UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A study of the phenotypic and genotypic variation of 545 single tree progenies of 38 provenances of the… Falkenhagen, Emil R. 1974

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

Item Metadata

Download

Media
831-UBC_1974_A1 F34.pdf [ 11.47MB ]
Metadata
JSON: 831-1.0093236.json
JSON-LD: 831-1.0093236-ld.json
RDF/XML (Pretty): 831-1.0093236-rdf.xml
RDF/JSON: 831-1.0093236-rdf.json
Turtle: 831-1.0093236-turtle.txt
N-Triples: 831-1.0093236-rdf-ntriples.txt
Original Record: 831-1.0093236-source.json
Full Text
831-1.0093236-fulltext.txt
Citation
831-1.0093236.ris

Full Text

A S T U D Y O F T H E P H E N O T Y P I C A N D G E N O T Y P I C V A R I A T I O N O F 545 S I N G L E T R E E P R O G E N I E S O F 38 P R O V E N A N C E S O F T H E 1970 I. U . F . R . O . S I T K A S P R U C E ( P I C E A S I T C H E N S I S ( B O N G . ) C A R R . ) C O L L E C T I O N by E M I L R. F A L K E N H A G E N A g r o n o m i c a l and F o r e s t Engineer (Gembloux, Belgium) A THESIS S U B M I T T E D IN P A R T I A L F U L F I L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R O F P H I L O S O P H Y in the F a c u l t y of F o r e s t r y We accept this thesis as conforming to the r e q u i r e d standard T H E 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 M a y , 1974 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree t h a t 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 study. 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 copying o f t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head o f my Department or by h i s r e p r e s e n t a t i v e s . I t i s understood t h a t c o p y i n g or p u b l i c a t i o n of 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 allowed without my w r i t t e n p e r m i s s i o n . Department o f Forestry. The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8 , Canada Date May 1st. 1974. A B S T R A C T D u r i n g the 1970 f a l l , the I.U.F.R.O. S e c t i o n 22 "Work-i n g group on p r o c u r e m e n t of seed f o r provenance r e s e a r c h " o r g a n i z e d an e x p e d i t i o n to c o l l e c t S i t k a s p r u c e cones f r o m B r i t i s h C o l u m b i a and A l a s k a . The l o c a t i o n s of the 39 p r o v e n a n c e s range f r o m 48. 38 ^ to 58". 37°latitude N and f r o m 121? 93° to 134''; 58°longitude W. The e l e v a t i o n v a r i e s f r o m 0 to 2, 200 feet above sea l e v e l . In m o s t ca s e s , the c o l l e c t i o n s w e r e made f r o m 15 t r e e s i n each l o c a t i o n . The s i n g l e t r e e p rogeny c o l l e c t i o n of 557 t r e e s r e p r e s e n t i n g 39 l o c a t i o n s c o n s t i t u t e the m a t e r i a l of t h i s t h e s i s . In a f i r s t p a r t , seed and cone m o r p h o l o g y w ere s t u d i e d on a s i n g l e t r e e b a s i s . T e n cones p er progeny w e r e r a n d o m l y s e l e c t e d and the length of e a c h cone m e a s u r e d to the n e a r e s t mm. F i v e r a n d o m l y s e l e c t e d seeds f r o m each t r e e w e r e mounted on a s p e c i a l sheet, and seed length, seed wid t h , wing l e n g t h and wing w i d t h w ere m e a s u r e d to the n e a r e s t 0. 01 mm. N e s t e d a n a l y s e s of v a r i a n c e and DUNCAN's m u l t i p l e range t e s t s f o r a l l the c h a r a c t e r i s t i c s s t u d i e d have been p e r -f o r m e d u s i n g f i v e s u b r e g i o n s . No d e f i n i t e c l a s s i f i c a t i o n of the p r o v e n a n c e s was p o s s i b l e b y u s i n g u n i v a r i a t e anova p r o c e d u r e s . A s i m p l e c o r r e l a t i o n m a t r i x has been c a l c u l a t e d between a l l the t r a i t s s t u d i e d and longitude, l a t i t u d e and a l t i t u d e of the p l a c e of o r i g i n of the p r o v e n a n c e s , u s i n g the provenance means. M u l t i p l e r e g r e s s i o n a n a l y s e s have been u s e d f o r i n -v e s t i g a t i n g t h i s c o r r e l a t i o n m a t r i x . The percentage of v a r i a t i o n (ii ) accounted for by the geographical co-ordinates v a r i e s between 10. 2 % and 43. 6%. Us i n g the seed and cone t r a i t s studied, a co m p a r i s o n of s e v e r a l m u l t i v a r i a t e s t a t i s t i c a l analyses which could be used for c l a s s i -f i c a t i o n purposes has been attempted. The s o - c a l l e d canonical a n a l y s i s , d i s c r i m i n a n t function a n a l y s i s and p r i n c i p a l component a n a l y s i s have been compared and applied for c l a s s i f y i n g the provenances. The sub-regions a l r e a d y used were analyzed separately. D e n d r o g r a m s were also constructed and analyzed. Advantages and disadvantages of each m u l t i v a r i a t e method have been d i s c u s s e d . It was found that the d i s -c r i m i n a n t function a n a l y s i s , its a s s o c i a t e d g e n e r a l i z e d distances of M A H A L A N O B I S and dendrograms p r o v i d e d the most r a t i o n a l c l a s s i f i c a -tion of the provenances. In a second part, the genetic v a r i a b i l i t y of 545 Sitka spruce single tree progenie s was studied in a n u r s e r y test during 1 971 a n d l 9 7 2 . A total of 545 single tree progenies grouped into 38 prov-enances was sown in A p r i l , 1971, using a ra n d o m i z e d complete block design with four r e p l i c a t i o n s and 24 seedlings per r e p l i c a t i o n or 96 seedlings per progeny. The seeds were pla c e d in the c a v i t i e s of s t y r o -b l o c k s using the method developed by the P a c i f i c F o r e s t R e s e a r c h Ce n t r e in co-operation with the B. C. F o r e s t S e r v i c e and they have been treated by the most recent n u r s e r y methods, i n the new B. C. F. S. n u r s e r y at S u r r e y (B.C.). G e r m i n a t i o n rate, bud set, length of the epi c o t y l and s u r v i v a l after the f i r s t growing season were a s s e s s e d i n (iii) 1971. The "seedlings w e r e t r a n s p l a n t e d i n p l a i n s o i l seedbeds i n May, 1972, to a d i s t a n c e of 6" to 6", each progeny b e i n g kept s e p a r a t e w h i l e r e s p e c t i n g the same s t a t i s t i c a l d e s i g n as i n 1971. B u d b u r s t , bud set, c o l o u r of the n e e d l e s and t o t a l height a f t e r the second g r o w i n g season, w e r e a s s e s s e d i n 1972. T h e r e was a c l i n a l v a r i a t i o n i n bud b u r s t , bud set, c o l o u r of the n e e d l e s and e p i c o t y l length. B u d b u r s t was n e g a t i v e l y c o r r e l a t e d w i t h longitude (r = -0. 50) and p o s i t i v e l y c o r r e l a t e d w i t h a l t i t u d e (r = 0.42). B u d set a p p e a r e d under s t r i c t g e netic c o n t r o l as i n d i c a t e d by the second e s t i m a t i o n of thi s t r a i t , at the end of the second g r o w i n g s e a s o n (with l a t i t u d e : r = 0. 88). L a t i t u d e and a l t i t u d e of the see d s o u r c e s e x p l a i n e d 6 5 % of the t o t a l v a r i a t i o n i n e p i c o t y l length. T o t a l height a f t e r the sec o n d g r o w i n g s e a s o n showed the same r e l a t i o n -ships as e p i c o t y l length. G e n e r a l equations f o r components of v a r i a n c e f o r u n b a l a n c e d data w e r e o r i g i n a l l y c a l c u l a t e d f o r a n e s t e d - c r o s s e d m o d e l . Components of v a r i a n c e and t h e i r s t a n d a r d e r r o r w e r e c a l c u l a t e d f o r e p i c o t y l l e n g t h and t o t a l height a f t e r the second g r o w i n g season. Depending on the sub-r e g i o n s , the genetic v a r i a n c e among p r o v e n a n c e s i s g e n e r a l l y l a r g e r than the t r e e to t r e e genetic v a r i a t i o n . The n a r r o w sense h e r i t a b i l i t y , on an i n d i v i d u a l b a s i s , and i t s s t a n d a r d e r r o r , f o r t o t a l height a f t e r the second g r o w i n g season, w e r e e s t i m a t e d o n a s u b r e g i o n b a s i s . H e r i t a b i l i t y was found g e n e r a l l y to be c l o s e to 0. 10, i n d i c a t i n g low g e n e r a l c o m b i n i n g ;ab'ility. (iv) The r e l a t i o n s h i p s between the seed, cone and seedling t r a i t s m e a s u r e d were studied. M u l t i p l e r e g r e s s i o n a n a l y s i s showed that a higher p r o p o r t i o n of the v a r i a t i o n of the seedling t r a i t s was accounted for by the geographical coordinates of the provenances than by the cone and seed t r a i t s studied. V a r i a t i o n in f o l i a r m a c r o - and m i c r o - nutrients of 10 Sitka spruce provenances was studied, but no geographical pattern of v a r i a t i o n detected in K, Ca, Mg, Fe, Mn, Zn, P, and N needle contents. Only K showed some provenance to provenance va r i a t i o n . P o s s i b l e p h y s i o l o g i c a l explanations for this absence of v a r i a t i o n are d i s c u s s e d . (v) T A B L E O F C O N T E N T S Page A B S T R A C T . i i T A B L E O F C O N T E N T S v i LIST O F T A B L E S i x LIST O F F I G U R E S , M A P S A N D P H O T O S x i i A C K N O W L E D G E M E N T S xv I N T R O D U C T I O N 1 1. Methodologica l outline 1 2. Autecology of Si tka spruce . 6 P A R T I. S I T K A S P R U C E AS G E N E C O L O G I C A L M A T E R I A L Chapter 1 A r e v i e w of the genetic l i t e ra ture perta ining to S i tka spruce 13 1.1 S i l v i c u l t u r a l studies . 13 1. Z P h y s i o l o g i c a l studies 21 1.3 Cytogenet ical studies 22 1.4 V a r i a t i o n i n wood proper t ies 23 " 2 Genecology def ines ; i ts re la t ionships with tree breeding 25 " 3 O r i g i n and c h a r a c t e r i s t i c s of the m a t e r i a l studied 27 " 4 C l i m a t o l o g y of the area of o r i g i n of the Sitka spruce provenances 33 P A R T II. S E E D A N D C O N E M O R P H O L O G Y AS P A R T O F T H E  B I O S Y S T E M A T I C S O F S I T K A S P R U C E Chapter 1 ' F o r e s t tree b iosys temat i cs defined 44 " 2 Seed and cone morphology . Univar ia te analyses of var iance and m u l t i v a r i a t e c o r r e l a t i o n and r e g r e s s i o n analyses 48 2. 1 Univar ia te analyses of var iance 48 (vi T A B L E O F C O N T E N T S (Continued) Page 2.2 Mult i p l e c o r r e l a t i o n and r e g r e s s i o n a n a l y s e s . R e l a t i o n s h i p s with the geo-g r a p h i c a l coordinates of the place of o r i g i n of the provenances studied 63 Chapter 3 M u l t i v a r i a t e s t a t i s t i c a l methods. A com-parative r eview of the mathematical theories behind the techniques f r o m a biologist's point of view 71 3. 1 M A N O V A 73 3. 2 P r i n c i p a l component analysis 74 3.3 F a c t o r a n a l y s i s • 79 3.4 D i s c r i m i n a n t function a n a l y s i s (DF) and canonical a n a l y s i s (CA) 80 3.5 R e s u l t s and conclusions 88 3. 5. 1 C A a c c o r d i n g to S E A L (1964) 90 3.5.2 D e n d r o g r a m an a l y s i s 102 3.5.3 Stepwise d i s c r i m i n a n t a n a l y s i s ..... I l l 3.5.4 P r i n c i p a l component a n a l y s i s (PCA) . 114 3.5.5 Conclusions 116 P A R T III. S T U D Y O F T H E S E E D L I N G S T A G E Chapter 1 The philosophy of the n u r s e r y test 121 " 2 M a t e r i a l and methods used i n the n u r s e r y 127 2. 1 Seedling c h a r a c t e r i s t i c s studied i n the n u r s e r y . Why those p a r t i c u l a r c h a r a c t e r s ? 127 2.2 N u r s e r y treatments 131 2.3 Measur e m e n t techniques 138 2.4 Methods used in the analyses 143 2.4.1 M u l t i p l e c o r r e l a t i o n and r e g r e s s i o n analyses 143 2.4.2 A n a l y s i s of va r i a n c e models 143 2.4.2. 1 M o d e l for m a x i m u m epic o t y l length 144 " 3 R e s u l t s and conclusions 157 3.1 G e r m i n a t i o n rate 157 3.2 P h e n o l o g i c a l observations 159 v i i T A B L E O F C O N T E N T S (Continued) Page 3, 3 Needle c o l o u r 179 3.4 G r o w t h c h a r a c t e r i s t i c s 179 Ch a p t e r 4 The h e r i t a b i l i t y p r o b l e m : i t s a p p l i c a t i o n to one c h a r a c t e r i s t i c : t o t a l growth a f t e r two g r o w i n g seasons 199 " 5 R e l a t i o n s h i p s B e t ween Seed Cone and See d l i n g T r a i t s S t u d i e d 202 " 6 G e n e t i c v a r i a t i o n i n f o l i a r m a c r o - and m i c r o - n u t r i e n t s of ten s e l e c t e d one y e a r o l d S i t k a s p r u c e p r o v e n a n c e s 206 6. 1 I n t r o d u c t i o n 206 6. 2 M a t e r i a l and methods 207 6.3 R e s u l t s 209 P A R T IV - S U M M A R Y A N D C O N C L U S I O N S 217 B I B L I O G R A P H Y 223 v i i i LIST O F T A B L E S Table Page I Geographica l coordinates of the Si tka spruce provenances studied and number of single tree progenies avai lable per provenance 28 II V a r i a n c e s and means of the seed and cone t ra i t s m e a s u r e d 50 III F values of the nested A N O V A p e r f o r m e d on the different t ra i ts m e a s u r e d 53 IV Number of provenances out of a total of 39 with heterogeneity of var iances for 4 seed c h a r a c t e r i s t i c s and cone length 55 V Nested analyses of var iance for Region 1 58 V I Nes ted analyses of var iance for Region 2 59 VII Nested analyses of var iance for Region 3 60 VIII N e s t e d analyses of var iance for Region 4 61 IX Nested analyses of var iance for Region 5 62 X Components of var iance i n percentages for the seed t ra i t s studied and cone length 64 X I C o r r e l a t i o n m a t r i x between the seed c h a r a c t e r i s t i c s and cone length and the geographical coordinates of the place of o r i g i n of the 39 provenances studied 65 XII Canonica l functions for the four regions studied. Coeff ic ients of the o r i g i n a l var iab les 91 XIII M a t r i x of the distances of M A H A L A N O B I S between the provenances of Region 2 109 X I V M a t r i x of the distances of M A H A L A N O B I S between the provenances of Region 3 109 ix LIST O F T A B L E S (Continued) Table Page X V Idem for Region 4 . . . 109 X V I Idem for Region 5 110 X V I I Number of trees c l a s s i f i e d into the different provenances according to their provenance of o r i g i n 112 XVII I Loadings of the components on the f ive var iab les studied 116 X I X C o r r e l a t i o n coeff ic ients for height r e g r e s s e d on latitude of o r i g i n 130 X X Expectat ions of sums of squares for the total ly unbalanced c r o s s e d model 149 X X I Expectat ions of mean squares for p a r t i a l l y unbalanced c r o s s e d model 153 X X I I M e a n number of cavi t ies with at least one germinant and coeff ic ients of v a r i a t i o n (C. V . ) for block A and the b locks B , C and D pooled 160 XXII I Geographica l coordinates of the provenances and average values of the n u r s e r y t ra i t s studied 166 X X I V C o r r e l a t i o n matr ix , between the t ra i t s studied and the geographical coordinates of the place of o r i g i n . 168 X X V V a r i a n c e components of some important sources of v a r i a t i o n and their standard e r r o r for m a x i m u m epicotyl length 188 X X V I C o r r e l a t i o n m a t r i x between the geographical coordinates of the provenances and H E T , SD, SE and C . V 189 X X V I I Second year growth var iance components and their standard e r r o r 196 x LIST O F T A B L E S (Continued) T a b l e Page XXVIII N a r r o w sense h e r i t a b i l i t y for total height after the second growing season, with their e r r o r - by region . Z01 X X I X C o r r e l a t i o n m a t r i x between seed, cone and seedling t r a i t s studied Z03 X X X F o l i a g e a n a l y s i s of 10 one-year o ld dormant Sitka spruce provenances Z l l X X X I Idem (second part) 212 x i LIST O F F I G U R E S , M A P S A N D P H O T O S F i g u r e Page 1 The range of Sitka spruce 7 2 Monthly v a r i a t i o n in p r e c i p i t a t i o n and temperature for two stations of the Lower M a i n l a n d 35 3 Monthly v a r i a t i o n i n for two stations of the Northwest coast of B. C 36 4 Idem for Queen Charlotte Islands 38 5 Idem for one station of Southern Vancouver Island • • • ' 40 6 Growing degree-days accumulated for three stations of B. C 42 7 The four measurements taken for studying seed morphology of Sitka spruce 49 8 Re l a t i o n s h i p s between the c o r r e l a t i o n s of seed length with seed width and latitude of the place of o r i g i n of the provenances 70 9 Can o n i c a l a n a l y s i s for Region 2. Pl o t t i n g made using the f i r s t two canonical axes 93 10 Idem, but plotting made using the f i r s t c anonical axis and the t h i r d axis 94 11 Idem, but plotting made using the f i r s t and the fourth axis 95 12 Ca n o n i c a l a n a l y s i s for Region 3. Pl o t t i n g made usin g the f i r s t two canonical axes 97 13 Idem for Region 4 99 14 Idem for Region 5 101 1 5 D e n d r o g r a m of the provenances of Region 2 . . . . 104 x i i LIST O F F I G U R E S , M A P S A N D P H O T O S (Continued) F i g u r e Page 16 D e n d r o g r a m of the provenances of Region 3 105 17 D e n d r o g r a m of the provenances of Region 4 106 18 D e n d r o g r a m of the provenances of Region 5 ..... 107 19 P r i n c i p a l component an a l y s i s of the provenances of Region 5. P l o t t i n g made using the f i r s t two p r i n c i p a l components 115 20 Drawing showing how the styro-foam containers were a s s e m b l e d and seeded in the n u r s e r y .... 133 21 Schematic r e p r e s e n t a t i o n of the bud b u r s t stages as o b s e r v e d i n A p r i l , 1972 140 22 G e r m i n a t i o n rates plotted against time for some selected provenances 158 23 Bud bur s t evolution of different Sitka spruce provenances growing in Germany in the sp r i n g of 1956 162 24 Re l a t i o n s h i p s between the one-day method to a s s e s s bud set and B U R L E Y ' s method 163 25 Relationships between bud set of the Sitka spruce provenances and latitude of place of o r i g i n at two dates, in 1971 169 26 E v o l u t i o n of day-length f r o m M a r c h to October for two latitudes, i n hours, 1/10 174 27 Re l a t i o n s h i p between m aximum epic o t y l length and latitude of place of o r i g i n of the provenances 181 28 Mean epi c o t y l length of the provenances plotted against their v a r i a n c e , i n vT a double l o g a r i t h m i c paper 187 29 Re l a t i o n s h i p between total height after the second growing season and latitude of place of o r i g i n of the provenances 190 x i i i LIST O F F I G U R E S , M A P S AND P H O T O S (Continued) F i g u r e Page 30 Mean total height of the provenances plotted against their v a r i a n c e s , i n a double l o g a r i t h m i c paper 193 Maps 1 Lo c a t i o n s of the Sitka spruce provenances 30 2 Loc a t i o n s of the Sitka spruce provenances 31 Photos 1 View of the container stage 132 2 View of the transplantation of the (1 -l- 0) Sitka spruce seedlings 135 3 View of separate progenies, identified by using cedar poles 137 4 V a r i a t i o n i n needle colour 178 5 Seedling a b n o r m a l i t i e s 198 xiv A C K N O W L E D G E M E N T S The author s i n c e r e l y wishes to thank the numerous persons who contr ibuted to this thes is : D r . O . S Z I K L A I , P r o f e s s o r at the F a c u l t y of F o r e s t r y who supplied the seeds and constantly sup-ported the author 's r e s e a r c h i n many different ways ; D r . S. N A S H , P r o f e s s o r i n the Mathemat i cs Department who d i scussed many of the p r o b l e m s that we encountered, the p r o g r a m m e r s Mesdames L . C O W D E L L , K . H E J J A S and E . P R I N C Z who patiently helped us in p r o g r a m m i n g the s ta t i s t i ca l techniques used, D r . A . K O Z A K and D r . P h . H A D D O C K , P r o f e s s o r s at the F a c u l t y of F o r e s t r y who p r o v i -ded us wi th helpful re ferences and guidances; D r . R. P E T E R S O N , P r o f e s s o r i n the A n i m a l Science Department and D r . J . W O R R A L L , P r o f e s s o r at the F a c u l t y of F o r e s t r y also gave us very useful advice . The author i s grateful to the B R I T I S H C O L U M B I A F O R E S T S E R V I C E R E F O R E S T A T I O N DIVISION that took care of the seedl ings . The Surrey N u r s e r y staff, p a r t i c u l a r l y M e s s r s . J I M S W E E T E N and H A N S E L I A S have been indispensable . We also thank. M e s s r s . R . S C H M I D T and A . F R A S E R f r o m the B . C . F . S . R E S E A R C H DIVISION who prov ided f inanc ia l support for the n u r s e r y study. The latter k i n d l y also prov ided invaluable help in m e a s u r i n g the seedling height after the second growing season. (xv) I N T R O D U C T I O N 1. Methodological outline Sitka spruce ( P i c e a s i t c h e n s i s (BONG. ) CARR. ) is the la r g e s t of a l l the spruces. In B r i t i s h Columbia, in volume cut, Sitka spruce ranks seventh in importance with m o re than 40, 000, 000 cubic feet cut in 1971, ( B r i t i s h C olumbia F o r e s t S e r v i c e Annual Report, 1971). It is well known that the Queen Charlotte Islands of B r i t i s h C o l u m b i a are one of the areas where Sitka spruce y i e l d s its m a x i m u m production. C o m m e r c i a l stands of Sitka spruce may occur at 1, 000 feet elevation. However, 3/4 of c o m m e r c i a l stands are estimated to be within 2-1/2 m i l e s of tide water, a fact which would fa c i l i t a t e the c o m m e r c i a l exploitation of this spruce. Single trees may contain up to 15, 000 b o a r d feet of timber. F u r t h e r m o r e , Sitka spruce wood i s the most valuable of a l l spruce woods and finds a multitude of uses (building construction, a i r c r a f t , pulp and paper, etc. ). Sitka spruce seems, therefore, to be a v e r y valuable m a t e r i a l to genetically improve for future extensive plantations in western N o r t h A m e r i c a . It n a t u r a l l y grows in a narrow s t r i p along the P a c i f i c coast f r o m northwestern C a l i f o r n i a to no r t h e r n A l a s k a . Its range is over 1, 800 m i l e s long and c h a r a c t e r i z e d by a pronounced oceanic climate. z A s an exotic species, pure stands of Sitka spruce have proven to be v e r y manageable and exceptionally productive, e s p e c i a l l y in B r i t a i n . Recently, international i n t e r e s t in this tree species has i n c r e a s e d and, during the f a l l of 1970, on behalf of IZ European countries, the I.U.F.R.O., Section 2Z "Working group on procurement of seed for provenance r e s e a r c h " o r g a n i z e d an expedition to c o l l e c t Sitka spruce cones f r o m B r i t i s h C olumbia and A l a s k a . Twenty cones f r o m each of the 557 trees r e p r e s e n t i n g 39 locations were pr o v i d e d to P r o f e s s o r O. S z i k l a i , F a c u l t y of F o r e s t r y , U n i v e r s i t y of B r i t i s h C olumbia. T h i s single tree progeny c o l l e c t i o n constituted the m a t e r i a l with which the author commenced the present study. The objectives of the present study were to d e s c r i b e the v a r i a t i o n pattern presented by the 557 parent trees and their progenies which were growing in a n u r s e r y environment, near Vancouver, B.C. It was decided to keep the single tree progenies separate and to start a simultaneous provenance and progeny test i n the n u r s e r y . Many c h a r a c t e r s have been studied on both the parent trees and their progenies, in many tree species. M uch d i s c u s s i o n o c c u r r e d on the pos s i b l e patterns of v a r i a t i o n of the tr a i t s studied (c l i n a l , ecotypic or mixed), and how best to d e s c r i b e these patterns, but only r e s u l t e d in v e r y general explanations ( G A L O U X and G A L K E N H A G E N , 1965). In provenance and open pollinated progeny r e s e a r c h in tree species, many b a s i c p r o b l e m s r e m a i n to be solved, both p r a c t i c a l and th e o r e t i c a l . D i v i s i v e techniques such as anovas and techniques of mul t ip le c o r r e l a t i o n and r e g r e s s i o n analyses were cons idered c o m p l e -mentary and it was decided to use both techniques to express in object-ive fo rms a l l the facets of the complex v a r i a b i l i t y of the m a t e r i a l s tudied. Bes ide univariate anova models and mul t ip le r e g r e s s i o n and c o r r e l a t i o n techniques, m u l t i v a r i a t e s ta t i s t i ca l methods are also used; however , it was decided to solve some bas ic problems re la ted to the choice of the best methods in m u l t i v a r i a t e s ta t i s t i cs , p r i o r to any definite c l a s s i f i c a t i o n of the biotas which might exist in the Sitka spruce natura l populat ions. Many techniques, more or less e m p i r i c a l , have been used to c l a s s i f y the na tura l populations of a tree spec ies . F o r instance, once a great number of t ra i t s has been assessed on both the parent m a t e r i a l and their progenies growing in a n u r s e r y test, c o r r e l a t i o n m a t r i c e s are calculated and used to choose the best t ra i t s which could lead to the ' .distinction of biotas (or races) through regrouping the provenances in l a r g e r ent i t ies . In a m o r e r igorous way, these c o r r e l a t i o n m a t r i c e s are also used i n m u l t i v a r i a t e s ta t i s t i ca l analyses i n order to f ind i n t e r -pretable "components" or to c luster the provenances . The f i r s t method which leads to the " s u m of dif ferences method" developed by W R I G H T and B U L L in 1962 has been c r i t i c i z e d by S T E R N (1964). L A N G L E T (1959) has also pointed out the danger of fo l lowing b l i n d l y a computer generated c l a s s i f i c a t i o n of provenances . A great number of sophist icated methods of n u m e r i c a l c l a s s i f i c a t i o n have been proposed and appl ied without paying too much attention to their re la t ive s i m i -l a r i t y , e f f ic iency, mathemat ica l theory, e t c . , ( D A G N E L I E , 1966). Thus , there is an urgent need to compare the techniques used in m u l t i -var ia te m o r p h o m e t r i e s . A s part of the b iosys temat ics of Si tka spruce , i t was decided to measure 5 reproduct ive cone and seed c h a r a c t e r s , on a single tree b a s i s , of the parent t rees , in order to compare the fol low ing m u l t i v a r i a t e s ta t i s t i ca l methods which were avai lable to us : p r i n c i -pal component a n a l y s i s , d i s c r i m i n a n t function analys is and the so -ca l l e canonical a n a l y s i s . M u l t i p l e c o r r e l a t i o n analyses and anovas were extensively used in order to get an accurate pic ture of the v a r i a b i l i t y of the 5 t ra i ts s tudied. T h i s p r o b l e m solving approach was chosen instead of assess ing a number of charac ters and then to apply a given m u l t i v a r i a t e method. Only 7 t ra i t s having adaptative values were studied on progenies growing i n the n u r s e r y ; among these t ra i t s , only height growth was m e a s u r e d on a single tree progeny b a s i s . The genetic v a r i a t i o n in f o l i a r m a c r o - and m i c r o - nutrient contents was also studie because it was r a r e l y attempted. Genecology te l ls us that the natura l populations of a tree species are l o c a l l y adapted to their environment of o r i g i n , at least to some degree. It was decided to subdivide the m a t e r i a l into 5 regions , at least f a i r l y homogeneous eco log ica l ly and to analyze these 5 regions separately except for the c o r r e l a t i o n and r e g r e s s i o n analyses which were p e r f o r m e d over a l l the provenances . Thus , the heterogeneity of 5 the var iances is l i k e l y to be reduced and safe mul t ip le c o m p a r i s o n tests of the means can be p e r f o r m e d on a reduced set of provenances : genera l ly the number of provenances is less than 12 except for reg ion 5 which contains a m a x i m u m of 16 provenances . V a r i a n c e components es t imat ion thus makes more sense on region bas is than, on a global b a s i s , by using 39 provenances and 10, 000 degrees of f r e e d o m . The objectives of the present study were thus two-pronged: on the one hand, a p r o b l e m solving approach was used instead of b l i n d l y applying some given methods, a f r e s h look at the prob lems was attempted and many questions were r a i s e d ; on the other hand, a large n u r s e r y test of m o r e than 500 single tree progenies of Sitka spruce represent ing 39 provenances was in i t ia ted and taken care of, for three y e a r s . B a s i c measurements were made on the progenies and different hypotheses re la ted to the v a r i a t i o n pattern of S i tka spruce were e x p r e s s e d . In 1974, the n u r s e r y test y ie lded severa l f i e ld tests for long t e r m r e s e a r c h purposes which were establ ished at strategic locations in B . C . 6 2. Autecology of S i tka spruce Sitka spruce (P icea s i tchensis ( B O N G . ) C A R R . ) is a member of the C a s i c t a group of the genus P i c e a . The C a s i c t a group is c h a r a c t e r i z e d by unequally quadrangular flattened leaves, with wide upper and lower surfaces bear ing the stomata. The cones scales are thin and f lexible ( D A L L I M O R E and J A C K S O N , 1966). Its range ( F i g . 1) is over 1,800 m i l e s long and extends along the coast of A l a s k a , B r i t i s h C o l u m b i a , Washington, Oregon and southwards to nor thern C a l i f o r n i a . The width of the range v a r i e s bet-ween a few m i l e s i n C a l i f o r n i a to about 130 m i l e s i n A l a s k a . S i tka spruce reaches m a x i m u m development on the O l y m p i c P e n i n s u l a of Washington and the Queen Charlot te Islands ( B r i t i s h C o l u m b i a ) . A c c o r d i n g to D A U B E N M I R E (1957), P i c e a s i tchensis would have roughly occupied its present range before the glaciat ion p e r i o d , but the range f o r m e r l y extended southward along the C a l i f o r n i a coast, be-yond the San F r a n c i s c o bay. A f t e r the g lac ia t ions , it r e - a p p e a r e d n e a r l y a l l over its present range. T h i s would indicate a pers is tence on nuna-taks w e l l scattered along the coast f r o m Puget Sound to Juneau. H o w -ever , it i s d i f f i cul t to jus t i fy this theory of r e f u g i a scattered f r o m A l a s k a to Washington state, because Sitka spruce is present ly adapted to a pronounced oceanic c l i m a t e . E a r l i e r authors ( H U L T E N , 1937; H A L L I -D A Y and B R O W N , 1 943 i n B U R L E Y , 1 965) have be sides proposed that S i tka spruce inhabited a southern " n i c h e " during the glaciations and that it fo l lowed the re t rea t ing ice to A l a s k a . 7 F i g . 1 . The r a n g e o f S i t k a s p r u c e ( FOWELS, 1 9 6 5 ) 8 In s l ight ly w a r m e r pre-plestocene t imes , Si tka spruce probably grew on the Aleut ians and was l inked to P . j ezoens is , to which it i s s i m i l a r s y s t e m a t i c a l l y . S i tka spruce is a member of the homogeneous northwest A m e r i c a group whose o r i g i n is recent (WRIGHT, 1955). P . s i tchensis is unique among the spruces in being a l o w -land conifer even in the southern port ion of its range. S i tka spruce is t y p i c a l l y associated with the "fog be l t " of the P a c i f i c coast of N o r t h A m e r i c a . It grows in a c l imate c h a r a c t e r i z e d by equable temperature , high prec ip i ta t ion f r o m 66 inches to 150 inches of annual prec ip i ta t ion , long vegetation p e r i o d (at least 130 days), c loudiness averaging 200 days per year and m i l d winters ( F O W E L L S , 1965). K R A J I N A (1969) has defined the b iogeoc l imat ic zones of B r i t i s h C o l u m b i a . In B r i t i s h C o l u m b i a , Sitka spruce thr ives p r i m a r i l y i n two b iogeoc l imat i c zones: (1) The coastal D o u g l a s - f i r zone, the dr ies t part of the m e s o t h e r m a l zone of B r i t i s h C o l u m b i a , which has been subdivided into two subzones, (2) The coastal western hemlock zone i t se l f also subdivided into two subzones. Soi ls are s i m i l a r throughout the d i s t r ibut ion area : they are high in organic matter with a p H of 4. 0 to 5. 7. D A Y (1957) has suggested that Si tka spruce is favoured, i n c o m p a r i s o n wi th its asso-ciated species , by s o i l conditions of high nutrient status and that it w i l l assume a dominant pos i t ion when supply of aerated water and nutrients is adequate. The frequent occurrence of S i tka spruce as a seashore species m a y have a nutr i t ive reason, for such seaside si tuation must rece ive a r e l a t i v e l y heavy deposit of sea s p r a y . T h i s idea has been recent ly defended again by K R A J I N A (1969). A s noticed e a r l i e r , Si tka spruce is a low elevat ion species except in Southern A l a s k a . . In B r i t i s h Columbia 1 ; it i s se ldom found at elevations exceeding 1, 000 feet. It is a major species in only two forest types of the pac i f i c west according to the A m e r i c a n c l a s s i f i -cation of forest types used by F O W E L L S . (1965). It i s a p r o l i f i c seed producer wi th good cone c rop , every three or four y e a r s and p a r t i a l ones in between. Seed w i l l germinate on a lmost any k i n d of seedbed i f m o i s t u r e i s abundant. It is a fast growing tree that r e a d i l y dominates its assoc ia tes : wes tern hemlock (Tsuga heterophyl la (RAF.) SARGJ and western r e d cedar (Thuja p l i ca ta DONN.). However , Si tka spruce is les shade tolerant than western hemlock, and is able to take up a dominant pos i t ion only when f e r t i l i t y of the site suf f ic ient ly favours its growth, ( D A Y , 1957). P u r e Si tka spruce stands are essent ia l ly subc l imax , the m o r e tolerant wes tern hemlock eventually r e p l a c i n g S i tka spruce . M a t u r e trees have straight boles with swol len and but-t r e s s e d bases , are as high as 285 feet and possess diameter up to 10 feet or m o r e . Two types have been dis t inguished: the bottomland type to and the slope type ( C A R Y (1922) in K A R L S B E R G , 1961). Si tka spruce is cons idered sha l low-rooted and susceptible to wind throw. It is subject to decay after in jury and is attacked by s e v e r a l insects that m a y k i l l or damage large volumes of t i m b e r . The most important pest h inder ing the extensive use of Si tka spruce , on the P a c i f i c Coast , is the Sitka spruce w e e v i l (P issodes s i t chens is ) . T h i s beetle only attacks trees 2-8 inches i n diameter and 5 to 25 feet t a l l , k i l l i n g or i n j u r i n g the t e r m i n a l shoot. F o r k e d or crooked stems often r e s u l t f r o m weevi l ing ( W R I G H T , I960). Only i n A l a s k a and on the Queen Charlot te Is lands, i s S i tka spruce total ly free f r o m w e e v i l attack. In B r i t i s h C o l u m b i a , K R A J I N A (1969) descr ibes Sitka spruce as a "hygrohylophi le to subhygrohylophi le , submontane to montane, a c i c u l i l i g n o u s , subeutrophophytic, stenotrophophytic spec ies " o c c u r r i n g i n three b i o g e o c l i m a t i c zones. Its shade tolerance would be low to n i l , depending on the habitat. It would r e q u i r e high quantities of avai lable c a l c i u m and m a g n e s i u m , requirements which would expla in its tolerance to ocean s p r a y . However , the seaside situation of Si tka spruce might also be explained by some res i s tance to sea salt deposit on its foliage as w e l l as to wind '. ( P e r s o n a l communicat ion of D r . P h . H A D D O C K ) . . P H E L P S has recent ly rev iewed the l i t e ra ture dealing with Si tka spruce ; p a r t i c u l a r l y in teres t ing are the volume and y i e l d tables which are bas i c ins truments i n managing Si tka spruce stands i n the P a c i f i c Nor thwest . Of interes t to the genetic ist , is the p o s s i b i l i t y of 11 propagating Sitka spruce by cuttings or by graft ing ( P H E L P S , 1973). A c c o r d i n g to F L E T C H E R and F A U L K N E R (1972), Sitka spruce is an excellent exotic species wherever introduced, provided i ts e c o l o g i c a l preferences are respected. It has surpassed i n volume product ion many of the native and introduced tree species i n B e l g i u m , B r i t a i n , D e n m a r k , F r a n c e , I re land, New Zealand, G e r m a n y and N o r w a y . The most noteworthy success is its introduct ion in Great B r i t a i n , where it can y i e l d up to 25 cubic meters per ha per y e a r . In that country, it a l ready occupied 265, 800 ha in 1970 and the propor t ion of Sitka spruce , according to the same authors, w i l l probably continue i n c r e a s i n g in B r i t a i n . To s u m m a r i z e the most important eco logica l facts , Si tka spruce is the larges t of a l l the spruces . It grows n a t u r a l l y i n a n a r r o w s t r i p along the P a c i f i c coast f r o m northwestern C a l i f o r n i a to ~no:rthern A l a s k a , which is c h a r a c t e r i z e d by i ts pronounced oceanic c l i m a t e . S i tka spruce is p r o l i f i c , i ts seed germinat ing in a lmost any k ind of seedbed, provided humidi ty i s p l e n t i f u l . It is shade tolerant to intolerant according to the habitat and grows r a p i d l y , but m o s t l y remains a secondary species m i x e d with wes tern hemlock and western r e d cedar . Its regenerat ion is best after stand opening. It i s , therefore , a sub-c l i m a x species , probably close to the " o p p o r t u n i s t i c " type ( S T E R N , 1963), which is defined as having an i r r e g u l a r reproduct ion and random dif ferent iat ion at the population l e v e l . It i s , however, d i f f i cul t to forecast " a p r i o r i " the genetic 12 p l a s t i c i t y of Si tka spruce , because its eco log ica l preferences are f a i r l y n a r r o w , at least in c o m p a r i s o n with other western tree species such as D o u g l a s - f i r . A s an exotic species , pure stands of S i tka spruce have proven to be v e r y manageable and except ional ly product ive , notably i n B r i t a i n , and there is no doubt that, i n suitably managed plantations, S i tka spruce could become an important conifer i n the P a c i f i c Northwest i f the w e e v i l was under c o n t r o l . P A R T I S I T K A S P R U C E AS G E N E C O L O G I C A L M A T E R I A L 13 C H A P T E R I: A R E V I E W O F T H E G E N E T I C L I T E R A T U R E P E R T A I N I N G TO S I T K A S P R U C E 1. 1 S i l v i c u l t u r a l s t u d i e s B U R L E Y (1965, ) has r e v i e w e d the s c i e n t i f i c l i t e r a t u r e b d e a l i n g w i t h the genetic v a r i a t i o n of S i t k a s p r u c e p u b l i s h e d b e f o r e the e a r l y 1960's. M o s t provenance r e s e a r c h on S i t k a s p r u c e has been done i n E u r o p e . B U R L E Y (1965^) c i t e s one N o r t h A m e r i c a n r e p o r t on p l a n t a -tions of P. s i t c h e n s i s and the h y b r i d P. X l u t z i i f r o m A l a s k a and D e n m a r k on the e a s t coast of Canada and one p r o v enance t r i a l i n New Z e a l a n d . C L A R K (1965) has s t u d i e d some cone and seed c h a r a c t e r i s -t i c s of 35 t r e e s o r i g i n a t i n g f r o m seven d i f f e r e n t l o c a t i o n s . He has found c o n s i d e r a b l e w i t h i n and between l o c a t i o n s v a r i a b i l i t y f o r a l l cone and seed t r a i t s studied. However, he has not r e l a t e d the v a r i a b i l i t y o b s e r v e d to the p l a c e of o r i g i n o f the p r o v e n a n c e s s t u d i e d . The p r o v e n -ances have been t r a n s p l a n t e d on the Queen C h a r l o t t e I s l a n d s and a r e b e i n g s t u d i e d by the F o r e s t company R A Y O N I E R . In E u r o p e , i n t e r e s t has c e n t e r e d on volume p r o d u c t i o n , height g r o w t h and f r o s t r e s i s t a n c e , of g e n e r a l l y a l i m i t e d n u m ber of p r o v e n a n c e s . R e s u l t s a r e m o r e o r l e s s c o n t r a d i c t o r y as f a r as what co u l d be the b e s t p r o v e n a n c e s f o r p l a n t i n g i n g i v e n c o n d i t i o n s i s c o n c e r n e d . R O B A K (1962) d i s c u s s e d the winter s u r v i v a l of 1~0 and 2M) Si tka spruce provenances f r o m A l a s k a and B r i t i s h C o l u m b i a i n N o r w e g i a n n u r s e r i e s . K A R L - B E R G (1961) rev iewed the southern Scandinavian exper iments wi th Si tka spruce provenances . Seed suitable for Danish conditions would have to be co l lec ted f r o m areas as far south as B r i t i s h C o l u m b i a and Washington state. Super ior resul t s have been obtained f r o m seed f r o m Tongass (Lat . 