T H E E V O L U T I O N O F I N B R E E D I N G I N W E S T E R N R E D C E D A R (THUJA PLICATA: C U P R E S S A C E A E ) by L I S A M A R I E O ' C O N N E L L B . A . University of Ottawa, 1993 B.Sc. Dalhousie University, 1995 M . S c . Queen's University, 1997 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L 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 T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Department of Forest Sciences) We accept this thesis as conforming to the required 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 2003 © Lisa Marie O'Connell, 2003 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of forfs't Sci e rt c*5 The University of British Columbia Vancouver, Canada Date April H , 2 ^ 0 0 3 DE-6 (2/88) Abstract L o n g - l i v e d w o o d y plants usua l ly show h igh levels o f outcross ing, inbreeding depression and genetic d ivers i ty compared to other plants. A r ev iew o f the literature showed a mean oucross ing rate o f 83.5 i n conifers , and a posi t ive , but weak, corre la t ion between outcross ing and genetic d ivers i ty . A m o n g conifers , western redcedar (Thuja plicata, Cupressaceae) has one o f the highest rates o f self-fer t i l izat ion and lowest amount o f genetic d ivers i ty , and thus offers the opportuni ty to study the evo lu t ion o f inbreeding i n a predominant ly outcross ing group o f plants. T h i s thesis l i n k s the evo lu t ion o f inbreeding i n redcedar w i t h a loss i n inbreeding depression and genetic d ivers i ty . U s i n g one p o l y m o r p h i c i s o z y m e marker , I obta ined an average popula t ion outcross ing estimate o f 7 1 % over s ix natural populat ions o f redcedar. I deve loped 13 h i g h l y p o l y m o r p h i c microsate l l i te markers to conduct a f iner-scale study o f the mat ing sys tem and genetic structure o f redcedar. A new method o f b u l k i n g seedlings to estimate outcross ing rates was used to identify eco log i ca l correlates o f outcrossing. Se l f ing rates increased s igni f icant ly w i t h tree height i n four different populat ions . P o l l e n f rom larger trees probably made up a larger propor t ion o f the surrounding po l l en c loud , increas ing se l f -pol l ina t ion . There was no var ia t ion , however , i n the amount o f inbreeding among c r o w n posi t ions w i t h i n trees. In a seed orchard, a combina t ion o f cont ro l led crosses and i s o z y m e markers showed evidence that pos t -pol l ina t ion compet i t ion between embryos w i t h i n an ovu le decreased self ing. I used eight microsate l l i te l o c i to study patterns o f range-wide genetic structure i n redcedar. A phy logeograph ic analysis suggests that redcedar probably su rv ived i n three separate refugia dur ing the last g lac ia t ion . These results also suggest that i f a species-wide bott leneck is at the root o f reduced genetic d ivers i ty i n redcedar, i t p robably predates the last g lac ia t ion . T h e combina t ion o f an inbreeding mode o f reproduct ion and a bot t leneck probably contr ibuted to the decrease i n genetic d ivers i ty presently observed i n redcedar. F i n a l l y , after screening 80 trees at eight microsate l l i te l o c i , a single stepwise mutation was observed, yielding a somatic mutation rate of 6.3 x 10"4 (95% CI: 3.0 x 10"5 - 4.0 x 10"3) mutations per locus per generation in western redcedar. i v Table of Contents Abst rac t i i Tab le o f Contents i v L i s t o f Tab les v i i i L i s t o f F igures x L i s t o f A p p e n d i c e s • X 1 i A c k n o w l e d g m e n t s X U 1 P u b l i s h e d papers X 1 V Chapter 1 Gene ra l in t roduct ion and ove rv i ew 1 T h e evo lu t ion o f plant mat ing systems 1 Con i f e r s 3 Patterns o f genetic d ivers i ty i n conifers 3 R e v i e w o f outcross ing rates in conifers 6 Outc ross ing rate and genetic d ivers i ty 8 Na tu ra l populat ions vs seed orchards 10 Inbreeding depression i n conifers 11 M a t i n g system o f conifers 11 The genus Thuja (Cupressaceae) 12 Wes te rn R e d Ceda r (Thuja plicata) 12 E c o l o g y o f Thuja plicata - 12 Genet ic d ivers i ty i n Thuja plicata 13 Inbreeding depression i n Thuja plicata 13 Spec ies -wide bott leneck 14 Thes is o v e r v i e w 15 Chapter 2 T h e mat ing system i n natural populat ions o f western redcedar 16 Int roduct ion 16 Ma te r i a l s and methods IV Sample col lec t ions IV I sozyme analyses 18 D a t a analysis 18 Resul t s 19 D i s c u s s i o n 20 Outc ross ing rates 20 Factors affecting mat ing systems 20 Cor re la t ion o f paternity 22 V Chapter 3 Charac ter iza t ion o f microsate l l i te l o c i in western redcedar 24 Introduct ion 24 Mate r i a l s and methods 24 C l o n e development 24 Screen ing for p o l y m o r p h i s m s 25 Resul t s and D i s c u s s i o n 26 Chapter 4 F ine-sca le est imation o f outcrossing in western redcedar w i t h microsate l l i te assay o f bu lked D N A 29 Introduct ion 29 Mate r i a l s and methods 31 B u l k i n g tests 31 Sample co l lec t ions 32 D N A assay o f bu lks 33 Es t ima t ion o f outcrossing f rom b u l k samples 34 Resul t s 36 A l l e l e detection 36 B u l k i n g tests 37 Genet ic d ivers i ty 39 Outcross ing rates 39 D i s c u s s i o n 43 V a r i a t i o n in outcrossing rates 43 Popu la t ion outcross ing rates 44 B u l k i n g samples 45 Chapter 5 P o l y e m b r y o n y and early inbreeding depression i n a self-fertile conifer , Thuja plicata (Cupressaceae) 46 Introduct ion 46 Mate r i a l s and M e t h o d s 48 Po l l ina t ions 48 Seed v i ab i l i t y 50 E m b r y o compet i t ion 51 E x p e c t e d seed set and self ing w i t h p o l y e m b r y o n y 52 Fi tness o f se l f -pol len 54 Resul t s 56 Seed set 56 R e a l i z e d self ing rates 6 0 Success o f se l f po l l en 62 v i D i s c u s s i o n 63 P o l y e m b r y o n y as a rescue m e c h a n i s m 63 E m b r y o compet i t ion 64 E a r l y inbreeding depression se l f - incompat ib i l i ty 65 P u r g i n g o f inbreeding depression 65 T h e importance o f pre-pol l ina t ion mechanisms 66 Chapter 6 R a n g e - w i d e genetic structure and divers i ty i n western redcedar 68 Introduct ion 68 Mathe r i a l s and methods 7 0 Sample co l l ec t ion 70 Mic rosa t e l l i t e screening 74 D i v e r s i t y analyses 74 Phy logeography 75 C l i n e s i n a l le le frequencies 76 L i n k a g e d i s e q u i l i b r i u m 7 6 Bot t l eneck test 76 Resul t s 78 Genet ic D i v e r s i t y 78 Genet ic Structure 80 Isolat ion by distance 82 Nor the rn vs southern populat ions 84 M a t i n g sys tem 85 A l l e l e size dis t r ibut ion 85 C l i n e s i n a l le le frequencies 89 Bot t l eneck test 92 D i s c u s s i o n 95 Phy logeograph ic structure 95 Popu la t ion differentiat ion 99 R e d u c t i o n i n genetic d ivers i ty 100 T i m i n g a species-wide bott leneck 101 Inbreeding and genetic d ivers i ty 102 Chapter 7 Soma t i c mutations at microsate l l i te l o c i i n western redcedar 104 Introduct ion 104 Mate r i a l s and methods 105 E s t i m a t i n g mutat ion rate 105 Sample co l lec t ions 107 v i i Mic rosa te l l i t e s 108 Resul t s 109 Mic rosa t e l l i t e mutations 109 T y p e o f mutat ion 110 Somat ic mutat ion rate estimate 110 D i s c u s s i o n U l Somat ic mutat ion rate U l M u t a t i o n mode l 112 T h e consequences o f somatic mutations i n redcedar 112 Genet ic m o s a i c i s m 113 Chapter 8 G e n e r a l d i scuss ion and conc lus ions 114 M a i n f indings 114 M a t i n g system 114 Inbreeding depression 115 Gene t ic structure and divers i ty 117 T y i n g it a l l together 118 Further research 119 R e v i e w o f mat ing systems i n trees 119 G l a c i a l refugia 119 A l l e l e dis t r ibut ion 120 M a t i n g sys tem at the edge o f the d is t r ibut ion 121 Re la ted species 121 References 123 Vlll List of Tables 1.1 M e a n levels o f w i th in popula t ion i s o z y m e var ia t ion i n three species o f Thuja compared to mean levels i n gymnosperms and other plants 5 1.2 M e a n outcross ing rate (t ± S D ) b y genus i n 52 species o f conifers 6 1.3 Outc ross ing rates i n 13 species o f conifers w i t h estimates f rom both natural and seed orchard populat ions 10 2.1 Loca t ions , sample sizes (AO, gene frequency o f the most c o m m o n al lele at locus G 6 p d , popula t ion outcross ing rates (0 and correla t ion o f paternity (r p ) o f Thuja plicata populat ions i n southwestern B r i t i s h C o l u m b i a 23 3.1 Charac ter iza t ion o f Thuja plicata microsatel l i tes i n a coastal and two inter ior popula t ions ... 27 4.1 Probabi l i t ies o f band patterns observed for a s ingle progeny, and for b u l k e d progenies o f sizes 2 and 3, condi t ioned upon maternal genotype (homozygous A , A , or heterozygous A , A , ) 35 4.2 T h e total number o f alleles detected i n samples b u l k e d before D N A extract ion i n two trees o f Thuja plicata. Samples were scored at four microsate l l i te l o c i ( T P 1 , T P 3 , T P 9 and T P 1 1 ) 38 4.3 Gene t ic d ivers i ty measures and total number o f al leles detected at four microsate l l i te l o c i ( T P 1 , T P 3 , T P 9 and T P 1 1 ) in four natural popula t ions o f Thuja plicata 40 4.4 Outc ross ing rates ( S E ) at different posi t ions w i t h i n the c r o w n o f trees i n four natural popula t ions o f Thuja plicata 41 4.5 M e a n tree heights and i n d i v i d u a l tree outcross ing rates (0 i n four popula t ions o f Thuja plicata 43 5.1 E x p e r i m e n t a l design o f a po l l ina t ion exper iment i n four trees i n a Thuja plicata seed orchard. O n e hundred percent se l f -pol len (selfed) and 0 % sel f -pol len (crossed) treatments as w e l l as po l l en mixtures w i t h three different ratios o f self/cross po l l en ( 2 5 % / 7 5 % ; 5 0 % / 5 0 % ; 75%/25%) were appl ied to each tree 49 5.2 Probabi l i t i es o f setting a f u l l seed (/•) and setting a selfed seed (s,) w i t h one, two or three embryos per ovu le (n) 53 5.3 T h e propor t ion o f fu l l seeds ( S E ) and inbreeding depression at the seed stage i n four western redcedar trees (181, 295, 431 and 432) 58 i x 5.4 T h e propor t ion o f f u l l seeds expected (/•) based on the number o f embryos per ovu le (n) and the propor t ion o f self po l l en appl ied (pk) i n four Thuja plicata trees w i t h v a r y i n g levels o f embryo v i ab i l i t y 59 5.5 T h e propor t ion f u l l seeds expected (j)) wi thout p o l y e m b r y o n y (when n = 1) and observed f u l l seeds for three different proport ions o f se l f -pol len i n four Thuja plicata trees 60 5.6 T h e propor t ion o f selfed seeds expected (when n = 1) and observed for three different proport ions o f se l f -pol len i n three Thuja plicata trees 62 5.7 F i tness o f se l f -pol len relat ive to outcross-pol len (ws) w h e n app l i ed at different proport ions i n three Thuja plicata trees 63 6.1 L o c a t i o n o f 23 sampled populat ions o f Thuja plicata '. 73 6.2 M e a n d ivers i ty at eight microsate l l i te l o c i i n 23 populat ions o f Thuja plicata 75 6.3 Gene t i c d ivers i ty measures i n 23 populat ions o f Thuja plicata at seven microsate l l i te l o c i . . . .' 79 6.4 F-stat is t ics at eight microsate l l i te l o c i i n Thuja plicata 80 6.5 A n a l y s i s o f molecu la r variance ( A M O V A ) o f the effects o f populat ions and groups (Nor th vs South) on the dis t r ibut ion o f genetic d ivers i ty i n Thuja plicata based on seven microsate l l i te l o c i 82 6.6 A n a l y s i s o f covar iance ( A N C O V A ) o f the effects o f geographica l distance between popula t ions and groups ( N vs N , S vs S, and N vs S) on pa i rwise genetic distance (6). . . . 84 6.7 A l l e l e s w i t h frequencies s igni f icant ly correlated w i t h latitude at the 0.05 l e v e l at seven microsate l l i te l o c i i n Thuja plicata . . . .90 6.8 Resul ts o f the Bo t t l eneck test for two mutat ion models i n Thuja plicata. Twenty- three popula t ions were d i v i d e d into three groups based on the phy logeograph ic analysis : southern popula t ions (exc lud ing Ca l i fo rn i a ) , northern popula t ions and C a l i f o r n i a on ly . I sozyme data for eight populat ions were obtained f rom Y e h (1988) 94 7.1 D e s c r i p t i o n o f eight microsate l l i te l o c i used to genotype 80 Thuja plicata trees f rom four natural popula t ions 108 X List of Figures 1.1 Trai ts that are pos i t ive ly correlated i n plants i n meta-analyses o f the literature 2 1.2 D i s t r i bu t ion o f outcrossing rates i n 52 species o f conifers 7 1.3 Cor re la t ion between outcross ing rate (?) and mean expected heterozygosi ty (H e p ) i n 40 species o f conifers. Species o f Thuja are indica ted by crosses 9 4.1 D i a g r a m o f s ix cone co l l ec t ion posi t ions i n a Thuja plicata tree: (1) top, v igorous branches (2) top, inner branches (3) m i d , outer branches (4) m i d , inner branches (5) lower , outer branches and (6) lower , inner branches 33 4.2 B a n d intensity prof i le o f three b u l k e d ind iv idua l s at locus T P 9 f rom lane 6 on the microsate l l i te ge l . The four detected al leles are ind ica ted by b lack arrows on the band intensity prof i le 37 4.3 I n d i v i d u a l tree outcross ing rate estimates regressed on tree height in four popula t ions o f Thuja plicata. N = 73 42 5.1 T h e propor t ion o f selfed seeds expected w i t h different propor t ion o f se l f -pol len and number o f embryos (n) w i t h i n an ovule . T w o different outcomes are shown: the outcrossed embryo a lways outcompetes the selfed embryo (outcross wins ) or se l f and outcross embryos are equal ly compet i t ive (chance). A l l embryos are v iab le and there is no inbreeding depression 55 5.2 T h e propor t ion o f fu l l seeds obtained i n four Thuja plicata trees w i t h different proport ions o f se l f -pol len appl ied . ./V = 495 57 5.3 T h e propor t ion o f selfed seeds expected and observed (± S E ) w i t h different propor t ion o f se l f -pol len . N = 2028. n, number o f embryos per ovu le . E m b r y o v iab i l i t i es for the expected selfed seeds is based on the mean o f three trees 61 6.1 Range map and loca t ion o f 23 sampled populat ions o f Thuja plicata. The shaded areas indicate the range o f western redcedar 72 6.2 N e i g h b o r - J o i n i n g tree o f 23 populat ions o f Thuja plicata based on N e i ' s standard genetic distance 81 6.3 P a i r w i s e genetic distance (0) as a funct ion o f geographic distance between populat ions o f Thuja plicata 83 6.4 A v e r a g e number o f al leles per locus ( A / L ) , mean expected heterozygosi ty (Hep) and mean inbreeding coefficients (F) i n 23 populat ions o f Thuja plicata as a funct ion o f lati tude. . 86 xi 6.5 a-d Allele size distribution over the range of western redcedar at four microsatellite loci: T P l , T P 3 , T P 4 a n d T P 6 : 87 6.5 e-h Allele size distribution over the range of western redcedar at four microsatellite loci: TP7, TP8, TP9 and TP11 88 6.6 Relative frequency of (a) 12 northern and (b) 23 southern alleles as a function of latitude in 23 populations of Thuja plicata 91 6.7 Proportion of alleles at eight loci in different frequency classes for southern populations and northern populations of Thuja plicata. (n = 189 alleles) 93 6.8 Hypothesized post-glacial colonization routes for Thuja plicata from three glacial refugia discussed in the text: (1) California, (2) Queen Charlotte Islands, and either (3) western Oregon or (4) northern Idaho 97 7.1 Image of a microsatellite gel showing the genotype at locus TP9 for two different heights within the same tree. Two collections of ten bulked megagametophytes were made from three heights in each tree 109 7.2 Allele distribution at locus T P 9 over four populations of Thuja plicata (N = 80 trees). The new allele, which increased from 34 to 35 dinucleotide repeats, is indicated by the white box, with the arrow showing the original allele 110 Xll L i s t o f Appendices I D e s c r i p t i o n o f genetic d ivers i ty parameters 143 II L i te ra ture r e v i e w o f genetic d ivers i ty i n 50 species o f conifers i n c l u d i n g 35 species i n the Pinaceae, 13 i n the Cupressaceae and one i n the Taxodiaceae 144 III Li tera ture r ev iew o f outcrossing rates i n 52 species o f conifers and 32 species o f ang iosperm trees 149 IV Wes te rn redcedar fol iage D N A C T A B extraction p ro toco l 156 V T w e l v e western redcedar D N A sequences f rom w h i c h p r imer pairs where designed and amp l i f i ed scorable and var iable microsatel l i te l o c i 158 VI Pa i rwi se genetic distances between 23 populat ions o f Thuja plicata based on eight microsatel l i te l o c i . N e i ' s (1972) genetic distance is above the d iagonal .and Fsl (0) b e l o w the d iagonal 162 Xl l l Acknowledgements Fi r s t o f a l l , I thank m y supervisor K e r m i t R i t l a n d whose o r ig ina l w a y o f th ink ing and his receptiveness to new ideas he lped approach science i n a new an interest ing way . I thank the members o f m y supervisory commit tee John R u s s e l l , S a l l y Ot to and Y o u s r y E l - K a s s a b y , w h o have not on ly kept me on the r ight track but also he lped me t remendously a long the way w i t h their words o f encouragement. A n d o f course the lab w o r k w o u l d never have gone so smoothly i f C a r o l R i t l a n d hadn't been there to keep everyth ing running . I thank Freder ique V i a r d w h o w o r k e d out the i sozymes protocols for redcedar and began the work , Jeff G l a u b i t z and G w e n a e l V o u r c ' h w h o p rov ided samples o f redcedar D N A . I thank John R u s s e l l for s h o w i n g me the ropes i n Cupressaceae research. H e p r o v i d e d thousands o f samples, set up the faci l i t ies and p rov ided the in format ion needed to conduct m y studies. I thank H e i d i C o l l i n s o n and T i m C r o w d e r for their help at the M t . N e w t o n Seed Orcha rd . I thank m y parents, R o d and Ro lande , w h o a lways supported and encouraged me through m y many , many , many years o f univers i ty and made sure I w o u l d get through i t . I also thank m y lab " f ami ly" for their fr iendship and support: B r y a n Ie, D a w n M a r s h a l l , C a r o l G o o d w i l l i e , D i l a r a A l l y , Char les " C h i n - L i n " C h e n , M a r i s s a L e B l a n c , Y a n i k Berube , H u g h W e l l m a n , W a s h i n g t o n Gapare , Jodie K r a k o w s k i , A l l y s o n M i s C a m p b e l l , D a w n Cooper , M a r k V a n K l e u n e n , J ac lyn B e l a n d and Jennifer W i l k i n . W e created a home far away f r o m home and were a lways there for each other. T h a n k y o u to M a r k , C a r o l , A l l y s o n , D i l a r a , M a r i s s a , Jodie for t ak ing the t ime to read this thesis and suggesting many improvements . I also thank Jeannette W h i t t o n , S a l l y A i t k e n and D a n Shoen for their careful reading and helpful comments on this thesis. F u n d i n g was p r o v i d e d b y a Na tu ra l Sciences and E n g i n e e r i n g Research C o u n c i l o f Canada post-graduate scholarship ( P G S B ) and an Isaak W a l t o n K i l l a m pre-doctoral F e l l o w s h i p and a research assistanceship f rom K . R i t l a n d . "After great pain, a formal feeling comes." E m i l y D i c k i n s o n x i v Published Papers Chapter 2 is a rev ised vers ion o f the f o l l o w i n g paper: O ' C o n n e l l , L . M . , F . V i a r d , J . R u s s e l l , and K . R i t l a n d . 2001 . The mat ing system i n natural populat ions o f western redcedar (Thuja plicata). Canad i an Journal o f B o t a n y 79: 753-756 . F o r this study Freder ique V i a r d co l lec ted the seeds f rom two populat ions , o p t i m i z e d the i s o z y m e pro tocol and genotyped the seedlings for the first year o f the study. John R u s s e l l co l l ec ted the seeds f r o m five populat ions dur ing the second year. I genotyped a l l the seedlings f rom the second year, conducted the analyses and wrote the paper. K e r m i t R i t l a n d supervised the study and the analyses, and edited the manuscript . K e r m i t R i t l a n d . . . Chapter 3 is a rev ised vers ion o f the f o l l o w i n g paper: O ' C o n n e l l , L . M . , and C . E . R i t l a n d . 2000. Charac ter iza t ion o f microsatel l i te l o c i i n western redcedar (Thuja plicata). M o l e c u l a r E c o l o g y 9: 1920-1922. T h e c lones conta in ing microsatel l i tes were obtained f rom C r a i g N e w t o n ( B C Research) . C a r o l R i t l a n d o p t i m i z e d the p ro toco l for sequencing the clones and supervised a l l the steps f rom pr imer design, D N A iso la t ion and microsate l l i te screening, and edi ted the paper. I conducted the majori ty o f the lab w o r k and wrote the paper. C a r o l R i t l a n d : . . . . 1 Chapter 1 General introduction and overview The evolution of plant mating systems A major trend i n the evo lu t ion o f plant mat ing systems is a t ransi t ion f rom cross-fer t i l iza t ion to self-fer t i l izat ion. Ident i fy ing the selective factors i n v o l v e d i n the evo lu t ion o f plant mat ing systems has been the subject o f a large amount o f both theoretical (e.g.: L a n d e and Schemske , 1985; Jarne and Char leswor th , 1993; U y e n o y a m a et ah, 1993) and e m p i r i c a l w o r k resul t ing i n estimates o f outcross ing rates for over 200 species o f plants ( rev iewed i n Barret t et al., 1996). Meta-ana lyses o f the literature have shown that a species' ma t ing system, genetic d ivers i ty , inbreeding depression and l i fe his tory are interdependent ( F i g . 1.1). Species w i t h h i g h self-fer t i l izat ion rates tend to show lower genetic d ivers i ty and less inbreeding depression than predominant ly outcross ing species (Char leswor th and Char l e swor th , 1995; H a m r i c k and Godt , 1996; H u s b a n d and Schemske , 1996). A n n u a l plants are often selfers, and l o n g - l i v e d w o o d y plants are predominant ly outcrossers (Barrett and Ecker t , 1990; Barret t et al, 1996). L o n g - l i v e d w o o d y plants also tend to have higher genetic d ivers i ty than other plants ( H a m r i c k et ah, 1992; H a m r i c k and God t , 1996). Inbreeding depression is seen as a major d r i v i n g force i n the evo lu t ion o f mat ing systems, favor ing many traits that prevent plants f r o m sel f - fer t i l iz ing (Char leswor th and Char leswor th , 1987). L o n g e r - l i v e d plants, i n c l u d i n g w o o d y plants, are expected to accumulate a h igher genetic l o a d through somatic mutations, consequently main ta in ing outcross ing ( M o r g a n , 2001) . H i g h l y deleterious recessive mutat ions affecting ear ly l i fe stages such as seed product ion , germinat ion and early su rv iva l are expected to be readi ly purged i n sel f ing plants w h i l e later act ing inbreeding depression affecting g rowth and fer t i l i ty should be more d i f f icu l t to purge (Husband and Schemske , 1996). In a r ev iew compar ing the t i m i n g o f inbreeding depression and the mat ing sys tem o f plants, H u s b a n d and Schemske (1996) found that sel f ing plants had less inbreeding depression dur ing the early stages o f their l i fe c y c l e than d i d outcrossers. H o w e v e r , 2 at later stages there was no difference i n inbreeding depression between se l f - fer t i l iz ing and outcross ing species. Shor t - l i ved plants m a y also be more successful at purg ing inbreeding depression than longe r - l ived species but ove ra l l the results are i nconc lu s ive (Byer s and W a l l e r , 1999). Nevertheless , most t rop ica l angiosperm trees and temperate conifers show h i g h l y reduced seed set f o l l o w i n g se l f -pol l ina t ion ( B a w a , 1974; K o r m u t ' a k and L i n d g r e n , 1996; H u s b a n d and Schemske , 1996). Outcrossing 4 Genetic diversity Inbreeding depression f Lifespan Fig. 1.1 Trai ts that are pos i t ive ly correlated i n plants i n meta-analyses o f the literature. References (a) H a m r i c k and G o d t 1992 (b) H u s b a n d and Schemske , 1996 (c) Barret t and Ecker t , 1990 (d) Barret t et al, 1996 (e) H a m r i c k and God t , 1996 (f) B y e r s and W a l l e r , 1999. A l t h o u g h inbreeding depression is important i n shaping select ion on mat ing systems, other theoretical models have also incorporated po l l en eco logy and l i fe-his tory (Hols inger , 1991; M o r g a n et al, 1997; Johnston, 1998). Different costs and benefits are associated w i t h both self ing and outcross ing. F o r example , the advantage o f reproduct ive assurance through self ing i n 3 annual and c o l o n i z i n g plants i n the absence o f outcross p o l l e n has l o n g been recogn ized (Stebbins, 1950). In perennia l plants, reduced surv ivorsh ip and fecundi ty i n later years m a y increase the cost o f sel f ing b y p roduc ing less fit inbred seeds, rather than m a x i m i z i n g fitness b y de l ay ing reproduct ion un t i l non- inbred p o l l e n is avai lable ( M o r g a n et al, 1997). T h e relat ive amounts o f se l f and outcross po l l en avai lable to a plant w i l l a lso affect select ion on its mat ing sys tem (Hols inger , 1991). F o r example , the large size o f trees can increase se l f -pol l ina t ion through ge i tonogamy (sel f -pol l inat ion between different f lowers on the same plant) and consequent ly lead to the evo lu t ion o f mechanisms to prevent self ing (Barrett et al, 1996). V a r i a t i o n i n space and t ime i n the ava i l ab i l i ty o f unrelated po l l en can lead to a stable m i x e d -mat ing system (Hols inger , 1991). A l t h o u g h traits correlated w i t h mat ing systems can be ident i f ied at the species l eve l , a large amount o f the var ia t ion i n outcross ing rates occurs w i t h i n species. T o better ident i fy traits associated w i t h plant mat ing systems, differences among c lose ly related species, or among populat ions and ind iv idua l s w i t h i n a species, need to be examined (Barrett and Ecker t , 1990; Barret t et al, 1996). In this thesis I w i l l obtain estimates o f self-fer t i l izat ion, ear ly inbreed ing depression and genetic d ivers i ty i n a conifer , western redcedar (Thuja plicata D o n n ex D . D o n , Cupressaceae), u s ing t w o classes o f genetic markers: i sozymes and microsatel l i tes . I w i l l show h o w the history o f western redcedar has shaped the evo lu t ion o f these three l i n k e d quantities. T o set the stage for m y study I w i l l first r ev i ew patterns o f genetic d ivers i ty and ma t ing systems i n conifers and other trees. I w i l l then present data f rom previous studies on the genetic d ivers i ty , ma t ing system, eco logy and evolu t ionary history o f Thuja plicata and c lose ly related species. Conifers Patterns of genetic diversity in conifers - In general , conifers have h igher genetic d ivers i ty than other plants at the popula t ion l eve l , as measured b y the propor t ion o f p o l y m o r p h i c 4 l o c i (PPL), al leles per locus (A/L) and average expected heterozygosi ty (Hep \ Tab le 1.1; genetic d ivers i ty parameters are descr ibed i n A p p e n d i x I). In conifers , genetic var ia t ion resides mos t ly w i t h i n , rather than between, populat ions. A t the species l eve l , H a r d y - W e i n b e r g expected heterozygosi ty (Hes, W e i r , 1990) is not s igni f icant ly higher for gymnosperms (most ly c o n i f e r s ) ( / / „ = 0.169) than for either monocots (Hes = 0.159) or dicots (Hes = 0.184; H a m r i c k and God t , 1996). The total genetic d ivers i ty res id ing among popula t ions is s igni f icant ly l o w e r i n gymnosperms (G , , = 0.073) than i n other plants (monocots : Gsl = 0 .157; dicots: Gs, = 0.184; H a m r i c k and God t , 1996). I gathered data on mean levels o f popula t ion genetic d ivers i ty i n conifers f rom 68 i s o z y m e studies ( A p p e n d i x II). Because most studies sampled on ly part o f a species ' range, I dec ided to on ly report mean popula t ion measures o f genetic d ivers i ty (H e p, PPL, A/L) rather than divers i ty at the species l eve l (Hes). Measures o f w i t h i n popula t ion genetic d ivers i ty were lower for 13 species o f Cupressaceae, than for 35 species o f Pinaceae (Hep: t - test = 2.08, n = 49, P = 0.043; PPL: t - test = 2.18, n = 40, P = 0 .036; and A/L: t - test = 2.15, n = 40 , P = 0 .038; Tab le 1.1). Table 1.1 M e a n levels o f w i t h i n popula t ion i s o z y m e var ia t ion i n three species o f Thuja compared to mean levels i n gymnosperms and other plants. Species P O P L PPL A/L Hep Reference T. plicata 49 trees 9 0 1 0 Copes , 1981 T. plicata 8 15 15.8 1.17 0.039 Y e h , 1988 T. plicata 1 9 12 1.22 0.04 E l - K a s s a b y et al, 1994 T. occidentalis 6 18 37 1.5 0.094 Perry et al, 1990 T. occidentalis 6 " 11 13.9 1.17 0.034 Mat thes-Sears et al, 1991 T. occidentalis 6 20 54.2 1.6 0.129 L a m y et al, 1999 T. orientalis 14 26 57 1.89 0.144 X i e et al, 1992 Coni fe r s /V = 50 10.0 19.8 50.4 1.72 0.154 A p p e n d i x II Pinaceae N -36 10.7 20.8 54.0 1.78 0.163 A p p e n d i x II Cupressaceae N =13 7.4 17.1 42.3 1.59 0.124 A p p e n d i x II G y m n o s p e r m s N = 102 8.9 17.3 53.4 1.83 0.151 H a m r i c k et al, 1992 A l l plants N = 669 12.3 17.3 34.6 1.52 0.113 H a m r i c k et al, 1992 N, number o f species sampled; P O P , N u m b e r o f populat ions sampled; L , N u m b e r o f l o c i sampled; PPL, percent p o l y m o r p h i c l o c i ; A/L, mean number o f al leles per locus ; Hep, mean expected heterozygosi ty w i t h i n populat ions. See A p p e n d i x I for more details. 