UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Heritabilities and genetic correlations for weight, length and survivability in fresh water and salt… Swift, Bruce D. 1991

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

Item Metadata

Download

Media
831-UBC_1991_A6_7 S94.pdf [ 2.53MB ]
Metadata
JSON: 831-1.0098725.json
JSON-LD: 831-1.0098725-ld.json
RDF/XML (Pretty): 831-1.0098725-rdf.xml
RDF/JSON: 831-1.0098725-rdf.json
Turtle: 831-1.0098725-turtle.txt
N-Triples: 831-1.0098725-rdf-ntriples.txt
Original Record: 831-1.0098725-source.json
Full Text
831-1.0098725-fulltext.txt
Citation
831-1.0098725.ris

Full Text

HERITABILITIES AND GENETIC CORRELATIONS FOR WEIGHT, LENGTH AND SURVIVABILITY IN FRESH WATER AND SALT WATER OF SO AND SI COHO SALMON, (Oncorhynchus kisutch). B.Sc, The University of Guelph, 1979 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF MASTER OF SCIENCE (Department of Animal Science) We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA by BRUCE D. SWIFT A p r i l 1991 Swift, 1991 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. Department of /IAIVW-AX Su^tntc The University of British Columbia Vancouver, Canada Date A DE-6 (2/88) i i ABSTRACT H e r i t a b i l i t i e s and Genetic Correlations For Weight, Length and S u r v i v a b i l i t y i n Freshwater and Saltwater of SO and SI Coho Salmon, (Oncorhynchus kisutch) Bruce Swift, M.Sc, Supervisor: Dr. R. Peterson University of B r i t i s h Columbia, 1991 H e r i t a b i l t i e s and genetic correlations using Individual Animal Model, (IAM), analysis for weight, length and s u r v i v a b i l i t y were determined i n both freshwater and s a l t -water rearing of sO and s i coho, (Oncorhynchus kisutch). The most important t r a i t to salmon farmers i s the saltwater market weight which was found to have h e r i t a b i l i t y estimates of 0.21 and 0.45 for the s i and sO smolts. Estimates of h e r i t a b i l i t y for s u r v i v a l were high, (0.29 and 0.21) for both sO and s i rearing strategies. Genetic correlations between freshwater and saltwater size were small, (0.33 to 0.59) i n d i c a t i n g s e l e c t i o n for market weight should be done during the saltwater market weight window. S i g n i f i c a n t differences were found for weight and length between stra i n s during freshwater and saltwater rearing for both sO and s i coho. The northern stra i n s , (Kitimat, Bella Coola and Pallant Creek) were s i g n i f i c a n t l y larger in weight and length than i i i the southern s t r a i n s , (Big Qualicum and Robertson Creek). The Kitimat s t r a i n was larger i n weight and length than the Bella Coola and Pallant Creek str a i n s but lower than these two s t r a i n s for percent s u r v i v a b i l i t y . When comparing f i s h from the Kitimat s t r a i n reared on both a sO and s i rearing strategy, i t was found that the s i f i s h were s i g n i f i c a n t l y larger than the sO f i s h at the market weight window. i v Table of Contents ABSTRACT i i LIST OF TABLES v i ACKNOWLEDGEMENTS v i i i INTRODUCTION 1 LITERATURE REVIEW 3 Strain Comparison 3 Estimates of Genetic Parameters 4 METHODS AND MATERIALS 9 Data C o l l e c t i o n 9 Data Analysis 15 RESULTS and DISCUSSION 20 85 Coho Yearclass - SI Rearing Strategy 20 Strain Comparison 20 Freshwater Rearing Period 2 0 Saltwater Rearing Period 24 H e r i t a b i l i t y Estimates for Weight and Length 25 Correlation Estimates for Weight and Length 27 S u r v i v a b i l i t y 29 H e r i t a b i l i t y Estimates for S u r v i v a b i l i t y 30 V 86 Coho Yearclass - SO Rearing Strategy 32 Strai n Comparison 32 Freshwater Rearing Period 32 Saltwater Rearing Period 3 3 H e r i t a b i l i t y Estimates for Weight and Length 35 Correlation Estimates for Weight and Length 36 S u r v i v a b i l i t y 39 H e r i t a b i l i t y Estimates for S u r v i v a b i l i t y 40 CONCLUSION 41 Strain Comparison 41 H e r i t a b i l i t y Estimates For Weight and Length 43 S u r v i v a b i l i t y 44 Combining S u r v i v a b i l i t y and Growth 46 BIBLIOGRAPHY 48 v i L i s t of Tables Table 1. Summary of egg c o l l e c t i o n , ponding and saltwater entry dates for each s t r a i n of the 1985 and 1986 yearclass coho 10 Table 2. Summary of number of males and females for each s t r a i n of the 1985 and 1986 yearclass coho 11 Table 3. Least square means for weight and length for each s t r a i n of the 1985 yearclass coho for the freshwater measurements 22 Table 4. Least square means for weight and length of the 1985 yearclass coho for the saltwater measurements... 23 Table 5. H e r i t a b i l i t y estimates for weight and length of the 1985 yearclass coho derived from DFREML 24 Table 6. Estimates of phenotypic and genetic correlations for weight from the 1985 yearclass coho 27 Table 7. Estimates of phenotypic and genetic correlations for length from the 1985 yearclass coho 27 Table 8. Estimates of genetic and phenotypic correlations for weight and length of the 1985 yearclass coho 28 Table 9. Percent s u r v i v a b i l i t y from the IMPLANT measurement for a l l strains of the 1985 yearclass coho 28 Table 10. H e r i t a b i l i t y estimates for s u r v i v a b i l i t y using DFREML analysis of the 1985 yearclass coho 29 Table 11. Least square means for weight and length at the IMPLANT measurement of the 1986 yearclass coho 31 Table 12. Least squares means for weight and length at the 7 month and 15 month saltwater measurement for the 1986 yearclass coho 33 Table 13 H e r i t a b i l i t y estimates for weight and length for the 1986 yearclass coho using DFREML analysis 34 v i i Table 14 Estimates of phenotypic and genetic correlations for weight for the 1986 yearclass coho 36 Table 15 Estimates of phenotypic and genetic correlations for length for the 1986 yearclass coho 37 Table 16 Estimates of genotypic and phenotypic correlations for weight and length of the 1986 yearclass coho 37 Table 17 Percent s u r v i v a b i l i t y for a l l s t r a i n s of the 1986 yearclass coho 38 Table 18 H e r i t a b i l i t y estimates for survival for the 1986 yearclass coho 39 Table 19 Summary of the percentage of f i s h for each s t r a i n of the 1985 and 1986 yearclass coho that f a l l into the fresh farm f i s h market categories 41 Table 2 0 S t r a i n comparisons combining s u r v i v a l and weight at the "market window",(15 and 16 month saltwater measurement), for the 1985 and 1986 yearclass coho...46 v i i i ACKOWLEDGEMENTS The author would l i k e to express appreciation to h i s supervisor, Dr. R. Peterson for his guidance, assistance and patience during the research. Thanks are also extended to Jane Charles, Jim McGeer and Anne Winkleman for t h e i r assistance throughout the research. Thanks to Tom and Linda May, the s t a f f at Royal P a c i f i c Sea Farms, Ruth Withler and the D.F.O. s t a f f of the Kitimat, Bella Coola, Pallant Creek, Big Qualicum and Robertson Creek hatcheries for t h e i r f i s h c u l t u r e expertise and assistance. Financial assistance for the f i r s t 8 months of t h i s project from B.C. Science council was also appreciated. Special thanks to my wife and family of whom the completion of t h i s project would not have been possible without t h e i r understanding and support. 1 I n t r o d u c t i o n In 1985, salmon farming i n B r i t i s h Columbia experienced a period of rapid expansion and development. Coho salmon, (Oncorhynchus kisutch), was one of the P a c i f i c salmon species used i n the industry at the time and preferred by some salmon farm operators for t h e i r desirable carcass c h a r a c t e r i s t i c s , resistance to saltwater v i b r i o s i s and short generation i n t e r v a l r e l a t i v e to the other species, (Novotny,1975). Based on the success of the Norwegian farmed A t l a n t i c salmon, (Salmo s a l a r ) , broodstock program, (Gjedrem, 1983; Refstie, 1990), the development of a genetic broodstock program for B.C. farmed coho with industry cooperation, (Royal P a c i f i c Sea Farms Ltd), was i n i t i a t e d . Eggs and mi l t were c o l l e c t e d from d i f f e r e n t s t r a i n s of coho salmon at the Deptartment of Fisheries and Oceans enhancement hatcheries throughout B.C. and reared under commercial freshwater and saltwater salmon farming conditions. Two of the str a i n s evaluated, (Big Qualicum and Robertson Creek), were already being used as production sources of coho eggs by the B.C. salmon farmers; however the three northern s t r a i n s , (Kitimat, B e l l a Coola and Pallant Creek), had not been used i n the industry. 