53°33 ) . In some exper iments , however , n o r t h e r l y s t ra ins suffered extensive damage due to late spr ing f ros t ; in other, the Washington o r i g i n was p r e f e r r e d . . K A R L B E R G (1961) concludes his r e v i e w by pointing out that the P a c i f i c coast has a v e r y compl ica ted physiography and that i t is important to know the p r e c i s e locat ion of the provenance and i f it came f r o m poor f o r m t rees , growing nearby shores and r i v e r s . S C H O B E R (1962) has presented extensive resu l t s of r e s e a r c h on Si tka spruce provenances . He studied ten Sitka spruce provenances coming f r o m close to the entire range: f r o m 56° to 41°50 L a t . N , i n two f i e l d tests near Hann. Munden (Germany) (Gahrenberg plantat ions, at the t ime of his study: 28 and 32 years o ld) . F r o s t damage and frequency of forked stems i n c r e a s e d with decreas ing latitude of place of o r i g i n of the provenances . F l u s h i n g rate and late f ros t damage were c l e a r l y c o r r e l a t e d . S C H O B E R used a 4 step scale to est imate bud burs t every 3-4 days, during the f lushing p e r i o d . The A l a s k a provenance was the latest to f lush , the Queen Charlot te one the e a r l i e s t . However , the difference i n f lushing date was s m a l l : 3 to 8 days . The length of the f lushing p e r i o d v a r i e d f r o m 33 to 46 days bet-ween provenances . The provenances d id not keep their rank during the p r o c e s s of bud burst and there was no unique re la t ionship between latitude and budiburs t . T r e e to tree , wi thin provenance, v a r i a t i o n i n bud burs t was l a r g e . Wood product ion v a r i e d f r o m provenance to provenance. Complete r e c o r d s of height, diameter and volume of wood above 7 c m . diameter showed that the Queen Charlot te Islands is best, the C a l i f o r n i a one of the poorest , in one test; i n the other test, 'the Quinault provenance proved the best . A n opt imum appears to be f o r m e d by the provenances f r o m south B r i t i s h C o l u m b i a and Washington state; nor thward the y i e l d decreased p r o g r e s s i v e l y , southward abrupt ly . The author concluded his studies by pointing out the necess i ty to match the c l i m a t i c conditions of the place of o r i g i n with those of the place of in t roduct ion . H i s observat ions also suggested that the vegetation p e r i o d and its number of accumulated day-degrees over 5 0 ° F would be important in forecas t ing the success of a given provenance. A L D H O U S (1962) has b r i e f l y studied the ear ly growth of 12 p r o v -enances of Si tka spruce col lec ted between 61° and 47° L a t . N , growing i n B r i t a i n . There was a c l i n a l v a r i a t i o n i n height growth. Height growth decreased wi th lat i tude. The southern provenances were the latest to set their bud. . 'The-t ime of height growth cessat ion v a r i e d up to 3 months . In genera l , therefore , the European provenance studies indicate 16 that southern sources of S i tka spruce prove m o r e susceptible to f ros t damage than sources f r o m B r i t i s h C o l u m b i a and A l a s k a and thus, unproduct ive . B U R L E Y has studied many aspects of the genetic v a r i a t i o n of Sitka spruce , using a m a t e r i a l cover ing the entire range of the tree species , f r o m L a t . 41° to 60° and f r o m L o n g . 122° to 152° , and f r o m altitude 0 to 1, 300 feet ( B U R L E Y , 1965 a , 1965 c , 1965 d , 1966 a , 1966 b , 1966 c ) . Us ing bulk co l lec t ion of 30 provenances, seed weight v a r i a -b i l i t y was studied. There was no s ignif icant re la t ionship with lat i tude, but a trend for nor thern provenances to have heavier seeds. Dif ferent explanations were proposed. F o r the most heavy seed sources , there was a re la t ionsh ip between embryo length and seed dry weight, seed volume, etc . In the case of the l ighter seed sources , none of these r e -lat ionships were s ignif icant at the 5% probabi l i ty l e v e l . There was no s ignif icant re la t ionship wi th altitude or lat i tude. However , it was quite poss ib le that nor thern provenances have a lower germinat ion capaci ty . R i g i d contro l of co l lec t ion and storage treatment would be n e c e s s a r y for p r e c i s e determinat ion of these r e l a t i o n s h i p s . C o l d , wet treatment, co ld soaking, a l ternat ing temperatures , c h e m i c a l treatment (gibberel l ic acid) have been c o m p a r e d . G e r m i n a t i o n rate was i m p r o v e d by cold soaking. Cotyledon number and length were studied in growth chamber : the cotyledon number v a r i e d between 5 and 7 and was not c o r r e l a t e d with latitude or seed weight. There was a s ignif icant re la t ionship between cotyledon length and seed weight. The genetic v a r i a t i o n i n seedling development of 47 p r o v -enances of Si tka spruce was studied in different environments : na tura l and a r t i f i c i a l . The morphology of bud format ion and f lushing has been studied i n a n u r s e r y at New Haven, Conn. ( U . S . A . ) . T h r e e stages i n bud development have been dis t inguished. D u r i n g the growing season, the apex of the t e r m i n a l bud i s not v i s i b l e and i s surrounded by needles i n close s p i r a l : this stage was ca l l ed the type I bud by B U R L E Y . . The type II bud was c h a r a c t e r i z e d by the fact that the last scales have been in i t ia ted and v i r t u a l l y no height growth o c c u r s . The type III bud was c h a r a c t e r i z e d by the facts that the outer scales turn brown and r e s i n o u s ; a l l needle p r i m o r d i a were then f o r m e d , but i n southernpprovenances, some needles could be in i t ia ted dur ing the f i r s t f lush of spr ing growth. No changes occur u n t i l s p r i n g when there is bud swel l ing and loss of the dark res inous colour of the bud scales and needle elongation. However , bud development i n Sitka spruce , as in many tree species , i s a continuous process with no w e l l - d e f i n e d stages. The types of bud dis t inguished by B U R L E Y are , therefore , somewhat a r b i t r a r y as w i l l be d i s c u s s e d l a t e r . It is also important to examine i f there is a re la t ionship between the number of p r i m o r d i a i n the t e r m i n a l bud, and its diameter and height growth. These charac ters were invest igated and it was found that the number of scales and p r i m o r d i a both contributed to bud diameter and that bud size and apex length or diameter were c o r r e l a t e d . . However , 18 bud diameter was not re la ted s igni f i cant ly to latitude of seed o r i g i n , but i n c r e a s e d with seedling height (r = 0.608). The number of days f r o m July 1, 1963 to bud format ion of type II decreased with increas ing latitude of seed o r i g i n (r = -0 .86Z) . The seedlings f r o m nor thern sources thus f o r m e d their buds before the southern provenances ( B U R L E Y , 1966 a ) . Sixteen provenances were studied under n o r m a l and exten-ded photoperiod (20 hours) . The author concluded that the time r e q u i r e d to f o r m buds (type II) decreased s igni f icant ly with i n c r e a s i n g latitude, but that under extended photoperiod, no buds developed unt i l the a r t i f i c i a l l ight was r e m o v e d and then bud format ion was much fas ter . The onset of dormancy was act ivated by an adequate photoperiod, but it could also be hastened by r e l a t i v e l y high temperature . It is important to note that B U R L E Y has found a posi t ive re la t ionship between latitude and time to change f r o m bud type II to type III. T h i s latter re la t ionship could, there-fore , e l iminate the previous negative re la t ionship found on the bas is of type II bud f o r m a t i o n . L o c a l temperature reg imes of the place of o r i g i n were be l i eved to expla in the indiv idua l reac t ion of the provenances studied. No re la t ionship between f lushing rate as m e a s u r e d by B U R L E Y and latitude of seed o r i g i n has been found, which c o n f i r m s S C H O B E R ' S observat ions . Seedling height decreased with increas ing latitude of seed o r i g i n , but r e g r e s s i o n slopes v a r i e d substant ial ly f r o m one environment to the other, apparently indicat ing genotype by environment in terac t ions . The s ize of the f i r s t t e r m i n a l bud was dependent on seedling height. L a r g e buds produced greater shoots than s m a l l buds, but 8% only of the v a r i a t i o n i n extension growth was accounted for by previous seedling height, under natura l photoperiod. Both wi th in and between provenances , the ta l ler seedlings had m o r e l a t e r a l branches than s m a l l e r seedlings and presumably would m a i n t a i n or increase their height d i f f e r e n t i a l . D a i l y growth rate was m e a s u r e d using a cathetometer on two seedlings per provenance. No re la t ionship was found between the different growth rates and latitude or seed weight of the seed sources studied. The author s u m m a r i z e d his developmental studies on Sitka spruce by suggesting that f lushing was contro l led by temperature and that i n t r a s p e c i f i c v a r i a t i o n i n this t ra i t o c c u r r e d that was associated not with lat i tude, but wi th temperature r e g i m e . Two weeks separated the ex-t reme provenances at any stage of f lushing and three months in dates of bud f o r m a t i o n . Height growth was a function of latitude and bud f o r m a -t ion . Photoper iod and temperature m o d i f i e d this re la t ionship with photo-p e r i o d i n c r e a s i n g height growth by delaying bud f o r m a t i o n . The pattern of adaptative v a r i a t i o n was? discontinuous. There was a strong i n t e r -act ion between genotype and environment . Co lours of the hypocotyl and needle glaucousness of 47 seed sources were qual i ta t ively assessed i n a contro l led environmenta l r o o m , dur ing the f i r s t year of growth. Co lour intensity i n c r e a s e d with lat i tude, the nor thern provenances developing r e d hypocotyl c o l o u r a -tions before the southern ones. The t ime of appearance of b l u e - g r e y colour v a r i e d w i d e l y , the A l a s k a provenances showing it one month before the most southern seed s o u r c e s . • D A U B E N M L R E (1968) studied the m o r p h o l o g i c a l v a r i a b i l i t y of adult trees in ten na tura l populations of Sitka spruce . The author c l a i m e d that there is a c l i n a l v a r i a t i o n i n cone s ize and s te r ig ina angle which decreased f r o m south to n o r t h . P o s s i b l e i n t r o g r e s s i o n s with P . glauca of some populations of the Skeena r i v e r watershed is d i s c u s s e d . Length width rat io of the cone scales could be used to differentiate i n -sular f r o m main land populat ions. The very p a r t i c u l a r sampling p r o c e -dure used by the author r e q u i r e s cautious interpreta t ion of his data. R O C H E (1969) has repor ted i n t r o g r e s s i v e h y b r i d i z a t i o n between the four species of spruce growing i n B r i t i s h C o l u m b i a , on the bas is of cone scale morphology . H i s invest igat ion of 150 juveni le spruce populations genetic v a r i a b i l i t y i s not applicable for studying Si tka spruce genecology because his r e s e a r c h was based on a pooled m i x t u r e of provenances or ig ina t ing f r o m different spruce complexes . L A C A Z E (1970) has studied the growth behaviour and f lushing rate of 24 Si tka spruce c o m m e r c i a l provenances , at the n u r s e r y stage, i n F r a n c e . The s o - c a l l e d Washington provenances were ranked the best i n growth. They f lushed the latest , too. It is w e l l known that Sitka spruce h y b r i d i z e s , in natura l condit ions, with s y m p a t r i c spruces such as P . glauca (yie lding the h y b r i d P . X l u t z i i ) , (WRIGHT, 1955). 21 P . s i tchensis has been success fu l ly c r o s s e d with P . abies and P . o m o r i k a ( W R I G H T , 1955). It has been i m p l i c a t e d i n a cur ious h y b r i d with Tsuga h e t e r o p h i l l a : X Tsuga hooker iana ( V A N C A M P O -D U P L A N and G A U S S E N , 1948). F L E T C H E R and F A U L K N E R (1972) have publ ished a detai led breeding and select ion plan for the genetic improvement of Si tka spruce i n Great B r i t a i n . Th is v e r y ambit ious plan covers every aspect of tree breeding : provenance studies, p l u s - t r e e se lect ion, tree banks , cone and seed s t imula t ion , etc. S A M U E L e t a l . (1972) have success fu l ly done a complete d i a l l e l c r o s s among s ix Si tka spruce t r e e s . The eight charac ters studied on one-year o ld seedlings were divided into two groups : those concerned wi th tree f o r m (stem stra ightness , dormant bud number and branch angle), and those re la ted to the vigour of the tree (height, dry weight, branch number and branch length). The tree f o r m c h a r a c t e r i s t i c s were predominant ly inher i ted i n an additive manner , but dominance effects were also opera t ive . F o r the vigour t r a i t s , a compl icated pattern of inheri tance appeared: the additive effects were s t i l l l a r g e , but dominant and, to a l e s s e r extent, m a t e r n a l effects were present . C o m p l i c a t e d gene interact ions could be present , but the heterozygous nature of the m a t e r i a l did not c l a r i f y the i s s u e s . 1. 2 P h y s i o l o g i c a l studies A D D I S O N (1966) grew Si tka spruce seedl ings in sand 22 cul tures with nutrient defic ient so lut ions . He studied s ix m a c r o -elements in seedlings or ig inat ing f r o m seeds of four locations and f r o m three trees per loca t ion . Seedlings were grown i n a contro l led environment . Signif icant di f ferences exis ted between treatments and between provenances for a l l the v a r i a b l e s studied: root length, shoot growth, root c o l l a r d iameter , etc. . In most cases , the response of the t ra i t s studied depended on the provenances . A lack of ni trogen was found to have the most de t r imenta l effect. Sul fur , magnes ium and c a l -c i u m def ic iencies had the least effect. F l u s h i n g rate was c o r r e l a t e d w i t h latitude of seed o r i g i n with the most southerly provenances f lushing the e a r l i e s t , an observat ion which contradicts B U R L E Y ' s f indings . Genetic v a r i a t i o n i n photoperiodic response has-been demon-strated by V A A R T A J A (1959) for Sitka spruce,, for growth t ra i t s and bud set. B R I X (1972) has studied the d r y matter product ion , stem height, b a s a l d iameter , etc. of two seed sources of S i tka spruce , under different temperature r e g i m e s and l ight in tens i t ies , i n c o m p a r i s o n with white spruce . Sitka spruce was m o r e shade tolerant than white spruce . Stem diameter growth was greatest i n the 18° to 24°C range for Sitka spruce . D r y matter product ion was m a x i m u m with constant day-night t empera tures . Responses v a r i e d with seed s o u r c e s . 1. 3 Cytogenet ical studies Ten Si tka spruce seed sources were used for chromosome study. Treatment with 1% colchic ine for five hours was adopted. The h a p l o i d number of c h r o m o s o m e s i s IZ. M o s t c h r o m o s o m e s w e r e m e t a o r s u b - m e t a c e n t r i c . T o t a l h a p l o i d c o m p l e m e n t le n g t h i n c r e a s e d w i t h l a t i t u d e (r = 0. 576). N u c l e a r volume was a l s o c o r r e l a t e d w i t h l a t i t u d e . T h e s e r e l a t i o n s h i p s c o u l d be i n t e r p r e t e d as an i n d i c a t i o n of some v a r i a -t i o n i n absolute k a r y o t y p e s t r u c t u r e o r a v a r i a t i o n i n r e a c t i o n to c o l -c h i c i n e . The k a r y o t y p e i n d i c a t e d three h e t e r o b r a c h i a l c h r o m o s o m e s and d i f f e r e d f r o m the k a r y o t y p e of other s p e c i e s of s p r u c e by the a r m lengths of two s p e c i f i c c h r o m o s o m e s ( B U R L E Y , 1965 c). M I K S C H E (1971) has u s e d b i o c h e m i c a l , F e u l g e n double w.ave-length c y t o p h o t o m e t r y and m i c r o d e n s i t o m e t r y to study D N A p e r c e l l v a r i a t i o n among seven seed s o u r c e s of P. s i t c h e n s i s . D N A e s t i m -ates w e re r e l a t e d to l a t i t u d e w i t h the n o r t h e r n seed s o u r c e s p o s s e s s i n g m o r e D N A per c e l l than the s o u t h e r n p r o v e n a n c e s . Changes i n n u c l e a r D N A l e v e l s due to e x t e r n a l e n v i r o n m e n t a l s t r e s s ( m a t e r n a l e f f e c t s ) a r e d i s -c u s s e d . M O I R and F O X (1972) have d e s c r i b e d the p r e s e n c e of s u p e r n u m e r a r y c h r o m o s o m e s (also c a l l e d B - c h r o m o s o m e s ) i n s e e d l i n g s d e r i v e d f r o m eight p r o v e n a n c e s of S i t k a s p r u c e . T h e y a l s o r e f i n e d the k a r y o t y p e p r e s e n t e d by B U R L E Y (1965^). New c o n s t r i c t i o n s w e r e ob-s e r v e d and a l l o w e d a s u b d i v i s i o n of the c o m p l e m e n t into f i v e groups. The p r e s e n c e of B - c h r o m o s o m e s should be taken into account f o r i n t e r -p r e t i n g the v a r i a t i o n o b s e r v e d i n growth, n u c l e a r volume, etc. 1'. 4 V a r i a t i o n i n wood p r o p e r t i e s ^ G e n e t i c 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 , t r a c h e i d length, c e l l 24 w a l l thickness and other wood proper t i es have been shown ( B U R L E Y , 1965D ; B r i t i s h F o r e s t P r o d u c t R e s e a r c h Organizat ion (1966): dif ferent r e p o r t s ) . 25 C H A P T E R 2 G E N E C O L O G Y D E F I N E D : ITS R E L A T I O N S H I P S W I T H T R E E B R E E D I N G The study of forest tree provenances inescapably r a i s e s the question of the re levance of the science of genecology to provenance r e s e a r c h . A number of provenance and progeny genetic v a r i a t i o n studies of forest trees have been recent ly publ ished as genecological studies, ( G A L O U X and F A L K E N H A G E N , 1 9 6 5 ; G A L O U X , 1966; F A L K E N H A G E N , 1968 b , R O C H E , 1969; H A G N E R , 1970). Di f ferent authors have attr ibuted different meanings to the t e r m genecology ( L A N G L E T , 1971). The t e r m genecology was coined by T U R E S S O N (1923), as synonym of race ecology or as the study of the species and i t s h e r e d i t a r y habitat types f r o m an eco log ica l point of v i e w . Throughout his v e r y in teres t ing paper , L A N G L E T (op. c i t . ) has t r i e d to r e s t r i c t the use of the t e r m genecology to the o r i g i n a l T u r e s s o n i a n meaning . However , most f o r e s t e r s have adopted the broader meaning developed by H E S L O P - H A R R I S O N (1964) and S T E R N (1964). Genecology is then cons idered as a synthetic science which b o r r o w s its tools f r o m a number of other sc ient i f i c f i e lds such as p h y s i o l o g i c a l ecology, quant i -tative genetics , etc. whi le also interact ing with these sc iences . Gene-cology i s thus defined as the science whose subject matter i s the study of the genet ical ly based i n f r a - s p e c i f i c v a r i a t i o n of plant species i n r e l a t i o n w i t h the d i v e r s i t y of their habitat of o r i g i n ( F A L K E N H A G E N , 1968^.). 2.6 The reader interes ted i n a v e r y comprehensive r e v i e w of forest genecology is r e f e r r e d to the paper by S T E R N (1964). . It is s u f f i -cient to say that forest genecology forms the sc ient i f i c bases of p r o v e n -ance r e s e a r c h : a complete genecological study of a tree species should be a p r e r e q u i s i t e to any se lect ion of seed s o u r c e s . When two tree species must be c r o s s e d , it i s common sense to c r o s s the races which w i l l give the most product ive and adapted progenies . E v e n the resul t s of c r o s s i n g s i n a seed o r c h a r d depend on a thorough knowledge of the genetic v a r i a b i l i t y of the adaptative t ra i t s of the tree species ( F A L K E N -H A G E N , 1968). One can ask why we did not name our study: genecology of Si tka spruce . The reasons are that we favoured L A N G L E T ' s concept of genecology and that we did not attempt to study exper imenta l ly the bases of adaptation of the genetic m a t e r i a l that we were studying, to their habitat of o r i g i n . We did not use contro l led environments to study the responses of the progenies to different eco logica l factors i n the hope to expla in their behaviour i n the n u r s e r y or e l sewhere . In this sense, we only r e a l i z e d a single tree progeny and provenance study i n one set of growing conditions as an obl igatory step to test this m a t e r i a l i n forest condit ions . 2.7 C H A P T E R 3 ORIGIN A N D C H A R A C T E R I S T I C S O F T H E M A T E R I A L S T U D I E D D u r i n g the 1970 f a l l , the I. U . F . R . O . , Section 22, " W O R K I N G G R O U P O N P R O C U R E M E N T O F S E E D F O R P R O V E N A N C E R E S E A R C H " organized an expedition to col lect S i tka spruce cones f r o m B r i t i s h C o l u m b i a and A l a s k a . . In most cases , the col lec t ions were made f r o m 15 trees i n each loca t ion . The procedure of the co l lec t ion has been outl ined by the Working Group on Provenance R e s e a r c h and T e s t i n g M e e t i n g " at P o n t - a - M o u s s o n , held on 6-9 September, 1965. The locat ions of the 39 provenances range f r o m 48° 38 to 58° 37 latitude N and f r o m 121° 93 to 134° 58 longitude W. The e l e v a -t ion v a r i e s f r o m 0 to 2, 200 feet above sea l e v e l . • See Table I and maps 1 and 2 for m o r e detai ls on the o r i g i n s of the provenances studied. Twenty cones f r o m each of the 557 trees represent ing 39 locat ions were p r o v i d e d to P r o f e s s o r O . S Z I K L A I . The seed was ex-tracted by hand dur ing the 1970-71 winter and stored at 3 2 ° F . T h i s single tree progeny co l lec t ion constitutes the m a t e r i a l with which the author commenced the present study as his P h . D . thes i s . It was decided to keep the single tree progenies separate and to study as m u c h as poss ib le the different seed or cone and seedling t ra i t s on a single tree bas is so as to be able to compare the two h i e r -a r c h a l l eve ls of v a r i a t i o n : populations (= provenances) and between T A B L E I G E O G R A P H I C A L C O O R D I N A T E S O F T H E S I T K A S P R U C E  P R O V E N A N C E S S T U D I E D A N D N U M B E R O F S I N G L E T R E E P R O G E N I E S A V A I L A B L E P E R P R O V E N A N C E 28 N o . of the provenance Name L a t i -tude L o n g i -tude ° tenth ° tenth E l e v a -tion feet N o . of single tree proge-nies 1 V E D D E R 49. 12 121.93 100 10 2 S Q U A M I S H R I V E R 49.92 123.25 100 15 3 B I G Q U A L I C U M R I V E R 49.38 124.62 0 15 4 S A L M O N B A Y 50. 38 125.95 0 15 5 C R A N B E R R Y R I V E R 55.47 128.23 1700 15 6 K I T W A N G A 55. 17 127.87 2200 15 7 U S K F E R R Y 54..6;3v l;28'-.40.. 450 15 8 S H A M E S 54.40 128.95 100 5 9 W E D E N E R I V E R 54. 13 128.62 550 15 10 K I T S U M K A L U M L A K E P A R K 54. 72 128.77 450 15 11 D E R R I C K L A K E 55.68 128.68 800 15 12 D R A G O N L A K E 55. 35 128.95 850 7 13 Z O L A P C R E E K 55. 15 129.22 50 15 14 F U L M A R C R E E K 55. 15 128.97 1300 15 15 A B E R D E E N C R E E K 54.20 129.92 0 15 18 C E D A R V A L E 55. 02 128.32 800 15 19 K A S I K S R I V E R 54. 28 129.42 100 15 20 H U M P B A C K C R E E K - P O R C H E R I S L A N D 54. 03 130.37 1000 11 21 I N V E R N E S S 54. 20 130.25 50 15 22 H A Y S M T N , P R I N C E R U P E R T 54. 27 130.32 2100 15 23 M O S S P O I N T , A N N E T T E I S L A N D 55.03 131.55 0 15 24 C R A I G 55. 50 133.13 0 15 25 O L D H O L L I S 55.47 132.67 0 15 26 W A R D L A K E 55.42 131.70 50 15 27 O H M E R C R E E K 56. 58 132.73 25 15 28 D U C K C R E E K 58. 37 134.58 100 15 29 A L L O U E T T E R I V E R , H A N E Y 49.25 122.60 650 15 30 M U I R C R E E K , S O O K E 48. 38 123.87 0 15 31 P O R T R E N F R E W 48. 58 124.40 25 15 32 TAHSIS I 50.08 127.50 100 15 33 TAHSIS 2 49. 83 126.67 10 .15 34 H O L B E R G 50. 62 128.12 100 15 35 M O R E S B Y IS, S K I N C U T T L E I N L E T 52. 28 131.22 50 8 36 M O R E S B Y IS, S E W E L L I N L E T 52. 87 132.08 50 15 37 M O R E S B Y IS, C U M S H E W A I N L E T 53. 05 132.08 200 15 29 T A B L E I - Continued N o . of single L a t i - L o n g i - E l e v a - tree N o . of the tude tude t ion proge-provenances Name ° tenth ° tenth, feet nies 38 S A N D S P I T , Q U E E N C H A R L O T T E I S L . 53.13 131.80 250 15 39 ' T U S K A T L A , Q U E E N C H A R L O T T E ISL. 53.50 132.17 300 15 40 M A S S E T I N L E T 53.92 132.08 0 15 41 B L E N H E I M , S A R I T A 48.90 124.95 700 11 T o t a l number of populat ions : T o t a l number of trees s tudied: 39 557 31' t rees , wi th in population v a r i a b i l i t y . Such a study offers the advantages of a provenance test and those of a m a t e r n a l test enabling, therefore , the compar isons of different nested levels of genetic v a r i a b i l i t y . These compar isons are fundamental i f we w i s h to r a t i o n a l l y determine which type of se lect ion is best for the m a t e r i a l studied (mass se lec t ion , f a m i l y se lect ion, etc. ). Ten Sitka spruce provenances selected f r o m the m a t e r i a l co l lec ted i n 1970, are being proposed for an internat ional exper iment ( D R I S C O L L , 1972). The twelve par t i c ipa t ing forest r e s e a r c h o r g a n i z a -tions w i l l prov ide the same treatments i n o r d e r to study provenance by site in terac t ion and genotype s tabi l i ty under different eco log ica l condi -t ions . C H A P T E R 4 C L I M A T O L O G Y O F T H E A R E A O F O R I G I N O F T H E S I T K A S P R U C E P R O V E N A N C E S F o r forest tree species , it is general ly admitted that the c l i m a t i c di f ferences between the places of o r i g i n of the provenances are m o s t l y responsib le for the genet ical ly based adaptative v a r i a b i l i t y that these provenances show i n comparat ive f i e l d tests ( S T E R N , 1964). The extreme values taken by the eco log ica l v a r i a b l e s rather than the average values would explain the act ion of the environment i n m o d i f y i n g the fine genetic m a k e - u p of the populations of t r e e s . T h e r m a l and p l u v i o m e t r i c extremes; ; their frequencies and their dates of o c c u r -rence would act as se lect ion f a c t o r s . B i o t i c and edaphic factors might also intervene to modi fy the tree populations, but these eco logica l v a r i a b l e s also depend on the c l i m a t e . The forest tree species are p e r e n -n i a l plants and, therefore , are most inf luenced by the c l imate and should be p r i m a r i l y adapted to the l o c a l c l i m a t i c conditions ( W R I G H T , 1963; F A L K E N H A G E N , 1968). T h e r e f o r e , it i s important to study the c l imate of the place of o r i g i n of the provenances of a tree species which is being s tudied, noting that the current c l imate m a y notbe the c l imate of evolut ion. Unfortunately , the network of the c l i m a t o l o g i c a l stations i n B r i t i s h C o l u m b i a and southern A l a s k a i s too loose to a l low an accurate d e s c r i p t i o n of the c l i m a t o l o g i c a l conditions of the place of o r i g i n of our Si tka spruce provenances . Not only does the w i l d mountainous p h y s i o -3.4 g r a phy of the r e g i o n p r e c l u d e any e x t r a p o l a t i o n , but some s t a t i o n s o n l y p r o v i d e p r e c i p i t a t i o n r e c o r d s , w h i l e o t h e r s have been e s t a b l i s h e d f o r too s h o r t a p e r i o d of t i m e . M o s t s t a t i o n s a r e along m a j o r r i v e r s o r n e a r the sea and r e p r e s e n t v a l l e y o r sea l e v e l c l i m a t e s . A c c o r d i n g to C H A P M A N (1952), two types of c l i m a t e p r e -dominate i n the a r e a of o r i g i n of the S i t k a s p r u c e p r o v e n a n c e s s t u d i e d : p r i m a r i l y the M a r i n e West C o a s t c l i m a t e and l o c a l l y , the C o o l S u m m e r M e d i t e r r a n e a n c l i m a t e . T h e s e c l i m a t e s r e s u l t f r o m the p r e v a i l i n g west-east movement of a i r o v e r the p r o v i n c e w h i c h i s due to s e v e r a l m a i n p r e s s u r e s y s t e m s . . D u r i n g the w i n t e r , the winds a r e southeast along the coast, at the s u r f a c e . However, the p e r s i s t e n c y of the winds i f f r e -quently i n t e r r u p t e d by the p a s s a g e of d e p r e s s i o n s and by the l o c a l topo-graphy. . D u r i n g the s ummer, the p r e v a i l i n g winds a r e n o r t h w e s t . The weather along the c o a s t i s m o s t v e r s a t i l e as i n a l l m a r i t i m e c l i m a t e s , and t h i s c h a n g e a b i l i t y i s due to the passage of a i r m a s s e s of P a c i f i c o r i g i n , o v e r the c o a s t . • A l t i t u d e and topography, changing a b r u p t l y i n the p r o v i n c e , r e s u l t i n r a p i d m o d i f i c a t i o n of the l o c a l c l i m a t e : t e m p e r a -t u r e i n v e r s i o n s , r a i n shadow e f f e c t s a r e w e l l known; the e a s t c o a s t of V a n c o u v e r I s l a n d i s d r i e r than the west c o a s t . N a t u r a l l y , a s p e c t i s v e r y i m p o r t a n t . The f o l l o w i n g s u b d i v i s i o n s m a y be u s e d ( C H A P M A N , 1952). (1) The west co a s t c l i m a t e : The m e a n annual range of the C o a s t a l s t a t i o n s i s 16°F i n the south to 23°F i n the n o r t h and i s evidence of the m i l d w i n t e r s (38°- 41°F) and c o o l s u m m e r s (56° - 58°F). The l e n g t h of the f r o s t - f r e e p e r i o d 35 3.6 v a r i e s f r o m 150 to 250 days . P r i n c e Rupert has the lowest annual total of sunshine hours i n Canada (1053 hours ) . Annual prec ip i ta t ions are high (50 to 100 inches) , m o s t l y r a i n . The number of days of p r e c i p i t a -t ion is v e r y high (201 to 262). The seasonal r a i n d i s t r i b u t i o n is m a r i -t i m e , but wi th a summer m i n i m u m w e l l m a r k e d . Re la t ive humidi ty i s h igh . The coastal reg ion may be subdivided i n : (a) Outer d i v i s i o n , (b) Inner d i v i s i o n , (c) F j o r d s , and (d) T h r o u g h v a l l e y s . (2) The c l imate of southeast Vancouver Is land, G e o r g i a Strai t and the L o w e r F r a s e r delta The m a i n c h a r a c t e r i s t i c s are summer def ic iencies of m o i s t u r e , lower prec ip i ta t ions (20 to 40 inches) and p a r t i c u l a r l y i n the V i c t o r i a -Saanich P e n i n s u l a a rea , high totals of sunshine (2, 207 hours at V i c t o r i a ) . The f ros t free p e r i o d exceeds 200 days . We have plotted the monthly averages for total prec ip i ta t ions (inches) and for temperature (degrees 0 Fahrenheit ) for some r e l i a b l e c l i m a t o l o g i c a l stations c lose enough to the locations of at least some of the provenances . (Temperature and prec ip i ta t ion tables for B . C . , 1971). (A) B . C . main land ( F i g . 2 and 3). F o u r stations have been plotted: the stations of C h i l l i w a c k , Haney U . B . C . forest , K i t i m a t and P r i n c e R u p e r t . The station of C h i l l i w a c k is i n the L o w e r F r a s e r v a l l e y whi le that of Haney is i n the coasta l range mountains . The temperature 3S\ Fig.4 . Monthly v a r i a t i o n i n p r e c i p i t a t i o n and temperature f o r two stations of the Queen Charlotte Islands. curves are s i m i l a r at both stations, but C h i l l i w a c k is a l i t t le w a r m e r except i n January . The p r e c i p i t a t i o n is heavier at Haney and the m i n i -m u m is less pronounced than at C h i l l i w a c k . It snows a l i t t l e m o r e at Haney. The c l imates of K i t i m a t and P r i n c e Rupert are quite dis t inct i n spite of their s i m i l a r lat i tude. The c l imate of K i t i m a t is m o r e c o n t i -nental with heavier prec ip i ta t ion during the winter months, a delayed and m o r e pronounced r a i n f a l l m i n i m u m dur ing the summer (in July instead of June). The seasonal temperature v a r i a t i o n is also m o r e pronounced. . The average temperature at P r i n c e Ruper t r e m a i n s above f r e e z i n g point which is not the case at K i t i m a t . (B) Queen Charlotte Islands stations ( F i g . 4). Two stations have been plotted: the stations of Sandspit and M a s set. Only the latitude differentiates the two stat ions. The temperature curves are i d e n t i c a l ; only the p r e c i p i t a t i o n curves are d i f -ferent wi th a m o r e pronounced m i n i m u m at Sandspit . The annual total is 49.56 inches at Sandspit (Lat . 53° 15) and 56.27 at M a s s e t (Lafe. 54° 02). (C) Vancouver Is land ( F i g . 5). V i c t o r i a International A i r p o r t . Some Si tka spruce provenances or iginate f r o m is lands i n southeast A l a s k a . The topography of this reg ion is rough and mounta in-ous and is cut by numerous in land w a t e r w a y s . The c l imate i s cool and m o i s t wi th a r e l a t i v e l y n a r r o w range between summer and winter t empera tures . The heavy prec ip i ta t ion is w e l l d i s t r ibuted throughout 40 F i g . 5 . Monthly v a r i a t i o n i n p r e c i p i t a t i o n and temperature at V i c t o r i a I n t e r -n a t i o n a l a i r p o r t . -4-7 + 6 70. I 0 D 60 J _ ;so o a E <D 40. 30 4-5 w a u C c o • mm '* 2 • mm a " u a . - f i H 1 1 1 1 1 I I I 1 1 r -J F M A M J J A S O N D Months "4.1 the growing season. There i s no pronounced summer drought ( A N D E R S E N , 1955). The extreme nor thern port ion and along the m a i n -land has a m o r e continental c l i m a t e . In winter , n o r t h e r l y winds f r o m the Inter ior lower the temperature of the is lands ly ing nearer the m a i n -land, but the P a c i f i c Ocean influences the c l imate of the outer i s l a n d s . There are great var ia t ions i n c l imate within short distances due to aspect, water m a s s and elevation effects as in B r i t i s h C o l u m b i a . The reader is r e f e r r e d to A N D E R S E N (1955) who has given tables and i s o t h e r m a l or i sohyeta l maps for the fol lowing f a c t o r s : p r e c i p i t a t i o n , mean t e m p e r a -tures dur ing different months, number of f r o s t - f r e e days and mean temperature dur ing the three coldest months . P r e c i p i t a t i o n exceeds that n e c e s s a r y for opt imum tree growth and low potential evapotranspirat ion r e s u l t s i n a water surplus and makes s o i l drainage,together wi th summer tempera tures , a c r i t i c a l factor i n affect ing tree product iv i ty i n southeast A l a s k a . T r e e growth u l t i m a t e l y depends p r i m a r i l y on the p a r t i c u l a r energy and water balances of a given r e g i o n . A parameter is commonly used to descr ibe the growth potential of an area when water supply is adequate: the number of growing degree-days which i s defined as the sum of the di f ferences between the da i ly mean temperature and 4 2 ° F or 4 3 ° F chosen as c r i t i c a l temperature for growth p r o c e s s e s . B O U G H N E R (1964) has rev iewed the concept of growing degree-days and presented accumulated d ;e;gree-days against t ime ( F i g . 6) for A g a s s i z , Nanaimo and S m i t h e r s . The curves of A g a s s i z and Nanaimo are i d e n t i c a l , but 3800 Fig.6. Growing degree-days accumulated for three stations of B.C. The dates when bud burst (Date 1) and bud set (Date 2) were estimated are shown. IV A3 differ sharply f r o m the curve for S m i t h e r s , indicat ing that there is m u c h less avai lable radia t ion energy for growth here ; thus, that latitude increases resu l t in decreas ing radia t ive energy usable by the plants . The accumulated day-degrees over 4 3 ° F are 2, 500 to 3, 500 for the Coasta l D o u g l a s - f i r zone and 1, 500 to* 2, 500 for the Coas ta l Western H e m l o c k zone ( K R A J I N A , 1969). The evaporative demand of the c l imate of B r i t i s h C o l u m b i a is l i t t le known. . The potential evapo- transpirat ion , according to the w e l l known T H O R N T H W A I T E ' s method, of the L o w e r F r a s e r va l l ey , during the p e r i o d M a y - September, could be 20 inches , while the p r e c i p i t a t i o n i s only 14 ( C H A P M A N and B R O W N , 1966). The potent ial e v a p o t r a n s p i r a -tion ( P E T ) has been calculated for a number of A l a s k a n and nearby Canadian stations ( P A T R I C and B L A C K , 1968). A belt of heavy r a i n f a l l extends along the southern shorel ine f r o m B r i t i s h C o l u m b i a westward to the end of the A l e u t i a n chain . Within this forest reg ion , the average P E T for 68 stations i s 21.27 inches . The heavy prec ip i ta t ion along the coasta l p e r h u m i d belt becomes m u c h l ighter wi th in r e l a t i v e l y short distances f r o m the P a c i f i c Ocean. C o n c u r r e n t l y , there is a change f r o m western h e m l o c k - S i t k a spruce to the s p r u c e - a s p e n - b i r c h forest of the i n t e r i o r . P E T ranges i n this reg ion f r o m 15.5 to 19.8 inches . However , c o m m e r -c i a l t imbers m a y use m o i s t u r e f r o m water tables i n addit ion to l o c a l p r e c i p i t a t i o n for their water supply. T h e r e f o r e , growth of the nor thern forest is p r i m a r i l y governed by temperature ( P A T R I C and B L A C K , op. c i t . ). P A R T II S E E D A N D C O N E M O R P H O L O G Y A S P A R T O F T H E B I O S Y S T E M A T I C S O F S I T K A S P R U C E 44 C H A P T E R 1 F O R E S T T R E E B I O S Y S T E M A T I C S D E F I N E D B i o s y s t e m a t i c s has r e c e i v e d different meanings vary ing with the authors who use this w o r d . O r i g i n a l l y created by, C A M P and G I L L Y (1943) as the combination of taxonomy and evolution, b iosys te -m a t i c s or b iosystematy was considered by these authors as the c l a s s i -f i ca t ion of l i v i n g o r g a n i s m s according to a taxonomic sys tem based upon phylogenetic re la t ionsh ips , wi th emphasis upon the factors responsib le for the segregation of the taxa i n v o l v e d . B i o s y s t e m a t i c s corresponds , therefore , to the "new s y s t e m a t i c s " as proposed by H U X L E Y (1940). B i o s y s t e m a t i c s (sensu str icto) thus a ims at: (1) d e l i m i t i n g the natura l b iot ic uni ts ; (Z) to apply to these units a system of nomenclature ade-quate to the task of conveying p r e c i s e in format ion r e -garding their defined l i m i t s , r e la t ionsh ips , v a r i a b i l i t y and dynamic s t ruc ture , ( C A M P and G I L L Y , 1943). H E S L O P - H A R R I S O N (1964) has proposed a s i m i l a r d e f i n i -t ion : b i o s y s t e m a t i c s is the study of evolut ionary processes i n plants and the bearings of this study on their taxonomy. C A L L A H A M (1963) has s i m i l a r l y defined b iosystematy , but has also confounded it with genecology according to T U R E S S O N , which needs further e laborat ion . M o s t forest genetic ists consider as part of b i o s y s t e m a t i c s , the study of the m o r p h o l o g i c a l v a r i a b i l i t y of the parent trees of the 45 progenies that they are studying. B i o s y s t e m a t i c a l studies were to c l a s s i f y the parent trees into natura l b iotas , us ing different m a t e r n a l adult c h a r a c t e r i s t i c s such as cone, twig and leaf t r a i t s . Despite the advocacy of parental morphology as useful or indispensable tools in provenance r e s e a r c h by C A L L A H A M (1963) and R O C H E (1968), many cone and seed m o r p h o l o g i c a l studies of forest trees have not given v e r y useful resul t s when the study was done at the i n f r a - s p e c i f i c l e v e l . • Some interes t ing resu l t s as an a id to c l a s s i f i c a t o r y purposes have been c l a i m e d for seed weight (more p r e c i s e l y 1, 000 seed weight) for different species (for example , for L a r i x decidua M I L L ( S I M A K , 1967) and for D o u g l a s - f i r (B IROT, 1972); and for a set of seed c h a r a c t e r i s t i c s ( A L L E N , I960). However , the use of these c h a r a c t e r i s -t ics is l i m i t e d to the d is t inc t ion of a v e r y b r o a d geographical v a r i a t i o n such as coastal and i n t e r i o r var ie t i e s of D o u g l a s - f i r . G e n e r a l l y , there is a c l i n a l v a r i a t i o n for each c h a r a c t e r i s -t ic studied, different for each reg ion dis t inguished. A fact which does not s i m p l i f y the del ineat ion of workable u n i t s . A n example of c o m p l i c a -ted pattern of v a r i a t i o n , despite extensive measurements i s given by Y A O ' s study of seed weight and cone scale morphology of 124 p r o v e n -ances and 1,818 trees of D o u g l a s - f i r ( Y A O , 1971). There are a number of poss ib le reasons to expla in the lack of power to c l a s s i f y the natura l populations of a tree species into useful b io tas : 1. A m i x t u r e of c l i n a l v a r i a t i o n and ecotypic v a r i a t i o n m a y complicate the d is t inc t ion of workable u n i t s . It is quite poss ib le that the t ra i ts studied are not so stable as c l a i m e d by their proponents : cone and seed tra i ts cer ta in ly foliage or twig charac ters might not be under s t r i c t genetic c o n t r o l : their h e r i t a b i l i t y might be v e r y l o w . The choice of the charac ters to measure has been so far e m p i r i c a l or subject ive : i t was often decided to measure seed width, different angles on the cone brac t , etc. , without knowing the d i s c r i m i n a t o r y power of the t ra i ts m e a s u r e d . One method would be to measure a great number of t r a i t s : 20 to 30 in an adequate sample of popu-lat ions and then to p e r f o r m a s tep-wise d i s c r i m i n a n t function analys is or a p r i n c i p a l component analys is to e l iminate any redundancy of in format ion in the t ra i t s studied. M o s t studies cons idered one t ra i t at a t i m e . A m u l t i v a r i a t e approach could be u s e f u l . Another important p r o b l e m genera l ly neglected is the sampl ing p r o b l e m : how many indiv iduals should be m e a s u r e d The p r o b l e m is v e r y complex as different nested leve ls of v a r i a t i o n exis t , and must be s imultaneously taken into account. A s t ra t i f i ed sampl ing procedure has shown that 15 brac ts were n e c e s s a r y to estimate c o r r e c t l y the catkin means and that four catkins were n e c e s s a r y to est imate the tree means with a low e r r o r , p r i o r to a m o r p h o l o g i c a l •4:7 study of the catkin of y e l l o w b i r c h (Betula a l leghaniensis BRITT.) . T h i s sampl ing study is the only one ever mentioned i n the forest genet ics - l i terature ( F A L K E N H A G E N , 1968 b ) . 