 31. R e s u l t s o f comparisons and c o n t r a s t s o f mean time t o 50 % hatch among the t e s t populations reared w i t h i n each temperature regime. Mean time t o hatch used i n comparisons was adjusted f o r temperature v a r i a t i o n . A plu s (+) i n d i c a t e s t h a t the population t o the l e f t of the minus s i g n took more time reach 50 55 hatch than the population or average of two populations t o the r i g h t . (An a s t e r i s k (*) i n d i c a t e s that P < 0.05. Two a s t e r i s k s (**) i n d i c a t e t h a t P < 0.01). TEMPERATURE REGIME 6 10 EARLY LATE Comparsion 1982 1983 1982 1983 1982 1983 1982 1983 AB - WB + * N.S. N.S. N.S. N.S. N.S. + * N.S. AB - W + ** + * N.S. N.S. N.S. + * + ** WB - W N.S. N.S. N.S. N.S. N.S. N.S. + * + ** AB - (WB + W) 2 + ** + * N.S. N.S. N.S. + * + * + ** W - (AB + WB) 2 N.S. N.S. N.S. N.S. N.S. _ * _ •** _ * WB - (AB + W) 2 N.S. N.S. N.S. N.S. N.S. N.S. N.S. N.S. (WBxW) - WB N.S. N.S. ' N.S. N.S. + * + * N.S. + * (WBxW) - W N.S. N.S. N.S. + * N.S. + ** N.S. N.S. (WBxW) - (WB + W) N.S. N.S. N.S. + * + * + ** N.S. N.S. 2 141 reach 50 % hatch. As well, the early spawning stock required s i g n i f i c a n t l y more days to reach 50 % hatch than the W stock and the combined average of the l a t e spawning stocks under a l l temperature treatments except 10 C and the autumn-winter-spring temperature regime of the 1982-83 experiment. This suggests that the early spawning stock has an i n t r i n s i c a l l y slower incubation rate to hatch than the l a t e spawning stocks at least under conditions of cool temperatures during the early stages of embryonic development. In experiments at 6 C, 10 C, and under the autumn-winter-spring temperature regime there was no s i g n i f i c a n t difference in the number of days to 50 % hatch between the two late spawning stocks, WB and W. However, the WB stock required s i g n i f i c a n t l y more days to reach 50 % emergence than the W stock when reared under the winter-spring temperature regime. S i m i l a r l y , the W stock required fewer days to reach 50 % emergence than the combined average of the Bush Creek stocks under the autumn-winter-spring regime in the 1983-84 experiment and under the winter-spring regime in both experiments. The more rapid incubation rate to hatch of the Walker Creek stock compared to the Bush Creek stocks under the winter-spring regime may r e f l e c t an adaptation for incubating at cold temperatures during early development. Time to Emergence The frequency histograms of time to emergence among the populations within each temperature regime for the 1982-83 and 1983-84 experiments are shown i n Figure 16 and 17. In the 1982-83 experiment both population ( P < 142 T I M E T O E M E R G E N C E A T (i C E L S I U S FREQUENCY 300 -j 700 -600 -500 -T I M E T O E M E R G E N C E A T 10 C E L S I U S FREQUENCY 800 700 -600 -1982-83, 8ENCH SIDE Figure 16. Time to emergence under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime for AB, WB, W, and W x WB in the 1982-83 experiment .' 143 TIME TO EMERGENCE AT SIMULATED WINTER SPAWNED REGIME FREQUENCY 300 -, 700 • 600 • 500 -TIME TO EMERGENCE AT SIMULATED AUTUMN SPAWNED REGIME FREQUENCY 800 -, 700 -600 -500 -140 144 148 152 156 160 164 168 172 176 180 184 DAY MIDPOINT . POP • • • 1 C X J 2 . E S S 3 17771 4 1982-83, BENCH SIDE Figure 16 continued 144 T I M E T O E M E R G E N C E A T 6 C E L S I U S FREQUENCY 800 -, 700 -600 -500 -120 124 128 132 136 140 144 148 152 156 DAY MIDPOINT POP • • • 1 2 C S S 3 LZZZJ 4 1 9 8 2 - 8 3 , WALL S I D E 7 0 0 -6 0 0 -5 0 0 -7 7 8 8 8 8 8 8 8 9 6 8 0 1 3 4 6 8 9 1 8 4 0 6 2 8 4 0 6 2 DAY M I D P O I N T POP mmm 1 2 c v s x i 3 r z z z j 4 1982-33, WALL S IDE gure 16 continued 145 TIME TO EMERGENCE AT SIMULATED WINTER SPAWNED REGIME FREQUENCY 800 H O 144 148 152 156 160 164 168 172 176 130 TIME TO EMERGENCE AT SIMULATED AUTUMN SPAWNED REGIME FREQUENCY 800 700 -600 -500 400 300 200 -100 -0 1 JZZL d i l l If-*-* 136 140 144 148 152 156 160 164 168 172 176 180 DAY MIDPOINT P O P mm i a a 2 ' E S S 3 r z z a 4 1 9 8 2 - 8 3 . WALL SIDE Figure 16 continued 146 T I M E TO E M E R G E N C E AT (3 C E L S I U S FREQ'iENC • T I M E TO E M E R G E N C E AT 10 C E L S I U S FREQUENCY 1983-34 , BENCH SIDE Figure 17. Time to emergence under 6°C, 10°C, simualted autumn spawning regime, simulated winter spawning regime for AB, WB, W and W x WB in the 1983-84 experiment . 147 IMK TO E M E R G E N C E AT S I M U L A T E D WINTER SPAWNED R E G I M E FREQUEHC <: _ 1 4 4 147 I50 153 156 159 162 165 168 ...171 174 177 T I M E T O E M E R G E N C E A T S I M U L A T E D A U T U M N S P A W N E D R E G I M E FREQUENCY 800 700 -600 -500 1 2 8 1 3 2 '36 140 144 148 152 156 160 164 168 172 DAY MIDPOINT POP • • • 1 Q K 3 2 ' CV^Cl 3 [ 7 7 7 1 4 1933-84 , BENCH SIDE Figure 17 continued 148 T I M E T O E M E R G E N C E A T 0 C E L S I U S FREQUENCY 300 -, T I M E T O E M E R G E N C E A T 1 0 C E L S I U S FREQUENCY 800 T 106 108 110 112 114 116 118 120 122 124 126 128 DAY MIDPOINT p H i O Q 2 3 LZZZJ 4 1 9 8 3 - 8 4 , WALL SIDE Figure 17 continued 149 TIME TO EMERGENCE AT .-SIMULATED WINTER SPAWNED REGIME r REO'JENC'.' aoc -, 700 -riME TO EMERGENCE AT SIMULATED AUTUMN SPAWNED REGIME FREQUENCY 1983 -84 , WALL SIDE Figure 17 continued 150 0.0001) and temperature regime (P < 0.0001) had s i g n i f i c a n t e f f e c t on time to emergence. However, population by temperature regime i n t e r a c t i o n occurred (P < 0.0001) and therefore interpretation of each main e f f e c t i s conditional on the state of the other factor (Table 32). S i m i l a r l y , population (P < 0.0001), temperature regime (P < 0.0001) and population by temperature regime i n t e r a c t i o n (P < 0.001) had a s i g n i f i c a n t e f f e c t on time to 50 % emergence in the 1983-84 experiment. Thus, in t e r p r e t a t i o n of each main e f f e c t i s conditional on the state of the other factor (Table 33). The early stock, AB, required s i g n i f i c a n t l y more days to reach 50 % emergence than the la t e stock, WB, of the same stream under the autumn-winter-spring temperature regime, the winter-spring regime and the 6 C regime i n the 1982 experiment (Tables 32, 33 and 34). The mean time to 50 % emergence did not d i f f e r among AB and WB in the 10 C treatment. Also, the early stock needed s i g n i f i c a n t l y more days to reach 50 % emergence compared to the l a t e stock of Walker Creek, W, under a l l temperature treatments except the 10 C treatment in the 1982-83 experiment. S i m i l a r l y , the early stock took s i g n i f i c a n t l y longer to reach 50 % emergence than the combined average of the lat e stocks under a l l temperature treatments except the 10 C treatment i n the 1982-83 experiment. In general, when both stocks are reared under the same temperatures, the early spawning stock required more days to reach 50 % emergence than the l a t e stocks. 151 Table 32. Mean emergence time f o r population by temperature regime treatments i n 1982-83 experiment. (Standard d e v i a t i o n s are i n parentheses). Temperature Regime Population 10 EARLY LATE AB 152.4 (2.44) WB 126.8 (2.29) WBxW 146.7 (2.09) W 135.3 (0.41) 88.7 (0.71) 86.4 (0.03) 87.9 (2.53) 83.0 (1.74) 168.5 (0.56) 169.0 (1.98) 170.7 (3.53) 149.2 (0.19) 173.2 (0.03) 155.7 (0.77) 163.0 (2.26) 147.2 (0.86) Table 33. Mean emergence time f o r pop u l a t i o n by temperature regime treatments i n 1983-84 experiment. (Standard d e v i a t i o n s are i n parentheses). . Temperature Regime Population 6 10 EARLY LATE AB 156.5 (3.02) 121.4 (1.35) 166.7 (0.27) 173.4 (0.20) WB 149.6 (3.89) 122.7 (1.06) 149.7 (4.08) 159.7 (1.08) WBxW 136.4 (1.99) 109.9 (0.15) 149.6 (4.13) 152.5 (2.29) W 133.8 (3.42) 110.6 (0.11) 150.7 (1.55) 146.3 (0.02) 152 Table 34. R e s u l t s o f comparisons and c o n t r a s t s o f mean time t o 50 % emergence among the t e s t populations reared w i t h i n each temperature regime. Mean time t o emergence used i n comparisons was adjusted f o r temperature v a r i a t i o n . A p l u s (+) i n d i c a t e s that the popu l a t i o n t o the l e f t o f the minus s i g n took more time reach 50 % emergence than the popu l a t i o n or average o f two populations t o the r i g h t . (An a s t e r i s k (*) i n d i c a t e s t h a t P < 0.05. Two a s t e r i s k s (**) i n d i c a t e that P < 0.01). TEMPERATURE REGIME 6 10 EARLY LATE Comparsion 1982 1983 1982 1983 1982 1983 1982 1983 AB - WB + * N.S. N.S. N.S. + * + ** + ** + * AB - W + *•* N.S. + * + ** + ** + ** WB - W + * + * N.S. + * N.S. + * + * AB - (WB + W) 2 N.S. + * + * + ** W - (AB + WB) 2 N.S. _ * N.S. _ * _ * N.S. _ _ WB - (AB + W) 2 N.S. N.S. N.S. N.S. N.S. _ * N.S. N.S. (WBxW) - WB N.S. _ * N.S. _ * + * N.S. + * N.S. (WBxW) - W + * N.S. N.S. N.S. + * N.S. N.S. N.S. (WBxW) - (WB + W) + * N.S. N.S. N.S. N.S. N.S. 2 153 WB required s i g n i f i c a n t l y more days to reach 50 % emergence compared to the other la t e stock, W, under a l l temperature treatments except the 10 C treatment in the 1982-83 experiment and the autumn-winter-spring treatment in the 1983-84 experiment. The Walker Creek stock also required s i g n i f i c a n t l y fewer days to reach 50 % emergence than the combined average number of days required by the stocks in Bush Creek, WB and AB under the winter-spring progression i n both experiments, under 6 and 10 C in the 1983-84 experiment, and under the autumn-winter-spring temperature regime in the 1982-83 experiment. This suggests that stocks spawning in d i f f e r e n t locations have i n t r i n s i c a l l y d i f f e r e n t incubation rates. However, the d i r e c t i o n of the differences corresponds to the differences in time of spawning among the stocks. The l a t e s t stock, W, requires the fewest number of days to reach 50 % emergence. Overall, the most consistent difference occurred between populations that were s p a t i a l l y and temporally i s o l a t e d or spawned at d i f f e r e n t times (Table 35). S u r v i v a l , 1982-83 Survival from f e r t i l i z a t i o n to hatching was variable for the populations and temperature regimes. Survival rates varied from a low of 2-49 % of WB embryos surviving under the "Late Spawned" regime to a high of 99% of W embryos surviving under the 6 Celsius regime. M o r t a l i t i e s occurred throughout 154 Table 35. Summary o f the number of s i g n i f i c a n t comparisons out of s i x t e e n f o r time t o hatch and time t o emergence i n the 1982-83 and 1983-84 experiments. Comparison B i o l o g i c a l Meaning S i g n i f i c a n t Comparisons Direction AB - WB Temporal Is o l a t i o n i n 7 + Same Stream AB - W Temporal and Spatial 12 + WB - W S p a t i a l Isolation 8 + Similar Timing WB - C 1 Parent vs Cross 6 0 W - C 1 Parent vs Cross 4 0 1 AB - - (WB + W) Early vs Late Spawning 12 + 2 1 W - - (AB + WB) Walker vs Bush Creek 8 2 1 WB - - (W + AB) Small Eggs vs Large Eggs 1 1 WBxW - - (W + WB) Cross vs Average of Parents 6 2 155 the progression from f e r t i l i z a t i o n to epiboly, epiboly to eye pigment stage and eye pigment stage to hatch (Table 36, Figure 18). At 6 C most of the egg mortality in the AB and WBxW populations occurred from the eye pigment stage to hatch (77 % and 70 %, re s p e c t i v e l y ) . Survival was f a i r l y constant from f e r t i l i z a t i o n to hatch in the WB and W populations ranging from 91.5-95 % in the former and 99.5-100 % in the l a t t e r . At 10 C, under the "early spawned" regime and under the " l a t e spawned" regime most of the egg mortality i n the WB and WBxW populations occurred from eye pigment stage to hatch. Survival was f a i r l y constant from f e r t i l i z a t i o n to hatch i n the AB and W populations at 10 C and under the "early spawned" regime. Sur v i v a l was also f a i r l y constant for W under the " l a t e spawned" regime but was lowest from eye pigment stage to hatch for AB. For f e r t i l i z a t i o n to epiboly, epiboly to eye pigment stage and eye pigment stage to hatch s u r v i v a l in a l l populations was dependent on the temperature of incubation (P < 0.01). Survival at each temperature was dependent on the population, location of spawning and egg size (P < 0.01). Generally, s u r v i v a l depended on season of spawning (P < 0.01) except for f e r t i l i z a t i o n to epiboly at 10 C, epiboly to eye pigment stage under the "early spawned" regime and eye pigment stage to hatch at 10 C. Ov e r a l l , from f e r t i l i z a t i o n to hatch s u r v i v a l was dependent upon temperature regime, population, season of spawning, lo c a t i o n of spawning and egg size (P < 0.01). Survival was independent of season of spawning under the early temperature regime. 156 Table 36. S u r v i v a l o f chum salmon f o r each p o p u l a t i o n , temperature regime, tank and year during d i f f e r e n t segments o f embryonic development from f e r t i l i z a t i o n t o emergence. (1) s P R YR N f-ep N2 ep-ey N3 ey-h f-h N4 h-e N5 1 1 82 759 0.86 653 0.80 522 0.77 0.53 402 0.98 394 2 1 1 82 522 0.93 486 0.91 442 0.77 0.65 340 0.97 330 1 2 1 82 934 0.96 896 0.97 870 0.92 0.86 800 0.96 768 2 2 1 82 814 0.87 708 0.86 609 0.98 0.73 597 0.62 370 1 3 1 82 495 0.99 490 0.89 436 0.70 0.62 305 1.00 305 2 3 1 82 337 0.98 330 0.87 287 0.70 0.60 201 1.00 201 1 4 1 82 1006 0.99 996 1.00 996 1.00 0.99 996 1.00 996 2 4 1 82 1019 1.00 1019 0.99 1009 1.00 0.99 1009 1.00 1009 1 1 2 82 657 0.95 624 0.94 586 0.95 0.85 557 0.98 546 2 1 2 82 768 0.97 745 0.91 678 0.97 0.86 658 0.95 625 1 2 2 82 543 0.93 505 0.99 500 0.81 0.75 405 1.00 405 2 2 2 82 397 0.90 357 1.00 357 1.00 0.90 357 1.00 357 1 3 2 82 410 0.95 389 0.97 378 0.54 0.50 204 1.00 204 2 3 2 82 548 0.80 439 0.94 412 0.98 0.74 404 0.99 400 1 4 2 82 1018 0.98 998 1.00 998 1.00 0.98 998 1.00 998 2 4 2 82 1039 1.00 1039 0.99 1029 1.00 0.99 1029 1.00 1029 1 1 3 82 559 0.96 537 0.97 521 0.98 0.91 510 0.98 500 2 1 3 82 857 0.94 806 0.95 765 0.98 0.88 750 1.00 750 1 2 3 82 566 0.97 549 0.96 527 0.77 0.72 406 1.00 406 2 2 3 82 542 0.88 477 0.90 429 0.93 0.74 399 1.00 399 1 3 3 82 349 0.86 300 1.00 300 0.58 0.50 174 0.92 160 2 3 3 82 429 0.78 335 0.83 278 0.80 0.52 222 0.90 200 1 4 3 82 1007 1.00 1007 0.99 997 0.99 0.98 987 1.00 987 2 4 3 82 1020 0.99 1010 1.00 1010 1.00 0.99 1010 1.00 1010 1 1 4 82 814 0.88 716 1.00 716 0.57 0.50 408 0.98 400 2 1 4 82 946 0.61 577 1.00 577 0.70 0.43 404 0.99 400 1 2 4 82 624 0.50 312 1.00 312 0.98 0.49 306 1.00 306 2 2 4 82 502 0.81 407 0.06 24 0.41 0.02 10 1.00 10 1 3 4 82 403 0.96 387 0.52 201 0.76 0.37 153 0.98 150 2 3 4 82 777 0.90 700 0.52 364 0.85 0.40 309 0.97 300 1 4 4 82 991 0.94 932 0.93 866 0.92 0.80 797 1.00 797 2 4 4 82 996 0.80 797 0.81 645 1.00 0.65 645 0.95 613 1 1 1 83 1050 0.98 1029 0.99 1018 1.00 0.97 1018 0.98 998 2 1 1 83 1068 0.99 1057 1.00 1057 1.00 0.99 1057 0.96 1015 1 2 1 83 1054 0.98 1033 1.00 1033 0.98 0.96 1012 1.00 1012 2 2 1 83 1046 0.99 1036 0.99 1025 0.99 0.97 1015 1.00 1015 1 3 1 83 472 0.84 397 0.98 389 0.36 0.30 140 0.40 56 2 3 1 83 622 0.57 354 0.58 206 0.76 0.25 156 0.32 50 1 4 1 83 1010 1.00 1010 1.00 1010 0.99 0.99 1000 1.00 1000 2 4 1 83 1016 1.00 1016 0.98 996 1.00 0.98 996 1.00 996 1 1 2 83 1029 0.99 1018 0.99 1008 0.99 0.97 998 1.00 998 2 1 2 83 1057 0.98 1036 1.00 1036 0.98 0.96 1015 1.00 1015 1 2 2 83 947 0.98 928 1.00 928 1.00 0.98 928 0.97 900 2 2 2 83 1058 0.98 1037 0.99 1026 0.99 0.96 1016 0.99 1006 1 3 2 83 542 0.83 450 0.99 446 0.66 0.54 294 0.17 50 2 3 2 83 563 0.77 434 0.75 325 0.29 0.17 94 0.53 50 157 Table 36. S u r v i v a l o f chum salmon f o r each p o p u l a t i o n , temperature regime, tank and year during d i f f e r e n t segments of embryonic development from f e r t i l i z a t i o n t o emergence. (1) Continued S P R YR N f-ep N2 ep-ey N3 ey-h f-h N4 h-e N5 1 4 2 83 1021 1.00 1021 1.00 1021 0.98 0.98 1001 1.00 1001 2 4 2 83 1047 1.00 1047 1.00 1047 0.98 0.98 1026 1.00 1026 1 1 3 83 1030 0.99 1020 1.00 1020 1.00 0.99 1020 0.97 989 2 1 3 83 1045 0.99 1035 1.00 1035 0.99 0.98 1024 0.98 1004 2 4 3 83 1022 1.00 1022 0.99 1012 0.98 0.97 992 1.00 992 1 1 4 83 1048 0.97 1016 0.98 996 1.00 0.95 996 1.00 996 2 1 4 83 1030 0.98 1009 1.00 1009 1.00 0.98 1009 0.99 999 1 2 4 83 1026 0.99 1015 1.00 1015 1.00 0.99 1015 0.98 995 2 2 4 83 1056 0.98 1035 1.00 1035 1.00 0.98 1035 0.99 1025 1 3 4 83 667 0.50 333 0.50 167 0.48 0.12 80 0.25 20 2 3 4 83 583 0.50 292 0.40 117 0.50 0.10 58 0.60 35 1 4 4 83 1023 0.98 1003 1.00 1003 1.00 0.98 1003 1.00 1003 2 4 4 83 1038 0.99 1028 1.00 1028 1.00 0.99 1028 0.98 1007 (1) (S = Tank) (P r Population: 1 = AB; 2 = WB; 3 = WBxW; 4 = W) (R r Temperature Regime: 1 = 6 C ; 2 = 10 C ; 3 = 'Early'; 4 r 'Late') (Stages: f-ep = f e r t i l i z a t i o n to epiboly; ep-ey = epiboly to eye pigment stage; ey-h = eye pigment stage to hatch; f-h = f e r t i l i z a t i o n to hatch; h-e = hatch to emergence) SURVIVAL IN LABORATORY EXPERIMENTS FROM FERTILIZATION TO EPIBOLY 1982-83 Figure 18. S u r v i v a l under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime for AB WB W, and W x WB i n the 1982-83 experiment from fertilLati to epiboly, epiboly to eyed and eyed to hatch. SURVIVAL IN LABORATORY EXPERIMENTS FROM EPIBOLY TO EYED, 1982-83 Figure 18 continued SURVIVAL IN LABORATORY EXPERIMENTS FROM EYED TO HATCH, 1982-83 Figure 18 continued 161 Survival from hatching to emergence was generally higher than from f e r t i l i z a t i o n to hatch. Survival of alevins to emergence was variable for populations and temperature regime but was generally less variable than from f e r t i l i z a t i o n to hatching. Survival rates varied from 62-96 % for WB at 6 C to 100% for WBxW and W at 6 C, WB and W at 10 C, W under the "early spawned" regime and WB under the " l a t e spawned" regime (Figure 20) Survival from hatching to emergence for a l l populations was dependent upon temperature regime (P < 0.01). Survival from hatching to emergence at each temperature regime was dependent upon population, season of spawning (except under the " l a t e spawned" regime), lo c a t i o n of spawning and egg s i z e (P < 0.01). S u r v i v a l , 1983-84 Survival from f e r t i l i z a t i o n to hatch r e l a t i v e l y constant among the within population matings (groups where s i r e s and dams were from the same population). Survival ranged from a minimum of 96.5 % for the WB population at 6 C and the AB population at 10 C and under the " l a t e spawned" regime compared to a maximum of 98.5 % shared by W at 6 C and the " l a t e spawned" regime and WB under the " l a t e spawned" regime. The WBxW cross suffered much higher m o r t a l i t i e s . Survival ranged from 10-12 % under the "l a t e spawned" regime to 50 % under the "early spawned" regime (Figure 19). SURVIVAL IN LABORATORY EXPERIMENTS FROM FERTILIZATION TO EPIBOLY 1983-84 PERCENT SURVIVAL 1 . O O Q 0.67 0.33 0.00 ON 1X3 WINTER AUTUMN TEMPERATURE REGIME POPULATION Figure 19. S u r v i v a l under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime f or AB, WB, W, and W x WB in the 1983-84 experiment from f e r t i l i z a t i o n to epiboly, epiboly to eyed, eyed to hatch. SURVIVAL IN LABORATORY EXPERIMENTS FROM EPIBOLY TO EYED, 1983-84 PERCENT SURVIVAL 1.00 0.67 • 0.33 0.00 WBxW WB POPULATION WINTER AUTUMN TEMPERATURE REGIME Figure 19 continued SURVIVAL IN LABORATORY EXPERIMENTS FROM EYED TO HATCH, 1983-84 PERCENT SURVIVAL 0.67 0.33 -0.00 WBxW WB POPULAT. I ON WI NTER AUTUMN TEMPERATURE REGIME Figure 19 continued SURVIVAL IN LABORATORY EXPERIMENTS FROM FERTILIZATION TO HATCH, 1982-83 Figure 20. S u r v i v a l under 6°C, 10°C, simulated autumn spawning regime, simulated winter spawning regime f o r AB, WB, W, and W x WB in the 1982-83 experiment from f e r t i l i z a t i o n to hatch, hatch to emergence, and f e r t i l i z a t i o n to emergence. SURVIVAL IN LABORATORY EXPERIMENTS FROM HATCH TO EMERGENCE, 1982-83 PERCENT SURV gure 20 continued SURVIVAL IN LABORATORY EXPERIMENTS FROM FERTILIZATION TO EMERGENCE, 1932-83 Figure 20 continued 168 Mortality i n AB, WB, and W was r e l a t i v e l y constant through f e r t i l i z a t i o n to epiboly, epiboly to eye pigment stage and eye pigment stage to hatch for the within population matings. Survival varied between 97.5 and 100 % from f e r t i l i z a t i o n to epiboly, between 99 and 100 % from epiboly to eye pigment stage and between 98 and 100 per cent from the eye pigment stage to hatching. Although combined s u r v i v a l from f e r t i l i z a t i o n to epiboly, epiboly to eye pigment stage, eye pigment stage to hatch depend upon the temperature regime, population, season of spawning, lo c a t i o n of spawning and egg s i z e ( P < 0.01) there were more cases of independence than for 1982-83 su r v i v a l s as one might expect with high stable s u r v i v a l rates. Survival from f e r t i l i z a t i o n to epiboly was independent of: (1) season spawning at 6 C, 10 C, under the 'early spawned' regime and under the 'late spawned' regime; (2) location of spawning under the 'late spawned' regime; egg s i z e at 6 C, under the 'early spawned' regime and under the 'late' spawned regime. Survival from epiboly to eye pigment stage was independent of: (1) season of reproduction at 6 and 10 C; (2) location of reproduction at 6 C and under the 'early spawned1 regime; (3) of egg s i z e at 6 and 10 C and under the 'early spawned' regime. Survival from eye pigment stage to hatch was independent of: (1) season of spawning at 10 C and under the 'late spawned1 regime; (2) l o c a t i o n of spawning at 6 C, 10 C and under the 'late spawned' regime; (3) egg s i z e under the 'early spawned' and 'late spawned' regime. Survival from f e r t i l i z a t i o n to hatch was dependent on temperature regime, population, season of spawning, location of spawning but there were several 169 n o n - s i g n i f i c a n t t e s t s . S u r v i v a l was independent of: (1) season of spawning at 6 and 10 C; (2) l o c a t i o n of spawning under the 'early spawned' and ' l a t e spawned' regimes; (3) egg s i z e at 10 C, under the 'early spawned 1 and ' l a t e ' spawned regimes. S u r v i v a l from hatching to emergence was i n most cases higher than s u r v i v a l from f e r t i l i z a t i o n to hatch. S u r v i v a l of a l e v i n s from the, w i t h i n population matings ranged from 97 % of AB at 6 C to 100 % for W at 6 C, 10 C and under the " e a r l y spawned" regime, WB at 6 C and AB at 10 C. A l e v i n s u r v i v a l of the WBxW cross ranged from 10-18 % under the " e a r l y regime" compared to 25-60 % under the " l a t e regime" (Figure 21). S u r v i v a l from hatch to emergence f o r a l l populations was dependent on temperature regime (P < 0.01). Hatch to emergence s u r v i v a l at each temperature was a l s o dependent on p o p u l a t i o n , season of spawning, l o c a t i o n of spawning (except under the " l a t e spawned" regime) and egg s i z e (except under the " e a r l y spawned" regime. V e r t e b r a l Counts (1983-84 experiment) Both population ( P < 0.002 ) and temperature regime ( P < 0.0001 had s i g n i f i c a n t e f f e c t on v e r t e b r a l number. However, population by temperature regime i n t e r a c t i o n occurred ( P < 0.0001 ) and the r e f o r e i n t e r p r e t a t i o n of each main e f f e c t i s c o n d i t i o n a l on the s t a t e of the other f a c t o r (Table 37). SURVIVAL IN LABORATORY EXPERIMENTS FROM FERTILIZATION TO HATCH, 1983-84 PERCENT SURVIVAL O O .00 1 0.67 -0.33 -0.00 WBxW WB POPULATION WINTER AUTUMN TEMPERATURE REGIME Figure 21. S u r v i v a l under 6*C, 10°C, simulated autumn spawning regime, simulated winter spawning regime f o r AB, WB, W, W x WB i n the 1983-84 experiment from f e r t i l i z a t i o n to hatch, hatch to emergence, and f e r t i l i z a t i o n to emergence. SURVIVAL IN LABORATORY EXPERIMENTS FROM HATCH TO EMERGENCE, 1983-84 PERCENT SURVIVAL 1 .00O 0.67 n.33 WINTER A I I T I 11 I k 1 \ Figure 21 continued SURVIVAL IN LABORATORY EXPERIMENTS FROM F E R T I L I Z A T I O N TO EMERGENCE, 1 9 8 3 - 8 4 gure 21 continued 173 6 Celsius Emergent fry of the early spawning population in Bush Creek had s i g n i f i c a n t l y fewer vertebrae than those of the late spawning population in the same creek and from W creek (P < 0.05) (Table 37). The early spawning population also had fewer vertebrae than the combined average of the l a t e spawning populations ( P < 0.05 ). The fry from the population with smaller eggs had more vertebrae than those from the populations with larger eggs (P < 0.05). 10 C e l s i u s There were no s i g n i f i c a n t comparisons or contrasts. Vertebral counts from a l l three stocks were s i m i l a r (Table 37). Autumn-Winter-Spring Regime Emergent fry of the early spawning population in Bush Creek had s i g n i f i c a n t l y more vertebrae than those of the l a t e spawning population in the same creek (P < 0.05) (Table 37). The e a r l y spawning population had more vertebrae than the combined average of the l a t e spawning populations ( P < 0.05 ). The fry from the population with smaller eggs had fewer vertebrae than those from the populations with larger eggs (P < 0.05). 174 Table 37. Mean v e r t e b r a l counts f o r popul a t i o n by temperature regime treatments i n 1983-84 experiment. (Standard d e v i a t i o n s are i n parentheses, N = 50). Temperature Regime Population 6 10 EARLY LATE AB 65.40 (0.871) 65.25 (0.840) 65.80 (0.687) 66.55 (0.876) WB 66.10 (1.194) 65.25 (0.954) 65.20 (0.758) 67.50 (0.934) W 66.05 (1.085) 65.50 (0.679) 65.50 (0.816) 65.60 (1.033) 175 Winter-Spring Regime Emergent fry of the early spawning population in Bush Creek had s i g n i f i c a n t l y fewer vertebrae than those of the l a t e spawning population in the same creek and more vertebrae than those from W (P < 0.05) (Table 37). W fry had fewer vertebrae than the average of WB and AB (P < 0.05). The fry from the population with smaller eggs had fewer vertebrae than those from the populations with larger eggs (P < 0.05). Overall, AB and WB were the l e a s t s i m i l a r populations while WB and W were the most s i m i l a r . There, was no d i s t i n c t trend with one population always having higher vertebral counts. The responses tended to f l i p back and f o r t h . WB was the most l a b i l e i n response to temperature regime while W appeared to be the least responsive. Higher temperature in the early development ( i . e . under the 10 Celsius and Early Spawning Regime) appeared to r e s u l t in lower vertebral counts but there was a great deal of population by temperature regime i n t e r a c t i o n . External Morphology (1983-84 experiment) Stepwise discriminant analysis of the populations with a l l temperature treatments pooled revealed that AB progeny were morphologically d i s t i n c t from those of WB and W. A l l paired population comparisons were s i g n i f i c a n t (P 0.001) but a marked separation appeared between the early spawning stock and the two l a t e spawning stocks (Tables 38, 39). Symptomatic of the r e l a t i v e l y 176 Table 38. Approximate transformation F statistic to compare population centroids with a l l temperatures pooled. AB WB WB 78.00 W 119.45 17.32 Table 39. Mahalanobis temperatures pooled. distance between population centroids with a l l AB WB WB 10.63 W 16.38 2.36 177 greater morphological overlap between WB and W was the much higher m i s c l a s s i f i c a t i o n among these two groups by the discriminant functions (Table 40). The discriminators were in order of value: ED, SNL, PM, PFL, HL, HD, STL. The percentage m i s c l a s s i f i c a t i o n s decreased when the discriminant functions were calculated using fry reared under a s i n g l e temperature (Tables 41, 42, 43, 44). In general, a marked d i s c o n t i n u i t y occurred between the early spawning populations progeny and those of the la t e spawning populations (Tables 45, 46, 47, 48, 49, 50, 51, 52) . The important discriminantors changed with the temperature of incubation. The discriminators in order of value were ED, SNL, HD, BD, HL, PM, WT at 6 c e l s i u s ; ED, PM, HL, PFL, SNL, BD, HD, at 10 c e l s i u s ; ED, SNL, HL, WT, HD, PFL, BD under the "autumn spawned" regime; and HL, PM, PFL, STL, ED, HD, TL under the "winter spawned" regime. Progeny of the AB stock reared at 6 c e l s i u s c h a r a c t e r i s t i c a l l y had large eyes, long snouts, deep heads, deep bodies, long heads, few parr marks and were heavier than the other stocks. WB progeny had small eyes, short snouts, deep heads, shallow bodies, short heads, a moderate number of parr marks and were intermediate i n weight compared to the other stocks. W progeny had small eyes, short snouts, shallow heads, shallow bodies, short heads, a large number of parr marks and weighed the least of a l l the stocks (Table 53, Figure 22). Progeny of the AB stock reared at 10 c e l s i u s c h a r a c t e r i s t i c a l l y had large eyes, many parr marks, intermediate length heads, long pectoral f i n s , long 178 Table 40. C l a s s i f i c a t i o n matrix f o r 1983 f r y samples using the d i s c r i m i n a n t f u n c t i o n a l l temperatures pooled. GROUP PERCENT CORRECT CASES AB CLASSIFIED INTO WB GROUP W AB 87.0 87 13 0 WB 72.0 6 72 22 W 75.0 1 24 75 TOTAL 78.0 94 109 97 Table 41. C l a s s i f i c a t i o n matrix f o r l a b o r a t o r y d i s c r i m i n a n t f u n c t i o n at 6 c e l s i u s . f r y samples usi n g the GROUP PERCENT CORRECT CASES AB CLASSIFIED INTO WB GROUP W AB 92.0 23 1 1 WB 88.0 0 22 3 W 88.0 0 3 22 TOTAL 89.3 23 26 26 179 Table 42. C l a s s i f i c a t i o n matrix f o r l a b o r a t o r y f r y samples using the d i s c r i m i n a n t f u n c t i o n at 10 C e l s i u s . G R O U P P E R C E N T C O R R E C T C A S E S A B C L A S S I F I E D I N T O WB G R O U P W A B 1 0 0 . 