UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Densitometric studies on the wood of young coastal Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) Cown, David John 1976

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

Item Metadata

Download

Media
831-UBC_1976_A1 C69.pdf [ 15.69MB ]
Metadata
JSON: 831-1.0075135.json
JSON-LD: 831-1.0075135-ld.json
RDF/XML (Pretty): 831-1.0075135-rdf.xml
RDF/JSON: 831-1.0075135-rdf.json
Turtle: 831-1.0075135-turtle.txt
N-Triples: 831-1.0075135-rdf-ntriples.txt
Original Record: 831-1.0075135-source.json
Full Text
831-1.0075135-fulltext.txt
Citation
831-1.0075135.ris

Full Text

DENSTTOMETRIC STUDIES ON THE WOOD OF YOUNG COASTAL DOUGLAS-FIR (Pseudotsuga menziesii (Mirb.) Franco). by DAVID JOHN COWN B.Sc. (Hons.), Aberdeen, 1967 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of Forestry We accept t h i s t h e s i s as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA June, 1976 David John Cown, 1 9 7 6 In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h C o l u m b i a , I ag ree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . It i s u n d e r s t o o d that c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department o f The U n i v e r s i t y o f B r i t i s h Co lumbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 ABSTRACT X-ray densitometry was used to investigate some environmental and genetic influences on wood formation i n young Douglas-fir trees growing i n the P a c i f i c Northwest. Sampling methods were determined from s e c t i o n a l analyses of even-aged stems and eleven growth-ring width and density v a r i a b l e s were measured i n several t r i a l s to elucidate' some of the causes of v a r i a t i o n . Breast height increment core samples were shown to give-va good represen-r t a t i o n of stand properties and juvenile-mature c o r r e l a t i o n s f or a 40-year-old crop confirmed the v a l i d i t y of determining i n t r a - r i n g density parameters on young material. Increment core samples from the Co-operative Douglas-fir Provenance Test (5!.locations; 5 provenances/location) were used for both gravimetric and densi-tometric analyses. The major components of v a r i a t i o n were found to be stand l o c a t i o n and the i n d i v i d u a l tree e f f e c t . Provenances c o n s i s t e n t l y accounted for l e s s than 5% of the t o t a l v a r i a t i o n . Genotype-environment i n t e r a c t i o n was shown to be small f o r a l l properties measured with the exception of the i n t r a - r i n g density range. Between s i t e s , earlywood widths were more v a r i a b l e than latewood widths, but latewood density properties (mean latewood and r i n g maximum densi t i e s ) were morevariable than those for earlywood (mean earlywood and r i n g minimum d e n s i t i e s ) . Earlywood and latewood parameters varied indep-endently of one another. It was suggested that genetic c o n t r o l was weak at the provenance l e v e l but strong for i n d i v i d u a l trees. Regression analyses using growth-ring components and monthly weather data f o r the outer f i v e rings at each s i t e uncovered some highly s i g n i f i c a n t e f f e c t s which helped to explain the observed year-to-year v a r i a t i o n . Density v a r i a b l e s were l e s s affected by weather conditions than earlywood and latewood widths. An examination of eight ramets from each of ten 13-year-old clones revealed highly s i g n i f i c a n t d i f f e r e n c e s i n a l l eleven i n t r a - r i n g parameters. H e r i t a b i l i t y estimates f o r i n d i v i d u a l growth rings showed a regular increase with tree age, and latewood properties (width, density and r i n g maximum density) were found to be under strong genetic c o n t r o l . On a clone-mean basis, density was not related s i g n i f i c a n t l y to growth rate, so that vigour and density properties could be selected f o r independently. Crown phenology (flu s h i n g and shoot growth c h a r a c t e r i s t i c s ) was not strongly correlated with growth-ring parameters, although there was an i n d i c a t i o n that early f l u s h i n g may be associated with higher latewood density. A study of four parent trees and t h e i r c o n t r o l - p o l l i n a t e d progeny proved unsatisfactory due to lack of adequate r e p l i c a t i o n and a t y p i c a l stand cond-i t i o n s , but nevertheless provided a v e h i c l e f o r discussion of problems i n -volved i n assessing plus-tree wood q u a l i t y and narrow-sense h e r i t a b i l i t i e s . The combined r e s u l t s were discussed i n terms of the genecology of Douglas f i r and the implications f o r forest management and u t i l i z a t i o n . iv TABLE OF CONTENTS Page ABSTRACT i i TABLE OF CONTENTS i v LIST OP TABLES v i i LIST OF FIGURES v i i i LIST OP APPENDICES -\ x ACKNOWLEDGEMENTS x i i Chapter 1: INTRODUCTION 1 Chapter 2: LITERATURE REVIEW 5 2.1 Natural variation in wood density 5 2.1.1 Variation within and between growth rings 5 2.1.2 Variation within and between geographic regions 9 2.2 Relationships between growth-ring components 14 2.3 Effects of environmental factors 18 2.4 Inheritance of ring-width and density components 24 2.4.1 Measurement of genetic control 25 2.4.2 Juvenile-mature correlations 28 2.4.3 Genetic studies of Douglas-fir wood density 30 2.5 Crown phenology and wood properties 33 2.6 X-ray densitometry 39 Chapter 3: MATERIALS AND METHODS 42 3.1 Preliminary studies 42 3.2 Co-operative Douglas-fir Provenance Test 46 3.2.1 Sample selection 47 3.2.2 Gravimetric analyses 51 3.2.3 Densitometric analyses 52 3.2.4 Climatological analyses 53 \ V Page 3.3 Clonal study 54 3.3.1 Sample s e l e c t i o n 3.3.2 Analyses 57 3.4 D i a l l e l cross study 57 3.5 Densitometric techniques . 59 Chapter 4:> RESULTS AND DISCUSSION - PRELIMINARY STUDIES 64 4.1 Between-tree v a r i a t i o n i n breast height wood density 64 4.2 Relationship between breast height and whole-tree density 65 4.3 Within-tree v a r i a t i o n i n t r e e - r i n g components 66 4.4 Juvenile-mature c o r r e l a t i o n s . 76 Chapter 5: RESULTS AND DISCUSSION - PROVENANCE TRIAL 82 5.1 Gravimetric analyses 82 5.1.1 S i t e and provenance e f f e c t s 82 5.1.2 Relationship between wood density and growth rate 88 5.2 Densitometric analyses 98 5.2.1 Ring width components 98 5.2.2 Ring density components 104 5.2.3 C o r r e l a t i o n between r i n g components 114 5.2.4 Influence of seasonal weather v a r i a t i o n 117 Chapter 6: RESULTS AND DISCUSSION - CLONAL STUDY 130 6.1 Gravimetric analyses 130 6.2 Densitometric analyses 133 6.3 Phenological observations 149 6.3.1 Clonal d i f f e r e n c e s 149 6.3.2 Relationships between phenological and growth- 158 r i n g c h a r a c t e r i s t i c s . 6.4 Influence of water " d e f i c i t during_1972 162 Chapter.7: RESULTS AND DISCUSSION - DIALLEL CROSS STUDY 166 7.1 Parent tree ring-component c h a r a c t e r i s t i c s 166 7.2 Progeny ring-component c h a r a c t e r i s t i c s 170 7.3 Parent-progeny r e l a t i o n s h i p s 175 v i Page Chapter 8: GENERAL SUMMARY, DISCUSSION AND CONCLUSIONS 184 8.1 Recapitulation 184 8.2 Discussion and conclusions 185 8.3 Recommendations for further research 193 REFERENCES 195 APPENDICES 208 v i i LIST OF TABLES Table Page 3.1 Selected sites and seed sources. 49 4.1 Summary of ring components by stem levels. 72 4.2 Summary ofinmean ring components for 20 U.B.C. Campus trees. 77 4.3 Coefficients of determination f o r regression analyses of juvenile- 78 mature relationships in 40-year-old trees. 5.1 Provenance t r i a l variance components. 85 5.2 Summary of provenance t r i a l site analyses. 87 5.3 Analyses of variance for individual sites. 101 5.4 Ranking of provenances by RNG values. 108 5.5 Site densities predicted from latewood percentage values. 118 5.6 Provenance t r i a l variance components as a percentage of the total 120 variance. 6.1 Simple linear regressions of wood density on mean ring width for 132 clonal cores. 6.2 Heritability estimates for clonal ring components. 139 6.3 Predicted gains from 10% clonal selection. 146 6.4 Phenological and density variables used in s t a t i s t i c a l analyses. 155 6.5 Duncan's Multiple Range tests on clonal means for phenological and 156 growth-ring variables. 6.6 Coefficients of determination for ring component/crown property 160 relationships. 6.7 Mean clonal ring components, 1969-73. 165 7/.1 Results of Duncan's Multiple Range tests on progeny ring components. 172 v i i i LIST OF FIGURES Figure Page 3.1 Sampling scheme for individual trees. 44 3.2 Sampling scheme for provenance replications. 50 3.3 Definition of growth-ring components. 63 4.1 Relationship between breast height increment core density and 67 weighted whole-tree density. 4.2 Ring component patterns within the sample trees. 69 4.3 Radial patterns of width-component variation in 40-year-old stems. 74 4.4 Radial patterns of density-component variation in 40-year-old stems. 75 5.1 Summary of provenance t r i a l diameter data. 83 5.2 Summary of provenance t r i a l increment core density data. 84 5.3 Summary of provenance t r i a l plantation data. 89 5.4 Increment core density distributions by sites. 90 5.5 Wood density/growth rate relationship - a l l sites. 92 5.6 Wood density/growth rate relationships by sites. 93 5.7 Wood density/growth rate relationships by provenances. 94 5.8 Linear and curvilinear regressions for the wood density/growth rate 96 relationship. 5.9 Ring-width components by sites. 100 5.10 Densitometric profiles by sites. 105 5.11 Standard densigrams by sites. 106 5.12'Standard densigrams by provenances and sites, 1969-73. 110 5.13 Variance components as a percentage of the total variance. 112 5.14 Relationships between mean ring density and a) mean ring width, and 116 b) percent latewood. i x Figure Page 5.15 Percentage change i n r i n g components associated with one standard 122 deviation change i n monthly weather data f o r ; -a) earlywood components c) whole-ring properties b) latewood components d) minimum and maximum d e n s i t i e s 5.16 Average temperature and p r e c i p i t a t i o n for the provenance t r i a l s i t e s , 129 1969-74. 6.1 Clonal whole-core ring-width and basic density data. 131 6.2 Clonal densitometric p r o f i l e s . 134 6.3 Standard densigrams by clones, 1969-73. 136 6.4 H e r i t a b i l i t y estimates by growth periods. 137 6.5 Clonal core d e n s i t i e s by growth zones. 141 6.6 Clonal core latewood percentages by growth zones. 142 6.7 Clonal variance components as a percentage of the between-ramet 143 component. 6.8 Cumulative shoot and r a d i a l growth by clones, 1975. 150 6.9 Span of cambial and phenological events by clones. 151 6.10 Progress of c l o n a l f l u s h i n g c h a r a c t e r i s t i c s . 152 6.11 Clonal densigrams and r i n g p r o f i l e s for the 1975 increment. 154 6.12 Weekly temperature and p r e c i p i t a t i o n data for the 1975 growing season. 164 7.1 Results of Duncan's Mult i p l e Range tes t s on parent tree outerwood 167 r i n g component data. 7.2 Parent tree corewood and outerwood r i n g p r o f i l e s . 171 7.3 Progeny p r o f i l e summaries - rings 6-10. 178 7.4 S i m i l a r i t y indices for parent trees - rings 6-10. 180 7.5 Parent-progeny s i m i l a r i t y indices - rings 6-10. 181 X LIST OF APPENDICES Appendix Page 1: Natural distribution of Douglas-fir (a), and location of provenance 209 t r i a l seed origins and plantations (b). 2: Heritability estimates for growth-ring components. 210 3: Sources of weather data. 212 4: Data on clone origins. 213 5: Flushing groups used in f i e l d assessment. 214 6: Data on U.B.C. Campus parent trees; '.'.: a) location of four selected Douglas-fir trees on U.B.C. Campus, 215 b) parent-tree data (1975)^, 216 c) sample numbers for the d i a l l e l cross. 216 7: Summary of density data for the felled sample trees. 217 8: Correlation matrix for felled-tree density values. 218 9: Provenance t r i a l increment core data - a) diameter summary, 219 b) wood density summary. 220 10: Analyses of variance models; a) combined sites (gravimetric samples) 221 b) individual sites ( » " ) 221 c) " " (densitometric samples) 222 d) combined sites ( •• ) 222 e) r.anova model for individual sites incorporating the year effect. 223 11: Coefficients of determination for the density/growth rate relationship 224 12: Summary of ring component values by sites and provenances; a) rings 1-5, 225 b) rings 1970-74. 226 13: Correlation matrices for provenance ring components; a) rings 1-5, 227 b) rings 1970-74. 227 14: Regression stat i s t i c s by sites; a) density/ring width, 228 b) density/percentage latewood. 228 x i Appendix (contd.) Page 15: Regression statistics for the effects of climatic variables on ring components,-a) and b) monthly precipitation, 229 c) and d) monthly temperature - mean, 230 e) and f) "" " - minimum 231 g) and h) " " - maximum 232 16: Climatic summaries for the provenance t r i a l sites, 1969-74. 233 17: Ring component summaries by clones and ring groups. 234 18: Ranges and variance components for clonal data. 235 19: Results of Duncan's Multiple Range tests on clonal ring components; a) rings 1-5, 236 b) rings 1969-73. 237 20: Analyses of variance models for clonal data; a) rings 1-5 (mixed model) 238 b) rings 1969-73 (random model) 238 21: Correlation matrices for clonal ring components; a) rings 1-5, 239 b) rings 1969-73 239 22:' Correlation matrix for phenological and growth-ring variables. 240 23: Correlation matrices for progeny ring components; a) rings 1-5, 241 b) rings 1970-74. 241 x i i ACKNOWLEDGEMENTS Thanks are due to many who helped at various stages of t h i s work. In p a r t i c u l a r , I wish to acknowledge the assistance and encouragement received from Dr. K.K. Ching, O.S.U. School of Forestry, both at the planning stage and i n l a t e r discussions. Of the committee members, s p e c i a l thanks are due to Dr. R.W. Kennedy and Mr. M.L. Parker of the Western Forest Products Laboratory for unlimited use of s p e c i a l i z e d equipment and materials. The B.C. Forest Service was also h e l p f u l i n allowing access to the clone bank at Cowichan Lake F i e l d Experiment Station, where the assistance of Chris Heamon and Ingemar Karlsson was much appreciated. F i e l d assistance with tedious increment core c o l l e c t i o n s was supplied at various times by Stan D i l l (O.S.U.) and graduate students Keith Mackie and Dave E l l i o t t . Comments and c r i t i c i s m on s t a t i s t i c a l techniques were f r e e l y given by Drs. A. Kozak and W. Warren (W.F.P.L.). Les Jozsa (W.F.P.L.) also devoted much time to my queries on densitometric methods, and prepared an a r t i s t i c i l l u s t r a t i o n f o r the t h e s i s . Typing was s k i l l f u l l y undertaken by Mrs. Marianne Moore. F i n a n c i a l support for my stay i n Canada was supplied both by the New Zealand Forest Service and by the Faculty of Forestry, USB.C. - 1 -Densitometric Studies on the Wood of Young Coastal Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) Chapter I. INTRODUCTION Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) i s the p r i n c i p a l f o r e s t species of the P a c i f i c Northwest region i n terms of i t s importance to the f o r e s t industry. Its natural d i s t r i b u t i o n i s one of the widest of the gymnosperms, extending i n l a t i t u d e from about the 19th to 55th p a r a l l e l s and i n width from the P a c i f i c coast to western Alberta i n the north and inland through Arizona, New Mexico and Texas i n the south (Appendix 1). Within t h i s range, s i t e a l t i t u d e can vary from sea l e v e l to over 10,000 feet and r a i n f a l l from about 18 inches to over 100 inches per year. Due to the magnitude of the v a r i a t i o n s i n geographic and c l i m a t i c conditions, morphological and p h y s i o l o g i c a l differences have been noted which have re s u l t e d i n the sub-division of the species i n t o more or lees d i s t i n c t types. Two main v a r i e t a l forms are recognized, namely the coast, var. menz-i e s i i , or green Douglas-fir and the i n t e r i o r , var. glauca (Beissn.) Franco, or blue Rocky Mountain Douglas-fir. Other intermediate types may occur l o c a l l y (Haddock, 1967). The conditions for optimum growth of D o u g l a s f f i r occur west of the Cascade mountains i n Washington and Oregon, i n which area over 60% of the standing volume e x i s t s (McArdle et at. , 1961). I t i s also one of the most highly p r i z e d timber species along the southern coastal areas of B r i t i s h Columbia. The p r e v a i l i n g maritime conditions i n these areas tend to moder-ate the usual l a t i t u d i n a l temperature gradients i n such a way that c l i m a t i c v a r i a t i o n i s more c l o s e l y associated with elevation than with l a t i t u d e (Lassen and Okkonen, 1969). - 2 -The timber of Douglas-fir i s much sought a f t e r f o r s t r u c t u r a l uses, pulp, and veneer, but q u a l i t y i s known to vary considerably between growing regions, as has been shown by a number of studies CKnigge, 1962; U.S. Forest Service> 1965; Snodgrass and Noskowiakj. 1968 ). i t has been described as being t y p i c a l l y straight-grained, moderately l i g h t to moderately heavy ( s p e c i f i c g ravity 0.43 to 0.45), and of intermediate d u r a b i l i t y . One of the c h a r a c t e r i s t i c features of the wood i s the strong contrast i n colour and texture between the earlywood and latewood zones which contributes to the f i g u r e i n veneer. During the past 15 years or so wood s c i e n t i s t s and tree breeders around the world have expressed increasing i n t e r e s t i n the i n c l u s i o n of wood pro-p e r t i e s i n tree improvement programmes (Zobel, 1964; Har r i s , 1970; Einspahr, 1972). This i s a l o g i c a l extension of the forest e r ' s d e s i r e to grow the most p r o f i t a b l e and desi r a b l e types of trees i n any given area. S e l e c t i o n of species, provenances and even i n d i v i d u a l trees f o r propagation on the basis of growth rate, stem form and disease resistance has been p r a c t i c e d for some time, and i t i s natural to seek methods by whichtthe wood charac-t e r i s t i c s themselves can be modified. Discussion of wood q u a l i t y often centres around the s e l e c t i o n of appropriate indices since many properties can be shown to influence s p e c i f i c end uses. However, the r e s u l t s of such research over a wide v a r i e t y of si t u a t i o n s have determined that wood s p e c i f i c g ravity (increasingly being r e f e r r e d to as wood density) i s by f a r the most widely accepted general i n d i c a t o r of q u a l i t y . For example, i t i s c l o s e l y c o r r e l a t e d with the major strength properties of timber, often i n an exponential fashion (Panshin and de Zeeuw, 1970) and l i n e a r l y r e l a t e d to pulp y i e l d ( M i t c h e l l , 1963). In - 3 -addition, pulp and paper q u a l i t y can be influenced by v a r i a t i o n s i n density i n so f a r as high density within one species i s normally associated with a high proportion of latewood with longer and l e s s conformable tracheids (van Buijtenen, 1964). Other properties a f f e c t e d by v a r i a t i o n i n density are machining, glueing and f i n i s h i n g c h a r a c t e r i s t i c s . Industry may be i n t e r e s t e d i n a l t e r i n g t h i s property e i t h e r upwards or downwards (van Buijtenen, 1962; Einspahr, 1972). Several studies have confirmed that wood density i n Douglas-fir, as i n other species, i s a highly v a r i a b l e property, being influenced by both tree age and growth rate as well as other l e s s well defined f a c t o r s . Within trees, density appears to increase outwards from the p i t h for at l e a s t 100 years i n natural stands (U.S. Forest Service, 1965) and a s i m i l a r trend has been shown i n planted trees (Harris and Orman, 1958; McKimmy, 1966). Cu r t i s et al.- (1974) described the progression i n the P a c i f i c Northwest from harvesting of v i r g i n stands to management of'second growth, o r i g i n a t i n g i n a haphazard manner, to modern methods of intensive f o r e s t r y . The l a t t e r approach involves crop establishment with g e n e t i c a l l y improved seedlings and c o n t r o l l e d spacing to maintain optimum growth. F e r t i l i z a t i o n and thinning are l i k e l y to become standard p r a c t i c e on a l l s i t e c l a s s e s , reducing economic rotations to between 30 and 60 years. A large proportion of the produce from such'stands w i l l - c o n s i s t of wood from the lower density inner core which i s generally regarded as having l e s s desirable properties than the more mature outerwood. Since the q u a l i t y of the f i n a l wood product i s often l a r g e l y dependent on the c h a r a c t e r i s t i c s of the raw material, i t i s important to examine the properties of plantation-grown timber to determine which are the factors - 4 -contributing most to the v a r i a t i o n and to what extent management prac-t i c e s can help to produce a higher q u a l i t y material base. The objective of the current study i s to determine the patterns of v a r i a t i o n of growth-ring width and density components i n young Douglas - r fir trees and to i s o l a t e some of the major factors a f f e c t i n g t h e i r v a r i a t i o n by use of X-ray densitometry (Parker and Kennedy, 1973). In view of the findings f o r other tree c h a r a c t e r i s t i c s , e.g.', height growth and "phenology, i t i s postulated that both l o c a l environment and seed source should c o n t r i -bute s i g n i f i c a n t l y to the observed patterns. I t i s also proposed to estab-l i s h estimates of h e r - i t a b i l i t y f o r the r i n g components using densitometric data from c l o n a l material and to t e s t some of the p r e v a i l i n g hypotheses regarding the relationships between tree crown phenology and wood formation. - 5 -Chapter 2. LITERATURE REVIEW 2.1 V a r i a t i o n i n -Tree-ring Components 2.1.1 V a r i a t i o n within and between growth rings Few aspects of wood structure have received as much attention as the sources of v a r i a t i o n of wood density. This i s p r i m a r i l y due to the strong a s s o c i a t i o n of density with many other wood properties and i s f a c i l i t a t e d by the ease of measurement. Most assessment methods use weight and volume data which can be obtained with the simplest of instruments. Basic wood density (oven-dry weight/saturated volume) i s a measure of the r e l a t i v e proportions of c e l l - w a l l substance and a i r i n wood and i s thus influenced by c e l l diameter, c e l l - w a l l thickness, and to a much l e s s e r extent i n coni-fe r s by the r e l a t i v e numbers of d i f f e r e n t c e l l types. Temperate zone coniferous xylem t i s s u e i s characterized by well-defined growth zones composed mainly of r a d i a l f i l e s of tracheids l a i d down during the course of one growing season. The regular organization of r i n g width patterns was extensively studied i n red pine (Pinus resinosa A i t . ) by Duff and Nolan (1953) and i n a s e r i e s of papers by Forward and Nolan (1962). These authors stressed the influence of crown development on wood formation. Later,'Larson (1969) u n i f i e d the'concept-of external influences regulating cambial a c t i v i t y through shoot growth and endogenous growth hormone concentrations. The dimensions of wood c e l l s vary i n a predictable manner, the pattern of which i s c o n t r o l l e d p a r t l y g e n e t i c a l l y (species differences) and p a r t l y environmentally (year-to-year v a r i a t i o n s ) . Wood formed i n the e a r l y part of the growing season i s t y p i f i e d by large diameter tracheids with r e l a t i v e l y - 6 -t h i n walls whereas towards the end of the growth period r a d i a l diameters decrease and wall thickness increases. Due to the uniformity of c o n i f e r wood t i s s u e these a l t e r a t i o n s i n c e l l dimensions are r e f l e c t e d i n an increase i n wood density across the growth r i n g . Macroscopically,these changes often appear as t e x t u r a l and colour d i f f e r e n c e s which can be used to delineate r i n g component boundaries. Many early researchers recognized the importance of these s t r u c t u r a l changes within growth increments and the d i f f e r e n t i a t i o n of springwood (now more appropriately c a l l e d earlywood) as opposed to summerwood ( l a t e -wood) became common p r a c t i c e i n wood q u a l i t y studies. The f i r s t and most widely used quantitative d e f i n i t i o n o f the earlywood-latewood boundary appears to be that of Mork (1928), who considered that latewood i n Norway spruce (Picea abies (L.) Karst.) s t a r t s where double the combined tangential c e l l - w a l l thickness i s equal to or greater than the r a d i a l lumen diameter. This c r i t e r i o n has remained the most popular i n microscopic work but suf f e r s the drawback of being exceedingly tedious i f large numbers of specimens have to be evaluated. I t has also been noted that t h i s d e f i n i -t i o n . f a i l s to recognize a latewood zone i n some European c o n i f e r s (Fry and Chalk,1957). Over the years, several other approaches have been used i n attempts to improve on Mork's d e f i n i t i o n using a v a r i e t y of techniques such as mechani-c a l probes (Marian and Stumbo, 1960), chemical stains (Wiksten, 1955), l i g h t transmission (Green and Worrall, 1964) and i o n i z i n g r a d i a t i o n ( P h i l l i p s et dl. , 1962; Polge, 1965; Harris, 1969). With the exception of the r a d i a t i o n methods, most schemes have not met with wide acceptance due to l i m i t a t i o n s both i n the techniques themselves and i n the concept of earlywood-latewood - 7 -d i f f e r e n t i a t i o n which i m p l i c i t l y suggests some degree of uniformity within the two types of t i s s u e . Gravimetric analyses of small samples ( I f j u , 1969) and r e s u l t s from scanning densitometry (Brazier, 1969) both show gradations i n properties within these zones and between trees such that f o r a given amount of latewood, wood density and the contrast between earlywood and latewood d e n s i t i e s can vary appreciably (Harris, 1967). In f a c t , Paul (1939) provided data for several southern pine species demonstra-t i n g v a r i a t i o n i n earlywood and latewood d e n s i t i e s and showed that e a r l y -wood i n some rings could be of higher density than the latewood i n others. The t e x t u r a l differences already mentioned i n regard to the growth rings of Douglas-fir a r i s e because of the large contrast i n density between earlywood and latewood i n comparison with many other species such as spruces, true f i r s and s o f t pines. Harris (1969) i l l u s t r a t e d d i f f e r e n t species patterns of intra-increment v a r i a t i o n and showed that Douglas-fir density 3 3 could range from less than 0.2 gi/cm to more than 0.8 g/cm within the same r i n g . Harris and Orman (1958) considered that the nature of the e a r l y -wood-latewood boundary i n t h i s species varied between wide rings and narrow rings to such an extent that the term ' t r a n s i t i o n latewood' should be applied to the intermediate type of wood predominant i n fast-grown material. They maintained that since slow-grown timber exhibits a rapid change from earlywood to latewood, the influence of such latewood on r i n g density i s greater than i n fast-grown wood where the proportion of l e s s dense t r a n s i -t i o n latewood i s high. Alexander (1935) had noted that second-growth Douglas-fir could have a high latewood percentage but s t i l l be of lower density than comparable o l d growth due to the lower latewood density. Regular patterns of wood density v a r i a t i o n have been found within the stems of many co n i f e r species (Turnbull, 1947; Zobel et at., 1959; Paul, - 8 -1963). In general, density increases outwards from the p i t h at a l l stem l e v e l s , the inner lower density zone being r e f e r r e d to as corewood (or sometimes juvenile wood). The extent of t h i s corewood i s known to vary between species and even between trees, but cannot be p r e c i s e l y defined as i t s c h a r a c t e r i s t i c s are not dependent s o l e l y on wood density (Cown, 1974a). The tendency i n pine species i s for a sharp increase i n density over the f i r s t 10 to 20 rings from the p i t h and thereafter a more or l e s s constant l e v e l . On the other hand, Douglas-fir tends to e x h i b i t a gradual increase f o r at l e a s t 100 rings (U.S. Forest Service, 1965). McKimmy (1966) measured the r a d i a l density trend i n 46-year-old planted trees at four l o c a t i o n s by gravimetric determinations on 5-ring samples. A f t e r an i n i t i a l decrease f o r "1.0-- rings or so, density increased outwards but tended to l e v e l o f f a f t e r about 20 r i n g s . The data of Smith (1956) showed that whereas earlywood density remained 3 constant at about 0.23 g/cm between rings 5-10 and 25-30 from the p i t h , 3 3 latewood density increased from 0.71 g../cm to 0.76 'g'./cm over the same distance. Harris and Orman (1958) found that excised earlywood and l a t e -samples gave v i r t u a l l y constant values a f t e r an i n i t i a l decrease over 5 to 10 rings . The r a d i a l increase i n mean r i n g density outwards was a t t r i b u t e d to both an increase i n the percentage of latewood and a decrease i n the amount of t r a n s i t i o n latewood. In a l a t e r p u b l i c a t i o n , Harris (1969) presented sum-maries of densitometric scans of Douglas-fir grown i n d i f f e r e n t l o c a l i t i e s showing that the major developmental trends i n minimum and maximum r i n g d e n s i t i e s occurred within f i v e to ten,rings from the p i t h . Another source of within-tree v a r i a t i o n , often overlooked i n sampling, i s that around the circumference of the stem. An examination of Douglas-fir stems grown i n New Zealand plantations revealed that t h i s could contribute - 9 -s i g n i f i c a n t l y to the observed v a r i a t i o n i n small wood samples (Cown, 1971a). I t i s 'often reported that mean density decreases upwards i n the stem, p r i m a r i l y as a r e s u l t of the increasing proportion of lower density core-wood (Panshin and de Zeeuw, 1970; Okkonen et al., 1972). Recognition of inherent trends i n r i n g component properties within stems has not always been apparent i n the l i t e r a t u r e , with the r e s u l t that comparatively recently a r t i c l e s have appeared urging that sampling be c a r r i e d out i n f u l l knowledge of the b i o l o g i c a l v a r i a t i o n present (Richard-son, 1961; Larson, 1962). 2.1.2 V a r i a t i o n between and within geographic areas Dating from the 189O'swhen R. Hartig studied Norway spruce i n Germany, the l i t e r a t u r e abounds with reports of v a r i a t i o n i n wood properties between and within s i t e s . Results vary considerably, depending on the species and s i t e studied, but the important point i s that i s has been con c l u s i v e l y established that l o c a l environment can, under c e r t a i n conditions, s i g n i f i - ; cantly a f f e c t wood density. At the same time, however, i t has long been recognized that differences between trees growing on apparently uniform s i t e s can be of such magnitude as to overshadow s i t e e f f e c t s (Markwardt, 1927; Koehler, 1939). Much of the e a r l i e s t systematic work on regional v a r i a t i o n was under-taken during the 1950's i n the southeastern states of the U.S.A., and r e s u l t s have been summarized by M i t c h e l l (1964) and Wahlgren and Schumann (1972). Several pine species showed an increase i n density from northwest to southeast throughout t h e i r natural range, associated with concomitant v a r i a t i o n i n percentage latewood. However, both Zobel and Rhodes (1955) and Perry and Wang (1958) stressed that between-tree v a r i a t i o n s could be j u s t as large or larger than between-region di f f e r e n c e s . Larson (1957) - 10 -studied the influence of environment on the density of slash pine (Pinus e t l i o t t i i Engelm. var. e t l i o t t i i ) and was unable to detect s i g n i f i c a n t e f f e c t s from growth rate, stand age, s i t e q u a l i t y or length of the growing season. Only those factors associated with s o i l moisture-holding capacity proved u s e f u l f o r p r e d i c t i o n of density. Koch (1972) concluded that the r e l a t i v e contributions of environmental and genetic factors remain l a r g e l y unresolved, while Wahlgren and Schumann (1972) conceded that environmental factors on the whole f a i l to explain much of the v a r i a t i o n i n southern pines and suggested that there must be a strong genetic component. Other more l i m i t e d studies i n North America and Europe have tended to confirm the trend of increasing density towards the equator ( E l l i o t t , 1970). Work on regional v a r i a t i o n i n the southern hemisphere i s l i m i t e d , but sur-veys of some exotic pine plantations i n New Zealand have revealed a d i s -t i n c t increase i n density from south to north (Harris, 1965, 1973; Cown, 1974b) . A l l of these l a t t e r studies were c a r r i e d out i n even-aged crops where increment core samples were c o l l e c t e d from 50 to 100 stems at each l o c a t i o n . The data show that v a r i a t i o n between trees at any one s i t e i s much greater than differences between s i t e s . Densitometric analyses of small subsamples of lodgepole pine (Pinus contovta Dougl.) and Corsican pine (Pinus nigra Arnold) revealed that r a d i a l patterns of density devel-opment (minimum r i n g density, mean r i n g density and maximum r i n g density) could vary appreciably between s i t e s . Examination of Corsican pine of the same seed l o t grown at four d i f f e r e n t locations showed strong environmental e f f e c t s , and i n f a c t the mean r i n g density l e v e l s recorded at the most northerly s i t e (37°S) were equivalent to the maximum r i n g d e n s i t i e s found at the most southerly s i t e (46°S) - a l a t i t u d i n a l separation of about 600 - 11 -miles (Cown, 1974b). In an extensive study of radiata pine (Pinus radiata D. Don), Harris (1965) found that mean density of the outerwood was very highly correlated with mean annual temperature i f nutrient deficient sites are disregarded (r = 0.87). Early studies on the strength properties of Douglas-fir resulted in the recognition of at least three broad areas, corresponding to the coast, intermediate and Rocky Mountain types (Markwardt and Wilson, 1935). In general, tifie timber characteristics of the coast type are substantially superior to those of the Rocky Mountain type and commercial Douglas-fir has been separated into these two categories. In terms of wood structure alone, there are no clear distinguishing features which allow easy identi-fication of geographic origin, except possibly the higher frequency of aspirated pits in the latter type. Drow (1957) set out to collect data for a comprehensive comparison of coast, intermediate (interior west) and Rocky Mountain (interior) material. The results confirmed that differences exist between zones , but, unfortunately/ the coastal samples were second-growth only whereas the others appeared to be from virgin stands. Mean density values were found to be as follows: The author stressed that the coastal samples may not have been representa-t i v e of the area as a whole due to the young age and also noted that the high v a r i a b i l i t y within s i t e s would make predictions of strength properties based on small sample numbers subject to large possible errors. V a r i a t i o n Coast I n t e r i o r (south) I n t e r i o r west Interior (north) •J 3 0.428'.jgi/cm 3 0.433 g/cm 3 0.415 g/cm 3 0.386 g/cm - 12 -i n percentage latewood p a r a l l e l e d that o f density, but i t was found that the latewood density could vary with l o c a l i t y . For example, latewood density i n the southern section of the i n t e r i o r zone was s u b s t a n t i a l l y lower than elsewhere. In f a c t , Drow (1957) t e n t a t i v e l y proposed that t h i s l a t t e r region could be considered to be subdivided i n t o north and south sections. A recent and more comprehensive survey (U.S. Forest Service, 1965) confirmed the above trends and stressed the considerable v a r i a t i o n between sample locations within the broad regions. Increment cores were c o l l e c t e d from 1512 p l o t s l a i d out on a 6.8 mile g r i d over the natural range to give a r e l i a b l e systematic sample s i m i l a r to those previously employed succes-f u l l y i n the southern pine areas ( M i t c h e l l , 1964). However, the strong association of density with geographic l o c a t i o n t y p i c a l of the southern pines was not found to occur i n Douglas-fir. Within the'two major subdiv-i s i o n s (Coast and Rocky Mountain) only weak c o r r e l a t i o n s were found between l a t i t u d e and density. West of the Cascade and S i e r r a ranges there was a tendency for wood density to increase from north to south, whereas i n the Rocky Mountain zone the reverse was found. The study upheld the v a l i d i t y of considering Douglas-fir as c o n s i s t i n g of four main regions as proposed by Drow (1957). Mean density values were: No explanation was given for the f a c t that these figures are consis-t e n t l y higher than those reported previously. Canada was not included i n t h i s l a t t e r survey, but data given by Smith (1970) showed the average Coast I n t e r i o r (south) I n t e r i o r west I n t e r i o r (north) 0.45 g/cm' 0.46 g/cm' 0.45 g/cm 0.43 g/cm' - 13 -density of incremental core samples from the i n t e r i o r of B r i t i s h Columbia 3 was 0.42 g/cm . This value would appear to counteract the trend of increasing density northweards detected by the two U.S. surveys. Snodgrass and Noskowiak (1968) assessed strength properties of Douglas-f i r boards c o l l e c t e d from sawmills and concluded that material from the coastal area was superior to that of the i n t e r i o r i n a l l properties measured except maximum crushing strength. This anomaly could not be explained on the basis of the data compiled. Hughes and A l l e n (1949) examined a t o t a l of 16 stems representing second-growth on s i t e classes I, II and III i n western Washington. Mean 3 3 d e n s i t i e s ranged from 0.42 g/cm on s i t e c l a s s I to 0.49 g/cm on s i t e c l a s s I I I . They reasoned, however, that the better s i t e s produced the more desirable timber since s p e c i f i c strength values were p o s i t i v e l y correlated with s i t e q u a l i t y . Wellwood (1952) also found density to be lower on better s i t e s at the University of B r i t i s h Columbia (U.B.C.) Research Forest by up to 12%, but used density d i r e c t l y as an i n d i c a t o r of wood q u a l i t y to con-clude that the poorer s i t e s w i l l y i e l d the more desirable product for both pulping and s t r u c t u r a l use. Within these stands, dominant stems tended to have s i g n i f i c a n t l y lower density values than others. In a more intensive study of 36 trees from second-growth crops i n Washington and Oregon, McKimmy (1959) examined the e f f e c t s of crown c l a s s , s i t e c l a s s , geographic l o c a t i o n and tree age on percentage latewood and wood density. S t a t i s t i c a l analyses showed that tree age was the most impor-tant factor a f f e c t i n g density and that the others had l i t t l e influence. In most of the above surveys, a t t e n t i o n was drawn to the large v a r i a -t i o n found between i n d i v i d u a l trees. For example, the tables supplied by Drow (1957) give minimum and maximum values f o r each shipment - which show - 14 -a range f a r i n excess of the r e l a t i v e l y minor differences between regions. L i t t l e f o r d (1961) studied density and. strength properties i n f i v e open-grown stems, and although the mechanical properties measured were f a i r l y c l o s e l y r e l a t e d to wood density, the observed v a r i a t i o n s i n mean density within and between stems were not i n themselves s u f f i c i e n t to explain the considerable v a r i a t i o n i n strengths. In the portion of the stems outside the f i r s t 15 rings, wood density was associated with 64% of the v a r i a t i o n i n modulus of rupture, 57% for maximum crushing strength and only 36% for modulus of e l a s t i c i t y . Drow (1957) had also noted wide v a r i a t i o n s i n strength properties at a given density l e v e l . The above r e s u l t s strongly suggest that mean density by i t s e l f , although a good general i n d i c a t o r of wood q u a l i t y , cannot always be r e l i e d upon to provide r e l i a b l e p r e dictions of strength p r o p e r t i e s . The most l i k e l y reasons f o r t h i s may be found i n the data of Harris and Orman (1958) which indicates that both earlywood and latewood d e n s i t i e s can vary apprec-i a b l y between trees. Thus, i t i s e n t i r e l y possible that trees having the same mean density could e x h i b i t e n t i r e l y d i f f e r e n t earlywood and latewood values, even at the same percentage latewood. In the absence of evidence to the contrary, Harris and Orman (1958) suggested that t h i s type of v a r i a -t i o n may be of genetic o r i g i n . 2.2 Relationships Between Growth Ring Components A l t e r a t i o n s i n r i n g width pev se cannot a f f e c t wood density unless accompanied by regular changes i n proportions or d e n s i t i e s of the e a r l y -wood and latewood components. Thus the observed patterns of r i n g width and density, c h a r a c t e r i s t i c of a given combination of s i t e and stand - 15 -conditions (long term trends) may a r i s e under the influence of a set of factors quite d i f f e r e n t from those a f f e c t i n g f l u c t u a t i o n i n properties from year to year (short term trends). Paul (1963) showed c l e a r l y that manipulation of the growing space a v a i l a b l e to southern pines could e i t h e r increase or decrease wood density depending on whether the production of latewood or earlywood was aff e c t e d . E a r l y workers with Douglas-fir described an a s s o c i a t i o n between rate of growth (rings per inch) and timber strength properties. Both f a s t (< 6 r.p.i.) and slow (> 20 r.p.i.) growth were held to contribute to low strength values. On t h i s basis, i t was advocated that such material be omitted from the better s t r u c t u r a l grades (Sterns, 1918). Later, Alexander (1935) examined logs from B r i t i s h Columbia and concluded that the upper l i m i t of 20 r . p . i . be discarded since much of the timber excluded would have higher strength values than material near the lower l i m i t . Further work by the same author (Alexander, 1950), confirmed the necessity of adhering to the lower l i m i t i n second growth as i n o l d growth. Drow (1957) observed, however, that adherence to the lower l i m i t would have l i t t l e e f f e c t on improving wood q u a l i t y i n the Rocky Mountain area since only a very small proportion of samples would contain l e s s than 6 r . p . i . His p l o t s of density against growth rate f o r d i f f e r e n t geographic regions i l l u s t r a t e a rapid increase between 2 and 12 r . p . i . and thereafter only a s l i g h t decrease out to 60 r . p . i . i n the Rocky Mountain groups. Coastal samples showed a continuing but diminishing increase i n density with increasing r . p . i . a f t e r 12 r . p . i . , and had higher density values at any given growth rate. Mozina (1960) studied 35 stems of Douglas-fir ranging i n age from 45-70 years from several s i t e s . Examinationsof t e s t samples revealed that about 50% of the t o t a l v a r i a t i o n i n density could be explained i n multiple regression using age, r i n g width and l o c a t i o n within the stem as independent - 16 -v a r i a b l e s . Ring width alone was associated with only 11%, l o c a t i o n with around 12% and age proved non-significant. The above r e l a t i o n s h i p s are a r e f l e c t i o n of inherent b i o l o g i c a l v a r i a -t i o n within trees rather than a true growth rate e f f e c t . I t i s well estab-l i s h e d that growth rings tend to become narrower outwards from the p i t h whereas density and percentage latewood increases, thus r e s u l t i n g i n a negative association. The narrow outermost rings of o l d stems may e x h i b i t the lower density c h a r a c t e r i s t i c s of "overmature" wood ( E l l i o t t , 1970). The true e f f e c t of growth rate can only be assessed by comparing trees of s i m i l a r age growing i n the same environment but at d i f f e r e n t rates. That the observed negative r e l a t i o n s h i p i s not a straightforward one has been shown by both Harris and Orman (1958) and Knigge (1962). These authors, working r e s p e c t i v e l y i n fast-grown plantations i n New Zealand, and natural stands i n the P a c i f i c Northwest, confirmed that wood near the p i t h has lower density than older wood of equivalent growth rate. In f a c t the f i r s t -mentioned authors determined that i f the samples were subdivided by 5-ring groups from the p i t h , a d i f f e r e n t r e l a t i o n s h i p was evident f o r each group, i l l u s t r a t i n g that cambial age has an important e f f e c t on the density/growth rate a s s o c i a t i o n . In an unpublished survey of unthinned plantation-grown D o u g l a s f f i r , the current author (Cown, 1971b) observed a consistent but non-significant negative r e l a t i o n s h i p between growth rate and wood density i n the outerr rings of stems 40-50 years o l d . Of p a r t i c u l a r i n t e r e s t was the observation that i n one such pla n t a t i o n where stocking was very v a r i a b l e , a p o s i t i v e c o r r e l a t i o n was evident between growth rate andddensity when stems were groupedd by three spacing c l a s s e s . Turnbull (1948) was one of the f i r s t researchers to s u c c e s s f u l l y demonstrate a r e l a t i o n s h i p between cambial age and wood density i n pine, - 17 -independent of growth rate. Chalk (1953), i n an attempt to v e r i f y t h i s pattern f or Douglas-fir growing i n Great B r i t a i n examined two fast-growing stems and found density to increase out to about the 6th r i n g and remain more or l e s s constant thereafter. Most l a t e r studies which have examined the e f f e c t of tree age have determined that there i s a d e f i n i t e p o s i t i v e r e l a t i o n s h i p but that i t may not be a simple l i n e a r one (Harris and Orman, 1958; McKimmy, 1966). Knigge (1962), i n a study of second-growth Douglas-fir under 100 years o l d , was of the opinion that when wood samples are divided into age groups, growth rate has an important influence on density through the f i r s t few decades whereas tre e age becomes more s i g n i -f i c a n t i n l a t e r l i f e . According to the U.S. Forest Service survey of 1965, tree age was by f a r the most important si n g l e v a r i a b l e a f f e c t i n g wood density although i t only accounted for 16% of the t o t a l v a r i a t i o n . I t has been c l e a r l y established that there i s a close r e l a t i o n s h i p between percentage latewood and wood density i n Douglas-fir, but some discrepancies appear i n the l i t e r a t u r e due to d i f f e r e n t methods of measure-ment, p a r t i c u l a r l y i n the rings close to the p i t h where the s o - c a l l e d "tran-s i t i o n " latewood may give r i s e to errors i n assessment. This may account for the f a c t that Drow (1957) found the density/percentage latewood r e l a t i o n -ship to be d i f f e r e n t f o r separate growth rate c l a s s e s . The r e l a t i o n s h i p was also shown to vary somewhat with geographic l o c a t i o n due to differences i n mean latewood density. McKimmy (1959) considered that both percentage latewood and tree age were more important than growth rate i n coastal second-growth since v a r i a t i o n s i n percentage latewood at constant growth rate had more influence on density than v a r i a t i o n s i n growth rate at constant percentage latewood. He cautioned, however, that i n no part of the trees studied could growth rate and percentage latewood be used for the r e l i a b l e - 18 -p r e d i c t i o n of density. These r e s u l t s were i n contrast to the e a r l i e r work of Smith (1956) who studied 16 butt logs from Washington and Oregon and concluded that percentage latewood was strongly r e l a t e d to density indep-endent of wood age and growth rate. Harris and Orman (1958) and L i t t l e -ford (1961) also agree that percentage latewood i s the s i n g l e most import-ant factor i n f l u e n c i n g density i n Douglas-fir. Recent work by Lassen and Okkonen (1969) on coastal Douglas-fir growing on a v a r i e t y of s i t e s showed that 68% of the o v e r a l l v a r i a t i o n i n density could be accounted f o r by changes i n latewood percentage. The above r e l a t i o n s h i p i s an i n t u i t i v e l y obvious one, p a r t i c u l a r l y i n Douglas-fir where latewood density i s about 2.5 times that of e a r l y -wood. However useful t h i s a s s o c i a t i o n may be i n v i s u a l grading of lumber i t i s of l i t t l e assistance i n explaining how or why v a r i a t i o n s i n density a r i s e between trees and s i t e s . 