6. N a t u r a l l y , the reasons just mentioned can combine to b l u r r the c l a s s i f i c a t i o n of the parent trees on the bas is of their phenotypic morphology . 48 . C H A P T E R 2 S E E D A N D C O N E M O R P H O L O G Y • U N I V A R I A T E A N A L Y S E S O F V A R I A N C E A N D M U L T I P L E C O R R E L A T I O N A N D R E G R E S S I O N A N A L Y S E S The geographical co-ordinates of the provenances studied for cone and seed morphology and the number of progenies studied per provenance are indicated i n Table I. F i v e seeds were randomly col lec ted f r o m each tree progeny and stuck on a spec ia l sheet. The measurements , as shown i n F i g . 7, were made us ing an A B A D D O ' s machine for m e a s u r i n g annual r i n g s , to the nearest 0.01 m m . Ten cones per progeny were randomly col lec ted and the length of each cone was m e a s u r e d to the nearest m m . 2. 1 U n i v a r i a t e analyses of var iance The var iance and the mean, for each provenance, of the different c h a r a c t e r i s t i c s m e a s u r e d on the seed and the cone in m m are shown in Table II. H i e r a r c h a l analyses of var iance have been p e r f o r m e d , using the total m a t e r i a l and cons ider ing three leve ls of v a r i a t i o n : p r o v -enances, trees wi th in provenances , seeds within t rees . The large heterogeneity of the m a t e r i a l resu l t s i n the non-homogeneity of the v a r i a n c e s , at the different leve ls of v a r i a t i o n , and in modi fy ing "the l e v e l of s igni f i cance , thus i n i n c r e a s i n g the type l a n d II e r r o r . However , 49 Fig. 7 . The four measurements taken f o r studying seed morphology of Sitka spruce. a : w i n g l e n g t h b : w i n g w i d t h c r s e e d l e n g t h d : s e e d 50 T A B L E II V A R I A N C E S A N D M E A N S O F T H E S E E D A N D C O N E C H A R A C T E R I S T I C S M E A S U R E D Number T , , Seed C h a r a c t e r i s t i c s „ of the Cone provenance a. _b c d. Length var iance 0.84 0.07 0.75 0. 03 70. 58 mean 7.73 3.38 2.93 1.74 54.26 2 0.50 0.11 0. 10 0.02 112.68 8,06 3. 58 3. 08 1.84 66.81 0.54 0.12 0.05 0.02 64.98 3 6.96 3.41 2 .95 1.77 64.90 4 0.43 0.22 0.07 0.03 77.02 6.82 3.41 2.82 1.83 60. 19 5 1.16 0.12 0.06 0.02 68.40 7.50 3. 50 2.73 1.68 40.65 6 1.19 0.23 0.05 0.04 53.44 7.53 3.41 2.77 1.70 42.60 ? 1.21 0.15 0. 12 0.04 111.35 8.01 3.62 2.89 1.81 59.59 8 0. 71 0 .14 0.06 0.02 66 .85 8. 00 3.91 3.15 1.97 66. 08 Q 0. 59 0. 12 0.09 0.03 126.18 7.18 3.58 2.74 1. 77 54.71 1 0 0 .96 0.14 0.06 0 .05 102.64 7.53 3.69 2.99 1.81 56.46 n 0 .67 0 .15 0.08 0.04 28.55 6.54 3.29 2.48 1.74 38.82 1 2 1.82 0.12 0.10 0 .02 117.29 7.48 3.36 2.56 1.76 41.56 13 0.86 0.09 0. 06 0.02 81.42 7.08 3.39 2.55 1. 88 51. 30 51 T A B L E II - Continued Number of the provenance 14 v a r i a n c e mean Seed C h a r a c t e r i s t i c s 0. 83 7. 31 0. 15 3.62 •0. 10 2. 74 0. 03 1.88 Cone Length 43. 14 53. 07 15 0.49 7. 30 0.10 3. 71 0. 06 2. 83 0. 02 1. 99 71. 96 62. 12 18 1.41 7. 82 0. 32 3. 72 0. 11 2.83 0. 05 1. 79 54. 53 44. 50 19 0. 35 7. 38 0. 11 3. 72 0. 10 2. 86 0. 04 1. 86 65. 17 60.39 20 1. 22 6. 53 0. 11 3. 39 0. 08 2. 81 0. 03 1. 80 155.54 55. 07 21 0. 75 7. 45 0. 33 3.95 0. 05 2.97 0. 04 2. 09 99.07 59.97 22 0. 29 6.46 0. 15 3.65 0. 07 2. 72 0. 05 1. 84 67.98 49.73 23 0. 71 6. 34 0. 11 3.45 0. 14 2. 50 0. 04 1. 76 47. 04 44. 52 24 0. 73 7. 77 0.13 4. 01 0. 05 3. 04 0. 03 1.99 116.05 64. 51 25 0. 39 7. 13 0. 17 3.81 0. 06 2. 81 0. 03 2. 01 95. 83 62.62 26 0. 37 7. 56 0. 13 3. 68 0. 06 3. 00 0. 03 1. 96 143.97 58. 83 27 0. 62 6. 84 0. 16 3. 60 0. 06 2. 86 0. 03 1.90 88.97 55. 57 28 0. 83 7. 73 0. 19 3.61 0. 03 2.96 0. 02 1. 93 34.90 58. 58 29 0044 7. 24 0.09 3. 52 0. 07 2.99 0. 03 1. 78 89.31 53. 35 52 T A B L E II - Continued Number of the provenance 30 va r i a n c e mean Seed C h a r a c t e r i s t i c s 0.44 7. 04 0.10 3.41 0. 12 3. 04 0. 04 1. 75 Cone Length 130.78 59. 31 31 0. 68 7. 25 0. 14 3. 54 0. 11 2.91 0. 03 1. 81 117. 30 63. 63 32 0.96 6. 35 0.12 3. 22 0. 07 2. 58 0. 03 1. 75 46. 39 :52. 51 33 34 0.93 7. 06 0. 50 6. 70 0. 23 3.45 0. 16 3. 31 0.09 2. 85 0. 08 2. 79 0. 06 1.86 0. 04 1.84 72. 19 57. 84 53. 24 56. 28 35 0.95 6.97 0. 26 3. 56 0. 15 2. 86 0. 05 1. 77 201.54 61. 17 36 0.93 6. 51 0.15 3. 51 0. 08 2. 86 0. 02 1. 86 53. 70 49. 20 37 0.64 7.94 0. 20 4. 07 0. 10 3. 11 0. 05 1.95 141.07 68.99 38 0. 81 6. 70 0. 13 3.41 0. 08 2. 75 0. 04 1. 77 59. 28 49. 33 39 1. 07 6. 95 0. 15 3. 68 0. 06 2. 83 0. 04 1. 90 80. 27 53. 83 40 0. 77 6.96 0. 12 3. 61 0. 07 2. 70 0. 03 1. 77 172.34 66. 26 41 0. 77 6.81 0. 13 3. 31 0. 13 2. 67 0. 03 1.69 64. 23 60. 77 53 because the F values calculated are v e r y high, the resul ts of the Anova are , nevertheless presented here (Table III). T A B L E III F V A L U E S O F T H E N E S T E D A N O V A P E R F O R M E D O N T H E D I F F E R E N T T R A I T S M E A S U R E D Seed T r a i t s Cone Length Sources of V a r i a t i o n D . F . a b c d D . F . Provenances 38 58.8 44. 5 3G. 24 24. 6 38 231. 51 T r e e s / P r o v . 517 9.8 8. 1 3.9 2. 6 517 15. 33 E r r o r 2224 5004 A l l the F values calculated are very highly s igni f i cant . The heterogeneous pattern of v a r i a t i o n of the total m a t e r i a l prec ludes any es t imat ion of var iance components or m u l t i p l e compar isons of the means of the provenances . It should be noted that between provenances and between trees w i t h i n p r o v e n a n c e , v a r i a b i l i t i e s do exist and that the cone length seems to be the most var iab le c h a r a c t e r i s t i c . It is important to r e v i e w the assumptions made i n the analys is of v a r i a n c e : (1) The treatment effects and the environmenta l effects must be addi t ive ; (2) The exper imenta l e r r o r s must a l l be independent; 5'4 (3) The exper imenta l e r r o r s must have a common v a r i a n c e ; (4) The exper imenta l e r r o r s should be n o r m a l l y d is t r ibuted ( C O C H R A N , 1947). G e n e r a l l y , a s m a l l departure of the bas ic data f r o m n o r -m a l i t y does not affect too m u c h the s ignif icance l e v e l or the power of the F test, but the heterogeneity of the e r r o r var iance does affect these p r o p e r t i e s . When the e r r o r var iance changes with the treatment means according to a definite re la t ionsh ip , then some t rans format ion of the data can be calculated ( B A R T L E T T , 1947). N a t u r a l l y , the data should also respect the supplementary assumptions of the p a r t i c u l a r model used . If some l e v e l of v a r i a t i o n suffers heterogeneous v a r i a n c e s , then the assumptions are unwarranted and the es t imat ion of var iance components do not make m u c h sense. To test the homogeneity of v a r i a n c e s , for each l e v e l of v a r i a t i o n and for each var iab le while us ing the large number of m e a s u r e -ments accumulated (more than 7, ZOO measurements) would have been a v e r y expensive p r o p o s i t i o n . T h e r e f o r e , we decided to test only the homo-geneity of var iances of the t r e e - p r o g e n i e s , for each provenance and for each t ra i t separate ly . The fo l lowing table s u m m a r i z e s the B A R T L E T T ' s tests p e r f o r m e d (Table I V ) . T h u s , between 15% to 28% of the provenances have hetero-geneous var iances for the t ra i t s s tudied. On the bas is of these low p e r -centages, i t was decided that it was not worth the work to f ind out which -i f any - t r a n s f o r m a t i o n w i l l s tabi l ize the v a r i a n c e s . F u r t h e r m o r e , a 55 plotting (not shown) of the var iances against their means for some charac ters did not show any definite r e l a t i o n s h i p . A n eas ier way to a l leviate the heterogeneity of var iance problems is to subdivide the m a t e r i a l and analyze each subset separate ly . T A B L E IV N U M B E R O F P R O V E N A N C E S OUT O F A T O T A L O F 39  W I T H H E T E R O G E N E I T Y O F V A R I A N C E S F O R F O U R  S E E D C H A R A C T E R I S T I C S A N D C O N E L E N G T H (Significance l e v e l . 005) Seed C h a r a c t e r i s t i c s Cone Length a b c d N u m b e r 7 7 10 6 11 ' It was decided, for the anovas and the m u l t i v a r i a t e s ta t i s t i ca l analyses , to analyze separately 5 eco logica l subregions . T h u s , the large a rea of o r i g i n of the Sitka spruce provenances was sub-divided on the bas is of the b i o c l i m a t i c and physiographic data a v a i l a b l e . N a t u r a l l y , wi th in each reg ion , c l i m a t i c gradients exis t because of their s i z e . These regions are working units which have the fo l lowing advantages: these regions are f a i r l y eco log ica l ly different f r o m each other, the heterogeneity of var iance p r o b l e m is l i k e l y to be l i m i t e d and safe mul t ip le c o m p a r i s o n tests of the provenance means are poss ib le because the number of treatments (here provenances) i s never greater than 12 (except for reg ion 5 represented by 16 provenances) . 56 Region 1: i s f o r m e d by a c l i m a t i c a l l y homogeneous region - the eastern coast of Vancouver Island and the L o w e r F r a s e r V a l l e y have s i m i l a r c l i m a t i c c h a r a c t e r i s t i c s , at least i n c o m p a r i s o n with the western coast of Vancouver Is land which is p e r h u m i d . i s f o r m e d by the western coast of Vancouver Is land, p e r -humid and influenced by the P a c i f i c Ocean, i s f o r m e d by the Queen Charlot te Islands whose oceanic Region 2: Region 3: Region 4: c l imate i s accentuated. i s less w e l l defined and v e r y b r o a d : it i s f o r m e d by the A l a s k a panhandle. Region 5: is f o r m e d by two major r i v e r bas ins , however close enough to be pooled as the P r i n c e Rupert region or the Skeena R i v e r r e g i o n . . It represents provenances scat tered along an a l t i tudinal gradient in two major va l leys of centra l coastal B . C . Provenance 9 is i so la ted and was a r b i t r a r i l y p laced i n reg ion 5. Some provenances were a r b i t r a r i l y p laced in some reg ions : for instance, provenance 4 was placed in region 1 despite its nor thern p o s i t i o n . Provenance 30 has been placed in reg ion 2 because the Sooke area belongs to the w e s t e r n - h u m i d coast of Vancouver Island according to the map of the b iogeoc l imat i c subzones of Vancouver Island publ ished by the forest company M a c M i l l a n B l o e d e l L t d . i n 1971. F o r each of the five reg ions , separate nested analyses of var iance were p e r f o r m e d as w e l l as D U N C A N ' s mul t ip le range test of the provenance means, for each t ra i t studied. See Tables V , VI , VII , VIII and I X . The D U N C A N ' s tests c l e a r l y show that no coherent c l a s s i f i ca t ion of the provenances i s poss ib le on the bas is of the different t ra i t s studied i f they are cons idered independently. T h i s e r r a t i c v a r i a tion resu l t s f r o m the fact that the c h a r a c t e r i s t i c s studied are weakly c o r r e l a t e d (see Table XI) , d i s r e g a r d i n g the reg ions . T A B L E V N E S T E D A N A L Y S E S O F V A R I A N C E F O R R E G I O N 1 Source of V a r i a t i o n D F Provenances 3 T r e e s / P r o v . 50 E r r o r 216 T O T A L 269 539 D U N C A N ' s m u l t i p l e range test ^ ( 0 ^ = 0.05) Seed c h a r a c t e r i s t i c s Code number of the provenances ranked i n decreas ing order a J 29 3 4 • • • b 29 3 4 1 • • • • c 29 3 1 4 • • • • d 4 29 3 1 • • • • Cone Length 3 4 1 29 0 . • • • F Values Cone Seed C h a r a c t e r i s t i c s Length a b £ d D F 34. 28 4. 73 2.83 2.93 3 98.02 6.21 10.07 3.59 2.65 50 . 10.03 486 (*) Under l ined!provenances are not s t a t i s t i c a l l y different at the e r r o r l e v e l of 5%. 59. T A B L E VI N E S T E D A N A L Y S E S O F V A R I A N C E F O R R E G I O N 2 F Values Cone Source of Seed C h a r a c t e r i s t i c s Length V a r i a t i o n D F a b £ d D F Provenances 5 25.95 12.08 28.09 8.53 5 57.69 T r e e / P r o v . 80 8.53 5.39 3.25 2.52 80 13.66 E r r o r 344 774 T O T A L 429 859 A l l the F values are s ignif icant at the leve l 0 .001. D U N C A N ' s mul t ip le range test (oC = . 0.05) Seed c h a r a c t e r i s t i c s Code number of the provenance ranked i n decreas ing order a 31 33 30 41 34 32 31 33 30 34 41 32 30 31 33 34 41 32 33 34 31 30 32 41 Cone Length 31 41 30 33 34 32 60 T A B L E VII N E S T E D A N A L Y S E S O F V A R I A N C E F O R R E G I O N 3 F Values Cone Source of Seed C h a r a c t e r i s t i c s Length V a r i a t i o n : D F a. b_ _c d D F Provenances 5 68.96 68.89 31.42 16.29 5 255.85 T r e e / P r o v . 77 12.82 11.12 4 .90 2.95 77 18.76 E r r o r 332 747 T O T A L 414 829 A l l the F values are s ignif icant at the l e v e l 0 .001. D U N C A N ' s mul t ip le range test ( (/\ = 0.05) Seed c h a r a c t e r i s t i c s Code number of the provenances ranked in decreas ing o r d e r 37 35 40 39 38 36 37 39 40 35 36 38 37 35 36 39 38 40 37 39 36 35 40 38 Cone Length 37 40 35 39 38. 36 61 ; T A B L E VIII N E S T E D A N A L Y S E S O F V A R I A N C E F O R R E G I O N 4 F Values Cone Source of Seed C h a r a c t e r i s t i c s Length V a r i a t i o n D F a. b e d D F Provenances 5 102.52 50.78 80.41 27.66 5 242.61 T r e e / P r o v . 84 9.26 9.50 5.57 3.12 84 20.63 E r r o r 360 810 T O T A L 449 899 A l l the F values are s ignif icant at the l e v e l = 0 .001. D U N C A N ' s m u l t i p l e range test ( (A = 0.05) Seed c h a r a c t e r i s t i c s Code number of the provenances ranked i n decreas ing order a 24 28 26 25 27 23 b 24 28 26 25 27 23 c 24 26 28 27 25 23 d 25 24 26 28 27 23 Length 24 25 . 26 28 27- 23 62 T A B L E IX N E S T E D A N A L Y S E S O F V A R I A N C E F O R R E G I O N 5 F Values Cone Source of Seed C h a r a c t e r i s t i c s Length V a r i a t i o n D F a b £ d D F Provenances 15 62.03 38.49 44.81 30.84 15 313.16 T r e e / P r o v . 220 11.71 9.04 3.89 2.45 220 14.29 >[c >[e 5[c . >^^<:'fi E r r o r 952 2167 T O T A L 1187 2402. D U N C A N ' s mul t ip le range test ( o< = 0.05) Seed c h a r a c t e r i s t i c s Code number of the provenances ranked i n decreas ing order a 7 8 18 6 10 5 12 21 19 14 15 9 13 11 20 22 * • * • • • . . . . . 1 . — b 21 8 19 18 15 10 22 14 7 9 5 6 13 20 12 11 c 8 10 21 7. 19 15 18 20 6 14 9 5 22 12 13 11 d 21 15 8 13 14 19 22 10 7 20 18 9 12 11 6 5 * * • • * • • • « • m m m » : » • Cone Length : 8 15 19 21 7 10 20 9 14 13 22 18 6 12 5 11 If the tree to tree v a r i a t i o n in seed and cone c h a r a c t e r i s t i c s is large enough i n c o m p a r i s o n wi th the provenance to provenance v a r i a -t ion, it i s l i k e l y that a coherent c l a s s i f i c a t i o n of the stands on the bas is of the characters studied would be d i f f i c u l t . Components of var iance were , therefore , ca lculated assuming a complete ly random model (Model II of E isenhart ) and e l i m i n a t i n g the provenances not represented by 15 tree progenies . We so i n s u r e d thatthe model was balanced which enables easy and unbiased est imat ions of the components of v a r i a n c e s . Table X s u m m a r i z e s the components of var iance i n % for the different seed and cone charac ters studied and for each reg ion separate ly . T symbol izes the tree to tree var iance , S the provenance to provenance v a r i a n c e . T and S are expressed as % of T f S. A close examinat ion of Table X shows that the v a r i a b i l i t y is m o s t l y confined to the between tree v a r i a b i l i t y , or wi thin provenance v a r i a b i l i t y , except for seed length and width and for cone length; i n this case, for some regions only . A hypothesis would be that seed length, seed width and cone length could be good d i s c r i m i n a t o r y var ia tes for c l a s s i f y i n g the provenances . 2 .2 M u l t i p l e c o r r e l a t i o n and r e g r e s s i o n analyses . Relat ionships wi th the geographical coordinates of the place of o r i g i n of the provenan-ces studied A s imple c o r r e l a t i o n m a t r i x has been calculated between the c h a r a c t e r i s t i c s a, b , c, d, cone length and longitude, latitude and altitiude of the place of o r i g i n of the provenances , us ing the provenance means (Table XI) , ignor ing reg iona l groupings . • 64 T A B L E X C O M P O N E N T S O F V A R I A N C E IN P E R C E N T A G E S F O R T H E "SEED T R A I T S S T U D I E D A N D C O N E L E N G T H Seed T r a i t s Region N o . a b c d Cone Length T S T S T S T S T S 1 71. 35 28. 64 100 - 100 - 100 - 57.89 42. 10 2 86. 13 13.86 90. 90 9. 09 55. 35 44. 64 75. 00 25. 00 80. 43 19.56 3 74. 31 2 5; .6 8-, 70£44V 29. 55 66. 66 33. 33 68. 75 31. 25 50. 70 49.29 4 57. 08 42.91 75. 78 24. 21 47. 82 52. 17 58.82 41 . 17 56.88 43. 11 5 75. 25 24. 74 79. 52 20.47 50. 00 50. 00 42. 10 57. 80 38. 82 61 . 17 T : tree v a r i a n c e ; S: provenance var iance A close examinat ion of this c o r r e l a t i o n m a t r i x shows that a l l t ra i t s are c o r r e l a t e d to some extent with each other, with the exception of the wing length (a) which is not c o r r e l a t e d with seed width (d). Cone length is c o r r e l a t e d with seed c h a r a c t e r i s t i c s , but the re la t ionships are not too s trong, the c o r r e l a t i o n coeff ic ients v a r y i n g between 0. 32 and 0 .66 . Only wing width and seed width are s igni f icant ly c o r r e l a t e d with the geographical coordinates . The percentage of v a r i a t i o n accounted for o s c i l l a t e s around 16%. There is a tendency for a l l the t ra i t s to decrease when the elevation i n c r e a s e s . Cone length decreases s igni f i cant ly with elevat ion (r = - 0. 57). T h i s c o r r e l a t i o n m a t r i x has been further invest igated by T A B L E XI C O R R E L A T I O N M A T R I X B E T W E E N T H E S E E D C H A R A C T E R I S T I C S A N D  C O N E L E N G T H A N D T H E G E O G R A P H I C A L C O O R D I N A T E S O F T H E P L A C E O F ORIGIN O F T H E 39 P R O V E N A N C E S S T U D I E D Cone L o n g . L a t . A l t . d Cone L o n g . L a t . A l t . 1. 00 0 . 5 7 * * * 0 . 62* * * 0. 30NS 0. 32* •0. 099NS 0. 17NS 1. 00 0 . 6 0 * * * 1. 00 0 .75* * * 0 .45** 1.00 0 .48** 0 .66*** 0.49** 0 .45** - 0 . 1 0 N S 0 .49** 0 .39* - 0 . 1 9 N S 0.34* 1. 00 •0.033NS 1.00 - 0 . 31NS 0 .78* * * 1.00 0.0054NS - 0 . 1 5 N S -0 .28NS - 0 . 4 0 * .0. 5 7 * * * - 0 . 0 9 N S 0.24NS 1.00 NS Not s ignif icant at 5% l e v e l s ignif icant at 5% significant at 1 % signif icant at 0. 1% means of the m u l t i p l e r e g r e s s i o n analys is which al lows the study of different influences independently of each other . A " b a c k w a r d step-w i s e " technique p r o g r a m m e d by D r . A . K O Z A K ( U . B . C . F a c u l t y of F o r e s t r y ) has been used . T h i s technique consists in ca lculat ing a mul t ip le r e g r e s s i o n equation between the dependent var iab le and a l l the v a r i a b l e s chosen as independent; and then, i n dropping s u c c e s s i v e l y the independent var iab le w i t h the least p a r t i a l F test corresponding to i ts p a r t i a l r e g r e s s i o n coeff ic ient f r o m a l l the r e m a i n i n g independent v a r i a b l e s . Then to calculate and test a new m u l t i p l e r e g r e s s i o n equation with the r e m a i n i n g v a r i a b l e s . The p a r t i a l F test is defined i n standard textbooks on m u l t i p l e r e g r e s s i o n such as D R A P E R and S M I T H ' S book (1966, p. 71) which also d iscusses the advantages and disadvantages of the different methods ex is t ing for select ing the best r e g r e s s i o n equation. The procedure is repeated u n t i l there r e m a i n s only one independent v a r i a b l e . The var iab les were cons idered as contr ibuting s igni f i cant ly to the v a r i a t i o n of the dependent v a r i a b l e when re jec t ing one of them s i g -n i f i cant ly l o w e r e d the m u l t i p l e c o r r e l a t i o n coeff ic ient . 1) Wing Length (a) The altitude does not contribute s igni f icant ly to wing length v a r i a -b i l i t y and the pred ic t ion equation becomes : a = 7. 29 - 0. 0099 (Long. - 128. 87) + 0.12 (Lat . - 53. 21) r 2 = 0.187 r - 0 .433* 2) Wing Width (b) Only the longitude contributes s igni f icant ly to b v a r i a b i l i t y and the pred ic t ion equation becomes : b = 3.56 4- 0.028 (Long. - 128.87) r 2 = 0.209 5 t . J> NI-r = 0.457 3) Seed Length (c) No re la t ionship whatever has been demonstrated. . 4) Seed Width (d) The p r e d i c t i o n equation reduces to: d = 1.83 ¥ 0.014 (Long. - 128.87) - 0.000060 ( A l t . - 400.25). r 2 = 0.377 r = 0 . 6 1 4 * * * 5) Cone Length The pred ic t ion equation reduces to cone = 55.90 - 0.0077 ( A l t . - 400.25) r 2 = 0.333 •A. .A. .J„ r = 0.577""'" To s u m m a r i z e the equations obtained, one can say that the pattern of v a r i a t i o n d i f fers f r o m one character to another. The percent -age of v a r i a t i o n accounted for by the geographical coordinates is not too high, leaving m u c h of the v a r i a t i o n not expla ined. The explanations are d i f f i c u l t : lack of c l i n a l v a r i a t i o n , p r e c i s i o n too low, etc. Wing length is a m u l t i p l e function of longitude and lat i tude, wino; width p r i m a r i l y 68 increases wi th longitude, cone length p r i m a r i l y increases when e l e v a -t ion decreases . Seed width i s re la ted to longitude and altitude and i s independently re la ted to longitude, latitude and al t i tude. The r e g r e s s i o n equations for seed width a r e : d = 1.83 H- 0.015 (Long. - 128.87); r = 0 .497** d = 1. 83 + 0. 012 (Lat . - 53.21); r = 0 .347* d = 1.83 + 0.000076 ( A l t . - 400.25); r = 0.406* The provenance means have been plotted against longitude, latitude and elevat ion of the place of o r i g i n of the provenances neglect ing the r e g i o n s . A close examinat ion of the graphs obtained shows that two groups of provenances could exist over lapping i n about the same range of latitude or longitude, for each t r a i t . T h e r e f o r e , a separate r e g r e s s i o n l ine has been adjusted for each group and for each c h a r a c t e r i s t i c s tudied. No s ignif icant r e g r e s s i o n has been found except that two r e g r e s s i o n l ines exist for cone length on longitude. The corresponding equations a r e : cone = 54.84 - 0.018 (Long. - 126.21); r = 0 .535* cone = 54. 71 f 0. 019 (Long. - 130. 38); r = 0. 423* These two r e g r e s s i o n l ines correspond to two groups of provenances of longitudes east of 127° and west of 1 3 0 ° . In spite of the fact that no s ta t i s t i ca l adjustment has been p os s ib le , it would be in teres t ing to further test the hypothesis of a large change i n v a r i a t i o n pattern in seed c h a r a c t e r i s t i c s between the latitudes 52° and 54° and the longitude 12 7° and 1 3 0 ° . Two large groups of p r o v -enances could be dis t inguished: one group with latitudes below 52° and one group wi th latitudes l a r g e r than 5 4 ° ; with none i n between. Note also that the latitude and longitude of the places of o r i g i n are c o r r e l a t e d to some extent (r = 0. 78). Table X I shows the c o r r e l a t i o n coeff ic ients of the seed and cone t ra i t s studied calculated on the bas is of the provenance m e a n s . These corre la t ions are n a t u r a l l y different f r o m those calculated on the bas is of tree m e a n s . T h i r t y - n i n e c o r r e l a t i o n coeff ic ients have been c a l -culated between the t ra i t s c and d, one for each provenance, and as an example plotted against latitude of the place of o r i g i n ( F i g . 8). On the bas is of latitude of o r i g i n , two groups of provenances seem to be d is t inc t . The -Vancouver Is land reg ion and the more nor thern provenances . It seems as i f , wi th in each group, the c o r r e l a t i o n coe f f i -cients increase with lat i tude, the slopes being s i m i l a r . r l.OOi. •23 • 3 2 •34 , ' • 2 9 •39 , - ' • 3 8 .18 ^ ' --<13 . 6 • 5 , ' ' . 2 6 , ' , 0 5 •27 .24 • . 1 1 . ' - " 3 0 . 3 1 , - • 3 3 •3 1 0 , ' ' i ' 4 « 4 9 SO It" 3 2 S 3 . 3 4 3 9 lot 0 '/<> 5b Fig.S. Relationships between the correlations of seed length with seed width and l a t i t u d e of the place of o r i g i n . o 71 C H A P T E R 3 M U L T I V A R I A T E S T A T I S T I C A L M E T H O D S A C O M P A R A T I V E R E V I E W O F T H E M A T H E M A T I C A L T H E O R I E S B E H I N D S O M E T E C H N I Q U E S F R O M A B I O L O G I S T ' S P O I N T O F V I E W Introduction The five t ra i t s studied did not resu l t i n a c lear c l a s s i f i c a -t ion of the parent stands of the Sitka spruce provenances and it was , therefore , decided to use a m u l t i v a r i a t e approach as the simultaneous considerat ion of s e v e r a l charac ters may resu l t in a meaningful c l a s s i f i c a t i o n ( C O O L E Y and L O H N E S , 1966; H O R T O N , et a l . , 1968). C l a s s i f i c a t i o n should be dist inguished f r o m d i s c r i m i n a t i o n or c l a s s i n g techniques and f r o m d i s s e c t i o n . The latter techniques are concerned wi th compar ing the p r o f i l e of an i n d i v i d u a l with that of groups a l ready def ined. The f o r m e r techniques a i m at c l a s s i f y i n g indiv iduals into different groups in the f o r m of a s ignif icant pattern or a few group conste l la t ions . The " g r o u p s " wi thin a constel lat ion must n e c e s s a r i l y be c l o s e r to one another, in some sense, than those belonging to different conste l la t ions . It i s not poss ib le to cover the numerous techniques designed for c l a s s i f y i n g i n d i v i d u a l s , groups or entit ies of higher h i e r a r c h a l rank. So many techniques designed for c l a s s i f y i n g i n d i v i d u a l s , groups, etc. have been proposed that confusion is widespread i n the f i e l d 72 ' of n u m e r i c a l c l a s s i f i c a t i o n as to the names , v a l i d i t y , usefulness , s i m i -l a r i t i e s , e f f i c iency , s ta t i s t i ca l or mathemat ica l theories of these methods. The opinions of the authors often d iverge . B L A C K I T H and R E Y M E N T ' s opinion (1971) on factor analys is is i n opposit ion to that of C A T T E L L (1965). The f o r m e r authors , f u r t h e r m o r e , propose in their book a b i o l o g i c a l in terpreta t ion of m u l t i v a r i a t e s tat is t ics by re jec t ing the l i m i t s which could be imposed by the s ta t i s t i ca l theory. D A G N E L I E (1966) has outl ined the charac ters of s e v e r a l methods of n u m e r i c a l c l a s s i f i c a t i o n , the needs and the possible r e s e a r c h approaches i n c l e a r i n g the f i e l d . Two approaches are b a s i c a l l y p o s s i b l e : on the one hand, the ex is t ing techniques can be compared ; on the other, 0 an opt imum solution can be worked out. In both cases , some " o p t i m u m " c r i t e r i a must be decided upon. A n opt imum c l a s s i f i c a t i o n ( D A G N E L I E , 1966) could be defined as the one which resul t s i n , for a given number of c l a s s e s , one of the fo l lowing - not exc lus ive - c h a r a c t e r i s t i c s : the genera l ized distances between individuals wi th in c lasses are m i n i m u m ; the genera l ized distances between c lasses are m a x i m u m ; the average genera l ized distance within c lasses is m i n i m u m ; the average genera l ized distance between c lasses i s m a x i m u m . Ideal ly, the c o m p a r i s o n of the methods should include the cost i n d o l l a r s of running these techniques on a computer and should be applied on as m u c h data as p o s s i b l e . A sound interpreta t ion of the resu l t s of the use of a given method is only poss ib le when one/completely understands the theoret ica l and s t a t i s t i ca l background of the techniques used. We s h a l l , therefore , 73 r e v i e w and compare the m a t h e m a t i c a l theory b u r i e d in the methods that were avai lable to u s . To avoid a lengthy exposi t ion, we w i l l suppose that the reader i s f a m i l i a r with m u l t i v a r i a t e s ta t i s t i c s . F o r a complete e x p o s i -t ion of the mathemat ica l theories of the methods used , the reader is r e f e r r e d to the books of M O R R I S O N (1967), K E N D A L L (1966) and A N D E R S O N (1958). The capi ta l le t ters A , B , C , e t c . , w i l l be used for m a t r i x and s m a l l le t ters x, y , z , e t c . , for vector notation; f u r t h e r m o r e , the le t ters w i l l be under l ined A , x, etc. 3. 1 M u l t i v a r i a t e Anova ( M A N O V A ) 3 .1 .1 M u l t i v a r i a t e genera l iza t ion of a one-way univar ia te A N O V A The steps are the fo l lowing (after C O O L E Y and L O H N E S , 1966): (1) Tes t of homogeneity of the d i s p e r s i o n s ; (2) Test of homogeneity of the population centroids by us ing W I L K S 1 L a m b d a c r i t e r i o n ("X ): A = | WS | / | TS | I I : determinant notation WS : pooled wi th in group sums of squares and c r o s s products m a t r i x TS : total deviat ion sums of squares and c r o s s products m a t r i x The test of s ignif icance uses a C h i - s q u a r e approximat ion of •i •74: B A R T L E T T or a F approximat ion of R A O . No r e a l grouping of the bas i c populations i s poss ib le without a m u l t i p l e c o m p a r i s o n test of the cent ro ids . 3 . 1 . 2 G e n e r a l case The general procedure for any M A N O V A is given by S E A L (1964). The bas i c l inear model i s : X = Z ' . B + E Thus , there are N p - v a r i a t e observat ions and Q e r r o r free observat ions ( = t reatments , e t c . ) . It involves the computation of a " reduced m o d e l " and of a var ia te approximate ly d is t r ibuted as C h i - s q u a r e which is used as a bas is for a test of s igni f i cance . S E A L gives an extension of the S C H E F F E ' s S method of m u l t i p l e c o m p a r i s o n of the mean v e c t o r s . T h e r e f o r e , it i s poss ib le to f ind out group constel lat ions by ca lculat ing contrasts between a l l the p a i r s of centro ids . The S test is v e r y robust and might be used m o r e frequent ly . 3. 2 P r i n c i p a l Component A n a l y s i s ( P C A ) P C A is the technique most used and most known f r o m the m a t h e m a t i c a l point of v i e w . T h i s method should be dis t inguished f r o m factor ana lys i s ( F A ) , ( S E A L , 1964 and C A T T E L L , 1965). P C A has been in terpre ted by S E A L (op. cit) as a procedure for applying a l inear t r a n s f o r m a t i o n to the o r i g i n a l var iates represented by the vector x . T h i s t r a n s f o r m a t i o n i s : 75 _y = A . x The t r a n s f o r m a t i o n is orthogonal and the new var ia tes are m u t u a l l y independent and f o r m new axes of coordinates of m a x i m u m v a r i a t i o n , c a l l e d p r i n c i p a l components. A sample of N observat ions of p var ia tes (N p - v a r i a t e observations) i s s u m m a r i z e d i n a few var ia tes which are not c o r r e l a t e d and account for a m a x i m u m of the total p r i m i - . tive v a r i a t i o n . CATTELL (op. c i t . ) has in terpre ted PCA as a v e r y p a r t i -cular case of FA where the s imple c o r r e l a t i o n m a t r i x R ca lculated f r o m the p var ia tes i s equated as f o l l o w s : R = v_ • v 1 5/2 V = factor m a t r i x and _V = M . L where M is the m a t r i x of latent 'vector s of R L is the diagonal m a t r i x of latent roots of R PCA i s based on the der iva t ion of V f r o m a s o - c a l l e d "non r e d u c e d " m a t r i x R, i . e . , a m a t r i x R where the diagonal coeff ic ients have a l l been equated to 1. T h i s approach i s s i m i l a r to that of DAGNELIE (i960) where the o r i g i n a l var iates z . are explained by means of r f a c t o r s ; J also ca l l ed components common to a l l the var ia tes Z j and a spec i f ic var ia te or factor V j . The m o d e l i s : z . = a. x +• +• a x +• u v . J J l 1 j r r j j 76 In m a t r i x notation, the m o d e l can be expressed as : Z = A . X' + U . V It should be noted that P C A can be c a r r i e d out on the var iance - c o v a r i a n c e m a t r i x of the observat ions or on the c o r r e l a t i o n m a t r i x obtained when the observat ions have been s tandardized. The components obtained f r o m these two different m a t r i c e s are not the same and it is not poss ib le to pass f r o m one set of components to the other by sca l ing the coeff ic ients of the components. The covar iance m a t r i x should be p r e f e r r e d when the measurements were made i n s i m i l a r units and when tests are forecast . . It i s also important to note that the jth p r i n c i p a l component of the sample of p var iate observat ions i s the l inear component: y. = a . . x. +• +• a . x 3 i j i P J P where the coeff ic ients are the elements of the c h a r a c t e r i s t i c vector of the sample covariance m a t r i x S corresponding to the j th larges t c h a r a c t e r i s -t ic root l j . T h i s i s , of course , equivalent to the D A G N E L L E ' s m o d e l of above, where the speci f ic factor i s not cons idered . The sample var iance of the j th component i s 1., and the total sys tem var iance i s thus 1^ . +- . . . i - 1 = tr S. The importance of the j th component i n a m o r e pars imonious d e s c r i p t i o n of the sys tem is thus m e a s u r e d by : 1. / tr S J If the components have been extracted f r o m the c o r r e l a t i o n m a t r i x R rather than S, the sum of the c h a r a c t e r i s t i c roots w i l l be tr R = p, and the propor t ion of the total var iance i n the scatter of d imens ionless standard scores attributable to the jth component w i l l be l - / p . The sum of the squared corre la t ions a . . V 1 of the v a r i a b l e s on that component w i l l , of course , be the component var iance l j (a:fter M O R R I S O N , 1967). F i n a l l y , i t must be r e m e m b e r e d that as the t rans format ion m a t r i x i s orthogonal , the t r a n s f o r m a t i o n i s distance p r e s e r v i n g between the points i n the new space. Since the angles of a t r iangle are determined by the length of i ts s ides , the p r i n c i p a l component t rans format ion also p r e s e r v e s the angles ( A N D E R S O N , 1958). The disadvantages of P C A are numerous : * S t r i c t l y , P C A should only be used to s u m m a r i z e a m a s s of p -variate observat ions resu l t ing f r o m the sampl ing of one m u l t i v a r i a t e n o r m a l u n i v e r s e . M a n y authors have appl ied P C A to a set of observat ions a l ready s t ruc tured i n groups or for testing the hypothesis that groups do ex i s t . It i s questionable i f P C A is the adequate tool for such endeavours . If we want to construct tests of hypotheses and confidence in terva ls for the population roots , the d i r e c t i o n cos ines , etc. , it i s necessary to sup-pose that the sample be drawn f r o m a m u l t i v a r i a t e n o r m a l population whose covar iance m a t r i x has a speci f ied covar iance s tructure ( M O R R I S O N , 1967). The f o r m s of the components are not invar iant under changes in the scales of the responses . Thus , it is doubtful whether P C A should be appl ied to var ia tes m e a s u r i n g different ent i t ies , e . g . , a combinat ion of 78 lengths, weights and dichotomous var ia tes (variates taking the values 0 or 1), ( M O R R I S O N , op. c i t . ; S E A L , dp. c i t ) . A n attempt to avoid this d i f f i cu l ty i s to standardize a l l the var iates by div iding each by its e s t i m -ated standard deviat ion. However , there are two m a i n disadvantages i n us ing s tandardized v a r i a t e s : (a) the s tandardized observat ions are only approximate ly n o r m a l , and (b) the new var ia tes are put on an equal footing, thus r e s u l t i n g i n a d i s tor t ion of the measurements ( S E A L , op. c i t . ) . No r a t i o n a l c r i t e r i a exist for deciding when a suff ic ient p r o p o r -tion of the var iance of a l l the observat ions has been accounted for by the components, nor can p r o v i s i o n be made for components that are a t t r i b u -table only to the sampling v a r i a t i o n of the ind iv idua l responses . P C A is thus m e r e l y an orthogonal t rans format ion rather than a m o d e l for study-ing covariance s tructure ( M O R R I S O N , 1967). F u r t h e r m o r e , i f the off diagonal elements of a c o r r e l a t i o n m a t r i x are approximate ly equal , the in terpre t ive value of P C A is dubious ( S E A L , op. c i t . ) . F i n a l l y , the common factors are d is tor ted because of the a r t i -f i c i a l conditions imposed of being n o n - c o r r e l a t e d ( C A T T E L L , op. c i t . ) . It i s often di f f i cul t or subjective to interpret b i o l o g i c a l l y the new factors found,as a look at the numerous applicat ions of P C A can show; The advantages are the f o l l o w i n g : - S i m p l i c i t y of the mathematics i n v o l v e d . - Orthogonal i ty of the new space. - Distance p r e s e r v i n g t r a n s f o r m a t i o n . 1 7:9 -- Rotat ion of the factors p o s s i b l e . - Test of the latent roots p o s s i b l e . - P o s s i b i l i t y of ca lculat ing the centroids i n the new coordinates and to get a v i s u a l representat ion of the possible grouping of the popula-tions to be c l a s s i f i e d . F o r an example , see J E F F E R S and B L A C K (1963). 3 .3 F a c t o r A n a l y s i s (FA) A c c o r d i n g to C A T T E L L (1965), F A is p e r f o r m e d when the latent vectors and latent roots of a s o - c a l l e d reduced m a t r i x R^ are es t imated . R^ is reduced when the s o - c a l l e d " c o m m u n a l i t i e s " , es t imated in some way, replace the diagonal coeff ic ients of the o r i g i n a l m a t r i x R . U s i n g this "open m o d e l " i s m o r e l o g i c a l , but resu l t s i n many m o r e d i f f i -cult ies than with P C A . . Such d i f f i cu l t i es a r e : the es t imat ion of the c o m -m u n a l i t i e s ; the es t imat ion of the number of fac tors to be used; the i n -determinat ion of the spec i f i c factors and of the e r r o r s . C A T T E L L (op. c i t . ) has given some ru les for determining the number of factors to start with and the c r i t e r i a for deciding which rotat ion should be used. . He has proposed to f i r s t decide upon the number of fac tors , then to adjust the c o m m u n a l i t i e s . A s there i s an inf in i ty of equivalent factor m a t r i c e s , the next p r o b l e m is to f i x up his m i n d on a meaningful rotat ion i n o r d e r to get a meaningful factor m a t r i x . The c r i t e r i a given have their own defects and d i f f i c u l t i e s . C A T T E L L bel ieves that oblique factors are the rule in nature . F A is an interes t ing tool , a hypothesis generating method. It is m o r e a method for c l a s s i f y i n g the var iab les but not, as such, a tool for generating c lus ters of i n d i v i d u a l s . C l u s t e r search or c l a s s i f i -cation of entit ies could be done after an analys is of the factors respons ib le for the s tructure of the universe under study ( C A T T E L L , op. c i t . ) . 3. 4 D i s c r i m i n a n t Funct ion A n a l y s i s (DF) and Canonica l A n a l y s i s (CA) The concept of l inear d i s c r i m i n a n t function was o r i g i n a l l y introduced by F I S H E R i n 1936 as a solution to the p r o b l e m of c l a s s i f y i n g an observat ion into one of : two predetermined groups . In the case of two populations with a common covariance m a t r i x , but different mean vectors i f x^ and x-, are the sample mean vectors and W the pooled est imate of the covar iance m a t r i x , the l inear d i s c r i m i n a n t function is y = (x - x 2 ) ' W _ 1 x _ x being the vector of var ia tes The l inear d i s c r i m i n a n t function corresponds to the l inear combinations of the var iab les which assures the m a x i m u m value to the r a t i o between the "between" sub-populat ion sample var iance of the new variate y and the " w i t h i n " sub-populat ion sample v a r i a n c e . A c t u a l l y , this rat io corresponds to the HOT E L L I N G 1 s T 2 s tat is t ic ( M O R R I S O N , 1967). The l inear d i s c r i m i n a n t function is connected with the 2 M A H A L A N O B I S ' genera l ized distance D by the re la t ionship D 2 = d ' a -81 where d ' is the difference between the two sample mean vectors and a the vector of coeff ic ients of the d i s c r i m i n a n t funct ion. 2 F o r a number h of p - v a r i a t e groups, D is defined as P p . . D = Z Z (x - x ) (x - x ) i=l j=l l k x * jk J* or D 2 = Z . Z • u) 1 J r i d d. i J l J i j a) i s the adequate element of the inver ted sample wi th in covariance m a t r i x . 