0 2 5 0 0 WB 8 0 . 0 0 2 0 5 W 8 4 . 0 0 4 2 1 T O T A L 8 8 . 0 2 5 2 4 2 6 Table 43. C l a s s i f i c a t i o n matrix f o r l a b o r a t o r y d i s c r i m i n a n t f u n c t i o n at E a r l y Regime. f r y samples using the G R O U P P E R C E N T C O R R E C T C A S E S A B C L A S S I F I E D I N T O WB G R O U P W A B 1 0 0 . 0 2 5 0 0 WB 9 6 . 0 1 2 4 0 W 1 0 0 . 0 0 0 0 2 5 T O T A L 9 8 . 7 2 6 2 4 2 5 180 Table 44. C l a s s i f i c a t i o n matrix f o r la b o r a t o r y f r y samples usi n g the di s c r i m i n a n t f u n c t i o n at Late Regime. GROUP PERCENT CORRECT CASES AB CLASSIFIED INTO GROUP WB W AB 100.0 25 0 0 WB 92.0 1 23 1 W 84.0 0 4 21 TOTAL 92.0 26 27 22 Table 45. ce n t r o i d s Approximate tran s f o r m a t i o n F s t a t i s t i c t o compare popu l a t i o n at 6 C e l s i u s . AB WB WB 54.46 W 63.97 10.75 Table 46. Mahalanobis d i s t a n c e between po p u l a t i o n c e n t r o i d s at 6 C e l s i u s . AB WB WB 27.45 W 32.24 5.42 Table 47. c e n t r o i d s at Approximate 10 C e l s i u s . t r a n s f o r m a t i o n F s t a t i s t i c t o compare po p u l a t i o n AB WB WB 67.97 W 71.67 6.64 181 Table 48. Mahalanobis distance between popu l a t i o n c e n t r o i d s at 10 C e l s i u s . AB WB WB 34.26 W 36.12 3.35 Table 49. Approximate c e n t r o i d s at E a r l y Regime. transformation F s t a t i s t i c t o compare population AB WB WB 38.25 W 119.24 51.54 Table 50. Mahalanobis d i s t a n c e between po p u l a t i o n c e n t r o i d s at E a r l y Regime. AB WB WB 19.28 W 60.10 25.97 Table 51. Approximate c e n t r o i d s at Late Regime. transf o r m a t i o n F s t a t i s t i c t o compare popu l a t i o n AB ' WB WB 47.49 W 48.37 10.06 Table 52. Mahalanobis distance between population centroids at Late Regime. AB WB WB 23.94 W 24.39 5.07 182 Table 53. Means and standard d e v i a t i o n s o f morphological c h a r a c t e r i s t i c s o f emergent f r y reared at 6 c e l s i u s during 1983-84 (S.M. Mean standardized to a common length among the samples). POPULATION AB WB W CHARACTERISTIC MEAN S.D. MEAN S.D. MEAN S.D. S.M. S.M. S.M. Head Length 83.32 2. 61 79.40 1. 83 78. 44 2.02 82.28 79.37 79. 45 Snout Length 13.32 1. 10 10.76 0. 66 10. 08 0.90 13.10 10.76 10. 21 Pectoral F in 42.72 2. 42 39.44 1. 83 38. 64 2.47 Length 42.25 39.34 39. 02 Eye Diameter 30.28 1. 13 27.24 0. 52 27. 24 0.96 30.02 27.22 27. 20 Head Depth 45.84 1. 99 45.12 1. 52 42. 64 1.37 45.06 45.09 43. 20 Body Depth 53.28 4. 31 48.52 1. 85 47. 68 1.45 51.87 48.45 47. 56 Weight (mgs) 438.32 64. 96 388.60 25. 98 360. 60 17.03 413.38 387.55 369. 81 Parr Marks 9.60 1. 18 10.84 0. 89 11. 80 1.40 9.61 10.84 11. 32 Trunk Length 243.64 9. 26 242.64 6. 71 236. 76 4.47 239.11 242.03 241. 95 Caudal Fin 46.88 3. 51 44.72 3. 10 43. 96 3.16 Length 46.43 44.75 44. 87 WT/TL 1.17 0. 14 1.06 0. 06 1. 00 0.04 1.12 1.06 1. 01 KD (Bams 1976) 2.03 0. 52 1.99 0. 39 1. 97 0.34 Figure 22. Schematic drawings to show the body form of progeny from AB, WB, and W rearings from 6°C treatment of the 1983-84 experiment. 184 snouts, shallow bodies and deep heads. Progeny of the WB stock had intermediate sized eyes, few parr marks, long heads, short pectoral f i n s , short snouts, deep bodies and heads of intermediate depth. The progeny of the W stock had small eyes, an intermediate number of parr marks, short heads, short pectoral f i n s , short snouts, deep bodies and shallow heads (Table 54, Figure 23). When reared under the 'early' regime AB progeny had large eyes, long snouts, long heads, shallow heads, long pectoral f i n s , shallow bodies and were l i g h t e r i n weight. The WB progeny had intermediate sized eyes, short snouts, intermediate length heads, deep heads, intermediate length pectoral f i n s , bodies of moderate depth and were of intermediate weight compared to the other stocks. W progeny had small eyes, short snouts, short heads, shallow heads, short pectoral f i n s , deep bodies and were heavier compared to the other stocks (Table 55, Figure 24). Under the 'late' regime the AB progeny had long heads, few parr marks, long pectoral f i n s , large eyes, and deep heads. The WB progeny had heads of intermediate length, few parr marks, intermediate length pectoral f i n s , small eyes and shallow heads. The W progeny had short heads, many parr marks, short pectoral f i n s , small eyes, and shallow heads (Table 56, Figure 25). 185 Table 5A. Means and standard d e v i a t i o n s o f morphological c h a r a c t e r i s t i c s o f emergent f r y reared at 10 c e l s i u s during 1983-84 (S.M. Mean standardized t o a common length among the samples). POPULATION AB WB W CHARACTERISTIC MEAN S.D. MEAN S.D. MEAN S.D. S.M. S.M. S.M. Head Length 82.52 3.58 84.04 2.82 81.80 2.77 82.83 83.26 82.65 Snout Length 12.96 1.38 11.60 1.17 10.76 1.15 13.03 11.30 10.91 Pectoral F in 47.40 3.53 40.36 2.56 39.28 2.10 Length 47.61 40.16 39.54 Eye Diameter 31.24 0.87 29.12 1.11 27.88 1.15 31.30 28.90 28.10 Head Depth 47.76 2.14 46.16 2.62 44.48 1.94 47.88 45.35 44.84 Body Depth 45.56 2.35 47.52 2.84 45.76 2.46 45.69 46.60 46.62 Weight (mgs) 340.44 40.61 361.92 49.41 339.80 36.47 343.66 342.32 353.64 Parr Marks 12.08 1.14 9.32 0.79 10.40 0.98 12.07 9.33 10.43 Trunk Length 231.48 10.56 236.68 10.28 229.56 7.27 232.53 232.10 232.70 Caudal Fin 49.12 3.53 49.44 3.88 49.12 4.65 Length 49.12 49.24 49.74 WT/TL 0.93 0.07 0.97 0.10 0.94 0.07 0.94 0.94 0.97 KD (Bams 1976) 1.91 0.41 1.91 0.46 1.93 0.41 Figure 23. Schematic drawings to show the body form of progeny from 10°C treatment of the 1983-84 experiment. 187 Table 55. Means and standard d e v i a t i o n s o f morphological c h a r a c t e r i s t i c s o f emergent f r y reared under the E a r l y Regime during 1983-84 (S.M. Mean standardized t o a common length among the samples). POPULATION AB WB W CHARACTERISTIC MEAN S.M. tS.D. MEAN S.M. S.D. MEAN S.M. S.D. Head Length 89.38 86.84 3.33 83.08 83.47 2.54 75.96 77.58 2.09 Snout Length 14.84 14.44 1.13 10.88 10.98 1.08 10.52 10.10 0.86 Pectoral F in 46.56 2.46 40.28 3.12 34.56 2.56 Length 45.39 40.49 34.34 Eye Diameter 31.72 31.41 0.78 29.44 29.43 0.91 27.24 27.14 0.52 Head Depth 48.26 46.54 2.13 45.80 46.03 1.35 41.68 41.28 1.07 Body Depth 51.28 48.02 4.14 49.20 49.46 1.83 51.08 51.92 1.89 Weight (mgs) 439.60 380.95 70.57 369.40 37 3.77 22.72 374.00 388.14 17.02 Parr Marks 9.92 9.57 1.45 9.60 9.66 1.34 11.60 11.91 1.51 Trunk Length 243.80 235.46 9.23 236.88 238.84 6.75 237.92 244.73 4.12 Caudal F i n Length 51.04 49.07 5.93 47.84 47.77 3.83 43.24 46.17 3.42 WT/TL 1.14 1.03 0.14 1.00 1.01 0.05 1.05 1.05 0.05 KD (Bams 1976) 1.97 0.52 1.94 0.37 2.02 0.37 AB W WB CD CO Figure 24. Schematic drawings to show the body form of progeny from AB, WB, and W rearings from simulated autumn spawning treatment of the 1983-84 experiment. 189 Table 56. Means and standard d e v i a t i o n s o f morphological c h a r a c t e r i s t i c s o f emergent f r y reared under Late Regime during 1983-84 (S.M. Mean standardized t o a common len g t h among the samples). POPULATION AB WB W CHARACTERISTIC MEAN S.M. S.D. MEAN S.M. S.D. MEAN S.M. S.D. Head Length 88.96 88.41 3.16 81.88 81.60 2.62 81.44 81.77 1.21 Snout Length 12.92 12.70 1.59 10.12 10.04 1.67 10.68 10.85 0.68 Pectoral F i n 44.56 3.16 37.88 1.99 38.72 1.65 Length 44.03 37.67 39.55 Eye Diameter 32.24 32.05 2.20 29.60 29.54 1.03 28.88 28.92 0.66 Head Depth 51.04 50.54 2.90 47.20 47.06 1.07 45.96 46.07 1.58 Body Depth 51.56 50.64 3.80 52.00 51.85 1.59 50.36 50.10 1.80 Weight (mgs) 417.04 395.91 77.88 391.08 386.31 25.16 377.64 383.10 18.76 Parr Marks 9.48 9.42 1.62 8.84 8.78 1.27 11.00 10.67 1.07 Trunk Length 240.72 236.71 13.70 245.32 243.52 6.97 237.04 243.35 5.77 Caudal F i n Length 54.60 54.10 5.36 46.60 46.47 4.68 48.52 48.98 3.86 WT/TL 1.08 1.04 0.15 1.05 1.04 0.05 1.03 1.03 0.05 KD (Bams 1976) 1.93 0.53 1.96 0.37 1.97 0.37 191 DISCUSSION To summarize the r e s u l t s : Survival to hatch and to emergence in the 1982-83 experiment was poor and highly variable compared to that of the 1983-83 experiment and r e s u l t s of the other tests must be considered with t h i s i n mind. No strongly defined pattern with respect to temperature regime, populations, season of spawning and egg size emerged in the 1982-83 experiment. The f a i l u r e of the cross-mated population to thriv e stands out in the 1983-84 experiment along with the high (96.5% - 98.5%) s u r v i v a l of the AB, WB, and W populations under a l l regimes. Temperature regime exerted a powerful e f f e c t on the incubation rates to hatch and emergence in both 1982-83 and 1983-84. The winter spawning populations generally took less time to reach hatch and emergence than the autumn spawning population except at 10 c e l s i u s where a genotype environment in t e r a c t i o n reversed the order of hatching. Differences were greatest among populations that were temporally and s p a t i a l l y i s o l a t e d in the wild. Vertebral counts varied depending on the temperature regime or the population. Overall WB and AB were the least s i m i l a r populations while WB and W were the most s i m i l a r but there was no clear l i n e a r trend by population or temperature i n the vertebral counts. 192 Discriminant analysis of external morphology showed a s t r i k i n g separation between the progeny of the autumn stock and the two winter stocks. Eye diameter, head length, parr marks and snout length were the most useful discriminating t r a i t s . For example the AB progeny generally had large eyes and longheads compared to to the W progeny. The expression of the t r a i t s within a population was also dependent upon temperature of incubation. Survival of eggs and alevins among the 1982-83 treatments did not conform to my expectations. For example, I expected that progeny of lat e spawning stocks would survive better at 6 C compared to 10 C. However, s u r v i v a l of WB eggs in the 1982-83 experiment ranged between 73 to 86 % at 6 C compared to 75 to 90 % at 10 C. Alevin s u r v i v a l was much less at 6 C (62 to 96 %) compared to 10 C (100 % ) . Both the W and WB eggs had higher s u r v i v a l under an autumn -winter - spring progression of temperatures (72 to 74 and 98 to 99 %, respectively) than under a winter - spring progression (2 to 49 and 65 to 80 %, r e s p e c t i v e l y ) . The general pattern for a l l populations in the 1982-83 experiment was that eggs survived better at'10 C and under the 'early spawned' temperature regime compared to 6 C and the 'late spawned' temperature regime. In general, in the 1982-83 experiment, W eggs had the highest s u r v i v a l among the stocks compared within temperatures. At 10 C, and under the 'early' and 'late spawned' regimes WB eggs had the lowest s u r v i v a l . 193 M o r t a l i t i e s could be due to mechanical or v i b r a t i o n a l shock of the embryos when they were added to the trays or when they were transferred into gravel (Smirnov 1955, Jensen and Alderdice 1983). Jensen and Alderdice (1983) noted that s e n s i t i v i t y of salmon eggs increased with time from the onset of f e r t i l i z a t i o n up to the stage where the eyes of the embryo are c l e a r l y v i s i b l e . After the 'eyed' stage the embryos were r e l a t i v e l y i n s e n s i t i v e to disturbance. Therefore, one might expect that the greatest mortality due to t h i s cause would occur from f e r t i l i z a t i o n to epiboly. Beacham and Murray (1986a, 1987b) found t h i s to be the case for chum salmon embryos reared at constant temperatures. The addi t i o n a l m o r t a l i t i e s that I observed af t e r epiboly suggest other factors might have influenced s u r v i v a l . M o r t a l i t i e s occurred throughout development from f e r t i l i z a t i o n to hatching. Deaths were not confined to any one period of development. The much lower o v e r a l l s u r v i v a l in 1982-83 experiment compared to the 1983-84 experiment suggests that the incubation conditions i n Experiment 1 were sub-optimal. The m o r t a l i t i e s a f t e r epiboly could be due to the exposure of the embryos to l i g h t or due to the e f f e c t of rearing upon screens. It i s perhaps not s u r p r i s i n g that there i s no c o r r e l a t i o n between s u r v i v a l by stocks and temperature regime. H e r i t a b i l i t y of s u r v i v a l during embryonic development has been shown to be les s than 0.05 i n Oncorhynchus (Withler et a l . 1987). The r e l a t i v e l y poor s u r v i v a l i n the WBxW cross i n both years suggests that there i s a f e r t i l i t y b a r r i e r between the W and WB populations. This 194 represents an unusual f i n d i n g . I n t r a s p e c i f i c hybrids of salmonids are generally v i a ble even when among d i f f e r e n t forms (Bakkala 1970). Incompatibility might be related to the differences in egg s i z e among the two groups. Although these populations tend to spawn lat e and have r e l a t i v e l y rapid incubation rates compared to the AB population they may not be achieving t h i s through the same means. Beacham and Murray (1986) found that egg s i z e was inversely related to the time to reach emergence of the progeny of early and late spawners. AB could have developed smaller eggs in response to the need for more rapid development while in W the change may be more a r e s u l t of differences i n the genetic material i n the nucleus. By achieving s i m i l a r r e s u l t s through d i f f e r e n t means the populations may not be g e n e t i c a l l y compatible. The presence of population by temperature i n t e r a c t i o n in the ANOVAs of hatch and emergence time suggests that the stocks are responding i n d i f f e r e n t ways to the variety of incubation regimes. This suggests that the populations are i n t r i n s i c a l l y d i f f e r e n t i n t h e i r incubation rate program. For example, progeny of the autumn spawning stock appear to develop more e f f i c i e n t l y r e l a t i v e to the o f f s p r i n g of the l a t e spawning stocks at the high constant temperature as opposed to the rest of the temperature regimes. The greatest gap between the populations occurred under the winter-spring progression where the l a t e season stocks developed much faster r e l a t i v e to the early season stock. 195 The r e s u l t s show that time to hatch and time to emergence d i f f e r between stocks that normally spawn during d i f f e r e n t seasons in the wild. As well, the populations which are both s p a t i a l l y and temporally i s o l a t e d show the greatest divergence o v e r a l l in time to hatch and time to emergence. If one eliminates the 10 c e l s i u s treatment on the grounds that t h i s i s outside the natural range of temperature for each stock then the comparison between early and late spawning stocks i s s i g n i f i c a n t 11 out of 12 times and between the two stocks that are temporally and s p a t i a l l y i s o l a t e d the comparison i s s i g n i f i c a n t 12 out of 12 times. The observed pattern of differences among the stocks in time to hatch and time to emergence probably has some genetic basis. However, the genetic mechanisms c o n t r o l l i n g these differences are not c e r t a i n . Natural genetic v a r i a t i o n in quantitative t r a i t s could involve many l o c i but i t i s possible that only a few regulatory genes are involved. Tauber and Tauber (1977) found that only 3 l o c i determined the development rate differences among seasonally i s o l a t e d insect species. Allendorf et a l . (1983) found that development rate in rainbow trout (Salmo gairdneri) was influenced by a mutant a l l e l e at a single regulatory locus c o n t r o l l i n g the expression of the phosphoglucomutase gene. Danzmann et a l . (1986) found that heterozygous i n d i v i d u a l s developed more r a p i d l y . They proposed that linkage d i s e q u i l i b r i u m between a l l e l e s at s p e c i f i c l o c i and dominant-acting genes that accelerate or retard development rate are responsible for i n t r a s p e c i f i c v a r i a t i o n in incubation rate. 196 Allendorf et a l . (1983) suggested that accelerated development could confer a competitive advantage throughout the f i s h ' s l i f e . They showed that rainbow trout that hatched at an e a r l i e r time became larger and reached sexual maturity sooner than t h e i r slower developing brethren. In the hatchery environment e a r l i e r hatch would probably allow a head s t a r t on feeding and hence growth. However, the rapid development that I observed in the late spawning stocks probably evolved to prevent t h e i r progeny from experiencing a competitive disadvantage. Tardy downstream migrants might experience increased thermal s t r e s s as temperatures in the stream r i s e during the spring. As the season progresses nights become shorter. The time a v a i l a b l e when the fry can migrate under the cover of darkness i s progressively reduced. As well, l a t e migrants would probably have a higher mortality rate because predators would be aggregated and experienced as a r e s u l t of previous chum salmon migrations. F i n a l l y , food resources on the estuary could be depleted by e a r l i e r a r r i v i n g f r y . Adjustment of emergence timing through a change in development rate was reported by Beacham and Murray (1986a) and Murray and Beacham (1987) for e a r l y and l a t e spawning stocks of chum salmon of the Fraser River. However, Beacham and Murray (1986) also reported no difference in time to hatch at 4, 8 , and 12 c e l s i u s between the progeny of early and l a t e spawners of the Vedder and Chehalis Rivers of the Fraser River system. This suggests that early and l a t e spawning populations d i f f e r in the genes c o n t r o l l i n g development rate from hatch to emergence. In contrast, my r e s u l t s show that both pre-hatch and post-hatch development rate can d i f f e r among early and late spawning 197 populations. Beacham et a l . (1985) indicated that chum salmon stocks on Vancouver Island were g e n e t i c a l l y d i s t i n c t from those of the Fraser River. Long term i s o l a t i o n of these major d i v i s i o n s i n the North American Chum salmon population could explain why The Fraser River stocks may not have evolved the hatch time v a r i a b i l i t y . S election can only be e f f e c t i v e i f there i s some degree of intra-population v a r i a b i l i t y i n a t r a i t (Falconer 1981). If the environment remains stable and predictable organisms w i l l experience s t a b i l i z i n g s e l e c t i o n (Wright 1987). Thus, genetic v a r i a t i o n w i l l be l o s t . In general, the mobility provided by the al e v i n stage in salmon i s thought to have s u r v i v a l value in preventing embryos from being caught in unsuitable areas when winter freshets occur or when ce r t a i n areas of the stream bed dry out. Where the stream bed i s stable throughout the winter, hatch time i s r e l a t i v e l y free of s e l e c t i o n pressure. The flow of these r i v e r s may be more moderated i n terms of winter freshets than the coastal streams on Vancouver Island. I f the gravel bases are more stable i n the Vedder and Chehalis Rivers early hatch time might not be selected in the late spawning stocks. However, the 1983-84 experiment, with more precise control of temperatures, f a i l e d to show any differ e n c e within the same system for time to hatch although the o v e r a l l comparison between early and l a t e spawners was s i g n i f i c a n t 5 out of 8 times. According to F. Velsen (pers. comm.) there i s a great amount of v a r i a t i o n in incubation rate to hatch among chum salmon stocks. Smoker (1982) also found that time to hatch d i f f e r r e d among three chum salmon stocks reared under the same conditions. My r e s u l t s support t h i s hypothesis that hatch time 198 varies depending upon the stock and rather than the hypothesis put forward by Beacham and Murray (1986a) that incubation rate to hatch does not vary among chum salmon populations. My re s u l t s generally agree with those of Beacham and Murray (1986) regarding differences in time to emergence under c o n t r o l l e d conditions among early and l a t e spawning stocks. However, I observed much more genotype -environment i n t e r a c t i o n than in t h e i r experiments. In p a r t i c u l a r , the ten c e l s i u s treatment showed l i t t l e or no difference among the populations. As well, the differences in time to emergence that I observed at the other temperatures are much greater than those that occurred among the vedder and Chehalis River populations. The observed differences between experiment 1 and experiment 2 in time to hatch and emerge may be accounted for by temperature of incubation except in the ten c e l s i u s regime. Only about 30 per cent of the difference observed at the high temperature appears to be d i r e c t l y a r e s u l t of mean temperature dif f e r e n c e s . Emergence timing can be greatly altered from the norm in suboptimal conditions (Bams 1969, Bams and Simpson 1977). For example, Bailey and Taylor (1974 i n Bams and Simpson 1977) and Bailey, P e l l a and Taylor (1975 in Bams and Simpson 1977) recorded premature emergence of pink salmon fry by 2 to 6 weeks compared to the wild stocks. Exposure to l i g h t during development may advance emergence s u b s t a n t i a l l y (Bams, unpublished data). Mead and Woodall (1968) 199 found that f r y produced on a f l a t surface were less photonegative than those produced in gravel beds. Bailey and Heard (1973) indicate that such fry w i l l enter t h e i r l a c u s t r i n e l i f e prematurely. Alevins on screens may emerge one to two weeks e a r l i e r than those supported by gravel and a s o l i d bottom (Bams 1982, pers comm. Nortvedt 1986). Bams (1982) reported that unsupported f r y emerged 14 to 17 days sooner than wild f r y . F i n a l l y , l a r v a l period can be influenced by surrounding i n d i v i d u a l s . For example, Plytyez et a l . (1984) found that l a r v a l periods were much shorter in Rana temporaria reared i n d i v i d u a l l y rather than i n groups. This might account for difference between my re s u l t s and those of Beacham and Murray (1986). Beacham and Murray (1986) reared embryos in groups of 10 in the presence of other stocks (on the same heath tray) while I reared the embryos in large interpopulational groupings in separate tanks. The year to year v a r i a t i o n i n emergence timing can be p a r t i a l l y explained by the differences in mean temperature of incubation of the f i r s t year from the planned values. The early emergence of f r y incubated at 10 Celsius during 1982-83 could be due a s y n e r g i s t i c e f f e c t from the exposure to l i g h t , combined with incubation on the screens with minimal gravel for support and the temperature f l u c t u a t i o n s . Fluctuating high temperatures could mimic the d a i l y warming and cooling that would occur during the spring months during the natural emergence period for these f r y . Thus, fr y could emerge ear l y . 200 The presence of substantial temperature by population i n t e r a c t i o n in the analysis of vertebral number suggests that the populations each possess a unique genetic program for vertebral development. Populations spawning during d i f f e r e n t seasons in the same r i v e r system are the most divergent in t h e i r vertebral counts at various temperatures. However, only h a l f of the contrasts between the early stock and the l a t e stocks were s i g n i f i c a n t l y d i f f e r e n t so late spawners do not always have higher vertebral counts. There also appeared to be a connection between the s i z e of the egg and the vertebral number. So, egg s i z e rather than genotype could c o n t r o l the response. Lindsey and Arnason (1981) suggested that the number of elements i n a m e r i s t i c character i s determined by two independent processes, probably growth and d i f f e r e n t i a t i o n , and t h e i r i n t e r a c t i o n s during development (The 'Atroposic Model'). Variation in the number of elements in a m e r i s t i c character may be dependent upon genetic and environmental influences on growth and d i f f e r e n t i a t i o n of the embryo, although the r e l a t i v e magnitude of the environmental component can d i f f e r among various characters. Beacham and Murray (1986b) found differences among progeny of early and la t e chum salmon spawners within a population i n dorsal f i n rays, anal f i n rays, pectoral and p e l v i c f i n rays and g i l l rakers but not among vertebrae. Lindsey et a l . (1984) found that temperature changes during development of rainbow trout could emphasize the expected pattern of development or cause a "paradoxical" reaction in the development program. This e f f e c t might account 201 for the large amount of genotype-environment i n t e r a c t i o n observed in my r e s u l t s . The inheritance of meristic v a r i a t i o n in salmonids has been discussed by Leary et a l . (1985) and Lindsey (1988). Leary et a l . (1985) found high h e r i t a b i l i t i e s for m e r i s t i c t r a i t s in rainbow t r o u t . They invoked the idea that much phenotypic v a r i a t i o n i n m e r i s t i c t r a i t s stems from the degree of heterozygosity and developmental s t a b i l i t y . This view has been challenged by Swain (1987) who argued that v a r i a b i l i t y i n me r i s t i c characters could not be properly measured against heterozygosity because these were threshold t r a i t s . Furthermore, Beacham and Withler (1987) tested t h i s idea empirically with chum salmon and found no r e l a t i o n s h i p between meristic developmental s t a b i l i t y and heterozygosity. The response of the genome to environmental change seems too complex for any description except by a quantitative model such as the "Atroposic Model" of Lindsey and Arnason (1981) (Lindsey 1988). Vertebral counts of AB fry reared under the autumn - winter - spring incubation regime were s i m i l a r to counts of W fry reared under the winter -spring incubation regime (X = 65.80 and X - 65.60, r e s p e c t i v e l y ) . As in the wild caught f r y , vertebral count of WB fry reared under the winter - spring incubation regime d i f f e r e d from the other populations (X - 67.50). This suggests that the i n t e r a c t i o n .between the temperature of incubation and the genome of the population i n the wild i s s u f f i c i e n t to explain the observed vertebral count d i s t r i b u t i o n . It further suggests that the AB and W stocks 202 are adapted to compensate e f f e c t i v e l y for the temperature of incubation in the wild so as to produce the same phenotype. Judged from the percentage of m i s c l a s s i f i c a t i o n s , the degree of separation between the laboratory reared f i s h was greater than between the wild even when a l l temperatures were pooled. The range of phenotypic v a r i a t i o n within lab populations (and hence the m i s c l a s s i f i c a t i o n between populations) was greater than with a sing l e temperature because the environmental variance contributes much more with four temperature regimes. The pooled temperatures cover a range of environmental v a r i a b i l i t y i n temperature regime that i s greater than in the wild. Yet the d i r e c t i o n of separation was consistent with the hypothesis that temporal i s o l a t i o n contributes to genetic divergence among the populations. Separation was greatest between stocks i n the wild which had the greatest separation i n time and space. The pattern of morphological differences among the stocks may be due to the a l t e r a t i o n of the timing of events or processes i n development (Balon 1979, Gould 1982). S l a t k i n (1987) has developed a model that l i n k s genetic and environmental v a r i a t i o n to both the rates of tissue growth and the timing of ontogenetic t r a n s i t i o n s . The model shows how genetic v a r i a t i o n in developmental parameters governs v a r i a t i o n and covar i a t i o n i n phenotypic t r a i t s . Selection on the phenotype a l t e r s the d i s t r i b u t i o n s of developmental parameters. Selection can therefore lead to changes in the average times of developmental events. Heterochrony i s most evident i n comparisons between AB 203 progeny and W progeny. At a l l temperatures of incubation the AB fry had larger eyes, longer snouts and pectoral f i n s . They also tended to have longer and deeper heads than the W progeny. In other c h a r a c t e r i s t i c s , such as the number of parr marks, weight and body depth, whose expression depended less on s k e l e t a l development, the d i r e c t i o n of interstock differences was less consistent. At 6 C the AB progeny were heavier with fewer parr marks and deeper bodies than the W progeny. Under the early regime AB progeny were l i g h t e r and had shallower bodies compared to W. The timing of emergence of the W progeny was accelerated r e l a t i v e to the rate of development of the eyes, snout, pectoral f i n s and head. The evidence presented here I believe demonstrates conclusively that when reared under the same environmental conditions populations that n a t u r a l l y reproduce during d i f f e r e n t seasons are g e n e t i c a l l y d i f f e r e n t . This difference i s most s t r i k i n g for the duration of the incubation period and external morphology. Discounting the 1982-83 experiment, because of technical problems, i t appears that there i s l i t t l e s e l e c t i o n by temperature regime during the embryonic period. Rather the r e s u l t s suggest that s e l e c t i o n may be strong in the post-emergent, downstream migrant phase of the l i f e c y c l e . Also, i t i s clear that, for a l l the quantitative t r a i t s studied, temperature of incubation has a profound e f f e c t on phenotype. In s p i t e of t h i s , when temperature e f f e c t s are removed, the populations show appropriate adaptation of t h e i r quantitative t r a i t s to season of reproduction. 