2.3 E f f e c t s of Environmental Factors Many workers have published r e s u l t s demonstrating e f f e c t s of the external environment on wood formation, both i n terms of geographic and c l i m a t i c v a r i a b l e s and as a r e s u l t of applied treatments (see Kozlowski, 1971 f o r r e f s . ) . The bulk of the work, however, has been concerned only with growth-ring width components since e x c i s i o n of small wood samples for q u a l i t y determination can be exceedingly tedious. The development of techniques i n r a d i a t i o n densitometry of wood has provided the p o t e n t i a l for rapid accumula-t i o n of data on both intra-increment width and wood density components, but to date few studies have been reported i n t h i s area. Studies on environmental factors i n f l u e n c i n g wood density i n Douglas-f i r have been l i m i t e d , perhaps due to a tendency to a t t r i b u t e geographical - 19 -v a r i a t i o n to genetic rather than c l i m a t i c sources. The present author i s not aware of any comparison of wood properties of, say, co a s t a l and i n t e r i o r types grown i n the same environment, apart from a small study by Haigh (1961). McKimmy (1959) was of the opinion that geographic l o c a t i o n was of l i t t l e s i g n i f i c a n c e i n western Washington and Oregon, but l a t e r work by the same author (McKimmy, 1966; McKimmy and Nicholas, 1971) showed l o c a l conditions to have a highly s i g n i f i c a n t e f f e c t . The l a t t e r r e s u l t s were achieved using a much more comprehensive sample, but no attempt was made to r e l a t e the differences to s p e c i f i c environmental f a c t o r s . Knigge (1963) extracted two increment cores from each of 256 trees growing on 51 s i t e s i n Washington, Oregon and northern C a l i f o r n i a . Also noted were tree s i z e and age and c l i m a t i c data. Mu l t i p l e regression tech-niques showed that at best only about 34% of the v a r i a t i o n could be accounted for by the parameters measured, i n contrast to around 70% for surveys i n the southern pines ( M i t c h e l l , 1964). Only minor e f f e c t s could be ascribed to a l t i t u d e , r a i n f a l l and temperature. O v e r a l l , there was a tendency towards increased density with increasing age, s i t e c l a s s and growing season tempera-ture, and a decrease with increasing a l t i t u d e and growing season r a i n f a l l . The U.S. Forest Service survey (1965) included an examination of the influence of geographic l o c a t i o n i n the area of the west slopeoof the Cascade Mountains i n Washington and Oregon. When data were grouped by age classes to remove the highly s i g n i f i c a n t age e f f e c t i t was shown that 2 l a t i t u d e ~ had i t s greatest e f f e c t i n the 150-249 yr group with an R value of 0. 8 (density increasing northwards). No e f f e c t of l a t i t u d e was apparent i n the 5-34 yr or 35-74 yr groups. El e v a t i o n was found to have a stronger - 20 -e f f e c t , f i r s t appearing i n the 35-74 yr. group (R = 0.6)and also reaching 2 a maximum i n the 150-249 yr . c l a s s (R = 0.25). The graphical presentation of the data suggests that at elevations of 4000 feet and over, density does not show the c h a r a c t e r i s t i c increase with age found at lower a l t i t u d e s , such that trees on these s i t e s tend to have r e l a t i v e l y high density when young and r e l a t i v e l y low density when o l d . A s i m i l a r trend can be observed i n data presented by McKimmy (1966) i n which the highest elevation s i t e showed a density pattern quite d i f f e r e n t fromtthe other plantations. The only reported study i n Douglas-fir designed s p e c i f i c a l l y to evaluate environmental influences i s that of Lassen and Okkonen (1969). Three sample locations were selected i n each of 15 categories representing a l l possible combinations of f i v e a r b i t r a r i l y selected r a i n f a l l classes (< 5" to > 15" per. summer) and three a l t i t u d i n a l groups between sea l e v e l and 3000 feet . At each l o c a t i o n , f i v e dominants of around 50 years o l d were sampled with increment borers and one growth r i n g measured for width, percen-tage latewood and wood density. Within each elevation c l a s s a negative r e l a -t i o n s h i p was established between density and summer r a i n f a l l . The highest mean density values were found i n the lowest elevation group (< 900 feet) and the lowest density at the intermediate elevation (1000 - 1750 f e e t ) . The r e s u l t s tend to confirm trends indicated i n the work of Knigge (1962) and the U.S. Forest Service (1965). Percentage latewood as a single v a r i a b l e was associated with 68% of the v a r i a t i o n i n density whereas p r e c i p i t a t i o n alone accounted f o r only 31% and a combination of p r e c i p i t a t i o n and elevation accounted f o r 53%. A measure of the importance of the r e l a t i o n s h i p s determined i s the range i n 3 predicted density values from 0.44 g/cm (high elevation, wet summer) to - 21 -0.55 g/cm (low elevation, dry summer), a spread much i n excess of that f or dif f e r e n c e s between broad geographic areas. The reason f o r the unaccounted v a r i a t i o n could not be explained, but i t was suggested that genetic sources as well as sampling errors might have been involved. Results from work done i n Douglas-fir plantations of New Zealand indic a t e that s i t e has an important e f f e c t on wood properties (Harris and Orman, 1958; Spurr, 1961), but attempts to i d e n t i f y c o n t r i b u t i n g factors have so f a r proven unsuccessful due to confounding of provenance and s i t e e f f e c t s (Harris, 1966). I t has been suggested by the above authors, however, that s o i l moisture conditions are very important. Some workers have attempted to explain environmental e f f e c t s on wood formation by examining year-to-year v a r i a t i o n s i n r e l a t i o n to weather records. Chalk (1930) studied wood of Dougl a s t f i r and showed that the amount of latewood produced i n any one year was r e l a t e d to a v a i l a b l e s o i l moisture during latewood formation, whereas earlywood production was negatively relatedt.to mean May and June temperature. Kennedy (1961) examined second-growth stems from the U.B.C. Research Forest i n r e l a t i o n to v a r i a t i o n i n percentage latewood over a 4-year period. Both earlywood and latewood widths varied independently between growing seasons, but i t was noted that A p r i l weather had a s i g n i f i c a n t influence on latewood. L o w r r a i n f a l l , high temperatures and long periods of sunshine at t h i s time were p o s i t i v e l y associated with latewood percentage. H a l l (1962) also measured earlywood and latewood widths over a 5-year period at the U.B.C. Research Forest. Highest percentage latewood values were found on mesic rather than extreme s i t e s , and there was a tendency for both early and l a t e season r a i n f a l l to be p o s i t i v e l y c o r r e l a t e d with l a t e -- 22 -wood percentage. The results appear contradictory to those of Kennedy (1961) and point out the dangers of extrapolating from a limited data base. A study of Douglas-fir growing on Vancouver Island (Smith, 1973) revealed important climatic influences. When earlywood and latewood widths were analyzed separately in relation to weather during the period of their formation, precipitation was shown to have a strong positive effect and temperature a negative effect. Differences were noted between sites in the magnitude of these influences, which were measured over a 30-year period. Heger et al. (1974) used a scanning densitometer to examine 20 trees from the U.B.C. Research Forest and related growing season weather to growth ring component variables over a 7-year period. Average latewood density and maximum ring density varied most from year to year and were correlated positively with temperature and negatively with precipitation. Neither of these ring properties was closely related to mean ring density however, despite the fact that average earlywood density remained almost constant throughout the period. The effect of weather on percentage late-wood was not reported. In summary, i t has been established that site effects and weather variations from year to year can influence the percentage latewood and density of the wood formed. However, earlywood and latewood widths can be altered independently of each other, so i t may be inappropriate to attempt to correlate single environmental factors with percentage latewood. Densi-tometric analyses can be expected to provide new opportunities for relating climatic and wood property parameters since ring component densities are measured in addition to widths. The bulk of the studies to date on Douglas-fir wood properties have - 23 -not been planned to account f o r genetic sources of v a r i a t i o n . Work on geographic e f f e c t s has provided valuable information on the wood resource, but inferences about the reasons f o r such v a r i a t i o n are purely speculative, since very often tree age, s i t e , stand conditions and genetic c o n s t i t u t i o n are confounded. Attempts to remove some of the obvious sources of error have sometimes been made, e.g. , sampling only trees of s p e c i f i e d age or crown classes. Other p r a c t i c e s , such as c o l l e c t i n g only the outer tten inches of wood at breast height (U.S. Forest Service, 1965), and f a i l u r e to remove wood extractives, can contribute to sampling e r r o r s . There i s c l e a r l y a need f o r more work on Douglas-fir i n view of i t s economic importance and since sources of v a r i a t i o n i n wood density have not been s a t i s f a c t o r i l y explained. No combination of independent v a r i a b l e s thus f a r employed has been associated with more than about 30% of the observed v a r i a t i o n . Very l i t t l e has been reported on the genetic contribution, apart from the consistent mention of large between-tree differences i n properties , and these are most often ascribed to genotype. M i t c h e l l (1964) was of the opinion that the lack of success i n accounting f o r v a r i a b i l i t y i n wood properties of Douglas-fir may be due to the very wide natural d i s t r i b u t i o n and a d a p t a b i l i t y of the species to d i f f e r e n t environments. I m p l i c i t i n hi s statement i s the p o s s i b i l i t y that genetic d i f f e r e n t i a t i o n has given r i s e to a number of geographic types, or provenances, within the major types already recognized, and that these may have evolved more or l e s s d i s t i n c t wood properties. To date, no conclusive evidence has been presented to support t h i s hypothesis, but the works of Gohre (1958), Harris (1966), McKimmy (1966) and McKimmy and Nicholas (1971) suggest that such may be the case. - 24 -2.4 Inheritance of Ring-Width and Density Components Studies such as the survey work discussed e a r l i e r have repeatedly shown v a r i a t i o n i n growth and wood properties due to geographic l o c a t i o n . However, i t cannot be d e f i n i t e l y ascertained whether these regional d i f f e r -ences are c o n t r o l l e d by genetic or environmental f a c t o r s . I t has long been recognized that genetic v a r i a t i o n can e x i s t within the natural d i s t r i b u t i o n s of tree species, a'.fact which becomes evident when seedlots from d i f f e r e n t geographic sources (provenances) are planted together i n the same environment. Over the past 20-30 years there has been a spec-t a c u l a r f l o u r i s h i n g of f o r e s t genetics research, much of which has centered around provenance experimentation with the object of determining the best seed-source/site combinations (Haddock, 1967). Of prime importance i s s u r v i v a l and volume production, and evidence of provenance differences i n height and diameter growth abound i n the l i t e r a t u r e of countries using exotic species. Hermann and Ching (1975) have compiled a bibliography of provenance research i n Douglas-fir. Increasingly, i n t e r e s t has been shown i n wood q u a l i t y , and wood density i n p a r t i c u l a r , since the report by F i e l d i n g (1953) on c l o n a l differences i n r a d i a t a pine. Provenance te s t s provide a means of comparing seed-source and environ-mental e f f e c t s under s p e c i f i e d conditions, and d i f f e r e n c e s i n wood density associated with seed o r i g i n have been shown for many species. Hoist (1958) ref e r r e d to v a r i a t i o n of density i n white spruce (Pioea glauca (Moench.) Voss.) i n Canada, while Parrott (1960) found highly s i g n i f i c a n t d i f f e r e n c e s i n Norway spruce {Pioea abies (L.) Karst.). Genetic and environmental influences have been measured i n ponderosa pine (Pinus ponderosa Laws) by Harris and Kripas (1959) and Echols and Conkle (1971) , and much work has also been c a r r i e d out on the major southern pine species (Zobel, 1961; Harris and B i r t , 1972) . - 25 -In some cases, genetic differences have been found to be e i t h e r small or lacking. For example, of 19 seed sources from the natural range of red pine {Firms vesinosa A i t . ) only one was s i g n i f i c a n t l y d i f f e r e n t from the others (Rees and Brown, 1954), suggesting that genetic d i f f e r e n t i a t i o n has been minimal i n t h i s species. Fowler (1964) i n a review of v a r i a b i l i t y i n red pine confirmed the o v e r a l l uniformity. ' Both Knudsen (1956) and Worrall (1975) found l i t t l e v a r i a t i o n i n the wood properties of Norway spruce provenances, and a comparison of several seedlots of Corsican pine (Pinus nigra Arn.) grown i n New Zealand also f a i l e d to reveal s i g n i f i c a n t e f f e c t s (Cown, 1974 b,). Whether or not s i g n i f i c a n t provenance differences were found, most authors have commented on the large i n d i v i d u a l tree v a r i a t i o n , even under comparatively uniform p l a n t a t i o n conditions, and several workers have ascribed t h i s phenomenon to strong genetic c o n t r o l . Since the early 1960's, many reports have appeared dealing with obser-vations r e l a t i n g to the inheritance of growth rate and density, some con-cerning both earlywood and latewood c h a r a c t e r i s t i c s . Of p a r t i c u l a r i n t e r e s t are the works of Polge and I l l y (1968), N i c h o l l s and Brown (1971), Polge (1971), Harris and B i r t (1972) and K e l l e r (1973), a l l of whom have used r a d i a t i o n densitometry to compare wood properties of g e n e t i c a l l y d i f f e r e n t groups of trees. 2.4.1 Measurement of genetic c o n t r o l The most commonly used measure of the r e l a t i v e degree of genetic con-2 t r o l of plant c h a r a c t e r i s t i c s i s the h e r i t a b i l i t y index, h , defined as that portion of the t o t a l v a r i a b i l i t y which can be i d e n t i f i e d as being of - 26 -genetic o r i g i n and capable of being transmitted from parents to o f f s p r i n g . In the case of sexual reproduction, only the additive component of the 2 genetic variance can be exploited i n breeding, so h becomes: 2 h = V /V-, where V i s the ad d i t i v e genetic variance and V i s the A P A ^ P t o t a l phenotypic variance. This expression i s known as h e r i t a b i l i t y i n the narrow sense. With vegetative reproduction'(eg. clones) the genotype remains unaltered between generations, and so the r e l a t i o n s h i p becomes: 2 h = V /V , where V i s the t o t a l genetic variance, i n c l u d i n g a d d i t i v e , G P G dominance and i n t e r a c t i o n e f f e c t s (Falconer, 1960). This i s known as the broad sense h e r i t a b i l i t y . Although these concepts are simple, the use of h e r i t a b i l i t y i n the for e s t r y l i t e r a t u r e has lead to much confusion. C a l c u l a t i o n of the variance components from analyses of variance requires a degree of s k i l l i n biomet-r i c s , and Namkoong et dl. (1966) point out that errors have been made i n some cases. Zobel (1964) comments that some authors present h e r i t a b i l i t y data without d e f i n i n g whether they are used i n the narrow or broad sense, or whether they were calculated on an i n d i v i d u a l tree or family mean b a s i s . Genetic information may be gathered from studies of parent-progeny r e l a t i o n s h i p s or from comparisons of groups of r e l a t e d p l a n t s . Zobel and Rhodes (1957) examined 12-year-old s e l f e d trees and open-pollinated groups of l o b l o l l y pine (Pinus taeda L.) and noted good c o r r e l a t i o n s between parent and o f f s p r i n g wood density. Later studies reported h e r i t a b i l i t y estimates f o r r i n g width and density components. Appendix 2 presents the bulk of the av a i l a b l e information. The reported h e r i t a b i l i t y estimates are quite v a r i a b l e and i t i s obvious that they depend very much on the type of plant material used, tree age, - 27 -experimental design and the environment studied. Each of these factors can influence e i t h e r the genetic or environmental variance components on which the c a l c u l a t i o n s are based. I t i s therefore important to r e a l i z e that published estimates r e f e r to s p e c i f i c conditions and need not be relevant to other s i t u a t i o n s . For example, Goggans (1964) studied young open-pollinated l o b l o l l y pine at two locations (Georgia and Louisiana) and determined some markedly d i f f e r e n t h e r i t a b i l i t y values f o r the same wood c h a r a c t e r i s t i c s at the two s i t e s . Tree age may be a p a r t i c u l a r l y important v a r i a b l e . Zobel (1964) pointed out that narrow sense h e r i t a b i l i t y of wood density i n l o b l o l l y pine appears to increase up to at l e a s t 15 years from planting. N i c h o l l s (1967) made density determinations on i n d i v i d u a l growth rings of 25-year-old r a d i a t a pine clones and reported a decrease i n h e r i t a b i l i t y f o r about ten years, followed by an increase out to the periphery of the stem. Values f o r the inner and outer rings were s i m i l a r . In a l a t e r p u b l i c a t i o n , (Nicholls and Brown, 1971), data were tabulated which suggested that h e r i t a b i l i t i e s of minimum, maximum and mean r i n g d e n s i t i e s a l l increased from the p i t h out to at l e a s t the 8th r i n g . Burdon and Harris (1973), on the other hand, found l i t t l e evidence of a change i n h e r i t a b i l i t y between rin g s 1 - 5 and 6 - 1 0 i n 12-year-old ra d i a t a pine clones r e p l i c a t e d on four s i t e s . The values presented i n Appendix 2 have been determined f o r a wide v a r i e t y of species, tree ages (ranging from 1-year-old seedlings (van B u i j -tenen, 1962) to 46-year-old trees (McKimmy, 1966)), and types of material. In some cases i n d i v i d u a l growth rings have been measured (Dadswell et al. , 1961; K e l l e r , 1973; Worrall, 1975) and i n others wood sample mean values have been used (Goddard and Cole , 1966). Some experimental designs were well - 28 -r e p l i c a t e d (Nicholls et al. , 1964; M a t z i r i s and Zobel, 1973) and some unreplicated (Nicholls and Brown, 1971; Dadswell et al. , 1961). The vast majority of reported values have been ca l c u l a t e d on very young trees, so that p r e d i c t i o n s regarding properties at r o t a t i o n age must be speculative only. The important point i s that genetic c o n t r o l of wood c h a r a c t e r i s t i c s appears to be high i n comparison with other t r a i t s such as growth rate and branching c h a r a c t e r i s t i c s (Campbell, 1964). 2.4.2 Juvenile-mature c o r r e l a t i o n s Forest g e n e t i c i s t s must u t i l i z e the r e s u l t s of e a r l y t e s t s as soon as they can be r e l i e d upon to give <a reasonable i n d i c a t i o n of the r e l a t i v e performance at maturity. For external tree c h a r a c t e r i s t i c s the necessary data may require a study of the properties throughout the r o t a t i o n or else an examination of parent-progeny r e l a t i o n s h i p s . Wood properties have also been studied on a parent-progeny basis by several workers (van Buijtenen, 1962; Polge and I l l y , 1968; N i c h o l l s and Brown, 1971; K e l l e r , 1973), who have generally been able to show good c o r r e l a t i o n s between o f f s p r i n g and parent properties, e s p e c i a l l y when allowance i s made f o r the d i f f e r e n c e i n age of wood formation. Since the xylem i s a permanent t i s s u e within stems, i t o f f e r s an opportunity to study juvenile-mature r e l a t i o n s h i p s within stems using wood of d i f f e r e n t ages. Zobel and Rhodes (1956) found high c o r r e l a t i o n s (r>.8) between the density of the f i r s t 8 rings and mean cr o s s - s e c t i o n a l density of l o b l o l l y pine trees 25-70 years o l d i n natural stands i n Texas. N i c h o l l s and Dadswell (1965) reported an absence of u s e f u l c o r r e l a t i o n s between the density of wood from near the p i t h and that near the bark i n 25-year-old - 29 -radiata pine trees from four populations i n A u s t r a l i a . On the other hand, Harris (1965), working with the same species i n New Zealand,found that mean density of the t h i r d r i n g from the p i t h was associated with about 50% of the v a r i a t i o n i n density of the outer ten rings i n trees 35 years of age. A small number of such studies have been c a r r i e d out using Douglas-fir. Northcott et al. (1964) found wood of the inner•ten rings to be of extreme v a r i a b i l i t y i n density patterns and recommended that such j u v e n i l e wood be excluded from between-tree comparisons and juvenile-mature c o r r e l a t i o n s . Harris (1966) showed that whereas the density of rings 1-5 was not s i g n i f i -cantly r e l a t e d to stem density at age 40 years, the density of rings 6-10 d i d give a s i g n i f i c a n t c o r r e l a t i o n (r= 0.36). An analysis of the 1912 Douglas-fir provenance t e s t by McKimmy (1966) f a i l e d to reveal a s i g n i f i c a n t juvenile-mature r e l a t i o n s h i p and the conclusion derived was that no u s e f u l p r e d i c t i o n s of outerwood density could be obtained before at l e a s t 25 years of age. Reck and S z i k l a i (1973) studied a large number of increment cores from plus trees growing throughout the P a c i f i c Northwest and determined r i n g width and density on samples of s p e c i f i c r i n g groups. C o r r e l a t i o n matrices showed that properties of rings 1-5 were r e l a t e d to mature wood properties but that for p r e d i c t i v e purposes older wood would be more s u i t a b l e . They recommended that samples f o r e a r l y assessment should contain at l e a s t eight growth ri n g s . Results of juvenile-mature c o r r e l a t i o n s f o r wood properties have been very v a r i a b l e , but t h i s i s perhaps not s u r p r i s i n g considering the d i f f e r e n t conditions under which the sample material had grown and the use of d i f f e r e n t wood ages i n the c o r r e l a t i o n s . I t might have been a n t i c i p a t e d that planta-tion-grown trees would give more s a t i s f a c t o r y r e s u l t s than trees from - 30 -natural stands since age and spacing are not so v a r i a b l e and hence corewood properties should be more uniform. However, the data of McKimmy {op ait.) proved disappointing whereas those of Reck and S z i k l a i {op ait.) working with stems from widely separated natural stands were encouraging. Most reports have attempted to r e l a t e properties of a group of inner rings to those of a group of outer r i n g s . This type of c o r r e l a t i o n would appear to be of l i m i t e d p r a c t i c a l value since early evaluation should be aimed at p r e d i c t i n g mature c h a r a c t e r i s t i c s of the merchantable stem, of which corewood may i t s e l f be a s i g n i f i c a n t portion. I t should also be remembered that highly accurate predictions are u n l i k e l y to be required, the object being i n most cases to separate trees or groups of trees into broad cl a s s e s . No reports are yet a v a i l a b l e on juvenile-mature c o r r e l a t i o n s for growth r i n g components other than r i n g width and r i n g density. 2.4.3 Genetic studies of Douglas-fir wood density Provenance research i n Douglas-fir has established beyond doubt that genetic d i f f e r e n t i a t i o n i n t h i s species i s s i g n i f i c a n t , even within the coastal v a r i e t y (Hermann and Ching, 1975). A number of reports have dealt with provenance differences i n seed and seedling c h a r a c t e r i s t i c s (Ching and Bever, 1960; Sweet, 1965), early height growth (Ching, 1965; Schmidt, 1973), flu s h i n g (Morris et al., 1957; S i l e n , 1962) and height growth i n plantations (Rowe and Ching, 1973; Namkoong et al. , 1972). Taken together, these studies suggest that c l i n a l v a r i a t i o n patterns e x i s t i n both north-south and east-west d i r e c t i o n s west of the Cascade Mountains. Genetic d i f f e r e n t i a t i o n i n the east-west d i r e c t i o n appears to be marked (Kleinschmidt et al., 1974). Work on the genetic aspects of wood q u a l i t y v a r i a t i o n i n Douglas-fir has been very sparse considering the world-wide importance of the species. Gohre (1958) presented data which showed that coastal and i n t e r i o r provenances growing i n Germany produced wood of d i f f e r e n t density and a l s o suggested that some sources from Washington and Oregon may be s l i g h t l y d i f f e r e n t . Haigh (1961) examined four trees from each of f i v e provenances from B r i t i s h Columbia, Washington and Colorado at age 7 years, and found that provenance differences were not s i g n i f i c a n t . Within each provenance, a strong negative r e l a t i o n s h i p . was observed between growth rate and density, but on a provenance mean b a s i s , growth rate and density were not r e l a t e d . Nicholas (1963) studied genetic and environmental sources of v a r i a t i o n i n the 1912 U.S. Forest Service provenance study described by Munger and Morris (1936). One hundred and twenty parent trees had been selected from 13 l o c a l i t i e s i n Washington and Oregon and h a l f - s i b progenies outplanted at s i x s i t e s . Nicholas (op. C-ut.) examined ten progeny from each of f i v e parent trees at three l o c a t i o n s by extracting a sin g l e increment core per tree. Growth r i n g s 1954-58 were used f o r wood density determination. S i g n i f i c a n t parent e f f e c t s were found but these were overshadowed by the v a r i a t i o n due to p l a n t a t i o n l o c a t i o n . Estimates of h e r i t a b i l i t y ranged from 0.17 to 0.52 between p l a n t a t i o n s . McKimmy (1966) obtained one increment core sample from four trees of each of 13 seed sources at four s i t e s of the same t r i a l . /Analyses of wood density were performed by 5-year growth i n t e r v a l s from bark to p i t h on resin-extracted wood. The dominant influence on density was p l a n t a t i o n en-vironment, with one high a l t i t u d e s i t e showing a markedly d i f f e r e n t r a d i a l - 32 -trend. Within plantations there was a tendency for seed sources to maintain similar rankings despite the large individual tree variation, and overall, the effect of race was highly significant. The observed differences between seed sources could not be satisfactorily explained in terms of elevation of the parent trees, but there was an indication that families from areas close to the plantations ranked near the average density for the site. Heri-t a b i l i t y values were calculated for each 5-year period at a l l plantations and estimates ranged from 0.00 to 0.66. Despite the division of the wood samples into age groups, no trends in h e r i t a b i l i t y were apparent with increasing tree age. Harris (1966) reported results of a survey of Douglas-fir plantations aged 34-40 years growing in New Zealand. At each of 19 sites, 100 trees were sampled with an increment borer and the outer ten rings used for wood density determinations. Density varied widely between sites and i t was sus-pected that some of the differences were due to provenance effects although seedlot records were not available. It was also noted, however, that large density differences could occur within the same plantation. For example, two 100-tree samples were collected less than 100 m apart, one in a gully 3 and one on a ridge 20 m above and the density difference of 0i.45 g/cm to 3 0.-53 g/cm was larger than between most sites. More work was advocated to examine provenance and environmental effects. Interest in the wood quality of Douglas-fir is rapidly increasing and research is currently being undertaken in New Zealand (J.M. Harris, pers. comm.) and in France (M. Thoby, pers. comm.) where this species is an impor-tant exotic. Results from these studies w i l l no doubt become available during the course of the work reported here and w i l l be discussed in the appropriate chapters. - 33 -2.5 Crown Phenology and Wood Properties Research r e s u l t s over the past 40 years have shown co n c l u s i v e l y that Douglas-fir trees can become adapted to s p e c i f i c l o c a l environments within the natural range. Munger and Morris (1936) and Morris et al. (1957) reported observations on the 1912 Douglas-fir progeny t e s t regarding height growth and bud burst of f a m i l i e s from 13 l o c a l i t i e s . S i g n i f i c a n t differences i n height at 20 years were apparent due to seed source with some high a l t i -tude sources showing strong genotype-environment i n t e r a c t i o n . In contrast, differences i n timing of f l u s h i n g were found between plantations and years but the ranking of the seed sources remained remarkably constant from year to year and between the growing s i t e s . The r e s u l t s suggested a strong genetic control over time of bud burst, with sources from areas of warm spring days and nights f l u s h i n g f i r s t , and those from areas of warm spring days and c o l d nights l a s t . Intermediate ranks consisted of sources from s i t e s of cold days and nights. The authors also noted that the time of f l u s h i n g appeared to be unrelated to height growth p o t e n t i a l . The studies of Irgens-Moller (1957, 1958, 1968) and Sorensen (1967) con-firmed that v a r i a t i o n s i n phenology between seedlings of d i f f e r e n t o r i g i n grown under uniform conditions could be r e l a t e d to seed source. For example, coastal low elevation seedlings r a i s e d at C o r v a l l i s , Oregon, were found to f l u s h e a r l i e r and continue height growth longer than seedlings from the i n t e -r i o r of B r i t i s h Columbia and the high Cascades. These dif f e r e n c e s were r e f l e c t e d i n t o t a l height growth and were ascribed to the response to dec-rease i n daylength during the growing season. This was seen as part of the mechanism by which plants become adapted to l o c a l temperature regimes. Ching and Bever,(1960) observed wide v a r i a t i o n i n the timing of f l u s h i n g between seedling groups i n a t e s t i n v o l v i n g 14 seed sources from B r i t i s h - 34 -Columbia, Washington and Oregon, but were unable to r e l a t e these s a t i s f a c -t o r i l y to seed o r i g i n . Later observations i n three of the outplanting areas o f the same t r i a l gave r e s u l t s somewhat s i m i l a r to those of the 1912 study (Walters and Ching, 1966). Within plantations,dates of bud burst v a r i e d from year to year,but ranking of seed sources remained f a i r l y consistent. Between plantations, many sources retained t h e i r r e l a t i v e p o s i t i o n but others showed a marked source-site i n t e r a c t i o n . S i l e n (1962) studied f l u s h i n g i n Douglas-fir c l o n a l material and c a l -culated that 94-96% of the observed v a r i a t i o n was due to a genetic compo-nent. A strong c o r r e l a t i o n was also established between time of bud burst and t o t a l seasonal elongation ,- early f l u s h i n g being associated with more growth. Walters and Soos (1963) stressed that while the pattern of shoot growth i s under strong genetic c o n t r o l , there i s s u b s t a n t i a l v a r i a t i o n i n the i n i t i a t i o n and duration of phenological events between i n d i v i d u a l s which may or may not be due to genetics. Sweet (1965) pointed out that confusion i n the l i t e r a t u r e could a r i s e from the d i f f e r e n t methods used to assess f l u s h i n g time. His r e s u l t s showed that the time d i f f e r e n t i a l between bud burst of l a t e r a l and terminal buds of seedlings could i n i t s e l f be a c h a r a c t e r i s t i c s subject to c l i m a t i c modification through natural s e l e c t i o n . Plants from 30 seed sources within the c o a s t a l region were grown i n a nursery i n New Zealand and assessed for phenological properties during two growing seasons. S i g n i f i c a n t d i f f e r -ences were found between provenances i n bud burst, bud set, t o t a l height and f l u s h i n g d i f f e r e n t i a l . The l a s t two features were s u f f i c i e n t l y c l o s e l y correlated to temperature of the seed o r i g i n to enable checks to be made on the stated o r i g i n s of commercial seedlots. Walters and Ching (1966) - 35 -also measured the l a t e r a l - t e r m i n a l f l u s h i n g d i f f e r e n t i a l i n older proven-ances and found i t to be highly dependent on geographic l o c a t i o n of the p l a n t a t i o n and very v a r i a b l e from year to year. Provenances d i d not show consistent ranking between s i t e s or years. G r i f f i t h (1968) studied phenology and r a d i a l growth of 154 open-grown Douglas-fir trees at the U.B.C. Research Forest over a period of ten growing seasons, from age 18 to 28 years. Average f l u s h i n g date varied from year to year, but the order of ranking f o r i n d i v i d u a l trees remained very consistent within a period of three to s i x weeks. No s i g n i f i c a n t r e l a -t i o n s h i p was found between order of bud burst and height or diameter growth. The current opinion expressed i n the l i t e r a t u r e i s that the timing of bud burst i s a provenance response to l o c a l temperature conditions (Wommack, 1960; Lavender and Hermann, 1970; Campbell, 1974) . The r e s u l t s of Campbell (op cit.) show that predicted dates of f l u s h i n g of l o c a l populations vary on average by 4.9 - 5.5 days for each 1° of l a t i t u d e and 400 feet of elevation. However, the s i g n i f i c a n c e of t h i s environmental e f f e c t must be considered against the within-crop v a r i a t i o n of up to s i x weeks'observed by G r i f f i t h (1968). The r o l e of the tree crown i n mediating the influence of the external environment on wood formation was stressed by Larson (1962). In a l a t e r paper, the same author (Larson, 1969) presented a concept of wood q u a l i t y v a r i a t i o n i n terms of c y c l i c a l patterns of vegetative a c t i v i t y and plant hormone a v a i l a b i l i t y . B r i e f l y , cambial growth i s seen as being i n i t i a t e d by auxins emanating from the developing buds i n spring, and earlywood pro-duction being maintained during shoot expansion and new needle development, i n an environment of high hormone concentration. Cessation of shoot - 36 -elongation and maturation of new needles d i v e r t s food supplies to the vascular cambium and simultaneous reduction of growth substance l e v e l s r e s u l t s i n the formation of smaller diameter, thick-walled c e l l s of the latewood type. Radial growth i s brought to a h a l t by the accumulation of growth-inhibiting compounds o r i g i n a t i n g from the apices. A l l external factors i n f l u e n c i n g rate or duration of xylem formation are considered to have a d i r e c t e f f e c t on shoot growth and thus only i n d i r e c t l y modify cambial a c t i v i t y . The above explanations of wood production have found general accep-tance i n the l i t e r a t u r e , b u t constant refinements are being added by phy-s i o l o g i s t s and biochemists working with growth processes, without a l t e r i n g the fundamental p r i n c i p l e s . The i n t e r a c t i o n s between shoot growth and wood formation have stimulated workers to seek explanations f o r the high degree of tree - t o - t r e e v a r i a b i l i t y i n wood properties i n terms of crown phenology. Mergen et al. (1964) reviewed some of the ea r l y l i t e r a t u r e on the subject and made obser-vations on troNorway spruce plantations growing i n Connecticut. The hypo-thesis to be tested was that e a r l y flushers would have a longer period a v a i l a b l e f o r earlywood production and therefore show lower latewood percentages and mean d e n s i t i e s . This r e l a t i o n s h i p was found to be s i g n i f i -cant i n the older of the two stands and i n 1-year-old vegetative propagules from both crops. C o e f f i c i e n t s of determination presented suggested that f l u s h i n g i n the f i e l d was associated with only about 1% of the v a r i a t i o n i n density and percentage latewood whereas i n the greenhouse up to 25% could be accounted f o r . Worrall (1970), also working with Norway spruce i n eastern U.S.A., found a weak c o r r e l a t i o n between fl u s h i n g date and wood density but showed that times of height growth cessation and r a d i a l growth - 37 -cessation were more c l o s e l y associated with r i n g density. O v e r a l l , 42% of the v a r i a t i o n i n wood density could be accounted f or by using phenologi-c a l v a r i a b l e s . Much of the remaining 58% was thoughtto be due to v a r i a b l e earlywood and latewood d e n s i t i e s . A considerable amount of work has been done on r e l a t i o n s h i p s between time of f l u s h i n g and wood density i n France, p a r t i c u l a r l y with Norway spruce and grand f i r (Abies grandis (Dougl.) L i n d l . ) . Polge (1968) established a s i g n i f i c a n t c o r r e l a t i o n i n the l a t t e r species ; l a t e f l u s h i n g provenances having lower mean d e n s i t i e s . Later work by Lacaze and Polge (1970) resulted i n s i m i l a r observations f o r Norway spruce. In a study of 16 provenances at age 4 years a c o r r e l a t i o n c o e f f i c i e n t of -0.68 was determined f or wood density with f l u s h i n g rank. Within provenances, the r e l a t i o n s h i p also held true. T h i e r c e l i n (1970) selected 18 early flushers and 32 l a t e flushers from a 150-year-old spruce stand i n order to t e s t the above r e l a t i o n s h i p s i n mature trees. Wood formed over a 10-year period was examined densito-m e t r i c a l l y and again the l a t e fl u s h e r s were found to have lower mean density 3 3 (.354 g/cm against .370 g/cm f o r the early f l u s h e r s ) . The di f f e r e n c e was a t t r i b u t e d to lower latewood d e n s i t i e s i n trees f l u s h i n g l a t e , since neither r i n g width nor latewood percentage were aff e c t e d . However, close examination of the i n d i v i d u a l tree data presented shows that the di f f e r e n c e s i n density means between the two fl u s h i n g groups are minor compared to v a r i a t i o n within each group and o v e r a l l bud burst can only account f or a small portion of the observed v a r i a t i o n . A very s i g n i f i c a n t aspect of the above r e s u l t s i s that the nature of the r e l a t i o n s h i p between f l u s h i n g and wood density i n Norway spruce i s quite d i f f e r e n t between locations i n the U.S.A. and i n France ; l a t e flushers - 38 -giving higher density i n the former case and lower density i n the l a t t e r . No discussion of t h i s apparent anomaly has appeared i n the l i t e r a t u r e but i t i s obviously an expression of an important environmental influence on the patterns of shoot growth and cambial a c t i v i t y . McKimmy (1966), i n h i s wood q u a l i t y study of the 1912 Douglas-fir provenance t e s t , compared provenance density ranks with the fl u s h i n g data given e a r l i e r by Morris et al. (1957). High c o r r e l a t i o n s were found 2 between the average f l u s h i n g order and mature wood density (R =0.67). He cautioned, however, that v a r i a b i l i t y was such that poor c o r r e l a t i o n s were obtained when the density values f o r s p e c i f i c 5-year periods were used. The o v e r a l l trend was for early f l u s h i n g trees to be of higher wood density. Similar r e s u l t s were obtained by Kennedy (1970), studying wood density over an 8-year period i n 40-year-old trees. Early f l u s h i n g trees 3 3 showed a mean density of 0.438 g/cm compared to 0.408 g/cm for the l a t e flushers. This difference,however, was s u f f i c i e n t to account f o r only about 27% of the t o t a l v a r i a t i o n i n density despite the f a c t that extreme trees were chosen on the basis of previous f l u s h i n g data ( G r i f f i t h , 1968) . In New Zealand, Harris et al. (1971) reported weak but s i g n i f i c a n t trends toward higher d e n s i t i e s i n early f l u s h i n g trees and expressed the opinion that the r e l a t i o n s h i p may become stronger with tree age. Smith (1973) measured earlywood and latewood widths along the stems of nine early and nine l a t e f l u s h i n g trees growing i n the U.B.C. Research Forest and con-firmed that early flushers tend to have higher latewood percentages at a l l stem l e v e l s (34% as against 27% for l a t e f l u s h e r s ) . Heger et al. (1974) c o l l e c t e d three breast height increment cores from each of 20 trees i n two flu s h i n g groups and compared the years 1960-67 densitometrically. Despite the - 39 -tendency f o r early flushers to have higher mean density the di f f e r e n c e was not s t a t i s t i c a l l y s i g n i f i c a n t at the 5% l e v e l . The ove r r i d i n g source of v a r i a t i o n f o r r i n g density components was again shown to be the i n d i v i d u a l trees within the f l u s h i n g groups. In summary, i t may be stated that i n Douglas-fir there i s a tendency f o r e a r l y f l u s h i n g trees to show s l i g h t l y higher mean density values but that date of bud burst by i t s e l f i s not a s a t i s f a c t o r y i n d i c a t o r of wood q u a l i t y . Within each fl u s h i n g group i t i s s t i l l p ossible to f i n d a very wide range of density values. It has been demonstrated that other pheno-l o g i c a l v a r i a b l e s can also be r e l a t e d to wood density (Worrall, 1970) but that the best r e l a t i o n s h i p s uncovered s t i l l f a i l to account for more than about 40% of the v a r i a t i o n i n density since earlywood and latewood widths and d e n s i t i e s appear to vary from tree to tree and independently of one another. 