UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Patterns of variation in Pacific silver fir (Abies amabilis [Dougl.] Forbes) on Vancouver Island Davidson, Roberta H. 1990

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

Item Metadata

Download

Media
831-UBC_1990_A1 D38.pdf [ 9.89MB ]
Metadata
JSON: 831-1.0100310.json
JSON-LD: 831-1.0100310-ld.json
RDF/XML (Pretty): 831-1.0100310-rdf.xml
RDF/JSON: 831-1.0100310-rdf.json
Turtle: 831-1.0100310-turtle.txt
N-Triples: 831-1.0100310-rdf-ntriples.txt
Original Record: 831-1.0100310-source.json
Full Text
831-1.0100310-fulltext.txt
Citation
831-1.0100310.ris

Full Text

PATTERNS OF VARIATION IN PACIFIC SILVER FIR (ABIES AMABILIS [DOUGL.] FORBES) ON VANCOUVER ISLAND b y R o b e r t a H. D a v i d s o n B . S c , Simon F r a s e r U n i v e r s i t y , 1980 A THESIS SUBMITTED I N PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES ( F a c u l t y o f F o r e s t r y ) We a c c e p t t h i s t h e s i s a s c o n f o r m i n g t o t h e r e q u i r e d s t a n d a r d THE UNIVERSITY OF B R I T I S H COLUMBIA A u g u s t , 1990 ©Roberta H. D a v i d s o n , 1990 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 Fbra The University of British Columbia Vancouver, Canada DE-6 (2/88) ABSTRACT T h i s t h e s i s d e s c r i b e s p a t t e r n s o f v a r i a t i o n i n s e v e r a l c o n e , s e e d a n d s e e d l i n g c h a r a c t e r i s t i c s o f P a c i f i c s i l v e r f i r ( A b i e s amabilis [Doug.] F o r b e s ) s a m p l e d f r o m V a n c o u v e r I s l a n d , B.C. Cone c o l l e c t i o n s k e p t s e p a r a t e b y t r e e w e re made a t e i g h t l o c a t i o n s d u r i n g t h e f a l l o f 1983 t o p r o v i d e m a t e r i a l f o r t h e s t u d y . The i n h e r i t a n c e p a t t e r n s o f 13 enzyme l o c i w e r e d e t e r m i n e d f r o m s e e d t i s s u e s o f 87 t r e e s , a n d s e v e n l o c i w e r e f o u n d t o p o s s e s s a t l e a s t t wo a l l o z y m e v a r i a n t s . T h e s e l o c i c o n f o r m e d t o t h e a s s u m p t i o n s o f M e n d e l i a n - t y p e i n h e r i t a n c e , a l t h o u g h AAT - 2 d i s p l a y e d m a r k e d s e g r e g a t i o n d i s t o r t i o n . No l i n k a g e g r o u p s c o u l d be e s t a b l i s h e d a t t h e s a m p l i n g i n t e n s i t y a v a i l a b l e (20 s e e d s p e r t r e e ) . S i g n i f i c a n t l e v e l s o f i n b r e e d i n g , b a s e d on a m u l t i l o c u s e s t i m a t e o f o u t c r o s s i n g , w e re d e t e c t e d i n f i v e o f s e v e n p o p u l a t i o n s a n d i n d i r e c t e v i d e n c e s u g g e s t s r e l a t e d m a t i n g s o t h e r t h a n s e l f i n g may be o c c u r r i n g . V a r i a t i o n among p o p u l a t i o n s i n o u t c r o s s i n g r a t e was e v i d e n t ( 0 .725 < t m < 1.0) a n d a p p e a r s p o s i t i v e l y c o r r e l a t e d w i t h s e e d s i z e ( m e a s u r e d b y 1 0 0 0-seed w e i g h t ) . H i g h l e v e l s o f a l l o z y m e v a r i a t i o n were f o u n d t o e x i s t w i t h i n p o p u l a t i o n s (95-98%) a n d e s t i m a t e s o f t h e e x t e n t o f p o p u l a t i o n d i f f e r e n t i a t i o n w e r e shown t o d i f f e r d e p e n d i n g upon t h e p a r t i c u l a r a n a l y t i c m e t h o d e m p l o y e d . M a t e r n a l ( e x t a n t ) t r e e s a p p e a r e d more h e t e r o z y g o u s t h a n d i d v i a b l e e m b r y o s a n d p o p u l a t i o n s s a m p l e d on s o u t h e r n V a n c o u v e r I s l a n d a p p e a r e d more g e n e t i c a l l y d i v e r s e t h a n d i d p o p u l a t i o n s s a m p l e d on n o r t h e r n V a n c o u v e r I s l a n d . A s u b - s a m p l e c o n s i s t i n g o f two p o p u l a t i o n s , e a c h w i t h s e v e n t r e e s , f r o m n o r t h e r n , m i d a n d s o u t h e r n V a n c o u v e r I s l a n d p r o v i d e d m a t e r i a l f o r a g e r m i n a t i o n t e s t a n d o p e n -p o l l i n a t e d p r o g e n y s t u d y . S e e d d o r m a n c y was n o t p r o n o u n c e d among p o p u l a t i o n s . L a r g e f a m i l y d i f f e r e n c e s i n g e r m i n a t i o n r e s p o n s e s w e re d e t e c t e d , i r r e s p e c t i v e o f p r e g e r m i n a t i o n t r e a t m e n t , s u g g e s t i n g a h i g h d e g r e e o f g e n e t i c c o n t r o l o f g e r m i n a t i o n i n P a c i f i c s i l v e r f i r . A n o m a l o u s g e r m i n a t i o n b e h a v i o r i n one p o p u l a t i o n was a t t r i b u t e d t o s u b - o p t i m a l s t r a t i f i c a t i o n c o n d i t i o n s a n d p r o l i f e r a t i o n o f m o l d . I m p r o v e m e n t i n p r o d u c t i o n o f s e e d l i n g s o f P a c i f i c s i l v e r f i r may b e a c h i e v e d b y c o l l e c t i n g a n d g e r m i n a t i n g s e e d s on a f a m i l y - b y - f a m i l y b a s i s . G e r m i n a n t s f r o m t h e f i r s t c o u n t o f t h e g e r m i n a t i o n t e s t p r o v i d e d o p e n - p o l l i n a t e d p r o g e n y f o r meas u r e m e n t o f g r o w t h v a r i a b l e s . S e e d l i n g s were grown i n a g r e e n h o u s e a l o n g s i d e p r o d u c t i o n Abies s t o c k f o r 29 weeks. P o p u l a t i o n d i f f e r e n c e s a c c o u n t e d f o r a c o n s i d e r a b l e p a r t o f v a r i a n c e i n c o n e a n d s e e d s i z e . The e f f e c t o f p o p u l a t i o n on h e i g h t o f s e e d l i n g s a t e i g h t weeks was s i g n i f i c a n t b u t d e c l i n e d t o v i r t u a l l y z e r o b y t h e e n d o f t h e t e s t . P o p u l a t i o n s h a d n e g l i g i b l e i n f l u e n c e on g r o w t h r a t e o f s e e d l i n g s a s w e l l . V a r i a t i o n i n g r o w t h r a t e among o p e n - p o l l i n a t e d f a m i l i e s was s t a t i s t i c a l l y s i g n i f i c a n t but accounted f o r only 20% of the t o t a l v a r i a t i o n . S i g n i f i c a n t p o p u l a t i o n d i f f e r e n c e s were d e t e c t e d i n r o o t weight of h a r v e s t e d s e e d l i n g s . Family d i f f e r e n c e s i n t h i s and other biomass v a r i a b l e s were at most 20%, with the m a j o r i t y of v a r i a t i o n i n s e e d l i n g growth t r a i t s r e s i d i n g w i t h i n f a m i l i e s . V TABLE OF CONTENTS P a 9 e ABSTRACT i i TABLE OF CONTENTS v LIST OF TABLES v i i i LIST OF FIGURES x i ACKNOWLEDGEMENTS . . . . x i i 1. INTRODUCTION 1 1.1 Study Objectives 8 2. MATERIAL COLLECTION AND PROCESSING 9 3. ISOZYME INHERITANCE 13 3.1 Introduction 13 3.2 Materials and Methods 17 3.3 Results and Discussion 21 3.3.1 Enzymes Classed as Monomorphic ... 21 3.3.2 Enzymes Exhibiting Polymorphism .. 28 3.4 Conclusion 40 4. LINKAGE 42 4.1 Introduction 42 4.2 Materials and Methods 44 4.3 Results and Discussion 47 4.4 Conclusion 53 5. MATING SYSTEM 55 5.1 Introduction 55 5.2 Materials and Methods 57 5.3 Results and Discussion 62 5.3.1 A l l e l e Frequencies 62 v i Page 5.3.2 Outcrossing Rates 68 5.4 Conclusions 75 6. ESTIMATES OF ELECTROPHORETIC VARIATION, ITS STRUCTURE AND RELATIONSHIP TO THE MATING SYSTEM 7 6 6.1 Introduction 76 6.2 Materials and Methods . ... 79 6.2.1 Sample c o l l e c t i o n and e l e c t r o -phoretic assay 79 6.2.2 Analytic methods 80 6.3 Results 86 6.3.1 A l l e l i c v a r i a t i o n 86 6.3.2 D i s t r i b u t i o n of v a r i a t i o n 91 6.3.3 Mating system e f f e c t s on v a r i a t i o n . 97 6.4 Discussion 101 6.5 Conclusions 109 7. SEED GERMINATION 110 7.1 Introduction 110 7.2 Materials and Methods 114 7.2.1 Co l l e c t i o n and t e s t i n g methods 114 7.2.2 Analytic methods 116 7.3 Results and Discussion 120 7.4 Conclusions 136 8. WIND-POLLINATED PROGENY NURSERY-STAGE GROWTH .. 138 8.1 Introduction 138 8.2 Materials and Methods 139 8.2.1 Test establishment and culture ... 139 8.2.2 Measurement and analysis 140 v i i Page 8.3 Results 145 8.3.1 Geographic and maternal tree v a r i a t i o n 147 8.3.2 Geographic influences on seed variables 148 8.3.3 Maternal and seed relationships .. 148 8.3.4 Relationships among seeds and seedlings 151 8.4 Discussion 163 8.5 Conclusions 172 9. SUMMARY AND CONCLUSIONS 173 10. LITERATURE CITED 176 APPENDICES 1. S c i e n t i f i c names of tree species 202 2. Buffers, solutions, gel preparation, running conditions and stain recipes for horizontal starch gel electrophoresis 203 3. Multiple regression equations involving geographic location of populations as predictor variables 207 v i i i LIST OF TABLES Table Page 1.1 Number of seedlings of Abies spp. grown or planted i n B.C. from 1978 to 1986. 5 2.1 Locations of populations of A. amabilis from which cone c o l l e c t i o n s were made in f a l l 1983. 10 3.1 Summary of enzyme systems studied, nomenclature and numbers of l o c i obtained-from electrophoresis. 20 3.2 Segregation patterns for seven variable l o c i . 31 4.1 Results of chi-square t e s t s for the detection of two-point linkage. 48 5.1 A l l e l e frequencies for maternal and outcrossing pollen gene pools. 63 5.2 Ratios of d i s c e r n i b l y outcrossed (DOC) embryos to embryos possessing maternal (homozygous) genotype for three variable l o c i . 66 5.3 Single- and multi-locus estimates of outcrossing rate (± 95% confidence i n t e r v a l s ) . 69 6.1 Pollen a l l e l e frequencies detected at six l o c i in eight populations of A. amabilis. 87 6.2 Actual and " e f f e c t i v e " numbers of a l l e l e per locus. 89 6.3 Mean expected heterozygosities for adult and progeny gene pools. 90 6.4 Estimates of genetic distances and corresponding geographic distances between pairs of A. amabilis stands. 92 6.5 Gene d i v e r s i t y estimates for six and thirteen l o c i i n eight stands of A. amabilis. 94 6.6 A l l e l i c d i f f e r e n t i a t i o n indices for six variable l o c i . 96 ix Page 6.7 Fixation indices for maternal and progeny gene pools. 98 6.8 Comparison of f i x a t i o n indices calculated using i n d i r e c t and d i r e c t methods fo r maternal and progeny gene pools. 100 6.9 Comparison between amounts of population d i f f e r e n t i a t i o n detected by two d i f f e r e n t measures (G S T and 5) at six l o c i . 104 6.10 Average G s x values determined for several Abies spp. 106 7.1a Results of ANOVA of germination capacity (GC) of s t r a t i f i e d seeds from six populations of A. amabilis. 120 7.1b Apportionment of t o t a l v a r i a t i o n and expected mean squares from ANOVA described in Table 7.1a. 122 7.2 Total number of germinants vs. non-germinants, i n d i c a t i n g treatment and cabinet e f f e c t s . 123 7.3a Apportionment of v a r i a t i o n for GC and germination value (GV) i n u n s t r a t i f i e d seeds. 126 7.3b Apportionment of v a r i a t i o n for GC and GV in s t r a t i f i e d seeds. 127 7.4 Average GC in six populations of A. amabilis, for u n s t r a t i f i e d and s t r a t i f i e d seeds. 128 7.5 Average GV i n six populations of A. amabilis , for s t r a t i f i e d and u n s t r a t i f i e d seeds. 129 7.6 Mean GC and GV for i n d i v i d u a l trees i n a single population of A. amabilis. 131 7.7 Linear correlations of average population seed size with GC, GV and l a t i t u d e of o r i g i n . 133 8.1 Level of sampling, number of measurements and variables measured i n nursery stage growth of A. amabilis. 141 Correlations among population, maternal tree and variables. Apportionment of va r i a t i o n for mean cone length, seed weight and cotyledon number in six populations of A. amabilis. Summary s t a t i s t i c s for cone seed variables. Correlations and variance patterns for seed and seedling growth variables. Summary s t a t i s t i c s of seedling growth variables. Results of ANOVA of cumulative and incremental seedling height growth over 2 weeks. Correlations and variance patterns for seed and seedling form variables. Correlation structure among population, maternal tree and seedling form variables. Ranking of population mean value for variables describing seedling form at 29 weeks. x i LIST OF FIGURES Figure Page 1.1 D i s t r i b u t i o n of P a c i f i c s i l v e r f i r i n western North America. 2 2.1 , Location of eight sampled populations on Vancouver Island, B r i t i s h Columbia. 11 3.1 Band patterns of seven polymorphic l o c i detected by starch gel electrophoresis. 29 6.1 Dendrogram representing genetic distances among eight populations of P a c i f i c s i l v e r f i r . 93 8.1 Relationships of variables examined i n wind-pollinated progeny of 42 families from six locations. 146 x i i ACKNOWLEDGEMENTS It i s with enormous respect and gratitude I acknowledge the contributions of the members of my thesis committee, Drs. D.G.W. Edwards, Y.A. El-Kassaby, D.T. Lester, J.R. Maze and B.J. van der Kamp, and the experiences and insights of my supervisor, Dr. 0 . S z i k l a i , throughout the preparation of t h i s d i s s e r t a t i o n . Thanks are also extended to my program advisor Dr. J.W. Wilson, for his continuing encouragement. The generous support of several forest product companies and the B.C. Ministry of Forests i n the c o l l e c t i o n of material was invaluable, as was the assistance of Lachlan Glenn and colleagues at Reid, C o l l i n s nursery, i n processing the volume of seeds required. Thanks are owed to Jack Maze and Rob Scagel, who arranged a f i e l d t r i p which helped me e s t a b l i s h objectives for the thesis and, along with Mishtu Banerjee, p a r t i c i p a t e d i n many discussions on how best to reach them. I also thank Dr. B.A. Bohm, who supported some preliminary work on needle flavonoids. I am grateful to Lee Charleson for both her technical assistance and friendship throughout the course of t h i s work. Lee and her husband, N e i l , also graciously provided accommodation and L e i l a and Yousry El-Kassaby, on numerous occasions, opened heart and home to me. Many thanks are due to George Edwards, Doug Taylor and Frank Portlock of the Canadian Forestry Service for help i n mounting the germination and progeny tests at PFC, V i c t o r i a and Vlad Korelus and the s t a f f of CP Forest Products forestry centre i n Saanich for supporting much of my work. I am e s p e c i a l l y g r a t e f u l to Joy Parkinson for her patience and dedication throughout the isozyme analyses. S t a t i s t i c a l advice was received from Mishtu Banerjee, Yousry El-Kassaby, Drs. H.-R. Gregorius, M. Grieg, A. Kozak, P.L. Marshall, M. Penner and K. Ritland, and Rob Scagel; as well as computing assistance from Rob Davidson, John Emmanuel, Dermot McCarthy, Steve McGillivray and Barry Wong; a l l of i t much appreciated. I cannot adequately acknowledge here the contributions of my family and friends, who gave freely of t h e i r time to a l l aspects of t h i s work, from dissecting cones to babysitting my daughter. The thesis was ably typed by Patsy Quay, Pat Johnson, and, at the eleventh hour, by L i s a Allenbach and Yolanda McGillivray. I could not have done any of t h i s without the u n f a i l i n g support of my husband, Rob. This t h e s i s represents the e f f o r t s of many, however, the mistakes remain my own. 1 1. INTRODUCTION P a c i f i c s i l v e r or amabilis f i r (Abies amabilis [Dougl.] Forbes) 1 occurs over a large portion of the west coast of North America (Figure 1.1) from the southern end of the Alaska panhandle (56°N) to northwestern C a l i f o r n i a (42°N) but i s r e s t r i c t e d i n i t s eastward d i s t r i b u t i o n to a r e l a t i v e l y narrow band, seldom found more than 300 km from the P a c i f i c Ocean (Schmidt 1957, Fowells 1965, Packee et al. 1982, Worrall 1983). The species reaches i t s greatest development and commercial productivity on the west sides of the Olympic and Cascade Mountains, the f o o t h i l l s of the Columbia River and the west coast of Vancouver Island (Handley 1982). In B r i t i s h Columbia, P a c i f i c s i l v e r f i r i s found at elevations from sea-level to 1500 m at the 49th p a r a l l e l and to 300 m at i t s northern l i m i t (Schmidt 1957). In the region of greatest productivity i t i s codominant with western hemlock (Pojar 1982). P a c i f i c s i l v e r f i r grows to higher elevations than does Douglas-fir, grand f i r , Sitka spruce, western hemlock, western red cedar and Alaska yellow-cedar but i s not found as high as mountain hemlock or subalpine f i r (Schmidt 1957). 1 Other t r e e s p e c i e s are mentioned i n the t e x t by common name, o n l y . S c i e n t i f i c names are l i s t e d i n Appendix 1. 2 125° 120° 115° Figure 1.1 D i s t r i b u t i o n of P a c i f i c s i l v e r f i r in western North America (after Fowells, 1965). 3 Shade tolerance of P a c i f i c s i l v e r f i r i s highest of a l l forest tree species,in B.C., which contributes to r e l a t i v e l y dense stocking because of the low space requirement of the spire-shaped crown (Krajina et a i . 1982). P a c i f i c s i l v e r f i r t olerates and even requires large quantities of water, as much as 6650 mm annual mean t o t a l p r e c i p i t a t i o n (Krajina et al. 1982), however s i t e s must be s u f f i c i e n t l y well-drained f o r the species to produce good growth (Schmidt 1957). Schmidt (1957) and Packee et al. (1982) emphasize the a b i l i t y of the species to occupy a wide range of cl i m a t i c and s i t e conditions, and on cool, north-facing slopes may outgrow Douglas-fir (Thornburgh 1969). Several other reports (Handley 1982, Husted and Korelus 1982, Pojar 1982) attest to the productivity of P a c i f i c s i l v e r f i r , due to higher stocking densities and better form but, by virt u e of i t s lower wood density, i t produces i n f e r i o r saw timber to Douglas-fir. Nevertheless, P a c i f i c s i l v e r f i r i s widely used i n general construction (known as "hem-bal"), as a corewood i n plywood production and i s a high-yielding pulpwood species (Handley 1982). U n t i l recently the importance of P a c i f i c s i l v e r f i r as a commercial species in coastal B.C. has been overshadowed by the d e s i r a b i l i t y and a v a i l a b i l i t y of Douglas-fir. S t a t i s t i c s from the B.C. Ministry of Forests Annual Report (1985-86) show that i n the coastal portion of the Vancouver Region the stumpage price for Abies species (including grand 4 and subalpine f i r and referred to c o l l e c t i v e l y as "balsam") i s about 50% of that received for Douglas-fir. The r a t i o of average timber sale prices i n coastal B.C. for the period 1915-1979 (calculated from Handley 1982) was approximately 0.5:1.0 for balsam to Douglas-fir, suggesting that the value placed on these species has remained v i r t u a l l y the same for the past 70 years. In the past decade, however, increasing attention has been paid to P a c i f i c s i l v e r f i r as a suitable tree f o r reforestation on higher-elevation or brush-prone s i t e s (Husted and Korelus 1982, Green et al. 1984), p a r t i c u l a r l y where Douglas-fir plantations are not or w i l l not be successful (Reuter 1973). As more and more e a s i l y accessed stands of lower-elevation species are logged P a c i f i c s i l v e r f i r i s being considered more favorably by the coastal forest industry. This i s evident in the sharp r i s e in the number of P a c i f i c s i l v e r f i r seedlings grown in B.C. nurseries from 1978 to 1986 (Table 1.1), from less than 386,000 in 1978 to 3.4 m i l l i o n . The increased sowing of P a c i f i c s i l v e r f i r r e f l e c t s not only the apparent need for suitably adapted seedlings for reforestation but also represents a response to the removal of a province-wide ban 2 on the nursery production of a l l true f i r s i n B.C., that had been i n e f f e c t since 1966 as a measure to control the spread of balsam O r d e r i n C o u n c i l 4 6 0 , P l a n t P r o t e c t i o n A c t , B.C. R e g . 5 8 / 6 6 . 5 Table 1.1 Number of Abies spp. seedlings grown or planted in B r i t i s h Columbia from 1978 to 1986. Year Thousands of seedlings grown® or planted® 1986 3, 400 1985 * 3,700 1984 1,400 1983 1, 600 1982 2, 900 1981 3, 400 1980 2, 245 1979 1,984® 1978 396® ® Information courtesy of M. Pelchat, B.C. Ministry of Forests, Silviculture Branch, Victoria, B.C. for 1981-1986. @ Information obtained from B.C. Ministry of Forests Annual Reports 1978-1980. @ Includes A. grandis and A. lasiocarpa. woolly aphid (Adelges piceae Ratz.). The present regulations 3, which are under review 4, attempt to l i m i t the i n f e s t a t i o n of balsam woolly aphid by r e s t r i c t i n g movement 3 Order in Council 44, Plant Protection Act, B.C. Reg. 7/77. 4 Peter Hall, Entomologist, B.C. Ministry of Forests, Protection Branch, Victoria, B.C. 6 of Abies species grown within a s p e c i f i e d quarantine zone and ensuring proper treatments to control the aphid are applied at the nursery. The role of P a c i f i c s i l v e r f i r i n future forests of coastal B.C. appears promising, but basic genetic studies of the species are lacking. A nine-provenance study was established at two test s i t e s on Vancouver Island i n 1980 by the Ministry of Forests. [To date, one, three and six year height data have been co l l e c t e d but not evaluated (CC. Ying, pers. comm., February 1990)]. A five-provenance test r e p l i c a t e d at four elevations near Squamish, B.C. was set up in 1978 and 1979 by the University of B.C. i n co-operation with the Ministry of Forests. These t r i a l s have been assessed for v a r i a t i o n i n bud-flushing dates and threshold temperatures (Worrall 1983). S i g n i f i c a n t l y lower threshold temperatures were found in high elevation provenances, suggesting adaptation to a short growing season. P a c i f i c s i l v e r f i r has been the subject of several recent systematic studies. Parker et al. (1979a) looked at va r i a t i o n i n needle and twig anatomy among f i v e populations growing i n northwestern B.C. (above 54°N). Monoterpene v a r i a b i l i t y in c o r t i c a l oleoresin has been assessed by Zavarin et al. (1973) as a means of detecting geographic differences in the species, and leaf and twig o i l terpenes were used to estimate r e l a t i v e v a r i a t i o n among trees and populations for 19 locations in the P a c i f i c Northwest (von 7 Rudloff and Hunt 1977). Needle flavonoids have been i d e n t i f i e d for use as potential taxonomic characters by Parker et al. (1979b). Somewhat more work on the ecology and s i l v i c a l c h a r a c t e r i s t i c s of P a c i f i c s i l v e r f i r has been published (Schmidt 1957, Dimock 1958, Krajina 1969, Kotar 1,972 and 1978, Murray and Treat 1979, Krajina et al. 1982, Packee et al. 1982, Martin 1985) and the biogeoclimatic units i t occupies have been well characterized (Packee 1974, Krumlik et al. 1982, Green et al. 1984). Many of these studies allude to or describe the v a r i a b i l i t y in s i t e conditions which the species occupies but, to date, l i t t l e work has been done to determine the genetic component of v a r i a t i o n i n adaptively important t r a i t s . The growth and s u r v i v a l of P a c i f i c s i l v e r f i r on d i f f e r e n t habitats at two elevations in the Cascade Range northwest of Seattle, WA, has been investigated by Kotar (1978) but conclusions regarding v a r i a b i l i t y i n seedling performance were r e s t r i c t e d because only one seed source was used. Seed source e f f e c t s on the germination of P a c i f i c s i l v e r f i r were part of a study by Leadem (1986) which examined several factors a f f e c t i n g what Edwards (1982) described as c h a r a c t e r i s t i c a l l y "poor" seed germinability i n t h i s and other Abies species. In Leadem's study, seedlot differences were found to affect germination s i g n i f i c a n t l y i n P a c i f i c s i l v e r f i r . 8 1.1 Study Objectives Given the increased u t i l i z a t i o n of P a c i f i c s i l v e r f i r for r e f o r e s t a t i o n and the lack of knowledge regarding the nature and extent of v a r i a t i o n i n the species, a c o l l e c t i o n of cones from i n d i v i d u a l trees was made i n the f a l l of 1983 to provide material for the present study. The o v e r a l l goal of the project was to elucidate patterns of v a r i a t i o n i n several biochemical, morphological and phys i o l o g i c a l characters of P a c i f i c s i l v e r f i r sampled from Vancouver Island. The s p e c i f i c objectives were: (i) to determine the mode of inheritance and linkage of e l e c t r o p h o r e t i c a l l y detectable enzyme l o c i , ( i i ) to estimate parameters of the mating system, ( i i i ) to estimate indices of genetic v a r i a t i o n using allozyme polymorphisms, (iv) to estimate the extent of genetic control over germination parameters, and (v) to describe h a l f - s i b progeny performance over one growing season and to examine variable i n t e r r e l a t i o n s h i p s , p a r t i c u l a r l y between maternal parent and progeny and c o r r e l a t i o n with geographic variables. 2. MATERIAL COLLECTION AND PROCESSING Coll e c t i o n s of cones from eight populations d i s t r i b u t e d on Vancouver Island from 49"18' to 50°42' l a t i t u d e and 150 to 945 m a l t i t u d e (Table 2.1) were made in August and September 1983. Locations are shown in Figure 2.1. Populations w i l l be referred to by t h e i r " l e t t e r " designation. Populations A, B and C each consisted of eight trees, population F - 10 trees, H, N and R - 13 trees and W - 17 trees. The c r i t e r i o n of 30 m spacing among sampled trees (Lines 1967) was adhered to for populations A, F and W where cones were c o l l e c t e d from the ground, however the remaining sampling was done using a cone rake suspended from a helicopter, and the minimum spacing could not be v e r i f i e d . However, P a c i f i c s i l v e r f i r was present as a codominant with western hemlock i n each of the f i v e stands and the age class i n a l l cases was mature(150+ years), so i t i s l i k e l y that s u f f i c i e n t distance between sampled trees was obtained i n these populations. The c o l l e c t i o n s were made over a four-week period, each population monitored for ripeness by f i e l d personnel. Nevertheless, there was a high degree of v a r i a b i l i t y i n cone maturity, among populations, among trees within a population and even within i n d i v i d u a l trees. Cones were kept i n mesh bags at 4°C u n t i l a l l c o l l e c t i o n s were obtained. They were Table 2.1 Loca t i o n of po p u l a t i o n s of P a c i f i c s i l v e r f i r from which cones were c o l l e c t e d f o r present study during a four-week p e r i o d (September 1983) . C o l l e c t i o n s i t e Reference code L a t i t u d e Longitude Average slevation (m) Average age (yr) Bio g e o c l i m a t i c designation® C o l l e c t i o n Agency® Ta y l o r R i v e r A 49' 18' 125* 22' 300 150+ CWH b 3 MacMillan B l o e d e l S e b a l h a l l Creek B ® 49- 57' 126* 25' 300 150+ CWH bl Canadian F o r e s t Products M a q u i l l a Creek c® 50" 03' 30" 126* 20' 30" 500 150 + CWH b l Canadian F o r e s t Products F l e e t R i v e r F 48- 39' 124- 06' 710 150+ CWH b 4 P a c i f i c Forest Products Hathaway Creek H ® 50* 34' 45" 127" 43' 45" 212 150+ CWH b l Western Forest Products NE62 (Holberg I n l e t ) N ® 50- 43' 30" 128* 0' 215 150+ CWHdl Western Forest Products Ronning Creek R® 50- 36' 30" 128* 11' 15" 275 150 + CWHdl Western Forest Products Mystery Creek W 48" 48' 15" 128" 09' 625 40 CWHbl B.C. Forest Products Green et al. (1984) and R.N. Green, Research P e d o l o g i s t , B.C. M i n i s t r y of Forests, pers. comm. Company names at the time of c o l l e c t i o n , September 1983. H e l i c o p t e r c o l l e c t i o n ; remaining s i t e s c o l l e c t e d from the ground (trees topped using r i f l e or climbed). F i g u r e 2.1 L o c a t i o n o f e i g h t s a m p l e d p o p u l a t i o n s o f P a c i f i c s i l v e r f i r on Vancouver I s l a n d , B.C. then a i r - d r i e d at 12 to 20°C for two weeks, during which time an average of eight intact cones were randomly selected for 79 trees from which there were s u f f i c i e n t cones, to enable seed y i e l d s per cone and morphological measures to be made. The hand separation of seeds from cone scales and y i e l d analyses using Softex-ray procedures were c a r r i e d out at the G.S. A l l e n Forest Genetics - Tree Seed Laboratory, Faculty of Forestry, University of B.C. Seeds from the remaining cones were extracted using the f a c i l i t y at Reid, C o l l i n s i n Aldergrove, B.C. These seeds were hand-dewinged and f i l l e d seed samples for weight determination, electrophoretic procedures and a germination/progeny test were obtained using a vacuum separation apparatus (Edwards 1979). Care was exercised at a l l stages of seed handling to avoid damage to the seed coats, as bursting of resin v e s i c l e s has been implicated i n reducing germinability of Abies species ( K i t z m i l l e r et a i . 1975). 13 3. ISOZYME INHERITANCE 3.1 Introduction Most, i f not a l l , economically important c h a r a c t e r i s t i c s of forest trees are known to be i n h e r i t e d quantitatively- These t r a i t s are polygenic i n nature, with large numbers of l o c i involved and where a l l genes may or may not contribute equally to t h e i r expression (Wehrhahn and A l l a r d 1965). Phenotypic v a r i a t i o n i n such polygenic characters as stem height, wood density and branch angle and size provide the raw material for a forest tree breeding program. Estimates of the extent to which observed v a r i a t i o n i n quantitative t r a i t s i s genetic i n o r i g i n and thus manipulable though breeding have t r a d i t i o n a l l y come from provenance and progeny tests and common garden experiments (Libby et al. 1969). These are, because of the long generation time of conifers, often slow and labour-intensive data c o l l e c t i o n methods (Mitton 1983). Forest genetics and tree improvement programs have benefitted from advances made in genetic analyses of agronomic species such as maize, as well as animal (including human) genetic studies. For the past three decades isozyme analyses, involving the electrophoretic separation of enzyme variants (Smithies 1955) combined with histochemical staining techniques (Hunter and Markert 1957), have provided an extremely useful set of molecular marker 14 l o c i to a l l these f i e l d s of study. Feret and Bergmann (1976) defined isozymes as e l e c t r o p h o r e t i c a l l y detectable enzyme variants which possess i d e n t i c a l or s i m i l a r c a t a l y t i c a c t i v i t i e s . Once the inheritance pattern for a given enzyme locus i s established, isozymes may be in f e r r e d to as allozymes (= a l l e l i c isozyme). Allozyme l o c i are characterized by t h e i r single-gene, "Mendelian" inheritance patterns, codominant expression (heterozygotes are distinguishable from homozygotes, thus, genotype i s i n f e r r e d d i r e c t l y from phenotype) and t h e i r occurrence i n a wide range of animal and plant species (Hartl 1980). Their u t i l i t y as markers of gene function and v a r i a t i o n i s r e f l e c t e d i n the dozens of population-based studies of enzyme polymorphisms which exist (reviewed by Selander 1976, Gottlieb 1979, Nevo 1978, Hamrick et al. 1979) and i n the numerous breeding applications i n forest trees (reviewed by Adams 1983, Paule 1990). P r i o r to the advent of molecular markers, g e n e t i c i s t s r e l i e d upon simply inherited morphological t r a i t s such as colour mutants i n attempts to associate marker locus with quantitative t r a i t v a r i a t i o n (Edwards et al. 1987). Single-locus morphological v a r i a t i o n i s r e l a t i v e l y uncommon i n gymnosperms (Mitton 1983). The use of monoterpene l e v e l s i n vegetative tissue, also shown to be controlled by single genes (Adams 1983), i s hindered by cumbersome a n a l y t i c a l 15 techniques and complicated by dominance in t h e i r expression (Squillace et al. 1980). Isozyme marker l o c i have been used to successfully associate quantitatively inherited t r a i t s i n tomato (Tanksley et al. 1982, Tanksley and Hewitt 1988) and maize (Edwards et al. 1987, Stuber et al. 1987) and a newer class of molecular (DNA) markers known as r e s t r i c t i o n fragment length polymorphisms (RFLPs) have also been used to detect linked genes c o n t r o l l i n g quantitative v a r i a t i o n i n tomato (Osborn et al. 1987, Tanksley and Hewitt 1988) and maize (Edwards et al., unpublished, c i t e d i n Helentjaris 1987). Gusella et al. (1983) reported a polymorphic DNA marker linked to Huntington's disease in humans. In fact, the model for u t i l i z i n g RFLPs to construct genetic linkages and linkage maps i n plant species stemmed from human genetic studies (Botstein et al. 1980, Helentjaris 1987). Tanksley and Hewitt (1988) suggest that once s u f f i c i e n t l y t i g h t linkage between genes c o n t r o l l i n g a molecular marker and a quantitative t r a i t locus (QTL) i s found, then the chromosomal segment with the desired QTL can be transferred into d i f f e r e n t l i n e s or v a r i e t i e s . Helentjaris (1987) predicts the eventual cloning of such l o c i by chromosome walking or jumping (Poustka et al. 1987) techniques. Such advancements i n the breeding of agronomic species have been aided by features of the organisms' biology which have enabled rapid characterization of t h e i r genomes. The 16 development of highly inbred l i n e s , e s p e c i a l l y e f f i c i e n t i n s e l f - p o l l i n a t i n g species such as tomato, and the presence of chromosomal aberrations such as d e f i c i e n c i e s ( i . e . , nullisomics i n maize) or e a s i l y i d e n t i f i a b l e translocations have resulted i n well developed linkage maps for these species. In contrast, the cytogenetics of most forest tree species are not well known, karyotypes are poorly characterized and construction of isozyme-based linkage maps only barely begun. To date even the a l l o c a t i o n of marker l o c i to p a r t i c u l a r chromosomes has not been possible. With current forestry research being steered toward "biotechnological" development of genetically improved species through various state-of-the-art techniques borrowed from plant and animal breeders, there may be great pressure to implement techniques which are predicated upon knowledge of the conifer genome not yet at hand. Prior to the u t i l i z a t i o n of molecular l o c i in assembling linkage maps, t h e i r inheritance patterns must be known (Cheliak and P i t e l 1985) . . , Inheritance studies in gymnosperms are p a r t i c u l a r l y e f f i c a c i o u s owing to the presence of haploid, megagametophyte tissue in seeds, which possesses the same genetic c o n s t i t u t i o n as the female gamete (Foster and G i f f o r d 1974). Assuming regular meiosis, no gametic selection and random sampling of mature seeds, the electrophoretic assay of s u f f i c i e n t megagametophytes of 17 mother trees heterozygous for a given enzyme locus should produce equal numbers of the two alternate ( a l l e l i c ) forms. A goodness-of-fit test i s then employed to v e r i f y the 1:1 segregation. Where enzymes are functional i n embryo (diploid) tissue, the quaternary structure may be deduced (El-Kassaby 1980, M i l l a r 1985). In t h i s chapter, isozyme band phenotypes are described for 13 putative l o c i i d e n t i f i e d in P a c i f i c s i l v e r f i r and compared to related species, balsam and Fraser f i r , and other conifers. Mode of inheritance i s either demonstrated or, for invariant enzymes, inferred. Isozyme origins and deviations from Mendelian segregation are discussed. 3.2 Materials and Methods Wind-pollinated seeds were obtained from cones of 87 P a c i f i c s i l v e r f i r trees c o l l e c t e d in eight locations on Vancouver Island (as described in Chapter 2). F i l l e d seeds were separated by soft x-ray and stored at 2°C. P r i o r to electrophoresis, seeds were soaked for 48 hr in deionized water at 20°C and then held at 3°C for one to four days. Embryos were separated from corresponding megagametophytes and i n d i v i d u a l l y homogenized i n one drop of extraction buffer (El-Kassaby et al. 1982b). Electrophoretic techniques followed those of Conkle et al. (1982) u t i l i z i n g two buffer systems: 'A' - T r i s - c i t r a t e (pH 7.0) following El-Kassaby et al. (1982b) and 'B' - Sodium borate (pH 8.0), 18 modified from Conkle et al. (1982) by d i l u t i o n of the gel buffer to one-half strength with deionized water. Horizontal starch gels (12.5% w/v) were used for both systems. The enzymes, t h e i r abbreviations and enzyme commission numbers, and the appropriate buffer systems used are l i s t e d i n Table 3.1. Stain recipes and substrate preparation followed Conkle et al. (1982). Detailed descriptions of electrophoretic procedures and recipes are given i n Appendix 2. Where possible, 20 seeds from each tree were assayed, however eight trees had 19 seeds with s u f f i c i e n t enzyme a c t i v i t y and i n two cases, only 18 seeds were ava i l a b l e . Assuming no errors i n detection, the p r o b a b i l i t y of mi s c l a s s i f y i n g a heterozygote at any one locus i s 0.5 for k megagametophytes (Tigerstedt 1973). Eighteen to 20 seeds affords a large enough sample to make t h i s l i k e l i h o o d extremely small, unless severe segregation d i s t o r t i o n exists (Millar 1985). For those l o c i where no v a r i a t i o n was observed, the band phenotypes in megagametophytes were compared to embryo ti s s u e and, where possible, to reports of inheritance i n rel a t e d species. Where heterozygous trees were i d e n t i f i e d , a l o g - l i k e l i h o o d G-test (Sokal and Rohlf 1981) was used to test the goodness-of-fit of the observed segregation r a t i o to the 1:1 a l l e l i c segregation predicted for l o c i conforming to Mendelian expectations (ME) and, where more than one tree 19 exhibited heterozygosity at a p a r t i c u l a r locus, a heterogeneity G-test (Sokal and Rohlf 1981) was performed to te s t the homogeneity of segregation among trees. Enzyme l o c i are referred to by t h e i r t h r e e - l e t t e r abbreviation (Table 3.1) and where more than one locus i s resolved for a given enzyme system, the most anodally-migrating i s assigned the number l . 5 Isozyme designation followed the method of El-Kassaby et al. (1987b). At a given locus, the most frequently occurring band was designated '1' and bands migrating slower and f a s t e r were assigned odd and even numbers, respectively. Preliminary experiments a) determined that there was no substantial change in isozyme presence or resolution i n seeds held at 3°C for up to four days and b) eliminated Lithium borate (pH 8.3) and T r i s - c i t r a t e (pH 6.2) buffer systems of Conkle et al. (1982), as too few l o c i could be r e l i a b l y scored. T h i s i s t h e most common but by no means s t a n d a r d p r a c t i c e i n i n h e r i t a n c e s t u d i e s . O'Malley et al. (1979) used a s c o r i n g s ystem which numbered t h e l e a s t - a n o d a l l o c u s as 1. H a r r y (1986) a d o p t e d t h e most common system i n h i s t e x t but showed m i g r a t i o n t o w a r d t h e o p p o s i t e e l e c t r o d e i n h i s f i g u r e s . T h i s s u g g e s t s t h a t c a r e must be e x e r c i s e d when comparing p a t t e r n s a c r o s s s p e c i e s and r e s e a r c h e r s . 