2 D is not v e r y sensi t ive to the n o n - n o r m a l i t y of the o r i g i n a l v a r i a t e s , but the within covariance m a t r i c e s ( = dispers ions) should be homogeneous ( M I L L L E R and T O M A S S O N E , 1969). The d i s c r i m i n a n t function is in t imate ly l inked with the p r o b -l e m of c l a s s i f y i n g an i n d i v i d u a l into one of two n o r m a l m u l t i v a r i a t e popu-lat ions prov ided that some " a p r i o r i " p r o b a b i l i t i e s , cost of m i s c l a s s i f i -cat ion, etc. are known. F o r a complete coverage of the problems of c l a s s i n g indiv iduals i n one of s e v e r a l groups, see A N D E R S O N (1958), (Chapter 6). The m u l t i p l e d i s c r i m i n a n t functions are computed as latent vectors associated with the latent roots of the determinantal equation J D J w i t h : r-l D | = | W " 1 B - Xj. | = 0 (1) " 82 T h i s equation can also be wr i t t en as : | D | = | B - X W j = 0 B i s the m a t r i x of var iance - covar iances between the groups or un iverses W is the m a t r i x of v a r i a n c e - c o v a r i a n c e s within the universes X i s a var iab le represent ing the latent roots of the d i s s y m e t r i c m a t r i x W B I i s the usua l identity m a t r i x The w o r d canonical analys is has been used i n connection wi th quite different techniques, adding to the confusion of the user of m u l t i v a r i a t e techniques. F o r K E N D A L L and S T U A R T (1966), it includes P C A , canonical cor re la t ions and F A . B L A C K I T H and R E Y M E N T (1971) separates the canonical var ia tes f r o m d i s c r i m i n a n t function analys is (their Chapters 7 and 8) and dis t inguish (p. 49) between d i s c r i m i n a n t analys is as d e s c r i b e d by A N D E R S O N (1958) and canonical d i s c r i m i n a n t funct ions. . In fact, their " canonica l v a r i a t e s " (Chapter 8, op. c i t . ) are only the usua l d i s c r i m i n a n t var ia tes corresponding to the usual d i s c r i m i -nant axes generated by the equation (B - A W) U = O i n our notation, which corresponds to their equation (8.2), p. 89: 83 S E A L (op. c i t . ) has proposed his canonical analys is by c r i t i c i z i n g the genera l ized distance technique which i s a by-product of the d i s c r i m i n a n t function a n a l y s i s . He contends that: (1) the compar isons of h p - v a r i a t e un iverses where h ( p should be made i n a space of h - l dimensions rather than in a space of p d imens ions , and (2) the distance technique depends on a l l p var ia tes being m e a s u r e d i n the same uni ts . T h i s last c r i t i c i s m does not hold as the M A H A L A N O B I S 1 genera l ized distance is independent of the scale of the var iab les ( C O O P E R , 1963). The M A H A L A N O B I S ' distance remains unchanged no matter what (non-singular) change of var iab les one m a k e s . . S E A L (op. c i t . ) proceeds as f o l l o w s : Suppose that we have h p - v a r i a t e n o r m a l u n i v e r s e s r e p r e s e n -ted by s a m p l e s . E a c h universe is c h a r a c t e r i z e d by its centro id and its v a r i a n c e - c o v a r i a n c e m a t r i x . Two steps are then dis t inguished: test for the equality of v a r i a n c e - c o v a r i a n c e m a t r i c e s c o m p a r i s o n of the h universes It i s poss ib le to f ind a t rans format ion such that the f i r s t axis is i n c l i n e d in the d i r e c t i o n of the greatest v a r i a b i l i t y between the mean v e c t o r s , then that the second ax i s , at r ight angle to the f i r s t , i s to be i n c l i n e d i n the d i r e c t i o n of the next greatest v a r i a b i l i t y , and so on. T h i s t rans format ion , i s , let us say: 84 x = vector of the o r i g i n a l measurements y_ = vector of the canonical var iates U = m a t r i x of the t rans format ion Because the elements of the t rans format ion m a t r i x are indeterminate , two conditions are i m p o s e d : the new canonical var ia tes should be u n c o r r e l a t e d ( i . e. , wi th zero covariances) and of unit v a r i a n c e s . C A thus reduces i t s e l f to f inding the latent roots V* and the associated eigenvectors of the fol lowing determinantal equation: W _ 1 / 2 B W " 1 / 2 - = 0 O (2) B i s the m a t r i x of v a r i a n c e - c o v a r i a n c e s between the universes or the groups W is the m a t r i x of v a r i a n c e - c o v a r i a n c e s wi th in the u n i v e r s e s V i s a var iab le represent ing the latent roots I i s the usual identity m a t r i x There are a m a x i m u m of p roots and the p roots are d i s -t inguishable when p ^. h - l . When p > h - 1, there are p - h f 1 zero roots and h - l dist inguishable roots , i . e . , different f r o m zero ( B L A C K -IT H and R E Y M E N T , 1971). If V is the m a t r i x of e igenvectors corresponding to equation (2), then V is generated by the equation ( W ~ l / 2 B W " 1 / 2 - A I) V = O o-5 A i s the diagonal m a t r i x of the c h a r a c t e r i s t i c roots of (2) Then the t rans format ion m a t r i x U i s obtained, according to S E A L (op. c i t . ) by the equation 1 IZ w U = V (3) The centroids of the different universes can be expressed in canonical f o r m and their mutual pos i t ion v i s u a l i z e d i n a d i a g r a m . C o n f i -dence c i r c l e s can be drawn around the points represent ing the u n i v e r s e s i n the different graphs . A . v i s u a l c l a s s i f i c a t i o n i s , therefore , p o s s i b l e . C i r c l e s are appropriate instead of e l l ipses because C A t rans forms the within sample e l l i p s o i d s of scatter into spheres , m o r e p r e c i s e l y h y p e r -spheres . A c c o r d i n g to S E A L , in C A , the p var ia tes may be different types of m e a s u r e m e n t s . The new dimensions r e q u i r e d would be less than those r e q u i r e d by P C A . C A is thus based on the ca lculat ion of the latent roots and the associated latent vectors of the m a t r i x C wi th -1 /2 -1 /2 C = W B W while D F is based on the es t imat ion of the eigenvalues and eigenvectors of the m a t r i x D with -1 D = W B -1 /2 . . -1 /2 -1 W B W is general ly different f r o m W B . H o w -ever , they have ths same c h a r a c t e r i s t i c roots A . 6-6 P r o o f : The n o n - z e r o c h a r a c t e r i s t i c roots of the product A B are equal to the non-zero roots of B A' (MORRIS'SON, 1967, p . 62); but the fo l lowing m a t r i c e s have the same c h a r a c t e r i s t i c r o o t s : W - 1 ^ 2 B W - " " / 2 , B W " 1 ^ 2 W " 1 ' ' 2 = B W " 1 ; therefore W~1^2 B W'1^2 and B W _ 1 or W _ 1 B have the same eigenvalues . However , the associated eigenvectors are general ly di f ferent . Le t U = u n u i l - P be a m a t r i x of e igenvectors c o r r e s -ponding to D F . U i s generated by the fo l lowing set of equations: ( W - 1 B - A I) U = O S i m i l a r l y , V is generated by the set of equations: -1 /? -1/2 (W B W A / - A . I ) V = O corresponding to C A . -1 N . - 1 /2 -1 /2 T h e r e f o r e , W B U = U A and W B W V = V ^ . But W ~ l / 2 B ( W _ 1 / 2 V) = V A , thus B ( W ' 1 / 2 V) = ( W 1 / 2 V) y\ = W ( W " 1 / 2 V) A. and W " 1 B ( W _ 1 / 2 V) = W " 1 / 2 V A. i therefore , U = W _ 1 / 2 v or y = w 1 / / 2 U. 1/2 V = W U But this equation is ident i ca l to equation (3). Consequently, C A . a c c o r d -ing to S E A L and D F in the usual meaning p e r f o r m exactly the same t rans -format ion and this t rans format ion is not orthogonal , in genera l . U is not orthogonal , i n genera l . £.7 P r o o f : If U was orthogonal , then | U | = - 1 and U ' U = 1. I being the identity m a t r i x as u s u a l ; f u r t h e r m o r e | U 1 U | = | Ij = 1 ( M O R R I S O N , 1967, p . 59). If | U | = - 1 and U 1 U = I, then | U ' W U | = | U ' | | w l l u | = | Wl but | W| 4 * 1. But U» W U = I, then |U* W U | = 1 which is con-t r a d i c t o r y . T h u s , i n spite of the fact that U 1 W U i s equal to I, U i s not general ly orthogonal . 1/2 But even i f W and U are not orthogonal , V = W . U is an orthogonal m a t r i x . The t r a n s f o r m e d values have thus the fo l lowing covar iance m a t r i c e s , as it i s easy to demonstrate : W i t h i n : U ' W U = I_ Between: U 1 B U = A The new v a r i a b l e s are thus uncorre la ted and of var iance 1. E a c h group has been so t r a n s f o r m e d as to take the shape of an hypersphere . One obvious consequence of this non-orthogonal t r a n s f o r m a -tion is that, unl ike i n P C A , the distances and angles i n the new space are changed. T h e r e f o r e , the graphs represent ing the indiv iduals i n the new space are pure conventions as the new axes are not p e r p e n d i c u l a r . Canonica l a n a l y s i s , according to S E A L , i s s i m p l y another in terpreta t ion of D F . C A or D F take into account the nested levels of v a r i a t i o n which might ex is t : between groups and within groups. F u r t h e r -m o r e , i t can be used to calculate the s o - c a l l e d M A H A L A N O B I S 1 d is tances . T h i s fact seems to have been over looked by S E A L who introduces his C A by c r i t i c i z i n g the genera l ized dis tances . A n important c h a r a c t e r i s t i c of D F or C A is that: t race of W _ 1 B = trace of W~1^2 W ' 1 ^ 2 B = -1 /2 -1 /2 -1 trace of W B W = trace of B W = t race of A Consequently, the trace of A is the sum of the var iances between groups of the different var iab les studied and the percent trace of a c h a r a c t e r i s t i c root can be in terpre ted as the f rac t ion of the total v a r i a t i o n between groups explained by that root . B L A C K I T H and R E Y M E N T (1971) contend that i t i s not because some roots do not contribute s igni f i cant ly to the total sum of squares in the analys is of d i s p e r s i o n that they must be d i s r e g a r d e d . The larges t of the canonical roots would generate a vector which often r e p r e -sents s ize v a r i a t i o n in m o r p h o m e t r i c analyses , but s ize v a r i a t i o n would se ldom be an object of study i n i t se l f . However , their argumentation is not convinc ing . If some roots are not s t a t i s t i c a l l y s igni f i cant , i t i s d i f f i cu l t to understand how one could p o s s i b l y base r e l i a b l e b i o l o g i c a l in terpre ta t ion on the v a r i a b i l i t y associated with the corresponding e igen-v e c t o r s . 3. 5 Resul ts and Conclus ions It is obvious that there is a p r e s s i n g need to compare the c l a s s i f i c a t i o n techniques under a l l poss ib le aspects , even those tech-niques which are supported by a substantial body of s ta t i s t i ca l theory. L a c k of t ime and of money have resu l ted in our r e s t r i c t i n g cT9 the c o m p a r i s o n to the fo l lowing techniques: C A and its d e r i v a t i v e s , the M A H A L A N O B I S 1 distances and the dendrograms ; Stepwise D i s c r i m i n a n t Funct ion A n a l y s i s ; P C A . The five cone and seed c h a r a c t e r i s t i c s m e a s u r e d have been used to do this c o m p a r i s o n . These t ra i t s are supposed to be n o r m a l l y d is t r ibuted and the re la t ionships between their means and var iances are not so s t r i c t as to impose some t rans format ion of the o r i g i n a l var ia tes (see Section Z. 1). The "opera t iona l taxonomic u n i t s " ( "OTU") i n the sense of S O K A L and S N E A T H (1963) are the provenances . Only the provenances wi th 15 trees w i l l be cons idered , unless otherwise mentioned. The trees are cons idered as repet i t ions . In o r d e r to be able to use C A as proposed by S E A L (1964), an o r i g i n a l computer p r o g r a m m e has been developed and tested. F u r -t h e r m o r e , a computer p r o g r a m m e developed at the B i o m e t r i c Station of the Nat iona l Centre for F o r e s t R e s e a r c h at Champenoux, F r a n c e , under the d i r e c t i o n of D r . C . M I L L I E R , has been adapted to the U . B . C . 's M i c h i g a n T e r m i n a l Sys tem. Both p r o g r a m m e s r e q u i r e d painstaking efforts during s e v e r a l months and further p r o g r a m m i n g on the other p r o g r a m m e s of m u l t i v a r i a t e analyses exis t ing i n the U . B . C . 's Computing Centre was p r e c l u d e d . The M A N O V A has not been inc luded i n the c o m -p a r i s o n as the avai lable p r o g r a m would have to be completed wi th a d i s -p e r s i o n test and a mul t ip le c o m p a r i s o n test of the centroids i n order to 90 become meaningful . F A has not been t r i e d as such because there i s no spec i f i c p r o g r a m to use this method and because this method is delicate to use and should be the subject of a thesis i n i t se l f . 3 .5 .1 C A according to S E A L (1964) The p r o g r a m m e wr i t t en in F O R T R A N IV for the I. B . M . System 360 - M o d e l 67 computer of the U . B . C . computing centre fol lows exactly the procedure outl ined by S E A L (1964, Chapter 7). It gives the fol lowing output: (1) the covar iance and c o r r e l a t i o n m a t r i c e s and their determinant for each group of i n d i v i d u a l s ; (2) the T , B , W m a t r i c e s ; (3) test of the homogeneity of the within covariance m a t r i c e s ; (4) test of the o v e r a l l di f ferences of the different groups M A N O V A ' s W I L K S L a m b d a c r i t e r i o n tested by R A O ' s F approximat ion ; (5) B A R T L E T T ' s test of the latent roots ( S E A L , op. c i t . , p . 135); (6) the canonical functions and the centro id vectors i n the new canonical space for each group; (7) the plotting of the groups in a chart f o r m e d by the axes with the greatest s ignif icant roots . The provenances have been grouped into zones which have been analyzed separately i n order to respect as much as poss ible the bas i c hypothesis of homogeneity of d i s p e r s i o n m a t r i c e s . These are the same regions but with some deletions because only the 91 ' • ' o T A B L E XII C A N O N I C A L F U N C T I O N S F O R T H E F O U R o R E G I O N S S T U D I E D C O E F F I C I E N T S O F T H E O R I G I N A L V A R I A B L E S The canonical functions are ranked according to the decreas ing o r d e r of magnitude of their root 0 Number of the Region canonical V a r i a b l e s functions 1 2 3 4 5 1 0.0306 0. 668 3. 518 -4.641 0. 663 2 .0. 331 -0 .557 3. 712 2. 789 -0 .140 2 3 0. 801 -0 .135 -1 .563 6'. 850 0.0286 4 0. 936 1. 539 -1.511 -3 .939 -0 .0450 5 -1 .222 . 3. 331 0.783 -0 .155 -0.0191 1 -0 .00502 1.477 -2 .285 -1 .996 0. 120 2 0, 151 -0 .112 3. 799 4. 079 -0 .0230 3 3 - 0 . 513 1.012 -3 .432 6. 332 0. 0109 4 1. 227 0. 548 -3 .056 0. 209 - 5 . 0990 1 0. 577 -0 .565 2. 345 1. 915 0.0488 2 -1 .051 1. 754 - 3 . 220 4. 805 -8 .038 4 3 0. 936 -3 .020 -3 .761 7. 312 -0 .00724 4 -1 .222 -1 .313 3. 257 1. 595 0. 141 5 0.0849 0. 777 0. 399 5.440 - 0 . 104 1 -0 .366 -0 ,0783 0. 243 4. 591 0. 120 2 -0 .386 -0 .276 -4 .405 6. 330 -0.0111 5 3 -0 .997 2. 262 2.849 0. 670. -0 .0720 4 1. 000 -0 .966 -0 .0140 5. 555 -0 .0652 5 0. 282 2. 538 -2 .790 -1 .835 0.0115 p o p u l a t i o n s w i t h 15 t r e e s have been i n c l u d e d . A s u n i t of o b s e r v a t i o n s , the means of each t r e e w e r e c o n s i d e r e d . T h e s e means w e r e c a l c u l a t e d f r o m the s a m p l e s of f i v e seeds and ten cones. The c a n o n i c a l functions f o r the fou r r e g i o n s s t u d i e d a r e shown i n T a b l e X I I . V a n c o u v e r I s l a n d Zone (= R e g i o n 1 f 2) = pr o v e n a n c e s 3, 4, 30, 31, 32, 33, 34. ° ° o a Number of groups: 7. N u m b e r of p - v a r i a t e o b s e r v a t i o n s : 105. , T e s t of e q u a l i t y of the group d i s p e r s i o n s : U = 3.93. T h i s i s v e r y h i g h l y s i g n i f i c a n t and the group d i s p e r s i o n s a r e not equal, nor ar.e the c o r r e l a t i o n m a t r i c e s . There°are thus, d i f f e r -ences i n s i z e and/or o r i e n t a t i o n of the d e n s i t y e l l i p s o i d s 0 f o r the d i f f e r e n t groups. T e s t of the o v e r a l l s i g n i f i c a n c e of the d i f f e r e n c e s between groups: •F (30, 378) = 2.84 B A R T L E T T's'Test of the L a t e n t R o o t s : . Root s R e m o v e d Root s • C h i - S q u a r e •D.F. S i g n i f i c a n c e P e r c e n t - ' age t r a c e 0 0.65- • 79.79 30 ' •A. »u f *T« ~£ . 65.97 • 1 • •.0.16 30.73 . 20 ' NS . 16. 51 ' • 2 ' • . 0. 14 15.96 ' 12 . ' . NS 14v 5,3 3 .• 0.027 2.85 °. 6 NS"' • 2. 78 . • • 4 .0.002 • • 0.20- ' 2 0.21 . • .. 'Only one r.oot, the g r e a t e s t , i s - s i g n i f i c a n t l y - d i f f e r e n t . • f r o m z e r o . The .canonical, v e c t o r s a s s o c i a t e d - w i t h the other r o o t s do'not canonical axes. The numbers are the provenance numbers. 9 5 % confidence c i r c l e s are shown. Canonical axis l g . 1 0 . C a n o n i c a l a n a l y s i s f o r R e g i o n 2. P l o t t i n g made u s i n g t h e f i r s t c a n o n i c a l a x i s a n d t h e t h i r d . a x i s . g . l l . Canonical analysis for Region 2. P l o t t i n g made using the f i r s t and the lourth axes. Ln c o n t r i b u t e s i g n i f i c a n t l y to the t o t a l v a r i a t i o n . S i x t y - s i x p e r c e n t of the v a r i a t i o n between groups i s e x p l a i n e d by the f i r s t c a n o n i c a l r o o t . F i g . 9 shows the o r d i n a t i o n r e a l i z e d by the f i r s t axes and (*) 9 5 % "confidence c i r c l e s " . "When the c i r c l e s do not o v e r l a p , the p r o v e n -ances a r e d i s t i n c t . A look at F i g . 9 shows that the: p r o v e n a n c e s 3 and 31 f o r m one group; p r o v e n a n c e s 33 and 34 f o r m another group; p r o v e n a n c e 4 i s i n t e r m e d i a t e between the two groups. The p r o v e n a n c e s 32 and 30 a r e d i s t i n c t f r o m a l l the other p r o v e n a n c e s . However, i f the f i r s t c a n o n i c a l a x i s i s o n l y c o n s i d e r e d , t h i s a x i s o r d i n a t e the p r o v e n a n c e s of V a n c o u v e r I s l a n d along a l a t i t u d i n a l g r a d i e n t c o n t r a s t -ing the s o u t h e r n p r o v e n a n c e s (3, 30 and 31) and the n o r t h e r n p r o v e n a n c e s (4, 32, 33 and 34) without d i f f e r e n t i a t i n g the west coast f r o m the east coast. F i g . 10 and 11 show that the p r o v e n a n c e s o v e r l a p m u c h along the other axes and the l a t t e r have not much: meaning. Queen C h a r l o t t e I s l a n d s Zone = R e g i o n 3 = p r o v e n a n c e s 36, 37, 38, 39, 40. N u m b e r of groups: 5. N u m b e r of p - v a r i a t e o b s e r v a t i o n s : 75. T e s t of e q u a l i t y of group d i s p e r s i o n s : U = 1.218 NS. The hypothe-s i s of e q u a l i t y of group d i s p e r s i o n s i s not r e j e c t e d . T e s t of o v e r a l l s i g n i f i c a n c e of the d i f f e r e n c e s between group cen-t r o i d s : F (20, 219) = 6.08***. T h e r e a r e d i f f e r e n c e s bet-ween the group mean v e c t o r s . (*) S t r i c t l y , the 9 5 % confidence c i r c l e s s h o u l d ' v a r y i n s i z e f r o m one p r o v e n a n c e to another as the d i s p e r s i o n s a r e not a l l i d e n t i c a l . C a n o n i c a l a x i s 2 F i g . 1 2 . Canonical analysis f o r Region 3 . P l o t t i n g made using the f i r s t two canonical axes. B A R T L E T T ' S T e s t of the L a t e n t Roots: 98 R o o t s * " \ - P e r c e n t -R e m o v e d Ro o t s C h i - S q u a r e ' D. F. S i g n i f i c a n c e age t r a c e 0 1.317 100.60 20 62.71 1 0.681 42.72 12 *** 32.40 2 0.069 6.90 6 NS 3! 27 3 0.034 2.32 2 NS 1.62 O n l y the two g r e a t e s t r o o t s a r e s i g n i f i c a n t l y d i f f e r e n t f r o m z e r o . M o s t of the v a r i a t i o n i n the new c a n o n i c a l space i s c o n f i n e d to the c a n o n i c a l v e c t o r s a s s o c i a t e d w i t h these r o o t s as the p l o t t i n g u s i n g the c o m b i n a t i o n o f the axes c o u l d show i t . A look at F i g . 12 shows that the p r o v e n a n c e s 36, 38 and 39 f o r m one c l u s t e r . P r o v e n a n c e 37 i s quite d i s t i n c t f r o m a l l the other p r o v e n a n c e s d e s p i t e i t s g e o g r a p h i c a l p r o x i m i t y f r o m the p r o v e n a n c e s 36, 38 and 39. The p r o v e n a n c e 40 i s w e l l i s o l a t e d and d i f f e r e n t f r o m a l l the p r o v e n a n c e s . A l a s k a Zone = R e g i o n 4 = p r o v e n a n c e s 23, 24, 25, 26, 27, 28. N u m b e r of groups: 6. N u m b e r of p - v a r i a t e o b s e r v a t i o n s : 90. T e s t of e q u a l i t y of the group d i s p e r s i o n s : U = 2.88**. The n u l l -h y p o t h e s i s of e q u a l i t y of the d i s p e r s i o n s i s r e j e c t e d . T e s t of the o v e r a l l s i g n i f i c a n c e of the d i f f e r e n c e s between group *** c e n t r o i d s : F (25, 298): 5. 56 Fig. 13. Canonical analysis f o r Region 4. P l o t t i n g made using' the f i r s t two canonical axes. vO 10.0 B A R T L E T T ' s T e s t of the Latent R o o t s : P e r c e n t -R e m o v e d R o o t s C h i - S q u a r e D.F. S i g n i f i c a n c e age t r a c e 0 1.20 118.54 25 *** ' 61.85 1 0.53 52.52 16 *** 27. 03 2 0.13 17.19 9 * 6.48 3 0. 08 7. 26 4 NS 4. 35 4 0. 01 0.41 1 NS 0. 29 The f i r s t three g r e a t e s t r o o t s a r e s i g n i f i c a n t l y differ.ent f r o m z e r o and t h e i r a s s o c i a t e d c a n o n i c a l v e c t o r s condense a l l the v a r i a -t i o n i n the new space. F i g . 13 shows the o r d i n a t i o n r e a l i z e d by the f i r s t a x e s . P r o v e n a n c e s 23 and 25 a r e quite d i f f e r e n t f r o m the p r o v e n a n c e s 24, 26, 27 and 28 w h i c h o v e r l a p m o r e or l e s s . P r o v e n a n c e 27 i s ' c o n t i -guous to a group f o r m e d by the p r o v e n a n c e s 26 and 28. Skeena R i v e r Zone = R e g i o n 5 = p r o v e n a n c e s 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 18, 19, 21, 22. N u m b e r of groups: 14. Number of p - v a r i a t e o b s e r v a t i o n s : 210. T e s t of e q u a l i t y of group d i s p e r s i o n s : U = 3.86 . The n u l l h y p o t h e s i s of e q u a l i t y of d i s p e r s i o n s i s r e j e c t e d . T e s t of o v e r a l l s i g n i f i c a n c e of the d i f f e r e n c e s between the group c e n t r o i d s : F (65, 911).= 8.21 . T h e r e a r e d i f f e r e n c e s . between the mean v e c t o r s of the groups. the f i r s t two canonical axes. 1-02 B A R T L E T T ' s T e s t of the L a t e n t R o o t s : Roots P e r c e n t -R e m o v e d Roo t s G h i - S q u a r e ' * D.F. S i g n i f i c a n c e age t r a c e 0 2.15 434.51 65 *** 62.39 1 0.82 205.46 48 *** 23.81 2 0. 27 85.83 33 *** 7.75 3 0.18 38.56 20 ** 5.33 4 0. 02 4. 90 9 NS 0. 72 The f i r s t f o u r r o o t s a r e s i g n i f i c a n t l y d i f f e r e n t f r o m z e r o . T h e i r a s s o c i a t e d c a n o n i c a l axes condense the v a r i a b i l i t y i n the new space. F i g . 14 shows that the p r o v e n a n c e s , on the b a s i s of the t r a i t s s t udied, f o r m d i f f e r e n t c l u s t e r s of g e o g r a p h i c a l p r o x i m i t y o r i g i n . P r o v e n a n c e s 7 and 10 f o r m one group; 14 and 22 another; 15 and 21 another; 5,6 and 18 s t i l l another group. The p r o v e n a n c e s 11, 13, 9, 19 and 8 a r e quite d i s t i n c t , h owever some o v e r l a p p i n g m a y o c c u r . P r o v e n a n c e s 11 and 9 ar e i n t e r i o r h i g h v a l l e y p r o v e n a n c e s and be l o n g to other r i v e r basins than the Skeena R i v e r b a s i n . 3.5.2 D e n d r o g r a m a n a l y s i s The p r o g r a m d e v e l o p e d by D r . C. M I L L I E R i n F r a n c e p e r f o r m s a c a n o n i c a l a n a l y s i s s i m i l a r to that p e r f o r m e d above, but a l s o c a l c u l a t e s , among many other things, the g e n e r a l i z e d d i s t a n c e s of M A H A L A N O B I S between the groups and c l a s s i f i e s the groups a c c o r d -i n g to two types of d e n d r o g r a m : one b a s e d on the " V A N D E N D R I E S S C H E ' s c l a s s i f i c a t i o n " , the other on the " u l t r a m e t r i c c l a s s i f i c a t i o n " d e v e l o p e d by R O U X ( 1968). ROUX's t h e s i s w o r k was not a v a i l a b l e and the u l t r a -m e t r i c c l a s s i f i c a t i o n w i l l not be c o n s i d e r e d . The d e n d r o g r a m s g i v e n by the two methods a r e quite d i f f e r e n t and the c o m p a r i s o n o f d e n d r o g r a m s i s p a r t i c u l a r l y c o m p l e x ( M I L L I E R and T O M A S S O N E , o£. _cit. ) and beyond the scope of the p r e s e n t study. D e n d r o g r a m s a r e b a s e d on "phenetic r e s e m b l a n c e " and should not i m p l y descent. The a b c i s s a of suc h a d e n d r o g r a m has no s p e c i a l m e a n i n g and i s i n d i f f e r e n t ; i t s e r v e s o n l y to s e p a r a t e the OTU's w h i l e the o r d i n a t e i s i n some s i m i l a r i t y c o e f f i c i e n t ( S O K A L and S N E A T H , 1963) . The d e n d r o g r a m s p r e s e n t e d h e r e have an o r d i n a t e i n a s c a l e p r o p o r t i o n a l to the M A H A L A N O B I S ' d i s t a n c e s between the p o p u l a t i o n s of t r e e s . The c l u s t e r i n g p r o c e d u r e i s b a s e d on common sense: the p r i n c i p l e i s that any two groups b e l o n g i n g to the same c l u s t e r s h o u l d at l e a s t on the average, have a s m a l l e r D than those b e l o n g i n g to two d i f f e r e n t c l u s t e r s ( V A N D E N D R I E S S C H E , 1965). P o i n t s of j u n c t i o n between stems along the o r d i n a t e m e a n that the two connected p o p u l a t i o n s f o r m a c l u s t e r at the value shown on the o r d i n a t e . The d e n d r o g r a m does not, however, r e p r e s e n t the d i s t a n c e s between the groups because i t i s u n i d i m e n s i o n a l . The d i s a d v a n t a g e s of the d e n d r o g r a m a r e that i t has an i m p o s e d h i e r a r c h a l s t r u c t u r e and that c o n s e q u e n t l y there i s no p r o v i s i o n f o r o v e r l a p between the e n t i t i e s c l a s s i f i e d ( B L A C K I T H and R E Y M E N T , 1971). The d i s t a n c e s of M A H A L A N O B I S between the p r o v e n a n c e s and f o r the four r e g i o n s s t u d i e d a r e shown i n T a b l e s X I I I , X I V , X V and X V I . The d e n d r o g r a m of the f o l l o w i n g r e g i o n s a r e r e p r e s e n t e d 1*04 F i g . 15.. Dendrogram of the provenances of Region 2. The numbers are the provenance numbers. o 30 3 31 32 4 .34 33 40 37 38 39 36 105 F i g . l 6 . Dendrogram of the provenances of Region 3. The numbers are the provenance numbers. 23 25 27 24 28 26 Fig.17. Dendrogram of the provenances of Region 4. The numbers are the provenance numbers. 107 F i g . 16". Dendrogram of the provenances of Region 5. The numbers are the provenance numbers. 18 6 5 22 13 10 7 8 21 15 19 14 9 108., ( F i g . 15, 16, 17 and 18). O n l y the p r o v e n a n c e s w i t h 15 t r e e s w e re c l a s s i f i e d . R e g i o n 2: V a n c o u v e r I s l a n d ; p r o v e n a n c e s 3, 4, 30, 31, 32, 33 and 34. R e g i o n 3: 36, 37, 38, 39 and 40. R e g i o n 4: p r o v e n a n c e s 23, 24, 25, 26, 27, 28. R e g i o n 5: p r o v e n a n c e s 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 18, 19, 21, 22. The d e n d r o g r a m f o r V a n c o u v e r I s l a n d c o n f i r m s that, on the b a s i s of the t r a i t s s t u d i e d , two groups of p r o v e n a n c e s appear: one group w i d e l y d i s t i n c t f o r m e d by the p r o v e n a n c e s 30, 31 and 3 ( s o u t h e r n part) and another group, w i t h the provenance 32 w e l l d i s t i n c t f r o m t h i s g r o up.formed by the p r o v e n a n c e s 32, 33, 34 and 4 ( n o r t h e r n p a r t of the Is l a n d ) . In r e g i o n 3, the p r o v e n a n c e s 36, 37, 38 and 39 c l u s t e r i n one group, the provenance,40 b e i n g d i s t i n c t . In r e g i o n 4, the p r o v e n a n c e s 24, 25, 26, 27 and 28 c l u s t e r i n one group, the p r o v e n a n c e s 23 b e i n g d i s t i n c t . In r e g i o n 5, the c l u s t e r s a r e of g e o g r a p h i c o r i g i n : the p r o v e n a n c e s g e o g r a p h i c a l l y c l o s e tend to c l u s t e r . One group d i v e r g e : the upper v a l l e y p r o v e n a n c e s 5, 6, 11 and 18. G e n e r a l l y , the d e n d r o g r a m s e s t a b l i s h e d a c c o r d i n g to the V A N D E N D R I E S S C H E m e t h o d tend to c l u s t e r the p r o v e n a n c e s a c c o r d i n g to t h e i r g e o g r a p h i c a l o r i g i n . The c l u s t e r s g e n e r a l l y d i f f e r f r o m the 109. T A B L E XIII M A T R I X O F T H E D I S T A N C E S O F M A H A L A N O B I S B E T W E E N T H E P R O V E N A N C E S O F R E G I O N 2 ovenances 3 4 30 31 32 33 34 3 0 123 140 51 512 279 295 4 123 0 220 73 195 56 49 30 140 220 0 168 574 264 293 31 51 73 168 0 439 147 213 32 512 195 574 439 0 243 123 33 279 56 264 147 243 0 33 34 295 49 293 213 123 33 0 T A B L E X I V M A T R I X O F T H E D I S T A N C E S O F M A H A L A N O B I S  B E T W E E N T H E . P R O V E N A N C E S O F R E G I O N 3 Provenances •'• 36 37 _38 39 40 36 0 714 108 101 923 37 714 0 760 412 572 38 108 760 0 131 618 39 101 412 131 0 557 40 923 572 618 557 0 T A B L E X V M A T R I X O F T H E D I S T A N C E S O F M A H A L A N O B I S B E T W E E N T H E P R O V E N A N C E S O F R E G I O N 4 Provenances 23 24 25 26 27 28 0 1068 799 835 409 866 1068 0 276 146 269 252 799 276 0 373 279 519 835 146 373 0 153 36 409 269 279 153 0 258 866 252 519 36 258 0 23 24 25 26 27 28 T A B L E X V I M A T R I X O F T H E D I S T A N C E S O F M A H A L A N O B I S B E T W E E N T H E P R O V E N A N C E S O F R E G I O N 5 P r o v e n -ance s 10 11 13 14 15 18 19 21 22 5 6 7 8 9 10 11 13 14 15 18 19 21 22 0 599 741 391 1881 110 1153 2098 799 26 872 1927 599 741 391 1090 731 1881 26 0 .684 1663 503 586 434 984 872 684 0 381 177 163 1201 646 1927 1663 503 586 381 177 163 0 657 325 434 1201 2270 1673 105 991 1902 766 508 571 159 828 697 389 1278 262 409 974 657 170 1090 984 646 1276 430 731 627 269 646 116 496 409 123 806 223 325 2270 1276 170 636 1139 614 437 168 738 430 636 430 0 1139 0 1498 483 1163 1770 505 897 506 897 506 0 307 551 174 442 840 529 770 499 646 116 496 280 420 158 458 202 0 1295 189 146 570 110 1153 2088 627 1673 105 991 1902 269 508 571 159 828 389 1278 262 409 409 123 806 307 614 437 168 551 1498 483 1163 1770 1295 0 749 1352 556 189 749 .0 408 343 146 1352 408 0 659 799 766 697 974 223 738 430 505 174 442 840 529 770 499 0 280 420 158 458 202 570 556 343 659 0 I l l c l u s t e r s c o n s t r u c t e d f r o m the c a n o n i c a l a n a l y s e s w h i c h s p r e a d the p r o v e n a n c e s a c c o r d i n g to s e v e r a l d i m e n s i o n s . 3. 5. 3 Stepwise d i s c r i m i n a n t a n a l y s i s The p r o g r a m u s e d was the one a v a i l a b l e at the U.B.C. computing c e n t r e l i b r a r y under the name B M D 0 7 M . It p e r f o r m s b a s i c a l l y a d i s c r i m i n a n t f u n c t i o n a n a l y s i s as e x p l a i n e d i n S e c t i o n 3.4, but i n a step-w i s e manner. A t e a c h step, one v a r i a b l e i s e n t e r e d into the set of d i s -c r i m i n a t i n g v a r i a b l e s . The v a r i a b l e i s e n t e r e d w h i c h g i v e s the g r e a t e s t d e c r e a s e i n the r a t i o of " w i t h i n " to " t o t a l " g e n e r a l i z e d v a r i a n c e s . A v a r i a b l e i s d e l e t e d i f the r e v e r s e o c c u r s . When a l l the v a r i a b l e s have been t e s t e d , a set of d i s c r i m i n a n t f u n c t i o n s i s obtained and u s e d to c a l -culate the M A H A L A N O B I S (D ) d i s t a n c e of an i n d i v i d u a l f r o m one group to another group and the c h i - s q u a r e and i t s a s s o c i a t e d "a p o s t e r i o r i " p r o b a b i l i t y f o r t e s t i n g i f a gi v e n i n d i v i d u a l of a gi v e n group c o u l d come f r o m another group. T h e s e s t a t i s t i c s a r e i n t e r e s t i n g to evaluate the d i s -c r i m i n a t o r y value of the v a r i a b l e s u s e d and the d i f f e r e n c e s between the groups. The r e s u l t s a l r e a d y o b t a i n e d b y C A w i l l not be r e p e a t e d and on l y the new r e l e v a n t r e s u l t s w i l l be p r e s e n t e d h e r e f o r two sets of p r o v e n a n c e s : r e g i o n 2 and r e g i o n 5. R e g i o n 2 = p r o v e n a n c e s 3, 4, 30, 31, 32, 33, 34. N u m b e r of groups: 7. N u m b e r of t r e e s pe/r g roups: 15. T A B L E X V I I • N U M B E R O F T R E E S C L A S S I F I E D INTO T H E D I F F E R E N T  P R O V E N A N C E S A C C O R D I N G TO T H E I R P R O V E N A N C E O F ORIGIN O r i g i n a l C l a s s i f i e d Into the Provenances  Provenance 3 4 30 31 32 33 34 3 5 4 4 2 0 0 0 44 1 3 1 3 3 3 1 30 3 2 7 0 0 1 2 31 7 1 2 2 2 0 1 32 0 1 0 0 10 1 3 33 1 2 3 1 1 3 4 34 0 1 1 0 4 4 4 113 Only t'Kriee var iables , were s u c c e s s i v e l y entered: i n decreas ing o r d e r of d i s c r i m i n a t i o n ; seed length, cone length, and f ina l ly seed width . Wing width and length did not s igni f i cant ly contribute to the d i s c r i m i n a t o r y power of the ca lculated functions. The F tests of the di f ferences between each p a i r of groups showed that 5 p a i r s out of 21 were not s igni f i cant ly different f r o m z e r o . Provenance 3 was not different f r o m provenance 31. " 4 was not different f r o m provenances 31, 33, 34. " 33 was not different f r o m provenance 34. G e n e r a l l y , the " p o s t e r i o r i " probabi l i ty of group appartenance was quite var iab le and genera l ly low. The number of trees c l a s s i f i e d into the group to whom they belong ranges f r o m 1 to 10, i . e . , between 7% and 66% of the trees were c l a s s i f i e d back into their own population attest-ing the weak population d i s c r i m i n a t i o n of the five var iab les studied, or that the groups so over lap that they might be c lus tered as one large group (Table X V I I ) . Reg ion 5 Number of groups : 14. Number of trees per group: 15. Number of p - v a r i a t e observat ions : 210. In this case, the five var iab les were s u c c e s s i v e l y entered i n the d i s c r i m i n a n t funct ion. In decreas ing o r d e r of d i s c r i m i n a t i o n : cone length, seed width, seed length, wing length and f i n a l l y , wing width . The F test of the dif ferences between each pa i r of groups showed that s i x p a i r s out of 91 p a i r s were not s igni f i cant ly different f r o m z e r o . G e n e r a l l y , again the "a p o s t e r i o r i " probab i l i ty of group appartenance 114 was quite var iab le and genera l ly low. The number of trees c l a s s i f i e d into their own group ranges f r o m 1 to 9, i . e . , f r o m 7% to 60% attesting the weak d i s c r i m i n a t o r y value of the 5 var iab les - s tudied . These r e s u l t s seem to c o n f i r m what we found i n analyzing the components of var iance for the t ra i t s s tudied: the best charac ters were supposed the ones w i t h the smal les t wi thin provenance v a r i a b i l i t y (see Section 2. 1, P a r t II). 3 . 5 . 4 P r i n c i p a l component analys is ( P C A ) The U . B . C . Computing centre p r o g r a m U B C F A C T O was t r i e d on r e g i o n 5. F o r a def ini t ion of Region 5, see Sect ion 3. 5. 2. The name of the p r o g r a m is m i s l e a d i n g because it p e r f o r m s a P C A fol lowed by a s o - c a l l e d V A R I M A X rota t ion . P C A is a v e r y pecul iar case of F A and should be separated f r o m the numerous techniques of F A ( D A G N I E L I E , . ; I960; C A T T E L L , 1964; S E A L , 1964 and M O R R I S O N , 1967). The V A R I M A X procedure used for rotat ing the p r i n c i p a l compo-nent axes is explained by C O O L E Y and L O H N E S (1966). A c c o r d i n g to C A T T E L L (1965), this method is inadequate because it imposes the orthogonali ty of the new axes and their r i g i d ro ta t ion . The eigenvalues are 3.47, 1.01; 0.27,0^20 and 0.04. T h e i r corresponding cumulat ive proport ions of the total var iance are 69. 5%, 89.8%, 95.2%, 99. 1% and 100%. Thus , 90% of the total v a r i a t i o n is explained by the f i r s t two e igenvectors . A look at Table XVIII shows that the f i r s t component i s p o s i t i v e l y and strongly c o r r e l a t e d with the five m e a s u r e m e n t s : it could component 2 2 2 4* .11 21 14 15 » 10 15 19 |2 ^» component 1 • 18 • 6 8 F i g . 1 9 . P r i n c i p a l component a n a l y s i s o f t h e p r o v e n a n c e s o f R e g i o n 5. P l o t t i n g made u s i n g t h e f i r s t two p r i n c i p a l c o m p o n e n t s , The M numbers a r e t h e p r o v e n a n c e numbers. . ' £ 116 T A B L E XVII I L O A D I N G O F T H E C O M P O N E N T S O N T H E  V A R I A B L E S S T U D I E D Components V a r i a b l e 2 3 4 5 1 0.63 - 0 . 72 - 0 . 2 0 - 0 . 20 0. 03 2 0.94 0. 09 0. 01 - 0 . 03 0. 13 3 0. 91 -0 .30 0. 18 0. 14 - 0 . 14 4 0. 76 0. 58 - 0 . 08 - 0 . 27 - 0 . 07 5 0.89 0. 23 - 0 . 31 0. 25 0. 04 be in terpre ted as a s ize component. The other components cannot be object ively in te rpre ted . The plott ing of the populations i n the new orthogonal space is shown i n F i g . 19. The shuffl ing of the populations is quite different f r o m the one p e r f o r m e d by C A or D F . Whi le a group of provenances (5, 6, 18) i s mainta ined, the other provenances are widely m i x e d up. No geographic p r o x i m i t y can expla in the scat ter ing of the provenances p e r f o r m e d by the P C A . 3. 5. 5 Conclus ions Before answering the question of what is the best m u l t i -var ia te s ta t i s t i ca l technique for c l a s s i f y i n g O T U ' s , of these we have t r i e d , we must define what is best, what i s c l a s s i f i c a t i o n , its a i m s , etc. C l a s s i f i c a t i o n is the bas ic method which man employs to 117. come to g r i p s with and organize the external w o r l d (DAVIS and H E Y W O O D , 1-965), but any c l a s s i f i c a t i o n of l i v i n g organisms must also have some b i o l o g i c a l meaning . It d i d happen that the methods used by the s y s t e m a -t ists in the past n a t u r a l l y resu l ted in coherent b i o l o g i c a l c l a s s i f i c a t i o n s where , i f a c lass of plants shares many charac ters (morphologica l or physio logica l ) in common, when some new t ra i t i s found i n one m e m b e r then the other m e m b e r s of the c lass also possess i t . Beyond the theore t i ca l studies of c l a s s i f i c a t o r y schemes (See C O L E , 1969 and J A R D I N E and SIB S O N , 1971), there is only one ul t imate c r i t e r i o n for judging a c l a s s i f i c a t i o n method: its re levance as an i n f o r m a t i o n storage and r e t r i e v a l s y s t e m . T h i s c r i t e r i o n cannot be used in m u l t i v a r i a t e s ta t i s t i ca l analyses of some m o r p h o l o g i c a l c h a r a c -ters which v a r y wi th in a spec ies . E v e n i f a great number of t ra i t s could be included i n the s ta t i s t i ca l a n a l y s i s , which is r a r e l y the case i n p r a c -t i ce , the adequacy of the method should be tested by us ing c r i t e r i a out-side the method i t se l f . M a t h e m a t i c a l ar t i fac ts are eas i ly produced by b l i n d l y fo l lowing the r e s u l t s of a computer generated a n a l y s i s . L A N G L E T (1959) has demonstrated it by pointing out the inadequacies of analyses of var iance of Scots pine (Pinus s y l v e s t r i s L . ) provenances as a " p r o o f " of the existence of ecotypes. It i s , therefore , p a r t i c u l a r l y important to understand the mechanics of a method as to its poss ible " d i s t o r t i o n " effects, not to ment ion the consequences of d i s r e s p e c t i n g the hypotheses under ly ing a p a r t i c u l a r m o d e l . 