204 EVIDENCE FOR SELECTION ON INCUBATION RATE INTRODUCTION According to evolutionary theory phenotypic variance of a population may be p a r t i t i o n e d into genetic and environmental or non h e r i t a b l e components (Falconer 1981). The reason for t h i s d i s t i n c t i o n i s that the rate of evolution under natural s e l e c t i o n depends on the r e l a t i v e magnitudes of ce r t a i n components of variance. According to B u l l (1987) phenotypic v a r i a t i o n may be s e l e c t i v e l y maintained i n a population according to i t s components: se l e c t i o n may favor the maintenance of only the environmental components, only the genetic components, or be i n d i f f e r e n t to the composition of the variance. Even when s e l e c t i o n i s shown to favor phenotypic v a r i a t i o n regardless of i t s components, the p o s s i b i l i t y e x i s t s that environmental variance w i l l evolve to displace genetic components or vice versa. Environmental and genetic factors may thus compete to produce a given selected l e v e l of phenotypic variance. I have hypothesized that natural s e l e c t i o n for synchronized emergence time has caused incubation rate differences to evolve among seasonally d i s t i n c t spawning populations of chum salmon. According to Falconer (1981) fi t n e s s related t r a i t s generally have low h e r i t a b i l i t i e s because constant s e l e c t i o n to a fixed optimum removes the additive genetic variance. Therefore, i f the rate of incubation of each population has been determined through natural s e l e c t i o n one would expect the h e r i t a b i l i t y to be low. 205 However, r e l a t i v e l y high estimates of additive genetic variance in f i t n e s s related t r a i t s have been recently recorded in aquatic organisms (Gjedrem 1983). MATERIALS AND METHODS To estimate the contribution of maternal and additive genetic variance to time to hatch and emergence within the populations samples from the 25 fami l i e s produced per population were also reared i n d i v i d u a l l y at 8 C during 1983-84. The s i r e , dam and s i r e plus dam h e r i t a b i l i e s at 8 C were estimated for time to hatch and time to emergence by c o r r e l a t i o n between r e l a t i v e s (Falconer 1981). H e r i t a b i l i t y estimates were calculated from i n d i v i d u a l family rearings by the f a c t o r i a l method described by Becker (1975). The lower confidence l i m i t of the h e r i t a b i l i t y estimates were generated by a Monte Carlo simulation technique s i m i l a r to that of Rodda et a l . (1977). I assumed that the error entered when I gathered the data for hatch and emergence time. The error in the estimates of hatch time and emergence time of each family I assumed to be normally d i s t r i b u t e d . 200 h e r i t a b i l i t y estimates were computed for hatch and emergence time of each sub-population from simulated random samples of 100 of f s p r i n g generated per family. The tenth lowest simulated h e r i t a b i l i t y estimate was taken as the lower confidence l i m i t . 206 RESULTS Families where mortality exceeded 10 per cent were not used to estimate h e r i t a b i l i t y for time to hatch and time to emergence. I did t h i s to eliminate the p o s s i b i l i t y suggested by Allendorf et a l . (1983) that developmental ontogeny and mortality could be linke d . Rather than contend with missing c e l l s in the analysis I chose to drop rows and columns as necessary to maintain a f a c t o r i a l design. The s i r e by dam (SxD) design for time to hatch of AB was reduced to 4x4; WB, 4x3;, W, 3x4. The SxD design for time to emerge was 4x4 for AB; 2x5 for WB; 3x4 for W. Estimated s i r e plus dam h e r i t a b i l i t y for time to hatch was 0.27 for AB, 0.35 for WB and 0.47 for W. H e r i t a b i l i t y for time to emerge exceeded the h e r i t a b i l i t y for time to hatch i n a l l populations (0.50 - AB, 0.40 - WB, 0.54 - W). In a l l cases, except emergence of AB, s i r e h e r i t a b i l i t y was greater than dam h e r i t a b i l i t y . The lower confidence l i m i t s estimated v i a simulations were greater than zero in a l l cases (Tables 57 and 58). DISCUSSION H e r i t a b i l i t y estimates may vary depending on the environment (Hartl 1981). Thus, a p p l i c a t i o n of h e r i t a b i l i t y estimates from laboratory rearings to discussions of genetic v a r i a t i o n i n natural populations must be treated with caution. 207 Table 57. H e r i t a b i l i t i e s and lower confidence l i m i t (LCL) (P = 0.05) f o r time t o 50 % hatch at 8 C. H e r i t a b i l i t y Sire Dam Sire Plus Dam Population Mean LCL Mean LCL * Mean LCL AB 0.38 0.250 0.16 0.066 0.27 0.203 WB 0.71 0.524 0.14 0.025 0.35 0.214 W 0.77 0.629 0.64 0.115 0.47 0.388 Table 58. H e r i t a b i l i t i e s and lower confidence l i m i t (LCL) (P = 0.05) f o r time t o 50 % emergence at 8 C. H e r i t a b i l i t y Sire Dam Sire Plus Dam Population Mean LCL Mean LCL Mean LCL AB 0.44 0.330 0.56 0.452 0.50 0.431 WB 0.73 0.684 0.10 0.047 0.40 0.377 W 0.93 0.807 0.52 0.101 0.54 0.474 208 Maternal e f f e c t s appear to be r e l a t i v e l y unimportant to within population differences i n time to hatch and time to emergence. This suggests that incubation rate i s con t r o l l e d d i r e c t l y by the genome. The r e l a t i v e l y high h e r i t a b i l i t i e s suggest that there i s much greater i n d i v i d u a l v a r i a t i o n i n incubation rate within each population than one might predict from a r i g i d model of early stock = slow incubation rate : la t e stock = fast incubation rate. There are several ways whereby h e r i t a b i l i t y could remain high even when under strong s e l e c t i o n . Gene flow between the populations could cause the mixing of genes for fast and slow incubation rates thereby increasing the within population genetic v a r i a b i l i t y . The r e l a t i v e l y high genetic v a r i a t i o n within a population could r e s u l t from s l i g h t s h i f t s in the optimum time of downstream emergence. In some years, the progeny of i n d i v i d u a l s with s l i g h t l y faster incubation rates have higher s u r v i v a l while i n other years the s i t u a t i o n may be reversed. The lack of p r e d i c t a b i l i t y due to annual v a r i a t i o n in the environment means that the genetic composition of the population would not come to equilibrium. The recombination between animals born i n d i f f e r e n t years would l i k e l y be a f a i r l y common occurrence since each year several ages of chum salmon spawn together. Lande (1975) suggested an a l t e r n a t i v e explanation for the presence of high additive genetic variance i n selected t r a i t s . He demonstrated that the 209 high spontaneous mutation rate of quantitative characters ( one per one hundred gametes) could maintain he r i t a b l e v a r i a t i o n even in the presence of strong s e l e c t i o n . In a species l i k e chum salmon with the p o t e n t i a l to produce thousands of young per f i s h a s ubstantial number of o f f s p r i n g each generation could be mutants. For example, from a female with a fecundity of 2000 eggs roughly 20 fry should have a mutation for a p a r t i c u l a r quantitative t r a i t . If these mutants mate together on return the v a r i a t i o n in the quantitative t r a i t w i l l increase in the population. Antagonistic plieotropy among f i t n e s s related t r a i t s may also prevent the loss of genetic v a r i a t i o n . E s s e n t i a l l y , one gene or genes may code for two f i t n e s s related t r a i t s so that an increase in f i t n e s s in one t r a i t i s accompanied by a decrease i n f i t n e s s in the other t r a i t and vice versa. Thus, the e f f e c t s of s e l e c t i o n on genetic v a r i a t i o n i n either t r a i t i s constrained. F i n a l l y , genetic v a r i a b i l i t y might be maintained through disruptive s e l e c t i o n on c h a r a c t e r i s t i c s g e n e t i c a l l y correlated with incubation rate. Disruptive s e l e c t i o n could i n t e r a c t with the s t a b i l i z i n g s e l e c t i o n on incubation rate so that at equilibrium a range of hatch and emergence times are equally favored. 210 An Electrophoretic Analysis of Genetic Variation among Seasonally Separated Populations 211 INTRODUCTION A considerable portion of the v a r i a b i l i t y of morphological, p h y s i o l o g i c a l , behavioural or m e r i s t i c characters i s determined by nongenetic factors (Falconer 1981). Assessment of evolutionary relationships or estimation of the amount and d i s t r i b u t i o n of genetic v a r i a t i o n cannot be based exc l u s i v e l y on analyses of phenotypic c h a r a c t e r i s t i c s (Ryman and Stahl, 1981). In some cases conclusions have been made by over-emphasizing the genetic control of phenotypic t r a i t s that are also influenced by the environment (Ryman and Stahl 1981). Recently in the l i t e r a t u r e and, as described i n section 4, d e t a i l e d experiments for the estimation of environmental and genetic variance components of e c o l o g i c a l l y important characters have been made (Gjedrem 1976, Gunnes and Gjedrem 1978, Naevdal et a l . 1978, Refstie and Stien 1978, R i d d e l l and Leggett, 1981, Smoker 1982). It must be emphasized, however, that in the experiments of other workers and to a lesser extent i n section 4 such estimates refer to p a r t i c u l a r experimental conditions and they constitute r e l a t i v e , and not absolute, measure of the amount of genetic v a r i a t i o n . The c l a s s i c a l techniques of animal breeding refer to " s t a t i s t i c a l " genes rather than to d i s t i n c t a l l e l l e s at p a r t i c u l a r l o c i . One must consider t h i s when in t e r p r e t i n g the r e s u l t s of quantitative genetic information for the discrimination of stocks or assessment of evolutionary r e l a t i o n s h i p s . 212 Electrophoretic techniques can allow a d e t a i l e d examination of i n t r a s p e c i f i c genetic r e l a t i o n s h i p s . Early electrophoretic analyses of v a r i a t i o n at single l o c i confirmed the existence of g e n e t i c a l l y d i f f e r e n t i a t e d subunits in several species eg. A t l a n t i c cod (Gadus morhua) (Frydenberg et. a l . 1965), A t l a n t i c salmon (Salmo salar) (Payne et a l . 1971) and A r t i e char (Nyman 1972). More recently d e t a i l e d studies based on a large number of l o c i have demonstrated the complex genetic structure of populations of species such as A t l a n t i c herring (Anderson et a l . 1981), A t l a n t i c salmon (Stahl 1981), brown trout (Allendorf et a l . 1977; Ryman et a l . 1979), and Whitefish (Coregonus spp.) (Vuorinen et a l . 1981). Electrophoresis has many a t t r a c t i v e properties for genetic analysis of populations. F i r s t l y , i t i s r e l a t i v e l y easy to perform and most of the techniques have been standardized. It i s now a regular part of the management strategy for the P a c i f i c Coast salmon f i s h e r i e s ( C C . Wood, D.F.O., Nanaimo, B.C., personal communication). Secondly, electrophoresis generally gives clear quantitative answers. Third, a sample of isozymes i s considered an unbiased sample of the genome because they are neutral to s e l e c t i o n (Roughgarden 1979, Lewontin and Hubby 1966, Harris 1966). However, t h i s l a s t point i s open to question. The amount of polymorphism may be underestimated because, at best, routine electrophoresis detects only those amino acid substituions that r e s u l t in charge differences in a p r o t e i n , and the procedure may miss even some of these. 213 For example, a change in technique r e s u l t e d . i n an increase from 6 to 37 in the number of i d e n t i f i e d a l l e l e s at the xanthine dehydrogenase locus i n Drosophila pseudo obscura and an increase in the estimate of average heterozygosity from 0.44 to 0.73 (Singh et a l . 1976). Electrophoresis could also over-estimate the amount of polymorphism because the enzymes t y p i c a l l y surveyed are those found in r e l a t i v e l y high concentration i n tissues or body f l u i d s . Such enzymes are often c a l l e d Group II enzymes to d i s t i n g u i s h them from more substrate - s p e c i f i c Group I enzymes involved in processes such as energy transformation. Since Group II enzymes do not include regulatory enzymes, also, they are not an adequate sample of the genome. The t o t a l sample of genes represented by a l l enzymes may be less than 1% of the genome. F i n a l l y , some authors have suggested that molecular evolution and organismal evolution are decoupled (Clayton 1981). Thus, while electrophoretic studies may give some insight into the genetic history of populations in an environment of mutation, genetic d r i f t , and migration, quantitative genetic analysis of phenotypic v a r i a t i o n provides information regarding s e l e c t i v e forces as w e l l . Within the Salmonidae, electrophoresis has been used extensively. For reviews of the usage refer to Withler et a l . (1982) and Allendorf and Utter (1979). I assayed a wide array of enzymes, some of which are Group I, some Group II and one, PGM-1, which has been suggested as a regulatory enzyme (Allendorf et a l . 1983). 214 However, I do not imply that the l o c i used here are a representative sample for that would require perfect knowledge of a l l systems. In t h i s chapter I w i l l compare heterozygosity at each locus within each population; test whether the l o c i are in Hardy-Wienberg equilibrium; estimate the genetic migration among the populations; determine the amount of genetic d i f f e r e n t a t i o n among the populations; determine e f f e c t i v e population s i z e ; the di r e c t i o n of genetic migration among the populations; genetic s i m i l a r i t i e s ; and construct a Wagner Tree of relatedness among the populations. MATERIALS AND METHODS Samples of hypaxial muscle, heart, l i v e r and eye tiss u e were c o l l e c t e d from spawning adults during the 1981, 1982, and 1983 seasons. Tissues were transported to the laboratory on ice and stored at - 20 C u n t i l the analysis was done i n the winter of 1983-84. Tissue samples were ground and centrifuged. The supernatants were electrophoresed using the starch gel procedure described by Allendorf and Utter (1979). A t o t a l of 39 l o c i were screened (Table 1). Fi f t e e n of these were monomorphic or did not s t a i n w e l l . The r e s u l t s from the remaining 24 enzymes were used for computations. The most common a l l e l e (see studies by Beacham in Lit e r a t u r e cited) was designated "100". Other a l l e l e s were designated according to the distance they t r a v e l l e d in the gel r e l a t i v e to the distance t r a v e l l e d by the most 215 Table 59. Enzymes w i t h i n t i s s u e s and b u f f e r systems used i n the e l e c t r o p h o r e t i c a n a l y s i s . USED IN TISSUE ENZYME BUFFER ANALYSIS Muscle IDH-1,2 AC YES Muscle ME-1 AC YES Muscle LDH-1,2 RW YES Muscle LDH-3,4 RW YES Muscle LGG-1 RW YES Muscle AGP-1 AC NO Muscle MDH-3,4 AC NO Muscle 6-PG AC NO Muscle PGI-1 RW NO Muscle PGI-2 RW NO Muscle PGI-3 RW YES Muscle PGM-1 RW NO Muscle GL-1 RW NO Heart IDH-2 AC YES Heart AAT-1,2 AC YES Heart MDH-1,2 AC NO Heart MDH-3,4 AC YES Heart PMI AC YES Heart AGP-1,2 AC YES Heart ACON-1,2 AC NO Heart ACON-3,4 AC NO Heart LDH-3 MF YES Heart PGM-1 AC NO Heart GL-1 MF YES Heart LGG-1 MF YES Liver IDH-3,4 AC YES Liver MDH-1,2 AC YES Liver PGM-1 AC YES Liver LDH-4 RW YES Liver PMI AC NO Liver SDH RW YES Eye IDH-3,4 AC YES Eye AAT-3 AC YES Eye MDH-1,2 AC NO Eye MDH-3,4 AC YES Eye GAP AC NO Eye LDH-5 MF NO Eye GL-2 MF YES Eye LGG-1 MF YES 216 common a l l e l e . For example, i f an a l l e l e t r a v e l l e d h a l f the distance t r a v e l l e d by the most common a l l e l e i t was designated as "50". If i t tr a v e l l e d twice the distance t r a v e l l e d by the most common a l l e l e i t was designated as "200". Average heterozygosity was calculated for each of the polymorphic l o c i within each population i n three ways: (1) the proportion of in d i v i d u a l s sampled that are a c t u a l l y heterozygous; (2) heterozygosity based on Hardy-Weinberg expectations (for 2 a l l e l e system H = 2pq, for a 4 a l l e l e system H = 2pq + 2pr + 2ps + 2qr -•- 2qs + 2rs where p, q, r , and s are the frequencies of d i f f e r e n t a l l e l e s ) ; (3) the unbiased estimate of H based on conditional expectations (Nei 1978). To test the hypothesis that each population i s in Hardy-Weinberg equilibrium at each locus a chi-square goodness-of-fit test was employed using the observed genotype frequencies and those expected under Hardy-Weinberg equilibrium. The chi-square test i s suspect i n cases where expected frequencies of some classes are low. Therefore, when more than two a l l e l e s were observed at a locus, the test i s repeated using the genotypes pooled into three c l a s s e s . Pooling i s accomplished by considering a l l a l l e l e s except the common a l l e l e as a sing l e a l l e l e . Three classes of genotype r e s u l t : (1 ) homozygotes for the most common a l l e l e ; (2) heterozygotes for the most common a l l e l e and one of the other a l l e l e s ; and (3) a l l other genotypes. The re s u l t i n g chi-square value i s used with one degree of freedom. 217 The Rate of Migration at Equilibrium Assuming the populations are at equilibrium with respect to migration and that s e l e c t i o n and mutation are n e g l i g i b l e the p r o b a b i l i t y , m, that a randomly drawn a l l e l e i s from a migrant depends on the degree of d i f f e r e n t i a t i o n among the populations, F s t > a n c j the e f f e c t i v e population s i z e , N, i s as follows: 2 F = (1/2N - (1 - 1/2N) x F ) (1 - m) ST ST Thus m can be estimated as: F 0.5 m 1 -1/2N + (1 - 1/2N) x F Calc u l a t i o n of E f f e c t i v e Population Size I used three d i f f e r e n t measures of e f f e c t i v e population s i z e . The f i r s t was simply the mean population si z e from the three years in which samples were co l l e c t e d (1981-1983). 218 N81 + N82 + N83 Ne As a more r e l i a b l e estimate I calculated the harmonic mean of the population s i z e s recorded from 1981-1985. I assumed that the fl u c t u a t i o n in population siz e from year to year during t h i s period would approximate the fluctuations from generation to generation. 1 1 1 1 1 1 1 N e 5 N81 N82 N83 N84 N85 For the f i n a l estimate I assumed that the average i n t e r v a l between generations to be 3 years and calculated the e f f e c t i v e population s i z e as the harmonic mean of the population s i z e s of 1981 and 1984. 1 1 1 1 N e 2 N81 N84 The sex r a t i o was assumed to be 1:1 in the population. 219 DIRECTION OF MIGRATION The pattern among the populations of monomorphism and polymorphism at i n d i v i d u a l l o c i may be used to i n f e r the most l i k e l y migration paths between the stocks. Six pathways e x i s t : Gene flow may be from AB to WB, WB to AB, AB to W, W to AB, WB to W and W to WB (See Figure 25). For the purpose of t h i s analysis I assume that these stocks are i s o l a t e d from other stocks or that contributions from other stocks are n e g l i g i b l e . There are 64 possible models of migratory exchange among these three populations ranging from complete exchange ( that i s , s i g n i f i c a n t migration in a l l six d i r e c t i o n s ) through various degrees of p a r t i a l and complete i s o l a t i o n of the populations. If a locus i s fixed in population 'A' and polymorphic in population 'B' then population 'B' w i l l remain polymorphic even with the introduction of migrants from population 'A1. Migration in the reverse d i r e c t i o n w i l l eventually result i n population 'A' becoming polymorphic. For example, i f straying occurred only from WB to W and W to WB there should not be l o c i that are fixed in AB, fixed i n WB and polymorphic in W or fixed i n AB, polymorphic in WB and fixed in W or polymorphic in AB, polymorphic in W and fixed in WB or polymorphic in both AB and WB and fixed i n W. GENETIC DISTANCE Nei (1971) developed a s t a t i s t i c a l method for u t i l i z i n g electrophoretic data for estimating the number of codon differences per gene and divergence time between c l o s e l y related species. This method i s useful for the study of Migration Paths B e t w e e n Bush and Walker Creeks Figure 26. Pathways for gene flow among AB, WB, and W. 221 gene differences between races or c l o s e l y related l o c a l populations within a species (Nei 1972). Nei (1971) defined the normalized i d e n t i t y between populations to take into account the e f f e c t of polymorphism within populations. This s t a t i s t i c related to the accumulated number of gene differences per locus, which was now c a l l e d genetic distance (D). D has several useful properties: (1) It i s related to Malecot's c o e f f i c i e n t of kinship i n a simple way; (2) It measures the accumulated number of gene substitutions per locus; (3) I f the rate of gene substituations per year i s constant i t i s l i n e a r l y related to evolutionary time; (4) In some migration models i t i s l i n e a r l y related to geographical distance or area. The simplest measure of distance for a sin g l e locus i s the Prevosti distance (Wright 1978) which i s h a l f the sum of the absolute differences between the a l l e l i c frequencies of the two populations, (D •= 0.5 This takes the value of 1.0 in the case of two populations in which d i f f e r e n t a l l e l l e s are f i x e d . The index of multiple l o c i i s the arithmetic mean of these for the separate l o c i (Wright 1978). Qx - qy . 222 The concept of genetic distance was f i r s t used in connection with the means of q u a n t i t a t i v e l y varying characters. A consistant set of distances requires that a l l of the subpopulations be located at points in a E u c l i d i a n hyperspace with an axis for each chosen variable. The distance between two populations i s the square root of the sum of the squared differences between thei r coordinates by the extended Pythagorean theorem. An example i s Pearson's (1926) c o - e f f i c i e n t of r a c i a l likeness. Mahalanobis (1936) generalized distance took account of c o r r e l a t i o n s among the characters by using obliquely i n c l i n e d axes. Rogers (1972) proposed the formula: D( Xy) = [0.5 £ k ( q x ( i ) - q y ( i ) ) 2 ] 0 . 5 for distance with respect to a locus, and defined distance with respect to multiple l o c i as the arithmetic average of the c o - e f f i c i e n t s for the separate l o c i D = (£) £D. Wright (1978) suggests a modification of Dt = C/L) 2D2]£. This gives less weight to l o c i i n which the difference in a l l e l i c frequencies are small. C a v a l l i - S f o r z a and Edwards (1967) proposed a measure of genetic distance which d i f f e r s from Roger's D in taking the square roots of the a l l e l i c frequencies >v/qx(i) and-\/qy(i) as the coordinates of the points representing the population instead of the frequencies themselves. 223 Any two populations which have no a l l e l e s in common are at the same distance, 0 , apart. The scale of are distances i s transformed to that of gene frequencies by the angular transformation: D = O/7J- cos-1 (1-2q) and thus i s stretched symmetrically near the extremes but condensed symmetrically near the middle. Another measure used i s the chordal distance: CD r [ 2 / v / ( 2 ) / y * c o s ) i n which cos Q = —K-^/CqxC i ) • q y ( i ) ) ' Nei (1978) modified the measures of genetic i d e n t i t y and genetic distance developed by Nei (1972) to remove the biases r e s u l t i n g from samples from a small number of i n d i v i d u a l s . Distance Wagner Procedure: F a r r i s (1970, 1972) developed a procedure c a l l e d the Distance Wagner procedure to construct evolutionary trees with minimal s e n s i t i v i t y to heterogeneities in rates of divergence. 224 A Wagner Tree for a c o l l e c t i o n S, of OTU's (operational taxonomic units) i s a tree with the following properties: W1, the c o l l e c t i o n S i s a subset of the set of a l l the nodes of the tree; W2, the length of the Wagner tree, as defined below, i s less than or equal to the length of any other tree s a t i s f y i n g condition W1 ( F a r r i s 1972). The length of a tree i s defined by F a r r i s (1972) as follows. Each node A of the tree i s assumed to be described by a well-defined value, or character state, x ( i , A), for each element of a set of characters indexed by i . The phenetic difference between any two such nodes i s defined to be: D (A,B) = | x ( i , A ) - x(i,B) | The length of a branch of a tree i s defined as the phenetic difference between the two nodes forming the end points of the branch. The length of the tree i s defined to be the sum of the branch lengths over a l l the branches of the tree. A tree i s said to be most parsimonius i f i t has minimum length according to the measure defined in the equation above. RESULTS A l l e l e frequencies at the polymorphic l o c i within each population are shown in Table 60 (a,b,c,d). Frequencies of the most common a l l e l e varied Table 60a. Sample s i z e s , and allozyme frequencies of polymorphic l o c i used i n the a n a l y s i s . MUSCLE IDH-1,2 ME-1 LDH-1,2* LDH-3,4 LGG-1 PGI-3* a l l e l e a l l e l e a l l e l e a l l e l e a l l e l e a l l e l e Ln STOCK N 100 64 N 100 120 N 100 136 N 100 136 N 100 75 N 100 80 AB 158 0.968 0.032 153 0.601 0.399 157 0.5 0.5 158 0.513 0.487 153 0.889 0.111 128 1.000 0.000 WB 81 0.907 0.093 81 0.648 0.352 81 0.5 0.5 81 0.500 0.500 81 0.821 0.179 50 1.000 0.000 W 92 0.995 0.005 90 0.572 0.428 92 0.5 0.5 92 0.500 0.500 85 0.847 0.153 58 0.991 0.009 Table. 