2.6 X-ray Densitometry The importance of wood density as an index of wood q u a l i t y and i t s s e n s i t i v i t y to weather v a r i a t i o n lead during the l a t e 1950 1s and early 1960's to the development of methods of providing continuous recordings of density across growth ri n g s . A v a r i e t y of approaches were used, ranging from the st y l u s probe of Marian and Stumbo (1960) to l i g h t trans-mission through t h i n sections (Green and Worrall, 1964; E l l i o t t and Brook, 1967) and r a d i a t i o n techniques (Cameron et al., 1959; Polge, 1963). These last-mentioned methods, adapted f o r use with increment core samples, proved to be highly successful and have undergone continuous refinement to become powerful t o o l s i n wood q u a l i t y studies. Within the f i e l d of r a d i a t i o n densitometry, two main groups have evolved, that employing beta rays (Cameron et al. , op. cit.} P h i l l i p s , - 40 -1960; P h i l l i p s et at. , 1962; H a r r i s , 1969) and that pursuing the X-ray technique (Polge, op. ait.; Echols, 1970; Parker, 1972). Although meth-ods of data a c q u i s i t i o n d i f f e r both between and within these groups (Harris and Polge, 1967; Polge, 1969; Parker and Kennedy, 1973) the na-ture of the information i t s e l f i n a l l cases i s s i m i l a r , i. e., a~continuous p r o f i l e of intra-increment wood density. Analysis of such data may involve manual i n t e r p r e t a t i o n of scan l i n e s (Harris, 1969; Harris and B i r t , 1972) or sophisticated on-line computer manipulation (Parker et al., 1973). Such continuous scanning o f f e r s new opportunities and challenges i n growth r i n g a n a l y s i s , the most important of which i s consideration of the latewood concept. T r a d i t i o n a l measures have received wide d i s -cussion i n the l i t e r a t u r e (Edlin, 1965; Harris, 1967) and disadvantages of t h e i r use outlined. However, a u n i v e r s a l l y acceptable c r i t e r i o n f o r describing i n t r a - r i n g density has not yet been developed. " P h i l l i p s (1960) proposed that the earlywood-latewood boundary should be a f i x e d density l e v e l , s i m i l a r to the popular d e f i n i t i o n of Mork (1928). Others considered that,since wood density varies within both earlywood and l a t e -wood zones, the latewood measure should be f l e x i b l e from r i n g to r i n g . (Green and Worrall, 1964; H a r r i s , 1969). Green and Worrall '.op.oitl used the mid-point of the density range between the f i r s t formed e a r l y -wood and the last-formed latewood, ensuring that every r i n g contained some latewood, while Harris op. cit. also used the minimum and maximum den s i t i e s to derive a "latewood r a t i o " . Rudman (1968) and Brazier (1969) maintained that t h i s type of measure, while perhaps useful i n p h y s i o l o g i c a l studies, gave l i t t l e i n d i c a t i o n of technological properties and advocated use of the f i x e d density l e v e l . The main disadvantage of such a c r i t e r -- 41 -ion i s that c e r t a i n types of rings may r e g i s t e r as wholly earlywood or wholly latewood on t h i s basis (Sylvander and Smith, 1973). Some workers have attempted to overcome the problem by using several d i f f e r e n t density l e v e l s f o r each r i n g (McKinnell, 1970; T h i e r c e l i n , 1970). In a -compari-son of various measures of latewood, Cown (1971c) determined that f i x e d density c r i t e r i a gave much superior c o r r e l a t i o n s with mean r i n g density i n both corewood and outerwood of radi a t a pine. Echols (1972,1973) used densitometric scans to compare species i n t r a -r i n g density p r o f i l e s and derived a uniformity index to quantify the dif f e r e n c e s . Radiation densitometry i s s t i l l a r a p i d l y developing f i e l d of study i n which new refinements are constantly appearing. I t has reached the stage where acceptance to standardized methods of analysis would be of great advantage to dendrochronologists and wood technologists i n general. - 42 -Chapter 3. MATERIALS AND METHODS Before embarking on a study of environmental and genetic sources of v a r i a t i o n i n young Douglas-fir trees, i t i s necessary to e s t a b l i s h some of the c h a r a c t e r i s t i c s of the properties to be assessed. In p a r t i c u l a r , i t i s important to have some knowledge of the quantitative and q u a l i t a t i v e aspects of between-tree and within-tree v a r i a t i o n and an appreciation of the extent to which data obtained from young trees can be used to p r e d i c t properties of mature crops. Two preliminary studies were undertaken; one to give information on the v a r i a t i o n of r i n g components i n trees of an age s i m i l a r to those to be examined i n the major studies, the other to investigate juvenile-mature c o r r e l a t i o n s i n plantation-grown Douglas f i r . 3.1 Preliminary Studies A 17-yr-old stand of natural regeneration was selected for intensive study since i t was located adjacent to the U.B.C. Research Forest r e p l i c a -t i o n of the Co-operative Douglas-fir Provenance Test (to be described below) and had been well tended. From each of 60 trees, two 4 mm increment cores were removed from the breast height p o s i t i o n along a diameter selected to minimize the l i k e l i h o o d of including compression wood. I d e n t i f i c a t i o n numbers were written on a l l cores with an i n d e l i b l e copy p e n c i l since previous experience had shown that the l a b e l s thus ins c r i b e d remain l e g i b l e throughout subsequent pro-cessing. - 43 -Extraneous materials within wood, e s p e c i a l l y i n heartwood, have been shown to contribute to anomalous wood density values (Harris, 1969; Parker et al. , 1974), so an extraction procedure was adopted for a l l wood samples p r i o r to density determination. Increment cores were placed i n a Soxhlet apparatus and treated with a 1:2 alcohol-benzene mixture for 72 hr. Basic density values were ca l c u l a t e d for the cores using the maximum moisture content method described by Smith (1954). This involved satura-t i n g the cores by repeated a p p l i c a t i o n of vacuum to the samples while under water for several days. Wet weights were obtained a f t e r removal of surface o water with a damp c l o t h , and the cores d r i e d overnight i n an oven at 105 C. A dessicator was used to carry f i v e cores at a time from the oven; a f t e r dry weights were obtained basic density values were derived from the expression: 1 Basic density = wet wt. - oven-dry wt. 1 oven-dry wt. 1.53 where 1.53 i s the assumed density of the c e l l - w a l l substance (Stamm, 1964). Combined core density values can then be used as an estimate of tree breast height density. To study the r e l a t i o n s h i p between breast height density and whole-tree density, the stems were a r b i t r a r i l y c l a s s i f i e d into f i v e density groups on the basis of the core values, and three trees from each group selected at random f o r f e l l i n g . Cross-sectional d i s c s , 5 cm t h i c k , were removed at 1.5 m i n t e r v a l s up the boles (Fig. 3.1). In the laboratory the d i s c s were divided into two samples, one 1 cm t h i c k f o r l a t e r X-ray a n a l y s i s . The remaining 4 cm disc was used for diameter measurement and, f i n a l l y , two pie-shaped sectors were removed for gravimetric density determination. These blocks were saturated under vacuum for one week p r i o r to wet volume measurement by immersion i n - 44 -Gravimetric density determination X-ray analysis Figure 3.1 Sampling scheme f o r i n d i v i d u a l t r e e s . - 45 -water, then d r i e d i n an oven at 105°C to constant weight. Basic densi-t i e s were c a l c u l a t e d from the conventional expression: „ .^  Oven-dry weight Basic density = — — J , „ !*— Saturated Volume Whole-tree d e n s i t i e s wer.e then c a l c u l a t e d using c r o s s - s e c t i o n a l d i s c areas, d i s c heights and mean d i s c d e n s i t i e s . Of the 15 trees f e l l e d , f i v e were chosen at random f o r a study of the v a r i a t i o n of i n t r a - r i n g width and density components. For t h i s purpose the 1 cm d i s c s retained from the above study were re-cut to y i e l d diametric sample s t r i p s 1 cm i n tangential width and 5 mm thick l o n g i t u d i n a l l y . A f t e r extraction for one week i n 1:2 alcohol-benzene they were prepared for X-ray analysis by procedures to be o u t l i n e d below .,(3.5). The study of juvenile-mature c o r r e l a t i o n s required an older group of sample trees representative of the type of produce to be expected from future plantations under intensive management. Unfortunately, i d e a l study material d i d not appear to be a v a i l a b l e so i t was decided to c o l l e c t increment cores from a stand of trees located on the U.B.C. campus, reputed to be one -of the oldest plantations of Douglas-fir i n B r i t i s h Columbia, having been established j i n 1935. j The crop remains unthinned to date (1974) and has a mean breast height diameter of only about 30 cm. Single 4 mm cores were c o l l e c t e d from 20 trees and extracted i n a Soxhlet apparatus f o r 48 hr with 1:2 alcohol-benzene mixture. Following e x t r a c t i o n the cores were d r i e d , mounted and cut for X-ray a n a l y s i s . Since the number of growth rings varied s l i g h t l y between cores, and the p i t h was absent from the majority, i t was decided to s t a r t scanning from the 1940 r i n g of each core. On average t h i s represented the fourth r i n g outwards from the p i t h . - 46 -3.2 The Co-operative Douglas-fir Provenance Test The importance of Douglas-fir to the economy of the P a c i f i c Northwest, combined with a general lack of knowledge of the sources of v a r i a t i o n within i t s range, lead i n 1954 to the formulation of a plan to set up a comprehensive provenance t e s t (Anon., 1955). The objective was to t e s t the hypothesis that d i s t i n c t races of Douglas-fir existed i n as s o c i a t i o n with geographic and c l i m a t i c variables such as temperature regime and to determine whether the v a r i a t i o n could be described as ecotypic or c l i n a l (Ching and Bever, 1960). Seed was c o l l e c t e d from 16 locatio n s i n B r i t i s h Columbia, Washington and Oregon during 1956-57 and grown i n a nursery at C o r v a l l i s , Oregon. A r e c i p r o c a l planting design was adopted i n which each of the 16 coastal provenances was planted on a s i t e near each of the o r i g i n s , and e s t a b l i s h -ment was c a r r i e d out i n the f a l l of 1959 or spring of 1960. At each location, two blocks c o n s i s t i n g of 32 p l o t s each and generally 1/2 to 1 km apart were l a i d out. Each seed source was r e p l i c a t e d twice within each block, spaced at 2 x 2 m i n the square p l o t s , with 11 seedlings i n each row. This t r i a l i s a useful source of material f o r a study of the c h a r a c t e r i s t i c s of coastal Douglas-fir as they are influenced by seed source and pla n t i n g l o c a t i o n . In contrast to the wood property surveys previously undertaken i n natural stands, provenance t r i a l s o f f e r samples of known o r i g i n s , of even age and at the same spacing, growing under a range of environmental conditions. Thus many of the variables confounding the i n t e r p r e t a t i o n of other studies are e l i -minated . The current study was planned to y i e l d information on the following topics of i n t e r e s t : - 47 -a) the influence of seed source on growth-ring width and density components, b) the influence of geographic l o c a t i o n on growth-ring width and density components, c) the extent of any genotype-environment i n t e r a c t i o n s on growth r i n g components. Previous reports by McKimmy (1966) and McKimmy and Nicholas (1971) on h a l f - s i b progeny from a l i m i t e d number of parent trees have indi c a t e d that both s i t e and genotype can have a s i g n i f i c a n t e f f e c t on average wood density. To date information i s lacking on the v a r i a t i o n of the growth-ring components. 3.2.1 Sample s e l e c t i o n Studies of t h i s type require c a r e f u l thought on the type and numbers of samples to be c o l l e c t e d and processed. Many workers i n the past have opted for intensive examination of r e l a t i v e l y small numbers of stems from several s i t e s (Hughes and A l l e n , 1949; Wellwood, 1952; McKimmy, 1959). How-ever, i n view of the number of reports s t r e s s i n g the magnitude of the between-tree v a r i a t i o n i n wood properties, a larger number of samples from each l o c a t i o n would be more s a t i s f a c t o r y . Increment core techniques developed by the U.S. Forest Service during the 1950's (M i t c h e l l , 1964) enabled a much more e f f i c i e n t c o l l e c t i o n of data from many trees non-destructively, and the r e l a t i o n s h i p s established between breast height and whole-tree properties allowed extrapolation from core data to tree values. The preliminary studies of young Douglas-fir trees proved the v a l i d i t y of employing breast height increment cores i n crops of t h i s age (Chapter 4), - 48 -so i t was planned to use t h i s technique on the provenance t r i a l i n con-junction with gravimetric and X-ray densitometric analyses. In consultation with Dr. Kim K. Ching, Oregon State U n i v e r s i t y , Cor-v a l l i s (Co-ordinator of the Co-operative Douglas-fir Provenance Test), i t was decided that the a v a i l a b l e time would allow the examination of f i v e seed sources each growing at f i v e widely spearated l o c a t i o n s . The d e t a i l s of the samples selected are given i n Table 3.1 and Appendix 1. The p l a n t i n g s i t e s represent a wide range of l o c a l i t i e s found through-out the coastal Douglas-fir region and, with the exception of Mola l l a , the plantations were situated on l e v e l and apparently uniform s i t e s . The Molalla plantings were located about 1 km apart near the bottom of a f a i r l y narrow v a l l e y at an elevation of around 1,100 m. Height growth had been very poor i n comparison to the o t h e r s s i t e s and i n t h i s respect there was also a v i s u a l d i f f e r e n c e between the two plantations. Some d i s t o r t i o n of the stems was apparent and presumed to be due to heavy accumulation of snow during the winter months. In contrast to the other four s i t e s , Dorena Dam had only one pl a n t a t i o n . In order to ensure a d e t a i l e d and balanced s t a t i s t i c a l a n a l y s i s , two 5 mm increment cores were c o l l e c t e d from a minimum of ten randomly selected stems i n the outer two rows (buffer s t r i p s ) of each provenance r e p l i c a t i o n (see F i g . 3.2). The choice of two cores was to enable a more accurate e s t i -mation of the tree mean values, since i t has been shown that there i s often s i g n i f i c a n t c i rcumferential v a r i a t i o n i n wood properties (Cown, 1971a). The cores were removed from the stems at a height of around 1 m from ground l e v e l i n order to obtain the maximum number of growth rings without entering the region of butt swell. Care was taken to avoid compression wood, where posr -s i b l e , by ignoring deformed stems and extracting the cores from a diameter perpendicular to any obvious sweep or lean. Alphanumeric characters - 49 -TABLE I 5.1: Selected S i t e s and Seed Sources Provenance S i t e Latitude Longitude E l e v a t i o n (m ) Nimpkish, B.C. Nimpkish 50°30' 127°20' 120-200 Haney, B.C. Haney 49°10' 123° 0' 150-220. Mo l a l l a , Ore. Mo l a l l a . 45°10' 122° 30' 1,000-1,200 Willamette V a l l e y , Ore. Va l l e y 44°50* 121°40' 60 '." j Butte F a l l s , Ore. 42°20' 800-1,000 Dorena Dam 43 045' 122°58* 250 - 5 0 -P R O V E N A N C E R E P L I C A T I O N © o « © « » e o o ® « •«r« • • • • • •)• • ® e • • ® * 9 0 • • •••••• • •••••• • a ® • • • • 9 • • © Buffer strip Main plot Sampla tra« G R A V I M E T R I C A N A L Y S E S 1 2 3 4 5 6 7 8 9 10 Traas Coras ( 20) D E N S I T O M E T R Y A N A L Y S E S Traas Coras Figure 3.2 Sampling scheme for provenance r e p l i c a t i o n s . - 51 -i d e n t i f y i n g s i t e , p l antation, provenance,' r e p l i c a t i o n , tree and core numbers were i n s c r i b e d on the samples with i n d e l i b l e copy p e n c i l . Cores from each p l o t were t i g h t l y wrapped i n p l a s t i c bags and stored i n a cool place to keep them green and fresh. In a l l , over 2,000 such cores were c o l l e c t e d during the f a l l and winter of 1974. 3.2.2 Gravimetric analyses Cores representing ten trees per p l o t were selected f o r measurement on the basis of q u a l i t y . That i s , they were screened f o r absence of such defects as p i t c h pockets, knots, compression wood, and proximity to the p i t h . Bark and cambial zones were c a r e f u l l y removed from the outsides and excess wood at the p i t h and excised with a razor blade. At t h i s stage i t was apparent that the samples from the M o l a l l a s i t e d i f f e r e d from the others i n that i t was p r a c t i c a l l y impossible to s e l e c t cores e n t i r e l y free from compression wood. This feature was so abundant that i t could be expected to s i g n i f i c a n t l y a f f e c t the density r e s u l t s . The lengths of these green prepared cores were measured to the nearest 0.1 mm with the a i d of a set of d i a l gauge c a l i p e r s and the number of growth rings counted. These data were entered d i r e c t l y onto IBM coding forms along with the i d e n t i f i c a t i o n codes. With such large numbers of cores to be processed, the most e f f i c i e n t means of density determination i s c l e a r l y the maximum moisture method (Smith, 1954). Wet and dry weights, obtained as described previously, were entered on to the IBM coding forms alongside the other core data. Two computer programmes were developed to process the gravimetric data as follows: - 52 -Programme 1 To read i n the basic raw data from cards and convert i t to a u s e f u l form, i.e. c a l c u l a t e diameter, mean r i n g width and basic density for each core, and summarize by trees. The i n d i v i d u a l core and tree data were stored i n separate d i s c f i l e s for conven-ience . Programme 2 To summarize output from Programme 1 into s p e c i f i e d groups, e.g. s i t e s , provenances, plantations, p l o t s by screening the i d e n t i f i c a t i o n codes. For each group, means, standard devia-t i o n s , standard errors, minima and maxima were c a l c u l a t e d for the growth rate and density data and l i s t e d on a l i n e p r i n t e r . Such information i s u s e f u l f or graphical presentation of the r e s u l t s . S t a t i s t i c a l analyses of the gravimetric data were c a r r i e d out using the U.B.C. MFAV analysis of variance/covariance programme. 3.2.3 Densitometric analyses The preparation and scanning of X-ray negatives from wood samples i s a time-consuming process and normally only small numbers are involved i n any one study. In t h i s case, however, i t was decided to take advantage of the randomized block design of the t r i a l and s e l e c t a subsample of cores from each p l o t f or densitometric a n a l y s i s . In t h i s way, any microsite v a r i a t i o n could be allowed f o r . Five cores per r e p l i c a t i o n were selected from those which had been used - 53 -i n the gravimetric study (see F i g . 3.2), based on the following c r i t e r i a : a) f i v e d i f f e r e n t trees had to be represented b) cores had to be free from obvious defects c) p i t h had to be present where possible to enable scanning of the en t i r e cross section In a l l , 450 cores were prepared and scanned by the methods to be described l a t e r . 3.2.4 C l i m a t o l o g i c a l analyses The densitometric analyses o u t l i n e d above enabled r i n g component sumi maries to be c o l l a t e d for each planting s i t e . These data can be used i n conjunction with l o c a l weather records to study the e f f e c t s of s p e c i f i c environmental factors on earlywood and latewood widths and d e n s i t i e s . Unfortunately, not a l l the provenance t r i a l s i t e s selected had meteoro-l o g i c a l stations nearby, but i n a l l cases records from the nearest stations were used. Appendix 3 gives the sources of weather data and approximate distances from the plantations. For each s i t e , values f o r monthly minimum, maximum and average temperature and t o t a l p r e c i p i t a t i o n were obtained for the period 1968-1974. The outer f i v e growth rings only were to be used i n order to avoid the problems of inherent v a r i a t i o n within the rings c l o s e s t to the p i t h . By making the assumptions that weather e f f e c t s are s i m i l a r between a l l s i t e s and that the 25 growth periods (5 s i t e s x 5 years) are independent, regression analyses can be performed using the ring-component and monthly weather data with 25 observations. In order to overcome, at l e a s t p a r t l y , the obvious bias i n using data from stations up to 100 km from the actual plantations, i t was decided to - 54 -work e n t i r e l y with indices rather than the actual data. Simple l i n e a r regression analyses were performed on indi c e s f o r r i n g components and meteorological observations based on deviations from the 5-year averages for each l o e a t i o n . A 12-month e f f e c t i v e season was assumed, ranging from October of the previous calendar year to September of the current year. Previously,such analyses have been mainly concerned with associations between weather and growth-ring v a r i a b l e s with a view to improving paleo-c l i m a t o l o g i c a l techniques ( F r i t t s et al. , 1971; Parker and Henoch, 1971). Thus, c o r r e l a t i o n c o e f f i c i e n t s were the most useful s t a t i s t i c s to be assessed. In the current study, however, the r e l a t i v e e f f e c t s of v a r i a t i o n s i n monthly weather were to be examined, i n which case the regression c o e f f i c i e n t s are of more value. A method of presenting the r e s u l t s g r a p h i c a l l y was devised i n which the percentage changes i n r i n g component values are given i n terms of a change of one standard deviation i n the independent v a r i a b l e . This takes account of the fa c t that some months are more va r i a b l e than others. 3.3 Clonal Study As noted above (Chapter 2.2), c l o n a l material o f f e r s unique opportuni-t i e s i n genetic studies since the observed v a r i a t i o n s i n such non-segregating populations are assumed to be of wholly environmental o r i g i n , thus enabling simple c a l c u l a t i o n of genetic and environmental variances. H e r i t a b i l i t y estimates used i n p r e d i c t i o n of genetic gain are often obtained from exami-nation of clones. Propagation of Douglas-fir by rooting of cuttings has not proved s u f f i -c i e n t l y successful to be used i n breeding programmes, but the g r a f t i n g of scion material to nursery rootstock i s standard p r a c t i c e i n the establishment fof seed orchards. However, breeding methods are designed f o r e f f i c i e n t - 55 -seed production and do not always provide s u i t a b l e circumstances f o r h e r i t a b i l i t y studies. Ideally, wood properties should be assessed on mature clones, well r e p l i c a t e d within a range of t y p i c a l f o r e s t environ-ments. Since published information on the extent of genetic c o n t r o l of growth-r i n g components i s almost e n t i r e l y lacking f o r Douglas-fir, i t was decided to use densitometric analyses to study genetic and environmental influences on wood formation i n c l o n a l material. Ring components have already been shown to vary s i g n i f i c a n t l y between geographical areas and within popula-t i o n s , and clones can be used to t e s t the hypothesis that genetics c o n t r i -bute a large p o r t i o n of the observed v a r i a t i o n . I t was also considered desirable to c o l l e c t some phenological information to examine any r e l a t i o n -ships with wood prope r t i e s . 3.3.1 Sample s e l e c t i o n The a v a i l a b i l i t y of Douglas-fir clones i n B r i t i s h Columbia i s such that only one group was considered at a l l s u i t a b l e for c o l l e c t i o n of wood samples. Scions from lower mainland and Vancouver Island plus-trees had been c o l l e c t e d from 1957 onwards and grafted to rootstock at the B.C. Forest Service Cowi-chan Lake Experimental Station i n l i n e s of up to 15 ramets (Orr-Ewing and S z i k l a i , 1960). Graft incompatability has been a continuing problem (Heaman, 1966), such that of the clones established up to 1960, only feen could be found with eight or more-apparently healthy ramets.- The o r i g i n s of these plus-trees-are given i n Appendix 4. During October, 1973, two d i a m e t r i c a l l y opposed 4 mm increment cores were c o l l e c t e d from each of 80 ramets (10 clones x 8 ramets), l a b e l l e d with i n d e l i b l e p e n c i l , and t i g h t l y wrapped i n p l a s t i c bags for ease of transport. - 56 -In the laboratory, core lengths and r i n g numbers were recorded so that growth rate comparisons could be made. The average number of complete annual rings a t breast height proved to be nine, but v a r i e d from eight to twelve, thus growth rates were ca l c u l a t e d as the mean r i n g width of the f i r s t nine rings to f a c i l i t a t e comparisons. A f t e r e x t r a c t i o n i n a Soxhlet apparatus for 72 hr i n 1:2 alcohol-benzene, basic d e n s i t i e s were determined by the maximum moisture content method (Smith, 1954). The cores were then prepared for X-raying and the f i l m negatives scanned i n the manner described i n Chapter 3.4 When preliminary examination of the data revealed s u b s t a n t i a l d i f f e r -ences between clones i n mean core density values, i t was decided to carry out a small-scale phenological study i n an attempt to reveal some of the underlying r e l a t i o n s h i p s . A group of s i x clones was selected to represent three a r b i t r a r y density c l a s s e s , low, medium and high and f i v e healthy ramets chosen for measurement of diameter and shoot growth during the 1975 growing season. Bi-weekly readings of two marked diameters were taken, with the a i d of d i a l gauge c a l i p e r s accurate to 0.05 mm, from the middle of A p r i l u n t i l the end of September. The same instrument was used to record the lengths of shoots on four l a t e r a l branches accessible from ground l e v e l . One set of readings was taken on A p r i l 18th of the length of the unexpanded buds, then bi-weekly measurements were c o l l e c t e d from May 30th when the f i r s t signs of f l u s h i n g were observed. Flushing c h a r a c t e r i s t i c s of the clones were assessed by c l a s s i f y i n g each ramet into one of s i x groups describing the progression of bud burst over the whole crown at each measurement date (Appendix 5). Increment core samples of the 1975 growth r i n g were removed (two per ramet) i n l a t e November for densitometric analyses. - 57 -3.3.2 Analyses In the analyses, data for the two cores from each ramet were pooled to give more r e l i a b l e r e s u l t s since they were c o l l e c t e d from one diameter and cannot be treated as wholly independent samples. The two growth zones recognized i n the preliminary studies (Chapter 4.3) were analyzed separately by bulking rings 1-5 from the p i t h and growth periods 1969-1973. This res u l t e d i n r i n g 5 (1969 i n most cases) being included i n both groups, but i t was considered that the advantage i n balancing the numbers of rings would overcome any disadvantages. H e r i t a b i l i t y estimates were obtained from analyses of variance on the gravimetric values (ring width and density only) and the densitometric data ( a l l r i n g components). In the l a t t e r case, analyses were performed a) on the i n d i v i d u a l growth periods, and b) on arithmetic means f o r the 5-year groups. In each case, the model used was as follows: Source of V a r i a t i o n d.f. Between clones 9 Within clones _70 Total 79 H e r i t a b i l i t y estimates were then c a l c u l a t e d on an i n d i v i d u a l tree basis, a S S V 2 2 / , 2 2 X h = a / 0 + a c c T On a clone mean basis the estimates become: h 2 = a 2 / ( a 2 + a 2/8) c c T 3.4 D i a l l e l Cross Study Studies of the v a r i a t i o n of wood density normally assume that t h i s property behaves as a metric character (Falconer, 1964) responsive to simple Expected Mean Square 2 ^ 2 a + 8 a T c 2 a T - 58 -s e l e c t i o n procedures. Few workers have dealt with parent-progeny r e l a t i o n -ships i n c o n t r o l l e d - p o l l i n a t e d material to determine general and s p e c i f i c combining a b i l i t i e s . Such work would y i e l d information on the mode of inheritance of wood density. Previous studies at the Univ e r s i t y of B r i t i s h Columbia have examined v a r i a t i o n and inheritance of some phenological c h a r a c t e r i s t i c s on stems located on the campus. Four such trees were selected f o r intensive study by S z i k l a i (1964) and, following c o n t r o l l e d p o l l i n a t i o n , seedlings from a complete d i a l l e l cross (except for one s e l f - s t e r i l e combination) were out-planted at the U.B.C. Research Forest at a spacing of 1 m x 1 m. The four parent trees (designated A, B, E and 11) are described i n d e t a i l by S z i k l a i (op oit.^i . A summary of some relevant information i s given i n Appendices 6(a) and 6(b) . Increment core samples were removed from the parent trees at four p o s i -tions around the breast height l e v e l , i n order to allow f o r some of the exposed c i r c u m f e r e n t i a l v a r i a t i o n i n wood density (Cown, 1971a). The ex-tracted cores were scanned densitometrically and summaries of the previous 25 growth rings used to characterize the outerwood properties of the i n d i -v i d u a l stems. Since i t was considered that varying environmental factors may possibly have had some influence on the absolute density l e v e l s , the i n t r a - r i n g p r o f i l e s were,given considerable a t t e n t i o n i n e f f o r t s to e l u c i -date parent-offspring properties. Seedlings from the c o n t r o l l e d crosses were planted i n the f i e l d between 1959 and 1963, but lack of c u l t u r a l treatment has lead to severe suppression of vigour.. Heights i n 1974 ranged from 6 - 10 m while diameters averaged only about 60 mm. Mo r t a l i t y , however, had been low. Progeny were randomly sample'd up to a maximum of eight trees per group, - 59 -but lack of replication restricted the numbers obtainable in several groups (Appendix 6c). In all,representatives from 12 out of the theoretical maxi-mum of 16 groups were available. From each of these, two increment cores were removed from about the 0.5 m level in order to obtain the maximum number of growth rings without being influenced by any butt swell effect. The cores were extracted and processed densitometrically as described below. 3.5 Densitometric Techniques X-ray densitometric scanning of increment core samples was carried out at the Canadian Forestry Service Western Forest Products Laboratory, Vancou-ver, using techniques developed specifically for wood quality assessment by members of the staff (Parker and Jozsa, 1973; Parker et al. , 1973). A brief outline of the procedures i s as follows: 1. The increment core samples were prepared for sawing to a uniform thickness by glueing them between pre-shaped supports in such a way that cross-sectional surfaces were exposed by the saw cuts (Parker and Meliskie, 1970). Experience showed that mounts made from yellow cypress (Chamaecyparis nootkatensis (Lamb.) Spach.), being even-grained and less l i k e l y to cause overheating of the saw blades, were more suitable than those from species with resin canals. 2. The cores and mounts were reduced to a thickness of 2 mm by passing them through a specially designed twin-blade saw (Kusec, 1972). Identification characters were then written on the mounts adjacent to the cores with an X-ray opaque lead-based paint. 3. The prepared samples were placed on X-ray film along with calibration wedges constructed from Douglas-fir groundwood pulp and passed under - 60 -a collimated X-ray source (20 kV, 2.5 mA) on a hydraulically-driven carriage (Parker and Jozsa, 1973). 4. The X-ray films were developed according to a s t r i c t l y controlled schedule to ensure uniform characteristics both within and between films (Parker and Jozsa, 1973). 5. The film negatives were scanned on a custom-built densitometer and information recorded on magnetic tape using a highly sophisticated interactive system (Parker et al., 1973). Density values obtained from this system depend on the values assigned to the steps of the calibration wedges and have previously been expressed i n oven-dry density units. The present author is of the opinion that basic density is a more meaningful measure since i t i s the most commonly used c r i -terion i n the literature and allows direct comparisons with gravimetrically-determined values. Consequently, basic density values were used in calibra-tion in a l l studies reported here. The programme used in the on-line computer for data acquisition was one developed by Messrs. Parker and Bruce at the Western Forest Products Labora-tory but included a refinement to overcome any possible non-linearity in the voltage-density relationship on which results are based. This was con-sidered necessary since i t had been shown that fluctuations in film quality could lead to significant non-linearity. A parabolic model was therefore adopted for wedge calibration since this proved to give highly satisfactory results in both linear and curvilinear cases. Each annual ring scanned was treated separately by the programme as follows: - 61 -a) 8 r i n g parameters were ca l c u l a t e d according to a pre-selected earlywood-latewood boundary l e v e l based on wood density. A 3 l e v e l of 0.50 g/cm (basic) was chosen to correspond roughly 3 to the 0.54 g/cm (oven-dry) used i n previous reports (Parker et dl. , 1973). The r i n g c h a r a c t e r i s t i c s stored on tape were: 1. Ring width (RW) 5. Earlywood density (ED) 2. Earlywood width (EW) 6. Latewood density (LD) 3. Latewood width (LW) 7. Minimum density (MND) 4. Mean r i n g density (RD) 8. Maximum density (MXD) b) A 100-point density p r o f i l e of the r i n g was obtained from the actual density configuration by means of a subroutine and stored on tape. The magnetic tapes produced at the Western Forest Products Laboratory on the HP 2100 A serie s computer were incompatible with the U.B.C. IBM 370 computer without modification, so a programme had to be constructed to copy a l l raw data to other tapes i n such a form as to be useable at U.B.C. A s e r i e s of computer programmes were written to summarize the data. Separate programmes were developed to handle the r i n g summary parameters and the density p r o f i l e data, each having the option of t r e a t i n g the r i n g data by r i n g number from!the p i t h or by year of r i n g formation. These general summary programmes produced means f o r tre e groups according to input information s p e c i f y i n g the number of groups (e.g. s i t e s , plantations, pro-venances, or p l o t s ) , the number of trees per group and the number of cores per tr e e . The summaries included not only the data r e t r i e v e d d i r e c t l y from the densitometer tapes but al s o three derived c h a r a c t e r i s t i c s thought to be of importance i n wood q u a l i t y studies. These are percentage latewood - 62 -(PLW, calculated from LW/RW) and two properties r e l a t e d to i n t r a - r i n g density contrast, designated as uniformity (UNI, derived from LD-ED) and range (RNG from MXD-MND). These are shown diagramtically i n Figure 3.3. Graphical presentation of the density p r o f i l e data was accomplished by f i r s t averaging according to groups of i n t e r e s t , s t oring the summaries on d i s c f i l e , and p l o t t i n g on the U.B.C. CALCOMP p l o t t e r . Although these graphs are very useful f o r examining year-to-year v a r i a t i o n within samples, they do not always provide a good means of comparing p r o f i l e shapes of d i f f e r e n t sample groups. For t h i s purpose the p r o f i l e data were used to 3 construct r e l a t i v e frequency histograms using a c l a s s i n t e r v a l of 0.05 g/cm . This method of presentation i s a modification of a technique used by Echols (1971), and w i l l henceforth be r e f e r r e d to as a standard densigram. When placed side by side, they give a good v i s u a l i n d i c a t i o n of the intra-increment density d i s t r i b u t i o n patterns. The U.B.C. MFAV (analysis of variance/covariance) was widely used i n data analyses. Legend RW = Ring width mm EW = Earlywood width mm LW = Latewood width mm PLW = Percent latewood % RD = Ring density g/cm" ED = Earlywood density '!> LD = Latewood density » MND = Minimum density » MXD = Maximum density » UNI = Uniformity » RNG = Density range •• Figure 3.3 D e f i n i t i o n of r i n g components. - 64 -Chapter 4. RESULTS AND DISCUSSION - PRELIMINARY STUDIES 4.1 Between-tree V a r i a t i o n i n Wood Density Of the 60 trees sampled at the U.B.C. Research Forest, 56 were reta i n e d f o r analyses, the other four having been discarded on the basis of poor sample q u a l i t y (fractured cores or presence of knots). Mean breast height density estimated from the two cores per tree was determined to be 0.3 94 3 g/cm with a standard deviation of 0.022 and a range of 0.341 to 0.448. The d i s t r i b u t i o n of tree values was approximately normal so that these data were used for c a l c u l a t i o n of the sample numbers required to obtain estimates of mean density to w'ithin s p e c i f i e d l i m i t s of accuracy. Freese (1962) out l i n e d an i t e r a t i v e process f o r determining sample numbers i n an i n f i n i t e population without replacement using the r e l a t i o n s h i p : 2 2 n - t S y  11 " "I E 2 where n i s the minimum number of samples, Sy i s the estimated variance of the character being observed, E i s the desired confidence i n t e r v a l and t i s the student's t value based on (n-1) degrees of freedom and the selected p r o b a b i l i t y l e v e l . The accuracy required n a t u r a l l y depends upon the objectives of the study, and the current author considers that, f o r comparisons of s i t e s or groups of trees, the confidence i n t e r v a l should be within at l e a s t ±5% of the mean,i.e. 3 3 ±0.02 g/cm i f a mean of around 0.40 g/cm can be assumed. A conservative 3 confidence i n t e r v a l of 0.015 g/cm was used i n the above formula, which at a p r o b a b i l i t y l e v e l of 0.05, gave a best estimate of n which was calculated to be around 10 to 11 trees. A good working p r i n c i p l e should therefore be - 65 -to aim f o r a minimum of 10 good samples to represent each experimental u n i t where the above conditions are appropriate. In p r a c t i c e , t h i s w i l l normally involve the c o l l e c t i o n of cores from 12 to 15 trees to allow for discards due to defects. 4.2 Relationship between Breast Height Density and Whole-tree Density Several authors'.have previously demonstrated the v a l i d i t y of using increment cores to estimate whole-tree density, employing varying numbers and portions of cores ( M i t c h e l l , 1964; Wahlgren and Maeglin, 1966; Smith and Wahlgren, 1971). The U.S. Forest Service survey of 1965 used multiple regression methods to determine which of several gross tree c h a r a c t e r i s t i c s i n a d d i t i o n to core density could improve the estimate of o v e r a l l density. For Douglas-fir of varying ages, the i n c l u s i o n of breast height diameter 2 increased the c o e f f i c i e n t of determination (R ) from 0.53 to 0.61, so that both v a r i a b l e s were used f o r p r e d i c t i v e purposes. Density values for the discs removed from the 15 f e l l e d trees are shown i n Appendix 7 along with the breast height increment core density and weighted whole-tree density data. Without exception, the i n d i v i d u a l trees show a trend of decreasing density upwards i n the stem. The increment core values are seen to be c o n s i s t e n t l y higher than the d e n s i t i e s of d i s c No. 2, 3 which corresponds to the 1.5 m height, by about 0.02 gg/cm . Such d i f f e r -ences are normally regarded as being a r e s u l t of compression of the sample by the increment borer, although i t i s also possible that some unremoved chemicals from the extraction process could contribute to the e f f e c t (Northcott et al. , 1964). The c o r r e l a t i o n matrix given i n Appendix 8 shows very good r e l a t i o n -ships between the d e n s i t i e s of d i s c s up to 6.0 m i n the sample trees, i n d i -cating a high degree of consistency i n within-tree density despite inherent - 66 -v a r i a t i o n patterns. The r e l a t i o n s h i p between increment core and whole-tree values gave an 2 R estimate of 0.83 (Fig.4 *Dwhich i s s u b s t a n t i a l l y greater than reported previously (U.S. Forest Service, 1965). This highly s a t i s f a c t o r y r e s u l t i s no doubt a r e f l e c t i o n of the uniform age and stand conditions i n the sample area. I t was not considered worthwhile to include tree diameter i n the regression equation since the objective of the study was not to provide a p r e d i c t i v e function but to confirm the v a l i d i t y of sampling populations by means of increment cores. Besides, i t might be expected that the actual r e l a t i o n s h i p would vary q u a n t i t a t i v e l y from s i t e to s i t e , even i n crops of the same age. The data presented here i l l u s t r a t e that v a r i a t i o n s i n breast height density are c l o s e l y r e l a t e d to v a r i a t i o n s i n whole-tree density within one crop, so that increment cores provide a good basis for i n d i v i d u a l tree comparisons. 4.3 Within-tree V a r i a t i o n i n Tree-ring Components Chapter 2 dealt with reports i n the l i t e r a t u r e r e f e r r i n g to v a r i a t i o n of t r e e - r i n g components i n Douglas-fir. Apart from the work of Harris and Orman (1958) and Harris (1969>, very l i t t l e mention of e i t h e r width or density components i s to be found i n the context of b i o l o g i c a l v a r i a t i o n . In the current study two wood'strips. from each rof s i x d i s c s up the stems of f i v e randomly selected trees from the U.B.C. Research Forest were analysed densitometrically. The data were summarized both by r i n g number from the p i t h and by year of formation,which revealed the presence of two well-defined features: - 67 -. 4 4 I 1 1 | 1 1 r o - 3 2 h .30 < ' ' 1 '—: 1 ' 4 • 3 4 - 3 6 . 3 8 - 4 0 - 4 2 . 4 4 . 4 6 - 4 S Breast H « i g h t Increment C o r e Density g / c m Figure 4.1. Relationship between breast height increment core density and weighted whole-tree density. - 68 -a) a strong c l i n a l trend from the p i t h out to about r i n g no. 5 at a l l stem l e v e l s , independent of the a c t u a l year of wood formation, and , b) an apparently random f l u c t u a t i o n a f t e r the 5th r i n g , but f o l -lowing a consistent pattern within and between trees according to year of formation. Since the numbers of growth rings present i n the discs from any one stem l e v e l varied s l i g h t l y from tree to tree, standardized graphs were con-structed to show both the above trends. These''are shown i n F i g s . 4.2(a-k), adjusted to accommodate the average number of rings at each l e v e l . The v e r t i c a l sequences within growth rings are stressed by the i n c l u s i o n of some hatched l i n e s outside the c e n t r a l zone. Earlywood width (EW), latewood width (LW) and r i n g width (RW) a l l showed the trends discussed above at the stem l e v e l s sampled, the increase from r i n g 1 to r i n g 5 being of the order of 300% for EW and RW and 150% for LW. In general, f l u c t u a t i o n s i n r i n g width components thereafter appeared to show no o v e r a l l trend i n s p i t e of the proximity to the p i t h . Mean r i n g density (RD) decreased i n the rings adjacent to the p i t h then fluctuated around a more or l e s s constant l e v e l . The other density components shown i n F i g s . 