6 Review of outcrossing rates in conifers - Con i fe r s are p redominant ly outcross ing, however several species show moderate levels o f se l f ing (the propor t ion o f se l f - fer t i l ized offspring). I gathered data f rom more than 100 studies es t imat ing the outcross ing rate i n both natural and seed orchard populat ions o f conifers and angiosperm trees ( A p p e n d i x III). Tab le 1.2 summarizes the outcrossing rates for a total o f 52 species o f conifers f r o m eight genera and three fami l ies . The majori ty o f species for w h i c h data were avai lable were i n the Pinaceae, spec i f ica l ly i n the genus Pinus. T h e average outcross ing rate for a l l conifer species was 0.835 ± 0.171 S D . W h i l e most species o f conifers have outcrossing rates o f over 8 0 % , a quarter o f the species are b e l o w this value and show signif icant amounts o f inbreeding ( F i g . 1.2). O n e species, C h i h u a h u a spruce (Picea chihuahuana) is a lmost ent irely inbred, but this species is l im i t ed to a few s m a l l i sola ted populat ions ( L e d i g et al., 1997). T h e mean outcross ing rate for 32 angiosperm tree species is not s igni f icant ly different than for conifers (mean t = 0.896 ± 0.164 S D ; t - test = 1.619, d f = 82, 2- tai led P = 0.109; data i n A p p e n d i x III.) Table 1.2 M e a n popula t ion outcross ing rate (t ± S D ) by genus i n 52 species o f conifers (see A p p e n d i x III). Genus N u m b e r o f species M e a n t ± S D F a m i l y Abies 5 0.889 ± 0.066 Pinaceae Larix 4 0.821 ± 0.062 Pinaceae Picea 9 0.732 ± 0.285 Pinaceae Pinus 28 0.878 ± 0 . 1 2 4 Pinaceae Pseudotsuga 1 0.880 Pinaceae Tsuga 1 0.975 Pinaceae Thuja 3 0.597 ± 0 . 1 3 5 Cupressaceae Cunninghamia 1 0.902 Taxodiaceae 7 Number of species 14 12 10 8 6 • Thuja spp. • Other conifers 0 Picea chihuahuana 2 0 0 0.2 0.4 0.6 Outcrossing rate 0.8 Fig. 1.2 Distribution of outcrossing rates in 52 species of conifers. Data are in Appendix III. 8 Outc ross ing rate and genetic d ivers i ty : Est imates o f both genetic d ivers i ty and outcross ing rates were avai lable for 40 species o f conifers (Append ices II and III). A species' outcross ing rate and genetic d ivers i ty (Hep) were pos i t ive ly , but w e a k l y , correlated (r = 0.406, N = 40 , P = 0 .009; F i g . 1.3). Species w i t h a h igh self ing rate showed reduced genetic d ivers i ty ; however , several species w i t h l o w genetic d ivers i ty s t i l l main ta ined h i g h outcross ing rates. T h e corre la t ion remained s ignif icant when the outlier, Picea chihuahuana, was exc luded f r o m the analysis ( r = 0.387, N = 38, P = 0.015). These results suggest that h i g h levels o f inbreeding can reduce levels o f genetic d ivers i ty at the popula t ion l e v e l i n conifers . In species w i t h higher outcross ing rates, however , other factors may be cont r ibut ing to a reduct ion o f genetic d ivers i ty . Other species o f conifers w i t h l o w levels o f genetic d ivers i ty m a y also show a l i n k between inbreeding and a reduct ion i n genetic d ivers i ty . T w o species o f conifers that show almost no i s o z y m e var ia t ion, red p ine (Pinus resinosa) and Tor rey pine (Pinus torreyana), were not i nc luded i n this analysis because no estimates o f outcross ing were avai lable . In red pine, on ly four p o l y m o r p h i c l o c i have been observed out o f 64 sampled (Hep - 0 .002, based on 27 enzyme systems; F o w l e r and M o r r i s , 1977; A l l e n d o r f et al, 1982; S i m o n et al, 1986; M o s s e l e r et al, 1991). T h e l a ck o f early inbreeding depression i n 46 red p ine trees f o l l o w i n g con t ro l led selfed ( 7 1 % seed set) vs outcrossed (72% seed set) po l l ina t ions , suggests that se l f ing is potent ia l ly h igh i n natural popula t ions (Fowle r , 1965). M a t i n g sys tem informat ion l a c k i n g for Tor rey p ine as w e l l , w h i c h exists i n on ly two populat ions showing no i s o z y m e var ia t ion (Hep - 0 based on 59 l o c i ; L e d i g and C o n k l e , 1983; a l though the two populat ions are f i x e d for different al leles at one locus) . 0.30 0.25 I 0.20 0.15 ^ 0 10 -0.05 -0.00 T^vrientaUs • x Picea chihuahuana :?xgr;occidentals x T. plicata • 0,00 0.20 0.40 0.60 0.80 1.00 © u k r a s s i r i g f r a t e Fig. 1.3 Cor re la t ion between outcrossing rate (0 and mean expected heterozygosi ty (H e p) i n species o f conifers . Species o f Thuja axe ind ica ted b y crosses. 10 Na tu ra l popula t ions vs seed orchards: D a t a on outcross ing rates f r o m both natural and seed orchard popula t ions were avai lable for 13 species o f conifers (Table 1.3). Because seed orchards are des igned to m i n i m i z e se l f -pol l ina t ion , outcross ing rates are expected to be higher i n orchards than i n natural populat ions ( A d a m s and B i r k e s , 1991). Indeed, ten o f the 13 species o f conifers showed a higher outcross ing rate in seed orchards compared to natural populat ions . B u t overa l l there was no s ignif icant difference i n outcross ing rates between the two popula t ion types ( W i l c o x o n s ign rank test: T = 19.5, N= 13, l - t a i l ed P = 0.092). Table 1.3 Outc ross ing rates i n 13 species o f conifers w i t h estimates f rom both natural and seed orchard populat ions . References are i n A p p e n d i x III. Popu la t ion type: Na tu ra l O r c h a r d Abies procera 0.94 1 Larix decidua 0.809 0.852 Larix occidentalis 0.894 0.803 Picea abies 0.895 0.937 Picea glauca 0.855 0.931 Picea mariana 0.827 0.837 Picea omorika 0.84 1 Pinus caribaea 0.921 1 Pinus leucodermis 0.802 0.86 Pinus sylvestris 0.965 0.975 Pinus tabulaeformis 0.864 0.957 Pseudotsuga menziesii 0.886 0.874 Thuja plicata 0.715 0.32 A v e r a g e = 0.867 ± 0.016 S E 0.873 ± 0.050 S E 11 Inbreeding depression in conifers - Inbreeding depression (8) i n conifers can be expressed through reduced seed v iab i l i t y , reduced germinat ion , s lower g rowth rate and higher morta l i ty f o l l o w i n g self-fer t i l izat ion. Coni fe r s usua l ly show strong inbreeding depression expressed mos t ly at early l i fe stages (Char leswor th and Char leswor th , 1987; Sorensen, 1999). T h e propor t ion o f f u l l seeds after se l f -pol l inat ion (ws) relat ive to f u l l seeds cross-pol l ina t ion (wx) i n 17 species o f conifers (a l l i n the Pinaceae) averaged 3 9 % (5 = 1 - ws I wx= 0 .61, r ev i ewed i n K o r m u t ' a k and L i n d g r e n , 1996). In 10 species o f outcross ing conifers , inbreeding depression at the seed stage (5 = 0.58) was m u c h higher than du r ing germinat ion (8 = 0.09) or dur ing g rowth and reproduct ion (8 = 0.18) ( rev iewed i n H u s b a n d and Schemske , 1996). Mating system of conifers - In conifers, p re-pol l ina t ion mechanisms such as m o n o e c y (separate male and female cones) and d i c h o g a m y (separation i n t ime between po l l en shedding and female recept iv i ty) can reduce se l f -pol l inat ion (Richards , 1986). U n l i k e angiosperms, gymnosperms seem to lack early se l f - incompat ib i l i ty mechanisms occur r ing between po l l ina t ion and fer t i l iza t ion. H o w e v e r , po lyembryony , the presence o f several embryos w i t h i n an ovule , can potent ia l ly decrease the number o f inbred progeny i n conifers . F e m a l e gametophytes can possess several a rchegonia w i t h the same hap lo id maternal genotype, but fe r t i l i zed b y different p o l l e n parents. Severa l embryos begin to develop w i t h i n an ovu le but on l y one e mbr yo eventual ly survives . I f both outcrossed and selfed embryos are v iab le , outcrossed embryos c o u l d outcompete less fit inbred embryos wi th in the same ovu le , decreasing the propor t ion o f self-fe r t i l i zed progeny (Sorensen, 1982; Savo la inen et al, 1992). P o l y e m b r y o n y can also potent ia l ly decrease the number o f empty seeds i n species w i t h h igh inbreeding depression. I f both selfed and outcrossed embryos are found w i t h i n the same ovule , the death o f an inv iab le selfed e mbryo w i l l not necessari ly cause the abort ion o f a seed i f a v iab le e mbr yo is also present (Sorensen, 1982). 12 The genus Thuja (Cupressaceae) There are s ix species i n the genus Thuja wor ldwide , t w o i n N o r t h A m e r i c a (T. plicata and T. occidentalis) and four i n eastern A s i a (T. koriaensis, T. orientalis, T. standishii and T. sutchuensis; V i d a k o v i c , 1991). The popula t ion genetic structure o f three o f these species (T. plicata, T. occidentalis and T. orientalis) has been studied us ing i sozymes . M e a n levels o f i s o z y m e divers i ty i n T. plicata and T. occidentalis are l o w e r than i n other conifers , w h i l e genetic d ivers i ty i n T. orientalis is s imi l a r to other species o f coni fers /gymnosperms (Table 1.1; F i g . 1.3). A l l three species o f Thuja have a m i x e d mat ing system w i t h outcross ing rates among the lowest i n conifers ( F i g . 1.2). Thuja orientalis showed an average o f 7 5 % outcross ing i n natural populat ions ( X i e et al, 1991), w h i l e Thuja occidentalis showed 6 4 % outcross ing i n Onta r io popula t ions (Perry and K n o w l e s , 1990) and 2 9 % outcross ing i n Quebec popula t ions ( L a m y et al, 1999). T h e outcross ing rate i n a seed orchard popula t ion o f Thuja plicata was 3 2 % ( E l -K a s s a b y et al, 1994). Thus , the genus Thuja offers the opportuni ty to study the evo lu t ion o f sel f ing i n a predominant ly outcrossing group o f plants, the conifers . Western Red Cedar {Thuja plicata) Ecology of Thuja plicata - The range o f Thuja plicata D o n n ex D . D o n (Cupressaceae) extends a long the P a c i f i c coast o f N o r t h A m e r i c a f rom southeastern A l a s k a to northern C a l i f o r n i a , and i n the inter ior f rom east-central B r i t i s h C o l u m b i a into the panhandle o f Idaho and western M o n t a n a . T h e coastal and inter ior parts o f the range are essent ial ly isolated f rom each other, but a few stands occur between the Coas t Ranges and the S e l k i r k M o u n t a i n s near the southern border o f B r i t i s h C o l u m b i a ( M i n o r e , 1990). Reproduc t ion i n redcedar usua l ly starts at 20-30 years but can start as early as 10 years i n trees exposed to sunl ight ( M i n o r e , 1983). Trees can often l i v e over 1000 years ( M i n o r e , 1983). Vegeta t ive reproduct ion b y l ayer ing or root ing o f fa l len branches is c o m m o n i n mature stands i n Idaho (Parker and Johnson, 1988) and m a y lead to 13 c l o n a l clusters. Pure stands o f T. plicata are rare, and it usua l ly g rows i n mixed-species , uneven-aged stands and occurs at a l l stages o f forest succession ( M i n o r e , 1983). Genetic diversity in Thuja plicata - Measures o f genetic d ivers i ty show l o w var ia t ion w i t h i n and l o w differentiation among populat ions o f western redcedar. F i rs t , there is l o w var ia t ion i n relat ive amounts o f leaf o i l terpenes over the entire range o f western redcedar, but some m i n o r differences have been detected between coastal and inter ior populat ions (von R u d l o f f and L a p p , 1979; v o n R u d l o f f et al., 1988). Second , Copes (1981) found no i s o z y m e var ia t ion at nine l o c i i n trees f rom W a s h i n g t o n and Oregon , and Y e h (1988) and E l - K a s s a b y et al. (1994) found very l o w var ia t ion i n B r i t i s h C o l u m b i a populat ions (Table 1.1). O v e r a l l , o f 21 i s o z y m e l o c i studied, f ive showed some var ia t ion i n at least one popula t ion , and on l y one, Glucose-6-phosphate dehydrogenase (G6pd) , was var iable i n a l l s tudied popula t ions ( Y e h , 1988; E l - K a s s a b y et al, 1994). T h i r d , G l a u b i t z et al. (2000) screened ind iv idua l s f r om throughout the range o f T. plicata w i t h R F L P (restriction fragment length p o l y m o r p h i s m ) probes and found li t t le differentiat ion among regions. F i n a l l y , studies that have examined var ia t ion i n phenotypic traits have found either no difference ( B o w e r and D u n s w o r t h , 1988) or l o w differentiat ion among popula t ions (Rehfeldt , 1994). H o w e v e r , s ignif icant differences i n resistance to K e i t h i a b l ight , caused b y the fungus Didymascella thujina, have been found among popula t ions o f redcedar (J. R u s s e l l , pers. c o m m . ) Inbreeding depression in Thuja plicata - U n l i k e most other conifers , Thuja plicata has shown li t t le inbreeding depression dur ing early l i fe stages. In one study, two self -pol l inated c lones set more seeds per cone than four cross-pol l inated clones (Owens et al, 1990). In a r ev iew o f inbreeding depression i n plants, H u s b a n d and Schemske (1996) found that predominant ly outcross ing conifers showed a cumula t ive inbreeding depression over their l i fespan o f 5 = 0.67 w h i l e i n western redcedar, l i fe- t ime inbreeding depression was on ly 8 = 0.30 14 (J. R u s s e l l c i ted i n H u s b a n d and Schemske , 1996). M o s t o f the inbreeding depression i n western redcedar occur red at the seed stage (8 = 0.28) but was m u c h lower than i n other conifers at the same stage (8 = 0.58). M o r e recently, inbreeding depression o f 10% i n g rowth rate after 11 years has also been measured i n western redcedar (J. R u s s e l l , pers. c o m m . ) Species-wide bottleneck - A reduct ion i n a species ' genetic d ivers i ty can occur through a reduct ion in the effective popula t ion size, either through a large h i s to r ica l bott leneck or sma l l recurr ing bott lenecks dur ing co lon iza t ion . Y e h (1988) suggested that a reduct ion i n species-wide genetic d ivers i ty i n redcedar was caused by a bott leneck dur ing the last g lac ia t ion . Pa leobotan ica l records suggest that western redcedar exper ienced a severe popula t ion bot t leneck dur ing the last ice age and the species was l i m i t e d to the extreme southern part o f its present day range ( H e b d a and Mat thewes , 1984). There are two possible scenarios for h o w a popula t ion bott leneck can lead to a reduct ion i n inbreeding depression. (1) T h e loss o f rare recessive mutat ions, some h igh ly deleterious, can occur dur ing a bot t leneck a l l o w i n g a subsequent swi tch to inbreeding ( K i r k p a t r i c k and Jarne, 2000) . (2) D u r i n g a severe popula t ion bott leneck the reduct ion to on ly a few related ind iv idua l s necessari ly leads to mat ing between relat ives and self-fer t i l iza t ion for reproduct ive assurance, caus ing deleterious mutat ions to be purged. Because both a swi tch to inbreeding and a reduct ion i n inbreeding depression are so t ight ly l i n k e d it m a y be imposs ib le to separate these two events i n Thuja plicata i f they stem f rom a popula t ion bott leneck. Thesis Overview In Chapter two, I w i l l first present estimates o f outcross ing rates i n s ix natural populat ions o f western redcedar based on one p o l y m o r p h i c i s o z y m e locus . In Chapter three, I w i l l out l ine the methods used to deve lop microsate l l i te markers i n western redcedar. Because o f 15 the l o w p o l y m o r p h i s m i n redcedar i sozymes , a more var iable marker was required for finer-scale studies. Mic rosa te l l i t e s are h igh ly p o l y m o r p h i c , codominant and usua l ly neutral markers , m a k i n g them w e l l suited for mat ing system and popula t ion genetic studies. In Chapter four, I w i l l address eco log i ca l factors affecting outcross ing rates i n natural popula t ions o f western redcedar. I used microsatel l i tes to obtain estimates o f outcross ing for i n d i v i d u a l trees and pos i t ion o f cones w i t h i n the c r o w n o f trees. I w i l l a lso describe a new method us ing b u l k e d D N A to estimate outcross ing rates. In Chapter f ive , I w i l l present the results o f a con t ro l l ed po l l ina t ion study conducted to assess early inbreeding depression and the role o f p o l y e m b r y o n y i n r educ ing sel f ing. In general , self ing rates are est imated f rom germinated seedlings, and i n conifers these rates are usual ly m u c h lower than the actual se l f -pol l ina t ion rate. E a r l y inbreeding depression and pos t -pol l ina t ion mechanisms, such as embryo compet i t ion , can potent ia l ly decrease the number o f selfed seedlings compared to se l f - fer t i l ized ovules . In Chapter s ix , I w i l l out l ine patterns o f genetic d ivers i ty over the range o f western redcedar and present evidence for separate southern and northern refugia dur ing the last g lac ia t ion , suggest ing that a species-wide bot t leneck predated at least the last g lac ia t ion . In Chapter seven I w i l l present an estimate for the rate o f somatic mutat ions at neutral microsate l l i te l o c i i n western redcedar based on an observed mutat ion. Chapter eight h ighl ights the m a i n f indings o f the different studies and ties together genetic d ivers i ty , ma t ing system and inbreeding depression i n western redcedar. In this f ina l chapter, I discuss the impl i ca t ions o f m y results and out l ine further research needed i n western redcedar. 16 Chapter 2 The mating system in natural populations of western redcedar Introduction M a t i n g systems, inbreeding depression and genetic d ivers i ty are inex t r icab ly l i n k e d ( U y e n o y a m a et al, 1993; Harder and Barrett , 1996; Ho l s inge r , 1996). Species w i t h higher sel f ing rates tend to show lower genetic d ivers i ty and less inbreeding depression, and the evo lu t ion o f these quantities invo lves their close interaction (Char leswor th et al, 1990; Cha r l e swor th and Char leswor th , 1995). One d r i v i n g force i n their evo lu t ion is the l i fespan o f the o rgan ism. L o n g - l i v e d plants are expected to accumulate a h igher genetic l oad through somatic mutat ions, consequent ly leading to the evo lu t ion o f outcrossing. T h e large size o f trees also increases the chance o f se l f -pol l ina t ion through gei tonogamy, and consequently, the evo lu t ion o f mechanisms to prevent such accidental se l f ing (Barrett et al, 1996). Indeed, studies have found that conifers are predominant ly outcrossing (5 - 10% self ing; A d a m s and B i r k e s , 1991), have h i g h genetic var iab i l i ty (15 - 2 0 % i s o z y m e heterozygosi ty , H a m r i c k and G o d t , 1996) and h igh inbreeding depression (mean = 64%, H u s b a n d and Schemske , 1996). H o w e v e r , these conc lus ions are based upon studies restricted to m a i n l y one f ami ly o f conifers , the Pinaceae. In the Cupressaceae, genetic d ivers i ty has been measured i n on ly a few genera and outcross ing rates have on ly been estimated i n the genus Thuja. In conspicuous except ion to other conifers, species o f Thuja d i sp lay h i g h self ing and l o w gene d ivers i ty . Thuja orientalis showed 2 5 % self ing ( X i e et al, 1991) and 14% i s o z y m e heterozygosi ty ( X i e et al, 1992), w h i l e T. occidentalis showed between 36 and 7 1 % self ing (Perry and K n o w l e s , 1990; L a m y et al, 1999), and 3 - 13% heterozygosi ty (Perry et al, 1990; Mat thes-Sears et al, 1991; L a m y et al, 1999). L i k e w i s e , western redcedar {Thuja plicata D o n n ex D . D o n ) shows m u c h reduced i s o z y m e var ia t ion (4% - 6% heterozygosi ty) , w i t h on ly one o f 21 l o c i ana lyzed be ing appreciably p o l y m o r p h i c (Copes 1981; Y e h , 1988; E l - K a s s a b y et al, 1994). In a seed orchard 17 setting, T. plicata had a self ing rate o f 6 8 % , one o f the highest measured i n a conifer ( E l K a s s a b y etal, 1994). In conifers , seed orchards are expected to exhib i t higher outcross ing rates than their natural popula t ion counterparts ( M u o n a , 1988), or at least be s imi l a r to natural populat ions ( A d a m s and B i r k e s , 1991). In natural populat ions, western redcedar trees tend to occur at l o w e r densi ty and are larger, leading to more ge i tonogamy (Farris and M i t t o n , 1984), and i n contrast to seed orchards where ne ighbour ing trees are unrelated, potent ia l f a m i l y structure w i t h i n natural populat ions o f T. plicata c o u l d further inflate natural levels o f inbreeding. K n o w l e d g e o f the rates o f self ing i n natural populat ions, and their var ia t ion among popula t ions , w i l l shed l ight on this puzz l e o f the evo lu t ion o f self ing i n predominate ly outbreeding conifers . In this study I estimate outcross ing rates i n s ix natural populat ions o f Thuja plicata i n southwestern B r i t i s h C o l u m b i a us ing enzyme electrophoresis. T h i s is the first study to document levels o f self ing i n natural populat ions o f western redcedar. Materials and methods Sample collections - C o n e s were co l lec ted i n the fa l l o f 1996 f rom two natural popula t ions o f Thuja plicata near V a n c o u v e r , B r i t i s h C o l u m b i a : the M a l c o l m K n a p p Research Forest i n M a p l e R i d g e and at M o u n t S e y m o u r (Table 2.1). In 1997, cones were co l lec ted f rom three popula t ions on southern V a n c o u v e r Is land ( M o u n t Bren ton , Shawnigan , and Reinhar t L a k e ) and f rom two populat ions on the ma in l and o f B r i t i s h C o l u m b i a ( M o u n t S e y m o u r and E n d e r b y ) (Table 2.1). Popula t ions consis ted o f trees o f less than 80 yrs except at Ende rby , where trees were over 100 yrs o ld . In 1996, cones were co l lec ted f rom the l o w e r to m i d - c r o w n o f trees, w h i l e i n 1997 cones were co l lec ted f rom the m i d - to upper -c rown o f trees. I co l lec ted seeds f rom at least 12 trees f rom each popula t ion , but erratic germinat ion reduced the number o f fami l i es eventual ly assayed (Table 2.1). 18 Isozyme analyses - Seeds were extracted f rom cones and stored at 4 ° C for up to eight months. T h e y were germinated on wet fi l ter paper at r o o m temperature for approximate ly 10 days. A few days after germinat ion , seed tissue was g round us ing the buffer o f M i t t o n et al. (1979), and the extracts were then stored at - 8 0 ° C . O n l y the embryos f r o m the 1996 co l lec t ions were assayed, w h i l e tissues f rom the 1997 col lec t ions were assayed separately as embryos and megagametophytes. T h e latter tissue is hap lo id and is genet ica l ly iden t ica l to the maternal al lele passed to the embryo . A s s a y o f megagametophytes a l lows more accurate inference o f maternal genotype i n a l l progeny arrays, and more accurate inference o f sel f ing rate and paternity i n arrays descended f rom heterozygous mothers (R i t l and and E l - K a s s a b y , 1985). M a t e r n a l genotypes f rom the 1996 co l lec t ions were inferred f rom the progeny arrays. F o l l o w i n g Y e h and O ' M a l l e y (1980) and E l - K a s s a b y et al. (1994), I assayed for the Glucose-6-phosphate dehydrogenase (G6pd) i s o z y m e locus on a morphol ine-c i t ra te buffer ( p H 8.0). Other i s o z y m e l o c i were k n o w n , f rom previous studies, to be m o n o m o r p h i c or not suff ic ient ly informat ive (gene frequency p < 0.05) for mat ing system est imat ion ( E l - K a s s a b y et al, 1994; Y e h , 1988). In the absence o f more markers , and to reduce statist ical variance, I increased the number o f i nd iv idua l s per f a m i l y assayed for outcross ing rates. I assayed f rom about 500 to 7 0 0 seed progeny per popula t ion , a l though the 1996 co l lec t ions y i e l d e d fewer germinants for assay (Table 2.1). Data analysis - S ing l e locus estimates o f popula t ion outcross ing rates, a l le le frequencies and correlated matings were obtained us ing a vers ion o f M L T R (R i t l and , 1990a) that incorporated megagametophyte informat ion . Corre la ted paternity (r p ) is def ined as the propor t ion o f fu l l sibs among outcrossed sibs (Ri t l and , 1989). Because o f the l o w number o f trees sampled per popula t ion , parental f ixa t ion indices c o u l d not be estimated, and hence were constrained by the est imation program to equal zero. A s w e l l , because there was on ly one 19 marker locus , I c o u l d not j o i n t l y estimate both components o f correlated matings (the corre la t ion o f paternity and the correla t ion o f self ing). T h e latter was assumed to be zero. E r ro r s o f estimates were computed w i t h the bootstrap method. Outc ross ing and corre la t ion o f paternity estimates were cons idered s igni f icant ly different f rom one or zero when the 9 5 t h percenti le o f the bootstrap values d i d not over lap w i t h these values. T h e mean outcross ing rate o f a l l populat ions was obtained b y we igh t ing each estimate i n propor t ion to the inverse o f its statistical variance. A chi-square heterogeneity test was used to test whether outcross ing rates differed among populat ions . Results The gene frequency o f the most c o m m o n al lele at the G 6 p d locus ranged w i t h i n a remarkably nar row interval across a l l populat ions, f rom 0.52 to 0.59 (Table 2.1). B y contrast, outcross ing rates (f) showed w i d e var ia t ion among populat ions, ranging f rom 0.173 to 1.257 (Table 2.1). Est imates o f outcross ing were s igni f icant ly l o w e r than uni ty i n three populat ions: Research Forest (t = 0.173), M o u n t S e y m o u r (t = 0.826) and E n d e r b y (t = 0.771). Est imates o f popula t ion outcross ing rates differed s igni f icant ly f rom each other (%2 = 40 , d f = 6, p < 0.001). H o w e v e r , when the Research Forest popula t ion was exc luded outcross ing rates d i d not differ f rom each other (%2 = 5.45, d f = 5, p > 0.05). The average weighted outcross ing rate for a l l popula t ions was 0.715, w h i l e the unweighted average was 0.837. Est imates o f the corre la t ion o f paternity general ly d i d not differ f r om zero, except for the 1996 Research Forest popula t ion , w h i c h showed a h igh value o f 0.917 but w i t h a h igh attached error (it shou ld be noted that estimates o f correlated paternity are statist ically independent f r om the estimates o f self ing). T h e weighted average o f correlated paternity estimates across popula t ions was 0.025, w h i c h d i d not differ s igni f icant ly f rom zero. 20 Discussion Outcrossing rates - The mean estimate o f outcross ing I obta ined for western redcedar is among the lowest i n conifers . Est imates o f outcrossing i n most conifers are above 8 0 % and except ions inc lude Larix laricina (mean t = 72 .9%, K n o w l e s et al, 1987), Pinus maximinoi (t = 6 5 % , M a t h e s o n , 1989), Picea glauca (mean / = 7 3 % , Innes and R i n g i u s , 1990) and Picea rubens (mean t = 59 .5%, Ra jo ra et al, 2000) . L o w outcross ing rates occur i n both Picea chihuahuana ( 0 % and 15 .3%, L e d i g et al, 1997) and i n Picea martinezii (mean 5 6 % , L e d i g et al, 2000), but these species are restricted to sma l l , ext remely isolated populat ions . T h e mean outcross ing rate for Thuja plicata i n this study (t = 71 .5%) was s imi la r to T. orientalis (mean t = 7 5 % , range: 68 -8 1 % , X i e et al, 1991) and higher than i n T. occidentalis (mean t = 6 3 % , range: 51 - 7 5 % , Perry and K n o w l e s , 1990; mean t = 3 0 % , range: 24 - 3 3 % , L a m y et al, 1999). O n e popula t ion , the Research Forest seems to be an out l ier i n regards to its l o w self ing estimate and heav i ly inf luences the mean outcrossing rate for a l l populat ions. O v e r a l l , the estimates o f outcross ing rates I obtained for natural populat ions o f Thuja plicata were higher than I expected. In a l l populat ions , except the Research Forest , outcross ing rates were higher than i n the seed orchard study (t = 3 2 % ; E l - K a s s a b y et al, 1994). Factors affecting mating systems - O u r results, together w i t h the h i g h self ing rate found i n the seed orchard ( E l Kas saby et al, 1994), indicate that the mat ing sys tem i n western redcedar is quite lab i le , w i t h marked among-popula t ion var ia t ion, and probably m a r k e d among-tree var ia t ion . A poss ib le reason for the l o w e r outcross ing rate i n the western redcedar seed orchard is assortative mat ing , caused b y asynchrony o f recept ivi ty to po l l en a m o n g trees f rom different geographica l locat ions i n the orchard. In a Doug la s - f i r seed orchard E l - K a s s a b y et al (1988) found that trees that were receptive earl ier or later had higher sel f ing than the r ema in ing trees. 