20 Table 3.1 Enzyme systems, t h e i r abbreviations as used i n the text, Enzyme Commission reference number, buffer system used and the number of l o c i observed and those consistently scorable for each system. Enzyme No. l o c i Enzyme system Commission Buffer (abbr.) Number System® Observed Scored Aspartate amino transferase (AAT) 2.6.1.1 B Glucose-6-phosphate dehydrogenase (G6P) 1.1.1.49 B Glutamate dehydrogenase (GDH) 1.4.1.3 B Is o c i t r a t e dehydrogenase (IDH) 1.1.1.42 A Malate dehydrogenase (MDH) 1.1.1.37 A 6-Phosphogluconate dehydrogenase (6PG) 1.1.1.4 4 A Phosphoglucose isomerase (PGI) 5.3.1.9 A Phosphoglucomutase (PGM) 2.7.5.1 A Superoxide dismutase (SOD) 1.15.1.1 B ® Composition of buffer solutions and s t a i n recipes are given in Appendix 2. 21 3.3 Results and Discussion Nine enzyme systems yielded clear and consistently scorable bands for 13 putative l o c i . An additional band phenotype was detected but could not be r e l i a b l y scored. 3.3.1 Enzymes classed as monomorphic Six l o c i (AAT-1 and 3, GDH, SOD, MDH-3 and 6PG) i n f i v e enzyme systems displayed invariant band patterns. (a) Aspartate-amino transferase (AAT) AAT, equivalent to glutamate oxalacetate transaminase (Cheliak and P i t e l 1985), was found to have three anodally-migrating l o c i . AAT-1 appeared invariant and single banded in a l l trees. A similar locus was detected i n balsam f i r (Neale and Adams 1981, Jacobs et al. 1984) and Fraser f i r (Jacobs et al. 1984), and found to be d i a l l e l i c in both species. In d i p l o i d heterozygotes, the presence of an intermediate, hybrid band in addition to the a l l e l i c forms contributed by both gametes indicates that the enzyme i s dimeric i n structure (Guries and Ledig 1978). Neale and Adams (1981) detected t h i s structure for AAT-1 in balsam f i r . Shea (1988) found AAT-1 to be invariant in a sample of subalpine f i r , but did not describe the band pattern. This locus was reported to be monomorphic in samples of blue and Engelmann spruce (Ernst et al. 1987), whitebark pine 22 (Furnier et al. 1986), tamarack (Cheliak and P i t e l 1985) and Douglas-fir (El-Kassaby et al. 1982b). In contrast, AAT-3 was found in P a c i f i c s i l v e r f i r to be double banded and monomorphic over a l l sampled trees. The bands appeared i n the same position on the gel for embryo tissue, but were poorly resolved. In balsam f i r t h i s locus also displayed a double band pattern, but two a l l e l e s were detected (Neale and Adams 1981). Embryos were also unclear, however the segregation patterns of heterozygous mother trees confirmed single-locus Mendelian inheritance. Double and t r i p l e band phenotypes in haploid megagametophyte tis s u e have been detected at t h i s locus in several other conifers (Cheliak and P i t e l 1985, M i l l a r 1985, Furnier et al. 1986, Harry 1986, Ernst et al. 1987). Dimeric structure has also been demonstrated (Rudin 1975, Guries and Ledig 1978). AAT-3 was found to be monomorphic in subalpine f i r (Shea 1988) but the structure was not reported. El-Kassaby et al. (1982b) reported a single band phenotype for Douglas-fir at AAT-3, with two a l l e l e s and detected a fourth, cathodally migrating zone which was found to covary with AAT-3. O'Malley et al. (1979) found a si m i l a r pattern i n ponderosa pine and postulated that both zones were controlled by a single locus. Two anodally-migrating bands, designated AAT-1 and AAT-2, were detected in p i t c h pine, and a t h i r d , cathodal zone was found to match the v a r i a t i o n i n AAT-2 exactly (Guries and Ledig 1978). 23 However, when t h i s species was assayed using the same running conditions for l o b l o l l y pine (Adams and Joly 1980a), t r i p l e banded allozymes were observed. These r e s u l t s suggest that buffer pH and/or composition as well as electrophoretic running conditions may a f f e c t band resolution and/or position in the gel. Further, Ernst et al. (1987) observed triple-banded phenotypes for AAT-3 in blue and Engelmann spruce haploid tissue while only two bands were detected in d i p l o i d embryo and bud tissues of both species. They determined that both double and t r i p l e -banded allozymes are inherited as a single a l l e l e and suggested that some post-translational modification of a single allozyme (apparently d i s s i m i l a r modifications i n d i f f e r e n t tissues) may be the cause of such complex phenotypes. Another possible explanation may be that multiple bands are the r e s u l t of duplicated and necessarily t i g h t l y linked l o c i (Harry 1986) . Duplication of l o c i and subsequent mutation i s considered by Markert and Whitt (1968), Scandalios (1975) and Gottlieb (1982) to be the probable o r i g i n of many isozymes. (b) Glutamate dehydrogenase (GDH) GDH was found to be single banded and considered invariant i n t h i s study, although one tree displayed what appeared to be a faster-migrating band in one of 20 24 megagametophytes assayed. The migration distances under the electrophoretic conditions of t h i s study were so close that the band pattern of the corresponding embryo was impossible to characterize. One zone of a c t i v i t y with no v a r i a t i o n was observed i n balsam, Fraser and subalpine f i r (Neale and Adams 1981, Jacobs et al. 1984, Shea 1988) and several other conifer species (El-Kassaby et al. 1982b, Cheliak and P i t e l 1985, M i l l a r 1985, El-Kassaby et al. 1987b, Furnier et al. 1986, Harry 1986, Strauss and Conkle 1986). However, t h i s locus was found to possess two a l l e l e s and possibly be multimeric in Engelmann spruce (Ernst et al. 1987),, in white spruce (Stewart and Schoen 1986) and in black spruce, where double-banded embryos suggested a monomeric structure (Boyle and Morgenstern 1985). Five a l l e l e s at GDH and an apparent multimeric structure were reported for l o b l o l l y pine (Guries and Ledig 1978). A larger sample of P a c i f i c s i l v e r f i r may confirm a l l e l i s m at t h i s locus also. (c) Superoxide dismutase (SOD) SOD appeared as a single, reverse-staining band on gels stained for GDH. It was monomorphic and occupied the same gel p o s i t i o n i n both tissue types. This enzyme was not studied i n balsam or Fraser f i r (Neale and Adams 1981, Jacobs et al. 1984). SOD has been detected i n several members of the Pineaceae, displaying one invariant band i n bishop pine (Millar 1985) and white spruce (King and Dancik 1983), and one zone with three a l l e l e s in Douglas-fir ( E l -25 Kassaby et al. 1982b). Three invariant bands were observed for t h i s enzyme i n whitebark pine (Furnier et al. 1986) and knobcone pine (Strauss and Conkle 198 6), although two of the three bands were described as f a i n t . Since no v a r i a t i o n existed, these bands could not be characterized as d i s t i n c t l o c i i n P a c i f i c s i l v e r f i r . (d) Malate dehydrogenase (MDH) Three l o c i were resolved for MDH and the slowest-migrating locus (MDH-3) appeared monomorphic and si n g l e -banded i n the megagametophytes of a l l trees sampled i n t h i s study. A c t i v i t y at t h i s locus was observed in embryos, but bands did not resolve c l e a r l y enough to score. Neale and Adams (1981) reported only two MDH l o c i in balsam f i r ; both single banded and invariant. MDH-2 was found to be inactive i n balsam f i r embryo tissue. Two zones were also found i n the balsam and Fraser f i r s sampled by Jacobs et al. (1984). MDH-2 was found to be polymorphic for three a l l e l e s . El-Kassaby (1981) reported three l o c i i n MDH for white f i r and grand f i r . Three to four l o c i are often observed i n t h i s enzyme system i n other conifers (Simonsen and Wellendorf 1975, O'Malley et al. 1979, El-Kassaby et a l . 1982b, Cheliak et al. 1984, Boyle and Morgenstern 1985, Cheliak and P i t e l 1985, M i l l a r 1985, El-Kassaby et al. 1987b, Harry 1986 and Ernst et al. 1987). In addition, a band described as a non-genetic inter-locus heterodimer has 26 been detected i n the MDH complex of many conifers (reviewed i n El-Kassaby 1981, El-Kassaby et al. 1982b, Boyle and Morgenstern 1985, Cheliak and P i t e l 1985 and Ernst et a i . 1987) . In a comparative study of four genera of Pineaceae, El-Kassaby (1981) found that the three l o c i of white and grand f i r were equivalent to MDH-1, 2 and 4 of Douglas-fir and that the least anodal locus, designated MDH-4, was not well resolved i n grand f i r . He also found that the heterodimer band was absent in both Abies species he assayed. Interestingly, no heterodimer band (usually forming between the second and t h i r d locus) was detected i n P a c i f i c s i l v e r f i r . As well as n o n - a l l e l i c interactions in t h i s enzyme system, genes which appear to a l t e r the migration patterns of MDH l o c i have been reported, in maize (Newton 1979), incense-cedar (Harry 1983) and bishop pine (Millar 1 9 8 5 ) . Guries and Ledig (1978) determined that in p i t c h pine there i s l o c a l i z a t i o n of MDH enzymes such that mitochondria contain only the MDH-2 locus, while crude tissu e preparations y i e l d both MDH-1 and MDH-2. Similar subcellular compartmentalization has been observed i n maize where three unlinked MDH l o c i produce mitochondrial isozymes and another two unlinked l o c i are cytoplasmic in o r i g i n (Goodman et al. 1980, Newton and Schwartz 1980). Enzyme 27 duplication and l o c a l i z a t i o n in plants has been examined i n an evolutionary context by Gottlieb (1982 and references c i t e d ) . He concluded that where gene products are produced i n the same subcellular compartment, both i n t r a - and i n t e r -locus hybrid enzymes form, whereas such hybrid molecules do not form between enzymes of d i f f e r e n t compartments. The band phenotypes of the three MDH l o c i observed i n P a c i f i c s i l v e r f i r suggest that they may be contained within d i f f e r e n t subcellular organelles. A c e l l f r a c t i o n a t i o n study may reveal information to support t h i s hypothesis. In addition, the presence of an enzyme locus i n megagametophyte but not embryo tissue (Adams and Joly 1980a) or vice versa (Millar 1985) suggests d i f f e r e n t i a l expression ontogenetically, which makes the MDH system a good candidate for studies of gene regulation. The complexity of the MDH enzyme system i s evidenced also by the existence of multimerism at some or a l l of the l o c i reported i n many conifers, however, t h i s cannot be confirmed i n the monomorphic locus found in P a c i f i c s i l v e r f i r . (e) 6-Phosphogluconate dehydrogenase (6PG) Gels stained for 6PG showed one zone of a c t i v i t y . Gametophytes were single banded and invariant across a l l trees sampled in t h i s study. A similar band pattern was found i n balsam f i r by Neale and Adams (1981), with two 28 a l l e l e s observed. However, two variable l o c i , each with two a l l e l e s , were detected in populations of balsam/Fraser f i r by Jacobs et al. (1984) and a single population of subalpine f i r by Shea (1988). These two d i f f e r i n g reports resolved 6PG on d i f f e r e n t buffer systems ( T r i s - c i t r a t e pH 6.2 vs. pH 8.0) which suggests that the detection of l o c i i n t h i s enzyme system i s pH-sensitive. This phenomenon was also observed i n tamarack by Cheliak and P i t e l (1985). Other conifers commonly exhibit two zones for 6PG (Guries and Ledig 1978, O'Malley et al. 1979, El-Kassaby et al. 1982b, King and Dancik 1983, Cheliak et al. 1984, Boyle and M'orgenstern 1985, M i l l a r 1985, Harry 1986, Stewart and Schoen 1986 and Ernst et al. 1987), and dimeric structures are often reported. Three l o c i were reported i n knobcone pine (Strauss and Conkle 1986) and a single zone i n l o b l o l l y pine (Adams and Joly 1980a). 3.3.2 Enzymes' exhibiting polymorphism Seven l o c i i n six enzyme systems displayed varying band patterns under the electrophoretic conditions of t h i s study. The r e l a t i v e p o s i t i o n i n g of the band phenotypes with respect to t h e i r o r i g i n a l placement on the starch gel i s pictured i n Figure 3.1. 29 r-F AAT G6P Aat-2 2 G6p 2 1 3 5 MDH PGI M d h - 1 2 Mdh-2 2 1 ^ ™ Pgi-2 1 3 L-0 PGM IDH P g m 1 2 ™ " " ™ ' 3 5 Idh 1 2 3 Figure 3.1 Representations of band patterns observed in seven polymorphic loci found in trees from eight popula-tions of Pacific silver f i r on Vancouver Island, British Columbia. Band positions indicate relative migration distance between the origin (0) and the buffer front (F). For each locus, the band desig-nated *1' occurred most frequently. 30 (a) Glucose-6-phosphate dehydrogenase (G6P) This enzyme exhibited a single band phenotype with a t o t a l of four allozymes when assayed i n P a c i f i c s i l v e r f i r . C l e a rly staining embryos showed t r i p l e band patterns i n heterozygotes, i n d i c a t i n g dimerism. G6P was not studied i n balsam or Fraser f i r (Neale and Adams 1981, Jacobs et al. 1984). One zone of a c t i v i t y with two to f i v e a l l e l e s has been found i n jack pine (Cheliak et al. 1984), western white pine (El-Kassaby et a i . 1987b), incense-cedar (Harry 1986), black spruce (Boyle and Morgenstern 1985) and Douglas-fir (El-Kassaby et al. 1982b). Two zones of a c t i v i t y were reported i n knobcone pine (Strauss and Conkle 1986), bishop pine (Millar 1985) and ponderosa pine (O'Malley et al. 1979), with only one being variable in each species. Cheliak et al. (1984) observed some ontogenetic v a r i a t i o n i n t h i s enzyme system in jack pine. A c t i v i t y was i n i t i a l l y detected i n one zone and as germination progressed (days four to ten) the a c t i v i t y disappeared from that zone to a more anodal region. This report suggests that more than one locus may be present, but may also be developmentally unstable. Seeds of P a c i f i c s i l v e r f i r were assayed at approximately the same state of germination, and a l l trees consistently presented only one zone of a c t i v i t y . Segregation analyses of the four d i f f e r e n t a l l e l i c combinations (Table 3.2) revealed no s i g n i f i c a n t deviation 31 from 1:1 r a t i o s . For the two combinations detected i n more than one tree heterogeneity tests showed s i g n i f i c a n t Table 3.2 Observed segregation in megagametophytes of heterozygous mother trees and G tests of goodness-of-fit to 1:1 r a t i o (df = 1) and heterogeneity among trees (df i n parentheses). Locus A l l e l i c Observed Deviation Hetero-Combinationt Segregation geneity G G AAT-2 1/2 224:53 113.57** 22.25 (13) G6P 1/2 12:8 0.81 — 1/3 44 :36 0.80 16.75M3) 1/5 315:321 0.06 64.79* (32) 2/5 8:12 0.81 — IDH 1/2 12:8 0.81 _ 1/3 22: 18 0.40 3.69 (1) 1/5 12:7 1.33 -MDH-1 1/2 116:102 0.90 28.02 (10) 1/3 9:11 0.20 -MDH-2 1/2 13:7 1.83 -PGI-2 1/3 409:404 0.03 63.09** (39) PGM 1/2 8:12 0.81 — 1/3 227:210 0. 66 44.02 (21) 1/5 21:19 0.10 0.10 (1) t Most common alle l e designated '1'; faster and slower alleles given even and odd numbers, respectively. * Significant, P < 0.05. ** Highly significant, P < 0.01. 32 v a r i a t i o n among trees. Individual goodness-of-fit G tests indicated that f i v e of the 33 trees (15%) heterozygous for a l l e l e s '1' and '5' did not conform to ME and one of the four trees possessing a l l e l e s '1' and '3' also exhibited a s i g n i f i c a n t deviation from 1:1 segregation. In 1/5 heterozygotes the deviant r a t i o s were not uniformly skewed in favour of one a l l e l e . Adams and Joly (1980a) suggest that when a p a r t i c u l a r genotype displays a deficiency of an a l l e l e consistently over several parents i t may indicate that the a l l e l e , or some portion of the chromosome linked to i t , i s being selected against. Where no p a r t i c u l a r pattern in the unequal a l l e l e frequencies presents i t s e l f , heterogeneity i n the pollen pool may be a cause for the v a r i a t i o n i n segregation d i s t o r t i o n (Cheliak et al. 1984, Stewart and Schoen 1986). The maternal genotype i s determined from viable seeds and, in turn, embryo v i a b i l i t y determines the survivorship of the female gametophyte. Germinable seeds are thus products of a maturation process wherein some form of embryonic selection might occur, giving r i s e to a disproportionate number of favoured a l l e l e s , depending upon pollen d i s t r i b u t i o n and v i a b i l i t y . Adams and Joly (1980a) also suggest that d i f f e r i n g genetic backgrounds of female gametes may produce d i f f e r e n t degrees of incompatibility among trees. Polyembryony i s known to occur i n P a c i f i c s i l v e r f i r (Owens and Molder 1977) and Sorensen (1982) maintains that in noble f i r the presence 33 o f more t h a n one o v u l e a l l o w s c o n s i d e r a b l e p o s t -f e r t i l i z a t i o n s e l e c t i o n t o t a k e p l a c e . Because t h e m a j o r i t y o f h e t e r o z y g o u s t r e e s conformed t o ME, i t seems r e a s o n a b l e t o assume t h a t G6P i s under s i n g l e - l o c u s c o n t r o l i n b o t h seed t i s s u e s o f P a c i f i c s i l v e r f i r , a l t h o u g h s i n g l e t r e e d e v i a t i o n s suggest t h i s l o c u s may be s u b j e c t t o some s e l e c t i v e i n f l u e n c e . (b) I s o c i t r a t e dehydrogenase (IDH) IDH e x h i b i t e d a s i n g l e band phenotype w i t h f o u r a l l e l e s , t h r e e o f which, a l t h o u g h v e r y c l e a r l y s t a i n i n g , o c c u r r e d i n v e r y few t r e e s (Table 3.2). Embryos, a l s o d i s t i n c t l y r e s o l v e d , showed a t r i p l e band p a t t e r n when h e t e r o z y g o u s , s u g g e s t i n g a d i m e r i c s t r u c t u r e f o r t h i s enzyme. S i m i l a r b a n d i n g was o b s e r v e d i n balsam f i r (Neale and Adams 1981) a l t h o u g h o n l y one h e t e r o z y g o u s t r e e was d e t e c t e d . Jacobs e t al. (1984) found t h e i r samples o f b a l s a m / F r a s e r f i r i n v a r i a n t at t h i s l o c u s as d i d Shea (1988) i n a sample o f s u b a l p i n e f i r . Dimerism i n IDH has been o b s e r v e d i n Engelmann s p r u c e ( E r n s t e t al. 1987), D o u g l a s -f i r ( E l - K a s s a b y e t al. 1982b), b l a c k spruce (Boyle and M o r g e n s t e r n 1985), w h i t e s p r u c e (King and Dancik 1983), ponderosa p i n e (O'Malley et al. 1979) and p i t c h p i n e ( G u r i e s and L e d i g 1978). More t h a n one zone of a c t i v i t y of IDH has been o b s e r v e d i n Engelmann s p r u c e ( E r n s t et al. 1987) and i n c e n s e - c e d a r 34 (Harry 1986). IDH exhibited a double band phenotype i n bishop pine (Millar 1985) and Strauss and Conkle (1986) found t h i s enzyme to be single or double banded depending on the buffer system employed. Further, Strauss and Conkle (1986) noted that one of two l o c i resolved i n staining of phosphoglucomutase (PGM) produced i d e n t i c a l v a r i a t i o n to IDH. O'Malley et al. (1979) reported that f a i n t IDH patterns appeared on many gels stained for enzymes requiring NADP (nicotinamide adenine dinucleotide phosphate). The c i t r i c acid found i n the buffer used was presumed to be the substrate. In Douglas-f i r , El-Kassaby et al. (1982b) v e r i f i e d that the second, f a i n t e r zone on PGM gels was, in fact, IDH, not a second PGM locus. These studies suggest that the composition of buffer systems should be considered whenever species comparisons are made for t h i s enzyme system. Table 3.2 shows that for the three a l l e l i c combinations detected i n IDH, there was no deviation from the expected segregation r a t i o and that homogeneity existed among trees, where possible to t e s t . The IDH locus in P a c i f i c s i l v e r f i r may then be assumed to exhibit simple, Mendelian inheritance. 35 (c) Malate dehydrogenase (MDH) MDH-1 was found to be single banded with three a l l e l e s i n P a c i f i c s i l v e r f i r . Eleven trees were found to be heterozygous for common and fast a l l e l e with homogeneous segregation r a t i o s among trees and no deviation from the expected 1:1 segregation (Table 3.2). A second, slower migrating zone of a c t i v i t y (MDH-2) was found to be variable in one tree only. The observed segregation for the common and fast a l l e l e at t h i s locus did not d i f f e r from ME (Table 3.2). For both l o c i , resolution was not d i s t i n c t enough to permit a st r u c t u r a l interpretation of embryo band patterns. Neale and Adams (1981) and Jacobs et al. (1984) both reported only two single banded l o c i for t h i s enzyme system i n balsam f i r and populations of balsam and Fraser f i r . MDH-1 was invariant i n both studies and, while Jacobs et al. (1984) found three a l l e l e s at MDH-2, Neale and Adams (1981) found megagametophytes to be monomorphic and no a c t i v i t y i n embryo tissue at t h i s locus. Three l o c i were observed i n P a c i f i c s i l v e r f i r , with, as previously noted, very poor resolution i n embryos at MDH-3. The assay of seeds from several Abies species under the same electrophoretic conditions (similar to El-Kassaby 1981) would enable more accurate comparisons across true f i r species and perhaps shed some l i g h t on the evolution of th i s enzyme system r e l a t i v e to other members of the Pineaceae. 36 (d) Phosphoglucose isomerase (PGI) Two regions of a c t i v i t y were observed on gels stained for PGI. What may be considered as PGI-1 was very blurry and could not be resolved well enough to score. One band was evident, but there were dif f u s e bands leading and t r a i l i n g the area of dark staining. This was less apparent i n embryos, but a l l e l i c v a r i a t i o n s t i l l could not be r e l i a b l y detected. Inconsistent staining at t h i s putative locus has been reported in blue and Engelmann spruce (Ernst et al. 1987), black spruce (Boyle and Morgenstern 1985), tamarack (Cheliak and P i t e l 1985) and Douglas-fir ( E l -Kassaby et al. 1982b). The locus was resolved in balsam, Fraser and subalpine f i r as a single banded invariant zone in a l l three species (Neale and Adams 1981, Jacobs et al. 1984, Shea 1988). PGI-1 exhibited apparent non-genetic v a r i a t i o n i n Douglas-fir controlled cross progeny (Neale et al. 1984). The slower migrating zone (PGI-2) had two a l l e l e s present i n P a c i f i c s i l v e r f i r . Heterozygous embryos, although somewhat blurry, suggested a dimeric structure for t h i s enzyme. Shea (1988) found 2 a l l e l e s at t h i s locus i n subalpine f i r . This locus was also detected i n both balsam and Fraser f i r , with three a l l e l e s , one very rare, found i n both studies (Neale and Adams 1981, Jacobs et al. 1984). Neale and Adams (1981) were not able to v e r i f y a dimeric structure for t h i s enzyme although i t i s demonstrated i n 37 several other conifers (Guries and Ledig 1978, El-Kassaby et al. 1982b, Neale et al. 1984, Boyle and Morgenstern 1985). The pooled segregation data for PGI-2 (Table 3.2) showed no deviation from the expected 1:1 r a t i o although heterogeneity among trees was s i g n i f i c a n t . Individual goodness-of-fit G tests revealed that only six of 40 heterozygous trees exhibited deviant segregation r a t i o s and there was no trend to one a l l e l e or the other among trees. This pattern of segregation suggests single-locus control of PGI-2 i n P a c i f i c s i l v e r f i r . (e) Phosphoglucomutase (PGM) PGM appeared as a clear, single banded region of a c t i v i t y with four a l l e l e s detected in P a c i f i c s i l v e r f i r . Double-banded phenotypes from heterozygous embryo tissue indicate a monomeric structure for t h i s enzyme. Two zones of a c t i v i t y were found by Neale and Adams (1981) i n balsam f i r . PGM-1 was single banded with two a l l e l e s and monomeric s t r u c t u r a l l y whereas PGM-2 was double banded and invariant. Two a l l e l e s were detected at PGM-1 in subalpine f i r but the PGM-2 locus was too f a i n t to be scored (Shea 1988). In white spruce, t h i s second locus was observed i n seed but not vegetative tissue (Stewart and Schoen 1986). Neale et al. (1984) determined that PGM-2 in Douglas-fir i s a d i s t i n c t , independent locus when resolved at pH's higher than that 38 used by El-Kassaby et al. (1982b), where IDH bands no longer cover the f a i n t e r PGM-2 bands. Table 3.2 shows no deviation from the expected r a t i o for any of the three a l l e l i c combinations and no heterogeneity among trees where possible to t e s t . These res u l t s suggest that, under the electrophoretic conditions of t h i s experiment, PGM appears to exhibit Mendelian inheritance i n P a c i f i c s i l v e r f i r . (f) Aspartate-animo transferase (AAT) AAT-2 was the only band among three resolved on gels stained for AAT that was variable with two a l l e l e s detected i n P a c i f i c s i l v e r f i r . Heterozygous embryos did not st a i n c l e a r l y enough to permit any inference on structure at t h i s locus. Neale and Adams (1981) report a single band with two a l l e l e s and dimeric structure at AAT-2 in balsam f i r . A sim i l a r pattern was detected by Jacobs et al. (1984) i n balsam/Fraser f i r . Of three l o c i resolved i n subalpine f i r , only AAT-2 displayed a l l e l i c v a r i a t i o n (Shea 1988). Three zones of a c t i v i t y have been observed for AAT i n several other conifers (O'Malley et al. 1979, El-Kassaby et al. 1982b, Cheliak and P i t e l 1985, M i l l a r 1985, El-Kassaby et al. 1987b, Furnier et al. 1986, Strauss and Conkle 1986 and Ernst et al. 1987). AAT-2 was found to be inactive i n embryos of black spruce (Boyle and Morgenstern 1985) and O'Malley et al. (1979) suggested isozyme l o c a l i z a t i o n i n • 39 s p e c i f i c tissues for t h i s enzyme system. T i s s u e - s p e c i f i c a c t i v i t y l e v e l s of AAT isozymes were detected in lodgepole pine ( P i t e l et a i . 1984) and org a n e l l e - s p e c i f i c a c t i v i t y of AAT isozymes has been observed in other plant species (Huang et al. 1976, Hart and Langston 1977). In AAT-2, two a l l e l e s , common and fast, were observed to segregate i n 14 of the 87 trees sampled. Segregation r a t i o s were found to be homogeneous among trees (Table 3.2). When pooled, however, these data exhibited a highly s i g n i f i c a n t deviation from the hypothesized 1:1 r a t i o . It was found that 13 of the 14 trees showed s i g n i f i c a n t i n d i v i d u a l deviations from ME and those i n d i v i d u a l r a t i o s were uniformly and heavily skewed in favour of the common a l l e l e (216:43). Segregation d i s t o r t i o n in AAT-2 has been reported i n several conifers (Rudin 1975, Rudin and Ekberg 1978, Witter and Feret 1978, O'Malley et al. 1979, Boyle and Morgenstern 1985, Adams and Joly 1980a, Harry 1986, Strauss and Conkle 1986). Although there i s a s t a t i s t i c a l l i m i t placed on the magnitude of deviation from ME a sample as small as 20 seeds i s able to detect (Mulcahy and Kaplan 1979), these data suggest strongly that either there are se l e c t i v e differences between the two isozymes themselves (Strauss and Conkle 1986, Cheliak et al. 1984) or a distorted or " s e l f i s h " a l l e l e (Strauss and Conkle 1986) may be linked to t h i s locus, r e s u l t i n g in an over-representation of the common 40 a l l e l e i n sampled megagametophytes. The apparent segregation d i s t o r t i o n was not confined to trees of just one or even a few populations i n th i s study. From t h i s evidence i t appears that AAT-2 i n P a c i f i c s i l v e r f i r samples from Vancouver Island does not exhibit the c l a s s i c a l Mendelian inheritance pattern. The homogeneity among trees for the di r e c t i o n of d i s t o r t i o n tends to support some genetic mechanism such as linkage to a so-called s e l f i s h a l l e l e , which acts to increase i t s own frequency and those to which i t may be linked (Strauss and Conkle 1986) , however non-genetic causes of the observed d i s t o r t i o n cannot be ruled out without further study. 3.4 Conclusions (a) Band patterns of invariant enzyme l o c i in P a c i f i c s i l v e r f i r (AAT-1 and 3, GDH, SOD, MDH-3 and 6PG) appear to be simi l a r to other conifers, but may not be "good" Mendelian markers without v e r i f i c a t i o n of inheritance. (b) The polymorphic enzymes IDH, MDH-1, MDH-2 and PGM appear to exhibit segregation patterns conforming to Mendelian expectations, while G6P, PGI-2 and AAT-2 show some segregation d i s t o r t i o n . It cannot be ascertained whether segregation d i s t o r t i o n i s due to female gametophytes only (i.e., gametic selection) or to the heterogeneous d i s t r i b u t i o n of l e t h a l or semi-lethal 41 a l l e l e s i n the pollen pool. But, for G6P and PGI-2, where there i s no apparent trend in segregation d i s t o r t i o n , i t i s assumed that these l o c i also represent Mendelian markers. (c) MDH and AAT possess multiple l o c i , which may be explained by gene duplication and subsequent mutation. There i s also i n d i r e c t evidence for compartmentalization for these enzymes. 42 4. LINKAGE 4.1 Introduction Enzyme polymorphisms have proven to be useful markers for the study of population structure (reviewed i n Hamrick 1982), genetic d i v e r s i t y (reviewed in El-Kassaby 1990) and mating systems (reviewed i n Adams and Birkes 1990) i n forest trees. The use of isozymes i n the estimation of genetic parameters requires certain assumptions, one of which being that they display regular Mendelian inheritance patterns (Rudin 1976). When isozymes are employed in estimating parameters of the mating system (Shaw et al. 1981, Ritland and El-Kassaby 1985) , i t i s also necessary to assume they are unlinked. Linkage i s defined by Hartl (1980) as a lack of independent assortment due to l o c i being located on the same chromosome. The greater the distance between two l o c i , the more frequently w i l l crossover events occur. The maximum frequency of recombination (RF) w i l l be 0.5, the s t a t i s t i c a l equivalent of l o c i behaving as i f they sort independently. Recombination frequencies between l o c i are also important i n establishing linkage maps, as mapping the location of allozyme l o c i i s the preliminary step i n the use of isozymes as markers of quantitatively controlled characters (Vallejos and Tanksley 1983). 43 Studies of linkage i n forest trees have also provided insights into t h e i r evolutionary r e l a t i o n s h i p s . Harry (1986) noted that t i g h t linkage between AAT and PGI has been reported for several genera of the Pinaceae (Pinus, Abies, Picea, Pseudotsuga and also Larix [Cheliak and P i t e l 1985]), and although the data i s i n s u f f i c i e n t to conclude that the same AAT and PGI genes are involved i n a l l cases, Harry's discovery of the same linkage group i n incense-cedar, suggests the maintenance of t h i s gene block since before the divergence of the Pinaceae and Cupressaceae. Other reports of linkage support the notion that gene arrangements and karyology of conifers are highly conserved (Guries et al. 1978, King and Dancik 1983, Strauss and Conkle 1986). Roberds and Brotschol (1986) contend that there i s substantial evidence that evolutionary processes involve multiple locus associations and that i t may be inappropriate to use single locus measures or means to describe e f f e c t s over several l o c i . The i d e n t i f i c a t i o n of non-random associations may provide clues to the mechanism(s) responsible for the maintenance of linkage groups ( E l -Kassaby et al. 1982a) and balanced polymorphisms i n natural populations (Adams and Joly 1980b). The re s u l t s presented in t h i s chapter summarize attempts to detect linkage among allozyme l o c i i n Abies amabilis sampled on Vancouver Island. 44 4.2 Materials and Methods Linked l o c i may be uncovered in a number of ways (Bailey 1961, H i l l 1974, Wright 1976, Squillace and Swindel 1986), however, linkage studies of allozyme l o c i i n conifers can be conducted without the necessity of controlled crossing experiments (Adams and Joly 1980b) because haploid megagametophytes of a tree found heterozygous at two l o c i represent meiotic products analogous to d i p l o i d progeny phenotypes from a double backcross (a doubly-dominant parent mated to a double recessive, [cf. Bailey 1961]). Linkage i s detected when deviation from the 1:1:1:1 joint segregation of the four a l l e l e combinations, expected under the assumption of independent assortment, i s s i g n i f i c a n t (Hattemer 1979). In addition, recombination distances between l o c i may be calculated from the two smallest a l l e l e combinations (Bailey 1961, Rudin and Ekberg 1978). Of 87 trees subjected to starch gel electrophoresis, as described i n Chapter 3, 42 trees were found to be heterozygous at two or more of the seven l o c i found to be polymorphic. Segregation analyses confirming single-locus inheritance i n those trees was also presented i n Chapter 3. Of 21 possible two-locus linkage groups, 16 are represented in t h i s study by at least one doubly-heterozygous tree and 12 are represented by more than one tree, allowing tests of homogeneity among trees to be made. Where homogeneity can be confirmed, the data may be pooled over trees to produce a 45 more powerful analysis (Zar 1984), thus increasing the p r e c i s i o n of linkage estimates (Neale and Adams 1981, Cheliak and P i t e l 1985, Muona et al. 1987). Linkage tests of several pairs of l o c i involved more than one a l l e l i c combination, because these l o c i were polymorphic for more than two a l l e l e s . In these cases, the a l l e l e s at each locus were treated as two classes (e.g., A and A') p r i o r to analysis, following Adams and Joly (1980b). Chi-square tests of heterogeneity were performed according to Zar (1984), where the observed two-locus segregation r a t i o s were tested against the n u l l hypothesis that there i s a 1:1:1:1 r a t i o of a l l e l i c segregants in the entire population from which the sampled trees came. If samples are homogeneous, then the t o t a l of the chi-square values from i n d i v i d u a l trees (X2n degrees of freedom (df) equal to the number of trees in the sample mu l t i p l i e d by three independent classes) should be of sim i l a r magnitude as the chi-square calculated from the pooled t o t a l s of each of the four a l l e l i c combinations (x2p, df = 3) . The difference between these two chi-square values i s i t s e l f a chi-square, c a l l e d the heterogeneity chi-square (X2H, df = d f ( T ) - d f ( p ) ) . Should X 2 H D e s i g n i f i c a n t , then performing a goodness-of-fit test on the pooled data (Sokal and Rohlf 1981, Zar 1984) i s not j u s t i f i e d . Partitioned, single degree of freedom goodness-of-fit tests, with X 2 A t e s t i n g the segregation of a l l e l e s at locus 46 A/ X2B t e s t i n g the r a t i o at locus B and % 2 h t e s t i n g for the existence of linkage, were carried out using the formulae of Bailey (1961) for the analogous double backcross model of linkage. Bailey (1961) notes that the % 2 h t e s t for linkage i s s t i l l v a l i d even i f one of the two single-locus r a t i o s deviates from the expected 1:1 segregation. The l o g - l i k e l i h o o d r a t i o test s t a t i s t i c (G) was not employed i n the linkage analyses presented here, despite i t s t h e o r e t i c a l advantages over chi-square (%2) tests (Sokal and Rohlf 1981), because i t s c a l c u l a t i o n precludes the use of any observed frequency class with a value of zero, the logarithm of which i s undefined. Although reports of G values derived from segregation classes with zero c e l l frequencies have been reported (Boyle and Morgenstern 1985, Stewart and Schoen 1986), use of j} tests was chosen. Cochran (1954) established that in order to avoid bias i n x2 calculations where sample sizes are small, no more than 20% of the expected c e l l frequencies should be less than 5.0. This condition i s encountered for any sample of less than twenty seeds. For the purpose of detecting linkage, trees which had less than 20 megagametophytes available were eliminated from the analysis. The c r i t e r i a for the detection of linkage adopted for t h i s study are e s s e n t i a l l y those of Strauss and Conkle (1986) , namely that l o c i would be considered linked i f % 2 L 47 was s i g n i f i c a n t i n more than one tree, or i f r e s u l t s from a single tree could be confirmed by reports i n other conifers. Further, Rudin and Ekberg (1978) note that, f o r the sample si z e employed i n t h i s study (20 seeds per doubly heterozygous tr e e ) , the maximum degree of linkage that may be r e l i a b l y uncovered corresponds to an RF of 0.30. Should X2 t e s t s reveal s i g n i f i c a n t linkage, i t was decided that any calculated RF values i n excess of 0.30 would be discounted. 4.3 Results and Discussion Of the 12 linkage pairs represented by more than one tree, 10 displayed homogeneity of segregation among trees, enabling data to be pooled within each p a i r (Table 4.1). Heterogeneous data and linkage pairs represented by only one tree w i l l be discussed separately. It i s evident that in addition to being homogeneous, the segregation data for 7 of the 10 pairs (PGI-2:G6P, PGI-2:PGM, PGI-2:IDH, G6P:PGM, G6P:IDH, G6P:MDH-1, and PGM:MDH-1) when pooled yielded X2 values too small to reject the . n u l l hypothesis of independent joint segregation (Table 4.1). P a r t i t i o n e d % 2 values were also non-significant when pooled and i n d i v i d u a l trees were largely non-significant as well. None of these combinations were found to covary in the balsam f i r studied by Neale and Adams (1981), so no comparisons within the genus are possible. In Douglas-fir, El-Kassaby et a l . (1982a) found weak linkage between G6P and T a b l e 4.1. C h i - s q u a r e a n a l y s e s f o r p a i r - w i s e c o m b i n a t i o n s o f l o c i i n A. amabilis. X2H t e s t s h o m o g e n e i t y o f l i n k a g e among t r e e s (where a p p l i c a b l e ) , X2P t e s t s d e v i a t i o n f r o m n u l l h y p o t h e s i s o f i n d e p e n d e n t a s s o r t m e n t . P a r t i t i o n e d % 2 t e s t s ( f o r m u l a e o f B a i l e y (1961)) t e s t d e v i a t i o n a t l o c u s A, B and j o i n t d e v i a t i o n f r o m 1:1:1:1 s e g r e g a t i o n {%2h) . L o c i c o m b i n a t i o n No. o f t r e e s X 2 H ( d f ) / x 2 p (df=3) X 2 A ( d f = l ) X 2 B(df=D % 2 L(df=l) PGI-2:G6P 15 53.20 (42) 0 . 65 0.33 0.21 1.20 PGI-2:PGM 11 39.09 (30) 0 .00 0. 66 0.66 1.31 PGI-2:IDH 2 4 . 60 (3) 0 . 90 0.40 0. 90 2.20 G6P:PGM 9 35. 64 (24) 0 .20 0.36 0.80 2.36 G6P:IDH 2 5. 60 (3) 0 .40 0.00 0.40 0.80 G6P:MDH-1 5 19. 68 (12) 1 .00 1.44 1.44 3.52 PGM:MDH-1 3 9.20 (6) 1 .07 1.67 0.07 2.80 PGI-2 :AAT-2 3 1. 90 (6) 0 . 