2 The objectives of t h i s study were to determine i f genetic differences i n growth and s u r v i v a b i l i t y existed between coho stra i n s grown under commercial conditions and to estimate genetic parameters for these t r a i t s for use i n a genetic s e l e c t i o n program for farm-reared coho. The f i s h from t h i s study would be selected at maturity for the creation of a improved synthetic coho s t r a i n to be used i n the B.C. salmon farming industry. 3 Review of Literature S t r a i n Comparison Evaluating important t r a i t s such as growth and su r v i v a l between s t r a i n s grown under production conditions i s a fundamental component of broodstock development practiced i n many f i s h farm industries, (Wilkins, 1981; Kinghorn, 1983, Refstie, 1990). In Norway, approximately 40 wild st r a i n s of A t l a n t i c salmon were evaluated for use as broodstock i n the Norwegian salmon farming industry. From t h i s study, Refstie et a l , (1977); Refstie and Steine, (1978) and Gunnes and Gjedrem, (1978) found s i g n i f i c a n t differences between s t r a i n s for s m o l t i f i c a t i o n and growth i n both the freshwater and saltwater rearing stages. Gjedrem and Aulstad, (1974) have reported s t r a i n differences with respect to resistance to v i b r i o s i s . In B r i t i s h Columbia, the salmon farming industry i s r e l a t i v e l y new compared to Norway, r e s u l t i n g i n l i t t l e information available on broodstock development. Withler et a l , (1987) found s i g n i f i c a n t v a r i a t i o n for freshwater growth and su r v i v a l between six strains of chinook salmon, (Oncorhynchus tshawytscha). These six str a i n s were further 4 evaluated i n s a l t water under f i s h farm conditions but s t r a i n differences have not been conclusive due to confounding e f f e c t s from nongenetic sources, (Krieberg,H., Dept. of Fisheries and Oceans, pers comm, 1991). Cheng et a l , (1987), found s i g n i f i c a n t differences between st r a i n s of chinook for growth during freshwater and saltwater rearing but only s i g n i f i c a n t differences during the freshwater period for s u r v i v a l . At the time t h i s study was i n i t i a t e d no information was available regarding differences i n weight and length or s u r v i v a b i l i t y between strains of B.C. coho salmon, (Oncorhynchus kisutch), grown under f i s h farm conditions. Since the inception of t h i s study, Withler, (1990), has undertaken a s i m i l a r study and found s i g n i f i c a n t s t r a i n differences for weight, length and s u r v i v a b i l i t y between three st r a i n s of net cage reared coho. Estimates of Genetic Parameters When establishing a s e l e c t i v e breeding program the estimation of genetic parameters such as h e r i t a b i l i t y i s fundamental i n determining the response to se l e c t i o n . H e r i t a b i l i t y i n the narrow sense i s defined as the r a t i o 5 of additive genetic variance to the phenotypic variance. The l e v e l of h e r i t a b i l i t y for a t r a i t i s of primary importance i n determining what selection method should be used for salmonids, (Falconer, 1981; VanVleck 1987 ; Gjedrem, 1983). Estimates of h e r i t a b i l i t y for farmed salmonids are well reviewed by Kinghorn,(1983), Gjedrem,(1983) and Tave (1986). Gjedrem, (1983) reports averaged h e r i t a b i l i t y estimates for body weight and length for A t l a n t i c salmon during the freshwater juvenile period of 0.08 - 0.14 and of 0.36 - 0.41 for the saltwater period. Gjerde and Gjedrem, (1984), found h e r i t a b i l i t i e s for saltwater weight varied between 0.38 - 0.44 and 0.19 - 0.32, and length between 0.3 3 - 0.3 5 and 0.16 - 0.2 6 i n cage reared A t l a n t i c salmon and rainbow trout respectively. Unfortunately published estimates of genetic parameters for P a c i f i c salmon reared under farm conditions are l i m i t e d . H e r i t a b i l i t y estimates of freshwater weight, (0.25 to 0.66) and eight month saltwater weight and length, (0.19 to 0.33 and 0.17 to 0.3 0 respectively), for pan s i z e sO coho over ten years of s e l e c t i o n have been reported by Hershberger et a l . , 6 (1990). Withler (1990), reports higher h e r i t a b i l i t y estimates , (0.55), for eight month saltwater weight of PIT (Passive Integrated Transponders), tagged sO coho. For s u r v i v a b i l i t y , estimated h e r i t a b i l i t i e s are generally low, and as indicated i n Gjedrem's review. However Withler, (1990), estimated h e r i t a b i l i t y values for Ba c t e r i a l Kidney Disease resistance i n sO coho ranging from 0.26 to 0.53. Estimation of genetic correlations are also important salmon broodstock programs for determining the effectiveness of i n d i r e c t s e l e c t i o n . As reported by Refstie and Steine, (1978); Gunnes and Gjedrem,(1978); Refstie,(1980) and Gunnes and Gjedrem,(1981) the genetic c o r r e l a t i o n between weight and length for A t l a n t i c salmon and rainbow trout i s close to unity. As length i s generally more her i t a b l e than weight, (Kinghorn,1982), the correlated response i n weight due to s e l e c t i o n on length i s expected to be greater than d i r e c t response to selection on weight. Estimation of genetic correlations for t r a i t s measured at d i f f e r e n t time periods throughout the production cycle w i l l also determine the most e f f i c i e n t period to select for the desired t r a i t . Most of these reported estimates of genetic parameters, 7 (Hershberger,1990; Rye et al.,1990; Standal and Gjerde,1987; Gjerde and Gjedrem, (1984); Refstie,1979; Gunnes and Gjedrem, 1978), have been derived from the analysis of measurements of f i s h i d e n t i f i e d to the family l e v e l by use of f i n c l i p s or freeze branding techniques, (Gunnes and Refstie, 1980), allowing variance component estimation to be derived using s i r e and dam models and least square estimates. As genetic s e l e c t i o n programs become established i n the salmon farming industry, modern analysis techniques presently used i n the dairy and swine industries w i l l be used to estimate variance components with increased accuracy. As an example, Gjerde and Schaeffer, (1989), estimated variance components for carcass t r a i t s of sea-cage reared Rainbow trout, (Oncorhynchus  mykiss), i n Norway using the i t e r a t i v e Minimum Variance Quadratic Unbiased Estimation (MIVQUE) based on family means. To allow accurate i d e n t i f i c a t i o n of accelerated coho and chinook smolts at a saltwater entry s i z e of 8 g and also to achieve individual f i s h i d e n t i f i c a t i o n , the use of Passive Integrated Transponders, (PIT tags), have recently been adopted from fishery assessment researchers for use i n broodstock development, (Prentice et a l . , 1987). Individual f i s h i d e n t i f i c a t i o n PIT tags, now allows variance 8 components to be estimated using the Individual Animal Model, (IAM). Meyers, (1988), describes variance component estimation using IAM analysis through the Derivative Free Restricted Maximum Likelihood, (DFREML) algorithm. The advantage of the IAM i s that i t f i t s an additive-genetic e f f e c t for a l l animals included i n the analysis, (Henderson, 1984). The IAM i s used extensively for variance component estimation in other domestic selection programs, (eg dairy industry), but has not been applied to the aquaculture industry u n t i l t h i s study. In summary, the present review of l i t e r a t u r e indicates that s u f f i c i e n t genetic v a r i a t i o n exists between and within s t r a i n of other salmonid species, (primarily A t l a n t i c salmon and rainbow t r o u t ) , used i n the f i s h farming industry and that s t r a i n evaluation and accurate estimation of genetic parameters would be a b e n e f i c i a l step i n the i n i t i a t i o n of a s e l e c t i v e breeding program for B.C. farm coho. 9 Materials and Methods Data C o l l e c t i o n In t h i s study f i s h a r i s i n g from gametes c o l l e c t e d i n 1985 are c a l l e d 1985 yearclass, ('85YC) and those from 1986 are referred to as 1986 yearclass, ('86YC). Differences i n the methods between the two yearclasses w i l l be discussed where appropriate. Gametes from mature broodstock for both the '85YC and '86YC were co l l e c t e d at the r i v e r source of which they originated. Each r i v e r system i n t h i s study was assumed to consist of a discrete breeding population which w i l l be defined as a s t r a i n , (Refstie and Steine, 1978). C r i t e r i a for s t r a i n s e l e c t i o n was based on geographic d i v e r s i t y and the a v a i l a b i l i t y of t h i r t y mature female and ten mature male coho, (Oncorhynchus kisutch), to be c o l l e c t e d on a single day. Eggs and mi l t were c o l l e c t e d from the Kitimat, Big Qualicum and Bella Coola strains i n 1985 and from the Kitimat, Pallant Creek and Robertson Creek s t r a i n s i n 1986. A l l gametes were co l l e c t e d from broodstock being retained at the Department of Fisheries and Oceans Enhancement Hatcheries which were located on each of the f i v e r i v e r systems. The time between the egg c o l l e c t i o n s for the 1 85YC 10 was considerably longer than the '86YC, (Table 1). Th i r t y mature females and ten mature males were randomly selected from a pooled group of ripe adults at the Kitimat, Big Qualicum and Robertson Creek hatcheries. The ten males and t h i r t y females used at the Bella Coola and Pallant Creek hatcheries were the only adult f i s h available for gamete c o l l e c t i o n at the time. Approximately 1000 eggs per female and 2 to 5 ml of m i l t were obtained per male during the spawning process. A l l c o l l e c t e d gametes from individual f i s h were kept separate and transported back to the Chapman Creek Hatchery, (Royal P a c i f i c Sea Farms Ltd, Sechelt, B.C.) for f e r t i l i z a t i o n . Table 1. Summary of egg c o l l e c t i o n , ponding and saltwater entry dates for each s t r a i n of the 1985 and 1986 coho. 1985 Yearclass, ('85YC) River Kitimat Big Qualicum Be l l a Coola Egg Take Nov.19/85 Dec.13/85 Dec.24/85 Pond Date Feb.28/86 Mar.01/86 Mar.10/86 May 28/87 May 28/87 May 28/87 Saltwater Entry 1986 Yearclass, ('86YC) River Kitimat Pallant Creek Robertson Creek Egg Take Oct.31/86 Nov.10/86 Nov.12/86 Pond Date Feb.02/87 Feb.08/87 Feb.09/87 June 16/87 June 16/87 June 16/87 Saltwater Entry 11 Weight and length measurements on a l l adult f i s h were co l l e c t e d a f t e r spawning. Tissue samples were also obtained from each adult used i n the study for b a c t e r i a l and v i r a l screening, (Gordon et a l . , 1987). Eggs from three randomly selected females were f e r t i l i z e d using the m i l t from one randomly selected male within the same s t r a i n . This nested hierarchal breeding design resulted in creating three i n d i v i d u a l h a l f s i b families per s i r e and 30 f u l l s i b families for each s t r a i n , (Falconer, 1981; Becker, 1984) p r i o r to disease screening. Table 2 summarizes the f i n a l number of males and females per s t r a i n a f t e r disease screening t e s t s . Table 2. Summary of the number of males and females for each s t r a i n for the 1985 and 1986 yearclass coho. Strain '85YC Males Females Kitimat 8 25 Big Qualicum 10 29 Bell a Coola 8 24 • 86YC Males Females Kitimat 9 30 Robertson Creek 10 29 Pallant Creek 10 29 Eggs from each f u l l - s i b family were hatched i n separate incubation compartments. Temperature was manipulated during 12 the incubation period i n an attempt to synchronize the ponding dates of the three st r a i n s i n each yearclass, Table 1. At ponding, fry from each f u l l - s i b family were randomly assigned into i n d i v i d u a l 40 L fiberglass tanks. The tanks were outdoors and divided into 3 banks with 3 0 tanks per bank. Families from one s t r a i n were ponded into one bank of tanks. Water source and temperature were the same i n a l l tanks for a yearclass but d i f f e r e d between the yearclasses. The water source for the '85YC was primarily from Chapman Creek and ranged i n water temperature from 4 to 18 °C. The f i s h from t h i s yearclass remained i n freshwater for a t o t a l of 18 months and w i l l be referred to as s i smolts, (Clarke and Shelbourne, 1988). Water temperature for the '86YC was increased, (8 to 13 C), using well and heated creek water for the production of sO smolts, (saltwater entry a f t e r 8 months of freshwater rearing), (Donaldson and Brannon, 1976; Clarke, 1986, pers. comm,; Clarke, 1989). A l l tanks were placed under an 8 h a r t i f i c i a l photoperiod regime during the f i r s t 8 weeks a f t e r ponding, to a s s i s t i n the uniform development of the sO smolts,(Clarke, 1989). Project f i s h were hand-fed 6 times d a i l y to s a t i a t i o n on a commercially produced semi-moist feed and reared i n the family tanks u n t i l a 4 g average si z e was reached by the entire yearclass. 13 I d e n t i f i c a t i o n of f i s h to family within s t r a i n was accomplished by rearing each family i n i s o l a t i o n u n t i l f i s h were large enough to i d e n t i f y i n d i v i d u a l f i s h using Passive Integrated Transponders, (PIT tags). When the average weight for the yearclass reached 4 g, randomly selected f i s h from each family tank were anaesthetized and a PIT Tag was implanted by the method described by Prentice et a l , (1987). A l l f i s h implanted with PIT tags were externally marked by removal of the adipose f i n . Immediately following PIT tag implant, the f i s h were weighed and measured for length. A l l PIT tagged f i s h from each family and s t r a i n were then "pooled" into a 5000 L tank for the remainder of the freshwater rearing period. Equal numbers of unmarked f i s h from each family were also added to the pooled PIT Tag group to increase the number of f i s h to production densities, (8 kg/m V at harvest). Approximately 9 f i s h from each family i n the '85YC were implanted with PIT tags r e s u l t i n g i n a t o t a l of 732 i d e n t i f i e d f i s h . Approximately 20 PIT tags were implanted per family for the '86YC r e s u l t i n g i n a t o t a l of 1810 i d e n t i f i e d f i s h . Measurements taken when the PIT tags were implanted w i l l be referred to as the IMPLANT measurement throughout the 14 resu l t s and discussion sections for both the *85YC and *86YC. The '85YC pooled group remained i n fresh water at the Chapman Creek Hatchery u n t i l May, 1987. This 18 month freshwater rearing period w i l l be referred to as the s i rearing strategy. A l l PIT tagged f i s h from t h i s group were i d e n t i f i e d , weighed and measured twice during t h i s time period; the f i r s t i n Jan 1987, (14 month freshwater measurement, FW14mo), and the second i n May, 1987, 18 month freshwater measurement, FW18mo), just p r i o r to saltwater entry. The '85YC, were transferred to s a l t water on May 28, 1987, ( t a b l e l ) . The f i s h from the '86YC, were transferred to s a l t water on June 16, 1987, (table 1), approximately two weeks aft e r implanted with PIT tags. This 8 month freshwater period w i l l be referred to as the sO rearing strategy. Both the '85YC and '86YC groups were transferred to the cooperator's saltwater net-pen farm located i n the Sechelt area. Each yearclass group was transferred into a separate 1700 m3 net-cage and reared under standard production management conditions. During the saltwater period, a l l PIT tagged f i s h from both the '85YC and '86YC groups were i d e n t i f i e d , weighed and measured for length on two d i f f e r e n t occasions. The f i r s t 15 s a l t water measurement was i n January, 1988. This measurement resulted i n 8 months of rearing i n s a l t water for the s i smolts from the '85YC ,(SW8mo), and 7 months of saltwater rearing for the sO smolts from the 186YC,(SW7mo). The second saltwater measurement took place i n September, 1988. This date coincides with the normal market window for no n s t e r i l i z e d coho which occurs before the i n i t i a t i o n of phenotypic changes caused by sexual maturation. This measurement w i l l be referred to as the 16 month saltwater measurement, (SW16mo) for the s i smolts from the *85YC and the 15 month saltwater measurement, (SW15mo), for the sO smolts from the '86YC. In both saltwater measurements for the '85YC and *86YC, a l l f i s h were dip netted from the s a l t water pens, anaesthitized and screened for an adipose c l i p , (physical marker for PIT tag implant). The tagged f i s h were i d e n t i f i e d , weighed and measured for length and then returned to the pen. Non PIT tagged f i s h , (no adipose c l i p ) , were returned to the pen without measurement. Data Analysis A mixed model, (Model 1), including a fixed e f f e c t for s t r a i n and random e f f e c t s for male within s t r a i n and female within male and error was used to te s t for differences between stra i n s for weight and length at each measurement. 