21 There are numerous factors that m a y affect the sel f ing rates o f western redcedar, and o f conifers i n general , descr ibed as f o l l ow s . (1) C l o n a l structure - Wes te rn redcedar probably exhibi t s c l o n a l structure, because o f extensive vegetative reproduct ion g i v i n g rise to groups o f genet ica l ly iden t ica l ramets, w h i c h can increase the rate o f inbreeding. (2) F a m i l y structure - L o c a l i z e d f a m i l y structure due to l i m i t e d dispersal can contribute to biparental inbreeding. M o r e than one p o l y m o r p h i c locus is needed to differentiate between self-fer t i l iza t ion and mat ing between relatives. (3) Tree size - In the populat ions w i t h the higher outcross ing estimates (Shawnigan , M o u n t Bren ton , Reinhar t ) , I co l lec ted seeds f rom young trees w i t h larger, mature trees nearby. In contrast, s ignif icant inbreeding was found i n E n d e r b y where seeds were co l l ec ted f rom older trees on ly . A higher outcross ing rate i n smal ler trees is expected because they produce less p o l l e n and therefore their po l l en constitutes a smal ler p ropor t ion o f the p o l l e n c l o u d and increases the chance o f be ing fe r t i l i zed by outcross po l l en . (4) C r o w n pos i t ion - Outc ross ing rates can also vary among different heights w i t h i n a tree. L o w e r outcross ing rates have been found i n l o w e r c rowns compared to upper c rowns i n D o u g l a s -f i r (Pseudostuga menziessii) and S i t k a spruce (Picea sitchensis)(Ovm and A d a m s , 1986; E l -K a s s a b y et al, 1986; Cha i su r i s r i et al, 1994). In this study, cones were co l lec ted f rom the upper -c rown o f trees i n most populat ions but seeds were co l l ec ted f r o m l o w e r branches i n the popula t ion w i t h the lowest outcross ing rate, the Research Forest . H o w e v e r , seeds co l l ec ted i n the same way i n the 1996 co l l ec t ion f rom M t . S e y m o u r d i d not also show l o w e r outcrossing. (5) Inbreeding depression - H i g h inbreeding depression at the seed stage (mean 5 = 0.58, H u s b a n d and Schemske , 1996) can lead to the h igh rates o f outcross ing observed i n conifers . A s seedlings are n o r m a l l y used to infer outcross ing rates, the selfed seeds that die are mi s sed i n the outcross ing rate estimate. H o w e v e r , T. plicata seems to l ack early depression at the seed stage 22 (Owens et al, 1990; H u s b a n d and Schemske , 1996), so m y estimates o f outcross ing are not as b iased b y this ear ly-act ing inbreeding depression as i n other conifer species. Correlation of paternity - V e r y few mat ing sys tem studies i n conifers report corre la t ion o f paternity estimates. In Pinus washoensis ( M i t t o n et al, 1997) and Tsuga mertensiana ( A l l y et al, 2000) , no s ignif icant correla t ion o f paternity was found. In two species, Picea martinezii ( r p = 0.389; L e d i g et al, 2000) and Abies borisii ( r p = 0.990; F a d y and W e s t f a l l , 1997), the h igh corre la t ion o f paternity estimates were attributed to the l o w number o f reproduct ive ind iv idua l s i n the popula t ion . In Larix occidentalis, correla t ion o f paternity estimates were s ignif icant i n high-densi ty populat ions ( r p = 0.062 and 0.104) but not s ignif icant i n low-dens i ty popula t ions ( r p = 0.001 and 0.02; E l - K a s s a b y and Jaquish, 1996), p robably because h i g h tree density l i m i t e d p o l l e n movement . O v e r a l l m y results showed no correla t ion o f paternity i n redcedar (mean weighted r p = 0.025) ind ica t ing that the outcrossed seeds w i t h i n a progeny array were fer t i l ized b y several different po l l en parents. The l o w outcross ing rate and h i g h corre la t ion o f paternity i n the Research Forest popula t ion suggest that trees m a y have been disproport ionate ly exposed to their o w n po l l en i n this popula t ion . In this study I have found signif icant amounts o f inbreeding i n natural populat ions o f western redcedar and var ia t ion i n outcross ing rates among populat ions . I have set up a f ramework for further studies on the mat ing system w i t h i n natural populat ions w i t h more p o l y m o r p h i c and informat ive genetic markers , microsatel l i tes ( O ' C o n n e l l and R i t l a n d , 2000; Chapter 3). a o o s 3 * T 3 Q. SO o o H 03 a CS > 4) -a 3 5) ^ a o 1) -a a o o OH OO oo T - H o i n so © co Os © Os CN © i n 00 © © CN CN 00 >n o CN o © © © © © © © © ' © r-T - H Os 00 CN o co oo © © © © © • T - H T - H Os i n © SO ^ H i n CN © © © 1 © © V © ©• © d " © i n Os o CN >n CN T - H r -© i n m T - « r -T - H T ^ CO CO T - H i n m CO SO © m © © © © © © © © • © © CO r -i—i 00 Os SO CN 00 00 so Os SO Os T - H r -r -r -i n CN CO 00 i n • t— © © ' © © ' © ' © ' T - H © © ' co T - H 00 © © © i n CO © CO 00 o ' N SO SO O © © © © © © O, CN co Os co >n oo SO U0 oo CO uo r -n CN T - H 00 00 Os Os 50-80 © oo 1 © >n o co i i n T—< © i n 1 T - H © i n 1 IT) T - H © © T—H A Os i r -o o • n S 0 ©s as o PH 43 I u CM1 o © SO 3 >> on B 3 O © © SO © i n o 00 so © © i n r -• i n CN r-in r - r- CN in T - H r - in in >n © •* © 0 o 0 o o o O CN CN CN CO 00 CN CN CN CN CN r-H CN T - H T - H T - H ^ T - H T ^ T - H o CN CN CO 00 © in CN CN CN >n in CO o o 0 o o o O Os OS Os 00 00 © 00 •si- >n - H 3 o a on B O B r-1 vi m t 4—» •*—> C3 OS OS s O a 3 o s s s 43 B 'S •e SP "8 43 u 8 T 3 JD 43 .SP B T3 OX) CO CN u 43 B OH _B a > '5b Q on 43 3 S3 u _3 > * 24 Chapter 3 Characterization of microsatellite loci in western redcedar Introduction Coni fe r s are among the most genet ica l ly diverse plants ( H a m r i c k and God t , 1996) and are predominant ly outcrossed (Barrett and Ecker t , 1990). In contrast, western redcedar (Thuja plicata D o n n ex D . D o n Cupressaceae) has shown l o w genetic d ivers i ty based on leaf o i l terpenes (von R u d l o f f and L a p p 1979), i s o z y m e l o c i ( E l K a s s a b y et al, 1994) and restr ict ion fragment length p o l y m o r p h i s m s (Glaub i t z et ah, 2000) . Popu la t i on outcross ing rates for western redcedar based on one i s o z y m e locus indica ted a m i x e d mat ing strategy i n this species ( E l -K a s s a b y et ah, 1994; O ' C o n n e l l et al, 2001 ; Chapter 2). Thuja plicata i s a widespread conifer found a long the west coast o f N o r t h A m e r i c a f rom southern A l a s k a to northern C a l i f o r n i a , and i n the inter ior f r om east-central B r i t i s h C o l u m b i a into northern Idaho ( M i n o r e 1990). T h e coastal and inter ior popula t ions are geographica l ly isola ted f rom each other and m a y be genet ica l ly differentiated. Es t ima t ing outcross ing rates i n plants usua l ly requires several p o l y m o r p h i c l o c i , however . In species w i t h l o w genetic d ivers i ty the l a ck o f i s o z y m e p o l y m o r p h i s m prevents us f rom obta in ing accurate estimates o f outcrossing rates w i t h this marker . Because microsatel l i tes are h igh ly p o l y m o r p h i c , codominant and usual ly neutral markers they are w e l l suited for mat ing sys tem studies. I des igned microsatel l i tes for T. plicata to study its popula t ion genetic structure and ma t ing system. Materials and Methods Clone development - Mic rosa t e l l i t e markers were isola ted f rom redcedar genomic D N A us ing modi f ica t ions o f b io t in-enr ichment strategies o f K i j a s and F o w l e r (1994). G e n o m i c D N A was digested w i t h H a e III, and i n d i v i d u a l fragments were l igated to double stranded 25 o l igonucleo t ide adapters ( M 2 8 , M 2 9 ) on their 5' and 3' ends, respect ively . A d a p t e d fragments were then denatured, hyb r id i zed w i t h 5' b io t in labeled ( T G ) 1 2 and enr iched b y select ion w i t h magnet ic s treptavidin affinity supports ( D y n a l M - 2 8 0 ) . B i o t i n selected genomic fragments were then amp l i f i ed us ing p r imer M 3 0 , and the resul t ing mix ture was cut w i t h E c o R I and l igated into standard c l o n i n g vectors ( p G E M 3 Z + , Promega) for propagat ion i n bacteria. I nd iv idua l microsatel l i tes con ta in ing clones were isolated by c o l o n y hybr id i za t ion w i t h P 3 2 - l a b e l e d ( A C ) 1 2 and p i c k e d into g l y c e r o l cultures for l ong term storage and i so la t ion . I sequenced 96 clones d i rec t ly f rom g l y c e r o l stocks us ing S e q u i T h e r m E X C E L ™ II L o n g - R e a d D N A Sequenc ing K i t s - L C (Epicentre Techno log ies ) on a L I - C O R 4200 sequencer ( L i n c o l n , Nebraska) . F r o m these, I chose 35 c lones to des ign microsate l l i te p r imer sets. In each p r imer pair , one o f the pr imers was ta i led (Table 3.1; Oet t ing et al, 1995). Screening for polymorphisms - T o t a l D N A f rom T. plicata fo l iage was isolated us ing a m o d i f i e d C T A B method ( D o y l e and D o y l e , 1987; A p p e n d i x I V ) . T o test for microsate l l i te p o l y m o r p h i s m I screened ind iv idua l s f r om one coastal popula t ion (southwestern B C , N = 22) and two inter ior populat ions (southeastearn B C , A^= 11; and northern Idaho, N = 11). P r e v i o u s l y isolated D N A samples for the inter ior populat ions were part o f another study (Glaub i t z et al, 2000) . Po lymerase cha in reactions ( P C R ) ampl i f ica t ions were per formed us ing 10 u\ total react ion vo lumes w i t h l x T a q buffer ( l O m M Tr i s , 1.5 m M M g C l 2 , 50 m M K C 1 , p H 8.3; Roche ) , 1 p m o l d N T P , 0.5 p m o l each o f fo rward and reverse pr imers , 0.5 p m o l M 1 3 I R D - l a b e l e d pr imer , l U n i t T a q D N A Polymerase (Roche) , and between 10-30 n g o f genomic D N A template. Samples were amp l i f i ed on a P T C - 1 0 0 thermocycler ( M J Research) denaturing at 95 °C for 5 m i n , f o l l o w e d b y 33 cyc les o f 95 °C for 45 s, anneal ing temperature (Table 3.1) for 45 s, 72 °C for 45 s and end ing w i t h one c y c l e o f 72 °C for 5 m i n . F o l l o w i n g ampl i f i ca t ion , 3 u\ o f load ing 26 dye (100% formamide , l m g / m l pararosanil ine basic red 9) was added to each react ion. F o r f ina l screening the microsatel l i tes were detected on a L I - C O R 4 2 0 0 sequencer w i t h a 7 % p o l y aery l amide ( L o n g R a n g e r ™ ) gels. Results and Discussion O f the 35 p r imer sets, 12 ampl i f i ed interpretable p o l y m o r p h i c l o c i . O n e o f these showed t w o l o c i for the pr imer pair ( T P 1 2 a & b ; Tab le 3.1). S i x l o c i were c o m p o s e d o f s imple d inucleot ide repeats, two o f c o m p o u n d repeats, and three had interrupted repeats. O n e o f the l o c i , T P 6 , i nc luded a hexanucleot ide repeat. Obse rved and expected heterozygosi t ies were higher for a greater number o f l o c i i n the coastal popula t ion than i n the inter ior popula t ion . There was a s ignif icant def ic iency i n heterozygotes at three l o c i ( T P 4 , T P 1 0 and T P 1 2 a ) i n the coastal popula t ion ind ica t ing possible n u l l a l leles . Because redcedar is k n o w n to self-fert i l ize i n natural populat ions, a heterozygote def ic iency can also be due to inbreeding. The number o f al leles per locus ranged f rom 3 to 36 for a total o f 189 alleles for the 13 l o c i . Together these l o c i have enough var iab i l i ty to conduct a detai led study o f the popula t ion genetic structure o f T. plicata. L o c i w i t h h igh number o f al leles w i l l be par t icular ly useful for mat ing sys tem studies i n this species. The 12 sequences w i t h interpretable microsatel l i tes have been deposi ted i n Genbank ( A p p e n d i x V ) . r - Os r -m r -r- o co in oo o m so m so d •§ S I a o o m o o o r -d ro r -r -d in in d O O CO r -d in in d o CN CN CN CN CN CN CN CN os Os CN CN SO d r -o Os CN O O CN CO 00 CO r -d so in oo CO CO oo r -oo CS o V Os no d CN CN in I / O Os d CN CN CO Os d o o SO o CN Os m d CN CN r -CN r -d CN CN r -d o r -d o CN SO CO CN SO CN SO Os co in CO CO I SO Os CO o CN I >n 00 CN I CO SO CN SO r-CN SO SO CN Os f-CN i >n CN CN CN I 00 CN CN Os CO CN m 00 c« CS I u S i u Q l 04 ' < a < o O H a a a < o < u o Os A u u u o o SO CN m in in in in in in oo >n m >n CN >n # O a a u u a §1 u u H u < U a < < < a u < < CD H a < < a H H U a < H U H n CM H so r - 00 OH H Si ^ ea o U 5 on oo oo oo O N © O N C N C N O N © >n i n O N © C N C N O C N N O © CO i T t T t C N o 1 CJ o | o s CJ •c OH < 00 i n a a H H < H H U E-i u < H H < <2 H a a b < u a O N H T t CO © © i n CO © C N O N IT) N O © CO i n CO C N r -00 N O U O o N O © r -C N r -C N C N © CO N O © O N r -i n C N i CO © CN U E-i U i n N O H a a a H H U a u u H < a CJ O < o u H U < 8 g < u o a < o < < < N O 00 © r -00 C N O N © O N d C N C O N O d O N C N C N ON < U >n N O H U H U a o a < < < < U O O < u H H < H U H < a a a u u H u H < u o u o P H H CN bZ H co C N C O O N N O in N O J O C N CU H oo C N o O D >> N o I i CJ X ) s 3 G O ^" i f \— 3 ej J 3 o c cj T3 -1—1 is 'S CJ cj "3 cj C c E-T •o CJ "S to ca =E cj o a CJ 3 CT1 CJ CJ > o 1 . O •s P H * , C/3 O 00 • & £ 2 o h o cj JS -a CJ > O T 3 CJ S 6 ca 3 T3 o •a ea B 3 u a . rv CJ ea ca o o £ • cj 'E8 • £ N o CJ T3 3 o ca CJ o cj cj O H X cj O H W ) tH CJ x> c "53 > f 29 Chapter 4 Fine-scale estimation of outcrossing in western redcedar with microsatellite assay of bulked DNA Introduction Weste rn redcedar (Thuja plicata D o n ex D . D o n n , Cupressaceae) shows quite h igh self ing rates for a conifer , w i t h estimates o f outcross ing i n natural popula t ions ranging f rom 17-100% (weighted mean = 71 .5%) ( O ' C o n n e l l et al, 2001 ; Chapter 2). The outcross ing rate for most coni fer species is above 8 0 % (52 species, mean 83 .5%; Chapter 1). W h i l e western redcedar produces abundant v iable selfed seeds (Chapter 5), inbred trees produced b y self-fer t i l iza t ion have shown signif icant reduct ion i n g rowth rate (ca. 10%) compared to non- inbred trees (J. R u s s e l l , unpubl i shed data). Because seeds for reforestation are n o r m a l l y co l lec ted f rom natural populat ions, knowledge o f var ia t ion i n sel f ing rates w i t h i n a tree is important for the co l l ec t i on o f seeds w i t h the highest expected outcross ing rate. In addi t ion , knowledge o f f ine-scale var ia t ion o f outcross ing rate w i l l enable deeper understanding o f the evo lu t ion o f factors med i a t i ng se l f ing rates (Barrett and Ecker t , 1990). The mat ing system o f conifers is d y n a m i c i n space and t ime, m a i n l y be ing affected by var ia t ion o f se l f -pol len ava i l ab i l i ty ( M i t t o n , 1992). P re -po l l ina t ion factors that can affect inbreeding rates i n conifers inc lude popula t ion density (Farr is and M i t t o n , 1984), c r o w n pos i t ion (Cha i su r i s r i et al, 1994; E l - K a s s a b y et al., 1986; O m i and A d a m s , 1986), f a m i l y structure and tree size ( M i t t o n , 1992). In redcedar both male and female cones are dis t r ibuted throughout the c r o w n o f trees and the l o w e r branches often reach the ground. H i g h e r sel f ing is expected in l o w e r branches or i n cones closest to the trunk because they are more l i k e l y to receive sel f -pol len f rom branches higher up i n the tree. L i k e w i s e , higher branches and branch tips are expected to receive more unrelated po l l en and show less self ing. In a previous study o f western redcedar, outcross ing estimates showed that l o w e r branches have higher levels o f se l f -pol l ina t ion than 30 higher branches ( K . R i t l a n d , unpubl i shed data). H o w e v e r , these were i n seed orchard populat ions , w h i c h may be unrepresentative o f natural populat ions . In another study, differences for outcross ing among natural populat ions suggested that populat ions w i t h larger trees showed l o w e r outcross ing rates ( O ' C o n n e l l et al, 2001 ; Chapter 2). Popu la t i on outcrossing rates are usual ly estimated us ing i s o z y m e markers, as i n O ' C o n n e l l et al. (2001) and Chapter 2, but i n western redcedar, on ly one i s o z y m e locus (glucose-6-phosphate dehydrogenase) o f 21 surveyed shows sufficient p o l y m o r p h i s m for es t imat ion o f outcross ing rate (Copes , 1981; Y e h , 1988; E l - K a s s a b y et al, 1994). M o r e l o c i are needed to study f iner-scale var ia t ion o f outcrossing rates w i t h i n a popula t ion . Severa l h igh ly var iable microsate l l i te l o c i have been deve loped for western redcedar, w h i c h should a l l o w us to conduct detai led studies o f var ia t ion o f sel f ing rate ( O ' C o n n e l l and R i t l a n d , 2000; Chapter 3). Mic rosa te l l i t e s can indeed be used to estimate outcrossing at different c r o w n posi t ions {Pinus densiflora, Pinacaeae; L i a n et al, 2001), to d iscr iminate between bi-parental inbreeding and self-fer t i l iza t ion (Caryocar brasiliense, Caryocaraceae; Co l l eva t t i et al, 2001) , and to study the var ia t ion i n sel f ing rates between immature and mature fruits (Shorea leprosula, Dipterocarpaceae; Nagami t su , 2001) . In the herbaceous plant, Zostera marina (Zosteraceae), outcross ing rates and relatedness among s ib l ings were estimated us ing microsatel l i tes (Reusch , 2000, 2001) . H o w e v e r , the disadvantage w i t h microsatel l i tes is that they are more expensive and labour intensive, compared to the less- informative i sozymes . T h e cost o f assay leads to a reduct ion i n sample s ize. F o r example , i n other microsate l l i te studies, seeds were sampled f rom o n ly one maternal tree i n Pinus densiflora ( L i a n et al, 2001) , f ive trees i n Shorea leprosula (Nagami t su , 2001) and outcross ing rates were estimated f rom a s ingle progeny per f ami ly i n Zostera marina (Reusch , 2000, 2001) . B y b u l k i n g offspr ing co l l ec ted f r o m the same loca t ion 31 w i t h i n a tree, the number o f D N A extractions and samples genotyped can be s igni f icant ly reduced, p r o v i d e d that al leles can s t i l l be detected i n b u l k e d samples. In this study I test for fine-scale var ia t ion o f outcross ing rates w i t h i n natural popula t ions o f Thuja plicata, at the l eve l o f i n d i v i d u a l trees and w i t h i n i n d i v i d u a l trees. T o increase exper imenta l e f f ic iency, I use a new est imat ion method based upon the b u l k i n g o f several i n d i v i d u a l progeny into one sample. T h i s method greatly increases the power to detect f ine-scale var ia t ion , as more ind iv idua l s can effect ively be i n c l u d e d for the same number o f genetic assays. T h i s me thod is then used to test for f ine-scale differences o f outcross ing rates w i t h i n popula t ions o f western redcedar, i n re la t ion to tree size and pos i t ion o f cones w i t h i n trees. Materials and Method Bulking tests - T o evaluate the feas ib i l i ty o f b u l k i n g D N A (extract ing D N A tissues f rom several i nd iv idua l s s imul taneously) , I performed three tests to assess whether a l l al leles c o u l d be detected, us ing samples b u l k e d either before or after D N A extract ion. F i rs t , I b u l k e d equal proport ions o f D N A separately extracted f rom three i nd iv idua l s w i t h k n o w n genotypes to test whether a l l al leles c o u l d be detected. Second , I b u l k e d D N A f r o m t w o ind iv idua l s i n 5:1 and 1:5 D N A ratios to test whether al leles occur r ing at a l o w frequency c o u l d be detected. T h i r d , I screened samples o f one, three or ten seedlings b u l k e d before D N A extract ion and i n this case the i n d i v i d u a l seedl ing genotype was not k n o w n but the maternal genotypes were obtained f rom separately screened h a p l o i d megagametophytes. I screened a l l i nd iv idua l s at four easi ly interpretable and robust microsate l l i te l o c i ( T P 1 , T P 3 , T P 9 , T P 1 1 , Chapter 3). P C R ampl i f i ca t ion and detection o f al leles on a L I - C O R 4200 ( L i n c o l n , Nebraska) were carr ied out as descr ibed i n O ' C o n n e l l and R i t l a n d (2000) and Chapter 3. T o score al leles, I used a p rogram that gives the intensi ty o f each band (Odyssey ver. 1.0.55, L I - C O R Inc., L i n c o l n , Nebraska) . T h e band intensity is equal to the s u m o f the intensi ty values 32 for a l l p ixe l s in an area covered by a band (integrated intensi ty) . T o test whether al leles were missed i n b u l k e d D N A samples compared to unbu lked samples, pa i red t-tests were used. Stat is t ical tests were performed us ing J M P vers ion 3.2.1 ( S A S Institute, 1997). Sample collections - In the autumn o f 1999, mature cones were co l lec ted f rom four southwestern B r i t i s h C o l u m b i a populat ions: three on eastern V a n c o u v e r Is land ( B C 1 2 , B C 1 4 and B C 1 5 ) and one on the ma in l and near Pember ton ( B C 1 3 ; see Chapter 6). S a m p l e d trees ranged i n height f rom 4.8 to 36.8 m , and were representative o f the height o f reproduct ive ind iv idua l s i n the populat ions . Cones were co l lec ted f rom up to three different c r o w n heights w i t h i n a tree: the top o f each tree, m i d w a y up the tree and f rom the lowermos t branches o f the tree. A t each height, cones were co l lec ted f rom two posi t ions: f r om the t ip o f the outer branches and f rom the hanging inner branches c loser to the trunk. In the three i s land populat ions , cones were co l lec ted f rom up to six posi t ions on each tree w h i l e i n the Pember ton popula t ion , seeds were co l lec ted f rom on ly two posi t ions (1 and 5) (Figure 4.1). Seeds were mechan ica l ly extracted f rom the cones and stored at 4 ° C . Seeds were germinated i n petri dishes on mois t fi l ter paper at r o o m temperature ( O ' C o n n e l l et al, 2001) . A few days after germinat ion the hap lo id megagametophytes and d i p l o i d seedlings were separated and p laced i n 1.5 m L microtubes . B u l k s o f ten megagametophytes were used to obtain maternal genotypes and bu lks o f three seedlings to obtain estimates o f outcross ing. Samples were kept at - 2 0 ° C un t i l D N A extract ion. E a c h sample was ground and D N A extractions were carr ied out us ing a m o d i f i e d C T A B method i n the microtubes ( D o y l e and D o y l e , 1987; A p p e n d i x I V ) . 33 Fig. 4.1 D i a g r a m o f s ix cone co l l ec t ion posi t ions i n a Thuja plicata tree: (1) top, outer branches (2) top, inner branches (3) m i d , outer branches (4) m i d , inner branches (5) lower , outer branches and (6) lower , inner branches. DNA assay of bulks - T o test for differences i n outcross ing among c r o w n posi t ions , two col lec t ions o f three b u l k e d seedlings each were screened f rom each pos i t ion . Seedl ings were screened at four l o c i : T P 1 , T P 3 , T P 9 and T P 1 1 . Megagametophytes were also scored at three addi t ional l o c i as part o f another study (Chapter 6). T h e total number o f seedlings sampled per tree ranged between 12 ( two posi t ions) and 36 (six posi t ions) . T o ensure that bands were u n i f o r m l y scored, I used an a l l e l i c ladder composed o f two to three ind iv idua l s w i t h al leles o f k n o w n sizes and spanning the range o f al lele sizes at a locus . T h e intensity and size o f microsate l l i tes were scored us ing the Odyssey software. T h i s extra caut ion i n scor ing bands was 34 needed, as many al leles are possible i n b u l k samples. F o r example , i n a b u l k o f three seedlings, up to four al leles can exist for a homozygous mother, and up to five al leles for a heterozygous mother. Estimation of outcrossing from bulk samples - P robabi l i t i es o f progeny, cond i t ioned upon maternal genotype, are the basic ingredients for the est imation o f mat ing systems. These are functions o f the popula t ion al lele frequencies and the outcross ing rate. T a b l e 4.1 gives these probabi l i t ies for cases o f s ingle progeny, two progeny and three progeny. T h e probabi l i t ies o f s ingle progeny are used i n the c lass ic method o f est imating outcrossing, w h i l e the two- and three-progeny probabi l i t ies represent the cases where two and three ind iv idua l s are bu lked , respect ively . 35 Table 4.1. P robabi l i t i e s o f band patterns observed for a s ingle progeny, and for b u l k e d progenies o f sizes 2 and 3, cond i t ioned upon maternal genotype (homozygous A , A , - o r heterozygous A , A y ) . N o t e : for outcross ing rate t = 1 - s, and for gene frequency pt, the f o l l o w i n g abbreviat ions for fo rmu la are used: (1) a,• = s + pf, (2) bt = pf, (3) c, = s/4 + pjt/2, (4) dt = ptt/2. B a n d s P r o b | A , A , B a n d s P r o b | A , A , 1 p rogeny (unbulked) 1 ai ij bj 2 progeny (bulked) •J 2apj+bf 2bjbk i,k or j,k l>] i,j,k or i,k or j,k i,j,k,l or i,k,l or j,k,l 3 p rogeny (bulked) i a? i i,j 'iafbj+'bapj+b- i,j i,j,k 6aibpk+?>bfbk+3bjbk2 i,j,k or i,k or j,k ij, k, I 6bjbkb, ij, k, I or i, k, I or j,k, I c, C,+Cj c,+2cicj+c/ 2Cjdk+2Cjdk 2dkdl c / + 3 c , 2 c y + 3 c , c / + c / 6cfjdk+3c12dk+3cJ2dk+3cjdk2+3cjdk2+dk 6cidkdl+6Cjdkdl+3dk2dl+3dkd2 ij, k, I, m or i, k, I, m or j, k, I, m 6dkdflm T h e procedure for est imating outcrossing then f o l l o w s the usual procedures, as for example , descr ibed i n R i t l a n d (1983). A computer p rog ram was wri t ten i n FORTRAN 95 b y K. R i t l a n d for es t imat ing outcross ing rate for the case o f bu lks o f size three. M a t e r n a l parentage was assumed k n o w n , and the bootstrap method was used to ascertain errors o f estimates. A l s o , another p rog ram was wri t ten to s imulate data, and hence to check the accuracy o f the est imat ion p rogram. Est imates indeed were obtained w i t h i n the statistical range o f the true values o f outcross ing used to generate the s imula ted data. R i t l a n d (2002) presented a m o d e l for est imating gene frequencies and heterozygosi t ies us ing b u l k e d samples. In that paper, a general p robabi l i ty was g i v e n us ing a "mask" funct ion, 36 but this notat ion is m u c h more d i f f icu l t w i t h these mat ing sys tem probabi l i t ies . In that paper, it was found that w i t h larger p o o l sizes, the higher non- l inear i ty o f the p robabi l i ty m o d e l resulted i n greater es t imat ion bias. L i k e w i s e , the s imula t ion p rogram d i d f i n d bias but the bias was not more than 5% o f the true va lue o f outcross ing rate, for the b u l k size o f three. Results Allele detection - In samples o f two and three seedlings that were b u l k e d after D N A extract ion a l l al leles were detected at each o f the four l o c i i n both the 1:5 and the 1:1:1 D N A ratios (results not shown) . Mic rosa t e l l i t e al leles t yp i ca l l y show addi t iona l non-a l l e l e l i c bands o f l o w e r intensity and smal ler size on gels. These stutter bands are the result o f s l ippage dur ing P C R ampl i f ica t ions . Tests per formed w i t h ind iv idua l s o f k n o w n genotypes showed that stutter bands were a lmost exac t ly h a l f the intensity o f the previous band and the intensi ty o f over lapp ing bands was addi t ive . U s i n g this informat ion , al leles can be disentangled f rom stutter bands and other al leles i n b u l k e d samples ( F i g . 4.2). T h e tests also showed that, for b u l k e d samples, the number o f copies o f a par t icular al lele c o u l d not be accurately inferred. T h i s is p robably because o f a l le le compet i t ion dur ing P C R reactions. B a n d s f rom larger size al leles are usual ly less intense than for shorter al leles (pers. obs.) 37 Locus TP9 Band intensity profile Fig. 4.2 B a n d intensity prof i le (right) o f three b u l k e d ind iv idua l s at locus TP9 f rom the lane 6 on the microsate l l i te gel (left). The four detected alleles are indica ted by b lack arrows on the band intensity prof i le . M , 10 b u l k e d megagametophytes f rom the maternal plant. S, three b u l k e d seedlings. Bulking test - T h e total number o f alleles detected for the same number o f seedlings was greater when samples were not b u l k e d (Table 4.2). Ma te rna l al leles and c o m m o n alleles in the popula t ion were detected i n both b u l k e d and non-bu lked samples. H o w e v e r l o w frequency alleles were probably missed i n b u l k e d samples. A l l e l e s at l o c i spanning a larger range o f al lele sizes were more l i k e l y to be detected because there was less band overlap. B u l k s o f three seedlings were chosen for outcross ing rate est imation because i n d i v i d u a l al leles were easi ly ident i f ied, yet b u l k i n g s t i l l p rov ided the advantage o f reduc ing the number o f samples o f D N A to extract and score. 