60 19.27* 0.07 32.90* G6PD:AAT-2 5 11.00 (12) 1 .44 38.44* 3.24 39.60* PGM:AAT-2 2 1. 60 (3) 0 .10 19.60* 0.90 20.60* MDH-1:PGI-2 7 37.15 (18) * - - -MDH-1:AAT-2 2 10.00 (3) * - - -PGI-2:MDH-2 1 -/9. 20* 7 .20* 1.80 0.20 PGI-2:IDH 1 -12. 80 0 .20 1.80 1.80 PGM:MDH-2 1 -12. 80 0 .20 1.80 0.80 IGH:MDH-1 1 -/ 5 . 20 3 .20 1.80 0.20 * = (P < 0.05) - = t e s t n o t a p p l i c a b l e . 49 IDH (RF = 0.33 ± 0.04) and O'Malley et al. (1986) discovered t i g h t e r linkage between these two l o c i i n p i t c h pine (95% confidence i n t e r v a l (CI) for recombination frequency, 0.20 -0.26). The linkage between G6P and PGM was considered inconclusive i n p i t c h pine (95% CI, 0.33-0.42; O'Malley et al. 1986). The remaining three pairs where data displayed among-tree homogeneity involve possible linkages with AAT-2 (PGI-2:AAT-2, G6P:AAT-2, and PGM:AAT-2). In a l l three cases, the pooled x2 for joint segregation (df = 3) was s i g n i f i c a n t (P < 0.05), suggesting linkage (Table 4.1). However, the p a r t i t i o n e d % 2 analyses strongly suggest that segregation d i s t o r t i o n at locus B (AAT-2 in a l l three cases) contributes to the deviant % 2 P value, as X2i (df = 1) values indicate that the p a i r s are unlinked. The X 2 H values for PGI-2:MDH-1 and MDH-1:AAT-2 are s i g n i f i c a n t and conclusions as to t h e i r possible linkages must be i n f e r r e d from individual tree data. In the f i r s t instance, two.of seven trees showed a s i g n i f i c a n t i n d i v i d u a l X2 (df = 3) value and, in the partitioned x2 analysis, these same trees showed deviations, one at the PGI-2 locus, the other at MDH-1. These deviations appear undirected and considering the small sample size, may be a t t r i b u t e d to chance. The fact that none of the seven trees exhibited a s i g n i f i c a n t X2L suggests that PGI-2 and MDH-1 are unlinked 50 at t h i s l e v e l of sampling. Only two trees were found to j o i n t l y vary for MDH-1 and AAT-2 but examination of the par t i t i o n e d x2 values reveals that unequal segregation at locus B (AAT-2 i n t h i s case) may be the cause of di s t o r t e d j o i n t segregation between these two l o c i , rather than actual linkage. Of the four linkage-pairs found in one tree only (PGI-2:MDH-2, PGI-2:IDH, PGM:MDH-2 and IDH:MDH-1), only PGI-2:MDH-2 yielded a s i g n i f i c a n t x2 (df = 3) value (Table 4.1). The p a r t i t i o n e d analysis revealed that linkage was not as much a factor as the single-locus deviation at PGI-2, which may be attri b u t e d to chance. The remaining three pairs showed no evidence of individual-locus deviation or linkage. Weak linkage (RF = 0.36 ± 0.05) has been reported i n Norway spruce for MDH-2 and PGM-2 in one tree only (Muona et al. 1987). Contrary to many published reports in other conifers, linkage was not evident in t h i s sample of P a c i f i c s i l v e r f i r between (PGI-2) and (AAT-2). Tight linkage between AAT-2 and PGI-2 (also referred to as GOT-2 and GPI-2) was detected in balsam f i r by Neale and Adams (1981) and also i n Douglas-f i r by El-Kassaby et al. (1982a). Linkage between AAT-1 and PGI-2 has been reported for several species of pine and spruce (reviewed i n Cheliak and P i t e l 1985). Simultaneous screening of AAT l o c i in fi v e conifer genera by Cheliak and P i t e l (1985) reveal possible homologies between AAT-1 and 51 AAT-2 and suggest the evolution of the AAT enzyme system by means of gene duplication. Observation of linkage between AAT and PGI i n several species and genera has strengthened arguments which suggest that gymnosperm evolution has been very conservative. Guries et al. (1978) note that p i t c h and ponderosa pine, although separate species for at least two m i l l i o n years, appear to possess the AAT:PGI linkage group. Harry (1986) also found close linkage between these two enzymes, but perhaps not i d e n t i c a l l o c i , i n incense-cedar, suggesting the maintenance of t h i s linkage since before the divergence of the Pinaceae and Cupressaceae, or some 160 m i l l i o n years ago. Recently, however, data have been published which show an order of magnitude difference i n recombination frequency between these two l o c i i n d i f f e r e n t samples of black spruce (Barrett et al. 1987). It was speculated that the large difference may be due to the c o l l e c t i o n l o cation of the samples, with marginal populations expected to show more v a r i a b i l i t y and have looser linkage than populations sampled in the center of a species' range, however, sampling errors could not be ruled out. The two trees for which linkage was s i g n i f i c a n t i n Barrett et al. (1987) had sample sizes of 51 and 144 seeds. The other published report (Boyle and Morgenstern 1985) established linkage based on eight trees with an average of 12 seeds per tree. The discrepancy i n RF 52 values may be a result of d i s s i m i l a r sample sizes, rather than true population differences. Rudin and Ekberg (1978) caution that because conifers have long chromosomes, l o c i may be so widely separated that ordinary % 2 analysis w i l l not uncover linkages unless sample sizes are very large. Alternate (Bayesian) methods for estimating recombination frequencies have been proposed by Nordheim et al. (1983), but they also note that "for recombination values close to 0.5 detection of linkage and good estimation w i l l require large sample sizes regardless of s p e c i f i c methodology." Perhaps the strength of evidence for highly conserved linkage groups l i e s not only i n the measurement of RF's for ind i v i d u a l trees using substantial sample sizes, but in the detection of linkage of s i m i l a r magnitude in a large proportion of trees heterozygous for a given pair of l o c i . Homogeneity of linkage can only be observed where resources permit the screening of s u f f i c i e n t trees to detect enough double heterozygotes for among-tree va r i a t i o n in recombination frequency to be r e l i a b l y estimated. Larger-scale studies could begin to address the causes of population differences in linkage such as those observed by Barrett et al. (1987) for AAT and PGI, and by Rudin and Ekberg (1978) for LAP (leucine amino peptidase) and AAT. More convincing data may reveal apparent differences in linkage among trees and populations of the same species in 53 fact have a genetic basis ( i . e . , genetic v a r i a t i o n i n frequency of crossing over) and/or are subject to environmental influences ( i . e . , temperature at time of meiotic a c t i v i t y ; Rudin and Ekberg 1978). Such information could prove useful in future mapping of quantitative t r a i t l o c i i n conifers. In the present study, no conclusive linkage i s revealed by X2 analyses i n any of the 16 locus-pairs tested. Although three linkages with AAT-2 indicated s i g n i f i c a n t deviation from independent assortment, suggesting linkage, p a r t i t i o n e d X2 tests showed the deviation to be influenced e n t i r e l y by segregation d i s t o r t i o n at AAT-2. As mentioned previously, segregation d i s t o r t i o n i t s e l f does not inva l i d a t e the test of independent assortment (Bailey 1961), however, i t can cause bias i n the estimation of both outcrossing rates and pollen pool a l l e l e frequencies (Cheliak et al. 1984). For t h i s reason, the apparent polymorphism observed in AAT-2 w i l l not be u t i l i z e d i n the estimation of parameters of the mating system of P a c i f i c s i l v e r f i r . 4.4 Conclusions (a) Independence amongst the 16 locus-pairs available for t e s t i n g could not be ruled out, thus providing six polymorphic l o c i on which mating system parameters may be estimated (IDH, MDH-1, MDH-2, PGM, G6P and PGI-2.) . Linkages with AAT-2 were not confirmed, but severe segregation d i s t o r t i o n (30% in excess of the 1:1 expectation for the most common a l l e l e ) observed at t h i s locus, suggests i t not be used as a marker locus i n a study of the mating system in P a c i f i c s i l v e r f i r . 55 5. MATING SYSTEM 5.1 Introduction The nature and extent of genetic v a r i a t i o n i n a species i s larg e l y determined by i t s pattern of breeding. Its system of mating constitutes the l i n k ..between successive generations whereby genetic information i s transferred, organized and di s t r i b u t e d among progeny (Clegg 1980). Plant species which p r a c t i s e a high degree of c r o s s - f e r t i l i z a t i o n are more ge n e t i c a l l y variable than species which reproduce vi a s e l f -f e r t i l i z a t i o n (Hamrick et al. 1979). Forest trees, conifers i n pa r t i c u l a r , are considered to be among the most heterozygous plants known, at least at the le v e l of enzyme v a r i a t i o n (Hamrick 1982). Mating system studies have shown most conifers to be high, although not obligate, outcrossers (see Adams and Birkes 1990 for a review). Because the reproductive process of many economically important conifers i s characterized by some natural s e l f - f e r t i l i z a t i o n and most exhibit large amounts of growth depression as a result of inbreeding (Franklin 1970, Sorensen 1982), accurate estimates of level s of inbreeding become important i n planning and implementation of tree improvement programs. The mating systems of gymnosperms have been studied extensively i n recent years, owing to the a v a i l a b i l i t y of electrophoretic single-gene markers and the unique structure of conifer seeds which eliminates the necessity of controlled 56 matings (Brown et al. 1975, Shaw and A l l a r d 1982a). Species-l e v e l estimates of outcrossing based on allozyme l o c i generally exceed 90% (Adams and Birkes 1990), although i n d i v i d u a l population estimates were found to be as low as 54% (tamarack, Knowles et al. 1987) and 78% (balsam f i r , Neale and Adams 1985b). As well, i n d i v i d u a l tree outcrossing rate v a r i a t i o n has been reported (60 to 100% in a natural population of western white pine, El-Kassaby et al. 1987b; 20 to 100% in an orchard population of Douglas-fir, El-Kassaby et al. 1986 and 78-100% i n a clonal orchard white spruce population, Denti and Schoen 1988). Both environmental and genetic factors have been shown to influence mating systems ' (Clegg 1980). Outcrossing rates were reported to vary with stand density (Farris and Mitton 1984, Shea 1987), elevation ( P h i l l i p s and Brown 1977, Neale and Adams 1985b) and population substructuring (Ritland and El-Kassaby 1985). The mating system of P a c i f i c s i l v e r f i r i s of interest given i t s p a r t i c u l a r s i l v i c a l and ec o l o g i c a l c h a r a c t e r i s t i c s . The high shade tolerance of P a c i f i c s i l v e r f i r permits the development of family structure within populations. The presence of family structure i s conducive to mating among r e l a t i v e s while the cone-bearing habit of the species (females r e s t r i c t e d to the top 20 to 30% of the canopy, most male cones in the lower portion of the crown) promotes c r o s s - f e r t i l i z a t i o n (Franklin and Ritchie 1970, Owens and Molder 1977). Noble f i r has been shown to possess r e l a t i v e l y high s e l f - f e r t i l i t y among 57 conif e r s (Sorensen et al. 1976). Although no c o n t r o l l e d mating studies have been reported for P a c i f i c s i l v e r f i r , the species i s characterized by very low yie l d s of f i l l e d seed (Franklin 1974, Owens and Molder 1977). In t h i s chapter, estimates of mating system parameters of seven populations of P a c i f i c s i l v e r f i r on Vancouver Island u t i l i z i n g allozyme markers are reported. Possible causes and implications of outcrossing rate v a r i a t i o n are discussed. Aspects of the reproductive biology and ecology of P a c i f i c s i l v e r f i r which may play a role in determining mating behaviour are also considered. 5.2 Materials and Methods Cones c o l l e c t e d from seven populations of P a c i f i c s i l v e r f i r on Vancouver Island (locations given i n Table 2.1, excluding population F) were used i n t h i s study. Eight trees w e r e sampled i n populations A, B and C, 13 in N, 11 in H and R and 17 i n population W, as described in Chapter 2. During cone and seed processing, the identit y of each mother tree was maintained. Two seed samples were randomly drawn from each of three to s i x (average 4.7) cones per tree. These samples were x-rayed (described in Chapter 2) and the percentage of f i l l e d seeds determined. Seed size was i n d i r e c t l y estimated using 1000-seed weight measures based on samples of f i l l e d seeds from i n d i v i d u a l trees. Starch gel electrophoresis of megagametophyte (same haploid genetic composition as the egg) and corresponding embryo tissues was conducted according to methods outlined i n section 3.2 and detailed i n Appendix 2. Five enzymes (PGI-2, G6P, PGM, IDH and MDH-1) were selected for estimation of matin system parameters based on segregation and linkage studies (Chapters 3 and 4) which indicated they were polymorphic, although not i n a l l populations, and inherited independently. Maternal genotypes were infer r e d from the segregation of allozymes i n megagametophyte tissue of 18 to 20 seeds per tree The p r o b a b i l i t y of in c o r r e c t l y c l a s s i f y i n g a heterozygote at any one locus i s (0.5) k _ 1 where k = number of seeds (Tigerstedt 1973), and for a sample of t h i s size, very close to zero. Tandem assay of megagametophyte and d i p l o i d embryo reveals the pollen contribution d i r e c t l y . The segregation patterns of marker l o c i i n these progenies provide the data upon which estimates of the extent of apparent outcrossing are based. Single- and multi-locus population estimates of outcrossing rate ( t s and t m) and outcrossed pollen a l l e l e frequencies (p) were calculated using the maximum l i k e l i h o o d procedure of Ritland and El-Kassaby (1985). The procedure i s based on a multi-locus, mixed mating system model which was shown to be s t a t i s t i c a l l y more e f f i c i e n t than 'observed outcross' models (e.g., Shaw and A l l a r d 1982a), es p e c i a l l y when r e l a t i v e l y few l o c i are assayed. A multi-locus estimate of outcrossing i s considered more accurate and less sensitive to v i o l a t i o n s of 59 model assumptions because a greater number of outcrosses may be i d e n t i f i e d with certainty ( i . e . the pollen a l l e l e i s distinguishable i n i t s genetic o r i g i n - see Shaw et al. 1981 for more d e t a i l ) . A d d i t i o n a l l y the model takes advantage of information contained i n the offspring of mothers which are heterozygous for a given locus. The mixed mating model i s usually employed to study patterns of mating in conifers because i t s attendant assumptions are considered more b i o l o g i c a l l y reasonable than to merely assume panmixis (Schoen and Clegg 1984), where the pr o b a b i l i t y of mating between individuals of a p a r t i c u l a r genotype i s s t r i c t l y equal to the product of t h e i r i n d i v i d u a l frequencies i n the population (Hedrick 1983) . The model assumes that (i) f e r t i l i z a t i o n events are a mixture of random outcrossing and s e l f - f e r t i l i z a t i o n ; ( i i ) there i s no selection between f e r t i l i z a t i o n and census; ( i i i ) the rate of outcrossing i s independent of the genotype of the mother tree; (iv) a l l e l e s at di f f e r e n t l o c i act independently (for multi-locus estimates) and (v) a l l e l e frequencies in the outcrossing pollen pool are i d e n t i c a l over the population of mother trees (Fyfe and Bailey 1951, Shaw et al. 1981). 60 The l a s t assumption i s one that, although viewed by Clegg (1980) as "fundamental", may not be e a s i l y met i n natural stands of trees. The d i s t r i b u t i o n of pollen i n a given population i s most l i k e l y subject to both phenological and s p a t i a l v a r i a t i o n , which could r e s u l t i n the pollen genotypes received by maternal trees being correlated (Schoen and Clegg 1984) . Population F (9 trees) was o r i g i n a l l y included i n t h i s study but no estimates for outcrossing rates could be obtained using the model of Ritland and El-Kassaby (1985). The procedure for estimating t and p i s i t e r a t i v e and, i n the case of F, f a i l e d to produce convergent values. Ritland and El-Kassaby (1985) found that i n d i v i d u a l female tree outcrossing rate estimates f a i l e d to converge for 15-25% of t h e i r orchard Douglas-fir sample and note (in the Appendix) that "estimates do not always converge i f parents are heterozygous at several l o c i " , y i e l d i n g too great a number of non-discernable mating events. For the three l o c i which were variable among a l l populations of P a c i f i c s i l v e r f i r , F had the greatest proportion of heterozygous mothers at two of them (PGI-2 and G6P). This may have contributed to the f a i l u r e of the model to produce estimates of t and p for t h i s population. A l t e r n a t i v e l y , the nonconvergence of estimates may be i n d i c a t i v e of a f a i l u r e of one or more model assumptions (Schoen 1988; Dr. K. Ritland, pers. comm., March 1989). The mixed mating model assumes that "successive outcross events 61 within a family a r i s e from independent draws of pollen from the t o t a l population of male plants" (Schoen and Clegg 1984). As previously stated, phenological and/or s p a t i a l v a r i a t i o n i n pollen d i s t r i b u t i o n may modify the f e r t i l i z a t i o n p r o b a b i l i t i e s of pollen genotypes such that pollen genotypes received by females are correlated to some degree. This scenario i s most l i k e l y to occur when r e l a t i v e l y few males are contributing to the pollen cloud (Schoen 1988). Population F i s the highest elevation sample in t h i s study, where high amounts of p r e c i p i t a t i o n and/or a short growing season may act to reduce the number of parents contributing to the pollen cloud. Franklin and Ritchie (1970) observed close synchrony•in•pollen dispersal and seed cone r e c e p t i v i t y in P a c i f i c s i l v e r f i r , but also noted considerable tree to tree v a r i a t i o n , where "the majority of P a c i f i c s i l v e r f i r trees began shedding pollen before female s t r o b i l i were receptive, and one tree actually shed the bulk of i t s pollen several days p r i o r to t h i s stage". Further support for non-uniform pollen d i s t r i b u t i o n in t h i s population comes from the fact that ten trees were o r i g i n a l l y sampled i n t h i s stand but one tree yielded less than 0.1% f i l l e d seed and was eliminated. Because estimates of t < 1.0 may be due to some amount of actual inbreeding or non-homogeneous d i s t r i b u t i o n of pollen (Ennos and Clegg 1982), a test for intrapopulation pollen a l l e l e heterogeneity (Brown et al. 1975) using homozygous mother trees i n each population was conducted for the three 62 commonly variable l o c i (PGI-2, G6P and PGM). The number of heterozygous progeny (= di s c e r n i b l y outcrossed) and the number of progeny with genotypes i d e n t i c a l to the maternal tree (includes undetectable outcrosses plus any selfs) were compared for each locus and population using contingency % 2 t e s t s . When forms of inbreeding other than d i r e c t s e l f i n g occur, the amount of s e l f i n g w i l l be overestimated (Shaw and A l l a r d 1982a). Since the multi-locus method i s less sensitive to v i o l a t i o n s of the assumptions of the mixed mating model (Shaw et al. 1981), when consanguineous matings occur, the multi-A locus estimates are presumed to be less biased, and thus t m A . A would exceed t s . A comparison of t m and the minimum variance A mean single locus outcrossing rate estimate (t s) provides an ind i c a t i o n of the extent of mating other than s e l f i n g (Shaw and A l l a r d 1982a, El-Kassaby et al. 1987b). 5.3 Results and Discussion 5.3.1 A l l e l e frequencies Estimates of a l l e l e frequencies at f i v e enzyme l o c i for both the maternal (ovule) and outcrossed pollen gene pools (p) over seven populations of P a c i f i c s i l v e r f i r , along with 95% confidence in t e r v a l s , are l i s t e d in Table 5.1. Only the most common a l l e l e i s reported. Differences between gene pools at any one locus were determined by comparing the overlap of confidence i n t e r v a l s (P <0.05; Jones and Matloff 1986, E l -Table 5.1 A l l e l i c frequencies (most common allele) and their 95% confidence intervals for the maternal (Ovule, 0) and outcrossing pollen (P) gene pools for the seven Pacific silver f i r populations on Vancouver Island. Locus Gene pool 16 + A 16 159 B 160 Population 16 26 Z H 159 258 22 N 219 26 R 258 34 W 335 PGI-2 0 0.750 ± 0.212 0.812 ± 0.191 0.750 ± 0.212 0.769 ± 0.162 0.773 ± 0;175 0.538 ± 0.192 0.559 ± 0.167 P 0.786 ± 0.064 0.825 ± 0.059 0.887 ± 0.049 0.868 ± 0.051 0.694 ± 0.061 0.605 ± 0.060 0.704 ± 0.049 G6P 0 0.750 ± 0.212 0.875 ± 0.162 0.812 ± 0.191 0.769 ± 0.162 0.727 ± 0.186 P 0.736 ± 0.067 0.775 ± 0.065 0.817 ± 0.060 0.822 ± 0.047 0.827 ± 0.050 0.923 ± 0.102 0.676 ± 0.157 0.818 ± 0.047 0.803 ± 0.043 PGM 0 0.812 ± 0.191 0.937 ± 0.119 1.000 ± 0.000 0.808 ± 0.151 0.909 ± 0.120 0.923 ± 0.102 0.647 ± 0.161 P 0.672 ± 0.073 0.944 ± 0.036 0.905 ± 0.046 0.837 ± 0.049 0.808 ± 0.052 0.845 ± 0.044 0.716 ± 0.048 IDH 0 P 1.000 ± 0.000 0.994 ± 0.012 0.962 ± 0.073 1.000 ± 0.000 MDH-1 O P 0.962 ± 0.073 0.964 ± 0.078 0.923 ± 0.102 0.996 ± 0.008 1.000 ± 0.000 0.988 ± 0.013 t Superscripts represent the number of maternal genes sampled and subscripts represent the number of embryos sampled. 64 Kassaby et al. 1987b). With the exception of the PGM locus i n population C, no s i g n i f i c a n t differences between the two gene pools were observed, i n d i c a t i n g that the outcrossed pollen pool i s representative of the maternal population, or conversely, that the maternal trees are representative of the stands i n which they were c o l l e c t e d (Brown et al. 1975, El-Kassaby et al. 1987b). It i s also noteworthy that the confidence i n t e r v a l s of pollen a l l e l e frequencies are smaller than that of ovule a l l e l e frequencies. This i s expected because of v a r i a t i o n in sample size (Brown et al. 1975) and reinforces use of the pollen pool (with larger 'n') estimates i n population genetic studies where the sample of maternal parents i s l i m i t e d (El-Kassaby 1990) . The penetrance of a l l e l e s across populations i s also apparent from Table 5.1. Populations R and W tend to be the most variable i n a l l e l i c composition for the PGI-2 locus and population W for G6P and PGM. These two stands d i f f e r by nearly 2 degrees latitude, with W being one of the most southerly c o l l e c t i o n s and R the most northern. The remainder of the populations possess quite s i m i l a r a l l e l e frequencies. This pattern has been described i n other Abies species. A l l e l e frequency differences at eight l o c i i n four stands of balsam f i r located along a steep elevational transect were viewed by Neale and Adams (1985b) as being small. The stands i n Neale and Adams' study spanned a distance of less than four km, but 65 represented a range i n a l t i t u d e of some 610 m. The apparent lack of v a r i a t i o n i n t h e i r maternal a l l e l e frequencies was contrasted with steep c l i n e s observed on the same transect by Fryer and Ledig (1972) in seedling quantitative t r a i t s . A wider-ranging study of balsam f i r by Jacobs et al. (1984) also revealed l o c i with s i m i l a r allozyme frequencies among 12 populations sampled from North Carolina to Maine. In two populations of subalpine f i r growing on contrasting s i t e s , only one of seven l o c i showed s i g n i f i c a n t differences in a l l e l e frequencies (Shea 1987). Table 5.1 also reveals that not a l l l o c i are variable across a l l seven populations, however a l l e l i c v a r i a t i o n exists, at least i n the pollen pool, at PGI-2, G6P and PGM in a l l stands. Vaquero et al. (1989) note that l o c i with r e l a t i v e l y rare a l l e l e s create a large number of empty genotypic classes in arrays of progeny from a single mother, which in turn, cause d i f f i c u l t i e s i n estimating outcrossing rates. Precision of single locus outcrossing rate estimates depend on a l l e l e and maternal genotype frequencies. Marker l o c i that are v i r t u a l l y monomorphic (pA > 0.970; Shaw and A l l a r d 1982a) add very l i t t l e information on outcrossing (because so few heterozygotes are recovered i n the progeny) and produce large variances (Shaw and A l l a r d 1982a, Ritland 1983). Brown et al. (1975) point out that the variance of a given single locus t value i s minimized when a l l e l e frequencies are equal. Given these considerations, multi-locus outcrossing rate estimates were calculated using Table 5.2 For three commonly variable loc i , the ratio of discernibly outcrossed (heterozygous) embryos to embryos possessing the maternal genotype, followed by the heterogeneity X2(df) value. Population Locus A W B C H N R P G I - 2 1 / 1 6 : 7 4 3 0 : 5 0 7 : 9 3 2 : 9 7 1 1 : 1 4 8 2 3 : 1 1 7 1 7 : 4 3 7 . 9 3 M 3 ) 6 . 5 6 (3) 3 . 9 9 (4) 8 . 1 6 (4) 9 . 0 2 (7) 3 . 4 2 (6) 1 5 . 8 7 * ( 2 ) 3 / 3 3 5 : 5 1 7 : 3 1 3 : 7 8 : 1 2 2 4 : 1 6 5 . 7 1 * ( 1 ) (0) (0) (0) 1 0 . 4 1 M D G 6 P 1 / 1 2 3 : 5 7 2 6 : 9 3 2 5 : 9 4 2 2 : 9 8 2 1 : 9 9 1 1 : 8 9 3 6 : 1 8 4 2 . 7 8 ( 3 ) 2 . 2 3 ( 5 ) 2 . 7 9 ( 5 ) 3 . 7 9 ( 5 ) 2 . 6 0 ( 5 ) 3 . 4 7 ( 4 ) 1 3 . 4 2 ( 1 0 ) PGM 1 / 1 2 9 : 7 1 2 8 : 1 8 3 1 0 : 1 3 0 1 0 : 1 3 0 2 0 : 1 4 0 3 2 : 1 4 3 3 6 : 1 8 4 3 . 1 1 ( 4 ) 7 . 7 2 ( 5 ) 5 . 8 0 ( 6 ) 4 . 3 9 ( 6 ) 5 . 4 9 ( 7 ) 1 1 . 3 2 ( 8 ) 3 1 . 2 8 * t ( 1 0 ) 3 / 3 3 2 : 7 4 . 6 8 * ( 1 ) * S i g n i f i c a n t at P <0.05. t This X2 value made s i g n i f i c a n t by one tree out of 11 ( > 50% of value of X 2 H ) (TI ON 67 a l l v a r i a b l e l o c i i n each population (t m) and as well, using only the three most polymorphic l o c i (PGI-2, G6P and PGM) which were also common to a l l populations (t m c) . Some pollen a l l e l i c heterogeneity within populations i s revealed by analyzing heterozygous'(= detectably outcrossed) genotype frequencies among homozygous mother trees, l i s t e d i n Table 5.2. The ra t i o s of heterozygous to maternal-type progeny (summed over trees) are given for the three commonly variable l o c i . Results are structured by la t i t u d e of population. These data suggest there i s some variable penetration of pollen a l l e l e s within populations, however most appear homogeneous (only s i x of 24 contingency % 2 tests were s i g n i f i c a n t ; P < 0.05). These re s u l t s should be interpreted with some caution as Cochran (1954) showed that the y} test i s v a l i d where no more than 20% of the classes have expected frequencies of less than 5.0. This c r i t e r i o n was met i n only four of the 24 tests performed. Brown et a i . (1975) also f a i l e d to meet t h i s c r i t e r i o n for some of t h e i r data, however the test was s t i l l used to i d e n t i f y l o c a l pollen heterogeneity. Non-uniform d i s t r i b u t i o n of pollen a l l e l e s over maternal trees results i n an underestimate of the extent of outcrossing (Brown et al. 1975). Intra-population heterogeneity may be the result of tree-to-tree v a r i a t i o n i n outcrossing rate (a v i o l a t i o n of assumption ( i i i ) of the mixed mating model) and/or non-uniformity of pollen a l l e l e d i s t r i b u t i o n ( v i o l a t i n g assumption 68 (v) , l i s t e d i n section 5.2) but Brown et al. (1985) assert that contingency % 2 tests cannot discriminate between causes. 5.3.2 Outcrossing rates Single- and multi-locus outcrossing rate estimates are l i s t e d i n Table 5.3. Single locus estimates were s i g n i f i c a n t l y d i f f e r e n t from t = 1.0 for at least one locus i n ,five of seven populations. Estimates of outcrossing varied from as low as 48% to 100% and the PGI-2 locus gave consistently lower estimates than other l o c i . Variation in single locus estimates of outcrossing i s common among conifers (Douglas-fir, E l -Kassaby et al. 1981, 1986, 1988; Shaw and A l l a r d 1982a; Neale and Adams 1985a; Ritland and El-Kassaby 1985; Yeh and Morgan 1987; balsam f i r , Neale and Adams 1985b; tamarack, Knowles et al. 1985; white spruce, King et al. 1984; Cheliak et al. 1985b; black spruce, Boyle and Morgenstern 1986; Barrett et al. 1987; jack pine, Cheliak et al. 1985a; Snyder et a l . 1985; lodgepole pine, Epperson and A l l a r d 1984; Perry and Dancik 1986; Jeffrey pine, Furnier and Adams 1986; Western white pine, El-Kassaby et al. 1987b; ponderosa pine, Mitton et al. 1981; F a r r i s and Mitton 1984; Scots pine, El-Kassaby et al. 1989; l o b l o l l y pine, Friedman and Adams 1985). This v a r i a t i o n i s most l i k e l y s t a t i s t i c a l (as discussed previously) because even i f a l l e l e frequencies (p's) are the same there w i l l be varying genotypes among mother trees sampled in each population, which w i l l possess d i f f e r e n t powers of detection of outcrossing events. Table 5.3 Single-locus and multi-locus estimates of outcrossing rate for seven populations of Pacific silver f i r from Vancouver Island, B.C. (95% confidence intervals). Population Locus A B C H N R W PGI-2 0.478 ± 0.206* 0.446 ± 0.213* 0.669 ± 0.234* 0.647 ± 0.175* 0.461 ± 0.154* 0.892 ± 0.167 1.204 ± 0.152 G6P 0.740 ± 0.174* 0.906 ± 0.209 0.913 ± 0.234 0.903 ± 0.178 0.803 ± 0.190* 0.947 ± 0.153 1.066 ± 0.151 PGM 0.853 ± 0.206 0.999 ± 0.220 0.842 ± 0.289 0.870 ± 0.179 0.555 + 0.212 0.999 ± 0.173 1.001 ± 0.107 IDH 0.900 ± 0.515 0.999 ± 0.175 MDH- 1 0.899 ± 0.269 0. 999 ± 0.188 0. 949 ± 0.214 A t s 0.696 + 0.112* 0.787 ± 0.120* 0.803 ± 0.144* 0.859 ± 0.083* 0.684 ± 0.091* 0.946 ± 0.087 1.068 + 0.076 t m 0.762 + 0.115* 0.888 ± 0.130 0.798 + 0.148* 0.869 ± 0.091* 0.725 ± 0.107* 0.993 ± 0.062 1.089 ± 0.073 t m r 0.762 ± 0.115* 0.867 ± 0.137 0.798 ± 0.148* 0.847 ± 0.100* 0.650 ± 0.116* 0.976 ± 0.080 1.089 ± 0.073 * Significant at P < 0.05. t s - minimum variance mean, t m - multi-locus outcrossing rate. t m c - multi-locus outcrossing rate based on three common l o c i . 70 S t a t i s t i c a l f l u c t u a t i o n i s reduced by adding more l o c i to t estimates. Yeh and Morgan (1987) postulate that some early zygotic s e l e c t i o n against s e l f e d genotypes may create v a r i a t i o n i n outcrossing rates among l o c i . Brown et al. (1985) suggest that t h i s may not be a desirable argument for electrophoretic l o c i , which are considered more remote from s e l e c t i o n than l o c i c o n t r o l l i n g morphological t r a i t s (Ritland 1983). Whether the v a r i a t i o n i s s t a t i s t i c a l or l o c i do not f i t assumptions of the mating model, the greater degrees of freedom provided by multi-locus estimates make them a less biased estimate of outcrossing (Shaw et al. 1981). It can be seen from Table 5.3 that outcrossing rates based on multi-locus estimates d i f f e r from 1.0 in a l l populations except B, R and W. Shaw and A l l a r d (1982a) suggest that where forms of inbreeding ( i . e . mating among relatives) i n addition to s e l f i n g occur, then t m i s expected to be higher than single-locus estimates. The minimum variance mean of single locus estimates were compared to corresponding multi-locus estimates for each population of P a c i f i c s i l v e r f i r . Multi-locus estimates A exceeded means of t s for a l l populations except for population C. Differences range from 1% to nearly 6% but a l l f a l l within the l i m i t s of the 95% confidence i n t e r v a l s . Neale and Adams (1985b) found a s i m i l a r range of differences i n four populations of balsam f i r and concluded that mating other than s e l f i n g was not a factor. There seems to be some v a r i a t i o n i n the i n t e r p r e t a t i o n of how substantial the difference i n 71 estimates must be to i n f e r consanguineous mating. A difference of 3% was not enough for Shaw et al. (1981) to att r i b u t e any inbreeding to mating among r e l a t i v e s , yet a 2.5% (also within 95% CI's) difference between t m and t s was considered as in d i c a t i v e of inbreeding other than s e l f i n g by El-Kassaby et al. (1987b). In P a c i f i c s i l v e r f i r , high shade tolerance and heavy seed (Franklin 1974) are conducive to the development of family c l u s t e r s and very l i k e l y render the assumption that a l l inbreeding i s due to s e l f - f e r t i l i z a t i o n i n v a l i d . Differences between t m and t s , while not great i n magnitude, are observed in 6 of 7 sampled populations, which suggests that some mating among r e l a t i v e s may be occurring and some assortative mating may be practised even when complete outcrossing i s apparent (populations B, R and W). Table 5.3 also gives multi-locus estimates for each population based on the three commonly variable l o c i . These differences are generally small and t m ' s with a reduced number of l o c i are always smaller, probably as a resu l t of reduction i n sample s i z e . The s i g n i f i c a n c e of deviation from complete outcrossing does not change. Estimates of outcrossing obtained by singl e and multi-locus methods exceed 1.0 i n one of eight populations (W). These values are not reasonable b i o l o g i c a l l y but are seen rather as an e f f e c t of sampling and not necessarily a v i o l a t i o n of model assumptions (Brown et al. 1985). These estimates may be interpreted as being i n fact t = 1.0. It i s known that 72 constraining estimates of t to b i o l o g i c a l l i m i t s (0-100%) during estimation procedures ( i . e . the Expectation-Maximization algorithm of Cheliak et al. 1983) biases estimates of t and i t s variance downward (Brown et al. 1985, Ritland and El-Kassaby 1985). Estimates of t > 1.0 should be permitted in order to provide unbiased population-level estimates of t and outcrossing pollen pool a l l e l e frequencies (Ritland and E l -Kassaby 1985). As previously noted, non-uniform s p a t i a l d i s t r i b u t i o n of pollen a l l e l e s within a population w i l l produce greater v a r i a t i o n i n the outcrossed progeny genotype d i s t r i b u t i o n s of sampled families and thus be detectable by a % 2 deviation s t a t i s t i c (Brown et al. 1975). If there i s substantial heterogeneity then a) pollen of d i f f e r i n g a l l e l e frequency i s a r r i v i n g at d i f f e r e n t trees and/or b) the propensity to outcross i s varying between trees. It i s expected that i f (a) were a cause of t < 1.0, then a p o s i t i v e relationship between heterogeneity and s e l f i n g rate would occur (Mitton et al. 1981). In P a c i f i c s i l v e r f i r populations, there i s no consistent association between heterogeneous l o c i (indicated by *'s i n Table 5.2) and low outcrossing rates (Table 5.3). It was found that three of four variable l o c i in population R ( t m = 1.0) are s i g n i f i c a n t l y heterogeneous, two of f i v e in A population W ( t m = 1.0) and most importantly only one (PGI-2 i n population A) i n the f i v e populations where there i s some inbreeding detected. The extent to which outcrossing rates 73 vary among ind i v i d u a l trees within a population could not be estimated given the number of progeny avai l a b l e to t h i s study (progeny arrays should exceed 30, Ritland 1983), however pollen pool heterogeneity does not appear to be influencing departures from panmixis. Given that some amount of inbreeding (as indicated by outcrossing rates < 1.0 in f i v e of seven populations) i s apparent, i t i s pertinent to ask whether * t ' r e f l e c t s the actual amount of inbreeding taking place. Ritland (1983) describes ' t ' as the " e f f e c t i v e " outcross rate, a summary variable representing the net e f f e c t of deviation from panmixis caused by correlations of maternal and paternal genotypes, v a r i a t i o n i n self-compatability, gametic selection and early zygotic l e t h a l i t y . Adams and Birkes (1990) caution that s (= 1-t) represents the proportion of viable progeny due to s e l f i n g , but i s not a measure of the actual frequency of s e l f -p o l l i n a t i o n , which may be considerably greater. Sorensen (1982) maintains that self-incompatability mechanisms appear to be lacking i n conifers and crossing experiments in noble f i r by Sorensen et a l . (1976) revealed that r e l a t i v e s e l f - f e r t i l i t y i s high i n that species. Embryo abortion i s common in P a c i f i c s i l v e r f i r (Owens and Molder 1977), but i t i s not known to what extent embryos resulting from s e l f i n g are aborted. The species i s known to produce large numbers of otherwise normal-appearing empty seeds (Franklin 1974). S e l f i n g reduced numbers of f i l l e d seed by 31% i n noble f i r (Sorensen et al. 1976). It may be 14 expected then, that a p o s i t i v e r e l a t i o n s h i p would exi s t between percentage of f i l l e d seed and outcrossing rate, with high seed y i e l d s correlated with high l e v e l s of outcrossing. In t h i s study, simple correlation between mean seed y i e l d per population and t m was very weakly negative (r = -0.165) and not s i g n i f i c a n t . The relationship has i n t u i t i v e appeal and had sample sizes permitted outcrossing rate estimates on an i n d i v i d u a l tree l e v e l then the c o r r e l a t i o n could have been based on i n d i v i d u a l variation rather than population means. However, El-Kassaby et al. (1987b) found the c o r r e l a t i o n between percent f i l l e d seed per tree and i n d i v i d u a l tree outcrossing rate i n a Western white pine population to be non-s i g n i f i c a n t (r = 0.012). Seed y i e l d s are subject to a number of environmental factors (climate, insects, etc.) and a clear relationship, although t h e o r e t i c a l l y p l a u s i b l e , may not be detectable from co l l e c t i o n s in natural stands. Seed size, however, i s considered to be one of the least p l a s t i c of plant characters (Sorensen and Franklin 1977) and known to be under a high degree of genetic control (Khalil 1986, Stoehr and Farmer 1986). A highly p o s i t i v e c o r r e l a t i o n of population mean seed size and multi-locus outcrossing rate was observed i n P a c i f i c s i l v e r f i r (r = 0.712, 0.05 < P < 0.10). Seed size was represented by mean thousand-seed weight estimates. This i s the reverse of that found in subalpine f i r by Shea (1987) where higher than average outcrossing rates were obtained from trees with smaller seeds. 