16 The '85YC and '86YC were analyzed separately. MODEL 1 Y i j k l = u + a i + s i j + d i j k + e i j k l Where Y i j k l = weight or length of the 1th f i s h of the kth dam mated to the j t h s i r e nested within the i t h s t r a i n . u = o v e r a l l mean a i = fixed e f f e c t of the i t h s t r a i n ; s i j = random e f f e c t of the j t h s i r e within the i t h s t r a i n ; d i j k = random e f f e c t of the kth dam within the j t h s i r e within the i t h s t r a i n ; e i j k l = the residual error e f f e c t , (o,n). Least Square means for s t r a i n at each measurement were calculated using GLM routines i n PC/SAS, (SAS,1985), under Model 1. Strain means were compared using the Student-Neuman-Keuls multiple range te s t i n PC/SAS, (SAS,1985), for Model 1. Variance components were estimated for both t r a i t s using a Derivative Free Restricted Estimate of Maximum Likelihood, (DFREML), analysis technique, (Meyers, 1988). Henderson's Method 3 estimates of variance components, (Henderson, 1984), were used to derive s t a r t i n g h e r i t a b i l i t y values, ( p r i o r s ) , for the i t e r a t i v e DFREML analysis. A l l solutions converged within 22 rounds of i t e r a t i o n based on the variance r a t i o 17 remaining i d e n t i c a l to 9 decimal places. F i n a l h e r i t a b i l i t y estimates were calculated by the DFREML program. The REML algorithm u t i l i z e d a univariate estimation of variance components and as follows; Model 2 y = Xb + Za + e y = a vector of observations of weight or length for a p a r t i c u l a r measurement b = a vector of fixed e f f e c t s , ( s t r a i n ) a = a vector of additive genetic values of ind i v i d u a l animals* e = i s a vector of residual e f f e c t s X and Y are known incidence matrices for fixed and random ef f e c t s *Inclusion of the additive genetic rel a t i o n s h i p between animals i s done by adding the inverse of the rel a t i o n s h i p matrix to the random block of the mixed model equation.The inverse of "A - 1" was formed d i r e c t l y using rules outlined by Henderson, (1984). Standard error of h e r i t a b i l i t i e s were approximated according to Becker, (1984). Such standard error approximations should be regarded as overestimates since DFREML estimates of variance components generally have lower standard errors than Least Squares estimates, (Gjerde and Schaeffer,1988). 18 Genetic, phenotypic and environmental covariances were estimated between d i f f e r e n t measurements for each yearclass using variance components derived from univariant DFREML analyses, Model 2, (Meyers, 1988) and using the i d e n t i t y ; Cov(X,Y) = [Var(X+Y) - Var(X) - Var(Y)]/2 . Simple product moment correlations ,(Van Vleck,1987), were used to estimate correlations between weight and length. Genetic, phenotypic and environmental correlations for weight or length between measurements were obtained from the appropriate variances and covariance estimates using equations given by Becker, (1984). Observations for survival for each sampling period were coded 0 for animals not found, (assumed dead), and 1 for a l l animals that were measured. Variance component and h e r i t a b i l i t y estimates for the binomial s u r v i v a b i l i t y data was analyzed with DFREML, (Meyers, 1988), using Model 2. These h e r i t a b i l i t y estimates were transformed to the underlying l i a b i l i t y scale, (h 2 x), as described by Dempster and Lerner,(1950). This transformation assumes that s u r v i v a l i s determined by an underlying l i a b i l i t y scale that i s normally 19 di s t r i b u t e d and inherited i n a polygenic manner. Survival percentages were calculated as the number of individu a l s with records at the end of a period, r e l a t i v e to the number of records at the IMPLANT period. Significance between s t r a i n for percent survival was tested using chi square analysis, (Malik and Mullen, 1973). 20 Results Results for the 1985 and 1986 yearclass, ('85YC and '86YC), w i l l be treated separately due to t h e i r d i f f e r e n t rearing strategies, ( s i vs sO). Comparisons between the two yearclasses w i l l be made where appropriate. 85 COHO YEARCLASS - SI REARING STRATEGY Strai n Comparison Freshwater Rearing Period Table 3 summarizes the least square means and standard errors using Model 1 for the three freshwater, measurements of the '85YC coho. A s i g n i f i c a n t difference between a l l 3 str a i n s for weight and length was found at the IMPLANT and a f t e r 14 and 18 months of freshwater rearing, (FW14mo and FW18mo). Strai n differences for weight and length at IMPLANT are confounded with maternal and tank e f f e c t s , Refstie and Steine, 1978) and as such are not r e l i a b l e estimates of genetic differences between s t r a i n s . The Kitimat s t r a i n , which was spawned and ponded the e a r l i e s t , was found to be s i g n i f i c a n t l y larger than the Big Qualicum or Bella Coola str a i n s at each freshwater measurement. The Be l l a Coola s t r a i n was s i g n i f i c a n t l y smaller than both the other s t r a i n s at IMPLANT was not s i g n i f i c a n t l y 21 d i f f e r e n t from the Big Qualicum s t r a i n at FW14mo, but was s i g n i f i c a n t l y larger than the Big Qualicum s t r a i n at the FW18mo measurement. Two d i s t i n c t groups of f i s h were present at the FW14mo and FW18mo measurements based on the bimodal frequency d i s t r i b u t i o n for the '85YC. The f i r s t group was comprised of 582 f i s h from a l l three st r a i n s which had an average weight of 27.4 g and length of 118.7 mm at FW14mo and 555 f i s h which had an average of 39.9 g and length of 153.8 mm at the FW18mo measurement. The second group was comprised of 60 f i s h from a l l 3 stra i n s with an average weight of 129.46 g and length of 210.18 mm at the FW14mo and 3 9 f i s h with an average weight of 174.7 g and length of 248.1 mm at the FW18mo measurement, (refer to table 3) . Although no external evaluation was made on the state of s m o l t i f i c a t i o n of the f i s h , i t i s speculated that t h i s difference between the two groups was attributed to d i f f e r i n g stages of s m o l t i f i c a t i o n , (Gorbman et al.,1982 and Folmar et a l . , 1982). The smaller f i s h of the f i r s t group were categorized as "sl"smolts i n d i c a t i n g that they remained as parr throughout the freshwater period. The s i g n i f i c a n t l y larger f i s h were catagorized as s"0" smolts indi c a t i n g they had prematurely smolted during the freshwater period. 22 Due to the small percentage of the yearclass that the s"0" smolts represent and t h e i r low s u r v i v a b i l i t y during the saltwater period, (0.4%) they have been excluded from analysis of s t r a i n differences and estimates of genetic parameters. Table 3 Least square means for weight and length for each s t r a i n of the 1985 yearclass coho for the freshwater measurements. Strain n Weight SE Length SE (gm) (mm) Implant Measurement Kitimat 229 5.60a* 0.06 79.02 a 0.40 Big Qualicum 264 4.30b 0.06 70.84 a 0.38 Bell a Coola 213 3.85c 0.07 68.79 c 0.45 14 Month Freshwater Weight (FW14mo)  s i smolts Kitimat 198 30.50 a 0.59 134.87 a 0.79 Big Qualicum 191 26.75 b 0.64 126.71 b 0.85 Bella Coola 193 25.01 b 0.59 126.73 b 0.79 s"0" smolts Kitmat 13 129.41a 9.93 213.65 a 5.96 Big Qualicum 45 117.21a 6.31 210.50 a 3.78 Bell a Coola 2 146.95a 20.86 224.50 a 12.50 18 Month Freshwater Weight (FW18mo) s i smolts Kitimat 182 43.86 a 0.69 159.39 a 0.79 Big Qualicum 186 36.14 b 0.72 146.58b 0.85 Bell a Coola 187 40.07° 0.67 155.39 c 0.82 s"0" smolts Kitimat 8 Big Qualicum 29 Be l l a Coola 2 202.55 a 18.36 162.29a 11.05 213.85a 34.70 243.25 a 7.79 243.85 a 4.69 262.50 a 14.73 * SNK t e s t - Means with the same superscript are not s i g n i f i c a n t l y d i f f e r e n t , (P > 0.05). 