00 co on o T 3 4) "3 4 2 4 ) 4 3 T3 T3 B 1) O = 3 cd x < Q 4) 1 B 1) 4 2 4 3 ^ T ^ .2. P H PH H S X> XI 4) -4-t B r ! 8 »i 2 '3 "£> 4 3 CM O 22 v-H CN 42 73 H 4) VH o o CTJ 60 ed -a 4) 60 _ B Xi 4) 4 ) 4 ) 1 3 B O H 60 _ B xi 4) 4 ) 4 ) 4 = 3 B X M ~3 X ro X ro co X CO X CO X X Os CO X CO X o o x o 1—I X 3 o o 4) 4 ) /-) T t T f i n CO CO CN 00 TT CN CN CN CN CN CN CN CN CN CN CN r-H <-H r-H CO CO Os Os OH OH OH OH OH OH H b« b« b-> b< b-< I-H CN *—i CN i—i CN CN ro CN CN CN CN CN © ro ro SO CN SO CN CN os SO SO CO i n CO * 0 3 1 -4 ) > 1) 4 3 43 'I _3 "o u ~G o >n o 0 V a. •a P H c In 4) 4 3 4 3 O 03 4 ) O 'a 60 U l OH •c o U l 4 ) PH 3 a 4) .3, X* 4) -83 I o o SH 4) L* O 0 a .TH XI -rt 4 ) 4 3 (fl 4) 1 3 * 4) 4 3 3 a 39 Genetic diversity - Mic rosa te l l i t e s were scored i n a total o f 2019 seedlings f rom 80 fami l ies i n four populat ions (20 famil ies /pop) . E x p e c t e d heterozygosi ty and inbreeding coefficients for each popula t ion based on seven l o c i were obtained f rom the megagametophytes (Table 4 .3 ; Chapter 6). Three o f the populat ions had inbreeding coefficients that were s igni f icant ly greater than zero. The four microsate l l i te l o c i used to score seedlings showed a large amount o f p o l y m o r p h i s m w i t h i n sampled populat ions . The number o f maternal al leles per popula t ion ranged f rom f ive to seven at locus T P 1 1 , and f rom 14 to 16 at locus T P 9 . A t each locus a lmost twice the number o f alleles were detected i n the seedlings compared to the maternal plants. Outcrossing rates - Outc ross ing rates d i d not s igni f icant ly differ among branch heights over a l l the populat ions (Table 4.4). In the three populat ions where cones were co l lec ted separately f rom inner and outer c rowns , outcross ing rates were higher i n inner branches compared to outer branches (6 o f 9 posi t ions) . H o w e v e r , the differences i n outcross ing rates between inner vs outer branches were not stat ist ically s ignif icant (paired t - test: 1.57, P = 0.077). O v e r a l l tree outcross ing rates decreased s igni f icant ly w i t h tree height (^=0.15; N =73; P = 0.0006; F i g . 4.3). A n analysis o f covar iance showed that overa l l the decrease i n outcross ing w i t h an increase i n tree height was h igh ly s ignif icant (F = 12.8; df= 1; P = 0 .0007) . B u t there was no difference among popula t ions i n the mean tree outcross ing rate (F = 0 .35; df = 3; P = 0.79) or for the slope o f outcross ing rate on tree height ( F = 0.19, df= 3, P = 0.90; F i g . 4.3). T h e mean outcross ing rate over a l l trees i n a popula t ion ranged f rom 6 6 % to 7 8 % (Table 4.5). 3 CM E-I ON CM H CO CM H CM o o o T t i -O C3 T3 0) •*-» CJ CJ -w CJ T3 JZ o l H CJ 1 3 G o X ) co cj i -3 C3 I X ) C J 3 cj o Tt ca H _B cj CJ O 60 a X ) cj cj 00 - M C O e CJ cd 6 <+H o a o a o 3 O H O O H C3 c OH 0 \ CM H CO CM H OH H =3 ^ CM H ON CM H T t CO O ^ 00 CO CO CN ^ >n oo r-o CN r - T t T t m CN CN o 1—I >n T t CO CN CN T t NO CO CO NO T t T t T f NO i—1 ^ M »™M r—t r- m NO NO o ^ M ON ON CN i—i ON NO 00 ON * * * CO ON NO CN o i n T t o ^ M o O d d d d i n ^ M i -H CN T t 00 00 d d d d o o o o CN CN CN CN r ^ T t ' V ^ M a a o CM CN 6 o o U U m a o cj CJ CM U o CM O 3 2 C J ON o CN O 00 o H c cj > C J C O a o •a CJ C O ai O 00 N o I M CJ ^ H CJ J 3 X ) CJ cj a, x CJ X> CJ On 00 _B xi CJ CJ O U i CJ 1 3 B to oo .fi cj CJ oo X ) CJ I C 3 C O C O CJ B CJ 3 3 O O o d V o, ©" A * o o £ ? cj O .-3 3 3 CJ 111 oo C O c/3 o a > CJ I/O B O -a CJ CO cd -o -*-» B °o "+H CD O O oo s '•3 CJ cj u-. O B O o o .B CJ -4—» co O hH o 42 1.0 0.8 L 0.6 L Outcrossing rate (t) 0.4 • o 0.2 0.0 •—Coombs Yellow Point ••• • Paldi ^Pember ton 0 10 15 20 25 30 Tree height (m) G 35 40 Fig. 4.3 Individual tree outcrossing rate estimates regressed on tree height in four populations of Thuja plicata. N = 13. 43 Table 4.5 M e a n tree heights and i n d i v i d u a l tree outcross ing rates (t) i n four popula t ions o f Thuja plicata. Popu la t i on M e a n tree height (range) t (range) S E C o o m b s 2 1 . 7 ( 1 4 . 0 - 3 5 . 5 ) 0.741 ( 0 . 2 7 - 1.00) 0 .040 Pember ton 20.2 (4.8 - 36.8) 0.778 (0.29 - 1.00) 0 .040 Y e l l o w Po in t 20 .2 (12 .1 - 3 4 . 5 ) 0 . 6 6 2 ( 0 . 2 2 - 0 . 9 8 ) 0.046 P a l d i 18.8 ( 8 . 3 - 3 2 . 8 ) 0,744 (Q.48 - 1.00) 0.033 Discussion Variation in outcrossing rates - Se l f i ng rates i n plants are usua l ly estimated us ing seedlings, and are l o w e r than the true se l f -pol l ina t ion rates because o f select ion against inbred embryos . Redcedar shows h igh self-fert i l i ty and the corre la t ion between selfed seeds and self-po l l i na t ion is c loser than for other conifers (Chapter 5). N o s ignif icant difference i n outcross ing was detected among c r o w n posi t ions . Fur thermore , a l l trends observed were i n the opposite d i rec t ion o f what was expected. Outc ross ing rates were higher i n the l o w e r branches vs the higher branches, and higher i n inner vs outer branches. U n l i k e these natural populat ions , trees i n a prev ious study i n a western redcedar seed orchard showed a s ignif icant decrease i n sel f ing i n higher branches compared to l o w e r branches ( K . R i t l a n d , unpubl i shed data). H i g h e r outcross ing i n upper c rowns vs l o w e r c rowns was also found i n S i t k a spruce (Picea sitchensis; Cha i su r i s r i et al., 1994) and Doug la s - f i r seed orchards (Pseudotsuga menziesii; E l - K a s s a b y et al., 1986; O m i and A d a m s , 1986). Factors i n seed orchards that differ f r om natural populat ions and that c o u l d enhance outcross ing inc lude h igher tree density, top p run ing and decreased f a m i l y structure. These factors c o u l d also contribute to the difference i n outcross ing among c r o w n heights observed the seed orchard. The 44 var ia t ion i n the number o f inbred seedlings among different posi t ions w i t h i n a tree is due m a i n l y to var ia t ion i n the amount o f self-pol len rece ived. In Thuja plicata, outcross ing rates decreased w i t h increas ing tree height i n a l l four populat ions. I nd iv idua l tree outcross ing rates var ied w i d e l y i n a l l populat ions (Table 4.5). L a r g e r trees were probably more l i k e l y to self-fert i l ize than shorter trees because their po l l en makes up a larger propor t ion o f the po l l en c loud . The tallest redcedar trees i n every stand towered above other trees. A l l four populat ions were s imi l a r i n terms o f tree height structure (Table 4.5). P h e n o l o g i c a l differences that occur w i t h tree s ize can a lso potent ia l ly affect outcrossing. In Doug las - f i r , shorter trees have a higher propor t ion o f male cones, and larger trees have more female cones so that outcrossing rate is expected to increase w i t h tree height ( M i t t o n , 1992). T h i s is opposite to the trend observed i n redcedar. N o data are avai lable on whether there is a change i n cone sex-ratio w i t h tree size i n redcedar. A l t e rna t ive ly , i f se l f ing is prevented b y a se l f - incompat ib i l i ty mechan i sm, a b reakdown i n this m e c h a n i s m w i t h tree age c o u l d exp la in the increase i n sel f ing in larger trees. A l t h o u g h there is no strong evidence for this scenario i t w o u l d be an interesting quest ion to pursue. Population outcrossing rates - U n l i k e a previous study conducted w i t h i sozymes ( O ' C o n n e l l et al, 2001 ; Chapter 2), there was no difference i n outcross ing rates among populat ions . In the current study, outcross ing rates were not s igni f icant ly different among populat ions and occurred w i t h i n a narrow range (t = 66.2 to 77 .8%) . In contrast to the i s o z y m e study, sampled trees in this study were more homogeneous i n size across a l l populat ions . I f se l f ing rates are constant over t ime the l eve l o f inbreeding should be reflected i n the inbreeding coefficient . There was no relat ionship between mean popula t ion outcross ing rates and inbreeding coefficients or genetic d ivers i ty . 45 Bulking samples - Est imates o f outcross ing obtained us ing b u l k e d D N A samples shou ld be l o w e r than when us ing un -bu lked samples for t w o reasons. F i r s t , al leles are more l i k e l y to be mis sed i n b u l k e d samples due to al lele compet i t ion dur ing P C R and over lapp ing stutter bands on gels. Second , s imula t ions showed a 5 % d o w n w a r d bias i n outcross ing estimates o f b u l k e d samples. Cor respond ing ly , i n the redcedar seed orchard study average outcross ing rates est imated us ing microsatel l i tes w i t h non-bu lked samples were about 5 % higher than those observed i n this study ( K . R i t l a n d , unpubl i shed data). L i k e w i s e , i n the i s o z y m e study outcross ing estimates were above 7 7 % for a l l popula t ions but one (5/6) ( O ' C o n n e l l et a l . , 2001 and Chapter 2). A t least, the compar i son o f trends i n e c o l o g i c a l factors affecting outcross ing rates are l i t t le affected b y systematic statistical bias. 46 Chapter 5 Polyembryony and early inbreeding depression in a self-fertile conifer, Thuja plicata (Cupressaceae) Introduction Inbreeding depression, expressed b y self - fer t i l ized progeny through reduced su rv iva l and reproduct ion, is a major d r i v i n g force i n the evo lu t ion o f plant mat ing systems (Lande and Schemske , 1985). Plants have e v o l v e d several pre- and pos t -pol l ina t ion mechanisms o f inbreeding avoidance. In general , conifers show very l o w seed set after se l f -pol l ina t ion ( rev iewed i n H u s b a n d and Schemske , 1986, and K o r m u t ' a k and L i n d g r e n , 1996), and traits f avor ing cross-fer t i l iza t ion are usual ly present. These inc lude asynchrony o f po l l en shedding and ovu le recept ivi ty (e.g. E r i k s s o n and A d a m s , 1990) and p h y s i c a l separation o f male cones f r o m female cones w i t h i n a tree (e.g. Pa rk and F o w l e r , 1984; O m i and A d a m s , 1986). U n l i k e most conifers , western redcedar (Thuja plicata D o n ex D . D o n n , Cupressaceae) has shown signif icant amounts o f inbreeding i n both orchards and natural popula t ions ( E l -K a s s a b y et al, 1994; O ' C o n n e l l et al, 2001 ; K . R i t l a n d , unpubl i shed data; C . N e w t o n , B . C . Research Inc., unpubl i shed data). Redcedar lacks the obvious inbreeding avoidance traits. P o l l e n shed and ovu le recept ivi ty over lap, and male and female cones are in te rming led w i t h i n a tree (Owens et al, 1990; E l - K a s s a b y , 1999). Thuja plicata a lso shows h i g h self-fert i l i ty, setting abundant seed after se l f -pol l ina t ion (Owens et al, 1990). Because pure stands o f western redcedar are rare and trees are often isolated f rom conspeci f ics ( M i n o r e , 1983), the species has p robab ly had to adapt to the l o w ava i l ab i l i ty o f unrelated po l l en and self-fert i l ize. A possible mechan i sm that can reduce the effects o f inbreeding i n conifers is p o l y e m b r y o n y , the occurrence o f mul t ip l e embryos w i t h i n an ovule . In conifers , several genet ica l ly ident ica l archegonia w i t h i n the same ovu le can potent ia l ly be sired b y po l l en grains f rom different parents. P o l y e m b r y o n y can potent ia l ly "rescue" ovules w i t h an inv iab le embryo . T h e abort ion o f an embryo does not necessari ly cause the abort ion o f the seed i f at least one 47 embryo w i t h i n the same ovu le is v iab le . T h e presence o f more than one v iab le e m b r y o per ovu le also offers the opportuni ty for compet i t ion between embryos and should favor those that are non -inbred. T h i s impl i e s that p o l y e m b r y o n y might increase ovu le success rates when embryos have l o w v i a b i l i t y and/or be a m e c h a n i s m to e l iminate se l f -pol l inated embryos (Sorensen, 1982; La t ta , 1995). W i t h o u t po l l en , redcedar cones start to deve lop but soon stop g r o w i n g and remain on the branches as sma l l , undeve loped cones; po l l en is required for the in i t ia t ion o f megagametophyte development . U n l i k e other conifers, redcedar cones w i l l deve lop w i t h on ly one fe r t i l i zed seed, and fer t i l i zed but aborted seeds are external ly indis t inguishable f rom v iab le seeds (Owens et al, 1990). Thuja plicata has archegonia l p o l y e m b r y o n y w i t h seven to n ine archegonia per ovule . O n average, two to three archegonia per ovu le are fe r t i l i zed b y different po l l en grains. E v e n t u a l l y one embryo becomes more mi to t i ca l ly active, and the other embryos i n the ovu le are aborted (Owens and M o l d e r , 1980). In western redcedar self-fer t i l izat ion can offer reproduct ive assurance when conspec i f ic po l l en is rare, and p o l y e m b r y o n y c o u l d favor outcross ing when a mix ture o f outcross and sel f -pol len is rece ived. P o l y e m b r y o n y is often suggested as a mechan i sm mi t iga t ing the effects o f self-fer t i l iza t ion and is incorporated i n theoretical models (Park and F o w l e r , 1984; C r o o k and F r i e d m a n , 1992; N a k a m u r a and Whee le r , 1992; Savo la inen et al, 1992; K a r k k a i n e n and Savo la inen , 1993; M o r g a n t e et al, 1993). H o w e v e r , there is a l ack o f e m p i r i c a l data testing this hypothesis . In this study, I used seed set f o l l o w i n g self- and cross-pol l ina t ions to obtain v i ab i l i t y estimates o f selfed vs outcrossed embryos i n western redcedar. F r o m these estimates, I ca lcula ted the propor t ion o f fu l l seeds and selfed seeds expected w i t h and wi thout p o l y e m b r y o n y f o l l o w i n g Sorensen (1982). I then used mixtures o f se l f and outcross po l l en to test the n u l l hypotheses that there is no difference i n seed set and propor t ion o f selfed seeds f rom that expected i n the absence o f p o l y e m b r y o n y . A n increase i n the propor t ion o f v iable seeds 48 (compared to expectat ions wi thout p o l y e m b r y o n y ) suggests that non- inbred v iab le embryos rescue ovules that also contain inv iab le inbred embryos . S i m i l a r l y , a decrease i n the propor t ion o f selfed seeds, indicates that outcrossed embryos are out -compet ing selfed embryos w i t h i n the same ovu le . Materials and Methods Pollinations -1 conducted a cont ro l led po l l ina t ion study at the M o u n t N e w t o n seed orchard i n Saanichton, B r i t i s h C o l u m b i a dur ing the win ter o f 1999. I chose four trees w i t h a large number o f both male and female cones. E a c h tree was genotyped at the G l u c o s e s -phosphate dehydrogenase (G6pd) i s o z y m e locus. Other i s o z y m e l o c i i n this seed orchard were either m o n o m o r p h i c or showed very l o w a l le l i c d ivers i ty ( E l - K a s s a b y et al, 1994). T w o o f these trees (181 & 395) were h o m o z y g o u s for the s low al le le " a " at this locus , one tree (432) was h o m o z y g o u s for the fast a l le le " A " and one tree (431) was heterozygous (Table 5.1). O n each tree 14 branches conta in ing a few hundred female cones were covered i n paper bags w i t h plast ic w i n d o w s i n mid-February , before they were receptive to po l l en . Be fo re bagg ing the branches, a l l male s t robi l i were removed . Other branches, w i t h po l l en cones, were co l l ec ted and p laced on d r y i n g racks at 1 5 - 1 7 ° C and 6 0 - 7 0 % relat ive humid i ty for two to three days un t i l they released po l l en . The co l lec ted po l l en was then stored at 4 ° C i n plast ic v ia l s for up to three days. M i x t u r e s o f po l l en f rom t w o trees were measured i n a graduated cy l inder , p l aced i n sma l l plast ic squeeze bottle w i t h an attached nozz l e , and m i x e d by shaking . 49 Table 5.1 E x p e r i m e n t a l des ign o f a po l l ina t ion exper iment i n four trees i n a Thuja plicata seed orchard. O n e hundred percent se l f -pol len (selfed) and 0 % sel f -pol len (crossed) treatments as w e l l as po l l en mixtures w i t h three different ratios o f self/cross po l l en ( 2 5 % / 7 5 % ; 5 0 % / 5 0 % ; 75%/25%) were appl ied to each tree. E a c h "selfed" treatment was per formed on three branches per tree and a l l other po l l ina t ion treatments were performed on one branch per tree. P o l l e n parents 431 ( A a ) F E M A L E P A R E N T (genotype at the G 6 p d locus) 4 3 2 ( A A ) 181 (aa) 395 (aa) 431 Se l fed Cros sed Cros sed Crossed 4 3 2 Cros sed Se l fed Cros sed Crossed 181 Cros sed Cros sed Se l fed Crossed 395 Cros sed Cros sed Cros sed Se l fed 25/75 25/75 431/181 50/50 50/50 75/25 75/25 25/75 25/75 432/181 50/50 50/50 75/25 75/25 25/75 25/75 431/395 50/50 50/50 75/25 75/25 25/75 25/75 432/395 50/50 50/50 75/25 75/25 50 O n each tree, 12 different treatments' were r andomly assigned to bagged branches w h i l e the two rema in ing branches served as controls . E a c h tree rece ived three se l f -pol len treatments and three outcross-pol len treatments (one f rom each o f the three other parent trees). P o l l e n mixtures w i t h three different ratios o f se l f to outcross po l l en were also app l i ed (Table 5.1). E a c h treatment was performed two to three t imes i n early M a r c h after female cones on a branch had become receptive as indica ted b y the presence o f po l l ina t ion drops on the m i c r o p y l e ( C o l a n g e l i and O w e n s , 1990). T o pol l ina te a branch, a sma l l hole was made i n the i so la t ion bag, the n o z z l e o f the squeeze bottle was inserted and a few puffs o f po l l en were b l o w n i n the bag. T h e bag was shaken to distribute po l l en evenly , and the hole was sealed w i t h tape. In June the paper bags were r e m o v e d and replaced w i t h mesh bags to protect the cones f rom insect damage. In late summer, once the cones were dry and had turned b r o w n , the branches were r e m o v e d f rom the trees. Seed viabUity - T o determine seed set for each treatment, 100 seeds f rom each branch were cut i n ha l f w i t h a razor blade, b l i n d to the identi ty o f the treatment. O n l y po l l ina ted ovules i n redcedar deve loped into seeds (Owens et al, 1990). F u l l seeds had a p l u m p whi te embryo , w h i l e aborted seeds had a shr iveled empty, b r o w n embryo . I use the term inbreeding depression to descr ibe the decrease i n self-fert i l i ty even though this c o u l d be due to late ac t ing self-i ncompa t ib i l i t y mechanisms (Seavey and B a w a , 1986). F o r each tree, /, inbreeding depression (5) at the seed stage was calcula ted as: b = l-(SJSJ where Sai is the propor t ion o f f u l l seeds for the se l f po l l ina t ion treatment, and Sci is the propor t ion o f fu l l seeds for the 100% outcrossed po l l en treatment. A n analysis o f covar iance ( A N C O V A ) was used to test for the effect o f maternal tree and po l l ina t ion treatment on the propor t ion o f f u l l 51 seeds. T h i s statistical analysis as w e l l as the others reported b e l o w were per formed us ing J M P (vers ion 3.2.1, S A S Institute, 1997). T h e propor t ion o f f u l l seeds part ly depends on the propor t ion o f se l f -pol len rece ived and inbreeding depression. P o l y e m b r y o n y c o u l d increase the number o f v iab le seeds i f v iab le embryos rescue ovules that also conta in inv iab le embryos . T h e expected propor t ion o f f u l l seeds for each tree, i, and treatment, k, i n the absence o f p o l y e m b r y o n y was ca lcula ted as: fit = ( l-Pk)Sci + Pk(Sai) where p is the propor t ion o f se l f -pol len i n the treament k, and Sc and Sa are values estimated for tree i. W i l c o x o n s ign-rank tests were used to test whether there was an increase i n the propor t ion o f observed f u l l seeds compared to the expected (/). Embryo competition - In m i x e d - p o l l e n treatments, enzyme electrophoresis was used to genotype the offspr ing at the G 6 p d locus . F o r each treatment, 100 seeds, when avai lable , were germinated and genotyped (b l ind to the identi ty o f the parent and treatment) u s ing the methods descr ibed i n O ' C o n n e l l et al. (2001) and Chapter 2. W h e n both parents i n the po l l en mix ture were h o m o z y g o u s for alternative alleles, the offspr ing were h o m o z y g o u s for the parental a l le le w h e n selfed, or heterozygous when outcrossed. In the heterozygous tree 4 3 1 , the propor t ion o f selfed seedlings, s, was estimated as s = 2 N A A / ( N A A + N a a ) where N A A is the number o f seedlings w i t h the " A A " genotype and N a a , the number o f seedlings w i t h the "aa" genotype. The paternal trees i n this case a lways had the "aa" genotype (Table 5.1). W h e n tree 431 was used as the c ross-pol len parent i n m i x e d po l l ina t ions , the propor t ion o f selfed seedlings was estimated as 5 = l - ( N A a / < ? ) 52 where N A a is the number o f heterozygous offspr ing and q is the propor t ion o f the alternative a l le le i n the outcross po l l en p o o l (i.e. q is f i x e d at 0.5). In this case, the maternal trees had an "aa" genotype. Expected seed set and selfing with polyembryony - T h e propor t ion o f f u l l and selfed seeds expected w i t h one, two or three embryos per ovu le were also ca lcula ted f o l l o w i n g Sorensen (1982). T h i s m o d e l incorporates estimates o f both selfed and outcrossed embryo v iab i l i t i es as w e l l as the probabi l i ty o f ovules conta in ing different combina t ions o f selfed and outcrossed embryos . Est imates o f selfed-embryo v iab i l i t i es were ca lcula ted as ai~\-(\-SaiY where Sai is the propor t ion o f v iab le seeds set i n tree i after con t ro l l ed sel f -pol l inat ions and n is the number o f embryos i n an ovule . Outcrossed embryo v iab i l i t i es were ca lcula ted as q = l - ( l - 5 d ) " where Sci is the propor t ion o f v iab le seeds after cont ro l led cross-pol l ina t ions (Sorensen, 1982). T h i s m o d e l assumes that i f an ovu le contains at least one v iab le embryo , it w i l l set a seed. A n ovu le contains n embryos f rom four different classes w i t h probabi l i t ies Pu P2, P3 and P 4 . T h e four classes w i t h their probabi l i t ies are: selfed and v iab le ( P , = apk), selfed and non-v iab le (P2 = (1 - a) pk), outcrossed and v iab le ( P 3 = c (l-pk)), and outcrossed and non-v iab le ( P 4 = (1 - c) (1 -pk)). T h e array o f embryo classes are m u l t i n o m i a l l y dis tr ibuted such that P(N{ = nv...,N, = nA) = x p,"' ...p* n,! . . .n 4 ! where /V„ N2, N3 and N4 are the number o f each respective embryo class and N, + N2 + N3 + N4 = n, the number o f embryos per ovu le . T h e propor t ion o f fu l l seeds expected i n a tree (ft), for each value o f n is equal to the probabi l i ty o f an ovu le conta in ing at least one v iab le embryo (Table 5.2). T h e propor t ion o f selfed seeds expected (s,) w i l l be equal the propor t ion o f ovules w i t h selfed seeds d i v i d e d b y the 53 propor t ion o f f u l l seeds. W h e n outcrossed embryos a lways outcompete selfed embryos w i t h i n an ovu le , on ly ovules that contain at least one v iab le selfed emb r y o and no v iab le outcrossed em b ryo w i l l g ive rise to selfed seeds (Table 5.2). I f selfed and outcrossed embryos have an equal chance o f o c c u p y i n g an ovu le (chance), then the increase i n the expected self ing rate depends on the propor t ion o f ovules w i t h both v iab le selfed and v iab le outcrossed embryos . W h e n the w i n n i n g e mbryo is determined by chance and when n = 2 the expected increase i n the propor t ion o f embryos g i v i n g rise to selfed seeds is l/2(2PlP3), and the increase i n the propor t ion o f embryos w i t h selfed seeds w h e n n = 3 is 2 / 3 ( 3 P , 2 P 3 ) + 1 / 2 ( 6 P , P 2 P 3 ) + 1 /3 (3P ,P 3 2 ) + 1 / 2 ( 6 P , P 3 P 4 ) (Table 5.2). Table 5.2 P robabi l i t i es o f setting a fu l l seed (f,) and setting a selfed seed (s,) w i t h one, two or three embryos per ovu le (n). Outcross wins , the ouctrossed embryos a lways outcompetes the selfed embryos w i t h i n an ovu le . Chance , both selfed and outcrossed embryos have equal chance o f o c c u p y i n g an ovu le . Abbrev i a t i ons o f probabi l i t ies are i n the text. P ropor t ion o f fu l l seeds expected: n = 1 / = P , + P 3 n = 2 / = 1 - ( P 2 2 + 2 P 2 P 4 + P 4 2 ) n = 3 / = 1- ( P 2 3 + P 4 3 + 3 P 2 2 P 4 + 3 P 2 P 4 2 ) P ropor t ion o f selfed seeds expected: n = l 5 , = P 1 / ( P 1 + P 3 ) Outcross w i n s n = 2 s, = [ P , 2 + 2 P , P 2 + 2 P , P 4 ] / [1- ( P 2 2 + 2 P 2 P 4 + P 4 2 ) ] n = 3 s,= [ P , 3 + 3 P , 2 P 2 + 3 P , 2 P 4 + 3 P , P 2 2 + 3 P , P 4 2 + 6 P , P 2 P 4 ] / [1- ( P 2 3 + P 4 3 + 3 P 2 2 P 4 + 3 P 2 P 4 2 ) ] Chance : n = 2 s,. = ( P , 2 + 2 P , P 2 + 2 P , P 4 + P , P 3 ) / [1- ( P 2 2 + 2 P 2 P 4 + P 4 2 ) ] n = 3 px+ 3pi2p2 + 3 P i 2 p 4 + 3pA2 + 3pip42 + 6PAP4 + 2pi2p3 + 3 / j P 2 P 3 + PrP 3 2 + 3 P t P 3 P 4 " 1 - ( P 2 3 + P 4 3 + 3 P 2 2 P 4 + 3 P 2 P 4 2 ) 54 W i l c o x o n s ign-rank tests for each treatment were used to test whether the observed propor t ion o f selfed seedlings differed f rom the propor t ion expected when n = 1. F igu re 5.1 illustrates the expected number o f self-fer t i l ized seedlings for n = 1, 2 or 3, w i t h no inbreeding depression caus ing the death o f an embryo , when (1) an outcrossed e mbr yo a lways outcompetes selfed embryos or (2) selfed or outcrossed embryos have equal chances o f f i l l i n g a seed. A n A N C O V A was used to test for a difference i n slopes for the propor t ion o f expected vs observed selfed seeds w i t h different proport ions o f self -pol len. Fitness of self-pollen - Se l f -po l l en can pe r fo rm differently depending on its frequency i n the po l l en p o o l . T h e fitness o f se l f -pol len relat ive to outcross po l l en was ca lcula ted as: w =PkWs + (l-pk)w0 where, w is the total fitness, pk is the propor t ion o f se l f -pol len appl ied and 1 - pk, the propor t ion o f outcross-pol len appl ied . The propor t ion o f selfed seeds observed, s, is therefore: s=PkWs/\pkws + (l-pk)w0] T o obtain the fitness o f se l f po l l en , vv,, relat ive to outcross po l l en , let w0 = 1 s=PkwJ\pkws + {\ -pk)] w h i c h when so lved for the relative fitness o f se l f po l l en , w s , g ives w s = s ( l -pk)lpk(\ -s) 55 1.00 co 0.80 T3 0.60 n = 1 n = 2, chance n = 2, outcross wins n = 3, chance n = 3, outcross wins / o 0.40 o Q . 2 °- 0.20 0.00 0.00 0.20 0.40 0.60 0.80 Proportion of self pollen 1.00 Fig. 5.1 The proportion of selfed seeds expected with different proportion of self-pollen and numbers of embryos (n) within an ovule. Two different outcomes are shown: the outcrossed embryo always outcompetes the selfed embryo (outcross wins) or self and outcross embryos are equally competitive (chance). All embryos are viable within an ovule (no inbreeding depression). 56 Results Seed set - In bags rece iv ing no po l l en , cones began to develop but remained s m a l l and no seeds were produced, ind ica t ing a l o w chance o f po l l en contaminat ion . S o m e o f the branches were b roken b y w i n d over the summer, and therefore some replicates are mi s s ing . A total o f 4954 seeds were checked for an embryo . Three o f the trees, 181, 395 and 432 , set a large number o f fu l l seeds under a l l po l l ina t ion treatments and inbreeding depression was around 3 0 % ( F i g . 5.2; T a b l e 5.3). H o w e v e r , i n tree 431 the propor t ion o f f u l l seeds decreased strongly w i t h an increas ing amount o f se l f -pol len, and inbreeding depression at the seed stage was 9 3 % . A n analysis o f covar iance showed that the propor t ion o f f u l l seeds va r i ed among trees ( F = 7.88, df= 3,P = 0 .0009) and among treatments (F = 5.68, df= 4,P = 0 .0025), but the interaction between tree and treatment was not s ignif icant ( F = 1.75, df = 12, P = 0.12). O n average, trees set 7 8 % o f their seeds when three o f the trees (181, 395, 431) were used as po l l en parents i n the 100% outcrossed treatments. Branches pol l ina ted w i t h tree 4 3 2 as an outcross po l l en parent set on ly 5 8 % o f their seeds. The difference i n s i r ing success among trees was not s ignif icant ( F = 1.872, N = 10, df= 3, P = 0.24). H o w e v e r , the power to detect a difference was l o w (Power = 0.28). There was no associat ion between the propor t ion o f f u l l seeds observed and the outcross po l len parent used i n m i x e d pol l ina t ions . A l t h o u g h tree 431 showed very l o w self-fert i l i ty, it d i d not show a decreased s i r ing success i n outcross and m i x - p o l l e n treatments when used on unrelated trees. 