75 5.4 Conclusions Although v a r i a t i o n i n a l l e l e frequencies i s not extensive over the range of P a c i f i c s i l v e r f i r sampled, there i s some apparent v a r i a t i o n i n the rate of outcrossing among the seven populations of P a c i f i c s i l v e r f i r i n t h i s study. The mating system i s known to be dynamic (Hamrick 1982) and given P a c i f i c s i l v e r f i r ' s e c ological status as a climax species (Krajina et al. 1982), i t i s not surprising that population estimates of inbreeding were variable (from zero to as much as 27 percent). Differences i n the magnitudes of outcrossing obtained by single and multi-locus estimation procedures suggest that some related matings are occurring, a result which i s not unexpected i n l i g h t of the high shade tolerance and l i m i t e d seed dispersal exhibited by t h i s species. At the sampling i n t e n s i t y available to t h i s study, deviation from panmixis could not be associated with mating behaviour however seed si z e was strongly related to the extent of apparent outcrossing i n P a c i f i c s i l v e r f i r . 76 6. ESTIMATES OF ELECTROPHORETIC VARIATION, ITS STRUCTURE AND RELATIONSHIP TO THE MATING SYSTEM 6.1 I n t r o d u c t i o n P r e r e q u i s i t e t o t h e a c h i e v e m e n t o f g e n e t i c g a i n i n a n y t r e e - b r e e d i n g p r o g r a m i s some k n o w l e d g e o f t h e n a t u r e a n d m a g n i t u d e o f v a r i a t i o n t h a t e x i s t s i n t h e s p e c i e s o f i n t e r e s t . A s p e c i e s v i r t u a l l y d e v o i d o f g e n e t i c v a r i a t i o n i s o b v i o u s l y n o t s u i t a b l e f o r i m p r o v e m e n t b u t most c o n i f e r s , w i t h few e x c e p t i o n s , a r e g e n e t i c a l l y d i v e r s e ( m e a n i n g r i c h i n number o f g e n e t i c t y p e s as d e f i n e d by G r e g o r i u s 1 9 8 7 ) , w h i c h i s , i n t u r n , e x p l o i t e d by b r e e d e r s . A s H e d r i c k (1983) d e s c r i b e s , " o u r f u n d a m e n t a l i n t e r e s t i n d e t e r m i n i n g t h e e x t e n t o f g e n e t i c v a r i a t i o n i s t o document t h e v a r i a t i o n t h a t r e s u l t s i n s e l e c t i v e d i f f e r e n c e s among p h e n o t y p e s " . B e c a u s e n a t u r a l p o p u l a t i o n s o f f o r e s t t r e e s a r e n o t l i k e l y t o b e h a v e as i d e a l , p a n m i c t i c b r e e d i n g u n i t s , e f f i c i e n t s a m p l i n g s t r a t e g i e s d e p e n d upon some e s t i m a t e o f t h e d e g r e e t o w h i c h p o p u l a t i o n s a r e s u b d i v i d e d g e n e t i c a l l y ( G u r i e s a n d L e d i g 1977, G r e g o r i u s a n d R o b e r d s 1 9 B 6 ) . S e l e c t i o n m ethods may t h e n be e m p l o y e d t o " t a k e a d v a n t a g e o f a c t u a l p o p u l a t i o n s t r u c t u r e " ( G u r i e s a n d L e d i g 1 9 7 7 ) . Where t h e r e i s e x t e n s i v e d i f f e r e n t i a t i o n among s t a n d s , t h e n s e v e r a l s t a n d s may be r e q u i r e d t o a d e q u a t e l y r e p r e s e n t t h e s p e c i e s , w h e r e a s l o w amounts o f among-77 population v a r i a t i o n i n a given character supports comparison-tree selection (Guries and Ledig 1977). However, baseline (individual) selection w i l l be more e f f i c i e n t than the comparison-tree method i f individuals within a population are inbred (Ledig 1974) . H i s t o r i c a l l y , methods of measuring genetic v a r i a t i o n i n forest trees have followed two avenues. The t r a d i t i o n a l approach r e l i e d on the assessment of some morphological and/or physiological t r a i t s according to an experimental model which enables among and within-population and family components of v a r i a t i o n to be estimated. Such studies usually e n t a i l the growth of progeny from wind-pollination or c o n t r o l l e d crossing i n one or more environments (Libby et al. 1969). The t r a i t s which are measured are mostly polygenic i n nature and often exhibit large amounts of environmental v a r i a t i o n (Mitton 1983). Such studies are expensive to conduct and require considerable lengths of time to obtain r e l i a b l e data (Libby et al. 1969). In the past twenty years a large number of studies estimating v a r i a t i o n i n natural and experimental populations of forest trees have u t i l i z e d enzyme polymorphisms revealed by starch gel electrophoresis (reviewed by El-Kassaby 1990). Their ease of detection, single-gene inheritance patterns and codominant expression make isozymes well-suited for studies of genetic v a r i a t i o n . Moreover, i t i s assumed that e l e c t r o p h o r e t i c a l l y detected gene products are less 78 susceptible to changes i n selection pressure and thus represent a more stable sample of the t o t a l genome than metric t r a i t s (Lewontin 1974, Ritland 1983, Gregorius and Roberds 1986, Plessas and Strauss 1986). Most conifers exhibit s i m i l a r l i f e h i s t o r i e s . They are long-lived, often occupy heterogeneous environments and necessarily must possess the capacity for adaptations which enable them to remain i n subsequent generations (Rehfeldt and Lester 1969, Muller-Starck and Gregorius 1986). Genetic v a r i a t i o n i s maintained by high lev e l s of outcrossing and mechanisms to promote outcrossing, such as inbreeding depression and s e l f - i n c o m p a t i b i l i t y , are expected to be c h a r a c t e r i s t i c of outbreeding species (Loveless and Hamrick 1984, Lande and Schemske 1985). Conifers i n general and true f i r s i n p a r t i c u l a r are known to exhibit inbreeding depression i n growth (Franklin 1970, Sorensen et al. 1976). To date, no such studies have been conducted with P a c i f i c s i l v e r f i r . The biology and ecology of P a c i f i c s i l v e r f i r are unique among western North American conifers. Its extremely high shade tolerance and late successional status (Krajina et al. 1982) coupled with production of heavy seeds which are often cached by frugivores (Franklin 1974) suggest at least the pot e n t i a l for some reproductive i s o l a t i o n and population substructure within the species. Further, the a b i l i t y of P a c i f i c s i l v e r f i r to occupy a variety of s i t e s 79 (Schmidt 1957) suggests exposure to variable s e l e c t i o n regimes. Narrow d i s t r i b u t i o n (east-west) and i s l a n d inhabitance may also act to r e s t r i c t gene flow and promote d i f f e r e n t i a t i o n (Loveless and Hamrick 1984). The organization of genetic v a r i a t i o n i n P a c i f i c s i l v e r f i r on Vancouver Island was studied using enzyme polymorphisms detected i n seeds sampled from eight populations. Several analyses based on a l l e l e and genotype frequencies were used to describe the extent and apportionment of v a r i a t i o n . Parameters of v a r i a t i o n and t h e i r r e l a t i o n s h i p to the mating system of P a c i f i c s i l v e r f i r (discussed i n Chapter 5) are also explored i n t h i s chapter. 6.2 Materials and Methods 6.2.1 Sample c o l l e c t i o n and electrophoretic assay Samples of wind-pollinated seeds c o l l e c t e d i n 1983 from eight populations of P a c i f i c s i l v e r f i r on Vancouver Island, B.C. were the source of tissues for the electrophoretic assays outlined i n Chapter 3 and detailed i n Appendix 2. Mendelian inheritance patterns for 13 putative l o c i are described i n Chapter 3 and t h e i r linkage relationships i n Chapter 4. Some evidence for segregation d i s t o r t i o n i n one locus, AAT-2, did not exclude i t i n estimating genetic v a r i a t i o n patterns (Adams 1983). Population F i s also included despite the fact that i t s mating system could not 80 be c h a r a c t e r i z e d u s i n g the mixed mating model (Fyfe and B a i l e y 1951) and i t s attendant assumptions. 6.2.2 A n a l y t i c methods (a) Caveats Adequate sample s i z e f o r e s t i m a t i n g a l l e l e f r e q u e n c i e s s h o u l d exceed nk = 100 (Brown and Moran 1981) where 'n' i s the number of maternal t r e e s i n the p o p u l a t i o n and 'k' equals the number of progeny per t r e e . T h i s number i s exceeded i n a l l e i g h t p o p u l a t i o n s i n my study, although none c o n s i s t of the i d e a l sample s i z e of 40 to 60 t r e e s recommended by El-Kassaby and S z i k l a i (1983) i n order t o o b t a i n r e l i a b l e estimates of p o p u l a t i o n a l l e l e f r e q u e n c i e s . A review by El-Kassaby (1990) of 55 papers p u b l i s h e d on c o n i f e r g e n e t i c v a r i a t i o n showed t h a t the number of i n d i v i d u a l s r e p r e s e n t i n g a p o p u l a t i o n v a r i e d from one to as many as 132. S e v e r a l s t u d i e s used bulk seed c o l l e c t i o n s w i t h the number of c o n t r i b u t i n g t r e e s unknown. Because the number of maternal t r e e s sampled i n the present study i s low (eight t o 17), p o l l e n a l l e l e frequency data were chosen t o r e p r e s e n t the e i g h t p o p u l a t i o n s i n analyses of a l l e l i c v a r i a t i o n and i n determining the d i v e r s i t y and g e n e t i c d i s t a n c e measures of Nei (1972, 1973, 1977 and 1978) and the d i f f e r e n t i a t i o n s t a t i s t i c s of Gregorius and Roberds (1986). P o l l e n a l l e l e f r e q u e n c i e s are d e r i v e d by " s u b t r a c t i n g " the maternal genotype (known with v i r t u a l c e r t a i n t y from a 81 sample of 20 seeds per tree) from the d i p l o i d embryo genotype. Pollen a l l e l e frequencies have been used i n several studies of electrophoretic v a r i a t i o n i n conifers because a larger portion of the population gene pool i s represented (Steinhoff et al. 1983, M i l l a r 1983, L i and Adams 1989, Surles et al. 1989). Where between-generation comparisons were desired, maternal tree and embryo a l l e l e frequencies were u t i l i z e d , acknowledging the dependency of these two sets of data. The o v e r a l l objectives of t h i s thesis (outlined i n Chapter 1) dictated certain sampling c r i t e r i a , namely that populations (stands) of trees represent a range of growing conditions and i n d i v i d u a l trees be sampled as randomly as possible within each population, following IUFRO guidelines (Lines 1967). However, certain l i m i t a t i o n s inherent i n the r e s u l t i n g material c o l l e c t i o n need to be i d e n t i f i e d . Some l i m i t s were set by the biology of the species i t s e l f as well as the cone crop year. The a c c e s s i b i l i t y of cones of P a c i f i c s i l v e r f i r and the 1983 crop quality both l i m i t e d the c o l l e c t i o n s i z e . Sampling of course was necessarily s t r a t i f i e d , as non-conebearing trees were excluded. In addition to the cost of sampling, the cost of analyses set l i m i t s to the amount of material which could be processed. And, given the exploratory nature of the study, I could not be assured that v a r i a t i o n would be detected i n any or a l l of the selected characters. Sampling decisions, 82 then, were affected by b i o l o g i c a l and resource r e s t r i c t i o n s and la r g e l y shaped by previous studies on related species of true f i r s . The "Catch-22" dilemma of sampling for genetic v a r i a t i o n estimation i s emphasized by Archie et al. (1989) i n that "...our a b i l i t y to detect and estimate the frequency of variants i s d i r e c t l y related to the number of individ u a l s sampled... The implications... are that an investigator must c o l l e c t and analyze large samples to discover the patterns of genetic d i v e r s i f i c a t i o n before recommendations can be made as to how many individuals should be sampled". The sampling properties of measures used to quantify genetic and es p e c i a l l y electrophoretic v a r i a t i o n have been studied by a number of investigators (Lewontin and Cockerham 1959, Brown 1970, Smith 1970, Ward and Sing 1970, Brown et al. 1975, Nei 1978, Mueller 1979, Curie-Cohen 1982, Nei and Chesser 1983, Weir and Cockerham 1984, Archie 1985, Simon and Archie 1985, Archie et al. 1989, are some). This body of work suggests that c l a s s i c a l methods of hypothesis t e s t i n g to determine how variable i s variable or how di f f e r e n t i s di f f e r e n t may not be appropriate because of non-normality and inherently large variances associated with these estimates. For example, to be 90% confident i n r e j e c t i n g a f i x a t i o n index of 0.0035 (a value comonly observed i n highly outcrossed species) using a = 0.05 a sample of at least 830,000 individuals i s required (Ward and Sing 1970). Shaw and A l l a r d (1982b) acknowledge the lack of 83 p r e c i s i o n involved i n studies of population v a r i a t i o n i n saying that t h e i r inclusion of standard errors i n the c a l c u l a t i o n of f i x a t i o n indices i s intended as an indicator of dispersion only, "as i t i s u n l i k e l y that the c r i t e r i a necessary for the v a l i d formation of confidence i n t e r v a l s from sample standard errors w i l l be met (e.g. homogeneity of variance for a l l observations)". However, publication of papers such as Kahler et a l . (1986) suggests that while estimates of genetic variation based on electrophoretic assays may only be point estimates (their paper contains no estimates of variance for either f i x a t i o n indices or genetic distance measures), they are s t i l l considered as evidence of patterns or trends in the data. Similarly, l i m i t a t i o n s encountered in the present study do not preclude data analysis and subsequent interpretation but introduce a degree of uncertainty which should and w i l l temper any conclusions. (b) Analytic methods A l l e l i c v a r i a t i o n of sampled populations of P a c i f i c s i l v e r f i r was quantified in several ways. The actual nurober of a l l e l e s per locus was averaged over a l l detected l o c i i n each population. This value i s considered to r e f l e c t a l l e l i c richness (Marshall and Brown 1975) but i n f l a t e s the contribution low-frequency a l l e l e s make to v a r i a t i o n . A more informative measure i s that of Crow and Kimura (1970) known as the " e f f e c t i v e " number of a l l e l e s per 84 locus, n e . Ne equals the inverse of the sum of squares of a l l e l e frequencies (Pi's) at a given locus. Ne i s greatest when a l l e l e s are of equal frequency and i s close to one i f only one a l l e l e i s i n very high frequency. Thus, n e r e f l e c t s both presence and frequency of a l l e l e s (Hedrick 1983). This measure i s i d e n t i c a l to the d i v e r s i t y measure ('v') proposed by Gregorius (1987). The average n e i s determined by c a l c u l a t i n g the geometric mean of i n d i v i d u a l locus values, including monomorphic l o c i (Lundkvist 1979). The quantity 1 - X p i 2 , the t h e o r e t i c a l heterozygosity obtained under Hardy-Weinberg assumptions of mating, i s very often used as a measure of genetic v a r i a t i o n (Hedrick 1983). It i s variously referred to in the population genetics l i t e r a t u r e as expected panmictic heterozygosity (Shaw and A l l a r d 1982b) or genie d i v e r s i t y (Brown et al. 1980, Falkenhagen 1985, Muona and Szmidt 1985). The average expected heterozygosity (He) i s the sum over a l l l o c i and indi v i d u a l s of ind i v i d u a l locus heterozygosities. S t a t i s t i c a l bias introduced by small sample sizes (less than 20 individuals) i s found to be reduced by applying a correction factor of 2N/2N-1 to He (Nei and Roychoudhury 1974). This correction was applied to estimates of He for a l l maternal gene pools in t h i s study. Further sampling properties of heterozygosity are considered i n section 6.4. In addition to the amount of var i a t i o n present, i t s d i s t r i b u t i o n among and within populations i s of in t e r e s t . A 85 distance measure, analogous to geometric distance, was developed by Nei (1972, 1978). The genetic distance between population X and Y (Dxy) represents the amount of between-population differences i n a l l e l e frequencies, averaged over monomorphic and polymorphic l o c i . A c l u s t e r analysis of the matrix of D values for a l l pairs of populations using an arithmetic averaging procedure was performed using SYSTAT (Wilkinson 1988) to obtain a graphical representation of the genetic relatedness of the eight sampled stands of P a c i f i c s i l v e r f i r . The apportionment of v a r i a t i o n among and within populations was estimated using the d i v e r s i t y s t a t i s t i c s devised by Nei (1973, 1977) and a population d i f f e r e n t i a t i o n measure ('8') developed by Gregorius and Roberds (1986). Nei's " c o e f f i c i e n t of genetic d i f f e r e n t i a t i o n " (GST) was c a l c u l a t e d using only polymorphic l o c i to make a comparison with 8 and also with the inclusion of monomorphic l o c i to allow comparison with d i v e r s i t y s t a t i s t i c s estimated for other true f i r s and conifers in general. Both G S T (which was shown to be equivalent to Wright's [1978] F S T) and 8 were o r i g i n a l l y devised to estimate population subdivision or m i c r o d i f f e r e n t i a t i o n but these s t a t i s t i c s have been applied to discrete populations i n a number of studies of forest trees (e.g. O'Malley et a l . 1979, Furnier and Adams 1986, L i and Adams 1989 for GST; Muller-Starck and Gregorius 1986 for 8). The 86 d i f f e r e n t i a t i o n estimates according to the concept of Gregorius and Roberds (1986) were provided by H.-R. Gregorius (pers. comm., March 1989) using pollen a l l e l e frequencies. Mating system ef f e c t s on population genetic structure were estimated using Wright's (1969) f i x a t i o n index (F). Individual locus estimates of F ( = 1 - [observed number of heterozygous genotypes/expected number of heterozygous genotypes]) were calculated as well as a minimum variance mean over l o c i (Shaw and A l l a r d 1982b) for maternal tree and embryo samples i n each population. • This quantity i s the same as the 'i n d i r e c t ' estimate of F referred to by Brown (1979). A 'direct' estimate of F using the equilibrium r e l a t i o n s h i p with s e l f i n g ( F e = s/2-s) where s i s equal to the rate of s e l f - f e r t i l i z a t i o n (Hedrick 1983) was also obtained f o r seven populations (excluding F) using the multilocus estimates of outcrossing rate (t = 1-s) reported i n Chapter 5. 6.3 Results 6.3.1 A l l e l i c v a r i a t i o n Pollen a l l e l e frequency estimates for in d i v i d u a l l o c i across eight populations are shown in Table 6.1. The inc l u s i o n of IDH and MDH-1 as polymorphic r e f l e c t i n d i v i d u a l frequencies where the incidence of the most common a l l e l e Table 6.1 Pollen a l l e l e frequencies for six variable l o c i in eight populations of P a c i f i c s i l v e r f i r on Vancouver Island, B.C. (Populations are arranged i n order of increasing l a t i t u d e and t h e i r l e t t e r designations are followed by sample size; represents a n u l l a l l e l e ) . Population Locus/Allele F (180) W(335) A (159) B(160) C (159) N(219) R(258) H(258) PGI-2 1 0.672 0 .704 0.786 0 .825 0.887 0.694 0.605 0.868 2 0.328 0.296 0.201 0.175 0 .113 0.306 0.395 0.132 'n' 0.000 0.000 0.013 0.000 0.000 0.000 0.000 0.000 G6P 1 0.572 0.803 0.736 0.775 0.817 0.827 0.818 0.822 2 0.044 0.003 0.000 0.000 0.000 0.000 0.000 0.000 3 0.006 0.000 0.031 0.012 0.000 0.000 0.000 0.008 5 0.378 0.191 0.223 0.213 0.183 0.164 0.182 0.169 •n' 0.000 0.003 0.000 0.000 0.000 0.009 0.000 0.000 PGM 1 0.783 0.716 0. 672 0. 944 0 . 905 0.808 0.845 0.837 2 0.011 0 . 027 0.006 0.000 0.013 0. 000 0.000 0.000 3 0.189 0.248 0.322 0.056 0.076 0.187 0.155 0.151 5 0.006 0.009 0.000 0.000 0.006 0.000 0.000 0.004 'n' 0.000 0.000 0.000 0.000 0.000 0.005 0.000 0.008 IDH 1 1.000 0. 997 1.000 0. 994 1.000 1.000 1.000 1.000 3 0.000 0.003 0.000 0.006 0.000 0.000 0.000 0.000 MDH-1 1 1.000 1.000 1.000 1.000 1.000 1.000 0. 988 0.996 2 0.000 0.000 0.000 0.000 0.000 0.000 0.012 0.004 AAT-2 1 0. 972 0.788 0.874 0.938 0. 924 0. 945 0.965 0.984 2 0.028 0.212 0.126 0.162 0.076 0.055 0.035 0.016 88 was l e s s t h a n 0.99 but when averaged over a l l i n d i v i d u a l s exceeds t h e 99% c r i t e r i o n ( H a r t l 1980). There a r e s e v e r a l g e o g r a p h i c t r e n d s apparent i n t h e t a b l e . A p r i v a t e a l l e l e , f o u n d o n l y i n p o p u l a t i o n A was d e t e c t e d a t t h e PGI-2 l o c u s , whereas a l l o t h e r a l l e l e s o c c u r r e d i n a t l e a s t two p o p u l a t i o n s . A l l e l e '2' i n G6P was d e t e c t e d o n l y i n t h e two southernmost p o p u l a t i o n s (F and W). The n u l l a l l e l e ('n') was d e t e c t e d a t t h e PGM l o c u s i n o n l y two o f t h e no r t h e r n m o s t p o p u l a t i o n s , N and H . V a r i a t i o n a t MDH-1 was found o n l y i n t h e n o r t h e r n samples (R and H ). P o p u l a t i o n s had e i t h e r f o u r o r f i v e v a r y i n g l o c i and t h r e e o f t h e o f 11 l o c i a s s a y e d were commonly v a r i a b l e (PGI-2, G6P and PGM). A c o n t i n g e n c y y} a n a l y s i s of a l l e l e f r e q u e n c i e s (Zar 1984) was a t t e m p t e d t o o b t a i n an e s t i m a t e o f a l l e l i c h e t e r o g e n e i t y among p o p u l a t i o n s , however o n l y two l o c i had s u f f i c i e n t l y l a r g e e x p e c t e d c e l l f r e q u e n c i e s t o p e r m i t % 2 e s t i m a t e s t o be made. Both l o c i showed s i g n i f i c a n t h e t e r o g e n e i t y among p o p u l a t i o n s (%2 f o r PGI-2 was 82.3 and f o r AAT-2 was 111.06, compared t o a c r i t i c a l v a l u e [7 d f , a = 0.05J o f 14.07) . The average a c t u a l and e f f e c t i v e number of a l l e l e s p e r l o c u s a r e l i s t e d i n Ta b l e 6.2. The g r e a t e s t d i s p a r i t y i n t h e s e two measures o c c u r s i n p o p u l a t i o n H where, a l t h o u g h a l a r g e r number o f a l l e l e s were d e t e c t e d (1.73 a l l e l e s p e r l o c u s on a v e r a g e ) , some o f t h e s e a l l e l e s o c c u r r e d a t r e l a t i v e l y low f r e q u e n c i e s , r e d u c i n g t h e e f f e c t i v e number t o 89 Table 6.2 Populations l i s t e d in order of increasing lati t u d e , sample size, actual and 'e f f e c t i v e ' numbers of a l l e l e s per locus for eight stands of P a c i f i c s i l v e r f i r . Population Latitude No. of trees No. of a l l e l e s per locus E f f e c t i v e no. of a l l e l e s per locus F 48 '39' 9 1.73 1.23 W 48 > 1 II 48 15 17 1. 82 1.22 A 49 'l8 ' 8 1. 64 1.21 B 49* 57 ' 8 1.55 1.11 C 50 '03'30" 8 1.55 1.09 H 50* _ 1 11 34 45 13 1.73 1.10 R 50* '3e'3o" 13 1. 45 1.16 N 50* 43'30" 11 ' 1.55 1.16 Weighted Mean1 1 . 64 1.16 1 Arithmetic average for actual number of a l l e l e s ; geometric mean for e f f e c t i v e number (Lundkvist 1979). 1 .10 . Actual a l l e l e counts did not show any trend with geography, however e f f e c t i v e numbers of a l l e l e s were negatively correlated (r = - 0 . 7 1 9 , P < 0 . 0 5 ) , i n d i c a t i n g that, i n general, a l l e l i c v a r i a t i o n decreases with increasing l a t i t u d e . Overall, populations possessed r e l a t i v e l y low le v e l s of a l l e l i c d i v e r s i t y with n e values showing a tendency for one a l l e l e to dominate. No population was devoid of v a r i a t i o n and n e estimates suggest population W i s the most 90 ge n e t i c a l l y diverse and population C the most depauperate. These populations also had the largest and smallest samples, respectively, however Spearman's rank c o r r e l a t i o n (Zar, 1984) f a i l e d to show any strong rel a t i o n s h i p between sample size and d i v e r s i t y (r = 0.241), as t h e o r e t i c a l l y predicted by Gregorius (1987) . Average expected heterozygosities (H) are l i s t e d i n Table 6.3 for maternal tree and embryo gene pools. Weighted Table 6.3 Average expected heterozygosities® for adult (maternal tree) and embryo gene pools for eight populations of P a c i f i c s i l v e r f i r . Population Adult® Embryo F 0. 156 0.108 W 0. 177 0.125 A 0. 097 0.091 B 0.068 0.054 C 0.889 0. 052 H • 0.124 0. 072 N 0.109 0.073 R 0.081 0.073 Mean® 0.118 0.085 ® Average expected heterozygosity = N-m XXHij where represents the heterozygosity of the i t h i n d i v i d u a l at the j t h locus, summed over m l o c i and N i n d i v i d u a l s . © Sample correction factor 2N/2N-1 applied to adult estimates (Nei and Roychoudhury 1974). ® Weighted by sample size. 91 mean values are also given. These measures of d i v e r s i t y showed a s i m i l a r geographic trend to ne, the c o r r e l a t i o n with l a t i t u d e being negative and just s i g n i f i c a n t at a = 0.05 (r = -0.643) for the maternal gene pool and s i g n i f i c a n t for embryos also (r = -0.755). This i s not unexpected since both quantities are derived from s i m i l a r data structures (p^'s). Average H's do d i f f e r between gene pools, and i n a l l populations maternal tree heterozygosity i s greater than that expected for the embryo population. 6.3.2 D i s t r i b u t i o n of v a r i a t i o n Estimates of Nei's (1978) genetic distance between pair s of P a c i f i c s i l v e r f i r populations along with t h e i r geographic distances (measured in kilometers) are presented in Table 6.4. The values of D are a l l very close to zero. Overall, the average genetic distance i s 0.0112. The smallest value joins populations B and C which are also i n the closest geographic proximity (9 km). Another two stands (N and R) which are physically near each other are also g e n e t i c a l l y very s i m i l a r (D = 0.00192). Population F i s distinguishable from the rest of the stands by being the most remote gene t i c a l l y and t h i s i s evident i n the graphical representation shown in Figure 6.1. Two other subgroups appear i n the cl u s t e r diagram but grouping does not f i t any strong geographic trend, as two of the northernmost samples (N and R) are grouped more closely with two of the southern populations (A & W) than with the other northern sample and Table 6.4 Estimates of genetic distances (Nei 1978) below the diagonal and geographic distances (measured i n kilometers and l i s t e d above the diagonal) between pairs of stands of P a c i f i c s i l v e r f i r . POPULATION F W A B C H N R F — 28 110 222 225 335 368 363 W 0. 01848 — 83 195 199 308 341 336 A 0. 01420 0 .00480 — 112 116 225 258 254 B 0. 01696 0 .01283 0 .01433 — 9 113 147 145 C 0. 02133 0 .01517 0 .01316 0 .00178 — 111 145 142 H 0. 01745 0 .01390 0 .00889 0 .00504 0.00192 — 33 32 N 0. 01173 0 .00596 0 .00783 0 .00830 0.00843 0 .00617 — 15 R 0. 01165 0 .01080 0 .01589 0 .01332 0.01681 0 .01359 0.00201 — 10 to 93 0.03 0.02 0.01 0.0 GENETIC DISTANCE Figure 6.1 Dendrogram from clu s t e r analysis based.on Nei's (1978) genetic distances between eight populations of P a c i f i c s i l v e r f i r , referenced by l e t t e r and described i n Table 2.1. 94 the two middle l a t i t u d e stands (B & C). The c o r r e l a t i o n among genetic and geographic distances was not s i g n i f i c a n t , with geographical distance, explaining only 1.4% of the v a r i a t i o n i n genetic distance among pai r s of populations. Individual locus and average values of Nei's (1977) d i v e r s i t y s t a t i s t i c s are l i s t e d in Table 6.5. Only polymorphic l o c i were used in order to emphasize the extent of population d i f f e r e n t i a t i o n (Brown 1979) and to make comparisons with the d i f f e r e n t i a t i o n index of Gregorius and Table 6.5 Gene diversity® estimates for six polymorphic l o c i averaged over eight populations of P a c i f i c s i l v e r f i r . Mean gene d i v e r s i t y Locus Total (HT) Within Population (Hs) Among Population (DST) G S T® PGI-2 0.381 0.363 0.018 0.047 G6P 0.349 0. 340 0.009 0.026 PGM 0.318 0.288 0.030 0. 094 IDH 0.002 0. 002 0.000 0.000 MDH-1 0.005 0. 005 0.000 1 0.000 AAT-2 0.150 0.140 0.010 0.067 Mean 0.039 Mean (including monomorphic loci ) 0.018 ® Nei (1975). Gene d i v e r s i t y based on estimates of expected heterozygosity using pollen a l l e l e frequencies HT = t o t a l gene d i v e r s i t y over a l l populations based on average a l l e l e frequencies for each locus; Hs = average gene density within populations; DST = HT - Hs. ® G S T = r e l a t i v e amount of gene d i v e r s i t y due to differences among populations ( = DST / HT) . 95 Roberds (1986). Three l o c i (PGI-2, G6P and PGM) showed moderate l e v e l s of d i v e r s i t y within populations (Hs) but the amounts were also very close to values shown by a l l populations (stands) considered together (HT) , r e s u l t i n g i n d i v e r s i t i e s attributable to differences among populations (D S T's) which were not much larger than D S T values for l o c i e x h i b i t i n g much lower levels of d i v e r s i t y (IDH, MDH-1 and AAT-2). This low l e v e l of population d i f f e r e n t i a t i o n i s manifest i n an average G S T value for polymorphic l o c i of 0.039, or less than 4% of the t o t a l d i v e r s i t y detected i n a l l the samples being attributed to genetic differences among populations of P a c i f i c s i l v e r f i r . Thus, the vast majority of a l l e l i c v a r i a t i o n (96%) resides within i n d i v i d u a l stands. Another measure of genetic disparateness proposed by Gregorius and Roberds (1986) i s summarized in Table 6.6. Individual locus Dj values, representing the amount of a l l e l i c d i f f e r e n t i a t i o n of the j t h population from a l l other populations are given in the table as well as 8 values for each locus. According to Gregorius and Roberds (1986) D^  can be interpreted as the proportion (expressed as a percentage i n Table 6.6) of the e f f e c t i v e number of genetic elements (in t h i s case genes or a l l e l e s ) by which the j t h population d i f f e r s from i t s complement ( a l l other populations pooled). The value 8 represents the mean (weighted by sample size) percentage of the e f f e c t i v e 96 Table 6.6 Individual a l l e l i c d i f f e r e n t i a t i o n values (Dj's) for each locus and population and t o t a l s (summed over population) and average 8 values for each varying locus detected i n eight populations of P a c i f i c s i l v e r f i r . Population Locus F W A B c H N R 8 (weighted) <£ PGI-2 9.6 6.0 4.8 7.9 15.0 12.9 7 .1 17.3 10.1 G6P 23.0 3.7 4.8 0.9 5.2 5.9 7 .2 5.3 6.7 PGM 2.8 11.4 17.0 14.7 11.3 3.3 1 .9 3.4 7.7 IDH 0.1 0.2 0.1 0.5 0.1 0.1 0 .1 0.1 0.1 MDH-1 0.2 0.2 0.2 0.2 0.2 0.2 0 .2 1.1 0.3 AAT-2 5.5 15.5 5.6 1.6 0.0 6.8 2 .4 4.7 6.3 Mean 6.9 6.2 5.4 4.3 5.3 4.9 3 .2 5.3 5.2 ® D i f f e r e n t i a t i o n values (5's) determined by taking arithmetic mean of Dj's weighted by sample size of each population. numbers of genes by which populations d i f f e r from t h e i r complements. It i s apparent from Table 6.6 that some l o c i discriminate better than others. The 8 values range from less than 1% for IDH and MDH-1 to over 10% for PGI-2. Individual Dj's reveal that no single locus i s p a r t i c u l a r l y discriminating, as large Dj values are not consistent for any one locus across a l l populations. This v a r i a t i o n i s es p e c i a l l y evident for G6P, where Dj's vary from 0.9 to 23.0. Comparison of Dj's for each population suggest population F i s the most d i f f e r e n t i a t e d (primarily as a re s u l t of the high value at G6P), and population N the le a s t . Calculation of mean Dj values shows population F d i f f e r s from the rest of the populations at nearly 7% of.the 97 e f f e c t i v e number of genes for these six l o c i . Mean Dj values also correlate negatively with l a t i t u d e (r = -0.778; P < 0.05) i n d i c a t i n g that populations are less d i f f e r e n t i a t e d i n the northern portion of the sampled range of P a c i f i c s i l v e r f i r . 6.3.3 Mating system ef f e c t s on v a r i a t i o n Individual locus f i x a t i o n indices and mean values (weighted by the inverse of the associated variance estimated according to Rasmussen 1964) for maternal tree and embryo gene pools i n each population are found in Table 6.7. These values were calculated in order to quantify the extent to which the observed number of heterozygotes i n each gene pool deviated from the number expected under Hardy-Weinberg assumptions. Where F i s greater than zero, fewer heterozygous genotypes than predicted are present and, where the value of F i s less than zero, an excess of heterozygotes r e l a t i v e to expectations i s observed. Table 6.7 shows that single locus F's are highly variable both among l o c i and gene pools. For the three commonly variable l o c i (PGI-2, G6P and PGM), 22 of 48 estimates are negative, 26 of 48 p o s i t i v e . Of the more invariant l o c i , MDH-1 showed a tendency to excess heterozygosity (eight out of 10 estimates are negative). Comparing gene pools reveals a somewhat stronger trend. Over 75% of the i n d i v i d u a l locus F's for Table 6.7 Individual locus and minimum variance mean estimates of the fixation index for maternal tree (M) and embryo (E) gene pools in eight populations of Pac i f i c s i l v e r f i r . POPN F • w A B c H N R LOCUS M E M E M E M E M E M E M E M E PGI-2 -0.275 0.149 -0.274 0.034 -0.250 0.091 -0.154 -0.056 0.375 0.184 0.165 0.123 0.258 0.302 -0.192 0.012 G6P -0.228 -0.058 -0.435 -0.092 -0.155 0.077 -0.068 0.034 -0.096 0.004 -0.252 0.007 -0.310 0.041 -0.020 0.004 PGM 0.640 0.050 0.039 -0.101 -0.154 0.045 0.008 -0.005 0.024 -0.196 0.063 -0.052 0.117 -0.041 0.123 IDH 0.009 -0.045 -0.035 -0.011 0.061 -0.049 MDH-1 -0.150 -0.088 -0.060 -0.049 -0.013 0.017 -0.052 -0.304 -0.041 0.130 AAT-2 0.128 -0.042 0.059 0.008 -0.040 -0.157 0.064 -0.137 0.002 -0.052 -0.111 MEAN -0.011 0.002 -0.052 -0.029 -0.183 -0.007 -0.057 -0.028 0.057 0.069 -0.063 0.050 -0.030 0.002 -0.078 0.047 00 99 maternal trees show excess heterozygosity, while less than 47% of embryo f i x a t i o n indices are negative. In spite of large inter-locus v a r i a t i o n i n F, mean values also r e f l e c t the pattern of greater homozygosity i n embryo gene pools than that of maternal trees. Seven of eight maternal tree populations have average f i x a t i o n indices that are negative while negative average values were obtained for only three embryo populations. Hedrick (1983) c l a s s i f i e s t h i s method of estimating f i x a t i o n indices (based on deviations in genotypic population structure) an i n d i r e c t one (after Brown [1979]). An alternate method, c a l l e d the dire c t approach, uses the rela t i o n s h i p F = 1-t/l+t, where t i s an estimate of the proportion of outcrossing based on progeny genotypes i d e n t i f i e d from outcrossing events. Both estimates assume that populations have reached inbreeding e q u i l i b r i a , meaning that the same amount of outcrossing reconstitutes the same l e v e l of heterozygotes at each round of mating. Given t h i s assumption, both estimates should be equivalent i f genotypic proportions are determined by no evolutionary factors other than the mating system. Table 6.8 shows F x and F D (calculated using the multilocus estimate of outcrossing, t m , derived i n Chapter 5) compared, where possible, for maternal and embryo gene pools. The dire c t estimates of F, determined from progeny Table 6.8 Estimates of average equilibrium f i x a t i o n indices based on an in d i r e c t method (F x, a f t e r Brown 1979) and a direct approach, where F D i s derived from the multilocus extimate of outcrossing, t m (Ritland and El-Kassaby, 1985) . POPULATION (estimated from 1-HD/He)® F D (estimated from l - t m / l + t m ) MATERNAL EMBRYO EMBRYO F -0.011 0,002 <3> W -0.052 -0.029 -0.043 A -0.183 -0.007 0.135 B -0.057 -0.028 0.059 C 0.057 0.069 0.112 H -0.063 0.050 0.070 N -0.030 0.002 0.159 R -0.078 0.047 0.004 ® Minimum variance mean values, calculated from Shaw and Alland 1982b, and l i s t e d in Table 6.7. No estimate of outcrossing rate obtained for population F. 101 arrays, are l a r g e l y p o s i t i v e . This suggests an o v e r a l l deficiency of heterozygotes i n the progeny generation. Indirect estimates of F, determined for both generations, match F D only i n nature and magnitude for population W. Embryo populations show less homozygosity than that predicted by the outcrossing rate estimate (F-j's are less p o s i t i v e than F D) , and maternal tree populations show even greater heterozygosity (F x maternal < F x embryo < F D) . 6.4 Discussion Given the ec o l o g i c a l amplitude and successional status of P a c i f i c s i l v e r f i r , combined with i t s very high shade tolerance and production of heavy seeds, the pot e n t i a l exists for substantial genetic v a r i a t i o n among populations. However, estimates of enzyme v a r i a b i l i t y are low across both populations and gene pools. The patterns of v a r i a t i o n which were detected may r e f l e c t c e r t a i n b i o l o g i c a l phenomena or may be the re s u l t of a largely unknown degree of sampling error inherent i n estimates of t h i s kind. In addition, there i s some dispute i n the forest genetics l i t e r a t u r e as to which measures are most appropriate for describing v a r i a t i o n when i t i s detected (Gregorius and Roberds 1986, Muller-Starck and Gregorius 1986). Apparent low l e v e l s of within-population v a r i a b i l i t y were based on estimates of e f f e c t i v e number of a l l e l e s (ne) and expected heterozygosity (He; referred to as gene • 102 d i v e r s i t y by Nei 1973). While n e i s considered a true d i v e r s i t y measure according to Gregorius (1987), r e f l e c t i n g the number of di f f e r e n t genetic types i n a population, i t s sampling properties are not considered "as good" as that of He by Nei (1975). Observed frequencies of a l l e l e s form the basis for both of these estimates, so i t i s not unexpected that they would show s i m i l a r trends. Statements as to the s t a t i s t i c a l significance of any differences among populations or between gene pools for these values have been avoided because of the large inter-locus variances inherent i n average heterozygosity estimates (Simon and Archie 1985) and the l i k e l i h o o d that parametric s t a t i s t i c a l tests may not be v a l i d for the sample sizes available and the l e v e l of heterozygosity present i n the data (Archie 1985). Point estimates of d i v e r s i t y do suggest that populations are more heterozygous in the south than in the northern portion of the sampled range and that extant populations, represented by maternal tree samples, appear to be more heterozygous than the poten t i a l (viable seed) populations of P a c i f i c s i l v e r f i r . Measures of He f a l l in the range of average heterozygosities for a number of conifers (Boyle and Morgenstern 1986) but below the average He (0.207) of 20 conifer species determined by Hamrick et al. (1981). Contingency % 2 tests of a l l e l e frequency differences suggest some evidence for genetic heterogenity among populations of P a c i f i c s i l v e r f i r but there are two cautions 103 accompanying t h i s observation. Only two of six varying l o c i could be tested by t h i s procedure owing to the presence of too many expected c e l l frequencies of less than f i v e i n the remaining l o c i (Cochran 1954). Muona and Schmidt (1985) also point out that a l l e l e frequency differences w i l l be s i g n i f i c a n t even i f they are r e l a t i v e l y small i f sample sizes are s u f f i c i e n t l y large. They considered a sample of 134 seeds (268 a l l e l e s ) per population "large". Sample sizes for populations in the present study ranged from 159 to 335 seeds (318 to 670 a l l e l e s ) per locus, so differences based on contingency % 2 tests may be exaggerated. Neale and Adams (1985b) found a l l e l e frequencies among four stands of balsam f i r to be r e l a t i v e l y homogeneous for eight l o c i and s i m i l a r to another balsam f i r population sampled from a d i f f e r e n t location and i n a d i f f e r e n t year (Jacobs et al. 1984). A consistently low e f f e c t i v e number of a l l e l e s was at t r i b u t e d by Bousquet et al. (1987) to a high l e v e l of gene flow among populations of green alder. Despite a number of references i n d i c a t i n g that the genetic distance measure of Nei (1972, 1978) performs poorly at values close to zero (Nei 1972, Gregorius 1978, Falkenhagen 1985), i t i s most commonly c i t e d i n assessing inter-population v a r i a t i o n in conifers. Commonly, D's are very small and do not correlate well with physical distances (Linhart et al. 1981, Boyle and Morgenstern 1986), which suggests i s o l a t i o n per se i s not responsible for population 104 d i f f e r e n t i a t i o n . This pattern i s observed in P a c i f i c s i l v e r f i r and when considered along with the apparent commonality of a l l e l e s may indicate few b a r r i e r s to gene flow are present, despite considerable e c o l o g i c a l v a r i a t i o n among sampled populations. The two measures of population d i f f e r e n t i a t i o n c i t e d i n t h i s paper, G S T (equivalent to F S T, but derived for d i f f e r e n t purposes [Nei 1975]) and 5 also suggest population d i f f e r e n t i a t i o n i s not strong in P a c i f i c s i l v e r f i r , although the measures themselves do not appear to be equivalent. Table 6.9 l i s t s i n d i v i d u a l locus d i f f e r e n t i a t i o n values Table 6.9 Comparison of the proportion of d i f f e r e n t i a t i o n explained by various l o c i for estimates of G S T (Nei 1977) and 8 (Gregorius and Roberds 1986). Values expressed as percentages. Locus G S T 8 PGI-2 4.7 10. 