118 P C A i s an o r t h o g o n a l t r a n s f o r m a t i o n a i m i n g at r e d u c i n g the d i m e n s i o n a l i t y of the o r i g i n a l h y p e r s p a c e into a few l i n e a r c o m b i n a -t i o n s of the o r i g i n a l v a r i a t e s . P C A s h o u l d only be a p p l i e d to a s ample whose o r i g i n can be s a f e l y a t t r i b u t e d to one m u l t i v a r i a t e n o r m a l p o p u l a -t i o n . C A i s i d e n t i c a l to D F . It r e s u l t s i n a new n o n - o r t h o g o n a l space. It can be u s e d when two h i e r a r c h a l l e v e l s of v a r i a t i o n m u s t be d i s t i n g u i s h e d and when a v i s u a l r e p r e s e n t a t i o n i n a few c h a r t s b a s e d on the m o s t i m p o r t a n t c a n o n i c a l axes i s c ontemplated. C A and D F e s s e n t i a l l y l e a d to the same p l o t t i n g into the new space. T h e y both l e a d to the c a l c u l a t i o n of d i s c r i m i n a n t f u n c t i o n s 2 and the a s s o c i a t e d g e n e r a l i z e d d i s t a n c e s D . A s i m p l e p l o t t i n g i n the c a n o n i c a l space even w i t h confidence c i r c l e s , i s not enough to c l a s s i f y the p r o v e n a n c e s into c l u s t e r s as some c r i t e r i a m u s t be d e c i d e d upon i n o r d e r to define what i s a c l u s t e r . The c a l c u l a t i o n o f the g e n e r a l i z e d d i s t a n c e s between a l l p a i r s of p r o v e n a n c e s r e s u l t i n a m a t r i x of d i s t a n c e s w h i c h can be u s e d f o r c o n s t r u c t i n g two o r three d i m e n s i o n r e p r e s e n t a t i o n s of the r e l a t i v e p o s i t i o n s of these p r o v e n -2 ances. To c l u s t e r "OTU's" u s i n g the d i s t a n c e s D between them, d i f f e r e n t methods have been pr o p o s e d . One m e thod has been t r i e d : the method of V A N D E N D R I E S S C H E (1965) w h i c h r e s u l t e d i n d e n d r o g r a m s f o r w h i c h an i n t e r p r e t a t i o n has been p o s s i b l e . However, the c l u s t e r s shown by the p l o t t i n g s of the p r o v e n a n c e s i n the c a n o n i c a l space a r e g e n e r a l l y d i f f e r e n t f r o m the c l u s t e r s d i s p l a y e d by the d e n d r o g r a m s / ^ (*) The c a n o n i c a l space shows the m u l t i - d i m e n s i o n a l a s p e c t s of the p o s s i b l e g r o u p i n g s w i t h the amount of o v e r l a p p i n g . 119 The d i s c r i m i n a n t functions are useful as a c l a s s i f i c a t i o n means when a new i n d i v i d u a l i s m e a s u r e d . If a stepwise d i s c r i m i n a n t analys is i s used, D F could be the most eff icient technique by : (1) reducing the o r i g i n a l number (if poss ible v e r y large) of v a r i a b l e s to a set of d i s c r i m i n a t o r y v a r i a b l e s , thus e l i m i n a t i n g any redundancy of i n f o r m a t i o n ; (2) by ca lculat ing the d i s c r i m i n a n t functions and test ing their c l a s s i f i c a t o r y power us ing the reduced set of v a r i a b l e s ; 2 (3) by ca lcula t ing the genera l ized distances D between a l l the O T U S , thus p e r m i t t i n g the construct ion of three d i m e n -sion or two d imens ion representat ions of the re la t ive p o s i -tions of these O T U S or p e r m i t t i n g their c lus ter ing and their representat ion as a d e n d r o g r a m . However , the den-d r o g r a m has the disadvantage of t runcat ing the s i m i l a r i t y re la t ionships of the elements to be c l a s s i f i e d . It must be noted that the plott ing of the provenances i n the dif ferent spaces generated by the techniques used ( C A , D F and P C A ) shows that when the axis used corresponded to a root which is not s i g -n i f i cant ly different f r o m zero , no meaningful v a r i a t i o n could be t ied up to i t . T h i s f inding contradicts the contention of B L A C K I T H and R E Y M E N T (19VI), that the v a r i a t i o n associated with the roots not s igni f i cant ly d i f -ferent f r o m zero can be more important in c lus ter ing the O T U S than the s ignif icant r o o t s . (See F i g . 10 and 11). 120 • A n o t h e r a p p r o a c h l e s s e f f i c i e n t and above a l l m o r e s u b j e c t -i v e than C A o r DF, w o u l d be to use P C A on a few s e l e c t e d p r o v e n a n c e s , e a c h provenance b e i n g a n a l y z e d s e p a r a t e l y . A g r e a t number of v a r i a b l e s s h o u l d be m e a s u r e d . T h e n the l o a d i n g s of the v a r i a b l e s on the p r i n c i p a l components c o u l d be u s e d as an i n d i c a t i o n of the i m p o r t a n c e of the v a r i a b l e s i n e x p l a i n i n g the m u l t i d i m e n s i o n v a r i a b i l i t y of these p r o v e n a n c e With, the few v a r i a b l e s s e l e c t e d , C A o r D F c o u l d be u s e d to o r d i n a t e the po p u l a t i o n s . P A R T III S T U D Y O F T H E S E E D L I N G S T A G E 1 2 1 . C H A P T E R 1 T H E P H I L O S O P H Y O F T H E N U R S E R Y T E S T The n u r s e r y test design of tree genetic exper iments i s d e r i v e d f r o m the s ta t i s t i ca l theory of exper imenta l design which has be-come c l a s s i c a l since the books of F I S H E R (1947, 1954) have been pub-l i s h e d . F o r a complete exposi t ion of the theory and appl icat ion of ex-p e r i m e n t a l design, see the books of C O C H R A N and C O X , 1957 or K E M P T H O R N E , 1952). The usual prac t i ce i s to use some r a n d o m i z e d complete block des ign . A t the M i c h i g a n State U n i v e r s i t y , the standard n u r s e r y experiment consists of four repet i t ions used for the m e a s u r e -ments and of a f i f th r e p l i c a t i o n , f r o m which most or a l l of the seedlings for f i e l d planting are d e r i v e d ( W R I G H T , 1970). However , each forest geneticist has his own pre ferences : some use s ix r e p l i c a t i o n s , others use plots of l inear shape (the s o - c a l l e d r o w or l ine plot) , some of quadrat shape. The s izes v a r y f r o m one author to the other . Recent ly , the p r o b l e m of opt imum size plot has been theore t i ca l ly attacked ( H U H N , 1970 a , 1970 b ) . Only W R I G H T (1963) has t r i e d to give some indicat ions as how to determine the best plot and sample s i z e s . L i t t l e is known in. pract ice about the poss ib le effects of genotype by genotype in terac t ions , as w e l l as genotype by environment interact ions on the dif ferences expressed i n the n u r s e r y . Only recent ly , the spacing effect has been studied i n a progeny test of D o u g l a s - f i r L e c -( C A M P B E L L and W I L S O N , 1973). It seems that spacing-genotype in terac t ion i n D o u g l a s - f i r is not l i k e l y to apprec iab ly affect the selt t ion r e s u l t s . However , genotype by n u r s e r y in terac t ion may be poss ib le ( V A N D E N D R I E S S C H E , 1973). The poss ib le long t e r m effects of these interact ions are not known for sure . L i t t l e i s known about the possible m a t e r n a l effects or p r e -condit ioning effect ( R O W E , 1964). In provenance test ing, this effect -i f any - i s susceptible to be repeated when col lec t ions are made f r o m the same t rees . The p r o b l e m could be m o r e ser ious as the r e s u l t s of a provenance test could be different f r o m those obtained f r o m a c r o s s r e a l i z e d i n a seed o r c h a r d composed of scions of the mother trees l i v i n g i n a new environment . The genetic in terpre ta t ion of a s ta t i s t i ca l mode l i n te rms of gene effects (addit ive, dominant, e tc . ) and the poss ib le gains under d i f -ferent types of se lect ion should be cautiously accepted. The s ta t i s t i ca l model used to estimate and test the dif ferences e x p r e s s e d i n a n u r s e r y or f i e l d test has essent ia l ly s ta t i s t i ca l a i m s . A good exper iment ensures : (1) The absence of sys temat ic e r r o r s . T h i s i s achieved by r a n d o m i z a t i o n . The estimate of a treatment contrast w i l l only di f fer f r o m its true value by random e r r o r s . The covariance between treat -ments and m i c r o e n v i r o n m e n t s is on the average, z e r o . (2) The standard e r r o r of a treatment should be s m a l l enough to be able to draw r e l i a b l e conc lus ions . The p r e c i s i o n of an exper iment 12$ w i l l depend on: (a) the i n t r i n s i c v a r i a b i l i t y of the e x p e r i m e n t a l m a t e r i a l and the a c c u r a c y of the e x p e r i m e n t a l work; (b) the number of e x p e r i m e n t a l u n i t s and the number of r e p e t i t i o n s ; (c) the d e s i g n of the e x p e r i m e n t . (3) The range of v a l i d i t y of the test s h o u l d be l a r g e enough: i . e. , the c o n d i t i o n s not too f a r f r o m the p r a c t i c e . (4) The d e s i g n i s s i m p l e . (5) A n adequate s t a t i s t i c a l a n a l y s i s be p o s s i b l e , i . e . , the o b s e r -v a t i o n s r e s p e c t the b a s i c a s s u m p t i o n s o f the m o d e l u s e d ( n o r m a l i t y , a d d i t i v i t y , e t c . ) , ( a f t e r COX, 1958). T h e r e f o r e , i t i s not because the test u s e d has. been so a n a l y z e d as to e l i m i n a t e the c o v a r i a n c e between the genetic e l e m e n t s m e a s u r e d (pr o v e n a n c e s o r p r o g e n i e s ) and the e n v i r o n m e n t a l v a r i a t i o n w h i c h c o u l d p o s s i b l y o c c u r i n a n u r s e r y , that there i s no p h y s i o l o g i c a l i n t e r a c t i o n between the genotypes and t h e i r m i c r o - e n v i r o n m e n t . In the c l a s s i c a l a d d i t i v e m o d e l , the phenotype of an i n d i v i d u a l i s c o n s i d e r e d as the s u m of two components: P = G f E P = phenotypic value G = genotypic value E = e n v i r o n m e n t a l d e v i a t i o n O r , i n t e r m s of v a r i a n c e s : 124* 2 2 2 £ * p = 6 ^ , H- 4- 2 .cov (G, E ) . G e n e r a l l y , the design is such that cov (G, E) = O. But even i f cov (G, E) = O, the e n v i r o n -menta l component of var iance cannot be d i r e c t l y es t imated unless the other component i s e l iminated . If vegetative propagation or highly i n b r e d l ines are used , the phenotypic var iance provides an estimate of the en-v i r o n m e n t a l v a r i a n c e . In any case, the es t imat ion of the environmental var iance res t s on the assumption that this var iance i s the same in a l l species genotypes. The environmenta l var iance m e a s u r e d i n one i n b r e d l ine or i n one vegetat ively reproduced indiv idua l i s that shown by that p a r t i c u l a r genotype ( F A L C O N E R , 1964). T h e r e f o r e , the par t i t ion into genotypic and environmenta l var iances depends on the genotypes studied and of the environments i n which they grow. The broad sense h e r i t a b i l i t y calculated for n u r s e r y or f i e l d test, thus, is but a measure of the p r e c i -sion of the test; i t depends on the e x p e r i m e n t e r ' s choice of the design or care in seeding, seedbed prepara t ion , etc. It i s a repeat ib i l i ty of the test as defined, by F A L C O N E R , (1964 ). It seems, therefore , m i s l e a d i n g to speak of "genotypic h e r i t a b i l i t y " ( N A N S O N , 1970). The usefulness of the "genotypic" gains for dif ferent " se lec t ion i n t e n s i t i e s " by using such (*) "genotypic h e r i t a b i l i t y " i s therefore dubious. (*) That the genotypic h e r i t a b i l i t y is a measure of repeat ib i l i ty of a progeny or provenance test and has no or l i t t l e genetic s ignif icance comes f r o m the fact that it is ca lculated f r o m the means of the popu-la t ions , the means of the progenies whatever the populations and the means of the progenies for one population (see the f o r m u l a s given by N A N S O N , 1970, p. 117). •128 We can also question the necess i ty of u n i f o r m test con-d i t i o n s . R E U T E R (1971) has recent ly analyzed the r e s u l t s of a proven- . ' ance test es tabl ished i n the forest conditions of a v e r y mountainous r e g i o n : coastal B r i t i s h C o l u m b i a . Provenance by b lock interact ions were noted for total height. No c o r r e l a t i o n was found with height m e a s u r e d e a r l i e r . It does not seem, however , that the conclusions of the author are se l f -ev ident : m o r e repet i t ions , m o r e u n i f o r m test s i te , e tc . In fact , it i s quite poss ib le that genotype by environment in terac t ion undetected at the n u r s e r y stage ( H A D D O C K , et a l . 1967) i s so important i n D o u g l a s - f i r and i n coasta l B r i t i s h C o l u m b i a conditions that, i n p r a c t i c e , severa l coastal provenances could grow as w e l l as any other, on a given s i te . T h i s resu l t i s as important as f inding the "best" provenance for a given station as it would r e l a x the ru les of seed t ransfer and al low f reer plus tree se lec t ion . In provenance or progeny tests , it i s thus important , because so many factors are i n in terp lay , to use as m u c h as poss ib le the n u r s e r y methods of the t i m e , as w e l l as the usual s i l v i c u l t u r a l methods i n p la in o r d i n a r y forest conditions because there i s some evidence i n the recent forest genetic l i t e r a t u r e , that different kinds of genotype by environment in terac t ions , do exist ( S I L E N , 1966; M E R G E N , et a l . 1967; V A N D E N (**) D R I E S S C H E , 1973). However , many European f o r e s t e r s be l ieve that (*'*) A . A n interes t ing r e v i e w of the poss ib le genotype by environment i n t e r -actions i n forest trees has been done by S Q U I L L A C E (1970). . WRIGHT (1973), has also r e v i e w e d the provenance tests establ ished i n N o r t h C e n t r a l Uni ted States, i n this respect . 12 6 t e s t s i te-genotype i n t e r a c t i o n s a r e not i m p o r t a n t enough to p r e c l u d e the ch o i c e of the best provenance f o r a l a r g e region. (See, among o t h e r s , N A N SON, 1970). The r e a s o n s f o r these d i f f e r e n c e s of o p i n i o n between . . j the A m e r i c a n f o r e s t g e n e t i c i s t s and the E u r o p e a n ones a r e not known f o r s u r e . The e n v i r o n m e n t c o u l d be d i f f e r e n t as w e l l as the t r e e s p e c i e s . N o r w a y s p r u c e ( P i c e a a b i e s (L} KARST.) provenance p e r f o r m a n c e s i n N o r t h E a s t A m e r i c a a r e s i m i l a r to the p e r f o r m a n c e s showed i n E u r o p e a n f i e l d t e s t s , p o i n t i n g to low o r weak genotype by e n v i r o n m e n t i n t e r a c t i o n s i n t h i s t r e e s p e c i e s ( B A L D W I N , et a l . 1973). 121 C H A P T E R 2 M A T E R I A L A N D M E T H O D S U S E D I N T H E N U R S E R Y 2. 1 Seedling c h a r a c t e r i s t i c s studied i n the n u r s e r y . Why those p a r t i - cular c h a r a c t e r s ? M u c h factual i n f o r m a t i o n on the i n f r a s p e c i f i c genet ical ly based v a r i a t i o n c o r r e l a t e d wi th the place of o r i g i n of the genetic elements tested has been accumulated i n the past decades for many forest tree spec ies . It i s not poss ib le to r e v i e w here a l l the facts accumulated i n the s p e c i a l i z e d l i t e r a t u r e . Since L A N G L E T (36/37) pioneered i n this f i e l d , a great number of m o r p h o l o g i c a l and p h y s i o l o g i c a l charac ters have been studied: f r o m dry mat ter , n i t rogen content to height, bud set, length of the greatest internode, etc. for a large number of tree spec ies . The m a t r i c e s of cor re la t ions between the phenotypical t ra i t s studied and bet-ween these t ra i t s and the c h a r a c t e r i s t i c s of the place of o r i g i n of the p r o v -enances (geographical coordinates , c l i m a t i c var iab les such as vegetation p e r i o d , mean annual temperature , etc.) can be explained by the select ive act ion of the environment of o r i g i n of the provenances s tudied. Other factors of d i f ferent ia t ion of the populations of t r e e s : random d r i f t , i s o l a -t ion (geographic or reproduct ive) , i n t r o g r e s s i o n between spec ies , etc. have been so far cons idered as of m i n o r importance ( G A L O U X and F A L K E N H A G E N , 1965). The p r o b l e m of the c l i n a l or ecotypic v a r i a t i o n of adaptative t r a i t s i n forest trees has been m u c h d i scussed i n the past . A s L A N G L E T 128 (1959) showed i t , the existence of s ignif icant cor re la t ions between t ra i t s and habitat of o r i g i n do not prec lude the p o s s i b i l i t y of showing a d i s c o n -tinuous v a r i a t i o n by us ing analyses of v a r i a n c e s . The two techniques do not oppose themse lves . T h i s fact does not mean that how a forest tree i s adapted to its environment is not important i n t e r m s of phenotypic f l e x i b i l i t y ( S T E R N , 1964) and i ts i m p l i c a t i o n s so far as the "genetic s y s t e m " of the tree species i s concerned. F o r a d i s c u s s i o n of these important p r o b l e m s , see S T E R N (1964) or F A L K E N H A G E N (1968). V e r y general explanations are only avai lable to expla in the cor re la t ions o b s e r v e d . S T E R N (1964) has extensively d i s c u s s e d the factors respons ib le for the v a r i a t i o n observed i n forest t rees . The c o r -re la t ions of the t ra i t s with the habitats of o r i g i n could be explained by some c l i n a l v a r i a t i o n i n gene frequencies c o n t r o l l i n g these t ra i t s which would p a r a l l e l the v a r i a t i o n i n the environment . The genetic c o r r e l a t i o n s between t ra i t s could be at tr ibuted to p l e i o t r o p i s m and/or l inkage . M o r e o v e r , as pointed out by G A L O U X and F A L K E N H A G E N (1965) an o r g a n i s m grows and p r o s p e r s only i f adequate c o r r e l a t i o n s between its organs , s t ructures and functions are mainta ined . There is quite a redundancy of in format ion i n m e a s u r i n g a great number of t ra i t s due to the highly integrated aspect of a l l the functions of a l i v i n g o r g a n i s m . T h e r e f o r e , a l l the v a r i a t i o n detected i n a provenance test i s not n e c e s s a r i l y of adaptative nature . The re la t ive importance and ro le of a t ra i t is often a matter of hypothesis as only c o r -re la t ions with other t ra i t s are avai lable most of the time i n a study not accompanied wi th p h y s i o l o g i c a l invest igat ions . 12:9, Recent ly , some authors have s e r i o u s l y misunders tood and m i s u s e d the c o r r e l a t i o n s calculated f r o m provenance tests ( F A L K E N -H A G E N , 1972). The cor re la t ions and the r e g r e s s i o n coeff ic ients which can be d e r i v e d f r o m them s t a t i s t i c a l l y relate the provenance or progeny character means to some c h a r a c t e r i s t i c s of the place of o r i g i n of these genetic e lements . They are pure s ta t i s t i c s . The p r e c i s e eco log ica l law of v a r i a t i o n of the r e g r e s s i o n or c o r r e l a t i o n coeff icients for a given species , in r e l a t i o n with the test site has never been es tabl i shed. These s tat is t ics w i l l v a r y f r o m one test site to another according to the age and according to the genotype by environment in terac t ion c h a r a c t e r i s t i c of the spec ies . Invers ion of the sign or disappearance of the r e g r e s s i o n coef f i -cients are w e l l known. See among others , K I N G (1965) and W A K E L E Y (1961). The fo l lowing table s u m m a r i z e s W A K E L E Y ' s fundamental findings (Table X I X ) . The v a r i a t i o n presented by these coeff ic ients is not regular or predic tab le . W A K E L E Y ' s useful planting sites (2 or 3) are not numerous enough to enable a p r e c i s e in terpre ta t ion . F u r t h e r m o r e , there seems to be an age effect. W A K E L E Y ' s ideas (1961, p . 23) are worth quoting: " A c l i n a l re la t ionship of height growth to latitude proved common to l o b l o l l y at 5 years and to separate plantings of shortleaf at 5 and 3 y e a r s . When l ike sets of stocks were compared , curves of height over latitude of seed source had n e a r l y s ignif icant to highly s ignif icant nega-tive slopes in southern plantations and highly s ignif icant posi t ive slopes i n nor thern plantat ions. In plantations at intermediate lat i tudes the slopes were var iab le i n d i r e c t i o n , and less s ignif icant or non-s ign i f i cant . The 130 dominant c h a r a c t e r i s t i c of t h i s c l i n a l r e l a t i o n s h i p i s the r e v e r s a l of the slope of the c u r v e when p l a n t -i n g of the same s t o c k s i s r e p l i c a t e d at the opposite end of the s p e c i e s 'north and south' range. " T A B L E X I X C O E F F I C I E N T S F O R H E I G H T R E G R E S S E D ON L A T I T U D E O F O R I G I N T r e e S p e c i e s N o r t h e r n P l a n t i n g Site Southern P l a n t i n g Site L o b l o l l y pine 0.93 - 0.48 NS S h o r t l e a f pine (1st t e s t , 5 y e a r s old) 0. 34 NS - 0 . 9 7 Idem (2nd test, 3 y e a r s old) 0.94 -0.88 and 0.82 A f t e r W A K E L E Y (1961). Southern Conf. on F o r e s t T r e e Improv. P r o c e e d . 6: 10-24. K I N G ' S i d e a s (KING, 1965, p. 146) b a s e d on Scots pine st u d i e s a r e a l s o w o r t h quoting because the i s s u e seems to be c o n f u s i n g to some: "Thus, s t u d i e s intended to d e t e r m i n e the p a t t e r n of genetic v a r i a t i o n w i t h i n a s p e c i e s m u s t be c a r r i e d out i n m o r e than a s i n g l e e n v i r o n m e n t i n o r d e r to a c c u r a t e l y a s s e s s the v a r i a t i o n p a t t e r n . F u r t h e r m o r e , c o r r e l a t i o n s between seed s o u r c e growth c h a r a c t e r s and seed s o u r c e c l i m a t e c h a r a c t e r s (mean t e m p e r a t u r e , length of g r o w i n g season, l a t i t u d e , etc. ) m u s t be c a r e f u l l y i n t e r p r e t e d . It i s quite c o n c e i v a b l e that c o r r e l a t i o n s between, f o r example, height g r o w t h and seed s o u r c e l a t i t u d e c o u l d be p o s i t i v e , negative o r non s i g n i f i c a n t depending upon the te s t e n v i r o n m e n t . " It s eems that, when one has o n l y l i m i t e d t i m e and r e s o u r c e s , some choi c e s h o u l d be made so f a r as the t r a i t s to study a r e c o n c e r n e d i n p r o v e n a n c e t e s t i n g . 131 We decided to measure only the t ra i ts which have proven to contain no redundant in format ion and to have important adaptative values , on the bas i s of the avai lable forest genetus l i t e r a t u r e . The fol lowing charac ters were thus es t imated: germinat ion rate , bud set, epicotyl length and s u r v i v a l at the end of the f i r s t growing season; bud burs t , bud set and total height at the end of the second growing season and a s t r i k i n g charac ter : colour of the needles , which is probably l inked with f ros t and drought res i s tance . 2 .2 N u r s e r y treatments A r a n d o m i z e d complete b lock design with four repl i ca t ions was chosen. E a c h provenance was randomized wi th in each block and wi th in each provenance plot , the single tree progenies kept separate and identi f ied by a number , were randomly placed i n two l ines (see F i g . 20 and photo 1). The seeds were placed in the cavi t ies of s t y r o - f o a m blocks and the s t y r o - b l o c k s so assembled as to group a l l the progenies p e r t a i n -ing to one provenance i n one plot . E a c h progeny, for a given repet i t ion , occupied 24 cav i t i es , thus a theoret ica l number of 24 seedlings per r e p l i c a t i o n or a total of 96 seedlings per progeny was a i m e d at. We did not determine the germinat ion capacity of the progenies because of the d i f f i cu l t ies involved - and the length of t ime -in p u r i f y i n g and testing 557 seed lots of Si tka spruce . X - r a y techniques avai lable did not enable us to accurate ly separate empty f r o m f i l l e d seed and it was decided to sow 2 or 3 seeds in each cavi ty . No pretreatment was appl ied to the seeds. The preparat ion of the containers has been descr ibed by 132 Photo 1. View o f the c o n t a i n e r s t a g e . P a r t i a l view o f one row o f c o n t a i n e r s . There were two rows a l o n g each s i d e o f a s p r i n k l e r l i n e . Each row c o n t a i n e d two r e p e t i t i o n s . Photo by the a u t h o r •Drawing showxng how the styro-foam containers were assembled and seeded i n the nursery. The containers were assembled four by four. Each progeny occupied 2 rows of 12 ca v i t i e s and was randomly placed. Two provenances were placed so as to occupy four containers. 134 M A T T H E W S (1971). The s o i l m e d i u m used i n the containers consis ted of a m i x of three parts of c o m m e r c i a l peat moss and one part of h o r t i -c u l t u r a l grade v e r m i c u l i t e . D u r i n g m i x i n g , dolomite was added to b r i n g the p H of the m i x to a value of about 5. The seed was covered wi th Grani te g r i t . B l o c k A was seeded at the U n i v e r s i t y of B r i t i s h C o l u m b i a . The containers of b lock A were labe l l ed , seeded, covered with g r i t , wetted and i m m e d i a t e l y put i n a co ld r o o m (t° = 30 - 3 2 ° F ) . The lack of f a c i l i t i e s resu l ted i n lengthening the seeding process of b lock A and four days were n e c e s s a r y to seed this b lock which was completed on the 28th of A p r i l , 1971. The containers of b lock A were then moved to the B . C . -F . S . S u r r e y n u r s e r y and s tored outside. B l o c k B was seeded using the S u r r e y n u r s e r y f a c i l i t i e s and stored outside, on the 30th of A p r i l , 1971. B l o c k C was seeded on the 1st of M a y , 1971. B l o c k D was seeded on the 2nd of M a y , 1971. The four b locks were then u n i f o r m l y t reated. They r e m a i n e d for severa l days outside unt i l they were moved into the germinat ion r o o m where a humid and hot atmosphere (70°F) was mainta ined for about two weeks . Then the containers were moved outside and were so assembled as to respect the s ta t i s t i ca l design (photo 1) with the four b locks i n two long r o w s . The containers were placed on pal lets to ensure adequate drainage and vent i la t ion . 135 Photo 2. View of the t r a n s p l a n t a t i o n of the (1 + 0) S i t k a spruce se e d l i n g s . Five workers worked f o r f i v e days to t r a n s p l a n t the progenies. Photo by the author. 1 3 6 E a r l y signs of germinat ion were slow to occur and the counting of the germinants could only begin i n June. The containers were then treated according to M A T T H E W S ! , (1971) p r o v i s i o n s , so far as shade, i r r i g a t i o n and f e r t i l i z a t i o n are con-cerned . The containers with the seedlings overwintered outs ide . In A p r i l , 1972, the seedlings were thinned so as to leave only one healthy seedling per cav i ty . In M a y , 1972, they were transplanted to a spacing of 6" to 6" i n p l a i n s o i l , i n a nearby area , i n the same n u r s e r y . The seedbeds were c a r e f u l l y p r e p a r e d and the s o i l t o p - d r e s s e d with m i x i n g of 100 l b s / acre of f e r t i l i z e r (21 - 0 - 0). No s o i l analys is of the site has been done so f a r . The s o i l l e v e l is m o s t l y hor izonta l wi th a slope in the western part (aspect: west) . The s o i l is l o a m i e r in the upper part , but m o s t l y loamy sand. F i v e w o r k e r s w o r k e d for five days to transplant the 1T0 seedlings (photo 2) while the same s ta t i s t i ca l design was as much as pos -sible respected . The single tree progenies were kept separate. Ident i -f i ca t ion was done by us ing cedar poles (photo 3). The seedlings grew there during the summer of 1972 w i t h -out any f e r t i l i z a t i o n to reduce their growth as m u c h as poss ib le because the seedlings were to be transplanted i n the f i e l d . In the spr ing of 1973, the seedlings were r o o t - p r u n e d to prepare them for the l i f t i n g which took place in October , 1973. Weeding was done by hand as many t imes as n e c e s s a r y . 137 Photo 3. E a c h p r o g e n y was k e p t s e p a r a t e and i d e n t i f i e d by u s i n g c e d a r p o l e s . P h o t o by t h e a u t h o r . 1 3 8 2. 3 Measurements techniques  G e r m i n a t i o n rate The number of cavi t ies w i t h at least one germinant was counted for each progeny and each r e p l i c a t i o n us ing toothpicks of d i f -ferent colour to differentiate the dates of es t imat ion , on the fo l lowing dates: 3 rd , 9th, 15th, 21st, 28th of June, 1971, the 5th, 12th of Ju ly , 1)971. T h i s i s equivalent, for a given progeny, to a r a n d o m sampl ing of 4 x 24 ce l l s quoted 0 for absence of germinants and one for presence of at least one germinant . Consequently, the frequency d i s t r i b u t i o n of the total number of cavi t ies with at least one germinant should be d i s t r ibuted according to the B i n o m i a l d i s t r i b u t i o n . A s b lock A had been p a r t l y submitted to some m i l d cold humid treatment, the germinat ion rates observed i n that b lock w i l l be contrasted to those of the other b locks B , C and D . Bud set o w -B u d setting was determined for each provenance and each b lock separately at two dates in 1971 and at one date i n 1972. The f o i l ing procedure was u s e d : bud set was observed at different dates dur ing late August and i n September unt i l the provenances displayed a var ie ty of stages of bud set. Then an e m p i r i c a l scale of s ix bud stages was deter -m i n e d by care fu l examinat ion of the provenance shoot apices . The f o l l o w -ing c lasses were d is t inguished. Stage 0 No t e r m i n a l bud v i s i b l e . Apex obviously s t i l l g rowing . 139 Stage G e n e r a l l y a c l u s t e r of t w i s t e d , s p i r a l i n g , p a l e , g r e e n n e e d l e s . 1 A p e x f l a t , not g r o w i n g any m o r e , g r e e n i s h o r w h i t i s h , c o v e r e d by t w i s t e d n e e d l e s . 2 S m a l l t e r m i n a l bud g r e e n i s h , c o v e r e d by t w i s t e d and s p i r a l i n g n e e d l e s . 3 S m a l l bud, b r o w n i s h , s m a l l e r than i n stage 4, a l w a y s m o r e or l e s s c o v e r e d by the n e e d l e s . 4 T e r m i n a l bud w e l l v i s i b l e , but s m a l l e r than i n stage 5, b r o w n i s h , some r e s i n o u s s c a l e s a r e v i s i b l e . The needles tend to f o r m a c r o w n a r o u n d the bud, but c o v e r m o r e of the bud than i n stage 5. 5 T e r m i n a l bud w e l l v i s i b l e , b r o w n w i t h some v i s i b l e s c a l e s , the n e e d l e s g e n e r a l l y f o r m i n g , m o r e o r l e s s , a c r o w n . The t e r m i n a l bud w i l l o v e r w i n t e r under t h i s a s pect. F o r each r e p l i c a t i o n and each p r o venance, at l e a s t 30 seed-l i n g s w e r e r a n d o m l y chosen and c l a s s e d a c c o r d i n g to t h i s s c a l e i n o r d e r to e s t a b l i s h a f r e q u e n c y d i s t r i b u t i o n of the s i x bud stages and to c a l c u l a t e a weighted m e a n f o r e a c h p r o v e n a n c e and each b l o c k , A, B, C, o r D. No attempt was made to e s t i m a t e the average bud stage on a p r o g e n y b a s i s b e c a u s e of the l e n g t h of t i m e i n v o l v e d and the i r r e g u l a r and often s m a l l n umber ( l e s s than 24) s e e d l i n g s r e p r e s e n t i n g each progeny. O n l y the apex of the m a i n shoot was c o n s i d e r e d b e cause l a t e r a l b r a n c h e s m a y be at a stage d i f f e r e n t f r o m the one of the l e a d i n g shoot. I 4 i : But set was est imated on the 26th of September, 1971 and on the 2nd of October , 1971. A l l the b locks were m e a s u r e d i n one day to avoid any effect due to a sudden weather change. Bud setting was also est imated on the 8th of September, 1972 using the same p r o c e d u r e . Bud burst F l u s h i n g rate was assessed using the same procedure as for bud set on the 3 rd of A p r i l , 1972. F i v e stages of bud burs t ing of the dominant (leader) bud were dist inguished ( F i g . 21): Stage s 1 Bud swol len , s t i l l b r o w n i s h . 2 B u d swol len , green pale spots appear ing. 3 Bud swol len , tips of leaves v i s i b l e , at the center, the bud is m o r e or less f r a c t u r e d . 4 The bud i s open, the leaves are w e l l v i s i b l e , but some scales are s t i l l v i s i b l e . 5 The bud is wide open, the young needles are elongating. C o l o u r of the needles Needle colour was est imated on the 13th of September, 1972. A s i m i l a r procedure as for bud set was used for studying this c h a r a c t e r i s -t i c . Three c lasses were d is t inguished : O = needles green pale ( c a r r o t -l i k e ) ; 1 = glaucous, in termediate ; 2 = b l u i s h . L e n g t h of the epi c o t y l _ The l e n g t h of the e p i c o t y l of the th r e e g r e a t e s t s e e d l i n g s , f o r e a ch progeny and f o r each r e p l i c a t i o n , was m e a s u r e d to the n e a r e s t m m d u r i n g the w i n t e r 1971/7Z, We t r i e d o n l y to m e a s u r e a constant upper p e r c e n t a g e (- 20%) of the s e e d l i n g s p r e s e n t i n a progeny. T h e r e -f o r e , we e l i m i n a t e d the r e p l i c a t i o n w h i c h were r e p r e s e n t e d by only a few s e e d l i n g s . T h i s length i s supposed to r e p r e s e n t the t o t a l height g r o w t h a f t e r the f i r s t g r o w i n g season, i r r e s p e c t i v e of the s i z e of the e m b r y o h y p o c o t y l . About 3, 000 m e a s u r e m e n t s were made. T o t a l height In F e b r u a r y , 1973, a r a n d o m s a m p l e of 5 (1 - 1) s e e d l i n g s was m e a s u r e d f o r each r e p l i c a t i o n and each progeny so as to e s t i m a t e the n a t u r a l v a r i a t i o n o c c u r r i n g i n a progeny and a prove n a n c e . • About 10, 000 m e a s u r e m e n t s of t o t a l height w e r e made to the n e a r e s t 1/2 cm. S u r v i v a l a f t e r the f i r s t g r o w i n g s e a s o n The number of l i v i n g s e e d l i n g s was d e t e r m i n e d f o r e a c h O progeny and each r e p l i c a t i o n i n o r d e r to have some i d e a of the m a t e r i a l p r e s e n t , i n O c t o b e r , 1971. L i t t l e i n f o r m a t i o n can be gained f r o m these f i g u r e s and they w i l l not be a n a l y z e d . S u r v i v a l at the end of the second g r o w i n g season The s e e d l i n g s (1 r 0) w e r e t r a n s p l a n t e d w i t h the s o i l a t t a ched to t h e i r r o o t s , under o p t i m u m c o n d i t i o n s and, there was no n o t i c e a b l e m o r t a l i t y at the end of 1972: a l l the t r a n s p l a n t s s u r v i v e d the wi:ri£e*r 72/73 as w e l l as the summer, 1973. 143 2. 4 Methods used i n the analyses 2 .4 .1 M u l t i p l e c o r r e l a t i o n and r e g r e s s i o n analyses A s i t has a l ready been pointed out in P a r t III, Chapter 2. 1, the c o r r e l a t i o n and r e g r e s s i o n analyses do not preclude the use of d i v i -sive methods l ike the A N O V A S in provenance tests . T h e r e f o r e , the A N O V A S and the mul t ip le r e g r e s s i o n techniques w i l l be s y s t e m a t i c a l l y used . A s imple c o r r e l a t i o n m a t r i x was calculated between the provenance means of the t ra i ts assessed and between the charac ters and lat i tude, longitude and altitude of the place of o r i g i n of the provenances , over a l l the provenances , ignor ing reg ions . The geographical coordinates were chosen because the c l i m a t i c data were not suff icient or warranted for each place of o r i g i n as it has been mentioned i n P a r t I. The t ra i ts were plotted against these geographical coordinates i n order to check v i s u a l l y the re la t ionships between the provenance means and these v a r i a t e s . A m u l t i p l e r e g r e s s i o n equation was calculated between each t ra i t and a l l the var iab les or between the trai t and a few selected v a r i a t e s . The backward s t e p - w i s e p r o g r a m m e descr ibed i n P a r t II was used . 2 . 4 . 2 A n a l y s i s of var iance models F o r a l l the t ra i t s studied, on a provenance b a s i s , i . e. , a l l the t ra i t s except the two growth measurements , univar ia te analyses of var iance were done, us ing a r a n d o m i z e d complete b lock design model -complete ly random - (Model II) with four r e p l i c a t i o n s . The provenances not represented by four b locks were not inc luded. 144 F o r the g r o w t h c h a r a c t e r i s t i c s , two e l a b o r a t e m o d e l s w e r e u s e d to take into account the f a m i l y ( s i n g l e t r e e progeny) l e v e l . A s the n u mber of f a m i l i e s v a r i e d f r o m one provenance to the o t h e r , the data w e r e u n b a l a n c e d f o r t h i s l e v e l and s p e c i a l c o m p u t i n g t e c h n i q u e s f o r u n b a l a n c e d m o d e l s were used. 2.4.2.1 M o d e l f o r m a x i m u m e p i c o t y l l e n g t h A n e s t e d m o d e l c r o s s e d w i t h the b l o c k s was u s e d so as to get the d i f f e r e n t i n t e r a c t i o n s c o r r e s p o n d i n g to the d i f f e r e n t l e v e l s of v a r i a t i o n . F o r s u c h a c o m p l i c a t e d m o d e l , no f o r m u l a s to o b t a i n the components of v a r i a n c e was found f o r unbalanced data, i n the s p e c i a l i z e d l i t e r a t u r e and the g e n e r a l f o r m u l a s w e re developed. It i s to be hoped that these o r i g i n a l f o r m u l a s w i l l be u s e f u l to o t h e r g e n e t i c i s t s . The a c t u a l c a l c u l a t i o n of the u n b a l a n c e d m o d e l was r e a l i z e d by c o m b i n i n g two t e c h n i q u e s : the l i n e a r h y p o t h e s i s m o d e l t e c h -nique s u i t e d to the a n a l y s e s of u n b a l a n c e d d e s i g n and the u s u a l A N O V A technique, because the p r o g r a m m e a v a i l a b l e at the U.B.C. com p u t i n g c e n t r e f o r a p p l y i n g the l i n e a r h y p o t h e s i s m o d e l c o u l d not handle the l a r g e number of data i n v o l v e d i n the a n a l y s e s . The l i n e a r h y p o t h e s i s m o d e l was u s e d to e s t i m a t e the sums of s q u a r e s c o r r e s p o n d i n g to the p r o v e n a n c e s , b l o c k s and b l o c k by p r o v e n a n c e i n t e r a c t i o n . T o e s t i m a t e the other sums of s q u a r e s , the data w e re r e - a r r a n g e d and the u s u a l A N O V A p r o g r a m m e was used. The f i n a l sums of s q u a r e s and d e g rees of f r e e d o m w ere got by a p p r o p r i a t e s u b s t r a c t i o n s . The s t a t i s t i c a l m o d e l , c o m p l e t e l y r a n d o m (or M o d e l II), i s as f o l l o w s : 145 (D y..., i j k l i r j W i k K i j P 0 i k / j / A - : g e n e r a l mean Q^. : provenance e f f e c t ; . i s N . D . (0, ) 1 1 « N 2 : b l o c k e f f e c t ; y S . i s N . D . (0, tf£) ^ ' progeny w i t h i n provenance e f f e c t ; ^ i s N. D. (0, o j 2 ) (oCjJ : pr o v e n a n c e by b l o c k i n t e r a c t i o n ; 0(/3 i s N . D . ijkl p r o g e n y w i t h i n provenance by b l o c k i n t e r a c t i o n ; 2 C : e r r o r t e r m ; fi i s N . D . (0, <r- ) C i j k l i j k S E A R L E (1971) has e x t e n s i v e l y r e v i e w e d the a c t u a l methods f o r e s t i m a t i n g the v a r i a n c e components f r o m u n b a l a n c e d data. The e s t i -m a t i o n of v a r i a n c e components f r o m b a l a n c e d data r e s t s a l m o s t e n t i r e l y on one method, the a n a l y s i s of v a r i a n c e method d e s c r i b e d as: f o r any m o d e l , c a l c u l a t e the A N O V A as i f the m o d e l was a f i x e d e f f e c t m o d e l and then d e r i v e the ex p e c t e d v a l u e s of the mean s q u a r e s . The ex p e c t e d v a l u e s w i l l be l i n e a r f u n c t i o n s of the v a r i a n c e components. E q u a t i n g these e x p e c t e d mean s q u a r e s to t h e i r c a l c u l a t e d v a l u e s l e a d s to l i n e a r equations i n the v a r i a n c e components, the s o l u t i o n s of w h i c h a r e taken as e s t i m a t o r s of these components. H A R T L E Y (1967) has gi v e n a 14'6-g e n e r a l p r o c e d u r e a p p l i c a b l e to any un b a l a n c e d m o d e l and y i e l d i n g d i r e c t l y the n u m e r i c a l v a l u e s of the c o e f f i c i e n t s i n the f o r m u l a s f o r expe c t e d m e a n s q u a r e s . In c o n t r a s t , t h ere a r e s e v e r a l methods a v a i l a b l e f o r use w i t h u n b a l a n c e d data, w h i c h r e d u c e to the A N O V A method when the data a r e b a l a n c e d . T h e s e methods have advantages and disa d v a n t a g e s and t h e i r p r o p e r t i e s a r e not w e l l known, nor a r e there o b j e c t i v e c r i t e r i a to judge them. The A N O V A method i s the s i m p l e s t one and m o s t used. However, t h i s method, f o r m i x e d m o d e l s , l e a d s to b i a s e d e s t i m a t o r s of v a r i a n c e components and cannot be u s e d as such f o r these m o d e l s ( S E A R L E , 1971). O u r s i s c o m p l e t e l y r a n d o m and i t was d e c i d e d to use the A N O V A method. However, i t sh o u l d be noted that the analogous sums of s q u a r e s u s e d i n t h i s m e t h o d f o r u n b a l a n c e d data do not, under n o r -m a l i t y a s s u m p t i o n s , have c h i - s q u a r e d i s t r i b u t i o n s n or a r e they d i s t r i b u -t e d independently of one another. D e s p i t e t h i s , v a r i a n c e s of these e s t i -m a t o r s , under n o r m a l i t y a s s u m p t i o n s , can be d e r i v e d ( S E A R L E , 1971, p. 433). It m u s t a l s o be noted that our d e s i g n i s t o t a l l y b a l a n c e d , but f o r one n e s t e d f a c t o r and h o p e f u l l y the e s t i m a t o r s s h o u l d be o n l y v e r y s l i g h t l y b i a s e d . F o r the t o t a l l y u n b a l a n c e d case of a c o m p l e t e l y n e s t e d d e s i g n , the m e a n s q u a r e s a r e not independent because the o b s e r v a t i o n s t h e m s e l v e s have a c o v a r i a n c e s t r u c t u r e ( S C H E F F E , 1959). The a p p l i c a t i o n of the a n a l y s i s of v a r i a n c e m ethod i s thus b a s e d on the c a l c u l a t i o n of the e x p e c t e d v a l u e s of the analogous s u ms of 147 squares corresponding to the unbalanced m o d e l . The mathemat i ca l mode l (1) can be translated by the fo l lowing equations: $ goes f r o m 1 to K l 1 K 2 ' ' ' K 5 a r e c o n s t a n t s to be determined Sums of Squares (SS) Analogous Sums of Squares «• if. V?.. - — i v - i * t . . . 