1.00 0.80 % ^ 0 . 6 0 . .3 o 0.40 o Q . 2 CL 0.20 0.00 i J ? \ V \ \ V • • ; . « ^ J '^i —^—181 --«--395 . - • -431 x 432 \ • \ 0.00 0.25 0.50 0.75 Proportion of self pollen 1.00 Fig. 5.2 The proportion of full seeds obtained in four Thuja plicata trees with different proportions of self-pollen applied. N = 495. 58 Table 5.3 The proportion of full seeds (SE) and inbreeding depression at the seed stage in four western redcedar trees (181, 395, 431 and 432). N, number of seeds sampled. Outcross treatment Self treatment Tree N Full seeds - Sa N Full seeds - Sa Inbreeding Depression 181 300 0.707 (0.026) 200 0.515 (0.035) 0.27* 395 100 0.870 (0.034) 300 0.600 (0.028) 0.31* 431 300 0.753 (0.025) 168 0.051(0.015) 0.93* 432 300 0.710(0.026) 300 0.527 (0.029) 0.26* *Inbreeding depression was significantly different from zero. P < 0.05. The proportion of full seeds expected, with different numbers of embryos within an ovule, are shown in Table 5.4. The increase in the expected number of seeds with an increase in n, is largest when 50% self-pollen is received. This is when the proportion of ovules with both selfed and outcrossed embryos is at its highest. The tree with the highest inbreeding depression, tree 431, shows the greatest increase in expected seed set between ovules with one and multiple embryos. Only one tree, 432, showed an increase in full seeds as expected after mixed pollinations if polyembryony contributes to increasing seed set (Wilcoxon sign rank test, n-6,T = 9.5, P = 0.031; Fig. 5.2). The three other trees showed no increase in seed set. None of the treatments (pooled over all four trees) showed a significant increase in full seeds with mixed pollinations compared to the seed set expected without polyembryony (Table 5.5). 59 Table 5.4 The propor t ion o f f u l l seeds expected (£) based on the number o f embryos per ovu le (n) and the propor t ion o f sel f -pol len appl ied (pk) i n four Thuja plicata trees w i t h v a r y i n g levels o f embryo v i ab i l i t y . N u m b e r o f Propor t ion o f se l f -pol len Tree embryos (n) 0 0.25 0.50 0.75 1 181 1 0.707 0.659 0.611 0.563 0.515 2 0.707 0.664 0.617 0.568 0.515 3 0.707 0.665 0.619 0.569 0.515 395 1 0.870 0.803 0.735 0.668 0.600 2 0.870 0.816 0.753 0.681 0.600 3 0.870 0.820 0.760 0.687 0.600 431 1 0.753 0.578 0.402 0.227 0.051 2 0.753 0.620 0.459 0.269 0.051 3 0.753 0.633 0.478 0.286 0.051 4 3 2 1 0.710 0.664 0.619 0.573 0.527 2 0 .710 0.668 0.624 0.577 0.527 3 0.710 0.670 0.626 0.578 0.527 60 Table 5.5 T h e propor t ion o f fu l l seeds expected (ft) wi thout p o l y e m b r y o n y (when n = 1) and observed f u l l seeds for three different proport ions o f se l f -pol len i n four Thuja plicata trees. N, number o f treatments. P ropor t ion o f F u l l seeds W i l c o x o n s ign-rank test se l f -pol len fa) N E x p e c t e d Obse rved ( S E ) Dif ference T P (2-tailed) P ( l - t a i l e d ) 0.25 8 0.676 0.591 (0.067) -0.085 8 0.30 0.16 0.50 8 0.592 0.660 (0.058) 0.068 7 0.25 0.15 0.75 7 0.548 0.598 (0.096) 0.05 4 0.46 0.27 A l l treatments 23 0.607 0.617 (0.410) 0.01 9 0.79 0.40 Realized selfing rates - A total o f 2553 seedlings were genotyped at the G 6 p d locus . Tree 431 y i e l d e d few germinants, resul t ing i n a large error o f se l f ing estimates. Therefore tree 431 was exc luded f r o m further analyses. F o r the r ema in ing three trees, the propor t ion o f selfed seedlings was s igni f icant ly larger i n the 7 5 % self -pol len treatment than i n the 2 5 % self -pol len treatment ( F i g . 5.3). T h e 5 0 % sel f -pol len treatment d i d not s igni f icant ly differ f r om the other two treatments (Tukey test over a l l treatments: TV = 18; F = 5.86, P = 0.013). T h e propor t ion o f selfed seeds was l o w e r than expected i n the 5 0 % and 7 5 % sel f -pol len treatments. In the 2 5 % self -pol len treatment there were more selfed seeds than expected, but this difference was not signif icant . T h e propor t ion o f selfed seeds was on ly s igni f icant ly l ower than expected i n the 7 5 % sel f -pol len treatment (Table 5.6). T h e slopes for the propor t ion o f selfed seeds observed, and expected when n - 1 and w i t h 2 8 % inbreeding depression, were s igni f icant ly different ( A N C O V A : F = 19.49, d f = 1, P = 0 .0001; F i g . 5.3). W i t h higher levels o f inbreeding depression, fewer ovules should contain both selfed and outcrossed embryos , and the expected decrease i n self ing is not as pronounced as when a l l embryos are v iab le , espec ia l ly w i t h l o w proport ions o f se l f -pol len. 61 0.80 0.70 to "S 0.60 CD CO Jj 0.50 CD co o 0.40 c o o 0.30 Q . 2 Q_ 0.20 0.10 -' — n •=: 1, no inbreeding depression - n ;= 1 Q - - n •= 2, chance ^ -• - - n = 2, outcross wins t n = 3, chance -t- - h = 3, outcross wins • selfed observed 0.25 0.50 Proportion of self pollen 0.75 Fig. 5.3 The proportion of selfed seeds expected and observed (± SE) with different proportions of self-pollen. N = 2028. n, number of embryos per ovule. Embryo viabilities for the expected selfed seeds are based on the mean of three trees. See Fig. 5.1 for more details. 62 Table 5.6 The propor t ion o f selfed seeds expected (when n = 1) and observed for three different proport ions o f se l f -pol len i n three Thuja plicata trees (181, 395 and 432) . N, number o f replicates. P ropor t ion o f Se l fed seeds W i l c o x o n S ign- rank sel f -pol len (pk) N E x p e c t e d Obse rved ( S E ) Di f fe rence T P (2-tailed) P ( l - t a i l e d ) 0.25 6 0.193 0 .312(0 .061) -0.119 7.5 0.156 0.078 0.50 6 0.419 0.390 (0.041) 0.029 2.5 0.68 0.34 0.75 6 0.683 0.527 (0.026) 0.156 10.5 0.031 0.016 A l l treatments 0.432 0.410 (0.032) 0 .022 17.5 0.468 Success of self-pollen - T h e relative success o f se l f -pol len changed depending on the propor t ion appl ied on a tree (Table 5.7). Se l f -po l l en had the greatest success when it was re la t ive ly rare, pe r forming 1.5 t imes better than outcross-pol len when it consti tuted 2 5 % o f the po l l en p o o l . H o w e v e r when sel f -pol len was c o m m o n ( 7 5 % self-pol len) it showed reduced success w i t h on ly 0.382 o f the relat ive success o f outcross po l l en . Se l f -po l l en performed s i m i l a r l y i n the three trees, w i t h the highest success i n the 2 5 % sel f -pol len treatment. Because o f unequal variance among treatments a W e l c h A N O V A was used to test for a difference among treatments ( S A S Institute, 1997). M e a n success o f se l f -pol len was s ign i f ican t ly different among treatments (F = 9 .192, df= 3, P = 0 .0027). W h e n the 2 5 % po l l en treatment was exc luded means for the other treatments s t i l l differed f rom each other ( A N O V A , F = 6.058, df= 2,P = 0.01). 63 Table 5.7 Fitness of self-pollen relative to outcross-pollen (ws) when applied at different proportions in three Thuja plicata trees. Proportion of self pollen applied (pk) Maternal tree 0.25 0.50 0.75 1.00 181 395 432 mean (SE) 2.083 1.366 1.148 1.533 (0.388) 0.547 0.667 0.811 0.675 (0.106) 0.441 0.365 0.340 0.382 (0.038) 0.729 0.690 0.742 0.720 (0.066) Discussion Polyembryony as a rescue mechanism -1 detected no significant increase in seed set in mixed-pollen treatments compared to the seed set expected based on pure self- and outcross-pollen treatments. Because the majority of western redcedar trees have high self-fertility, if polyembryony contributes to an increase in seed set, this increase is probably too small to detect. In fact, Table 5.4 shows expected increases in seed set of less than 1% between n = 1 and n = 3 for two of the trees, 181 and 432. Overall there were only 1% more full seeds observed than expected when n = 1. We would expect trees with the highest level of inbreeding depression to show the greatest increase in seed set. However, contrary to expectation tree 432, which had the lowest level of inbreeding depression (26%, Table 5.3) showed a significant increase in seed set following mixed pollinations. On the other hand, tree 431, which set only 7/200 seeds after pure self-pollinations and had 93% inbreeding depression showed no difference between expected and observed seed set. If polyembryony does play a role in rescuing ovules these differences should only be detectable in trees with low self-fertility but this was not the case for tree 431 in western redcedar. The limited number of trees makes it difficult to make firm conclusions. On the other 64 hand, i f embryo death occurs late i n seed development , after one e mbr yo has already outcompeted the others, p o l y e m b r y o n y should not affect seed set. Embryo competition - T h e propor t ion o f selfed seedlings decreased s igni f icant ly on ly w h e n the highest propor t ion o f se l f -pol len was appl ied (p k = 0.75). W h e n l o w proport ions o f se l f -pol len were appl ied 0 A = 0.25) there was actual ly an increase i n the propor t ion o f selfed seedlings, this difference, however , was not s ignif icant and a large error was associated w i t h the self ing estimate. W i t h the observed l eve l o f inbreeding depression over three o f the trees (5 = 28%) the compet i t ion among embryos and the relat ive decrease i n sel f ing should be strongest when 5 5 - 6 5 % self -pol len is rece ived (for n = 2 or 3). Sorensen (1982) found s imi la r results i n Doug la s - f i r (Pseudotsuga menziesii) and noble f i r (Abies procerd), and stated that self-po l l ina t ion rates o f around 5 0 % should confer the greatest advantage to p o l y e m b r y o n y and that at l o w proport ions o f se l f -pol len, p o l y e m b r y o n y should contribute l i t t le to increas ing outcrossing rates. A decrease i n self ing is not on l y due to a higher compet i t ive ab i l i ty o f outcrossed embryos . F igu re 5.3 shows that even when selfed and outcrossed embryos have the same compet i t ive ab i l i ty (i.e. chance) there is s t i l l a decrease i n the propor t ion selfed seeds due to l o w e r v i a b i l i t y o f selfed embryos . Because a delay o f a lmost four months occurs between po l l en germinat ion and fer t i l iza t ion o f ovules (Owens and M o l d e r , 1980), redcedar has the potent ia l for strong post-po l l i na t ion select ion for outcrossed offspr ing v i a differential po l l en tube g rowth rates o f self vs outcross po l l en . T h i s s i r ing advantage o f unrelated over related p o l l e n i n conifers w o u l d be s imi l a r to the c ryp t ic se l f - incompat ib i l i ty system found i n angiosperms. H o w e v e r , i n angiosperms different ial po l l en tube g rowth leads to an advantage o f very large magni tude for outcross po l l en . Ba teman (1956) obtained 9 0 % outcrossed seeds after se l f and outcross po l l en were app l i ed i n a 1:1 ratio i n Cheiranthus cheiri (Brassicaceae) , but there was no difference i n seed set when each po l l en type was appl ied separately. In this study I obta ined outcross ing rates 65 o f 6 1 % after a 1:1 ratio o f se l f to outcross po l l en was appl ied , and a lmost a l l o f the reduct ion i n selfed seeds can be accounted for b y differences i n self-fert i l i ty. E c k e r t and A l l e n (1997) found a s imi l a r advantage o f outcross over se l f -pol len i n Decodon verticillatus (Lythraceae) , but on ly ha l f o f i t c o u l d be attributed to early inbreeding depression. T h e y found that differences i n po l l en tube g rowth rates l i k e l y contr ibuted to the addi t iona l advantage o f outcross ing over se l f ing. Early inbreeding depression vs self-incompatibility - M o s t conifers exh ib i t l o w seed set f o l l o w i n g se l f -pol l ina t ion (Sorensen, 1969; F r a n k l i n , 1972; N a m k o o n g and B i s h i r , 1987; Savo la inen et al., 1992). Seed abort ion f o l l o w i n g self-fer t i l izat ion is thought to be caused b y deleterious recessive al leles (Char leswor th and Char leswor th , 1987). In this study, I use the term " inbreeding depress ion" to describe early e mbr yo death, even though this c o u l d actual ly be a case o f late se l f - incompat ib i l i ty (J. O w e n s , U n i v e r s i t y o f V i c t o r i a , pers. com. ) Pos t -zygo t ic self-i ncompa t ib i l i t y has been found i n several angiosperm trees and may be occu r ing i n gymnospe rms (Seavey and B a w a , 1986). In Pinus taeda, W i l l i a m s et al. (2001) found that a lethal factor l i n k e d to a microsate l l i te marker c o i n c i d e d w i t h the separation o f the e mbr yo f rom the megagametophyte, and this lethal factor acted i n an overdominant fashion suggesting a self-recogni t ion mechan i sm. T h i s sys tem paral lels late se l f - incompat ib i l i ty found i n some ang iosperm trees and may operate i n other conifer species that show ext remely l o w seed set f o l l o w i n g sel f -pol l inat ions . I f such a factor exists i n redcedar, however , it must have l o w potency g iven the h igh self ing success that I observed overa l l . Purging of inbreeding depression - A l t h o u g h one tree i n our study showed very l o w self-fer t i l i ty , the other three showed levels t yp i ca l for western redcedar (J. R u s s e l l , c i ted i n H u s b a n d and Schemske , 1996). It has been suggested that the h igh l eve l o f inbreeding depression, usual ly 66 associated w i t h conifers , has been purged i n T. plicata ( E l - K a s s a b y et al, 1994). In order for plants to purge their inbreeding depression inbred plants must surv ive to reproduce (Lande et al, 1994). La te act ing inbreeding depression such as a reduct ion i n g rowth rate can lead to the death o f inbred trees before they reach the reproduct ive stage. Sorensen (1999) l o o k e d at the su rv iva l rate o f three early successional species o f conifers (Abies procera, Pinus ponderosa, Pseudostuga menziesii, Pinaceae) w i t h 10% inbreeding depression i n g rowth rate. Because these species are not shade tolerant, late inbreeding depression can successful ly e l iminate inbred ind iv idua l s w h i c h w i l l be outcompeted b y faster g r o w i n g outbred trees. Wes te rn redcedar has also shown 10% inbreeding depression for g rowth rate (J. R u s s e l l , pers. com. ) but u n l i k e the species l is ted above it occurs at a l l stages o f forest succession and tolerates shade ( M i n o r e , 1990). S l o w e r g rowth caused b y inbreeding depression w i l l not necessar i ly lead to the death o f inbred ind iv idua l s i n western redcedar so that they can surv ive and reproduce. S u r v i v a l o f inbred ind iv idua l s i n western redcedar is reflected i n pos i t ive inbreeding coefficients at the adult stage i n natural popula t ions (Chapter 6). T h e h igh l eve l o f self-fert i l i ty i n western redcedar is faci l i ta ted by the purg ing o f early inbreeding depression i n natural populat ions . The importance of pre-pollination mechanisms - In this study a large number o f self-po l l ina ted ovules su rv ived to the seedl ing stage i n m i x e d po l l ina t ions , espec ia l ly in the 7 5 % self-po l l ina t ion treatment. T h i s indicates that the amount o f se l f -pol len rece ived (pr imary self ing rate) i n natural populat ions o f western redcedar is an important determinant o f the propor t ion o f selfed seedlings. In a previous study a mean self ing rate o f 2 8 . 5 % over s ix popula t ions was measured i n seedlings ( O ' C o n n e l l et al, 2001 ; Chapter 2). B a s e d on the results o f this study w e w o u l d expect se l f -pol l ina t ion rates i n natural populat ions to be s imi l a r to the measured self ing rate. In contrast, the self-fertile conifer , Picea omorika, showed 6 9 % seed set after con t ro l led self- po l l ina t ions but very h igh outcross ing rates (nearly one) i n natural popula t ions (Kui t t inen 67 and Savolainen, 1992). The authors suggest that pre-pollination mechanisms such as protogyny are responsible for the low levels of natural selfing. In western redcedar, polyembryony can contribute to a decrease in selfed seedlings when levels of self-pollen are high, but in general low post-pollination competition means that pre-pollination mechanisms will play an important role in determining selfing rates in most trees. Other trees that show very low levels of self-fertility, such as tree 431 in this study, should have high outcrossing rates but low seed-set with high levels of self-pollination. A combination of the degree of self-pollination and individual tree self-fertility will determine outcrossing rates in redcedar. Correspondingly, outcrossing rates in redcedar should vary among trees and populations as has been found in other studies (El-Kassaby et al, 1994; O'Connell et al, 2001, Chapters 2 and 4). Based on data from Owens and Molder (1980) which is mostly descriptive, I assumed that two to three embryos per ovule is typical in western redcedar. However, redcedar can potentially produce more embryos per ovule as the number of archegonia per ovule is higher. If outcrossed embryos are preferentially selected, then the decrease in selfing rates should be greater with a larger number of embryos. In this case, the high number of selfed seedlings observed in the low pollen treatment (Pk = 0.25) would seem to contradict the competitive advantage of outcrossed embryos. 68 Chapter 6 Range-wide genetic structure and diversity in western redcedar Introduction T h e amount and par t i t ioning o f genetic d ivers i ty i n a species is related to l i fe-his tory and eco log i ca l traits but is mos t ly the result o f species-specif ic his tory ( H a m r i c k et al., 1992). A r e v i e w o f the plant i s o z y m e literature has shown that l o n g - l i v e d w o o d y perennials have s igni f icant ly more genetic d ivers i ty at the species l e v e l than annuals, shor t - l ived perennials and l o n g - l i v e d herbaceous perennials ( H a m r i c k et al., 1992). O n e l o n g - l i v e d conifer , western redcedar (Thuja plicata D o n n ex D . D o n , Cupressaceae) is among the least genet ica l ly diverse trees. Wes te rn redcedar has shown l o w genetic var ia t ion w i t h i n and among populat ions for several traits. These inc lude relat ive amounts o f leaf o i l terpenes (von R u d l o f f and L a p p , 1979; v o n R u d l o f f et al, 1988), i sozymes (Copes , 1981; Y e h , 1988; E l - K a s s a b y etal, 1994), restr ict ion fragment length p o l y m o r p h i s m s ( R F L P ; G l a u b i t z et al, 2000) , and phenotypic traits (Rehfeldt , 1994; B o w e r and D u n s w o r t h , 1988). A reduct ion i n genetic d ivers i ty can occur through a reduct ion i n the effective popula t ion s ize, either through a large h is tor ica l bott leneck or sma l l , r eoccur r ing bott lenecks dur ing the co lon iza t ion o f new areas. A species-wide bott leneck i n Thuja plicata c o u l d have l ed to the l o w amount o f genetic var ia t ion observed i n present day populat ions ( Y e h , 1988; E l - K a s s a b y et al, 1994). T h e present range o f redcedar extends f rom southeastern A l a s k a to Nor the rn C a l i f o r n i a a long the coast o f western N o r t h A m e r i c a , and f rom southeastern B r i t i s h C o l u m b i a to northern Idaho i n the inter ior ( F i g . 6.1). The coastal and inter ior parts o f the range are essential ly geographica l ly isolated f r o m each other ( M i n o r e , 1990). D u r i n g the last 600 ,000 years o f the Quaternary (2.4 m i l l i o n years ago to the present) a series o f g l a c i a l cyc les o f approximate ly 100,000 years each, separated b y warmer in terg lac ia l periods o f about 10,000 years, have had a profound impact on the d is t r ibut ion o f N o r t h A m e r i c a n conifers 69 (Cr i t ch f i e ld , 1984; Hewi t t , 2000) . Pa leobotan ica l records indicate that western redcedar exper ienced a severe reduct ion i n range size dur ing the last ice age. A l o n g the P a c i f i c coast, the Fraser g lac ia t ion reached its m a x i m u m southern extent i n Nor the rn W a s h i n g t o n dur ing the V a s h o n stage, 15,000 y B P (years before present). P o l l e n records suggest that western redcedar was found m u c h further south, i n C a l i f o r n i a , and when glaciers began retreating redcedar s l o w l y r eco lon ized the coast, reaching the Puget S o u n d area 10,000 y B P and Nor the rn V a n c o u v e r Is land 3,000 y B P (Cr i t ch f i e ld , 1984; H e b d a and Mat thewes , 1984). Wes te rn redcedar may have persisted i n more than one re fug ium dur ing the last g l ac i a l per iod . G l a c i a l refugia o f western mes ic forest p robably occurred in more than one loca t ion a long the Pac i f i c coast, i n c l u d i n g central C a l i f o r n i a i n the south, the Queen Char lot te Islands i n the north, and i n north-central Idaho i n the inter ior (Bruns fe ld et al, 2001) . U n l i k e the previous genetic markers used to study western redcedar, microsatel l i tes show extensive p o l y m o r p h i s m , and can be used to retrace a more detai led his tory o f the species ( O ' C o n n e l l and R i t l a n d , 2000; Chapter 3). The present patterns o f genetic structure i n a species are a combina t ion o f both h is tor ica l events and current gene f l o w . Even t s such as the number and loca t ion o f g l ac i a l refugia, and the severity o f a bottleneck, w i l l affect patterns o f al lele frequency dis t r ibut ion. Gene t ic differentiat ion among regions f o l l o w i n g deglac ia t ion can arise through two different processes: p r imary intergradation and secondary contact. D u r i n g p r imary intergradation a front o f co lon izers occupy n e w l y avai lable habitats through long-dis tance dispersal and f o r m a leading edge o f co lonizers w h i c h is more l i k e l y to p rov ide migrants for subsequent co lon iza t ion (Hewi t t , 1993). Secondary contact i nvo lves the pos t -g lac ia l contact o f popula t ions isolated i n separate g l ac i a l refugia. P r i m a r y intergradation and secondary contact can be d i f f icu l t to d is t inguish f rom each other but each process should show different patterns o f genetic structure i n the zone o f differentiat ion. W i t h secondary contact, c l ines i n a l le le frequency at different l o c i should co inc ide , w h i l e w i t h p r imary intergradation they w i l l not 70 necessari ly co inc ide (Bar ton and Hewi t t , 1985). Fur thermore, dur ing secondary contact l inkage d i s e q u i l i b r i u m between l o c i should also persist for a few generations i n the contact zone (Durrett et al, 2000) . A severe reduct ion i n the effective popula t ion size can leave its mark on patterns o f a l le le d ivers i ty for several generations. D u r i n g a bott leneck, popula t ions lose rare alleles more q u i c k l y than c o m m o n alleles, so that the number o f al leles per locus is reduced more q u i c k l y than expected heterozygosi ty. T h i s results i n an excess o f heterozygosi ty relat ive to the number o f al leles observed i n recently bot t lenecked populat ions ( N e i et al, 1975; M a r u y a m a and Fuerst, 1985; C o m u e t and L u i k a r t , 1996). In contrast, i n a popula t ion q u i c k l y expand ing f rom a s m a l l effective popula t ion , such as du r ing range expans ion , mutat ions w i l l increase the number o f rare al leles , and a heterozygosi ty def ic iency should be observed ( M a r u y a m a and Fuerst , 1984; C o m u e t and L u i k a r t , 1996). B o t h o f these events c o u l d have occurred, but a heterozygosi ty excess should be detectable for a shorter t ime per iod f o l l o w i n g a bot t leneck than a heterozygosi ty def ic iency . T h e w i n d o w o f t ime dur ing w h i c h an excess or def ic iency i n heterozygosi ty can be detected w i l l a lso depend on the mutat ion rate o f the genetic markers used. In this study I sampled trees f rom throughout the range o f western redcedar and used hyper-var iable microsatel l i tes to retrace its g l ac i a l and pos t -g lac ia l his tory to answer two questions: W a s there a s ingle re fug ium or mul t ip l e refugia dur ing the last g lac ia t ion? Is there evidence o f a h is tor ica l bot t leneck i n western redcedar? Materials and methods Sample collection - T w e n t y to 37 Thuja plicata trees per popula t ion were sampled f rom 23 populat ions (mean 27 trees) for a total o f 620 trees. L o c a t i o n s o f the sampled populat ions are i n F i g . 6.1 and Tab le 6.1. F o r most populat ions, fresh fol iage was co l lec ted , transported i n l i q u i d ni t rogen and stored at - 8 0 ° C un t i l the D N A was extracted us ing a m o d i f i e d C T A B method ( A p p e n d i x I V ; D o y l e and D o y l e , 1987). F o r four o f the populat ions ( B C 1 2 , B C 1 3 , B C 1 4 and 71 B C 1 5 ) , D N A was extracted f rom 20 to 60 poo led megagametophytes/tree, w h i c h had been r e m o v e d f rom germinated seedlings (see Chapter 4) . D N A extracted for a previous study was used for popula t ions I D A , O R 1 , O R 2 , B C 3 and some ind iv idua l s o f B C 2 (Glaub i t z et al, 2000) . Samples f rom throughout the range o f Thuja plicata were grouped into 23 populat ions . Throughou t most o f its range western redcedar has a cont inuous d is t r ibut ion and separation into popula t ions is arbitrary. S o m e o f the populat ions i n this study are composed o f trees f rom different, nearby locat ions (wi th in 150 k m o f each other). I f there is l o c a l genetic structure w e expect an increase i n the f ixa t ion index (Fis) when two sub-populat ions are grouped. I compared Fis values o f grouped and ungrouped trees and found no increase i n Fis for any o f the grouped populat ions , ind ica t ing no signif icant substructure among the grouped trees ( B a l l o u x and L u g o n -M o u l i n , 2002) . 72 Fig. 6.1 Range map and location of 23 sampled populations of Thuja plicata. The shaded areas indicate the range of western redcedar. Filled circles represent populations from the southern clade and open circles, populations from the northern clade. Abbreviations are found in Table 6.1. Table 6.1 L o c a t i o n o f 23 sampled populat ions o f Thuja plicata. n, number o f trees sampled. Popu la t ion L o c a t i o n n Lat i tude ( °N) L o n g i t u d e ( ° W ) B C 1 M a p l e R i d g e , B C 30 4 9 ° 1 3 ' 1 2 2 ° 3 5 ' B C 2 Castlegar, B C 37 49° 19' 1 1 7 ° 4 0 ' I D A M o s c o w , I D 31 4 6 ° 4 3 ' 1 1 6 ° 5 6 ' B C 3 Reve l s toke , B C 34 5 0 ° 5 8 ' 1 1 8 ° 1 2 ' B C 4 V a l e m o u n t , B C 30 5 2 ° 4 9 ' 1 1 9 ° 1 5 ' B C 5 M c B r i d e , B C 30 5 3 ° 1 7 ' : 120° 10' C A E u r e k a , C A 30 4 0 ° 4 8 ' 1 2 4 ° 0 9 ' O R 1 N o r t h B e n d , O R 25 4 3 ° 2 4 ' 1 2 4 ° 1 3 ' O R 2 L i n c o l n C i t y , O R 26 4 4 ° 5 7 ' 1 2 4 ° 0 1 ' W A 1 St. He lens , W A 26 4 6 ° 2 0 ' 1 2 2 ° 3 1 ' W A 2 Port A n g e l e s , W A 27 4 8 ° 0 7 ' 1 2 3 ° 2 5 ' B C 1 2 C o o m b s , B C 20 49° 19' 124° 19' B C 1 3 Pember ton , B C 20 50° 19' 1 2 2 ° 4 7 ' B C 1 4 Y e l l o w Poin t , B C 20 4 8 ° 5 8 ' 1 2 3 ° 4 9 ' B C 1 5 P a l d i , B C 20 4 8 ° 4 7 ' 1 2 3 ° 5 0 ' B C 6 T o f i n o , B C 34 49°08" 1 2 5 ° 5 4 ' B C 7 Port H a r d y , B C 30 5 0 ° 4 3 ' 1 2 7 ° 2 8 ' B C 8 B e l l a C o o l a , B C 30 5 2 ° 2 2 ' 1 2 6 ° 4 6 ' B C 9 Queen Charlot te , B C 20 5 3 ° 1 6 ' 1 3 2 ° 0 4 ' B C 1 0 P r ince Rupert , B C 30 5 4 ° 1 9 ' 1 3 0 ° 1 9 ' B C 1 1 N e w Haze l ton , B C 30 5 5 ° 1 5 ' 1 2 7 ° 4 0 ' A L 1 Petersburg, A L 20 5 6 ° 4 8 ' 1 3 2 ° 5 7 ' A L 2 K e t c h i k a n , A L 20 5 5 ° 2 0 ' 1 3 1 ° 3 8 ' T o t a l 620 74 Microsatellite screening -1 screened 620 samples at eight p o l y m o r p h i c microsate l l i te l o c i . I car r ied out P C R reactions and fragment detection on a L I - C O R 4200 sequencer ( L i n c o l n , Nebraska) as descr ibed i n O ' C o n n e l l and R i t l a n d (2000) and Chapter 3. T o ensure that bands were u n i f o r m l y scored, on every ge l I used an a l l e l i c ladder o f two to three ind iv idua l s w i t h al leles o f k n o w n sizes and spanning the range o f a l le le sizes at a locus . F i v e l o c i , T P 1 , T P 3 , T P 4 , T P 7 and T P 9 , were composed o f s imple d inucleot ide repeats w h i l e T P 6 , T P 8 , and T P 1 1 had interrupted, c o m p o u n d or more c o m p l e x repeat mot i fs (Table 6.2). T h e number o f d inucleot ide repeats were obtained b y subtracting the number o f base pairs i n the f l ank ing regions o f the microsate l l i te plus the length o f the ta i l f r om the a l le le length, and d i v i d i n g i t b y two ( O ' C o n n e l l and R i t l a n d , 2000) . Because locus T P 6 contains a hexanucleot ide repeat, as w e l l as d inucleot ide repeats, the number o f repeats used i n our analyses does not accurately represent the number o f repeats for this locus (i.e. one hexanucleot ide repeat equals 3 d inuc leo t ide repeats). Diversity analyses - Heterozygosi t ies , a l le le frequencies, pa i rwise mul t i l ocus Fst (0, W e i r and C o c k e r h a m , 1984), inbreeding coefficients ( F ) and F-stat is t ics were obta ined us ing Genepop vers ion 3.3 (an updated vers ion o f 1.2, R a y m o n d and Rousset , 1995). G e n e p o p was also used to estimate P -va lues f rom exact tests o f departure f rom H a r d y - W e i n b e r g e q u i l i b r i u m us ing the M a r k o v cha in method w i t h 1000 iterations ( G u o and T h o m p s o n , 1992). 