1 G6P 2.6 6. 7 PGM 9.4 7. 7 IDH 0.0 0. 1 MDH-1 0.0 0. 3 AAT-2 6.7 6. 3 for both G S T and 8. Although differences might be negated i f some indices of dispersion were known with any r e l i a b i l i t y , i t appears that in general, the ranking of l o c i 105 i n t h e i r a b i l i t y to detect population d i f f e r e n t i a t i o n i s very close when a l l e l e frequencies are extreme (IDH, MDH-1, AAT-2) but where a l l e l e frequencies .are more moderate the rank and magnitude of the contribution of l o c i varies widely (e.g. the 8 value of PGI-2 suggests i t i s twice as discriminating as the G S T value would i n d i c a t e ) . Perhaps t h i s arises from larger variances associated with a l l e l e frequencies which are approximately equivalent (Brown 1979, El-Kassaby and S z i k l a i 1983). This res u l t suggests the importance of including as many l o c i as possible i n estimation of population d i f f e r e n t i a t i o n and that consideration be given to the degree of polymorphism expressed by l o c i when species or population comparisons are desired. In P a c i f i c s i l v e r f i r , average 8 was 5.2% compared to average G S T (using polymorphic l o c i only) of 3.9%. Comparisons between F S T and 8 ave been made for ponderosa pine subpopulations (Gregorius and Roberds 1986), beech sub-populations (Gregorius et al. 1986) and discrete populations of Scots pine (Miiller-Starck and Gregorius 1986) and i n a l l cases 8 was at least twice the magnitude of F S T (= GST) . This may r e f l e c t a greater s e n s i t i v i t y of 8 i n detecting population differences (El-Kassaby 1990), or may be the r e s u l t of comparing apples and oranges - F S T and G S T measures are based upon assumptions of random samples of the genome 106 including both mono- and polymorphic l o c i , whereas 8 i s defined for variable l o c i only. Table 6.10 shows average G S T values of several Abies species. The v a r i a t i o n i n G S T among rela t e d species p a r a l l e l s results for most other conifer genera (reviewed by El-Kassaby 1990). Nei (1973) stresses that G S T values are population-specific and non-comparable unless breeding systems are si m i l a r . Given that v a r i a t i o n i n mating systems among populations of P a c i f i c s i l v e r f i r was detected (see Table 6.10 Estimates of G S T for several species of Abies. Species No. of Populations G, ST References A. balsamea 5 0.012 Neale 1978 (cited i n Guries and Ledig [1981]) A. lasiocarpa 1 0.015 Grant and Mitton 1977 A. amabilis 8 0.018 Present study A. procera 22 0.123 Yeh unpublished (cited i n Adams 1983) A. grandis 23 0.140 Yeh unpublished (cited i n Adams 1983) 107 Chapter 5) and i s often encountered in other conifers (Adams and Birkes 1990) the concept of a species average G S T may be inappropriate, yet Nei (1975) reports these values himself. Roberds and Conkle (1984) maintain that even though a l l e l e frequencies may not d i f f e r s ubstantially, population structure may s t i l l be present. The mating system can a f f e c t population structure by promoting or r e s t r i c t i n g h ybridization (Ritland 1983). Mating systems characterized by some degree of inbreeding w i l l exhibit reduced recombination and increased homozygosity which ultimately r e s t r i c t s gene flow and decreases the e f f e c t i v e population size (Ritland 1983, Loveless and Hamrick 1984). Results of the present study suggest that some evolutionary forces i n addition to the mating system may be acting in populations of P a c i f i c s i l v e r f i r . Individual f i x a t i o n indices were highly variable over l o c i and populations, which may indicate that some genotypes are favored in some environments, but not in others. Also, there was more heterozygosity detected in most of the populations than expected from the mating system alone (F T < F D) . In addition gene pools representing extant P a c i f i c s i l v e r f i r populations showed greater heterozygosity than that exhibited by potential populations. Thus, i t may be that there i s selection favoring heterozygosity i n P a c i f i c s i l v e r f i r , as was postulated for balsam f i r by Neale and Adams 108 (1985b), and l i k e l y s e l ection pressures are d i f f e r e n t i n d i f f e r e n t environments (Linhart et al. 1981). Ritland (1983) contends that the assumption of inbreeding e q u i l i b r i a i s not unreasonable i n highly outcrossed species but the known i n s e n s i t i v i t y of estimates of f i x a t i o n indices to the detection of inbreeding at l e v e l s believed to be e x i s t i n g i n most conifer stands (F < 0.10; Ward and Sing 1970, Brown 1979, Shaw and A l l a r d 1982b), would weigh i n favour of the inbreeding lev e l s predicted by the multilocus outcrossing rate. Inbreeding l e v e l s estimated from t m should be more precise for the sample sizes available to t h i s study because of i t s greater s t a t i s t i c a l e f f i c i e n c y (Ritland and El-Kassaby 1985). An e f f i c i e n t s t a t i s t i c i s defined by Zar (1984) as one that when obtained from any single sample w i l l be very close to the value of the parameters being estimated. The method of estimating f i x a t i o n used in t h i s study possesses a r e l a t i v e l y large inherent bias (Weir and Cockerham 1984) unless sample sizes are very large. The f i x a t i o n index based on samples of viable seed i s by nature a conservative estimate, as i t contains the e f f e c t s of any early embryonic competition which may occur p r i o r to sampling (Brown and A l l a r d 1970). For t h i s reason alone, i n d i r e c t estimates of F may be biased downward. 109 6.5 Conclusion The expectation of a high degree of population d i f f e r e n t i a t i o n based on the biology of P a c i f i c s i l v e r f i r was not borne out i n the analysis of. electrophoretic v a r i a t i o n , but t h i s result may not be that surprising considering the l i m i t s to detection placed on i t by the nature of the variables being estimated. Nonetheless there are several l i n e s of evidence which point to some l i m i t e d heterogeneity among populations. Heterozygosity appears to be reduced i n northern la t i t u d e s and i n embryo gene pools. Heterozygosity appears to be higher i n extant populations, suggesting selection i s acting to eliminate inbred seeds in nature. Greater genetic d i v e r s i t y , as measured by n e or He, i s seen in populations sampled i n southern Vancouver Island. Interestingly, a more southerly population, F, i s also the most genetically d i f f e r e n t i a t e d , as revealed by both D and 8. However, there i s a lack of agreement between G S T and 8 in the detection of population genetic structure in P a c i f i c s i l v e r f i r . Lack of agreement between dire c t and i n d i r e c t estimates of the f i x a t i o n index suggests that other evolutionary mechanisms i n addition to the mating system may be influencing the d i s t r i b u t i o n of genotypes within populations of P a c i f i c s i l v e r f i r on Vancouver Island. 110 7. SEED GERMINATION 7.1 Introduction At present, reforestation i n B r i t i s h Columbia r e l i e s almost exclusively upon planting stock grown from seeds (Konishi et al. 1989). Ever-increasing costs of producing containerized stock demand the use of high q u a l i t y seeds. Edwards (1982) reported that seeds of Abies species are frequently of lower q u a l i t y than other conifers, the range i n nursery germination being t y p i c a l l y 20-50% (Franklin 1974). Although improvements i n the c o l l e c t i o n and handling of Abies seeds have increased germination, true f i r seeds are known to exhibit varying degrees of dormancy, which hampers nursery production (Leadem 198 6). According to Wang (1981) the components of seed quality at the physiological l e v e l include seed v i a b i l i t y , germinability and vigour. The v i a b i l i t y of a seed i s simply i t s capacity for growth and development (Bewley and Black 1978). Germinability i s a measure of the a b i l i t y of a population of seeds to germinate, or as Bewley and Black (1978) describe i t , "the maximum percentage of seeds that w i l l germinate under favorable conditions". Vigour i s more problematic i n i t s d e f i n i t i o n because of i t s complexity. The vigour of seeds was seen by Heydecker (1969) to have as many "shades of meaning" as seed q u a l i t y . The Association of O f f i c i a l Seed Analysts (AOSA) stated that vigour i s operating on at least two l e v e l s - at the biochemical l e v e l , I l l as the coordination of several metabolic events and at the macroscopic l e v e l , i n the speed and completeness of germination over a range of environmental conditions (Anon. 1976a). Vigour i s controlled by two major factors - one being genetic and the other consisting of various environmental conditions which may occur during seed development, maturation, processing and storage (Heydecker 1969, Maguire 1977). Seed vigour i s then the sum of a l l "those properties which determine the p o t e n t i a l for rapid, uniform emergence and development of normal seedlings under a wide range of f i e l d conditions" (Bonner 1984). Germination i n coniferous seeds i s the culmination of a complex of metabolic a c t i v i t y involving two d i s t i n c t genomes (the d i p l o i d embryo surrounded by the n u t r i t i o n a l and osmotic envelopes of the haploid megagametophyte) and s p e c i f i c environmental t r i g g e r s . In addition, germination responses are l i k e l y conditioned by environments encountered by seeds throughout t h e i r development (Rowe 1964). The process of seed development which involves the accumulation of nutrient reserves and the eventual suspension of embryo growth. As a seed matures, the water content can drop dramatically (to about 10%; Bewley and Black 1978) whereupon normal metabolism i s disrupted. In t h i s so-called quiescent state the embryo can often remain a l i v e for extended periods of time. Jann and Amen (1977) maintain that a quiescent seed i s readily germinable - that 112 i s , growth w i l l resume when the seed i s exposed to favourable conditions. This i s i n marked contrast to a dormant seed defined by Amen (1963) as one that w i l l not germinate under conditions normally considered favorable for i t s growth. In dormant seeds, the temporary suspension of growth i s credited to some endogenous i n h i b i t o r y mechanism(s) for which there may be s p e c i f i c environmental t r i g g e r s (Jann and Amen 1977). Thus, both germination and dormancy may be viewed as gross manifestations of a seed's genetic program being engaged (or disengaged) i n response to the p a r t i c u l a r environment which i t encounters. Germination responses are subject to endogenous control, presumably through the mediation of hormones contained i n the seed i t s e l f , as well as external cues (Jann and Amen 1977, Wang et al. 1982). S t r a t i f i c a t i o n - moist, low-temperature storage for a few to several weeks either i n some medium (Allen 1941) or "naked" (Allen and Bientjes 1954) - i s a commonly-used dormancy-breaking treatment i n temperate zone conifer species (Wang et al. 1982). Conditions of s t r a t i f i c a t i o n (or p r e c h i l l i n g ) are set to approximate the environments that autumn-ripening seeds might f i n d themselves exposed to upon dissemination (Krugman et al. 1974). The degree of dormancy may be expected to show some v a r i a t i o n related to climate of o r i g i n (Levins 1969, Thompson 1981). Campbell and Ri t l a n d (1982) found populations of western hemlock at 113 higher l a t i t u d e s to exhibit e a r l i e r and more rapid germination, a pattern which was detected i n other forest tree species also inhabiting climates where cold temperatures l i m i t the growing season (see references t h e r e i n ) . S t r a t i f i c a t i o n has been shown to improve the. germination (in terms of capacity and/or speed) of several Abies species, which i s also taken as evidence that dormancy ex i s t s i n these seeds (Edwards 1962). P r e c h i l l i n g produces ameliorating effects on the germination of grand, subalpine and P a c i f i c s i l v e r f i r in degrees varying with both species and seedlot (Edwards 1982, Leadem 1986). Single-tree seed c o l l e c t i o n s available to t h i s study make i t possible to analyze germination responses on a l e v e l below that of the seedlot (where seedlot i s seen as equivalent to population). A germination test of six populations of P a c i f i c s i l v e r f i r was designed to obtain some estimate of the magnitude of family v a r i a t i o n i n germination response, r e l a t i v e to that of population and c o l l e c t i o n region (populations grouped by l a t i t u d e ) , and how a given s t r a t i f i c a t i o n regime might aff e c t t h e i r variance structure. As well, the. size of the test required that r e p l i c a t i o n s be s p l i t between two germination cabinets, which was also incorporated into the germination model. 114 7.2 Materials and Methods 7.2.1 C o l l e c t i o n and t e s t i n g methods Cones were co l l e c t e d from eight populations of P a c i f i c s i l v e r f i r as described i n Chapter 2. Owing to space r e s t r i c t i o n s for germination and subsequent progeny t e s t i n g , six populations, two each from north, middle and southern l a t i t u d e c o l l e c t i o n regions of Vancouver Island were chosen and a subset of seven trees within each population were randomly selected, where there were s u f f i c i e n t f i l l e d seeds. Six random samples of 50 f i l l e d seeds were obtained for each tree (42 trees i n total) by X-ray. A double germination test, modified from the International Seed Testing Association's (Anon. 1976b) rules for the t e s t i n g of A. amabilis seeds, was conducted at the Canadian Forestry Service, P a c i f i c Forestry Centre, V i c t o r i a B.C. One half of the seeds were subjected to a 28-day s t r a t i f i c a t i o n (prechill) period p r i o r to incubation of a l l seeds on February 7, 1985. P r e c h i l l i n g e n t a i l e d placing dry seed samples i n clear p l a s t i c , sealed germination boxes (12x12x3 cm) on three layers of Whatman #1 f i l t e r paper over "Kimpak" c e l l u l o s e towelling wetted with 43 ml d i s t i l l e d water. Dishes were immediately placed i n darkness at 1-4° C for 28 days. 115 U n s t r a t i f i e d seeds were set into germination boxes i n a s i m i l a r manner but placed immediately along with p r e c h i l l e d seeds into two upright germination cabinets. The experiment was too large to be c a r r i e d out i n one cabinet but was structured such that three samples of both p r e c h i l l e d and untreated seeds from each tree were placed at random on trays i n each cabinet. Temperatures were maintained a l t e r n a t e l y at 30° C for eight hours and 20° C for 16 hours, with l i g h t at approximately 1000 lux being provided during the higher temperature period using cool-white fluorescent tubes. Germinants were counted eight times during the 28 day germination period with two observations during the f i r s t week to assess onset of germination. The test period was extended and additional counts were made at 35 and 42 days to reduce the truncation e f f e c t on seeds from l a t e r -germinating trees. Germinants were removed when the r a d i c l e had reached the length of the seed coat (Edwards 1982). The number of germinants that would be considered abnormal according to ISTA (Anon. 1976b) rules ( i . e . cotyledon emergence, twin radicles) was extremely small (less than 0.05%). This number may have been higher had counts been made when a l l "normal" structures could be assessed as recommended by ISTA (Anon 1985). However, t h i s procedure requires a longer period of incubation which would increase the l i k e l i h o o d of both fungal contamination and counting errors as a res u l t of crowding. For the purpose of t h i s 116 experiment, the few abnormal germinants observed were classed as normal. Germination counts were summarized as two response variab l e s : germination capacity (GC), the number of germinants, expressed as a percentage of f i l l e d seeds, at the end of the t e s t ; and germination value (GV) , computed according to Czabator (1962). This index of germination i s the product of two quantities: The mean d a i l y germination (MDG), obtained by d i v i d i n g the t o t a l number of germinants by the length of the test period (in days), and the peak value (PV) which i s determined by c a l c u l a t i n g a cumulative germination percentage for each successive count and d i v i d i n g by the number of elapsed days. The maximum quotient corresponds to PV, and GV = MDG x PV. The higher the value of GV, the more complete and/or the more rapid the germination process. 7.2.2 Analytic methods Previous work on p r e c h i l l i n g treatments on P a c i f i c s i l v e r f i r (Davidson et al. 1984) suggested that the s t r a t i f i c a t i o n process would a l t e r germination patterns to the extent that i t would be very u n l i k e l y that homogeneity of variances between treatment groups would be achieved. Given t h i s p r o b a b i l i t y , preliminary analysis of the germination data was ca r r i e d out using a multi-way contingency table approach (Fienberg 1970) i n order to get 117 some estimate of the relationships among the hypothesized sources of v a r i a t i o n without invoking the analysis of variance (ANOVA) assumptions of homoscedasticity and normality. The method also accommodates censored data sets. The analysis i s based on f i t t i n g a l o g - l i n e a r model to i n d i v i d u a l c e l l frequencies, i n t h i s case, the number of germinants per day. The r e l a t i v e importance of a given factor i n the model i s determined by obtaining an approximate % 2 value for the f i t of a p a r t i c u l a r model containing the factor of interest and then r e f i t t i n g the model without that factor and observing the change i n % 2. The magnitude of the difference r e f l e c t s the r e l a t i v e importance of the term of interest (Schoener 1970). Not a l l factors were testable by t h i s method, however, which i s not uncommon where models contain both fixed and random e f f e c t s (M. Grieg, UBC Computing Center, pers. comm. November 1985). The analysis did c l e a r l y reveal that the s t r a t i f i c a t i o n treatment was the greatest single factor a f f e c t i n g germination patterns in sampled P a c i f i c s i l v e r f i r . Further analyses were conducted on s t r a t i f i e d and u n s t r a t i f i e d seeds separately. Recognizing that employing censored data would ultimately lead to an underestimate of error variances, a decision to use ANOVA was made because of the reasonably high germination exhibited by most trees i n the study and because ANOVA i s frequently used i n studies of seed source 118 v a r i a t i o n i n conifers. An ad-hoc procedure for finding suitable transformations to normalize the calculated response variables and achieve homogeneity of variances (A. Kozak, UBC Faculty of Forestry; pers. comm. December 1985) was u t i l i z e d for GC and GV of s t r a t i f i e d seed and for GC of u n s t r a t i f i e d seed. Box's (1949) test for equality of variances was used i n conjunction with an appropriate power transformation ( a l l performed using MIDAS s t a t i s t i c a l software, Fox and Guire 1976) to obtain variables suitable to ANOVA. Lack of normality (Hicks 1982) and heteroscedasticity (Glass et al. 1972) do not seriously a f f e c t ANOVA for balanced designs. Attempts to achieve homogeneity by some data transformation usually improves normality and a d d i t i v i t y of ef f e c t s (Zar 1984), however the censored nature of the data remains inherent in the study. Germination value for u n s t r a t i f i e d seeds f u l f i l l e d ANOVA assumptions without transformation. Where transformation was of benefit, results using untransformed variables are included for comparison. A nested-factorial analysis was based on the following model: i jklm = Ji. + Ri + P + T k ( i j ) + C1 + CR n + CP j l ( i ) + CT k l ( i j ) + ^ ( i j k l ) where ji = o v e r a l l mean germination response RA = clim a t i c region (i=l,2,3) P j ( i ) = population within c l i m a t i c region (j = l,2) T k(ij) = tree within population (k=l,...,7) 119 C 2 = cabinet (1=1,2) em(ijki) = error (m=l,2,3) A l l e f f e c t s i n the model were considered random except R and C which were deemed to be fixed. Expected mean squares were included to indicate appropriate terms for s i g n i f i c a n c e t e s t i n g and to enable estimation of variance components. In t h i s chapter, the r e l a t i v e magnitudes of v a r i a t i o n which may be ascribed to factors i n the model are presented as r a t i o s of the appropriate variance components to t h e i r sum (CV, expressed as a percentage). The c o e f f i c i e n t of i n t r a c l a s s c o r r e l a t i o n ( r I # Sokal and Rohlf 1981), which i n these analysis measures the proportion of v a r i a t i o n among maternal trees,, i s also computed. This value i s referred to by Falconer (1981, p. 126) as r e p e a t a b i l i t y and may be viewed as an upper l i m i t of h e r i t a b i l i t y i n the broad sense. In addition, the apportionment of v a r i a t i o n based on a percentage of the t o t a l sums of squares (%SS or eta 2, af t e r Fisher 1932, L i t t l e 1981, Hicks 1982) i s presented. This method of apportioning v a r i a b i l i t y i s appealing because there i s no chance of obtaining negative variance components when source contributions are very small ( i . e . Huehn et al. 1987). Numerically, the s i m i l a r i t y between %SS and respective %CV was shown for Douglas-fir growth variables by Maze et al. (1989). However, when compared with the equivalent variance components, residual v a r i a t i o n i s 120 usually underestimated using %SS, so that the proportion of the t o t a l v a r i a t i o n i n the data accounted for by other terms should be considered maximal (Hicks 1982 p. 135). 7.3 Results and discussion Results of /ANOVA for germination capacity (transformed) of s t r a t i f i e d seeds are presented i n Tables 7.1a and 7.1b. It i s evident that the cabinet e f f e c t and i t s interactions are n e g l i g i b l e . A s i m i l a r lack of s i g n i f i c a n c e was obtained using response variables for u n s t r a t i f i e d seeds, prompting the decision to remove the cabinet terms from the model, thus improving the error degrees of freedom. The region of c o l l e c t i o n (R) - northern, mid and southern Vancouver Island was retained in the model despite i t s small contribution because of i t s implication for seed crop management. Populations within regions and trees within populations account f o r a substantial portion of t o t a l variance, as revealed by either %SS or %CV (Table 7.1b). The frequency of germinants vs. non-germinating seeds compiled i n Table 7.2 further i l l u s t r a t e s both cabinet and s t r a t i f i c a t i o n e f f e c ts on germination capacity. There was very l i t t l e difference in the number of germinants of either u n s t r a t i f i e d or s t r a t i f i e d seed between the two germination cabinets (less than 2% of the t o t a l number of germinants i n the t e s t ) . There was, however, substantial difference i n the number of seeds l e f t ungerminated at the end of the test Table 7.1a Sources of v a r i a t i o n , associated degrees of freedom, sum of squares, mean squares, F values and associated p r o b a b i l i t i e s for a mixed effects ANOVA model of germination capacity for s t r a t i f i e d seeds (transformed values). Source of Degree of Sum.of Mean F Probability V a r i a t i o n Freedom Squares Squares Region (R) 2 0. .515 0, .257 0. .155 0 .86 Population (P(R)) 3' 4. .991 1, .664 11. .228 < 0 .00 Tree (T(PR)) 36 5, .335 0. .148 12. .268 < 0 .00 Cabinet (C) 1 0. .003 0. .003 0. .085 0 .78 C X R 2 0. . 053 0. .027 0. .695 0 .57 C X P (R) 3 0. ,115 0. .038 2. .456 0 .08 C X T (PR) 36 0, .560 0, .016 1. .288 0 .15 Residual 168 2. .029 0. .012 Total 251 13. . 601 Table 7.1b Sources of v a r i a t i o n , percentage of t o t a l v a r i a t i o n based on sums of squares (%SS), components of variance (%CV) and expected mean squares based on a mixed e f fec ts ANOVA model of germination in P a c i f i c s i l v e r f i r on Vancouver I s land . Calculat ions based on ANOVA described in Table 7.1a. Source of %SS %CV Expected V a r i a t i o n Mean Squares Region (R) 3. 8 3.3 <T2e + 6 o 2 T P R + 42a 2 P R + Populat ion (P(R)) 36. 7 48.8 ° 2 e + 6 c 2 T P R + 42a2pR Tree (T (PR) ) 39. 2 28.4 ° 2 e + 60" 2 T P R: Cabinet (C) < 0. 0 < 0.0 ° 2 e + 3o"2CTPR + 21C 2 C P R + 126())c C X R 0. 4 0.7 ° 2 e + 3o2 C T P R + 2 l a 2 C P R + 42a 2 C R C X P (R) 0. 9 2.2 ° 2 e + 3G 2 C T P R 2 l a 2 C P R C X T(PR) 4 . 1 1.5 <52e + 3o2 C T P R Residual 14 . 9 15.1 Tota l 100. 0 100.0 Table 7.2 Total number of germinants and non-germinants i n a l l 42 trees involved in a paired germination test of six populations of P a c i f i c s i l v e r f i r . One half of the seeds were subjected to a 28-day s t r a t i f i c a t i o n p r i o r to t e s t i n g ; seeds from each pre-treatment were further divided between two germination cabinets. PRETREATMENT CABINET No. Germinated No. Ungerminated. Total U n s t r a t i f i e d A 4718 1260 5978 B 4890 1213 6103 TOTAL 9608 2473 12081 S t r a t i f i e d A 5429 607 6096 B 5317 705 6022 TOTAL 10746 1372 12188 as a r e s u l t of pretreatment, with over 20% of the seeds remaining for u n s t r a t i f i e d compared to just over 11% for seeds subjected to the 28-day p r e c h i l l . This r e s u l t suggests that t h i s sample of P a c i f i c s i l v e r f i r seeds possess some degree of dormancy. Overall means for germination capacity were 79.6 ± 1 . 8 % (95% CI) and 89.9 ± 1.8% for u n s t r a t i f i e d and s t r a t i f i e d seeds, respectively (calculated from untransformed data). The influence of s t r a t i f i c a t i o n on both the t o t a l amount and r a p i d i t y of germination i s r e f l e c t e d more dramatically i n the o v e r a l l increase i n germination value, from an average of 4.42 ±. 124 0.22 to 11.37 ± 0.44. These re s u l t s suggest that dormancy i s indeed a factor influencing germination and i t i s evident that the s t r a t i f i c a t i o n treatment i s e f f e c t i v e i n overcoming dormancy i h t h i s species. S t r a t i f i c a t i o n has been found to improve t o t a l germination i n several seedlots of P a c i f i c s i l v e r f i r (Edwards 1980, Leadem 1986). The r e s u l t s also show that the variances associated with average GC i n u n s t r a t i f i e d and s t r a t i f i e d seeds are equivalent, and perhaps more revealing, the variance i n GV for p r e c h i l l e d seeds i s twice that of seeds not subjected to p r e c h i l l i n g . This i s not the t y p i c a l pattern one would expect from s t r a t i f i c a t i o n . The usual e f f e c t of s t r a t i f i c a t i o n i n conifers i s a hastening of germination with a concomitant reduction i n v a r i a b i l i t y (Allen and Bientjes 1954, Edwards 1969) . Examination of the results of h i e r a r c h i c a l ANOVA for both u n s t r a t i f i e d and s t r a t i f i e d seeds (Table 7.3a and b respectively) reveals that the largest source of v a r i a b i l i t y i n both GC and GV i s associated with differences among trees. Consistently high values of i n t r a - c l a s s c o r r e l a t i o n s , obtained for both u n s t r a t i f i e d and s t r a t i f i e d seeds, further emphasize the large i n t e r - t r e e v a r i a t i o n . Regions are i n most analyses r e l a t i v e l y unimportant, and the major e f f e c t of s t r a t i f i c a t i o n on the apportionment of v a r i a t i o n i s seen i n the s h i f t i n g of nearly h a l f of the r e l a t i v e variance associated with i n d i v i d u a l trees to the 125 population l e v e l . I ntra-individual v a r i a t i o n remains approximately the same (less than 25% of t o t a l SS). It i s apparent that seed pretreatment by s t r a t i f i c a t i o n f or 28 days has varying e f f e c t s on populations within the same l a t i t u d i n a l band. There does not seem to be an immediate b i o l o g i c a l explanation for the apparent change in the r e l a t i v e importance of in d i v i d u a l and population e f f e c t s r e s u l t i n g from seed pretreatment. Seedlot v a r i a t i o n i n the degree of dormancy i n Abies i s well known (Edwards 1962) and t h i s might be attributed to the timing of c o l l e c t i o n , as Edwards (1969) observed increased dormancy from early to la t e c o l l e c t i o n s of noble f i r . Cones from both populations within any one c o l l e c t i o n region in the present study were c o l l e c t e d at most one day apart. However, Edwards (1982) reports that maturity differences can exist among cones within the same tree and even among seeds within any one cone. Cones were monitored for embryo maturity v i a cone-cutting t e s t s but the l o g i s t i c s of c o l l e c t i o n required that cones be picked from a l l trees in a given stand on the same day. Thus i t i s unlikely that the sampled trees represent the same degree of seed maturity within each population. A geographic trend in dormancy i s not evident i n samples of P a c i f i c s i l v e r f i r . None of the samples could be considered deeply dormant as average germination capacities (Table 7.4) for u n s t r a t i f i e d seeds i n a l l six populations Table 7.3a The percentages of t o t a l sums of squares and equivalent variance components from ANOVA associated with each source of var i a t i o n for germination response variables in u n s t r a t i f i e d seeds of P a c i f i c s i l v e r f i r (TRANS = transformed values). Source of Degrees of % Sums of Squares % Components of Variance Var i a t i o n Freedom GC (TRANS)GC GV GC (TRANS)GC GV Region 2 1, . 1 1. 0 9. 7*® 0. .0 0. ,0 11. .7 Population 3 5, .7 6. . 6 1. 2 0. .0 0. ,0 0, .0 Tree 36 74 , .7 69. .3* 71. 5* 79. .1* 66. ,2* 69, .8* Residual 210 18 , .5 23. ,1 17 . 6 20. .9 33, ,8 18, .5 Total 251 100. .0 100. .0 100. 0 ' 100. .0 100, ,0 100, .0 rx® -- 0. ,79 0. , 66 0, .79 ® *, s i g n i f i c a n t at P < 0.05 ® r : , i n t r a - c l a s s c o r r e l a t i o n c o e f f i c i e n t = 0 2 T P R / o*2TPR + o 2 e Table 7.3b The percentages of t o t a l sums of squares and equivalent variance components from ANOVA associated with each source of va r i a t i o n for germination response variables in s t r a t i f i e d seeds of P a c i f i c s i l v e r f i r (TRANS = transformed values). Source Degrees % Sums of Squares % Components of Variance of of Variation Freedom GC (TRANS)GC GV (TRANS)GV GC (TRANS)GC GV (TRANS)GV R 2 7, .5 3. ,8 4 , , 8 4 , 4 0. .0 0. .0 0. .0 0, .0 P(R) 3 33, . 8*® 36. , 7* 29. .7* 28, ;6* 47, . 9* 50, .2* 41. .0* 39, .4* T (PR) 36 42 . 4* 39. .2* 48. .5* 50, ,1* 36. , 6* 31. .4* 42. .7* 44. .3* Residual 210 16. , 3 20. ,3 17 . 0 16. .9 15. .5 18, .4 16. ,3 16, .3 Total 251 100. , 0 100. ,0 100. ,0 100. ,0 100. .0 100, .0 100. .0 • 100. .0 rr® -- -- 0. ,70 0, , 63 0. ,72 0. ,73 ® *, s i g n i f i c a n t at P < 0.05 ® r t , i n t r a - c l a s s c o r r e l a t i o n c o e f f i c i e n t = 0"2TPR / C 2 T P R + 0 2 E 128 are consistently high. In response to the s t r a t i f i c a t i o n regime applied i n t h i s study, t o t a l germination (GC) was s i g n i f i c a n t l y (P < 0.05) increased and v a r i a b i l i t y reduced in a l l populations except W. Fewer germinants were obtained and germination was more variable for s t r a t i f i e d seeds than u n s t r a t i f i e d seeds i n t h i s population. The speed of germination (inferred from GV) improved the least i n population W (Table 7.5). Its l a t i t u d i n a l counterpart, F, on the other hand, exhibits a more c l a s s i c response to s t r a t i f i c a t i o n . Sound seeds, already possessing Table 7.4 Average germination capacity (±95% confidence interval) for six populations of P a c i f i c s i l v e r f i r for both u n s t r a t i f i e d and s t r a t i f i e d seeds. Region Population > PREGERMINATION TREATMENT No 28-day S t r a t i f i c a t i o n S t r a t i f i c a t i o n North R 82.2 (3.7) 92.5 (2.9) H 76.2 (4.9) 91.0 (2.9) Middle B 76.2 (5.5) 89.2 (3.5) C 79.7 (3.7) 94 .7 (2.4) So-uth F 86.6 (3.9) 97.8 (1.4) W 76.8 (3.7) 68. 9 (5.9) a reasonably high capacity for exposed only to the conditions (i. e . r e l a t i v e l y non-dormant), germination aft e r of c o l l e c t i o n and show higher, more being processing uniform 129 germination with a marked increase i n GV after p r e c h i l l i n g 28 days. However, the variance of GV increased for s t r a t i f i e d seeds i n a l l six populations, although t h i s increase was le a s t i n population F (Table 7.5). This unexpected r e s u l t merits further examination. Table 7.5 Average germination values (Czabator, 1962) followed by 95% confidence i n t e r v a l for six populations of P a c i f i c s i l v e r f i r , for unstrat-i f i e d and s t r a t i f i e d seeds. REGION POPULATION PREGERMINATION TREATMENT No S t r a t i f i c a t i o n 28-day S t r a t i f i c a t i o n North Middle South R H B C F W 4.55 (0.51) 3.93 (0.51) 4.36 (0.63) 4.16 (0.43) 5.48 (0.49) 5.34 (0.45) 11.06 (0.94) 11.63 (0.96) 11.82 (1.10) 12.83 (0.69) 13.70 (0.59) 7.15 (0.90) Populations R and H, sampled from northern Vancouver Island, exhibit s i m i l a r germination values for both u n s t r a t i f i e d and s t r a t i f i e d seeds and both populations exhibit v i r t u a l l y the same le v e l s of v a r i a b i l i t y among trees (Table 7.5). In the mid-latitude c o l l e c t i o n region (B and C), germination behaviour p a r a l l e l s that in the north, although population C appears to respond more rapidly and 130 uniformly to s t r a t i f i c a t i o n . Germination values are highest i n the southernmost c o l l e c t i o n region without s t r a t i f i c a t i o n , suggesting that germination i s more rapid at lower l a t i t u d e s . However, the average GV for s t r a t i f i e d seeds of population W i s considerably less than that of F. This r e s u l t when considered along with the substantial i n d i v i d u a l tree component of v a r i a t i o n i n germination response (Table 7.3b) prompted a closer look at germination behavior within population W. The average GC and GV of six 50-seed r e p l i c a t i o n s for each tree representing W are l i s t e d in Table 7.6. Germination percentages range from 54 to 92 for u n s t r a t i f i e d seeds and 38 to 95 for s t r a t i f i e d seed. The s t r a t i f i c a t i o n treatment appeared to be detrimental to seeds from four of the seven trees i n the sample (mean germination capacity s i g n i f i c a n t l y reduced). Germination value was s i g n i f i c a n t l y improved only i n the three trees in which germination capacity did not diminish in response to s t r a t i f i c a t i o n (Table 7.6). Several studies on tree seed maturity reviewed by Edwards (1980) have revealed that immature seeds tend to: a) be l i g h t e r i n weight; b) germinate slowly i f at a l l ; c) show reduced germination as a result of p r e c h i l l i n g ; and d) be more susceptible to disease. Populations F and W were the l a s t to be co l l e c t e d (September 30, vs September 8/9 for B and C and September 28 for R and H) and also produced the 131 Table 7.6 Mean germination capacity (GC) and germination value (GV) for i n d i v i d u a l trees of population W (each mean based on six 50-seed rep l i c a t i o n s ) and t h e i r associated 95% confidence i n t e r v a l s , and population mean values for both germination responses (*, s i g n i f i c a n t treatment response, P < 0.05). MEAN GERMINATION CAPACITY Tree No. No S t r a t i f i c a t i o n S t r a t i f i c a t i o n 1 75. . 1 (2. .6) 58. .7 (4. .6)* 2 77. , 7 (3. . 6) 64. .5 (8. .9)* 3 85. .7 (4. .2) 85. .2 (4. .5) 4 54 . 5 (8. .1) 38. .2 (7. • D* 5 77. . 0 (4. .3) 64. .7 (7. .0)" 6 92. .0 (3. .2) 95. . 6 (1. .9) 7 75. .5 (3. .9) 75. .7 (9. .9) Mean 76. .8 (3. .7) 68. . 9 (5. .9) MEAN GERMINATION VALUE Tree No. No S t r a t i f i c a t i o n S t r a t i f i c a t i o n 1 5 .23 (0 .25) 5 . 07 (0 . 69) 2 5 .86 (0 .82) 6 .99 (1 .73) 3 6 . 95 (0 .89) 9 .49 (0 .54) * 4 2 .54 (0 .83) 2 .36 (0 . 81) 5 5 .54 (0 . 65) 7 . 15 (1 .62) 6 6 .23 (0 .39) 10 .67 (0 .51) * 7 5 .03 (0 .48) 8 .29 (1 .52) * ;an 5. 34 (0. 45) 7. 15 (0. 90) * heaviest seeds (42.8 ± 5 . 