60b. Sample s i z e s , and allozyme frequencies o f polymorphic l o c i used i n the a n a l y s i s . HEART IDH-1,2 AAT-1,2 MDH-3,4 PMI AGP-1,2 LDH-3 a l le le a l l e l e al le le a l le le al le le al le le STOCK N 100 110 N 100 115 N 100 120 N 100 92 N 100 93 N 100 136 AB 148 0.970 0.030 138 0.819 0.181 137 0.007 0.993 127 0.807 0.193 41 0.976 0.024 156 0.997 0.000 WB 76 0.941 0.059 7.6 0.757 0.243 80 0.000 1.000 72 0.840 0.160 5 1.000 0.000 81 0.994 0.006 W 88 0.966 0.034 79 0.823 0.177 89 0.000 1.000 72 0.889 0.111 45 0.856 0.144 90 1.000 0.000 Table 60c. Sample sizes, and allozyme frequencies of polymorphic loci used in the analysis. STOCK N 100 80 N 100 75 N 100 27 40 83 N 100 200 N 100 110 N 100 136 AB . 120 0.983 0.017 153 0.886 0.114 85 0.682 0.035 0.176 0.106 153 0.931 0.069 1 1.000 0.000 103 0.927 0.073 WB 73 1.000 0.000 810.9010.099 48 0.563 0.063 0.135 0.240 80 0.8810.119 6 1.000 0.000 80 0.994 0.006 W 87 1.000 0.000 89 0.876 0.124 56 0.393.0.036 0.429 0.143 91 0.775 0.225 21 0.905 0.095 91 0.956 0.044 Table 60d. Sample sizes, and aHozyme frequencies of polymorphic loci used i n the analysis. LIVER EVE SDH* IDH-3,4 AAT-3 ICH-3,4 GL-2* LGC-1 allele allele allele allele allele allele ho 00 STOCK N 100 90 80 N 100 27 40 83 N 100 115 N 100 120 N 100 80 N 100 75 AB 22 0.500 0.500 0.00 123 0.402 0.020 0.492 0.085 120 0.837 0.162 1 0.000 1.000 152 0.987 0.013 1 20 0.946 0.054 WB 3 0.500 0.500 0.000 63 0.452 0.032 0.397 0.119 74 0.932 0.068 20 0.000 1.000 77 1.000 0.000 62 0.903 0.097 W 11 0.500 0.455 0.045 74 0.324 0.014 0.595 0.068 80 0.781 0.219 38 0.000 1.000 89 1.000 0.000 69 0.957 0.043 229 from f i x a t i o n to 0.429 in the m u l t i - a l l e l i c IDH-3,4 locus expressed in l i v e r tissue of Walker Creek spawners. The percentage of polymorphic l o c i was s i m i l a r among the populations. AB had the highest percentage of polymorphic l o c i ( 53.85 % ) compared to WB (43.59 ?o ) and W ( 48.72 % ) . However, the frequency of the most common a l l e l e was 0.95 or l e s s at 38.46 % of the l o c i of WB samples compared to 35.89 % of the l o c i of AB samples and 35.90 % of the l o c i of W samples. There were several l o c i that showed polymorphism in one or two of the populations and not the remainder ( Table 60a-d ). For example, the PGI-3 locus expressed i n the muscle t i s s u e and the PGM-1 locus expressed in the l i v e r were polymorphic in the W stock but not in the WB or AB stocks. S i m i l a r l y , the MDH-3,4 and GL-1 l o c i expressed in the heart t i s s u e , and the GL-2 locus expressed in the eye were monomorphic in WB and W but polymorphic in AB. The LDH-3 locus expressed i n heart ti s s u e was monomorphic in W and polymorphic i n the other two stocks. The AGP-1,2 locus was monomorphic in WB and polymorphic i n the other two stocks. However, only f i v e f i s h from WB were scored for t h i s locus. Polymorphism might be detected as more in d i v i d u a l s were sampled. The patterns of polymorphism and f i x a t i o n among the populations could occur i f there was no interbreeding among the populations or i f migration occurred from the WB stock into the AB stock while there was no migration i n the other possible d i r e c t i o n s . 230 Among the polymorphic l o c i , the average heterozygosity was less than 0.30. Mean heterozygosity was lowest when calculated from a l l e l e frequencies ( For AB, H r 0.206 with standard error of 0.041; For WB, H = 0.213 with standard error of 0.044; For W, H = 0.231 with standard error of 0.043). The mean unbiased heterozygosity estimates were 0.208 ( S.E. = 0.042) for AB, 0.218 ( S.E. = 0.046 ) for WB, and 0.234 ( S.E. = 0.044 ) for W. The estimates obtained from d i r e c t count of heterozygotes were highest with 0.257 ( S.E. = 0.069) for AB, 0.277 ( S.E. = 0.072 ) for WB and 0.289 ( S.E. -0.069) for W. Heterozygosity varied greatly from locus to locus ( Table 61). At some l o c i H was very low or near f i x a t i o n ( e.g. LGG-1 in eye whereas i n other l o c i H might be 1.000 by a d i r e c t count ( e.g. LDH in muscle) (Table 61). Several l o c i were not in Hardy-Weinberg equilibrium (Table 61). ME-1, LDH-1,2 and LDH-3,4 in muscle tis s u e and IDH-3,4 in eye tissue did not conform to Hardy-Weinberg equilibrium i n any of the populations. The AB population did not conform to Hardy-Weinberg equilibrium at the MDH-3,4 and GL-1 l o c i in heart t i s s u e , SDH i n l i v e r tissue and GL-2 and AAT-3 i n eye tiss u e . The W population did not conform to Hardy-Weinberg equilibrium at the IDH-3,4, MDH-1,2, LDH-4 and SDH l o c i in the l i v e r t i s s u e and the AAT-3 locus i n the eye. Loci that were not in Hardy-Weinberg equilibrium f e l l into three categories. Those with a deficiency of heterozygotes r e l a t i v e to 231 Table 61. Heterozygosity and test of conformance to Hardy-Weinberg equilibrium of polymorphic l o c i within each population. ( H = Heterzygpsity calculated uaing Hardy-Weinbecg expectations* HUB = Unbiased Heterozygosity Estimate (Nei 1978); HDC = Proportion o f Heterozygotes; P = P value for Chi-Square Test ). POPULATION AB WB LOCUS H HUB HDC H HUB HDC H HUB HDC MUSCLE IDH-1,2 0.061 0.061 0.063 0.681 ME-1 0.479 0.481 0.078 0.000 LDH-1,2* 0.500 0.502 1.000 0.000 LDH-3,4 0.500 0.501 0.975 0.000 LGG-1 0.198 0.198 0.209 0.467 PGI-3* 0.000 0.000 0.000 1.000 0.168 0.169 0.185 0.358 0.011 0.011 0.011 0.958 0.456 0.459 0.037 0.000 0.490 0.492 0.011 0.000 0.500 0.503 1.000 0.000 0.500 0.503 1.000 0.000 0.500 0.503 1.000 0.000 0.500 0.503 1.000 0.000 0.294 0.296 0.333 0.228 0.259 0.261 0.282 0.408 0.000 0.000 0.000 1.000 0.017 0.017 0.017 0.947 HEART IDH-1,2 AAT-1,2 MDH-3,4 PMI AGP-1,2 LDH-3 GL-1* LGG-1 0.059 0.059 0.297 0.298 0.014 0.015 0.311 0.313 0.048 0.048 0.006 0.006 0.033 0.033 0.203 0.203 0.061 0.275 0.000 0.339 0.049 0.006 0.000 0.229 0.703 0.399 0.000 0.325 0.873 0.968 0.000 0.110 0.111 0.368 0.000 0.268 0.000 0.012 0.000 0.178 0.112 0.371 0.000 0.270 0.000 0.012 0.000 0.179 0.118 0.382 0.000 0.292 0.000 0.012 0.000 0.198 0.583 0.754 1.000 0.462 1.000 0.955 1.000 0.324 0.066 0.292 0.000 0.198 0.247 0.000 0.000 0.217 0.066 0. 0.293 0. 0.000 0. 0.199 0. 0.250 0. 0.000 0. 0.000 0. 0.218 0. 068 0.741 304 0.711 000 1.000 194 0.895 289 0.257 000 1.000 000 1.000 247 0.183 LIVER IDH-3,4 MDH-1,2 PGM-1 LDH-4 SDH* 1 0.491 0.494 0.435 0.102 0.128 0.128 0.111 0.106 0.000 0.000 0.000 1.000 0.135 0.136 0.126 0.508 1 0.604 0.610 0.667 0.720 0.209 0.211 0.237 0.228 0.000 0.000 0.000 1.000 0.012 0.013 0.013 0.955 1 0.640 0.646 0.625 0.349 0.351 0.451 0.172 0.177 0.190 0.084 0.085 0.022 0.500 0.512 1.000 0.000 0.500 0.600 1.000 0.083 0.541 0.018 0.006 0.630 0.000 2 0.567 1.000 0.012 EYE IDH-3,4 0.588 0.591 0.789 0.000 AAT-3 0.272 0.273 0.325 0.034 MDH-3,4 0.000 0.000 0.000 1.000 GL-2* 0.026 0.026 0.000 0.000 LGG-1 0.102 0.103 0.092 0.248 0.623 0.628 0.841 0.000 0.126 0.127 0.135 0.533 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 0.175 0.176 0.194 0.399 0.537 0.540 0.689 0.030 0.342 0.344 0.438 0.012 0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000 0.083 0.084 0.087 0.706 1 Chi-Square Test with Pooling, P = 0.087 for AB, P = 0.486 for WB, P - 0.043 for W. 2 Chi-Square Test with Pooling, P r 0.001 3 Chi-Square Test with Pooling, P = 0.0000 for AB, P = 0.003 for WB, P = 0.003 for W. 232 Hardy-Weinberg expectations, those with an excess of heterozygotes r e l a t i v e to Hardy-Weinberg expectations and those that were nearly monomorphic. In the l a t t e r case the expected frequency of the l e s s common homozygote was so low that the chi-square value was unusually high. ME-1 i n muscle had a d e f i c i e n c y of heterozygotes. In c o n t r a s t , LDH-1,2 and LDH-3,4 from muscle, IDH-3,4 from the eye i n a l l populations and SDH from the l i v e r , AAT-3 from the eye i n AB and W, and MDH-12 from l i v e r t i s s u e from W samples had an excess of heterozygotes. Loci c l o s e to f i x a t i o n included the MDH-3,4 and GL-1 l o c i from the heart t i s s u e , the GL-2 from the eye t i s s u e from the samples of the AB population as w e l l as the LDH-4 from l i v e r t i s s u e from the W samples. O v e r a l l , the m a j o r i t y of the polymorphic l o c i i n each population had more heterozygotes than p r e d i c t e d under Hardy-Weinberg c o n d i t i o n s . Twelve out of 21 polymorphic l o c i found i n the t i s s u e s sampled from the AB spawners had an excess of heterozygotes r e l a t i v e to Hardy-Weinberg expectations. Sixteen out of 17 polymorphic l o c i found i n t i s s u e s from the WB spawners had an excess of heterozygotes. F i f t e e n out of 19 l o c i showed an excess of heterozygotes i n W samples. The amount of genetic d i f f e r e n t i a t i o n among the populations r e l a t i v e to a h y p o t h e t i c a l group of subpopulations, each homozygous, but having the same o v e r a l l frequency as the r e a l populations v a r i e d from locus to locus (Table 62). Three l o c i , heart AGP-1,2, l i v e r IDH-3,4 and PGM-1, had F ( s t ) values that i n d i c a t e d moderate d i f f e r e n t i a t i o n ( Wright 1978, H a r t l 1980). The other F ( s t ) values were l e s s than 0.05. The o v e r a l l mean of 0.016 suggested that 233 Table 62. Summary o f F - S t a t i s t i c s at a l l l o c i . TISSUE LOCUS F(IS) F(IT) F(ST) MUSCLE IDH12 -0.080 -0.045 0.032 MUSCLE ME 0.911 0.912 0.004 MUSCLE LDH12 -1.000 -1.000 0.000 MUSCLE LDH34 -0.984 -0.983 0.000 MUSCLE LGG-1 -0.099 -0.092 0.006 MUSCLE PGI-3 -0.009 -0.003 0.006 HEART IDH12 -0.047 -0.043 0.004 HEART AAT12 -0.004 0.001 0.006 HEART MDH34 1.000 1.000 0.005 HEART PMI -0.061 -0.052 0.009 HEART AGP12 -0.146 -0.060 0.075 HEART LDH-3 -0.005 -0.003 0.002 HEART GL 1.000 1.000 0.011 HEART LGG-1 -0.128 -0.126 0.001 LIVER IDH34 0.005 0.060 0.056 LIVER MDH12 -0.165 -0.123 0.036 LIVER PGM-1 -0.105 -0.033 0.066 LIVER LDH-4 0.306 0.319 0.019 LIVER SDH -0.946 -0.943 0.002 EYE IDH34 -0.327 -0.305 0.017 EYE AAT-3 -0.213 -0.176 0.031 EYE GL-2 1.000 1.000 0.009 EYE LGG -0.032 -0.023 0.009 MEAN -0.265 -0.244 0.016 234 only a limited amount of genetic d i f f e r e n t i a t i o n in l o c i detectable by electrophoresis had occurred among the populations. S i g n i f i c a n t differences in a l l e l e frequency among the populations occurred at 6 l o c i , muscle IDH-1,2, heart AGP-1,2, l i v e r IDH-3,4, MDH-1,2 and LDH-4 and eye AAT-3 (Table 63). EFFECTIVE POPULATION SIZES AND MIGRATION RATES The e f f e c t i v e population s i z e s estimated using the three methods varied within populations from plus or minus 7 % to 22 % of the combined average ( Table 64) The estimates of the e f f e c t i v e population s i z e of WB varied least while the estimates of W e f f e c t i v e population size were the most depedent upon the method. The lowest estimates were by the algebraic mean method for WB, the harmonic mean method for W and the modified harmonic mean method for AB. The AB population was about 10 times larger than the WB population and about 6 times larger than the W population. The mean p r o b a b i l i t y that an a l l e l e drawn at random w i l l be from a migrant was highest i n the WB population (4.1 to 4.5 ? o ) , next highest in the W population (1.9 to 3.1 %) and lowest in the AB population (0.38 to 0.45 %) (Table 65). A l l measures of genetic distance or genetic i d e n t i t y except Nei's (1978) unbiased minimum distance show that AB and WB are the c l o s e s t in t h e i r enzyme 235 Table 63. Contingency chi-square a n a l y s i s at a l l l o c i . Chi Square and p r o b a b i l i t y values are f o r the hypothesis that the samples were drawn from the same p o p u l a t i o n . TISSUE LOCUS NO. OF ALLELES CHI-SQUARE D.F. P MUSCLE IDH12 2 18.277 2 0.00011 MUSCLE ME-M 2 2.089 2 0.35193. MUSCLE LDH12 2 0.000 2 1.00000 MUSCLE LDH34 2 0.106 2 0.94844 MUSCLE LGG-1 2 4.417 2 0.10985 MUSCLE PGI-3 2 3.076 2 0.21486 HEART IDH12 2 2.380 2 0.30425 HEART AAT12 2 2.906 2 0.23391 HEART MDH34 2 2.475 2 0.29007 HEART PMI-H 2 4.532 2 0.10374 HEART AGP 12 2 9.128 2 0.01042 HEART LDH-3 . 2 1.070 2 0.58570 HEART GL-H 2 5.372 2 0.06816 HEART LGG-1 2 0.533 2 0.76618 LIVER IDH34 4 42.329 6 0.00000 LIVER MDH12 2 25.679 2 0.00000 LIVER PGM-1 2 1.436 2 0.48775 LIVER LDH-4 2 9.529 2 0.00853 LIVER SDH-L 3 2.338 4 0.67392 EYE IDH34 4 11.522 6 0.07353 EYE AAT-3 2 13.819 2 0.00100 EYE GL-2- 2 4.396 2 0.11102 EYE LGG-E 2 3.659 2 0.16052 (TOTALS) 171.065 56 0.00000 236 Table 64. Estimates of e f f e c t i v e p o p u l a t i o n s i z e f o r each popu l a t i o n u s i n g the a l g e b r a i c mean, harmonic mean and modified harmonic mean methods described i n the t e x t . Population METHOD 1 Algebraic Mean METHOD 2 Harmonic Mean METHOD 3 Modified Harmonic AB 3800 4078.14 3416.77 WB 318.33 325.30 355.56 W 775.33 478.36 593.55 237 Table 65. P r o b a b i l i t y t h a t a randomly chosen a l l e l e w i l l be from a migrant i n d i v i d u a l assuming random mating, no s e l e c t i o n , or mu l t a t i o n . (See t e x t f o r d e t a i l s o f c a l c u l a t i o n o f m from F and N). EFFECTIVE POPULATION SIZE CALCULATED BY METHOD 1 METHOD 2 METHOD 3 LOCUS AB WB W AB WB W AB WB W Muscle IH-1,2 0.00198 0.02294 0.00961 0.00185 0.02247 0.01544 0.00221 0.02061 0.01250 ME 0.01599 0.15215 0.07175 0.01492 0.14958 0.10922 0.01774 0.13939 0.09082 LDH-1,2 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 LDH-3,4 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 LGG-1 0.01072 0.10920 0.04949 0.01000 0.10723 0.07675 0.01191 0.09942 0.06323 PGI-3 0.01072 0.10920 0.04949 0.01000 0.10723 0.07675 0.01191 0.09942 0.06323 Heart IDH-1,2 0.01599 0.15215 0.07175 0.01492 0.14958 0.10922 0.01774 0.13939 0.09082 AAT-1,2 0.01072 0.10920 0.04949 0.01000 0.10723 0.07675 0.01191 0.09942 0.06323 MDH-3,4 0.01284 0.12715 0.05858 0.01198 0.12492 0.09016 0.01425 0.11606 0.07456 PMI 0.00717 0.07666 0.03372 0.00668 0.07520 0.05301 0.00796 0.06945 0.04338 AGP-1,2 0.00081 0.00955 0.00395 0.00076 0.00935 0.00638 0.00090 0.00856 0.00515 LDH-3 0.03130 0.25126 0.13020 0.02925 0.24771 0.18931 0.03463 0.23342 0.16092 GL 0.00586 0.06391 0.02779 0.00547 0.06267 0.04392 0.00651 0.05779 0.03585 Liver LGG-1 0.05988 0.37611 0.22014 0.05613 0.37199 0.30058 0.06595, 0.35515 0.26310 IDH-3,4 0.00111 0.01298 0.00539 0.00103 0.01271 0.00870 0.00123 0.01165 0.00703 MDH-1,2 0.00176 0.02039 0.00852 0.00164 0.01997 0.01371 0.00195 0.01831 0.01109 PGM-1 0.00093 0.01093 0.00453 0.00087 0.01070 0.00731 0.00103 0.00980 0.00591 LDH-4 0.00338 0.03824 0.01624 0.00315 0.03746 0.02594 0.00376 0.03444 0.02106 SDH 0.03130 0.25126 0.13020 0.02925 0.24771 0.18931 0.03463 0.23342 0.16092 Eye IDH-3,4 0.00378 0.04254 0.01814 0.00353 0.04168 0.02892 0.00420 0.03833 0.02350 AAT-3 0.00205 0.02368 0.00993 0.00191 0.02319 0.01595 0.00228 0.02128 0.01291 GL 0.00717 0.07666 0.03372 0.00668 0.07520 0.05301 0.00796 0.06945 0.04338 LGG 0.00717 0.07666 0.03372 0.00668 0.07520 0.05301 0.00796 0.06945 0.04338 MEAN 0.00402 0.04506 0.01926 0.00375 0.04416 0.03067 0.00447 0.04062 0.02494 238 patterns. WB and W, the l a t e spawning stocks, are the least s i m i l a r . Nei's (1978) unbiased minimum distance, which i s the only measure corrected for the bias from sampling a small number of i n d i v i d u a l s , shows that AB and WB are most s i m i l a r while AB and W are the least s i m i l a r populations (Table 66, Figure 27). A locus by locus comparison using Nei's (1978) unbiased genetic i d e n t i t y shows a wide variety of possible r e l a t i o n s h i p s . WB and W were least s i m i l a r using Muscle IDH-1, 2, ME, Heart AGP - 1, 2, Liver and Eye IDH - 3, 4, Eye AAT - 3, LGG. AB and WB were least s i m i l a r using Muscle LGG and Liver CDH - 4, AB and W were least s i m i l a r using Heart PMI and; Liver MDH - 1 , 2 (Table 67). Figure 28 shows a Wagner Tree using Prevosti distance. The figure suggests that the populations in Bush Creek s p l i t from the population i n Walker Creek before s p l i t t i n g into an e a r l y and l a t e spawning stock. DISCUSSION There are several reasons why l o c i may not conform to the expectations of Hardy-Weinberg equilibrium. It i s possible that non-random mating takes place with respect to the locus i n question. The sexes may d i f f e r in a l l e l e frequency at a locus. Progeny that d i f f e r i n genotype may have unequal v i a b i l i t y . 239 Figure 27. Genetic s i m i l a r i t y using Nei's (1978) Unbiased Genet S i m i l a i r i t y . SIMILARITY 0.95 6.96 0.97 0.98 0.99 1.00 - - H 1 H H + H H H + H + * AB ****** * * WB * ****** ^ + + + + + + + + + + + + 0.95 0.96 0.97 0.98 0.99 1.00 240 Table 66. Genetic d i s t a n c e and genetic i d e n t i t y over a l l l o c i , BELOW DIAGONAL: ROGERS "D" (WRIGHT 1978) ABOVE DIAGONAL: NEI (1972) GENETIC IDENTITY POPULATION 1 2 AB WB W ****** 0.0497 0.0785 0.9969 ****** 0.0876 0.9922 0.9902 ****** Table 67. Genetic d i s t a n c e over a l l l o c i . BELOW DIAGONAL: PREVOSTI DISTANCE (WRIGHT,1978) ABOVE DIAGONAL: NEI (1972) GENETIC DISTANCE POPULATION 1 AB WB W ****** 0.0376 0.0484 0.0031 ****** 0.0622 0.0078 0.0098 ****** Table 68. Genetic d i s t a n c e over a l l l o c i . BELOW DIAGONAL: CAVALLI-SFORZA & EDWARDS (1967) CHORD DISTANCE ABOVE DIAGONAL: NEI (1978) UNBIASED MINIMUM DISTANCE POPULATION 1 2 3 1 AB ****** 0.0000 0.0043 2 WB 0.0623 ****** 0.0036 3 W 0.0871 0.1045 ****** 2 4 1 Table 69. Genetic d i s t a n c e over a l l l o c i . BELOW DIAGONAL: CAVALL I -SFORZA & EDWARDS ( 1 9 6 7 ) ARC DISTANCE ABOVE DIAGONAL: N E I ( 1 9 7 2 ) MINIMUM DISTANCE POPULATION 1 2 3 1 AB * * * * * * 0 . 0 0 2 5 0 . 0 0 6 2 2 WB 0 . 0 6 2 2 * * * * * * 0 . 0 0 7 7 3 W 0 . 0 8 6 9 0 . 1 0 4 2 * * * * * * Table 70. Genetic d i s t a n c e over a l l l o c i . BELOW DIAGONAL: EDWARDS ( 1 9 7 1 , 1 9 7 4 ) " E " DISTANCE POPULATION 1 2 3 1 AB * * * * * * 2 WB 0 . 0 6 7 6 * * * * * * 3 W 0 . 0 9 4 0 0 . 1 1 2 5 * * * * * * 242 Table 71. IDENTITY). SINGLE-LOCUS GENETIC SIMILARITY (NEI 1978 UNBIASED GENET! COMPARISON LOCUS TISSUE AB-WB WB-W AB-W IDH12 Muscle 0.9983 0.9998 0.9960 ME Muscle 1.0000 1.0000 1.0000 IDH12 Heart 1.0000 1.0000 1.0000 AAT12 Heart 0.9983 1.0000 0.9983 MDH34 Heart 1.0000 1.0000 1.0000 PMI Heart 1.0000 0.9957 1.0000 AGP12 Heart 1.0000 0.9920 0.9879 LDH-3 Heart 1.0000 1.0000 .1.0000 GL-1 Heart 0.9999 0.9999 1.0000 LGG-1 Heart 1.0000 1.0000 1.0000 IDH34 Liver 0.9800 0.8506 0.8495 Figure 28. WAGNER TREE PRODUCED BY ROOTING AT MIDPOINT OF LONGEST PATH (Base measure used Prevosti Distance (Wright 1978)) ****************************************** DISTANCE FROM ROOT 0.00 0.00 0.01 0.01 0.01 0.02 0.02 0.02 0.02 0.03 0.03 *************************************** ****************** * ************************************************************************************ \tfQ * ***************************************************************************************************** ^ 0.00 0.00 0.01 0.01 0.01 0.02 0.02 0.02 0.02 0.03 0.03 TOTAL LENGTH OF TREE = 0.074 244 Assortative mating and mating between r e l a t i v e s are examples of s i t u a t i o n s with non-random mating. If parents with s i m i l a r genotypes mate there w i l l be an excess of homozygous ind i v i d u a l s in the progeny. If negative assortative mating occurs then an excess of heterozygotes might occur in the next generation. A c l a s s i c a l example of p o s i t i v e assortative mating occurs with respect to flowering season in plants (Hartl 1980). If the length of time an i n d i v i d u a l flowers i s l i m i t e d r e l a t i v e to the length of the o v e r a l l breeding season then plants which p r e f e r e n t i a l l y flower during the early part of the breeding season w i l l be pollenated by plants with a s i m i l a r p r o c l i v a t y for early flowering and those plants that flower l a t e i n the breeding season w i l l be pollenated mainly by other l a t e blooming i n d i v i d u a l s . In humans, po s i t i v e assortative mating occurs for height and I.Q. score. Negative assortative mating i s demonstrated in plants, such as primroses, where there i s heterostyly (Stebbins 1976). Rarely are t r a i t s coded for by a single locus involved in assortative mating although the phenomenon of "rare male advantage" observed in Drosophila spp. by Ehrman (1970), Parsons (1977) and others might q u a l i f y as an example. Mate choice based on enzyme differences has not been demonstrated i n fishes but might be possible given fishes great prowess in o l f a c t i o n (Hasler 1954). P a c i f i c salmon have frequently been noted for t h e i r a b i l i t y to d i s t i n g u i s h between t h e i r natal t r i b u t a r y and confleunt waters by differences in odor (Groot et a l . 1986, Quinn and Tolson 1986, Olsen 1986). S l i g h t differences in metabolic enzymes might allow a returning f i s h to i d e n t i f y a stray from another population. 245 I f males and females have d i f f e r e n t a l l e l e frequencies or are fixed for opposing a l l e l e s then an excess of heterozygotes from the expected may result in the next generation. This s i t u a t i o n does not seem to be l i k e l y to l a s t for more than one generation unless accompanied by unequal v i a b i l i t y of d i f f e r e n t genotypes dependent on the sex of the i n d i v i d u a l . F i n a l l y , unequal v i a b i l i t y of genotypes may be responsible for the excess or deficiency of heterozygous i n d i v i d u a l s . Overdominance has been noted for the two a l l e l e s coding for the Beta chain of human hemoglobin ( C a v a l l i - S f o r z a 1974). The normal type homozygote i s s e n s i t i v e to malaria while the other homozygote r e s u l t s in s i c k l e - c e l l anemia. The heterozygote i s mildly anemic and r e s i s t a n t to malaria. Heterozygote i n f e r i o r i t y occurs p r i m a r i l y in conjunction with chromosomal abnormalities such as translocations. Ideally, e f f e c t i v e population si z e should be calculated under the following conditions : 1) non-overlapping generations; 2) using a s e r i e s of population s i z e s from generation to generation; 3) accounting for divergence of the sex r a t i o from 1:1. Although chum salmon are semelparous organisms there i s a degree of overlap i n the generations due to v a r i a t i o n in age at maturity. T y p i c a l l y , a spawning population w i l l contain f i s h that were born in several d i f f e r e n t years. Thus, to some extent the f i r s t condition i s violated in my c a l c u l a t i o n s . The average i n t e r v a l from b i r t h to death i s usually between three and four years so that one cannot r e l i a b l y use population s i z e s from a progression of years ( i . e . one cannot use population siz e every fourth year as a d i s c r e t e estimate). However, interbreeding 246 between f i s h born during d i f f e r e n t years should tend to homogenize the gene pool from year to year. Therefore, average population s i z e could be calculated from data from successive years. As well, I have assumed equal sex r a t i o . Generally, during the early part of the spawning period there are more males than females while l a t e r on females predominate in numbers. However, sex r a t i o i s usually 1:1 during the peak of the run when the majority of the f i s h are spawning. The Distance Wagner procedure has been c r i t i c i z e d by Prager and Wilson (1978) as not minimizing SD as well as other procedures. F a r r i s (1978) acknowledged that improvements in t h i s aspect of the procedure could be made but concluded that the procedure was s t i l l the best method for a r r i v i n g at geaneologies. Swofford (1978) independently concluded that the distance Wagner method was superior in both finding the most parsimonious so l u t i o n as well as minimizing the standard deviation. For a d e t a i l e d discussion of a l t e r n a t i v e methods see Prager and Wilson (1978), F a r r i s (1978) and Swofford (1978). The pattern of genetic distance measures and the distance Wagner procedure suggest that the populations are divided by geography rather than timing of spawning. This r e s u l t i s i n t e r e s t i n g because i t may suggest an explanation for the lack of success of many investigators to f i n d genetic differences among seasonally spawning stocks using electrophoretic a n a l y s i s . Given that isozyme v a r i a t i o n i s r e l a t i v e l y neutral to s e l e c t i o n these r e s u l t s may r e f l e c t an h i s t o r i c a l separation among the populations. In recent times quantitative genetic v a r i a t i o n has been subjected to s e l e c t i o n caused by the 247 separation i n spawning timing. In e f f e c t , the late populations have possibly undergone a form of convergent evolution towards having s i m i l a r quantitative genetic responses. 248 Summary and Synthesis of empirical r e s u l t s . "...the same old arguments are c i t e d again and again i n favor of sympatric speciation, no matter how d e c i s i v e l y they have been disposed previously...Sympatric speciation i s l i k e the Lernaean Hydra which grew two new heads whenever one of i t s old head was cut o f f . " Ernst Mayr (1963) 249 Chum salmon, Oncorhynchus keta, i s an anadromous, north temperate salmonid species with considerable phenotypic v a r i a t i o n among populations. Phenotypic v a r i a t i o n among chum salmon populations i s associated with differences i n location and season of spawning. While the genetic basis of differences associated with d i f f e r e n t spawning lo c a l e s i s well established the genetic basis for phenotypic v a r i a t i o n among seasonally separated spawning populations has received l i t t l e a t t e n t i o n . Chum salmon l i f e h istory can be divided into s i x phases: coastal adult, spawning adult, egg to f r y , downstream migrant, coastal juvenile and pelagic j u v e n i l e . Mortality and se l e c t i o n are s i g n i f i c a n t at each phase but most of the mortality occurs during the early l i f e h i s t o r y in the egg to f r y , downstream migrant and coastal j u v e n i l e phases. Phenotypic d i v e r s i t y among sexes, populations and "races" suggests that there are several l e v e l s of genetic organization in chum salmon below the l e v e l of the species. In p a r t i c u l a r , seasonal races may d i f f e r in location of spawning, time of spawning migration, time of spawning, siz e at maturity, morphology, distance migrated from the sea, average weight of spawners, age at maturity, and absolute fecundity. Cold season spawners can migrate further upstream, be longer, heavier, more fecund, mature at older ages, have slimmer bodies, slimmer heads and longer pectoral f i n s than warm season spawners. 250 Several hypotheses have been proposed to account for the evolution of seasonally separated spawning populations i n f i s h e s . However, each theory has i t s flaws and only Frost (1966) has approached the problem with a genetical perspective. The analyses of phenotypic and genetic d i v e r s i t y among seasonally separated chum salmon populations has been confounded with geographic separation since seasonally separated stocks may also spawn 100's of km apart. The l i n k between season of reproduction and temperature of development raises the p o s s i b i l i t y that a l l phenotypic d i v e r s i t y observed i s caused by environmental rather than genetic factors. The r e s u l t s of my investigations at Bush and Walker creeks suggest that seasonally separated populations of chum salmon in the same locale may not be very d i f f e r e n t in phenotype. I established that there were three spawning stocks in these r i v e r s : an autumn spawning stock, AB, in Bush Creek that spawned mainly from l a t e September to mid-November; a winter spawning stock, WB, also in Bush Creek that spawned from the end of November to the middle of December i n Bush Creek; and a winter spawning stock, W, in Walker Creek that spawned from the end of November to the end of December. WB spawners had fewer vertebrae than s p a t i a l l y separated W and temporally separated AB. Age of spawners was s i m i l a r in 1981 and 1982 with exception that the l a t e stocks combined had younger spawners than AB during 1982. Length at age was also s i m i l a r among the populations in 1981 and 1982 except 251 that in 1982 the l a t e stock, WB, had larger spawners than the temporally separated stock AB. Females of the WB stock had smaller eggs at. length than either of the other two stocks. Differences in the incubation environment between the autumn stock and the winter stocks were s u b s t a n t i a l . Since the streams did not d i f f e r greatly in temperature regime the autumn stock must experience r e l a t i v e l y warm temperatures i n early development while the winter stocks would experience much cooler temperatures. the temporal pattern of fry run timing suggested a synchronous emergence pattern among the stocks. Thus, the incubation period to emergence of the two winter stocks, W and WB, was much shorter than that of the autumn stock, AB. This pattern was consistent even though the years were quite d i f f e r e n t c l i m a t i c a l l y and could not be accounted for due to environmental differences among the streams. Calculations of the number of thermal units (TU's) used to reach emergence showed that AB required more TU's to emerge than the two winter stocks. The winter stocks had s i m i l a r TU requirements. The number of TU's required for each stock was consistent from year to year. The WB progeny had more vertebrae in 1982 and fewer vertebrae in 1983 than the s p a t i a l l y separated W and temporally separated AB stocks. Overlap in external morphology was great among the stocks. The WB stock was intermediate between the two others. 