4.2 (e-j) generally exhibited s i m i l a r trends of more ra p i d change near the p i t h . V a r i a t i o n s i n percent latewood (PLW) followed very c l o s e l y those of r i n g density. V e r t i c a l l y i n the stems, the patterns remained q u a l i t a t i v e l y s i m i l a r although the quantitative r e l a t i o n s h i p often changed s l i g h t l y . Table 4.1 gives means f o r the r i n g components at each stem l e v e l sampled and c l e a r l y demonstrates a decrease i n RD and PLW with height. Closers;examination of the data revealed that these trends are influenced most by decreasing percent Figure 4.2 Ring component patterns within the sample trees. Figure 4.2 (contd.) Ring component patterns within the sample trees. 71 -10 "i ' i D-6 I I r -• (j) Rang« i—i—i 1 — i — i -3 5 '07 '69 '71 '73. Ring No. I Year Figure 4.2 (contd.) Ring component patterns within the sample trees. 71. -i i i • 0-6 I ( i ) U n i f o r m i t y r ' V — i — i — i — i — i — i -10 •5 T — — l "T~ D-6 - i 1 1 r-1 ( j ) Range E 0-5 i \ i \ - i — i — H \ 0-3 i'X' ' * N •— D-1 | \ \ v v 1 3 5 Ring No. ! igure 4.2 (contd.) Ring component patterns'within.the sample trees. TABLE 4.1: Summary of Ring Components by Stem Levels COMPONENTS Disc ' NO. Height M. No. of Rings RW mm EW mm LW mm PLW o, "o RD g/cm ED . - 3 . g/cm . LD 3 . g/cm MND g/cm MXD . 3 ' g/cm UNI „ „ / 3 9/crn RNG „ •/ 3 g. /cm 1 0.2 13 559 358 201 36.0 .452 .327 .642 .271 .783 .315 .512 2 1.5 11 565 377 188 33.3 .445 .320 .686 .259 .858 .366 .599 3 3.0 10 621 430 191 30.6 .443 .330 .681 .273 .840 .351 .567 4 4.5 8 651 483 168 25.8 .424 .337 .665 .281 .833 .328 .552 5 6.1 6 635 501 134 21.1 .389 .320 .647 .255 .764 .327 .509 6 7.6 4 548 445 103 18.8 .393 .347 .585 .260 .676 .238 .416 Component abbreviations as given i n F i g . 3.3. - 73 -latewood i n the inner f i v e rings rather than by the increasing proportion of such wood (corewood). Between 0.2 m and 6.1 m, PLW for the rings adjacent to 3 3 the p i t h decreased from 51.7% to 21.1% and RD from .498 g/cm to .392 g/cm . Within t h i s zone the other density components remained r e l a t i v e l y unchanged. The v a r i a t i o n i n PLW i s accounted for by an increase i n earlywood width with tree height, since the latewood component remained almost constant between tree l e v e l s . A more complete i n d i c a t i o n of the h o r i z o n t a l r i n g component patterns i s given by the breast height samples from the 40-year-old U.B.C. campus trees (Figs. 4.3 - 4.4). These graphs were not standardized to Separate, p i t h trends and c l i m a t i c e f f e c t s since some rings were missing where cores d i d not pass through the centre of the tree, but the o v e r a l l patterns are c l e a r . Ring-width components (EW, LW and RW) decreased from the p i t h outwards, but more r a p i d l y i n the zone formed p r i o r to about 1950, i . e . w i t h i n the f i r s t 15 rings from the p i t h . Some density components (MND, ED) remained v i r t u a l l y unchanged with age whereas others showed d e f i n i t e trends. Ring density v a r i a -t i o n p a r a l l e l e d that of PLW i n t h a t i t increased out to about the 15th r i n g and thereafter fluctuated with no d e f i n i t e pattern. LD and MXD showed signs of increase out to the bark but with greater rates of change i n the f i r s t 15 rings. Trends i n uniformity (UNI) and range (RNG), not shown here, followed those of the latewood parameters. The two groups of samples together enable some conclusions to be drawn regarding v a r i a t i o n of t r e e - r i n g components: 1) Between trees, consistent and s i g n i f i c a n t d ifferences i n wood c h a r a c t e r i s t i c s can occur. 2) Within trees, predictable patterns of v a r i a t i o n e x i s t . I f corewood i s defined as the zone within which the maximum 10.0 8.0 2.0 0.0 \ I \ P L W L W E W 1940 1950 1960 Y e a r 1970 Figure 4.3 Radial patterns of width-component v a r i a t i o n i n 40-year-old trees. I I I i 1 1 ! ' — 1940 1950 1960 1970 Year Figure 4.4 Radial patterns of density component v a r i a t i o n i n 40-year-old stems. - 76 -recognizeable b i o l o g i c a l v a r i a t i o n occurs, i t could be considered to extend to about the 15th r i n g from the p i t h . 3. Within the corewood zone as defined above, two d i s t i n c t regions can be i d e n t i f i e d , i.e. rings 1-5 from the p i t h where strong inherent patterns dominate, and r i n g 6 outwards. 4. Corewood properties are not constant up the stem, as i l l u s t r a t e d i n the decreasing density values i n rings 1-5 between the stump and 6.1 m l e v e l s . 4.4 Juvenile-mature Correlations A summary of the r i n g component c h a r a c t e r i s t i c s of the 40-year-old U.B.C. campus trees i s given i n Table 4.2. To compare the i n t r a - r i n g properties at various stages of growth, the i n d i v i d u a l tree data were handled i n seven 5-ring groups from 1940 outwards. Cross c o r r e l a t i o n s f or the means of a l l r i n g groups were ca l c u l a t e d , together with meaningful aggregations of groups. Wood formed during the period 1940-49 can be considered as the corewood, for which the r e s u l t s are presented i n Table 4.3. In general, i t can be seen that the density components were much more r e l i a b l e f o r p r e d i c t i o n than e i t h e r EW, LW or RW. This would'have been a n t i -cipated since r i n g widths are much more susceptible to environmental modifi-2 cation than density values. However, the very low R estimates for c o r r e l a -t i o n of corewood and outerwood widthi data are s u r p r i s i n g for plantation-grown material. Better c o r r e l a t i o n s were usually obtained when more rings were used i n the assessment, the best r e s u l t s being for corewood and the e n t i r e cross - 77 -TABLE 4.2: Summary of Mean Ring Components for 20 U.B.C, Campus Trees. RING COMPONENTS RW EW LW PLW RD ED ; LD MND MXD UNI RNG Mean 322 177 145 45.0 .527 .297 .793 .238 .902 .496 .663 Minimum 223 110 76 21.8 .413 .258 .712 .200 .693 .404 .429 Maximum 437 273 217 56.1 .628 .338 .934 .283 .984 .605 .735 s.d. 61 50 38 8.3 .057 .023 .055 .024 .072 .047 .071 c. v.% 18.9 28.2 26.2 18.4 10.8 7.7 6.9 10.1 8.0 9.5 10.7 s.d. = standard deviation c.v.% = coefficient of variation, % Component abbreviations and units as explained i n F i g . 3.3. - 78 -TABLE 4.3: Coefficients of Determination for Regression Analyses of Juvenile-mature Relationships in 40-year-old Trees. Intra-increment Variable -Coefficient Determination, % Regression RW EW : LW PLW MND ED RD LD MXD UNI RNG 1970-73 on 1940-44 5 1 0 14 33 10 19 12 38 3 26 1970-73 on 1940-49 9 3 1 7 23 13 17 7 40 2 23 1970-73 on 1940-44 0 0 1 8 30 23 20 11 45 6 28 1955-73 on 1940-44 10 0 0 29 41 14 49 37 65 14 49 1965-73 on 1940-49 14 2 2 19' 28 17 48 32 65 11 23 1965-73 on 1940-54 0 0 4 20 37 28 51 35 65 18 28 1960-73 on 1940-44 9 0 1 38 42 16 63 60 76 31 65 1960-73 on 1940-49 11 0 1 34 32 20 64 55 77 27 61 1960-73 on 1940-54 1 4 6 37 42 32 68 61 80 38 65 1940-73 on 1940-44 1 38 12 62 65 44 83 81 83 58 80 1940-73 on 1940-49 3 41 18 75 65 54 89 80 89 61 . 83 1940-73 on 1940-54 5 70 63 81 76 68 93 89 95 76 90 Component abbreviations as given i n F i g . 3.3. - 79 -s e c t i o n . I t i s n o t e w o r t h y t h a t an i n c r e a s e i n t h e number o f r i n g s i n t h e 2 o u t e r w o o d p o r t i o n has a much g r e a t e r i n f l u e n c e on R v a l u e s t h a n an i n c r e a s e i n t h e corewood r e g i o n . In f a c t , f o r most d e n s i t y components , c o r r e l a t i o n s b a s e d on t h e i n n e r r i n g g r o u p a r e o n l y m a r g i n a l l y p o o r e r t h a n t h o s e u s i n g t h e t h r e e i n n e r m o s t g r o u p s . The p r o p e r t i e s most amenable t o p r e d i c t i o n f r o m • j u v e n i l e v a l u e s a r e RD, L D , MXD and RNG, no d o u b t a r e f l e c t i o n o f t h e g r e a t e r b e t w e e n - t r e e v a r i a t i o n 2 i n l a t e w o o d t h a n i n e a r l y w o o d c h a r a c t e r i s t i c s . R v a l u e s f o r t h e above components r a n g e between 0 .80 and 0 .83 f o r c o r r e l a t i o n s o f r i n g s 1940-44 a g a i n s t t h e whole c r o s s s e c t i o n . In d e s c e n d i n g o r d e r t h e c o e f f i c i e n t s o f d e t e r m i n a -t i o n f o r t h e o t h e r p r o p e r t i e s w e r e : MND 0.-65; PLW 0 . 6 2 ; UNI'0.58; ED 0 .44 EW 0 . 3 8 ; L^W 0 . 1 2 ; -RW 0 . 0 1 . , T h e s a b o v e c a l c u l a t i o n s were b a s e d on a r i t h m e t i c a v e r a g e s o f t r e e r i n g s and a l l r i n g g r o u p s were g i v e n e q u a l w e i g h t . In p r a c t i c e , i t i s l i k e l y t h a t w e i g h t e d c r o s s - s e c t i o n a l means would be o f more i n t e r e s t a s t h e p r o p e r t y t o be p r e d i c t e d , i n w h i c h c a s e t h e i n f l u e n c e o f t h e i n n e r r i n g s wou ld be a f f e c t e d by b a s a l a r e a d i s t r i b u t i o n , i. e., t h e r e l a t i v e p r o p o r t i o n o f c o r e w o o d . I t i s i n t e r e s t i n g t h a t UNI and RNG show good c o r r e l a t i o n s between c o r e -wood and c r o s s - s e c t i o n a l means, s i n c e t h i s i m p l i e s some c o n t r o l o f t h e s e c h a r a c t e r i s t i c s w i t h i n t r e e s and more o r l e s s c o n s i s t e n t d i f f e r e n c e s between t r e e s . The o b s e r v e d r a n g e i n tree.:*mean v a l u e s was c o n s i d e r a b l e ( n e a r l y 100% d i f f e r e n c e between e x t r e m e s ) . Rank o r d e r c o r r e l a t i o n c o e f f i c i e n t s (Moroney, 1951) p r o v e d n o n - s i g n i f i c a n t a t t h e '5% l e v e l f o r b o t h ED w i t h LD and MND w i t h MXD, i l l u s t r a t i n g t h a t a l t h o u g h : t h e d i f f e r e n c e between e a r l y w o o d and l a t e -wood l e v e l s may be c o n t r o l l e d w i t h i n and between t r e e s , t h e l e v e l s t h e m s e l v e s a r e n o t r e l a t e d . T h a t i s , t h e r e i s no s i g n i f i c a n t t e n d e n c y f o r t r e e s w i t h - 80 -low or high earlywood density c h a r a c t e r i s t i c s to have correspondingly low or high latewood c h a r a c t e r i s t i c s . These fac t s suggest that earlywood and latewood properties may be under independent genetic c o n t r o l . In summary, i t may be stated that several tree density properties appear to be highly amenable to p r e d i c t i o n from measurement of rings adjacent to the p i t h . In the samples used, the innermost r i n g group represented rings numbering 4 - 8 from the p i t h and gave very good estimates of c r o s s - s e c t i o n a l values for RD, LD, MXD and RNG. Other density properties were l e s s s u i t a b l e but could s t i l l be associated with about 50% or more of the v a r i a t i o n . Of the width components measured, only PLW gave good r e s u l t s , but t h i s could be very important i n that i t i s a property which can be e a s i l y assessed without complex equipment. C o r r e l a t i o n analyses of the r e s u l t s f o r cores on an i n d i -v i d u a l r i n g basis, and also using core means, showed that PLW was associated with from 70% - 90% of the within-tree and between-tree v a r i a t i o n i n mean r i n g density. I t must be stressed, however, that the r e s u l t s apply to one p a r t i c u l a r group, of stems from an even-aged unthinned p l a n t a t i o n and need not represent the s i t u a t i o n i n crops growing under d i f f e r e n t c l i m a t i c and stand conditions. A recent report from France (Thoby, 1975) also deals with juvenile-mature cor-r e l a t i o n s i n Douglas-fir. In t h i s case, properties of the inner ten rings were compared with those of the outer ten rings i n three crops of d i f f e r e n t ages. I t i s encouraging that the r e s u l t s are e s s e n t i a l l y the same as those reported here f o r the properties which are d i r e c t l y comparable (RW, RD, MND, MXD, RNG). Densitometric r e s u l t s have proven more s a t i s f a c t o r y than those using gravimetric techniques, not only i n the larger number of v a r i a b l e s measured but i n the estimated c o r r e l a t i o n s f o r Douglas-fir wood pr o p e r t i e s . Contrary to the opinions of Northcott et al. (1964) and McKimmy (1966), i t would appear - 81 -that u s e f u l e a r l y predictions of growth r i n g density components can be made from corewood samples, p a r t i c u l a r l y i f , as i s most l i k e l y , only broad c l a s s i -f i c a t i o n s are required. - 82 -Chapter 5. RESULTS AND DISCUSSION - PROVENANCE TRIAL 5.1 Gravimetric Analyses 5.1.1 Site and provenance effects Summaries of the diameter and wood density values determined on the whole increment cores are given in Appendix 9 (a and b), with the data grouped by plantations (plots A,B and C,D combined), provenances (plots A,B, C,D) and sites. Graphical presentations of the site and provenance results are given in Figs. 5.1 and 5.2 Clearly, highly"significant differences occurred in bo'th growth rate and wood density between sites, and Molalla stands out as being of par t i -cularly slow growth and high density. At the other extreme, Valley showed the largest diameters and the lowest density levels. On the whole, the densities are lower than the average values published for coastal Douglas-3 3 f i r (0.43 g/cm - Drow (1957); 0.45 g/cm - U.S. Forest Service (1965)). This effect would be more pronounced i f whole-tree values were determined according to the relationship shown in Fig. 4.1, and is undoubtedly due to the fact that the inner rings are always found to be of comparatively low density. However, the Molalla site has produced corewood values well above average, which may be due to the presence of compression wood (mean = 0.47 g/cm^). An analysis of the four sites having two plantations each was carried out in order to obtain an estimate of the relative contributions from the various identifiable sources of variation to the observed growth rate and wood density variation. Mean ring widths were used in addition to diameters, to allow for any possible bias in the numbers of growth rings at the level sampled. The model used is given in Appendix 10a and the results tabulated in Table 5.1. S t r i c t l y speaking,' i t is not really valid to use the individual Nimpkish L O C A T I O N S Haney Mo la l l a Val ley Do rena E E 200 O 150 *-> E <TJ b 1 0 0 50 N H M V B N H M V B Legend T Max. 2 « s.e. Mean •*• Min. r - f - - f -N H M V B N H M V B N H M V B CO to N : Nimpkish H=Haney M = Molalla V = Valley B= Butte Falls P r o v e n a n c e s Figure 5.1 Summary of provenance t r i a l diameter data. Nimpkish LOCATIONS Haney Molalla Valley Dorena • 55 ro E u .50 • .45 -C a a .40 -u in fO rll • 35 • • 30 N H M V B N H M V B Lcqgnd T Max. ^ 2 « s.e. Mean Min. N H M V B N H M V B N H M V B N = Nimpkish H = Haney M = Molalla V = Valley B= Butte Falls Provenances Figure 5.2 Summary of provenance t r i a l increment core density data. - 85 -TABLE .5.1: Variance Components - Provenance T r i a l Diameter Density Ring Width Source Component % Component o, "o Component % 2 S i t e s , 6 ST 1493.8 65.7 .0014771 56.7 2.3254 62. 9 2 Provenances, 6 PR 4.8 0.2 .0000301 1.2 .0075 0. 2 2 Si t e s x provenances, 6gp^ 8.6 0.4 .0000000 0.0 .0070 0. 2 2 Plantations, 6 PL 48.3 2.1 .0000332 1.3 .1151 3. 1 2 Plantations x provenances, 6 PPL 0.0 0.0 .0000603 2.3 .0149 0. 4 2 Plo t s , 6p>r 55.7 2.5 .0000289 1:1 .0862 2. 3 2 Trees, 6 T 577.2 25.4 .0007124 27.3 .9849 26. 7 2 Cores, 6 e 84.7 3.7 .0002625 10.1 .1535 4. 2 - 86 -cores i n t h i s way since they were c o l l e c t e d along one diameter and are therefore not independent random samples. However, the analyses are pre-sented i n t h i s form to give an i n d i c a t i o n of the r e l a t i v e importance of within-tree v a r i a t i o n . The variance component f or s i t e accounted f o r the greatest part of the v a r i a t i o n i n both r a d i a l growth and wood density, ranging from 56.7% to 65.7%. The provenance component measures the extent of the t o t a l genetic variance between the selected populations, and i s seen to be of minor importance i n t h i s study (0.2% f o r r a d i a l growth and 1.2% for density). Tree-to-tree v a r i a t i o n within p l o t s contributed a s u b s t a n t i a l p o r t i o n of the t o t a l v a r i a t i o n (25.4% to 27.3%) and represents both genetic d i f f e r -ences within populations and microsite e f f e c t s . Between-core v a r i a t i o n i s also r e l a t i v e l y high, p a r t i c u l a r l y f o r wood density (10.1%), r e f l e c t i n g a combination of true c i r c u m f e r e n t i a l v a r i a t i o n and measurement e r r o r s . This component i s much l a r g e r than that a t t r i b u t a b l e to provenances i n a l l cases. The contributions of plantations and p l o t s , although r e l a t i v e l y small, also together exceed the provenance e f f e c t . Genotype -environment i n t e r a c t i o n terms ( s i t e x provenance and block x provenance) were of minor importance, i n d i c a t i n g l i t t l e change i n the ranking of provenances between pl a n t a t i o n lo c a t i o n s . The s i m i l a r i t y i n the magnitudes of the variance components f o r growth and density c h a r a c t e r i s t i c s suggests that the balance between genetic and environmental c o n t r o l i s also of the same order f o r both properties. Analyses of variance were performed f o r each i n d i v i d u a l s i t e according to the model given i n Appendix 10b(except Dorena with only one p l a n t a t i o n ) . Diameter and wood density r e s u l t s are given i n Table 5.2. In no case was a d e f i n i t e provenance e f f e c t found. This would be con-TABLE 5.2 Summary of Provenance T r i a l S i t e Analyses F-ratios DIAMETER WOOD DENSITY Locations Locations Source d.f. Nimpkish Haney Molalla Valley Dorena Nimpkish Haney Molalla Valley Dorena Provenances 4 i . o i n s 1.43 n S 3 . 6 i n S 5.08X 0.15 n S 0.45 3.25 n S , „ ns 1.39 0.67 n s ns 0.88 Plantations 1 0.13 n S i . i o n s 120.33 0.26 0.05 i . i o n s 14•00 X X 0.15 n S Provenances x plantations 4 3.37 n S 0.74 n S 0.33 n S . „,ns 0.26 „ _ .ns 2.54 1.65 n S 0.85 n S 9. SI 3"™ Plots within provenances 10 0.82 n S i . 8 1 X 3.13 X X X 2.41 X X X 13. Zl™* 1.77X 1.84 1.35 n S „ __ns 0.37 2.29^ Trees within plots 180 12.72 X X X 19.36 X X X 19.ez*** _ . . X X X 7.52 10.83 X X X 15. Bl™* 3.74 X X X 8.26 Error 200 Total 399 CO ns = not s t a t i s t i c a l l y s i g n i f i c a n t x = s i g n i f i c a n t at the 10% l e v e l xx " s i g n i f i c a n t at the 5% l e v e l xxx = s i g n i f i c a n t at the 1% l e v e l - 88 -eluded i n most cases from the non-significant F - r a t i o s , but the magnitudes of the provenance x p l a n t a t i o n i n t e r a c t i o n s precluded assessment of the main e f f e c t s i n two cases. Plantation e f f e c t s were not s i g n i f i c a n t except at M o l a l l a , where appreciable d i f f e r e n c e s were found between the two planted areas. This e f f e c t i s i l l u s t r a t e d i n F i g . 5.3. The d i f f e r e n c e i n 3 density between the two blocks i s 0.02 g/cm , almost as great as that between the lowest and highest density s i t e s excluding M o l a l l a (Valley -3 3 0.384 g/cm ; Nimpkish - 0.411 g/cm ). The p l o t e f f e c t s were s i g n i f i c a n t i n several cases, showing that even within apparently uniform s i t e s , microsite can have an influence on growth and wood properties. The r e s u l t s i n d i c a t e that provenance differences i n r a d i a l growth and mean wood density are i n s i g n i f i c a n t i n t h i s study i n comparison to s i t e e f f e c t s (both geographic l o c a t i o n and microsite) and tree-to-tree v a r i a t i o n s . It i s noteworthy that the s i t e means for corewood d e n s i t i e s cover a much wider range than the averages published for the p r i n c i p a l geographic zones (Drow, 1957; U.S. Forest Service, 1965). The wide tree-to-tree v a r i a t i o n i s of i n t e r e s t since i t appears to o f f e r the only r e a l opportunity f o r s e l e c t i v e breeding to improve both r a d i a l growth and wood density. Figure 5.4 shows the d i s t r i b u t i o n of i n d i v i d u a l tree data bulked by s i t e s . Within each s i t e , the frequency d i s t r i b u t i o n s are approximately normal and the density range i s much larger than between extreme s i t e s . C o e f f i c i e n t s of v a r i a t i o n v a r i e d between 7.0% and 7.8%. In view of the many reports l i n k i n g growth rate and wood density i n a negative manner, i t i s important to examine the contribution of r a d i a l growth rate to the observed phenotypic values. 5.1.2 Relationship between growth rate and wood density Several workers have reported a negative c o - r e l a t i o n between growth N a Nimpkish H * Han«y M = Mola l la V = Valla y 1 = P l a n t a t i o n 1 2 « » 2 Figure 5.3 Summary of provenance t r i a l p l antation data, - 90 -30 10 o 30 10 30 10 — 30 10 30 10 Nimpkish Mean - 0-411 Haney Molal la Dorena i 0.393 J - 0.4 70 - 0 4 0 5 J •30 -40 -50 00 B a s i c D e n s i t y g Jc m Figure 5.4 Increment core density d i s t r i b u t i o n s by s i t e s . - 91 -rate and wood density in Douglas-fir as in other species (Harris and Orman, 1958; Paul, 1963; E l l i o t t , 1970). On the other hand, i t has been noted in some studies that wider growth rings may be associated with higher density (Wellwood and Smith, 1962). Since the latter comparison was based on pairs of increment cores from individual.«trees, i t is not clear to what extent the wider rings were related to compression wood formation. Data from the current study allows an examination of the growth rate effect for the corewood zone as i t is affected both by site and by provenance. Preliminary plots of the bulked data (Fig. 5.5) indicated that the relationship may be curvilinear rather than linear. Consequently, equations were fi t t e d using linear,hyperbolic, parabolic and logarithmic functions. Tabulation of the coefficients of determination i s given in Appendix 11. In a l l cases, the curvilinear functions accounted for a greater portion of the va r i a b i l i t y than the linear, and the logarithmic was chosen as the basis for comparison. Figures 5.6-5.7 show the relationships, back-transformed to the original values. Within sites, radial growth and wood density were correlated in a highly significant manner, coefficients of determination ranging from 0.06 to 0.26. That i s , up to 26% of the variation in wood density was associated with fluctuation in mean ring width. The Molalla plantations gave consis-tently higher density values at a l l growth rates and at least part of this effect must be due to the high incidence of compression wood. Tests of parallel-ism and coincidence showed that there were significant differences in the slopes of the regression lines even when tested without the Molalla data. From Fig.5.6 i t would appear that the differences occur mainly at mean ring widths of less than 6 mm, %..e. , in trees growing more slowly than the site average. Above this level of growth thereiis practically no influence on wood density. f -0.60 «t VI C a 10 10 M 0.55 0.50 ta 0.45 0.40 0.35 0.30 0,25 0,0 3 ** * * * * 2 * 4 * ** * ** 2* *234* ** ** 3*2 *3* * **2 * 4 *32222*2** * ** 4****323 * 4* 2 **32*22 *** * ** * * *4*32332* 322** * * *233 24** 2423*2643 333 ** 2 2 3 2***2353232*2222 2 * ** 22**43*25562533 35*25 422**2 . * 2* 22**2*452222263532*52224* * * **** 2*5255*53402*4932554*3 * *2* 3 * * * *3* 2522250658343865* 3**4 * * * ** **** **575544 45*84*22**** ** * * 2 222*444*7 657422*6*324*42*** * * ****4*35*4722264*2332422.** * *** 3 2.2 32 ***2* * 423.2.* * * * 2 *2**2* * *2 * * ** 2 * * * * * * * * * * * - one observation 0 - ten observations • • • **2 * * * * ** • •• . • • • • • * * 3.0 6.0 9.0 Mean Ring Width mm to • • . • 12.0 Figure 5.5 Wood density - growth rate r e l a t i o n s h i p , •55 •50h •45h •40 X * ^ x x^v 2 ^ " "'•^g-r. X""X""X~X~X-X-X-X-X-X-X-X-X Z-7 o-o Nimpkish • - • Haney x-x Mo la l l a v-v Val ley z-z 0 o r e n a v v—»-jt-v-v-v-v-v-y-v-y-v-v-v-v CO •35 •30' _1_ 1-0 2-0 30 4-0 5-0 60 Mean Ring W i d t h mm 70 8-0 9'0 Figure 5.6 Wood density/growth rate r e l a t i o n s h i p by s i t e s . Figure 5.7 Wood density/growth rate r e l a t i o n s h i p by provenances. - 95 -The data bulked by provenances over a l l s i t e s (Fig. 5.7) shows essen-t i a l l y the same r e l a t i o n s h i p as that found within s i t e s . Tests of p a r a l l e l i s m and coincidence showed that the slopes were not s i g n i f i c a n t l y d i f f e r e n t but that intercepts were j u s t s i g n i f i c a n t at the 5% l e v e l . In p r a c t i c a l terms, the provenance differences over the range of values being considered here are of l i t t l e or no importance. The i n d i v i d u a l r e l a t i o n s h i p s account for between 50% and 61% of the v a r i a t i o n i n tree density, compared to 55% for the complete data over a l l s i t e s (Appendix 11). In F i g . 5.8 a comparison i s given of the l i n e a r and logarithmic r e l a t i o n -ships, with and without the high-density M o l a l l a s i t e . I t can be seen that the use of a s t r a i g h t l i n e would give appreciable discrepancies at the extremes of growth rate. These c o r r e l a t i o n s between mean r i n g width and density are s u r p r i s i n g l y high i n view of the f a c t that previous workers have c o n s i s t e n t l y f a i l e d to explain more than about 30% of the v a r i a t i o n even when using several indepen-dent variables such as age, s i t e c l a s s , a l t i t u d e , l a t i t u d e , crown c l a s s , etc., (McKimmy, 1959; Knigge, 1962; U.S. Forest Service, 1965). This may be i n large part a r e s u l t of the type of material used i n the current study, i.e. , young plantations where age i s not a v a r i a b l e and stand conditions are as uniform as p o s s i b l e . Since these r e s u l t s apply to the inner core of fast-grown wood (within 10 to 12 rings from the p i t h ) , i t i s d i f f i c u l t to assess t h e i r relevance to older, more mature wood. For example, McKimmy (1966), i n h i s study of 46-year-o l d f a m i l i e s , found that r i n g number from the p i t h and percentage latewood accounted for about 33% of the v a r i a t i o n i n density, and that addition of growth rate as another v a r i a b l e i n the multiple regression analysis had no i 1 1 1 1 1 1 1 — r _ i i ' i , i i i , i i i 1.0 2.0 3.0 4.0 50 6.0 7.0 8.0 ' 9.0 10.0 Mean Ring Width mm Figure 5.8 Linear and c u r v i l i n e a r regressions f o r the wood density/growth rate r e l a t i o n s h i p . - 97 -extra e f f e c t . On t h i s basis he concluded that rate of growth had no s i g n i f i c a n t e f f e c t on the density of the trees. However, i t i s commonly found that both distance from the p i t h and percentage latewood are c l o s e l y r e l a t e d to growth rate within stems due to inherent growth patterns. Harris and Orman (1958) allowed f o r t h i s trend by examining the e f f e c t of growth rate i n d i f f e r e n t r i n g groups from the p i t h and produced a se r i e s of curves, but gave no s t a t i s t i c a l information on t h e i r v a l i d i t y . I t appears from the graphs that the nature of the r e l a t i o n s h i p i s s i m i l a r at each age studied. McKimmy and Nicholas (1971) gave c o r r e l a t i o n c o e f f i c i e n t s for the tree diameter-wood density r e l a t i o n s h i p within h a l f - s i b f a m i l i e s at three s i t e s . A negative c o r r e l a t i o n was found i n a l l cases but s i g n i f i c a n t at only two of the l o c a t i o n s . The r e l a t i v e l y strong r e l a t i o n s h i p s uncovered i n the current study have implications for the tree breeder interested i n manipulating wood density. For example, the phenotypic d i s t r i b u t i o n s shown i n F i g . 5.4 contain a component of v a r i a t i o n due to growth rate and i t becomes a problem to separate the genetic and environmental influences. However, while there i s a highly s i g n i f i c a n t r e l a t i o n s h i p between growth rate and density, two important factors must be borne i n mind: a) i t i s possible to s e l e c t trees of r e l a t i v e l y high or low density within a l l growth rate classes, and b) the e f f e c t of growth rate on corewood density i s of minor importance i n trees growing f a s t e r than the average at each s i t e . - 98 -5.2 Densitometric analyses The gravimetric results presented above showed that significant d i f f e r -ences in growth rate and basic wood density occurred between the sites studied. The objective of the densitometric analyses is to partition the average ring width and density values into their component parts in an effort to explain how and possibly why the differences arise. The preliminary densitometric studies (Chapter 4.3) showed that two wood zones could be recognized within the corewood of young Douglas-fir trees, so i t was decided to treat the provenance t r i a l cores according to two 5-ring groups; rings 1-5 and rings 1970-74. These groups did not overlap since the cores had an average of 12 growth rings. For each of the 5-ring groups, core mean values were calculated for ring components, and summarized by sites and provenances. This information is tabulated in Appendix 12. Analyses of variance were performed on these data according to the models given in Appendices 10 (c-d). 5.2.1 Ring width components Appendix 12(a) shows that considerable site differences were found for mean ring width (RW) in rings 1-5 from the pith. The averages ranged from 1.79 mm at Molalla to 4.85 at Valley, a difference of 271%. Ranking of sites by earlywood width (EW) follows the same pattern as RW, but latewood width (LW) ranks are slightly different, and this i s reflected in the percentage latewood (PLW) values which range from 35.1% at Haney to 54.8% at Molalla.. Site ranking by ring components in the outer five rings gave essentially the same patterns as for the inner rings (Appendix 12 (b)) . EW and LW again showed different ranks, PLW ranging from 24.5% at Valley to 32.8% at Molalla. In summary, i t can be said that the relative ranking of sites tends to - 99 -remain the same between the two growth zones but that d i f f e r e n c e s between s i t e s are of l e s s e r magnitude i n the outer r i n g s . Figure 5.9 shows the trends i n EW, LW and RW from p i t h to bark i n stan-dardized s e r i e s . Four of the s i t e s e x h i b i t an increase i n component widths out to about the s i x t h r i n g and f l u c t u a t i o n t h e r e a f t e r . The exception i s the M o l a l l a s i t e where widths were found to increase gradually throughout the s e r i e s . C l e a r l y , the high a l t i t u d e environment has r e s u l t e d i n a d i f f e r e n t pattern of growth. In general, i t would appear that i n trees of t h i s age, v a r i a t i o n s i n RW are accounted f o r l a r g e l y by v a r i a t i o n s i n EW. Analyses of variance were performed f o r each s i t e and r i n g group accord-ing to the model given i n Appendix 10(c) and r e s u l t s are given i n Tables 5.3 (a - e ) . None of the s i t e s showed a s i g n i f i c a n t provenance e f f e c t at the 5% l e v e l or better for any of the r i n g width components. Thus, genetic e f f e c t s a t t r i -butable to provenances were not apparent, e i t h e r during the period of forma-t i o n of rings 1-5, or during 1970-74. These r e s u l t s confirm the whole-core r i n g width data given previously and extend the conclusions to EW, LW and PLW. Plantations within s i t e s likewise had non-significant e f f e c t s , except at Molalla i n the outer r i n g group where RW differences were a t t r i b u t a b l e mainly to v a r i a t i o n i n LW. This observation helps elucidate the p l a n t a t i o n d i f f e r -ences shown i n F i g . 5.3. The highly s i g n i f i c a n t differences i n LW were not of s u f f i c i e n t magnitude, however, to influence the PLW values strongly, r e f -l e c t i n g the predominant influence of EW i n wood of young t r e e s . In a few cases there was a tendency for the provenance x p l a n t a t i o n i n t e r -a c t i o n to appear s i g n i f i c a n t at the 10% l e v e l but only at V a l l e y was a greater s i g n i f i c a n c e l e v e l achieved, and then for LW and PLW i n the outer rings only. S i g n i f i c a n t differences between provenance r e p l i c a t i o n s within planta-tions occurred frequently f o r a l l width components and can probably be i n t e r -- 100 -1 2 3 4 5 1968 1970 1972 1974 Ring From Pi th Growth Par iod Figure 5.9 Ring width components by s i t e s . Analyses of variance f o r i n d i v i d u a l s i t e s (a) NIMPKISH 1-5 r-ratios Source d.f. RW EW LW • PLW RD ED LD MUD MXD UNI RN'G Provenances 4 0.76 0.83 3.59 1.22 1.G5 1.54 0.69 2.07 3.09 1.64 4.01 Plantations 1 0.31 0.01 4.47 1.72 1.26 6.10* 3.44^ 10.8S<:< 20.02^ ' 26. 02^ Provs. x plantations 4 2.74 2.12 2.77* 1.97 1.5S 1.12 , 2.52 1.08 1.75 1.45 1.47 Plots within provo. 10 1.41 1.65* 1.11 1.77X 1.76x 2.44 X X 0.84 ^.S2 X X 0.S8 1.66 1.40 Trees within plots 80 Rings 1970-74 Provenances 4 0.74 0.77 0.62 0.73 0.40 0.26 0.89 0.18 1.60 0.94 2.94 Plantations 1 0.60 0.30 2.64 0.00 0.04 C.15 3.08 0.38 1.49 3.73 3.11 Provs. x plantations 4 2.11 1.83 2.08 1.30 1.13 0.76 0.44 0.81 0.29 0.77 0.35 Plots within provs. Trees within plots 10 80 2.02** 2.33** 0.92 1.93 2.86 X X X 6.03 2.22** fa. 96 4.47 1.50 3.60 (b) I 1 A N E Y Rings 1-S , F-ratios 's Source d.f. EW EW LW PLW RD ED LD MND MXD UNI RNG Provenances 4 0.54 0.71 • 0.77 0.45 0.78 1.87 1.99 0.83 2.42 1.57 3.88 Plantations 1 1.47 2.92 1.32 6.81X 8.51 7.70X 12.IS** 5.53* 10.71** 0.85 3.64 Provs. x plantations 4 1.67 1.64 2.61X 2.65* 4.59 X X 1.62 1.34 1.84 1.29 0.71 0.45 Plots within provs. 10 1.48 1.49 1.05 0.64 0.69 1.46 1.24 1.97X 2.33 K X 3.65*** , _„xxx 6.50 Trees within plots 80 'Rings 1970-74 Provenances 4 0.18 0.37 0.73 1.93 1.95 5.75X 0.22 2.82 0.42 0.39 0.73 Plantations 1 0.04 0.26 0.95 2.09 4.28 8.52 X X 0.48 S.80X 2.04 0.06 0.76 Provs, x plantations 4 1.57 2.12 1.43' 10.7850"5 12. SG 3 0 0 1 1.06 3.72 3.12X 3.53** 2.02 2.04 Plots within provs. 10 1.78X 1.31 2.75 X X X 0.23 • 0.25 1.74X 1.76 1.84X 1.92** 2. S O " 2.56 X X Trees within D i e t s 80 x = significant at 10% lev o l xx " significant at 5* leve l xxx » significant at 1% lev o l TABLE 5.3 (contd.) (c) M o l a l l a Source d.f. RW EW LW PLW R D E D LD M N D K X D UNI RNG Provenances 4 0.48 0.39 1.20 ' 1.08 1.23 0.84 5.71 X 1.51 8.46 1.21 5.57 X P l a n t a t i o n s 1 2.02 1.49 2.90 0.00 0.10 0.13 11. s e 3 0 1 0.25 7.84 3.34 Provs. x p l a n t a t i o n s 4 2.10 3.00 X 0.31 0.83 0.62 0.90 0.19 0.80 0.12 0.58 0.21 P l o t s w i t h i n p r o v s . 10 2 . 7 6 X X X 1.48 „ „_xx 2.06 2.80 . , „xx 2.40 2.07 2.42 , ...XXX 4.58 2.28 4.25J O Q' Trees w i t h i n p l o t s 80 Rings 1970-74 • Provenances 4' 0.71 0.53 1.03 0.46 1.62 3.08 2.34 3.91 1.62 1.83 1.54 P l a n t a t i o n s 1 13.78 S C X 5.74 X 4 2 . 8 4 X X X 0.00 0.18 0.11 5.08 X 0.51 1.60 5.59 X 2.14 Provs. x p l a n t a t i o n s 4 1.52 2.12 0.30 1.97 0.73. 0.27 0.37 0.55 0.59 0.43 0.67 P l o t s w i t h i n provs. 10 1.38 . 1.94 X X 1.13 1.71* 2.83 „ ,„xxx 3.18 _ __xxx 5.22 , _-XXX 3.73 3 . 3 3 X X X 6.43 Trees w i t h i n p l o t s 80 (d) VALLEY Rings 1-5 F - r a t i o s Source d.f. ' RW EW LW PLW RD ED LD MND HXD UNI RNG Provenances 4 1.22 1.62 0.15 0.71 1.33 0.90 1.04 1.17 1.07 0.05 0.26 P l a n t a t i o n s 1 0.09 0.24 0.07 0.21 1.34 1.02 1.37 0.37 3.35 1.09 1.93 Provs. x p l a n t a t i o n s 4 1.01 1.14 1.37 1.82 0.87 1.09 1.86 0.81 0.88 2.37 1.38 P l o t s w i t h i n p r o v s . 10 1.86 X 1.69 1.76 X 1.00 1.64 1.33 2 . 5 0 X X 2.18 s.eo300* 2 . 1 1 X X xxx 4.46 Trees w i t h i n p l o t s 80 Rings 1970-74 Provenances 4 . 0.40 0.55 0.24 1.35 0.62 0.40 0.42 0.49 0.25 0.39 0.18 P l a n t a t i o n s 1 1.84 2.51 0.63 0.14 0.07 0.56 0.86 0.02 0.79 0.69 0.96 Provs.. x p l a n t a t i o n s 4 1.94 1.28 8 . 9 0 X X X 1.38 6.01 3.64 X X 3.18 X 1.75 • 4.65 2.38 3.40* P l o t s w i t h i n p rovs. 10 4 . 7 8 X X X S . 3 5 X X X 0.87 2.23™ 1.18 , . .XX 1.94 3.56 3.63 3 0 0 1 „ ..XXX 2.90 xxx 3.43 XXX 3.75 Trees w i t h i n p l o t s • 80 x » s i g n i f i c a n t a t 10% l e v e l xx » s i g n i f i c a n t a t 5% l e v e l xxx - u i ' j n l f i c n i i t a t 1% l c v u l TABLE 5.3 (contd.) (e) DORENA Rings 1-5 F - r a t i o s Source d . f . RW EW LW PLW RD ED LD MND MXD UNI RNG Provenances 4 0.52 0.68 0.50 1.15 1.04 0.48 4 . 4 4 X X X 0.6S 7 . 3 6 X X X 2.49 X P l o t s w i t h i n p r o v s . 5 B.91*** 6 . 7 2 X X X , ._xxx 3.o7 0.70 0.99 1.45 1.00 2.38 X 1.21 1.08 0.83 Trees w i t h i n p l o t s 40 Rings 1970-74 Provenances 4 2.56* 4 . 6 5 X X X 0.36 1.38 1.03 0.38 4.61 0.69 2.44 X 4 . 3 4 X X X 9 . 3 5 x x x P l o t s w i t h i n p r o v s . • 5 4 . 0 7 X X X 3.21** 1.85 1.29 2.04 X . 2.42 X 3.92 1.74 1.26 Trees w i t h i n p l o t s 40 -x •> s i g n i f i c a n t a t 10% l e v e l xx » s i g n i f i c a n t a t 5% l e v e l xxx - s i g n i f i c a n t at 1% l e v e l - 104 -preted as expressions of microsite heterogeneity. As a general trend, i t may be noteworthy that the outer rings are more v a r i a b l e i n t h i s regard, suggesting that some environmental factors exert more influence as the trees age. A notable exception to t h i s , however, i s the M o l a l l a s i t e where more v a r i a b i l i t y was found i n the inner growth zone. In t h i s case the harsh high a l t i t u d e environment may have allowed e a r l i e r expression of microsite d i f f e r -ences where c r i t i c a l or l i m i t i n g factors such as s o i l moisture, ground temp-erature and snow re t e n t i o n might be expected to be more v a r i a b l e due to the broken topography. 5.2.2 Ring density components Appendices 12(a) and (b) tabulate the provenance mean values for density components summarized over rings 1-5 and 1970-74, while F i g . 5.10 gives the stan-dardized density p r o f i l e s obtained from the 100 cores per s i t e (50 for Dorena). Trends of r a d i a l density v a r i a t i o n were quite s i m i l a r at a l l s i t e s , with maximum latewood density (MXD) increasing s l i g h t l y throughout the s e r i e s while minimum earlywood (MND) and r i n g mean (RD) d e n s i t i e s decrease, p a r t i c u l a r l y i n the rings close to theppith. The M o l a l l a s i t e c o n s i s t e n t l y showed the highest earlywood density (ED) and RD values, but latewood density (LD) f e l l within the range found for the other s i t e s , r e s u l t i n g i n low earlywood-latewood density contrast (UNI). Figure 5.10 demonstrates c l e a r l y that within s i t e s , minimum density (MND) remains more or l e s s constant from year to year whereas r i n g maxima (MXD) can vary appreciably. I t i s also apparent that the pattern of MXD v a r i a t i o n d i f f e r s from s i t e to s i t e , probably r e f l e c t i n g weather v a r i a t i o n . Differences i n intra-increment density d i s t r i b u t i o n are not r e a d i l y apparent from these graphs. Figure 5.11 presents the standard densigrams (Chapter 3.5) derived from the - 106 -Rings 1-5 Rings 1970-1974 >» o c o cr o c > 0 25 0 25 0 25 O 25 O 25 O I H J M J SEE I I ' * " I 1 ' " ' " f o •2 • 2 • 8 Dcnsi ty g /cm Figure 5.11 Standard densigrams by s i t e s . N = Nimpkish V = V a l l e y H = Haney D = Dorena M = M o l a l l a Ew - Earlywood Lw - Latewood - 107 -p r o f i l e s of the two growth zones. These show very c l e a r l y that d i f f e r e n c e s i n intra-increment density d i s t r i b u t i o n are much greater than the d i f f e r -ences i n RD mentioned above and discussed i n d e t a i l i n Chapter 5.1. For example, M o l a l l a has l e s s density contrast (RNG) than Nimpkish and Haney, which i n turn have l e s s than V a l l e y and Dorena. The range i n values extends 3 3 from 0.535 g/cm at M o l a l l a to 0.647 g/cm at Dorena. I t i s also apparent that MND and MXD means vary appreciably between s i t e s , Haney and V a l l e y having low MND and Dorena having highest MXD. Figures 5.10 - 5.11 use s i m i l a r basic data but present i t i n d i f f e r e n t ways to emphasize p a r t i c u l a r points. In F i g . 5.10, the conventional scan data were standardized to a f i x e d x-axis to give r a d i a l trends and allow v i s u a l comparison between tree groups by s p e c i f i c growth periods. The current study demonstrates the value of standard densigrams i n summarizing r i n g groups to show differences between samples which are not immediately apparent from other methods of presentation. As for r i n g width components, provenance e f f e c t s were analysed by analyses of variance (Appendix 10(c)) and the r e s u l t s given i n Tables 5.3 (a) to 5.3(e). S i g n i f i c a n t d ifferences a t t r i b u t a b l e to provenances d i d not occur c o n s i s t e n t l y i n e i t h e r growth zone, except at Dorena where highly s i g n i f i c a n t d ifferences were found for LD and RNG i n both analyses. This s i t e also showed s i g n i f i c a n t e f f e c t s for MXD i n rings 1-5 and f o r UNI i n rings 1970-74. At the other s i t e s , MXD and RNG sometimes appeared to be aff e c t e d by provenance whereas MND showed no i n d i c a t i o n of varying. I t was therefore considered worthwhile to examine the ranking of RNG values at each s i t e by provenance. Table 5.44contains the r e s u l t s . Two features are immediately apparent from the table, i.e. : - 108 -TABLE 5.4: Ranking of Provenances by RNG Values Site Ring Group 1 2 Rank 3 4 5 Nimpkish 1-5 H M N V B 1970-74 H M V B N Haney 1-5 V H B M N 1970-74 V H B M N Molalla 1-5 H B V N M 1970-74 H B V M N Valley 1-5 N H M B V 1970-74 H M N B V Dorena 1-5 B H V N M 1970-74 B H V M N Provenance N = Nimpkish H = Haney M = Molalla V = Valley B = Butte Falls - 109 -a. the ranking remains v i r t u a l l y the same between the two growth zones, whether or not s i g n i f i c a n t differences occur between provenances. b. there i s a tendency f o r Nimpkish (N) and M o l a l l a (M) provenances to have high RNG values and for Haney to have low values. On the other hand, both V a l l e y (V) and Butte F a l l s (B) provenances vary from highest at one s i t e to lowest at another. Figure 5.12 presents provenance densigrams by s i t e s (N, H, M and V) for the outer, f i v e growth ri n g s . These confirm that intra-increment density d i s t r i b u t i o n patterns are quite s i m i l a r between provenances at a p a r t i c u l a r l o c a t i o n , e s p e c i a l l y i n contrast to the s i t e d ifferences shown i n F i g . 5.12. The V a l l e y s i t e i s p a r t i c u l a r l y uniform i n t h i s regard. In summary, i t may be surmised that there i s a tendency for small provenance differences i n i n t r a - r i n g contrast to p e r s i s t from s i t e to s i t e but that c e r t a i n provenances appear to e x h i b i t genotype x environment i n t e r a c t i o n . S i g n i f i c a n t differences between plantations within s i t e s were found f o r some density components, predominantly i n the rings near the p i t h . At three of the four s i t e s providing such comparisons, LD and MXD were shown to be s i g n i f i c a n t l y v a r i a b l e ; MND, ED and RD appeared to be l e s s susceptible to microsite influences. I t i s of i n t e r e s t that the gravimetric analyses showed s i g n i f i c a n t differences for mean density between plantations at Molalla whereas t h i s i s not apparent i n Table 5.3(c). The most obvious explanation f o r t h i s i s that, since the cores for densitometric study were screened f o r presence of compression wood, and a large p o r t i o n of the Molalla cores were rejected on t h i s basis, the differences between plantar tions may have been influenced by varying proportions of reaction wood. - 110 -50 25 0 25 O 25 O 25 0 25 0 50 25 0 25 O 25 0 25 0 25 0 N impkish Ew 1 Lw 1 N mean | • > -*• 1 , H . • 1 M . - V* i i 1 v -1 1 I B . Vallay w*3 -A Ew J Lw N J H J M J i—1„ ..nrp B '8 Dansi ty 50 25 0 25 0 25 0 25 0 25 O 50 25 O 25 0 25 0 25 O 25 O Han«y H 4 M J I V J B J •4 -O M o l a l l a • 8 Ew J Lw •A' »;•:•>: i M . m a \» 1 RSS?| T 1 PL ! v ' 1 • 2 .4 • 8 g / c m Figure 5.12 Standard densigrams by s i t e s and provenances , 1969-73. N = Nimpkish V = V a l l e y H = Haney . B = Butte F a l l s M = M o l a l l a Ew - Earlywood Lw - Latewood - I l l -The provenance x pl a n t a t i o n i n t e r a c t i o n proved to be s t a t i s t i c a l l y s i g n i f i c a n t f o r RD, LD and MXD at Haney and RD, ED and MXD at V a l l e y i n the outer.ring group. V a r i a t i o n between provenance r e p l i c a t i o n s was also shown to be s i g -n i f i c a n t i n several cases, p a r t i c u l a r l y i n rings 1970-74. The character-i s t i c s most s e n s i t i v e to such microsite e f f e c t s were MND, MXD and RNG, which might have been ant i c i p a t e d since they r e f e r to d i s c r e t e points on the density p r o f i l e and could be more e a s i l y influenced than properties based on means of many points. The above r e s u l t s confirm the gravimetric wood density analyses i n that v a r i a t i o n s i n r i n g width and mean density are found to be more strongly influenced by environmental factors than by seed source. Other r i n g width and density components are a f f e c t e d i n a s i m i l a r manner. At an e a r l y age, s i g n i f i c a n t d i f f e r e n c e s between plantations were apparent for some wood c h a r a c t e r i s t i c s , but as the trees grew older the e f f e c t of p l o t environ-ment became r e l a t i v e l y more important. In a l l the above analyses the s i g n i f i c a n c e of the sources of v a r i a -t i o n was assessed by F-test against a mean square term containing a compo-nent f o r error variance (tree-to-tree variance within p l o t s ) . Analyses of variance were also performed using the combined data for 4 s i t e s (N, H, M, V) and components of variance c a l c u l a t e d for each term i n the model (Appendix 10(d))in order to estimate the r e l a t i v e contributions of a l l recognizable sources of v a r i a t i o n . Figure-5 .:13 presents the r e s u l t s g r a p h i c a l l y . O v e r a l l , the factors which contribute most to the t o t a l v a r i a t i o n are s i t e s , p l o t s (replications) and i n d i v i d u a l t r e e s . Smaller e f f e c t s of plantations and provenance x pl a n t a t i o n i n t e r a c t i o n are sometimes present i n a d d i t i o n . Provenance and provenance x s i t e terms, where present, u s u a l l y - 112 -100 50 1 0 50 i 0 50 0 50 0 50 0 °/o 50 o 50 0 50 0 50 0 50 0 50 0 RW EW LW M I PLW RD EO LD MND MXD UNI RNG —-1 I »™—• X Z L t e 2 s ! e* e 2 e 2 e 2 ST PR SPR PL PPL PT T H Rings 1 5 Q Rings 1970 74 F i g u r e 5 . 