3 and 53.1 ± 12.5 g are the mean values and t h e i r standard deviations for 1000-seed weights of populations F and W, respectively) of a l l six populations 132 (overall average 33.3 ± 13.1 g). Ackerman and Gorman (1969) found that l i g h t e r weight seeds had lower germination percentages i n lodgepole pine. Seed weights of ponderosa pine were also correlated with germination capacity and among stand differences were found to be s i g n i f i c a n t i n a study by Wang and Patel (1974). In l o b l o l l y pine, heavier seeds were shown to have better germination with s i g n i f i c a n t family differences i n seed size (Hodgson 1980, c i t e d i n Wang et al. 1982). Correlation of seed weight with germination capacity was also high in wide-ranging samples of white spruce and seed weight was also found to correlate s i g n i f i c a n t l y with population l a t i t u d e in t h i s species ( K h a l i l 1986). The 1000-seed weight of provenances of Douglas f i r was found to correlate with a l t i t u d e of o r i g i n (Birot 1972) and i n jack pine, Chalupa and Durzan (1972) found seed sizes highly correlated with climate of seed o r i g i n . Simple l i n e a r correlations among population mean seed weight and germination response variables and also l a t i t u d e for P a c i f i c s i l v e r f i r (Table 7.7) indicate that population W, with i t s large seeds and more southerly l a t i t u d e , appears anomalous i n i t s germination behavior. Without W, germination speed (inferred from GV, Table 7.5) i s c l o s e l y related to seed weight and given the negative c o r r e l a t i o n of weight with lati t u d e , faster germination appears c h a r a c t e r i s t i c of more southern populations, the reverse of that found by Campbell and Ritland (1982) in 133 Table 7.7 Simple l i n e a r correlations between average seed siz e per population and germination response variables and latitude of populations, calculated with and without values for population W. The c r i t i c a l values of r for three and four degrees of freedom are 0.878 and 0.811 (* P < 0.05), respectively. Without W With W Seed size and i) GC a) u n s t r a t i f i e d 0.872 0.238 b) s t r a t i f i e d 0.763 -0.624 i i ) GV a) u n s t r a t i f i e d 0.959* 0.921* b) s t r a t i f i e d 0.731 -0.537 i i i ) l a t i t u d e of -0.872 -0.882* population western hemlock. Thus, c o l l e c t i o n date, seed size and/or maturity per se do not adequately account for the differences observed in germination behaviors between F and W. In both untreated and treated seeds, differences i n germination responses are most strongly associated with in t e r - t r e e (family) v a r i a t i o n (Tables 7.3 and 7.4). Farmer and Rienholt (1986) observed large family differences i n germination responses of tamarack seeds exposed to d i f f e r e n t temperature and l i g h t combinations, and stand and provenance differences were i n a l l cases non-significant, prompting them to speculate a high degree of genetic control over seed q u a l i t y and germination c h a r a c t e r i s t i c s , since environments were presumed to be r e l a t i v e l y uniform within stands. 134 S i g n i f i c a n t family v a r i a t i o n was also found i n yellow poplar (Barnett and Farmer 1978). Both germination capacity and germinative energy were related to parental genotype i n con t r o l l e d crosses of V i r g i n i a pine (Bramlett et al. 1983). The majority of var i a t i o n in germination percentage was att r i b u t e d to differences among 19 black spruce clones by Stoehr and Farmer (1986) . They also observed that a few clones with "weak, decay-prone" seed contributed heavily to t h i s variance. Leadem (1986) found v a r i a t i o n among seedlots of P a c i f i c s i l v e r f i r in response to s t r a t i f i c a t i o n and suggested differences may be the resu l t of vigour differences among seeds. An additional f i n d i n g of Leadem (1986) was that the ISTA (Anon. 1976b) prescribed incubation temperature was too high for P a c i f i c s i l v e r f i r seeds, which germinated better at alternating temperatures of 15 and 10° C. This finding reinforces results from a seedlot of P a c i f i c s i l v e r f i r germinated on a thermogradient plate at f i v e constant temperatures (Davidson et al. 1984). It was determined that greater germination occurred at and below 21° C due i n part to the prevalence of seed coat fungi at 24 and 27° C which infected emerging r a d i c l e s . Mold growth i s a freguent problem i n germination studies of Abies species (Edwards 1969, 1982; K i t z m i l l e r et al. 1973, Adkins 1983, Blazich and Hinesley 1984) and high temperatures increase s u s c e p t i b i l i t y to fungal attack (Leadem 1986). 135 Every e f f o r t was made to avoid damage to the re s i n v e s i c l e s present i n the seed coats of P a c i f i c s i l v e r f i r obtained for the present study. The actual function of the r e s i n i s unknown in P a c i f i c s i l v e r f i r but i t was found to i n h i b i t germination i n seeds of white and red f i r s ( K i t z m i l l e r et al. 1975) and to be a good medium for fungal development (Kitzmiller et al. 1973). When the cones c o l l e c t e d from population W arrived at UBC, molds were v i s i b l e on cones of two of the 17 trees. Neither of these two trees were included in the germination test, however molds were present i n every germination dish for a l l trees in population W, without exception, by the fourth week of the germination t e s t . Molds occurred in other populations as well, but none were infected to the l e v e l seen i n population W, p a r t i c u l a r l y for s t r a t i f i e d seed. The s t r a t i f i c a t i o n conditions, employed in t h i s test i n terms of moisture, temperature and duration, may not have been optimal for population W, permitting molds to f l o u r i s h while i n h i b i t i n g germination. S t i l l , i n d i vidual trees varied i n t h e i r propensity to germinate despite the presence of fungi, as seen i n Table 7.6. Although the germination patterns revealed by t h i s study may have resulted i n part from seed preconditioning p r i o r to harvest, as suggested by Campbell and Ritland (1982) for western hemlock, the large i n d i v i d u a l tree e f f e c t would argue for a s i g n i f i c a n t genetic component i n germination c h a r a c t e r i s t i c s , at least in the 1983 cone-crop 136 year. Sorensen and Franklin (1977) found large year e f f e c t s on seed weight i n noble f i r , which suggests that any preconditioning e f f e c t s associated with parent tree environment may be temporally variable. Environmental preconditioning can mimic genetic differences among populations (Campbell and Ritland 1982) and, as A l l e n (1958) pointed out, the genetic component may be masked by the kind of handling seeds receive. It i s not known to what extent these factors affected germination performances observed i n t h i s t e s t . Repeating the c o l l e c t i o n in another year as well as varying after-ripening and s t r a t i f i c a t i o n conditions would help to further define the factors influencing i n t e r -and intra-population v a r i a t i o n . 7.4 Conclusions This study does not reveal any strong geographic trend i n dormancy over the sampled range, but, the high degree of family v a r i a t i o n has important implications for nursery germination i n that there may be inadvertent selection for rapid-germinating families within a given seedlot. In addition, s t r a t i f i c a t i o n , i s l i k e l y to affect seedlots d i f f e r e n t l y and i t s effectiveness may be a complex function of the environments of developing seed, handling conditions as well as i n t e r - and intra-tree vari a t i o n in dormancy and germination responses. 137 The large f a m i l i a l component of v a r i a t i o n detected i n t h i s study suggests that seedlings of P a c i f i c s i l v e r f i r may be more e f f i c i e n t l y obtained by single tree cone c o l l e c t i o n and germination/planting on a family basis. The apparent c l i n a l trend i n seed size and i t s rel a t i o n s h i p to germination performance could also have implications for nursery prac t i c e . Separation of f i l l e d and empty seeds i s d i f f i c u l t i n Abies species (Edwards 1979, 1982) and f i l l e d but l i g h t seeds present in bulked seedlots could be l o s t , r e s u l t i n g i n under-representation of some fam i l i e s . 138 8. WIND-POLLINATED PROGENY NURSERY-STAGE GROWTH 8.1 I n t r o d u c t i o n Some k n o w l e d g e o f t h e p a t t e r n o f v a r i a t i o n e x h i b i t e d b y a s p e c i e s i n t r a i t s w h i c h a r e a d a p t i v e a n d / o r o f c o m m e r c i a l i m p o r t a n c e i s o f v a l u e t o b r e e d e r s i n o r d e r t o d e t e r m i n e a p p r o p r i a t e s e l e c t i o n s t r a t e g i e s f o r i m p r o v e m e n t . P r o v e n a n c e ( o r p o p u l a t i o n ) t e s t s u s i n g b u l k e d s e e d c o l l e c t i o n s h a v e p r o v i d e d t h e m a j o r i t y o f i n f o r m a t i o n on t h e e x t e n t o f v a r i a t i o n i n g r o w t h c h a r a c t e r i s t i c s o f n o r t h e r n c o n i f e r s ( E l - K a s s a b y e t al. 1 9 8 7 a ) . V a r i a t i o n i n h e i g h t g r o w t h f o r s e v e r a l Abies s p e c i e s h a s b e e n e s t i m a t e d ( A r b e z 1969, L e s t e r 1970, H a m r i c k 1976, L e s t e r e t a l . 1976, L i n e s 1978, K l e i n s c h m i t 1986, F u n c k e t al. 1 9 9 0 ) . W i t h l i t t l e e x c e p t i o n , a h i g h p r o p o r t i o n o f t h i s v a r i a t i o n h a s b e e n shown t o r e s i d e w i t h i n p o p u l a t i o n s . The f a m i l y s t r u c t u r e o f t h e m a t e r i a l c o l l e c t e d f o r t h i s t h e s i s e n a b l e s f u r t h e r r e f i n e m e n t o f w i t h i n - p o p u l a t i o n v a r i a t i o n t o a l e v e l w h i c h a c c o u n t s f o r d i f f e r e n c e s among i n d i v i d u a l p a r e n t t r e e s . To t h i s e n d , a w i n d - p o l l i n a t e d p r o g e n y n u r s e r y t r i a l was e s t a b l i s h e d u s i n g o f f s p r i n g f r o m t h e 42 t r e e s e m p l o y e d i n t h e g e r m i n a t i o n t e s t r e p o r t e d i n C h a p t e r 7. 139 8.2 M a t e r i a l s and Methods 8.2.1 Test establishment a n d - c u l t u r e 7An o p e n - p o l l i n a t e d progeny t e s t was p l a n t e d i n p o l y s t y r e n e b l o c k s (Beaver P l a s t i c s L t d . , each with 160 c a v i t i e s , 65ml per c a v i t y ) i n a completely randomized design w i t h f o u r r e p l i c a t i o n s of e i g h t s e e d l i n g s per t r e e . Each e i g h t - s e e d l i n g r e p l i c a t i o n formed a row p l o t i n one of the b l o c k s , with non-test s e e d l i n g s p l a n t e d around the p e r i m e t e r of each block, to reduce edge e f f e c t s . Germinants were o b t a i n e d from the f i r s t count of the germination t e s t d i s c u s s e d i n Chapter 7. The count was delayed u n t i l day 12 of the t e s t t o ensure an adequate supply of germinants from every t r e e . Germination had been observed on day f o u r of the t e s t i n two t r e e s i n p o p u l a t i o n B, one i n H, and t h r e e i n W. By day s i x , some germination had o c c u r r e d i n a l l p o p u l a t i o n s but i n only two t r e e s i n p o p u l a t i o n R. The p l a n t i n g of even-aged germinants became i m p r a c t i c a b l e , and i t was assumed t h a t p l a n t i n g germinants of uniform r a d i c l e l e n g t h ( i d e a l l y , one times the l e n g t h of the seed) would reduce the age d i f f e r e n c e . N e v e r t h e l e s s , germinants of s e v e r a l e a r l y - g e r m i n a t i n g t r e e s had a l l exceeded the i d e a l l e n g t h . The mean age of the p l a n t a t i o n on establishment was c o n s i d e r e d to be one week. Blocks were p l a c e d i n a c o n t r o l l e d growth chamber at the Canadian F o r e s t r y S e r v i c e , P a c i f i c F o r e s t r y Centre, V i c t o r i a , B.C. and exposed to a growing regime of h i g h 140 humidity and 18°C, 14-hour days (light provided by cool-white florescent tubes at 1000 lux at the center of the bench) and 10 hours of darkness at 15°C. At week three, when a l l the seed coats were shed (or removed), incandescent l i g h t s at 200 lux were used to extend the photoperiod to 18 hours. During week eight, the test was moved to Canadian P a c i f i c Forest Products Ltd. (formerly P a c i f i c Forest Products) Saanich Forestry Centre nursery in Saanichton, B.C. and set up alongside other true f i r seedlings being grown operationally. The experimental seedlings were exposed to the same c u l t u r a l regime employed by the nursery for container production of Abies species, including supplemental l i g h t i n g u n t i l week 26, whereupon the t r i a l was exposed to an eight-hour photoperiod (controlled using black-out curtains) and moderate moisture stress to encourage bud formation. The t r i a l was measured weekly during the black-out period, i n an attempt to assess v a r i a t i o n i n bud set, but no buds were detected at week 29 when the t r i a l was harvested. 8.2.2 Measurement and analysis The variables selected for measurement are l i s t e d i n Table 8.1. Cotyledon numbers were based on 40 seedlings per family, as the two edge trees for each row plot came from the same family. Individual seedling heights were taken along the main stem above the cotyledons to the t i p s of the 141 Table 8.1 Variables used i n analysis of wind-pollinated progeny nursery growth t r i a l of s ix populations of P a c i f i c s i l v e r f i r from Vancouver Island, B r i t i s h Columbia. Sampling Level (n) Variables measured (abbreviations i n brackets) Population (POPN) (6) Tree/POPN (7) Replication/tree/POPN (4) Seedling (6-10/replication) l a t i t u d e (LAT), longitude (LONG) elevation (ELEV) mean rachis length (RACHLEN), mean seed weight (SDWT), observed enzyme heterozygosity (ENZHET) mean shoot dry weight (SHWT), mean root dry weight (RTWT) height at 8, 16, 22, 26, 27, 28, and 29 weeks (HT), f i n a l diameter (DIAM), cotyledon number (COTY), growth rate (RATE) uppermost needles (held v e r t i c a l l y ) , measured to the nearest mm. Root c o l l a r diameter of each seedling was determined at harvest using microcalipers. Biomass measurements were obtained aft e r oven-drying each r e p l i c a t i o n separately and c a l c u l a t i n g a mean dry weight per seedling. Maternal tree variables consisted of average cone rachis length, based on measures of seven to ten cones (to the nearest mm) , average seed weight (mean of 10 samples of 100 f i l l e d seeds) and observed enzyme heterozygosity based on 11 l o c i (reported in Chapter 3). Populations were 142 characterized by t h e i r geographic location (latitude, longitude, and elevation). Plots of seedling height values against time showed, on average, a pattern of exponential growth (Y = a • Xb, where Y i s seedling height and X i s time). A logarithmic transformation, In Y = In a + b •In X, was selected to best l i n e a r i z e the data (Steel and Torrie 1980) given that measurement in t e r v a l s were i r r e g u l a r . The slope, b, was calculated i n a simple l i n e a r regression analysis using MIDAS (Fox and Guire 1976) on logarithmically transformed data for each seedling surviving to the end of the test (n = 1219). These i n d i v i d u a l b values were taken to represent seedling growth rates (RATE in Table 8.1). This approach i s s i m i l a r to that of Maze et al. (1989) for analysis of branch growth i n Douglas f i r seedlings. The discussion of variable i n t e r r e l a t i o n s h i p s i s based upon the r e l a t i v e magnitudes of c o e f f i c i e n t s obtained from l i n e a r c o r r e l a t i o n and regression analyses. Simple b i v a r i a t e correlations do not account for interactions of any other variables with the p a i r of i n t e r e s t (Zar 1984), however they serve as a useful s t a r t i n g point for determining r e l a t i v e magnitudes and, thus, strength of r e l a t i o n s h i p s . P a r t i a l c o r r e l a t i o n c o e f f i c i e n t s r e f l e c t the c o r r e l a t i o n between any two variables while holding the values of other i n t e r a c t i n g variables constant. In a 143 multiple c o r r e l a t i o n analysis, the square of the multiple c o r r e l a t i o n c o e f f i c i e n t , or the c o e f f i c i e n t of multiple determination (R 2), i s interpreted as the amount of v a r i a b i l i t y i n one of M variables which i s accounted for by c o r r e l a t i n g i t with M - 1 other variables. In a multiple regression, R2 r e f l e c t s that portion of the t o t a l v a r i a t i o n i n the dependent variable (Y) which can be att r i b u t e d to the l i n e a r r e l a t i o n s h i p with the independent or predictor variables ( X j ' s ) . I f the X i ' s are themselves highly i n t e r c o r r e l a t e d , then conclusions regarding t h e i r s i g n i f i c a n c e may be "spurious" (Zar 1984, p. 338) . Simple and multiple c o r r e l a t i o n analyses assume a normal d i s t r i b u t i o n of values for each v a r i a b l e . The variables employed i n t h i s study (Table 8.1) meet t h i s c r i t e r i o n with the exception of those describing populations (LAT, LONG and ELEV). Although Zar (1984, p. 311) states that t e s t i n g the hypothesis that the parametric c o r r e l a t i o n c o e f f i c i e n t i s zero remains v a l i d when only one of two variables i s normally distributed, i t was decided that multiple regression analysis using LAT, LONG and ELEV as independent variables would more closely respect the structure of the available data. However, r e s u l t s from eithe r multiple regression or correlation are s i m i l a r because the significance test of the p a r t i a l regression c o e f f i c i e n t i s " i d e n t i c a l " to that of the corresponding p a r t i a l c o r r e l a t i o n c o e f f i c i e n t (Zar 1984, p. 340). 144 Because variables were measured at d i f f e r e n t sampling l e v e l s (Table 8.1), data sets were constructed such that the same value of a population- or maternal t r e e - l e v e l variable was paired many times with variables measured on i n d i v i d u a l seedlings, which resulted i n very low c r i t i c a l values of co r r e l a t i o n c o e f f i c i e n t s (r c) , due to the large number of degrees of freedom (df). Therefore, correlations reported i n t h i s chapter are based on family mean values (n = 42 for a l l variables except RACHLEN where n = 38; df = n - 2 for single correlations, n - M for p a r t i a l and multiple correlations involving M vari a b l e s ) , unless stated otherwise. This allows comparison of the magnitude of r or i t s square, the c o e f f i c i e n t of determination (r 2) , a measure of the proportion of v a r i a t i o n i n one variable that can be explained by v a r i a t i o n i n another. ANOVA was used to apportion v a r i a t i o n among hypothesized sources, namely, population (geography), maternal trees (families), r e p l i c a t i o n s and i n d i v i d u a l seedlings (where appropriate). A f u l l y nested random ef f e c t s ANOVA was based on the following model: Y i j k m = \i + p i +  F)ii) + -f^uj) + em(ijk) where Y i j k m i s the measure of the mth seedling i n the k t h r e p l i c a t i o n i n the j t h family i n the i t h population, (1 the o v e r a l l mean, Pi the ef f e c t associated with the i t h population (i = 1,...,6), Fj the e f f e c t of the j t h family i n the i t h population (j = 1,...,7), Rk the e f f e c t associated with the k t h r e p l i c a t i o n i n the j t h family and i t h 145 population (k = 1,...,4) and the residual (seedling) term. The number of seedlings per r e p l i c a t i o n (those surviving u n t i l the end of the test) varied from 6 to 8 (except i n the case of the variable COTY where m = 10) making t h i s design unbalanced. Where components of variance were calculated, c o e f f i c i e n t s were obtained using the method described by Sokal and Rohlf (1981, pp. 3 0 2 - 3 0 5 ) . For variables measured at the l e v e l of r e p l i c a t i o n within maternal tree ( i . e . , SHWT) and measures based on mean values for each maternal tree ( i . e . , SDWT), the appropriate e f f e c t s were removed from the model s p e c i f i e d above. El-Kassaby et al. (1987a) point out that estimates of genetic variance and h e r i t a b i l i t y require common gene frequencies across populations. Where population differences i n seedling growth t r a i t s were not s i g n i f i c a n t i t was considered appropriate to calculate h e r i t a b i l i t y (narrow-sense) as h 2 = 4 a 2 F / P / o~2F/P + O"2R/F/P + °"2e- Except for ANOVA of growth rate, which was conducted using the variance components approach, a l l apportionment of v a r i a t i o n was based on a percentage of the total.sums of squares (%SS, see section 7.2.2). 8.3 Results Seedling losses during the 29 week t r i a l were low , (9.3%) and were primarily as a re s u l t of handling during measurement. The material did not show any evidence of disease or insect problems and shoot growth appeared lush throughout the experiment. The relationships among factors presumed to influence the growth responses of the open-pollinated progeny sampled are summarized i n Figure 8.1. The magnitude of the ef f e c t s of these hypothesized sources of v a r i a t i o n w i l l be determined by describing both c o r r e l a t i o n and variance structures and any changes i n these structures over the course of the experiment. Source e f f e c t s w i l l be discussed using the numbering scheme i n Figure 8.1. (1-2) 1. population ( 1 - 3 ) (1-4) L . 2. maternal parent — (2-3) 3.seed (2-4) ( 3-4 ) Growth (a) pattern (b) form 4. seedling . _ l Figure 8.1 Relationships of the variables examined i n the analysis of wind-pollinated from 42 families of P a c i f i c s i l v e r f i r c o l l e c t e d in six locations on Vancouver Island, B r i t i s h Columbia. 147 8.3.1 Geographic and maternal tree v a r i a t i o n The c o l l i n e a r i t y among geographic variables describing the s i x sampled stands (Table 8.2:1) clos e l y r e f l e c t s the ov e r a l l d i s t r i b u t i o n pattern of the species. As one moves north and west, P a c i f i c s i l v e r f i r i s found at increasingly lower elevations (Schmidt 1957, Green et al. 1984). Multiple regression equations describing the contributions of l a t i t u d e , longitude and elevation of population to enzyme heterozygosity and cone rachis length (also seed and seedling variables) are reported i n Appendix 3. Although the multiple c o r r e l a t i o n c o e f f i c i e n t s (R's) are s i g n i f i c a n t (Table 8.2:11), the standard errors of the p a r t i a l regression c o e f f i c i e n t s ( l i s t e d beneath the appropriate c o e f f i c i e n t s i n Appendix 3) are r e l a t i v e l y large, which also suggests the presence of substantial i n t e r c o r r e l a t i o n among predictor variables (Zar 1984, p. 338). For both enzyme heterozygosity and cone size (represented by rachis length), l a t i t u d e of population exhibits the largest p a r t i a l c o r r e l a t i o n c o e f f i c i e n t among the variables describing geographic location (Table 8.2:11). The p a r t i a l correlations among ENZHET and RACHLEN indicate that both protein v a r i a b i l i t y and size of cones are reduced i n more northerly populations, but increase from east to west. Associations with elevation are not s i g n i f i c a n t . Enzyme heterozygosity and rachis length were themselves poorly i n t e r c o r r e l a t e d (r 2 = 11%) . One-way ANOVA (Table 8.3a) showed that differences among populations accounted for 41% of the v a r i a t i o n i n rachis length. Table 8.3b indicates that the largest cones are produced i n the southernmost locations. 8.3.2 Geographic influences on seed variables Seed weight and cotyledon number had the same pattern of r e l a t i o n s h i p with geography as did maternal tree variables (Table 8.2: 1-3). The strength of the rela t i o n s h i p among geographic variables and seed weight (R2 = 63%) i s greater than with the number of cotyledons (R2 = 47%). This i s also r e f l e c t e d i n population differences i n ANOVA (Table 8.3a) which explain 67% of the v a r i a t i o n i n seed weight and less than 10% i n cotyledon numbers, suggesting that the number of cotyledons shows far greater intra-population v a r i a t i o n than does seed weight. Table 8.3b shows that heavier seeds, associated with larger numbers of cotyledons, tend to come from southern Vancouver Island (exception: population R) . 8.3.3 Maternal tree and seed relationships Table 8.2 (2-3) indicates that rachis length and seed weight are highly correlated, suggesting that large cones are producing heavier seeds, and that to a lesser extent, embryos from heavier seeds produce a larger number of cotyledons, as indicated by the lower c o r r e l a t i o n between rachis length and cotyledon number. Similar trends are Table 8.2 Cor r e l a t i o n c o e f f i c i e n t s for geographic, maternal tree and seed progeny va r i a b l e s . C r i t i c a l value for simple correlations a) among LAT, LONG and ELEV i s 0.811 (df=4); b) with RACHLEN i s 0.320, a l l others r c = 0.304. For p a r t i a l and multiple correlations, r c = 0.329 for RACHLEN, otherwise r c = 0.320 I. Simple c o r r e l a t i o n s among population and maternal tree variables: LAT ELEV RACHLEN LONG 0.987 -0.933 ENZHET 0.329 ELEV -0.927 II. P a r t i a l and multiple c o r r e l a t i o n c o e f f i c i e n t s for population, maternal tree and seed variables (1-2, 1-3): LAT LONG ELEV R (multiple) ENZHET -0 .366 0 .312 0 . 038NS 0 .472 RACHLEN -0 . 649 0 .57 6 0 .202NS 0 .802 SDWT -0 .507 0 .390 0 . 1 0 5 N S 0 .792 COTY -0 .22 6 N S 0 . 138 N S 0 . 150 N S 0 . 688 II I . Simple c o r r e l a t i o n s among maternal tree and seed variables (2-3): ENZHET RACHLEN COTY SDWT 0.574 0.798 SDWT 0.747 COTY 0.304 0. 622 Table 8.3a Variance structures for mean cone rachis length (RACHLEN) , mean seed weight (SDWT) and cotyledon number (COTY). Differences among means determined to be significant (*, P < 0.05) based on F test (not presented). I. RACHLEN: SOURCE df Sum of Squares %SS II. SDWT: POPULATION 5 7358.64 41.1' ERROR 33 10533.50 58.9 TOTAL 38 17892.14 100.0 SOURCE df Sum of Squares %SS POPULATION 5 ERROR 36 TOTAL 41 40.738 67.0* 20.047 33.0 60.785 100.0 III. COTY: SOURCE df %SS POPN 5 9.1* T R E E / P O P N 36 8.5* R E P / T R E E / P O P N 126 8.0* ERROR 1495 74.4 TOTAL 1662 100.0 151 Table 8.3b Population average values of rachis length, 1000-seed weight and cotyledon number, with associated 95% confidence i n t e r v a l s . POPULATION RACHLEN SDWT COTY R 88.7 ± 2.0 3.08 ± 0.22 4.78 + 0.17 H 68.1 ± 3.4 2.33 ± 0.13 4 . 69 + 0.16 B 79.5 ± 1.9 2. 95 ± 0.15 4 .82 ± 0.17 C 76.2 ± 2.1 2.77 ± 0.21 4 . 87 ± 0.18 F 95.3 ± 3.1 4 . 44 + 0.16 5.12 ± 0.18 W 112.5 ± 3.1 5. 10 ± 0.33 5.24 ± 0.17 evident among population means (Table 8.3b). Enzyme heterozygosity showed a po s i t i v e relationship with seed weight however, the correlation with cotyledon number i s just s i g n i f i c a n t (r = r c = 0.304). Seed weight and cotyledon number are themselves p o s i t i v e l y correlated (r 2 = 56%). A h i e r a r c h i c a l ANOVA (Table 8.3a) of cotyledon number showed s i g n i f i c a n t (P < 0.05) family differences for t h i s t r a i t however t h i s source accounts for less than 10% of the t o t a l variance. The r e p l i c a t i o n e f f e c t was also s i g n i f i c a n t (8% of the t o t a l variance), which w i l l be considered further i n section 8.4. 8.3.4 Relationships among seeds and seedlings The average seed weight (per tree) among the 42 trees sampled in t h i s test varied from 15.4 to 70.9 mg, a 360% difference. Cotyledons were less variable in number, 152 r a n g i n g from t h r e e t o seven. The r e l a t i o n s h i p s among seed v a r i a b l e s and v a r i a b l e s d e s c r i b i n g t h e growth o f s e e d l i n g s a r e summarized i n Table 8.4 ( 3 - 4 a ) . The c o n t r i b u t i o n o f seed weight and c o t y l e d o n number t o s e e d l i n g h e i g h t a t e i g h t weeks i s h i g h but drops d r a s t i c a l l y and becomes n e g a t i v e , but not s i g n i f i c a n t l y so, a t 29 weeks growth i n h e i g h t . Average growth r a t e s have a n e g a t i v e r e l a t i o n s h i p w i t h seed w e i g h t and c o t y l e d o n number, i n d i c a t i n g t h a t s e e d l i n g s d e r i v e d from h e a v i e r ( l a r g e r ) seeds d i d not grow as r a p i d l y as s e e d l i n g s produced from l i g h t e r ( s m a l l e r ) seeds. The n e g a t i v e c o r r e l a t i o n between i n d i v i d u a l s e e d l i n g h e i g h t s a t e i g h t weeks and c o r r e s p o n d i n g growth r a t e s (r = -0.658) s u g g e s t s t h a t s h o r t e r s e e d l i n g s i n t h e e a r l y s t a g e s o f development p o s s e s s e d h i g h e r o v e r a l l r a t e s o f growth. T h i s r e l a t i o n s h i p was r e v e r s e d by th e end o f t h e t e s t p e r i o d , as i n d i c a t e d by th e h i g h l y p o s i t i v e c o r r e l a t i o n between h e i g h t a t 29 weeks and RATE. How growth r a t e was a p p o r t i o n e d among d i f f e r e n t s o u r c e s was e s t i m a t e d by ANOVA (Table 8 .4) . The growth r a t e p a t t e r n e x h i b i t e d by P a c i f i c s i l v e r f i r s e e d l i n g s i n t h i s t e s t appears t o have v e r y l i t t l e g e o g r a p h i c i n f l u e n c e (the p o p u l a t i o n term i s n o n - s i g n i f i c a n t , P < 0.05). T r e e - t o - t r e e d i f f e r e n c e s w i t h i n p o p u l a t i o n a r e s i g n i f i c a n t but account f o r l e s s v a r i a t i o n than do d i f f e r e n c e s among r e p l i c a t i o n s w i t h a s i n g l e t r e e . And by f a r t h e g r e a t e s t v a r i a t i o n i n Table 8.4 Correlation and variance structures for variables describing seeds and seedling growth. ( N S = non-significant; P < 0.05) I. C o r r e l a t i o n among seed and seedling growth variables (3-4a): HT8 RATE HT8 HT29 RATE HT29 RATE -0 . 049 N S -0.658 0. 677 SDWT COTY 0.546 0.511 -0.108Ns -0 . 117NS -0.426 -0.302 II. Nested ANOVA for growth rate of seedlings (RATE): Source of Vari a t i o n df SS Among populations (P) 5 2.19 Among families (F/P) 36 12.12 Among r e p l i c a t i o n (R/F/P) Among seedlings (error) TOTAL 126 14.60 %SS® %vc® 3 . 1 N S 17.3 20.8 1051 41.21 58.8 1218 . 70.12 100.0 0. 8 N S 13.7 68. 3 100. 0 Components of Variance® cy + k«o2 R / F / p + k 5G2 F / p + k 6 o 2 P Ge2 + k20-2R/F/p + k 3C2 F / p 17.2 Oc2 + k ^ F / p c y ® V a r i a t i o n accounted for by each source expressed as a percentage of the t o t a l sum of squares ® V a r i a t i o n expressed as a percentage of the t o t a l variance ( a 2 p + a 2 F / p + a 2 R/F/P + a2 ,) ® kx = 7.338, k 2 = 6.955, kA = 7.346; k3 = 29.010, k5 = 29.990, k 6 = 203.230 calculated according to the method of Sokal and Rohlf (1981) 154 growth rate resides among in d i v i d u a l seedlings. When separate ANOVA*s of growth rate were conducted on each population, the amount of v a r i a t i o n accounted for by i n t e r -tree differences was not the same within each population. Differences among trees i n population R accounted for 3.8% of the v a r i a t i o n , 8.6 i n population H, 7.6 i n B, 25.1 i n C, 20.0 i n F and 21.9 i n W . Tree-to-tree e f f e c t s on seedling growth rate appear to be'stronger in more southerly populations than those sampled from northern Vancouver Island. Table 8.5 l i s t s some population-level s t a t i s t i c s regarding height growth. The ov e r a l l minimum seedling height at eight weeks was 13 mm and the maximum, 60 mm. The range i n seedling heights for population H was 13-49 mm and for W , 21-60 mm. At 29 weeks, the minimum seedling height was 65 mm and the maximum was 229 mm. The range of seedling heights as 29 weeks was 74-228 mm for population H, and for W , 68-218 mm. Population W had the heaviest seeds (5.08 ± 0.15 mg, mean ± 95% CI), highest number of cotyledons (5.25 ± 0.09), and highest mean seedling height at eight weeks but not at 29 weeks, and average growth rate was the slowest of a l l six populations i n the te s t . Population H, on the other hand, had the l i g h t e s t seeds (2.32 ± 0.06 mg) , lowest cotyledon number (4.67 ± 0.09), lowest average seedling height at eight weeks, but showed the second highest height 155 at the end of the test and exhibited the fastest average rate of growth of a l l populations. Examination of the r e l a t i v e v a r i a b i l i t y of the six populations i n terms of the three growth response variables (Table 8.5) shows that, whether the variance or range of values i s used to describe dispersion, some populations show dramatic changes i n v a r i a b i l i t y over time. At eight weeks growth population R i s the least variable (and second t a l l e s t ) i n height whereas at 29 weeks i t exhibits the greatest v a r i a t i o n . Population F, on the other hand, behaves i n exactly the opposite manner. Growth rate i s also least variable i n population F while varying most i n population C. Table 8.6 l i s t s the r e l a t i v e contributions (based on %SS) of populations and trees within each population to the t o t a l v a r i a t i o n i n seedling height growth over the duration of the t r i a l . Contributions, unless noted, are a l l s i g n i f i c a n t (P < 0.05) but of much greater inte r e s t i s the magnitude and s t a b i l i t y of the v a r i a t i o n ascribed to these terms. V a r i a t i o n in cumulative seedling height associated with population differences i s most pronounced at eight weeks and d e c l i n i n g to v i r t u a l l y zero at 29 weeks. The same pattern i s seen i n height increment, except between 26 and 27 weeks when population contribution r i s e s . This i s 156 T a b l e 8.5 Summary s t a t i s t i c s o f s e e d l i n g h e i g h t s a n d r a t e s o f g r o w t h b a s e d on p o p u l a t i o n means, f o l l o w e d b y 95% c o n f i d e n c e i n t e r v a l s . P o p u l a t i o n A v e r a g e HT a A v e r a g e HT 2 9 A v e r a g e RATE R 37.29 ± 0.74 149.05 ± 4.47 1.038 ±0.033 H 33.39 ± 0.84 148.07 ± 4.12 1.127 ± 0.033 B 35.44 ± 0.83 143.37 ± 4.50 1.055 ± 0.033 C 36.55 ± 0.83 146.20 ± 4.43 1.045 ± 0.034 F 36.98 ± 0.88 143.02 ± 3.96 1.025 ± 0.030 W 39.77 ± 0.95 145.79 ± 4.11 0.987 ± 0.031 INDICES OF DISPERSION f o r 6 p o p u l a t i o n s : RANK (RANGE) HT 8 HT 2 9 RATE 1 H R C 2 F C B 3 B B W 4 W W R 5 C H H 6 R F F RANK (VARIANCE) HT 8 HT 2 9 RATE 1 W B C 2 F R R 3 B C B 4 H W H 5 C H W 6 R F F 157 coincident with the i n i t i a t i o n of the "stress" culture regime i n the nursery. The differences among trees within each population account for a range of v a r i a t i o n and appear to fluctuate somewhat with time. A l l family differences do decline over time except i n population B and C where the contribution remains more or less stable. Tree-to-tree differences are of more importance in incremental height growth i n the f i r s t h a l f of the test than i n the l a t t e r h a l f . The relationships among seeds and seedlings in terms of form or biomass are l i s t e d i n Table 8.7 (3-4b). Variables describing f i n a l seedling form are intercorrelated, the highest c o r r e l a t i o n between shoot weight and r o o t - c o l l a r diameter. Over a l l , seed weight and cotyledon number account for si m i l a r amounts of v a r i a t i o n in shoot and root weight and in diameter at the r o o t - c o l l a r . ANOVA of seedling biomass and diameter measures are also given i n Table 8.7. With respect to the a l l o c a t i o n of biomass (ratio of shoot to root weight) large family differences are apparent, but populations account for much less v a r i a t i o n . This p a r a l l e l s the pattern for top growth alone, whereas root weight varies among population to a greater degree. Individual seedling diameter varies the least among populations, and family differences are much less (10% compared to an average of 30%) than for the other Table 8.6 Source contributions presented as %SS from ANOVA of seedling heights over time (cumulative and incremental) among 6 populations of P a c i f i c s i l v e r f i r and among trees within each population. CUMULATIVE SEEDLING HEIGHT S O U R C E df 8 wk 16 wk 22 wk 26 wk 27 wk 28 wk 29 wk P O P N (all) 5 10.0 4.4 2.5 0 . I N S 0. 0 N S 0.0NS 0.1NS T R E E S (H) 6 22.2 10 . 9a 9.2* 7 . 7* 9.0a 9.0a 7.5a (R) 6 25.8 16. 0a • 15.3 11 . 1 11.2 10.7 11.2 (B) 6 11.0 25.7 21.7* 21 . 6 21.2 21.3 21.6 (C) 6 18.3 11. l a 22.7* 20 . 9 21.0 20.8 18.7 (F) 6 27 . 3 20.3 17.4 9 .9 9.9 9.7 8.5 (W) 6 17.3 9. 9a 10.3 11 . 0 12.7 12.0 13.4 SEEDLING HEIGHT INCREMENTS SOURCE df 8-16 wk 16-22 wk 22-26 wk 26- 27 wk 27-28 wk 28-29 ' POPN (all) 5 3.7 0 .5 N S 1.8 3. 1 0. 6 N S a 0. 6 N S a TREES (H) 6 9.5a 8 . 7a 0. 6NS 0. 7 NS 0.2 10. 7* (RJ 6 24 . 9 9.7 3.3 N S a 5. 6 21. l a 3.2NS* (B) 6 12.5 10.2 10 . 9 2 . 4NSa O . O N S 6.9 (C) 6 18.5a 27 . 2 a 8.6 3. 8 "0. 0 N S 10.8 a (F) 6 31 .2 a 8.2 0.0NS 0. QMS 1 . 7NSa 5.9 (W) 6 14 . 8a 6.8 7.9 9. 5 4 . 3a 16.7 N S Contribution i s not s i g n i f i c a n t at P < 0.05; a l l remaining values are considered s i g n i f i c a n t using approximate F t e s t s . a Homogeneity of variance not achieved in the ANOVA. variables describing f i n a l seedling form. The majority of v a r i a t i o n i n diameter i s also within families and r e p l i c a t i o n s contribute s i g n i f i c a n t l y to the intra-family variance. Further relationships among populations and maternal trees and seedling form at 29 weeks development are l i s t e d i n Table 8.8 and Appendix 3. Geographic influences on seedling form are n e g l i g i b l e except for root weight (1-4). P a r t i a l c o r r e l a t i o n reveals that elevation i s associated with v a r i a t i o n in root weight, suggesting that at a given l a t i t u d e , root weight of seedlings increases with a l t i t u d e . This i s not borne out by population ranking of mean root weights (Table 8.9), where overlap of confidence i n t e r v a l s obscure any differences among means. Based on population ranking of mean values, smaller seedlings (less mass of either roots and shoots) are tending to come from more northerly populations (Table 8.9). Cone size and enzyme heterozygosity of maternal trees (Table 8.8:2-4) appear to have a p o s i t i v e but non-significant influence on seedling biomass accumulation. Table 8.7 Correlation and variance structures for seed and f i n a l seedling form variables (seedlings harvested at 29 weeks). Variable c o r r e l a t i o n s among seed variables and seedling variables at harvest (3-4b): SHWT R T W T DIAM SHWT DIAM SDWT 0 .304 0. 355 0. 253NS COTY 0 .316 0. 342 .0. 370 R T W T 0. 639 0.525 DIAM 0. 651 I I . ANOVA of seedling biomass and diameter (%SS associated with each SOURCE): SOURCE DF SHWT RTWT SHWT:RTWT POPN TREE/POPN ERROR 5 36 126 5 . 7NS 32 .1* 62 .2 12.2* 27.4* 60.4 7 . 3 N S 34 . 