23 S a l t w a t e r Rearing P e r i o d A summary of least square means and standard errors for both weight and length measurements for the '85YC during the saltwater period are given i n Table 4. Table 4. Least square means and standard errors for weight and length for the 1985 yearclass coho for the saltwater measurements. Strain n Weight SE Length SE (gm) (mm) 8 Month Saltwater Measurement (SW8mo) Kitimat 81 666.28a* 16.99 367.92a 3.12 Big Qualicum 92 432.83 b 14.00 318.95 b 2.57 Bell a Coola 94 6 1 4 . l l c 14.48 3 6 0 . l l c 2.66 16 Month Saltwater Measurement (SW16mo) Kitimat 38 4380.99a* 175.58 629.43 a 10.67 Big Qualicum 48 2885.56b 162.51 545.59 b 9.88 Bell a Coola 63 4061.22a 153.33 624.65 a 9.32 * SNK t e s t - Means with the same superscript are not s i g n i f i c a n t l y d i f f e r e n t , ( P > 0.05). After 8 months of saltwater rearing, (SW8mo), a l l three stra i n s remained s i g n i f i c a n t l y d i f f e r e n t for both weight and length. However by the SW16mo measurement, (Table 4) the two northern s t r a i n s , (Kitimat and Bella Coola), were no longer s i g n i f i c a n t l y d i f f e r e n t from each other, but were s i g n i f i c a n t l y heavier and longer than the Big Qualicum s t r a i n . In the f i n a l 8 months of SW rearing, a l l three 24 str a i n s gained 85% of t h e i r t o t a l market weight. H e r i t a b i l i t y Estimates for Weight and Length Table 5 presents h e r i t a b i l i t i e s estimated from the '85YC for both weight and length at each sampling period within the freshwater and saltwater rearing phase. Table 5. H e r i t a b i l i t y estimates for weight and length for the 1985 yearclass coho derived from DFREML.  Freshwater Period Saltwater Period Implant h 2 SE Weight 0.7 2 0.2 7 Length 0.7 2 0.2 5 FW14mo h 2 SE 0.77 0.31 0.75 0.30 FW18mo h 2 SE 0.61 0.27 0.66 0.31 SW8mo h 2 SE 0.55 0.35 0.64 0.37 SW16mo h 2 SE 0.21 0.38 0.21 0.38 The h e r i t a b i l i t y estimates for freshwater weight and length for the '85YC coho are higher, (0.61 to 0.75), compared to values for sO coho, (0.25 to 0.63) reported by Hershberger et a l , (1990). In other species such as A t l a n t i c salmon and Rainbow trout, reported h e r i t a b i l i t y values for freshwater weight range from 0.05 to 0.29, (Kinghorn, 1982). One reason for the high freshwater h e r i t a b i l i t y estimates obtained i n t h i s study was due to tank e f f e c t s which could not be eliminated. Tank e f f e c t i n t h i s study decreased v a r i a t i o n within s t r a i n by increasing the environmental c o r r e l a t i o n between the progeny 25 within the same family tank as compared to between family tanks. Refstie and Steine, (1978), report that tank e f f e c t accounted for 4.5% of the t o t a l variance within s t r a i n s for weight and 4.1% for length. H e r i t a b i l i t y estimates for SW8mo weight and length, (0.55 and 0.64 respectively), were also higher than 8 month SW estimates, (0.17 to 0.30), for sO coho found by Hershberger et a l , (1990), but agree with estimates for 8 month saltwater weight, (0.55), reported by Withler,(1990) for sO, PIT tagged coho. H e r i t a b i l i t i e s for the SW16mo weight and length, (0.21) were lower than for the SW8mo measurement, (0.55 and 0.64 respec t i v e l y ) , but were i n agreement with those reported for SW18mo weight i n cage reared trout, (0.17 to 0.23), by Gjedre and Gjedrem,(1984) and Kinghorn, (1983). One problem with the SW16mo data was the low s u r v i v a b i l i t y to t h i s measurement, (refer to table 9). A t o t a l of 8 families had 0 individuals survive and 21 families had only 1 indi v i d u a l survive to the SW16mo measurement. Although the standard errors are approximately the same for the h e r i t a b i l i t y estimates at the SW8mo and SW16mo measurements, i t can be speculated that the low s u r v i v a b i l t y w i l l e f f e c t the accuracy of the h e r i t a b i l i t y estimates derived for both weight and length for the SW16mo measurement. 26 Correlation Estimates for Weight and Length Genetic and phenotypic correlations are given for weight and length for the '85YC i n Tables 6 and 7. Table 6. Estimates of phenotypic (below the diagonal) and genetic (above the diagonal) correlations for weight from the 1985 yearclass coho. Impl FW14mo FW18mo SW8mo SW16mo Impl 1 0.710 0.780 0.529 -10.30 FW14mo 0.567 1 0.929 0.706 -1.15 FW18mo 0.561 0.893 1 0.588 -0.57 SW7mo 0.362 0.448 0.410 1 3.45 SW16mo -1.018 0.349 0.315 0.735 1 Table 7. Estimates of phenotypic (below the diagonal) and genetic (above the diagonal) correlations for length from the 1985 yearclass coho. Impl FW14mo FW18mo SW7mo SW16mo Impl 1 0.795 0.770 0.479 0.486 FW14mo 0.600 1 0.950 0.645 0.872 FW18mo 0.550 0.902 1 0.567 0.747 SW7mo 0.300 0.436 0.383 1 0.761 SW16mo 0.269 0.387 0.385 0.703 1 27 The c o r r e l a t i o n estimates for weight between the SW18mo and the other measurements of the '85YC, (Table 6), f a l l outside the parameter space and do not have a sensible b i o l o g i c a l explanation. As indicated previously, low s u r v i v a b i l i t y i s reasoned to e f f e c t the accuracy of parameter estimates for the SW18mo measurement. Phenotypic and genetic correlations for both weight and length were generally higher for two measurements within the freshwater rearing period compared to the correlations determined for two measurements between the freshwater and saltwater rearing periods. The genetic c o r r e l a t i o n , (rg=0.58) between smolt weight at saltwater transfer, (FW18mo) and weight at SW8mo was i n agreement with the genetic c o r r e l a t i o n , (rg=0.55) reported for sO coho by Hershberger et a l , (1990),. However the phenotypic c o r r e l a t i o n , (rp=0.41), for the same time period was higher than those, (rp=.10 to .18), reported by Hershberger et a l . , (1990). 28 As shown i n Table 8, there was a high phenotypic c o r r e l a t i o n between weight and length at a l l measurements of the '85YC . Genetic correlations between weight and length were also found to be close to unity during the freshwater period however decreased , (0.82 for the SW8mo and 0.75 for the SW16mo) over the saltwater period. Table 8 Estimates of genetic and phenotypic correlations for weight and length of the 1985 yearclass coho.  Correlation IMPLANT FW14mo FW18mo SW8mo SW16mo Genetic 0.92 0.99 1.00 0.82 0.75 Phenotypic 0.93 0.96 0.95 0.96 0.94 S u r v i v a b i l i t y Table 9 summarizes percent s u r v i v a b i l i t y from IMPLANT to each measurement for a l l three st r a i n s of the '85YC. Table 9. Percent s u r v i v a b i l i t y from the IMPLANT measurement for a l l s t r a i n s of the 1985 yearclass coho  Strai n FW14mo FW18mo SW8mo SW16mo Kitimat 93% a 86% a 35% a 21% a Big Qualicum 89% a 86% a 35% a 23% a B e l l a Coola 92% a 89% a 44% a 31% b *Percent s u r v i v a b i l i t y with the same superscript are not s i g n i f i c a n t l y d i f f e r e n t , (P > .05) 29 S u r v i v a b i l i t y of the '85YC coho i s shown to decrease rapid l y within the f i r s t 8 months a f t e r saltwater entry. The cause of t h i s mortality i s assumed due to predation, Ba c t e r i a l Kidney Disease and unexplained losses which were inherent i n the B.C. salmon farming industry at the time. The Be l l a Coola s t r a i n exhibited s i g n i f i c a n t l y higher s u r v i v a b i l i t y than the Kitimat or Big Qualicum s t r a i n s to SW18mo, but i s s t i l l below the le v e l s acceptable for p r o f i t a b l e commercial production. H e r i t a b i l i t y Estimates for S u r v i v a b i l i t y Estimates of h e r i t a b i l i t y , (h 2 p) , for s u r v i v a b i l i t y using the IAM ,(Individual Animal Model), approach through DFREML programs are summarized i n Table 10. Transformed h e r i t a b i l i t y estimates, ( h 2 x ) , on an underlying l i a b i l i t y scale, (Dempster and Lerner, 1950), are also given i n Table 10. Table 10. H e r i t a b i l i t y estimates for s u r v i v a b i l i t y using DFREML analysis and adjusted for the 1985 yearclass coho  Observed Scale (h 2p) 0. 164 L i a b i l i t y Scale (h 2 x) FW14mo 0.435 FW16mo 0. 137 0. 302 SW8mo 0. 097 0.161 SW16mo 0. 103 0.210 30 H e r i t a b i l i t y estimates for s u r v i v a b i l i t y analyzed by DFREML were moderate, (0.435 to 0.3 02) for the FW14mo and FW18mo measurements and low (0.161 to 0.210) for the SW8mo and SW18mO measurements. No published estimates for s u r v i v a b i l i t y to FW14mo or Fwl8mo measurements could be found but estimates for SW8mO and SW18mo were i n agreement with values, (0.16 to 0.27) reported by Standal and Gjedre, (1987) for SW18mo A t l a n t i c salmon. Transformation to the underlying l i a b i l i t y scale increased h e r i t a b i l i t y estimates for a l l measurements. Estimates derived by Rye et a l , (1990) and Standal and Gjedre, (1987), also increased a f t e r transformation to the underlying l i a b i l i t y scale. 