14.8 i | o . . . bit ^ i ^ -^ - 2 T A B L E O F D E G R E E S O F F R E E D O M Sums of Squares Degrees of F r e e d o m ti c 0 - i if JO . . . - I r Sums of squares j T A B L E X X E X P E C T A T I O N S O F SUMS O F S Q U A R E S F O R T H E T O T A L L Y  U N B A L A N C E D C R O S S E D M O D E L Expectat ions re X X - Continued quares Expectat ions Y + [ tc i^co - f i . o) 3 *J T A B L E X X - Continued Sums of squares Expectat ions Mi}.. 4 5cCO - ^ 6 ) + 0 s , 1 « T A B L E X X - Continued Sums of squares Expectat ions Net 4 [ l c i > 1 f c , ' ) ~ f ; ^ < r , ) -^-CtcO^^iC ' O Ul T A B L E X X I E X P E C T A T I O N S O F M E A N S Q U A R E S F O R T H E Sums of squaresj P A R T I A L L Y U N B A L A N C E D M O D E L Expectat ions [ M... - J _ u ^ 3 r rao - 1 3 + \ a. <£• Y + 3 <s^J, + s e + 3 rf-Jf* + 1 [3 10.}.. - 3 i£ K h l 1 3. ^  +' *T 4 154" The e x p e c t a t i o n s of the sums of s q u a r e s a r e given i n T a b l e X X . B y d i v i d i n g these e x p e c t a t i o n s by the c o r r e s p o n d i n g d e g r ees of f r e e d o m and by ta k i n g into account the fact that the data w e re ch o s e n so that n - j ^ = 3 and j = 1, .... 4, the e x p e c t a t i o n s of the m e a n sq u a r e s f o r the p a r t i a l l y u n b a l a n c e d m o d e l u s e d to a n a l y z e the e p i c o t y l l e n g t h m e a s u r e m e n t s a r e ob t a i n e d (Table X X I ) . The equations of the T a b l e X X I have been d o u b l e - c h e c k e d b y u s i n g H A R T L E Y ' S method "by s y n t h e s i s " ( H A R T L E Y , 1967), on a sample of 252 data. The t e s t s of the d i f f e r e n t f a c t o r s a r e s e l f e vident i f the ex p e c t a t i o n s of the mean s q u a r e s a r e c o n s i d e r e d . The t e s t s of the d i f -f e r e n t f a c t o r s a r e exact, i n our p a r t i a l l y b a l a n c e d case, p r o v i d i n g there i s no i n t e r a c t i o n s between the provenance and b l o c k e f f e c t s . If the l a t t e r i n t e r a c t i o n c o n t r i b u t e s s i g n i f i c a n t l y to the d i f f e r e n t mean s q u a r e s , i t i s n e c e s s a r y to use c o m p l i c a t e d l i n e a r c o m b i n a t i o n s of the m e a n s q u a r e s i n o r d e r to i s o l a t e the m a i n e f f e c t s f r o m t h e i r i n t e r a c t i o n and e q u a l i z e the c o e f f i c i e n t s of the components. The t e s t s a r e then o n l y a p p r o x i m a t e , ( S C H E F F E , 1959). Anyhow, the F t e s t s l o s e t h e i r i n t e r e s t because of the p r e s e n c e of i n t e r a c t i o n . The e r r o r s of the v a r i a n c e components have been e s t i m a t e d a c c o r d i n g to A N D E R S O N and B A N C R O F T (1952, p. 321). However, the f o r m u l a u s e d i s s t r i c t l y v a l i d f o r b a l a n c e d data. The e r r o r of the v a r i a n c e component i s g i v e n as: ^ 2 2 • , r. v ~T f. + 2 c i 155 c : c o e f f i c i e n t of the v a r i a n c e component V. : mean square u s e d to get the component f. : degrees of f r e e d o m c o r r e s p o n d i n g to V. T o t a l height a f t e r the second g r o w i n g s e a son was a n a l y z e d by u s i n g a c o m p l e t e l y n e s t e d m o d e l ( M o d e l II) w i t h the b l o c k s c o n s i d e r e d as s i m p l e r e p l i c a t i o n s : i j k m ' 1 r i j tf i j k * i j k m i = 1.2. . • • • cL j = 1.2, . . . . b. I k = 1.2, , . . c.. m = 1.2, . • ' ' n i j k A l l the c'element's?-, a r e n o r m a l l y d i s t r i b u t e d w i t h mean O and some a p p r o p r i a t e v a r i a n c e . A g a i n , the d e s i g n i s c o m p l e t e l y b a l a n c e d except f o r the f a m i l y l e v e l . The data w e r e so s e l e c t e d as to e n s u r e that a l w a y s k = 1, .... 3: c. - = 3 and n.., =5. 3 i j i j k The components of v a r i a n c e and t h e i r e r r o r w e r e ob t a i n e d b y u s i n g the f o r m u l a s g i v e n by M A H A M U N U L U (1963). The f o r m u l a s a r e e x t r e m e l y c o m p l i c a t e d and a s p e c i a l p r o g r a m m e was w r i t t e n i n o r d e r to compute the components of v a r i a n c e and t h e i r e r r o r s once a c e r t a i n number of p a r a m e t e r s have been c a l c u l a t e d . The p r o g r a m m e i s a v a i l a b l e to the i n t e r e s t e d r e a d e r . The e r r o r s of the components a r e s l i g h t l y b i a s e d b e c ause the c o v a r i a n c e of the components have not been 156 calculated as the formulas used o r i g i n a l l y were based on S E A R L E ' s book (1971) which does not mention the covariance f o r m u l a s . The pooled sums of squares used in M A H A M U N U L U ' s f o r m u l a are not ch i - square d is t r ibuted , hence the e r r o r var iance i s not c o r r e c t l y der ived e i ther . In spite of the fact that for a totally u n -balanced case of a complete ly nested design, the mean squares are not independent, independent sums of squares - not ch i - square d is t r ibuted -can be d e r i v e d (personal communicat ion of D r . G . N a m k o o r g , N o r t h C a r o l i n a State U n i v e r s i t y ) . 1 5 7 C H A P T E R 3 R E S U L T S A N D C O N C L U S I O N S 3. 1 G e r m i n a t i o n r a t e F o r each provenance, the t o t a l number of c a v i t i e s w i t h at l e a s t one g e r m i n a n t was c a l c u l a t e d f o r each date. A s the number of f a m i l i e s sown v a r i e d f r o m one pr o v e n a n c e to another, a c o r r e c t i o n was made by d i v i d i n g t h i s t o t a l number by the t o t a l n umber of c a v i t i e s sown. A s the e x a c t number of seeds sown i n each c a v i t y was not known, the r e s u l t s have o n l y a c o m p a r a t i v e v a l u e . The r a t e s have been p l o t t e d a g a i n s t t i m e i n number of days, f o r some s e l e c t e d p r o v e n a n c e s . The g e r m i n a t i o n r a t e c u r v e s a f f e c t the f o r m of a s i g m o i d ( F i g . 22). T h e s e c u r v e s a r e d i f f e r e n t , but the d i f f e r e n c e s cannot be e x p l a i n e d . T h e r e i s no c l e a r g e o g r a p h i c a l p a t t e r n of v a r i a n c e ; f o r i n s t a n c e , the c u r v e f o r pro v e n a n c e 1 ( L o w e r M a i n l a n d , B.C.) '.- ~ g a c..:. v v e r y s i m i l a r to the one of provenance 23 (Southeast A l a s k a ) . The p r o v e n a n c e s s l o w to g e r -m i n a t e m a y or m a y not e v o l v e s l o w l y and the t o t a l g e r m i n a t i o n m a y o r may not r e m a i n s m a l l (provenances 1, 23, 27, 30). T h e r e w e r e s t r i k i n g d i f f e r e n c e s between b l o c k A and the ot h e r b l o c k s . B l o c k A was f a s t e r to be c o v e r e d by s e e d l i n g s and the b l o c k B, C and D p r e s e n t e d l a t e g e r m i n a n t s throughout the g r o w i n g season. The m e a n number of c a v i t i e s o c c u p i e d by at l e a s t one g e r m i n -ant, i t s s t a n d a r d e r r o r and c o e f f i c i e n t of v a r i a t i o n have been c a l c u l a t e d f o r e a c h date and e a c h provenance, f o r b l o c k A and f o r the b l o c k s B, C 15,8 Fig.22. Germination rates plotted against time f o r some selected provenances. The numbers are the provenance numbers. Days in June 1971 159 and D pooled. The dif ferences are s t r i k i n g between block A and the other and need not be studied s t a t i s t i c a l l y . The di f ferences are most important at the beginning of the germinat ion p r o c e s s : there are often two t imes as many seedlings in b lock A as i n the other b l o c k s . The coeff icient of v a r i a t i o n is much lower in b lock A attesting that the few days of cold and humid treatment might have equal ized the g e r m i n a -tion rates of the Sitka spruce f a m i l i e s for a given provenance, while i n c r e a s i n g the rates themselves . The di f ferences , however, tend to disappear with time and by the end of the germinat ion p r o c e s s , the number of germinants and their coeff icient of v a r i a t i o n are m u c h c loser i n the two ser ies of b l o c k s . With t ime, the coeff icient of v a r i a t i o n decreases , point ing out a l e v e l l i n g of the rates of the different f a m i l i e s w i t h i n the same provenance. Table X X I I shows a few germinat ion s tat is -t ics for some provenances . A c c o r d i n g to our data, the rate of growth does not seem to be affected by the germinat ion ra te : the di f ferences between the p r o v e n -ances in germinat ion rates tend to disappear . F u r t h e r m o r e , there is no re la t ionship between height growth and the germinat ion ra tes : there is no apparent geographic t rend in the dif ferences between the provenances while there is a strong one for height growth. 3. 2 Phenolog ica l observat ions B u d burst and bud set rates are important adaptative t ra i t s as they determine the co-adaptat ion of the vegetation p e r i o d of the tree to the p e r i o d i c v a r i a t i o n of the c l i m a t i c components of the environment 160 T A B L E X X I I M E A N N U M B E R O F C A V I T I E S W I T H A T L E A S T O N E  G E R M IN A N T A N D C O E F F I C I E N T O F V A R I A T I O N (C. V . )  F O R B L O C K A A N D T H E B L O C K S B , C , A N D D P O O L E D B l o c k B , C B l o c k A and D Provenance Date M e a n C . V . M e a n C . V . 1 f i r s t 2.8 96.0 0. 07 460. 7 las t 8. 5 87. 3 5. 5 99 .5 5 f i r s t 6. 1 55. 5 1.5 144. 9 last 18. 9 20. 6 16. 0 30. 7 36 f i r s t 0. 7 159.4 0. 02 648. 3 last 16. 2 21.8 6.6 60.8 39 f i r s t 0. 5 279.0 0. 09 323. 7 last 15. 1 26. 5 12.4 48. 41 where the tree t h r i v e s . T h i s co-adaptat ion i s p a r t i c u l a r l y important for the species growing i n the temperate regions where important c y c l i c v a r i a t i o n i n day length, temperature r e g i m e , r a i n f a l l , etc . m a y ex is t . Bud set i s l inked with the cessat ion of the elongation of the s tem, but i t i s a lso r e l a t e d to the adaptation to the co ld season and the entrance i n dormancy which general ly accompanies i t , i n the temp-a-erate tree spec ies . B u d set and bud f lushing have been p a r t i c u l a r l y studied i n E U R O P E , for dif ferent tree species , i n connection w i t h height growth, late and e a r l y f rost damage (see S C H O B E R , 1962 or N A N S O N , 1964). In a m a r i t i m e , i r r e g u l a r c l imate as the one of the d i s t r ibut ion area of S i tka spruce , these t ra i t s could be of paramount importance , e s p e c i a l l y 161" i f t h i s m a r i t i m e c l i m a t e i s s u p e r i m p o s e d to a w i l d , mountainous topo-g r a phy w i t h abrupt changes i n aspect, a l t i t u d e , r a i n y w i n d w a r d and d r i e r l e e - s l o p e s o r f r o s t p o c k e t s due to t e m p e r a t u r e i n v e r s i o n . The m e t h o d o l o g y of the study of bud set of bud b u r s t i s p a r t i c u l a r l y i m p o r t a n t i f p r e c i s e c o m p a r i s o n s a r e needed. S C H O B E R (1962) has s t u d i e d the late f r o s t and w i n t e r f r o s t damage v a r i a b i l i t y o f t e n S i t k a s p r u c e p r o v e n a n c e s . F l u s h i n g r a t e was a s s e s s e d e v e r y 3-4 days u s i n g an e m p i r i c a l s c a l e of four bud st a g e s . The development of the t e r m i n a l bud takes the shape of a s i g m o i d ( F i g . 23). However, t r e e to t r e e v a r i a t i o n was n o t i c e d . B u d set a l s o takes a s i g m o i d a l f o r m . S e v e r a l methods can be u s e d to a s s e s s bud set and bud b u r s t . One c a n e s t i m a t e the percentage of bud at a g i v e n stage of development e v e r y two o r th r e e days and d r a w a c u r v e e x p r e s s i n g the e v o l u t i o n of the percentage w i t h t i m e , o r count the number of days u n t i l some type of bud a p p e a r s . T h i s l a t t e r m ethod was u s e d by B U R L E Y (196 6 a ) . T h e s e p r o c e d u r e s m a y be lon g and tedious, depending on the s i z e of the m a t e r i a l and a l s o i m p r e c i s e b e cause of the l a c k of c l e a r d e f i n i t i o n of the type of bud c o n s i d e r e d or the s m a l l number of t r e e s a s s e s s e d : B U R L E Y (1966 a) a s s e s s e d two s e e d l i n g s p e r t r e a t m e n t . A n o t h e r m e t h o d u s e d i n t h i s study, c o n s i s t s of t a k i n g into account the c o n t i n u i t y of the bud f o r m a t i o n p r o c e s s , i n d i s t i n g u i s h i n g d i f f e r e n t stages and i n e s t a b l i s h i n g , f o r e ach p o p u l a t i o n , the f r e q u e n c y d i s t r i b u -t i o n of the stages and i n c a l c u l a t i n g the ave r a g e bud stages. The £nfw/cA fangs -Tjeben.yp /'."••"> rtifht ya// en/fetter fy>C'rw !/T A'lr/' 'pf?ft;/hj 7? am FwMni* un4&S ta.tfemr. Q(hsfumt A'ftospenhati'e jetprenft f\ Ahospc gescMossen (f A/r 23 A'affforKftn, $/$Amm Mr 24 Orepen. Sws/aw A/r 25 k'ashwfhv?, G/ympir \'r 0 ' Vr 99 ' .Sn/educk Wr 90 * Qumautf A/r 27 fi/ Co/umt/a, Qutm OxrrfoPe Ms A/r /C4 A/a**(2. Baranaf -jrtsi/n PS /SS f6S 2V S 245 285S/5 So 66 fS6 m 206 246 fpf6 A b b . 33: Austreiben der Sitka im Friihjahr 1956 im Mittel v o n je 9 Stammen Versuch Forstamt Gahrenberg A b t . 174 Flushing of S i tka spruce in spring 1956, mean of 9 trees in each sample. Experiment Forstamt Gahrenberg, compt. 174. F i g . 23. Bud b u r s t e v o l u t i o n o f d i f f e r e n t S i t k a s p r u c e p r o v e n a n c e s g r o w i n g i n Germany i n t h e s p r i n g o f 1956. ( SCHOBER, 1962 ) 163' variable estimated by Burley F i g . 24» Relationships between tne one-day method to assess bud set and BURLEY1s method. 1 6 4 . r e l a t i o n s h i p between the two methods are shown i n F i g . 24, f o r an i d e a l i z e d e x a m p l e of three p r o v e n a n c e s . The method u s e d i n t h i s study-e s t i m a t e s the a v e r a g e bud set of a g i v e n p r o v e n a n c e at a g i v e n date. It i s a one-day e s t i m a t i o n ; t h e r e f o r e , the d i f f e r e n c e s e x p r e s s e d by the p r o v e n a n c e s v a r y w i t h the date c h o s e n f o r the e s t i m a t i o n . The d i f f e r -ences e s t i m a t e d a r e i n d i c a t e d a l o n g the v e r t i c a l l i n e i n F i g . 24. The d i f f e r e n c e s i n days g i v e n by B U R L E Y ' s m e t h o d a r e i n d i c a t e d b y the a r r o w s c o r r e s p o n d i n g to the h o r i z o n t a l l i n e . B e f o r e bud set s t a r t s , a l l the p r o v e n a n c e s a r e g r o w i n g and the d i f f e r e n c e s i n bud s e t t i n g , between them, a r e equal to z e r o ; when a l l the p r o v e n a n c e s a r e r e s t i n g , a g ain, the d i f f e r e n c e s i n bud set a r e equal to z e r o ; consequently, t h e r e m u s t be some date where, on the a v e r a g e , the d i f f e r e n c e s a r e m a x i m u m and con-sequently, the genetic v a r i a n c e m a x i m u m ( F A L K E N H A G E N , 1968). The n a t u r e of the f r e q u e n c y d i s t r i b u t i o n of bud f l u s h i n g stages w i t h i n d i f f e r e n t p r o v e n a n c e s of E u r o p e a n b e e c h (Fagus s y l v a t i c a L.) has been i n t e n s i v e l y s t u d i e d by G A L O U X and F A L K E N H A G E N ( G A L O U X , 1966). Two t h e o r e t i c a l d i s t r i b u t i o n s , n o r m a l and l o g n o r m a l , have been a d j u s t e d to the data a c c u m u l a t e d f o r d i f f e r e n t b e e c h p r o g e n i e s . D i f f e r e n t n u m b e r s of b u d stages: 5, 6 or 7 have been c o n s i d e r e d . The number of stages or c l a s s e s to be d i s t i n g u i s h e d i s a p r o b l e m i n i t s e l f . G e n e r a l l y , the d i s t r i b u t i o n was found to be skewed and not n o r m a l , r a r e l y l o g - n o r m a l . A c o r r e l a t i o n m a t r i x (Table X X I V ) between a l l the t r a i t s w h i c h p r e s e n t e d some s y s t e m a t i c v a r i a t i o n has been c a l c u l a t e d and bet-ween these t r a i t s and the g e o g r a p h i c a l c o o r d i n a t e s of the p l a c e of o r i g i n 165 of the provenances . The means of the provenances have been used to calculate this c o r r e l a t i o n m a t r i x , d i s r e g a r d i n g the r e g i o n s . The legend of the m a t r i x i s as f o l l o w s : L A T = latitude ( ° , 1/10). longitude ( ° , 1/10). altitude (feet). bud set est imated on the 26th of September, 1971 bud set es t imated on the 2nd of October , 1971. BST72 = bud set es t imated on the 8th of September, 1972. B B T = bud burs t es t imated on the 3rd of A p r i l , 1972. m a x i m u m epicoty l length. mean total height after the second growing season. L O N A L T B S T 1 BST 2 E P L H E T H M N H M X - mean m i n i m u m total height after the second growing season. = mean m a x i m u m total height after the second growing season. C O L = needle colour es t imated i n 1972. Table X X I I I , wi th s i m i l a r legends, shows the average values for the same var iates s tudied. A look at this m a t r i x shows that latitude and longitude of the place of o r i g i n of the provenances are p o s i t i v e l y c o r r e l a t e d (r - 0. 78). T h i s fact resu l t s f r o m the p a r t i c u l a r physiography of the nor thern part of the west coast of N o r t h A m e r i c a . T h e r e f o r e , mul t ip le r e g r e s s i o n techniques or p a r t i a l c o r r e l a t i o n coeff ic ients are n e c e s s a r y to d issoc ia te the effects of these two v a r i a t e s . T A B L E XXIII G E O G R A P H I C A L C O O R D I N A T E S O F T H E P R O V E N A N C E S A N D A V E R A G E V A L U E S O F T H E N U R S E R Y T R A I T S S T U D I E D B U D -B U D - B U D - B U D S E T P R O V . L A T . L O N G . A L T . S E T 1 S E T 2 B U R S T C O L . 72 E P L H E T H M N H M X (mm) (mm) 1 49. 12 121.93 100. 0.84 1. 10 2.67 0.81 0. 53 80. 24 196.92 150.37 238.70 2 49.92 123.25 100. 1. 10 1.41 3. 21 1. 25 1.61 79. 77 206.58 148.53 263*. 47 3 49. 38 124.62 0. 1. 58 1.64 3. 10 1.09 1. 53 87. 37 230.50 171.41 283.39 4 50. 38 125.95 0. 1. 14 1.99 2. 85 1. 18 1. 84 85.01 202.59 143.08 260.53 5 55.47 128.23 1700. 2. 52 2.96 3. 05 1. 36 3. 28 60.29 143.70 99.45 194.88 6 55. 17 127.87 2200. 2. 55 2. 85 3. 35 1. 34 3. 20 59.95 125. 61 83. 52 175.75 7 54.63 128.40 450. 2. 57 2. 79 2. 89 1. 39 2. 47 72.47 181.23 128.15 234.43 8 54. 40 128.95 100. 2. 80 2. 98 2. 86 1. 37 2. 79 72.63 167. 60 119.36 217.59 9 54. 13 128.62 550. 2. 48 2. 75 2. 90 1. 55 2. 89 66.96 167. 15 123.96 214.92 10 54. 72 128.77 450. 2. 37 2. 79 2. 92 1. 27 2. 32 71.48 184.91 130.33 244.79 11 55.68 128.68 800. 2. 44 2. 95 3. 13 1. 20 3. 07 57.66 135. 63 90. 07 184.56 12 55. 35 128.95 850. 2. 38 2. 74 2. 94 1. 19 2. 79 64. 31 153. 38 106.04 205. 26 13 55. 15 129.22 50. 2. 24 2. 68 3. 01 1. 28 2. 80 67. 03 155.09 99.64 209.70 14 55. 15 128.97 1300. 2. 50 2. 87 2. 84 1. 29 3. 00 66. 11 151.96 105.87 205.97 15 54. 20 129.92 0. 1. 94 2. 39 2. 56 1. 42 2. 05 81. 71 184.39 135.34 238.34 18 55. 02 128.32 800. 2. 30 2. 81 3. 44 1. 25 3. 13 59. 18 141.28 95.09 193.55 19 54.28 129.42 100. 2. 35 2. 79 2. 91 1. 33 2. 29 65.25 174. 05 121.53 228.10 20 54. 03 130.37 1000. 2. 01 2. 49 2. 06 1. 49 2. 49 65.90 142.91 92.47 192.94 21 54.20 130.25 50. 1. 91 2. 00 2. 24 1. 35 2. 20 75. 02 172.38 125.47 222.09 23 55. 03 131.55 °- 1. 71 2. 22 2. 09 1. 40 1. 85 67. 85 153.44 106.73 204.27 24 55. 50 133. 13 0. 2. 05 2. 86 2. 35 1. 54 2. 79 73. 13 170.18 116.91 222.89 T A B L E X X I I I - Continued BUD-BUD- BUD- B U D S E T P R O V . L A T . LONG. A L T . S E T 1 S E T 2 B U R S T C O L . 72 E P L H E T H M N H M X ((mm) (mm) 25 55.47 132.67 0. 2. 56 3. 03 2. 70 1.64 2. 84 73.66 173.30 120.48 324.35 26 55.42 131.70 50. 2. 10 2.69 2. 34 1.60 2. 30 67. 56 156. 20 103.37 211.09 27 56. 58 132.73 25. 2.48 2.95 2. 54 1.68 2.97 66. 29 154.94 113.32 205. 00 28 58. 37 134.58 100. 2. 71 3. 08 2. 18 1. 54 3. 37 68.61 144.76 95. 72 200.87 29 49.25 122.60 650. 0.89 0.96 2. 75 0. 74 1. 59 82. 89 205. 65 156.38 251.04 30 48. 38 123.87 o. 0. 51 0. 55 3. 13 0. 61 0. 51 91. 37 217. 58 162.32 279.87 31 48. 58 124.40 25. 0. 56 0. 34 2.43 1. 33 0. 60 76. 87 202.95 148.84 259.68 32 50. 08 127.50 100. 1. 28 1. 51 2.95 1. 35 1.68 79. 32 193. 57 145.76 239.69 33 49.83 126.67 10. 0. 55 0. 74 2. 54 1. 32 1. 69 78. 87 191.25 132.14 249.29 34 50.62 128.12 100. 1. 35 1. 72 2. 60 1. 42 1. 78 74.89 169.88 126.07 213.73 35 52. 28 131.22 50. 1.17 1.49 2. 34 1. 21 1. 66 77. 21 208.62 151.33 266.87 36 52. 87 132.08 50. 1. 35 1.91 2. 47 1. 33 2. 03 71. 53 158. 17 106.05 211.04 37 53. 05 132.08 200. 1. 54 2.43 2. 55 1. 49 2. 16 87.96 199.86 137.24 258.03 38 53. 13 131.80 250. 1. 61 2. 08 2. 19 1. 29 2. 04 77. 03 170.57 114.08 233.84 39 53. 50 132. 17 300. 1. 52 1.99 2. 77 1. 20 1. 96 72. 62 186.95 133.07 234.97 40 53.92 132.08 0. 1. 22 1. 60 2. 60 1. 24 1. 86 89. 55 207. 83 148.13 263.69 41 48. 90 124.95 700. 0.95 0. 80 2. 90 1. 12 1. 34 81. 77 211. 13 154.39 263. 95 T A B L E X X I V C O R R E L A T I O N M A T R I X B E T W E E N T H E T R A I T S S T U D I E D A N D T H E G E O G R A P H I C A L C O O R D I N A T E S O F T H E P L A C E O F O R I G I N L A T L O N A L T BST1 BST2 E P L B B T BST72 HETo H M N H M X C O L L A T 1.0 L O N .78 1.0 A L T .26 - . 1 2 1.0 36 D . F . BST1 .89 .53 .39 1.0 r . 05 - . 32 BST2 .92 .63 .32 .96 1.0 r .01 = .41 E P L - .71 - . 3 4 - . 56 - . 7 4 - . 7 0 1.0 r .001 = .51 B B T - . 1 4 - .51 .42 .13 .066 - . 1 3 1.0 BST72 .88 ., 59 .47 .91 .92 - . 7 5 .12 1.0 H E T - . 7 8 - . 4 9 - . 5 2 - . 7 4 - . 74 .91 .037 - . 8 0 1.0 H M N - . 8 0 - . 5 4 - . 49 - . 7 2 - . 76 .88 .069 - . 81 .98 1.0 H M X - . 7 4 - . 4 3 - . 5 3 - . 7 2 - . 7 2 .90 .019 - . 7 7 .98 .95 1.0 C O L .64 .72 - .039 .58 .62 - . 4 4 - . 3 7 .61 - . 5 0 - . 5 3 - . 4 8 1.0 oo. F i g . 2 5 . R e l a t i o n s h i p s b e t w e e n bud s e t o f t h e S i t k a s p r u c e p r o v e n a n c e s and l a t i t u d e o f p l a c e o f o r i g i n a t two d a t e s , i n 1971. The numbers 1 t o 41 a r e t h e p r o v e n a n c e numbers. The 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 o f l a t i t u d e w e r e 79% and 83% f o r bud s e t e s t i m a t e d on t h e 26th S e p t . 1971(crosses) and on t h e 2nd O c t . 1 9 7 1 ( c i r c l e s ) r e s p e c t i v e l y . 4 ^ ^ 5^ 5T 5^ 5^ 5^ lts° 5 V 5 ^ 58^ 5 V 6 ^ 2 f 6T 6 ^ L A T I T U D E N O R T H 170 B u d set 1 and 2 and bud set 72 a r e s t r o n g l y c o r r e l a t e d w i t h l a t i t u d e (r = 0.89 o r 0.92 and 0.88). T h e r e a r e a l s o at l e a s t s i g n i f i c a n t l y c o r r e l a t e d w i t h longitude and a l t i t u d e ; however, the r e l a t i o n s h i p s a r e somewhat weaker. What i s m o s t i m p o r t a n t i s the c o n s t a n c y of a l l c o r -r e l a t i o n s o v e r the y e a r s '71 and ?'72. T h i s would i n d i c a t e a s t r i c t g e netic c o n t r o l of bud set, at l e a s t i n c o m p a r i s o n w i t h the d i f f e r e n c e s i n the g r o w i n g c o n d i t i o n s of the-two y e a r s and the two n u r s e r y l o c a t i o n s . F i g . 25 shows the r e l a t i o n s h i p between bud set 1 and 2 and l a t i t u d e . The g e n e r a l r e l a t i o n s h i p i s l i n e a r . M u l t i p l e r e g r e s s i o n and c o r r e l a t i o n s a n a l y s e s show that a l t i t u d e does not c o n t r i b u t e s i g n i f i c a n t l y to the v a r i a t i o n o f bud set 1 and 2. 86. 2 p e r c e n t of the v a r i a t i o n i n bud set i s e x p l a i n e d by longitude and l a t i t u d e , 79. 0 p e r c e n t by l a t i t u d e alone. 84. 8 p e r c e n t of the v a r i a t i o n i n bud set two i s e x p l a i n e d by longitude and l a t i t u d e , 83.0 p e r c e n t by l a t i t u d e a l o n e . The r e g r e s s i o n equations a r e : budset 1 = -3. 28 - 0. 0936Long +• 0. 322Lat o r = -0. 0105 f 0. 232Lat. budset 2 = -8. 16 - 0. 052 Long +• 0. 3 2 L a t o r = -12. 20 +• 0. 2 6 9 L a t F o u r t y - f o u r p e r c e n t o'febud b u r s t i s e x p l a i n e d by l a t i t u d e and l o n g i t u d e and 26 p e r c e n t by l o n g i t u d e alone. The m u l t i p l e r e g r e s s i o n equations a r e : b u d b u r s t = 12. 8 4- 8. 9 4 L a t - 0. 001 5Long or = 1 0 . 0 - 0 . 0 5 6 6 ong 171. P a r t i a l c o r r e l a t i o n c o e f f i c i e n t s a r e v e r y u s e f u l i n r e p r e s e n t -i n g m u l t i p l e r e l a t i o n s h i p s . If bud b u r s t i s s y m b o l i z e d by y If a l t i t u d e i s s y m b o l i z e d by X If longitude i s s y m b o l i z e d by X^ I f l a t i t u d e i s s y m b o l i z e d by X^ then: r y X 1 / ' X 2 X 3 = 0.20 NS >|: % s}: ryV xi x3 = - 0- 5 1 W x i x 2 = ° ' 2 7 N S Thus, t h e r e i s a negative c o r r e l a t i o n of 0. 51 between bud f l u s h i n g and longitude f o r f i x e d l a t i t u d e and a l t i t u d e , but not w i t h the other p a r a m e t e r s : the m o r e m a r i t i m e the l o c a l c l i m a t e , the l a t e r the p r o v e n -ances f l u s h . Thus, t h e r e i s some evidence that the d i f f e r e n c e s i n bud f l u s h i n g a r e r e l a t e d to the t h e r m i c c o n d i t i o n s of the p l a c e of o r i g i n , p r o b a b l y w i t h the l o c a l l a t e f r o s t d i s t r i b u t i o n as a l r e a d y o b s e r v e d by S C H O B E R (1962). N i n e t y - o n e p e r c e n t of the t o t a l v a r i a t i o n i n bud set 72 can be a t t r i b u t e d to the i n t e r a c t i o n of l a t i t u d e , a l t i t u d e , bud set 2, bud b u r s t and c o l o u r ; A l l these v a r i a b l e s c o n t r i b u t i n g s i g n i f i c a n t l y to the v a r i a t i o n i n bud set 72. E i g h t y - s e v e n p e r c e n t i s e x p l a i n e d by l a t i t u d e , a l t i t u d e and bud b u r s t ; e i g h t y - f o u r p e r c e n t b y l a t i t u d e and bud b u r s t ; s e v e n t y - e i g h t p e r c e n t by l a t i t u d e alone. I f 2 -Thus, there i s a c o v a r i a n c e s t r u c t u r e between bud set 72 and bud b u r s t , bud set 2 and needle c o l o u r . A s these t r a i t s a r e s t r i c t l y -h e r i t a b l e , i t i s p o s s i b l e that genetic c o r r e l a t i o n s e x i s t between these t r a i t s , p o s i t i v e o r n e g a t i v e . T h e r e f o r e , i n a s e l e c t i o n e x p e r i m e n t , c a r e s h o u l d be taken of the c o r r e l a t e d r e s p o n s e s when s e l e c t i n g f o r a g i v e n t r a i t s uch as la t e f l u s h i n g o r e a r l y bud s e t t i n g . B u d set 1 and 2 may be c o n s i d e r e d as one t r a i t b ecause of the s t r o n g c o r r e l a t i o n c o e f f i c i e n t between them and the other t r a i t s . B u d set 72 i s s t r o n g l y c o r r e l a t e d w i t h e p i c o t y l l e n g t h and t o t a l height (r = -0.75 o r -0.81). B u d set, i n g e n e r a l , i s p o s i t i v e l y c o r r e l a t e d w i t h needle c o l o u r . The h i g h e r the l a t i t u d e of the p l a c e of o r i g i n of the p r o v e n -ances, the e a r l i e r these p r o v e n a n c e s set t h e i r bud and the s h o r t e r the height of the p r o v e n a n c e s . Thus, the e a r l i e r they set t h e i r bud, the s h o r t e r the p r o v e n a n c e s . B u d b u r s t i s not s i g n i f i c a n t l y c o r r e l a t e d w i t h bud set or e p i c o t y l l ength. Our data, t h e r e f o r e , c o n f i r m B U R L E Y ' s (1966 ) f i n d i n g s cL that s e e d l i n g s f r o m n o r t h e r n s o u r c e s f o r m t h e i r buds b e f o r e the s o u t h e r n p r o v e n a n c e s , f i n d i n g s b a s e d on the study of the f o r m a t i o n of a c e r t a i n type of b u d (Type II, a c c o r d i n g to B U R L E Y ) . However, some of h i s ob-s e r v a t i o n s on the Type III bud c o u l d c o n t r a d i c t h i s r e l a t i o n s h i p s b a s e d on h i s Type II bud. However, our r e s e a r c h does not c o n f i r m B U R L E Y ' s con-tention that f lushing rate i s not re la ted to the place of o r i g i n of the Si tka spruce provenances . It i s also important to note that B U R L E Y has not studied the re la t ionship between the number of leaf p r i m o r d i a and i n t e r -nodes and the next y e a r ' s shoot growth. A c c o r d i n g to R O M B E R G E R (1963), for some species of P i n u s or P i c e a , the number of internodes te lescoped in the t e r m i n a l bud determines the total shoot growth of the next growing season. It i s not known i f S i tka spruce has t e r m i n a l buds w i t h p r e d e t e r m i n e d shoots and f u l l y p r e f o r m e d needles and internodes . A t the S u r r e y n u r s e r y , the Si tka spruce provenances grew continuously u n t i l f a l l . No repeated f lushes have been not iced by us , which would indicate ( K O Z L O W S K I , 1971) that S i tka spruce has no shoots p r e f o r m e d i n the dormant buds. B u d set i n trees is cons idered by many authors as often set i n mot ion by dec l in ing photoperiod during late summer through some " p h y t o c h r o m e " m e c h a n i s m ( R O M B E R G E R , 1963; K O Z L O W S K I , 1971). The data presented here do not support this theory, as F i g . 26 seems to show, because at the t ime of m e a s u r e m e n t s , the di f ferences between the photoperiods at lat i tudes 50° and 58° are too s m a l l to expla in the di f ferences i n bud setting expressed by the provenances growing at S u r r e y . P e r h a p s the m e c h a n i s m t r i g g e r i n g the onset of dormancy is set i n mot ion e a r l i e r , dur ing the s u m m e r . The same holds true for bud f lushing as ob-served i n the n u r s e r y . Thus , it i s d i f f i cul t to attribute the dif ferences i n bud f lushing and bud set observed i n the n u r s e r y as due to different adap-tations to different photoperiods at the place of o r i g i n of the provenances . F i g . 2 6 . E v o l u t i o n o f d a y - l e n g t h f r o m M a r c h t o O c t o b e r f o r two 175 However, the t h e r m a l e n e r g y a c c u m u l a t e d o v e r the g r o w i n g season v a r i e s m u c h i n quantity and r a t e of a c c u m u l a t i o n w i t h l a t i t u d e , as a look at F i g . 5 m a y suggest. The ef f e c t of a l t i t u d e c o u l d be the same. It has been suggested by G A L O U X (1966) that the r a t e of bud f l u s h i n g i n F a g u s s y l v a t i c a L. i s a p h y s i o l o g i c a l p r o c e s s depending on an adequate a c c u m u l a t i o n of t h e r m a l e n e r g y i n the e n v i r o n m e n t - m o s t p a r t i c u l a r l y the s o i l , once d o r m a n c y has been b r o k e n . T h i s a c c u m u l a t i o n of c a l o r i e s i s l i n k e d w i t h the number of degree-days a c c u m u l a t e d i n the s p r i n g . T h e r e w o u l d be a need f o r r e a c h i n g some a c c u m u l a t e d t h r e s h o l d b e f o r e f l u s h i n g can take p l a c e , once the buds a r e q u i e s c e n t and r e a d y to f l u s h . F i g . 6 migh t i n d i c a t e that the d i f f e r e n c e s i n bud b u r s t m i g h t be l i n k e d w i t h adaptation to d i f f e r e n t t e m p e r a t u r e t h r e s h o l d s : the h i g h e r the a l t i t u d e of the place of o r i g i n , the q u i c k e r to f l u s h when g r o w i n g i n a m i l d e r e n v i r o n m e n t . Note that the c o r r e l a t i o n of bud b u r s t w i t h a l t i t u d e i s +• 0.42 '. G e n e t i c c o n t r o l of time of bud b r e a k i s w e l l known and the b e g i n n i n g and c e s s a t i o n of shoot g r o w t h often a r e g e n e t i c a l l y f i x e d i n r e l a t i o n to a g i v e n p h o t o p e r i o d ( K O Z L O W S K I , 1971). B u d b u r s t i n F a g u s s y l v a t i c a L. has been shown to be i m p o s s i b l e without an adequate photo-p e r i o d f o r some t i m e ( L A V A R E N N E - A L L A R Y , 1965 i n G A L O U X , 1966). C o m p l e x t e m p e r a t u r e - p h o t o p e r i o d i n t e r a c t i o n s may p l a y some r o l e i n t r i g g e r i n g both bud f l u s h i n g and bud set i n f o r e s t t r e e s (VEGIS, 1965). N a t u r a l l y , the study o f the gr o w t h r e s p o n s e s of S i t k a s p r u c e p r o v e n a n c e s to s h o r t e n e d and extended p h o t o p e r i o d s and under d i f f e r e n t t e m p e r a t u r e r e g i m e s s h o u l d p r o v i d e some c l u e s as to the m e c h a n i s m s 17 6 of bud burs t and bud set prevalent i n this tree spec ies . A N O V A s and D U N C A N ' s mul t ip le range tests (c< = 0.05) have been c a l -culated on bud set 1 and 2 and bud b u r s t , for each reg ion separate ly . The fo l lowing tables s u m m a r i z e the analyses done for some regions as examples . The complete analyses are avai lable on request . The model used is a complete ly r a n d o m i z e d block design (Model II) with four b l o c k s . Region 1 = provenances 1, Z, 3, 4, Z9 F Values and Signif icance  Sources of V a r i a t i o n D . F . B u d set 1 B u d set Z B u d burst % afc ajc ?|< >|- ^ Provenances 4 13.67 7. Z9 4. Z0o B l o c k s 3 0 . Z 4 N S 0. Z3 NS 4.13"' E r r o r IZ T O T A L 19 Duncan's tests : (c< = 0.05). The greatest value is always on the left . B u d set 1 P r o v . N o : _3 4 2 29 1 Value 1.'58 1'. 14 l . ' l O 0J89 C~84 B u d set Z P r o v . N o : 4 3 Z 1 T * — r - - * -V a l u e : 1.99 1.64 1.41 1.10 0.96 B u d burst P r o v . N o : 2 3 4 29 1 V a l u e : 3*. ZI 3^ 10 2?85 zi 75 2.'67 Region 2 = provenances 30, 31, 32, 33, 41 F Values and Signif icance Sources of V a r i a t i o n D . F . Bud set 1 Bud set 2 B u d burs t Provenances 4 1 5 . 6 5 * * * 11.46 **'* 5. 02 B l o c k s 3 3.01 NS 3 . 6 2 * 1.82NS E r r o r 12 T O T A L 19 177 Duncan's tests : ( ©< = 0.05), B u d set 1 B u d set 2 B u d burs t P r o v . N o : 32 41 33 31 30 V a l u e : l i 28 0*95 ot~55 0?~55 0?51 P r o v . N o : 32 41 33 30 31 V a l u e : l i 51 o'. 80 (K14 0*755 5T34 P r o v . N o : 30 32 41 33 31 V a l u e : 3| 13 2.95 2I9O 2t~54 2?43 Region 5 = provenances 5, 6, 7, 9, 10, 11, 12, 13, 14, 15, 18, 19, 20, 21 F Values and Signif icance  Sources of V a r i a t i o n D . F . B u d set 1 B u d set 2 B u d bur st Provenances 13 4 .68 9.86 9 .49 , u i B l o c k s 3 1 . 7 7 N S 0.87 NS 11 .49* * * E r r o r 39 T O T A L 55 Duncan's tests : (o< = 0 .05) . B u d set 1 P r o v . N o : 7 6 14 5 9 11 10 12 19 18 13 20 15 21 V a l u e : 2.51 2.55 2.50 2.50 2|48 2.44 2.38 2.38 2^35 2^30 ZZ4 2.01 L94 1.91 B u d set 2 P r o v . N o : 5 11 14 6 18 7 10 19 9 12 13 20 15 21 Va lue : 2.96 2.95 2.87 2.85 2J81 2!79 £ 7 9 Jl3 2!75 z!74 2^68 2.40 2^39 2!oQ B u d burs t P r o v . N o : 18 6 11 5 13 12 10 9 ,7 14 15 21 20 V a l u e : 3.44 X35 3ll3 3.05 3.01 2^4 2^92 2.'90 2^89 2l84 2!56 2.24 2io6 G e n e r a l l y , i t can be seen that the D U N C A N ' s tests for the two bud sets 1 and 2 can be different despite the s i m i l a r c o r r e l a t i o n coeff ic ients with latitude or longitude. The combined effects of sampl ing e r r o r and speci f ic rates of evolution of the bud setting could be one explanation. 178 Photo 4 . V a r i a t i o n i n needle c o l o u r was s t r i k i n g . The h i g h e r the l a t i t u d e or the g r e a t e r the longitude of the p l a c e of o r i g i n , the m o r e b l u i s h the p rovenance. Photo by the author. 179 3. 3 N e e d l e c o l o u r B U R L E Y (1966^) has o b s e r v e d genetic v a r i a t i o n i n needle and h y p o c o t y l c o l o u r of d i f f e r e n t S i t k a s p r u c e p r o v e n a n c e s . The b l u i s h b l o o m o c c u r r i n g on p l u m f r u i t s , e u c a l y p t l e a v e s and needles of blue s p r u c e , i s w e l l known and has been l i n k e d w i t h c u t i c u l a r wax s t r u c t u r e s . It was d e c i d e d to study the needle g l a u c o u s n e s s on a pr o v e n a n c e b a s i s , u s i n g an e m p i r i c a l s c a l e of b l u i s h b l o o m i n t e n s i t y c o n s i s t i n g of three c l a s s e s , i n o r d e r to a c c u r a t e l y check the q u a l i t a t i v e o b s e r v a t i o n s of B U R L E Y . Indeed, the d i f f e r e n c e s w e r e s t r i k i n g (photo 4). N e e d l e c o l o u r i s p o s i t i v e l y c o r r e l a t e d w i t h l a t i t u d e , l ongitude, (r = 0.64 and 0.72), w i t h bud set 1,bud set 2, n e g a t i v e l y c o r r e l a t e d w i t h e p i c o t y l l e n g t h o r t o t a l h eight, (Table X X I V ) . Thus, the h i g h e r the l a t i t u d e o r longitude, the m o r e b l u i s h the p r o v e n a n c e s a r e and the s h o r t e r they a r e . When the i n f l u e n c e s of l a t i t u d e , longitude and a l t i t u d e of the p l a c e of o r i g i n a r e c o n s i d e r e d s i m u l t a n e o u s l y , only longitude c o n t r i b u t e s to c o l o u r v a r i a t i o n : 51.8 p e r c e n t of the v a r i a t i o n i n needle c o l o u r i s ex-p l a i n e d by lo n g i t u d e . 3. 4 G r o w t h c h a r a c t e r i s t i c s M a x i m u m e p i c o t y l l e n g t h and t o t a l height a f t e r the second g r o w i n g season of the t r a n s p l a n t s (1-1 se e d l i n g s ) w i l l be s t u d i e d u s i n g the w e l l known tec h n i q u e s : m u l t i p l e r e g r e s s i o n a n a l y s i s and anovas. 3.4.1 M a x i m u m e p i c o t y l l e n g t h ( E P L ) E P L has been r e l a t e d to a l l the other v a r i a b l e s and w i t h an a d d i t i o n a l one not i n c l u d e d i n T a b l e X X I V : a m e a s u r e of ave r a g e d e n s i t y 180 obtained by dividing the number of seedlings at the end of the f i r s t growing season by the number of cavi t ies sown, for each progeny and each r e p l i c a t i o n ; then the average was ca lculated for each provenance. The corre la t ions are ca lculated over a l l provenances, neglecting the r e g i o n s . The average density so calculated only contributed 1.08% of the total coeff icient of determinat ion of a l l the independent var iab les cons idered , thus the average number of seedlings per cavity did not influence the growth of the f a m i l i e s during the f i r s t y e a r ; the a v a i l a b i l i t y of water and m i n e r a l nutr ients being suff ic ient . E ighty percent of E P L is explained by the l inear combina-tion of latitude, longitude, al t i tude, co lour , bud burs t and density. S i x t y - f i v e percent by latitude and altitude only . F i f t y percent by latitude alone. F i g . 27 indicates the re la t ionship between E P L and lat i tude. The mult ip le r e g r e s s i o n equations a re : E P L = 182.58 - 2.00 L a t . - 0.00682 A l t . or E P L = 198.44 - 2. 35 L a t . Another m u l t i p l e r e g r e s s i o n analys is shows that 74% of E P L is explained by altitude and bud set 1. S i x t y - s i x percent by bud set 1 alone. T h i s might indicate that the rate of bud setting determines some-how height growth. The ca lculat ion of p a r t i a l c o r r e l a t i o n coeff ic ients shows that i f : y = E P L X n = altitude F i g . 2 7 . R e l a t i o n s h i p between maximum e p i c o t y l l e n g t h and l a t i t u d e o f p l a c e o f o r i g i n o f t h e p r o v e n a n c e s . The c o e f f i c i e n t o f c o r r e l a t i o n i s - 0 . 7 1 . •30 •41 •3 •29 . 4 0 . 4 •37 •1 .2 •31 3 2 15 •34 •35 •38 •36 •39 2 1 8.7, 2 0 - T 1 9 10 2 5 -* 2 4 1 4 . 1 2 18. * 6 * 5 11 •27 •28 30 4 -49° 50° 51° 52° 53° 54° LATITUDE NORTH 55" 56° 57c 58° 59° ft 182; X 2 = longitude x 3 = l a t i t u d e f o r f i x e d X^ and X = 3 = -0.365 NS r y X 2 f o r f i x e d X^ and X^ = = + 0. 275 NS r y X 3 f o r f i x e d X 2 and X = 1 = -0.535* The p a r t i a l c o r r e l a t i o n a n a l y s i s shows - as i t shou l d be -the p r i m o r d i a l i n f l u e n c e of l a t i t u d e i n e x p l a i n i n g E P L v a r i a t i o n . F o r a giv e n longitude and a l t i t u d e , E P L s i g n i f i c a n t l y d e c r e a s e s w i t h l a t i t u d e : the p a r t i a l c o r r e l a t i o n c o e f f i c i e n t i s - 0.535, w h i c h i s r e m a r k a b l e b e c a u s e the c o n t r i b u t i o n s of longitude and a l t i t u d e have been e l i m i n a t e d . If o n l y longitude i s kept constant, the c o r r e l a t i o n w i t h l a t i t u d e i s i n c r e a s e d f° r f i x e d X^ = -0.75 . F i x i n g longitude a l s o i n c r e a s e s the c o r r e l a t i o n w i t h a l t i t u d e r ,.. f o r f i x e d X = -0.627 Yxl 2 The a n a l y s e s of v a r i a n c e , the components of v a r i a n c e and (*) the D U N C A N's te s t s have been c a l c u l a t e d f o r each r e g i o n s e p a r a t e l y , by u s i n g the m o d e l a n d methods d e s c r i b e d i n P a r t III, S e c t i o n 2.4.2. The d e t a i l s of the c a l c u l a t i o n s a r e g i v e n f o r one r e g i o n only: r e g i o n 1. The other r e g i o n s w i l l be p r e s e n t e d under the f o r m of sh o r t t a b l e s . In o r d e r to e s t i m a t e the v a r i a n c e components, a c o m p l e t e l y r a n d o m m o d e l was as s u m e d . However, the p r a c t i c a l t r e e b r e e d e r w i l l be i n t e r e s t e d i n knowing how a p a r t i c u l a r p r o v enance o u t p e r f o r m e d , i n the n u r s e r y , another provenance; t h e r e f o r e , D UNCAN's te s t s f o r the pr o v e n a n c e means were c a l c u l a t e d w h i c h would, t h e o r e t i c a l l y , i m p l y that the p r o v e n a n c e effect i s f i x e d . 