75 Table 6.2 M e a n divers i ty at eight microsate l l i te l o c i i n 23 populat ions o f Thuja plicata L o c u s n Repeat m o t i f Repeat N L e n g t h i n bp A A/P F T P 1 620 ( C A ) , 13-34 166-208 17 8.57 0.702 0 .124* T P 3 614 ( T G ) „ 8-39 176-238 24 9.70 0.796 0 .043* T P 4 619 ( T G ) „ 9-24 280-310 14 6.78 0.604 0 .103* T P 6 606 ( G C U G T U A T 78-111 231-297 30 15.70 0.900 0 .063* A T G T ) „ . . . ( G T ) „ T P 7 601 ( C A ) „ 6-32 231-283 22 7.83 0.764 0 .169* T P 8 578 ( C A ) J V C G ( C A ) J V 17-49 202-266 30 11.52 0.842 0 .271* T P 9 617 ( A C ) „ 20-65 263-351 42 15.00 0.857 0 .040* T P 1 1 613 ( C T U C A ) „ 15-33 200-236 10 5.48 0.663 0.0491 n, number o f trees sampled; Repeat N, number o f d inucleot ide repeats; bp, range o f a l le le lengths i n base pairs; A, total number o f al leles; A/P, mean number o f al leles per popula t ion ; Hep, mean expected heterozygosi ty; F, mean inbreeding coefficient . E x a c t test o f departure f rom H a r d y -W e i n b e r g e q u i l i b r i u m : * P - v a l u e < 0 .00001, | P - v a l u e = 0.055. Phylogeography -1 used Gendis t i n P h y l i p 3.573c (Felsenstein, 1995) to calculate N e i ' s genetic distance ( N e i , 1972) among a l l pairs o f populat ions . T h e N e i g h b o r p rog ram was used to construct a N e i g h b o r - j o i n i n g tree based on N e i ' s distances. The tree was v i s u a l i z e d w i t h T r e e v i e w 1.5 (Page, 1998). T h e Seqboot , Gendis t and Consense programs ( in P h y l i p ) were used to generate 1000 datasets and trees to obtain bootstrap values for each branch. T o test for s ignif icant differentiat ion among groups o f populat ions, I per formed an A n a l y s i s o f M o l e c u l a r V a r i a n c e ( A M O V A ) w i t h A r l e q u i n vers ion 2.000 (Schneider et al, 2000) . T o test for i so la t ion b y distance, I used the R - P a c k a g e vers ion 4 (Casgra in and Legendre , 2001) to convert latitudes and longitudes into k m between populat ions, and to per form a M a n t e l test on the corre la t ion between pa i rwise Fs, and pa i rwise geographic distances. Corre la t ions between latitude and mean 76 number o f al leles per locus (A), expected heterozygosi ty (Hep) and inbreeding coefficients (F) were ca lcula ted us ing J M P vers ion 3.2 ( S A S Institute Inc, 1997). Clines in allele frequencies - Secondary contact and p r imary intergradation should show different patterns i n c l i n a l a l le le frequency. I f separate clades are the result o f different g l ac i a l refugia a steep c l ine i n a l le le frequencies should occur i n the zone o f secondary contact and c l ines at different l o c i shou ld co inc ide . T o identify contact zones, pa i rwise correlat ions between latitude and al le le frequencies were per formed for each o f the 189 alleles. A l l e l e s that increased i n frequency w i t h latitude were termed northern al leles , and al leles that decreased i n frequency, southern al leles. F o l l o w i n g Turgeon and Berna tchez (2001) the s u m o f frequencies o f a l l northern alleles 0^,N) and the s u m o f a l l southern al leles (ptS) at the ith locus were ca lcula ted for each popula t ion . T h e relat ive frequency o f northern or southern al leles i n a popula t ion were obta ined b y d i v i d i n g p,N or p ,S b y the mean o f for a l l popula t ions (pf Ipt). T h e relat ive frequency o f al leles for each locus was then regressed on latitude. P i ecewi se regressions were used to ident i fy contact zones, as shown b y an increase i n the slope o f a l le le frequencies on latitude. Linkage disequilibrium -1 used G E N E T I X vers ion 4 .02 ( B e l k h i r et al, 2000) to calculate l inkage d i s e q u i l i b r i u m between pairs o f northern or southern al leles at different l o c i . G E N E T I X adapts the a lgor i thms f rom L I N K D I S o f B l a c k and K r a f s u r (1985). T h e p rog ram a l l ows the ca lcula t ions o f Dy, the l inkage d i s e q u i l i b r i u m between pairs o f al leles i and j (i = a l le le at locus 1 and j = a l le le at locus 2). F o r each popula t ion , I ca lcula ted D, the mean value o f Dtj for a l l pairs o f al leles at the different l o c i . Bottleneck test - T h e p rog ram Bot t l eneck (vers ion 1.2.02; Cornue t and L u i k a r t , 1996) was used to test for a h i s to r ica l reduct ion i n the effective popula t ion size o f western redcedar. 77 T h i s p rogram calculates the heterozygosi ty expected at mutat ion-drif t e q u i l i b r i u m (Heq) for the observed number o f al leles (ka) i n a popula t ion and tests whether there is a excess or def ic iency i n expected heterozygosi ty (He). Di f ferent levels o f heterozygosi ty are expected at mutat ion-drift e q u i l i b r i u m depending on whether the l o c i evo lve under the Infini te A l l e l e M o d e l ( I A M ) , the S tepwise M u t a t i o n M o d e l ( S M M ) or the T w o Phase M o d e l ( T P M ) . U n d e r the I A M each new mutat ion gives rise to a new al lele different f r om a l l ex i s t ing ones ( K i m u r a and C r o w , 1964). U n d e r the S M M new mutations are one size larger or smal ler than the o r ig ina l a l lele (Ohta and K i m u r a , 1973). Mic rosa te l l i t e s are expected to evo lve under the T P M , a combina t ion o f the I A M and the S M M , where most mutat ions are stepwise and a s m a l l number o f mutations are mult is tep ( D i R i e n z o et al, 1994). B a s e d on the differentiat ion patterns observed i n the phy logeograph ic analysis , populat ions were ana lyzed as three separate groups: the southern c lade ( exc lud ing Ca l i fo rn i a ) , the northern clade and C a l i f o r n i a . F o r compar i son , the data were tested for an excess or def ic iency i n expected heterozygosi ty under a l l three mutat ion-models . F o r the T P M analysis , var iance o f the length dis t r ibut ion o f mult is tep mutat ions was set at 30 ( D i R i e n z o et al., 1994), and the percentage o f single-step-mutations at 7 0 % (the p rog ram default values) . A l l results are based on 1000 iterations. T o test for a def ic iency or excess i n H e , two different statistical tests were performed. The "s ign test" has l o w statistical power but the W i l c o x o n s ign-rank test can be used w i t h as few as four p o l y m o r p h i c l o c i (Cornuet and L u i k a r t , 1996). L o w e r heterozygosi ty (Heq) is expected under the I A M than under the S M M for the same number o f observed alleles so that microsatel l i tes w i l l often show a heterozygosi ty excess when ana lyzed under the I A M m o d e l . I sozyme data f rom a previous study o f eight popula t ions o f western redcedar f rom southern B r i t i s h C o l u m b i a were also ana lyzed for compar i son ( Y e h , 1988). I sozymes are expected to f o l l o w the I A M so I ana lyzed the data under this m o d e l on ly . I f we assume that the populat ions are at mutation-drif t e q u i l i b r i u m , the Bo t t l eneck test can also ident i fy w h i c h mutat ion m o d e l a locus fo l l ows (Cornuet and L u i k a r t , 1996). 78 Results Genetic Diversity - A total of 189 alleles were scored at eight loci ranging from 10 to 42 alleles per locus (Table 6.2). Mean expected heterozygosity at each locus ranged from 0.604 to 0.900. Inbreeding coefficients (Fis) were positive and significantly different from zero for all but one locus (Table 6.2). With microsatellites, positive inbreeding coefficients can be the result of null alleles. This is probably the case for locus TP8, which had the largest Fis value and 6.8% of the individuals did not amplify a single band. If we assume that all individuals that did not amplify are homozygous recessive (r) for a null allele, the estimated frequency of null alleles at this locus is p = Vr = V0.068 = 0.26. Locus TP8 was excluded from other analyses below. Other loci still showed a positive inbreeding coefficient even though all 620 individuals produced bands. The results for genetic diversity at seven loci at the population level are presented in Table 6.3. The mean expected heterozygosity over all populations was 0.755 and the observed heterozygosity was 0.692. A total of 23 private alleles were scored. 79 Table 6.3 Genetic diversity in 23 populations of Thuja plicata at seven microsatellite loci Locus TP8 was excluded. Population A/L PA F South B C 1 11.3 3 0.831 0.743 0.105* B C 2 11.0 1 0.817 0.756 0.077N S I D A 11.1 1 0.733 0.613 0.164* B C 3 10.9 2 0.767 0.761 0.009 N S B C 4 9.7 0 0.755 0.694 0.088* B C 5 10.7 1 0.811 0.714 0.123* C A 9.1 1 0.760 0.703 0.078 N S OR1 11.3 1 0.775 0.654 0.150* OR2 12.1 5 0.819 0.602 0.271* W A 1 10.6 0 0.797 0.659 0.186* W A 2 10.9 1 0.806 0.758 0.058* BC12 10.6 0 0.817 0.805 0.013 N S BC13 9.9 0 0.825 0.743 0.109* BC14 8.9 1 0.771 0.729 0.056* BC15 9.6 3 0.741 0.707 0.042* Mean south (SD) 10.5(0.1) Total=20 0.788(0.113) 0.709(0.154) 0.102* (0.146) North B C 6 10.4 1 0.733 0.693 0.064 N S B C 7 10.0 1 0.715 0.690 0.037 N S B C 8 9.9 1 0.721 0.648 0.107 N S B C 9 8.0 0 0.660 0.631 0.040 N S B C 1 0 9.0 0 0.678 0.711 -0.043 N S BC11 8.3 0 0.695 0.669 0.035 N S A L 1 6.4 0 0.670 0.655 0.025 N S A L 2 7.3 0 0.669 0.580 0.147* Mean north (SD) 8.7(0.1) Total=3 0.693(0.147) 0.659(0.176) 0.052* (0.137) Overall mean (SD) 9.9(0.1) Total=23 0.755(0.134) 0.692(0.163) 0.085* (0.144) A/L, average number of alleles per locus; PA, number of private alleles; Hep, average expected heterozygosity; H0, average observed heterozygosity; F, average inbreeding coefficient; Exact test of departure from Hardy-Weinberg equilibrium *P < 0.005, N S , not significant after sequential Bonferroni correction (Rice, 1989). 80 Genetic structure - Differentiation among populations was low (mean multilocus Fst = 0.062; Table 6.4). Patterns of genetic structure coincided with geographical locations (Fig. 6.2). Overall, bootstrap values were low except for grouping made up of the eight northern populations (six north-western British Columbia populations: B C 6 - B C 1 1 , and the two Alaska populations: A L 1 & A L 2 ) from the rest of the populations. These populations from the northern clade are represented by open circles in Fig. 6.1. I will refer to these populations as the "northern" populations and the 15 others as the "southern" populations from now on. Results from an A M O V A also showed that the northern and southern groups were significantly different from each other (FCT= 0.039; P-value <0.0001; Table 6.5). The interior populations did not form a separate clade from the coastal populations and the branch lengths among central populations were relatively short in the dendrogram (Fig. 6.2). T a b l e 6.4 F-statistics at eight microsatellite loci in Thuja plicata Locus F , F,, F, TP1 0.133 0.066 0.190 TP3 0.046 0.064 0.107 TP4 0.117 0.083 0.191 TP6 0.059 0.032 0.088 TP7 0.171 0.048 0.212 TP8 0.280 0.058 0.322 TP9 0.042 0.062 0.101 TP11 0.042 0.057 0.096 Multilocus 0.112 0.064 0.169 Without TP8 0.085 0.062 0.142 CD o ro Z ' o o 3 . X J o ro co o 4= u c CS 43 CJ 3 "cS > X) B CS CJ a o Xt CJ 3 P H o P H cn CM O CJ B 60 B X) B CS •s ^ CS VO u cj * s O cS *u E-1 a o *> '5. C J CS 5. "3 co O XI °< , * ^ H .y o a cj § .* X) C J hi O 4= cj a cS V H X ) CJ .CJ C J 43 60 g ' & H 3 o O M to CJ B - B 5 ~ cS s o P H O H 3 CO to M O H > CS ^ o 43 43 60 CM NO X! B cj O H P H < CJ cS B CJ es g Q CJ 4= NO 0 X 1 cj 60 82 Table 6.5 A n a l y s i s o f molecu la r var iance ( A M O V A ) o f the effects o f populat ions and groups (Nor th vs South) on the dis t r ibut ion o f genetic d ivers i ty i n Thuja plicata based on seven microsate l l i te l o c i . Source o f var ia t ion df S u m o f V a r i a n c e Percentage o f squares components var ia t ion A m o n g groups 1 69.875 0.10917 3.89 A m o n g populat ions w i t h i n groups 21 177.437 0.10911 3.89 W i t h i n populat ions 1217 3148.675 2.58724 92.22 T o t a l 1239 3395.986 2.80553 F i x a t i o n Indices 2-tai led P F C T (Groups to Tota l ) 0.03891 <0.00001 F S C (Populat ions to G r o u p ) 0.04047 <0.00001 F S T (Populat ions to Tota l ) 0.07781 <0.00001 Isolation by distance - O v e r a l l , pa i rwise genetic distance (0) between popula t ions increased l inear ly w i t h geographica l distance ( M a n t e l ' s t = 7.206, r = 0 .788, P - v a l u e <0.00001, N = 253) . T h i s re la t ionship was found both among populat ions w i t h i n a reg ion (Nor th or South) and between populat ions f rom different regions (Pa i rwise correlat ions: N o r t h vs N o r t h : r = 0.412 P < 0.029, N = 28; South vs South : r = 0 .737, P < 0 .0001, N = 105; N o r t h vs Sou th : r = 0 .749, P < 0 .0001, N = 120; F i g . 6.3). The slope between genetic and geographica l distance differed among groups and was s igni f icant ly steeper i n the southern populat ions than the northern popula t ions ( A N C O V A : F- ra t io = 5.95, P = 0.016; Tab le 6.6). 83 2000 2000 1000 1500 Kilometres 2000 Fig. 6.3 Pa i rw i se genetic distance (0) as a funct ion o f geographic distance between populat ions o f Thuja plicata. Dis tances are shown between northern popula t ions ( N vs N ) , southern popula t ions (S vs S ) , and between northern and southern popula t ions ( N vs S) . 84 Table 6.6 A n a l y s i s o f covar iance ( A N C O V A ) o f the effects o f geographica l distance between popula t ions and groups ( N vs N , S vs S, and N vs S) on pa i rwise genetic distance (9). W h o l e m o d e l : R2 = 0 .701, N = 253, F - ratio = 62.73, P< 0 .0001. df F - rat io M S P - va lue Dis t ance 1 75 .4394 <0.0001 G r o u p s 2 12.4201 <0.0001 Dis tance x Groups 2 6.9467 0 .0012 M o d e l 5 115.673 0.028353 <0.0001 E r r o r (residual) 247 0.000245 Northern vs southern populations - P a i r w i s e genetic distances were lower between populat ions f rom the northern c lade (mean 6 = 0.027 ± 0.013 S D ; range 0 .0075-0.0529, N = 28; A p p e n d i x V I ) than between populat ions f rom the southern clade (mean 0 = 0.043 ± 0.023 S D ; range 0.0043-0.1137) . The mean genetic distance between populat ions f rom different clades (mean 9 = 0.066 ± 0.016 S D ; range 0.0309 - 0 .111; N = 81) was s igni f icant ly larger than between popula t ions w i t h i n a clade (t = 10.73, df= 251 , P < 0 .0001). Gene t ic d ivers i ty i n northern populat ions was lower than i n southern popula t ions . B o t h mean expected heterozygosi ty (r =-0.556, N = 23,P = 0 .0059) and the mean number o f al leles per locus ( r = -0.626, N = 23 , P = 0 .0014) decreased w i t h latitude ( F i g . 6.4). T h i s decrease i n d ivers i ty was largely dr iven by differences i n genetic d ivers i ty between the northern and southern groupings . A s a group, southern populat ions had both s igni f icant ly more al leles per popula t ion (P < 0 .0005) and higher mean expected heterozygosi ty (P < 0 .0001) than northern popula t ions (Table 6.3). Seventy-three percent o f southern popula t ions (11/15) contained between one and f ive private al leles w h i l e on ly 3 7 % o f the northern popula t ions (3/8) contained one private a l le le . 85 Mating system - A l l populat ions but one ( B C 1 0 ) had a pos i t ive inbreeding coeff icient and the mean inbreeding coeff icient (F = 0.085) was s igni f icant ly different f r om zero (Table 6.3). T h e inbreeding coeff icient was negat ively correlated w i t h latitude ( r = -0 .451, N = 23 , P = 0.031; F i g . 6.4) and it was not s igni f icant ly different f r o m zero i n seven out o f eight northern popula t ions (Table 6.3). Allele size distribution - O n l y one locus ( T P 1 1 ) out o f the eight l o c i d i d not show a b i m o d a l or m u l t i m o d a l d is t r ibut ion i n al lele sizes ( F i g . 6.5). There was an over lap i n al lele size dis t r ibut ion between Nor the rn and Southern populat ions w i t h the most frequent alleles at a locus often be ing the same i n both groups. S i x t y - t w o al leles were on ly found i n the southern popula t ions w h i l e four alleles were found on ly i n northern populat ions . A l l these alleles occur red at frequencies b e l o w 1% except two al leles at locus T P 8 w i t h respect ive frequencies o f 2 .7% and 1.7% over a l l southern populat ions. 86 13 •12 11 A/bo 9 8 7 • - ,* • • • 9 • • - • • o; .... o o H 'i 0 o o 40 42 44 46 48 50 52 54 56 58 0.82 • • • 0.78 H • - • • • ep • 0.74 • . • : o o o 0.70 • :.o o . . . o o 0.66 i i i i 40 42 44 46 48 50 52 54 56 58 40 42 44 46 48 50 52 54 56 58 Latitude (*N) Fig. 6.4 Average number of alleles per locus (A/L), mean expected heterozygosity (Hep) and mean inbreeding coefficients (F) in 23 populations of Thuja plicata as a function of latitude. Filled circles represent populations from the southern clade and open circles, populations from the northern clade. r-• CM .. 43 O O M m 0) cd ^ > ° s§ 0 "3 a u oo 3 43 II 43 O " •C cd c 3 U ^ •» I n « ^ I o u .a a -3 1 an, O CO D H § ^ a [ -1 u -a *w B 33 ON « OH -3 H 0 0 OH H S3 f—I co . y O 1 s 2 3 •a "a B CN 3 0 0 42 ll i5 B « . a u -a T3 JS 8 S _ p-i E ° 1 6 £ 43 o 5 4 ) W 4 ) C S ^ , , - H «3 43 4 ) u > * T t 4 ) O J £ o H _43 0 0 ° ^ 2 • H C 3 r j -II 3 43 • C ccj C a 43 ^ W t— tfl f 1 CO N — B J 2 B U 3 If --7 6 43* « « t ; • 2H 43 89 Clines in allele frequencies - Seven l o c i showed at least one a l le le that was s igni f icant ly correlated w i t h latitude. Twenty- three al leles (i = 7 l oc i ) decreased i n frequency w i t h latitude (southern al leles) and 12 al leles (/ = 6 loc i ) increased i n frequency (northern alleles; T a b l e 6.7). W h e n the relat ive frequency o f northern al leles (pt I pt) was plot ted on latitude there was a steep c l ine i n al lele frequency a long the coast that corresponded w i t h the f ive V a n c o u v e r Is land populat ions ( F i g . 6.6a). The slope o f p, / p , on latitude for the V a n c o u v e r Is land popula t ions ((3 = 0.42; r2 = 0 .325; N = 30; P = 0 .001; 9 5 % C I = 0.186, 0.655) was s igni f icant ly different f rom the slope for either the southern populat ions ((3 = 0.044; r2 = 0 .179; N = 12;P = 0 .0002; 9 5 % C I = 0 .021, 0.066; A N C O V A : F = 14.45; N = 102; P = 0 .0003), or the northern populat ions (p = 0.018; r2 = 0 .002; N = 36; P = 0.77; 9 5 % C I = - 0 . 1 0 6 , 0 .141; A N C O V A : F = 8.126, N= 66; P = 0.006). T h i s same c l ine d i d not occur i n the inter ior popula t ions at the same latitudes. A steep c l i ne i n the relat ive frequency o f southern alleles occurred between the C a l i f o r n i a and O r e g o n populat ions, m u c h further south than the c l ine i n northern al leles ( F i g . 6.6b). The slope o f relat ive a l le le frequency on latitude for the four southernmost popula t ions (P = -0 .352; r2 = 0.388; N = 28; P = 0 .0004; 9 5 % C I = -0.530, -0.174) was s igni f icant ly steeper than for the 19 other popula t ions (p = -0.080; r 2 = 0.200; N = 133; P <0.0001; 9 5 % C I = -0 .107, -0 .052; A N C O V A : JV= 161; F = 25 .243; P < 0 .0001). T h e mean relat ive frequency o f southern al leles was s igni f icant ly larger i n the C a l i f o r n i a popula t ion than i n a l l other popula t ions ( T u k e y test; a = 0.05 over a l l compar isons for a l l popula t ion pairs). There was no s ignif icant l inkage d i s e q u i l i b r i u m between the southern al leles, nor the northern al leles i n any o f the populat ions, and popula t ions i n the contact zone d i d not show higher mean l inkage d i s e q u i l i b r i u m ( D ) values than other populat ions . 90 Table 6.7 A l l e l e s w i t h frequencies s igni f icant ly correlated w i t h latitude, at the 0.05 l eve l , at seven microsate l l i te l o c i i n Thuja plicata. A l l e l e = number o f repeats, r, Pearson corre la t ion coefficient . L o c u s A l l e l e F requency r A l l e l e type T P 1 17 0.053 -0.5434 south T P 1 18 0.44 0.8021 north T P 1 19 0.168 -0.6034 south T P 1 20 0.084 -0.4201 south T P 1 21 0.044 -0.6863 south T P 3 8 0.032 -0.5921 south T P 3 16 0.103 -0.5351 south T P 3 18 0.274 0.5501 north T P 3 24 0.067 0.5692 north T P 6 84 0.069 -0.4688 south T P 6 90 0.032 -0.5468 south T P 6 101 0.0103 0.6697 north T P 7 12 0.001 -0.5088 south T P 7 16 0.327 0.4538 north T P 7 17 0.169 -0.4154 south T P 7 18 0.017 -0.5593 south T P 7 20 0.215 0.6868 north T P 7 21 0.101 -0 .7650 south T P 8 18 0.011 -0 .6099 south T P 8 25 0.060 -0.5761 south T P 8 31 0.199 0.5083 north T P 8 46 0.056 0 .5252 north T P 9 25 0.008 -0.5619 south T P 9 27 0.173 0 .7072 north T P 9 28 0.068 0.4719 north T P 9 31 0.06 -0 .4512 south T P 9 34 0.036 -0.4285 south T P 9 35 0.036 -0.4153 south T P 9 36 0.125 -0.5174 south T P 9 37 0.143 0 .6020 north T P 9 43 0.006 -0.4144 south T P 9 53 0.004 -0.4467 south T P 9 58 0.005 -0 .4380 south T P 1 1 26 0.107 -0 .4232 south 91 Fig. 6.6 Relative frequency of (a) 12 northern and (b) 23 southern alleles as a function of latitude in 23 populations of Thuja plicata. Filled circles represent mean allele frequencies for southern populations and open circles represent means for northern populations. Error bars represent standard error of the mean. Dotted lines represent steep clines in allele frequencies, and populations with crosses are part of those slopes. 92 Bottleneck test - Tab le 6.8 summarizes the results o f the Bo t t l eneck test. U n d e r the Infinite al lele m o d e l ( I A M ) the southern group showed a heterozygosi ty excess at a l l eight l o c i and the C a l i f o r n i a popula t ion at seven l o c i . In the northern group, ha l f o f the eight microsate l l i te l o c i showed a heterozygosi ty excess and the other ha l f a def ic iency. In contrast, under the T w o -phase m o d e l ( T P M ) both the northern and southern groups showed a heterozygosi ty def ic iency at seven o f the eight l o c i , and s igni f icant ly differed f r o m the T P M mutation-drif t e q u i l i b r i u m as shown by both the s ign test and the W i l c o x o n test. The C a l i f o r n i a popula t ion however d i d not differ f r om the expected T P M e q u i l i b r i u m and on ly ha l f o f the l o c i showed a heterozygosi ty def ic iency . U n d e r the Stepwise mutat ion m o d e l ( S M M ) a higher heterozygosi ty at muta t ion-drift e q u i l i b r i u m {Heq) is expected than for the other two mode ls for the same number o f al leles (Cornuet et L u i k a r t , 1996). The results for the S M M are not shown but a l l groups showed a s ignif icant heterozygosi ty def ic iency {He < Heq) under this m o d e l . F o r i sozymes , four o f the f ive l o c i showed a heterozygosi ty def ic iency under the I A M . T h i s def ic iency was not, however , stat ist ically s ignif icant due to the sma l l number o f p o l y m o r p h i c l o c i avai lable . Recen t ly bot t lenecked populat ions should also show a mode-shif t d is t r ibut ion i n al lele frequencies so that al leles i n l o w frequency classes (<0.1) become less abundant than intermediate and h igh frequency classes (Lu ika r t et al., 1998). T h e Bo t t l eneck p rog ram d i d not show any s ignif icant mode-shift toward higher al lele frequency classes i n redcedar. In fact most al leles (>70%) i n both southern and northern populat ions were found i n the lowes t frequency class (<0.05; F i g . 6.7). O n l y the northern group had al leles that had frequencies o f over 6 0 % . These were a l le le 16 at locus T P 1 ( F i g . 6.5a) and al lele 16 at locus T P 4 ( F i g . 6.5c). 93 0 7-1 Allele frequency class Fig. 6.7 Proportion of alleles at eight loci in different frequency classes for southern populations (filled bars) and northern populations (open bars) of Thuja plicata. (n = 189 alleles). cu CU T 3 CU T 3 TJ t o fi O T J "33 CO c o 3 o PH CU o c i cs CD r— OO co <2 43 'S ^ 3 P H .5 P H CO fi 15 <3 T J O e o 4 rt CN NO 60 O <2 "rt fi co rt 4^ cj CU B CU O PQ cu 43 CU 00 NO Ji 3 ca H C*H T J CU B ' r t 43 O CU CO G O OH -3 3 PH O a 43 .60 'S u rt 60 o . cu rt 6o -a 4? cj CJ O r H CO *3 i—i § >> T3 CJ in rt 43 B O rt B 60 U cu T J O CU CO rt 43 P H 6 c-i CU T J o E J D "rt B O X o o s bp C O o o CU 43 6 3 55 B O X o o s .60 B 00 cj O cj 43 E 3 V cy A cy c/) CJ CJ X CU cu &: cy v cy A cy CU o x CU 43 o 43 o a o I CM ON O N ON o o d o d o r-O N o o o d o 00 CN IT) * E cu 43 » 3 O oo O N O N O N tn o o d o d T t co IT) d CO 00 CO N O 00 00 CN T t E CU 43 t! o T t d oo T t d 00 T t O N o o O N O N O N O N o d T t o NO rt U O N 00 CO d co 00 00 T t - M N O T t O N »H CU 1 3 B T J CU > 43 O 43 O 3 a" CJ 3 fi o cu B O cy V cj B CU O 60 N o T J CU CU T J 3 cu 1 3 B T J CU CJ cu P H X CU P H X a o 60 >. N o CU *-» CU 43 cy cu 1 3 B T J CU CJ 43 T > O O CJ O H X CJ CO CU CJ X cu •8 43 O i -P H ' r t cu B O A cj B CJ 'u WH CJ T J 1-o CJ CJ X CJ CJ X ° H ^ ' S CO o 60 >, N O 43 O l H P H CU CJ X CU cu &: TO 4—» 43 rt 43 I '5 O CU -4—* cy HS hC rt cu O « : P H G o 3 O H o P H rt u CU CU T ) _3 "u X W # 95 Discussion Phylogeographic structure - The dis t r ibut ion o f genetic d ivers i ty i n western redcedar suggests the presence o f at least three refugia dur ing the last g l ac i a l per iod . Three m a i n patterns stand out f rom the range-wide phylogeographic analysis . F i rs t , the northern populat ions a long the coast o f B r i t i s h C o l u m b i a and A l a s k a fo rm a dis t inct clade f rom the southern coastal and inter ior populat ions . Second , the C a l i f o r n i a popula t ion is divergent f rom a l l the other populat ions . T h i r d , the inter ior and coastal populat ions, a l though geographica l ly disjunct, are not genet ica l ly dist inct . T h e steep c l ine i n the frequency o f northern al leles and the greater genetic distance o f populat ions f rom different clades, compared to popula t ions w i t h i n a clade, support V a n c o u v e r Is land as a zone o f secondary contact between trees f r o m southern and northern refugia. The popula t ion B C 6 near T o r i n o appears to be composed o f a mix ture o f al leles f rom the northern and southern clades. A probable r e fug ium for the northern populat ions o f western redcedar is near the Queen Charlot te Islands, B r i t i s h C o l u m b i a ( F i g . 6.8). The strongest evidence for a northern g l ac i a l r e fug ium o f mesic forests o f f the coast o f the Queen Char lo t te Islands are 16,000 year o l d fossi ls o f several plant species, i n c l u d i n g S i t k a spruce (Picea sitchensis) (Warner and Ma thewes , 1982). Sea levels were l o w e r du r ing the last g l ac i a l pe r iod and forests w o u l d have persisted i n an area n o w under water. T h e occurrence o f a north/south spli t on V a n c o u v e r Is land co inc ides w i t h the results o f a prev ious study where restr ict ion fragment length p o l y m o r p h i s m s ( R F L P ) were used to infer the genetic structure o f western redcedar over its range (Glaub i t z et al, 2000) . In this R F L P study, trees f rom southern coastal B r i t i s h C o l u m b i a were poo led w i t h trees f rom northern V a n c o u v e r Is land. The results showed that the southern B r i t i s h C o l u m b i a coastal group was as c lose ly related to the Queen Char lot te Islands group, as to the groups f rom the coastal U n i t e d States and B r i t i s h C o l u m b i a interior. These results are not surpr is ing as the B r i t i s h C o l u m b i a coastal group w o u l d have i n c l u d e d ind iv idua l s f rom both the northern and southern clades ou t l ined i n this study. 96 A north/south spli t i n genetic structure occurs i n several other plant species a long the west coast o f N o r t h A m e r i c a . U s i n g c p D N A markers So l t i s et al. (1997) found that popula t ions o f s ix plant species (three herbaceous perennials , a tree, a shrub and a fern) were a l l geographica l ly structured into northern and southern clades that met i n south-central Oregon . T h e y attributed this north/south spli t to the presence o f more than one g l ac i a l re fugium. T h e loca t ion o f the steep c l ine i n southern a l le le frequencies i n western redcedar co inc ides w i t h the loca t ion o f the north/south par t i t ioning o f the species r ev i ewed b y So l t i s et al. (1997). T h e l ack o f differentiat ion between inter ior and coastal populat ions and the d ivergence o f the C a l i f o r n i a popula t ion suggest a th i rd possible refugium, i n either the inter ior i n northern Idaho, or a long the coast o f Oregon ( F i g . 