252 R e t u r n s o f marked f i s h showed t h a t s t r a y i n g among t h e t h r e e p o p u l a t i o n s was s u b s t a n t i a l . S t r a y i n g i n t o t h e W and AB s t o c k s was l e s s t h a n 1 2 - 1 3 %. However , more t h a n 70?o o f t h e s p a w n e r s i n WB c o u l d be f r o m t h e o t h e r two p o p u l a t i o n s . To summar i ze t h e r e s u l t s f o r p h e n o t y p i c and e n v i r o n m e n t a l - v a r i a t i o n among t h e s t o c k s i n t h e w i l d : 1) t h e i n c u b a t i o n e n v i r o n m e n t s o f t h e autumn and w i n t e r s p a w n i n g s t o c k s a r e v e r y d i f f e r e n t . S p a w n i n g t i m e does not appear t o be t o c o m p e n s a t e f o r d i f f e r e n c e s i n t e m p e r a t u r e o f i n c u b a t i o n e n v i r o n m e n t s a s has been n o t e d by B r a n n o n (1984) f o r s o c k e y e s a l m o n ; 2 ) WB a p p e a r e d t o be t h e most d i v e r g e n t p o p u l a t i o n p h e n o t y p i c a l l y . The p o p u l a t i o n s s e p a r a t e d b o t h t e m p o r a l l y and s p a t i a l l y , W and A B , had c o n v e r g e n t p h e n o t y p i c t r a i t s . In g e n e r a l a l l p o p u l a t i o n s were q u i t e s i m i l a r i n age and l e n g t h o f s p a w n e r s , t i m i n g o f emergence and e x t e r n a l m o r p h o l o g y o f p r o g e n y , ; 3) s t r a y i n g f r o m t h e o t h e r two p o p u l a t i o n s i n t o t h e WB s p a w n i n g a r e a i s v e r y g r e a t . Whether t h e s t r a y s s u c c e s s f u l l y mate w i t h t h e r e s i d e n t s i s a n o t h e r m a t t e r b u t t h e p r o b a b i l i t y t h a t a r e s i d e n t f i s h m a t e s w i t h an o u t s i d e r must be s u b s t a n t i a l l y g r e a t e r f o r WB f i s h compared t o W o r A B ; 4) i n c u b a t i o n r a t e s appear t o be a d a p t e d t o s e a s o n o f r e p r o d u c t i o n t o compensate f o r t h e d i f f e r e n c e s i n t e m p e r a t u r e e x p e r i e n c e d by t h e s t o c k s d u r i n g e m b r y o n i c d e v e l o p m e n t . I n v e s t i g a t i o n u s i n g l a b o r a t o r y r e a r i n g s o f p r o g e n y o f t h e t h r e e p o p u l a t i o n s showed t h a t f o r i n c u b a t i o n r a t e , v e r t e b r a l number and e x t e r n a l m o r p h o l o g y o f f r y a l l t h r e e s t o c k s a r e g e n e t i c a l l y d i f f e r e n t . However , e a c h o f t h e t h r e e c a t e g o r i e s o f t r a i t s i s a f f e c t e d a l s o by t h e t e m p e r a t u r e o f 253 incubation. Survival during embryonic development appeared to be unrelated to stock or temperature of incubation. Survival was much lower in the 1982-83 experiment. This appears to be due to technical problems in rearing and not necessarily related to the treatments. Therefore, I consider the results from the 1982-83 experiment to be less r e l i a b l e than those of the 1983-84 lab rearings. In both t r i a l s the o f f s p r i n g of autumn stock matings required more days to reach 50% hatch and 50% emergence than the progeny of winter stock matings when reared at 6 C, under a simulated "autumn spawned" regime and under a simulated "winter spawned" regime. At 10 C the autumn spawning stock required fewer days to hatch than the winter stocks. In o v e r a l l t e s t s of s i g n i f i c a n c e , populations that were both s p a t i a l l y and temporally i s o l a t e d or spawned during d i f f e r e n t seasons were the most divergent. AB and WB were the least s i m i l a r populations with respect to vertebral counts while WB and W were the most s i m i l a r . There was no d i s t i n c t ordering of the populations in mean vertebral count. WB was the most responsive to d i f f e r e n t incubation temperature while W was the least responsive. Discriminant analysis of pooled and separated samples by temperature treatment revealed a marked separation by external morphology between the AB progeny and the progeny of each of W and WB. The differences were much greater than those observed among the wild progeny. When the samples were pooled over a l l temperatures the discriminators in order of importance were: Eye diameter, snout length, parr marks, pectoral f i n length, head length, head depth and standard length. At a l l temperatures the AB progeny had larger 254 eyes. The AB progeny generally had longer snouts than the winter stocks. From the r e s u l t s i t i s clear that the genetic program for external morphology d i f f e r s among a l l three stocks. The difference appears to be greatest between the autumn spawning stock, AB, and each of the two winter spawning stocks, W and WB. The t r a i t s time to hatch and time to emergence proved to be heritable in a l l stocks. S i r e plus dam h e r i t a b i l i t i e s for time to hatch were 0.27, 0.35 and 0.47 for AB, WB, and W, r e s p e c t i v e l y . Sire plus dam h e r i t a b i l i t i e s for time to emergence were 0.50, 0.40 and 0.54 for AB, WB, and W, respectively. Maternal e f f e c t s were not important. In a l l cases the h e r i t a b i l i t i e s were s i g n i f i c a n t l y greater than zero. Electrophoretic analysis of 39 l o c i revealed that many l o c i in the populations were polymorphic and that heterozygosity was higher than expected i f the populations were i n Hardy-Wienberg equilibrium. F(st) values suggest that only a l i m i t e d amount of genetic d i f f e r e n t i a t i o n i n biochemical t r a i t s has occurred among the populations. The patterns of polymorphism and f i x a t i o n could not eliminate the p o s s i b i l i t i e s that there was no interbreeding among the populations or that migration occurs from the WB stock into the AB stock. Genetic distance measures indicated that AB and WB were more c l o s e l y related to each other than they were to the W population. A Wagner Tree was constructed that indicated that the W population had diverged f i r s t and then the AB and WB populations had s p l i t up. 255 The r e s u l t s suggest a s i t u a t i o n as follows: S t a b i l i z i n g s e l e c t i o n on the phenotypes of conspecific populations that inhabit the same locale i s operating so that there i s convergence in t r a i t s required to survive in the common environment. Certai n l y these populations do inhabit b a s i c a l l y the same environment after they emerge and migrate downstream. However, because the incubation environments of the seasonally separated populations are so d i f f e r e n t there e x i s t s a kind of d i s r u p t i v e s e l e c t i o n between them. Thus at the l e v e l of the genetics of development the seasonally separated populations have evolved in d i f f e r e n t d i r e c t i o n s to compensate g e n e t i c a l l y for the environmental differences experienced during development so that they may a t t a i n the r e l a t i v e l y uniform phenotypes needed to survive in the post-emergent environment. An explanation such as t h i s i s s u f f i c i e n t to account for the r e l a t i v e l y s i m i l a r phenotypes among the populations in the wild. However, how can one resolve the r e s u l t s of the tagging experiments that show a large amount of straying among the populations or the electrophoretic r e s u l t s that indicate that the populations i n the same stream, AB and WB, are more c l o s e l y related than those of the same season? The high rate of straying seems contradictory to the idea that the early and l a t e populations are s u f f i c i e n t l y i s o l a t e d to diverge. Spieth (1974) demonstrated that, in the absence of s e l e c t i o n , migration of less than one i n d i v i d u a l per generation was s u f f i c i e n t to homogenize the gene pools of two populations. In the past many studies suggesting that phenotypic differences 256 among populations were a r e f l e c t i o n of genetic differences among the populations were d i s c r e d i t e d using the l o g i c of Spieth (1974). However, recently i t has been suggested that rates of migration as high as 6 adults per generation were not s u f f i c i e n t to impede s e l e c t i o n in populations of greater than 1000 i n d i v i d u a l s (Wehrhahn and Powell 1987). Endler (1986) suggests that s e l e c t i o n and gene flow i n t e r a c t to set the l e v e l of genetic d i f f e r e n t i a t i o n . It i s important recognize that migration can be both an e c o l o g i c a l phenomenon involving the movement of i n d i v i d u a l s from one area to another and a genetic one involving the transfer of genetic material between populations. Individuals may stray into the breeding time and place of the other populations but i t does necessarily follow that they w i l l s u c c e s s f u l l y find a mate i n the resident population. E c o l o g i c a l migration does not influence population structure (by d e f i n i t i o n ) and natural s e l e c t i o n therefore acts on ec o l o g i c a l migration i n a simple manner: the movement y i e l d i n g the highest, reproductive success i s favored ( B u l l et a l . 1987). Genetic migration influences population structure and i t s evolution i s complex ( B u l l et a l . 1987). C e r t a i n l y there i s no doubt that e c o l o g i c a l migration occurs between the three populations but i t i s not c e r t a i n that there i s much gene exchange. The analysis of polymorphic enzyme patterns i n section 6 indicates that there i s l i t t l e genetic migration between the populations. The only p o s s i b i l i t y i s from WB to AB. This may represent the true l e v e l of genetic i s o l a t i o n i n the system. Of the populations, WB, with i t s small population s i z e and the high rate of straying from other populations i s the most l i k e l y to recieve enough migrants to disrupt i t s response to the surrounding s e l e c t i v e forces. It also shows the most f l u c t u a t i o n from the norm in phenotype. From t h i s I postulate 257 that genetic migration into WB from the other populations might prevent i t r'rom responding e f f e c t i v e l y to s t a b i l i z i n g s e l e c t i o n for phenotype. One must also note that I was unsuccessful in rearing cross mated progeny of WB and W in the laboratory. WB may not recieve many viable migrants at a l l i f there i s a f e r t i l i t y b a r r i e r among the populations. It i s important to remember that there are probably a ser i e s of b a r r i e r s to gene flow among the populations with temporal i s o l a t i o n being just one aspect. It i s the product of the reduction in the p r o b a b i l i t y of crossmating at each b a r r i e r that gives the true l e v e l of genetic i s o l a t i o n (Figure 29). The degree of congruence between the results from analyses of the v a r i a t i o n i n polymorphic enzymes and taxonomic c h a r a c t e r i s t i c s such as morphological, e c o l o g i c a l , and behavioural t r a i t s i s variable from study to study. For example, Lindenfelser (1984) was able to use the agreement between morphometric and allozymic data to define subspecies in the prawn, Macrobrachium rosenbergii. S i m i l a r l y , C h i l d (1980) was able to separate the morphologically d i s t i n c t seasonal races of Windermere charr by electrophoretic analysis of isozymes. In contrast, great divergence in morphological, me r i s t i c and e c o l o g i c a l features among populations of A r t i e charr in Ireland was not matched by the r e s u l t s from analysis of s t r u c t u r a l protein morphs (Ferguson 1980). According to Clayton (1981) many investigations have shown that there i s only a tenuous correspondence between the rates of organismal (polygenic t r a i t s ) and molecular evolution (evolution of s t r u c t u r a l proteins detectable by e l e c t r o p h o r e s i s ) . Clayton (1981) proposes that the dilemma that t h i s creates can be resolved by postulating that the rates of molecular and organismal evolution d i f f e r in nature. 258 Pj= percent isolation SEASONAL OR HABITAT ISOLATION n ETHOLOGICAL ISOLATION P 2 MECHANICAL ISOLATION p 3 GAMETIC MORTALITY p 4 ZYGOTIC MORTALITY p c 5 HYBRID INVIABILITY p 0 6 HYBRID STERILITY p? Probability of a successful migrant equals P, • P 2 P3- P4- P5- ^ Figure 29. B a r r i e r s to crossmating showing the cumulative e f f e c t on genetic migration (After Mayr 1970). 259 Clayton (1981) invokes the "molecular clock" hypothesis that molecular evolution proceeds at a steady rate. In contrast, modern interpretations of the f o s s i l record have emphasized the e r r a t i c pace of organismal evolution (Gould and Eldridge 1977, Stanley 1979). Clayton (1981) proposes that the d i s p a r i t i e s between divergences estimated from molecular and organismal data may be interpreted as r e f l e c t i o n s of the e r r a t i c pace of organismal evolution. If one accepts t h i s as a reasonable explanation for the differences between the incubation rate, meristic and morphological, r e s u l t s compared to those from electrophoretic analysis for these populations one may make some in t e r e s t i n g conclusions. F i r s t one must conclude that the divergence in quantitative t r a i t s among the seasonally separated populations must have taken place a f t e r the separation of the Walker Creek population from the Bush Creek populations. I conclude t h i s because i f the seasonal populations had separated f i r s t we would expect that WB isozyme frequencies would be more s i m i l a r to W than to AB. Given that there i s r e l a t i v e l y l i t t l e genetic d i f f e r e n t i a t i o n among the populations by electrophoretic data one must conclude that the divergence in quantitative t r a i t s i s r e l a t i v e l y recent. Thus, one must modify the o r i g i n a l hypothesis for t h i s system. O r i g i n a l l y there may have been only one population, probably spawning only i n Bush Creek as i t appears to be the larger, more stable system. At some point there may have developed a group that returned somewhat l a t e r i n the year than the main group in the population. These f i s h may have found most of the spawning s i t e s occupied or 260 conditions in Bush Creek less than ideal for spawning and thus been attracted to nearby Walker Creek. The s l i g h t l y warmer water in Walker Creek may have allowed t h e i r progeny to catch up just enough to make them v i a b l e . This could allow the late f i s h to colonize further into the winter months and thus increase the temporal separation among the stocks. The W stock continued to adapt to the new l a t e season niche. Some time l a t e r the Bush stock spawned a new winter spawning stock in Bush Creek i t s e l f . The two winter stocks have adapted to t h e i r season of reproduction but since they had d i f f e r e n t s t a r t i n g points they have solved the problem in d i f f e r e n t ways. This would elegantly explain the apparent i n f e r t i l i t y b a r r i e r among the two stocks as well as the difference in egg s i z e among them. An a l t e r n a t i v e scenario i s possible taking into account conditions at the end of the Wisconsin G l a c i a t i o n . One might suppose that populations with rapid incubation rates would be selected for in the cold c l i m a t i c period 8,000 to 10,000 years ago. Also, Ladysmith Harbor may have been much shorter than i t i s now and Bush and Walker creeks may have been one system. Thus, the two l a t e populations could have evolved as stocks spawning in d i f f e r e n t t r i b u t a r i e s of the same r i v e r . As the climate warmed and Ladysmith Harbor lengthened these stocks became separated by geography and were forced to spawn l a t e r and l a t e r in the year. F i n a l l y an e a r l i e r spawning population evolved in Bush Creek. One other i n t e r e s t i n g r e s u l t that seems d i f f i c u l t to resolve i s that t r a i t s thought to be related to f i t n e s s , time to hatch and time to emergence 261 have r e l a t i v e l y high, non-zero h e r i t a b i l i t i e s . Fisher's fundamental theorem of natural s e l e c t i o n postulates that the mean rate of evolutionary change i n mean f i t n e s s equals the additive genetic variance in f i t n e s s i t s e l f (Fisher 1930). Natural s e l e c t i o n w i l l drive the population to equilibrium where fi t n e s s differences w i l l be equal to zero and consequently the genetic variance in f i t n e s s w i l l be soon depleted (Robertson 1955). According to Falconer (1981) f i t n e s s related t r a i t s should have very low or zero h e r i t a b i l i t i e s . Further, when t r a i t s are under s t a b i l i z i n g s e l e c t i o n , as I have proposed here, additive genetic variance should diminish (Falconer 1981). There are several possible ways out of t h i s quandry. The explanation may be quite simply that since I measured h e r i t a b i l i t y in the laboratory I simply affected the norm of reaction for these t r a i t s (Hartl 1981). Thus by switching the animals from the natural environment to an a r t i f i c i a l one I might s u b s t a n t i a l l y a l t e r the h e r i t a b i l i t y of the t r a i t . In other words my lab r e s u l t s do not apply to the f i e l d . Another p o s s i b i l i t y i s that the system does not come to equilibrium. There i s continual noise i n the environment that does not allow s e l e c t i v e processes to work nearly as e f f e c i e n t l y as Fisher (1930) envisioned. Many deaths are due to stochastic processes that cannot be e f f e c t i v e l y tracked by s e l e c t i o n ever though the t r a i t s are important to the f i t n e s s of the organism. A t h i r d p o s s i b i l i t y i s that there e x i s t s antagonistic plieotropy such that the more optimal genes for incubation rate also code for e f f e c t s that are negatively correlated with f i t n e s s . Thus equilibrium cannot be achieved. Individuals with optimal incubation rate genes die because these genes produce l e t h a l phenotypes for other t r a i t s . A 262 fourth p o s s i b i l i t y i s i f the populations have recently occupied d i f f e r e n t seasonal breeding niches then natural s e l e c t i o n has not had time to bring the gene pools to equilibrium. F i n a l l y , Lande (1975) and Arnold (1986) have argued on t h e o r e t i c a l grounds that additive genetic variance could be maintained in wild populations even given that there i s strong s e l e c t i o n occurring. Lande (1975) argued that in quantitative t r a i t s there must be thousands of genes segregating and that by virtue of mutation alone genetic variance would be replaced at the same rate i t was l o s t . Arnold (1986) determined mathematically that there are l i m i t s to s t a b i l i z i n g s e l e c t i o n . For example he found that in sexual s e l e c t i o n i n male salamanders s t a b i l i z i n g s e l e c t i o n could reduce the variance of a normally d i s t r i b u t e d t r a i t by no more than about 35%. Perhaps salmonids by v i r t u e of t h e i r t e t r a p l o i d ancestry can maintain higher l e v e l s of v a r i a t i o n in f i t n e s s related t r a i t s . Both Gjedrem (1983) and G a l l et a l . (1988) have noted high h e r i t a b i l i t y l e v e l s in f i t n e s s related t r a i t s in salmonid populations. The two main t r a i t s of i n t e r e s t , seasonal timing of spawning and the seasonal timing of fry migration may be considered as a function of heterchrony and the quantitative genetics of development. According to Wright (1969, 1988) evolution consists of movements on a genotypic adaptive landscape. Within Wright's (1931) hypothesis the genomes of the seasonally separated stocks occupy d i f f e r e n t adaptive peaks of coadapted gene complexes (parts of the adaptive complex are the genes for season of spawning and the 263 genes for the speed of embryonic development). Wright (1988) points out that the s e l e c t i v e peaks do not depend on the s t r i c t a d d i t i v i t y of the e f f e c t s of component genes. There may be gene interactions ( e p i s t a s i s ) such as heterochrony. Heterochronic e f f e c t s can profoundly a l t e r the phenotype without a large change in the genome. Atchley (1987) has examined how a change in the ontogenetic t r a j e c t o r y of a t r a i t can re s u l t in s t r i k i n g divergence in phenotype. S l a t k i n (1987) developed a model for the quantitative genetics of heterchrony to show how genetic v a r i a t i o n in developmental parameters governs v a r i a t i o n and covar i a t i o n in phenotypic t r a i t s and how s e l e c t i o n on the phenotype a l t e r s the d i s t r i b u t i o n s of developmental parameters. According to Semlitsch et a l . (1988) timing of transformations ( i . e . to reproductive adult or to emergent fry) during the l i f e h istory can a f f e c t the f i t n e s s of the indi v i d u a l during the en t i r e l i f e h i s t o r y . It i s d i f f i c u l t to investigate genetic divergence among ecotypes, races or d i f f e r e n t forms of a species without considering how the re s u l t s might impinge on the theories of sp e c i a t i o n . As stated i n the introduction, the a l l o p a t r i c model of speciation i s the most widely accepted mode of speciation (Futuyma 1979). However, given: 1) the existence of " b i o l o g i c a l races", "host races", " e c o l o g i c a l races" or "a l l o c h r o n i c races" of a species within the same t e r r i t o r y (White 1978) and 2) the r e s u l t s of laboratory experiments on disruptive s e l e c t i o n by various researchers (Thoday and Gibson 1962, Thoday and Boam 1959, Thoday 1972, Pimentel et a l . 1967) and evidence for s e l e c t i o n 264 against hybrids in the wild (Dowling and Moore 1985) sympatric speciation must be considered as a d e f i n i t e p o s s i b i l i t y (Dickson and Antonovics 1973, Bush 1974). Races or ecotypes are a common feature of salmonid species (Saviattova 1983). Some examples are kokanee and sockeye salmon in the species, Oncorhychus nerka, (Foote 1987, Foerster 1968), F-charr, normal and S-charr in A r t i e charr, Salvelinus alpinus (Nyman 1980), and the numerous examples of seasonal races (Frost 1966, Berg 1934, 1959, Ricker 1972). According to Maynard-Smith (1966) the c r u c i a l step in sympatric speciation i s the establishment of a stable polymorphism i n a heterogenous environment. Given a polymorphism, such as observed with seasonal races, Maynard-Smith (1966) proposed that reproductive i s o l a t i o n need only evolve for sympatric speciation to be po s s i b l e . In seasonally a l l o c h r o n i c populations one has both c r i t e r i a r o l l e d into one. The seasonal differences i n physiological readiness to spawn represents a s t r i k i n g polymorphism and also serves to i s o l a t e the populations. Although the step of e s t a b l i s h i n g a stable polymorphism by dis r u p t i v e s e l e c t i o n seems formidable, one can develop a mechanism for t h i s with a simple genetic model. Suppose that s e l e c t i o n of spawning season i s due to one s p e c i f i c locus T. Also suppose that reproductive success (the s u r v i v a l of progeny) during a p a r t i c u l a r season depends upon one locus E. In the present example E could control time to emergence. Individuals that emerged too early 265 or late in the spring could be selected against. "Tl—T1" i n d i v i d u a l s spawn in the autumn. "T1-T2" i n d i v i d u a l s spawn either during autumn or during the winer. "T2-T2" i n d i v i d u a l s spawn during the winter. "E1-E1" i n d i v i d u a l s develop slowly - time to emergence i s longer. "E1-E2" i n d i v i d u a l s also develop slowly. "E2-E2" i n d i v i d u a l s develop r a p i d l y - time to emergence i s shorter. This arrangement w i l l cause a steady increase i n T1T1E1E1 ind i v i d u a l s spawning in the autumn and T2T2E2E2 i n d i v i d u a l s spawning in the winter. Is t h i s p l a u s i b l e in terms of empirical information? Personally, I believe the system must be more complex than above and involve polygenic t r a i t s . However, Tauber et a l . (1977) showed that only two genes c o n t r o l l e d seasonal i s o l a t i o n among two species of Chrysopa. Allendorf et a l . (1983) found that the PGM-1 locus c o n t r o l l e d the rate of embryonic development and time to hatching in rainbow trout, Salmo g a i r d n e r i . Thus, the l i t e r a t u r e contains some undisputed examples of simple genetic control over the key t r a i t s needed for a model of speciation by allochrony. As well, my r e s u l t s show that among seasonally separated groups the rate of development i s adapted to the season of spawning and that in the wild the timing of emergence and downstream migration i s synchronous among the progeny of the seasonally separated populations. That s t a b i l i z i n g s e l e c t i o n operates on fry timing so that a narrow window of time e x i s t s for the fry to migrate s u c c e s s f u l l y has been demonstrated by Fresh et a l . (1985), Taylor (1980), B i l t o n (1980), Walters et a l . (1978) and Brannon (1984). 266 Aside from the implications for evolutionary theory these r e s u l t s impinge on the management of aquaculture and f i s h e r i e s . In a recent review Gerking (1988) concluded that two f i e l d s badly need attention i f we are to move ahead in fishery biology: genetics and behaviour. The reduction of genetic variance in a f i s h species can r e s u l t in a serious loss of adaptability to environmental change (Meffe 1986, 1987, Kapuscinski and Lannan 1986). These res u l t s suggest a l i n k between a behaviour: season of spawning, and genetics: the rate of incubation and r e s u l t i n g morphology of the f r y . Spawning time i s c l o s e l y correlated with annual migration timing which in turn i s connected to the a v a i l a b i l i t y to the f i s h e r y . Heavy e x p l o i t a t i o n during a p a r t i c u l a r time of the year may extinguish a unique stock even when the annual e x p l o i t a t i o n rate on the species i s at low l e v e l s . Another concern i s the introgression of hatchery stock into wild populations. For example, Leider et a l . (1986) found that the success of hatchery steelhead trout i n producing smolt o f f s p r i n g was only 28% of that of wild f i s h . Yet due to the reduction i n mortality by rearing a r t i f i c a l l y 62% of the na t u r a l l y produced smolts were o f f s p r i n g of hatchery spawners. Chilcote et a l . (1986) concluded that hatchery operations could threaten the genetic i n t e g r i t y of wild populations. The result that populations are adapted by incubation rate to t h e i r season of reproduction means that hatchery operators should take care to use broodstock with the appropriate genetic make-up to both the season of reproduction and the l o c a l e where the progeny w i l l be released. On the other hand the high h e r i t a b i l i t y of the time to hatch and emergence suggests that broodstocks could be selected for emergence at a p a r t i c u l a r time as needed by the hatchery operator. 