183 Region 1 = provenances 2, 3, 4, 29 Sources of V a r i a t i o n D . F . : S .S . M . S . F B l o c k s 3 454.53895 151.51298 NS Provenances 3 3, 387. 58675 1,129.19558 240 NS T r e e s wn p r o v . 41 19, 248.41325 469.47349 *X, 4. 29 P r o v . X block 9 2, 163.65625 240.40625 2. 20 NS T r e e wn p r o v X block 123 13,455.57393 109.39491 1. 26 NS E r r o r 360 31,283 86.89722 T O T A L 539 No s ignif icant di f ferences between provenances , thus no D U N C A N ' s test. Components of var iance N = 540 1 / N N . = 1 3 5 • J. 2 1 / N . SI. N • J. i i j 2 •i N i 146.4 36. 6 The equations become: 2 2 2 2 2 131.2 +• 32.68 C ^ t 12 0 ^ + 3 ( X ^ +• cr 135 a \ , 3 6 . 6 ^ 3 o - 2 12 CT +• 3 <5 32.8 <5\ n +• 3 cr, 2 <*(>> ^ 3 CT 2 c r e cr = 1,129.19558 =151.51298 = 469.47349 = 240.40625 = 109.39491 = 86.89722 184 Solving these equations g ive : 2 0~ = 86.89722 e (^b = ?- 4" 2 3 _ 2 ° ^ = 30.00654 2 ^ = 4.029807, etc. See Table X X V for the e r r o r of the components of v a r i a n c e . Region 2 = provenances 30, 32, 34, 41 A n a l y s i s of var iance Sources of V a r i a t i o n D . F . S .S . M . S . F B l o c k s 3 1,343.51915 447.83971 1. 64 Provenances 3 6,062.39993 2,020.79997 3 .98* T r e e wn p r o v . 17 8,631.60007 507.74118 3. 77 P r o v . X block 9 2, 458, 30680 273.14520 2. 03 NS T r e e wn prov X b l o c k 51 6 ,862. 26898 134.55429 1. 30 NS E r r o r 168 17,277 102.83928 T O T A L 251 D U N C A N ' s test: (oc = 0.05) • E P L = 90. 5 78. 8 78.5 78.5 m m P r o v . nr = ~3*0 4 1 - 34" 3*2 Components of v a r i a n c e : see Table X X V . 1&5 R e g i o n 3 = p r o v e n a n c e s 35, 36, 37, 39, 40 A n a l y s i s of v a r i a n c e S o u r c e s bf V a r i a n c e D.F. S.S. M.S. F B l o c k s 3 2, 165. 02593 721.67531 2. 84 NS P r o v e n a n c e s 4 27~ 397.26486 6,849.31621 16. 16 T r e e wn pr o v . 24 10,173.73514 423.90563 *l- -«t> •V* 3. 08 P r o v . X b l o c k 12 3,051.92880 254.32740 1. 85 NS T r e e wn p r o v . X b l o c k 72 9,907.62696 137.60593 1. 38 NS E r r o r 232 23,201 100.00431 T O T A L 348 D U N C A N ' S te s t : (ot = 0.05). P r o v . No. 37 40 35 39 36 E P L = 92'. 0 86:8 83: 6 701.8 70.'5 Components of v a r i a n c e : see T a b l e X X V R e g i o n 4 = p r o v e n a n c e s 24, 25, 26. 27 A n a l y s i s of v a r i a n c e S o u r c e s of v a r i a t i o n D.F. S.S. M.S. F B l o c k s 3 1,042.15836 347.38612 2. 62 NS P r o v e n a n c e s 3 3,844.71402 1,281.57134 1. 81 NS T r e e wn p r o v . 18 12,770.28598 709.46033 6.44 *** P r o v . X b l o c k 9 1, 193.48817 132.60979 1. 20 NS T r e e wn p r o v . X b l o c k 54 5, 947.41183 110.13725 1. 57 NS E r r o r 176 12, 313 69.96022 T O T A L 263 No s i g n i f i c a n t d i f f e r e n c e s ; thus no DUNCAN's test, f o r the provenance e f f e c t . 186 Components of variance :: see Table X X V . Region'5 = provenances 5, 6, 7, 8, 9, 10, 19, 21 11, 12, 13, 14, 15, 18, A n a l y s i s of variance Sources of V a r i a t i o n D . F . : s.s. • ' ' • M . S . F B l o c k s 3 3,958.18031 1,319.39343 12 .04*** Provenances 13 74', 524.47710 5,732.65185 10 .68** * Tree wn p r o v . 83 44,535.52290 536.57256 6 . 9 0 * * * P r o v . X block 39 4,274.27871 109.59689 1. 41 NS Tree wn p r o v . X b l o c k 249 19,;369. 45602 77.78898 1. 19 NS E r r o r 776 51,243 66.03479 T O T A L 1, 163 D U N C A N ' s test: ( o<: = 0.05). P r o v . N o . 15 21 10 8 7 13 14 9 12 19 5 6 11 18 E P L '= 81.0 78.8 75.0 72, 6 72.0 66.4 65.6 65.5 62.7 61.9 60.4 59.4 58.6 57.0 Components of v a r i a n c e : see Table X X V . In the fol lowing and preceding analyses , only the provenances represented by at least three f a m i l i e s and the f a m i l i e s represented by four complete block measurements were cons idered . To see if some t rans format ion of the o r i g i n a l epicotyl m e a s u r e -ments was necessary , in order to s tabi l ize the var iances of the means , the means of the E P L of the provenances have been plotted against their var iance on a b i - l o g a r i t h m i c paper . The re la t ionship ( F i g . 28) is too weak to ensure a useful t rans format ion which would be close to a r e c i p r o -ca l t ransformat ion if we use the technique descr ibed by J E F F E R S (1959, p. 75). 187. Fig.28. Mean epi c o t y l length of the provenances plotted against t h e i r variance, on a double logarithmic paper. 400 4-300 200-C O u > ioo4— 504-• . • . r — I — [ - H - I I 1 I 1 H \-50 100 200 Epicotyl l e n g t h — log.mm. T A B L E X X V V A R I A N C E C O M P O N E N T S O F S O M E I M P O R T A N T S O U R C E S  O F V A R I A T I O N A N D T H E I R S T A N D A R D E R R O R F O R M A X I M U M E P I C O T Y L L E N G T H S o u r c e s of the E f f e c t N u m b e r of the R e g i o n P r o v e n a n c e s T r e e s E r r o r 1 4. 0 t 5.5 30. 0 t 8.5 8 6 . 9 t 6.4 2 23.2 t 21.8 31. 1 - 13. 9 102.8 - 11.2 3 95. 0 t 59.6 23.8 ± 10.0 100.0 t 9.2 4 8. 7 t 13.4 49.9 - 18.8 70. 0 t 7.4 5 ' 63. 0 t 25. 5 38. 2 ± 6.9 66.0 t 3.3 3.4.2 T o t a l height a f t e r the second g r o w i n g s eason ( H E T ) H E T was m e a s u r e d i n 1972 on the b a s i s of a r a n d o m s a m p l e of f i v e s e e d l i n g s p e r r e p l i c a t i o n and per progeny. In o r d e r to study the r e l a t i o n s h i p s between H E T and E P L , the aver a g e m a x i m u m height ( H M X ) and the a v e r a g e m i n i m u m height (HMN), f o r each provenance, have been c a l c u l a t e d by ta k i n g the g r e a t e s t and the s m a l l e s t s e e d l i n g of each p r o g e n y and s u m m i n g up o v e r a l l the r e p l i c a t i o n s and p r o g e n i e s f o r each p r o v e n a n c e . The s t a n d a r d d e v i a t i o n (SD), the s t a n d a r d e r r o r (SE) and the c o e f f i c i e n t of v a r i a t i o n (CV) of H E T have a l s o been c a l c u l a t e d f o r e a c h p r o v e n a n c e . The c o r r e l a t i o n m a t r i x ( T a b l e X X I V ) shows that the c o r r e -l a t i o n s between HMX, HMN, E P L and H E T a r e v e r y high as w e l l as 189, T A B L E X X V I C O R R E L A T I O N M A T R I X B E T W E E N T H E G E O G R A P H I C A L C O O R D I N A T E S O F T H E P R O V E N A N C E S A N D H E T , SD, SE A N D C T V . Lat Long A l t H E T C . V . SD SE L a t 1.00 Long 0.78 1.00 r . 0 5 = 0 ' 3 2 A l t 0.26 -0 .12 1.00 H E T -0 .78 -0 .49 -0 .52 1.00 C . V . 0.59 0.39 0.30 - 0 . 7 4 1.00 SD -0 .16 -0 .009 - 0 . 3 3 0.25 0.45 1.00 SE -0 .41 -0 .24 -0 .26 0.34 00.100 0.62 1.00 highly s ignif icant (r = 0 .88; 0 .90; 0 .91 , e t c . ) . Consequently, the con-c lusions concerning E P L should be equally valuable for H E T , at least as far as the c o r r e l a t i o n analyses are concerned. A c o r r e l a t i o n m a t r i x has been ca lculated between the geo-graphica l coordinates and H E T , C . V . , SD and SE (Table X X V I ) . Th is m a t r i x shows that C . V . increases with latitude (r = 0.59), but that SE decreases wi th la t i tude. T h i s would indicate that the populations become m o r e homogeneous with lat i tude. H E T is negatively c o r r e l a t e d wi th C . V . (r = - 0. 74). Seventy-three percent of the total v a r i a t i o n i n H E T is ex-pla ined by latitude and al t i tude. Longitude does not contribute s igni f i cant ly to the total v a r i a t i o n in H E T . Sixty-one percent is explained by latitude alone. 10+ 49° 5 0 0 510 5 2 o 5 3 o 540 5 5 o 5 6 o 5 7 o Latitude - 0 North F i g . 2 9 . Relationship between t o t a l height after the second growing season and latitude of place of o r i g i n of the provenances. 191. E i g h t y - n i n e p e r c e n t of the v a r i a t i o n i s e x p l a i n e d by bud set 72, E P L and bud b u r s t . E i g h t y - t w o , point four p e r c e n t by E P L alone. The p l o t t i n g of H E T on l a t i t u d e shows that i f the r e l a t i o n -s h i p i s l i n e a r , -'..ite l i s ^ t r l a l i s o s i v e r y c o m p l e x x i f i the p l a c e of o r i g i n of the p r o v e n a n c e s i s taken into account (see F i g . 29). The p r o v e n a n c e s of V a n c o u v e r I s l a n d can be s p l i t t e d into two sub-groups: the west coast and the east coast groups of the I s l a n d . On t h i s I s l a n d , H E T s h a r p l y d e c r e a s e s w i t h l a t i t u d e . A l t i t u d e cannot be r e s p o n s i b l e f o r thi s b e h a v i o u r , b e c a u s e the a l t i t u d e of the p r o v e n a n c e s r e m a i n s constant. The p r o v e n a n c e s of the Queen C h a r l o t t e I s l a n d s s e e m to behave d i f f e r e n t l y f r o m the other groups of p r o v e n a n c e s : H E T i n c r e a s e s w i t h l a t i t u d e . A l t i t u d e cannot be the f a c t o r r e s p o n s i b l e f o r t h i s odd be-h a v i o u r . The A l a s k a and the Skeena R i v e r w a t e r s h e d p r o v e n a n c e s do not f o r m two d i s t i n c t groups, but some d i v e r g e n c e s a r e n o t i c e a b l e . H E T a g a i n d e c r e a s e s s h a r p l y w i t h l a t i t u d e . The a n a l y s e s of v a r i a n c e , the components of v a r i a n c e and t h e i r s t a n d a r d e r r o r , and the D UNCAN's tests of the means of the p r o v e n -ances have been c a l c u l a t e d f o r e a c h r e g i o n s e p a r a t e l y . In t h i s case, o n l y the p r o v e n a n c e s r e p r e s e n t e d by at l e a s t f i v e f a m i l i e s and the f a m i l i e s w i t h t h r e e r e p l i c a t i o n s w e r e c o n s i d e r e d . It i s i m p o r t a n t to r e m i n d the r e a d e r that the m o d e l and the f o r m u l a s f o r the v a r i a n c e components and t h e i r e r r o r a r e quite d i f f e r e n t 19-2 . f r o m those u s e d to a n a l y z e E P L because i t was b e l i e v e d that t h e r e was no need to use a c o m p l i c a t e d m o d e l to a n a l y z e data as no i n t e r a c t i o n -not i m p o r t a n t at l e a s t - was noted and no exact f o r m u l a f o r the e r r o r of the components e x i s t e d f o r t h i s c r o s s e d m o d e l . Thus, we u s e d a three l e v e l n e s t e d m o d e l w h i c h has been w e l l w o r k e d out by S E A R L E and h i s students ( S E A R L E , 1971). A g a i n , to see i f some t r a n s f o r m a t i o n of the o r i g i n a l height m e a s u r e m e n t s was n e c e s s a r y , the m e a n H E T of the p r o v e n a n c e s have been p l o t t e d a g a i n s t the c o r r e s p o n d i n g v a r i a n c e s , i n a double l o g a r i t h -m i c graph. The square r o o t t r a n s f o r m a t i o n w o u l d be the c l o s e s t p o s s i b l e t r a n s f o r m a t i o n - u s i n g J E F F E R S ' method ( J E F F E R S , 1959), but again, the r e l a t i o n s h i p i s too l o o s e to w a r r a n t any u s e f u l r e s u l t s ( F i g . 30). The f o l l o w i n g t a b l e s s u m m a r i z e the a n a l y s e s done: R e g i o n 1 = p r o v e n a n c e s 2, 3, 4 and 29 A n a l y s i s of v a r i a n c e S o u r c e s of V a r i a t i o n D.F. S. S. M.S. F P r o v e n a n c e s 3 129,681.4 4,322.713 8.25 *** T r e e wn p r o v . 43 225,185.8 5,236.879 1.19 NS Rep wn t r e e wn p r o v . 94 410,880.1 4,371.063 1.75 *'** E r r o r 564 1, 406, 640 2, 494. 043  T O T A L 704  DUNCAN's t e s t : ( c r t = 0 . 0 5 J . P r o v . No. ' 4 2 29 3" H E T = 20!3. 3 206'.3 21 z[ 5 234.'4 m m Components of v a r i a n c e : see T a b l e X X V I I . Fig.30. Mean t o t a l height of the provenances, plotted against t h e i r variances, on a double logarithmic paper. 1 1 — 2 3 Total height — log.mm. 1 9 4 . R e g i o n 2 = p r o v e n a n c e s 30, 32, 34, 41 A n a l y s i s of v a r i a n c e S o u r c e s of V a r i a t i o n D.F. S. S. M.S.' F P r o v e n a n c e s 3 249,628.2 82, 209.328 14.91 T r e e wn p r o v . 30 167, 355. 2 5,578.504 1. 17 NS R e p wn tr e e wn p r o v . 68 321,681.2 4,730.605 2.47 E r r o r 408 778,323.2 1,907.655 T O T A L 509 DUNCAN's test : (o< = 0 .05). P r o v . No. 30 41 32 34 H E T 220. 7 215.5 198."9 164. 6 m m Components of v a r i a n c e : : see T a b l e X X V I I . R e g i o n 3 = p r o v e n a n c e s 36, 37, 39, 40 A n a l y s i s of v a r i a n c e S o u r c e s of V a r i a t i o n D. F. S.S. M.S. F P r o v e n a n c e s 3 236, 654.9 78,884.94 11.29*** T r e e wn p r o v . 48 335, 373. 5 6, 986.945 0. 94 NS R e p wn t r e e wn p r o v . 104 772,471. 5 7,427.609 >!= * !l< 3. 19 E r r o r 624 1, 450,453 2, 324.444 T O T A L 779 D U N C A N's test : ( c< = 0 . 05). P r o v . No. 40 37 3? 36 H E T 2.(55. 7 200; 1 187:2 158.'5 m m Components df v a r i a n c e : see T a b l e X X V I I . Region 4 = provenances 23, 24, 25, 26, 28 195 Sources of V a r i a t i o n D . F . S .S . M . S. F Provenances 4 78, 252.9 19.563.22 * 3.21 T r e e wn p r o v . 45 273, 915. 2 6 ,087.004 1. 23 NS Rep wn tree wn p r o v . 100 493230. 7 4 ,932 .30? 2. 26 E r r o r 600 1,306,665 2 ,177.775 T O T A L 749 D U N C A N ' s test: (c* = 0.05). P r o v . N o . H E T 25 24 23 28 26 16b. 7 1681.6 152.0 147.4 144] 5 Components of v a r i a n c e : see Table X X V I I . Region 5 = provenances 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 18, 19 A n a l y s i s of var iance Sources of V a r i a t i o n D . F . S .S . M . S . F Provenances 11 T r e e wn p r o v . 118 Rep wn tree wn p r o v . 260 E r r o r 1560 835,438.9 75,948.94 11.83 * >|< * 757,167.6 6 ,416.672 1.20 NS 1,3 86 , 7 0 4 5 3 33.476 2 . 7 1 * * * 3,068,979 1,967.294 T O T A L 1949 D U N C A N ' s test: ( © < = 0 . 0 5 ) . 1.5 10, 19. 8 14 13 11 18 —fc 1 1 fc-P r o v . N o . — 4 - -—i 4 h H E T = 188.5.183.1 1.8.1.9 18.0.0 171.5 170.2 15.0.7 150.7 142,0 133.8 131.9 128.2 Components of v a r i a n c e : see Table X X V I I . 196. - T A B L E X X V I I S E C O N D Y E A R H E I G H T V A R I A N C E C O M P O N E N T S  A N D S T A N D A R D E R R O R Sources of the Ef fec t  Region Provenances T r e e s Repeti t ions E r r o r s 1 2 2 4 . 6 ± 218.0 57. 7 t 86 .5 375.4 ± 130.9 2, 494.0 ± 148. 5 2 639.1 ± 587.9 56. 5 t H O . 2 564.6 ± 164.6 1, 907. 7 ± 133. 6 3 370.2 t 105.2 - 2 9 . 4 * 117.3 1, 020. 6 ± 207. 7 2, 324. 4 ± 1 31. 6 4 91.6 t 96.0 77.0 ± 97.4 550.9 ± 141.8 2, 177. 8 ± 125. 7 5 430. 0 * 204.6 72.2 t 63.8 673. 2 t 94.6 1 ,967.3 ± 70.4 F o r both E P L and H E T , the components of var iance dif fer a c c o r d i n g to the r e g i o n . E P L i s c h a r a c t e r i z e d by s m a l l e r components of v a r i a n c e , i n absolute value, as i t should be expected, than those for H E T . . The components of var iance est imated for H E T indicate that, genera l ly , the genetic var iance between provenances i s l a r g e r than the tree to tree within provenance genetic v a r i a t i o n , while i n the case of E P L , the wi th in provenance v a r i a t i o n can be l a r g e r than the between v a r i a t i o n . The s o - c a l l e d e r r o r var iance i s the larges t component. T h i s large component might suggest that the wi th in progeny group genetic v a r i a t i o n is quite l a r g e , attesting a genetic p l a s t i c i t y in harmony wi th the idea that Si tka spruce might be an " o p p o r t u n i s t i c " species (see P a r t I, Chapter 3). However , i n the absence of data on the frequency of se l f -pol l inat ion and the inbreeding processes taking place i n Si tka spruce , no 197 definite in terpreta t ion is poss ib le (personal communicat ion with D r . R . P E T E R S O N ) . 1 9 8 Photo 5. Abnormal s e e d l i n g w i t h s t u n t e d growth and s h o r t and t h i c k n e e d l e s . P r o b a b l y a t r i s o m i c i n d i v i d u a l . Photo by the a u t h o r . 199 C H A P T E R 4 T H E H E R I T A B I L I T Y P R O B L E M : ITS A P P L I C A T I O N T O O N E C H A R A C T E R I S T I C : T O T A L G R O W T H A F T E R T W O G R O W I N G S E A S O N S C l a s s i c a l l y , the nested s tructure of the height data: r e p l i -cations within progeny/progenies wi thin provenances/provenances is in terpre ted as f o l l o w s : the single tree progenies come f r o m open p o l -l inated mother trees and can be a s s i m i l a t e d to f a m i l i e s of half s ibs . T h e r e f o r e , the var iance between tree progenies wi th in provenances i s equivalent to the covariance among half sibs (after F A L C O N E R , 1964). But : (1) <j- 2 = cov = 1/4 V f 1/16 V f 1/64 V T r e e half sibs A A A A A with = additive genetic var iance = additive by additive interact ion var iance V = additive by additive by additive i n t e r -A A A . act ion var iance (after B E C K E R , 1967). 2 . G e n e r a l l y , 6" i s taken to be equal to 1/4 of V . But T r e e A assuming that this last f o r m u l a is true, by neglect ing the non-addit ive gene effects, gives an overes t imat ion of the additive genetic v a r i a n c e . N A M K O O N G (1966) has d i scussed different sources of b ias i n us ing open pol l inated seed for es t imat ing the additive genetic v a r i a n c e : dominance effects, relatedness of the parent trees (inbreeding), r e s t r i c t e d number ZOp: of pol len parents , a l l are factors which might lead to an overes t imat ion of the true additive genetic v a r i a n c e . T h e r e f o r e , the es t imat ion : (2) v = 4 . <r 2 A tree can always be accepted as an upper l i m i t of the true v a r i a n c e . A n es t imat ion of the h e r i t a b i l i t y i s given by: ^-2 (3) l / = t r e e total 2 with <y = sum of the components of var iance es t imated; it i s the total total phenotypic var iance (after F A L C O N E R , 1964 and B E C K E R , 1967). The h e r i t a b i l i t y defined by the f o r m u l a (3) i s the s o - c a l l e d n a r r o w sense h e r i t a b i l i t y , on an i n d i v i d u a l b a s i s ; i t i s the parameter used in the formulas to est imate possible genetic gains under different se lect ion schemes. T h i s h e r i t a b i l i t y should not be confused with the h e r i t a b i l i t y of the f a m i l y means as defined by W R I G H T (1963, f o r m u l a 61), (WRIGHT, 1963). A useful approximat ion of the standard e r r o r of the h e r i t a -b i l i t y i s given by: ? 4 • . / var <T (4) S . E . (h ) = V t r e e total (According to B E C K E R , 1967) The n a r r o w sense h e r i t a b i l i t y has been es t imated only for the second year height growth which takes into account the natura l v a r i a -201 T A B L E X X V I I I N A R R O W S E N S E H E R I T A B I L I T Y F O R T O T A L H E I G H T A F T E R T H E S E C O N D G R O W I N G S E A S O N : W I T H T H E I R E R R O R - B Y R E G I O N Region H e r i t a b i l i t y Standard E r r o r 1 0.07 0.11 2 0.07 0.14 3 negl igible 4 0.11 0.13 5 0.09 0.08 b i l i t y encountered in n u r s e r y prac t i ces with Sitka spruce . The h e r i t a b i l i t y - n a r r o w sense - has been est imated for each region separate ly . Table X X V I I I indicates their values and their e r r o r . The formulas (2), (3) and (4) were appl ied . The study of H E T was based on a random sample of five ( 1 - 1 ) seedlings per r e p l i c a t i o n and per f a m i l y . The n a r r o w sense h e r i t a b i l i t y calculated is genera l ly less than 10% and is affected by a large e r r o r . However , four regions give consistent est imations for h ^ . That the n a r r o w sense h e r i t a b i l i t y , i . e . , the additive gene-tic var iance could be s m a l l for growth trai ts is conf i rmed by the r e s u l t s of the d i a l l e l c ross p e r f o r m e d on Sitka spruce a l ready mentioned ( S A M U E L , et a l . 1972). T h e r e f o r e , f a m i l y se lect ion would be appropriate ( F A L -C O N E R , 1964). The general combining ab i l i ty seems to be low, therefore speci f ic combining ab i l i ty should be used through cros sing between outstand-ing individuals and populat ions. 202 C H A P T E R 5 R E L A T I O N S H I P S B E T W E E N S E E D , C O N E A N D S E E D L I N G T R A I T S S T U D I E D It i s interest ing to relate the four seed tra i ts studied and cone length to the seedling charac ters m e a s u r e d in the n u r s e r y in o r d e r to know if the f o r m e r parenta l c h a r a c t e r i s t i c s could be used to accura te ly predic t some juvenile charac ters of the progenies growing in the n u r s e r y . If this was the case, then some of these parent charac ters could be used in an " e a r l y test" to predic t the behaviour of these progenies . A s imple c o r r e l a t i o n m a t r i x has been ca lculated between a l l the seed, cone and seedling c h a r a c t e r s , as w e l l as with lat i tude, longitude and altitude of the place of o r i g i n of the provenances, (see T A B L E X X I X ) . The legend is the same as the legends of Tables X I and X V . A look at Table X X I X shows that the seed and cone t ra i t s are general ly not s igni f i cant ly re la ted wi th the seedling charac ters studied. However , wing width (b) is p o s i t i v e l y c o r r e l a t e d wi th bud set m e a s u r e d in 1971 (r = 0. 32 and 0. 37), needle colour (r = 0. 39); seed length (c) is p o s i t i v e l y c o r r e l a t e d with epicotyl length and total height (r = 0. 36); seed width (d) is negatively c o r r e l a t e d (r = - 0 . 40) with bud burs t , pos i t ive ly with needle colour (r = 0.48); cone length is negatively c o r r e l a t e d wi th bud set 72 (r = - 0 . 38) and p o s i t i v e l y c o r r e l a t e d wi th epicotyl length and total height (r - 0. 69 and 0. 67). T A B L E X X I X C O R R E L A T I O N M A T R I X B E T W E E N S E E D , G O N E A N D S E E D L I N G T R A I T S S T U D I E D E P L H E T a b c L A T L O N G A L T B S T 1 B S T 2 B B T C O L B S T 7 2 L A T 1. 00 L O N G 0. 78 1. 00 A L T 0. 26 -0. 12 1. 00 r 0 . 05 BS T 1 0. 89 0. 53 0. 39 1. 00 B S T 2 0.92 0. 63 0. 32 0.96 1. 00 B B T -0. 14 -0. 51 0.42 0. 13 0. 07 1. 00 C O L 0. 64 0. 72 -0.039 0. 58 0. 62 -0. 37 1. 00 BST 72 0. 88 0. 59 0.47 0.91 0. 92 0. 12 0.61 1. 00 E P L -0. 71 -0. 34 -0. 56 -0. 74 -0. 70 -0. 13 -0..44 -0. 75 H E T -0. 78 -0.49 -0. 52 -0. 74 -0. 74 0. 037 -0. 50 -0. 80 a -0. 17 -0.099 0.0054 0. 30 0. 27 0. 24 0. 053 0. 22 b 0. 39 0. 45 -0. 15 0. 32 0. 37 -0. 21 0. 39 0. 28 c -0. 19 -0. 10 -0. 28 -0. 13 -0. 12 -0. 11 -0.052 -0. 21 d 0. 34 0.49 -0.40 0. 25 0. 30 -0.40 0.48 0. 22 cone -0. 30 -0.033 -0. 57 -0. 28 -0. 28 -0. 24 0. 10 -0. 38 cone = 0. 32 1.00 0.91 1.00 -o .093 -o.017 . i ; o o 0.0041 -0.010 0.57 1.00 0.36 0.36 0.62 0.62 1.00 0.066 -0.075 0.30 0.75 0.451 1.00 0.69 0.67 0.32 0.48 0.66 0.49 1. 00 ,g 204 The mul t ip le r e g r e s s i o n technique used to study the other c o r r e l a t i o n m a t r i c e s has been appl ied to test how w e l l the seed and cone t ra i t s predict the seedling t ra i ts m e a s u r e d . F i f t y , point two percent of the v a r i a t i o n i n bud set 1 i s explained by the five seed and cone t r a i t s , 37. 7% by wing width and cone length alone. F o r bud set 2, these f igures become 53.3% and 43.0%. F i f t y - f o u r , point nine percent of bud set 72 i s explained by the five t ra i t s studied, 44. 3% by wing width and cone length alone. T h i r t y - f i v e , point zero percent of the var ia t ion i n bud burst is explained by the five t ra i t s studied, 31. 5% by wing length and weed width alone. T h i r t y - s e v e n , point nine percent of needle colour i s ex-pla ined by the five t ra i t s studied; 33.3% by seed length and seed width alone. S ix ty-e ight , point four percent of epicotyl length is ex-pla ined by the five t ra i ts studied; 47. 7% by cone length alone. F o r total height, these f igures become 71.2% and 44.5%. However , the re la t ionships establ ished between lat i tude, longitude and altitude of the place of o r i g i n of the provenances and the seedling t ra i ts studied (see Chapter 3, P a r t II) show that a higher percent -age of the v a r i a t i o n i n these t ra i t s i s explained by the place of o r i g i n of the provenances . F o r example , up to 86. 2% of the v a r i a t i o n i n bud set i s explained by longitude and latitude of the place of o r i g i n . S i x t y - f i v e percent of epicotyl length v a r i a t i o n is explained by latitude and altitude 205 alone. There fore , the seed and cone t ra i t s studied are not as eff ic ient as the geographical coordinates of the provenances as predic tor of the progeny c h a r a c t e r i s t i c s . They are also more time consuming to obtain and should be used only when the o r i g i n of the provenance is unknown and as gross p r e d i c t o r s . It would be interes t ing to use the canonical corre la t ions to study the re la t ionship between the two sets of v a r i a t e s . 206. C H A P T E R 6 G E N E T I C V A R I A T I O N IN F O L I A R M A C R O A N D M I C R O N U T R I E N T S O F T E N S E L E C T E D O N E Y E A R O L D S I T K A S P R U C E P R O V E N A N C E S 6. 1 Introduction So many t ra i t s present genetic di f ferences between p r o v e n -ances, that it is tempting to t r y f inding dif ferences i n m i n e r a l content that could be assoc ia ted w i t h and perhaps explain these d i f ferences . However delicate i s the interpretat ion of the resul t s of plant ana lys i s ( G O U N Y , 1956), m o r e p a r t i c u l a r l y f o l i a r a n a l y s i s , the recent years have seen studies of plant t issue m i n e r a l composi t ion in r e l a t i o n with seed sources ( M E R G E N and W O R R A L L , 1965; S T E I N B E C K , 1966; V A N D E N D R I E S S C H E , 1969, 1973). The work of A D D I S O N (1966) has a l ready been mentioned i n Chapter 3, P a r t I. There i s a lso some need to know the average m i n e r a l c o m -posi t ion of the most important tree species i f f e r t i l i z e r appl icat ions must be done i n the n u r s e r y . T h e r e f o r e , f o l i a r ana lys i s of ten selected provenances of Si tka spruce were p e r f o r m e d i n order to detect any genet ica l ly based v a r i a t i o n i n m i n e r a l content w h i c h could be re la ted to their geographic o r i g i n or which could find some explanations and also to provide some data on the m i n e r a l composi t ion of the foliage of S i tka spruce seedlings under opt imum n u r s e r y condit ions. 207 6. 2 M a t e r i a l and methods The m a t e r i a l u s e d comes f r o m 1-0 s e e d l i n g s w h i c h w ere g r o w i n g i n the s t y r o - f o a m c o n t a i n e r s i n 1971. D u r i n g the f i r s t g r o w i n g season, the s e e d l i n g s w e re f e r t i l i z e d r e g u l a r l y . The f e r t i l i z e r s u s e d w e r e d i f f e r e n t h i g h l y s o l u b l e c o m m e r -c i a l p r e p a r a t i o n s ( P l a n t - P r o d : P l a n t - P r o d u c t s Co. L t d . P o r t C r e d i t , Ont. ), f o r m u l a t e d as f o l l o w s : 28 - 14 - 14 1 5 - 1 5 - 3 0 21 - 0 - 0 F o r i n s t a n c e , the guaranteed m i n i m u m c o m p o s i t i o n a d v e r -t i s e d was, f o r the f e r t i l i z e r 28 - 14 - 14. T o t a l n i t r o g e n : 2 8 % A v a i l a b l e p h o s p h o r i c a c i d : 1 4 % Sol u b l e potash: 1 4 % C h e l a t e d t r a c e e l e m e n t s : Manganese: 0.045% I r o n : 0.03 % Copper: 0. 0 0 2 5 % Z i n c : 0. 014 % M o l y b d e n u m : 0.001 % B o r o n : 0.04 % The f e r t i l i z e r s w e r e fed to the s e e d l i n g s t h r o u g h the i r r i -g a t i o n s y s t e m . The guidi n g p r i n c i p l e s w e r e to supply the s e e d l i n g s w i t h h i g h l e v e l s of n i t r o g e n t h r o u g h m i d - s u m m e r and to r e d u c e n i t r o g e n supply and i n c r e a s e p o t a s s i u m t o w a r d s the end of the g r o w i n g season. To r e d u c e h i g h s a l t a c c u m u l a t i o n , the f e r t i l i z e r s 28 - 14 - 14 and 15 - 15 -30 w e r e a l t e r n a t e d w i t h a m m o n i u m s u l f a t e (21 - 0 -.0). L e a c h i n g w i t h 208 water avoided b u i l d - u p of f e r t i l i z e r to any toxic l e v e l i n the plug s o i l (after A R N O T T , 1971). A complete r e c o r d of the f e r t i l i z e r applicat ions w h i c h began in June, 1971 shows that the applicat ions were made weekly or b i - w e e k l y and v a r i e d f r o m 1.03 gr to 9.44 gr of a given f e r t i l i z e r per s t y r o - f o a m container . The last appl icat ion was made on the 30th of N o v e m b e r , 1971 (15 - 15 - 30). M o s t applicat ions were of the f e r t i l i z e r 28 - 14 - 14. To avoid any growth art i fact effects as evidenced by the studies of M E R G E N and W O R R A L L (1965) and other authors ( G O U N Y , op. c i t . ), the ten selected provenances were sampled on the 23rd of F e b r u a r y , 1972, s t i l l quiescent. The provenances were chosen a c c o r d -ing to their place of o r i g i n along a la t i tudinal gradient , to study a major source of v a r i a t i o n . Three b locks A , B and D were sampled . Healthy seedlings were c l ipped, i m m e d i a t e l y put in p las t i c bags and freeze d r i e d the same day in a " T h e r m o v a c Industries C o r p o r a -tion F r e e z e D r i e r " . The condenser temperature was at - 6 4 ° F , the shelves at +• 7 0 ° F and the vacuum reached 80-90 m i c r o n s of M e r c u r y . The freeze dry ing process lasted five days and the freeze d r i e d seedlings were put in sealed p las t i c bags and kept i n deep f r e e z e unt i l the analyses were done. The dry needles were separated f r o m the stems and m i l l e d i n a W I L E Y m i l l unt i l the powder obtained passed through a 60 m e s h s ieve . Then digest ion in p e r c h l o r i c a c i d m i x t u r e was p e r f o r m e d . 20-9 I r o n c o n t a m i n a t i o n by the m i l l was c o n s i d e r e d n e g l i g i b l e and i n any case, random. P h o s p h o r u s was d e t e r m i n e d by the m o l y b d e n u m blue method of D I C K M A N and B R A Y . P o t a s s i u m , C a l c i u m , M a g n e s i u m , Manganese, I r o n and Z i n c w e r e d e t e r m i n e d u s i n g an a t o m i c a b s o r p t i o n f l a m e - e m i s s i o n J A R R E L L - A S H s p e c t r o - p h o t o m e t e r . N i t r o g e n was d e t e r m i n e d by the w e l l known s e m i - m i c r o K J E L D A H L method. The methods u s e d a r e a l l o u t l i n e d by C H A P M A N and P R A T T (1961). 6. 3 R e s u l t s The r e s u l t s of the a n a l y s e s a r e shown i n T a b l e s X X X and X X X I . T h e y a r e quite r e m a r k a b l e . T h e r e a r e no d i f f e r e n c e s i n m a c r o and m i c r o n u t r i e n t s contents w h i c h c o u l d be a t t r i b u t a b l e to the g e o g r a p h i c o r i g i n of the p r o v e n a n c e s . However, the e p i c o t y l l e n g t h of the ten p r o v e n -ances v a r i e s between 66. 5 m m (provenance 27) and 89.6 m m (provenance 40). The a n a l y s e s of v a r i a n c e show that t here a r e no s i g n i f i c a n t d i f f e r e n c e s between the p r o v e n a n c e s f o r a l l the e l e m e n t s a n a l y z e d except f o r P o t a s s i u m . The anova m o d e l was the r a n d o m i z e d c o m p l e t e b l o c k d e s i g n w i t h three b l o c k s . The DUNCAN's test f or P o t a s s i u m i s as f o l l o w s : 2 1.0 8, 100 ppm 6, 500 ppm P r o v e n a n c e N u m b e r 3 32 40 15 25 4 37 27 28 23 —I 1 1 1 1 1 1 1 i 1-The i n t e r p r e t a t i o n of these r e s u l t s i s d i f f i c u l t , but no m o r e than i f we had found genetic v a r i a t i o n i n m i n e r a l contents of the n e e d l e s of S i t k a s p r u c e p r o v e n a n c e s . Thus, t h e r e a r e no d i f f e r e n c e s i n a b s o r p -t i o n and sto r a g e of the n u t r i e n t s s t u d i e d i n the f o l i a g e of one y e a r o l d S i t k a s p r u c e s e e d l i n g s o r th e r e a r e d i f f e r e n c e s w h i c h appear at some t i m e of the y e a r ; but these d i f f e r e n c e s d i s a p p e a r b e c a u s e t h e r e a r e c o u n t e r b a l a n c i n g d i f f e r e n c e s i n r a t e s of l e a c h i n g o r i n r a t e s of i n t e r n a l t r a n s l o c a t i o n of the same element. A n o t h e r hy p o t h e s i s would be that t h e r e a r e d i f f e r e n t r a t e s of change i n c a r b o h y d r a t e content as, a c c o r d i n g to R E U T H E R and S M I T H ( i n C H I L D E R S , ed. 1954), a change i n concen-t r a t i o n of a m i n e r a l n u t r i e n t i n l e a v e s of c i t r u s s p e c i e s m a y be c a u s e d by an a c c u m u l a t i o n o r d e p l e t i o n of d r y m a t t e r . T h i s l a t t e r h y p o t h e s i s i s d i f f i c u l t to v i s u a l i z e i n o ur ca s e . N u t r i e n t a b s o r p t i o n , t r a n s l o c a t i o n and l o s s i n f o r e s t t r e e s a r e not w e l l u n d e r s t o o d (VOIGT, 1968). D i s c u s s i o n s a r e g e n e r a l l y c a r r i e d out at a g e n e r a l ! l e v e l that g i v e s l i t t l e o r no i n d i c a t i o n of the r e a s o n s for the d i f f e r e n c e s - i f any - i n the a b i l i t y of pl a n t s to a c c u m u l a t e n u t r i e n t s . G O U N Y (op. c i t . ) has r e v i e w e d the d i f f i c u l t i e s i n i n t e r -p r e t i n g the r e s u l t s of f o l i a r and other plant t i s s u e a n a l y s i s . However, T A B L E X X X F O L I A G E A N A L Y S I S O F 10 O N E Y E A R O L D D O R M A N T S I T K A S P R U C E P R O V E N A N C E S 100 P P M D R Y N E E D L E S (*)" F e M n K M g C a N r P R O V A B D A B D A B D A B D A B D 3 1.4 3.5 1.4 1. 5 1. 5 1 . 5 78 79 87 .58 .58 .62 16 15 18 4 1.4 1.4 1.4 1.6 1.4 1.3 71 70 74 .62 .58 .54 14 17 17 15 1.4 2.2 1.4 1.5 1.7 1. 5 71 78 74 .62 .58 .58 10 14 18 23 1.4 3.3 1.4 1.5 1.4 1.5 62 62 70 .66 .62 .54 17 13 18 25 1.4 5.1 1.4 1.6 1.5 1.5 72 80 66 .62 . 71 .50 17 16 16 27 1.4 3.3 1.4 1.5 1.5 1.5 62 75 70 .66 .62 .58 15 17 18 28 3.3 1.4 1.4 1.7 1.5 1.5 76 69 62 .75 .66 .58 14 18 15 32 1.4 2.2 1.4 1.6 1.5 1.6 71 78 81 .62 .56 . 54 14 16 15 37 1.4 1.4 1.4 1.5 1.5 1. 5 72 76 66 .58 .58 .62 15 17 18 40 1.4 1.4 2.6 1.6 1.5 1.5 74 76 76 .62 .56 . 58 15 17 15 Z N : content l e s s than 20 P P M (*) The f i g u r e s f or the M g content must be m u l t i p l i e d by 5, 000. T A B L E X X X I F O L I A G E A N A L Y S I S O F 10 O N E Y E A R O L D D O R M A N T S I T K A S P R U C E P R O V E N A N C E S % N E E D L E D R Y W E I G H T P N P R O V A B D A B D 3 . 53 .44 .49 1.24 1. 16 1.26 4 .45 .39 . 51 1.27 1. 12 1.46 15 .35 .46 . 57 1.26 1. 34 1. 38 23 .49 .42 .32 1. 08 1. 24 1.22 25 . 50 .35 1.24 1. 34 1.35 27 . 58 . 53 . 50 1.44 1.21 1.40 28 . 50 .39 .43 1.40 1. 36 1. 35 32 . 50 .42 . 55 1.18 1. 16 1. 32 37 .43 .46 .49 1. 33 1.25 1.24 40 .42 .47 . 58 1. 14 1. 32 1. 22 2 13 L E Y T O N (in I . H . R . O . , 1956) has found s ignif icant l inear corre la t ions between tree height and concentrations of N , P , K and ash i n needles of Japanese l a r c h . •Authors diverge as to the s ignif icance of m i n e r a l concentra-tion of plant t i s sue : t ime is important as w e l l as the p r e c i s e p h y s i o l o g i c a l stage of the plant ( G O U N Y , o_p. c i t . ) ; L A V E N D E R and C A R M IC H A E L (1966), W A R I N G and Y O U N G B E R G (1972), L E A F (1968), bel ieve that m u c h b i o l o g i c a l i n f o r m a t i o n is lost by sampl ing foliage during the dormant season or in the f a l l . The best t ime to charac ter ize the di f ferences would be while growth takes p lace . Fo l iage might not n e c e s s a r i l y re f lec t the total absorpt ion of the elements by the seedlings ( L E A F , op. c i t . ) . However , V A N D E N D R I E S S C H E (1969 b) does not think so on the bas is of his studies on m i n e r a l nut r i t ion of D o u g l a s - f i r and Sitka spruce . L A V E N D E R (1970) bel ieves that i f the m i n e r a l content of a given plant population could be shown to be c o r r e l a t e d with growth, this would be evidence of a def ic iency . Despite the d i f f i cu l t i es in in te rpre t ing the r e s u l t s of plant t issue a n a l y s i s , s e v e r a l authors have r e s e a r c h e d genet ical ly based dif ferences i n m i n e r a l composi t ion of plant t i s s u e s . M E R G E N and W O R R A L L (1965) havefbund^differences i n m i n e r a l content of seed sources of Jack pine: the dif ferences might be due to di f ferences i n ab i l i ty of u t i l i z i n g the m a t e r i a l s f r o m the s o i l m e d i u m and p a r t l y to di f ferences i n growth responses to different e n v i r o n -ments . The two authors do not mention the p r e c i s e development stage at 214 the t ime of harves t : 90 days o l d . N i t r o g e n content was c o r r e l a t e d wi th weight. S T E I N B E C K (1966) has found great dif ferences between provenances of Scots pine for N , P , N a , M g and B . However , i n t e r -actions were noted and no explanations have been proposed for the di f -ferences o b s e r v e d . V A N D E N D R I E S S C H E (1969) has studied the re la t ionships between the growth of D o u g l a s - f i r seedlings and leve ls of some s o i l and tissue nutr ients . Only 19% of the two year o ld shoot d r y weight v a r i a -b i l i t y was accounted for by the f i r s t year concentrations i n P , K , M g . The same author (1969^), has extensively studied the t issue nutrient con-centrations of Sitka spruce and D o u g l a s - f i r under different nutrient s o l u -t ions . G e n e r a l l y , needle concentrations p a r a l l e l e d stem and root concen-t ra t ions . Seed m i n e r a l composi t ion was i r r e l e v a n t . Di f ferences i n a l l nutr ients except P were found in four seed sources of D o u g l a s - f i r . The same author (1973) has further studied the dif ferences i n nutrient concen-trat ions of foliage of dif ferent provenances of D o u g l a s - f i r . Signif icant di f ferences were found, but different kinds of interact ions ex is ted . Deductions about n u r s e r y , site nutrient a v a i l a b i l i t y made f r o m f o l i a r analyses would only be applicable for a given provenance grown on that site because the dif ferences between provenances v a r i e d f r o m one n u r s e r y to another, as w e l l as the growth r e l a t i o n s h i p s . It i s important to note that a l l the studies mentioned were based on seedlings s t i l l growing or sampled at the beginning of the r e s t -215, ing p e r i o d (October to December ) . No p r e c i s e d e s c r i p t i o n of the p h y s i o l -o g i c a l stage has ever been attempted. Our resu l t s show that time is important and the di f ferences , i f they ever ex is ted , could disappear late in the cold season. The seed-l ings were sampled i n F e b r u a r y . They were s t i l l dormant . If the seed-l ings were i n a state of res t imposed by the environmenta l conditions or s t i l l i n a state of i r r e v e r s i b l e d o r m a n c y . i s not known. The m e c h a n i s m s of such u n i f o r m i z a t i o n of the m i n e r a l composi t ion of the foliage of S i tka spruce are d i f f i cu l t to v i s u a l i z e or to understand as to their poss ib le s i g -ni f icance in the p h y s i o l o g i c a l processes of the t ree . The resu l t s of some authors seem to point out that, even when nutrient supply is abundant, there is higher use of m i n e r a l s by the trees growing the fastest, as attested by their higher m i n e r a l content. Consequently, our data would show that the dif ferences in m i n e r a l c o m -posi t ions detected by the authors mentioned, would be due to higher metabol ic ac t iv i t i e s connected wi th higher growth r a t e s . It would be in teres t ing to study the m i n e r a l composi t ion of provenances of Sitka spruce under different m i n e r a l nutr i t ion s t resses to further test the hypothesis that S i tka spruce provenances might have d i f -ferent a b i l i t i e s to absorb the m i n e r a l e lements . The m i n e r a l contents of the needles of ten Si tka spruce p r o v e n -ances are wi th in the range of m i n e r a l composit ions of S i tka spruce foliage c o m p i l e d by V A N D E N D R I E S S C H E (1969). However , it seems that the concentrations obtained by us are f a i r l y high, i n c o m p a r i s o n with those shown by the latter author. 216 The r e c e n t a n a l y s e s of S i t k a s p r u c e s e e d l i n g s by B E N Z I A N and S M I T H (1973) a r e not d i r e c t l y c o m p a r a b l e as they a r e b a s e d on whole p l a n t s . P A R T IV S U M M A R Y A N D C O N C L U S I O N S 217 S U M M A R Y A N D C O N C L U S I O N S (1) In a f i r s t part , seed and cone morphology of 557 Sitka spruce trees represent ing 39 locations were studied on a single tree b a s i s . Ten cones per progeny were randomly selected and the length of each cone m e a s u r e d to the nearest m m . F i v e randomly selected seeds f r o m each tree were mounted on a spec ia l sheet, and seed length, •seed width, wing length and wing width were m e a s u r e d to the nearest 0. 