31 198 6 Yearclass Coho - sO Rearing Strategy Str a i n Comparison Freshwater Rearing Period The IMPLANT measurement i s the only freshwater data c o l l e c t e d before saltwater entry because of the sO rearing strategy, (Clarke et a l , 1989) of the '86YC coho. A l l f i s h were transferred to s a l t water approximately 3 weeks following PIT tag implant and vaccination for saltwater v i b r i o s i s . Least square means for both weight and length for each s t r a i n at the IMPLANT measurement are given i n Table 11. S i g n i f i c a n t differences between a l l three s t r a i n s were found for both weight and length. Table 11. Least squares means and standard errors for weight and length at the IMPLANT measurement for the 1986 yearclass coho.  Strain n Weight SE Length SE (gm) (mm) Kitimat 622 4.49a* 0.30 73.69 a 0.17 Pallant Crk 600 3 . 67 b 0. 03 67.20 b 0.17 Robertson Crk 588 4 . 06 c 0. 03 68.16 c 0. 01 *SNK t e s t - Means with the same superscript are not s i g n i f i c a n t l y d i f f e r e n t , ( P > 0.05). 32 The f i s h from the Kitimat s t r a i n were s i g n i f i c a n t l y larger in weight and length at the IMPLANT measurement than the Pallant and Robertson Creek s t r a i n s . This s u p e r i o r i t y of the Kitimat s t r a i n was also demonstrated at the IMPLANT measurement for the '85YC, however the spawning and ponding times for the stra i n s of the '86YC were more s i m i l a r , (see Table 1), ind i c a t i n g a v a l i d s t r a i n difference. The difference between stra i n s at the IMPLANT measurement w i l l be la r g e l y attributed to tank e f f e c t s which was also evident i n the '85YC. Salt Water Rearing Period The frequency d i s t r i b u t i o n s for both weight and length at the 7 month saltwater measurement, (SW7mo), were bimodal. No evaluation was made for the physiological status of these f i s h , (Gorbman et a l , 1982 and Folmar, 1982), but i t i s thought by the author that the d i s t i n c t i o n between the two groups can be attributed to differences i n s m o l t i f i c a t i o n . The mean weight and length for the f i s h categorized as sO smolts were 2 7 6.4 gm and 281.6 gm, whereas the mean weight and length for the sO nonsmolts was 21.8 gm and 123.7 gm . The nonsmolts were not used for the analysis of s t r a i n differences or estimate of genetic parameters. As indicated in Table 12, the stra i n s were a l l 33 s i g n i f i c a n t l y d i f f e r e n t from each other for weight and length for the SW7mo measurement. The Kitimat s t r a i n was s i g n i f i c a n t l y larger than the other two st r a i n s for both saltwater measurements. The Pallant Creek s t r a i n which was smaller than the Robertson Creek s t r a i n at IMPLANT was s i g n i f i c a n t l y heavier and longer for both the saltwater measurements. Table 12. Least squares means and standard errors for weight and length at the 7 month and 15 month saltwater measurement for the 1986 yearclass coho.  Strain n Weight SE Length SE (gm) (mm) 7 Month Saltwater Measurement (SW7mo) sO smolts Kitimat 237 352 . 64 a* 5. 82 304. 68 a 1. 99 Pallant Crk 316 264 . 12 b 4 . 01 279. 00 b 1. 38 Robertson Crk 262 199 .75 c 4. 92 257. 01 c 1. 69 nonsmolts Kitimat 36 18 .9 a 1. 31 120. 66 a 2 . 01 Pallant Crk 15 22 .8 a 1. 91 122. 89 a 2 . 95 Robertson Crk 30 22 .3 a 1. 55 124. 83 a 2. 39 15 Month Saltwater Measurement (SW15mo)  sO smolts Kitimat 139 3303.70a* 64.11 566.08a 4.76 Pallant Crk 199 2842.58b 56.73 542.47 b 4.21 Robertson Crk 107 1996.07c 81.58 472.84 c 6.05 *SNK t e s t - Means with the same superscript are not s i g n i f i c a n t l y d i f f e r e n t , ( P > 0.05) 34 A l l but 4 of the "nonsmolts" from the SW7mo measurement disappear by the SW15mo measurement. The loss of these f i s h w i l l be the r e s u l t of production management practices being directed to the larger f i s h and mortality due to osmoregulatory f a i l u r e which can be c h a r a c t e r i s t i c of the nonsmolt i n saltwater, (Folmar et a l , 1982). Other causes such as BKD and predation would have also been a factor. H e r i t a b i l i t y estimates H e r i t a b i l i t y estimates for weight and length for the '86YC using DFREML analysis are summarized i n Table 13. Table 13. H e r i t a b i l i t y estimates and standard errors for weight and length for the 1986 yearclass coho using DFREML analysis. Freshwater Period Saltwater Period Implant SW7mo SW15mo h 2 SE h 2 SE h 2 SE Weight 0.98 0.24 0.48 0.19 0.45 0.20 Length 0.72 0.24 0.40 0.19 0.35 0.23 H e r i t a b i l i t y estimates for weight and length at the IMPLANT measurement are considerably higher than the saltwater measurements for the '86YC. As i n the '85YC, the high h e r i t a b i l i t y estimates for weight and length at IMPLANT for the '8 6YC ,(0.98 and 0.72 ) were influenced by tank e f f e c t s , which r e s u l t i n an increased environmental 35 c o r r e l a t i o n among f u l l s i b r e l a t i v e s . H e r i t a b i l i t y estimates for weight and length at the SW7mo, (0.48 and 0.40), were moderately high but less than the estimates for weight and length, (0.55 and 0.64) at the SW8mo measurement for the '85YC* Withler, (1990), also reports high h e r i t a b i l i t y estimates, (0.55) for 8 month saltwater weight of PIT tagged sO coho, but estimates reported by Hershberger et a l . , (1990), for 8 month saltwater weight and length of sO coho over 10 years of selec t i o n were low, (0.17 - 0.33, 0.10 - 0.30 res p e c t i v e l y ) . H e r i t a b i l i t y estimates at the SW15mo measurement for weight and length, (0.45 and 0.35), of the '86YC were also moderately high and considerably greater than the those determined for the '85YC at the SW16mo measurement, (0.21 and 0.21). The standard errors for the h e r i t a b i l i t y estimates of the '86YC are lower than those of the '85YC ind i c a t i n g that the higher number of PIT tags implanted per family did r e s u l t i n increased accuracy during the saltwater measurements. Genetic and Phenotypic Correlations A summary of genetic and phenotypic correlations for weight and length between measurements of the 86YC are given i n Tables 14 and 15. Genetic and phenotypic correlations 36 were found to be p o s i t i v e and the highest, (rg = 0.674, rp = 0.619) between the SW7mo and SW15mo saltwater weights and the lowest or negatively correlated, (rg = -0.402, rp = 0.054), between the IMPLANT measurement and the 15 month saltwater measurement. The genetic c o r r e l a t i o n between the SW7mo and SW15mo for length, (0.858) i s higher than that for weight, (0.674), between the same measurement periods. Genetic correlations from the saltwater entry weight, (IMPLANT), to the SW7mo weight, (0.32) was lower than the genetic correlations between saltwater entry and the 8 month saltwater measurement, (re = 0.4 2 to 0.70) reported by Hershberger et a l . , (1990). Table 14. Estimates of phenotypic (below the diagonal) and genetic (above the diagonal) correlations for weight for the 1986 yearclass coho.  Implant SW7mo SW15mO Implant 1 0.320 -0.402 SW7mo 0.227 1 0. 674 SW15mo -0.054 0. 619 1 37 Table 15. Estimates of phenotypic (below the diagonal) and genetic (above the diagonal) correlations for length of the 1986 yearclass coho.  Implant SW7mo Swl5mo Implant 1 0.410 0.108 SW7mo 0.444 1 0.878 SW15mo 0.167 0.677 1 As shown i n Table 16 there was a high phenotypic and genetic c o r r e l a t i o n between weight and length at a l l measurements of the '86YC. Table 16. Estimates of genotypic and phenotypic correlations for weight and length of the 1986 yearclass coho.  Freshwater Saltwater Correlation Implant SW7mo SW15mo Genetic 0.95 0.92 0.94 Phenotypic 0.91 0.92 0.92 The genetic and phenotypic correlations between weight and length appear to be close to unity which has been found i n both A t l a n t i c salmon and rainbow trout, Refstie and Steine,1978; Refstie 1980; Gunnes and Gjedrem, 1981). 38 If both the genetic c o r r e l a t i o n and h e r i t a b i l i t y estimates for length are higher than for weight, a correlated response in weight due to i n d i r e c t s e l e c t i o n on length i s expected to be greater than d i r e c t response to sel e c t i o n on weight. Kinghorn, (1983), also states t h i s opinion a f t e r finding that in most studies length i s generally more her i t a b l e than weight. S u r v i v a b i l i t y Table 17 summarizes percent s u r v i v a b i l i t y for a l l three st r a i n s of the 86YC for a l l sampling periods. Comparison of percent s u r v i v a b i l i t y between the '86YC and '85YC found the two yearclass.es were almost i d e n t i c a l for s u r v i v a b i l i t y to the second saltwater measurement, (26% and 24% res p e c t i v e l y ) . Table 17. Strai n SW7mo SW15mo Kitimat 45% a* 23% a Pallant Crk. 56% b 33% b Robertson Crk. 51% c 21% a *Percent s u r v i v a l with the same superscript are not s i g n i f i c a n t l y d i f f e r e n t , (P > 0.05) 39 H e r i t a b i l i t y Estimates for S u r v i v a b i l i t y Estimates of h e r i t a b i l i t y for s u r v i v a b i l i t y (h 2p) of the '86YC using DFREML analysis are summarized i n Table 18. These h e r i t a b i l i t y estimates were transformed to the underlying l i a b i l i t y scale, h 2x, (Dempster and Lerner 1950) and are presented i n Table 18. Table 18 H e r i t a b i l i t y estimates for survival for the 1986 yearclass coho.  