1 3 V a r i a n c e components as a p e r c e n t a g e o f t h e t o t a l v a r i a n c e ( s e e T a b l e 5 . 1 f o r l e g e n d ) . - 113 -contribute less than 5% of the observed variation. The general trends are as follows: 1. Provenance effects are insignificant. 2. Among the ring width components, site exerts a greater effect in the inner rings than in the outer rings, whereas tree-to-tree variation becomes more important with age. 3. Among the ring density components, the opposite trends tend to hold, i.e., site effects seem to be similar or greater in the outer rings and the relative contribution from individual trees decreases (RD may be an exception to thi s ) . 4. For most properties measured, the contribution from individual trees i s greater than that for sites. Notable exceptions to this are RW and EW i n the inner rings where site accounts for about 60% of the variation. 5. ED and MND vary much more between sites than LD and MXD in comparison to tree-to-tree variation. These variance components are estimates only, based on four sites, five provenances and one core from each of five trees per provenance replication, and so i t is inadvisable to place too much reliance in individual values. The effect of sampling technique may be gauged by comparing the values for RW and RD with those i n Table 5.1 based on two cores from each of 900 trees. The densitometric analyses could not account for within-tree variation since time restrictions limited sampling to one core per tree. The result i s that tree-to-tree variation i s increased by an unknown amount and appears to be more influential in the analyses. The important features are the general trends between growth zones and theefact that the density components as a group appear to behave similarly. - 114 -Environmental factors associated with geographic l o c a t i o n , p l a n t a t i o n s i t i n g and microsite (plots) account for a large portion of the t o t a l v a r i a t i o n (usually more than 50%), but tree-to-tree v a r i a t i o n within p l o t s i s often almost as great an e f f e c t , suggesting strong genetic c o n t r o l over a l l growth r i n g components. 5.2.3 Correlations between r i n g components C o r r e l a t i o n matrices for a l l growth r i n g components were obtained f o r each s i t e and r i n g group (rings 1-5 and years 1970-74). Since c o r r e l a t i o n c o e f f i c i e n t s were generally s i m i l a r i n sign and magnitude both within s i t e s and between a l l s i t e s , only the bulked data w i l l be presented here (Appendix 13). I t i s also apparent that both r i n g groups behave s i m i l a r l y and so a general discussion w i l l s u f f i c e . Relationships of s p e c i a l i n t e r e s t w i l l be dealt with i n more d e t a i l following the introductory statements. Most of the c o r r e l a t i o n c o e f f i c i e n t s obtained are s t a t i s t i c a l l y s i g n i -f i c a n t at the 1% l e v e l , but many are of no i n t e r e s t i n t h i s study. Ring width i s highly c o r r e l a t e d with both earlywood and latewood widths, but earlywood and latewood widths themselves are not so c l o s e l y r e l a t e d , suggesting a f a i r degree of independence. In rings of these ages, t o t a l r i n g width i s more highly r e l a t e d to earlywood than latewood width, a f a c t which i s also r e f l e c t e d i n the negative c o r r e l a t i o n s between r i n g width and percentage latewood. Mean r i n g density i s much more c l o s e l y r e l a t e d to earlywood density than to latewood density when the data i s bulked over a l l s i t e s , but i t i s worth noting that the reverse i s true i n the outer r i n g groups at Nimpkish and Haney. As expected, earlywood and latewood d e n s i t i e s are highly corre-l a t e d with UNI (= LD - ED) as are minimum and maximum densites with RNG - 115 -(= MXD - MND). I t i s c l e a r that i n most cases the latewood c h a r a c t e r i s t i c s are of more value i n p r e d i c t i n g intra-increment density contrast parameters. Mean r i n g density i s seen to be highly correlated with minimum density and l e s s well with maximum density. Minimum and maximum d e n s i t i e s are poorly correlated. O v e r a l l , mean r i n g density i s poorly r e l a t e d to the i n t r a - r i n g density contrast properties, UNI and RNG, although within some s i t e s there i s a tendency towards a stronger p o s i t i v e c o r r e l a t i o n i n the outer r i n g s . In some cases, up to 25% of the v a r i a t i o n i n RNG i s associated with v a r i a t i o n i n mean density, e.g. , years of 1970-74 at Nimpkish and Haney. The gravimetric analyses of the mean density/growth rate r e l a t i o n s h i p (Chapter 5.1.2) showed a highly s i g n i f i c a n t negative c u r v i l i n e a r associa-t i o n . A l l provenances behaved s i m i l a r l y , but there was a d i s t i n c t s i t e e f f e c t on the regressions. Analyses were performed on the densitometric data by r i n g groups using tree mean values bulked by s i t e s and r i n g groups. P l o t t i n g of the raw data showed that within each r i n g group' a l i n e a r model was the most appropriate. The regression s t a t i s t i c s are given i n Appen-dix 14. Ring density was more c l o s e l y r e l a t e d to growth rate i n the inner rings than i n the outer rings where only two s i t e s showed a s i g n i f i c a n t r e l a t i o n -2 ship. Nimpkish and Haney gave the best c o r r e l a t i o n s i n both cases with R values of up to 0.28. Tests of p a r a l l e l i s m and coinidence were used to determine that the regression slopes were s i g n i f i c a n t l y d i f f e r e n t i n the outer r i n g groups and intercepts were d i f f e r e n t i n both cases. Nevertheless, the o v e r a l l regressions were computed and given g r a p h i c a l l y i n F i g . 5.14(a). The slopes show that v a r i a t i o n i n growth rate i s associated with l e s s - 116 -Figure 5.14 Relationships between mean r i n g density and a) mean r i n g width, b) percent latewood 117 -v a r i a t i o n i n mean density as the trees age. Within the range of the data c o l l e c t e d the inner rings tend to have higher d e n s i t i e s than the outer at the same growth rate. The c o r r e l a t i o n matrices c o n s i s t e n t l y gave f a i r l y high c o r r e l a t i o n c o e f f i c i e n t s f o r the r e l a t i o n s h i p between r i n g density and percentage latewood. Linear regression s t a t i s t i c s are shown i n Appendix 14. within s i t e s , latewood percentage accounts f o r between 75% and 88% of the v a r i a -t i o n i n density. The slopes of the regressions were not s i g n i f i c a n t l y d i f f e r e n t within age groups and the intercepts only s l i g h t l y d i f f e r e n t . The regression f o r the bulked data are shown i n F i g . 5.14(b). In t h i s case, changes i n latewood percentage influence wood density to a greater extent as the trees age. Mult i p l e regression analyses were performed for each age group at a l l s i t e s using both r i n g width and percentage latewood as independent va r i a b l e s , but i n no case d i d the i n c l u s i o n of r i n g width s i g n i f i c a n t l y improve the density/percentage latewood r e l a t i o n s h i p . Table 5.5 shows the di f f e r e n c e between the measured s i t e d e n s i t i e s and the density estimated using percentage latewood only on the basis of the o v e r a l l regressions given i n F i g . 5^ 14 (o)l) As can be seen, percentage latewood i s a highly r e l -i a b l e i n d i c a t o r of mean density despite the s i t e v a r i a t i o n s i n r i n g density components shown i n Appendices .12(a)'-and (b). 5.2.4 Influence of seasonal weather v a r i a t i o n One of the advantages of densitometric techniques f o r measurement of r i n g components over gravimetric methods i s the ease with which data can be accumulated for s p e c i f i c growth rings and even parts of r i n g s . Modern - 118 -TABLE 5.5: S i t e Densities Predicted from Latewood Percentage Values Ring Group S i t e Mean Density Percent Latewood Predicted Density Difference ( 1 ) 3 g/cm (2) 3 g/cm (2) - y.) g/cm 1-5 N .461 39.2 .454 - .007 H .434 35.1 .440 + .006 M .510 54.8 .509 - .001 V .440 36.9 .446 + .006 D .466 41.3 .461 - .005 1970-74 N .412 31.3 .412 .000 H .394 30.7 .409 + .015 M .430 32.8 -.420 - .010 V .373 24.5 .376 + .003 D .406 27.6 .393 - .013 N = Nimpkish H = Haney M = M o l a l l a V = V a l l e y D = Dorena - 119 -dendrochronological studies often r e l y h e avily on density data obtained by densitometer f o r the matching of r i n g s e r i e s from d i f f e r e n t wood samples (Polge, 1970; Parker, 1971). Whether or not there i s evidence of v a r i a t i o n i n r i n g widths associated with yearly weather, density scans may y i e l d a d d i t i o n a l information. In the current study, the cores do notpprovide a s u f f i c i e n t l y long seri e s of rings to j u s t i f y year by year comparisons with weather records, but analyses can be performed i n such a way as to measure the r e l a t i v e e f f e c t of weather v a r i a t i o n on growth-ring components i n the 1970-74 r i n g s . The model given i n Appendix 10(e) was used t r e a t i n g years as a random v a r i a b l e , and variance components estimated. Of the eleven components thus derived, seven were found to c o n s i s t e n t l y account f o r over 90% of the observed v a r i a -t i o n and these are l i s t e d i n Table 5.6 by s i t e s . The r e l a t i v e importance of the weather e f f e c t varied between s i t e s over the period considered, as do contributions from the other factors involved. In general, the year e f f e c t i s considerably smaller than the tree e f f e c t and larger than the provenance and p l a n t a t i o n terms. Interaction components (provenance x p l a n t a t i o n and year x tree) are high i n some cases, p a r t i c u l a r l y at M o l a l l a . At the B r i t i s h Columbian s i t e s (Nimpkish and Haney) a l l r i n g width com-ponents were found to vary with growth period to about the same extent (5-6% and 10-13% of the respective t o t a l s ) whereas at the other s i t e s , RW and EW were r e l a t i v e l y much more affected than LW (10-17% as compared to 2-5%). Among the r i n g density components, ED and MND showed very l i t t l e weather e f f e c t (0-1% except at Molalla where the incidence of compression wood may have aff e c t e d the r e s u l t s ) . MXD and sometimes LD tend to vary much more with years (up to 18% of the t o t a l v a r i a t i o n ) . UNI and RNG also vary appreciably as a r e s u l t of f l u c t u a t i o n i n LD and MXD l e v e l s . TABLE 5 . 6 : Variance.Components as Percentage of the Total Variance SITE NIMPKISH HANEY MOLALLA VALLEY Source PR PL PR X PL YR PT T YR X T Total % PR PL . PR x PL YR PT T YR X T Total % PR PL PR X PL YR PT T YR X T Total % • PR PL PR X PL YR PT T YR X T Total % Ring COEDor.ent RW 0 0 12 6 12 55 12 97 0 0 5 11 9 58 14 97 0 27 3 10 3 38 23 100 0 4 14 10 23 29 11 91 EW * 0 0 10 5 15 54 13 97 0 0 9 10 3 60 15 97 0 14 8 17 7 32 21 99 0 6 4 11 26 28 14 89 LW 0 4 5 5 0 58 24 96 0 .0 5 *3 13 33 31 • 95 0 11 0 5 1 32 50 99 0 0 31 2 0 41 23 97 PLW 0 0 3 2 11 52 27 95 6 3 7 13 0 45 23 97 0 0 6 13 5 30 41 95 2 0 3 5 9 32 42 93 RD' 0 0 2 3 22 55 16 98 ' 7 10 11 9 0 44 16 97 . 1 0 0 ' 11 17 41 28 98 0 0 26 2 1 37 23 94 ED 0 0 0 1 42 38 17 98 18 11 0 0 6 35 26 96 0 0 0 4 16 25 55 100 0 0 20 1 7 34 32 94 LD 0 1 0 10 15 58 14 98 0 0 22 13 6 42 14 97 0 4 0 2 34 36 22 98 0 0 21 6 13 23 24 87 MND 0 0 0 1 45 35 18 99 17 18 13 0 5 30 14 97 8 0 0 8 19 31 23 94 0 0 11 0 20 36 23 90 MXD 0 0 0 16 29 40 12 97 0 6 19 18 6 36 12 97 0 0 0 4 41 31 23 99 0 0 29 11 11 25 19 95 UNI 0 3 0 10 6 58 17 94 0 0 12 11 14 44 14 95 0 4 0 1 18 30 45 ' 98 0 0 13 6 13 26 26 84 RNG 0 0 0 20 22 39 15 96 0 0 11 20 12 38 14. 95 0 2 0 5 35 26 30 98 0 0 23 12 15 24 19 93 PR = Provenance . . YR - Years '. T Trees • PL - Plantations PT » Plots (Replications) Component abbreviations as given i n Fig. 3.3. - 121 -Thus even i n the comparatively uniform growth of the young trees examined here, density v a r i a t i o n from year to year can be picked up. Earlywood and latewood c h a r a c t e r i s t i c s appear to vary independently, so that s i t e s compla-cent for one can be s e n s i t i v e f o r the other. Indices f o r growth r i n g components and monthly weather data for the period 1969-74 were c a l c u l a t e d as described i n Chapter 3.2.4 and subjected to simple l i n e a r regression analyses using the U.B.C. TRIP package. Appendix 15 gives the c o e f f i c i e n t s of determination and the values of the regression c o e f f i c i e n t s x standard deviation of the independent v a r i a b l e s . Some of these l a t t e r data are given g r a p h i c a l l y i n F i g . 5.15(a-d) for the important r i n g components. Since the three temperature values (minimum, maximum and mean) behaved somewhat simi -l a r l y , only p r e c i p i t a t i o n and average temperature are shown i n the f i g u r e s . The y-axes represent the percentage change i n the r i n g component accompanying a change of one standard deviation (upwards) of the monthly weather v a r i a b l e s . This i s somewhat s i m i l a r to the beta c o e f f i c i e n t described by E z e k i a l and Fox (1959). Figure 5.15(a) shows that only June weather gave s i g n i f i c a n t c o r r e l a t i o n s with earlywood width; an increase i n temperature having a weak negative e f f e c t 2 (R = 0.11). Earlywood production i s thus favoured by cool wet weather i n June. Latewood i s also stimulated by summer and f a l l r a i n f a l l but winter p r e c i p i t a t i o n (December through March) appears to encourage earlywood formation at the expense of latewood, probably by helping to induce water d e f i c i t s . S i t e s with r e l a t i v e l y warm dry winters, low r a i n f a l l during earlywood formation, and wet cool summers would favour high percentage latewood. Earlywood and latewood d e n s i t i e s were much l e s s subject to v a r i a t i o n than the widths but signi'f icant'icorrelations were revealed nonetheless. A p r i l r a i n f a l l and temperature had s i m i l a r but opposing e f f e c t s , i.e. , wetter cooler years g i v i n g 122 -• 8 • 6 " *4 • 2 • O f - 2 - 4 • 4 • 2 O -- 2 - 4 P r e c i p. H h Ave. Temp. E W • 4 • 2 0+-- 2 -4 • 4 • 2 P r e c i p. E D XX A v e Temp. o - k - — -2 4 XXX _ vzzs I 1 O N D J F M A M J J M O N T H Figure 5.15(a) Precentage change i n earlywood components associated with a one standard deviation change i n monthly weather data. - 123 -• 6 •.4 • 2 O - 2 - 4 • 4 • 2 O - 2 -4 LW Pr G C i p. ITS? Said x xx x PI 1 Ave. Temp. £2S H L D • 4 • 2 r P r c c i p . - 2 - 4 • 4 • 2 Ave . T e m p . 2 4 O N D J F M A M O N T H M J A S Figure 5.15(b) Percentage change i n latewood components associated with a one standard deviation change i n monthly weather data. - 124 -RW • 6 • 4 2 O - 2 - 4 V • 4 • 2 04-- 2 -4 P r c c i p. -* h A V G . Temp. H h W 1 3 l i -4— h • R D 4 2 2 • 4 P r c c i p . * 4 • 2 A v e Temp. o-i- -2 • 4 O N D J F M A M J J A S M O N T H Figure 5.15(c) Percentage change.in whole-ring pro p e r t i e s associated with a one standard d e v i a t i o n change i n monthly weather data. - 125 -M N D • 4 • 2 O - 2 - 4 • 4 • 2 O - 2 - 4 ! P r e c i p . 1 1 - - ^  - ^  ^ 1 33 j 1 1 i ! 1 1 1 1 1 1 1 '. Ave . T e m p . _ f£U _ Egg] _ _ s n . _ _ t GSB9 BSD KKJi 1 1 1 I 1 ' . . ' XXX 1 i : i M X D • 4 • 2 0 - 2 - 4 • 4 •2 0 - 2 - 4 [ P r e c i p . _ rasa _ Ave . Temp. -i i_ X X I I I I O N D J F M A M J J A S M O N T H Figure 5.15(d) Percentage change i n i n t r a - r i n g density extremes associated with.a one standard d e v i a t i o n change monthly weather data. - 126 -higher density. High July r a i n f a l l was associated with lower earlywood density. July weather was also the most influential i n affecting latewood density, high temperatures and r a i n f a l l resulting in lower density. In this case, the summer (July and August) minima were more closely related to density than the mean values (Appendix 15(e)). Summaries of weather effects on total ring widths and mean ring densities are shown in Fig. 15'(cK Widths were significantly related only to June 2 weather (R =0.33 for r a i n f a l l and 0.17 for minimum temperature), clearly as a result of the influence on earlywood, since fluctuations in this month's weather alter earlywood width by about twice as much as latewood width. Ring density 2 was most closely related to April temperatures (R = 0.16) , again through the effect on earlywood. Fig. 5.15(d) shows the relative influence of r a i n f a l l and average temperature on minimum and maximum ring densities. Only April temperatures were closely 2 related to MND (R =0.27), high values giving lower densities. Average July and September temperatures were related to MXD but minimum temperatures gave higher correlations for the three months July to September (Appendix 15(e)). The regres-sions were negative for the former two months and positive for the latter, pro-bably a reflection of a tendency for increasing moisture stress towards the end of summer. High temperatures would normally encourage cambial growth when ade-quate water i s available but in times of stress would result in reduced c e l l division and higher density, as evidenced by the occurrence of-., false, rings in areas subject to drought (McKinnell and Shepherd, 1971). Rainfall data themselves were not significantly related either to MND or to MXD. Interpretation of the above results in terms of the physiology of tree growth is not a simple matter since there are undoubtedly many complex interactions between the processes of photosynthesis, respiration and transpiration, as well as direct and indirect effects on food storage and u t i l i z a t i o n . It i s also - 127 -inadvisable to attempt to i s o l a t e the e f f e c t s of temperature and r a i n f a l l since they are intimately r e l a t e d . The analyses were performed a f t e r c e r t a i n assumptions were made, i.e. that the twenty-five observations were independent and that each s i t e sould behave s i m i l a r l y . Whether or not these are s t r i c t l y v a l i d , most of the c o r r e l a t i o n s revealed were i n t u i t i v e l y reasonable. The f a c t that maximum and minimum tempera-tures were i n some cases more c l o s e l y r e l a t e d to the r i n g components may be e i t h e r a r e s u l t of the method of determining the average ( m i n* + max.^^ w h i c h may not be the most su i t a b l e i n d i c a t o r i n terms of growing conditions, or may be a r e f l e c t i o n of temperature e f f e c t s on the balance of r e s p i r a t i o n and photo-synthetic processes. For example, high June minima are associated with s i g n i f i -cantly reduced earlywood widths and t h i s may be a r r e s u l t of increased night r e s p i r a t i o n leading to reduced food reserves a v a i l a b l e f o r cambia-1 a c t i v i t y . Despite the l i m i t a t i o n s of the data, some of the c o r r e l a t i o n s were s u r p r i s -i n g l y high. For example, 33% of the v a r i a t i o n i n t o t a l r i n g width was associated with June r a i n f a l l and 39% of the v a r i a t i o n i n earlywood density was r e l a t e d to minimum A p r i l temperatures. O v e r a l l , width components were subject to greater environmental modification than density components, i n several cases up to 5-7% accompanying a change of one standard deviation i n the monthly weather data. Density components varied only up to 2% under the same conditions. Subsequent analyses i n which the weather data were grouped by three- and six-month periods f a i l e d to account f o r more of the v a r i a t i o n i n r i n g components. These r e s u l t s confirm that weather fl u c t u a t i o n s are associated with a much smaller portion of the o v e r a l l v a r i a t i o n than the inherent tree-to-tree d i f f e r -ences shown i n Table 5.6. The amount of work done on c l i m a t i c e f f e c t s on Douglas-fir growth i s as yet i n s u f f i c i e n t to enable d e f i n i t e general patterns to be stated with c o n f i -dence since observations cover v a s t l y d i f f e r e n t tree ages and time spans (Kenn-- 128 -edy, 1961; Lassen and Okkonen, 1969; Smith, 1973). The current study was c a r r i e d out on stems much younger than those i n the above mentioned works. With more r e l i a b l e weather records (especially f o r Molalla) i t may have been possible to extend the analyses to t r y to account for the s i t e d ifferences i n r i n g components given i n Appendix 12(b) i n terms of the l o c a l climate. How-ever, other unrecorded factors such as s o i l f e r t i l i t y , moisture retention capa-c i t y and length of growing season are known to have an important e f f e c t on growth (Strand, 1964). I t has also been suggested that vegetational cover may influence the wood q u a l i t y of Douglas-fir (Hursey, 1970). One prominent feature only i n d i r e c t l y r e l a t e d to climate i s the occurrence of compression wood i n the Mo l a l l a growth ri n g s . This has been a t t r i b u t e d to the e f f e c t s of snow bending i n the young stems and should become l e s s evident as the trees age. Figure 5.16 and Appendix 16 give summaries of the p r e c i p i t a t i o n and average temperatures for the f i v e s i t e s over the period 1969-74. The patterns of v a r i a -t i o n are c l e a r l y s i m i l a r i n a l l cases,and, for the reasons o u t l i n e d above, i t i s dangerous to use s i t e d i f f e r e n c e s to explain differences i n r i n g components. For example, Nimpkish and Haney have very s i m i l a r monthly weather values, except that June and J u l y r a i n f a l l at the l a t t e r s i t e i s much greater. Without obtaining more information on the pl a n t i n g areas i t would be tempting to explain the differences i n earlywood widths (Nimpkish 3.4 mm; Haney 3.8 mm) i n terms of the highly s i g n i f i c a n t r e l a t i o n s h i p shown for June p r e c i p i t a t i o n i n F i g . 5:15 (a). The d i f f e r e n c e i n earlywood density (Nimpkish 0.292; Haney 0.284) might also be a t t r i b u t e d to the higher J u l y r a i n f a l l at Haney. On the other hand, Vall e y and Dorena with lower r a i n f a l l throughout the growing season have wider earlywood (4.4 mm and 4.0 mm, r e s p e c t i v e l y ) . In these cases i t might be hypo-thesized that the higher summer temperatures were responsible. These examples i l l u s t r a t e , however, that much more information i s required before general statements can be made about the influence of climate on wood formation. 129 -500 400 e £ c 300 o .- 200 a u *» c a. 100 z N i mpk i sh x Hanay a Mola l la v V o 11 ay o D o r t n a t. 3 et a. e t— F M A M J J A S O N D M o n t h Figure 5.16 Average temperature and p r e c i p i t a t i o n data f o r the provenance t r i a l . s i t e s , 1969-74. - 130 -Chapter 6. RESULTS AND DISCUSSION - CLONAL STUDY 6.1 Gravimetric Analyses Mean tree (ramet) values for average r i n g width and g r a v i m e t r i c a l l y determined wood density are given i n F i g . 6.1 with clones arranged i n ranks and r e s u l t s of Duncan's Mul t i p l e Range t e s t s shown. S i g n i f i c a n t differences occur i n both growth rate and density between clones, but s u b s t a n t i a l differences are also apparent within clones due to random environmental e f f e c t s . The core d e n s i t i e s ranged from .290 to .413 3 g/cm and the o v e r a l l average of .345 was much lower than that of any of the provenance t r i a l s i t e s . I t would therefore appear that the findings of Burdon and Harris (1973) i n r a d i a t a pine also hold true f o r Douglas-fir, i.e., that grafted material produces wood of s i g n i f i c a n t l y lower density than planted stock. H e r i t a b i l i t y values were ca l c u l a t e d to be 0.34 and 0.66 for r i n g width and density, r e s p e c t i v e l y , on an i n d i v i d u a l tree b a s i s . Regression analyses were performed on these data despite the small number of r e p l i c a t i o n s per clone and r e s u l t s are given i n Table 6.1. Eight regres-sion c o e f f i c i e n t s were found to be negative, and of these, f i v e proved to be s i g n i f i c a n t at the 10% l e v e l or better. C o e f f i c i e n t s of determination f o r these f i v e clones ranged from 0.45 to 0.80 and a t e s t of p a r a l l e l i s m and coin-cidence showed that although the y-axis intercepts were s i g n i f i c a n t l y d i f f e r e n t , the slopes were not. The mean slope was such that an increase i n mean r i n g 3 width of 1 mm would be associated with a density decrease of .012 g/cm . A l l other clones gave non-significant regressions, as d i d the bulked data for a l l trees. From the l i m i t e d data, i t would appear that .whereas there i s a tendency - 131 -•10.0 8-0 2 6.0 . - 4-0 01 2-0 E .40 u •35 § -30 a •25 R i n g W i d t h * 4-82 60 28 83 70 62 36 55 59 25 B a s i c D e n s i t y 4-82 36 59 83 25 60 55 28 70 62 C l o n e s Figure 6.1 Clon a l whole-core r i n g width and density data. (Horizontal l i n e s enclose means found to be not s i g n i f i c a n t l y d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l by Duncan's Mu l t i p l e Range t e s t s ) . - 132 -Table .6/1: Simple Linear Regression of Wood Density on Mean Ring Width f or Clonal Cores Clone No. Regression C o e f f i c i e n t R2 x 100 Sig n i f i c a n c e Level 25 - 0.01158 68.6 ** 28 - 0.01187 44.9 * 36 - 0.01629 80.1 *** 5.5 0.00181 2.9 h. s. 59 - 0.00467 14.7 n.s. 60 - 0.01076 45.0 * 62 - 0.00404 8.5 n.s. 70 - 0.01277 58.4 ** 82 - 0.00744 24.4 n.s. 83 0.11963 13.3 n.s. A l l - 0.05396 3.2 n.s. * = : s i g n i f i c a n t at the 10% l e v e l ** = s i g n i f i c a n t at the 5% l e v e l *** = s i g n i f i c a n t at the 1% l e v e l n.s. = not s i g n i f i c a n t - 133 -towards an environmentally-induced negative relationship within clones, there is no significant relationship between growth rate and density on a clone mean basis. For example, clones 82 and 83 could be classified as relatively slow-growing low density material, in contrast to clones 55, 62 and 70 which are fast-growing high density clones. The difference in wood weight production per unit height at breast height between clones 82 and 55 is almost 55%, due to growth rate and density differentials of 35% and 15% respectively. 6.2 Densitometric Analyses Densitometric scanning produced a series of yearly width and density com-ponent data. Figure 6.2 shows the density profile summaries, adjusted to give the two ring groups to be analysed, while Appendices 17 and 18 give tabulated data on ring component variation for clone means together with variance compo-nents for the between and within clone effects. Results of analyses of variance and Duncan's Multiple Range tests for the growth-ring components are given in Appendix 19. Standard densigrams give a better impression of clonal differences in intra-increment density distribution in Fig. 6.3. It i s immediately apparent that there are greater differences at the latewood side of the density scale than i n the earlywood. For instance, only clone 36 does not have i t s minimum 3 values in the range .20-.25 g/cm , whereas maxima can occur anywhere from class 3 .70-.75 to .85-.90 g/cm . Both lower mainland clones (82 and 83) are seen to have the least overall ranges, covering 11 density classes, in contrast to clones 25 and 62 which each cover 14 classes. In fact, the similarity in growth rate and density patterns between clones 82 and 83 suggests the possibi-l i t y that they may be related. Heritability estimates based on individual growth periods (see Chapter 3.3.2) are given graphically in Fig. 6.4 for a l l ring components. In a l l cases there is a pronounced trend of increasing values with age over the period under examination, ranging from<0.20 in the inner rings to between 0.3 and 0.7 in the outer rings, with - 134 -C L O N K 25 C L O N E 36. 1 9 6 9 1570 1971 1972 1973 2 : Lui a I I J J 1969 1970 1971 1972 197X 1969 1970 1971 1972 1973 C L O N E 59 UJ a l i 1969 1970 1971 1972 1973 Figure 6.2 Clonal densitometric p r o f i l e s . - 135 -Figure 5.2 (contd.) C l o n a l d e n s i t o m e t r i c p r o f i l e s . - 136 -Figure 6.3 Standard'densiarams for a l l clones, growth p e r i o d 1969-73. - 137 -Figure 6.4 H e r i t a b i l i t y estimates by growth periods. - 138 -apparent random f l u c t u a t i o n from r i n g to r i n g . These r e s u l t s are i n strong contrast to those of N i c h o l l s (1967) who found a decrease i n h e r i t a b i l i t y out-wards from the p i t h f o r basic density i n radiata pine. A betterrconcept of the r e l a t i v e magnitudes of the estimates can be obtained from Table 6.2. Here the c a l c u l a t i o n s were done on the 5-ring means and expressed both on an i n d i v i d u a l tree basis and on a clone mean ba s i s . Both sets of figures are included since reports i n the l i t e r a t u r e almost i n v a r i a b l y c i t e the former while the l a t t e r are of more p r a c t i c a l use i n estimating genetic gains from c l o n a l s e l e c t i o n . As indicated i n F i g . r6.4' s u b s t a n t i a l l y 2 higher h estimates were obtained for the outer r i n g s . This was p a r t i c u l a r l y noticeable for mean r i n g density (RD) and percent latewood (PLW). 2 The h estimates for mean r i n g width and density i n the outer rings (0.33 and 0.64 are i n very good agreement with the whole-core values given e a r l i e r (0.34and 0.66) and with the r e s u l t s of other workers, and so i t i s reasonable to assume that the other values are also f a i r l y representative. Latewood c h a r a c t e r i s t i c s are more strongly under genetic control than earlywood i n a l l cases, but despite the medium to high estimates f o r LD and MXD (0.57and 0.61) neither UNI or RNG, describing i n t r a - r i n g contrasts, was found to be very high (0.23and 0.38). The increase of h e r i t a b i l i t y values with age w i l l have important p r a c t i c a l consequences i n that assessments of young material w i l l give underestimates. There i s no i n d i c a t i o n from the current data that the trends have l e v e l l e d o f f so i t i s quite possible that even the data for the outer r i n g group may 2 underestimate the h values applicable to mature crops. Another method of estimating broad sense h e r i t a b i l i t y i s to use the pheno-t y p i c variance of a natural stand growing under s i m i l a r conditions i n conjunc-t i o n with the environmental variance from c l o n e s , i . e . , ,2 2 2,2 h = c - a /a a , r r p TABLE 6/2':: H e r i t a b i l i t y Estimates for Clonal Ring Components Ring Group Model 2 h Estimates RW EW LW PLW RD ED LD MND MXD UNI RNG 1-5 Indi v i d u a l Tree .19 .19 y.50 .44 .44 .29 .43 .26 .52 .17. .23 Clone Mean .65 .65 .88 .86 .86 .77 .86 .74 .90 .62 .71 6-10 Indi v i d u a l Tree .33 .24 .67 .69 .64 .35 .57 .35 .61 .23 .38 * (.36) (.37) (.44) (.79) (.69) (.17) (.66) (.00) (.58) (.38) (.36) Clone Mean .80 i l l .94 .95 .93 .81 .91 .81 .93 .70 .83 * Values derived using provenance t r i a l phenotypic variances. Component abbreviations as given i n F i g . 3.3. - 140 -where.of i s the phenotypic variance of the natural population and °^ i s the 2 ramets within clones variance. In the absence of c estimates f o r the area p under study, i t seems reasonable to use the pooled variance f o r trees within p l o t s from the bulked provenance data. These trees were of s i m i l a r age and also growing i n planted l i n e s . H e r i t a b i l i t y estimates using t h i s information are given i n brackets i n Table 6.2 for comparision and prove to be s u r p r i s i n g l y close i n some cases to the c l o n a l data. This must be considered highly s a t i s -factory i n view of the d i f f e r e n t environments covered by the provenance study and i s a good i n d i c a t i o n that the two t r i a l s tend to complement one' another. V i s u a l examination of the c l o n a l data suggested that genetic differences between clones were being more strongly expressed with age while the environ-mental variance within clones remained the same or even decreased (Figs. 6.5 and 6.6). The /'^2 values for the pooled standard deviation of ramets within clones and variance components i n Appendix 18 J confirm t h i s . The a v a i l a b i l i t y of yearly data allows the c a l c u l a t i o n of variance compo-nents f o r factors contributing to the t o t a l observed v a r i a t i o n . For t h i s pur- \ ; pose the two r i n g groups were treated separately and s l i g h t l y d i f f e r e n t models were employed. With rings 1-5, the growth periods were treated as a f i x e d e f f e c t since strong inherent trends are found i n t h i s region, whereas growth periods 1968-72 were analysed as random e f f e c t s , according to the models given i n Appendix 20.- In t h i s case, the two cores per tree were treated as random samples i n order to obtain an estimate of the component for year x tree i n t e r -action which otherwise , would have been included i n the er r o r term. Figure\6!:7 shows the contributions of each f a c t o r , expressed as a percentage 2 of the r e l a t i v e l y constant term f o r within-clone variance, . Apart from the increasing genetic contribution from c l o n a l e f f e c t s , i t can be seen that the influence of growth period i s greatly reduced for most properties outside - 141 -• 4 5 •40 ro - 3 5 £ u CO • 3 0 to - 2 5 c Q o - 4 0 to CO • 3 5 • 3 0 • 2 5 R i n g s 1 - 5 t 3 = .023 3 6 8 2 8 3 5 9 2 5 5 5 2 8 7 0 6 2 6 0 R i n g s 1 9 6 9 - 7 3 J 6 = .021 8 2 3 6 8 3 5 9 6 0 2 8 5 5 7 0 2 5 6 2 C l o n e s Figure 6.5 Clona l core d e n s i t i e s f o r the inner (rings 1-5) and outer (rings 1969-73) growth zones. (Horizontal l i n e s enclose means not s i g n i f i c a n t l y d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l ) . a = variance component f o r ramet-to-ramet v a r i a t i o n . - 142 -°/o 40 30 20 10 0 30 %» 20 10 0 Rings 1-5 82 36 59 83 25 60 62 28 5 5 70 R i n g s 6 - 1 0 61 = 3-2 T 82 83 36 60 59 25 28 70 62 5 5 C l o n e s Figure 6.6 Clonal core latewood percentages by growth zones. (Horizontal l i n e s enclose means not s i g n i f i c a n t l y d i f f e r e n t at the 5% l e v e l ) . 0"T = variance component f o r ramet-to ramet v a r i a t i o n . - 143 -4 • RW 2 0 EW 2 o LW 2 o PLW 2 o RD 2 o ED 2 o L D 2 o MND 2 o MXD 2 o Ifpf M I L I T " """^  i a n i - i r n _ESS£ZZL n ^2 ^2 6 T 6 C — Rings 1-5 — i — i wm—, 'YT — Rings 19(59-73 Figure 6. 7 Variance components as a percentage of 6 (Component abbreviations as given i n F i g . 3.3) - 144 -the inner f i v e rings. In general, the i n t e r a c t i o n terms contribute only a small amount to the t o t a l v a r i a t i o n s . Since the main objective of the c l o n a l study was to obtain h e r i t a b i l i t y estimates f o r r i n g component characters, i t i s necessary to discuss some of the l i m i t a t i o n s of the r e s u l t s . F i r s t l y , the data are s t r i c t l y speaking only applicable to the population studied under the given environmental condi-t i o n s . The population consisted of ten clones planted i n unreplicated l i n e s at Cowichan Lake, but i n t h i s case the ramets were confined to an area of l e s s than one acre on l e v e l ground, and. .there was no reason to suspect that e n v i r -onmental influences between the l i n e s were any d i f f e r e n t from those along l i n e s . The l i m i t e d numbers of clones and ramets, and the lack of r e p l i c a t i o n must be considered a disadvantage, but i n the absence of a l t e r n a t i v e material t h i s was unavoidable. There had been some suspicion that i n c i p i e n t g r a f t incompatability i n apparently healthy trees may have aff e c t e d growth and wood properties of some ramets. However, the f a c t that i n almost a l l cases the within-clone v a r i a t i o n i n the outer r i n g group d i d not exceed that of the inner rings was taken as an i n d i c a t i o n that t h i s was not a serious source of e r r o r . On the other hand, there may have been scion x rootstock i n t e r a c t i o n s of a l e s s serious nature cont r i b u t i n g to the observed ramet v a r i a t i o n . The properties assessed here applied to the breast height p o s i t i o n only and may not represent the s i t u a t i o n throughout the e n t i r e stem. I t does seem l i k e l y that c l o n a l differences would become l e s s marked with increasing height, r e f l e c t i n g the r a d i a l / t r e n d observed at breast height. The success of s e l e c t i o n for any property depends not only on the v a l i d i t y of the h e r i t a b i l i t y estimate but also on the magnitude of the genotype x envir-onmental i n t e r a c t i o n s . This could not be assessed d i r e c t l y i n the current - 145 -study i n v o l v i n g one l o c a t i o n only, but i n d i c a t i o n s from the l i t e r a t u r e are that i t i s of small s i g n i f i c a n c e f o r wood properties except perhaps where extreme s i t e types are used (McKimmy and Nicholas, 1971; Burdon and H a r r i s , 1973). Some support for t h i s conclusion can be obtained from the analyses of variance performed on the c l o n a l data which showed the year x clone i n t e r a c t i o n compo-nent to be of minor importance. Given these l i m i t a t i o n s , i t i s possible to c a l c u l a t e the expected gain from c l o n a l s e l e c t i o n from: Gain = h . 0 . i c where i s an estimate of the between-clone variance, and i i s the i n t e n s i t y of s e l e c t i o n expressed i n units of phenotypic standard deviation. Estimates of gain from c l o n a l s e l e c t i o n were calculated using the h e r i -t a b i l i t y of clone means, assuming a conservative s e l e c t i o n proportion of 10% ( i - 1.6 according to Falconer (1964)). Results are given i n Table 6.3 for rings 1969-73. Among the width components, LW and PLW show the greatest p o t e n t i a l f o r improvement through s e l e c t i o n , with estimated predicted gains of 45% and 36% r e s p e c t i v e l y . EW and RW give values of 12% and 15%. I f the strong genetic co n t r o l of latewood width i s maintained i n older wood where the percentage of latewood i s much higher, i t might be expected that the gain i n breeding for growth rate would be greater than that found i n the corewood. Of the density components, i t i s i n t e r e s t i n g that RD gives the highest estimated gain at 12%, despite the complex nature of the property. This i s of great importance, since numerous studies have determined the influence of density on product y i e l d and q u a l i t y , whereas the e f f e c t s of other c h a r a c t e r i s -t i c s remain l a r g e l y obscure. Predicted gains f o r the other components range from 6% f o r UNI to 11% f o r MND, showing that any one could be a l t e r e d by - 146 -TABLE 6.3: Predicted Gain from 10% Clonal S e l e c t i o n Ring Component Mean Estimated Estimated 6 2 c Predicted Gain % Gain RW mm 6.82 .80 .50 3J.01 15 EW 5.40 .72 .23 .65 " 12 LW „ 1.41 .94 .17 .64 45 PLW % 20.-66 .95 22.22 7.35 36 ™ / 3 RD g/cm .358 .93 .000806 .044 12 ED .275 .81 .000195 .020 7 LD .680 .91 .000650 .039 6 MND .208 .81 .000259 .023 11 MXD .. .839 .93 .002085 .070 8 UNI » .405 .70 .000322 .024 6 RNG „ .630 .83 .001118 .049 8 * on a clone mean basis (8 ramets/clone) Component abbreviations as shown i n F i g . 3.3. - 147 -s e l e c t i v e breeding should t h i s be desired. For instance,the contrast between earlywood and latewood density extremes could be reduced by using clones of low RNG values, such as 82 and 83. I t i s d i f f i c u l t to assess the r e l a t i o n s h i p s between v a r i a b l e s on a clone mean basis with a sample of only ten clones, but the c o r r e l a t i o n matrix given i n Appendix 21 shows the r e s u l t s of such an analysis for the two r i n g groups. S i g n i f i c a n t c o r r e l a t i o n s between width and density variables. -.are seen to be few apart from the already well established RD/PLW r e l a t i o n s h i p 2 (R = 0.75and 0.77). i n the inner rings, MND i s negatively r e l a t e d to RW and EW, but t h i s e f f e c t was not apparent i n rings 1969-73. Mean r i n g density i s highly correlated with earlywood density i n rings 1-5 but not with latewood density. In the older r i n g s , both earlywood and latewood d e n s i t i e s exert a s i m i l a r influence on the mean r i n g value. These r e l a t i o n s h i p s are no doubt a r e s u l t of the inherent patterns of v a r i a t i o n i n ED, LD and PLW outwards from the p i t h and point out the dangers of g e n e r a l i z i n g about r i n g component i n t e r - r e l a t i o n s h i p s . In the outer r i n g group, mean r i n g density i s seen to be c l o s e l y r e l a t e d to the maximum i n t r a - r i n g contrast (RNG) i n a p o s i t i v e manner, but not to the d i f f e r e n t i a l between earlywood and latewood d e n s i t i e s . I f these r e s u l t s are r e l i a b l e i t would appear that s e l e c t i o n for high density clones would r e s u l t i n wood with higher values for a l l r i n g density components except UNI. However, UNI and RNG are seen to be c l o s e l y c o r r e l a t e d i n both r i n g groups. Thus more data would be required to determine the true genetic c o r r e l a t i o n s between the wood density c h a r a c t e r i s t i c . - 148 -The current study has shown that wood density of Douglas-fir grafted material i s considerably lower than that of planted trees, although the values fo r i n t r a - r i n g density contrasts were found to be comparable to those found at the provenance t r i a l s i t e s described e a r l i e r . Between clones, there was no apparent r e l a t i o n s h i p between growth rate and density, but within c e r t a i n clones there was a tendency towards a nega-t i v e c o r r e l a t i o n . I t i s not known whether the lack of r e l a t i o n s h i p i n the other clones was a genuine e f f e c t or a r e f l e c t i o n of the l i m i t e d number of ramets used. Broad sense h e r i t a b i l i t i e s , c a l c u l a t e d by growth periods for a l l r i n g components, were shown to increase outwards from the p i t h . Environmental influences appeared to remain more or less constant with age while clone differences increased. Interactions of years with clones and ramets were found to be comparatively small e f f e c t s , but probably accounted for the random f l u c t u a t i o n of h e r i t a b i l i t y values around the general trend. The h e r i t a b i l i t y estimates were such that s i g n i f i c a n t gains could be achieved through c l o n a l s e l e c t i o n f o r r a d i a l growth and wood c h a r a c t e r i s t i c s . Based on s e l e c t i o n of 10% of clones i t was estimated that mean density could be a l t e r e d by 12% and other properties by between 6% and 11% i n the f i r s t s e l e c t i o n . Percent latewood, with a h e r i t a b i l i t y estimate of 0.69 (0.95 on a clone mean basis) was shown to be highly g e n e t i c a l l y c o n t r o l l e d and c l o s e l y r e l a t e d 2 to mean r i n g density (R about 0.75). This may be fortunate f o r those i n t e r -ested i n wood q u a l i t y but lacking the time or equipment to measure density by conventional methods. Even v i s u a l screening of increment cores f o r the propor-t i o n of latewood could lead to s i g n i f i c a n t modification of wood density. - 149 -Many of the wood density components were c l o s e l y r e l a t e d , but more data would be required to e s t a b l i s h the e f f e c t s of s e l e c t i o n for more than one character at the same time. 6.3 Phenological Observations 6.3.1 Clonal differences Clonal mean patterns of shoot growth and r a d i a l increment are shown i n F i g . 6.8. Cambial a c t i v i t y at breast height commenced about s i x weeks p r i o r to the f i r s t signs of f l u s h i n g , and by the time a l l buds had burst, r a d i a l growth was already about 30% complete. These observations support the findings of Dimock (1964) who reported that r a d i a l growth i n 16-year-old Douglas-fir preceded i n i t i a t i o n of a p i c a l growth. In Figv 6.V8 i t i s shown that, although there were dif f e r e n c e s between clones i n the rate of diameter increment, i t appeared that i n i t i a t i o n and cessation occurred at about the same time i n a l l groups. The ranking by magnitude of 1975 diameter growth corresponds well to the mean r i n g width data given i n F i g . 6.1. Clonal differences were observed both i n time of f l u s h i n g and i n duration of shoot growth. Figure 6.95,shows the v a r i a t i o n between ramets for the crown and diameter properties measured. In general, the timing of the f i r s t bud burst i s more c l e a r l y defined as a c l o n a l c h a r a c t e r i s t i c than the other parameter, suggesting strong genetic c o n t r o l . Flushing data i s shown i n more d e t a i l i n F i g . 