6.8). R i cha rdson et al. (2002) suggested an east to west co lon iza t ion o f the coast f r om a re fugium i n north-central Idaho for Pinus albicaulis. The poss ib le existence o f an inter ior mes ic forest r e fug ium is based on the present disjunct range for several different western plant and a n i m a l species (Bruns fe ld et al, 2001). N o mes ic forest r e fug ium has been conf i rmed , however , south o f g l ac i a l l imi t s as a source for in l and popula t ions because no Quaternary po l l en sites have been sampled i n this area (Mehr inger , 1985). A l t e rna t ive ly , the inter ior c o u l d have been c o l o n i z e d b y trees f rom a coastal refugium. S a m p l e d locat ions i n foss i l studies south o f the glaciers are also sparse a long the coast. H o w e v e r , Got tesfe ld et al. (1981) ident i f ied late Ple is tocene (>35,000 radiocarbon y B P ) macrofoss i ls o f Thuja plicata f r o m the western Cascade mountains , 55 k m east o f Eugene Oregon . F igu re 6.8 illustrates the poss ib le reco lon iza t ion routes f rom three separate g l ac i a l refugia. 97 Fig 6.8 Hypothesized post-glacial colonization routes for Thuja plicata from three glacial refugia discussed in the text. (1) California, (2) Queen Charlotte Islands, and either (3) western Oregon or (4) northern Idaho. 98 The s imi la r i ty i n the genetic structure o f several co -occur r ing species w i l l he lp identify the loca t ion o f g l ac i a l refugia and pos t -g lac ia l co lon iza t ion routes (Bruns fe ld et al, 2001) . T h e genetic structure o f other conifers w i t h a geographic dis t r ibut ion s imi l a r to Thuja plicata has been studied us ing i sozymes . These inc lude S i t k a spruce (Picea sitchensis, Y e h and E l - K a s s a b y , 1980), lodgepole pine (Pinus contorta, W h e e l e r and Gur ie s , 1982), western whi te p ine (Pinus monticola, S te inhof f et al, 1983), pac i f ic y e w (Taxus brevifolia, E l - K a s s a b y and Y a n c h u k , 1994), mounta in h e m l o c k (Tsuga mertensiana, A l l y et al, 2000) , and y e l l o w cedar (Chameacyparis nootkatensis, R i t l a n d et al, 2001) . In many o f these studies, the area sampled was l i m i t e d to B r i t i s h C o l u m b i a and no clear patterns o f geographic structure arose f rom the analyses. Excep t ions inc lude Pinus monticola, w h i c h showed a north/south spli t near the C a l i f o r n i a - O r e g o n border corresponding to the pattern observed i n redcedar (Ste inhoff et al, 1983) . In Pinus contorta sub. latifolia, northern and southern groups had a contact zone i n central B r i t i s h C o l u m b i a (Whee le r and Gur ie s , 1982). These results are not supported, however , b y chloroplas t microsate l l i te marker data ( M a r s h a l l et al, 2002) . I sozymes revealed different clades for Chamaecyparis nootkatensis (Cupressaceae) suggesting mul t ip l e g l a c i a l refugia a long the P a c i f i c coast (R i t l and et al, 2001) . It w o u l d be par t icular ly interest ing to obtain a more deta i led genetic study on the pos t -g lac ia l movement o f C. nootkatensis because w i t h i n the Cupressaceae, foss i l po l l en among species can not be differentiated ( H e b d a and Mat t ewes , 1984) . T h i s w o u l d help d is t inguish whether T. plicata or C . nootkatensis was more l i k e l y to be i n an area at a par t icular t ime f o l l o w i n g deglacia t ion. Species sharing a s imi la r range dis t r ibut ion w i l l not necessari ly show a s imi l a r genetic structure. The rate o f co lon iza t ion o f new areas w i l l result i n differences i n genetic structure. F o r example , a somewhat different structure f rom western redcedar was observed w i t h a l l ozymes i n Alnus rubra (Betulaceae), where V a n c o u v e r Is land and Queen Char lot te Is land popula t ions d ive rged f rom ma in l and populat ions ( H a m a n n et al, 1998). The authors hypothes ized that trees 99 f rom a northern g l ac i a l re fugium c o l o n i z e d V a n c o u v e r Is land when the sea levels were l o w e r and a land br idge connected the Queen Char lot te Islands to V a n c o u v e r Is land approximate ly 12,000 y B P . Trees f rom a southern re fug ium later c o l o n i z e d the B r i t i s h C o l u m b i a ma in land . Alnus rubra, u n l i k e Thuja plicata, is an early successional species that m o v e d into recently deglaciated areas more q u i c k l y than redcedar. Population differentiation - T h e amount o f differentiat ion among popula t ions o f western redcedar i n this study (mul t i locus Fst = 0.062) was s imi l a r to other gymnosperms as measured b y i sozymes (mean Gst = 0.073 for 102 species; H a m r i c k and God t , 1996). A previous i s o z y m e study cove r ing eight populat ions f rom southern and eastern B r i t i s h C o l u m b i a y i e l d e d a l o w e r estimate o f differentiat ion (Fsl = 0 .033; Y e h , 1988). W h e n on ly popula t ions f rom the same area covered b y the i s o z y m e study were i n c l u d e d a s imi la r result was obta ined (Fs, = 0 .027). D i v e r g e n c e is p robab ly underestimated b y Fsl for two reasons: (1) S tepwise mutat ions i n microsate l l i tes lead to h igh homoplasy , decreasing the magni tude o f Fsl ( B a l l o u x and L u g o n -M o u l i n , 2002) . (2) I f i nd iv idua l s o f a species are isolated into separate refugia dur ing g lac ia t ion and undergo separate bott lenecks, most alleles lost i n both groups w i l l be rare al leles . C o m m o n al leles should be the same i n the different groups and w i l l p robab ly surv ive the bott leneck and increase i n frequency independent ly (Lat ta and M i t t o n , 1999). I f l o w Fsl is due to the same al leles s u r v i v i n g i n both refugia, values o f Fs, shou ld vary among l o c i . H o w e v e r , i n d i v i d u a l locus Fst for redcedar had a nar row range, between 3 .2% to 8.3%, ind ica t ing that Fsl may not have been too biased by this effect. A n argument against p r imary intergradation and i n support for mul t ip l e g l ac i a l refugia as a cause for differentiat ion among regions i n western redcedar, is that trees are less l i k e l y to undergo founder effects than annual plants. Aus t e r l i t z et al. (2000) ou t l ined two m a i n reasons w h y trees show l o w e r differentiat ion i n nuclear markers among populat ions than annual plants. Fi rs t , because o f over lapp ing generations i n trees, mu l t i p l e 100 co lon izers w i l l reach an area before the first arr ivals have begun p roduc ing offspr ing. Second , the long-dis tance gene-f low through po l l en dispersal i n w i n d po l l ina ted trees w i l l decrease differentiat ion among populat ions . Reduction in genetic diversity - The results o f the bott leneck test ind ica ted that a r ap id expans ion f rom a restricted popula t ion size occurred i n redcedar. O f course, this test can on ly reveal the dynamics i n popula t ion growth since the last bott leneck. Nevertheless , it indicates the potent ial impac t o f g lac ia t ion on a species ' l eve l o f genetic d ivers i ty . B o t h the northern and southern groupings showed heterozygosi ty def ic iencies at microsate l l i te l o c i under the T P M ind ica t ing a popula t ion expansion. I sozyme l o c i , w h i c h were ana lyzed under the I A M , also showed a heterozygosi ty def ic iency at four o f the f ive l o c i , but the overa l l results were not s ignif icant . I sozymes are not as w e l l suited for the bott leneck test because they are less l i k e l y than microsate l l i tes to be neutral (Cornuet and L u i k a r t , 1996). T h i s is p robably the case for locus G 6 p d (Glucose-6-phosphate-dehydrogenase) w h i c h showed a near ly equal frequency o f the t w o al leles i n a l l sampled populat ions ( Y e h , 1988; E l - K a s s a b y et al, 1994; O ' C o n n e l l et al, 2001) . T h i s is also probably w h y a heterozygosi ty excess, instead o f a def ic iency , was observed at this locus . U n l i k e i sozymes , microsatel l i tes i n western redcedar are h i g h l y var iable . The higher mutat ion rate for microsatel l i tes , compared to i sozymes , can exp l a in w h y a larger number o f al leles per locus was found at this marker (see Chapter 7). B o t h types o f markers showed excesses i n rare al leles, a l though they showed different amounts o f genetic d ivers i ty . F o r i sozyme , a severe bott leneck probably reduced i s o z y m e divers i ty to one a l le le per locus at a l l but one locus (G6pd) , w h i l e for microsatel l i tes several o f the most c o m m o n alleles p robably su rv ived the bott leneck. T h i s is suggested i n the m u l t i m o d a l d is t r ibut ion at seven o f the eight microsate l l i te l o c i . In i t i a l ly , a few c o m m o n alleles w o u l d have persisted f o l l o w i n g a bot t leneck and a heterozygosi ty excess may have occurred unt i l new mutat ions q u i c k l y accumulated, 101 lead ing to the heterozygosi ty def ic iency presently observed. A l t h o u g h northern and extreme southern popula t ions have not been studied for genetic var ia t ion at i s o z y m e l o c i , they showed reduced genetic var ia t ion at microsate l l i te l o c i compared to popula t ions at the center o f the range. A decrease i n d ivers i ty w i t h latitude is sometimes an argument i n favor o f recently c o l o n i z e d area (Hewi t t 2000) . T h e decrease i n d ivers i ty i n redcedar c o u l d also be the result o f a more severe bott leneck i n the northern populat ions or l ower genetic d ivers i ty i n these popula t ions preceding the last g lac ia t ion . Timing a species-wide bottleneck - The last g l ac i a l pe r iod lasted approximate ly 100,000 years. A s s u m i n g a generation t ime o f about 50-100 years, popula t ions o f western redcedar w o u l d have been isola ted f rom each other for about 1000-2000 generations. T h e evidence o f mu l t i p l e refugia dur ing the last g lac ia t ion , c o m b i n e d w i t h the l ack o f strong differentiat ion over the range o f western redcedar suggests that i f a bot t leneck reduced species-wide genetic d ivers i ty i n western redcedar it predates the last g lac ia t ion . Because o f the long generation t ime i n western redcedar, on ly 100-200 generations have occurred since deglac ia t ion and even fewer i n recently c o l o n i z e d areas. A c c o r d i n g to po l l en records, western redcedar d i d not reach Nor the rn V a n c o u v e r Is land un t i l 3000 y B P (Cr i t ch f i e ld , 1984). T h i s means that the northern and southern clades have been i n contact for no more than 30 to 60 generations. T h e last in terglacia l pe r iod (Eemian) occurred between 130,000 - 115,000 y B P and was about the same length as the current w a r m pe r iod (Holocene ; A d a m s et al, 1999; F r o g l e y and Tzedak i s , 1999). I f a severe species-w i d e bot t leneck occur red dur ing a previous g l ac i a l per iod, p receding the last in terg lac ia l , levels o f genetic d ivers i ty dur ing the E e m i a n w o u l d not have been any higher than the present levels o f d ivers i ty . It is quite p laus ib le that western redcedar has exper ienced several dramatic contractions and expansions i n range and effective popula t ion size, and a severe loss i n genetic d ivers i ty has occur red more than once. T h e l o w differentiat ion between geographica l ly distant 102 southern and northern populat ions i n microsatel l i tes , R F L P (G laub i t z et al, 2000) and leaf o i l terpenes (von R u d l o f f and L a p p , 1979; v o n R u d l o f f et al, 1988) indicates that the separate refugia have probably not persisted for more than one g l ac i a l cyc le . T h i s i s i n contrast to a species such as ponderosa pine {Pinus ponderosa) that has differentiated to the l e v e l o f subspecies, suggesting that different parts o f the range have been isolated for more than one g l a c i a l c y c l e (Lat ta and M i t t o n , 1999). Inbreeding and genetic diversity - Other conifer species i n N o r t h A m e r i c a have exper ienced a reduct ion i n range size dur ing g lac ia t ion , but few have shown a reduct ion i n genetic d ivers i ty as severe as western redcedar. Excep t ions inc lude two conifer species, red p ine (Pinus resinosd) and Tor rey pine (Pinus torreyana), that have almost no i s o z y m e var ia t ion ( F o w l e r and M o r r i s , 1977; A l l e n d o r f et al, 1982; S i m o n et al, 1986; M o s s e l e r et al, 1991; L e d i g and C o n k l e , 1983). Gene t ic d ivers i ty is also associated w i t h a species ' ma t ing system. S e l f i n g and mixed -ma t ing species general ly show l o w e r genetic d ivers i ty than outcross ing species ( H a m r i c k and God t , 1996; Chapter 1). U n l i k e most conifers Thuja plicata shows h igh self-fert i l i ty and a m i x e d mat ing system (Chapters 2, 4 and 5). I found that popula t ions f rom throughout the range o f western redcedar showed a pos i t ive inbreeding coeff icient w i t h a mean inbreeding coeff icient o f F = 0 .085. The self ing rate, s, based on such a value w o u l d be s = 2(F I (1 + F)) = 0.156. T h i s is ac tual ly l o w e r than self ing rates observed i n natural populat ions (s = 29 % , O ' C o n n e l l et al, 2001 ; Chapter 2) and seed orchards o f western redcedar s = 25% ( K . R i t l a n d , unpubl ished) . T h i s indicates that a l though there is some e l imina t ion o f inbred ind iv idua l s after the seed census, many selfed seeds also surv ive and reproduce as trees. S i m i l a r l y , red pine (Pinus resinosd), has also shown a severe reduct ion i n genetic d ivers i ty at i s o z y m e l o c i . The species has been sampled through most o f its range i n eastern N o r t h A m e r i c a , and on ly four p o l y m o r p h i c l o c i have been observed (Hep = 0 .002 based on 27 enzyme systems 103 and 64 l o c i ; F o w l e r and M o r r i s , 1977; A l l e n d o r f et al, 1982; S i m o n et al, 1986; M o s s e l e r et al, 1991). L i k e redcedar, red pine also shows a potent ial l i n k between inbreeding and a reduct ion i n genetic d ivers i ty . T h e h igh self-fert i l i ty i n 46 trees f o l l o w i n g con t ro l led po l l ina t ions suggests that sel f ing is potent ia l ly h igh i n natural populat ions o f red p ine (Fowle r , 1965). A reduct ion i n popula t ion size i n western redcedar, f o l l o w e d b y forced mat ing w i t h relat ives and self-fer t i l iza t ion , and purg ing o f deleterious mutations probably added further to a reduct ion i n genetic d ivers i ty at neutral l i n k e d l o c i . 104 Chapter 7 Somatic mutations at microsatellite loci in western redcedar Introduction In plants, somatic mutations, i.e., mutations ar is ing f rom mi tos is , can be a s ignif icant source o f new genetic var ia t ion , both w i t h i n and between ind iv idua l s . Gene t i c var ia t ion w i t h i n i nd iv idua l s offers the opportuni ty for ce l l l ineage select ion (Otto and Has t ings , 1998) and c o u l d be important i n plant defense by creating a mosa ic o f different environments for insect pests ( W h i t h a m and S lobodch ikof f , 1981; A n t o l i n and Strobeck, 1985). A t the popula t ion leve l , somat ic mutations can potent ia l ly change al lele frequencies (Or ive , 2001) . Somat ic mutations are important i n the evo lu t ion o f plant mat ing systems, par t icular ly i n l o n g - l i v e d species such as forest trees, as they contribute to mutat ional l oad and inbreeding depression, f avor ing the evo lu t ion o f predominant ly outcross ing mat ing systems (Barrett et al., 1996; M o r g a n , 2001). Somat i c mutations can be detected at either genetic marker l o c i such as R A P D s (as i n aspen c lones Populus tremuloides; T u s k a n et al, 1996) or at conspicuous m o r p h o l o g i c a l l o c i such as c h l o r o p h y l l def ic iency (as i n s ix species o f Cupressaceae, K o m , 2001) . Mic rosa te l l i t e s , or s imp le sequence repeats ( S S R s ) , offer a specia l opportuni ty to observe and study somatic mutat ions, as their rate o f mutat ion is several orders o f magni tude greater than other D N A markers (E l legren 2000a). Mic rosa te l l i t e s consist o f tandemly repeated units o f D N A o f one to s ix base pairs, and their tandem nature results i n mutat ions due to rep l ica t ion s l ippage or s l ipped-strand mi spa i r i ng dur ing D N A repl ica t ion ( L e v i n s o n and G u t m a n , 1987). Mic rosa te l l i t e s are popular markers i n popula t ion genetic studies, as they are h igh ly var iable and co-dominant . U l t i m a t e l y , observat ions o f microsate l l i te mutations w i l l enable more accurate inferences based upon microsate l l i te mutat ion models , as such observations p rov ide in format ion about the sizes (change i n repeat number) and rates (numbers per mi tos is or per generation) o f microsate l l i te mutat ions. 105 Wes te rn redcedar (Thuja plicata D o n n ex D . D o n : Cupressaceae) is a coni fer w i t h a m i x e d mat ing system and l o w i s o z y m e divers i ty ( Y e h , 1988; E l - K a s s a b y et al. 1994; O ' C o n n e l l et al. 2001) . I n d i v i d u a l trees can l i v e up to 1,000 years and attain heights o f more than 50 m ( M i n o r e , 1990). Therefore, western redcedar provides a g o o d opportuni ty to detect and characterize new microsate l l i te mutations ar is ing v i a somatic processes. In this study, I sampled hap lo id megagametophyte tissue i n Thuja plicata to detect mutat ions i n a l o n g - l i v e d plant. A n advantage o f us ing megagametophytes to detect mutations is that they are part o f the g e r m l ine , so that the new mutations are heritable. Observat ions o f microsate l l i te mutations w i l l p rov ide informat ion on the generational somatic mutat ion rate and the magni tude o f s ize changes o f microsate l l i te mutations. Materials and methods Estimating mutation rate - M i t o t i c mutat ion rates can be est imated b y ca lcu la t ing the number o f c e l l d iv i s ions leading to a new mutat ion, but this i nvo lves several assumptions such as constancy o f c e l l sizes and the f ide l i ty o f ap ica l meris tems. In addi t ion , w h e n several samples are made throughout a tree, the uncertain o r ig in o f c e l l l ineages leading to different sampled tissues further compl ica tes these calcula t ions . Instead, i n this study I e m p l o y a s imple method to estimate per-generation mutat ion rates, as opposed to per-mitosis rates. O n a per-generation basis, the mutat ion rate is found b y s i m p l y observ ing the frequency o f new mutat ions i n seed-p roduc ing tissue. T h i s is s imi l a r to methods based upon the number o f genomes sampled (Schlbtterer et al. 1998; V a z q u e z et al. 2000; U d u p a and B a u m , 2001 ; V i g o u r o u x et al. 2002) . The mutat ion rate per locus per generation is est imated as U = m INLK, where m is the number o f mutations observed (number o f t imes that genetic differences among sampled tissues w i t h i n a tree was observed), N is the number o f tissues sampled per tree, L is the number o f trees sampled, and K is the number o f l o c i sampled. T h i s estimator is de r ived as fo l l ow s . I f u is the 106 expected per-generation somatic mutat ion rate, the dis t r ibut ion o f the number o f mutations found i n a sample are the terms i n the expansion o f LK(u + (l-u))N. I f u is sma l l , on ly two terms predominate: LK(l-u)N (trees w i t h no mutations) and LKNu(l-u)N'' = LKNu (trees w i t h one sample o f N mutant). The estimator is then obtained b y equating this latter te rm to the observed numbers o f mutants i n the total sample (m), and s o l v i n g for u. T h i s estimator for per-generation mutat ion rate makes two major assumptions: (1) that the seed-producing tissue sampled represents the h is tor ica l average age o f reproduct ion for the species, and (2) that new mutations are ident i f ied i n an unbiased manner. R e g a r d i n g (1), trees o f average mature age should be sampled. R e g a r d i n g (2), w e need to sample at least two tissues per tree to detect mutat ional changes. W e assume that mutant sectors are suff ic ient ly sma l l such that a l l samples are not a l l mutant; co l l ec t ion o f tissues at points mutua l ly separated b y the largest number o f mitoses should m i n i m i z e this poss ib i l i ty o f sampl ing on ly mutant tissues. O v e r a l l , to the extent that the trees represent the average age o f reproduct ion, this estimator w o u l d s l ight ly underestimate the true mutat ion rate, because o f the sl ight poss ib i l i ty that a l l samples w i t h i n a tree were new mutants. It does not matter i f a mutant sector was missed , as the expected estimate o f the frequency o f mutant sectors equals the observed fract ion o f trees w i t h mutant sectors; it does not matter that a l l mutant sectors have been sampled, on ly that they have been sampled i n an unbiased manner. W e also assume that mul t ip le independent mutat ions do not occur i n the same tree; g i v e n the re la t ive ly l o w mutat ion rate (ca. 1 i n 1000) this event should be h igh ly improbab le and not s igni f icant ly affect m y estimates. T h e 9 5 % confidence interval was calcula ted us ing the W i l s o n score method w i t h cont inui ty correct ion w h i c h is appropriate for samples sizes above 4 0 0 ( W i l s o n , 1927; N e w c o m b e , 1998). T h e lower and upper confidence l imi t s were ca lcula ted as: 107 2np + z2 -1 - zJz2 - 2 - 1/n + 4p(nq +1) L o w e r = — 2(n + z2) 2np + z2 +1 + zJz2 + 2 - l/n + 4p(nq - 1 ) Upper = - = 2(n + z2) where n is the sample size, p the propor t ion o f mutations observed (LO, q = 1 - p, and z is the standard N o r m a l deviate associated w i t h a two- ta i led probabi l i ty ( N e w c o m b e , 1998). Sample collections - D u r i n g the autumn o f 1999 mature cones were co l lec ted f rom a total o f 80 trees i n four natural populat ions i n southwestern B r i t i s h C o l u m b i a (20 trees/population; populat ions B C 1 2 , B C 1 3 , B C 1 4 and B C 1 5 , Chapter 6). Wes te rn redcedar trees produce seed cones throughout the c r o w n i n c l u d i n g the lower branches that often reached the g round (pers. obs.) C o n e s were co l lec ted f r o m reproduct ive trees ranging f rom 4.8 to 36.8 m i n height (average height = 20.1 m + 6.7 S D ) . In three o f the populat ions , cones were co l l ec ted f rom two branches f rom three different heights (top, midd le , and lower ) i n most trees, but on ly f rom the top and l o w e r branches i n the shorter trees ( F i g . 7.1). In one popula t ion cone co l lec t ions were made f rom one branch f rom each o f two posi t ions (top and lower ) . T h u s m y sampl ing attempted to m i n i m i z e the probabi l i ty that a l l samples w i t h i n a tree come f rom the same mutant sector ( i f the sector exists) . T o ensure that trees representative o f the average age o f reproduct ion were sampled, I also attempted to sample reproduct ive trees spanning a l l heights i n a popula t ion f rom the shortest and the tallest. T h e trees sampled are therefore representative o f the range i n size o f mature redcedar trees i n the geographica l area studied so that a mutat ion rate per tree, based on these trees, is reasonable. Seeds were mechan ica l ly extracted f rom cones and stored at 4 ° C un t i l germina t ion . Seeds were germinated f o l l o w i n g O ' C o n n e l l et al. (2001), and hap lo id megagametophytes were 108 separated f rom the seedlings. F r o m each co l l ec t ion pos i t ion ten megagametophytes were bu lked . A total o f 20 to 60 megagametophytes per tree were sampled, and D N A was extracted us ing a m o d i f i e d C T A B method ( D o y l e and D o y l e , 1987). T h i s use o f megagametophytes effect ively captures the somatic tissue just p r io r to its entrance into the g e r m l ine (new mutat ions a r i s ing f r o m meios i s are not detected i n these bu lks ) (See Chapter 4). Microsatellites - E a c h sample o f 10 b u l k e d megagametophytes per branch was genotyped at eight p o l y m o r p h i c microsate l l i te l o c i deve loped for Thuja plicata ( O ' C o n n e l l and R i t l a n d , 2000 ; Chapter 3). Mic rosa t e l l i t e repeat motifs ranged f rom s imple d inucleot ide repeats to more c o m p l e x and interrupted motifs (Table 7.1). P C R reactions and a l le le scor ing on a L I - C O R 4 2 0 0 sequencer ( L I - C O R Inc., L i n c o l n , Nebraska) were car r ied out as descr ibed i n O ' C o n n e l l and R i t l a n d (2000) and Chapter 3. Table 7.1 D e s c r i p t i o n o f eight microsatel l i te l o c i used to genotype 80 Thuja plicata trees f rom four natural populat ions . N, number o f d inucleot ide repeats; bp, range o f al lele lengths i n base pairs; A, total number o f al leles detected. L o c u s Repeat m o t i f N bp A T P 1 ( C A ) N 13-34 166-208 12 T P 3 ( T G ) N 9-36 178-232 15 T P 4 ( T G ) N 10-23 282-308 9 T P 6 ( G C ) N ( G T ) N ( A T A T G T ) N . . . ( G T ) N 78-110 231-295 28 T P 7 ( C A ) N 11-29 241-277 15 T P 8 ( C A ) N C G ( C A ) N 23-49 208-266 21 T P 9 ( A C ) N 20-59 263-341 27 T P 1 1 ( C T ) N ( C A ) N 25-33 220-232 7 109 Results Microsatellite mutations - Af te r screening the mater ial at eight microsatel l i te l o c i , I found a single new al lele at locus T P 9 . A l l e l e s 281 and 291 were found i n the lower and m i d part o f a tree and al leles 281 and 293 i n the top part ( F i g . 7.1). T h e new al lele (293) l i k e l y arose i n the upper part o f the tree. T o c o n f i r m that the new al le le was not a P C R artifact a l l the samples f rom the tree w i t h the new al lele were re -ampl i f ied and re-scored f rom the same D N A extract ion. T h e same al lele sizes were observed each t ime. Locus TPS Seed eotoetion Mid Top position Fig. 7.1 Image o f a microsate l l i te ge l showing the genotype at locus T P 9 for two different heights w i t h i n the same tree (left). T w o col lec t ions o f ten b u l k e d megagametophytes were made f rom three heights i n each tree (right). 281 bp = 29 dinucleot ide repeats, 291 bp = 34 repeats and 293 = 35 repeats. 110 Type of mutation - T h e size o f the new al le le corresponded to an increase i n one d inucleot ide repeat: f r om 34 to 35 repeats. T h e new al le le s ize already exis ted i n the sampled populat ions and was near the m i d d l e o f the dis t r ibut ion o f a l le le sizes at locus T P 9 ( F i g . 7.2). Somatic mutation rate estimate - T w o posi t ions were sampled i n 42 trees and three posi t ions were sampled i n 38 trees. U s i n g the above est imator and its assumptions, the estimated rate o f somatic mutat ion rate is U = m INLK = 1 / ((2 x 42 x 8) + (3 x 38 x 8)) = 1 / 1584 = 6.3 x 10"4 mutations per locus per generation w i t h a 9 5 % conf idence in terval o f 3.0 x 10"5 to 4 .0 x l O " 3 . •.-3JQ;.,r . Mi - I Number I 1 I I " i & -I I I I Fig 7.2 A l l e l e d is t r ibut ion at locus T P 9 over four popula t ions o f Thuja plicata (N = 80 trees). T h e new al lele , w h i c h increased f rom 34 to 35 d inucleot ide repeats, is indica ted by the whi te box , w i t h the arrow s h o w i n g the o r ig ina l al lele . I l l Discussion Somatic mutation rate -1 observed a s ingle somatic mutat ion occur r ing in the upper c r o w n o f a redcedar tree. T h e estimated mutat ion rate o f 6.3 x 10"4 per locus per generation (or 3.1 x 10"4 per a l le le per generation) is w i t h i n the expected range o f 10"3 to 10"4 mutations per generation general ly reported for microsatel l i tes (E l l eg ren , 2000a). In plants, microsate l l i te mutations rates have been estimated f rom mutations accumula ted i n inbred l ines. In maize (Zea mays subsp. mays) the estimated mutat ion rate for 142 microsate l l i te l o c i was 7.7 x 10"4 mutations per a l le le per generation ( V i g o u r o u x et al., 2002) . U d u p a and B a u m (2001) est imated microsate l l i te mutat ion rates o f 1.0 x 10"2 and 3.9 x 10"3 mutations per a l le le per generation i n t w o annual varieties o f ch i ckpea (Cicer arietinum: Fabaceae) . These methods w o u l d have captured both somatic and me io t i c mutations and thus mutat ion rates should be higher for somat ic mutations on ly . T h e actual mutat ion rate per generation i n western redcedar is l i k e l y h igher then reported here, because the method I used was not sensit ive enough to detect me io t i c mutat ions. A l t h o u g h the mater ia l sampled was part o f the ge rm l ine and offered the opportuni ty to detect me io t i c mutat ions, this was d i f f icu l t because the megagametophyte mater ia l was b u l k e d . Resul t s f r om Chapter 4 showed that al leles occur r ing at a l o w frequency, such as a new mutat ion ar i s ing dur ing meios is , w o u l d l i k e l y not be detectable i n a b u l k o f 10 megagametophytes. S t r ic t ly speaking, a generat ion 's wor th o f somatic g rowth should mean something quite precise i f w e take the demographic def in i t ion o f the term "generat ion" into account. It should reflect the su rv ivorsh ip and fecundi ty schedule o f the popula t ion . These values are quite d i f f icu l t to estimate outside o f a con t ro l led setting and for such a l o n g - l i v e d plant. D u r i n g m y sampl ing I attempted to sample f rom trees representative o f the age structure o f reproduct ive ind iv idua l s i n a popula t ion to approximate a generation. 