267 Returning to the questions posed in the introduction of the thesi s . . . There are not great phenotypic differences among these seasonally separated populations. However, since the difference in incubation environment would lead one to expect differences the lack of difference suggests that phenotypes are constrained to an optimum. Therefore the differences can be considered adaptive. The s i m i l a r i t i e s in phenotype are achieved by differences in the genomes among the populations. The populations are c l e a r l y g e n e t i c a l l y d i f f e r e n t . The casual factors for t h i s difference are not e n t i r e l y clear although I suggest that s t a b i l i z i n g s e l e c t i o n for phenotype of the emerging fry among the populations and d i s r u p t i v e s e l e c t i o n on the genotype for incubation rate and f r y morphology are responsible. The general wisdom among salmon b i o l o g i s t s i s that adult migratory timing has evolved to maximize successful reproduction (Mundy 1982, M e r r i t t and Roberson 1986). In p a r t i c u l a r , time of migration and spawning was selected to ensure a time of fry migration that w i l l maximize the p r o b a b i l i t y of s u r v i v a l (Merritt and Roberson 1986). M i l l e r and Brannon (1982) suggested that temperature regime of the home stream i s most responsible for the e c o l o g i c a l i s o l a t i o n of stocks. Brannon (1984) observed that sockeye salmon (Oncorhynchus nerka) spawning occurred l a t e r in the year in warmer streams of the Fraser River drainage. M i l l e r and Brannon (1982) proposed that spawning time has evolved to o f f s e t temperature differences among streams. Sheridan (1962) found that spawning of pink salmon (Oncorhynchus gorbuscha) in c o l d , inland streams of southeastern Alaska occurred e a r l i e r in the year than spawning in the warmer coastal streams. Sheridan (1962) demonstrated that 268 despite a 46 day difference in spawning times between pink salmon using Kadashan Creek and those using Klawock Creek, the ova received about the same number of temperature units by spring, so that fry of both streams emerged at about the same time. However, spawning time i n chum salmon has not evolved to off s e t the temperature of the spawning stream. Early spawners lay t h e i r eggs at higher temperatures than l a t e r spawners and must accumulate more temperature units to complete embryonic development. The l a t e spawners have faster r e l a t i v e and absolute incubation rates than the early spawners. According to Lynch (1986) there i s much inte r e s t in the factors that promote uniformity among conspecific populations. Population d i f f e r e n t i a t i o n may be prevented by interpopulational gene flow or by operation of s i m i l a r s e l e c t i o n pressures among i s o l a t e s (Lynch 1986). E h r l i c h and Raven (1969) argued that uniform s e l e c t i o n , not gene flow, i s the primary cohesive force i n evolution. More recently i t has been appreciated that gene flow and s e l e c t i o n may interact to set the degree of divergence (Endler 1977). In both these approaches s e l e c t i o n for phenotypic uniformity among populations i s considered to r e f l e c t genetic uniformity. However, Cohan (1984) argued that weak uniform s e l e c t i o n operating on f i n i t e populations would cause more genetic divergence than expected by genetic d r i f t among i s o l a t e s . Cohan (1984) suggested that i n t r a i t s with a high component of genetic a d d i t i v i t y , populations could a r r i v e at the same conclusion of uniformity i n phenotype through d i f f e r e n t genetic pathways. Favored a l l e l e s could become fixed at al t e r n a t i v e l o c i . This 269 mechanism might account for the incompatability of cross matings of l a t e stocks even though they appear to possess a s i m i l a r adaptation to season of spawning. The temporal i s o l a t i o n of stocks does not necessarily preclude any gene flow between them. The progeny of f i s h spawning in a p a r t i c u l a r season and location generally show f i d e l i t y to t h e i r natal l o c a t i o n and season of spawning but a s i g n i f i c a n t number of strays may return to spawn during other times of the year. In fact, gene flow i s p o t e n t i a l l y high enough to prevent genetic d i f f e r e n t i a t i o n in the absence of s e l e c t i o n (Spieth 1974). Differences in season of reproduction do not appear to be s u f f i c i e n t to i s o l a t e populations so that genetic divergence takes place as a result of genetic d r i f t . However, the e f f e c t of environmental differences on development due to the seasonal progression of climate coupled with s e l e c t i o n for phenotypic uniformity in a given locale may r e s u l t in genetic divergence among populations. The key event determining the pattern of v a r i a t i o n in the re s u l t s i s the timing of fry emergence. The synchronous appearance of fry on the estuary meant that common s e l e c t i v e pressures could exert t h e i r influence. Interpopulation synchrony of migration may not only be important i n promoting uniformity of the f r y phenotype among populations but also may influence adult t r a i t s . For example, Helle (1979) found that age at maturity could be related to the conditions the f i s h experienced as fry on the nursery grounds. The wild phenotypes are r e l a t i v e l y s i m i l a r in spite of t h e i r developing under quite d i f f e r e n t environmental conditions. In contrast, phenotypes were quite d i f f e r e n t when the embryos of the populations were reared under the same regime. 1 270 The phenotypic and genetic structure of chum salmon i s influenced by both the location and season of reproduction. For example, Kubo ( 1 9 5 6 ) noted that the mean vertebral number d i f f e r e d among s p a t i a l l y separated chum stocks. This phenotypic v a r i a t i o n was demonstrated to have a genetic basis. Stocks separated by both l o c a t i o n and season of spawning, such as the summer and autumn chum salmon of the Amur River, d i f f e r greatly in many aspects of the phenotype (Bakkala 1970). However, the genetic basis of the differences has not been confirmed. F i n a l l y , the r e s u l t s of t h i s study suggest that when the season of spawning d i f f e r s but the l o c a l e of spawning i s s i m i l a r , stocks may not diverge i n phenotype but s t i l l w i l l be divergent in genotype. If these r e s u l t s are applicable to a l l chum salmon populations then f i s h e r i e s management must consider that stocks with d i f f e r e n t seasons of reproduction may represent d i f f e r e n t production units. Ideally, these should be managed separately (Larkin 1981). This has been done for some time now in chum salmon. The pr a c t i s e did not have a confirmation of genetic divergence among the stocks u n t i l t h i s work. 271 LITERATURE CITED Abakumov, V.A. 1961. Seasonal races of diadrornous f i s h e s . Vopr. I k h t i o l . No. 17. Alderdice, D.F., W.P. Wickett and J.R. Brett. 1958. Some e f f e c t s of temporary exposure to low dissolved oxygen l e v e l s on P a c i f i c salmon eggs. J. F i s h . Res. Bd. Can. 15:229-250. Alexander, R.D. and R.S. Bigelow. 1960. Allochronic speciation in f i e l d c r i c k e t s and a new species, Acheta v e l e t i s . Evolution 14: 334-346 Allendorf, K.W., K.L. Knudsen, R.F. Leary. 1983. Adaptive s i g n i f i c a n c e of differences in the expression of a phosphoglucomutase gene in rainbow trout. Proc. N a t l . Acad. S c i . U.S.A. 80: 1397-1400. Allendorf, F.W. and F.M. Utter. 1979. Population genetics p. 407-454. In W.S. Hoar, D.J. Randall and J.R. Brett [ed.] Fish Physiology Academic Press, New York, N.Y. Allendorf, F.W., N. M i t c h e l l , N. Ryman, and G. Stahl. 1977. Isozyme l o c i i n brown trout (Salmo t r u t t a L.): detection and int e r p r e t a t i o n from population data Hereditas 86: 179-190. Altukov, Yu.P. 1981. The stock concept from the viewpoint of population genetics. Can. J. F i s h . Aquat. S c i . 38: 1523-1538. Altukov, Yu.P. and E.A. Salmenkova. 1981. Applications of the stock concept to f i s h populations i n the U.S.S.R.. Can. J. F i s h . Aquat. S c i . 38: 1591-1600 Andersson, L., N. Ryman, R. Rosenberg and G. Stah l . 1981. Biochemical genetic v a r i a t i o n i n A t l a n t i c herring (Clupea harengus L.): description of isozyme l o c i and population data. Hereditas 95: 69-78. Arnold, S.J. 1986. Limits on s t a b i l i z i n g , d i s r uptive and c o r r e l a t i o n a l s e l e c t i o n set by the opportunity for s e l e c t i o n Am. Nat. 128: 143-146. Aro, K.V. and M.P. Shepard 1967. Salmon of the North P a c i f i c Ocean - Part IV. Spawning populations of North P a c i f i c salmon. 5. P a c i f i c salmon i n Canada. Int. N. Pac. F i s h . Comm., B u l l . 23: 225-327 Atchley, W.R. 1987. Developmental quantitative genetics and the evolution of ontogenies. Evolution 41: 316-330. Atkinson, C.E., J.H. Rose, and T.0. Duncan. 1967. Salmon of the North P a c i f i c Ocean - Part IV. Spawning populations of North P a c i f i c salmon. 4. P a c i f i c salmon in the United States. INt. N. Pac. F i s h . Comm., B u l l . 23: 43-223. 272 Bakkala, R.G. 1970. Synopsis of B i o l o g i c a l Data on the Chum Salmon, Oncorhynchus keta (Walbaum) 1792. United States Department of the In t e r i o r , U.S. Fish and W i l d l i f e Service, Bureau of Commercial F i s h e r i e s , C i r c u l a r 315. Bailey, J.E., S.D. Rice, J . J . P e l l a and S.G. Taylor. 1980. E f f e c t s of seeding density of pink salmon, Oncorhynchus gorbuscha, eggs on water chemistry, f r y c h a r a c t e r i s t i c s , and fry s u r v i v a l in gravel incubators. Fishery B u l l e t i n 78(3): 649-657. Bailey, J.E. and W.R. Heard. 1973. An improved incubator for salmonids and resu l t s of preliminary tests of i t s use. NOAA Tech. Memo, NMFS ABFL-17p. Bailey, J.E., B.L. Wing and CR. Mattson. 1975. Zooplankton abundance and feeding habits of f r y of pink salmon, Oncorhynchus gorbuscha, and chum salmon, Oncorhynchus keta, i n Tr a i t o r s Cove, Alaska, with speculations on the carrying capacity of the area. F i s h . B u l l . 73(4): 846-861. Balon, E.K. 1979. The j u v e n i l i z a t i o n process in phylogeny and the a l t r i c i a l to p r e c o c i a l forms i n the ontogeny of fi s h e s . Env. B i o l . Fishes 4: 193-198. Bams, R.A. 1969. Adaptations i n sockeye salmon associated with incubation i n stream gravels. Univ. B r i t i s h Columbia, H.R. Macmillan Lectures i n Fis h e r i e s , Symposium on salmon and trout in streams. T.G. Northcote ed. pp. 71-88. Bams, R.A. 1976. Survival and propensity for homing as affected by the presence or absence of l o c a l l y adapted paternal genes in two transplanted populations of pink salmon (Oncorhynchus gorbuscha). J. F i s h . Res. Bd. Can. 33: 2716-2725. Bams, R.A. 1982. Experimental incubation of chum salmon (Oncorhynchus keta) in Japanese-style hatchery system. Can. Tech. Rep. Fish, and Aquatic S c i . No. 1101: 65 p. Bams, R.A. and K.S. Simpson 1977. Substrate incubators workshop-1976. Report on current s t a t e - o f - t h e - a r t . Env. Canada, F i s h . Mar. Serv. Tech. Rep. 689 68p. Barlow, G.W. 1961. Causes and s i g n i f i c a n c e of morphological v a r i a t i o n in fis h e s . Syst. Zool. 10: 105-117. Bax, N.J. 1982. Seasonal and annual vari a t i o n s i n the movement of juvenile chum salmon through Hood Canal, Washington, p. 208-218. In Proc. Salmon and Trout Migratory Behaviour Symposium. E.L. Brannon and E.0. Salo (eds). Sch. of Fish . , Univ. Washington, Seattle, WA. 273 Beacham, T.D. 1982. Fecundity of coho salmon (Oncorhynchus kisutch) and chum salmon (CL keta) in the northeast P a c i f i c Ocean. Can. J. Zool.60: 1463-1469. Beacham, T.D. 1984. Age and morphology of chum salmon i n southern B r i t i s h Columbia. Trans. Am. F i s h . Soc. 113 (6): 727-736. Beacham, T.D. 1987. Genotype - environment in t e r a c t i o n s i n growth of chum salmon. Can. Tech. Rep. F i s h . Aquat. S c i . (1517). Beacham, T.D., A.P. Gould, R.E. Withler, C.B. Murray, and L.W. Barner. 1987. Biochemical genetic survey and stock i d e n t i f i c a t i o n of chum salmon (Oncorhynchus keta) i n B r i t i s h Columbia. Can. J. Fi s h . Aquat. S c i . 44: 1702-1713. Beacham, T.D. and C.B. Murray 1985. Ef f e c t of female s i z e , egg size, and water temperature on the developmental biology of chum salmon (Oncorhynchus keta) from the Nitinat River, B r i t i s h Columbia. Can. J . Fi s h . Aquat. S c i . 42: 1755-1765. Beacham, T.D. and C.B. Murray 1986. Comparative developmental biology of chum salmon (Oncorhynchus keta) from the Eraser River, B r i t i s h Columbia. Can. Journal F i s h . Aquat. S c i . 43(2): 252-262. Beacham, T.D. and P. S t a r r . 1982. Population biology of chum salmon, Oncorhynchus keta, from the Fraser River, B r i t i s h Columbia. United States National Marine Service Fishery B u l l e t i n 80: 813-825. Beacham, T.D., and C.B. Murray, 1986b. The e f f e c t of spawning time and incubation temperature on me r i s t i c v a r i a t i o n in chum salmon (Oncorhynchus  keta). Can. J. Zool. 64: 45-48. Beacham, T.D., and C.B Murray. 1987a. Adaptive v a r i a t i o n i n body s i z e , morphology, egg s i z e , and developmental biology of chum salmon (Oncorhynchus keta) i n B r i t i s h Columbia. Can. J. F i s h . Aquat. S c i . 44: 244-261. Beacham, T.D., and C.B. Murray. 1987b. E f f e c t s of t r a n s f e r r i n g pink (Oncorhynchus gorbuscha) and chum (Oncorhynchus keta) salmon embryos at d i f f e r e n t developmental stages to a low incubation temperature. Can. J. Zool. 65: 96-105. Beacham, T.D. and P. S t a r r . 1982. Population biology of chum salmon, Oncorhynchus keta, from the Fraser River, B r i t i s h Columbia. United States National Marine Service Fishery B u l l e t i n 80: 813-825. Beacham, T.D. and R.E. Withler. 1987. Developmental s t a b i l i t y and heterozygosity in chum (Oncorhynchus keta) and Pink (Oncorhynchus  gorbuscha) salmon. Can. J. Zool. 65: 1823-1826. 274 Beacham, T.D., R.E. Withler, and A.P. Gould. 1985. Biochemical genetic stock i d e n t i f i c a t i o n of chum salmon (Oncorhynchus keta) i n southern B r i t i s h Columbia. Can. J. F i s h . Aquat. S c i . 42: 437-448. Becker, W. 1975. A manual of quantitative genetics. 3rd. Ed. Wash St. Univ., Pullman, Wash. Bengtson, D.A., R.C. Barkman and W.J. Berry. 1987. Relationship between maternal s i z e , egg diameter, time of spawning season, temperature, and length at hatch of a t l a n t i c s i l v e r s i d e , Menidia medidia. J. Fish. B i o l . 31: 697-704. Berg, L.S. 1934. Yarovye i ozimye rasy u prokhodnykh ryb. I z v e s t i i a Akademii Nauk SSSR, Otdel Matematicheskish i Estestvennykh Nauk 5: 711-732. Berg, L.S. 1959. Vernal and heimal races among anadromous f i s h e s . Translation in J. F i s h . Res. Bd. Can. 16: 515-537. Bie t t e , R.M., D.P. Dodge, R.L. Hassinger and T.M. Stauffer. 1981. L i f e history and timing of migrations and spawning behaviour of rainbow trout (Salmo gairdneri) populations of the Great Lakes. Can. J. F i s h . Aquat. S c i . 38: 1759-1771. B i l t o n , H.T. 1980. Returns of adult coho salmon in r e l a t i o n to mean s i z e and time of release of juveniles to the catch and the escapement. Can. Rep. Fis h . Aquat. S c i . 941. Birman, I.B. 1956. Local stocks of autumn chum salmon in the Amur Basin. Fis h . Res. Bd. Can. Transl. Ser. 349. Birman, I.B. 1959. More about the influence of the Kurio-shio on the dynamics of abundance of salmon. Fi s h . Res. Bd. Can. Transl. Ser. 269. Birman, I.B. 1977. Intrapopulation groupings of the Amur Autumn chum salmon, Oncorhynchus keta. J . Ichthyol. 17: 741-750. Birman, I.B. 1981. Ocean migration and o r i g i n of seasonal races of anadromous salmon (Salmonidae). J. Ichthyol. 21: 35-47. B l a i r , W.F. 1941. V a r i a t i o n , i s o l a t i n g mechanisms, and hy b r i d i z a t i o n i n c e r t a i n toads. Genetics 26: 398-417. Blaxter, J.H.S. 1958. Racial problem in herring from the viewpoint of recent p h y s i o l o g i c a l , evolutionary and genetical theory. Rapp. Cons. Explor. Mer. 143(2): 10-19. Brannon, E.L. 1982. Orientation mechanisms of homing salmonids. In E.L. Brannon and E.O. Salo (eds). Salmon and trout migratory behaviour symposium, Univ. Wash., Seattle. 219-227. 275 Brannon, E.L. 1984. Influence of stock o r i g i n on salmon migratory behaviour. IN Mechanisms of migration in fis h e s . McCleave, J.D., G.P. Arnold, J . J . Dodson and W.H. N e i l l . (eds). NATO Conference Series, Vol. 14: 103-112. Buklis, L.S. 1982. Anvik, Andreafsky and Tanana River salmon escapement studies. 1981. Yukon Salmon Escapement Report No. 15. Alaska Dept. Fish Game, Commercial F i s h e r i e s D i v i s i o n , Anchorage, Alaska. 40p. Buklis, L.S. and L.H. Barton. 1984. Yukon River chum salmon biology and stock status. Alaska Dept. Fi s h Game Info Leaflet No. 239. 67p. B u l l , J.J. 1987. Evolution of phenotypic variance Evolution 41(2): 303-315. B u l l , J . J . , C. Thompson, D. Ng and R. Moore. 1987. A model for natural s e l e c t i o n of genetic migration. Am. Nat. 129: 143-157. Bush, G.L. 1974. The mechanism of sympatric host race formation in the true f r u i t f i l e s (Tephritidae). In Genetic mechanisms of speciation in insects. M.J.D. White (ed), Sydney, A u s t r a l i a and New Zealand Book Co. pp. 3-23. Bush, G.L. 1975. Modes of animal s p e c i a t i o n . Ann. Rev. E c o l . Systemat. 6: 339-364. Burger, C.V., R.L. Wilmot, and D.B. Wangaard. 1985. Comparison of spawning areas and times for two runs of chinook salmon (Oncorhynchus tshawytscha) in the Kenai River, Alaska. Can. J . F i s h . Aquat. S c i . 42: 693-700. Burner, C.J. 1951. C h a r a c t e r i s t i c s of spawning nests of Columbia River salmon. U.S. Fish and W i l d l i f e Service Fishery B u l l . 52: 97-110. Calaprice, J. 1969. Production and genetic factors in managed salmonid populations In Symp. on salmon and trout in streams T.G. Northcote (ed) 377-385. Carson, H.L. and A.R. Templeton. 1984. Genetic revolutions in r e l a t i o n to speciation phenomena: The founding of new populations. Ann. Rev. E c o l . Syst. 15: 97-131. C a v a l l i - S f o r z a , L.L. 1969. "Genetic d r i f t " in an I t a l i a n population. S c i . Am. 221: 30-37. Ca v a l l i - S f o r z a , L.L. and A.W.F. Edwards. 1967. Phylogenetic analysis models and estimation procedures. Evolution 21: 550-570. Chatwin, B.M. 1953. Tagging of chum salmon i n Johnstone S t r a i t , 1945 and 1950. F i s h . Res. Bd. Can. B u l l . 96, 33 p. Chebanov, N.A. 1986. Factors c o n t r o l i n g spawning success in pink salmon T. Ichthyol. 26(3): 69-78. 276 C h i a r e l l o , N. and J. Roughgarden. 1984. Storage a l l o c a t i o n in seasonal races of an annual plant: optimal versus actual a l l o c a t i o n . Ecology 65(4): 1290-1301. Chilcote, M.W., S.A. Leider and J . J . Loch. 1986. D i f f e r e n t i a l reproductive success of hatchery and wild summer-run steelhead under natural conditions. Trans. Am. F i s h . Soc. 115: 726-735. Ch i l d , A.R. 1980. Electrophoretic analysis of A r t i e char of cambrian lakes. In proceedings of the f i r s t ISACF workshop on a r t i c charr. Institute Freshwater Research, Drottningholm (Sweden) (Ed): 5-8. Chilton, D.E. and R.J. Beamish. 1982. Age determination methods for fishes studied by the Groundfish Program at the P a c i f i c B i o l o g i c a l S t a t i o n . Can. Spec. Publ. F i s h . Aquat. S c i . 60. 102 p. Chugunova, N.I. 1959. Age and growth studies in f i s h . T r a n s l . TT61-31036, Clearinghouse Fed. S c i . Tech. Inform., S p r i n g f i e l d , Va., U.S.A. 22151. 164 pages. Clayton, J. 1981. The stock concept and the uncoupling of organismal and molecular evolution. Can. J. F i s h . Aquat. S c i . 38: 1515-1522. Clemens, W.A. and G.V. Wilby. 1946. Fishes of the P a c i f i c coast of Canada. F i s h . Res. Bd. Can., B u l l . 68, 368 p. Cohan, F.M. 1984. Can uniform s e l e c t i o n retard genetic divergence between isol a t e d c o n s p e c i f i c populations? Evolution 38: 495-504. Cole, L.C. 1954. The population consequences of l i f e history phenomena. Quarterly Rev. B i o l . 29: 103-137. Congleton, J.L. 1978. Feeding patterns of juvenile chum in the Skagit River s a l t marsh, p. 141-150. In Proc. Second Northwest Tech. Workshop of Fish and Food Habit Studies, S.J. Lipovsky and C A . Simenstad (eds). Wash. Sea. Grant. Univ. Washington, S e a t t l e , WA. Danzmaan, R.G., M.M. Ferguson, and F.W. Allendorf. 1986. Does enzyme heterozygosity influence developmental rate in rainbown trout? Heredity, 56(3): 417-425. Dickinson, H. and J. Antonovics. 1973. Theoretical considerations of sympatric divergence. Am. Nat. 107: 256-274. D i l l , L.M. and T.G. Northcote. 1970. E f f e c t s of gravel s i z e , egg depth and egg density on in t r a - g r a v e l movement and emergence in coho salmon (Oncorhynchus kisutch) a l e v i n s . J. F i s h . Res. Board Can. 27: 1191-1199. Dowling, T.E. and W.S. Moore. 1985. Evidence for s e l e c t i o n against hybrids in the family Cyprinidae (genus Notropis). Evolution 39: 152-158. 277 Dunnett, C.W. 1980. Pairwise multiple comparisons in homogeneous variance, unequal sample siz e case. J. Amer. Stat. Assoc. 75: 789-795. Ehrman, L. 1970. The mating advantage of rare males in Drosophila. Proc. Natl. Acad. S c i . U.S.A. 515: 345-348. Elson, M.S. 1975. Enumeration of spawning chum salmon (Oncorhynchus keta) in the Fishing Branch River in 1971, 1972, 1973 and 1974. IN: Northern Yukon F i s h e r i e s Studies, 1971-1974, Vol. 1, Chapter I I . Steigenberger, L.W., M.S. Elson and R.T. Delury. eds. Northern Oper. Branch, Fish Mar. Serv. Dept. Environ. Can., PAC/T-75-19 45p. Endler, J.A. 1977. Geographic Variation, Speciation and Clines. Princeton Univ. Press, Princeton, New Jersey. 239p. Endler, J.A. 1986. Natural s e l e c t i o n in the wild. Princeton Univ. Press, Princeton 126 p. E r l i c h , P.R. and P.H. Raven. 1969. D i f f e r e n t i a t i o n of populations. Science 165: 1228-1232. Evelyn, T.P.T., J.E. Ketcheson, L. Prosperiporta. 1986. Use of erythromycin as a means of preventing v e r t i c a l transmission of Renibacterium  salmoninarum. Dist. Aquat. Org. 2(1): 7-11. Falconer, D.S. 1981. Introduction to quantitative genetics. Longman, N.Y. F a r r i s , J.S. 1972. Estimating phylogenetic trees from distance matrices. Am. Nat. 106: 645-668. F a r r i s , J.S. 1970. Methods for computing Wagner trees. Syst. Zool. 19: 83-92. F a r r i s , J.S. 1981. Distance data i n phylogenetic a n a l y s i s . In V.A. Funk and D.R. Brooks (eds). Advances in c l a d i s t i c s ; proceedings of the f i r s t meeting of the W i l l i e Hennig Society. New York, NY pp 3-23. Fast, D.E. and Q.J. Stober. 1984. Intragravel behavior of salmonid alevins in response to environmental changes. F i n a l Report for Wash. Water Research Center and City of S e a t t l e , S e a t t l e , WA. FRI-UW-8414:103 p. Ferguson, A. 1980. Electrophoretic analysis of t i s s u e proteins for I r i s h char. In Proceedings of the f i r s t ISACF workshop on A r t i e charr. Inst. Freshwater Research, Drottningholm, Sweden pp. 12-13. Fisher, R.A. 1930. The genetical theory of natural s e l e c t i o n . Dover, New York. Fowler, J.A. 1970. Control of vertebral number in t e l e o s t s - an embrological problem. Quart. Rev. B i o l . 45: 148-167. 278 Fresh, K.L., R.D. Cardwell, B.P. Snyder and E.O. Salo. 1982. Some hatchery st r a t e g i e s for reducing predation upon juvenile chum salmon (Oncorhynchus  keta) in freshwater. IN Proceedings of the North P a c i f i c Aquaculture Symposium. B.R. Melteff and R.A. Neve (eds). Alaska Sea Grant Rep., Alaska Sea Grant Program, Alaska Univ. pp. 79-89. Fresh, K.L. and S. Schroder. 1987. Influence of abundance, s i z e , and yolk reserves of juvenile chum salmon (Oncorhynchus keta) on predation by freshwater fishes in a small coastal stream. Can. J . F i s h . Aquat. S c i . 44: 236-243. Frolenko, L.A. 1970. Feeding of chum and pink salmon juveniles migrating downstream i n the main spawning r i v e r s of the northern coast of the Sea of Okhotsk. F i s h . Res. Bd. Transl. Ser. 2416. Frost, W.E. 1966. Breeding habits of Windermere char, Salvelinus w i l l u g h b i i (Gunther), and th e i r bearing on speciation of these f i s h . Proc. Royal soc. s e r i e s B: 232-284. Frydenburg, 0., D. Moller, G. Naevdal, and K. Sick. 1965. Haemoglobin polymorphism i n Norwegian cod populations. Hereditas 53: 257-271. Futuyma, D.J. 1979. Evolutionary biology. Siriauer Assoc., Inc. Sunderland, Mass. 565 p. Gabriel, K.R. 1964. A procedure for te s t i n g the homogeneity of a l l sets of means in analysis of variance. Biometrics 20: 459-477. Gadgil, M. and W.H. Bossert 1970. L i f e h i s t o r i c a l consequences of natural s e l e c t i o n . Amer. Natur. 104: 1-24. Ga l l , G.A.E. 1974. Influence of s i z e of eggs and age of female on ha t c h a b i l i t y and growth i n rainbow trout. C a l i f . Dep. Fish Game 00(1): 26-35. G a l l , G.A.E. 1975. Genetics of reproduction in domesticated rainbow t r o u t . J. Animal S c i . 40(1): 19-27. G a l l , G.A.E., J. Baltodana, and N. Huang. 1988. H e r i t a b i l i t y of age at spawning for Rainbow trout Aquaculture 68: 93-102. Gerking, S.D. 1988. . Fishery biology: Past, present and future. F i s h e r i e s 13: 13-17. G i l b e r t , C H . and W.H. Rich. 1927. Investigations concerning the red-salmon runs to the Karluk River, Alaska. U.S. Bur. F i s h . B u l l . 43: 1-69. Gjedrem, T. 1976. P o s s i b l i t i e s of genetic improvement i n salmonids. J . F i s h . Res. Board. Can. 33: 1094-1099. 279 Gjedrem, T. 1983. Genetic v a r i a t i o n in quantitative t r a i t s and s e l e c t i v e breeding i n f i s h and s h e l l f i s h . Aquaculture 33: 51-72. Grigo, L.D. 1953. Morphological differences between summer and autumn chum salmon Oncorhynchus keta (Walbaum) and 0. keta (Walbaum) infraspecies autumnalis Berg. IN " P a c i f i c salmon: selected a r t i c l e s from Soviet p e r i o d i c a l s . Clearinghouse Fed. S c i . Tech. Inform., S p r i n g f i e l d , Va. 22151. Godin, J-G.J. 1982. Migrations of salmonid fishes during early l i f e history phases: d a i l y and annual timing. IN: Salmon and trout migratory behaviour symposium. E.L. Brannon and E.O. Salo, eds. U. Wash., Seattle, pp. 22-50. Gould, S.J. 1982. Change in developmental timing as a mechanism of macroevolution. IN. Evolution and Development. J.T. Bonner (ed). Springer-Verlag, New York, NY. Gould, S.J. 1982. Is a new and general theory of evolution emerging? Paleobiology 6: 119-130. Gould, S.J. and N. Eldridge. 1977. Punctuated e q u i l i b r i a : tempo and mode of evolution reconsidered. Paleobiology 3: 115- 151. Groot, C , T.P. Quinn, and T.J. Hara. 1986. Response of migrating adult sockeye salmon (Oncorhynchus nerka) to pop u l a t i o n - s p e c i f i c odours. Can. J. Zool. 64: 926-932. Grotnes, P. 1980. P l a s t i c i t y of char population structures and the char complex. In Proceedings of the f i r s t ISACF workshop on A r t i e char. Instit u t e Freshwater Research, Drottningholm (Sweden): 14. Gunnes, K., and T. Gjedrem. 1978. Selection experiments with salmon. IV Growth of A t l a n t i c salmon during two years in the sea. Aquaculture 15: 19-33. Gustaffson, L. 1986. Lifetime reproductive success and h e r i t a b i l i t y : Empirical support for Fisher's fundamental theorem. Am. Nat. 128: 761-764. Harris, H. 1966. Enzyme polymorphisms in man. Proc. Roy. Soc. Land. B164: 298-310. Harlan, J.R. 1981. Who's in charge here. Can. J. F i s h . Aquat. S c i . 38: 1459-1463. Harrison, R.G. 1985. B a r r i e r s to gene exchange among c l o s e l y related c r i c k e t species. I I . L i f e c y c l e v a r i a t i o n and temporal i s o l a t i o n . Evolution 39(2): 244-259. 280 H a r t l , D. 1980. P r i n c i p l e s of population genetics. Sinauer Assoc. Inc., Sunderland, Mass. 488 p. H a r t l , D. 1981. A Primer of Population Genetics. Sinauer and Assoc. Inc., Sunderland, Mass. 191p. Hartt, A.C. 1980. Juvenile salmonids in the oceanic ecosystem - the c r i t i c a l f i r s t summer, p. 25-58. In Salmonid Ecosystems of the North P a c i f i c . W.J. McNeil and D.L. Himsworth (eds). Oregon St. Univ. Press, C o r v a l l i s , OR. Hasler, A.D. 1954. Odour perception and o r i e n t a t i o n in f i s h e s . J. F i s h . Res. Bd. Can. 11(2): 107-128. Healey, M.C. 1978. The d i s t r i b u t i o n , abundance, and feeding habits of juvenile P a c i f i c salmon in Georgia S t r a i t , B r i t i s h Columbia. Fish. Mar. Serv. Tech. Rep. 788. 