6.10 by measurement dates. Here i t i s c l e a r that the time of f i r s t f l u s h i n g and the rate of progress from group I (up to 10% of the buds burst) to group V (over 75% flushed) d i f f e r s between clones. The period between the f i r s t clone (62) and the l a s t (82) to reach f u l l f l u s h i n g was 21 days and there appeared to be a strong p o s i t i v e c o r r e l a t i o n between f l u s h i n g date and rate of devel-- 150 -F i g u r e 6.8 C u m u l a t i v e s h o o t and r a d i a l g r o w t h b y c l o n e s , 1 9 7 5 . -151 -CLONE 33 CA I 1 " BB H- sc -H cc H CLONE 55 H H C L O N E 62 H H H C L O N E 70 I 1-B H H C L O N E 82 H H H CLON E 83 H 1—I H :'- • l-H APRIL MAY JUNE JULY AUGUST SEPT, F i g u r e 6 . 9 Span o f c a m b i a l and p h e n o l o g i c a l e v e n t s b y c l o n e s BB - 10% o f buds b u r s t SC - 95% s h o o t g r o w t h c o m p l e t e d CA - 5% o f c a m b i a l g r o w t h CC - 95% o f c a m b i a l g r o w t h - 1 5 2 -CL O c. O Ol c IY m TT I O 12 m n i o 12 m n T o Tv TTi TT I o Tv in TT T o Tv ml n T : (Days Group I-Clone 02 Group Y ) DG = 3 T 1 1 1 1 r Clone 36 DG = 5 -i -i 1 1 1 r Clone 83 DG = 5 "I"'" )"' T 1 1 r Clone 55 DG = 5 "I" " "1 Clone 70 0 6 ' 7 i i 1 i " a T i •V„„i,,.|.„ t,„t Clone 82 DG = 9 27/4 . 3/5 , 10/5 , 17/5 . 24/% 30/4 ' 7/5 I 13/5 ' 20/5 7 Day and Month Figure 6.10 Progress of c l o n a l f l u s h i n g c h a r a c t e r i s t i c s . (Based on the a l l o c a t i o n of ramets to f l u s h i n g groups as out l i n e d i n Appendix 5). - 153 -opment from group I to group V. This l a t t e r c h a r a c t e r i s t i c v a r i e d from three days for clone 62 to nine days f o r clone 82. While the timing of f l u s h i n g i s s c l e a r l y a genetic response, the time to reach f u l l f l u s h i n g could be a f f e c t e d by current weather conditions and may therefore be more of a genotype x environment i n t e r a c t i o n . Cessation of shoot growth varied between clones (Fig. 6.9) although the actual patterns of growth were s i m i l a r (Fig. 6.8). There was some i n d i c a t i o n that the clones which were f i r s t to f l u s h also completed shoot growth ea r l y . Densitometric scans of the 1975 growth increment (2 cores/ramet, 2 scans/ core) were summarized to give r i n g component data and mean c l o n a l p r o f i l e s . Figure 6.11 gives the p r o f i l e data along with the derived standard densigrams. The patterns were s i m i l a r to those already given i n F i g . 6.3 for rings 1969-73 combined, and again i l l u s t r a t e the large c l o n a l d ifferences i n density d i s t r i b u t i o n i n the latewood portion. From the shoot and diameter measurements, a seri e s of parameters was calculated for each ramet to be used as independent variables i n regression analyses with r i n g components as dependent v a r i a b l e s . The r i n g p r o f i l e s were used to determine the density l e v e l s at time of shoot growth cessation and i n conjunction with diameter measurements to derive the duration of earlywood 3 and latewood formation based on the density boundary of 0.50 g/cm . P r i o r to c o r r e l a t i o n and regression analyses, each v a r i a b l e was tested for c l o n a l differences using the U.B.C. MFAV package (analysis of variance/ covariance). A complete l i s t of the variables i s given i n Table 6.4 while Table 6.5 summarizes the r e s u l t s of Duncan's t e s t s on the analysis of variance data. Appendix 2 2 gives the c o r r e l a t i o n matrix f o r a l l the parameters. The densitometric data complement the r e s u l t s presented e a r l i e r i n that the rankings of clone means remained almost i d e n t i c a l to those i n the complete 1 - 154 -40 20 0 40 20 0 40 20 % 0 40 20 0 40 20 0 40 20 M a a n Ew I Lw Clona 82 I Clona 83 "l"' ""i ' i •••• [ Clona 36 Clona 70 Clona 62 I K •8 D e n s i t y g / c m ' P r o f i l e s Figure 6.11 Clonal densigrams and r i n g p r o f i l e s f o r the 1975 increment. SC = 95% of shoot growth completed. Ew - Earlywood Lw - Latewood - 155 -TABLE 6.4: Phenological and Density Variables Used i n S t a t i s t i c a l Analyses a) From densitometric analyses Y Ring width , mm ' RW Y^ Earlywood width , mm EW Y 3 Latewood width , mm LW Y 4 Percentage latewood ( ¥ 3 / ^ x 1 0 ° ) » % P L W Y Ring density , g/cm^ RD 3 Y„ Earlywood density , g/cm ED 3 Y Latewood density , g/cm LD 3 Y Minimum density , g/cm MND 3 Y Maximum density , g/cm MXD •x Y,„ Uniformity (Y -Y ) g/cm UNI 10 7 6' Y Range i^g-YQ) , g/cm RNG b) From f i e l d observations X l T o t a l diameter increment, mm X2 Number of days of diameter growth , days X 3 Rate of diameter increment (X^/X^), mm/day X4 Diameter growth to 95% shoot growth ( t h e o r e t i c a l . earlywood), mm X 5 Number of days f o r X^ , days X 6 Rate of earlywood production (X^/X,-), mm/day X7 Theoretical latewood (X^-X^) , mm X 8 Duration of latewood growth (X -X ), days <* 5 X 9 Rate of latewood production (X -X ), mm/day x i o •: T h e o r e t i c a l latewood percentage ((X^X^) x 100), % X l l T o t al shoot growth to 30th July , mm X12 Duration of shoot growth to 30th July, days X13 Rate of shoot growth to 30th J u l y , mm/day X14 Date of f i r s t f l u s h i n g - days a f t e r A p r i l 25th X15 Number of days from f l u s h i n g group I to group V, days c) From density p r o f i l e s and f i e l d observations combined 3 X Number of days of earlywood production (up to 0.50 g/cm ), days 3 X Number of days of latewood production (over 0.50 g/cm )> days TABLE 6.5: Duncan's Multiple Range Tests on Clonal Means for Phenological and growth-ring variables. Means Means Means Means RW mm 82 83 70 36 62 55 6.04 6.24 7.00 7.20 8.09 8.20 7.14 MND g/cm 82 36 70 55 83 62 .216 .219 .222 .223 n - s .224 .233 .223 x 4 mm 82 83 70 36 55 62 7.8 8.4 8.9 10.6 11.0 11.2 9.6 x 11 mm 83 65.3 62 82.3 82 84.1 36 97.5 70 111.6 55 120.5 93.5 EW mm 82 83 70 62 36 55 4.94 5.25 5.29 6.04 6.07 6.14 5.62 MXD g/cm 82 83 36 70 55 62 .729 .768 .825 .835 .916 .935 .835 x 5 days 83 62 55 70 82 36 62 63 66 67: 68 68 66 x 12 days 82 33 83 34 62 35 70 36 55 38 36 39 36 LW mm 83 0.99 83 15.0 36 1.14 36 15.6 82 1.16 82 18.9 70 1.71 70 24.5 62 2.05 55 25.2 55 2.06 62 25.6 1.52 UNI g/cm 82 83 36 70 62 55 .368 .382 .412 .429 .432 .482 .417 x 6 mm/day 82 70 83 36 55 62 0.103 0.121 0.123 0.141 0.150 0.159 0.133 x 13 mm/day 83 1.6 62 2.1 82 2.1 36 2.4 70 2.8 55 3.0 2.3 PLW % 20.8 RNG g/cm 82 .513 83 .545 36 .606 70 .614 55 .693 62 .702 .612 x 7 mm 83 82 36 55 62 70 4.2 4.2 5.2 5.3 n.s. 5.6 5.7 5.0 x 14 days 62 3.0 36 3.6 83 4.2 55 5.6 70 9.0 82 17.4 7.1 82 83 36 70 62 55 .339 .340 .340 .388 .406 .406 .370 x 1 mm 82 83 70 36 55 62 12.1 12.6 14.7 15.8 16.3 16.9 14.7 x 8 days 82 55 70 36 62 83 67 72 73 73 74 77 73 x 15 days 62 55 83 36 70 82 3.0| 4.8 5.0 5.0 7.0 8.6 5.6 ED g/cm 82 .270 82 36 .276 83 83 .281 36 70 .283n.s. 70 55 .285 62 62 .296 55 .282 x 2 days 82 63 55 83 70 36 134 138 138 139 n.s. 140 141 138 x 9 mm/day 83 82 36 55 62 70 0.049 0.057 0.004 0.066n. 0.068 0.070 0.062 x 16 days 62 55 82 70 36 83 79 I 97 116 117 119 125 109 LD g/cm .638 .663 .688 .712 .728 .767 .699 x 3 mm/day 82 83 70 36 55 62 0.081 0.082 0.095 0.102 0.106 0.110 0.096 10 »/•> 55 36 83 62 82 70 32.3 32.7 32.8 32.9 36.2 39.0 34.3 x 17 days 83 14 82 18 36 20 70 23 55 42 62 59 29 n.s. = not s i g n i f i c a n t d i f f e r e n t at the 5% l e v e l . V e r t i c a l l i n e s enclose means not s i g n i f i c a n t l y d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l . - 157 -c l o n a l analysis (Appendix 19) . Differences i n r i n g width (RW) were mainly accounted f o r i n terms of latewood widths (LW) i n so far as di f f e r e n c e s i n earlywood (EW) were no n - s i g n i f i c a n t . S i m i l a r l y , r i n g density (RD) ranking i s a f f e c t e d more by latewood density (LD) than by earlywood density (ED) since differences i n the l a t t e r were non - s i g n i f i c a n t . The s i g n i f i c a n t d i f f -erences i n UNI and RNG were thus a r e s u l t of the latewood c h a r a c t e r i s t i c s . Larson (1969) showed that from a t h e o r e t i c a l viewpoint earlywood could be defined as that p o r t i o n of the growth r i n g formed p r i o r to cessation of shoot and needle growth. On t h i s basis, a measure of earlywood width was derived for each ramet by c a l c u l a t i n g the diameter growth to 95% shoot growth. The remaining p o r t i o n of the r i n g was designated as latewood. In contrast to the densitometric r e s u l t s , the analyses showed s i g n i f i c a n t d i f f e r e n c e s i n earlywood width (x 4) and no n - s i g n i f i c a n t differences i n latewood width (x 7). Ranking of clone means was s i m i l a r when the two sets of widths are compared. The earlywood and latewood values derived from the shoot growth pattern and diameter increment were used to derive a value f o r percentage latewood (x 10) which i n turn was employed i n conjunction with the 100-point density p r o f i l e of each ramet to determine the density of the wood at the earlywood-latewood boundary. These density l e v e l s are shown i n Fig."'.15.11 where i t can be seen that the phenological t r a n s i t i o n (i.e. completion of 95% shoot growth) corresponded to d i f f e r e n t density l e v e l s i n the clones. In the low density groups (clones 82, 83 and 36) the derived density was between 0.31 and 0.32 3 3 g /cm whereas i n the others i t was between 0.37 and 0.39 g/cm . This i l l u s -t r a t e s that phenological events are not n e c e s s a r i l y associated with a consis-tent l e v e l of wood c h a r a c t e r i s t i c s . The f a c t that density values at the completion of shoot growth are s i g -n i f i c a n t l y lower than the selected densitometric earlywood,-latewood boundary - 158 -i s of l i t t l e s i g n i f i c a n c e and i n a l l p r o b a b i l i t y r e f l e c t s the proximity to the l i v e crown. I t has been c o n c l u s i v e l y demonstrated that i n i t i a t i o n of cambial a c t i v i t y proceeds b a s i p e t a l l y whereas latewood formation occurs acr o p e t a l l y (Fraser, 1952; Larson, 1969). Thus cessation of crown a c t i v i t y may be associated with i n i t i a t i o n of latewood at the base of the stem but w i l l precede t h i s event at higher l e v e l s . The important feature of the observations i s that the t h e o r e t i c a l c r i t e r i o n f o r the earlywood-latewood boundary represented a v a r i a b l e density l e v e l i n the clones examined. I t may also be of s i g n i f i c a n c e that the two methods of assessing percentage latewood give d i f f e r e n t absolute and r e l a t i v e values. Based on shoot growth patterns the figures are considerably higher (32.3 - 39.0% vs. 15.0 - 25.6%) and are l e s s discriminatory. 3 Using the a r b i t r a r i l y f i x e d density boundary of 0.50 g/cm i t was found that the duration of both earlywood (x 16) and latewood (x 17) forma-t i o n v a r i e d s i g n i f i c a n t l y between clones. High density trees tended to complete earlywood deposition i n a shorter time and to take longer for latewood formation. 6.3.2 Relationships between phenological and growth r i n g c h a r a c t e r i s t i c s In view of the reports i n d i c a t i n g an a s s o c i a t i o n between time of f l u s h -ing and mean wood density (Chapter 2.5), c o r r e l a t i o n and regression analyses were performed using the v a r i a b l e s l i s t e d i n Table 6.4. Ramet values were treated as independent observations since there were not enough clone mean values to provide a s u f f i c i e n t number of degrees of freedom for e f f i c i e n t t e s t i n g of hypotheses. As expected, many of the properties were found to be i n t e r r e l a t e d (Appendix 22.) but only a summary table w i l l be discussed here. C o e f f i c i e n t s - 159 -of determination f o r simple l i n e a r regressions l i n k i n g the growth-ring components with some crown c h a r a c t e r i s t i c s measured are given i n Table 6.6 Ring-width components tended to be p o s i t i v e l y c o r r e l a t e d with shoot growth data,'A.e.,, trees with vigorous crowns had l a r g e r diameters. Time of f i r s t f l u s h and rate of f l u s h i n g were negatively c o r r e l a t e d where the r e l a t i o n s h i p s were s t a t i s t i c a l l y s i g n i f i c a n t . These l a t t e r e f f e c t s accounted for only 14-16% of the v a r i a t i o n i n width components and would be of l i t t l e use f o r p r e d i c t i o n of growth c h a r a c t e r i s t i c s . Percentage latewood was not s i g n i f i c a n t l y r e l a t e d to any of these crown measures. ' The best i n d i v i d u a l a s s o c i a t i o n was between earlywood width (EW) and duration of shoot growth (x 12), which accounted f o r 32% of the earlywood v a r i a t i o n . Among the r i n g density components latewood density (LD), and hence uniformity (UNI), was the only one s i g n i f i c a n t l y r e l a t e d to shoot growth. The p o s i t i v e c o r r e l a t i o n suggests that the l a r g e r crowns tended to contribute more to c e l l - w a l l deposition r e l a t i v e to growth rate during latewood forma-t i o n . Again, however, the r e l a t i o n s h i p , although s i g n i f i c a n t , d i d not account for a large p o r t i o n of the v a r i a t i o n . Neither r i n g (RD), earlywood (ED), nor minimum (MND) d e n s i t i e s were r e l a t e d to time of f l u s h i n g , despite the f a c t that Figures 6.9 - 6.10 show the highest/ density clone (62) to f l u s h f i r s t and the lowest density clone (82) to f l u s h l a s t . In contrast, both latewood and maximum (MXD) d e n s i t i e s were negatively r e l a t e d to f l u s h i n g date. Thus,trees which flushed l a t e tended to develop lower latewood density c h a r a c t e r i s t i c s , hence lower i n t r a - r i n g contrast values (UNI and RNG). This c o r r e l a t i o n may be the important one i n that i f the phenology/ wood density r e l a t i o n s h i p becomes more pronounced with age as suggested by Harris et al. (1971) , then mean r i n g density may be s i g n i f i c a n t l y a l t e r e d . The age e f f e c t could be due to the f a c t that latewood percentage increases - 160 -TABLE 6.6: C o e f f i c i e n t s of Determination f o r Ring Component/ Crown Property Relationships Ring Crown Property Component X 11 X 12 X 13 X 14 X 15 RW (+) .20** (+) .25*** (+) .20** (-) .15** (-) .15** EW (+) .17** ( ±) .32*** (+) .17** (-) .16** ( — -LW (+) .13** (+) .13** -— (-) .14** PLW -— — -RD -— (-) .16** ED -— (-) .11* LD (+) .14** (+) .11* (+) .14** (-) .20** (-) .25*** MND -— — -MXD (-) .28*** (-) .42*** UNI (+) .21*** (+) .10* (+) .21*** (-) .12* (-) .14** RNG — (-) .26*** (-) .39*** C +) = p o s i t i v e l i n e a r c o r r e l a t i o n (-) = negative l i n e a r c o r r e l a t i o n = not s t a t i s t i c a l l y s i g n i f i c a n t * = s i g n i f i c a n t at the 10% l e v e l ** = s i g n i f i c a n t at the 5% l e v e l *** = s i g n i f i c a n t at the 1% l e v e l /Abbreviations as given i n Table 6.4. - 161 -with distance from the p i t h and hence latewood density exerts an increasing influence on r i n g density. The r e s u l t s are thus i n general agreement with those of workers dealing with older Douglas-fir trees (Kennedy, 1970; Smith, 1973). C o n f l i c t a r i s e s , however, when the density component data of Heger et al. (1974)are" examined for two f l u s h i n g groups. In that case, differences i n mean r i n g density appear to be a t t r i b u t a b l e more to earlywood density l e v e l s . With the exception of MND, a l l density components were negatively r e l a t e d to the rate of progress from f l u s h i n g group I to group V (x 15). I t i s possible that t h i s shoot c h a r a c t e r i s t i c i s an expression of crown vigour since the clone mean ranks are very s i m i l a r to those for rate of shoot growth (x 13 - Table 6*5). on t h i s b a s i s , i t might be hypothesized that those trees with l e s s vigorous crowns supply a smaller amount of car-bohydrates for c e l l wall formation.per u n i t of r a d i a l growth and hence tend to have lower d e n s i t i e s across the r i n g . Latewood maximum density i s most highly c o r r e l a t e d with t h i s c h a r a c t e r i s t i c . In conclusion, i t can be stated that although both phenological and growth r i n g properties appear to be under various degrees of genetic con-t r o l , c o r r e l a t i o n s between the two sets of v a r i a b l e s are i n general of l i t t l e value for p r e d i c t i v e purposes. Crown observations are. rapid to obtain but must be repeated at short regular i n t e r v a l s i n order to give quantitative data on tree d i f f e r e n c e s . Inferences regarding wood pro-p e r t i e s from such phenological data would be imprecise at best and of l i t t l e p r a c t i c a l value. Mean r i n g density was again shown to be very c l o s e l y r e l a t e d to per-centage latewood as defined densitometrically (*r = 0.888, Appendix 22). However, t h i s measure of latewood was not s i g n i f i c a n t l y r e l a t e d to any - 162 -crown c h a r a c t e r i s t i c examined. The t r a n s i t i o n from earlywood to latewood i s also important i n in f l u e n c i n g i n t r a - r i n g density contrast since the greater the length of time a v a i l a b l e f o r latewood formation (x 17); (1) the greater the percentage latewood (r = 0.692), (2) the greater the latewood density (r = 0.500), (3) the greater the uniformity (r = 0.361) and range (0.638). 6.4 Influence of water d e f i c i t during 1972 The density p r o f i l e s i n F i g . 6.2 show developmental trends, e.g. the general increase i n latewood maximum density i n the f i r s t few rings from the pith, and also h i g h l i g h t some weather e f f e c t s , such as the lower maximum density i n 1971. Of p a r t i c u l a r i n t e r e s t i s the i r r e g u l a r i t y i n the 1972 p r o f i l e near the earlywood-latewood boundary. This i s well defined i n some clones (36, 82, 83) and l e s s c l e a r or absent i n others (28, 60). Examination of the X-ray negatives revealed the presence of a band of f a l s e latewood more or l e s s well developed i n a l l cores. Standardization of the density p r o f i l e s has resulted i n the appearance of the s t e p - l i k e anomaly, the size of which i s a good i n d i c a t i o n of the degree of v i s u a l d e f i n i t i o n of the latewood band. The factor most l i k e l y to lead to f a l s e latewood formation i s s o i l moisture deficiency (Shepherd, 1964; Cown, 1973). The r o l e of water i n infl u e n c i n g the earlywood-latewood t r a n s i t i o n and intra-increment density patterns has been discussed by Zahner (1963), Kramer (1964) and McKinnell and Shepherd (1971). Conditions conducive to f a l s e latewood formation are those r e s u l t i n g i n a period of high i n t e r n a l moisture stress during e a r l y -wood production, followed by stress r e l i e f , i.e. prolonged hot dry weather p r i o r to a s p e l l of r a i n f a l l . Warrack and Joergensen (1950) demonstrated that s o i l water d e f i c i e n c y could occur during the summer at the Cowichan Lake Experimental Station and i t would appear from the density data that 1972 - 163 -was a season of greater than normal s t r e s s . The following b r i e f examination of c l i m a t i c records shows t h i s to have been the case. Figure 6.12 shows weekly t o t a l r a i n f a l l and average maximum temperature for the growing seasons of years 1971-75. During t h i s period, only the 1972 cambial growth produced a band of f a l s e latewood. The growing season was characterized by very low May and June p r e c i p i t a t i o n combined with higher than average temperatures. July had about 80 mm of r a i n but a l l i n a period of s i x days, followed by 33 hot days devoid of any p r e c i p i t a t i o n . During the week of 14th - 22nd of August {i.e. at the time of earlywood-latewood t r a n s i -tion) 40 mm of r a i n f e l l , s u f f i c i e n t to at l e a s t p a r t l y r e p l e n i s h depleted s o i l moisture supplies. This, together with increased atmospheric humidity, would reduce i n t e r n a l plant moisture st r e s s , and at the c r u c i a l earlywood-latewood t r a n s i t i o n could favour production of c e l l s more l i k e earlywood than latewood, providing that moisture d e f i c i e n c y has a r o l e i n determining the timing of latewood formation. That the l a t t e r i s the case i s borne out by the f a c t that latewood width and latewood percentage i n 1972 were greater than i n other rings during 1969-73 (Table 6.7). The occurrence of the f a l s e r i n g again shows beyond doubt that d i f f e r e n t clones are producing wood of d i f f e r e n t density at the same time, since i n some groups i t i s present above, and i n the others below, the boundary l e v e l 3 of 0.50 g/cm . I t i s also of i n t e r e s t that the clones respond i n d i f f e r e n t ways to the period of s t r e s s , some developing a d i s t i n c t f a l s e latewood band and others showing no change i n the density p r o f i l e . Since t h i s i s u n l i k e l y to be an expression of microsite d i f f e r e n c e s , i t may be an i n t e r n a l i n d i c a t i o n of the a b i l i t y of the clone to withstand drought conditions; those clones showing a d i s t i n c t band being more susceptible to possible damage i n dry environments. -.164 -50 25 25 E E _i < u. z < ce 25 25 25 1971 —rt / \ - /Tamp. I . . . i 1972 / \ / \ / A \ ' \ \ / / \ \ \ I \ o J I Q o 1 973 Q Q o n o n Q I I _ — \ / \ \ >—\ 1_L a_J o l i o 1974 /~-v Q 0 I \ ' \ \ / V Q 0 1975 / \ / \ I f • I • 11 \ ' v I • • I, I I MAY JUNE JULY AUGUST SEPT. 30 20 10 20 10 — j a 3 X3 a -> a 20 = 10 20 10 20 1Q 0 — No ra in 1 - > 50 mm Figure 6.12 Weekly temperature and p r e c i p i t a t i o n - d a t a f o r the 1975 growing season at Cowichan Lake Forest Experiment Station. - 165 -TABLE 6.77..: Mean Clonal Ring Components 1969- 1973 (10 clones) Year RW EW LW PLW RD ED LD MND MXD UNI RNG 1969 5.92 4.58 1.34 22.4 .374 .293 .661 .215 .823 .368 .608 1970 6.32 4.97 1.35 21.5 .362 .271 .700 .208 .853 .429 .645 1971 7.51 6.08 1.43 18.7 .346 .268 .687 .213 .834 .419 .621 1972 7.09 5.43 1.66 23.1 .369 .275 .680 .206 .853 .405 .647 1973 7.30 5.99 1.31 17.7 .341 .269 .675 .201 .831 .406 .630 Abbreviations as given i n Table 6.4. - 166 -Chapter 7. RESULTS AND DISCUSSION - DIALLEL CROSS STUDY 7.1 Parent tree ring-component c h a r a c t e r i s t i c s Appendices 6(a) and (b) g,4ve the loca t i o n s , ages and dimensions of the parent trees used i n t h i s study. Despite possible d i f f e r e n c e s i n growing conditions, analyses of the r e s u l t s require that a very important assumption be made,i.c3i that observed phenotypic di f f e r e n c e s i n the c h a r a c t e r i s t i c s measured are an expression of genotypic rather than environmental influences. This would appear to be quite reasonable even although three of the trees are of natural o r i g i n and open grown and the fourth (tree no. 11) i s of unknown provenance growing i n a p l a n t a t i o n . The l a t t e r i s i n f a c t a corner tree, exposed on three sides, and thus has a growth h a b i t , c l o s e l y r e l a t e d to the open grown trees. The four increment cores from each stem were analysed separately and treated as independent random samples to enable comparisons of tree means to be c a r r i e d out by analyses of variance, using cores- within-trees as the error term. For such analyses, summaries of the outer .25 growth rings were used as being representative of the outerwood of these trees. Results are shown i n F i g s . 7,?l(a)-(k) with i n d i v i d u a l core values given. Mean r i n g width for tree E was about twice that for tree 11, with A and B intermediate. I t was i n t e r e s t i n g , however, that tree A, with the l e a s t earlywood and l e s s than the average latewood width,had s i g n i f i c a n t l y greater latewood percentage than the others and also the highest mean density. Tree E, with s i g n i f i c a n t l y higher percentage-latewood values than tree 11 never-theless had the lowest mean density. The reason for t h i s l i e s i n the very low latewood density shown by tree E i n comparison with the others (Fig. 7.1 On the other hand, tree 11, which had the l e a s t latewood percentage,,had the highest latewood density. - 167 -4 E E W i < > 4-- T • -* r-t 5 -< 1 -A 11 B E . P L W 11 E B A •40 Results of Duncan's Multiple Range tests on outerwood ring component data for parent trees A,B,E and 11. (Horizontal lines enclose means not significantly different at the 5% probability level). . - 168 -• 25 .35 E o .25 .60 h E .50 h .30 Figure 7.1 (contd.) Results of Duncan's Mu l t i p l e Range t e s t s on the outerwood r i n g component data f o r parent trees A,B,E and 11. (Horizontal l i n e s enclose means not s i g n i f i c a n t l y d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l ) . - 169 -Since latewood and maximum d e n s i t i e s varied s i g n i f i c a n t l y between trees whereas earlywood and minimum values d i d not, the i n t r a - r i n g contrast d i f f e r -ences were s t a t i s t i c a l l y s i g n i f i c a n t . Tree E showed the most uniform wood, 3 3 with an extreme contrast (RNG) of 0.548 g/cm compared to 0.734 g/cm for tree 11. Figure 7.1 i l l u s t r a t e s several important features which are l i s t e d below: 1. S i g n i f i c a n t d i f f e r e n c e s occur i n several growth r i n g components between trees growing i n s i m i l a r environments. In t h i s case only ED and MND were found to be s i m i l a r between the four stems. 2. Ring components may vary independently of one another, e.g. earlywood widths and d e n s i t i e s rank d i f f e r e n t l y from latewood widths and de n s i t i e s . Tree A, which has a comparatively narrow latewood zone, ranks f i r s t f o r percentage latewood, and has the lowest earlywood density and the highest mean density. 3. Considerable v a r i a t i o n can occur between the values f o r the four cores. In some cases {e.g. LW, ED) the range found within an i n d i -v i d u a l tree exceeded that between the tree means. This could be due to eit h e r of two possible sources, i,<*e. natural v a r i a t i o n or sampling errors i n the densitometric technique. The observed differences here are large enoughtto suggest that c i r c u m f e r e n t i a l v a r i a t i o n i n the stem i s l a r g e l y responsible, which has already been demonstrated i n Douglas-fir (Cown, 1971a). C l e a r l y , i f i t i s desirable to obtain values f o r growth-ring components representative of i n d i v i d u a l stems, several cores would be required. The analyses of variance discussed above demonstrate that the density p r o f i l e s of the parent trees d i f f e r at several points. One of the parent-progeny comparisons to be attempted involves these p r o f i l e s but, since age - 170 -has a predictable i n f l u e n c e , i t would not be advisable to deal d i r e c t l y with outerwood for the parents and corewood f o r the progeny. The p r o f i l e summary programme (Chapter 3.5) was used on the parent tree data to gener-ate p r o f i l e s f o r rings no. 6-10 from the p i t h as well as f o r the outer 25 growth periods. These are shown i n F i g . 7.2. A s t r i k i n g feature of the summaries i s that,despite the obvious increase i n percentage latewood and maximum density from corewood to mature wood i n a l l groups, tr e e E maintains i t s r e l a t i v e uniformity of structure. Given that the assumption of s i m i l a r environments i s a v a l i d one, t h i s phenomenon i s further evidence f o r genetic c o n t r o l of wood c h a r a c t e r i s t i c s within trees being maintained through the various growth phases. 7.2 Progeny ring-component c h a r a c t e r i s t i c s The progeny core samples which were analyzed densitometrically y i e l d e d data for an average of ten growth rings from p i t h to bark. In accordance with procedure adopted f o r the previous studies, the r e s u l t s were treated as two growth zones. Since i n many cases the 1975 increment showed extreme sup-pression of growth, the zones selected were rings 1-5 and rings 1970-74. The number of r e p l i c a t i o n s per progeny group ranged from 4 to 6 (Appendix 6c). and tree summaries for the two wood zones were analyzed by the U.B.C. ANOVAR programme which takes account of unequal c e l l numbers. Results of Duncan's Mu l t i p l e Range t e s t s are given i n Table 7.1(a) and (b). The analysis f or rings 1-5 showed that, of the eleven components tested, eight gave s i g n i f i c a n t progeny group differences of some sor t . Among the width parameters, only latewood width showed no s i g n i f i c a n t d i f f e r e n c e s , whereas of the density properties earlywood and minimum d e n s i t i e s were the l e a s t v a r i a b l e . The density components tended to show d i s t i n c t groups despite the f a c t that d i v i s i o n s on the basis of r i n g width and earlywood width were very broad. No obvious pattern existed i n the composition of the groups and - 171 -- 1 • . TREE A e O tr. 4J -W CO c CD Q Rings 6 - 1 0 Outerwood (25 rings) Figure 7.2 Summary p r o f i l e s f o r corewood and outerwood of parent trees. TABLE 7.1: Results of Duncan's Multiple Range Tests on Progeny Ring Components, a) Rings 1 - 5 RW nm EW mm LW mm PLW % RD . 3 g/cm ED . 3 g/cm B x 11 3.35 B X 11 2.11 B X E 1.06 E x A 26 .5 E X A .406 11 X B .329 E X B 3.57 E X B 2.35 11 X B 1.14 B x E 31 .1 B X E .423 A x E .340 11 X E 3.67 11 X E 2.36 E X A 1.19 A x E 34 .7 A X E .437 E X B .341 E X E 3.77 11 X A 2.39 E X B 1.21 11 x B 36 .0 E X 11 .444 B x A .341 11 X B 3.85 B X A 2.49 E X E 1.22 E x 11 38 1 11 X B .446 E X 11 .342 11 X A 3.89 E X E 2.55 A X E 1.22 E X E 38 1 E X E .449 E X E .345 E X 11 3.94 E X 11 2.69 E X 11 1.24 "- S- A x B 39 2 • A X B .451 E x A .347 n ' S -B X E 4.01 11 X B 2.72 B X 11 1.25 11 X E 41 4 B X 11 .458 E X E .347 B X A 4.02 A X B 2.91 11 X E 1.30 E X B 41 9 E X B .460 B X 11 .350 A x E 4.47 B X E 2.95 11 x A 1.49 B x A 42 0 11 x E .460 11 X E .351 A X B 4.56 A X E 3.25 B X A 1.53 B X 11 42 9 11 X A .467 11 X A .353 E X A 5.30 E X A 4.11 A x B 1.66 11 X A 43 4 B X A . .468 A X B .363 4.03 2.74 1.29 37 9 .447 .346 LD , 3 g/cm MND g/cm MXD , 3 g/cm UNI , 3 g/cm RNG . 3 g/cm E x A .558 11 X B .254 E X A . .632 A x B .207 E x A .367 A x B .571 E x 11 .261 B X E . .660 E x A .211 E X B .378 B x E .578 E x A .265 A X B .667 B x 11 .230 A X E .391 B x 11 .580 B x E .267 B X 11 .668 B x. E .233 B x E .392 E X B .582 B x . l l .269 E X B .676 E x B .241 A X B .398 A X E .586 A X B .269 A x E . .683 A X E .245 B X 11 .399 E X E .596 B X A .277 n - s - E X E .684 11 X A .249 E X E .405 11 X A .603 E X E .278 11 X A .709 E X E .249 11 X A .422 E X 11 .603 11 X A .206 E X 11 .713 11 X E .261 11 X E .434 11 X E .612 1 1 X E . 2 9 1 B X A .720 E X 11 .261 B X Ji .444 B X A .619 A x E . 2 9 2 11 X E .725 B x A .278 E x 11 .451 11 X B .632 E X B . 2 9 3 11 X B .740 11 X B .303 11 X B .486 . 5 9 3 . 2 7 6 . 6 9 0 .247 . .. ,.414 n.s. - differences not s i g n i f i c a n t at the 5% l e v e l V e r t i c a l l i n e s enclose means not s i g n i f i c a n t l y d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l . TABLE 7.1: (contd.) b) Rings 1970-74 RW mm EW mm LW mm PLW % RD g/cm ED g/cm3 B X 11 1.88 B x 11 0.82 E X 11 1 02 E x A 37 7 E X A • .421 B x A .283 E X 11 1.91 E X 11 0.89 11 x B 1 04 A x E 38 8 A x E .429 A x B .284 E X B 2.38 E X B 1.24 11 X E 1 04 11 x E 42 5 11 X E .438 B x E .289 11 X E 2.56 11 X A 1.41 B X 11 1 06 11 x B 43 9 B X E .454 E X 11 .290 11 X B 2.58 E X E 1.45- A X E 1 13 B X E 47 1 11 X B .457 11 x B .291 ' B X E 2.66 B X E 1.47 E X B 1 14 A X B 47 3 A X B , .460 E x A .293 11 X A 2.67 n * S - 11 X E 1.51 B X E 1 19 n ' S - . B X A 48 8 B X A .472 n - S > 11 x E .293 n - S -E X E 2.72 11 X B 1.53 11 X A 1 26 11 X A 50 2 11 X A .481 B x 11 .300. B X A 2.93 B X A 1.61 E X E 1 27 E X E 51 0 E X B .482 E X E .301 "• 'A X E 3.38 A x B 2.04 B X A 1 32 E X B 52 7 E X 11 .485 11 X A .304 E X A 3.68 A X E 2.24 E X A . 1 33 E X 11 55 5: E X E .487 E X B .305 A X B 3.72 E X A 2.35 A X B 1 68 B X 11 58 8 . B X 11 .502 A X E .314 2.76 1.55 1 21 47 9 .464 .296 LD g/cm3 MND . 3 g/cm MXD . 3 - g/cm UNI . 3 g/cm RNG g/cm A X E .603 A x B .217 A X E . .699 A X E .289 A X E .455 E X A .627 B X E .222 E X A .727 E x B .332 B x 11 .486 11 X E .634 E X A .223 B X 11 .728 E X A .334 E x B .500 B X E .634 11 X B .226 D X E .743 B X 11 .337 11 X E .503 E x B .637 E x 11 .231 11 x E .476 11 x E .341 E x A .504 B. X 11 E X 11 .637 .648 n - S -B X A B X 11 .238 .242 n ' S -E X B 11 X A .752 .758 n - S \ B X E 11 x A .345 .348 n ' S ' 11 X A B X E •5 0 5 n.s. .521 A X B .653 11 X E .242 E X 11 .758 E X 11 .358 E X 11 .527 11 X A .653 E X E .244 11 X B .766 E X E .360 11 X B .539 11 X B .659 A X E .244 A X B .769 11 X B .368 B x A .539 E X E .662 E X B .252 B X A .776 A X B .369 E X E .540 • B X A .666 11 X A .253 E X E .784 B X A .382 A X B .552 .641 .236 .750 .347 .514 . - differences not s i g n i f i c a n t at the 5% l e v e l V e r t i c a l l i n e s enclose means not s i g n i f i c a n t l y d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l . - 174 -i n many cases r e c i p r o c a l cross p a i r s f e l l i nto d i f f e r e n t categories. In contrast to the r e s u l t s f o r the inner rings, Table 7.1(b) (rings 1970-74) shows that only earlywood widths and percentage latewood showed s i g n i f i c a n t differences i n the outer rings. This i s i n strong contrast to the r e s u l t s of the c l o n a l study which demonstrated increasing differences between clones with age, and i n a l l p r o b a b i l i t y i s due to the onset of severe competition between the c l o s e l y spaced ( 1 m„ x 1 m) .stems up tOi.l7 -:years o l d . Thus i t would appear that the unfavourable environment has overcome the expression of genetic d i f f -erences i n most r i n g components. Where s i g n i f i c a n t progeny differences p e r s i s t , the ranking remains s i m i l a r to that for the inner r i n g group. T h e o r e t i c a l l y , the d i a l l e l r e s u l t s could be treated according to the s t a t i s t i c a l model out l i n e d by K r i e b e l et al. (1972) to y i e l d estimates of narrow sense h e r i t a b i l i t y f o r the r i n g components. However, the current author considers that the paucity of r e p l i c a t i o n , missing c e l l s and the r e l i a n c e on important genetic and s t a t i s t i c a l assumptions, together with the small number of parent trees involved, would render any such c a l c u l a t i o n s worthless from a p r a c t i c a l point of view. Data from the l i m i t e d genetic material and the unusual environment would not be applicable to the general r e f o r e s t a t i o n case;. V a r i a t i o n within progeny groups was marked 'and the severe competition between stems undoubtedly contributed towards t h i s , p a r t i c u l a r l y i n the case of width components. Some of the edge trees, for instance, were noticeably larger than those i n the i n t e r i o r of the stand. Appendix 23 gives the corre-l a t i o n matrices for the r i n g components. I t i s seen that several of the density components are highly r e l a t e d to earlywood width which i s the charac-t e r i s t i c : most aff e c t e d by competition .'.'-\g?he growth rate/wood density r e l a t i o n -2 ship i s p a r t i c u l a r l y strong i n the outer r i n g group (R f o r mean r i n g density and earlywood width = 0.57). - 175 -In contrast to the provenance and c l o n a l cores i n which earlywood width increased and percentage latewood decreased from the inner to the outer 5-ring groups, the progeny show a large decrease i n earlywood width and an increase i n percentage latewood from 37.9% to 47.9% respec-t i v e l y . With largernnumbers of r e p l i c a t e s covariance analyses could be used to remove some of t h i s v a r i a t i o n but the current data do not j u s t i f y t h i s approach. 7.3 Parent-progeny r e l a t i o n s h i p s In F i g . 7.1 the s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e s between the parent trees were shown, based on the assumption that they were expressions of genotypic t r a i t s . Comparisons with the progeny means given i n Table 7.1, however, in d i c a t e that these c h a r a c t e r i s t i c s have not been transmitted to the progeny i n a cl e a r - c u t manner. The progeny from tree E s e l f e d (group E x E) serve to i l l u s t r a t e the discrepancies between parent and o f f s p r i n g properties. The parent was characterized by low mean, latewood and maximum d e n s i t i e s and the most u n i -form i n t r a - r i n g p r o f i l e s (Figs. 7.1 and 7.2). The progeny, however, are ranked among the top 50% of groups f o r a l l density components and show the highest maximum de n s i t i e s i n the outer r i n g zone. In other words, the ranking which might have been anticipated on the basis of parent tree pro-p e r t i e s and the assumption of a simple mode of inheritance d i d not eventuate. Explanations of the discrepancies o u t l i n e d above are not hard to f i n d . F i r s t of a l l , there were c l e a r l y i n s u f f i c i e n t numbers of samples a v a i l a b l e with which to properly evaluate and rank the progeny groups. With only four trees present i n several categories, the means could not be accurately estimated. The c l o n a l analyses have previously demonstrated large differences between ramets grown under uniform conditions and i t might be expected that - 176 -sexually reproduced s i b l i n g s would show even larger with'in-group v a r i a t i o n . This same argument can be extended to apply equally to the parent trees, i n that the phenotype a c t u a l l y measured i s an expression of genotype x environment i n t e r a c t i o n , and how close t h i s comes to the mean value f o r the s i t e i s a matter f o r speculation. On the basis of the c l o n a l r e s u l t s (e.g. F i g . 6.5) i t must be concluded that the environmental influence can be substantial and that the genotype can only be confidently assessed on a number of progeny, preferably at l e a s t ten. This being the case, parent-progeny comparisons with few parents are extremely suspect. The foregoing r e s u l t s have shown that a) progeny ranks do not corres-pond to parent ranks, and, b) the r e c i p r o c a l cross p a i r s can often have s i g n i f i c a n t l y d i f f e r e n t p roperties. I t would s t i l l be of i n t e r e s t to d i s -cover whether the progeny tend to c o n s i s t e n t l y resemble one or the other of the parent trees and a method to achieve t h i s was developed using the i n t r a - r i n g density p r o f i l e summaries. I t had been demonstrated i n Chapter 6 (Figs. 6.3 and 6.11) that these p r o f i l e s , averaged over a group of trees, are strongly g e n e t i c a l l y c o n t r o l l e d . V a l i d comparisons of density c h a r a c t e r i s t i c s can only be performed on wood samples of s i m i l a r age. and,since the inner few rings were missing, from the parent tree cores,best r e s u l t s could be obtained by comparing p r o f i l e s of the outer 5-ring group of the progeny with the inner f i v e rings measured on the parents. I t i s unfortunate that progeny differences were not s t a t i s t i c a l l y s i g n i f i c a n t i n t h i s zone, but the p r o f i l e c h a r a c t e r i s t i c s of the younger wood (high minimum and low maximum density) precluded the use of such wood i n comparisons with the parent p r o f i l e s . Relative frequencies were derived for density groups (class i n t e r v a l -3 0.05 g/cm ) as for the densigrams, using the p r o f i l e summaries over the periods concerned. Differences (d .)in the c l a s s frequencies between parent - 177 -and progeny d i s t r i b u t i o n s were then weighted by a m u l t i p l y i n g factor (m.), Is the magnitude of which was defined as: x 1 - observations occur i n the c l a s s f or both parent and progeny, x 2 - the density values of one p r o f i l e are one c l a s s removed from the nearest values i n the other p r o f i l e , x 3 - the values of one p r o f i l e are two classes removed from the nearest values i n the other p r o f i l e . n The sums of the products ( ,H d-m., where n i s the number of density . _ 1s 1s ^=l classes represented i n the p r o f i l e s ) were then used as a s i m i l a r i t y index to compare parent-progeny shapes. I t w i l l be noted that most emphasis i s given to differences i n density extremes between p r o f i l e s since these seem more l i k e l y to be a r e s u l t of genetic factors than di f f e r e n c e s i n magnitude within the same c l a s s . Progeny p r o f i l e summaries are given i n F i g . 7.3 and the r e s u l t s of the p r o f i l e s i m i l a r i t y analyses are shown diagramatically i n F i g s . 7.4 and 7.5, High index values represent large discrepancies between the p r o f i l e s being compared. Indices for parent tree p r o f i l e comparisons substantiated the findings of the analysis of variance given i n F i g . 7.1, i n that trees A and B showed the highest degree of s i m i l a r i t y whereas trees E and 11 (with the minimum and maximum i n t r a - r i n g density contrasts, respectively) showed the greatest d i v -ergence. In the case of parent-progeny comparisons, there d i d not appear to be a strong tendency f o r the progeny to resemble one parent rather than the other. Where there was a large d i f f e r e n c e i n index between r e c i p r o c a l cross p a i r s (such as with AE and EA to parent A, and BE, EB to parent B) neither the female parent nor the p o l l e n parent was favoured c o n s i s t e n t l y . Of p a r t i c u l a r i n t e r e s t were the progeny from tree E r e s u l t i n g from s e l f -p o l l i n a t i o n . F i g . 7.5 (f) gives the s i m i l a r i t y indices f or group EE with - 178 -Figure 7.3 Progeny p r o f i l e summaries - rings 6-10. - 179 -Figure 7.3 (contd.) Progeny p r o f i l e summaries - rings 6-10 - 180 -- 181 -(a) PARENTS 148 167 104 44 60 107 / \/ \ PROGENY A B BA A B (b) 217 188 \155 6 2 89 99 A E E A A E PARENTS B . X T 172 224 64 108 124 100 / \ / \ PROGENY B E E B B E Cd) B^_x__11 150 229 84 66 81 148 / \ / A B11 11 B B11 Figure 7.5 Parent-progeny s i m i l a r i t y indices - r i ngs 6-10. - 182 -(e) PARENTS E x ?11 •< 283 330 160 123 119 161 / w V PROGENY E 11 11 E E 1 1 (f) B 107 112 \ / E E 160 161 11 Figure 7.5 (contd.) Parent progeny s i m i l a r i t y i n d i c e s - rings 6-10. - 183 -a l l four parent trees. I t can be seen that smaller indices r e s u l t from the comparisons with trees A and B than with E and 11, confirming what had been shown i n Table 7.1, Le. that the c h a r a c t e r i s t i c low-contrast p r o f i l e of the parent (Fig. 7.2) has not been passed on to the progeny. Although the r e s u l t s of the analyses i n t h i s chapter have been un s a t i s -factory f o r the reasons previously mentioned, two major conclusions can be drawn: a) with the evidence at hand, i t appears that the mode of inheritance of wood density i n Douglas-fir may not conform to a simple multiple factor concept. I t may be that dominance and e p i s t a t i c e f f e c t s are s i g n i f i c a n t . b) the method of comparing density p r o f i l e s could prove u s e f u l f o r quantifying genetic and environmental e f f e c t s provided appropriate numbers of sample r e p l i c a t i o n s are used. Given the inherent v a r i a t i o n already observed within clones and f u l l - s i b groups, i t i s suggested that a minimum of ten trees be used to represent a given genotypic or environmental influence. - 184 -Chapter 8. GENERAL SUMMARY, DISCUSSION AND CONCLUSIONS 8.1 Recapitulation Wood density i n young coastal Douglas-fir has been shown densitometri-c a l l y to be highly v a r i a b l e , both within and between'individual stems. On the basis of ring-component v a r i a t i o n , i t would appear that corewood extends outwards from the p i t h for about 15 r i n g s . Within t h i s zone, however, two d i s t i n c t regions can be distinguished. The innermost f i v e rings show rapid and predictable component v a r i a t i o n f o r most width and density parameters whereas, outside t h i s zone, v a r i a t i o n i s much les s marked and some components show s e n s i t i v i t y to f l u c t u a t i o n s i n environmental influences. Such patterns occur at d i f f e r e n t l e v e l s upwards i n the stem and there i s a high degree of constancy i n the r e l a t i v e ranking of trees by mean density at each l e v e l . Wood density estimated from two breast height increment cores gives a good i n d i c a t i o n of weighted tree mean density (r = 0.91). Samples from 40-year-old trees showed that juvenile density component c h a r a c t e r i s t i c s c o r r e l a t e rea-sonably well with mature wood and c r o s s - s e c t i o n a l properties. The r e s u l t s from these preliminary studies were used to determine optimum sampling schemes for the broader examination of environmental and genetic factors a f f e c t i n g r i n g component v a r i a t i o n . Analyses of provenance t r i a l data (5 s i t e s , 5 provenances) showed that a) r i n g width components are more s e n s i t i v e to environmental modification than are density components, b) earlywood width i s more s e n s i t i v e than l a t e -wood width (at l e a s t i n trees up to 20 years of age), and c) the r e l a t i v e influence of genetic and environmental e f f e c t s can change with age. Variations i n growth rate accounted for a s i g n i f i c a n t portion of the phenotypic density v a r i a t i o n (up to about 30% i n some cases), although t h i s e f f e c t was shown to become les s apparent with tree age. In contrast, the - 185 -r e l a t i o n s h i p between density and percentage latewood became stronger with age. Correlations between other r i n g component parameters were les s con-s i s t e n t and appeared to be at l e a s t p a r t l y site-dependent. For example, at the B r i t i s h Columbia locations i n t r a - r i n g density contrast (RNG) was highly p o s i t i v e l y correlated to mean r i n g density. Weather data from stations near the provenance t r i a l s i t e s were used i n regression analyses with r i n g component values for the outer rings and a new method was employed to present the r e l a t i o n s h i p s . Some highly s i g n i f i c a n t c o r r e l a t i o n s were revealed which r e l a t e d v a r i a t i o n i n both width and density components to monthly weather records. Wood samples from c l o n a l stock were examined to determine the extent of genetic control over r i n g component variables and to i n v e s t i g a t e possible c o r r e l a t i o n s of wood properties with phenological c h a r a c t e r i s t i c s . H e r i t a -b i l i t y estimates for i n d i v i d u a l growth rings showed a d i s t i n c t increase with tree age up to ten rings from the p i t h . Width components allow considerable scope f o r improvement through tree breeding as d i d mean r i n g density. Clonal d i f f e r e n c e s i n crown phenology were appreciable but d i d not show strong asso-c i a t i o n s with wood properties. I t d i d appear, however, that there was a ten-dency for l a t e - f l u s h i n g trees to produce lower-density latewood. Analyses of 13- to 17-year-old d i a l l e l cross material, obtained from c o n t r o l l e d p o l l i n a t i o n of four parent trees, revealed that density patterns c h a r a c t e r i s t i c of the parents were not transmitted to the progeny i n a simple manner. However, i t was also found that the p l a n t a t i o n environment was un-favourable to the expression of genetic d i f f e r e n c e s due to intense competi-t i o n between the c l o s e l y spaced stems. 