Observed Scale L i a b i l i t y Scale (h 2b) (h 2x) SW7mo 0.218 0.371 SW15mo 0.157 0.291 H e r i t a b i l i t y estimates of s u r v i v a b i l i t y analyzed by DFREML were moderate, (0.37 and 0.29) for both saltwater measurements of the '86YC. Standal and Gjerde report s i m i l a r estimates, (h2x=0.27), for trout reared for 15 months i n saltwater. H e r i t a b i l i t y estimates for s u r v i v a b i l i t y at both the SW7mo and SW15mo measurements, (0.37 and 0.29), for the '86YC were higher than those estimated from the SW8mo and SW16mo measurements, (0.16 and 0.21) of the '85YC. Transformation to the underlying l i a b i l i t y scale increased the '86 h e r i t a b i l i t y estimates as was reported by Rye et a l , (1990) and Standal and Gjedre, (1987). 40 Discussion St r a i n Comparison It was shown from t h i s study that s i g n i f i c a n t differences between s t r a i n s of coho did e x i s t i n both the freshwater and saltwater rearing periods for both weight and length. As the f i s h farmer i s paid on the basis of f i s h weight at the "market window", (SW16mo weight for the *85YC and SW15mo weight for the '86YC), t h i s section w i l l concentrate on t h i s measurement. The northern s t r a i n s , (Kitimat, B e l l a Coola and Pallant Creek), were s i g n i f i c a n t l y heavier and longer than the southern str a i n s , (Big Qualicum and Robertson Creek) at the saltwater market window measurement. Fish from the Kitimat s t r a i n were the largest, (for both sO and s i rearing s t r a t e g i e s ) , followed by the B e l l a Coola and Pallant Creek s t r a i n s . Withler, (1990) , also found the Kitimat s t r a i n to be larger at 8 and 15 months of saltwater rearing than two southern s t r a i n s , (Robertson Creek and Quinsam River). St r a i n differences for saltwater weights i n farmed A t l a n t i c salmon and chinook have also been reported by Gunnes and Gjedrem, (1978); Cheng et al,(1987). St r a i n differences i n weight at the saltwater market window are e s p e c i a l l y important when considering the market price per pound, Table 19. 41 Table 19. Summary of the percentage of f i s h for each s t r a i n of the 1985 and 1986 yearclass coho that f a l l into the fresh farm f i s h market categories.  Weight Category* l-21b 2-41b 4-61b 6-9lb 9-121b 12+lb Price per pound** $1.45 $2.55 $3.00 $3.25 $3.35 $3.40 1985 Yearclass-sl rearing strategy Kitimat 0%*** 2.0% 7.9% 42.1% 42.1% 2.6% Big Qualicum 3.0% 10.4% 35.4% 47.9% 0% 0% Be l l a Coola 0% 4.8% 11.1% 52.4% 31.7% 0% 1986 Yearclass-sO rearing strategy Kitimat 1. 4% 5. 0% 33 . 1% 57 . 5% 2.8% 0% Robertson Creek 13 . 4% 52 . 1% 30. 3% 4. 2% 0% 0% Pallant Creek 0. 9% 14 . 9% 48. 8% 35. 3% 0% 0% *Weight categories are based on dressed head on with g i l l out. **Prices are based on #1 qua l i t y P a c i f i c salmon prices,FOB Vancouver for the week of March 11, 1991. ***A11 percentages have been adjusted for 15% dressing weight los s . As shown i n table 19, sel e c t i o n of the Kitimat s t r a i n would obtain the greatest number of f i s h i n the highest weight categories for both the '85YC and '86YC. This table also i l l u s t r a t e s that a higher price per pound i s associated with the higher weight categories r e s u l t i n g i n the Kitimat s t r a i n having a further economic advantage over the other str a i n s for weight. The freshwater rearing strategy, ( s i vs sO) was also shown to have a important e f f e c t on weight at the market window. Fish reared by the s i strategy were approximately 33% heavier 42 than the f i s h reared by the sO strategy. The reason for t h i s difference may be attributed to the larger smolt si z e at saltwater entry. H e r i t a b i l i t y Estimates H e r i t a b i l i t y estimates for both weight and length were high for a l l measurements, ( p a r t i c u l a r l y the IMPLANT measurement), during the freshwater period for both the '85YC and '86YC. The reason for these high values can be par t l y attributed to maternal and family tank e f f e c t s which have also been reported by Refstie and Steine, (1978), Kaines et al,(1976);). Estimates of h e r i t a b i l i t y for weight at the market window d i f f e r s considerably for the '85YC, ( 0.21) and the '86YC, (0.45). The difference i n accuracy of these h e r i t a b i l i t y estimates w i l l primarily be attributed to the low number of PIT tagged f i s h that survived to the market window i n the '85YC. A t o t a l of 8 families had no i d e n t i f i e d i ndividuals surviving and 21 families had only one in d i v i d u a l surviving to the SW18mo measurement. It i s important to r e a l i z e that using the IAM to estimate variance components for the '85YC at the SW18mo allowed the information from the single record families to be used. With the Least Squares method the information from these families would have been l o s t due to i n s u f f i c i e n t number of degrees of freedom. 43 For both yearclasses the h e r i t a b i l i t i e s for weight were s u f f i c i e n t l y large to allow for s i g n i f i c a n t response to s e l e c t i o n . (Falconer, 1981; Van Vleck, 1987). Hershberger et a l , (1990), found that moderate h e r i t a b i l i t y estimates,(.19 -.33), resulted i n large increases i n weight, (average 10.1% gain per generation), for 8 month saltwater weight of net-pen reared coho. Gjedrem, (1983), reports a genetic change of 3.6% and 2.7% for body weight a f t e r 2 years of saltwater rearing compared with the control groups. S u r v i v a b i l i t y The o v e r a l l s u r v i v a b i l i t y from the IMPLANT measurement to the market window of 17 - 18 months of saltwater rearing was s i m i l a r for both yearclasses and well below the l e v e l of p r o f i t a b i l i t y . The Pallant Creek and B e l l a Coola s t r a i n s had a 10% higher s u r v i v a l than the Kitimat, Big Qualicum and Robertson Creek st r a i n s , (32 % vs 22%). These differences were only found during the saltwater rearing period, as differences between stra i n s i n freshwater were not s i g n i f i c a n t . S u r v i v a b i l i t y of the two yearclasses of the Kitimat s t r a i n reared as s i smolts for the '85 YC and sO smolts for the '86YC was s i m i l a r for the two groups, (21% and 23% from IMPLANT to the market weight). This indicates that the method of rearing ( s i vs sO) did not influence s u r v i v a b i l i t y . 44 This i s consistent with the finding that differences between s t r a i n arose i n the s a l t water and not the fresh water. During the c o l l e c t i o n of PIT tags from the project's moribund f i s h , B a c t e r i a l Kidney Disease, was observed as one of the causes for low s u r v i v a b i l i t y during the saltwater period. McGeer et a l . , (1991) and Withler, (1990), found s i g n i f i c a n t s t r a i n differences for BKD resistance when comparing coho s t r a i n s . Withler, (1990) , also determined the h e r i t a b i l i t y estimate for BKD resistance to range from 0.26 to 0.53. H e r i t a b i l i t y estimates of s u r v i v a b i l i t y were low to moderate, (.16 to 0.43 for the '85YC and 0.29 to 0.37 for the '86 YC), but generally higher than estimates reported for A t l a n t i c salmon, and rainbow trout, (.14 and .18) by Gjedrem, (1983) or those, (0.00 to 0.35) reported by Standal and Gjerde, (1987). I t can be concluded that considerable additive genetic v a r i a t i o n exists for s u r v i v a b i l i t y . If s u r v i v a l cannot be increased through f i s h culture management techniques for coho, and in d i v i d u a l i d e n t i f i c a t i o n i s retained for successive generations, then s e l e c t i o n for s u r v i v a b i l i t y would be a economically important t r a i t i n coho broodstock development. 