01 m m . Nested analys is of var iance and D U N C A N ' s mul t ip le range tests for a l l the c h a r a c t e r i s t i c s studied have been p e r f o r m e d using f ive working subregions based on b iogeoc l imat i c data. No definite c l a s s i f i c a t i o n of the provenances was possible by using univariate anova p r o c e d u r e s . A s imple c o r r e l a t i o n m a t r i x has been ca lculated between a l l the t ra i t s studied and longitude, latitude and altitude of the place of o r i g i n of the provenances, us ing the provenance means . M u l t i p l e r e -g r e s s i o n analyses have been used for invest igat ing this c o r r e l a t i o n m a t r i x . The percentage of v a r i a t i o n accounted for by the geographical coordinates v a r i e s between 10.2% to 43.6%. Components of var iance calculated for the t ra i t s studied indicate that genera l ly the v a r i a b i l i t y i s m o s t l y confined to the between tree , wi thin provenance v a r i a b i l i t y . T h e r e f o r e , the t ra i t s studied are not l i k e l y to be good d i s c r i m i n a t o r y charac ters for dis t inguishing the provenances . Tentat ive ly , the t ra i t s with the best d i s c r i m i n a t i n g power were chosen as seed length, seed width and' cone length because they show a greater v a r i a b i l i t y between provenance s. 218 To see if the simultaneous considerat ion of the five reproduct ive t ra i ts studied could resu l t in a better c l a s s i f i c a t i o n of the Sitka spruce provenances, some mul t ivar ia te s ta t i s t i ca l techniques were t r i e d . P r i o r to any definite c l a s s i f i c a t i o n , the bas ic p r o b l e m of f inding the best c l a s s i f i c a t i o n method was attempted. (2) U s i n g the seed and cone trai ts studied, a c o m p a r i s o n of s e v e r a l m u l t i v a r i a t e s ta t i s t i ca l analyses which can be used for c l a s s i f i -cat ion purposes has been attempted. The s o - c a l l e d canonical a n a l y s i s , d i s c r i m i n a n t function analys is and p r i n c i p a l component analysis have been compared and applied for c l a s s i f y i n g the provenances . The sub-regions were analyzed separate ly . Dendrograms were also constructed and analyzed. Advantages and disadvantages of each mul t ivar ia te method have been d i s c u s s e d . It was found that the discriminant function a n a l y s i s , i ts associated genera l ized distances of M A H A L A N O B I S and dendrograms provided the most r a t i o n a l c l a s s i f i c a t i o n of the provenan-c e s . A c o m p a r i s o n of the techniques with the c l a s s i f i c a t i o n possible by a M A N O V A fol lowed by a m u l t i v a r i a t e mul t ip le c o m p a r i s o n of the centroids would be use fu l . Another approach to be t r i e d would be to apply the techniques studied to a " p l a s m o d e " for a def ini t ion, see C A T T E L L (1965), i . e . , a set object whose proper t ies and c lus ter ing are known, in order to compare the c lass i f i ca t ions made by these techniques. (3) In a second part , the genetic v a r i a b i l i t y of 545 Sitka spruce 219 single tree progenies was studied i n a n u r s e r y test during the 1971 and 1972 growing seasons. A total of 545 single tree progenies grouped into 38 p r o v e n -ances was sown in A p r i l , 1971, us ing a randomized complete block design with four repl i ca t ions and 24 seedlings per r e p l i c a t i o n or 96 seed-l ings per progeny. The seeds were placed i n the cavi t ies of s t y r o -b locks using the method developed by the P a c i f i c F o r e s t R e s e a r c h Centre in cooperation with the B . C . F o r e s t Service and they have been treated by the most recent n u r s e r y methods, in the new B . C . F . S. n u r s e r y at S u r r e y ( B . C . ) / G e r m i n a t i o n rate, bud set, length of the epicotyl and s u r v i v a l after the f i r s t growing season were assessed i n 1971. The seedlings were transplanted in p l a i n s o i l seedbeds i n M a y , 1972, to a distance of 6" to 6", each progeny being kept separate while respect ing the same s ta t i s t i ca l design as i n 1971. Bud burs t , bud set, colour of the needles and total height after the second growing season were assessed in 1972. There was a c l i n a l v a r i a t i o n i n bud burs t , bud set, colour of the needles and epicotyl length. Bud burst was negatively c o r r e l a t e d with longitude (r = - 0 . 50) and p o s i t i v e l y c o r r e l a t e d with altitude (r = 0.42). Bud set appeared under s t r i c t genetic contro l as indicated by the second est imat ion of this t ra i t , at the end of the second growing season (with lat i tude: r = 0.88). Latitude and altitude of the seed sources explained 65% of the total v a r i a t i o n i n epicotyl length. T o t a l height after the second growing season showed the same r e l a t i o n -ships as epicotyl length. 220 T h e r e f o r e , there i s a c l i n a l v a r i a t i o n , m o s t l y w i t h l a t i t u d e , f o r a l l the t r a i t s s t u d i e d as w e l l as a c o v a r i a n c e s t r u c t u r e between t r a i t s w h i c h m i g h t r e s u l t i n c o r r e l a t e d r e s p o n s e s when s e l e c t i o n f o r one c h a r a c t e r i s attempted. B u d set and bud b u r s t a r e i m p o r t a n t adaptative c h a r a c t e r i s t i c s w h i c h m i g h t be a d v e r s e l y a f f e c t e d when s e l e c t i o n f or height growth i s t r i e d . B u d set d e t e r m i n e s somehow the c e s s a t i o n of height g rowth as w e l l as the c o l d h a r d i n e s s of the S i t k a s p r u c e p r o v e n a n c e s . (4) The i m p o r t a n t q u e s t i o n of the p h y s i o l o g i c a l m e c h a n i s m c o n t r o l l i n g bud set and bud b u r s t i n S i t k a s p r u c e r e m a i n s to be s o l v e d as our data do not support d i r e c t l y the i d e a that bud set i s c o n t r o l l e d b y the d e c l i n i n g d a y - l e n g t h of late summer. B U R L E Y ' s growth chamber stu d i e s suggest that d e c l i n i n g p h o t o p e r i o d m i g h t hasten bud set t i n g i n S i t k a s p r u c e . The p r e s e n t study shows that, i n n a t u r a l c o n d i t i o n s , 86. 2 % of the v a r i a t i o n i n bud set can be e x p l a i n e d by longitude and l a t i -tude, 7 9 % by l a t i t u d e alone. F o r t y - f o u r p e r c e n t of bud b u r s t v a r i a t i o n i s e x p l a i n e d by l a t i t u d e and longitude, 2 6 % by longitude alone. O ur r e s e a r c h shows that there i s some evidence that the d i f f e r e n c e s i n bud f l u s h i n g a r e r e l a t e d to the l o c a l late f r o s t d i s t r i b u t i o n of the place of o r i g i n . The study of the growth r e s p o n s e s of S i t k a s p r u c e p r o v e n -ances to sh o r t e n e d and extended p h o t o p e r i o d s and under d i f f e r e n t t e m p e r a t u r e r e g i m e s should p r o v i d e some c l u e s as to the m e c h a n i s m s of bud b u r s t and bud set p r e v a l e n t i n thi s t r e e s p e c i e s . 221 (5) G e n e r a l equations f o r components of v a r i a n c e f o r un-b a l a n c e d data w e re o r i g i n a l l y c a l c u l a t e d f o r a n e s t e d - c r o s s e d m o d e l . Components of v a r i a n c e and t h e i r s t a n d a r d e r r o r w e r e c a l c u l a t e d f o r e p i c o t y l l e n g t h and t o t a l height a f t e r the second growing season. Depending on the s u b r e g i o n s , the genetic v a r i a n c e among p r o v e n a n c e s i s g e n e r a l l y l a r g e r than the t r e e to t r e e genetic v a r i a t i o n . (6) The n a r r o w sense h e r i t a b i l i t y , on an i n d i v i d u a l b a s i s , and i t s s t a n d a r d e r r o r , f o r t o t a l height a f t e r the second g r o w i n g season w e r e e s t i m a t e d on a s u b r e g i o n b a s i s . H e r i t a b i l i t y v a r i e d between 0.07 and 0.11. T h e s e r e s u l t s , as w e l l as those of S A M U E L , et a l . (1972), suggest that f a m i l y s e l e c t i o n would be a p p r o p r i a t e and that the s p e c i f i c c o m b i n i n g a b i l i t y s h o u l d be u s e d th r o u g h c r o s s i n g between outstanding i n d i v i d u a l s and p o p u l a t i o n s . (7) The c o r r e l a t i o n s between the seed, cone and s e e d l i n g t r a i t s w e r e studie d . Wing width was p o s i t i v e l y c o r r e l a t e d w i t h bud set (r = 0.32 and 0. 37), and needle c o l o u r . Seed length was p o s i t i v e l y c o r r e l a t e d w i t h e p i c o t y l l e n g t h and to t a l height (r = 0.36), cone length i s n e g a t i v e l y c o r r e l a t e d w i t h bud set 72 and p o s i t i v e l y w i t h e p i c o t y l length and t o t a l height (r = 0.69 and 0.67). However, the g e o g r a p h i c a l c o o r d i n a t e s of the p l a c e of o r i g i n of the p r o v e n a n c e s e x p l a i n m o r e of the v a r i a t i o n of the s e e d l i n g t r a i t s than the cone and seed c h a r a c t e r s w h i c h appear to be l e s s u s e f u l as " e a r l y t e s t c h a r a c t e r s " . The r e l a t i o n s h i p s e s t a b l i s h e d m i g h t i n d i c a t e some " m a t e r n a l " e f f e c t s . 222 (8) V a r i a t i o n i n fo l i a r m a c r o - and m i c r o - nutrients of ten provenances was studied but no geographical pattern of v a r i a t i o n detected in K , C a , M g , F e , M n , Z n , P and N needle contents. Only K showed some provenance to provenance v a r i a t i o n s . P o s s i b l e p h y s i o l o -g i c a l explanations for this absence of v a r i a t i o n are d i s c u s s e d . The 'data obtained i n this study would show that the dif ferences i n m i n e r a l composi t ion detected by some authors might be due to different m e t a -b o l i c ac t iv i t ies connected with different growth ra tes . It would be in teres t ing to study the m i n e r a l composi t ion of different S i tka spruce provenances under different m i n e r a l nutr i t ion s t resses to further test the hypothesis that Sitka spruce provenances might have different a b i l i t i e s to absorb the m i n e r a l e lements . (9) The phenotypic v a r i a t i o n of the parent trees of 545 S i tka spruce progenies has been studied on the bas is of f ive seed and cone t ra i t s i n order to begin c l a s s i f y i n g the natura l populations of this tree spec ies . Important adaptative c h a r a c t e r i s t i c s of the progenies growing i n a n u r s e r y near Vancouver (B. C . ) have also been studied as e a r l y test c h a r a c t e r s . The progenies have been outplanted in October , 1973 and spr ing 1974 on the Queen Charlot te Islands and on Vancouver Island by the B . C . F o r e s t Serv ice , to study their genetic v a r i a b i l i t y in different f i e l d environments . Genotype by environment interact ions and suscept i -b i l i t y to w e e v i l attack w i l l be studied. 223 B I B L I O G R A P H Y A D D I S O N , J . W . (1966). Nutr ient experiments with Sitka spruce . B a c h , s c i . F o r e s t r y thesis . F a c u l t y of F o r e s t r y . U . B . C . Vancouver , B . C . 1 v o l . , 98 pp. A L D H O U S , J . R . (1962). Provenances of Sitka spruce . A n account of the n u r s e r y stage of experiments sown i n 1958. F o r e s t r y c o m m i s s i o n . Report on F o r e s t r e s e a r c h . M a r c h 1961: 147-154. A L L E N , G . S . (I960). A method of dis t inguishing Coas ta l and Inter ior Douglas f i r seed. U . B . C . F a c u l t y of F o r e s t r y R e s . Note , no. 28. 3 pp. A N D E R S E N , H . E . (1955). C l i m a t e i n southeast A l a s k a in r e l a t i o n to tree growth. A l a s k a F o r e s t R e s e a r c h C e n t r e . Station paper no. 3. 5 pp. A N D E R S O N , T . W . (1958). A n introduct ion to m u l t i v a r i a t e s ta t i s t i ca l a n a l y s i s . J . W i l e y fk Sons, Inc. New Y o r k . 1 v o l . , 374 pp. A N D E R S O N , R . L . and B A N C R O F T , T . A . (1952). S ta t i s t i ca l theory i n r e s e a r c h . M c G r a w H i l l C y . New Y o r k . 1 v o l . , 400 pp. A R N O T T , J . T . (1971). P r o g r e s s report on f i e ld per formance of Douglas f i r and Western hemlock container seedlings on Vancouver Is land, B . C . P a c i f i c F o r e s t R e s e a r c h C e n t r e . C . F . S . Information report B C - X - 63. 59 pp. B A L D W I N , H . I . , E . J . E L I A S O N a n d D . E . C A R L S O N . (1973). I U F R O N o r w a y Spruce provenance tests in New H a m p s h i r e and New Y o r k . Si lvae genetica 22 (4): 93-114. B A N G , C . (1968). A study of the progeny of some forest trees i n D e n m a r k . Dansk Skovforen. T i d s s k r . 53 (11): 351-373. B A R T L E T T , M . S . (1947). The use of t rans format ions . B i o m e t r i c s 3: 39-52. B E C K E R , W . A . (1967). Manual of procedures i n quantitative genetics . The p r o g r a m in genetics . Washington State U n i v e r s i t y . Washington. 1 v o l . , 128 pp. 224 B E N S O N , L . (1962). P lant taxonomy. The Ronald press cy. New Y o r k . 1 v o l . 494 pp. B E N Z I A N , B . and S M I T H , H . A . (1973). Nutr ient concentrations of healthy seedlings and transplants of P i c e a s i tchensis and other conifers grown in E n g l i s h forest n u r s e r i e s . F o r e s t r y 46 (1): 55-69. B I R O T , Y . (1972). V a r i a b i l i t e i n f r a specifique du poids de l a graine chez le Douglas (Pseudotsuga m e n z i e z i i M i r b (Franco) . Si lvae genetica 21 (6): 230-243. B L A C K I T H , R . E . and R E Y M E N T , R . A . (1971). M u l t i v a r i a t e m o r p h o -m e t r i e s . A c a d e m i c p r e s s . London and New Y o r k . 1 v o l . 412 pp. B O U G H N E R , C . C . (1964). The d is t r ibut ion of growing-degree days i n Canada. C a n . M e t e o r o l . m e m o i r s 17. M e t e o r o l . b r a n c h . Dept . of T r a n s p o r t . 40 pp. B R A Z I E R , J . D . (1967). T i m b e r i m p r o v e m e n t . I. A study of the v a r i a t i o n i n wood c h a r a c t e r i s t i c s i n young Sitka spruce . F o r e s t r y 40 (2): 117-28. B R A Z I E R , J . D . (1967). Select ion of candidate trees on the bas is of t imber p r o p e r t i e s : assessment of wood quality i n breeding stock. F o r e s t products r e s e a r c h l a b o r a t o r y . P r i n c e s R i s b o r o u g h . 5 pp. B r i t i s h F o r e s t P r o d u c t R e s e a r c h Organizat ion. ( 1966). Home grown t imber invest igat ion . T i m b e r improvement by se lect ion and b r e e d -i n g . (Sitka spruce) . R e p . D i r e F o r . P r o d . R e s . London. 1965 - 66. 10 - 1. B R I X , H . (1972). G r o w t h response of Sitka spruce and White spruce seedlings to temperature and light intensi ty . P a c i f i c F o r e s t r e s e a r c h centre . C S F . B C - X - 74. 17 pp. B U R L E Y , J . (1965 a ) . Genetic v a r i a t i o n i n P i c e a s i tchensis (Bong) C a r r . P h . D . T h e s i s . 203 pp. D i s s e t T A b s t r . 26 (4): 1847. (1965, ). Genetic v a r i a t i o n in P i c e a s i tchensis (Bong) C a r r . _ _ _ _ _ _ _ _ _ _ _ _ £j _______ CT A l i t e ra ture r e v i e w . C o m m o n w . f o r , r e v . 44: 47-59. (1965 ). Karyotype analys is of Sitka spruce , P i c e a s i tchen-sis (Bong) C a r r . Si lvae genetica 14 (4): 127 - 132. 225 (1965^). V a r i a t i o n in seed c h a r a c t e r i s t i c s of Si tka spruce . Advanc ing f r o n t i e r s of plant sc iences 10: 11 - 24. (1966 a ) . Provenance v a r i a t i o n i n growth of seedling apices of Si tka spruce . F o r , S c i . 12 (2): 170-4. (1966^). V a r i a t i o n i n colour of Si tka spruce seedl ings . Quar t . J l . F o r . 60 (1): 51-4. (1966 c ) . Genet ic v a r i a t i o n i n seedling development of Si tka spruce , P i c e a s i tchensis (Bong) C a r r . F o r e s t r y X X X I X (1): 68-94. C A L L A H A M , J . P . (1963). Provenance r e s e a r c h - Investigation of genetic d i v e r s i t y assoc ia ted wi th geography. F A O / F O R G E N 63 - 3 / 0 . 24 pp. C A M P , L . and G I L L Y , P . (1943). The structure and o r i g i n of spec ies . B r i t t o n i a 4: 323-385. C A M P B E L L , R . K . and W I L S O N , B . C . (1973).. Spacing-genotype i n t e r -act ion in Douglas f i r . Si lvae genetica 22 (1-2) : 15-20. Canada Department of T r a n s p o r t . (1967). Tempera ture and p r e c i p i t a -tion tables for B r i t i s h C o l u m b i a . Meteoro logy branch r e p o r t . 46 pp. C A T T E L L , R . B . (1965). F a c t o r a n a l y s i s : an introduct ion to essent ia l s . I. The purposes and under ly ing m o d e l s . II. The role of F a c t o r ana lys i s i n r e s e a r c h . B i o m e t r i c s 21 (1 and 2): 190-215 and 405-435. C H A P M A N , J . D . (1952). The c l imate of B r i t i s h C o l u m b i a . T r a n s a c -tions of the 5th B r i t i s h C o l u m b i a N a t u r a l R e s o u r c e s Conf-erence . 1 v o l . 356 pp. p. 8 to 54. C H A P M A N , H . D . and P R A T T , P . F . (1961). Methods of ana lys i s for s o i l s , plants and w a t e r s . U n i v e r s i t y of C a l i f o r n i a . D i v i s i o n of A g r i c u l t u r a l sc iences . 1 v o l . 309 pp. C H A P M A N , L . J . and B R O W N , D . M . (1966). The c l imates of Canada for a g r i c u l t u r e . Report no. 3. The Canada land inventory . Queen's p r i n t e r . 36 pp. C H I L D E R S , N . F . rl, (1 954)'^, M i n e r a l nutr i t ion of f ru i t c r o p s . H o r t i -c u l t u r a l publ . Rutgers u n i v e r s i t y . New B r u n s w i c k , N . J . 1 v o l . 907 pp. 226 C L A R K , B . J . (1965). V a r i a t i o n in cone and seed c h a r a c t e r i s t i c s of Sitka spruce i n B r i t i s h C o l u m b i a . B a c h . s c i . F o r e s t r y thes i s . F a c u l t y of F o r e s t r y . U n i v e r s i t y of B r i t i s h C o l u m b i a . 1 v o l . 78 pp. C O C H R A N , W . G . (1947). Some consequences when the assumptions for the analys is of var iance are not sa t i s f i ed . B i o m e t r i c s 3: 22-38. C O C H R A N , W . G . and C O X , G . M . (1957). E x p e r i m e n t a l des ign . J . W i l e y and Sons, Inc. New Y o r k . 1 v o l . 611 pp. C O L E , A . J . ed. (1969). N u m e r i c a l taxonomy. A c a d e m i c p r e s s . London and New Y o r k . 1 v o l . 324 pp. C O O L E Y , W . W . and L O H N E S , F . . R . (1966). M u l t i v a r i a t e procedures for the behav iora l sc iences . J . W i l e y ck Sons, Inc. New Y o r k . 1 v o l . 211 pp. C O O P E R , P . W . (1963). S ta t i s t i ca l c l a s s i f i c a t i o n with quadratic f o r m s . B i o m e t r i k a 50: 439-448. C O X , D . R . (1958). P l a n n i n g of exper iments . J . W i l e y & Sons, Inc. New Y o r k . 1 v o l . 308 pp. D A G N E L I E , P . (I960). Contr ibut ion a l 'etude des communaute's vegetales par 1'analyse f a c t o r i e l l e . B u l l . Ser . Carte  P h y t o g . S.eries B . K . T o m e s I and II. 325 pp. D A G N E L I E , P . (1966). A propos des differentes methode-de c l a s s i f i -cation numer ique . Revue de statistique appliquee X I V (3): 55-75. D A L L I M O R E , W. and J A C K S O N , A . B . (1966). A handbook of coniferae and ginkgoaceae. (Edi t ion r e v i s e d by H A R R I S O N , S . G . ). A r n o l d publ . London. 1 v o l . 728 pp. D A U B E N M I R E , R . (1968). Some geographic v a r i a t i o n in P i c e a s i tchensis and their ecologic in terpre ta t ion . C a n . J l of Botany 46: 787-798. D A V I S , P . H . and H E Y W O O D , V . H . (1965). P r i n c i p l e s of A n g i o s p e r m ' taxonomy. O l i v e r and B o y d , Edinburgh and London. 1 v o l . 575 pp. D A Y , W . R . (1957). S i tka spruce in B r i t i s h C o l u m b i a . A study i n forest r e l a t i o n s h i p s . F o r e s t r y c o m m i s s i o n b u l l . no. 28. 109 pp. 227 D R A P E R , N.R. and S M I T H , H. (1966). A p p l i e d r e g r e s s i o n a n a l y s i s . J . W i l e y & Sons, Inc. New Y o r k and London, 1 v o l . 407 pp. D R I S C O L L , J.O. (1972). I. U. F. R. O. S2 - 02 - 1 2. - S i t k a s p r u c e p r o v e n a n c e w o r k i n g p l a n for i n t e r n a t i o n a l ten-provenance e x p e r i m e n t . S t e n c i l l e d notes. 68 pp. F o r e s t and w i l d l i f e s e r v i c e . D u b l i n 2. I r e l a n d . F A L C O N E R , D.S. (1964). I n t r o d u c t i o n to q u a n t i t a t i v e g e n e t i c s . O l i v e r & Boyd. E d i n b u r g h & London. 1 v o l . 356 pp. F A L K E N H A G E N , E.R. (1968 a). L a genecologie f o r e s t i e r e : une i n t r o -d u c t i o n . A n n a l e s de Gefnbloux 74: 181-192. (1968^). G e n e c o l o g y of y e l l o w b i r c h . P r o c . of the 11th m e e t i n g of the Comm.: on f o r e s t t r e e b r e e d i n g i n Canada. P a r t 2. p. 29. Dpt. of F i s h e r i e s and F o r e s t r y . (1972). Is t h i s the r i g h t a p p r o a c h to seed zone d e l i m i t a t i o n i n Canada? The F o r e s t r y C h r o n i c l e 48 (1): 9-10. F I S H E R , R.A. (1947). The d e s i g n of e x p e r i m e n t s . O l i v e r & Boyd. E d i n b u r g h & London. 1 v o l . 185 pp. _ " (1954). S t a t i s t i c a l methods f o r r e s e a r c h w o r k e r s . O l i v e r &; B o y d. E d i n b u r g h &; London. 1 v o l . 356 pp. F L E T C H E R , A.M. and F A U L K N E R , R. (1972). A p l a n f o r the i m p r o v e -ment of S i t k a s p r u c e by s e l e c t i o n and b r e e d i n g . . F o r e s t r y c o m m i s s i o n . London. R e s e a r c h and D evelopment paper no. 85. 31 pp. F O W E L L S , H.A. (1965). S i l v i c s of F o r e s t T r e e s of the U n i t e d Sta t e s . A g r i c u l t u r e Handbook No. 271. U.S.D.A. F o r e s t S e r v i c e . 1 v o l . 762 pp. G A L O U X , A. (1966). L a v a r i a b i l i t e genecologique du h e t r e commun (Fagus s y l v a t i c a L. )en B e l g i q u e . S t a t i o n de r e c h e r c h e s des E a u x et f o r e t s . G r o e n e n d a a l - H o e i l l a r t . T r a v a u x . S e r i e A. no. 11. 121 pp. G A L O U X , A. and F A L K E N H A G E N , E.R. (1965). R e c h e r c h e s sur l a v a r i a b i l i t e genecologique de l'e'rable s y c o m o r e ( A c e r pseudoplatanus L. Jen B e l g i q u e . P r e m i e r e p a r t i e . L a s a m a r e , l e s c o r r e l a t i o n s et l a c r o i s s a n c e au stade i n f a n t i l e . S t a t i o n de r e c h e r c h e s des E a u x et F o r e t s . G r o e n e n d a a l -H o e i l l a r t . T r a v a u x . S e r i e A. no. 10. B e l g i u m . 1 v o l . 133 pp. GOUNY, P. (1956). O b s e r v a t i o n s e n t r e l a c o m p o s i t i o n m i n e r a l e de l a plan t e et le r e n d e m e n t i n I.R.H.O. (1956). P l a n t a n a l y s i s and f e r t i l i z e r p r o b l e m s , p. 83-108. 228 H A D D O C K , P . G . , J . W A L T E R S , and A . K O Z A K . (1967). Growth of coastal and i n t e r i o r provenances of D o u g l a s - f i r (Pseudotsuga  m e n z i e s i i ( M i r b . Franco) at Vancouver and Haney i n B r i t i s h C o l u m b i a . U . B . C . R e s e a r c h no. 79. N o v . 1967. F a c u l t y of F o r e s t r y . 32 pp. H A G N E R , M . (1970). A genecological invest igat ion of the annual r h y t h m of P i n u s contorta D o u g l . and a c o m p a r i s o n with P i n u s s y l v e s t r i s L . Studia fores ta l i a suecica 81. 26 pp. H A N O V E R , J . W . and B A R N E S , B . V . (1962). H e r i t a b i l i t y of height growth i n 1 year o ld western white pine. P r o c . F o r e s t genetics workshop. S A F committee on forest tree i m p r o v e -ment . Southern F o r e s t tree improvement c o m m i t . Oct . 25-27, 1962. H A R M A N , H . (I960). M o d e r n factor a n a l y s i s . U n i v e r s i t y of Chicago p r e s s . Chicago . 1 v o l . 354 pp. H A R R I S , A . S . and R U T H , R . H . (1970). Si tka spruce - a b ib l iography wi th abs t rac ts . U S D A forest s e r v . r e s . pap. P N W - 105, 251 pp. P a c i f i c northwest forest and range exp. sta . , P o r t l a n d , Oregon . H A R T L E Y , H . O . (1967). Expectat ions , var iances and covar iances of A N O V A means squares by " s y n t h e s i s " . B i o m e t r i c s 23 (1): 105 - 114. " H A T T E M E R , H . H . (1963). E s t i m a t e s of h e r i t a b i l i t y publ ished i n forest tree breeding r e s e a r c h . F A O : F O R G E N , 63-2a:3. 14 pp. H E S L O P - H A R R I S O N , J . (1964). F o r t y years of genecology, in Advances i n eco logica l r e s e a r c h . V o l . 2. A c a d e m i c p r e s s . London and New Y o r k . 1964. 159-241 pp. H O R T O N , I . F . , R U S S E L L , J . S . and M O O R E , A . W. (1968). M u l t i -var iate - covariance and canonical a n a l y s i s : a method for select ing the most effective d i s c r i m i n a t o r s in a m u l t i v a r i a t e s i tuat ion. B i o m e t r i c s 24 (4): 845-858. H U H N , M . (1970a). The competi t ive environment and its genetic reac t ion v a r i a t i o n s . 2nd meeting of the working group on quantitative genetics sect . 22. I . U . F . R . O . R a l e i g h . A u g . 18-19, 1969. (1970D). Untersuchungen zur K o n k u r r e n z zwischen veschieden Genotypen in Pf lanzenbestaenden: I, II, III, I V . Silvae gene-t i ca 19 (Different i s sues ) . Idem 20 (5-6): 218-22*6"! 229 H U X L E Y , J . (1940). The new sys temat i cs . Clarendon p r e s s . O x f o r d . 1 v o l . 453 pp. I . R . H . O . (1956). P lant analys is and f e r t i l i z e r p r o b l e m s . I . R . H . O . P a r i s . 1 v o l . 410 pp. J A R D I N E , N . and SIBSON, R . (1971). . M a t h e m a t i c a l taxonomy. J . W i l e y and Sons, Inc. New Y o r k . 1 v o l . 286 pp. J E F F E R S , J . N . R . (1959). E x p e r i m e n t a l design and analys is in forest r e s e a r c h . A l m q u i s t and W i k s e l l . S tockholm. 1 v o l . 1 72 pp. J E F F E R S , J . N . R . and B L A C K , T . N . (1963). A n analys is of v a r i a b i l i t y i n P i n u s contorta . F o r e s t r y X X X V I (1):199-218. K A R L B E R G , S. (1961). Development and y i e l d of Douglas f i r (Pseudot-suga tax i fo l ia (Poir.) B r i t t . and Sitka spruce (P icea s i tchen-sis (Bong) C a r r . i n southern Scandinavia and on the P a c i f i c coast . K u n g l . Skogshogskolans s k r i f t e r nr 34. 141 pp. K E M P T H O R N E , O . (1952). D e s i g n and analys is of exper iments . J . W i l e y and Sons, Inc. New Y o r k . 1 v o l . 534 pp. K E M P T H O R N E , O . (1957). A n introduct ion to genetic s ta t i s t i c s . J . W i l e y and Sons, Inc. New Y o r k . 1 v o l . 545 pp. K E N D A L L , M . G . and S T U A R T , A . (1966). The advanced theory of s t a t i s t i c s . V o l . 3. Des ign and a n a l y s i s . T i m e s e r i e s . Hafner cy . New Y o r k . 1 v o l . 528 pp. K E N D A L L , M . G . (1969). A course i n m u l t i v a r i a t e a n a l y s i s . 1 v o l . 185 pp. Hafner cy . New Y o r k . K I N G , J . P . (1965). Seed source by environment in terac t ion i n Scots pine . 1. Height growth. 2. Needle length and c o l o u r . Si lvae  genetica 14 (4): 105-115; (5) 141-148. K O Z L O W S K I , T . T . (1971). Growth and development of t rees . V o l . I. Seed germinat ion , ontogeny and shoot growth. A c a d e m i c p r e s s . New Y o r k and London. 1 v o l . 443 pp. K R A J I N A , V . J . (1969). E c o l o g y of forest trees i n B r i t i s h C o l u m b i a in E c o l o g y of western N o r t h A m e r i c a . U n i v e r s i t y of B r i t i s h C o l u m b i a . V o l . 2. N o . 1, 147 pp. K R I E B E L , H . B . , G . N A M K O O N G , and R. A . U S A N I S . (1972). A n a l y s i s of genetic v a r i a t i o n i n 1-2 and 3 year o ld eastern white pine i n incomplete d i a l l e l c ross exper iments . Silvae genetica 21 (1-2): 44-48. L A C A Z E , J . F . (1970). Comportement de d iverses provenances d 'epicea de Sitka en F r a n c e . Revue fores t ier e f r a n qaise 22 ( l ) :45-54 . 230 L A N G L E T , O . (1936-37). Studier over tal lens f y s i o l o g i s k a v a r i a b i l i t e t och dess samband med K l i m a t e t . M e d d . Statens Skogs-tfbrs&ksantalt 29:219-470. (1959). A cl ine or not a c l i n e . A question of Scots pine . Si lvae genetica 8:13-21. • (1971). Two hundred years genecology. T A X O N 20 (5/6): 653-722. L A V E N D E R , D . P . and C A R M I C H A E L , D . L . (1966).. E f fec t s of three v a r i a b l e s on m i n e r a l concentrations i n Douglas f i r needles . F o r e s t science 12 (4):441-446. L A V E N D E R , D . P . (1970).. F o l i a r analys is and how it is used. A r e v i e w . O . S . U . School of F o r e s t r y . R e s e a r c h note 52. 8 pp. L E A F , A . L . (1968). K , M g and S def ic iencies in forest t rees . In Tennessee V a l l e y Author i ty (1968). F o r e s t f e r t i l i z a t i o n ; theory and p r a c t i c e , p . 88-122. L I N E S , R . (1964). E a r l y exper iments on the provenances of Si tka spruce . R e p . F o r . R e s . F o r . C o m m i s s i o n . London 1962/63: 126-146. L I N E S , R . and M I T C H E L L , A . F . (1966). Di f ferences i n phenology of Si tka spruce provenances . R e p . F o r . R e s . F o r . C o m m i s s i o n . London. 1964/65. 1966: 173-84. L I N S E R , H . ed . (1969). Handbuch der P f l a n z e n - e r n a e h r u n g and Duengung. E r s t e r B a n d : P f l a n z e n e r n a h r u n g . E r s t e Hael f te . 594 pp. Springer V e r l a g . Wien and New Y o r k . 3 v o l . M A H A M U N U L U , D . M . (1963).. Sampl ing var iances of the est imates of var iance components in the unbalanced 3-way nested c l a s s i -f i ca t ion . A n n , .math. . . stat. 34: 521-527. M A T T H E W S , R . G . (1971). Container seedling product ion: a p r o v i s i o n a l m a n u a l . P a c i f i c forest r e s e a r c h centre . C . F . S. Information r e p o r t . B . C . - X - 58. 57 pp. M E R G E N , F . and W O R R A L L , J . (1965). Ef fects of environment and seed sources on m i n e r a l content of Jack pine seedl ings . F o r e s t science 11 (4): 393-400. M E R G E N , F . and T H I E L G E S , B . A . (1967). In t ra - spe c i f i c v a r i a t i o n i n nuclear volume i n four conifers . - Evo lut ion 21 (4): 720-724. 231 M E R G E N , F . , J . W O R R A L L and G. M . F U R N I V A L . (1967). Genotype by environment interact ions i n 50 sources of jack pine seed-l i n g s . X I V I . U . F . R . O . K o n g r e s s . Muenchen papers . V o l . III. sect . 22. 459-466. M I L L I E R ' , C . and T O M A S S O N E , R . E . (1969). Me'thodes d 'ordinat ion et de c l a s s i f i c a t i o n : leur efficacite' 'et leurs l i m i t e s . C o l l . centre nat. de l a r e c h . sc ient . M a r s e i l l e s . 2-12 aout 1969. 34 pp. M O I R , R . B . and F O X , D . P . (1972). Supernumerary chromosomes in P i c e a s i tchensis (Bong.) C a r r . Si lvae genetica 21 (5): 182-186. M I K S C H E , J . P . (1971). Interspec i f i c v a r i a t i o n of D N A per c e l l between P i c e a s i tchensis (Bong.) C a r r . provenances . C h r o m o s o m a 32 (4): 343-352. M O L L , R . H . and R O B I N S O N , H . F . (1966). Observed and expected response in four se lect ion experiments i n M a i z e . C r o p science. ( 6 ) : 319-324. M O R R I S O N , D . F . (1967). M u l t i v a r i a t e s ta t i s t i ca l methods. M a c G r a w H i l l . New Y o r k . 1 v o l . 338 pp. M O R R I S O N , I . K . (1970). The absorpt ion of nutrients by roots of coniferous seedlings in r e l a t i o n to root c h a r a c t e r i s t i c s and s o i l condit ions . A b s t r of thesis in D i s s e r t . A b s t . 1970. 31B N A M K O O N G , G . (1966). Inbreeding effects on est imat ion of genetic additive v a r i a n c e . F o r e s t science 12 (1): 9 -13 . N A M K O O N G , G . , S N Y D E R , E . B . and S T O N E C Y P H E R , R . W . (1966). H e r i t a b i l i t y and gain concepts for evaluating breeding systems such as seedling o r c h a r d s . Si lvae genetica 15 (3): 76-84. N A N S O N , A . (1964). Donnees complementa ires au sujet de I'expe'rience internationale sur I 'or igine des graines d 'epicea en Be lg ique . Station de recherches des Eaux et F o r e t s . Groenendaa l -H o e i l l a r t . T r a v a u x B . no. 28. 38 pp. B e l g i u m . (1970). L 'her i tab i l i t e ' et le gain d 'or ig ine ge'netique dans quelques types d 'exper iences . Silvae genetica 19 (4): 113-121. P A T R I C , J . H . and B L A C K , P . E . (1968). Potent ia l evapotranspirat ion. and c l imate i n A l a s k a by Thornthwaite 1 s c l a s s i f i c a t i o n . U . S . D . A . F o r e s t s e r v . r e s . pap. P N W - 7 1 , 28 pp. i l l u s . P a c i f i c northwest forest and range experiment station, P o r t l a n d , Oregon . 232 P H E L P S , V . H . (1967). Should Sitka spruce be planted? T i m b e r talks no. 25. Department of F o r e s t r y . Canada I pp. (1973). Si tka spruce . A l i t e ra ture r e v i e w with spec ia l re ference to B r i t i s h C o l u m b i a . E n v i r o n m e n t - C a n a d a . F o r e s t r y s e r v i c e . B C - X - 83. 39 pp. P I N T A R I C , K . (1972). Ef fect of s t ra t i f i ca t ion on the germinat ion of seeds of P i c e a s i t chens i s . S u m a r s k i L s t . 96 (1/2): 63-68. P I R C H N E R , F . (1969). Popula t ion genetics i n a n i m a l breeding . F r e e m a n W . H . and cy . San F r a n c i s c o . 1 v o l . 274 pp. R A O , C . H . (1952). Advanced s ta t i s t i ca l methods i n b i o m e t r i c r e s e a r c h . J . W i l e y and Sons, Inc. New Y o r k . 1 v o l . 390 pp. R E U T E R , F . (1971). Some problems in testing provenances with spec ia l re ferences to the co-operat ive Douglas f i r provenance test at the U n i v e r s i t h of B r i t i s h C o l u m b i a r e s e a r c h fores t . M a s t e r ' s thes is . The U n i v e r s i t y of B r i t i s h C o l u m b i a . 1 v o l . 93 pp. R I C H A R D S O N , R . H . , K . K O J I M A and H . L . L U C A S . (1968). A n analys is of short t e r m select ion exper iments . Heredi ty 23 (4): 493-505. R O B A K , H . (1962). O v e r v i n t r i n g e n av en -og to a r i g s i tkagran i planteskolene. A r s s k r i f t 1961 for N o r s k e Skogplanteskoler . 27 pp. R O B E R T S O N , F . W . (1955). Select ion response and the proper t ies of genetic v a r i a t i o n . C o l d Spring H a r b o r Symp. on Quant, b iology X X X , : 166-177. R O C H E , L . (1968). The value of short t e r m studies i n provenance r e s e a r c h . The commonwealth f o r e s t r y r e v i e w 47 (1), no. 131: 14-26. (1969). A genecological study of the genus P i c e a i n B r i t i s h C o l u m b i a . New P h y t o l . 68 (2): 505-554. R O M B E R G E R , J . A . (1963). M e r i s t e m s , growth, and development in woody plants . U . S . Dep . agr . techn. b u l l . 1293. 1 v o l . 212 pp. R O U X , M . (1968). Un a lgor i thme pour cons t ru i re une h i e r a r c h i e p a r t i -c u l i e r e . These . 3eme c y c l e . U n i v e r s i t e de P a r i s . Not ava i lab le . 233 R O W E , J . S . (1964). E n v i r o n m e n t a l precondit ioning with spec ia l re ference to f o r e s t r y . Eco logy 45 (2): 399-403. R U T H , R . H . (1964). S i l v i c u l t u r e of the coastal S i tka spruce - western hemlock type. P r o c . soc. of A m e r . f o r e s t e r . Denver , C o l o r a d o : 32-36. S A L T , G . A . (1966). Diseases of Si tka spruce seedlings in forest n u r s e r i e s . R e p . Rothamst . exp. s ta . 133-134. S A M U E L , C . J . A . , R . C . B . J O H N S T O N E and A . M . F L E T C H E R . (1972). A d i a l l e l c r o s s i n Sitka spruce . A s s e s s m e n t of f i r st year charac ters in an e a r l y glasshouse test. T h e o r e t i c a l and appl ied genetics 42: 53-61. S C H E F F E , H . (1959). The analys is of v a r i a n c e . J . W i l e y and Sons, Inc. New Y o r k , 1 v o l . 346 pp. S C H O B E R , R . (1962). Die S i t k a - F i c h t e . Schr i f tenreihe der F o r s t l i c h e n Facul taet der U n i v e r s i t a e t Goett ingen. B d 24/25/230 pp. S E A L , H . (1964). M u l t i v a r i a t e s ta t i s t i ca l analys is for b i o l o g i s t s . Methuen and C o . London. 1 v o l . 207 pp. S E A R L E , S . R . (1971). L i n e a r m o d e l s . J . W i l e y and Sons, Inc. 1 v o l . New Y o r k , 532 pp. S I L E N , R . R . (1966). A 50 year r a c i a l study of Douglas f i r i n western Oregon and Washington. Western forest genetics a s s o c i a -tion proceedings 1965: 6-7. S I M A K , M . (1967). Seed weight of l a r c h f r o m different provenances ( L a r i x decidua M i l l ) . Studia f o r e s t a l i a suec ica 57. 51 pp. S P I C K E T T , S . G . and T HOD A Y , J . M . (1966). Regular responses to se lec t ion . 3. Interact ion between located polygenes. Genet, r e s . 7: 96-121. S Q U I L L A C E , A . E . (1970). Genotype - environment interact ions in forest t r e e s . 2nd meeting of the Working group on quantitative genetics . Sect. 22. I. U . F . R . O . R a l e i g h . A u g . 18-19, 1969. S M O U S E , P . E . (1972). The canonical analys is of mul t ip le species h y b r i -d i z a t i o n . B i o m e t r i c s 28: 361-371. S O K A L , R . R . and S N E A T H , P . H . (1963). P r i n c i p l e s of n u m e r i c a l taxonomy. F r e e m a n W . H . and cy . San F r a n c i s c o . 1 v o l . 359 pp. 234 S T E I N B E C K , K . (1966). Site, height and m i n e r a l nutrient content re lat ions of Scotch pine provenances . Si lvae genetica 15 (2): 42-50. S T E R N , K . (1964). Herkunftsversuche fuer Zwecke der F o r s t - P f l a n z e n -zuechtung er laeuter t a m B e i s p i e l zweier M o d e l l v e r s u c h e . P e r Zue enter 34 (5): 182-219. S T E R N , K . (1962). P r e l i m i n a r y est imates of the genetic s t ructure of two s y m p a t r i c populations of b i rches as determined by random effects and natura l se lec t ion . P r o c e e d , of the 9 th n o r t h -eastern forest tree Imp. Conf. A u g . 23-25, 1961. Syracuse , N . Y : 25-34. Tennessee V a l l e y A u t h o r i t y . (1968). F o r e s t f e r t i l i z a t i o n theory and p r a c t i c e . P a p e r s presented at the Sympos ium on forest f e r t i l i z a t i o n . A p r i l , 1967. G a i n e s v i l l e . F I . 306 pp. T U R E S S O N , G . (1923). The scope and importance of genecology. Heredi tas 4: 171-176. V A A R T A J A , K . (1959). Evidence of photoperiodic ecotypes i n t rees . E c o l o g i c a l monographs 29: 91-111. V A N C A M P O - D U P L A N and G A U S S E N , H . (1948). Sur quatre hybrides de genres chez les A b i e t i n e e s . T r a v . L a b . F o r e s t . Toulouse 1. 4 (24). 14 pp. V A N D E N D R I E S S C H E , R . (1965). L a recherche des constel lat ions de groupes a p a r t i r des distances ge'neralise'e s D ^ de Mahalano-b i s . B i o m e t r i e - P r a x i m e t r i e (6): 36-47. (1969). T i s s u e nutrient concentrations of Douglas f i r and Sitka spruce . B . C . F . S. R e s e a r c h note no. 47, 42 pp. (1971). Response of conifer seedlings to ni trate and a m m o n i u m sources of n i t rogen . P l a n t and s o i l 34: 421-439. (1973). F o l i a r nutrient concentrat ion d i f fer -ences between provenances of Douglas f i r and their s i g n i -f icance to fo l ia r ana lys i s in terpre ta t ions . C a n . J l . f o r . r e s . 5 (2): 323-328. V E G I S , A . (1965). Die Bedeutung von p h y s i k a l i s c h e n und chemischen Aussenfaktoren be i der Induktion und Beendigung von Ruhezustanden being Organen und Geweben hoherer P f l a n z e n . In Hanbuch der P f lanzenphys io log ie , herausgegeben von W . R U H L A N D . Band X V . D i f f e r e n z i e r u n g und E n t w i c k l u n g . T e i l 2, pp. 534-668, Spr inger - V e r l a g . 235 V O I G T , G . K . (1968). V a r i a t i o n in nutrient uptake by t rees . In Tennessee V a l l e y A u t h o r i t y . (1968). F o r e s t f e r t i l i z a t i o n theory and p r a c t i c e , p. 20-27. W A K E L E Y , P . C . (1961). Resul t s of the southwide pine seed source study through 1960-61. P r o c . s ix th southern conf. on forest tree improvement . G a i n e s v i l l e , F l o r i d a . 7-8 June, 1961, p . 10-24. W A R I N G , G . H . and Y O U N G B E R G , C . T . (1972). - E v a l uating for est sites for potential growth response of trees to f e r t i l i z e r , Northwest science 46 (1): 67-75. W R I G H T , J . W . (1955). Species c r o s s a b i l i t y in spruce in re la t ion to d i s t r ibut ion and taxonomy. F o r e s t science 1 (4): 319-349. (1963). Aspec ts genetiques de 1 'amel iorat ion des a r b r e s f o r e s t i e r s . F A O no. 16. 1 v o l . 431 pp (1970). A n i m p r o v e d r e c o r d system for forest genetic n u r s e r y and plantation studies . Silvae genetica 19 (2-3): 64-68. (1973). Genotype-environment interact ion i n N o r t h C e n t r a l Uni ted State s. Fo res t science 19(2): 113-123. W R I G H T , K . H . (1960). S i tka spruce w e e v i l . F o r e s t pest leaf let 47. U . S . D . A . F o r e s t s e r v i c e . - 66 pp. Y A O , C . (1971). Geographic v a r i a t i o n i n seed weight, some cone scale measurements and seed germinat ion of Douglas f i r (Pseudotsuga m e n z i e s i i ( M i r b . ) F r a n c o . ) M a s t e r ' s thes is . The U n i v e r s i t y of B r i t i s h C o l u m b i a . 88 pp. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0093236/manifest

Comment

Related Items