8.2 Discussion and conclusions Hermann and Ching (1975) have annotated evidence of r a c i a l v a r i a t i o n i n - 186 -Douglas-fir, from such diverse study areas as karyotype analyses, chemo-taxonomy, plant morphology and provenance t r i a l analyses. Apart from the indisputable d i f f e r e n t i a t i o n between i n t e r i o r and coastal forms, considerable v a r i a t i o n has been shown within the coastal type, associated with geographic o r i g i n . Certain c h a r a c t e r i s t i c s appear to ex h i b i t c l i n a l trends from south to north, e.g. nuclear volume (El-Lakany and S z i k l a i , 1971) and early planta-t i o n height growth (Ching and Bever, 1960). As for wood properties, the most clea r c u t evidence of v a r i a t i o n comes from studies of heartwood permea-b i l i t y (Bublitz and Blackman, 1971) and of wood density l e v e l s as influenced by geographic l o c a t i o n (Drow, 1957; U.S. Forest Service, 1965). U n t i l recently, there has been a noticeable lack of emphasis on the gene-t i c aspects of wood q u a l i t y v a r i a t i o n i n Douglas-fir. Work on c e l l u l o s e content by Kennedy and Jaworsky (1960), and on tracheid length patterns by D u f f i e l d (1964), suggested that important i n d i v i d u a l tree d i f f e r e n c e s e x i s t and that such information should be put to use i n tree breeding programmes. Only i n 1966 d i d the f i r s t comprehensive study of environmental and genetic factors a f f e c t i n g wood density appear i n the l i t e r a t u r e (McKimmy, 1966). Interest i n Douglas-fir as an exotic species has lead to provenance studies i n many countries and recently two reports have emerged dealing with density (Wilcox, 1974; Thoby, 1975). In contrast to the current study, each of the above authors conclude that s i g n i f i c a n t differences occur i n coastal proven-ances. Wilcox (1974) gave data for 45 provenances at each of three s i t e s i n New Zealand at age 13 years. Although s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e s could be demonstrated f o r density, the range between extreme values was only 3 about 0.05 g/cm and the variance component f o r provenances was 5% of the t o t a l , compared to 35% f o r geographic l o c a t i o n , 8% for r e p l i c a t i o n s within s i t e s and 48% for tree-to-tree v a r i a t i o n . Thoby (1975) examined 25 coas t a l - 187 -provenances i n one p l a n t a t i o n at age 10 years and found a range of 0.035 3 . . . . . g/cm i n mean density. Information on i n d i v i d u a l tree v a r i a t i o n was not given i n t h i s case but the provenances were c l a s s i f i e d i n t o several groups on the basis of Duncan's Mul t i p l e Range t e s t s (5% l e v e l ) . The predominant feature of a l l density studies i n Douglas-fir tends to be the large tree-to-tree v a r i a t i o n present, which often overshadows other e f f e c t s . In view of t h i s c h a r a c t e r i s t i c , i t may be worthwhile to re-assess the p r a c t i c a l implications of s t a t i s t i c a l l y s i g n i f i c a n t d i f f e r e n c e s between provenances. In a l l tree improvement work the aim i s to e f f i c i e n t l y produce a healthy crop with d e s i r a b l e growth and wood c h a r a c t e r i s t i c s . Each addi-= t i o n a l t r a i t considered f o r s e l e c t i o n adds a great deal to the cost of a pro-gramme and so a c l e a r need for modification of the property must be established. I f , as has been suggested i n the above-mentioned reports, wood density i s to be included i n provenance evaluation, a large amount of assessment work would be involved f o r l i t t l e p o t e n t i a l gain i n strength properties and pulp y i e l d . Several studies have demonstrated important provenance differences i n height growth and volume production (Namkoong et al. , 1972; Rowe and Ching, 1973; Wilcox, 1974), so the l o g i c a l approach to improvement would be to f i r s t of a l l ensure that the seed source selected f o r a p a r t i c u l a r area i s healthy and vigorous. Modifications i n wood properties could be considered an a d d i t i o n a l bonus i f advantage can be taken of the large i n d i v i d u a l tree v a r i a t i o n . With the exception of the 1912 heredity study (Munger and Morris, 1936; Namkoong et at. , 1972), based on h a l f - s i b f a m i l i e s , most provenance work on Douglas-fir has been reported on young material. Several researchers have found evidence of genetic d i f f e r e n c e s i n morphology and phenology r e l a t e d to the climate of o r i g i n , i n p a r t i c u l a r to mean growing season temperature, length of the growing season, and p r e c i p i t a t i o n (Ching and Bever, 1960; Sweet/ - 188 -1965; Kung and Wright, 1972). This i s not unexpected i n a species covering such diverse conditions and with a l i m i t e d capacity f or seed d i s p e r s a l (Silen, 1962b). I t might be surmised that d i s r u p t i v e s e l e c t i o n would promote divergence between d i f f e r e n t polymorphic types i n t h i s s i t u a t i o n . I t has also been pointed out by Kung and Wright (1972), however, that i s o l a t e d popu-l a t i o n s of the i n t e r i o r type often f a i l to show genetic adaption to l o c a l conditions through natural s e l e c t i o n , even a f t e r thousands of years. The hypothesis was advanced that the renowned ' p l a s t i c i t y ' of t h i s species i s so great that genetic adaptions are often unnecessary. Work reported i n t h i s t h e s i s and elsewhere by the current author (Cown, 1971b) tends to support the above hypothesis for wood density, although with some reservation. The provenances studied were widely separated i n both l a t i -tude and a l t i t u d e such that any gene flow between the source locations i s most u n l i k e l y , and yet there was l i t t l e evidence of differences between them i n r a d i a l growth or wood density. Some p h y s i o l o g i c a l processes, phenological events and cambial a c t i v i t y , were monitored for t h i s same provenance t r i a l at three s i t e s and using four seed sources by Emmingham (1974). The M o l a l l a and V a l l e y plantations and M o l a l l a and Butte F a l l s sources were inccommon with the study reported here. S i t e e f f e c t s were highly s i g n i f i c a n t but i n no case were seed source differences found to be s i g n i f i c a n t . The t o t a l height growth ranks ou t l i n e d by Rowe and Ching (1974) were shown to be the r e s u l t of small cumulative annual e f f e c t s . Thus, there i s l i t t l e evidence to support the early contentions of s p e c i f i c adaptions to l o c a l environments for the provenances used i n the Co-operative Douglas-fir Provenance Test. E i t h e r the environmental factors have not been s u f f i c i e n t l y d i f f e r e n t to cause genetic divergence or gene flow i s such as to overcome the tendency towards natural s e l e c t i o n . - 189 -I t would therefore seem that the ' p l a s t i c i t y ' e f f e c t i s predominant, since a strong environmental influence can be seen i n the performance of the plantations and p o l l e n d i s p e r s a l has been shown to be r e l a t i v e l y res-t r i c t e d ( S i l e n , 1962b). That l o c a l environment has a s i g n i f i c a n t e f f e c t on wood density i s demonstrated by the differences between the two planta-t i o n blocks at M o l a l l a and by the s i g n i f i c a n t p l o t - t o - p l o t v a r i a t i o n recorded for some properties at a l l s i t e s . These l o c a l i z e d e f f e c t s are often small compared to the i n d i v i d u a l tree v a r i a t i o n encountered, but are large i n r e l a -t i o n to the o v e r a l l geographic v a r i a t i o n . Despite the f a c t that several workers have q u a n t i f i e d phenological differences i n coastal seed l o t s , both at the stage of nursery performance (Sweet, 1965; Irgens-Moller, 1968) and i n semi-mature f a m i l i e s (McKimmy, 1966; McKimmy and Nicholas, 1971), such consistency was not observed i n the small study c a r r i e d out by Walters and Ching (1966) i n some of the Co-operative Douglas-fir Provenance Test p l a n t a t i o n s . On the basis of the l i m i t e d informa-t i o n a v a i l a b l e i t appears that seed source di f f e r e n c e s i n phenology have become les s marked with tree age (K.K. Ching, pers. comm.). The r e s u l t s of Emmingham (1974) showed f l u s h i n g at the V a l l e y s i t e to precede that at Molalla by about 42 days i n 1973, so that considerable environmental forces can act on Douglas-fir without leading to well-defined genetic adaption. As pointed out by Stern and Roche (1974), wood properties may be subject to natural s e l e c t i o n e i t h e r d i r e c t l y , through t h e i r f u n c t i o n a l r e l a t i o n s h i p s (support, conduction), or i n d i r e c t l y through c o r r e l a t i o n s with other tree c h a r a c t e r i s t i c s such as phenology. Neither of these forces has been found to have a s i g n i f i c a n t e f f e c t i n the current study, p a r t i c u l a r l y since the M o l a l l a provenance, from an o r i g i n where f l u s h i n g i s l a t e and the l o c a l conditions induce the formation of high density wood, i s found to have properties simi-l a r to the other provenances at a l l locations sampled. I t seems pos s i b l e - 190 -that the f l u c t u a t i o n i n environmental f a c t o r s (especially p r e c i p i t a t i o n ) from year to year at any one s i t e i s a major component of the o v e r a l l v a r i a -t i o n , such that long term s i t e mean diff e r e n c e s have r e l a t i v e l y l i t t l e e f f e c t on natural s e l e c t i o n . Even the phenological differences noted by the above-mentioned authors are not necessarily r e l a t e d to differences i n growth r i n g components, as shown by the intensive study of selected high and low density clones. Although some phenological events such as bud bursting are strongly geneti-c a l l y c o n t r o l l e d , the r e l a t i o n s h i p s with wood properties i n young trees are nebulous at best. In f a c t , i t seems that the concepts of wood formation as espoused by Larson (1969) do not apply d i r e c t l y to Douglas-fir. Bud burst, f o r instance, was found to occur about 40 days a f t e r the i n i t i a t i o n of cambial a c t i v i t y , when roughly 30% of the growth r i n g had been formed. Studies by many workers, notably B.J. Zobel and associates, leave l i t t l e doubt that genotype strongly determines wood density i n c o n i f e r s , but the findings of t h i s and other reports with respect to environmental e f f e c t s i n d i c a t e that growth rate may exert a predictable influence. In the c l o n a l and provenance material examined, up to 80% of the observed v a r i a t i o n i n mean density was r e l a t e d to v a r i a t i o n i n growth rate. Although there was a p o s s i b i l i t y that within clones the rootstock-scion i n t e r a c t i o n may have accounted f o r some of t h i s , the analyses pointed towards other unspecified environmental e f f e c t s . The c l o n a l r e s u l t s demonstrate c l e a r l y that large d i f f e r e n c e s i n geno-type can occur i n a l l r i n g components measured. In contrast to the environ-mental e f f e c t s discussed above, genetic c o n t r o l i s most strongly expressed i n the latewood portion of the growth r i n g . On a clone mean basis the h e r i t a b i l i t y estimates f o r earlywood, latewood and mean r i n g d e n s i t i e s ranged - 191 -between 0.77 and 0.93 when the data were analysed according to two 5-ring groups. This, together with the between-.tree v a r i a t i o n q u a n t i f i e d i n the provenance study, suggests that considerable manipulation of wood density values would be possible i n an improvement programme. The secondary density characters studied (UNI and RNG) a l s o showed high h e r i t a b i l i t i e s but, since they depend on both earlywood and latewood proper-t i e s , there i s a danger that they would be more susceptible to genotype-environment i n t e r a c t i o n . For example, the c l i m a t o l o g i c a l analyses showed earlywood and latewood and minimum and maximum de n s i t i e s to be influenced by d i f f e r e n t weather va r i a b l e s , so that the UNI and RNG values r e f l e c t a s p e c i f i c combination of genotypes and weather patterns. For phenotypes to maintain the same r e l a t i v e ranks between s i t e s , both i n d i v i d u a l components must react s i m i l a r l y f or a l l genotypes. Evidence presented i n Chapter 5 sug-gests that, at l e a s t for RNG, the genotype-environment i n t e r a c t i o n may be s i g n i f i c a n t . The d i a l l e l cross study h i g h l i g h t s some of the d e f i c i e n c i e s i n regard to e x i s t i n g knowledge of inheritance of wood properties i n Douglas-fir. Narrow sense h e r i t a b i l i t i e s were not c a l c u l a t e d due to the lack of adequate r e p l i c a -t i o n and the condition of the stand. The influence of tree competition was evident i n the strong growth rate/wood density r e l a t i o n s h i p but there were i n s u f f i c i e n t samples for covariance a n a l y s i s . Parent-progeny density pro-f i l e comparisons yi e l d e d inconclusive r e s u l t s and, i n fact, i t was c l e a r that most of the s i g n i f i c a n t d i f f e r e n c e s i n parental phenotypes were not apparent i n the progenies. There were s i g n i f i c a n t differences i n progeny r i n g compo-nents i n the early growth increments but the data shed no l i g h t on the mode of transmission from parent to o f f s p r i n g . Insofar as the breeding value of a selected plus tree i s determined by the c h a r a c t e r i s t i c s of i t s progeny, i t - 192 -would appear that accurate assessment of parent tree properties may be i n e f f i c i e n t from a genetic point of view. The parents were examined i n t h i s study by means of four increment cores from the breast height p o s i t i o n of each tree. Consistent density p r o f i l e patterns were observed within trees and these expressed themselves i n the s i g n i f i c a n t between-tree pheno-t y p i c d i f f e r e n c e s . However, the low density tree when s e l f p o l l i n a t e d produced progeny of average to high density. There i s thus a strong case for progeny t e s t i n g of plus trees before assumptions are made as to the geno-type. The trend towards i n t e n s i v e l y managed sho r t - r o t a t i o n crops i s l i k e l y to continue with increased momentum and the benefits of considering wood density have been stressed by several authors. Very l i t t l e work has been reported i n t h i s f i e l d f o r Douglas-fir but a glance at the c l o n a l r e s u l t s shown i n F i g . 6.1 should convince foresters that considerable improvement i n wood production could be attained through s e l e c t i o n of fast-grown high-density stock. The juvenile-mature c o r r e l a t i o n s discussed e a r l i e r i n d i c a t e that such e a r l y evaluations have p r a c t i c a l merit at l e a s t as f a r as density i s concerned. Further work of a s i m i l a r nature should attempt to determine the combined growth rate and density r e l a t i o n s h i p s i n terms of f i b r e y i e l d . X-ray densitometry i s a very v e r s a t i l e t o o l f o r analysing i n t r a - r i n g density v a r i a t i o n s within and between trees and f o r quantifying environmental and genetic factors as they may e f f e c t r i n g components. However, due to the independent development of systems at several l a b o r a t o r i e s , there i s a need fo r standardization of methods and presentation of r e s u l t s . In p a r t i c u l a r , the concept of earlywood-latewood s t i l l poses problems which researchers see f i t to deal with i n d i f f e r e n t ways (Rudman, 1968; H a r r i s , 1969; Parker and Kennedy, 1973). The densitometer y i e l d s much more information than can be - 193 -obtained by gravimetric methods, but latewood i s often expressed i n the form of a percentage or index which does not take f u l l advantage of the data a v a i l a b l e . The present author i s of the opinion, moreover, that attempts to characterize i n t r a - r i n g density patterns i n terms of mathematical func-t i o n s , (e.g. Kawaguchi, 1969) serve to confuse rather than s i m p l i f y the problem. The standard densigram approach adopted i n t h i s study i s proposed as a superior means of group comparison, i n that i t allows v i s u a l assessments to be made over the range of i n t r a - r i n g density values present, independent of any method of de f i n i n g latewood. I t was found to be e s p e c i a l l y useful i n describing genotypic di f f e r e n c e s between clones, and formed the basis f or c a l c u l a t i n g s i m i l a r i t y indices for parent-progeny r e l a t i o n s h i p s . Use of an earlywood-latewood boundary c r i t e r i o n enables the separation of the growth r i n g into component parts which can be treated separately i n many types of analyses and which help i n i d e n t i f y i n g sources of v a r i a t i o n . 3 The f i x e d density l e v e l chosen f o r these studies (0.50 g/cm basic density) gave values for percentage latewood which were highly correlated with mean ri n g density and was thus of considerable merit f o r comparisons of technologi-c a l p roperties. 8.3 Recommendations f o r further research The studies reported i n t h i s t h e s i s deal, of necessity, with r e l a t i v e l y small numbers of young trees and provide an i n s i g h t i n t o what might be expected i n terms of wood q u a l i t y at more advanced ages. The most important recommendations therefore concern r e p l i c a t i o n of the t r i a l s both i n space and i n time. Tree-to-tree and regional v a r i a t i o n s are f a i r l y well documented for Douglas-fir, although the e f f e c t s of s p e c i f i c environmental factors on q u a l i t y - 194 -are as yet poorly understood. The Co-operative Douglas-Fir Provenance Test provides a good opportunity to extend the work on provenance research and f o r studying s i t e e f f e c t s . Every e f f o r t should be made to encourage the agencies involved to maintain upkeep of the plantations and, i f p o s s i b l e , to i n s t a l l and service weather recording instruments. Plans should be made to carry out p e r i o d i c assessments of wood q u a l i t y , perhaps in v o l v i n g some form of multi-stage sampling to cover thewwhole range of provenances. Improved seed f o r P a c i f i c Northwest forests w i l l i n future come mainly from plus trees selected i n t h e i r natural environment and propagated vege-t a t i v e l y i n seed orchards around the region. Since wood q u a l i t y w i l l undoub-tedl y become a f a c t o r of greater importance i n years to come, e f f o r t s should be made to obtain more data on both narrow-sense and broad-sense h e r i t a b i l i t i e s . This could be done p a r t l y by analyses of established progeny t r i a l s and p a r t l y by planning c a r e f u l l y designed t r i a l s s p e c i f i c a l l y f o r wood q u a l i t y assessment. Again, adequate r e p l i c a t i o n i n space and time i s the key to sound r e s u l t s . Hopefully, developments i n the f i e l d of r a d i a t i o n densitometry w i l l bring more standardization i n methods of data a c q u i s i t i o n and analyses, such that more co-operative work would be possible between lab o r a t o r i e s around the world. - 195 -REFERENCES Alexander, J.B. (1935) The effect of rate of growth upon the specific gravity of Douglas f i r . Can. Dep. Interior For. Serv. Circular No. 44. pp. 8. .. (1950) The physical and mechanical properties of second-growth Douglas f i r . Am. Soc. Test Mat. Bull. No. 169:33-38. Anon., (1955) A cooperative seed provenance study for Pacific Northwest Douglas f i r . U.S.D.A., For. Serv. Mimeo. pp. 8. Brazier, J.D. (1969) Some considerations in appraising within-ring density. F.A.O., Second World Consultation on Forest Tree Breeding. Washington, 1969. FO-FTB 69 - 4/2:381-391. Bublitz, W.J., and Blackman, G. (1971) Kraft pulping qualities of refractory and permeable Douglas-fir heartwood. Tappi 5_4(6) :981-984. Buijtenen, J.P., van (1962) Heritability estimates of wood density in loblolly pines. Tappi 45(7):602-605. (1964) Anatomical factors influencing wood specific gravity of slash pine and the implications for the development of a high-quality pulpwood. Tappi 47_(7) : 401-404. Burdon, R.D., and Harris, J.M. (1973) Wood density in radiata pine clones on four different sites. N.Z. J. For.Sci. 3 (3) :286-303. Cameron, J.F., Berry, P.F., and Ph i l l i p s , E.W.J. (1959) The determination of wood density using beta rays. Holzforschung 13_(3) :78-84. Campbell, R.K. (1964) Recommended traits to be improved in a breeding program for Douglas f i r . Weyerhaeuser Company For. Res. Note No. 57. pp.19. (1974) Use of phenology for examining provenance transfers in reforestation of Douglas-fir. J. Appl. Ecol. 11(3):1069-1080. Chalk, L. (1930) The formation of spring and summer wood in ash and Douglas f i r . Oxford Forestry Memoir No. 10. pp. 48. (1953) Variation of density in stems of Douglas f i r . Forestry 26:33-36. Ching, K.K. (1965) Early growth of Douglas-fir in a reciprocal planting. Oreg. State Univ. For. Res. Lab. Res. Pap. No. 3. pp. 20. • , and Bever, D. (1960) Provenance study in Douglas-fir in the Pacific Northwest region. Silvae Genet. 9_(1):11-17. , and Hermann, R.K. (1973) Biosystematics of Douglas-fir. I.U.F.R.O. Proc. Working Party on Douglas-fir Provenances. Gottingen W. Germany Sept. 1973:23-36. - 196 - . Ching, K.K., Rowe, K.E., and D i l l , S.B. (1974) Provenance study of Douglas-f i r i n the P a c i f i c Northwest region. F i e l d performance at age fourteen. Oreg. State Univ. Sch. For. For. Res. Lab. Project F 628 (unpublished). Cown, D.J. (1971a) Circumferential v a r i a t i o n i n wood density of Douglas-fir. N.Z. For. Serv. For. Res. Inst. Rept. No. FP/WQ/5. pp. 10. (unpublished). (1971b) Wood density survey i n selected Douglas-fir compartments i n Kaingaroa Forest. N.Z. For. Serv. For. Res. Inst. Rept. No. FP/WQ/4. pp. 8. (unpublished). (1971c) Density-related latewood measurements from P. radiata densitometer s t r i p s and charts. N.Z. For. Serv. For. Res. Inst. ; Rept. No. FP/WQ/7. pp. 6. (unpublished). (1973) Resin pockets: Their occurrence and formation i n New Zealand: forests.. N.Z. J . For. 18J2) :233^251. (1974a) The wood density of r a d i a t a pine: I t s v a r i a t i o n and manipulation. N.Z.J. For. 19(1):84-92. (1974b) Physical properties of Corsican pine grown i n New Zealand. N.Z. J . For. S c i . 4_(1) :76-93. C u r t i s , R.O., Reukema, D.L., S i l e n , R.R., Fight, R. , and Romancier, R.M. (1974) Intensive management of Douglas-fir. For. Industries 101(-1):48-50. Dadswell, H.E., F i e l d i n g , J.M. N i c h o l l s , J.W.P., and Brown, A.G. (1961) Tree-to-tree v a r i a t i o n s and gross h e r i t a b i l i t y of wood c h a r a c t e r i s t i c s of Pinus radiata. Tappi 44 (3):174-179. Dimock, E.J. (1964) Simultaneous v a r i a t i o n s i n seasonal height and r a d i a l growth of young Douglas-fir. J . For. 62 (4):252-255. Dixon, J.R. (1970) Studies i n v a r i a t i o n i n the green Douglas-fir. Scott. For. 24_(1) :22-30. Drow, J.T. (1957) Relationship of l o c a l i t y and rate of growth to density and strength of Douglas f i r . U.S.D.A. For. Serv. For. Prod. Lab. Rep. No. 2078. pp. 56. Duff, G.H., and Nolan, N.J. (1953) Growth and morphogenesis i n the Canadian fo r e s t species. I. The controls of cambial and a p i c a l a c t i v i t y i n Pinus resinosa A i t . Can. J . Bot. 31:471-513. D u f f i e l d , J.W. (1964) Tracheid length v a r i a t i o n patterns i n Douglas-fir and s e l e c t i o n of extreme v a r i a n t s . Tappi 47 (2) :122-124. Echols, R.M. (1970) Moving-slit radiography of wood s'amples for incremental measurements. Iri Tree-ring analyses with s p e c i a l reference to Northwest America. Eds. J.H.G. Smith and J.GG Worrall. Univ. of B.C. Fac. For. B u l l . No. 7:34-36. • (1971) Patterns of wood density d i s t r i b u t i o n and growth rate i n Ponderosa pine. Proc. Symposium on the E f f e c t of Growth Acceleration on the Properties of Wood. Madison, Wis. 1971. pp. 16 (1972) Product s u i t a b i l i t y of wood determined by density gradients across growth rings. U.S.D.A. For. Serv. Res. Note PSW-273. pp. 6 - 197 -Echols, R.M. (1973) Uniformity of ••wood density assessed from X-rays of increment cores. Wood Sci. and Tech. 7_:34-44. , and Conkle, M.T. (1971) The influence of plantation and seed source elevation on wood specific gravity of 29-year-old ponderosa pine. For. Sci. 17 (3).388-394. Edlin, H.L. (1965) The summerwood:springwood ratio in conifers. Forestry 38_(1) :91-112. Einspahr, D.W. (1972) Forest tree improvement and the pulp and paper industry. Tappi 55(12):1660-1669. , Goddard, R.E., and Gardner, H.S. (1964) Slash pine wood and fiber h e r i t a b i l i t y study. Silvae Genet. 13 (4).103-109. El-Lakany, M.H., and Sziklai, 0. (1971) Intraspecific variation in nuclear characteristics of Douglas-fir. Advancing Frontiers of Plant Sciences 28:363-378. E l l i o t t , G.K. (1970) Wood density in conifers. Commonw. For. Bur. Tech. Commun. No.-8. pp. 44. , and Brook, S.E.G. (1967) A microphotometric technique for growth" ring analysis. Inst. Wood Sci. 18:24-43. Emmingham, W. (1974) Physiological responses of four Douglas-fir populations in three contrasting f i e l d environments. Ph.D. Thesis. Oreg State Univ. pp. 162. Ezekiel, M., and Fox, K.A. (1959) Methods of correlation and regression analysis. John Wiley & Sons Inc., New York. 3rd edition, pp. 548. Falconer, D.S. (1964) Introduction to quantitative genetics. Oliver and Boyd, Edinburgh, pp/ 365. Fielding, J.M. (1953) Variations in Monterey pine. Bull. For. Timb. Bur. Aust. No. 31. pp. 43. and Brown, A.G. (1960) Variations in the density of the wood of Monterey pine from tree to tree. Bull. For. Timb. Bur. Aust. No. 77. pp. 28. Forward, D.F., and Nolan, N.J. (1962) Growth and morphogenesis in the Canadian forest species. IV. The significance of specific increment of cambial area in P -UUL6 H<L&A,n0i>0L Ait. Can. J. Bot. 40(1) : 9 5 - l l l . Fowler, D.P. (1964) Effects of inbreeding in red pine. Silvae Genet. 13; 170-177. Franklin, E.C., and Squillace, A.E. (1973) Short-term progeny tests and second generation breeding in slash pine. Can. J. For. Res. 3_: 165-169. Fraser, D.A. (1952) Initiation of cambial activity in some forest trees in Ontario. Ecol. 33:259-273. - 198 -Freese, F. (1962) Elementary forest sampling. U.S.D.A. Handbook No.232. pp. 91. F r i t t s , H.C., Biasing, T.J., Hayden, B.P., and Kutzbach, J.E. (1971) Multi-variate techniques for specifying tree-growth and climate relationships and for reconstructing anolalies in paleoclimate. J. Appl. Meteor. 10: 845-864. Fry, G., and Chalk, L. '(1957) Variation of density in wood of P-inu.6 patiita grown in Kenya. Forestry 30(1):29-45. Goggans, J.F. (1964) Correlation and herita b i l i t y of certain wood properties in loblolly pine (PlmxA tdQAa. L.). Tappi 47_(6) : 318-322. Goddard, R.E., and Cole, D.E. (1966) Variation in wood production in six-year-old progenies of selected slash pines. Tappi 49(8):359-362. GOhre, K. (1958) Die Douglasie und ihr Holz. Akademie-verlag, Berlin. pp.596v Green, H.V., and Worrall, J.G. (1964) Wood quality studies. I. A scanning micro-photometer' for automatically measuring and recording certain wood characteristics. Tappi 47_(7) :419-427. G r i f f i t h , B.G. (1968) Phenology, growth and flower and cone production of 194 Douglas f i r trees on the University Research Forest as influenced by climate and f e r t i l i z e r , 1957-1967. Univ. of B.C., Fac. For. Bull. No. 6. pp. 70. Haddock, P.G. (1967) The importance of provenance in forestry. Proc. International Plant Propagators Society. Annual Meeting, 1967:91-98. Haigh, R.W. (1961) The effect of provenance and growth rate on specific gravity and summerwood percentage of young Douglas-fir. B.S.F. Thesis. Univ. of B.C. Fac. For. pp.' 57. Hall, G.S. (1962) Relationship between Douglas-fir latewood and some environ-mental factors. M.F. Thesis. Univ. of B.C. Fac. For. pp. 41. Harris, J.M. (1965) A survey of the wood density, tracheid length and latewood characteristics of radiata pine grown in New Zealand. N.Z. For. Serv. For. Res. Inst. Tech. Pap. No. 47. pp. 31. (1966) A survey of the wood density of Douglas-fir grown in New Zealand. N.Z. For. Serv. For. Res. Inst., For. Prod. Rep. No. 194. pp.22, (unpublished). (1967) Latewood, earlywood and wood density. Proc. I.U.F.R.O. Congress, Munich, 1967 : 56-69. (1969) The use of beta rays in determining wood properties. Parts 1-5. N.Z. J. Sci. 12_(2) : 395-451. (1970) Breeding to improve wood quality;. Unasylva 24 (2/3) :32-36. - 199 -Harris, J.M. (1973) Physical properties, r e s i n content and tracheid length of lodgepole pine grown i n New Zealand. N.Z. J . For. S c i . 3^(1)):91-109. , and B i r t , D.V. (1972) Use of beta rays f o r ea r l y assessment of wood density development i n provenance t r i a l s . S i l v a e Genet. 21^(1/2): 21-25. , and Kripas, S. (1959) The p h y s i c a l properties of two provenances of ponderosa pine grown i n Kaingaroa State Forest. N.Z. For. Serv. For. Res. Note No. 16. pp. 16. , and Orman, H.R. (1958) The p h y s i c a l and mechanical properties of New Zealand-grown Douglas-fir. N.Z. For. Serv., For. Res. Inst. Tech. Pap. No. 24. pp. 88. , and Polge, H. (1967) A comparison of X-ray and beta ray techniques for measuring wood density. J . Inst. Wood S c i . 19_ (4.1):34-42. , B i r t , D.V., and Cown, D.J. (1971) Wood properties of Douglas-f i r . N.Z. For. Serv. Rep. For. Res. Inst., 1971 : 79-80. Heamon, J.C. (1963) A provenance study of coastal Douglas-fir i n the P a c i f i c Northwest. B.C. Registered Forester's Thesis No. 411. pp. 85. (unpublished). (1966) Grafting i n Douglas-fir. B.C. For. Serv. Res. Rev., 1966. p. 14. Heger, L., Parker, M.L., and Kennedy, R.W. (1974) X-ray densitometry: A technique and an example of a p p l i c a t i o n . Wood S c i . 7_(2):140-148. Hermann, R.K., and Ching, K.K. (1975) Bibliography of Douglas-fir provenance studies, 1907-1974. Oreg. State Univ. Sch. For. Res. Lab., Res. Pap. No. 25. pp. 30. Hoist, M.J. (1958) Thoughts on wood density. Proc. 6th Meeting of the Committee on Forest Tree Breeding i n Canada. Part I I . pp. S-31 -32. Hughes, W.C, and A l l e n , J.W. (1949) Some ^strength properties of second-growth Douglas-fir grown i n Western Washington. M.F. Thesis. Univ. of Wash, pp. 60. Hursey, R. (1970) Wood q u a l i t y of Douglas-fir i n northern Idaho correlated to vegetational associations. Ph.D. Thesis. Univ. of Idaho Grad. Sch. pp. 28. I f j u , G. (1969) Within growth r i n g v a r i a t i o n i n some p h y s i c a l properties of southern pine wood. Wood S c i . 2_: 11-19. Irgens-Moller, H. (1957) Ecotypic response to temperature and photoperiod i n Douglas-fir. For. S c i . _3_(l):79-83. _ (1958) Genotypic v a r i a t i o n i n the time of cessation of height growth i n Douglas-fir. For. S c i . 4_(4) :325-330. (1968) Geographic v a r i a t i o n i n growth patterns of Douglas-f i r . S i l v a e Genet. 17(2/3):106-110. - 200 -Jackson, L.W.R., and Warren, B.J. (1962) V a r i a t i o n and inheritance i n s p e c i f i c g r a v i t y of slash and l o b l o l l y pine progeny. Georgia For. Res. Council Res. Pap. No. 14. Kawaguchi, M. (1969) (A Fourier analysis of growth r i n g p h o t o e l e c t r i c analysis curves). J . Jap. Wood Res. Soc. 15_(1):6-10. (For. Abs. 3_0(4) - 6383). K e l l e r , R. (1973) Caracteristiques du bois de p i n maritime; v a r i a b i l i t e " et transmission he'reditaire. Ann. S c i . For. 30 (1) :31-62. Kennedy, R.W. (1961) V a r i a t i o n and p e r i o d i c i t y of. summerwood i n some second-growth Douglas-fir. Tappi £4(3):161-166. (1966) Intra-increment v a r i a t i o n and h e r i t a b i l i t y of specifics g r a v i t y , p a r a l l e l - t o - g r a i n t e n s i l e strength, s t i f f n e s s , and tracheid length i n c l o n a l Norway spruce. Tappi 49(7):292-296. (1970) S p e c i f i c g r a v i t y of ear l y - and l a t e - f l u s h i n g Douglas-fir trees. Tappi 53(8):1479-1481. , and Jaworsky, J.M. (1960) V a r i a t i o n i n c e l l u l o s e content of Douglas-fir. Tappi 43 (1);25-27. Kleinschmit, J . van, Racz, J . , Weisgerber, H., Dietze, W., D i e t e r i c h , H., and Dimpflmeier, R. (1974) Ergebnisse aus dem internationalen Douglasien-Herkunftsversuch von 1970 i n der Bundesrepublik Deutschland. Silvae Genet. 23 (6):167-176. Knigge, W. (1962) Untersuchungen liber die Abhangigkeit der mi t t l e r e n Rohdichte nordamerikanischer Douglasienstamme von unterschiedlichen Wuchsbedingungen. Holz a l s Roh und Werkstoff 20:352-360. Knudsen, M.V. (1956) A comparitive study of some technological properties of Norway spruce i n a provenance t e s t . XII I.U.F.R.O. Congress, Oxford, 1956. No. 56/41/2. pp. 7. Koehler, A. (1939) Heredity versus environment i n improving wood i n f o r e s t trees. J . For. 37:683-687. Koch, P. (1972) U t i l i z a t i o n of the southern pines. V o l . I - The raw material. U.S.D.A. Agri c . Handb. No. 420. pp. 734. Kozlowski, T.T. (1971) Growth and development of trees. Academic Press, New York. Vols. I and I I . pp. 957. Kramer, P.J. (1964) The r o l e of water i n wood formation. In Formation of wood i n f o r e s t trees. Ed. M.M. Zimmermann. pp. 519-532. Kr i e b e l , H.B., Namkoong, G., and Usanis, R.A. (1972) Analysis of genetic v a r i a t i o n i n 1-, 2-, and 3-year-old eastern white pine i n incomplete d i a l l e l cross experiments. S i l v a e Genet. 21_(l/2) :44-48. Kung, F.H., and Wright, J.W. (1972) P a r a l l e l and divergent evolution i n Rocky Mountain trees. Silvae Genet. 21(3/4):77-85. - 201 -Kusec, D.J. (1972) Twin-blade saw for precision machining of increment cores. Wood and Fiber <4(1):44-49. Lacaze, J.F., and Polge, H. (1970) Relations phenotypiques au stade juvenile entre l a densite du bois et divers caracteres phenologiques et de vigeur chez Pioea abies Karst. Ann. Sci. For. 27(3):231-242. Larson, P.R. (1957) Effect of environment on the percentage of summerwood and specific gravity of slash pine. Yale Univ. Sch. For. Bull. No. 63. pp. 78. (1962) A biological approach to wood quality. Tappi £5 (6): 443-448. (1969) Wood formation and the concept of wood quality. Yale Univ. Sch. For. Bull. No. 74. pp. 54. Lassen, L.E., and Okkonen, E.A. (1969) Effect of r a i n f a l l and elevation on specific gravity of coast Douglas-fir. Wood and Fiber X_(3) :227-235. Lavender, D.P., and Hermann, R.K. (1970) Regulation of the growth potential of Douglas-fir seedlings during dormancy. New. Phytol. 69(3).675-694. Littleford, T.W. (1961) Variation of strength properties within trees and between trees in a stand of rapid-grown Douglas-fir. Can. Dept. For. For. Prod. Lab. Mimeo. V-1028. pp. 20. Marian, J.E., and Stumbo, D.A. (1960) A new method of growth ring analysis and the determination of density by surface texture measurements. Holz als Roh und Werkstoff 18:287-296. Markwardt, L.J. (1927) Density of wood as a guide to i t s properties. South. Lumberman 129:51. M a t z i r i s , M.I., and Zobel, B.J. (1973) Inheritance and c o r r e l a t i o n s of juvenile c h a r a c t e r i s t i c s i n l o b l o l l y pine (Pinus taeda L . ) . S i l v a e Genet. 22(1/2):38-45. McArdle, R.E., Meyer, W.H., and Bruce, D. (1961) The y i e l d of Douglas f i r i n the P a c i f i c Northwest. U.S.D.A. Tech. B u l l . No. 201. pp. 74. McKimmy, M.D. (1959) Factors r e l a t e d to v a r i a t i o n of s p e c i f i c g r a v i t y i n young-growth Douglas f i r . Oreg. State Univ. For. Res. Lab. Res. B u l l . No. 8. pp. 52. (1966) A v a r i a t i o n and h e r i t a b i l i t y study of wood s p e c i f i c g r a v i t y i n 46-year-old Douglas-fir from known seed sources. Tappi 4_9(12): 542-549. , and Nicholas, D.D. (1971) Genetic d i f f e r e n c e s i n wood t r a i t s among half-century-old f a m i l i e s of Douglas-fir. Wood and Fiber 2{A):347-355. McKinnell, F.H. (1970) Wood density studies i n Pinus radiata D. Don. Ph.D. t h e s i s , Aust. Nat. Univ., Canberra. pp. 188. - 202 -McKinnell, F.H., and Shepherd, K.R. (1971) The e f f e c t of moisture a v a i l -a b i l i t y on density v a r i a t i o n within the annual r i n g of r a d i a t a pine. J . Inst. Wood S c i . 5_(6) :25-29. Mergen, F., Burley, J . , and Yeatman, C.W. (1964) V a r i a t i o n i n growth c h a r a c t e r i s t i c s and wood properties of Norway spruce. Tappi 47_(8) : 499-504. M i t c h e l l , H.L. (1964) Patterns of v a r i a t i o n i n s p e c i f i c g r a v i t y of southern pines and other coniferous species. Tappi 47 (5):276-283. Mork, E. (1928) Die Qualitat des Fichtenholzes unter besondere Rucksicht-name auf Scheid-und Papierholz. Papierfabricant 48:741-747. Moroney, M.J. (1951) Facts from f i g u r e s . Penguin Books Ltd., Harmondsworth, England, pp. 472. Morris, W.G., S i l e n , R.R., and Irgens-Moller, H. (1957) Consistency of bud bursting i n Douglas f i r . J . For. 55:208-210. Mozina, I. (1960) Relation;:between annual r i n g width and density i n Douglas-fir. Holz. a l s Roh. und Werkstoff 18(11):409-413. Munger, T.T., and Morris, W.G. (1936) Growth of Douglas-fir trees of known seed source. U.S.D.A. Tech. B u l l . No. 537. pp. 40. Namkoong, G.R., Usanis, A., and S i l e n , R.R. (1972) Age-related v a r i a t i o n i n genetic c o n t r o l of height growth i n Douglas-fir. Theor. Appl. Genet. 42:151-159. Nicholas, D.D. (1963) The influence of environment, genetics and growth c h a r a c t e r i s t i c s on tracheid length and s p e c i f i c g r a v i t y v a r i a t i o n i n a Douglas-fir provenance study. M.Sc. t h e s i s . Oreg. State Univ. Sch. For. pp. 43. N i c h o l l s , J.W.P. • (1967) Preliminary observations on the change with age of the h e r i t a b i l i t y of c e r t a i n wood characters i n Pinus radiata clones. Silvae Genet. 16 (1):18-20. , and Brown, A.G. (1971) The ortet-ramet r e l a t i o n s h i p i n wood c h a r a c t e r i s t i c s of Pinus radiata. Appita 25(3):200-209. , and Dadswell, H.E. (1965) Assessment of wood q u a l i t y for tree breeding. I l l In Pinus radiata D. Don. C.S.I.R.O. Melbourne, Div. For. Prod. Tech. Pap. No. 37. pp. 34. , and F i e l d i n g , J.M. (1964) The h e r i t a b i l i t y of wood c h a r a c t e r i s t i c s of r a d i a t a pine. Silvae Genet. 13(1):68-71. Northcott, P.L., Hancock, W.V., and Kozak, A. (1964) The anatomical c h a r a c t e r i s t i c s of wood from selected trees. Can. Dep. For., For. Prod. Res. Branch Mimeo pp. 52. - 203 -Okkonen, E.A., Wahlgren, H.E., and Maeglin, R.R. (1972) Relationships of s p e c i f i c g r a v i t y to tree height i n commercially important species. For. Prod. J . 22(7):37-42. Orr-Ewing, A.L., and S z i k l a i , O. (1960) Plus-tree s e l e c t i o n f o r Douglas-f i r seed orchards. B.C. For. Serv., For. Res. Rev. 1960:26-27. Panshin, A.J., and de Zeeuw, C. (1970) Textbook of wood technology. V o l . I Structure, i d e n t i f i c a t i o n , uses and properties of the commercial woods of the United States and Canada. McGraw-Hill Book Company, New York. pp. 705. Parker, M.L. (1971) Dendrochronological techniques used by the Geological Survey of Canada. Can. Dep. Energy, Mines and Resources, Geol. Surv. Can. Pap. 71-25. pp. 30. (1972) Techniques i n X-ray densitometry of t r e e - r i n g samples. Paper presented at the 45th Meet. Northwest S c i . Assoc., For. Sect., West Wash. State C o l l . , Bellingham. pp. 11. , and Henoch, W.E.S. (1971) The use of Engelmann spruce latewood density f or dendrochronological purposes. Can. J . For. Res. _l:90-98. , and Jozsa, L.A. (1973) X-ray scanning machine for tree r i n g width and density a n a l y s i s . Wood and Fiber _5 (3) : 192-197 . , and Kennedy, R.W. (1973) The status of r a d i a t i o n densitometry for measurement of wood s p e c i f i c g r a v i t y . Proc. I.U.F.R.O. Meet. Div. 5, S. A f r i c a , Sept.-Oct. 1973. pp. 17. , and M e l i s k i e , K.R. (1970) Preparation of X-ray negatives of tree r i n g specimens for dendrochronological a n a l y s i s . Tree-Ring B u l l . 30(1/4):11-22. , Schoorlemmer, J . , and Carver, L.J. (1973) A computerized scanning densitometer for automatic recording of tree r i n g width and density data from X-ray negatives. Wood and Fiber 5_(3) :237-248. , Barton, G.M., and Jozsa, L.A. (1974) Detection of lignans i n western hemlock by radiography. Wood S c i . and Tech. 8_:229-232. Parrot, L. (1960) De l a v a r i a b i l i t y genefique de l a densite du bois chez l'epicea (Picea exoelsa L i n k . ) . Ann. S c i . For. 17:269-334. Paul, B.H. (1939) V a r i a t i o n i n the s p e c i f i c gravity of the springwood and summerwood of four species of southern pines. J . For. 37:478-482. (1963) The a p p l i c a t i o n of s i l v i c u l t u r e i n c o n t r o l l i n g the s p e c i f i c g r a v i t y of wood. U.S.D.A. Tech. B u l l . No. 1288. pp. 97. Perry, T.O., and Wang, C.W. (1958) . V a r i a t i o n i n the s p e c i f i c g r a v i t y of slash pine wood and i t s genetic and s i l v i c u l t u r a l i m p l i c a t i o n s . Tappi 41(4):178-180. P h i l l i p s , E.W.J. (1960) The beta ray method of'determining the density of wood and the proportion of summerwood. J . Inst. Wood S c i . 5:16-28. - 204 -P h i l l i p s , E.W.J., Adams, E.H., and Hearmon, R.F.S. (1962) The measurement of density v a r i a t i o n within growth rings i n t h i n sections of wood using beta rays. J . Inst. Wood S c i . 10:11-28. Polge, H. (1963) L'ananyse densitometrique de c l i c h e s radiographiques: Une nouvelle methode de determination de l a texture du bois. Ann. Ec. N a t l . Eaux Forets St. Rech. Exper. 20_(4) :530-581. (1965) Study of wood density v a r i a t i o n s by densitometric analysis of X-ray negatives of samples taken with a P r e s s l e r auger. Proc. Meet. I.U.F.R.O. Sect. 41, Melbourne, Oct. 1965. pp. 19. (1968) Tests precoces de l a q u a l i t e du bois sur 25 provenances d'Abies grandis. Ann. S c i . For. 25_(1) :3-23. (1970) The use of X-ray densitometric methods i n dendrochronology. Tree-Ring B u l l . 30=(l/4) :1-10. (1971) H e r i t a b i l i t y de l a densite" du bois de sapin pectine. Ann. S c i . For. 28 (2):185-194. , and I l l y , G. (1968) H e r i t a b i l i t e de l a densite" du bois et c o r r e l a t i o n s avec l a croissance etudiees a l ' a i d e de t e s t s non d e s t r u c t i f s sur plants de pins maritimes de quatre ans. Silvae Genet. 3/7(5/6):173-181. Reck, S.G., and S z i k l a i , 0. (1973) Juvenile-mature c o r r e l a t i o n i n wood c h a r a c t e r i s t i c s of Douglas-fir. Proc. I.U.F.R.O. Meet, of Working Party on Douglas-fir Provenances, Gottingen, W. Germany, Sept. 1973. Rees, L.W., and Brown, R.M. (1954) Wood density and seed source i n young pl a n t a t i o n red pine. J . For. 52;662-665. Richardson, S.D. (1961) A b i o l o g i c a l basis for studies of wood properties. Tappi 44 (3):170-173. Rowe, K.E., and Ching, K.K. (1973) Provenance study of Douglas-fir i n the P a c i f i c Northwest region. I I . F i e l d performance at age nine. Silvae Genet. 22 (4):115-119. Rudman, P. (1968) Growth r i n g a n a l y s i s . J . Inst. Wood S c i . 