45 Combining S u r v i v a l and Growth Rate To indicate which s t r a i n i s best when considering both weight and s u r v i v a l , weight at the 15-16 month saltwater measurement was adjusted for a sO rearing strategy, (86 Kitimat weight/85 Kitimat market weight), and then weighted by percent s u r v i v a l to give the harvest r e l a t i v e to the number of f i s h at the s t a r t , (Table 30). This value i s termed the Y i e l d which indicates that the B e l l a Coola and Pallant Creek str a i n s produced over 900 gms of f i s h for each f i s h started. Although the Kitimat s t r a i n was greater i n weight than the other s t r a i n s at each measurement the s u r v i v a b i l i t y was approximately 10% less than the Bella Coola and Pallant Creek st r a i n s . Table 3 0 shows that the increased s u r v i v a b i l i t y for the Bella Coola and Pallant Creek str a i n s has compensated for the l i g h t e r weight per f i s h and resulted with i n a higher calculated "Yield " for each f i s h started i n the t r i a l . Taking into account differences i n price per pound for the d i f f e r e n t market weight categories, (table 29), the B e l l a Coola and Pallant Creek str a i n s were the most productive of a l l f i v e s t r a i n s . 46 Table 20 Strai n comparisons combining survival and weight at the "market window",( 15 and 16 month saltwater measurement), for the 1985 and 1986 yearclass coho.  % Adj % St r a i n Survival Market Weight Y i e l d of Best Bel l a Coola 31 3061 949 100 Big Qualicum 23 2175 500 53 Kitimat '85 21 3303 693 73 Kitimat '86 23 3303 742 79 Pallant Ck. 33 2842 917 98 Robertson Ck. 21 1996 383 41 The Kitimat s t r a i n had only 74% and 79% of t h i s y i e l d under s i and sO rearing strategies. On the same basis the Big Qualicum and the Robertson Crk. st r a i n s exhibited the poorest performance ,(53 % and 41% of the B e l l a Coola s t r a i n ) , which demonstrates that neither of these two strai n s excelled i n s u r v i v a b i l i t y or growth rate. It i s apparent from t h i s study that evaluation of s t r a i n for performance t r a i t s , (weight and s u r v i v a b i l i t y ) i s important i n establishing the basis for a genetic broodstock sel e c t i o n program for farm reared coho i n B.C. The three northern s t r a i n s from B.C. were proven to be ge n e t i c a l l y superior compared to southern B.C. stocks for a l l t r a i t s measured i n t h i s study, (weight.length and s u r v i v a b i l i t y ) . 47 S u f f i c i e n t additive genetic v a r i a t i o n for these t r a i t s was found to e x i s t within and between s t r a i n to predict that s i g n i f i c a n t genetic improvement can be accomplished through a future s e l e c t i o n program. By retaining i n d i v i d u a l animal i d e n t i f i c a t i o n of the progeny from selected parents from t h i s study and allowing further analysis using the IAM, s i g n i f i c a n t genetic improvement i n coho growth performance and p a r t i c u l a r l y s u r v i v a b i l i t y i s anticipated. 48 BIBLIOGRAPHY Becker, W.A. , 1985. Manual of Quantitative Genetics. Academic Enterprises, Pullman , WA, 186 pp. Brannon, E., Feldmann, C. and Donaldson, L., 1982. University of Washington zero-aged coho smolt production. Aquaculture, 28: 195-200. Cheng, K.M., McCallum, I.M., McKay, R.I. and March, B.E., 1987. A comparison of survival and growth of two str a i n s of chinook salmon (Oncorhynchus tshawytscha) and t h e i r crosses reared in confinement. Aquaculture, 67: 301-311 Clarke, W.C.,1989. Rearing strategies for zero-aged coho salmon, Oncorhynchus kisutch. Clarke, W.C., 1986. Delayed photoperiod produces more uniform growth and greater seawater adaptability i n underyearling coho salmon, (Oncorhynchus kisutch). Aquaculture 56: 287-299. Dempster, E.R. and Lerner, I.M., 1950. H e r i t a b i l i t y of threshold characters. Genetics, 35: 212-236. Falconer, D.S., 1981. An Introduction to Quantitative Genetics. Longman Group Limited, New York, NY, 340 pp. Folmar, L.C., Dickhoff, W.W., Mahnken, C.V.W. and Waknitz,W. 1982. Aquaculture, 28: 91-104 G a l l , G.A.E. and Cross, S.J., 1978. A genetic analysis of the performance of three rainbow trout broodstocks. Aquaculture, 15: 113-127 Gjerde, B. and Gjedrem, T., 1984. Estimates of phenotypic and genetic parameters for carcass t r a i t s i n A t l a n t i c salmon and rainbow trout. Aquaculture, 1983: 97-110 Gjerde, B., and Schaeffer, L.R., 1989. Body t r a i t s i n rainbow trout II.Estimates of h e r i t a b i l i t i e s and of phenotypic and genetic cor r e l a t i o n s . Aquaculture, 80: 25-44 Gjedrem, T., 1975. P o s s i b i l i t i e s for genetic gain i n salmonids. Aquaculture, 6: 23-29 49 Gjedrem, T., 1983. Genetic v a r i a t i o n i n 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 Gjedrem, T. and Aulstad, D., 1974. Selection experiments with salmon I. Differences i n resistance to v i b r i o disease of salmon parr (Salmo s a l a r ) . Aquaculture, 3: 51-59 Gorbman, A., Dickhoff,W.W., Mighell J.L., Prentice,E.F. and Waknitz, F.W., 1982. Morphological i n d i c i e s of developmental progress i n the parr-smolt coho salmon, Oncorhynchus kisutch. Aquaculture, 28: 1-19 Gordon, M.R., K.C. Klotins, Campbell V.M,, Cooper, M.,1987. Farmed Salmon Broodstock Management. B.C. Research Vancouver, B.C. Gunnes, K. and Gjedrem,T., 1978. Selection experiments with salmon.IV.Growth of A t l a n t i c salmon during two years i n the sea. Aquaculture, 15: 19-23 Gunnes, K. and Gjedrem, T., 1981. A genetic analysis of body weight and length i n rainbow trout reared i n seawater for 18 months. Aquaculture, 24: 161-174 Gunnes, K. and Refstie, T., 1980. Cold-branding and f i n -c l i p p i n g for marking of salmonids. Aquaculture, 19: 295-299 Hershberger, W.K., Meyers, J.M., Iwamoto, R.N.,Mcauley, W.C., 1990. Genetic changes i n the growth of Coho salmon (Oncorhynchus kisutch) i n marine net pens, produced by ten years of sele c t i o n . Aquaculture, 85: 187-197 Henderson, C.R., 1984. Applications of Linear Models i n Animal Breeding. Univ. Guelph Press, Guelph, Ont., Can. Kanis, E., Refstie, T., and Gjedrem, T., 1976. A genetic analysis of egg, alevin and fry mortality i n salmon (salmo s a l a r ) , sea trout (salmo trutta) and rainbow trout (salmo g a i r d n e r i ) . Aquaculture, 8: 259-268 Kinghorn, B.P., 1983. A review of quantitative genetics i n f i s h breeding. Aquaculture, 31: 283-304 50 Klein, W.T., DeFries, J.C. and Finkbeiner, CT., 1973. H e r i t a b i l i t y and genetic c o r r e l a t i o n : standard errors of estimates and sample s i z e . Behavior Genetics, Vol.3, No.4. :355-364 McGeer, J . C , Baranyi, L. and Iwama, G.K., 1991. Physiological reponses to challenge t e s t s i n fixed stocks of coho salmon.Can. J . Fish. Aq. 48:in p r i n t Novotny, A.J., 1975. Net pen culture of P a c i f i c salmon i n P a c i f i c marine waters. Mar. Fish. Rev., 37: 36-47 Malik, H.J. and Mullen, K., 1973. A F i r s t Course i n Probabil i t y and S t a t i s t i c s . Addison-Wesley Publ. Co., Don M i l l s , Ont. 358 pp. Prentice, E.F., Flagg, T.A. and McCutcheon, S., 1987. A study to determine the b i o l o g i c a l f e a s i b i l i t y of a new f i s h tagging system 1986-87. Annual Report of Research. U.S. Dept. of Energy, Bonneville Power Administration, 112pp Refstie, T., 1980. Genetic and environmental sources of v a r i a t i o n i n body weight and length of rainbow trout f i n g e r l i n g s . Aquaculture, 19: 351-357 Refstie, T., 1990. Application of breeding schemes. Aquaculture, 85: 163-169 Refstie, T. and Steine, T.A., 1978. Selection experiments with salmon.III.Genetic and environmental sources of va r i a t i o n i n length and weight of A t l a n t i c salmon i n the freshwater phase, Aquaculture. 14: 221-234 Rye, M., Lillevik,K.M. and Gjerde, B., 1990. Survival i n early l i f e of A t l a n t i c salmon and rainbow trout:estimates of h e r i t a b i l i t i e s and genetic c o r r e l a t i o n s . Aquaculture, 89: 209-216 Standal, M and Gjedre, B., 1987. Genetic v a r i a t i o n i n sur v i v a l of A t l a n t i c salmon during the sea-rearing period. Aquaculture, 66: 197-207 SAS In s t i t u t e Inc., 1985. SAS/STAT Guide for Personal Computers, Version 6 Ed i t i o n . SAS In s t i t u t e Inc., Cary, NC, 378 pp. Wilkins, N.P., 1981. The r a t i o n a l and relevance of genetics in aquaculture: An overview. Aquaculture, 22: 209-228 51 Withler, R.E., Clarke, W.C., R i d d e l l , B.E. 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 Withler, R.E., 1990. Selective breeding for coho. Aquaculture update. Vol 4 pp. Van Vleck, L.D., Pollak,E.J. and Oltenacu, E.A.B., 1987. Genetics for the Animal Sciences. W.H. Freeman and Co., New York, NY, 391 pp. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0098725/manifest

Comment

Related Items