49 p. Healey, M.C. 1979. Detritus and j u v e n i l e salmon production in the Nanaimo Estuary: I. Production and feeding rates of juvenile chum salmon (Oncorhynchus keta). J. Fish. Res. Board Can. 36: 488-496. Healey, M.C. 1980. The ecology of juvenile salmon in Georgia S t r a i t , B r i t i s h Columbia, p. 203-229. Insalmonid ecosystems of the North P a c i f i c . W.J. McNeil and D.C. Himsworth (eds.). Oregon St. Univ. Press. C o r v a l l i s , OR. Healey, M.C. 1982. Timing and r e l a t i v e i n t e n s i t y of s i z e - s e l e c t i v e mortality of juvenile chum salmon (Oncorhynchus keta) during early sea l i f e . Can. J. F i s h , and Aquat. S c i . 39: 952-957. Healey, M.C. 1983. Coastwide d i s t r i b u t i o n and ocean migration patterns of stream- and ocean-type Chinook salmon, Oncorhynchus tshawytscha. Canadian F i e l d - N a t u r a l i s t 97(4):427-433. Helle, J.H. 1979. Influence of marine environment on age and s i z e at maturity, growth, and abundance of chum salmon, Oncorhynchus keta (Walbaum), from Olsen Creek, Prince William Sound, Alaska. Ph.D. Thesis, Oregon State University. 118p. Helle, J.H. 1981. Significance of the stock concept in a r t i f i c i a l propagation of salmonids in Alaska. Can. J. F i s h . Aquat. S c i . 38: 1665-1671. Hitchcock, C L . and A. Cronquist. 1973. F l o r a of the P a c i f i c Northwest. University of Washington. Seattle and London. 730 p. Hoar, W.S. 1958. The evolution of migratory behaviour among juvenile salmon of the genus Oncorhynchus. J. F i s h . Res. Bd. Can. 15: 391-428. Hochberg, Y. 1976. A modification of the T-method in simultaneous inference. J. Multivar. Anal. 4: 224-234. 281 Hubbs, C.L. and K.F. Lagler. 1958. Fishes of the Great Lakes Region. Univ. Mich. Press, Ann Arbor, Mich. 213 p. Hunter, J.G. 1959. Survival and production of pink and chum salmon in a coastal stream. J. Fish. Res. Bd. Can. 16: 835-886. Hurlbert, S.H. 1984. Pseudoreplication and the design of e c o l o g i c a l f i e l d experiments. Ecolo g i c a l Monographs 54(2): 187-211. Hutchison, G.E. 1959. Homage to Santa Rosalia, or why are there so many kinds of animals. Am. Nat. 93: 145-159. Hynes, H.B.N. 1976. The Ecology of Running Waters. Waterloo University Press, Waterloo, Ontario. 526p. Ihssen, P.E. 1976. Selective breeding and h y b r i d i z a t i o n in f i s h e r i e s management. J. Fi s h . Res. Board Can. 33: 316-321. Ihssen, P.E., H.E. Booke, J.M. Casselman, J.M. McGlade, N.R. Payne, and F.M. Utter. 1981. Stock i d e n t i f i c a t i o n : materials and methods. Can. J. F i s h . Aquat. S c i . 38(12): 1838-1855. I r i e , T. 1982. Migration and ecology of young salamon in t h e i r early marine l i f e . 11th UJNR Aquaculture Panel, 1982, Tokyo, Japan. 17 p. Ivankov, V.N. 1967. On seasonal races of pink salmon. Izv. Tikhook. n . - i . i n - t a rybn. kh-va i okeanogr., 61. - 1968. The P a c i f i c salmon of Iturup Island ( K u r i l Islands). Izv. Tikhookeansk. n.- i . i n - t a rybn. kh-va i okeanogr., 65. Iwamoto, R.N., E.O. Salo, M.A. Madej, and R.L. McComas. 1978. Sediment and water q u a l i t y : a review of the l i t e r a t u r e including a suggested approach for water q u a l i t y c r i t e r i a ; with Summary of Workshop conclusions and recommendations, by E.O. Salo and R.L. R u l i f s o n . The U.S. Environmantal Protection Agency, Reg. 10, EPA 910/9-78-048: 150 p. Iwata, M. 1982. Downstream migration and seawater a d a p t a b i l i t y of chum salmon (Oncorhynchus keta) f r y , p. 51-59. In Proc. No. Pac. Aquaculture Symp., Aug. 1980, Anchorage, Ak. Univ. Alaska, Alaska Sea Grant Rep. 82-83. Jensen, H.M. 1956. Recoveries of immature chum salmon tagged in southern Puget Sound. Wash. Dept. Fis h . , F i s h . Res. papers 1(4):32. Jensen, J. and D. Alderdice. 1983. Changes in mechanical shock s e n s i t i v i t y of coho salmon (Oncorhynchus kisutch) eggs during incubation. Aquaculture 32: 303-312. Kapuscinski, A.R.D. and J.E. Lannan 1986. A conceptual f i t n e s s model for f i s h e r i e s management. Can. J. F i s h . Aquat. S c i . 43(8): 1606-1616. 282 Katayama, M. 1935. Biometric study of dog salmon (Oncorhynchus keta (Walbaum)). B u l l . Jap. Soc. S c i . F i s h . 4: 171-173. Transl. Bur. Commer. Fish . , B i o l . Lab., S e a t t l e , Wash., Transl. Ser. 1. Kettlewell, H.B.D. 1956. Further s e l e c t i o n experiments on i n d u s t r i a l melanism i n the Lepidoptera. Heredity 10: 287-301. K i l l i c k , S. 1955. The chronological order of Fraser River sockeye salmonduring migration, spawning and death. International P a c i f i c Salmon F i s h e r i e s Commission B u l l e t i n 7. Kimura, S. 1981. Records of chum salmon, Oncorhynchus keta from northern Kyushu. Jap. J. Ichthyol. 28(2): 193-196. Kneib, R.T. 1987. Predation r i s k and use of i n t e r t i d a l habitats by young Fishes and shrimp. Ecology 68(2): 379-386. Kobayashi, T. 1960. An e c o l o g i c a l study of the salmon f r y , Oncorhynchus keta (Walbaum). VI. Note on the feeding a c t i v i t y of chum salmon f r y . B u l l . Jpn. Soc. S c i . F i s h . 26: 577-580 (English Abstr.) Kobayashi, T. and Y. Ishikawa. 1964. An e c o l o g i c a l study on the salmon f r y , Oncorhynchus keta (Walbaum) - VIII. The growth and feeding habit of the fry during seaward migration. S c i . Rep. Hokkaido Salmon Hatch. 18: 7-15. Koo, T.S.Y. 1962. Age designation in salmon. p. 37-48. In T.S.Y. Koo (ed.), Studies of Alaska red salmon. Univ. Wash. Press, S e a t t l e . Koski, K.V. 1975. The s u r v i v a l and f i t n e s s of two stocks of chum salmon (Oncorhynchus keta) from egg deposition to emergence in a controlled-stream environment. Ph.D. Thesis, Univ. Wash., Seattle, Wash. 212p. Kostarev, V.L. 1973. Variation of s u r v i v a l of Okhotsk keta. F i s h . Res. Board. Can T r a n s l . Ser. 2536: 1-16. Krogius, F.V., V.S. Bool and A.S. Baranenkova. 1934. A note on the biology of the salmonid fishes of Kamchatka. Byul. Kamchatskogo Otdel TINRO, No.1. Kubo, T. 1950. A preliminary report of the study on the groups of Oncorhynchus keta (Walbaum) (dog salmon) and the numbers of t h e i r segments. B u l l . Fac. F i s h e r i e s , Hokkaido Univ., 1(1): 1-11 Kubo, T. 1956. P e c u l i a r i t y of population of chum salmon i n the Shuriuchi River i n respect to vertebral count. B u l l . Fac. F i s h e r i e s Hokkaido Univ., 6(4): 266-270. U.S. Fish W i l d l i f e Serv., Translation No. 12 . Kulikova, N.I. 1971. I n t r a s p e c i f i c v a r i a b i l i t y of karyotype of the chum salmon (Oncorhynchus keta (Walb.)). J. Ichthyol. 11: 977-983. 283 Lalanne, J.J. 1971. Marine growth of chum, salmon. Int. N. Pac. F i s h . Comm. B u l l . 27: 91p. Lande, R. 1975. The maintenance of gametic v a r i a b i l i t y by mutation in a polygenic character with linked l o c i . Genet. Res. 26: 221-235. Larkin, P.A. 1972. The stock concept and management of P a c i f i c salmon. Univ. B r i t i s h Columbia, H.R. MacMillan Lectures, The stock concept in P a c i f i c salmon. 11-15. Larkin, P.A. 1981. A perspective on population genetics and salmon management. Can. J. F i s h . Aquat. S c i . 38(12): 1469-1475. Leary, R.F., F.W. Allendorf and K.L. Knudsen. 1985. Inheritance of meristic v a r i a t i o n and the evolution of developmental s t a b i l i t y i n rainbow trout. Evolution 39(2): 308-314. Leider, S.A., M.W. C h i l c o t e , and J.J. Loch. 1984. Spawning c h a r a c t e r i s t i c s of sympatric populations of steelhead trout (Salmo g a i r d n e r i ) : evidence for p a r t i a l reproductive i s o l a t i o n . Can. J. F i s h . Aquat. S c i . 41: 1454-1462. Leider, S.A., M.W. C h i l c o t e , J.J. Loch. 1986. Comparative l i f e h istory c h a r a c t e r i s t i c s of hatchery and wild steelhead trout (Salmo gairdneri) of summer and winter races in the Kalama r i v e r , Washington. Can. J. F i s h . Aquat. S c i . 43(7): 1398-1409. Lewontin, R.C. and J.L. Hubby. 1966. A molecular approach to the study of genie heterozygosity i n natural populations of Drosophila  pseudo-obscura. Genetics 54: 595-609. Levins, R. 1963. Theory of f i t n e s s i n a heterogeneous environment I I . Developmental f l e x i b i l i t y and niche s e l e c t i o n . Am. Nat. 97: 75-90. Levy, D.A. and T.G. Northcote 1981. The d i s t r i b u t i o n and abundance of juvenile salmon in marsh habitats of the Fraser River Estuary. Univ. B r i t i s h Columbia Tech. Report No. 25, Vancouver, B.C. Lindenfelser, M.E. 1984. Morphometric and allozymic congruence: Evolution i n the prawn, Macrobrachium rosenbergii (Decapoda: Palaemonidae). Syst. Zool. 33(2): 195-204. Lindsey, C C . 1975. Pleomerism, the widespread tendency among related f i s h species for vertebral number to be correlated with maximum body length. J. F i s h . Res. Board Can. 32: 2453-2469. Lindsey, C C 1988. Factors c o n t r o l l i n g m e r i s t i c v a r i a t i o n . F i s h physiology XIB: 197-274. Lindsey, C C , A.M. Br e t t , D.P. Swain. 1984. Responses of vertebral numbers in rainbow trout to temperature changes during development Can. J. Zool. •62(3): 391-396. 284 Lindsey, C C . and A.N. Arnason. 1981. A model for repsonses of vertebral numbers i n f i s h to environmental influences during development. Can. J . F i s h . Aquat. S c i . 38: 334-347. L i s t e r , D.B., D.G. Hickey and I. Wallace. 1981. Review of the e f f e c t s of enhancement strategies on the homing, straying and s u r v i v a l of P a c i f i c salmonids. Volume 1. D.B. L i s t e r and Associates Ltd., West Vancouver, B.C. 51p. Live l y , CM. 1986. Canalization versus developmental conversion in a s p a t i a l l y variable environment Am. Nat. 128: 561-572. Loftus, K.H. 1981. Where are we now? Can. J . F i s h . Aquat. S c i . 38: 1464-1465. Lyamin, K.A. 1949. Results of tagging P a c i f i c salmon in the Gulf of Kamchatka In P a c i f i c Salmon: Selected a r t i c l e s from Soviet p e r i o d i c a l s . Transl. from Russian, IPST Cat. No. 341 (1961) 184-190. Lynch, M. 1986. Random d r i f t , uniform s e l e c t i o n , and the degree of population d i f f e r e n t i a t i o n . Evolution 40: 640-643. Mahalanobis, P.C. 1936. On the generalized distance in s t a t i s t i c s . Proc. Nat. Inst. S c i . India 2: 49-55. Main, K.L. 1987. Predator avoidance in Seagrass Meadows: Prey Behaviour, microhabitat s e l e c t i o n , and c r y p t i c c o l o r a t i o n . Ecology 68(1): 170-180. Mason, J.C. 1974. Behavioural ecology of chum salmon fry (Oncorhynchus keta) i n a small estaury. J. F i s h . Res. Board Can. 31(1): 83-92. Mattson, CR. and R.G. Rowland. 1963. Chum salmon studies at T r a i t o r s Cove F i e l d S t a t i o n — J u n e 1960 to March 1963. Bur. Commer. Fi s h . , B i o l . Lab., Auke Bay, Alaska, Ms. Rep. 63-11, 32p. Mayr, E. 1963. Animal Species and Evolution. Belknap Press, Harvard Univ., Mass. 771p. Maynard-Smith, J. 1966. Sympatric s p e c i a t i o n . Am. Nat. 100: 637-650. McNamara, J.M., and A.I. Houston. 1987. Starvation and Predation as factors l i m i t i n g population s i z e . Ecology 68(5): 1515-1519. McNeil, W.J. 1966. E f f e c t of the spawning bed environment on reproduction of pink and chum salmon. U.S. F i s h . W i l d l . Serv., F i s h . B u l l . 65: 495-523. Mead, R.W. and W.L. Woodall. 1968. Comparison of sockeye salmon fry produced by hatcheries, a r t i f i c i a l channels and natural spawning areas. Int. Pac. Salmon Fi s h . Comm. Prog. Rep. 20 41p. 285 Meffe, G.K. 1986. Conservation genetics and the management of endangered f i s h e s . F isheries 11: 14-23. Meffe, G.K. 1987. Conserving f i s h genomes: philosophies and pr a c t i s e s . Env. B i o l . Fishes 18: 3-9. Me r r i t t , M.F. and J.A. Raymond 1983. Early l i f e h istory of chum salmon in the Noatak River and Kotzebue Sound. Alaska Dept. Fish and Game, Div. of Fish . Rehab. Enhan. and Dev., Report No. 1: 56p. Mer r i t t , M.F. and K. Roberson. 1986. Migratory timing of upper Copper River sockeye salmon stocks and i t s implications for the regulation of the commercial f i s h e r y . North American Journal of Fi s h e r i e s Management 6(2): 216-225. Mihara, T., S. Ito, T. Hachiya, and M. Ichikawa. 1951. Studies on the change of f i s h i n g conditions of salmon in Hokkaido. (1) The f i s h i n g conditions on salmon. S c i . Rep. Hokkaido Fish Hatch. 6: 27-133. M i l l e r , R.J. and E.L. Brannon 1982. The o r i g i n and development of l i f e h i s t o r y patterns in P a c i f i c salmonids. IN Proceedings of the Salmon and Trout Migratory Behaviour Symposium. E.L. Brannon and E.O. Salo (eds). 1st International Symposium, School of F i s h e r i e s , Univ. Washington, Se a t t l e , Washington, 98195. pp. 296-309. Mundy, P.R. 1982. Computation of migratory timing s t a t i s t i c s for adult chinook salmon in the Yukon River, Alaska and t h e i r relevance to f i s h e r i e s management. North American Journal of Fis h e r i e s Management 2(4): 359-370. Murray, C.B. 1980. Some e f f e c t s of temperature on zygote and ale v i n s u r v i v a l , rate of development and s i z e at hatching and emergence of P a c i f i c salmon and rainbow trout. M.Sc. Thesis, Univ. B r i t i s h Columbia, Vancouver, B.C. 156p. Murray, C.B., and T.D. Beacham. 1987. The development of chinook (Oncorhynchus tshawytscha) and chum salmon (Oncorhynchus keta) embryos and alevins under varying temperature regimes. Can. J. Zool. 65: 2672-2681. Naevdal, G., M. Holm, 0. Ingebrightsen, and D. Moller. 1978. Variation i n age at f i r s t spawning i n A t l a n t i c salmon (Salmo salar) J. F i s h . Res. Board Can. 35: 145-147. Neave, F. 1949. Game f i s h populations of the Cowichan River. B u l l . F i s h . Res. Bd. Can. 84. 32 p. Neave, F. 1953. P r i n c i p l e s a f f e c t i n g the s i z e of pink and chum salmon populations i n B r i t i s h Columbia. J. Fish. Res. Bd. Can. 9: 450-491. Neave, F. 1955. Notes on the seaward migration of pink and chum salmon f r y . J. F i s h . Res. Bd. Can. 12: 369-374. 286 Neave, F. 1966. Salmon of the North P a c i f i c Ocean - Part I I I . A review of the l i f e h istory of North P a c i f i c salmon. 6. Chum salmon in B r i t i s h Columbia. Int. N. Pac. F i s h . Comm., B u l l . 18:81-86 Neave, F., T. Yonemori, and R. Bakkala. 1976. D i s t r i b u t i o n and or i g i n of chum salmon in offshore waters of the North P a c i f i c Ocean. Int. North Pac. Fish. Comm. B u l l . 35: 79p. Nicola, S.J. 1962. Scavenging by A l l o p e r l a (Plecoptera: Chloroperlidae) nymphs on dead pink (Oncorhynchus gorbuscha) and chum (0. keta) salmon embryos. Can. J. Zool. 46: 787-796. Nei, M. 1971. I n t e r s p e c i f i c gene differences and evolutionary time estimated from electrophoretic data on protein i d e n t i t y . Am. Nat. 105: 385-398. Nei, M. 1972. Genetic distance between populations. Am. Nat. 106: 283-292. Nei, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of i n d i v i d u a l s . Genetics 89: N i k o l ' s k i i , G.V. 1952. The type of dynamics of stocks and the character of spawnig of the chum Oncorhynchus keta (Walb.) and pink salmon Oncorhynchus gorbuscha (Walb.) in the Amur River. In P a c i f i c Salmon: Selected a r t i c l e s from Soviet p e r i o d i c a l s . Transl. from Russian, IPST Cat. No. 341 (1961):9-12. Noerenberg, W.H. 1971. Earthquake damage to Alaskan f i s h e r i e s . Pp. 170-193 IN The great Alaska earthquake of 1964, Biology. N a t l . Res. Counc, Natl. Acad. S c i . , Washington, D.C. Norman, J.R. 1975. A History of Fishes. (3rd Edition) London. 465p. Northcote, T.G. 1978. Migratory s t r a t e g i e s and production i n freshwater f i s h e s . IN: Ecology of freshwater f i s h production, S.D. Gerking, ed. Blackwell S c i . Publ. Oxford: 326-329. Northvedt, R. 1986. The timing of emergence of A t l a n t i c salmon and rainbow trout, incubated of d i f f e r e n t substrates. IN Proceedings of counc. Meet, of the i n t . counc. for the Exploration of the sea. ICES, Copenhagen (eds): 33 pp. Nyman, L. 1972. A new approach to the taxonomy of the "Salvelinus alpinus species complex" Rep. Inst. Freshwater Res. Drottningholm 52: 103-131. Nyman, L. 1980. The population genetics of A r t i e char: A state of the a r t . In Proceedings of the f i r s t ISACF workshop on Ar t i e charr. Inst. Freshwater Research, Drottningholm, Sweden pp. 46-48. . Okazaki, T. 1978. Genetic differences of two chum salmon (Oncorhynchus keta) populations returing to the Tokachi River. B u l l . Far Seas F i s h . Res. Lab. 16: 121-127. 287 Okazaki, T. 1981. Geographical d i s t r i b u t i o n of a l l e l i c v ariations of enzymes in chum salmon, Oncorhynchus keta, populations in North America. B u l l . Japan. Soc. S c i . F i s h . 47: 507-514. Okazaki, T. 1983. Genetic structure of chum salmom, Oncorhynchus keta, r i v e r populations. B u l l . Jap. Soc. S c i . F i s h . 49(2): 189-196. Olsen, K.H. 1986. Chemo-attraction between juveniles of two sympatric stocks of a r t i c charr, Salvelinus alpinus (L.) and t h e i r gene frequency of serum esterases. J. Fish. B i o l . 28(2): 221-231. Parsons, P.A. 1977. Genes, behaviour and evolutionary processes: The genus Drosophila Adv. Genetics. 19: 1-32. Parker, R.R. 1962. A concept of the dynamics of pink salmon populations. In: N.J. Wilimovsky (ed.), Symposium on pink salmon, pp 203-211. H.R. Macmillan Lectures in F i s h e r i e s , Univ. B r i t . Columbia, Vancouver. Payne, R.H.' A.R. C h i l d , and A. Forrest. 1971. Geographical v a r i a t i o n i n the A t l a n t i c salmon. Nature 231: 250-252. Pearson, K. 1926. On the c o e f f i c i e n t of r a c i a l likeness. Biometrika 8: 105-117. Peterman, R.H. and M. Gatto. 1978. Estimation of functional responses of predators on juvenile salmon. J. F i s h . Res. Board. Can. 28: 1503-1510. Pimentel, D., G.J.C. Smith and D. Scans. 1967. A population model of sympatric s p e c i a t i o n . Am. Nat. 101: 493-504. Plytyez, B., J. Dulak, and Pecio. 1984. Genetic control of length of the l a r v a l period i n Rana temporaria. F o l i a B i o l . 32(3): 155-166. Prager, E.M. and A.C. Wilson. 1978. Methods of phylogenetic analysis. J. Mol. Evol. 11: 129-142. Pritchard, A.L. 1943. The age of chum salmon taken in commercial catches in B r i t i s h Columbia. F i s h . Res. Bd. Can., Progr. Rep. Pac. Coast Sta. 54: 9-11. Quinn, T.P. 1984. Homing and straying in P a c i f i c salmon. IN: J.D. McCleave, G.P. Arnold, J.J. Dodson and W.H. N e i l l . (Eds.) Mechanisms of migration i n f i s h e s , pp. 357-362. Quinn, T.P. and K. Fresh. 1984. Homing and straying in chinook salmon (Oncorhynchus tshawytscha) from Cowlitz River Hatchery, Washington. Can. J. F i s h . Aq. S c i . 41(7): 1078-1082. Quinn, T.P. 1985. Homing and the evolution of sockeye salmon. IN: Migration: Mechanisms and adaptive s i g n i f i c a n c e . Rankin, M.A., D. Checkley, J. Cullen, C. K i t t i n g , and P. Thomas (eds.): 353-366. 288 Quinn, T.P. and G.M. Tolson. 1986. Evidence of chemically mediated population recognition in coho salmon (Oncorhynchus kisutch) Can. J. Zool. 64: 84-87. Quinn, T.P. and R.F. Tallman. 1987. Seasonal environmental p r e d i c t a b i l i t y and homing in r i v e r i n e f i s h e s . Env. B i o l . Fishes 18(2): 155-159. Refstie, T. and T.A. Steine. 1978. Selection experiments with salmon I I I . Genetic and environmental sources of va r i a t i o n in length and weight of A t l a n t i c salmon i n the freshwater phase. Aquaculture 14: 221-234. Ricker, W.E. 1959. Vernal and heimal races among anadromous f i s h e s . J. Fish. Res. Bd. Can. 16: 515-537. Ricker, W.E. 1964. Ocean growth and mortality of pink and chum salmon. J. Fish . Res. Bd. Can. 21: 905-931. Ricker, W.E. 1972. Hereditary and environmental factors a f f e c t i n g c e r t a i n salmonid populations p.19-160. In R.C. Simon and P.A. Larkin ed. The Stock Concept in P a c i f i c Salmon. H.R. MacMillan Lectures in F i s h e r i e s , Univ. B r i t i s h Columbia, Vancouver, B.C.. Ricker, W.E. 1976. Review of the rate of growth and mortality of P a c i f i c salmon in saltwater, and noncatch mortality caused by f i s h i n g . J. F i s h . Res. Bd. Can. 33(7): 1483-1524. Rid d e l l , B.E. and W.C. Leggett. 1981. Evidence for an adaptive basis for geographic v a r i a t i o n in body morphology and time of downstream migration in j u v e n i l e A t l a n t i c salmon (Salmo s a l a r ) . Can. J. Fish. Aquat. S c i . 38: 308-320. Rid d e l l , B.E., W.C. Leggett and R.L. Saunders. 1981. Evidence of adaptive polygenic v a r i a t i o n between two populations of A t l a n t i c salmon (Salmo  salar) native to t r i b u t a r i e s of S.W. Miramichi River, N.B., Can. Can. J . F i s h . Aquat. S c i . 38: 321-333. Robertson, A. 1955. Selection i n animals: synthesis. Cold Spring Harbor Symp. Quant. B i o l . 20: 225-229. Rodda, D., L. Schaffer, K. Mullen and G. F r i a r s . 1977. Measuring the p r e c i s i o n of genetic parameters by simulation technique. Theor. App. Genet. 51: 35-40. Rosly, Yu. 1972. The scale structure of the Amur chum (Oncorhynchus keta (Walb.)) as an indicator of growth and l i v i n g conditions i n the freshwater stage. J. Ichthyol. 12(3): 483-497. Roughgarden, J. 1979. Theory of population genetics and evolutionary ecology: An Introduction. MacMillan, New York 634 p. 289 Rugh, R. 1952. Experimental embryology. Burgess Publishing Co., Minneapolis, MN. 480p. Rutberg, A.T. 1987. Adaptive hypotheses of b i r t h synchrony in ruminants: an i n t e r s p e c i f i c t e s t . Am. Nat. 130(5): 692-710. Ryman, N., F.W. Allendorf and G. S t a h l . 1979. Reproductive i s o l a t i o n with l i t t l e genetic divergence in sympatric populations of brown trout (Salmo  t r u t t a ) . Genetics 92: 247-262. Ryman, N. and G. Stahl. 1981. Genetic perspectives of the i d e n t i f i c a t i o n and conservation of Scandinavian stocks of f i s h . Can. J. F i s h . Aq. S c i . 38(12): 1562-1575. Salo, E.O. 1986. The l i f e h i s t o r y of chum salmon (Oncorhynchus keta). FRI Univ. Washington, Seattle, Washington. 238 p. MS. Salo, E.O. and R.E. Noble 1952. Minter Creek B i o l o g i c a l Station, Progress Report (Sept-Dec. 1952). Part I: Chum salmon. Washington Dept. Fis h , Olympia, WA. 33p. Sano, S. 1959. The ecology and propagation of genus Oncorhynchus found i n northern Japan. S c i . Rep. Hokkaido Salmon Hatch. 14:21-90 Sano, S. 1966. Salmon of the North P a c i f i c Ocean. Part I I I . A review of the l i f e h istory of North P a c i f i c salmon, 3. chum salmon in the Far East, Int. N. Pac. F i s h . Comm., B u l l . 18: 41-57. Saviaatova, K.A. 1983. The a p p l i c a t i o n of the b i o l o g i c a l species concept to an evaluation of the systematic p o s i t i o n of chars of the genus Salvelinus (Salmonidae). J. Ichthyol. 23(6): 1-12. Schmidt, P.Yu. 1947. Fish migrations. Moscow, Leningrad, U.S.S.R. Acad. S c i . Press. Schroder, S.L. 1973. E f f e c t s of density on spawning success of chum salmon (Oncorhynchus keta) i n an a r t i f i c i a l spawning channel. M.Sc. Thesis, Univ. Wash., Seattle, Wash. 78p. Schroder, S.L. 1974. Assessment of production of chum salmon f r y form the Big Beef Creek spawning channel. Univ. Washington, Fi s h . Res. Inst., Anadromous Fi s h Project, Annu. Rep. FRI-UW-7413. 56 p. Schroder, S.L. 1977. Assessment of production of chum salmon fry from the Big Beef Creek spawning channel. Completion Report. F i s h . Res. Inst., Univ. Washington, S e a t t l , WA. FRI-UW-7718: 77p. Schroder, S.L. 1981. The ro l e of sexual s e l e c t i o n in determining o v e r a l l mating patterns and mate choice in chum salmon. Ph.D. Thesis, College of Fi s h e r i e s , University of Washington. 290 Semlitsch, R.D., D.E. Scott and J.H.K. Pechmann. 1988. Time and si z e at metamorphosis related to adult f i t n e s s i n Ambystoma talpoideum. Ecology 69: 184-192. Seymour, A. 1959. E f f e c t s of temperature upon the formation of vertebrae and f i n rays in young chinook salmon. Trans. Am. F i s h . Soc. 88(1): 58-69. Sheridan, W.L. 1962. Relation of stream temperatures to timing of pink salmon escapements in Southeast Alaska. IN: Symposium on Pink Salmon, H.R. Macmillan Lectures i n F i s h e r i e s , Univ. B r i t i s h Columbia, Vancouver pp. 87-102 Sil l i m a n . 1975. Selective and unselective f i s h i n g . Fishery B u l l e t i n Vol. 73: 495-Shine, R. 1988. The evolution of large body siz e i n Females: A c r i t i q u e of Darwin's "Fecundity advantage" model. Am. Nat. 131: 124-131. Sibert, J.R. 1979. Detritus and j u v e n i l e salmon production in the Naniamo Estuary: I I . Meiofauna av a i l a b l e as food to juvenile chum salmon (Oncorhynchus keta). J. F i s h . Res. Board Can. 36: 497-503. Simenstad, C.A., W.J. Kinney, S.S. Parker, E.O. Salo, J.R. C o r d e l l , and Heuchner. 1980. Prey community structure and trophic ecology of outmigrating juvenile chum and pink salmon in Hood Canal, Washington: A synthesis of three years' studies, 1977-1979. F i n a l Rep., F i s h . Res. INst., Univ. Washington, S e a t t l e , WA. FRI-UW-8026. 113 p. Simenstad, C.A., K.L. Fresh, and E.O. Salo. 1982. The role of Puget Sound and Washington coastal estuaries i n the l i f e h istory of P a c i f i c salmon: and unappreciated function, p. 343-364. In Estuarine Comparisons, V.S. Kennedy (ed.). Academic Press. Singh, R.S., R.C. Lewontin, and A.A. Felton. 1976. Genetic heterogeneity within electrophoretic " a l l e l s " of xanthine dehydrogenase in Drosophila  pseudoobscura. Genetics 84: 609-629. S l a t k i n , M. 1987. Quantitative genetics of heterochrony. Evolution 41: 799-811. Smirnov, A.I. 1955. The e f f e c t of mechanical a g i t a t i o n at d i f f e r e n t periods of development on the eggs of autumn salmon (Oncorhynchus keta) i n f r a s p . autumnalis Berg, Salmonidae). F i s h . Res. Board Can. T r a n s l . Ser. No. 230, 1959. Smirnov, A.I. 1975. The biology, reproduction and development of the P a c i f i c salmon. F i s h . Mar. Serv. Transl. Ser. 3861. 335 p. Smith, S.B. 1969. Reproductive i s o l a t i o n in summer and winter races of steelhead trout. p. 21-38. In T.G. Northcote ed. Symposium on Salmon and Trout in Streams. H.R. MacMillan Lectures in F i s h e r i e s , Univ. B r i t i s h Columbia. 291 Smoker, W.W. 1982. Quantitative genetics of chum salmon, Oncorhynchus keta (Walbaam) Oregon State Univ. Ph.D. Thesis. 170 p. Sokal, R.R. and F.J. Rohlf. 1981. Biometry, 2nd Ed. W.H. Freeman and Co. New York, N.Y. 859 p. Spalding, D.J. 1964. Comparative feeding habits of the fur s e a l , sea l i o n and harbor seal on the B r i t i s h Columbia coast. F i s h . Res. Bd. Can., B u l l . 146: 52 p. Spieth, P.T. 1974. Gene flow and genetic d i f f e r e n t i a t i o n . Genetics 78: 961-965. Stahl, G. 1981. Genetic d i f f e r e n t i a t i o n among natural populations of A t l a n t i c salmon (5almo salar) i n northern Sweden. E c o l . B u l l 34: Stanley, S.M. 1979. Macroevolution: Pattern and process. W.H. Freeman and Co. San Francisco. 332 p. Stearns, S.C. 1977. The evolution of l i f e history t r a i t s . Ann. Rev. Ecol. Syst. 8: 145-172. Stebbins, G.L. 1976. Chromosome, DNA and plant evolution. Evol. B i o l . 9: 1-34. Svetovidova, A.A. 1961. Local stocks of summer keta, Oncorhynchus keta (Walbaum), of the Amur Basin. F i s h . Res. Board. Can. T r a n s l . Ser. No. 347. 11 p. Swain, D.P. 1987. A problem with the use of meristic characters to estimate developmental s t a b i l i t y . Am. Nat. 129: 761-768. Swain, D. and C C . Lindsey. 1984. S e l e c t i v e predation for vertebral number of young sticklebacks (Gasterosteus aculeatus). Can. J. Fi s h . Aq. S c i . 41: 1231-1233. Swofford, D.L. 1981. On the u t i l i t y of the distance Wagner procedure. IN V.A. Funk and D.R. Brooks (eds) Advances in c l a d i s t i c s : proceedings of the f i r s t meeting of the W i l l i e Hennig Society. New York, N.Y. 25-43 p. Taguchi, K. 1961. A t r i a l to estimate the instantaneous rate of natural mortality of adult salmon (Oncorhynchus sp.) and the consideration of r a t i o n a l i t y of offshore f i s h i n g . I. For chum salmon (Oncorhynchus  keta). B u l l . Jpn. Soc. S c i . F i s h . 27: 963-971. Taning, A.V. 1952. Experimental study of meristic characters in f i s h e s . B i o l . Rev. 27: 169-193. Tauber, M.J. and C A . Tauber. 1976. Experimental control of univoltinism and i t s evolution. Ann. Rev. Ecol. Syst. 12: 281-308. 292 Tauber, C A . and M.J. Tauber. 1981. Insect seasonal cycles: genetics and evolution. Ann. Rev. Ecol. Syst. 12: 281-308. Tauber, C A . and M.J. Tauber, J.R. Nechols. 