9* 57.8 SOURCE POPN TREE/POPN REP/TREE/POPN ERROR DF DIAM 5 1.4NS 36 9.9* 126 18.3* 1051 70.4 N S Non-significant (P > 0.05) Si g n i f i c a n t (P < 0.05) o Table 8.8 Correlations among population and maternal tree variables with those describing f i n a l seedling form. C r i t i c a l values are the same as those s p e c i f i e d i n Table 8.2. I. Simple Correlations ( 2 - 4 ) : SHWT RTWT DI7AM ENZHET 0 22 6Ns 0 10 9 N S 0.277NS RACHLEN 0 264Ns 0 146NS 0.298NS II. P a r t i a l and Multiple Correlations (1 -4) : LAT LONG ELEV R SHWT -0 .230NS 0. 135NS -0 . 110 N S 0 .358NS RTWT -0 .058NS -0. 164NS -0 .372 0 .538 DIAM -0 .244NS -0. 073NS -0 .244NS 0 .361NS Table 8 . 9 Ranking of population mean values for variables describing seedling biomass accumulation at 29 weeks (± 95% CI) . I. Rank of population mean SHWT (±95% CI) and RTWT(±95% CI) : 1 - W 1. 970 ± 0 . 097 1 -- B 1 . 0 6 9 ± 0 . 0 5 1 2 - F 1 . 880 ± 0 . 0 9 6 2 -- W 1 .061 ± 0 . 0 3 6 3 - B 1 . 831 ± 0 . 1 0 6 3 -- F 1. 057 ± 0 . 0 5 3 4 - H 1 .827 ± 0 . 1 1 3 4 -- H 0. 997 ± 0 . 0 3 9 5 - R 1 .814 ± 0 . 0 9 3 5 -- C 0. 988 ± 0 . 0 3 1 6 - c 1 . 765 ± 0 . 0 8 0 6 -- R 0. 967 ± 0 . 0 2 8 of population mean SHWT:RTWT rat i o (±95% CI) and DIAM (±95% CI): 1 -- R 1 . 873 ± 0 . 0 7 6 1 -- W 2 . 3 2 7 ± 0 . 0 3 5 2 -- W 1 .857 ± 0 . 0 7 3 2 -- B 2 . 2 9 2 + 0 . 0 3 6 3 -- H 1 .827 ± 0 . 0 7 7 3 -- F 2 . 2 9 0 ± 0 . 0 3 7 4 -- C 1 .787 ± 0 . 0 5 4 4 -- H 2 . 2 8 0 ± 0 . 0 3 8 5 -- F 1 . 782 ± 0 . 0 5 7 5 -- C 2 . 247 + 0 . 0 3 6 6 -- B 1 . 7 1 9 ± 0 . 0 7 7 6 -- R 2 . 2 3 3 ± 0 . 034 163 8.4 Discussion The very high i n t e r c o r r e l a t i o n among geographic variables representing the six populations i n t h i s study i s an a r t i f a c t of sampling. However, l a t i t u d e appeared to be the most important predictor among the three available geographic variables with respect to v a r i a t i o n i n enzyme heterozygosity, cone size and seed weight, whereas only elevation contributed s i g n i f i c a n t l y to root growth v a r i a t i o n i n harvested seedlings. Acknowledging the r e s t r i c t e d sample range of t h i s study and imprecision inherent i n the variables describing location of populations, i t remains that the structure of v a r i a t i o n i n cone, seed and seedling t r a i t s i s not consistent. In p a r t i c u l a r , v a r i a t i o n among populations i s marked i n some c h a r a c t e r i s t i c s and n e g l i g i b l e i n others. Cone (rachis) length showed a high degree of geographic v a r i a t i o n i n P a c i f i c s i l v e r f i r , with populations accounting for over 40 percent of the t o t a l variance. This contrasts with v a r i a t i o n i n cone length i n balsam f i r where geographic location accounted for less than one-third of the t o t a l v a r i a t i o n (Lester 1968). Rachis length and cotyledon number show l i m i t e d covariation with enzyme heterozygosity 164 (r 2 = 11% and 9% respectively). The rel a t i o n s h i p between 1000-seed weight and heterozygosity i s stronger (r 2 = 33%) but a larger sample of both enzyme l o c i and in d i v i d u a l s i s required to i n f e r any relationship between cone and seed size as any other polygenic t r a i t and protein v a r i a b i l i t y (Chakraborty 1981, Simon and Archie 1985). That cotyledon number per se appears to be a more variable c h a r a c t e r i s t i c i n P a c i f i c s i l v e r f i r than does the weight of f i l l e d seeds was not a surprising r e s u l t . Sorensen and Franklin (1977) found the residual variance for seed weight i n noble f i r to be 9% compared to 49% for cotyledon number. Helium (1968) also found the number of cotyledons to be more variable than seed weight i t s e l f i n populations of white spruce and observed that cotyledon number varied to the same extent over trees as among populations, which he presumed indicated that cotyledon number i s being determined by a large number of pollen parents. Helium (1968) also speculated that d i f f e r e n t parental contribution may be one of the reasons for the ov e r a l l low co r r e l a t i o n between seed weight and cotyledon number i n white spruce. Sorensen and Franklin (1977) also found poor c o r r e l a t i o n between seed weight and number of cotyledons but observed that cotyledon number exhibited less temporal v a r i a t i o n than did seed weight in noble f i r , which led them to i n f e r that cotyledon number i s determined independently of seed weight. 165 It i s known that seed weight i s influenced by s i t e environmental variables, c o l l e c t i o n year, timing of c o l l e c t i o n and storage conditions (Baldwin 1942, Bjornstad 1981) . Because at least 80% of t o t a l seed weight i s composed of tissue derived from maternal parent (Buchholz 1946, Perry 1976), then seed weight correlations with other variables w i l l contain a genetic component that i s ine x t r i c a b l e from macro- and micro-environmental factors influencing seed development. Seed size i s viewed as being one of the least p l a s t i c reproductive t r a i t s (Palmblad 1968, Harper et al. 1970), however Fins and Libby (1982) and Salazar (1986) are correct in t h e i r caution that genotype-environment int e r a c t i o n i s confounded in any progeny test where seeds are formed at di f f e r e n t locations. This was recognized by Sorensen and Campbell (1985) in a seed-seedling size study of Douglas-fir, where seeds of d i f f e r i n g mean weights were produced on the same tree (and p o s i t i o n within the crown) by manipulating the micro-environment of the developing cones. Their results showed that a 10% increase i n seed weight was associated with a 9% increase i n f i r s t year e p i c o t y l length. They also point out that seed weight differences i n seedlots of most conifers are l i k e l y to be greater than 10%, so that the poten t i a l for growth differences r e s u l t i n g from seed size differences i s present in wind-pollinated seed c o l l e c t i o n s . Seed weight differences among trees sampled in t h i s thesis was as large as 360%. Sweet and Wareing (1966) report differences i n 166 seed weight of 250% from a single radiata pine tree. No doubt the range would have been even larger i n P a c i f i c s i l v e r f i r had seed weight measures been based on uncleaned seed. Seed weight was seen to be p o s i t i v e l y associated with early height growth i n P a c i f i c s i l v e r f i r , but by the end of the 29 week t r i a l , the association had become negative and non-significant. Cannell et al. (1978) found that seed weight of l o b l o l l y pine no longer correlated with seedling height at two months a f t e r sowing. Mann (1979) found seed weights and seedling heights at 60 and 120 days for several Pinus species to be largely unrelated. A s i m i l a r r e s u l t was reported by Lavender (1958) for Douglas-fir. Lines (1978) found no seed weight influence on f i r s t year seedling height of grand f i r grown in B r i t a i n . However, Perry (1976) maintained that seed weight or "maternal" e f f e c t s i n some conifers can account for a simple proportion of v a r i a t i o n i n seedling size and that t h i s influence may extend for a number of years. Arbez (1969) found that seed weight influenced height growth in European s i l v e r f i r at age four, but that height was independent of cotyledon number at four years. K h a l i l (1981) also detected a s i g n i f i c a n t influence of seed weight on four-year height in white spruce. Faster growth rates among seedlings derived from larger seeds i s often attributed to higher nutrient reserves in these seeds (Mikola 1980) . However, t h i s study showed 167 that heavier seeds of P a c i f i c s i l v e r f i r displayed slower rates of growth than lighter-weight seeds. This i s s i m i l a r to the findings of Sweet and VJareing (1966) in radiata pine and European larch, where a slower growth rate among seedlings derived from heavier seeds was att r i b u t e d to a r e s t r i c t i o n i n growth caused by container s i z e . Sweet and Wareing (1966) also contend that any genetic v a r i a t i o n i n growth rate may be obscured by seed size differences. This would have no doubt occurred in the present study had population (provenance) based bulk seed c o l l e c t i o n s been made. G r i f f i n and Ching (1977) considered seed weight, cotyledon number, and hypocotyl length to be i n t e r - r e l a t e d expressions of embryo size in Douglas-fir, and advocated the measurement of height growth be r e s t r i c t e d to elongation above the cotyledons (length of epicotyl) to reduce the influence of seed si z e . This procedure was adopted i n the present study. Seedling growth in height was p o s i t i v e l y influenced by seed size (r = 0.546) at eight weeks but was negatively correlated with seed size at 29 weeks (r = -0.108 N S). Also, overa l l family influences on seedling height increments declined over the duration of the test (Table 8.6). Nonetheless, a s i g n i f i c a n t r e l a t i o n s h i p between seed size and average growth rate of seedlings (r = -0.426) remains. This result has implications for the 168 in t e r p r e t a t i o n of h e r i t a b i l i t y , which w i l l be discussed l a t e r i n t h i s section. The f i t n e s s , as measured by reproductive success, of trees with a cone-bearing habit such as P a c i f i c s i l v e r f i r , i s l a r g e l y determined by the attainment of at least codominant status within the stand. Suppressed in d i v i d u a l s of P a c i f i c s i l v e r f i r are not known to produce cones (J. Maze, UBC Department of Botany, pers. comm. June 1990). Growth i n height i s considered an important component of fi t n e s s i n Douglas-fir (King et al. 1988). Cannell et al. (1978) found seedling growth rates to r e f l e c t future y i e l d s i n l o b l o l l y pine. Because of the poten t i a l p r e d i c t i v e use of seedling growth rate, i t s relationship to o v e r a l l f i t n e s s (Bush et al. 1987) and because population differences i n growth rate appeared n e g l i g i b l e (see ANOVA i n Table 8.4) the h e r i t a b i l i t y of growth rate for t h i s sample of P a c i f i c s i l v e r f i r was calculated as follows: h 2 = 4 o 2 F / p / (o-2F /P + < j 2 R / F / p + <5\) , from appropriate variance components l i s t e d i n Table 8.4. H e r i t a b i l i t y of growth rate was estimated as 0.55. However, some inbreeding was detected i n four of the populations used in the t r i a l (Chapter 5). Thus, the assumption that a l l o f f s p r i n g from one maternal tree are related as h a l f -s i b l i n g s i s unl i k e l y to be true. Further, given the high shade tolerance of P a c i f i c s i l v e r f i r , there i s also the 169 p o s s i b i l i t y t h a t m a t e r n a l p a r e n t s may t h e m s e l v e s be r e l a t e d , d e s p i t e s a m p l i n g a t t e m p t s t o reduce t h a t l i k e l i h o o d ( L i n e s 1967, F l e t c h e r and B a r n e r 1978). Thus, 0.55 i s no doubt an o v e r - e s t i m a t e o f t h e h e r i t a b i l i t y o f s e e d l i n g growth r a t e , a l t h o u g h t h e d a t a do s u p p o r t t h e h y p o t h e s i s t h a t growth r a t e i s a d a p t i v e , as s m a l l e r seeds, coming from more n o r t h e r l y p o p u l a t i o n s , e x h i b i t f a s t e r r a t e s o f growth, w h i c h would be f a v o r e d i n s h o r t e r growing seasons. G i v e n t h a t g e r m i n a t i o n i n P a c i f i c s i l v e r f i r i s o f t e n low, t h e common n u r s e r y p r a c t i c e o f m u l t i p l e s e e d i n g and subsequent t h i n n i n g o f c o n t a i n e r s t o c k may i n a d v e r t e n t l y r e s u l t i n t h e s e l e c t i o n o f mal a d a p t e d s e e d l i n g s . Maze e t al. (1989) found t h a t o f f o u r D o u g l a s - f i r f a m i l i e s s t u d i e d i n t e n s i v e l y f o r l a t e r a l shoot growth r a t e , t h e f a m i l y w i t h t h e h i g h e s t mean growth r a t e a l s o had t h e l o w e s t a m o n g - i n d i v i d u a l v a r i a t i o n i n growth r a t e . I n P a c i f i c s i l v e r f i r , v a r i a n c e s t r u c t u r e was not t h e same f o r growth r a t e s among f a m i l i e s w i t h i n each p o p u l a t i o n , when a n a l y z e d s e p a r a t e l y ( S e c t i o n 8.3.3), and f a s t e r growth r a t e s were not a s s o c i a t e d w i t h l o w e r v a r i a b i l i t y a t any l e v e l . I n f a c t , t h e o p p o s i t e appears t o be t r u e , g i v e n t h e p o s i t i o n o f p o p u l a t i o n F i n r a n k i n g o f growth r a t e and v a r i a b i l i t y i n growth r a t e (Table 8.5). A l t h o u g h t h e t r i a l was t e r m i n a t e d p r i o r t o t h e f o r m a t i o n o f dormant buds, t h e s e e d l i n g form v a r i a b l e s d i s p l a y e d some a s s o c i a t i o n w i t h geography which s u g g e s t s 170 growth i n terms of biomass i s also adaptive in P a c i f i c s i l v e r f i r (Table 8.7:11 and Table 8.8). Sorensen (1983) suggests smaller top growth and smaller shoot:root biomass r a t i o s are associated with s i t e s having shorter growing seasons i n Douglas-fir. Ranks of populations for mean shoot weight and mean shoot:root r a t i o do not f i t t h i s trend i n P a c i f i c s i l v e r f i r . The highest r a t i o of shoot:root production i s obtained i n one of the most northerly d i s t r i b u t e d populations (R), but given the large v a r i a b i l i t y associated with these mean values, biomass measures are not d r a s t i c a l l y d i f f e r e n t across the entire sampling region. Within-family v a r i a b i l i t y in early seedling growth i s very high i n t h i s sample of P a c i f i c s i l v e r f i r . This may be best i l l u s t r a t e d by the magnitude of the r e p l i c a t i o n e f f e c t s , found to be s i g n i f i c a n t i n a l l models where the term was testable, including cotyledon number. There are at least three possible reasons for t h i s . When the t r i a l was planted, the four r e p l i c a t i o n s per family were randomized throughout 12 styrofoam blocks, and every e f f o r t was made to minimize bias by planting rows as encountered i n the design layout, rather than by tree, so that i t was unl i k e l y that the fourth r e p l i c a t i o n was always planted with the least desirable germinants. However, "planter's choice" cannot be completely ruled out as a source of error. The 12 styroblocks were re-randomized six times during the 29-week test to reduce any systematic e f f e c t of position with 171 respect to i r r i g a t i o n and f e r t i l i z e r regimes, but t h i s may not have been s u f f i c i e n t , in that i n t e r v a l s were not equally spaced throughout the growing season. The r e p l i c a t i o n e f f e c t may also be s t a t i s t i c a l , due to stochastic e f f e c t s r e s u l t i n g from the small number of seedlings within each r e p l i c a t i o n . Perry (1976) proposed that family-row p l o t s are l i k e l y to themselves create competitive e f f e c t s a r i s i n g from seed s i z e v a r i a t i o n . Yeh and Rasmussen (1985) discussed the a l l o c a t i o n of seedlings to open-pollinated progeny tests and the potential l a b i l i t y of family variances. In t h i s study, family variances for growth rates and consequent h e r i t a b i l i t y estimates, are l i k e l y biased upwards to an unknown degree by environmental influences on seed s i z e . As well the detection of inbreeding in some populations and potential genetic relatedness among maternal trees themselves suggests that h 2 estimates are i n f l a t e d . Cannell et a l . ' s (1978) study of l o b l o l l y pine revealed that family differences i n growth rate over one season accounted for 30 percent of the t o t a l variance, compared to 14 percent in P a c i f i c s i l v e r f i r . The inaccuracy inherent i n h 2 and the lower contribution of families to t o t a l v a r i a t i o n i n growth rate observed in P a c i f i c s i l v e r f i r suggests that i t may be unwise to use the h e r i t a b i l i t y of height growth based upon among- to within-family variance as a predictor of future growth performance in P a c i f i c s i l v e r f i r . 172 8.5 Conclusions (a) Populations account for a large proportion of the t o t a l variance of cone size (estimated by rachis length) and seed size (estimated by 1000-seed weight). (b) Seed weight and cotyledon number decrease with l a t i t u d e . Geographic influence on variance of seedling growth c h a r a c t e r i s t i c s i s much less evident. (c) Differences among open-pollinated families are greatest for seedling biomass variables and least for elongation (height) variables. (d) The nature of va r i a t i o n i n open-pollinated progeny growth t r a i t s , the i n t r a c t a b i l i t y of environmental influences on seed size, and the l i k e l i h o o d of inbreeding and genetic c o r r e l a t i o n among parent trees places some r e s t r i c t i o n s on the usefulness of h e r i t a b i l i t y of height growth rate among 0-P seedlings of P a c i f i c s i l v e r f i r . 9. SUMMARY AND CONCLUSIONS Variat i o n patterns in several cone, seed and seedling c h a r a c t e r i s t i c s were studied i n P a c i f i c s i l v e r f i r sampled on a single-tree basis at eight locations on Vancouver Island, B.C. It was found that: (a) of 13 enzyme l o c i that could be r e l i a b l y scored, six exhibited no detectable v a r i a t i o n while seven appeared to each possess at least two isozyme variants. The mode of inheritance for these polymorphic l o c i conformed to Mendelian expectations, although AAT-2 displayed marked segregation d i s t o r t i o n in 13 of 14 trees i n which v a r i a t i o n at t h i s locus was observed; (b) at the sampling i n t e n s i t y available to t h i s study, no evidence of linkage between pairs of enzymes was detected. Linkage between AAT and PGI was anticipated, given that i t has been detected in balsam f i r (Neale and Adams 1981) and several other conifers (Guries et a i . 1978, El-Kassaby et al. 1982b, Boyle and Morgenstern 1985, Cheliak and P i t e l 1985, Harry 1986, Barrett et al. 1987). Much larger sample sizes (of both trees and number of seeds per tree) might reveal the presence of t h i s apparently highly-conserved linkage group i n P a c i f i c s i l v e r f i r ; (c) small but s i g n i f i c a n t levels of inbreeding, based on a multilocus estimate of outcrossing, exist i n f i v e of seven populations of P a c i f i c s i l v e r f i r in the 1983 c o l l e c t i o n 174 year. Evidence for inbreeding other than s e l f i n g , based on differences between mean single locus and multilocus outcross rate estimates, i s weak, but some consanguineous matings may be occurring i n populations of P a c i f i c s i l v e r f i r . Outcrossing rate v a r i a t i o n among populations i s evident and appears p o s i t i v e l y correlated with seed s i z e ; (d) high lev e l s of allozyme v a r i a b i l i t y existed within populations of P a c i f i c s i l v e r f i r (95-98%), based on the d i v e r s i t y measures of Nei (1977) and the d i f f e r e n t i a t i o n index of Gregorius and Roberds (1986), a pattern which i s very common i n wild conifer populations (El-Kassaby 1990) . The extent of among-population d i f f e r e n t i a t i o n d i f f e r e d depending upon which index (Nei's G S T or 8 of Gregorius and Roberds) was employed. Population F was shown to be the most d i f f e r e n t i a t e d genetically by both Nei's (1978) genetic distance measure and 8 (Gregorious and Roberds 1986). Maternal trees appeared more heterozygous then did viable embryos, and southern populations appeared more ge n e t i c a l l y diverse than populations of P a c i f i c s i l v e r f i r sampled from the northern end of Vancouver Island; (e) strong family differences i n germination responses exist i n P a c i f i c s i l v e r f i r , i r r e s p e c t i v e of pregermination treatment. C o l l e c t i o n region had ne g l i g i b l e association with germination performance except in the case of germination value, where i t i s evident that more southerly populations germinate more rapidly without s t r a t i f i c a t i o n 175 than populations at higher l a t i t u d e s . Response to the p a r t i c u l a r s t r a t i f i c a t i o n conditions employed i n the germination test appeared to have a substantial population component, which was manifest as a d i f f e r e n t i a l a b i l i t y to withstand fungal i n f e c t i o n . These findings have important implications for nursery germination, as inadvertent s e l e c t i o n for more rapid-germinating and for mold-resistant f a m i l i e s could reduce the genetic base of the resultant planting stock; (f) a sizable c l i n a l trend exists for seed weight i n P a c i f i c s i l v e r f i r , with lighter-weight seeds found i n more northerly populations. Cotyledon number also decreases with l a t i t u d e , but population differences are much less for t h i s variable than for seed weight (9% vs 67%). Population influences on seedling height growth rapidly decline over a period of 29 weeks while the family component remains reasonably stable i n four of six populations. The magnitude of v a r i a t i o n among families i s greater for seedling biomass variables than for variables describing elongation. Seedling growth rate appears highly heritable, although the large within-family component of variance m i l i t a t e s against using early seedling height growth rate as a basis for s e l e c t i o n . Larger seeds were found to produce seedlings with slower rates of growth, a result which may have some implication for nursery practice where there i s deliberate or inadvertent selection for seed si z e . 176 10- LITERATURE CITED Ackerman, R.F. and J. R. Gorman. 1969. E f f e c t of seed weight on the size of lodgepole pine and white spruce container-planting stock. Pulp and Pap. Mag. Can. Convention issue, 1969:167-169. Adams, W.T. 1983. Applications of isozymes i n tree breeding. In: Isozymes in Plant Genetics and Breeding, Part A. E l s e v i e r S c i . Publishing Company. (S.D. Tanksley and T.J. Orton, eds). pp. 381-400. Adams, W.T. and D.S. Birkes. 1990. Estimating mating patterns i n forest tree populations. In: International Workshop of Plant Biology, Biochemical Markers in the Population Genetics of Forest Trees. Orvieto, I t a l y . (in press). Adams, W.T. and R. J. Joly. 1980a. Genetics of allozyme variants i n l o b l o l l y pine. J. Hered. 71:33-40. Adams, W. T. and R. J. Joly. 1980b. Linkage relationships among twelve allozyme l o c i i n l o b l o l l y pine. J. Hered. 71:199-202. Adkins, CR. 1983. Effects of selected fungicides, surface s t e r i l a n t s and environmental factors on ' germination of Fraser f i r seed. M.Sc. thesis, North Carolina St. Univ., Raleigh N.C. Allen, G.S. 1941. A standard germination test for Douglas f i r seed. For. Chron. 17:75-78. Allen, G.S. and W. Bientjes. 1954. Studies on coniferous tree seed at the University of B r i t i s h Columbia. For. Chron. 30:183-196. Amen, R.D. 1963. The concept of seed dormancy. Amer. S c i . 51:408-427. Anonymous. 1976a. International rules for seed t e s t i n g . Seed S c i . and Technol. 4:4-180. Anonymous. 197 6b. Progress report on the seed vigor t e s t i n g handbook. Newsl. of the Assoc. of O f f i c . Seed Anal. 50:1-78. Anonymous. 1985. International rules for seed t e s t i n g . Seed S c i . Technol. 13:300-520. 177 Arbez, M. 1969. Etude comparative en pepiniere de quelques provenances francaises de sapin pectine (Abies alba M i l l . ) Premier apercu de l a v a r i a b i l i t y i n f raspecifique et mise au point sur l e "sapin de l'aude." Ann. S c i . forest 26:475-509. ' Archie, J.W. 1985. S t a t i s t i c a l analysis of heterozygosity data: Independent sample comparisons. Evolution 39:623-637. Archie, J.W., C. Simon and A. Martin. 1989. Small sample size does decrease the s t a b i l i t y of dendrograms calculated from allozyme - frequency data. Evolution 43:678-683. Bailey, N.T.J. 1961. Introduction to the mathematical theory of genetic linkage. Oxford Univ. Press, London. Baldwin, H.I. 1942. Forest tree seed. Chronica Botanica Co. Waltham, Mass. Barnett, P.E.and R.E. Farmer, Jr. 1978. A l t i t u d i n a l v a r i a t i o n in germination c h a r a c t e r i s t i c s of yellow-poplar i n the southern Appalachians. Silvae Genet. 27:101-104. Barrett, J.W., P. Knowles and W.M. Cheliak. 1987. Variation i n the PGI/AAT linkage group between populations of black spruce. Can. J. For. Res. 17:756-758. Bewley, J.D. and M. Black. 1978. Physiology and biochemistry of seeds in r e l a t i o n to germination. Vol. 1. Springer-Verlag. B e r l i n . B i r o t , Y. 1972. V a r i a b i l i t y intraspecifique du poids de l a graine chez le Douglas (Pseudotsuga menziesii [Mirb.] Franco). Silvae Genet. 21:230-243. Birot, Y. and C. Christophe. 1983. Genetic structures and expected genetic gains from m u l t i t r a i t s e l e c t i o n i n wild populations of Douglas-fir and Sitka spruce. Silvae Genet. 32:141-150. Bjornstad, A. 1981. Photoperiodical a f t e r - e f f e c t s of parent plant environment in Norway spruce. (Picea abies [L.] Karst.) seedlings. Medd. Nor. Inst. Skogforsk. 36:1-30. Blazich, F.A. and L.E. Hinesley. 1984. Low temperature germination of Fraser f i r seed. Can. J. For. Res. 14:948-949. 178 Bonner, F.T. 1984. Glossary of seed germination terms for tree seed workers. USDA For. Serv. Gen. Tech. Rep. SO-4 9, 4pp. Botstein, D., R.L. White, M. Skolnick and R.W. Davis. 1980. Construction of a genetic linkage map i n man using r e s t r i c t i o n fragment length polymorphisms. Am. J. Hum. Genet. 32:314-331. Bousquet, J., W.M. Cheliak and M. Lalonde. 1987. Allozyme v a r i a b i l i t y in natural populations of green alder (Alnus crispa) i n Quebec. Genome 29:345-352. Box, G.E.P. 1949. A general d i s t r i b u t i o n theory for a class of l i k e l i h o o d c r i t e r i a . Biometrika 36:317-346. Boyle, T.J.B. and E. K. Morgenstern. 1985. Inheritance and linkage relationships of black spruce i n New Brunswick. Can. J. For. Res. 15:992-996. Boyle, T.J.B. and E.K. Morgenstern. 1986. Estimates of outcrossing rates in six populations of black spruce in central New Brunswick. Silvae Genet. 35:102-106. Bramlett, D.L., T.R. Dell and W.D. Pepper. 1983. Genetic and maternal influences on V i r g i n i a pine seed germination. Silvae Genet. 32:1-2. Brown, A.H.D. 1970. The estimation of Wright's f i x a t i o n index from genotypic frequencies. Genetica 41:399-406. Brown, A.H.D. 1979. Enzyme polymorphisms in plant populations. Theoret. Pop. B i o l . 15:1-42. Brown, A.H.D. and R.W. A l l a r d . 1970. Estimation of the mating systems in open-pollinated maize populations using enzyme polymorphisms. Genetics 66:133-145. Brown, A.H.D., S.C.H. Barrett, and G.F. Moran. 1985. Mating system estimation in forest trees: models, methods and meanings. In: Population Genetics i n Forestry. Lecture Notes in Biomathematics 60 (H.-R. Gregorius, ed.). Springer-Verlag, N.Y. pp. 32-49. Brown, A.H.D., M.W. Feldman and E. Nevo. 1980. Multilocus structure of natural populations of Hordeum spontaneum. Genetics 96:523-536. 179 Brown, A.H.D., A.C. Matheson and K.G. Eldridge. 1975. Estimation of the mating system of Eucalyptus obliqua L'Herit. using allozyme polymorphisms. Aust. J. Bot. 23:931-949. Brown, A.H.D. and G.F. Moran. 1981.' Isozymes and the genetic resources of forest trees. In: Proc. Symp. Isozymes of N.A. For. Trees and For. Insects (M.T. Conkle, ed.) USDA For. Serv. Gen. Tech. Rep. PSW-48. pp. 1-10. Buchholz, J.T. 1946. Volumetric studies of seed, endosperm and embryos i n Pinus ponderosa during embryonic d i f f e r e n t i a t i o n . Bot. Gaz. 108:232-244. Bush, R.M., P.E. Smouse and F.T. Ledig. 1987. The f i t n e s s consequences of multiple-locus heterozygosity: the r e l a t i o n s h i p between heterozygosity and growth rate i n p i t c h pine (Pinus r i g i d a M i l l . ) . Evolution 41:787-798. Campbell, R.K. and S.M. Ritland. 1982. Regulation of seed-germination timing by moist c h i l l i n g in western hemlock. New Phytol. 92:173-182. Cannell, M.G.R., F.E. Bridgwater and M.S. Greenwood. 1978. Seedling growth rates, water stress responses and root-shoot relationships related to eight-year volumes among families of Pinus taeda L. Silvae Genet. 27:237-248. Chakraborty, R. 1981. The d i s t r i b u t i o n of the number of l o c i i n an ind i v i d u a l in natural populations. Genetics 98:461-466. Chalupa, V. and D.J. Durzan. 1972. Unterschiede i n den Ausmassen und i n der biochemischen Zusammensetzung der Samen von Provenienzen Pinus banksiana Lamb. L e s n i c i t v i 45:1089-1106. Cheliak, W.M., and B.P. Dancik, K. Morgan, F.C.H. Yeh, and C. Strobeck. 1985a. Temporal v a r i a t i o n of the mating system i n a natural population of jack pine. Genetics 109:569-584. Cheliak, W.M., K. Morgan, B.P. Dancik, C. Strobeck and F.C.H. Yeh. 1984. Segregation of allozymes i n megagametophytes of viable seed from a natural population of jack pine, Pinus banksiana Lamb. Theor. Appl. Genet. 69:145-151. Cheliak, W.M., K. Morgan. C. Strobeck, F.C.H. Yeh and B.P. Dancik. 1983. Estimation of mating system 180 parameters i n plant populations using the EM algorithm. Theor.Appl. Genet. 65:157-161. Cheliak, W.M., and J.A. P i t e l . 1985. Inheritance and linkage of allozymes i n Larix l a r i c i n a . Silvae Genet. 34:142-148. Cheliak, W.M., J.A. P i t e l and G. Murray. 1985b. Population structure and the mating system of white spruce. Can. J. For. Res. 15:301-308. Clegg, M.T. 1980. Measuring plant mating systems. BioSci. 30:814-818. Cochran, W.G. 1954. Some methods for strengthening the common % 2 t e s t s . Biometrics 10:417-451. Conkle, M.T., P.D. Hodgkiss, L.B. Nunelly and S.C. Hunter. 1982. Starch gel electrophoresis of conifer seeds: a laboratory manual. Gen. Tech. Rep. PSW-64. Pac. S.W. For and Range Exp. Sta. USDA For. Serv. Berkeley, C a l i f . 18pp. Crow, J.F. and M. Kimura. 1970. An introduction to population genetics theory. Harper and Row, N.Y. Curie-Cohen, M. 1982. Estimates of inbreeding i n a natural population: a comparison of sampling properties. Genetics 100:339-358. Czabator, F.J. 1962. Germination value: an index combining speed and completeness of pine seed germination. For. S c i . 8:386-396. Davidson, R.H., D.G.W. Edwards and 0. S z i k l a i . 1984. Treatment and temperature e f f e c t s on the germination of P a c i f i c s i l v e r f i r . Northwest S c i e n t i f i c Assoc. Mtg., Missoula MT, March 23, 1984. 12pp. Denti, D. and D.J. Schoen. 1988. S e l f - f e r t i l i z a t i o n rates i n white sprvce: E f f e c t of pollen and seed production. J. Hered. 79:284-288. Dimock, E.J. 1958. S i l v i c a l c h a r a c t e r i s t i c s of P a c i f i c s i l v e r f i r . USDA for. Serv. Pac. Nor. West For. and Range Exp. Sta. S i l v . Series No. 4. 12pp. Edwards, D.G.W. 1969. Investigations on the delayed germination of noble f i r . Ph.D. thesis, Univ. Washington, Seattle WA. 231pp. 181 Edwards, D.G.W. 1979. An improved a i r seed-sorter for laboratory use. Env. Can., For. Serv. Rep. BC-X-188, 11pp. Edwards, D.G.W. 1980. Maturity and quality of tree seeds -a state-of-the-art review.. Seed S c i . and Technol. 8:625-657. . Edwards, D.G.W. 1981. A new p r e c h i l l i n g method for true f i r seeds. In: Proc. Joint Meeting Intermountain Nurseryman's Association and Western Forest Nursery Association. Boise, Idaho, Aug 12-14, 1980. Intermountain For. and Range Exp. Sta. Gen. Tech. Rep. INT-109:58-66. Edwards, D.G.W. 1982. Collection, processing, t e s t i n g and storage of tree f i r seeds - a review. In: The Biology and Management of True F i r i n the P a c i f i c Northwest. (CD. Oliver and R.M. Kenady, eds.) USDA For. Serv. Pac. Nor. West Range and Exp. Sta. and Univ. of Wash. Seattle, WA. pp. 113-137. Edwards, G. 1962. The germination requirements of Abies species. In: Proc. Int. Seed Test. Assoc. 27:142-180. Edwards, M.D., C.W. Stuber and J.F. Wendal. 1987. Molecular-marker-facilitated investigations of quantitative t r a i t l o c i in maize. I. Numbers, genomic d i s t r i b u t i o n and types of gene action. Genetics 116:113-125. El-Kassaby, Y.A. 1980. Isozyme patterns of a selected Pseudotsuga menziesii (Mirb.) Franco population. Ph.D. thesis. Univ. of B r i t i s h Columbia, Vancouver, B.C. 137pp. El-Kassaby, Y.A. 1981. Genetic interpretation of malate dehydrogenase isozymes in some conifer species. J. Hered. 72:451-452. El-Kassaby, Y.A. 1990. Genetic v a r i a t i o n within and among conifer populations: Review and evaluation of methods. In: International Workshop of Plant Biology, Biochemical Markers i n the Population Genetics of Forest Trees. Orvieto, I t a l y . (in press). El-Kassaby, Y.A., A.M.K. Fashler and 0. S z i k l a i . 1987a. Ef f e c t of family size and number on the accuracy and precision of the estimates of genetic parameters in the IUFRO Douglas-fir provenance-progeny t r i a l . For. Ecol. Manage. 18:35-48. 182 El-Kassaby, Y.A., M.D. Meagher, J. Parkinson and F.T. Portlock. 1987b. Allozyme inheritance, heterozygosity and outcrossing rate among Pinus monticola near Ladysmith, B r i t i s h Columbia. Heredity 58:173-181. El-Kassaby, Y.A., J. Parkinson and W.J.B. Devitt. 1986. The e f f e c t of crown segment on the mating system i n a Douglas-fir {Pseudotsuga menziesii [Mirb.J Franco) seed orchard. Silvae Genet. 35:149-155. El-Kassaby, Y.A., K. Ritland, A.M.K. Fashler and W.J.B. Devitt. 1988. The role of reproductive phenology upon the mating system of a Douglas-fir seed orchard. Silvae Genet. 37:76-82. El-Kassaby, Y.A., D. Rudin and R. Yazdani. 1989. Levels of outcrossing and contamination in two Scots pine seed orchards. Scand. J. For. Res. 4:41-49. El-Kassaby, Y.A., 0. S z i k l a i and F.C. Yeh. 1982a. Linkage relationships among 19 polymorphic allozyme l o c i i n coastal Douglas-fir [Pseudotsuga menziesii var. menziesii (Mirb.) Franco]. Can. J. Genet. Cytol. 24:101-108. El-Kassaby, Y.A., F.C. Yeh and 0. S z i k l a i . 1982b. Inheritance of allozyme variants i n coastal Douglas-fir (Pseudotsuga menziesii var. menziesii). Can. J. Genet. Cytol. 24:325-335. El-Kassaby, Y.A. and 0. S z i k l a i . 1983. Ef f e c t of sample size on the precision of the estimate of allozyme frequencies i n a natural stand of Douglas-fir. Egypt J. Genet. Cytol. 12:345-359. El-Kassaby, Y.A., F.C. Yeh, and 0. S z i k l a i . 1981. Estimation of the outcrossing rate of Douglas-fir [Pseudotsuga menziesii (Mirb.) Franco] using allozyme polymorphisms. Silvae Genet. 30:182-184. Ennos, R.A. and M.T. Clegg. 1982. Effect of population substructuring on estimates of outcrossing rate i n plant populations. Heredity 48:283-292. Epperson, B.K. and R.W. A l l a r d . 1984. Allozyme analysis of the mating system in lodgepole pine populations. Heredity 48:283-292. Ernst, S.G., D.E. Keathley and J.W. Hanover. 1987. Inheritance of isozymes in seed and bud tissues of blue and Engelmann spruce. Genome 29:239-246. 183 Falconer, D.S. 1981. Introduction to Quantitative Genetics. 2nd ed. Longman Press, N.Y. Falkenhagen, E.R. 1985. Isozyme studies i n provenance research of forest trees. Theor. Appl. Genet. 69:335-347. Farmer, R.E. Jr . and R.W. Reinholt. 1986. Seed qua l i t y and germination c h a r a c t e r i s t i c s of tamarack i n northwestern Ontario. Can. J. For. Res. 16:680-683. F a r r i s , M.A. and J.B. Mitton. 1984. Population density, outcrossing rate, and heterozygote s u p e r i o r i t y i n ponderosa pine. Evolution 38:1151-1154. Feret, P.P. and F. Bergmann. 1976. Gel electrophoresis of proteins and enzymes. In: Modern Methods i n Forest Genetics. (J.P. Miksche, ed.) Springer-Verlag, N.Y. pp. 49-77. Fienberg, S.E. 1970. The analysis of multidimensional contingency tables. Ecology 51:419-433. Fins, L. and W.J. Libby. 1982. Population v a r i a t i o n i n Sequoiadendron: Seed and seedling studies, vegetative propagation and isozyme v a r i a t i o n : Silvae Genet. 31:102-110. Fisher, R.A. 1932. S t a t i s t i c a l methods for research workers (4th ed.). Oliver and Boyd, Edinburgh. Fletcher, A.M. and H. Barner. 1978. The procurement of seed for provenance research with p a r t i c u l a r reference to c o l l e c t i o n s in NW America. In: Proc. IUFRO Joint Mtg. of Working Parties I. Background papers and Douglas-fir provenances. B.C. Min. of Forests, Info. Serv. Branch, V i c t o r i a , B.C. Canada. pp.141-154. Foster, A.S. and E.M. Gi f f o r d . 1974. Comparative morphology of vascular plants. Freedman and Co., San Fancisco. Fowells, H.A. 1965. S i l v i c s of forest trees of the United States. USDA Agric. Handb. No. 271. pp.546-556. Fox, D.J. and K.E. Guire. 1976. Documentation for MIDAS, 3rd. ed. S t a t i s t i c a l Research Laboratory, Univ. of Michigan, Ann Arbor, Michigan. Franklin, E.C. 1970. Survey of mutant forms and inbreeding depression i n species of the family Pinaceae. USDA For. Serv. Res. Pap. SE-61, 21pp. 184 Franklin, J.F. 1974. Abies M i l l . F i r . In: Seeds of Woody Plants of the United States. (CS. Schopmeyer, Tech. Coord.) U.S.D.A. Agric. Handb. No. 450:168-183. Franklin, J.F. and G.A. R i t c h i e . 1970. Phenology of cone and shoot development of noble f i r and some associated true f i r s . Forest S c i . 16:356-364. Friedman, S.T. and W.T. Adams. 1985. Levels of outcrossing i n two l o b l o l l y pine seed orchards. Silvae Genet. 34:157-162. Fryer, J.H. and F.T. Ledig. 1972. Microevolution of the photosynthetic temperature optimum i n r e l a t i o n to the elevational complex gradient. Can. J. Bot. 50:1231-1235. Funck, P., S. Banerjee and J. Maze. 1990. The structure of v a r i a t i o n in Abies amabilis as assessed through a provenance t e s t . (Submitted to Can. J. Bot.) Furnier, G.R. and W.T. Adams. 1986. Mating system i n natural populations of Jeffrey pine. Amer. J. Bot. 73:1002-1008. Furnier, G.R., P. Knowles, M.A. Aleksiuk and B.P. Dancik. 1986. Inheritance and linkage of allozymes i n seed tissue of white bark pine. Can. J. Genet. Cytol. 28:601-604. Fyfe, J.L. and N.T.J. Bailey. 1951. Plant breeding studies in leguminous forage crops. In: Natural cross-breeding i n winter beans. J. Agric. S c i . 41:371-378. Glass, G.V., P.D. Peckham and J. R. Sanders. 1972. Consequences of f a i l u r e to meet assumptions underlying the fixed e f f e c t s analysis of variance and covariance. Rev. Educ. Res. 42:239-288. Goodman, M.M., CW. Stuber, C.-N. Lee and F.M. Johnson. 1980. Genetic control of malate dehydrogenase isozymes i n maize. Genetics 94:153-168. Gottlieb, L.D. 1979. Electrophoretic evidence and plant populations. In: Progress in phytochemistry Vol. 7 (L. Reinhold and Y. Linschita, eds.) Interscience Press, NY. pp.1-46. Gottlieb, L.D. 1982. Conservation and duplication of isozymes in plants. Science 216:373-380. 185 Grant, M.C. and J. B. Mitton. 1977. Genetic d i f f e r e n t i a t i o n among growth forms of Engelmann spruce and subalpine f i r at tree l i n e . A r c t i c ad Alpine Res. 9:259-263. Green, R.N., P.J. Courtin, K. Klinka, R.J. Slaco and C.A. Ray. 1984. Site diagnosis, tree species selection and slashburning guidelines for the Vancouver Forest Region. Abridged version. B r i t i s h Columbia Ministry of Forests Land Management Handbook No. 8. Gregorius, H.