20:58-63. Schmidt, R.L. (1973) A provenance t e s t of coastal Douglas-fir i n B r i t i s h Columbia. Preliminary r e s u l t s at 6 years. Proc. I.U.F.R.O. Meet, of Working Party on Douglas-fir Provenances, Gottingen, W. Germany, Sept. 1973. p. 68-81. Shepherd, K.R. (1964) Some observations on the e f f e c t of drought on the growth of Pinus radiata D. Don. Aust. For. 28(1):7-22. S i l e n , R.R. (1962a) A study of genetic c o n t r o l of bud bursting i n Douglas-f i r . J . For. 60(7):472-475. (1962b) Pollen d i s p e r s a l considerations for. Douglas-fir. J . For. 60:790-795. (1966) A 50-year r a c i a l study of Douglas-fir i n western Oregon - 205 -and Washington. In Proc. West. For. Genet. Assoc. ©lympia, Wash. p. 6-7. Smith, D.M. (1954) Maximum moisture content method for determining s p e c i f i c g r a v i t y of small wood samples. U.S.D.A. For. Serv., For. Prod. Lab. Rep. 2014. pp. 8. (1955) Relation of s p e c i f i c g r a v i t y to percent summerwood i n wide-ringed Douglas-fir. M.Sc. t h e s i s . State Univ. of N.Y., Syracuse College of Forestry, pp. 114. , and Wahlgren, H.E. (1971) Half a core gives better r e s u l t s i n core-to-core s p e c i f i c g r a v i t y r e l a t i o n s h i p s . Tappi 54_(1):60-62. Smith, J.H.G. (1973) Influence of nitrogen f e r t i l i z a t i o n on widths and s p e c i f i c g r a v i t y of earlywood and latewood and on growth i n volume and weight of young Douglas-fir trees. Univ. of B.C., Fac. For. Mimeo. pp. 65. Smith, W.J. (1970) Wood density survey i n western Canada. Can. For. Serv., Inform. Rep. For. Prod. Lab., Vancouver. No. VP-X-66. pp. 29. Snodgrass, J.D., and Noskowiak, A.F. (1968) Strength and r e l a t e d properties of Douglas-fir from m i l l samples. Oreg. State Univ., For. Res. Lab. B u l l . No. 10. pp. 40. Sorensen, F. (1967) Two-year r e s u l t s of a west-east transect provenance t e s t of Douglas-fir i n Oregon. U.S.D.A. For. Serv. Res. Note PNW-72. Spurr, S.H. (1961) Observations on Douglas-fir i n New Zealand. N.Z. For. Serv., For. Res. Inst. Tech. Pap. No. 38. Squ i l l a c e , A.E., and Bengston, G.W. (1961) Inheritance of gum y i e l d and other c h a r a c t e r i s t i c s of slash pine. Proc. Sixth South. Conf. For. Tree Improvement:85-96. Stamm, A.J. (1964) Wood and c e l l u l o s e science. Ronald Press Co. N.Y. pp. 549. Stern, K., and Roche, L. (1974) Genetics of fo r e s t ecosystems. Springer-Verlag, N.Y. pp. 330. Sterns, R.W. (1918) Canadian Douglas f i r . I t s mechanical and p h y s i c a l properties. Can. Dep. I n t e r i o r , For. Branch B u l l . No. 60. pp. 84. Stonecypher, R.W. (1966) The l o b l o l l y pine h e r i t a b i l i t y study. Southlands Exp. For., Woodlands Dep. Tech. B u l l . No. 5. pp. 128. , and Zobel, B.J. (1961) Inheritance of s p e c i f i c gravity i n f i v e - y e a r - o l d seedlings of l o b l o l l y pine. Tappi 49:303-305. , Cech, F.C., and Zobel, B.J. (1964) Inheritance of s p e c i f i c g ravity i n two- and three-year o l d seedlings of l o b l o l l y pine. Tappi 47_(7) :405-407. Strand, R.F. (1964) S o i l and growth studies i n Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) stands near M o l a l l a , Oregon. Ph.D. t h e s i s . Oreg. State Univ. Sch. For. pp. 193. - 206 -Sweet, G.B. (1965) Provenance differences i n P a c i f i c coast Douglas-fir. I. Seed and seedling c h a r a c t e r i s t i c s . Silvae Genet. 14:46-56. Sylvander, R.B., and Smith, J.H.G. (1973) Widths and s p e c i f i c g r a v i t y of earlywood and latewood components of annual rings from I n t e r i o r B r i t i s h Columbia. Univ. of B.C., Fac. For. Mimeo. pp. 61. S z i k l a i , O. (1964) V a r i a t i o n and inheritance of some p h y s i o l o g i c a l and morphological t r a i t s i n Pseudotsuga menziesii (Mirb.) Franco var. men-ziesii. Ph.D. t h e s i s . Univ. of B.C., Fac. For. pp. 129. T h i e r c e l i n , F. (1960) T a r d i v i t e du'- debourrement et densite du bois dans une population adulte de Pieea abies Karst. Ann. S c i . For. 27_(3) : 243-254. Thoby, M. (1975) V a l i d i t e de t e s t s precoces d'appreciation des caracteres technologiques du bois de Douglas. Ap p l i c a t i o n a une pl a n t a t i o n compara-t i v e de 24 provenances de Douglas(Peyrat de chateau) en l i a s o n avec des caracteres mofphologiques et auxometriques. Inst. N a t l . Rech. Agron. Centre N a t l . Rech. For. June 1975. pp. 59. Turnbull, J.M. (1974) Some factors a f f e c t i n g wood density i n pine stems. Proc. 5th B r i t . Emp. For. Conf., Oxford, pp. 22. U.S. Forest Service (1965) Western wood density survey; report no. 1. U.S.D.A., For. Serv. Res. Pap. FPL-27. pp. 60. Wahlgren, H.E., and Maeglin, R.R. (1966) Estimating tree s p e c i f i c g r a v i t y of Maine c o n i f e r s . U.S.D.A., For. Serv. Res. Pap. FPL-61. pp. 22. , and Schumann, D.R. (1972) Properties of major southern pines. Part I. Wood density survey. U.S.D.A., For. Serv. Res. Pap. FPL-176. Walters, J . , and Soos, J . (1962) The v e r t i c a l and h o r i z o n t a l organization of growth i n some conifers of B r i t i s h Columbia. Univ. of B.C., Fac. For. Res. Pap. No. 51. pp. 22. , and Ching, K.K. (1966) Pattern of bud burst i n a Douglas-fir provenance t r i a l . Unpublished Report (pers. comm.). Warrack, G., and Joergensen, C. (1950) P r e c i s i o n measurement of r a d i a l growth and d a i l y r a d i a l f l u c t u a t i o n s i n Douglas-fir. For. Chron. 26_(1): 52-66. Wellwood, R.W. (1952) The e f f e c t of several variables on the s p e c i f i c g r a v i t y of second-growth Douglas-fir. For. Chron. 28:34-42. , and Smith, J.H.G. (1962) Variations i n some important q u a l i t i e s of wood from young Douglas f i r and hemlock tr e e s . Univ. of B.C., Fac. For. Res. Pap. No. 50. pp. 15. - 207 -Wiksten, A. (1955) (A method for measuring springwood and summerwood i n annual r i n g s . I I I . D i f f e r e n t i a t i o n of springwood and summerwood through staining.) Meddel. fran Statens Skogsforsoksanstalt 34_:476-488. Univ. of B.C. Translation No. 19. E.Kvarv, 1962. Wilcox, M.D. (1974) Douglas f i r provenance v a r i a t i o n and s e l e c t i o n i n New Zealand. N.Z. For. Serv., For. Res. Inst., Genet. Tree Impr. Rep. No. 69. pp. 42. Wommack, D.E. (1960) E f f e c t s of winter c h i l l i n g and photoperiod on growth resumption i n Douglas f i r . Abstr. i n B u l l . E c o l . Soc. Amer. 41_(2):57-58. Worrall, J.G. (1970) I n t e r r e l a t i o n s h i p s among some phenological and wood property-variables i n Norway spruce. Tappi 53_(1) : 58-63. (1975) Provenance and c l o n a l v a r i a t i o n i n phenology and wood properties of Norway spruce. Silvae Genet. 24_(1) :2-5. Zahner, R. (1963) Internal moisture stress and wood formation i n c o n i f e r s . For. Prod. J . 13_(6) :240-247. Zobel, B.J. (1961) Inheritance of wood properties i n c o n i f e r s . Silvae Genet. 10(3):65-70. (1964) Breeding f o r wood properties i n f o r e s t trees. Unasylva 18(2/3):89-103. , and Rhodes, R.R. (1955) Relationship of wood s p e c i f i c g r a v i t y i n l o b l o l l y pine {Pinus taeda L.) to growth and environmental f a c t o r s . Tex. For. Serv. Rep. No. 11. pp. 32. (1956) S p e c i f i c g r a v i t y estimation of mature l o b l o l l y pine from juvenile wood and seedling limb sections. For. S c i . 2_(2) :107-112. (1957) S p e c i f i c gravity indices f o r use i n breeding l o b l o l l y pine. For. S c i . 3_(3) :281-285. , Cole, D., and Stonecypher, R.W. (1962) Wood properties of clones of slash pine. Proc. For. Genet. Workshop, Macon, Georgia. South For. Tree Impr. Comm. Sponsored Publ. No. 22:32-39. , Webb, C , and Henson, F. (1959) Core or juvenile wood of l o b l o l l y and slash pine trees. Tappi 42 (5):345-356. - 208 -APPENDICES - 209 -SEED SOURCE PLANTATION N NIMPKISH N H HANEY H M MOLALLA M V VALLEY V DORENA 0 B BUTTE F A L L S Appendix 1 Natural d i s t r i b u t i o n of Douglasr-f i r (a) , and l o c a t i o n of provenance t r i a l seed o r i g i n s and plantations (b) . - 210 -Appendix 2 : H e r i t a b i l i t y Estimates for Growth Ring Components Species Age H e r i t a b i l i t y Narrow sense Broad Sense Reference P. radiata 12 10 10 3 6 2-3 5 Pioea albies 24-29 P. pinaster •i P. taeda .00 - .05 Ring Width .50 - .63 .20 .53 .02 - .09 .14 - .35 .20 .50 .69 Dadswell et al. (1961) Nicholls et al. (1964) Nicholls and Brown (1971) Polge and*Illy (1968) Keller (1973) Stonecypher et al. (1964) Matzirisand Zobel (1973) Worrall (1975) P. taeda 7-8 Earlywood Width .25 - .26 Goggans (1964) P. taeda 7-8 Pioea abies 24-29 Latewood Width .28 - .88 .51 .65 Goggans (1964) Worrall (1975) P. taeda P. radiata Pioea abiea 7-8 12 . 10 18 Percentage Latewood .25 - .81 .45 - .72 .14 .85 Goggans (1964) Dadswell et al. (1961) Nicholls et al. (1964) Kennedy (1966) Wood Density P. radiata 13-19 12 10 12 P. elliottii 12-14 3 • " • 5 " 6 3 .21 - .56 .43 .43 .43 .50 - .70 .74 - .75 .53 - .75 .57 - .90 .73 .46 - .62 .44 - .49 Fielding and Brown (1960) Dadswell et al. (1961) Nicholls et al. (1964) Burdon and Harris (1973) Squillace et al. (1962) Zobel et al. (1962) Einspahr et al. (1964) Goddard and Cole (1966) Franklin and Squillace (1973) - 211 -Appendix 2: Continued H e r i t a b i l i t y Species Age Narrow sense Broad sense Reference Wood Density (continued) P. taeda 2-6 .37 - 1.00 .7 - .84 van Buijtenen (1962) II 7-8 .76 - .87 Goggans .(1964) it 2-3 .56 - .72 Stonecypher et al. (1964) II 5 .47 Matziris and Zobel (1973) P. pinaster 3 .65 - .75 Polge and I l l y (1968) it 6 .30 - .50 Keller (1973) Pioea abies 18 .84 Kennedy (1966) ii 24-29 .61 .51 - .70 Worrall (1975) Pseudotsuga McKimmy (1966) menziesii 46 .00 - .66 Abies alba 4 .22 - .33 Polge (1971) Earlywood Density P. taeda 7-8 .04-- .-35 Goggans (1964) Pioea abies 18 .86 Kennedy (1966) ii 24-29 .80 .63 - .86 Worrall (1975) . Latewood Density P. radiata 12 .45 Dadswell et al. (1961) P. taeda 7-8 .63 - .67 Goggans (1964) Pioea abies .56 Kennedy (1966) P. radiata P. pinaster 10 6 Minimum Ring Density .25 - .56 .10 - .16 Nicholls and Brown (1971) Kelle r (1973) P. radiata P. pinaster 10 6 Maximum Ring Density .16 - .42 .03 - .11 Nicholls and Brown (1971) Kelle r (1973) - 212 -Appendix 3: Sources of Weather Data Site Weather Station Distance from site (km) Nimpkish Nimpkish (precip.) 10 Strathcona Dam (temp.) 100 Haney Haney 1 Molalla Scott's Mills 50 Valley Dallas 10 Dorena Dorena Dam 1 - 213 -Appendix A : Data on Clone O r i g i n s Clone No. S e l e c t i o n Date Parent Tree Information Latitude Longitude Land D i s t r i c t 25 1957 48°52' 124°04' Cowichan Lake, V.I., B.C. 28 1957 48°56' 123°53' Cowichan Lake, V.I., B.C. 36 1958 48°51' 123°47' Seymour, V.I., B.C. 55 1959 49°53' 126 O05 , Nootka, V.I., B.C. 59 1960 N.R. N.R. Denmark, o r i g i n unknown 60 1960 50°15' 125°24' Sayward, V.I., B.C. 62 1960 49°18' 125°13' Clayoquot, V.I., B.C. 70 1960 48°35' 123°58' Malahat, V.I., B.C. 82 1959 49°18' 122°34' New Westminster, L.M., B.C. 83 1959 49°18' 122°34' New Westminster, L.M., B.C. N.R. = not recorded V.I. = Vancouver Island L.M. B.C. = Lower mainland = B r i t i s h Columbia - 214 -Appendix 5: Flushing Groups Used i n F i e l d Assessment Group Percentage of Buds Burst 0 0 I 1 - 1 0 II 11 - 25 III 26 - 50 IV 51 - 75 V 76+ - 215 -Appendix 6(a): Location of Four Selected Douglas-fir Trees on the U.B.C. Campus. Trea A M Age/Yrs Tree E PBH/m 694 - 216 -Appendix 6(b): Parent Tree Data (1975) Tree A B E 11 O r i g i n Natural Natural Natural P l a n t a t i o n Age (yr.) 55 53 55 42 Height (m) 28.3 24.7 29.3 22.9 Diameter (mm) 655 694 695 390 Ave. date of f l u s h i n g (1959-63) May. 26 May 19 May 22 June 4 -Appendix . 6(c): Sample Numbers for the D i a l l e l Design A B ' E 11 A 0 6 6 6 B 5 0 6 4 . E 6 6 5 6 11 0 4 4 0 - 217 -Appendix 7: Summary of Density Data f o r F e l l e d Trees Tree Core Density g /c 3 m^ f o r d i s c no: Tree Density g /cm3 No. Density g/cm3 1 2 3 4 5 6 7 1 .372 .355 .355 .343 .326 .309 .312 .299 .338 2 .382 .362 .365 .346 .325 .342 .344 .350 3 .428 .393 .387 .353 .346 .334 .338 .367 4 .390 .411 .383 .338 .323 .309 ' .361 5 .363 .327 .344 .320 .328 .309 .296 .327 6 .441 .472 .450 .412 .387 .371 .345 .419 7 .418 .403 .387 .375 .368 .344 .342 .376 8 .411 .402 .382 .372 .352 .342 .344 .337 .368 9 .448 .449 .431 .405 .388 .374 .365 .350 .407 10 .411 .397 .388 .362 .344 .342 .331 .368 11 .400 .384 .368 .352 .338 .337 .327 .319 .355 12 .380 .382 .366 .354 .334 .316 .304 .324 .350 13 .367 .367 .365 .339 .339 .330 .306 .348 14 .399 .388 .390 .368 .360 .348 .337 .372 15 .393 .400 .376 .340 .360 .334 .366 Means .400 .393 .382 .359 .348 .336 .330 .326 .365. - 218 -Appendix 8:- C o r r e l a t i o n Matrix f o r F e l l e d - t r e e Density Values Core Density Density at height: m Whole-tree Density 0.2 1.5 3.0 4.5 6.0 Core Density 1.000 0.2 0.866 1.000 1.5 0.888 0.955 1.000 3.0 0.871 0.866 0.919 1.000 4.5 0.822 0.806 0.860 0.872 1.000 6.0 0.823 0.752 0.843 0.883 0.871 1.000 Whole-tree Density 0.911 0.963 0.988 0.932 0.904 0.879 1.000 Appendix 9(a): Provenance T r i a l Increment Core Data - Diameter Summary Location PROVENANCE AND PLOT GRAKD M5AXS NIMPKISH HANEY MOLALLA VALLEY BUTTE A B C D Mean A B C D Mean A B C D Mean A B C D Mean A B C D j Mean N 123.7 116.0 112.2 117.2 107.6 120.1 114.1 115.0 102.9 121.1 102.3 112.1 119.9 109.8 138.9 140.4 113.3 124.0 108.3 119.2 j i 119. 8 114 .7 117.3 113 .8 114 .5 114.2 112 .0 107.2 109.6 114 8 139 .6 127.3 11 8.9 113. 7 116.3 j 116.9 H 143.7 143.8 150.3 139.2 124.8 122.5 130.4 135.5 140.7 147.4 120.4 143.7 134.5 158.2 148.2 115.1 136.5 150.0 135.5 136.7 143 .7 144 .7 144.2 123.6 132 .9 128.3 144.1 132 .1 138.1 146. 3 131 .6 139.0 14: .2 13 3.6 139.9 137.9 M 06.1 71.9 54.7 | 46.3 61.3 66.8 41.5 48.2 80.4 68.0 40.4 40.9 78.1 65.0 61.0 48.5 71.4 63.0 51.0 30.9 69. 0 50. 5 59.7 64. 0 44. 8 54.4 74. 2 48.6 61.4 71.5 54. 8 63.2 67.2 41 .0 54.1 58.6 V 141.6 139.2 140.8 132.9 157.5 154.9 143.4 151.1 144.1 139.7 133.0 170.0 155.4 147.5 171.1 13C.2 154.6 117.9 139.7 116.2 140 .4 136 .8 138.6 156 .2 147 .2 151.7 11] 137.5 .9 85.7 151.5 146.7 151 4 153.6 152.5 13C 127.9 132.0 M4.3 D 9B.6 110.3 94.6 100.8 123.6 06.9 91.5 100.1 10 4.4' 104.4 10 1.7 101.7 111.6 111.6 105.2 105.2 95.8 103.7 GRAND MEA NS 112.8 110.1 113.5 117.4 107.6 • nr.3 All data expressed in mm. Plantation 1 - Plots A £ B. Plantation 2 - Plots C S D. Appendix 9(b): Provenance T r i a l Increment Core Data - Wood Density Summary. Location (Site) PROVENANCE DATA BY PLOTS (A - D) AND PLANTATIONS' (ASBlCSD) GRAVIMETRIC DENSITIES g/cm3 GRAND KEAtl NIMPKISH HANEY MOLALLA VALLEY BUTTE A B C 0 Mean A B C D Mean A B C D Mean A B C D Mean A B C D Mean • H .421 .406 .425 .413 .429 .409 .417 .392 .415 .401 .414 .404 .409 .419 .393 .387 .404 .407 .442 .414 114 .419 .416 119 . i 105 .412 .4 08 .409 .409 .414 .390 .402 .406 .428 .417 .411 H .395 .397 .413 .397 .379 .371 .373 .356 .386 .392 .412 .384 .395 .367 .402 .418 .408 .401 .396 .409 396 .4 35 .400 375 !65 .370 .3 89 .3 98 .393 .3 81 i .4 10 .395 .4 05 .403 .404 .393 M .476 .460 .474 .492 .455 .459 .486 .465 .456 .437 .464 .403 .460 .476 .455 .489 .457 .463 .497 .492 . t 168 .4 33 .475 .457 .466 .447 .474 .460 .468 .472 .470 .460 .495 .477 .470 V .385 .390 .390 .406 .362 .270 .382 .385 .394 .390 .377 .372 .381 .391 .377 .379 .394 .394 .378 1.377 1 !87 .398 .392 366 .383 .374 .3 92 .374 .303 .386 .378 .382 .394 .377 .385 .3 84 D .405 .385 .390 .411 .413 .417 .386 .4.14 .419 .407 .395 .395 .401 .401 .415 .415 .400 .400 .413 .413 .405 GRAND MEANS .416 .405 .412 .410 .419 .412 - 221 -Appendix 10: Analyses of Variance Models a) Combined Sites (gravimetric samples) Source of Variation .. d.f. E.M.S. Sites 3 a 2 e + + 2 0°PL + 40o 2 + PPL 200a 2 PL + 80o 2 + 4OO02 STP ST Provenances 4 a 2 e + + 2 0 ° P L + 40o 2 PPL + 80a 2 „ + 320a 2 STP PR Sites x prov-•enances 12 a 2 e + H + 2 0°PL + 40 o 2 PPL + 80a 2 STP Plantations 4 • a 2 e + 202 T + 2 0 ° P L + 4O02 + PPL 2000^ Provenances x plantations 16 0 2 e + 202 T + 2 0 ° P L + 40a 2 PPL Plots within provenances 40 a* e + T + 2 0°PL Trees within 2 2 plots 720 a e + 20 T Error 800 2 °e TOTAL 1599 b) Individual Sites (gravimetric samples) Source of Variation d.f. E.M.S. Sites 4 2 «e + 24 + 2 0 4L + 4 0 a 2  aPPL Plantations- 1 < + 20 2 + 2°°PL + 4 0 a 2 + 200(T2 °PPL °PL Provenances x plantations 4 < + 24 + 2°°PL + 4 0 a 2 °PPL Plots within provenances 10 0 2 e + 24 + 2°°PL Trees within plots 180 a 2 e + 24 Error 200 2 e TOTAL 399 - 222 -Appendix 10:(cont'd) c) Individual Sites (densitometric samples) Source of Variation d.f. E.M.S. • Provenances Plantations Provenances x plantations Plots within provenances Trees within p l o t s 4 1 4 10 80 4 * 5 02 T + l O o ^ + 2 0 3 2 r ° T + 5 4 r + 1 0 C T P P L + 5 0 O P L 4 + 5 0 2 T + 10a2 p L a 2 + 5a 2 T PT 2 °T TOTAL 99 d) Combined s i tes (densitometric samples) Source of Variation d.f. E.M.S. Sites Provenances Sites x provenances Plantations Provenances x plantations Plots within provenances Trees within plots a 2 > 5 a 2 „ + 1 0 a 2 p L + 5 0 a p L + 2 0 0 ^ + l o o a ^ °T + 5 aP T + 1 0 0 P P L + 2 0 a S T r - + 8 0 a p R a 2 + 5 a 2 + 10a 2 + 200-2 T PT PPL STP a 2 + 5 a 2 + i o a 2 + 5QO-2 T PT PPL PL ° 2 + 502 + ioa-2 -T PT PPL ' -°l + 5a 2 T PT O2 T N TOTAL Appendix 10: (cont'd) e) Anova Model for Individual Sites Incorporating Year Effect Source d.f. E.M.S. Provenances 4 0-2 + T + 5a 2 YPT > PT + 10a 2 YPP + 2 0 C 7YPR 5 ° a 2 T + 100a 2 PPL PR Plantations 1 a 2 e + T + 5a 2 YPT + PT + 10a2 YPP + 50a 2 YPL + 5 0 a 2 p L + 250a 2 L Provenances x plantations 4 a 2 e + 5 ° 2 T + 5a2 YPT + 2 5 ° P T + 10a 2 YPP + 5 0*PPL Years 4 a 2 e + 5a2 YPT + 10a2 YPP + 50a 2 YPL + 20a 2 p R- + 100a 2 Years x provenances 16 e + 5a2 YPT + 10a2 YPP + 20a 2 YPR Years x plantation's . 4 a2 e + 5<J2 YPT + 1 0 aYPP + 50a2 YPL Years x provs. x plants. • 16 + YPT + 1 0 a 2 p p Plots within provenances 10 < + + 5 aYPT + 2 54T Years x plots 40 a 2 e + 5 cYPT -Trees within plots 80 a2 e + Years x trees 320 a 2 e TOTAL 499 - 224 -Appendix 11: C o e f f i c i e n t s of Determination f o r the Density/Growth Rate Relationship Relationship Property Group Linear Hyperbolic Logarithmic Parabolic Ring Width Provenances N .47 .57 .57 .61 H .45 .49 .50 .50 M .52 .53 .56 .56 V .50 .51 .53 .55 B .55 -.59 .61 .62 S i t e s N .25 .26 .26 .26 H .20 .23 .22 .23 M .16 .20 .18 .17 V .05 .07 .06 .06 D .15 •15 .16 .16 A l l .49 .53 .55 .56 Diameter Provenances N .45 .52 .52 .54 H .52 .49 .54 .54 M .53 .53 .56 .56 V .47 .49 52 .53 B .51 .54 .58 .58 S i t e s N .23 .24 .24 .24 H .17 .19 .18 .19 M .18 .19 .19 .18 V .07 .08 .07 .07 D .02 .03 .03 .04 A l l .49 .50 .53 . .54 Appendix 12: Summary of Ring Component Values by S i t e s and Provenances, a) Rings 1-5. . — - •- . ... _. — . . - - _.. . . . _ .— — — — — . — .._ Site NIMPKISH HANEY MOLALLA VALLEY DORENA fTcver-ar.ee . ii H M V B Mean N II M V B Mean N H M V B Mean N H M V B Mean N i H K V . B . Heaa G^ T j c r i e r i t rJJ 223 X 100 469 549 443 519 443 484 537 547 519 567 511 536 275 282 323 305 288 295 631 647 521 584 501 577 519 704 496 588 439 549 I>* £3= X 100 321 394 309 365 303 338 374 413 360 412 346 381 187 189 226 213 189 201 472 510 401 451 375 442 394 545 329 440 301 402 LW ca. x 100 148 155 134 154 140 146 163 '134 159 155 165 155 88 93 97 92 99 94 159 137 120 133 126 135 125 159 167 148 138 147 PLW v 31.9 23.7 30.9 31.4 33.4 31.3 33.0 25.3 32.2 29.0 34.1 30.7 32.8 32.7 31.2 30.5 36.5 32.8 26.1 20.0 24.5 24.9 26.9 24.5 24.3 23.4 33.4 25.5 31.3 27.6 TO g/cn3 .414 .399 .409 .420 .421 .412 .395 .361 .401 .394 .419 .394 .446 .426 .<:13 .412 .451 .430 .382 .349 .366 .384 .306 .373 .395 .384 .420 .405 .417 .4 06 ED g/cm3 . 233 .292 .290 .290 .291 .292 .269 .271 .285 .297 .296 .204 .325 .319 .:no .300 .332 .319 .276 .267 .266 .283 .274 .271 .288 .295 .287 .298 .299 .293 LI/ g/<n=3 .634 .662 .674 .600 .672 .674 .640 .624 .642 .620 .649 .637 .691 .648 .(.59 .651 .661 .662 .680 .625 .671 .689 .691 .671 .720 .681 .712 .717 .671 .702 JC.'D c/crn 3 .236 .246 .242 .242 .229 .239 .198 .218 .221 .236 .237 .222 .282 .282 .::49 .251 .291 .271 .219 .205 .205 .214 .220 .213 .229 .230 .227 .240 .252 .236 MXD g/cK 3 .829 .785 .813 .830 .820 .015 .759 .736 .773 .744 .780 .760 .851 .779 ,t:04 .791 .005 .806 .044 .009 .825 .071 .063 .842 .912 .045 .904 .916 .836 .882 ir.i y/ca3 .396 .370 .384 .381 .300 .382 .371 .352 .357 .332 .353 .353 .366 .329 .149 .343 .329 .343 .404 .358 .404 .407 .417 .398 .440 .386 .425 .419 .372 .403 fr.C g/cc 3 .593 .539 .572 .588 .591 .576 .562 .518 .552 .500 .551 .538 .569 .498 .554 .540 .513 .535 .626 .604 .620 .656 .643 .630 .682 .614 .677 .676 .583 .647 Appendix 12: Summary of Ring Component Values by S i t e s and Provenances, b) Rings 1970-1974. s r r s KISI! HAI1EY MOLALLA VALLEY DORENA P rover.ar.ee H M V B Mean N H M V B Mean N H M V B Mean H H M V B Mean N • H M V B Xear. Cor-aoner-.t r.W c o X 100 335 326 333 389 379 352 384 386 431 418 420 408 194 155 195 179 172 179 469 456 499 542 458 485 360 470 432 368 313 389 EW = x 100 205 207 237 261 260 234 253 275 307 308 294 287 92 74 102 92 79 88 313 315 354 396 318 339 231 338 270 246 213 260 I.W TOT X 100 130 119 96 128 119 118 131 110 124 110 12G 120 102 80 97 87 92 91 156 141 145 146 140 146 129 132 162 122 100 129 FI.W \ 45 .4 4 1 . 2 3 4 . 3 3 S . 0 3 6 . 9 39 .2 38.1 34.'7 33 .7 33 .1 35 .9 35 .1 58 .3 56 .0 4 9 . S 50 .8 58.4 54.8 40 .2 36 .7 34 .7 34 .2 38 .7 36 .9 44 .7 36 .6 4 2 . 9 4 1 . 2 41 .1 4 1 . 3 .•180 .473 .438 .461 .454 .461 .445 .420 .431 .426 .447 .434 .525 .507 .496 .494 .530 .510 .452 .435 .426 .430 .451 .440 .479 .446 .471 .473 .434 .466 a . 303 .400 .366 .381 .378 .384 .361 .347 .352 .366 .372 .360 .423 .416 .403 .401 .433 .415 .365 .351 .353 .363 .369 .360 .390 .378 .372 .385 .392 . 3o4 LD g / c r . 2 .613 .600 .602 .613 .601 .605 .589 .579 .601 .577 .602 .589 .629 .596 .615 .592 .625 .612 .624 .607 .614 .633 .630 .622 .649 .623 .657 .657 .618 .641 :C;D q / c a 2 .336 .353 .310 .319 3 .10 .326 .282 .288 .284 .302 . .304 .292 .390 .382 . 35? .357 .403 .377 .305 .207 .285 .291 .312 .296 .331 .316 .306 .333 .345 .326 KXD q/CK 2 .770 .713 .725 .762 .753 .745 .726 .677 .723 .689 .736 .710 .787 .718 .761 .727 .762 .752 .785 .768 .772 .806 .B04 .707 .845 .763 .850 .833 .767 .811 L1JI q / c i i 2 . 220 .196 .234 .232 .223 .221 .232 .232 .248 .212 .229 .231 .206 .100 .219 .202 .192 .201 .260 .261 .261 • .270 .261 .263 .259 . 246 .285 .272 .225 .257 .434 .360 .414 .443 .443 .419 .444 .389 .439 .387 .432 .418 .398 .336 .412 .370 .359 .370 .480 .481 .487 .514 .491 .491 .514 .447 . 544 .500 .421 .486 Appendix 13: C o r r e l a t i o n M a t r i c e s f o r Provenance Ring Components. a) RINGS 1 - 5 Name Mean Std. Dev. Name Mean Std. Dev. RW 359.751 142.562 MND 0.323098 O.561071E-01 EW 259.520 120.548 MXD 0.755573 0.682426E-01 LW 119.829 43.1203 PLW 41.4656 13.7600 RD 0.462002 0.519344E-01 UNI 0.232151 0.482661E-01 ED 0.380193 0.418596E-01 RNG 0.432456 0.793429E-01 • LD 0.611998 0.358545E-01 Correlation Matrix Variable RW EW LW RD ED LD MND MXD PLW UNI RNG RW 1.0000 1.000 EW 0.9600 1.0000 LW . 0.6218 0.3777 1.0000 RD -0.5774 -0.7377 0.1541 1:0000 ED -0.5808 -0.6793 -0.0207 0.8384 1.0000-LD 0.0336 -0.0665 0.2986 0.4706 0.2175 1.0000 MND -0.6366 -0.7272 -0.0711 0.8013 0.9697 0.2114 1.0000 MXD 0.0685 -0-.0494 0.3660 0.4603 0.2205 0.9180 0.1971 1.0000 PLW -0.5948 -0.7523 0.1370 0.9334 0.8421 0.3071 0.8599 0.3217 1.0000 UNI 0.5218 0.5325 0.2372 -0.4112 -0.6890 0.5550 -0.6681 0.4907 -0.4897 1.0000 RNG 0.5092 0.4719 0.3652 -0.2277 0.4965 0.6398 '-0.5381 0.7203 -0.3316 0.8947 1.0000 b) RINGS 1970-74 Name Mean Std. Dev. Name Mean Std. Dev. RW 481.738 185.214 MND 0.230102 0.354091E-01 EW 347.140 154.348 MXD 0.814471 0.796599E-01 LW 134.224 49.1257 PLW 29.5633 8.14131 RD 0.402776 0.477160E-01 UNI 0.375129 'J.537347E-01 ED 0.292064 0.294014E-01 RNG 0.578338 0.813997E-01 LD 0.667189 0.481951E-01 Correlation Matrix ' Variable RW EW LW RD ED LI) UND MXD PLW UNI ' RNG RW 1.0000. EW 0.9744 1.0000 LW 0.7080 0.5313 1.0000 RD -0.3717 -0.5173 0.2243 1.0000 ED -0.2806 0.3281 -0.0260 0.7201 1.0000 LD 0.0167 . -0.0380 0.1824 0.4806 0.1064 1.0000 MND r0.4393 0.4881 -0.1222 0.7461 0.8809 0.1443 1.0000 MXD 0.0569 0.0084 0.1883 0.4634 0.1626 0.9*77 0.1727 1.0000 PLW -0.4126 0.5817 0.2731 0.8887 0.4727 0.2379 0.5524 6.2110 1.0000 UNI 0.1689 0.1459 0.1783 0.0363 -0.4525 0.8';85 -0.3531 0.7607 -0.0450 1.0000 1<NG 0.2461 0.2198 0.2371 0.1296 -0.2337 0.8050 -0.2653 0.9039 -0.0331 0.8981 1. 0000 - 228 -Appendix 14>_ Regression s t a t i s t i c s by s i t e s - d e n s i t y / r i n g width and density/percent latewood Ring S i t e Intercept Slope S.E. 2 Group b o b l (y) X\. Ring Density/Ring Width (a) 1-5 N .538 - .022 .035 .22** H .527 - .023 .034 .26** M .531 - .012 .056 •Oln.s. V .507 - .014 .034 .18** D .507 - .010 .038 .12* A l l .538 - .021 .042 .33** 1970-74 N .465 - .011 .038 .14** H .473 - .015 .037 .28** M .462 - .011 .050 .OGn.s. V .369 . 008 .040 .OOn.s. D .441 - .006 .042 •05n.s. A l l .449 - .010 .044 .14** Ring Density/Percent Latewood (b) 1 - 5 N .331 .0033 .017 .82** H .316 .0034 .017 .81** M .320 .0035 .023 .84** V .317 .0033 .016 .83** D .318 .0036 .014 .88** A l l .316 .0035 .019 .87** 1970-74 N .229 .0059 .019 .78** H .240 .0050 . 017 .85** M .263 .0051 .023 .80** V .250 .0050 .019 .76** . D .273 .0048 .019 .80** A l l .249 .0052 .022 .80** N = Nimpkish * = s i g n i f i c a n t at the 5% l e v e l H = Haney ** = s i g n i f i c a n t at the 1% l e v e l M = M o l a l l a n.s. = not s i g n i f i c a n t V = V a l l e y D = Dorena Appendix 15: Regression Statistics for the Effects of Climatic Variables on Ring Components (a) Coefficients of Determination for Monthly Precipitation Component Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. RW .004 .002 .003 .001 .001 .020 .008 .001 * * * .335 .011 .084 .048 EW .006 • OC-^  .003 .018 .011 .025 .000 .004 *** .278 .001 LW .001 .032 * .121 ** .173 .102 .001 .061 .002 * .136 .059 ** .190 *** .304 RD .029 .026 .093 * .111 .088 .049 .024 .000 .076 .070 .000 .080 ED .109 .099 .026 .011 .004 .014 *** .153 .030 .024 ** .250 LD .009 .019 .001 .002 .013 .046 .036 .005 .029 .043 .017 .032 MND .063 .071 .000 .049 .003 .002 .063 .020 .000 MXD .000 .016 .001 .004 .016 .078 .056 .007 .065 .040 .055 .047 (b) Regression Coefficients x 1 standard deviation of monthly precipation indices RW -0.75 -0.57 -0.59 0.39 0.38 • . 1.46 1.16 0.43 *** 6.00 -1.15 3.26 2.62 EW -1.12 0. 00 0.59 1.93 1.54 2.20 0.00 0.86 *** 6.86 0.00 LW -0.37 -1.72 * -3.27 ** -3.86 -3.08 -0.37 2.32 -0.43 * 3.00 -2.30 ** 4.07 * ** 5.25 RD 0.37 -0.57 -0.89 * -1.16 -0.77 -0.73 0.39 0.00 -0.86 -1.15 0.00 1.05 ED 0.75 -0. 57 -0.59 0.00 0.00 -0.37 ** 0.77 0.43 0.43 ** -1.15 LD 0.37.A -0.57 0.00 . 0.00 •fO.38 -0.37 -0.39 0.00 -0.43 -1.15 0.00 -0.52 MND 0.75 -0.57 0.00 0.39 0.00 0.00 0.77 -0.43 0.00 MXD - 0.00 -6.57 0.00 0.39 -0.38 -1.10 -0.77 0.43 -0.86 -1.15 0.81 -0.52 * Statistically significant at the 10% level ** Statistically significant at the 5% level ** Statistically significant at the 1% level Appendix 15: (Cont'd) (c) Coef f ic ients of Determination for Mean Monthly Temperatures Component • Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June ' July Aug. RW . .033 ^ .013 .003 .022 .001 .062 .044 .023 * .125 .007 .022 .024 EW .021 .000 .014 .010 .003 . . 048 .066 .004 * .113 .000 LW .031 .127* .027 .038 .014 .027 . .000 .105 .024 .095 .000 * * .238 RD .017 .038 .070 .002 .026 .016 * * .161 .028 .057 .010 .027 .064 ED .002 .095 .010 .110 .095 .014 * * * .334 .013 .003 .026 LD .009 .013 .022 .009 .004 .022 .003 .000 .000 * .138 .004 .028 MND .003 .047 .004 .047 . 019 .002 * * * .272 .026 .006 MXD .016 .024 .015 .018 .013 .002 .001 .017 .000 * * .253 .035 .120* (d) Regression Coef f ic ien ts x 1 standard deviat ion of monthly temperature RW - 2.03 1.32 0.61 1.69 -0.28 -2.79 2.37 • 1.67 * -3.96 0.96 1.65 -1.75 . EW 2.06 0.13 it 1.63 1.45 -0.74 -3.13 3.66 0.86 -4 .81* 0.00 LW 1.62 3.28 -1.53 1.80 1.08 -1.54 0.08 3.00 -1.42 2.83 -0.69 * * -4.45 RD -0.41 0.63 -0.81 -0.12 0.51 0.38 * * -1.28 0.51 0.77 0.30 -0.51 -0.81 ED -0.09 0.57 0.20 0.64 . 0.57 0.24 * * * -1.13 0.21 0.11 0.30 LD -0.24 -0.31 -0.41 -0.23 0.17 -0.38 -0.16 0.00 0.00 -0.96* -0.63 0.44 MND -0.15 0.57 -0.20 0.52 0.34 0.10 * * * -1.32 0.38 0.19 MXD -0.47 0.57 -0.41 -0.46 0.40 -0.19 -0.12 -0.47 0.04 * * -1.81 -1.05 . * 1.25 * S t a t i s t i c a l l y s i g n i f i c a n t at the 10% leve l * * S t a t i s t i c a l l y s i g n i f i c a n t at the 5% leve l * * * S t a t i s t i c a l l y s i g n i f i c a n t at the 1% leve l Appendix 15: (Cont'd) (e) Coefficients of Determination for Minimum Monthly Temperatures Component Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. RW EW LW RD ED LD MND MXD • .004 .009 .003 .005 .007 .005 .001 .000 .004 .000 .050 .015 .020 .012 .004 . .010 .000 .004 .028 .048 .003 .014 .003 .021 .016 .006 .044 .000 * .121 .008 .053 .030 .024 .032 .003 .070 .082 .004 .028 .015 .107 .088 .039 .044 .001 .015 .000 .000 .001 .009 .013 .090 *** .390 .001 ** .215 .005 .006 ** .170 it it .001 . .170 .125* .016 .047 .123 .001 .004 .001. .004 .010 .021 .028 .012 .005 .000 .036 .006 .013 .261 *** .390 .093 .009 .083 .141 ** .165 (f) Regression Coefficients x 1 Standard Deviation i n Monthly Temperature Indices .004 ** .222 .076 .069 ** .212 RW EW ' LW RD , ED. LD MND MXD -0.74 -1.36 • 0.53 0.25 -0.16 0.16 0.04 0.00 0.70 -0.09 2.00 0.35 0.26 -0.26 0.17 -0.35 0.12 0.87 -1.61 -0.74 -0.12 -0.24 -0.12 -0.48 1.44 1.11 3.03 0.00 0.64* -0.24 0.56 -0.56 -1.80 -2.60 0.48 0.85 0.53 0.16 0.42 0.42 -3.70 -4.25 -1.80 0,65 0.05 -0.30 0.00 0.00 0.34 1.36 -1.02 -0.94 *** -1.21 -0.07 ** -1.17 -0.26 A it 0.88 -4.62 ** -0.36 -5.90 3.26* -1.18 it it 0.72 1.12 -0.05 0.13 -0.10 -0.16 0.26 0.35 -0.62 -0.41 0.82 0.23 1.72 -0.26 -0.23 *** -1.32 *** -2.27 3.44 0.86 -0.90 0.94* ** -1.48 -0.74 ** -4.33 -0.85 0.68 ** 1.65 * ** ***' S t a t i s t i c a l l y s i g n i f i c a n t at the 1.0% l e v e l S t a t i s t i c a l l y s i g n i f i c a n t at the 5% le v e l S t a t i s t i c a l l y s i g n i f i c a n t a t the 1% le v e l Appendix 15: (Cont'd) (g) Coefficients of Determination for Maximum Monthly Temperatures Component Oct. Nov. Dec. Jan. Feb. Mar. Apr. May June July Aug. Sept. RW * .134 .024 .008 .020 .001 .030 .088 .023 .057 .001 .003 .027 EW .093 .001 .023 .013 .000 .021' .110 .004- .051 .003 LW .055 ** .190 .031 .016 .027 .021 .005 .093 .009 .080 .008 ** .208 RD .054 .056 .096 .010 • .015 .004 ** .172 .024 ,.015 .033 .001 .055 ED .001 ** .174 . .010 .067 .080 .028 ** .213 .044 .002 .060 LD .029 .006 .026 .005 .002 .015 .001 .000 .007 .038 .017 .015 MND .017 .109 .007 .023 .022 .005 ** .242 .073 .004 MXD .019 .023 .009 .002 .005 .001 .000 .009 . .005 * .113 .026 .083 (h) Regression Coefficients x 1 Standard Deviation of Monthly Temperature Indices RW 3.80 1.74 0.98 1.59 0.40 -1.95 3.34 1.07 -2.65 . 0.35 -0.58 -1.83 EW 4.36 0.39 2.22 1.64 -0.07 -2.10 4.72 0.91 -3.22 -0.84 LW 2.17 *# 4.00 -1.60 1.17 1.53 -1.31 0.67 2.81 -0.85 2.62 -0.84 ** -4.18 RD -0.74 0.73 -0.98 -0.32 0.40 0.21 ** -1.33 0.48 0.38 0.59 0.08 -0.73 ED -0.03 ** 0.79 0.18 0.48 0.53 0.32 ** -0.89 0.38 . 0.09 .0.49 LD . -0.44 -0.17 -0.44 -0.16 0.13 -0.16 -0.09 0.00 -0.24 -0.49 -0.34 0.29 MND ' -.33 0.84 -0.18 0.37 0.40 0.16 ** -1.25 0.67 0.14 MXD 0.50 -0. 56 -0.36 -0.16 0.27 -0.10 0.04 -0.33 -0.24 -1.22 0.58 1.03 * S t a t i s t i c a l l y s i g n i f i c a n t at the 10% le v e l ** S t a t i s t i c a l l y s i g n i f i c a n t at the 5% le v e l *** S t a t i s t i c a l l y s i g n i f i c a n t at the 1% l e v e l Appendix 16: Climatic Summaries for Provenance T r i a l Sites, 1969-1974 Data Site . Jan. Feb. Mar. Apr. Month May June July Aug. Sept. Oct. Nov. Dec. • Precip. Nimpkish 284 208 221 147 66 66 66 41 • 109 175 300 • 251 mm Haney 300 236 254 147 ' 94 112 . 112 33 127 188 251 297 Molalla 462 234 267 193 102 71 20 28 119 145 333 429 Valley 302 145 168 86 36 25 8 8 48 84 239 317 Dorena 213 . 114 157 112 51 38 2 13 58 86 201 211 Ave. Temp. Nimpkish -0.9 2.0 3.7 6.6 10.6 13.7 . 16.2 •16.5 13.5 7.7 . 3.8 0.6 °c Haney 0.6 3.7 4.9 7.3 11.7 14.3 16.4 17.0 13.3 8.5 3.9 1.6 Molalla 1.8 4.5 -1.3 5.4 9.8 12.6 15.9 15.8 13.6 8.8 5.2 2.6 Valley 3.8 6.1 7.5 8.9 13.3 16.2 19.3 18.9 16.2 11.0 7.0 4.1 Dorena 3.8 6.0 7.0 8.2 12.3 15.4 18.5 18.4 15.2 10.2 6.9 4.4 Min. Temp. • Nimpkish -3.9 -1.2 -0.4 1.8 4.9 8.0 9.5 10.0 8 .1 3.3 1.2 -1.9 °C Haney• -1.7 0.7 ' 1.3- 3.0 6.7 9.5 11.2 11.6 8.7 4". 7 0.7 -0.4 Molalla -1.'4 0.8 0.3 0.9 4.4 7.2 9.1 9.0 8.0 4.0 1.8 -0.4 Valley 0.3 1.5 2.2 2.5 5.3 7.9 9.5 8.8 7.3 4.2 2.7 0.8 Dorena -0.3 0.8 1.8 2.3 5.5 8.5 9.9 9.4 6.7 3.7 3.0 0.4 Max. Temp. Nimpkish . 1.9 5.3 .7.8 11.3 16.3 19.3 22.8 23.0 18.9 11.9 6.3 3.1 °C Haney 2.9 6.8 ' 8.6 11.7 16.7 18.8 21.7 22.3 18.0 12.5 7.2 3.6 Molalla 4.9 8.3 8.3 9.9 . 15.0 18.0 22.7 22.5 19.2 13.5 8.6 5.5 Valley 7.2 10.7 12.8 15.4 21.2 24.3 29.1 29.1 25.1 17.4 11.0 7.3 Dorena 7.8 11.2 12.3 14.2 19.0 22.3 27.2 27.4 23.7 16.8 10.9 8.4 _ Appendix 17: Ring Component Summaries by Clones and Ring Groups Rings 1-5 Rings 1968- 72 Clone RW EW LW PLW RD ED LD MND MXD UNI RNG RW EW LW PLW RD ED LD MND MXD UNI RNG . 25 5:67 4.61 1.05 19.7 .371 .306 .657 .252 .816 .351 .563 7.63' 5.96 1.66 21.9 .384 .292 .730 .234 .927 .439 .692 28 5.15 3.90 1.25 26.2 .389 .303 .660 .245 .812 .357 .568 6.13 4.75 1.38 23.0 .373 .276 .705 .211 .871 .429 .659 36 6.10 5.10 0.99 16.4 .330 .276 .604 .203 .712 .328 .509 7.49 6.25 1.23 16.7 .318 .253 .646 .179 .768 .392 .589 55 5.90 4.34 1.55 24.8 .376 .288 .639 .223 .774 .350 .550 7.57 5.50 2.06 27.1 .375 .261 .684 .197 .841 .423 .644 59 5.83 4.77 1.05 18.1 .367 .305 .639 .230 .762 .334 .531 7.57 6.11 1.46 19.2 .341 . .260 -.680 .186 .827 .620 .660 60 4.49 3.51 0.98 23.2 .397 .327 .646 .269 .794 .320 .525 5.71 4.64 1.07 18.6 .369 .294 .690 .233 .857 ..396 .625 62 5.44 4.17 1.27 23.2 .396 .325 .632 .254 .801 .306 .547 7.23 5.33 1.89 26.5 .400 .296 .694 .216 .884 .398 .667 70 5.67 4.16 1.50 26.0 .391 .312 .616 .253 .738 .304 .484 6,81 5.13 1.67 25.2 .376 .281 .663 .216 .814 .302 .590 82 4.36 3.70 0.65 15.5 .346 .298 .615 .24 a .740 .317 .490 5.64 4.91 0.72 13.2 .317 .264 .661 .207 .805 .397 .598 83 5.56 4.52 1.04 19.3 .361 .302 .619 .239 .730 .317 .500 6.42 5.40 0.95 15.1 .330 .274 .647 .205 .792 .373 .507 Mean 5.42 4.28 1.13 21.04 .372 .304 .633 .242 .769 .328 .527 6.82 5.40 1.41 20.66 .358 .275 .680 .208 .839 .405 .630 Abbreviations and uni t s as shown i n F i g . 3.3. Appendix 1 8 : Ranges and Variance Components for Clonal Data Ring Component Rings 1 -5 Rings 1969-73 Min. Max. Mean 2 a C a 2 T Min. Max. Mean a2 C a 2 T RW 4.36 6.10 5.42 0.222 0.944 5.64 7.63 6.82 0.500 1.020 EW 3.51 5.10 4.28 0.158 0.678 4.64 6.25 5.40 0.230 0.716 LW 0.98 1.55 1.13 0.067 0.071 0.72 2.06 1.41 0.171 0.082 PLW 15.46 25.96 21.04 :11.88 14.97 13.21 27.10 20.66 22.22 9.96 RD .330 .397 .372 .000426 .000546 .317 .400 .358 .000806 .000444 ED .276 .327 .304 .000179 .000433 .253 .296 .275 .000195 .000359 LD . .604 .660 .633 .000306 .000397 .646 . .730 .680 .000650 .000495 MND .203 .269 .242 .000263 .000743 .179 .235 .208 .000259 .000484 MXD . .712 .816 .769 .001181 .001088 .768 .927 .839 .002085 .001315 UNI .303 .357 .328 .000224 .001117 .373 .439 .405 .000322 .001095 RNG .484 .568 .527 .000641 .002137 .587 .693 .630 .001118 .001846 2 . 0 = Variance component c , for clone e f f e c t Component abbreviations and units as shown i n F i g . 3.3. = Component f o r trees within clones Appendix 19(a): Results of Duncan's M u l t i p l e Range Tests on Clonal Ring Components - Rings 1-5 RW mm EW mm 1 mm PLVi % RD g/cm3 ED , 3 g/cm 82 4.36 60 3.51 82 0.65 | 82 15.5 36 .330 36 .'276 60 4.49 82 3.70 60 0.98 36 16.4 82 .346 55 .288 28 5.15 28 3.90 36 0.99 59. 18.1 83 .361 82 .298 62 5.44 70 4.16 83 1.04 83 19.3 59 .367 83 .302 83 5.56 62 4.17 25 1.05 25 19.7 25 .371 28 .303 25 5.67 55 4.34 59 1.05 62 23.2 55 .376 59 .305 70 5.67 83 4.52 28 1.25 60 23.2 28 .389 25 .306 59 5.83 25 4.61 62 1.27 28 24.2 70 .391 70 .312 55 5.90 59 4.77 70 1.50 55 24.8 62 .396 62 .325 36 6.10 36 5.10 55 1.55 70 26.0 60 .397 60 .327 5.42 4.28 1.13 21.0 .372 .304 Means LD , 3 g /cm MND g/cm MXD g /cm 3 UNI , 3 g/cm RNG . .3 g/cm 36 .604 36 .203 36 .712 70 .304 70 .484 82 .615 '55 .223 70 .738 62 .306 82 .490 70 .616 5 9 .230 83 .738 83 .317 83 .500 83 .619 83 .239 82 .740 82 .317 36 .509 62 .632 28 .245 59 .762 60 .320 60 .525 55 .639 82 .249 55 .774 36 .328 59 .531 59 .639 25 .252 60 . 7 9 4 59 . 3 3 4 62 . 5 4 7 60 .646 70 .253 62 .801 55 .350 55 .550 25 .657 62 .254 28 .812 25 .351 25 .563 28 .660 60 .269 25 .816 28 1357 28 .568 .633 .242 .769 .328 .527 V e r t i c a l l i n e s enclose means not s i g n i f i c a n t l y d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l . Appendix 19 (b): Results of Duncan's Multiple Range Tests on Clonal Ring Components - Rings 1969-73 Means RW mm 82 5.64 60 4.64 60 5.71 28 4.75 28 6.13 82 4.91 83 6.42 70 5.13 70 6.81 60 5.33 62 7.24 83 5.46 36 7.49 55 5.50 59 7.57 25 5.96 55 7.57 59 6.11 25 7.63 36 6.25 6.82 5.40 EW mm LW 82 83 60 36 28 59 25 28 62 55 mm 0.72 0.95 1.07 1.23 1.38 1.46 1.66 1.67 1.89 2.06 1.41 PLW 82 83 36 60 59 25 28 70 62 55 13 15 16 18 19 21 23.0 25.2 26.5 27.1 20.7 RD 82 36 83 59 60 28 55 70 25 62 g/cm~ .317 .318 .330 .341 .369 .373 .375 .376 .384 .400 .358 ED 36 59 55 82 83 28 70 25 60 62 g/cm" ,253 .260 .261 ,264 ,274 ,276 ,281 .292 .294 .296 .275 LD g/cm^ MND g/cm3 MXD g/cm 3 UNI , 3 g/cm RNG g/cm3 36 .646 36 .179 36 .768 83 .373 83 .587 83 .647 • 59 .186 83 .792 70 .382 36 .589 82 .661 55 .197 82 .805 36 .392 82 .598 70 .663 83 .205 70 .814 60 .396 70 .598 59 .680 82 .207 59 .827 82 .397 60 .625 55 .684 28 .211 55 .841 62 .398 59 .640 60 .690 70 .216 60 .857 59 .420 55 .644 62 .694 62 .216 28 .871 55 .423 28 .659 28 .705 60 .233 62 .884 28 .429 62 . 667 25 .730 1 25 .234 25 .927 1 25 .439 25 .693 .680 .208 .839 .405 . .630 V e r t i c a l l i n e s enclose means not s i g n i f i c a n t l y d i f f e r e n t at the 5% p r o b a b i l i t y l e v e l . - 238 -Appendix 20: Analysis of Variance Models for Clonal Data Rings 1-5 (mixed model) Source d.f. E.M.S. Clones 9 O 2 e + 10a 2 + 80a2 Ramets 70 a 2 e + Years 4 a 2 e + 2a 2 + YT 16a2c + 160a 2 Years x clones 36 a 2 e + 2a2 + YT 1 6 4 c Years x ramets 280 2 a e + 2a 2 YT Error 400 2 °e TOTAL 799 Rings 1969-73 (random model) Source d.f. . E.M.S. Clones 9 a 2 e + 2 4 T + 1 6 4 c + i o < + 80° C Ramets 70 a 2 e + 2 a 2 T + 10a 2 Years 4 2 a e + 2a 2 + °YT 1 6a 2 c + 160a 2 Years x clones 36 a 2 e + 2 4 T + 1 6 4 c Years x ramets 280 a 2 e + 2 4 T Error 400 a 2 e TOTAL 799 Appendix 21: Co r r e l a t i o n Matrices f o r Clonal Ring Components RINGS 1 - 5 _ 10 observations t o t a l „ Std. Dev. 9 degrees of Name Mean Name Mean Std. Dev. freedom 5.41700 RW 0.583782 LD 0.632700 0.188447E-01 EW 4.27800 0.492472 MND 0.241700 0.188152E-01 LW 1.13300 0.266794 MXD 0.763700 0.361971E-01 PLW 21.0400 3.70801 UNI 0.328400 0.190158E-01 RD 0.372400 0.222421E-01 RNG 0.526700 0.3O1774E-01 Correlation Matrix Variable RW EW LW PLW RD ED LD MND MXD UNI RNG RW 1.0000 EW 0.8907** 1.0000 LW 0.5367 0.0945 1.0000 PLW 0.0595 •0.3872 0.8513** .1.0000 RD •0.2184 •0.5596 0.5651 0.8662** 1.0000 ED •0.4765 -0.5935 0.0636 0.4538 0.8126** 1.0000 LD •0.1507 0.2826 0.1978 0.4108 0.5960 0.3925 1.0000 UND 0.6964* 0.7801** -0.0748 0.3920 0.7135* 0.8953** 0.4178 1.0000 MXD -0.2264 0.3763 0.2073 0.4642 0.6946* 0.5592 0.9293** 0.5376 1.0000 UNI 0.2209 0.1903 0.1292 0.0349 -0.0658 -0.4111 0.6766* -0.2981 0.4677 1.0000 RNG 0.1596 0.0326 0.2938 0.3149 0.3910 0.1146 0.8564** 0.0450 0.8544** 0.7474** 1.0000 RINGS 1969-73 Name Mean Std. Dev. Name Mean Std. Dev. 10 observations t o t a l RW 6.82000 0.792941 LD 0.680000 0.265246E-01 9 degrees of EW 5.40400 0.564864 MND 0.208400 0.179270E-01 freedom LW 1.40900 0.424538 MXD 0.838600 0.474861E-01 PLW 20.6500 4.85140 UNI 0.404900 0.215945E-01 RD 0.358300 0.293487E-01 RNG 0.630000 0.36633E-01 '. ED 0.275100 0.154736E-01 Correlation Matrix Variable RW EW LW PLW RD ED LD MND MXD UNI RNG KW 1.0000 EW 0.8561** 1.0000 LW 0.7253" 0.2651 1.0000 PLW 0.4660 0.0540 0.9416** 1.0000 RD 0.2307 0.2501 0.7645** 0.8707** 1.0000 LU 0.2044 0.4333 0.1955 0.3562 0.741U** 1.0000 LD • 0.1791 0.0994 0.4703 0.5095 0.7419** 0.5931 1.0000 MND 0.3702 0.5507 0.0420 0.2186 0.6223* 0.9099** 0.G210* 1.0000 MXD 0.152U 0.150U 0.4IJU2 0.5115 0.U12U** 0.7259* 0.9751** 0.7142* 1.0000 UNI 0.3073 0.1857 0.4427 0.3761 0.3914 0.0250 0.019C** 0.1276 0.6940* 1.0000 RNG 0.3731 0.0700 0.6063' 0.5094 0.7462** 0.4901 0.9564** 0.4426 0.9437** 0.8325** 1.0000 Appendix 22 C o r r e l a t i o n Matrix f o r Phenological and Growth Ring Variables variable RW EW LW PLW RD ED LD MND RW 1.0000 EW 0.9417 1.0000 LW 0.7822 0.5271 1.0000 PLW 0.4014 0.0789 0.8676 1.0000 RD 0.3803 0.0857 0.0011 0.8833 1.0000 ED -0.0561 -0.2044 0.2362 0.3599 0.6884 1.0000 LD 0.6628 0.4999 0.7469 0.6234 0.7142 0.1935 1.0000 MND -0.1123 -0.2303 0.1419 0.2G88 0.5984 0.9432 0.1527 1.0000 MXD 0.7176 0.5586 0.7765 0.5875 0.7154 0.3062 0.9044 0.2594 UNI 0.6032 0.5780 0.6542 0.4825 0.4439 -0.1996 0.9227 -0.2189 RNG 0.7G42 0.6242 0.7732 0.5516 0.6162 0.1226 0.9032 0.0630 X 1 0.8350 0.7975 0.6323 0.3140 0.3241 0.0198 0.5370 -0.0415 x 2 -0.1506 -0.1066 -0.1833 -1339 -0.0367 0.0384 • 0.1349 0.0685 x 3 0.8477 0.8044 0.6518 0.3265 0.3272 0.0170 0.5166 -0.0467 x 4 0.8290 0.8352 • 0.5473 0.1806 0.2173 -0.0765 0.5395 -0.1167 x 5 -0.1382 -0.0630 -0.2298 -0.1829 -0.2523 -0.2347 -0.1271 -0.1795 x 6 0.8354 0.8210 0.5892 0.2.:U6 .' 0.2807 -0.0034 0.5601 0.0553 x 7 0.5856 0.4868 0.5781 0.4388 0.4040 0.1813 0.3639 0.1044 x 8 -0.0148 -0.0503 0.0524 0.0554 0.2466 0.3127 0.3005 0.2842 x 9 0.5875 0.4993 0.5656 0.4235 0.3497 0.1106 0.3060 0.0396 x 10 -0.1929 -0.3147 0.0960 0.3410 0.2380 0.2160 -0.0820 0.1926 x 11 0.4453 0.4104 0.3670 0.2688 0.1631 -0.2180 0.3794 -0.2477 x 12 0.4963 0.5623 0.2139 -0.0287 0.0717 0.0102 0.3258 -0.0330 x 13 0.4450 0.4100 0.3671 0.2692 0.1636 -0.2172 0.3791 -0.2475 x 14 -0.3893 -0.4020 -0.2381 -0.0294 -0.2614 -0.2721 -0.4515 -0.1925 x 15 -0.3879 -0.3289 -0.3691 -0.2077 -0.3982 -0.3259 -0.5014 -0.2454 x 16 -0.5094 -0.2829 -0.7629 -0.6976 -0.6676 -0.3381 -0.4398 -0.2307 x 17 0.4815 0.2551 0.7435 0.6925 0.6862 0.3547 0.5001 0.2525 MXD RNG 1.0000 .0.7833 1.0000 0.9802 0.8544 1.0000 0.5578 0.5275 0.5842 0.0335 0.1178 0.0201 0.5553 0.5036 0.5828 0.S720 0.5682 0.6146 -0.2472 • -0.0352 -0.2196 0.6285 0.5602 0.6606 0.3606 0.2909 0.3502 0.3214 0.1755 0.2745 0.2946 0.2611 0.2953 -0.1399. -0.1675 -0.1847 0.2075 0.4637 0.2636 0.2802 0.3210 0.2956 0.2073 0.4631 0.2633 -0.5326 -0.3433 -0.5111 -0.6497 -0.3720 -0.6216 -0.6221 -0.3082 -0.5953 0.6676 0.3615 0.6376 Variable x 6 x .7 . x 8 x 9 x 10 x 1 1.0000 X 2 0.0255 1.0000 2 3 0.9905' -0.1054 1.0000 X 4 0.9338 -0.0288 0.9334 X 5 0.0072 0.6198 -0.0699 X 6 0.9003 -0.1617 0.9181 X 7 0.8017 0.1058 0.7803 X 8 0.0211 0.4394 -0.0410 X 9 0.8030 0.0194 0.7941 X 10 -0.141.0 0.0990 -0.1615 X 11 0.4453 0.1781 0.4184 X 12 0.6147 0.2599 0.5811 X 13 0.444G 0.1766 0.4178 X 14 -0.4789 -0.3302 -0.4365 X 15 -0.4050 -0.2665 -0.3671 X 16 -0.4311 0.2883 -0.4743 X 17 0.4113 -0.0568 0.4248 X 11 1.0000 0.0047 0.9658 0.5351 -0.0385 0.5477 -0.4704 0.3831 0.5832 0.3825 -0.5231 -0.4571 -0.4414 0.4213 1.0000 -0.2446 0.0026 -0.4326 0.0845 -o.oino 0.4932 0.3S24 0.4924 0.2646 0.3190 0.2910 -0.1522 1.0000 0.5143 0.0942 0.5019 -0.CM3 0.2247 0.4712 0.2241 -0.5921 -0.5402 -0.5137 0.4662 1.0000 0.1186 0.9808 0.4G16 0.4072 0.4803 0.4067 -0.2644 -0.2012 -0.2835 0.2687 1.0000 -0.0744 0.1343 -0.3600 -0.1048 -0.3G07 -0.6823 -0.6713 -0.0019 0.1090 1.0000 0.4332 0.4709 0.4995 0.47G6 • -0.1320 -0.0670 -0.2776 0.2401 1.0000 0.0701. -0.1227 0.0706 0.3055 0.2693 0.1341 -0.1330 1.0000 0.6642 1.0000 -0.0090 0.0992 -0.0099 0.0560 Variable x 12 x 13 x 15 x 16 X 17 12 13 14 15 16 17 1.0000 0.6638 -0.4921 -0.2664 -0.0312 0.0693 1.0000 -0.0081 0.1000 -0.0100 . 0.0558 1.0000 0.8824 0 ''841 -0.3579 1.0000 0.5137 -0.5951 1.0000 -0.9672 1.C0OO o Appendix 23: C o r r e l a t i o n matrices f o r Progeny Ring Components a) Rings 1 - 5 Name Mean Std. Dev. Name Mean Std. Dev. RW 404.766 87.4524 MtID 0.277234 0.242001E-01 EW 275.719 82.9057 MXD 0.638828 0.452189E-01 LW 129.094 - 31.8552 PLW 37.8141 8.54972 RD 0.447328 0.307350E-01 UNI 0.246969 0.377006E-01 ED 0.345859 0.198605E-01 RNG 0.411469 0.438793E-01 LD • 0.592969 0.296327E-01 Correlation Matrix Varia-ble RW EW LW RD ED LD MND MXD PLW UNI RNG RW 1.0000 EW 0.9313 1.0000 LW 0.3226 -0.0444 1.0000 RD -0.6104 -0.7985 0.4012 1.0000 ED 0.1219 0.0688 0.1582 0.1423 1.0000 LD -0.5093 -0.5982 0.1566 0.6741 -0.1270 1.0000 MND -0.4804 -0.5182 0.0289 0.6228 0.2219 0.2038 1.0000 MXD -0.5053 -0.6152 0.2122 0.7064 -0.0321 0.9603 0.3215 1.0000 PLW -0.6317 -0.8360 0.4401 0.9279 -0.0662 ' 0.5599 0.5743 0.5952 1.0000 UNI -0.4659 -0.5072 0.0380 0.4554 -0.6250 0.8536 0.0427 0.7721 0.4747 1.0000 RNG -0.2548 -0.3478 0.2048 0.3846 -0.1538 0.8763 -0.2210 0.8523 0.2963 0.7706 1.0000 b) Rings 1970-74 Namo - Moan Std. Dev. Name Mean Std. Dev. RW 278.453 104.511 MND 0.237172 0.2792^ 4E-01 EW 157.703 80.6973 MXD 0.758031 0.662333E-01 LW 120.766 32.3928 PLW 47.4312 9.69511 RD 0.462766 6.484826E-C UNI 0.345766 0.45&669E-01 ED 0.296172 0.251639E-01 RNG 0.520766 0.670117E-01 LD 0.642063 0.356147E-01 Correlation Matrix. Varia-ble RW • EW LW RD ED LD MND MXD PLW UNI RNG RW 1.0000 EW 0.9710 1.0000 LW 0.8026 0.6368' 1.0000 RD -0.60B9 -0.7577 -0.2663 1.0000 ED -0.2359 -0.1781 -0.3183 0.4656 1.0000 LO -0.379B -0.4069 -0.0UU7 0.CU57 -0.10U5 1.0000 MND -0.5909 -0.5G63 -0.4036 0.70'JO 0.7974 0.2319 1.0000 MXD -0.2077 -0.2596 -0.0196 9.4473 0.0089 0.7089 0.1831 1.0000 I'l.W -0.7 202 -0.U3O5 -0.2191) 0.9399 0.2951 O.SUOl 0.51140 0.3219 1.0000 UNI -0.1653 -0.2804 0.1686 0.2770 -0.6348 0.8360 -0.2560 0.5465 0.2887 1.0000 RNG 0.0421 -0.0201 0.1886 •0.1462 -0.3253 0.6035 -0.2371 0.9116 0.0746 0.6486 1.0000 PUBLICATIONS Cown, D.J. 1973. Resin pockets: t h e i r occurrence and formation i n New Zealand f o r e s t s . N.Z. J . For. 18(2): 233-251. 1973. E f f e c t s of severe thinning and pruning treatments on the i n t r i n s i c wood pro-p e r t i e s of young r a d i a t a pine. N.Z. J. For. S c i . 3(3): 379-389. 1974. Wood density of r a d i a t a pine: i t s v a r i a t i o n and manipulation. N.Z. J. For. 19(1): 84-92. 1974. P h y s i c a l properties of Corsican pine grown i n New Zealand. N.Z. J . For. S c i . 4(1): 76-93. 1974. Comparison of the e f f e c t of two t h i n -ning regimes on some wood properties of r a d i a t a pine. N.Z. J . For. S c i . 4(3): 540-551. 1975. V a r i a t i o n i n tracheid dimensions i n the stem of a 26-year-old r a d i a t a pine tree. Appita 28(4): 237-245. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0075135/manifest

Comment

Related Items