1977. Two genes control seasonal i s o l a t i o n i n s i b l i n g species. Science 197: 592-593. Taylor, S.G. 1980. Marine s u r v i v a l of pink salmon fry from early and la t e spawners. Trans Amer. F i s h . Soc. 109: 79-82. Taylor, E.B. and J.D. McPhail. 1985. Variations i n body morphology among B r i t i s h Columbia populations of coho salmon, Oncorhynchus kisutch. Can. J. F i s h . Aquat. S c i . 42: 2020-2028. Templeton, A.R. 1981. Mechanisms of speciation - a population genetic approach. Ann. Rev. E c o l . Syst. 12: 23-48. Thoday, J.M. and T.B. Boam. 1959. E f f e c t s of d i s t r u p t i v e s e l e c t i o n I I I : Polymorphism and divergence without i s o l a t i o n . Heredity 13: 205-218. Thoday, J.M. and J.B. Gibson. 1962. Isolation by disruptive s e l e c t i o n . Nature 193: 1164-1166. Thoday, J.M. 1972. Disruptive s e l e c t i o n . Proc. Roy. Soc. Lond. (B) 182: 109-143. Thorsteinson, F.V., W.H. Noerenberg, and H.D. Smith. 1963. The length, age, and sex r a t i o of chum salmon in the Alaska Peninsula, Kodiak Island, and Prince William Sound areas of Alaska. U.S. Fish. Wildl. Serv., Spec. S c i . Rep. F i s h . 430. 84 p. Todd, I.S. 1966. A technique for the enumeration of chum salmon fry in the Fraser River, B r i t i s h Columbia. The Can. F i s h . C ult. 38: 3-35. Trasky, L.T. 1974. Yukon River anadromous f i s h i n v e s t i g a t i o n s . Anadromous Fish Conservation Act, Completion Report for period July 1, 1973 to June 30, 1974. Alaska Dept. Fish and Game, Comm. Fish. Div., Juneau, AK. 111 P-Tyler, A.V. 1966. Some l e t h a l temperature r e l a t i o n s of two minnows of the genus Chrosomus. Can. J. Zool. 44: 349-361. Van den Berghe, E.P. and M.R. Gross. 1984. Female competition, l i f e - t i m e reproductive success, and i n t e n s i t y of natural s e l e c t i o n i n coho salmon (Oncorhynchus kisutch). Submitted to Evolution. Van Vleck, L.D. 1973. Selection for d i r e c t and maternal genetic e f f e c t s . Genetics Lectures 3: 133-148. Velsen, F.P. 1980. Embryonic development in eggs of sockeye salmon, Oncorhynchus nerka. Can. Spec. Pub. Fish. Aquat. S c i . no. 49. 293 Vladimirov, V.I. 1975. C r i t i c a l periods in the development of f i s h . Vopr. I k t i o l . 15(6): 955-975. Vourinen, J. , M. Himberg, and P. Lankinen. 1981. Genetic d i f f e r e n t i a t i o n in Coregonus albula (L.) (Salmonidae) populations i n Finland. Hereditas 94: 113-121. Walters, C.J., R. Hilborn, R.M. Peterman, and M.J. Staley. 1978. Model for examining early ocean l i m i t a t i o n of P a c i f i c salmon production. J. F i s h . Res. Board. Can. 35(10): 1303-1315. Wehrhahn, C F . and R. Powell. . 1987. Electrophoretic v a r i a t i o n , regional differences, and gene flow in the coho salmon (Oncorhynchus kisutch) of southern B r i t i s h Columbia. Can. J. F i s h . Aquat. S c i . 44: 822-831. White, M.J.D. 1978. Modes of speciation. Freeman and Co. San Francisco, U.S.A. 455 p. Whitmus, C J . and S. Olson. 1979. The migratory behaviour of juvenile chum salmon release in 1977 from the Hood Canal Hatchery at Hoodsport, Washington. Univ. Washington, F i s h . Res. Inst,, FRI-UW-7916. 46 p. Wickett, W.P. 1954. The oxygen supply to salmon eggs in spawning beds. J. Fish. Res. Board Can. 11(6): 933-953. Wickett, W.P. 1958. Review of c e r t a i n environmental factors a f f e c t i n g the production of pink and chum salmon. J. Fish. Res. Bd. Can. 15: 1103-1126. Withler, R.E. 1988. Genetic consequences of f e r t i l i z i n g Chinook salmon (Oncorhynchus tshawytscha) eggs with pooled m i l t . Aquaculture, 68: 15-25. Withler, R.E., M.C. Healey, and B.E. R i d d e l l . 1982. Annotated bibliography of genetic v a r i a t i o n i n the family Salmonidae. Can. Tech. Rep. F i s h . Aquat. S c i . No. 1098. 161 p. Withler, R.E., B.E. R i d d e l l , and H. Kreiberg. 1987. Genetic v a r i a t i o n i n freshwater s u r v i v a l and growth of chinook salmon (Oncorhynchus  tshawytscha). Aquaculture, 64: 85-96. Wright, S. 1931. Evolution i n mendelian populations. Genetics 16: 97-159. Wright, S. 1969. Evolution and the genetics of populations. Volume 2: The theory of gene frequencies University of Chicago Press, Chicago. Wright, S. 1978. Evolution and the genetics of populations volume 4: V a r i a b i l i t y within and among natural populations. Univ. Chicago Press 580 p. Wright, S. 1988. Surfaces and s e l e c t i v e value r e v i s i t e d . Am. Nat. 131: 115-123. 294 APPENDIX 1 Analysis of variance tables estimating the ef f e c t of population and temperature regime for mean time to hatch and emergence during the 1982-83 and 1983-84 incubation experiments. 295 Table aa. 1982-83 Mean Time t o Hatch ( B a r t l e t t ' s Test - Chi Square = 18.28) ANALYSIS OF VARIANCE TAIL SOURCE SUM OF SQUARES DF MEAN SQUARE F VALUE PROBABILITY POP 581.9249 3 193.9750 40.55 0.0000 REGIME 13706.8584 3 4568.9526 955.14 0.0000 INTERACTION 796.1072 9 88.4564 18.49 0.0000 ERROR 76.5368 16 4.7835 ANALYSIS OF VARIANCE; VARIANCES ARE NOT ASSUMED TO BE EQUAL WELCH 15, 6 1198.75 0.0000 BROWN-FORSYTHE* POP 3, 4 40.55 0.0019 REGIME 3, 4 955.14 0.0000 INTERACTION 9, 4 18.49 0.0065 ALL GROUPS COMBINED (EXCEPT CASES WITH UNUSED VALUES FOR VARIABLES POP AND REGIME) MEAN 86.920 STD. DEV. 22.115 S. E. M. 3.909 MAXIMUM 128.262 MINIMUM 53.150 CASES EXCLUDED ( 0) CASES INCLUDED 32 •ROBUST S.D. 22.137 296 Table bb. 1983-84 Mean Time to Hatch (Bartlett's Test - Chi Square = 25.01) SOURCE ANALYSIS OF VARIANCE SUM OF SQUARES DF MEAN SQUARE F VALUE TAIL PROBABILITY POP 705.7712 3 235.2571 39.10 0.0000 REGIME 12729.1387 3 4243.0459 705.19 0.0000 INTERACTION 1031.5887 9 114.6210 19.05 0.0000 ERROR 96.2698 16 6.0169 ANALYSIS OF VARIANCE; VARIANCES ARE NOT ASSUMED TO BE EQUAL WELCH 15, 6 6010.63 0.0000 BROWN-FORSYTHE* POP 3, 3 39.10 0.0066 REGIME . 3, 3 705.19 0.0001 INTERACTION 9, 3 19.05 0.0169 ALL GROUPS COMBINED (EXCEPT CASES WITH UNUSED VALUES FOR VARIABLES POP AND REGIME) MEAN 90.419 STD. DEV. 21.674 S. E. M. 3.831 MAXIMUM 131.190 MINIMUM 53.995 CASES EXCLUDED ( 0) CASES INCLUDED 32 *ROBUST S.D. 21.780 297 Table c c . 1982-83 Mean Time t o Emergence ( B a r t l e t t ' s Test - Chi Square = 19.21). SOURCE ANALYSIS OF VARIANCE SUM OF SQUARES DF MEAN SQUARE F VALUE TAIL PROBABILITY POP 1401.7726 3 467.2575 153.29 0.0000 REGIME 30626.8750 3 10208.9580 3349.24 0.0000 INTERACTION 775.4931 9 86.1659 28.27 0.0000 ERROR 48.7703 16 3.0481 ANALYSIS OF VARIANCE; VARIANCES ARE NOT ASSUMED TO BE EQUAL WELCH BROWN-FORSYTHE* POP REGIME INTERACTION 15, 6 3, 7 3, 7 9, 7 1.97E+5 153.29 3349.24 28.27 0.0000 0.0000 0.0000 0.0001 ALL GROUPS COMBINED (EXCEPT CASES WITH UNUSED VALUES FOR VARIABLES POP AND REGIME) MEAN STD. DEV. S. E. M. MAXIMUM MINIMUM CASES EXCLUDED CASES INCLUDED *ROBUST S.D. 137.717 32.554 5.755 173.225 81.715 ( 0) 32 34.749 2 9 8 Table dd. 1983-84 Mean Time t o Emergence ( B a r t l e t t ' s Test - Chi Square r 22.5 9 ) . ANALYSIS OF VARIANCE T A I L SOURCE SUM OF SQUARES DF MEAN SQUARE F VALUE PROBABIL ITY POP 1 8 4 7 . . 0 8 4 1 3 6 1 5 . 6 9 4 7 1 1 5 . 3 0 0 . 0 0 0 0 REGIME 8 5 6 4 . . 9 6 8 7 3 2 8 5 4 . 9 8 9 7 5 3 4 . 6 5 0 . 0 0 0 0 INTERACTION 3 6 2 . , 8 1 5 2 9 4 0 . 3 1 2 8 7 . 5 5 0 . 0 0 0 3 ERROR 8 5 . , 4 3 9 1 16 5 . 3 3 9 9 ANALYSI S OF VARIANCE; VARIANCES ARE NOT ASSUMED TO BE EQUAL WELCH 1 5 , 6 9 0 5 4 . 6 0 0 . 0 0 0 0 BROWN-FORSYTHE* POP 3 , 6 1 1 5 . 3 0 0 . 0 0 0 0 REGIME 3 , 6 i 5 3 4 . 6 5 0 . 0 0 0 0 INTERACTION 9 , 6 7 . 5 5 0 . 0 1 1 5 ALL GROUPS COMBINED (EXCEPT CASES WITH UNUSED VALUES FOR VARIABLES POP AND REGIME ) MEAN 1 4 3 . 1 0 0 S T D . DEV. 1 8 . 7 1 7 S . E . M. 3 . 3 0 9 MAXIMUM 1 7 3 . 4 9 0 MINIMUM 1 0 9 . 8 2 6 CASES EXCLUDED ( 0 ) CASES INCLUDED 32 *ROBUST S . D . 1 9 . 7 0 2 299 APPENDIX 2 Comparisons and contrasts among means of time to hatch and time to emergence of embryos reared in the the 1982-83 and 1983-84 experiment. Calculation of C r i t i c a l Value for SS-STP Test C. V. = (a - 1) x MSW x F. (a-1), a(n-1), alpha 300 Table ee. Comparisons using the Sum of Squares Simultaneous Test Procedure (SS-STP) ( G a b r i e l 1964) of populations w i t h i n temperature regime f o r 1982-83 time t o hatch. ( C r i t i c a l Value = 293.03, P = 0.01 or 93.40, P r 0.05 ). Temperature Regime Comparison 6 10 EARLY LATE AB - WB 542.89** 1.69 6.25 190.44* AB - W 309.76** 5.29 0.04 556.96** WB - W 32.49 12.96 7.29 96.04* WB - C 106.09* 4.41 81 1.21 W - C 21.16 32.49 136.89* 75.69 AB - (WB + W) 557.60** 0.33 1.76 466.25** W - (AB + WB) 47.20 11.60 2.80 371.85** WB - (W + AB) 280.33** 8.00 9.01 5.33 WBxW - (WB + W) 74.00 20.28 142.83* 19.25 301 Table f f . Comparisons using the Sum of Squares Simultaneous Test Procedure (SS-STP) ( G a b r i e l 1964) o f populations w i t h i n temperature regime f o r 1982-83 time t o emergence. ( C r i t i c a l Value = 152.71, P = 0.01 or 60.30, P = 0.05 )3. Temperature Regime Comparison 6 10 EARLY LATE AB - WB 655.36** 5.29 0.25 306.25** AB - W 292.41** 32.49 372.49** 676.00** WB - W 72.25 11.56 392.04** 72.25 WB - C 396.01** 2.25 2.89 53.29 W - C 129.96* 24.01 462.25** 249.64** AB - (WB + W) 607.76** 21. 117.81* 630.75** W - (AB + WB) 24.65 27.60 509.60** 396.75** WB - (AB + W) 387.60** 0.40 137.36* 27 C - (WB + W) 326.56** 13.65 179.41** 177.87** 302 Table gg. Comparisons using the Sum of Squares Simultaneous Test Procedure (SS-STP) ( G a b r i e l 1964) of populations w i t h i n temperature regime f o r 1983 -84 time t o hatch. ( C r i t i c a l Value - 301.42, P = 0.01 or 119.01, P = 0.05 ). Comparison 6 Temperature Regime 10 EARLY LATE AB - WB 70.56 100 70.56 30.25 AB - W 156.25* 18.49 196.00* 533.61** WB - W 16.81 32.49 31.36 309.76** WB - C 1.69 92.16 210.25* 136.89* W - C 29.16 234.09* 404.01** 34.81 AB - (WB + W) 145.60* 68.16 167.25* 272.65* W - (AB + WB) 91.85 0.65 128.05* 552.16** WB - (AB + W) 6.16 82.16 2.61 48.80 C - (WB + W) 14.96 206.67* 399.05** 11.21 303 Table hh. Comparisons using the Sum of Squares Simultaneous Test Procedure (SS-STP) ( G a b r i e l 1964) o f populations w i t h i n temperature regime f o r 1983-84 time t o emergence. ( C r i t i c a l Value = 267.37, P = 0.01 or 105.57, P = 0.05 ). Temperature Regime Comparison 6 10 EARLY LATE AB - WB 47.61 1.69 289** 187.69* AB - W 515.29** 116.64* 256* 734.41** WB - W 249.64* 146.41* 1 179.56* WB - C 174.24* 163.84* 0.01 51.84 W - C 6.76 0.49 1.21 38.44 AB - (WB + W) 292.05** 30.08 363** 554.88** W - (AB + WB) 494.08** 174.80* 75 546.75** WB - (AB + W) 26.40 59.85 108*' 0.03 C - (WB + W) 37.45 60.75 0.48 0.33 304 APPENDIX 3 G-tests of independence of s u r v i v a l from f e r t i l i z a t i o n to epiboly of embryos reared during 1982-83 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg s i z e . 305 Table i i . Ho: S u r v i v a l o f 1982-83 progeny from f e r t i l i z a t i o n t o e p i b o l y Independent of Temperature of Incubation. Population d.f. G TOTAL 12 1198.88 ** AB 3 437.81 ** WB 3 422.31 ** WBxW 3 166.41 ** W - 3 132.55 ** 306 Table j j . Ho: S u r v i v a l o f 1982-83 progeny from f e r t i l i z a t i o n to e p i b o l y Independent o f P o p u l a t i o n . Temperature d.f. G TOTAL 12 1189.16 ** 6 3 269.58 ** 10 3 206.18 ** Early 3 319.31 ** Late 3 394.09 ** J 307 Table kk. Ho: S u r v i v a l o f 1982-83 progeny from f e r t i l i z a t i o n to e p i b o l y Independent o f Season o f Spawning. Tempera tu re d . f . G TOTAL 4 8 5 . 4 1 ** 6 1 6 0 . 4 8 ** 10 1 1 . 1 5 E a r l y 1 9 . 9 4 ** L a t e 1 1 3 . 8 4 ** 308 Table 11. Ho: S u r v i v a l o f 1982-83 progeny from f e r t i l i z a t i o n t o e p i b o l y Independent o f Loc a t i o n o f Spawning. Temperature d.f. G TOTAL 4 846.87 ** 6 1 226.89 ** 10 1 82.13 ** Early 1 122.48 ** Late 1 415.37 ** 309 Table mm. Ho: S u r v i v a l o f 1982-83 progeny from f e r t i l i z a t i o n t o e p i b o l y Independent o f Egg S i z e . Temperature d.f. G TOTAL 4 188.64 ** 6 1 33.11 ** 10 1 36.83 ** Early 1 46.06 ** Late 1 72.64 ** 310 APPENDIX 4 G-tests of independence of s u r v i v a l from epiboly to eye pigment stage of embryos reared during 1982-83 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg s i z e . 311 Table nn. Ho: S u r v i v a l o f 1982-83 progeny from e p i b o l y to eye pigment stage Independent of Temperature o f Incubation. Population d.f. G TOTAL 12 2308.06 ** AB 3 281.69 ** WB 3 766.05 ** WBxW 3 729.90 ** W 3 530.52 ** 312 Table oo. Ho: S u r v i v a l o f 1982-83 progeny from e p i b o l y to eye pigment stage Independent o f P o p u l a t i o n . Temperature d.f. G TOTAL 12 1981.41 ** 6 3 324.64 ** 10 3 170.51 ** Early 3 147.16 ** Late 3 1339.10 ** 313 Table pp. Ho: S u r v i v a l o f 1982-83 progeny from e p i b o l y t o eye pigment stage Independent o f Season o f Spawning. Temperature d.f. G TOTAL 4 818. 09 ** 6 1 130. 66 10 1 161. 35 ** E a r l y 1 4. 81 Late 1 521. 27 ** 314 Table qq. Ho: S u r v i v a l o f 1982-83 progeny from e p i b o l y t o eye pigment stage Independent o f Loc a t i o n o f Spawning. Temperature d.f. G TOTAL 4 465.48 ** 6 1 270.72 ** 10 1 85.03 ** Early 1 99.52 ** Late 1 10.21 ** 315 Table r r . Ho: S u r v i v a l of 1982-83 progeny from e p i b o l y t o eye pigment stage Independent o f Egg S i z e . Temperature d.f. G TOTAL 4 668. 36 ** 6 1 13. 95 * * 10 1 27. 55 *-*• Early 1 56. 03 •** Late 1 570. 83 * * 316 APPENDIX 5 G-tests of independence of s u r v i v a l from eye pigment stage to hatch of embryos reared during 1982-83 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg s i z e . 317 Table s s . Ho: S u r v i v a l o f 1982-83 progeny from eye pigment stage t o hatch Independent of Temperature o f Incubation. Population d.f. G TOTAL 12 1180.92 ** AB 3 814.03 ** WB 3 175.07 ** WBxW 3 27.87 ** W 3 163.95 ** 318 Table t t . Ho: S u r v i v a l o f 1982-83 progeny from eye pigment stage to hatch Independent of P o p u l a t i o n . Temperature d.f. G TOTAL 12 2571.87 ** 6 3 832.57 ** 10 3 545.53 ** Early 3 648.86 ** Late 3 544.91 ** 319 Table uu. Ho: S u r v i v a l o f 1982-83 progeny from eye pigment stage to hatch Independent of Season o f Spawning. Temperature d.f. G TOTAL 4 806.71 *•* 6 1 416.57 10 1 3.75 Early 1 24.07 Late 1 362.32 * * 320 Table vv. Ho: S u r v i v a l o f 1982-83 Independent o f Location o f Spawning. progeny from eye pigment stage t o hatch Temperature d.f . G TOTAL 4 1233.67 ** 6 1 364.69 ** 10 1 174.63 ** Early 1 156.67 ** Late 1 537.68 ** 321 Table ww. Ho: S u r v i v a l o f 1982-83 progeny from eye pigment stage t o hatch Independent o f Egg S i z e . Temperature d.f. G TOTAL 4 414.53 ** 6 1 9.76 ** 10 1 105.53 ** E a r l y 1 277.81 ** Late 1 21.43 ** 322 A P P E N D I X 6 G-tests of independence of s u r v i v a l from f e r t i l i z a t i o n to hatch of embryos reared during 1982-83 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg s i z e . 323 Table xx. Ho: S u r v i v a l o f 1982-83 progeny Independent o f Temperature o f Incubation. from f e r t i l i z a t i o n t o hatch Population d.f. G TOTAL 12 3481. 06 -X--X-AB 3 983. 55 WB 3 1079. 10 WBxW 3 201. 34 ** W 3 1217. 07 ** 324 Table yy. Ho: S u r v i v a l o f 1982-83 progeny from f e r t i l i z a t i o n t o hatch Independent o f Po p u l a t i o n . Temperature d.f . G TOTAL 12 3824.06 ** 6 3 1182.93 ** 10 3 758.39 ** Early 3 1071.57 ** Late 3 811.17 ** 325 Table z z . Ho: S u r v i v a l o f 1982-83 progeny from f e r t i l i z a t i o n t o hatch Independent o f Season o f Spawning. Temperature d.f . G TOTAL 4 628.48 ** 6 1 522.47 ** 10 1 69.61 ** Early 1 0.0007 Late 1 36.41 ** 326 Table aaa. Ho: S u r v i v a l o f 1982-83 Independent o f Location o f Spawning. progeny from f e r t i l i z a t i o n t o hatch Temperature d . f . G TOTAL 4 2070.84 6 1 860.17 10 1 315.17 ** Early 1 382.96 ** Late 1 566.54 ** 327 Table bbb. Ho: S u r v i v a l o f 1982-83 progeny from f e r t i l i z a t i o n t o hatch Independent of Egg S i z e . Temperature d.f . G TOTAL 4 886.63 ** 6 1 11.98 * * 10 1 86.39 ** Early 1 357.16 ** Late 1 431.10 ** 328 APPENDIX 7 G-tests of independence of s u r v i v a l from hatch to emergence of embryos reared during 1982-83 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg s i z e . 329 Table ecc. Ho: S u r v i v a l o f 1982-83 progeny from hatch t o emergence Independent of Temperature of Incubation. Population d.f. G TOTAL 12 762.77 ** AB 3 21.11 ** WB 3 539.44 ** WBxW 3 83.05 ** W 3 119.17 ** 330 Table ddd. Ho: S u r v i v a l o f 1982-83 progeny from hatch t o emergence Independent of P o p u l a t i o n . Temperature d.f. G TOTAL 12 952.28 ** 6 3 676.60 ** 10 3 105.79 ** Early 3 153.44 ** Late 3 16.45 ** 331 Table eee. Ho: S u r v i v a l o f 1982-83 progeny from hatch t o emergence Independent o f Season o f Spawning. Temperature d.f. G TOTAL 4 162.45 ** 6 1 38.93 ** 10 1 97.32 ** Ear l y 1 25.48 ** Late 1 0.72 332 Table f f f . Ho: S u r v i v a l o f 1982-83 progeny from hatch to emergence Independent o f Loc a t i o n o f Spawning. Temperature d.f . G TOTAL 4 504.70 ** 6 1 428.93 ** 10 1 55.67 ** Early 1 13.42 ** Late 1 6.68 333 Table ggg. Ho: S u r v i v a l o f 1982-83 progeny from hatch to emergence Independent o f Egg S i z e . Temperature d.f. G TOTAL 4 574.59 ** 6 1 548.31 ** 10 1 14.25 ** Ear l y 1 2.91 Late 1 9.12 ** 334 APPENDIX 8 G-tests of independence of s u r v i v a l from f e r t i l i z a t i o n to epiboly of embryos reared during 1983-84 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg s i z e . 335 Table hhh. Ho: S u r v i v a l o f 1983-84 progeny from f e r t to e p i b o l y Independent of Temperature of Incubation. Population d.f. G TOTAL 12 135.40 ** AB 3 14.66 ** WB 3 7.10 ** WBxW 3 27.52 ** W 3 86.12 ** 336 Table i i i . Ho: S u r v i v a l of 1983-84 progeny from f e r t i l i z a t i o n t o e p i b o l y Independent o f P o p u l a t i o n . Temperature d.f. G TOTAL 12 4605.16 ** 6 3 1042.47 ** 10 3 628.29 ** Early 3 975.92 ** Late 3 1958.48 ** 337 Table j j j . Ho: S u r v i v a l o f 1983-84 progeny from f e r t i l i z a t i o n to e p i b o l y . Temperature d.f. G TOTAL 4 18.39 ** 6 1 . 6.95 10 1 0.19 Early 1 4.34 Late 1 6.91 338 Table kkk. Ho: S u r v i v a l o f 1983-84 progeny from f e r t i l i z a t i o n t o e p i b o l y Independent o f Loc a t i o n o f Spawning. Temperature d.f. G TOTAL 4 128.60 ** 6 1 44.62 ** 10 1 53.23 ** Ear l y 1 29.12 ** Late 1 1.63 339 Table 111. Ho: S u r v i v a l o f 1983-84 progeny from f e r t i l i z a t i o n to e p i b o l y Independent of Egg S i z e . Temperature d.f. G TOTAL 4 27.17 ** 6 1 5.91 10 1 15.35 ** Early 1 4.08 Late 1 1.83 340 APPENDIX 9 G-tests of independence of s u r v i v a l from epiboly to eye pigment stage of embryos reared during 1983-84 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg s i z e . 341 Table mmm. Ho: S u r v i v a l o f 1983-84 progeny from e p i b o l y t o eye pigment stage Independent o f Temperature o f Incubation. Population d.f. G TOTAL 12 468.17 ** AB 3 28.12 ** WB 3 28.88 ** WBxW 3 354.04 ** W 3 57.13 ** 342 Table nnn. Ho: S u r v i v a l o f 1983-84 progeny from e p i b o l y to eye pigment stage Independent of P o p u l a t i o n . Temperature d.f. G TOTAL 12 3272.99 ** 6 3 578.70 ** 10 3 396.86 ** Ear l y 3 549.95 ** Late 3 1747.48 ** 343 Table ooo. Ho: S u r v i v a l o f 1983-84 progeny from e p i b o l y t o eye pigment stage. Temperature d.f. G TOTAL 4 70.76 ** 6 1 1.06 10 1 1.61 Ear l y 1 29.20 ** Late 1 38.89 ** 344 Table ppp. Ho: S u r v i v a l o f 1983-84 progeny from e p i b o l y t o eye pigment stage Independent o f Location o f Spawning. Temperature d.f. G TOTAL 4 33.57 ** 6 1 4.39 10 1 12.62 ** Ear l y 1 4.34 Late 1 12.22 ** 345 Table qqq. Ho: S u r v i v a l o f 1983-84 progeny from e p i b o l y t o eye pigment stage Independent of Egg S i z e . Temperature d.f. G TOTAL 4 19.53 ** 6 1 0.98 10 1 2.00 Ea r l y 1 4.15 Late 1 12.40 ** 346 APPENDIX 10 G-tests of independence of s u r v i v a l from eye pigment stage to hatch of embryos reared during 1983-84 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg s i z e . 347 Table r r r . Ho: S u r v i v a l o f 1983-84 progeny from eye pigment stage t o hatch Independent o f Temperature of Incubation. Population d.f. G TOTAL 12 446.36 ** AB 3 68.36 ** WB 3 46.53 ** WBxW 3 253.98 ** W 3 77.59 ** 348 Table s s s . Ho: S u r v i v a l o f Independent o f P o p u l a t i o n . 1983-84 progeny from eye pigment stage to hatch Temperature d.f. G TOTAL 12 4108.34 ** 6 3 1207.50 ** 10 3 1595.39 ** Early 3 315.40 ** Late 3 990.05 ** 349 Table t t t . Ho: S u r v i v a l o f 1983-84 progeny from eye pigment stage t o hatch. Temperature d.f. G TOTAL 4 42.94 ** 6 1 29.23 ** 10 1 0.45 Early 1 13.26 ** Late 1 0.00 350 Table uuu. Ho: S u r v i v a l o f 1983-84 Independent of Location o f Spawning. progeny from eye pigment stage t o hatch Temperature d.f. G TOTAL 4 25.48 ** 6 1 0.98 10 1 8.22 Early 1 16.28 ** Late 1 0.00 351 Table vvv. Ho: S u r v i v a l o f 1983-84 progeny from eye pigment stage to hatch Independent of Egg S i z e . Temperature d.f. G TOTAL 4 46.06 ** 6 1 28.80 ** 10 1 16.82 ** Ear l y 1 0.44 Late 1 0.00 352 APPENDIX 11 G-tests of independence of s u r v i v a l from f e r t i l i z a t i o n to hatch of embryos reared during 1983-84 from the e f f e c t s of temperature, population, season of spawning, location of spawning and egg s i z e . 353 Table www. Ho: S u r v i v a l o f 1983-84 progeny from f e r t i l i z a t i o n t o hatch Independent o f Temperature of Incubation. Population d.f . G TOTAL 12 538.54 ** AB 3 27.46 ** WB 3 19.50 ** WBxW 3 476.50 ** W 3 15.08 ** 354 Table xxx. Ho: S u r v i v a l o f 1983-84 progeny from f e r t i l i z a t i o n t o hatch Independent o f P o p u l a t i o n . Temperature d.f. G TOTAL 12 11666.70 ** 6 3 2963.82 ** 10 3 2379.34 ** Early 3 1730.17 ** Late 3 4593.37 ** 355 Table yyy. Ho: S u r v i v a l o f 1983-84 progeny from f e r t i l i z a t i o n t o hatch. Temperature d.f. G TOTAL 4 43.42 ** 6 1 1.44 10 1 4.71 E a r l y 1 13.31 ** Late 1 23.96 ** 356 Table z z z . Ho: S u r v i v a l o f 1983-84 progeny from f e r t i l i z a t i o n t o hatch Independent o f Lo c a t i o n o f Spawning. Temperature d.f. G TOTAL 4 26.74 ** 6 1 9.51 ** 10 1 8.04 ** E a r l y 1 2.85 Late 1 6.34 357 Table aaaa. Ho : Survival Independent of Egg Size. of 1983 -84 progeny from fertilization to hatch Temperature d.f. G TOTAL 4 26.63 ** 6 1 16.85 ** 10 1 0.22 Ear l y 1 2.80 Late 1 6.76 358 APPENDIX 12 G - t e s t s o f i n d e p e n d e n c e o f s u r v i v a l f rom h a t c h t o emergence o f embryos r e a r e d d u r i n g 1 9 8 3 - 8 4 f r o m t h e e f f e c t s o f t e m p e r a t u r e , p o p u l a t i o n , s e a s o n o f s p a w n i n g , l o c a t i o n o f s p a w n i n g and egg s i z e . 359 Table bbbb. Ho: S u r v i v a l o f 1983-84 progeny from hatch t o emergence Independent o f Temperature o f Incubation. Population d.f. G TOTAL 12 326.99 ** AB 3 114.78 ** WB 3 59.78 ** WBxW 3 96.16 ** W 3 55.27 ** 360 Table c c c c . Ho: S u r v i v a l o f 1983-84 progeny from hatch to emergence Independent o f P o p u l a t i o n . Temperature d.f. G TOTAL 12 5845.38 ** 6 3 1172.55 ** 10 3 1479.98 ** Early 3 2713.96 ** Late 3 478.89 ** 361 Table dddd. Ho: S u r v i v a l o f 1983-84 progeny from hatch to emergence. Temperature d.f. G TOTAL 4 205. 79 6 1 128. 40 10 1 27. 43 ** Early 1 42. 24 ** Late 1 7. 72 ** 362 Table eeee. Ho: S u r v i v a l o f 1983-84 progeny from hatch to emergence Independent of Loc a t i o n o f Spawning. Temperature d.f. G TOTAL 4 123.86 ** 6 1 44.36 ** 10 1 27.70 ** Ear l y 1 51.80 Late 1 0.00 363 Table f f f f . Ho: S u r v i v a l o f 1983-84 progeny from hatch t o emergence Independent of Egg S i z e . Temperature d.f. G TOTAL 4 135.07 ** 6 1 45.29 ** 10 1 82.13 ** Ear l y 1 0.72 Late 1 6.93 ** 364 Appendix 13 Comparisons and contrasts among means of vertebral counts of progeny from the 1983-84 experiment using Scheffe's method. 365 Table gggg. Comparisons of v e r t e b r a l counts o f emergent f r y using Scheffe's Method of populations w i t h i n temperature regime f o r 1983-84. ( C r i t i c a l Value = 0.444, f o r f i r s t comparisons, P = 0. three 05 ). comparisons, C r i t i c a l Value = 0.3135, f o r l a s t three Comparison 6 10 EARLY LATE AB - WB -0.70* 0.00 0.60* -0.95* AB - W -0.65* -0.25 0.30 0.95* WB - W 0.05 -0.25 0.30 1.90 AB - (WB + W) -0.90* -0.13 -0.45* 0.00 W - (AB + WB) -0.30 0.25 0.00 -1.40* WB - (AB + W) -0.38* -0.13 -0.45* 1.40* 366 APPENDIX 14 Population s i z e estimated by surv i v a l time and fish-days.. 367 Table hhhh. Estimate o f s i z e o f adult p o p u l a t i o n i n 1981 and 1982. Population Year Fish-Days Survival Time Population Size AB 1981 13523 3.26 D 4148 WB 1981 1984 3.26 D 609 W 1981 5749 8.04 D 715 WB 1982 890.5 2.96 D 301 W 1982 10570 5.94 D 1779 368 APPENDIX 15 Temperature changes (planned) in laboratory experiments. 369 Table i i i i . Planned Weekly Changes i n Temperature i n Laboratory Experiments. Temperature Treatment Week Number Constant 6 C Constant 10 C ' E a r l y Spawned' 'Late Sp 1 6 10 8 4 2 6 10 8 4 3 6 10 7 3 4 6 10 7 3 5 6 10 6 3 6 6 10 6 3 7 6 10 5 4 8 6 10 5 4 9 6 10 4 4 10 6 10 4 5 11 6 10 3 5 12 6 10 3 6 13 6 10 3 6 14 6 10 3 7 15 6 10 3 7 16 6 10 3 8 18 6 10 4 8 19 6 10 4 9 20 6 10 5 9 21 6 10 5 9 22 6 10 5 9 23 6 10 6 9 24 6 10 6 9 25 6 10 7 9 26 6 10 7 9 27 6 10 7 9 28 6 10 8 9 29 6 10 8 9 370 APPENDIX 16 Egg s i z e s and siz e of females use in laboratory experiments. 371 Table j j j j - O r b i t - h y p u r a l p l a t e length o f females and egg s i z e s used i n 1982-83 and 1983-84 experiments. Year P o p u l a t i o n Female Length Egg weight(gms) (cm) Mean S t a n d a r d D e v i a t i o n 1982 AB 1 56.0 0.3227 0.0143 2 55.5 0.3785 0.0236 3 52.7 0.3148 0.0161 4 52.8 0.2555 0.0131 5 52.8 0.2671 0.0070 WB 1 59.2 0.3223 0.0137 2 59.3 0.2973 0.0125 3 57.0 0.2952 0.0161 4 51.0 0.2085 0.0135 5 52.0 0.2623 0.0067 W 1 55.0 0.3359 0.0210 2 55.0 0.3238 0.0211 3 55.0 0.3002 0.0137 4 55.0 0.2820 0.0129 5 55.0 0.3262 0.0133 1983 AB 1 60.0 0.3180 0.0176 2 50.0 0.2458 0.0115 3 49.0 0.2459 0.0113 4 52.5 0.2755 0.0080 WB 1 52.0 0.2551 0.0143 2 49.5 0.2274 0.0198 3 51.5 0.2421 0.0155 4 53.0 0.2710 0.0098 5 51.5 0.2465 0.0123 W 1 54.0 0.3060 0.0140 2 55.5 0.3067 0.0157 3 52.5 0.2691 0.0097 4 49.5 0.2542 0.0106 5 59.0 0.3519 0.0168 3 7 2 APPENDIX 17 Water temperatures in r e l a t i o n to month and l a t i t u d e i n streams u t i l i z e d chum salmon stocks. 373 Table kkkk. L a t i t u d e s and monthly mean water temperatures from August to De cember i n 28 streams u t i l i z e d by chum salmon s t o c k s . Mean Temperature in C e l s i u s Location Latitude August September October November December Pallant Creek, B.C. 53.00 N 17.0 Nass River, B.C. 55.00 N 9.8 Skeena River, B.C. 54.50 N 14.3 Kemano River, B.C. 53.50 N 9.3 Kitimat River, B.C. 54.00 N 10.0 Be l l a Coola, B.C. 52.50 N 10.0 Dean River, B.C. 53.00 N 12.0 Wannock River, B.C. 51.50 N 13.7 Homathko River, B.C. 51.00 N 8.4 P h i l l i p s River, B.C. 50.50 N 13.6 Salmon River, B.C. 50.50 N 14.3 L i t t l e Qualicum 49.25 N 17.8 River, B.C. Puntledge River, 49.25 N 19.0 B.C. Big Qualicum 49.25 N 15.9 River, B.C. Tsable River, B.C. 50.00 N 15.6 Theodosia River, 50.00 N 15.0 B.C. Chemainus River, 49.00 N 18.0 B.C. Nanaimo River, B.C. 49.00 N 17.2 Cowichan River, B.C. 49.00 N 19.2 Koksilah River, B.C. 48.75 N 17.4 Goldstream River, 48.50 N 10.8 B.C. Sar i t a River, B.C. 49.00 N 16.8 Zebellos River, B.C. 50.00 N 13.5 Cheakumus River, 49.50 N 11.4 B.C. Sqaumish River, B.C. 49.50 N 10.0 4.3 Harrison River, B.C. 49.25 N 15.4 Yukon River, 65.00 N 12.9 Yukon Terr. 12.5 8.0 7.0 6.0 9.3 4.0 2.6 1.2 8.0 2.5 1.0 0.5 6.8 6.3 5.0 1.0 9.7 6.6 1.5 1.8 7.4 5.3 3.5 2.0 9.8 5.8 0.8 0.2 11.4 10.3 7.6 5.0 7.7 6.7 2.5 9.7 4.9 2.8 1.3 12.5 8.9 5.8 4.7 14.1 12.5 7.6 12.3 10.3 9.0 4.5 12.8 12.4 8.5 4.7 12.0 8.6 6.2 2.5 12.3 9.1 6.9 4.1 14.3 10.0 6.0 4.1 12.8 10.5 7.3 4.4 13.6 11.9 8.4 5.2 14.4 9.1 5.4 3.4 9.4 5.1 1.7 0.3 13.0 11.5 8.0 6.3 10.7 7.8 6.2 3.6 9.5 9.0 5.9 3.8 9.4 7.1 3.1 2.8 14.5 11.4 8.1 5.9 8.6 3.2 3.5 0.0 3 7 4 Regressions o f mean water temperature p r e d i c t e d by l a t i t u d e f o r each month. Point estimates o f mean water temperature at 49 and 65 degrees l a t i t u d e . Month Regression Equation 4 9 6 5 August Temperature = 3 1 . 3 - 0 . 3 3 9 (r-squared - 1 2 . 7 %) L a t i t i d e 1 4 . 7 0 9 . 2 7 September Temperature r 2 9 . 9 - 0 . 3 6 8 (r-squared = 2 8 . 3 %) L a t i t i d e 1 1 . 8 7 5 . 9 8 October Temperature = 3 6 . 6 - 0 . 5 5 7 (r-squared = 4 3 . 7 8 ) L a t i t i d e 9 . 3 1 0 . 4 0 November Temperature = 2 1 . 3 - 0 . 3 1 5 (r-squared = 1 7 . 0 %) L a t i t i d e 5 . 8 7 0 . 8 3 December Temperature = 1 9 . 8 - 0 . 3 2 2 (r-squared = 3 1 . 7 55) L a t i t i d e 4 . 0 2 - 1 . 1 3 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0098366/manifest

Comment

Related Items