-R. 1987. The relat i o n s h i p between the concepts of genetic d i v e r s i t y and d i f f e r e n t i a t i o n . Theor. Appl. Genet. 74:397-401. Gregorius, H.-R., J. Krauhausen and G. Miiller-Starck. 1986. Spatial and temporal genetic d i f f e r e n t i a t i o n among the seed i n a stand of Fagus s y l v a t i c a L. Heredity 57:255-262. Gregorius, H. -R. and J. H. Roberds. 1986. Measurement of genetical d i f f e r e n t i a t i o n among subpopulations. Theor. Appl. Genet. 71:826-834. G r i f f i n , A.R. and K.K. Ching. 1977. Geographic v a r i a t i o n i n Douglas-fir from the coastal ranges of C a l i f o r n i a . I. Seed, seedling growth and hardiness c h a r a c t e r i s t i c s . Silvae Genet. 26:149-157. Guries, R.P. and F.T. Ledig. 1977. Analysis of population structure from allozyme frequencies. In: Proc. 14th South. For. Tree Improv. Conf. G a i n s v i l l e , F i a . June 14-16, 1977. pp.246-253. Guries, R.P. and F.T. Ledig. 1978. Inheritance of some polymorphic isozymes i n p i t c h pine {Pinus r i g i d a M i l l . ) . Heredity 40:27-32. Guries, R.P., S.T. Friedman and F.T. Ledig. 1978. A megagametophyte analysis of genetic linkage i n p i t c h pine (Pinus r i g i d a M i l l . ) . Heredity 40:309-314. Guries, R.P. and F.T. Ledig. 1981. Genetic structure of populations and d i f f e r e n t i a t i o n i n forest trees. In: Proc. Symp. Isozymes of North Amer. For. Trees and For. Insects (M.T. Conkle, coord.) USDA For. Serv. Gen. Tech. Rep. PSW-48. pp. 42-47. Gusella, J.F., N.S. Wexler, P.M. Conneally, S.L. Naylor, M.A. Anderson, R.E. Tanzi, P.C. Watkins, K. Ottina, M.R. Wallace, A.Y. Sakaguchi, A.B. Young, 186 I. Shoulson, E. B o n i l l a and J. B. Martin. 1983. A polymorphic DNA marker gene t i c a l l y linked to Huntington's disease. Nature 306:234-238. Hamrick, J.L. 1976. Variation and selection i n Western montane species II. Variation within and between populations of white f i r on an elevational transect. Theor. Appl. Genet. 47:27-34. Hamrick, J.L. 1982. Plant population genetics and evolution. Amer. J. Bot. 69(10):1685. Hamrick, J.L., B. Linhart and J.B. Mitton. 1979. Relationships between l i f e history c h a r a c t e r i s t i c s and e l e c t r o p h o r e t i c a l l y detectable genetic v a r i a t i o n i n plants. Ann. Rev. Ecol. Syst. 10:173-200. Hamrick, J.L., J.B. Mitton and B. Linhart. 1981. Levels of genetic v a r i a t i o n i n trees: the influence of l i f e history c h a r a c t e r i s t i c s . In: Proc. Symp. Isozymes of North Amer. For. Trees and For. Insects (M.T. Conkle, coord.) USDA For. Serv. Gen. Tech. Rep. PSW-4 8. pp. 35-41. Handley, D.L. 1982. The B r i t i s h Columbia industry's inter e s t i n the true f i r s . In: The Biology and Management of True F i r in the P a c i f i c Northwest. (CD. Oliver and R. M. Kenady, eds.) USDA For. Serv. Pac. Nor. West Range and Exp. Sta. and Univ. of Wash. Seattle, WA. pp.261-262. Harper, J.L., P.H. Lovell and K.G. Moore. 1970. The shapes and sizes of seeds. Ann. Rev. Ecol. Syst. 1:327-356. Harry, D.E. 1986. Inheritance and linkage of isozyme variants i n incense-cedar. J. Hered. 77:261-266. Harry, D.E. 1983. I d e n t i f i c a t i o n of a locus modifying the electrophoretic mobility of malate dehydrogenase isozymes i n incense-cedar (Calocedrus decurrens) and i t s implications for population studies. Biochem. Genet. 21:417-434. Hart, G.E. and P.J. Langston. 1977. Glutamate oxaloacetate transaminase isozymes of the Triticinae: d i s s o c i a t i o n and recombination of subunits. Theor. Appl. Genet. 50:47-51. Hartl, D.L. 1980. Pri n c i p l e s of population genetics. Sinauer Assoc. Sunderland, Mass. 187 Hattemer, H.H. 1979. Problems in population genetic conclusions from biochemical marker studies. In: Proc. Conf. on Biochem. Genet, of For. Trees. Umea, Sweden, 1978. pp. .73-85. Hedrick, P.W. 1983. Genetics of populations. S c i . Books Int. Portola Valley, CA. Helentjaris, T. 1987. A genetic linkage map for maize based on RFLPs. Trends in Genetics 3:217-221. Helium, A.K. 1968. Variation in cotyledon number and seed weight i n white spruce in Alberta. In: Proc. 11th Mtg. Comm. on For. Tree Breeding i n Canada, Part 2. Rep. and Pap. pp. 65-76. Heydecker, W. 1969. The 'vigour' of seeds-a review. Proc. Int. Seed Test. Assoc. 34:210-220. H i c k s , C R . 1982. F u n d a m e n t a l c o n c e p t s i n t h e d e s i g n o f e x p e r i m e n t s (3rd e d . ) . H o l t , R i n e h a r t a n d W i n s t o n , New Y o r k . H i l l , W.G. 1974. E s t i m a t i o n o f l i n k a g e d i s e q u i l i b r i u m i n r a n d o m l y m a t i n g p o p u l a t i o n s . H e r e d i t y 33:229-239. H o d g s o n , T.J. 1980. P i n e s e e d g r a d i n g : t h e i m p l i c a t i o n s f o r o r c h a r d s e e d . S v i d - A f r i k a a n s e B o s b o u t y d s k r i f 112:10-14. Huang, A.H.C, K.D.F. Liu and R.J. Youle. 1976. Organelle-s p e c i f i c isozymes of aspartate-alpha-ketoglutarate transaminase in spinach leaves. Plant Physiol. 58:110-113. Huehn, M . , J. Kleinschmit and J. Svolba. 1987. Some experimental results concerning age dependency of d i f f e r e n t components of variance i n t e s t i n g Norway spruce (Picea abies (L.) Karst.) clones. S i l v . Genet. 36:68-71. Hunter, R.L. and C L . Markert. 1957. Histochemical demonstration of enzymes separated by zone electrophoresis i n starch gels. Science 125:1294-1295. Husted, L. and V. Korelus. 1982. Management of true f i r species for timber production - south coastal B r i t i s h Columbia and the Olympic peninsula. In: The Biology and Management of True F i r in the P a c i f i c Northwest. (CD. Oliver and R. M. Kenady, eds.) USDA For. Serv. Pac. Nor. West Range and Exp. Sta. and Univ. of Wash. Seattle, WA. pp.277-283. 188 Jacobs, B.F., CR. Werth and S.I. Guttman. 1984. Genetic relationships in Abies ( f i r ) of eastern United States: an electrophoretic study. Can. J. Bot. 62:609-616. Jann, R.C. and R.D. Amen. 1977. What i s germination? In: The physiology and biochemistry of seed dormancy and germination (A.A. Khan, ed.). North-Holland, Amsterdam. pp. 7-28. Jones, D. and N. Matloff. 1986. S t a t i s t i c a l hypothesis t e s t i n g i n biology: a contradiction i n terms. J. Econ. Entomol. 7 9:1156-1160. Kahler, A.L., A.R. Hallauer and CD. Gardner. 1986. Allozyme polymorphisms within and among open-p o l l i n a t e d and adapted exotic populations of maize. Theor. Appl. Genet. 72:592-601. K h a l i l , M.A.K. 1981. Correlation of juvenile height growth with cone morphology and seed weight i n white spruce. Silvae Genet. 30:179-181. K h a l i l , M.A.K. 1985. H e r i t a b i l i t y of juvenile characters of white spruce (Pica glavca [Moench.] Voss.) i n central Newfoundland, Canada. Theor. Appl. Genet. 69:247-251. K h a l i l , M.A.K. 1986. Variation in seed quality and some juvenile characters of white spruce (Picea glauca [Moench] Voss). Silvae Genet. 35:78-85. King, J.N. and B.P. Dancik. 1983. Inheritance and linkage of isozymes in white spruce (Picea glauca) . Can. J. Genet. Cytol. 25:430-436. J.N., B.P. Dancik and N.K. Dhir. 1984. Genetic structure and mating system of white spruce (Picea glauca) in a seed production area. Can. J. For. Res. 14:639-643. J.N., F.C Yeh and J.C. Heaman. 1988. Selection of growth and y i e l d t r a i t s in controlled crosses of coastal Douglas-fir. Silvae Genet. 37:158-164. K i t z m i l l e r , J.M., J.M. B a l t i g i n and J.A. Helms. 1973. Eff e c t of resin v e s i c l e damage on germination of stored Abies concolor seed. Int. Rep. No. 1. True F i r Mgmt. Co-op. School of Forestry and Conservation. Univ. of C a l i f . Berkeley, 16pp. Ki t z m i l l e r , J.M.., J.A. Helms and P. Stover. 1975. Influence of seed treatment on germination of King, King, 189 Abies concolor and A. magnifica. Int. Rep. No. 3. True F i r Mgmt. Co-op. School of Forestry and Conservation Univ. of C a l i f . , Berkeley, 39pp. Kleinschmit, J . 1986. Nursery results of the Abies grandis (Lindl.) provenance experiment i n northern Germany. In: IUFRO Abies grandis provenance experiments: nursery stage r e s u l t s (A;M. Fletcher, ed.) Forestry Commission Research and Devel. Pap. 139. pp. 39-58. Knowles, P., G.R. Furnier, M.A. Aleksiuk and D.J. Perry. 1987. Si g n i f i c a n t l e v e l s of s e l f - f e r t i l i z a t i o n i n natural populations of tamarack. Can. J. Bot. 65:1087-1091. Konishi, J., M. Crown and M. A l b r i c h t . 1989. Co-operative seed orchards in B r i t i s h Columbia, 1987-1989. In: Proc. 22nd Mtg. of the Can. Tree Impr. Assoc., Part 1. (S. Magnussen and T.J.B. Boyle, eds.) Edmonton, A l t a . , 1989. pp. 136-139. Kotar, J. 1972. Ecology of Abies amabilis in r e l a t i o n to a t t i t u d i n a l d i s t r i b u t i o n and in contrast to i t s common associate Tsuga heterophylla. Ph.D. Thesis, Univ. of Washington, Seattle, WA. 171 pp. Kotar, J. 1978. Relationships of early seedling development to a l t i t u d i n a l d i s t r i b u t i o n of Abies amabilis. B u l l . Torrey Bot. Club 105:289-295. Krajina, V.J. 1969. Ecology of forest trees i n B r i t i s h Columbia. Ecol. Western Nor. Amer. 2:1-147. Krajina, V.J., K. Klinka and J. Worrall. 1982. Di s t r i b u t i o n and ecological c h a r a c t e r i s t i c s of trees and shrubs of B r i t i s h Columbia. Faculty of Forestry, UBC. 13.1 pp. Krugman, S.L., W.I. Stein and D.M. Schmitt. 1974. Seed biology. In: Seeds of woody plants of the United States. C S . Schopmeyer (coord.) USDA Agric. Handb. No. 450. pp. 5-4 0. Krumlik, G.J., L. Husted and J.P. Kimmins. 1982. Nutrient cy c l i n g i n true f i r stands in western North America. In: The Biology and Management of True F i r i n the P a c i f i c Northwest. (CD. Oliver and R. M. Kenady, eds.) USDA For. Serv. Pac. Nor. West Range and Exp. Sta. and Univ. of Wash. Seattle, WA. pp.145-156. 190 Lande, R. and D.W. Schemske. 1985. The evolution of s e l f -f e r t i l i z a t i o n and inbreeding depression i n plants. I. Genetic models. Evolution 39:24-40. Lavender, D.P. 1958. E f f e c t of seed size on Douglas-fir seedlings. Greg. For. Lands Res. Cent. Note No. 32. Leadem, C L . 1986. S t r a t i f i c a t i o n of Abies amabilis seeds. Can. J. For. Res. 16:755-760. Ledig, F.T. 1974. An analysis of methods for selection of trees from wild stands. For. S c i . 20:2-16. Lester, D.T. 1968. Variation i n cone morphology Of balsam f i r , Abies balsamea. Rhodora 70:83-94. Lester, D.T. 1970. Variation in seedling development of balsam f i r associated with seed origin.. Can. J. Bot. 48:1093-1097. Lester, D.T., R.M. J e f f e r s and J.W. Wright. 1976. Provenance and family v a r i a t i o n i n balsam f i r from Michigan and Wisconsin. In: Pro. 12th Lakes States forest tree improvement conf. Aug. 1975. USDA For. Serv. North Cent. For. Exp. Sta. St. Paul Minn. pp.74-79. Levins, R. 1969. Dormancy as an adaptive strategy. Symp. Soc. Exp. B i o l . 23:1-10. Lewontin, R.C. 1974. The genetic basis of evolutionary change. Columbia Univ. Press, New York. Lewontin, R.C. 1984. Detecting population differences i n quantitative characters as opposed to gene frequencies. Amer. Nat. 123:115-124. Lewontin, R.C. and C C . Cockerham. 1959. The goodness-of-f i t test for detecting selection i n random mating populations. Evolution 13:501-564. L i , P. and W.T. Adams. 1989. Range-wide patterns of allozyme v a r i a t i o n i n Douglas-fir (Pseudotsuga menziesii). Can. J. For. Res. 19:149-161. Libby, W.J. 1982. Biology of true f i r s : genetic factors, problems and opportunities In: The Biology and Management of True F i r in the P a c i f i c Northwest. (CD. Oliver and R. M. Kenady, eds.) USDA For. Serv. Pac. Nor. West Range and Exp. Sta. and Univ. of Wash. S e a t t l e , WA. pp.15-18. 191 Libby, W.J., R.F. S t e t t l e r and F.W. Seitz. 1969. Forest genetics and forest-tree breeding. Ann. Rev. For. Genet. 3:469-494. Lines, R. 1967. Standardization of methods for provenance research. In: Proc. XIV IUFRO Congress 3(22). J.D. Matthews, ed. BVFFA Sekretariat fur den IUFRD - Kongress. pp. 672-718. Lines, R. 1978. The IUFRO experiments with Abies grandis i n B r i t a i n - nursery stage. In: Proc. IUFRO Joint Mtg. of Working Parties II. Lodgepole pine, Sitka spruce and Abies provenances. B.C. Min. of For., Info. Serv. Branch, V i c t o r i a , B.C. Canada, pp.339-345. Linhart, Y.B., J.B. Mitton, K.B. Sturgeon and M.L. Davis. 1981. Genetic v a r i a t i o n in space and time i n a population of ponderosa pine. Heredity 46:407-426. L i t t l e , T.M. 1981. Interpretation and presentation of r e s u l t s . Hort. S c i . 16:637-640. Loveless, M.D. and J.L. Hamrick. 1984. Ecolog i c a l determinants of genetic structure in plant populations. Ann. Rev. Ecol. Sys. 15:65-95. Lundkvist, K. 1979. Allozyme frequency d i s t r i b u t i o n s i n four Swedish populations of Norway spruce (Picea abies ¥,.) . Hereditas 90:127-143. Madsen, J.L. and G.M. Blake. 1977. Ecological genetics of ponderosa pine in the northern Rocky mountains. Silvae Genet. 26:1-8. Maguire, J.D. 1977. Seed quality and germination. In: The physiology and biochemistry of seed dormancy, and germination. (A.A. Khan, ed.) North-Holland. Amsterdam, pp. 219-235. Mann, W.F., J r . 1979. Relationships of seed size, number of cotyledons and i n i t i a l growth of southern pines. Tree Planters' Notes 30:22-23. Markert, C L . and C S . Whitt. 1968. Molecular v a r i e t i e s of isozymes. Experentia 24:977-991. Marshall, D.R. and A.H.D. Brown. 1975. Optimum sampling strategies in genetic conservation. In: Crop Genetic Resources for Today and Tomorrow (J.H. Frankel and J.G. Hawkes, eds.) Cambridge Univ. Press, London, pp. 53-80. 1 9 2 Martin, W.L. 1985. Post-clearcutting forest f l o o r nitrogen dynamics and regeneration response i n the Coastal Western Hemlock wet subzone. Ph.D. thesis Fac. For. Univ. of B.C., Vancouver, B.C. Maze, J., S . Banerjee and Y.A. El-Kassaby. 1989. V a r i a t i o n of growth rate within and among f u l l - s i b families of Douglas-fir (Pseudotsuga menziesii). Can. J. Bot. 67:140-145. Mikola, J. 1980. The eff e c t of seed size and duration of growth on the height of Scots pine (Pinus s y l v e s t r i s L.) and Norway spruce (Picea abies [L.] Karst) provenances and progenies at the nursery stage. S i l v a Fennica 14:84-94. M i l l a r , C.I. 1983. A steep c l i n e in Pinus muricata. Evolution 37:311-319. M i l l a r , C.I. 1985. Inheritance of allozyme variants i n Bishop pine (Pinus muricata D. Don). Biochem. Genet. 23 (11/12) :933-946. Mitton, J.B. 1983. Conifers. In: Isozymes in Plant Genetics and Plant Breeding, Part B (S.D. Tanksley and T.J. Orton, eds.). Elsevier, Amsterdam. pp. 443-472. Mitton, J.B., Y.B. Linhart, M.L. Davis and K.B. Sturgeon. 1981. Estimation of outcrossing i n ponderosa pine, Pinus ponderosa Laws., from patterns of segregation of protein polymorphisms and from frequencies of albino seedlings. Silvae Genet. 30:117-121. Mueller, L.D. 1979. A comparison of two methods for making s t a t i s t i c a l inference on Nei's measure of genetic distance. Biometrics 35:757-763. Muller-Starck, G. and H.-R. Gregorius. 1986. Monitoring genetic v a r i a t i o n in forest tree populations. In: 18th IUFRO World Congress, Ljubljana. Yugoslavia, 1986. Divi s i o n 2, Vol. II (E. Donaubauer, ed.) Sept. 7-21, pp. 589-599. Mulcahy, D.L. and S.M. Kaplan. 1979. Mendelian r a t i o s despite non-random f e r t i l i z a t i o n ? Amer. Nat. 113:419-425. Muona, 0. and A.E. Szmidt. 1985. A multi-locus study of natural populations of Pinus s y l v e s t r i s . In: Population genetics in forestry. Lecture notes i n biomathematics 60 (H.-R. Gregorius, ed.). Springer-Verlag. pp. 226-240. 193 Muona, 0., R. Yazdani and G. Lindqvist. 1987. Analysis of linkage in Picea abies. Heredity 106:31-36. Murray, M.D. and D.L. Treat. 1979. P a c i f i c s i l v e r f i r i n the Coast Range of southwestern Washington. Northwest S c i . 54:119-120. Neale, D. 1978. Allozyme studies in balsam f i r . M.Sc. the s i s . Univ. of New Hampshire. Durham, N.H. Neale, D.B. and W.T. Adams. 1981. Inheritance of isozyme variants i n seed tissues of balsam f i r (Abies balsamea) . Can. J. Bot. 59:1285-1291. Neale, D.B. and W.T. Adams. 1985a. The mating system i n natural and shelterwood stands of Douglas-fir. Theor. Appl. Genet. 71:201-207. Neale, D.B. and W.T. Adams. 1985b. Allozyme and mating-system v a r i a t i o n in balsam f i r (Abies balsamea) across a continuous elevational transect. Can. J. Bot. 63:2448-2453. Neale, D.B., J.C. Weber and W.T. Adams. 1984. Inheritance of needle tissue isozymes in Douglas-fir. Can. J. Genet. Cytol. 26:459-468. Nei, M. 1972. Genetic distance between populations. Amer. Nat. 106:283-292. Nei, M. 1973. Analysis of gene d i v e r s i t y i n subdivided populations. Proc. Nat. Acad. S c i . USA 70:3321-3323. Nei. M. 1975. Genetic v a r i a b i l i t y in natural populations. In: Molecular population genetics and evolution. North-Holland Publ. Amsterdam, Oxford. pp. 127-174. Nei. M. 1977. F - s t a t i s t i c s and analysis of gene d i v e r s i t y in subdivided populations. Ann. Hum. Genet. 41:225-233. Nei, M. 1978. Estimation of average heterozygosity and genetic distance from a small number of ind i v i d u a l s . Genetics 89:583-590. Nei, M. and R.K. Chesser. 1983. Estimation of f i x a t i o n indices and gene d i v e r s i t i e s . Ann. Hum. Genet. 47:253-259. 194 Nei, M. and A.K. Roychoudhury. 1974. Sampling variances of heterozygosity and genetic distance. Genetics 76:379-390. Nevo, E. 1978. Genetic v a r i a t i o n i n natural populations: patterns and theory. Theor. Pop. B i o l . 13:121-177. Newton, K.J. 1979. A gene which a l t e r s the electrophoretic m o b i l i t i e s of maize mitochondrial malate dehydrogenase isozymes. Genetics 91(Suppl.):S88-S89(Abs.) Newton, K.J. and D. Schwartz. 1980. Genetic basis of the major malate dehydrogenase isozymes in maize. Genetics 95:425-442. Nordheim, E.V., D.M. O'Malley and R.P. Guries. 1983. Estimation of recombination frequency i n linkage studies. Theor. Appl. Genet. 66:313-321. O'Malley, D.M., F.W. Allendorf and G.M. Blake. 1979. Inheritance of isozyme v a r i a t i o n and heterozygosity in Pinus ponderosa. Biochem. Genet. 17:233-250. O'Malley, D.M., R.P. Guries and E.V. Nordheim. 1986. Linkage analysis for 18 enzyme l o c i i n Pinus 'rigida M i l l . Theor. Appl. Genet. 72:530-535. Osborn, T.C., D.C Alexander and J.F. Fobes. 1987. I d e n t i f i c a t i o n of r e s t r i c t i o n fragment length polymorphisms linked to genes c o n t r o l l i n g soluble s o l i d s content in tomato f r u i t . Theor. Appl. Genet. 73:350-356. Owens, J.N. and A.A. Molder. 1977. Sexual reproduction of Abies amabilis. Can. J. Bot. 55:2653-2667. Packee, E.C. 1974. The biogeoclimatic zones of Vancouver Island and the adjacent mainland and islands (Rev.). MacMillan Bloedel Ltd. For. Div. For. Res. Note No. 1. 24pp. Packee, E.C, CD. Oliver and P. D. Crawford. 1982. Ecology of P a c i f i c s i l v e r f i r . In: The Biology and Management of True F i r in the P a c i f i c Northwest. (CD. Oliver and R.M. Kenady, eds.) USDA For. Serv. Pac. Nor. West Range and Exp. Sta. and Univ. of Wash., Seattle, WA. pp.19-34. Palmblad, I.G. 1968. Competition in experimental populations of weeds with emphasis on the regulation of population size. Ecology 49:26-34. 195 Parker, W.H., G.E. Bradfield, J. Maze and S.C. L i n . 1979a. Analysis of v a r i a t i o n in leaf and twig characters of Abies lasiocarpa and A. amabilis from north-coastal B r i t i s h Columbia. Can. J. Bot. 57:1354-1366. Parker, W.H.,. J. Maze and D.G. McLachlan. 197 9b. Flavonoids of Abies amabilis needles. Phytochemistry 18:508-510. Paule, L. 199.0. Bibliography: Isozymes and Forest Trees. Swedish University of Agric. S c i . Umea, Sweden. Perry, D.J. and B.P. Dancik. 1986. Mating system dynamics of lodgepole pine in Alberta, Canada. Silvae Genet. 35:190-195-Perry, T.O. 1976. Maternal effects on the early performance of tree progenies. In: Tree Physiology and Y i e l d Improvement. (M.G.R. Cannell, and F.T. Last, eds.) Acad. Press London and New York. pp. 473-481. P h i l l i p s , M.A. and A.H.D. Brown. 1977. Mating system and hybridity in Eucalyptus p a u c i f l o r a . Aust. J. B i o l . S c i . 30:337-344. P i t e l , J.A., W.M. Cheliak and B.S.P. Wang. 1984. Changes i n isozyme patterns during imbibition and germination of lodgepole pine (Pinus contorta var. l a t i f o l i a ) . Can. J. For. Res. 14:743-746. Plessas, M.E. and S.H. Strauss. 1986. Allozyme d i f f e r e n t i a t i o n among populations, stands and cohorts i n Monterey pine. Can. J. For. Res. 16:1155-1164. Pojar, J. 1982 . Management of western hemlock - amabilis f i r forests for timber production: north coast. In: The Biology and Management of True F i r i n the P a c i f i c Northwest. (CD. Oliver and R. M. Kenady, eds.) USDA For. Serv. Pac. Nor. West Range and Exp. Sta. and Univ. of Wash. Seattle, WA. pp.262-276 . Poustka, A., T.M. Pohl, D.P. Barlow, A.-M. Frischauf and H. Lehrach. 1987. Construction and use of human chromosome jumping l i b r a r i e s from Note I -digested DNA. Nature 3 2 5 : 3 5 3 - 3 5 5 . Rasmussen, D.I. 1964. Blood group polymorphisms and inbreeding i n natural populations of the deer 196 mouse Peromyscus maniculatus. Evolution 18:219-229. Rehfeldt, G.E. 1983. Genetic v a r i a b i l i t y within Douglas-f i r populations: implications for tree improvement. Silvae Genet. 32:9-14... Rehfeldt, G.E. and D.T. Lester. 1969. S p e c i a l i z a t i o n and f l e x i b i l i t y i n genetic systems of forest trees. Silvae.Genet. 18:118-123. Reuter, F. 1973. High elevation reforestation problems i n the Vancouver Forest D i s t r i c t . B.C. For. Serv. Research Branch Rep. p. 45. Ritland, K. 1983. Estimation of mating systems. In: Isozymes in Plant Genetics and Plant Breeding, Part A, (S.D. Tanksley and T.J. Orton, eds.). Els e v i e r , Amsterdam. pp. 289-302. Ritland, K. and Y.A. El-Kassaby. 1985. The nature of inbreeding i n a seed orchard of Douglas-fir as shown by an e f f i c i e n t multilocus model. Theor. Appl. Genet..71:375-384. Roberds, J.H. and M.T. Conkle. 1984. Genetic structure i n l o b l o l l y pine stands: allozyme v a r i a t i o n i n parents and progeny. Forest S c i . 30:319-329. Roberds, J.H. and J.V. Brotschol. 1985. Linkage d i s e q u i l i b r i a among allozyme l o c i i n natural populations of Liriodendron t u l i p i f e r a L. Silvae Genet. 34:137-141. Rowe, J.S. 1964. Environmental preconditioning, with special reference to forestry. Ecology 45:399-403. Rudin, D. 1975. Inheritance of glutamate - oxalate -transaminase (GOT) from needles and endosperms of Pinus s y l v e s t r i s . Hereditas 80:296-300. Rudin, D. 1976. Biochemical genetics and selection, a p p l i c a t i o n of isozymes in tree breeding. In: Proc. IUFRO Joint Meeting of Working Parties on Population Genetics, Breeding Theory, Biochemical Genetics and Progeny Testing. (Bordeaux, France. June 14-16, 1976.). pp. 145-164. Rudin, D. and I. Ekberg. 1978. Linkage studies i n Pinus s y l v e s t r i s L. using macro gametophyte allozymes. Silvae Genet. 27:1-12. 197 Salazar, R. 1986. Genetic v a r i a t i o n i n seeds and seedlings of ten provenances of G l i r i c i d i a sepium (Jacq.) Steud. For. Ecol. Manage. 16:391-401. Scandalios, J.G. 1975. Genes, isozymes and evolution. In: Isozymes (IV): Genetics and evolution ( C L . Markert, ed.). Acad. Press, N.Y. pp. 1-7. Schmidt, R.L. 1957. The s i l v i c s and plant geography of the genus Abies. In: The coastal forests of B r i t i s h Columbia. B.C. For. Serv. Tech. Publ. T46. 31pp. Schoen, D.J. 1988. Mating system estimation v i a the one pollen parent model with the progeny array as the unit of observation. Heredity 60:439-444. Schoen, D.J. and M.T. Clegg. 1984. Estimation of mating system parameters when outcrossing events are correlated. Proc. Natl. Acad. S c i . USA 81:5258-5262. Schoen, D.J. and M.T. Clegg. 1986. Monte Carlo studies of plant mating system estimation models: the one-pollen parent and mixed mating models. Genetics 112:927-945. Schoener, T.W. 1970. Nonsynchronous s p a t i a l overlap of •lizards i n patchy habitats. Ecology 51:408-418. Selander, R.K. 1976. Genie v a r i a t i o n in natural populations (Ch. 2). In: Molecular Evolution (F.J. Ayala, ed). Sinauer Assoc. Sunderland Mass. pp. 21-59. Shaw, D.V. and R.W. A l l a r d . 1982a. Estimation of outcrossing rates in Douglas-fir using isozyme markers. Theor. Appl. Genet. 62:113-120. Shaw, D.V. and R.W. A l l a r d . 1982b. Isozyme heterozygosity i n adult and open-pollinated embryo samples of Douglas-fir. Silvae Fenn. 16:115-121. Shaw, D.V., A.L. Kahler and R.W. A l l a r d . 1981. A multilocus estimator of mating system parameters i n plant populations. Proc. Nat. Acad. S c i . U.S.A. 78:1298-1302. Shea, K.L. 1987. Effects of population structure and cone production on outcrossing rates i n Engelmann spruce and subalpine f i r . Evolution 41:124-136. Shea, K.L. 1988. Segregation of allozyme l o c i i n megagametophytes of Engelmann spruce and subalpine f i r . Genome 30:103-107. 198 Simon, C. and J. Archie. 1985. An empirical demonstration of the l a b i l i t y of heterozygosity estimates. Evolution 39:463-467. Simonsen, V. and H. Wellendorf. 1975. Some polymorphic isoenzymes in the seed endosperm of Sitka spruce (Picea s i t c h e n s i s [Bong.] Carr.). Forest Tree Improvement 9:5-20. Smith, C.A.B. 1970. A note on te s t i n g the Hardy-Weinberg law. Ann. Hum. Genet. 33:377-383. Smithies, 0. 1955. Zone electrophoresis in starch gels: group v a r i a t i o n i n the serum proteins of normal human adults. Biochem. J. 61:629-663. Snyder, T.P.-, D.A. Stewart and A.F. S t r i c k l e r . ' 1985. Temporal analysis of breeding structure i n jack pine (Pinus banksiana Lamb.). Can. J. For. Res. 15:1159-1166. Sokal, R.R. and F.J. Rohlf. 19-8-1. Biometry (2nd ed.). W.H. Freeman and Co. N.Y. F.C. 1982. The roles of polyembryony and embryo v i a b i l i t y i n the genetic system.of conifers. Evolution 36:725-733. F.C. 1983. Geographic v a r i a t i o n i n seedling Douglas-fir (Pseudotsuga menziesii) from the western Siskiyou Mountains of Oregon. Ecology 64:696-702. F.C. and R.K. Campbell. 1985. Ef f e c t of seed weight on height growth of Douglas-fir (Pseudostuga menziesii [Mirb.] Franco) seedlings in a nursery. Can. J. For. Res. 15:1109-1115. F.C. and J.F. Franklin. 1977. Influence of year of cone c o l l e c t i o n on seed weight and cotyledon number i n Abies procera. Silvae Genet. 26:41-43. F.C, J.F. Franklin and R. Woollard. 1976. S e l f -p o l l i n a t i o n e f f e cts on seed and seedling t r a i t s on noble f i r . For. S c i . 22:155-159. Squillace, A.E. 1974. Average genetic correlations among offs p r i n g from open-pollinated forest trees. Silvae Genet. 23:149-156. Squillace, A.E. and B.F. Swinde.l. 1986. Linkage among genes c o n t r o l l i n g monoterpene constituent l e v e l s in l o b l o l l y pine. For.. S c i . 32:97-112. Sorensen, Sorensen, Sorensen, Sorensen, Sorensen, 9 199 Squillace, A.E., 0.0. Wells and D.L. Rockwood. 1980. Inheritance of monoterpene composition i n c o r t i c a l oleoresin of l o b l o l l y pine. Silvae Genet. 29:141-151. Steel, R.G.D. and J.H. Torrie. 1980.. P r i n c i p l e s and procedures of s t a t i s t i c s . 2nd ed. McGraw-Hill, New York. Steinhoff, R.J., D.G. Joyce and L. Fins. 1983. Isoyzme va r i a t i o n i n Pinus monticola. Can. J. For. Res. 13:1122-1131. Stewart, S.C. and D.J. Schoen. 1986. Segregation at enzyme l o c i i n megagametophytes of white spruce, Picea glauca. Can. J. Genet. Cytol. 28:149-153. Stoehr, M.V. and R.E. Farmer, Jr. 1986. Genetic and environmental variance in cone size, seed y i e l d and germination properties of black spruce clones. Can. J. For. Res. 16:1149-1151. Strauss, S.H. and M.T. Conkle. 1986. Segregation, linkage and d i v e r s i t y of allozymes in knobcone pine. Theor. Appl. Genet. 72:4 83-4 93. Stuber, C.W., M.D. Edwards and J. F. Wendel. 1987. Molecular-marker f a c i l i t a t e d investigations of q u a n t i t a t i v e - t r a i t l o c i in maize. II. Factors influencing y i e l d and i t s component t r a i t s . Crop S c i . 27:639-648. Surles, S.E., J.L. Hamrick and B.C. Bongarten. 1989. Allozyme v a r i a t i o n in black locust (Robinia pseudoacacia) . Can. J. For. Res. 19:471-479. Sweet, G.B. and P.F. Waring. 1966. The r e l a t i v e growth rates of large and small seedlings in forest tree species. In: 6th discussion meeting of the society of foresters of Great B r i t a i n on physiology i n forestry. Forestry Suppl. pp.110-117. Tanksley, S.D., H. Medina-Filhio and CM. Rick. 1982. Use of naturally-occurring enzyme variations to detect and map genes c o n t r o l l i n g quantitative t r a i t s i n an i n t e r s p e c i f i c backcross of tomato. Heredity 49:11-25. Tanksley, S.D. and J. Hewitt. 1988. Use of molecular markers i n breeding for soluble s o l i d s content i n tomato - a re-examination. Theor. Appl. Genet. 75:811-823. 200 Thompson, P.A. 1981. Ecological aspects of seed germination. In: Adv. i n Res. and Technol. of Seeds. Part 6. Centre for A g r i c u l t u r a l Publishing and Documentation, Wageningen, The Netherlands, pp. 9-42. Thornburgh, D.A. 1969. Dynamics of true f i r - hemlock forests of western Washington. Ph.D. thesis, Univ. of Wash. Seattle, WA. 210pp. Tigerstedt, P.M.A. 1973. Studies on isozyme v a r i a t i o n i n marginal and central populations of Picea abies. Hereditas 75:47-60. Va l l e j o s , C E . and S.D. Tanksley. 1983. Segregation of isozyme markers and cold tolerance i n an i n t e r s p e c i f i c backcross of tomato. Theor. Appl. Genet. 66:241-247. Vaquero, F., F.J. Vences, P. Garcia, L. Ramirez and M. Perez de l a Vega. 1989. Mating system i n rye: v a r i a b i l i t y i n r e l a t i o n to the population and plant density. Heredity 62:17-26. von Rudloff, E. and R.S. Hunt. 1977. Chemotaxonomic studies i n the genus Abies. I I I . Leaf and twig o i l analysis of amabilis f i r . Can. J. Bot. 55:3087-3092. Wang, B.S.P. 1981. Measuring quality of tree seed. In: High-Quality C o l l e c t i o n and Production of Conifer Seed (R.F. Huber, Compiler). Can. For Serv. NFRC Edmonton, Alberta. Inf. Rep. NOR-X-235. pp. 73-78. Wang, B.S.P., J.A. P i t e l and D.P. Webb. 1982. Environmental and genetic factors a f f e c t i n g tree and shrub seeds. In: Adv. in Res. and Technol. of Seeds, pt. 7. Centre for Agric. Publ. and Doc, Wageningen, The Netherlands. pp. 87-135. Wang, C.W. and R.K. Patel. 1974. Variation i n seed c h a r a c t e r i s t i c s and seedling growth of open-p o l l i n a t e d ponderosa pine.progenies. Univ. of Idaho, C o l l . of Forestry, W i l d l i f e and Range S c i . Sta. Pap. No. 15. 10pp. Ward, R.H. and C F . Sing. 1970. A consideration of the power of the % 2 test to detect inbreeding e f f e c t s i n natural populations. Amer. Nat. 104:355-363. Wehrhahn, C. a n d R.W. A l l a r d . 1965. The d e t e c t i o n a n d measurement o f t h e e f f e c t s o f i n d i v i d u a l g e n e s 201 involved i n the inheritance of a quantitative character i n wheat. Genetics 51:109-119. Weir, B.S. and C.C. Cockerham. 1984. Estimating F-s t a t i s t i c s for the analysis of population structure. Evolution 38:1358-1370. Wilkinson, L. 1988. SYSTAT: The System for S t a t i s t i c s . Evanston, IL:SYSTAT, Inc. 1988. Witter, M.S. and P.P. Feret. 1978. Inheritance of glutamate oxalo-acetate transaminase isozymes i n V i r g i n i a pine megagametophytes. Silvae Genet. 27:129-134. Woods, J.H. and J.C. Heaman. 1989. Effe c t of d i f f e r e n t inbreeding levels on f i l l e d seed production i n Douglas-fir. Can. J. For. Res. 19:54-59. Worrall, J. 1983. Temperature-bud burst relationships i n amabilis and subalpine f i r provenance tests r e p l i c a t e d at d i f f e r e n t elevations. Silvae Genet. 32:203-209. Wright, J.W. 1976. Introduction to Forest Genetics. Academic Press, N.Y. Wright, S. 1969. Evolution and the genetics of populations, Vol. 2. The theory of gene •frequencies. Univ. of Chicago Press, Chicago. Wright, S. 1978. Evolution and the genetics of populations, Vol. 4. Univ. of Chicago Press, Chicago. Yeh, F.C. and K. Morgan. 1987. Mating system and multilocus associations in a natural population of Pseudotsuga menziesii (Mirb.) Franco. Theor. Appl. Genet. 73:799-808. Yeh, F.C. and S. Rasmussen. 1985. H e r i t a b i l i t y of height growth i n 10-year old Sitka spruce. Can. J. Genet. Cytol. 27:729-734. Zar, J.H. 1984. B i o s t a t i s t i c a l analysis (2nd ed.) Prentice H a l l . Englewood C l i f f s , N.J. Zavarin, E., K. Snajberk and W.B. C r i t c h f i e l d . 1973. Monoterpene v a r i a b i l i t y of Abies amabilis c o r t i c a l r e s i n . Biochem. Syst. 1:87-93. Common Name Appendix 1 S c i e n t i f i c Name A l a s k a y e l l o w - c e d a r Balsam f i r Bishop p i n e B l a c k s p r u c e B l u e s p r u c e D o u g l a s - f i r Engelmann s p r u c e European beech F r a s e r f i r Grand f i r Green a l d e r I n c e n s e c e d a r J a c k p i n e Knobcone p i n e L o b l o l l y p i n e Lodgepole p i n e M o u n t a i n hemlock Noble f i r Norway s p r u c e P i t c h p i n e Ponderosa p i n e S i t k a s p r u c e S u b a l p i n e f i r Tamarack V i r g i n i a p i n e Western hemlock Western r e d ce d a r Western w h i t e p i n e White f i r White s p r u c e W h i t e b a r k p i n e Y e l l o w - p o p l a r Chamaecyparis nootkatensis D. Don Abies balsamea (L.) M i l l . Pinus muricata D. Don Picea mariana M i l l . Picea pungens Engelm. Pseudotsuga menziesii (Mirb.) F r a n c o Picea engelmannii ( P e r r y ) Engelm. Fagus s y l v a t i c a L. Abies f r a s e r i (Pursh.) P o i r . Abies grandis (Dougl.) L i n d l . Alnus crispa ( A i t . ) P u r s h . Calocedrus decurrens (Torr.) F l o r i n Pinus banksiana Lamb. Pinus attenuata Lemm. Pinus taeda L. Pinus contorta v a r . l a t i f o l i a Englem. Tsuga mertensiana (Bong.) C a r r . Abies procera Rehd. Picea abies (L.) K a r s t Pinus r i g i d a M i l l . Pinus ponderosa Laws. Picea s i t c h e n s i s (Bong.) C a r r . Abies lasiocarpa (Hook.) N u t t . Larix l a r i c i n a (Du Roi) K. Koch. Pinus v i r g i n i a n a M i l l . Tsuga heterophylla (Raf.) S a r g . Thuja p l i c a t a Donn Pinus monticola Dougl. A b i e s concolor (Gord. & Glend.) L i n d l . Picea glauca (Moench) Voss Pinus a l b i c a u l i s Engelm. Liriodendron tulipifer'a L. 203 APPENDIX 2: Buffers, solutions, gel preparation, running conditions and stain recipes for horizontal starch gel electrophoresis. i ) Buffer System ' A ' ; Electrode Buffer (Tris -Citrate pH 7.0): 62.97 g TRISMA base (Sigma Chemical Co . , St. Louis, MO). 33.04 g c i t r i c acid Dissolve up to 4L with deionized water; run at 120 mA and 120 volts for 4 hr. Gel preparation; 60 mL electrode buffer 540 mL deionized water 50 g Electrostarch lot 392 (Electrostarch Co . , Madison, WI) 10 g hydrolyzed potato starch (Sigma) 20 g hydrolyzed starch (Connaught Labs. , Willowdale, Ont.) Mix starches with 150 mL solution to form lump-free s lurry; heat remaining solution to boiling and add a l l at once to s lurry . Cook and remove a ir using vacuum apparatus. Makes two gels. i i ) Buffer System ' B ' ; Electrode Buffer (Sodium Borate pH 8.0) 8.0 g sodium hydroxide 37.2 g boric oxide T i t r a t e with AN NaOH to pH 8.0; dissolve up to 4L with deionized water. Run at 120 mA and 100 volts for 4 hr. Gel preparation: 200 mL Tr i s -c i t ra te (pH 8.8)* 400 mL deionized water 35 g Electrostarch 40 g hydrolyzed potato starch (Sigma) Mix as instructed for gels i n 'A' above. *Tr i s - c i t ra te (pH 8.8) 48.4 g TRISMA base (Sigma) Ti trate with 0.2M c i t r i c acid solution to pH 8.8; dissolve up to 4L with deionized water. 204 i i i ) Stain Buffer; 96.88 g TRISMA base (Sigma) T i t r a t e to pH 8.0 with concentrated HC1; dissolve up to 4L with deionized water. i v ) Extraction Buffer; 10 mL T-C electrode buffer (pH 7.0) 80 mL deionized water 5 mL NADP sol u t i o n 5 mL NAD solution 0.018 g ascorbic acid 0.034 g EDTA 0.100 g bovine serum albumin 5 drops 2-mercaptoethanol (to bind phenolics) Solutions; ( f o r extraction buffer and ind i v i d u a l enzyme staining): Solution (aqueous) Pyridoxal 5' phosphate Glucose 1,6 diphosphate MTT MgCl 2 NADP NAD PMS NBT Malic acid G6PDH Concentration 1 mg/mL 0.1 mg/mL 10 mg/mL 10 mg/mL 10 mg/mL 10 mg/mL 5 mg/mL 10 mg/mL 0.5 M (to pH 7.0 with 10N NaOH) 1000 /20 mL T-C electrode buffer (pH 7.0) and 80 mL deionized water + 5.0 mL NADP + 0.5 mL NAD Stain recipes; AAT (buffer system 'B' ); 50 mL Tris-HCl s t a i n buffer 200 mg L-aspartic acid 100 mg -ketoghitoric acid 200 mg Fast Blue BB s a l t 1 mL pyridoxal 5' phosphate solution Mix i n tray i n above order. 205 G6P (buffer system 'B'): 50 mL Tris-HCl stain buffer 200 mg glucose-6-phosphate 1 mL NADP solution 1 mL MgCl2 solution 1 mL MTT solution 1 mL PMS solution Mix In tray i n above order. GDH (and SOD - buffer system 'B*): 50 mL Trls-HCl stain buffer 400 mg L-glutamic acid 1 mL NAD solution 1 mL MTT solution 1 mL PMS solution Mix In tray In above order (SOD appears as clear l i n e on blue stained g e l ) . IDH (and PGM - buffer system 'A'): 50 mL Trls-HCl stain buffer 100 mg l s o c l t r l c acid 100 mg glucose-l-phosphate 0.5 mL glucose 1,6 diphosphate solution 3 mL G6PDH solution 1 mL NADP solution 1 mL MgCl2 solution 1 mL MTT solution 1 mL PMS solution Mix i n tray i n above order; PGM resolves most anodally and without Interference with IDH. MDH (buffer system 'A'): 25 mL Tris-HCl stain buffer 25 mL DL-malic acid (pH 7.0) 1 mL NAD solution 1 mL NBT solution 1 mL PMS solution Mix i n tray In above order. 206 6PG (buffer system 'B'): 5 mL Tris-HCl s t a i n buffer 10 mg phosphogluconic acid 1 mL NADP solution 1 mL MgCl2 solution 1 mL MTT solution 0.5 mL PMS solution Mix in above order i n amber flask; apply dropwl6e to g e l . PGI (buffer system ' A ' ) : 50 mL Tris-HCl s t a i n buffer 25 mL fructose-6-phosphate 1 mL G6PDH solution 1 mL NADP solution 1 mL MgCL£ solution 1 mL MTT solution 1 mL PMS solution Mix i n tray i n above order. Incubate a l l enzyme stains at 37* for 15 to 30 min. 207 APPENDIX 3 Multiple regression equations involving l a t i t u d e (LAT), longitude (LONG) and elevation (ELEV) as predictor variables (associated standard errors printed below c o e f f i c i e n t s ; * s i g n i f i c a n t , P < 0.05): 1. Enzyme heterozygosity (R2 = 0.22*): ENZHET = -1.74 - 0.229 (LAT)* + 0.105(LONG)* + 0.0004 (ELEV) N s (2.99) (0.094) (0.052) ' (0.0002) 2. Cone rachis length (R2 = 0.64*): RACHLEN = -4 96.1 - 69.2(LAT)* + 31.8(LONG)* + 0.031(ELEV) N S (452.5) (13.9) (7.7) (0.026) 3. Seed weight (R2 = 0.63*): SDWT = 2.45 - 3.67 (LAT)* + 1.45 (LONG)* + 0.0012 (ELEV) N s (32.0) (1.01) (0.56) (0.0018) 4. Cotyledon number (R2 = 0.48*): COTY = 7.79 - 0.387 (LAT) N S + 0.128 (LONG)Ns + 0.0004 (ELEV) N S (8.60) (0.271) (0.149) (0.0005) 5. Seedling shoot weight at 29 wk (R2 = 0.13 N S): SHWT = 4.87 - 0.309 (LAT) N S + 0.097 (LONG) N s - 0 . 0003 (ELEV) N S (6.73) (0.212) (0.117) (0.0004) Seedling root weight at 29 wk (R2 = 0.29 N S): RTWT = 8.18 - 0.028 (LAT) N S - 0.044 (LONG)NS - 0.0003 (ELEV) (2.50) (0.079) (0.043) (0.0002) 7. Seedling diameter at 29 wk (R2 = 0.13 N S): DIAM = 8.20 -(3.54) 0.04 6 (LAT) NS -(0.112) 0.028 (LONG) N S -(0.062) 0.0003 (ELEV) N s (0.0002) 

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.831.1-0100310/manifest

Comment

Related Items