112 Mutation model - Informat ion on the mutat ion processes o f microsate l l i tes w i l l help p rov ide more accurate mutat ion models for popula t ion genetics. D i s t ance measures and t i m i n g o f evolu t ionary events depend on accurate mutat ion models . T h e observed mutat ion was stepwise, increas ing i n size b y one base pa i r ( F i g . 7.2). S i m i l a r l y , o f 71 observed microsatel l i tes mutations i n maize , V i g o u r o u x et al. (2002) found that changes o f a s ingle repeat ( 8 3 % o f mutations) were more c o m m o n than mul t ip le repeats (17%), and a higher propor t ion o f alleles mutated to a larger a l le le (79%) than a smal ler a l ler (21%). T h e same d i rec t iona l biases have been reported i n birds and humans (P r immer et al. 1996; E l l e g r e n , 2000b) . Mic rosa te l l i t e s are k n o w n to exhib i t extensive homoplasy (unrelated al leles o f the same size) , and cor respondingly the new al lele i n this study mutated to an already ex i s t ing a l le le size i n the popula t ions sampled. Different l o c i , and even different al leles, p robab ly mutate at different rates (E l l eg ren 2000a). Muta t ions seem to occur i n longer l o c i or al leles, and at l o c i w i t h s imple repeats vs more c o m p l e x l o c i . The locus w i t h the observed mutat ion, T P 9 , is one o f the most var iable l o c i i n western redcedar, w i t h 27 al leles detected i n 80 ind iv idua l s (Table 7.1). In a range-wide study, 41 al leles were detected i n 620 ind iv idua l s at locus T P 9 (Chapter 6). Changes i n a l le le size can also be caused by changes i n the D N A regions f l ank ing the microsate l l i te . T h i s is u n l i k e l y i n this study because the new al le le corresponded exact ly to an increase i n one d inucleot ide repeat ( two extra base pairs) longer than the o r ig ina l al lele . The consequences of somatic mutations in redcedar - Genera t ion t ime can be anywhere f r o m 30-1000 years i n Thuja plicata. A h igh per-generation mutat ion rate at microsate l l i te l o c i can account for the h igh divers i ty at these markers compared to other markers , despite there be ing re la t ive ly few generations since a popula t ion bott leneck dur ing the last g lac ia t ion (Chapter 6). In a l o n g - l i v e d species such as redcedar, deleterious somat ic mutations also have the potent ial o f increas ing inbreeding depression. 1 Genetic mosaicism - The new al le le was found i n megagametophytes co l l ec ted f rom branches at the top o f the tree, but not i n any o f the l o w e r co l lec t ions . K o r n (2001) showed evidence o f a s ingle ap ica l i n i t i a l c e l l i n several Cupressaceae species. T h e mutat ion probably occur red i n the ap ica l c e l l , and a l l cel ls above this ap ica l mer i s tem conta ined the new al lele leading to a large sector w i t h a n o v e l genotype. Tha t trees can be mosa ics o f ce l l s o f different genotypes poses interesting questions about plant defense strategies and plant evo lu t ion ( W h i t h a m and S lobodch ikof f , 1981; A n t o l i n and Strobeck, 1985; G i l l 1995). 114 Chapter 8 General discussion and conclusions In previous studies, meta-analyses o f the literature have helped out l ine traits associated w i t h l o n g - l i v e d w o o d y plants, such as l o w genetic d ivers i ty , h igh inbreeding depression and h igh outcross ing rates. It is by s tudying var ia t ion o f these quantities and their associat ion w i t h i n a species, however , that w i l l ga in a better understanding o f the h o w they are interconnected. T o better understand the evo lu t ion o f self ing i n a l o n g - l i v e d plant, Thuja plicata, I have assembled three pieces o f a puzz le : its ma t ing system, l eve l o f inbreeding depression, and genetic structure. T o p lace western redcedar i n a phylogenet ic context, I first compared its l eve l o f se l f ing and genetic d ivers i ty to other species o f conifers (Chapter 1). A l t h o u g h western redcedar seems to stand out among trees i n terms o f its l o w genetic d ivers i ty and h i g h self ing rates, it is not except iona l among more c lose ly related conifers . In fact, its closest relat ive, Thuja occidentalis, also showed l o w genetic d ivers i ty and some o f the highest levels o f inbreeding among conifers (Chapter 1). M y study focussed on the evolu t ionary history o f western redcedar, and h o w it has l ed to the present mat ing system, patterns o f genetic d ivers i ty and inbreeding depression. I have shown that self-fer t i l izat ion e v o l v e d w i t h reduced inbreeding depression and reduced genetic d ivers i ty . Main findings Mating system - P rev ious ly , outcross ing rates i n western redcedar were on ly avai lable for a s ingle seed orchard popula t ion . I obtained estimates o f inbreeding for six natural populat ions o f redcedar us ing i sozymes (Chapter 2). I found s ignif icant amounts o f inbreeding, and popula t ion outcross ing estimates ranged f rom 17% to >100% (weighted mean 71%) . Outc ross ing rates differed s igni f icant ly among populat ions, and the results o f this study suggested that eco log ica l differences among popula t ions and among trees w i t h i n populat ions 115 were part ly responsible for the var ia t ion i n outcross ing rates. T o examine the mat ing system o f redcedar at a f iner scale, I deve loped new h igh ly var iable microsate l l i te markers (Chapter 3). I used microsatel l i tes to study the var ia t ion i n outcross ing rates w i t h i n an among redcedar trees (Chapter 4). There was no difference i n outcross ing among c r o w n posi t ions w i t h i n trees. In contrast, i n d i v i d u a l tree outcross ing rates decreased w i t h tree height i n a l l four populat ions surveyed. T h i s c o u l d be the result o f larger trees r ece iv ing more se l f -pol len as their po l l en makes up a larger propor t ion o f the surrounding po l l en c l o u d . U n l i k e many other conifers , redcedar was p rev ious ly shown to be h igh ly self-fertile (Owens et al, 1990). I set out to test whether post-fer t i l izat ion mechanisms c o u l d increase the propor t ion and number o f outcrossed seedlings f rom a redcedar tree (Chapter 5). T o m y knowledge this is the first study that has tested for early e mbr yo compet i t ion i n a conifer . W h e n a h igh propor t ion o f se l f -pol len (75%) was appl ied , unrelated po l l en was favoured, suggesting that embryo compet i t ion increased the propor t ion o f outcrossed seeds. There was no evidence that p o l y e m b r y o n y increased seed set i n redcedar but the power to detect an increase was l o w . A l t e rna t i ve ly , i f a late act ing se l f - incompat ib i l i ty mechan i sm f o l l o w i n g e mbr yo compet i t ion is responsible for seed death (see W i l l i a m s et al, 2001) p o l y e m b r y o n y should not affect seed set. Inbreeding depression - Inbred ind iv idua l s can be e l imina ted at different stages o f the l i f e -cyc le , either through mechanisms that promote outcross ing or through the reduced v i ab i l i t y o f inbred ind iv idua l s due to recessive deleterious mutations. Measures o f inbreeding, c o m b i n e d w i t h the propor t ion o f s u r v i v i n g ind iv idua l s f r om several life-stages in western redcedar, a l l o w e d me to estimate inbreeding depression. A t the seed stage, I ca lcula ted a measure o f self-fert i l i ty f rom the propor t ion o f fu l l seeds f o l l o w i n g se l f -pol l ina t ion vs c ross -pol l ina t ion (Chapter 5). I found that i n d i v i d u a l redcedar trees var ied w i d e l y i n the propor t ion o f v iab le seeds set after self-fer t i l iza t ion (Chapter 5). Three out o f four trees showed about a 3 0 % reduct ion i n seed set 116 f o l l o w i n g se l f -pol l ina t ion compared to c ross -pol l ina t ion . O n e tree, however , showed a 9 3 % reduct ion i n f u l l seeds i n selfed vs outcrossed treatments, setting on l y 7/200 f u l l seeds after self ing. A t the seedl ing stage, I obtained an estimate o f inbreeding depression based on o f the propor t ion o f selfed seedlings, relat ive to the amount o f se l f -pol len app l i ed (Chapter 5). Est imates o f inbreeding depression ranged f rom -0.533 w i t h 2 5 % se l f -pol len to 0.618 w i t h 7 5 % sel f po l l en . I also obtained self ing rates (s) measured at the seedl ing stage i n natural populat ions (Chapter 2) and estimates o f the number o f inbred ind iv idua l s that su rv ived to adul thood through popula t ion inbreeding coefficients (F ) (Chapter 6). B y us ing these last two quantities (s and F), and assuming that they are constant over t ime, inbreeding depression caus ing mor ta l i ty between the seedl ing and adult stage can be estimated i n natural populat ions . Pos i t i ve inbreeding coefficients i n natural populat ions indicate that at least some inbred trees surv ive to maturi ty . T o obtain an inbreeding coeff icient o f F = 0.085 (Chapter 6) se l f ing rates o f about s = 2(F I (1 +F)) = 16% are expected. H o w e v e r , the measured self ing rate at the seedl ing stage in natural popula t ions is a lmost twice that (s - 29%; Chapter 2). T h e adult inbreeding coefficients obtained are l o w e r than expected i f a l l selfed seeds survive as trees. There is p robably e l imina t ion i n inbred ind iv idua l s i n natural populat ions f o l l o w i n g the seed stage du r ing germinat ion , seedl ing establishment and early growth . Inbreeding depression between the seedl ing and adult stages is measured as: 5 = 1 - wsl w0 and, w s / w 0 = 2 x [(1 - s) F) I s (1 - F ) ] , where ws is the fitness o f selfed trees and w0 is the fitness o f outcrossed trees as expressed b y the number o f su rv iv ing ind iv idua l s at maturi ty (R i t l and , 1990b). The estimate o f inbreeding depression (5) caus ing the death o f i nd iv idua l s between the seedl ing and adult stage is about 5 5 % . A l t h o u g h self-fert i l i ty i n western redcedar is h i g h for a conifer , a large propor t ion o f the inbred ind iv idua l s are s t i l l e l imina ted after the seed stage in natural populat ions . L e t h a l mutations w i l l cause the death o f i nd iv idua l s early i n the l i fe cyc l e w h i l e late act ing inbreeding depression such as a 10% reduct ion i n g rowth rate i n redcedar (J. R u s s e l l , unpubl i shed data) w i l l s t i l l e l iminate inbred 117 ind iv idua l s that are eventual ly outcompeted by faster g r o w i n g trees. L a s t l y , I found an heritable mutat ion that arose i n the somatic tissue o f a redcedar tree, s h o w i n g the potent ial to accumulate a h igher genetic load i n a l o n g - l i v e d o rgan i sm through somatic mutations, compared to a short l i v e d o rgan i sm (Chapter 7) . Genetic structure and diversity - B y deve lop ing h igh ly p o l y m o r p h i c microsate l l i te markers for Thuja plicata, I deve loped a too l for addressing patterns o f genetic d ivers i ty i n a species w i t h l o w genetic d ivers i ty at other markers (i.e., leaf o i l terpenes, i sozymes and R F L P s ) (Chapter 3). Cur ren t patterns o f genetic structure i n a species are a combina t ion o f past events and current levels o f gene f l o w . B y s tudying the genetic structure o f redcedar, I have been able to elucidate h is tor ica l events w h i c h l ed to a reduct ion i n the l eve l o f d ivers i ty . Mic rosa te l l i t e s con f i rmed the l o w inter-populat ion differentiat ion observed w i t h other genetic markers. A study o f range-wide genetic structure w i t h microsatel l i tes also revealed more than one g l ac i a l re fugium for western redcedar (Chapter 6). T h i s study is the clearest evidence o f mul t ip l e coastal refugia for a tree species i n western N o r t h A m e r i c a . Because Thuja plicata is often a dominant species i n m a n y coastal forests, other associated species w i l l p robably show a s imi l a r pattern o f genetic structure. T h e al lele frequency dis t r ibut ion patterns over a l l microsate l l i te l o c i showed that western redcedar has been expand ing i n size f rom sma l l bot t lenecked populat ions and has accumula ted a large number o f new alleles through mutat ion (Chapter 6). T h e l o w , range-wide d ivers i ty i n redcedar, c o m b i n e d w i t h the evidence o f mul t ip l e refugia, suggest that a species-w i d e bot t leneck probably predates the last g lac ia t ion . The evidence o f an expand ing popula t ion suggests the potent ial for recurrent bot lenecks w i t h each g l ac i a l cyc l e . Gene t ic structure and d ivers i ty patterns ou t l ined us ing microsate l l i te l o c i w i l l be affected b y the mode and rate o f mutat ion o f these markers . I observed a s ingle mutat ion at a microsate l l i te locus cor responding to a single-step increase i n al lele size i n western redcedar. I 118 est imated a somat ic (mitot ic) mutat ion rate for western redcedar o f 6.3 x 10"4 mutations per locus per generation. (Chapter 7). The type and rate o f mutat ion observed f o l l o w e d the pattern expected for microsatel l i tes (E l l eg ren 2002a). Tying it all together - A reduct ion i n species-wide genetic d ivers i ty i n western redcedar was probably not due to a s ingle event (i.e., a bott leneck), but the interact ion o f a reduct ion in popula t ion s ize w i t h an inbreeding system o f mat ing. Other species o f Thuja a lso share a m i x e d mat ing system, but not such a severe reduct ion i n genetic d ivers i ty (see Chapter 1). T h i s leads us to be l ieve that inbreeding is not a direct consequence o f a bott leneck, but may have further accentuated the decrease i n genetic d ivers i ty i n redcedar. H i g h levels o f inbreeding are also found i n Thuja occidentalis, the closest related species to T. plicata, as w e l l as other species o f the Cupressaceae ( l is ted i n R i t l a n d et al, 2001) . T h e h igh self fert i l i ty o f Thuja plicata (Chapter 5) and h igh levels o f se l f fer t i l iza t ion measured i n natural popula t ions o f T. plicata and T. occidentalis (Chapter 1), leads us to conc lude that h i g h levels o f inbreeding have contr ibuted to a reduct ion i n genetic d ivers i ty i n both these species. E c o l o g i c a l traits i n western redcedar that differ f rom other conifers and that c o u l d further exp la in some differences i n patterns o f inbreeding inc lude shade tolerance and a c o l o n i z i n g l ife strategy ( M i n o r e , 1990). I n d i v i d u a l redcedar trees w i t h s lower g rowth due to inbreeding are not e l imina ted as readi ly as i n other species because they can g row i n the shade wi thout d y i n g . T h i s a l l ows some inbred adults to surv ive to maturi ty and reproduce. T h e evo lu t ion o f a m i x e d -mat ing sys tem i n Thuja plicata c o u l d also be due to the need for reproduct ive assurance i n an uncertain environment . Wes te rn redcedar usua l ly occurs i n mixed-spec ies stands so trees must be able to self-fert i l ize when conspeci f ics are rare. 119 Further research Review of mating systems in trees - The literature r ev iew o f outcross ing rates and genetic d ivers i ty i n trees presented i n Chapter 1 (and A p p e n d i c e s II & III) c o u l d be expanded by i n c l u d i n g other parameters. Outc ross ing rates for several tree species are not avai lable but indirect measures o f inbreeding can be obtained through the inbreeding coeff icient (F). R i t l a n d et al, (2001) l is ted a number o f species i n the Cupressaceae w i t h pos i t ive inbreeding coefficients suggesting that m i x e d mat ing is c o m m o n i n this f ami ly . H o w e v e r , there is no comprehens ive r e v i e w o f inbreeding coefficients to test whether pos i t ive //-estimates are unusual for trees. T h e combina t ion o f popula t ion inbreeding coefficients and outcrossing rates can also be used to estimate inbreeding depression w i t h i n natural populat ions i f these values are assumed to be constant over t ime (Ri t l and , 1990b). H o w e v e r , F also measures popula t ion substructure and can be inf lated by l i m i t e d dispersal and f ami ly structure. Genet ic d ivers i ty and a mixed -ma t ing system seem to be traits associated w i t h a longer l i fespan and woodiness rather than the t axonomic group i n w h i c h a species is found (i.e., G y m n o s p e r m s vs A n g i o s p e r m s ) (Barrett and Eckre t , 1990; H a m r i c k and G o d t , 1996; Barret t et al, 1996). I nc lud ing more ang iosperm trees i n a r ev iew o f mat ing systems may b r ing out other patterns associated w i t h inbreeding. A n o t h e r quantity w h i c h is increas ingly be ing reported in studies o f mat ing systems i n trees is the correla t ion o f paternity (rp), def ined as the propor t ion o f fu l l sibs among outcrossed sibs (R i t l and , 1989). Est imates o f rp for a few conifer species are l is ted i n Chapter 2, and a more comprehens ive r ev iew o f this quantity i n trees w o u l d p rov ide informat ion on the source o f the paternal cont r ibut ion to seeds. Glacial refugia - Patterns o f microsate l l i te d ivers i ty across the species range p rov ide strong evidence that western redcedar su rv ived i n at least three g l ac i a l refugia dur ing the last ice age (Chapter 6). T o c o n f i r m the locat ions o f g l ac i a l refugia and zones o f secondary contact in 120 western redcedar, uniparental ly inheri ted, non- recombin ing and h igh ly var iable genetic markers such as chloroplas t microsatel l i tes ( c p S S R ) c o u l d be used. These markers have been used to retrace pos t -g lac ia l movemen t and popula t ion dynamics i n another genet ica l ly depauperate species, red pine (Pinus resinosa; E c h t et al, 1998) as w e l l as other conifer species (e.g. Pinus contorta, M a r s h a l l et al, 2002) . Spec i f i c geographic locat ions to concentrate on i n redcedar w o u l d be areas o f possible secondary contact between different refugia. These inc lude populat ions a long the coast o f the B r i t i s h C o l u m b i a (between S q u a m i s h and B e l l a C o o l a on the main land , as w e l l as on V a n c o u v e r Island) and popula t ions i n northern C a l i f o r n i a to central Oregon . N o c p S S R markers presently exist for Thuja plicata so these markers w o u l d first need to be developed . T h e s imi la r i ty i n the genetic structure o f several species w i t h s imi la r geographic d is t r ibut ion patterns w i l l help identify and c o n f i r m the locat ion o f g l ac i a l refugia. T h e data for western redcedar increases the l is t o f plant species for w h i c h range-wide phy logeograph ic data is avai lable i n western N o r t h A m e r i c a (Sol t is et al, 1997). Other conifer species that share a s imi l a r geographic pattern inc lude S i t k a spruce (Picea sitchensis), ye l low-ceda r (Chamaecyparis nootkatensis), western h e m l o c k (Tsuga heterophylla) and pac i f i c y e w (Taxus brevifolia) (Burns and H o n k a l a , 1990). A l t h o u g h i s o z y m e data suggest mul t ip l e refugia for some o f these species, more p o l y m o r p h i c markers can prov ide a better resolut ion. Allele distribution - O n e method that can be used to estimate the t ime since the last bot t leneck is to assume that d ivers i ty at a l l l o c i was reduced to a s ingle al lele (Meno t t i and O ' B r i e n , 1993). T h i s method may be used w i t h i sozymes , but it is unreal is t ic to assume that on ly one a l le le per locus su rv ived a bott leneck at microsate l l i te l o c i because o f their h igh var iab i l i ty . It is more l i k e l y that a few o f the more c o m m o n alleles su rv ived at most l o c i . T h e a l le le size d is t r ibut ion pattern i n redcedar ranged f rom a s imple u n i m o d a l d is t r ibut ion at one locus to 121 b i m o d a l and m u l t i m o d a l dis tr ibut ions at other l o c i (Chapter 6). M u l t i m o d a l dis t r ibut ions c o u l d be a consequence o f genetic drift, popula t ion bott lenecks or occas iona l mult is tep mutat ions (Two-phase mutat ion model ) . S imula t ions o f bott lenecks w o u l d be useful for understanding their effect on al lele size dis t r ibut ion i n western redcedar: for example , to test whether bot t lenecks lead to a m u l t i m o d a l dis t r ibut ion i n al lele frequencies. Mating system at the edge of the distribution - Nor the rn popula t ions showed l o w inbreeding coefficients compared to more southern popula t ions (Chapter 6). M a t i n g sys tem dynamics at the edge o f the range might be quite different f r om the central populat ions . Est imates o f outcross ing have on ly been obtained for southwestern B r i t i s h C o l u m b i a populat ions . It w o u l d be interesting to see i f levels o f self ing or inbreeding depression i n northern popula t ions differ f r om populat ions at the centre o f the range. Related species - M a n y general izat ions about conifer mat ing systems have been predominant ly based on species i n the genus Pinus (Pinaceae). Other genera and fami l ies o f conifers may show higher levels o f inbreeding, self-fert i l i ty and/or l o w e r genetic d ivers i ty . M o r e data on other species o f Thuja and other members o f the Cupressacea m a y show that redcedar is not an out l ier among conifers . L e v e l s o f inbreeding have been est imated for very few conifer species outside the Pinaceae, so no conc lus ion can be reached about the mat ing system o f the Cupressaceae i n general . R i t l a n d et al. (2001) l i s ted species i n the Cupressaceae that had pos i t ive inbreeding coefficients suggesting a mixed -ma t ing system i n these species. A l t h o u g h the ext remely l o w levels o f self-fert i l i ty observed i n several conifers are often attributed to genetic load , w e migh t actual ly be observ ing late act ing se l f - incompat ib i l i ty mechanisms ( W i l l i a m s et al., 2001; J . Owens , pers. c o m m . ) The Cupressaceae m a y lack this hypothes ized se l f - incompat ib i l i ty mechan i sm found i n the Pinaceae a l l o w i n g them to inbreed. Tes t ing 122 whether p o l y e m b r y o n y promotes an increase i n seed set and outcross ing i n other groups o f conifers m a y y i e l d very different results f r om those observed i n Thuja plicata. F i n a l l y , microsatel l i tes deve lop for one species can often be used for other related species ( K a r h u et al, 2000). T h e microsatel l i tes I deve loped for Thuja plicata migh t be used to study the mat ing sys tem and genetic structure o f Thuja occidentalis and other related species. D a t a on the re la t ionship between mat ing system and genetic structure i n related species w i l l p rov ide informat ion on the longer- term dynamics o f these quantities. 123 References A d a m s , J . , M . M a s l i n , and E . Thomas . 1999. Sudden c l imate transitions dur ing the Quaternary. Progress i n P h y s i c a l Geography 23: 1-36. A d a m s , W . T . , and D . S. B i r k e s . 1991. Es t ima t ing mat ing patterns i n forest tree populat ions . In B i o c h e m i c a l markers i n the popula t ion genetics o f forest trees. S. F i n e s c h i , M . E . M a l v o l t i , F . Cannata , and H . H . Hat temer (Eds.) . S P B A c a d e m i c P u b l i s h i n g , The Hague , T h e Netherlands, pp. 157-172. A l l e n d o r f , F . W . , K . L . K n u d s e n , and G . M . B l a k e . 1982. Frequencies o f n u l l al leles at enzyme l o c i in natural populat ions o f ponderosa and red pine. Gene t ics 100: 497-504 . A l l y , D . , Y . A . E l - K a s s a b y , and K . R i t l a n d . 2000. Gene t ic d ivers i ty , differentiat ion and mat ing system i n mounta in h e m l o c k (Tsuga mertensiana) across B r i t i s h C o l u m b i a . Forest Genet ics 7: 97-108. A n t o l i n , M . F . , and C . Strobeck. 1985. T h e popula t ion genetics o f somat ic mutat ion i n plants. A m e r i c a n Natura l i s t 126: 52-62. Aus te r l i t z , F . , S. Mar ie t te , N . M a c h o n , P . - H . G o u y o n , and B . G o d e l l e . 2000 . Effects o f co lon iza t ion processes on genetic d ivers i ty : differences between annual plants and tree species. Genet ics 154: 1309-1321. B a l l o u x , F . , and N . L u g o n - M o u l i n . 2002. The est imat ion o f popula t ion differentiat ion w i t h microsate l l i te markers. M o l e c u l a r E c o l o g y 11: 155-165. Barrett , J . W . , P . K n o w l e s , and W . M . C h e l i a k . 1987. The mat ing system i n a b l ack spruce c lona l orchard. Canad ian Journal o f Forest Research 17: 379-382. Barrett , S. C . H . , and C . G . Ecker t , 1990. V a r i a t i o n and evo lu t ion o f mat ing systems i n seed plants. In B i o l o g i c a l Approaches and E v o l u t i o n a r y Trends i n Plants . S. K a w a n o (Ed.) A c a d e m i c Press, L o n d o n , pp. 229-254. Barrett , S. C . H . , L . D . Harder , and A . C . W o r l e y . 1996. T h e compara t ive b i o l o g y o f po l l ina t ion and mat ing i n f l ower ing plants. P h i l o s o p h i c a l Transact ions o f the R o y a l Soc ie ty o f L o n d o n , Series B 351 : 1271-1280. Bar ton , N . H . , and G . M . Hewi t t . 1985. A n a l y s i s o f h y b r i d zones. A n n u a l R e v i e w o f E c o l o g y and Systemat ics 16: 113-148. Ba teman , A . J . 1956. C r y p t i c se l f - incompat ib i l i ty i n the w a l l f l o w e r : Cheiranthus cheiri L . Hered i ty 10: 257-261 . B a w a , K . S. 1974. B r e e d i n g systems o f tree species o f a l o w l a n d t rop ica l c o m m u n i t y . E v o l u t i o n 85-92. B e a u l i e u , J . , and J . P . S i m o n . 1994. Gene t ic structure and var iab i l i ty i n Pinus strobus i n Quebec . Canad ian Journa l o f Forest Research 24: 1726-1733. B e a u l i e u , J . , and J . P . S i m o n . 1995. M a t i n g system i n natural popula t ions o f eastern whi te p ine i n Quebec . Canad ian Journal o f Forest Research 25: 1697-1703. 124 B e l k h i r , K . , P . B o r s a , J . Goudet , L . C h i k h i , and F . B o n h o m m e . 2000. G E N E T I X vers ion 4.02. Labora to i re G e n o m e , Popula t ions , Interactions, C N R S U P R 9060, Un ive r s i t e de M o n t p e l l i e r II, M o n t p e l l i e r , France . B l a c k W . C . , and E . S. Kra f sur . 1985. A F O R T R A N program for the ca lcu la t ion and analysis o f two- locus l inkage d i s e q u i l i b r i u m coefficients. Theore t i ca l and A p p l i e d Genet ics 70: 4 9 1 -496 . B o s c h e r i n i , G . , G . G . V e n d r a m i n , and R . G i a n n i n i . 1993. M a t i n g sys tem analysis i n two I tal ian populat ions o f N o r w a y spruce. Journal o f Genet ics and B r e e d i n g 47: 45-48 . B o y l e , T . J . , and E . K . Morgens te rn . 1986. Est imates o f outrcross ing rates i n s ix populat ions o f b l ack spruce i n central N e w B r u n s w i c k . S i l v a e Gene t i ca 35: 102-106. B o y l e , T . J . B . , C . L i e n g s i r i , and C . P i e w l u a n g . 1990. Gene t i c structure o f b l ack spruce on two contrast ing sites. Hered i ty 65: 393-389. B o y l e , T . J . B . , C . L i e n g s i r i , and C . P i e w l u a n g . 1991. Gene t ic studies i n a t ropica l pine - Pinus kesiya III. T h e mat ing system in four populat ions f rom northern Tha i l and . Journa l o f T r o p i c a l Forest Sc ience 4: 37-44. B o w e r , R . C . , and B . G . D u n s w o r t h , 1988. Provenance test o f western red cedar on V a n c o u v e r Is land. In Wes te rn red cedar—does it have a future? N . J . S m i t h (Ed.) 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 , Facu l ty o f Forestry, V a n c o u v e r , Canada , pp 131-135. B runs f e l d , S. J . , J . S u l l i v a n , D . E . Sol t i s , and P . S. So l t i s . 2001 . C o m p a r a t i v e phylogeography o f north-western N o r t h A m e r i c a : a synthesis. In Integrating e c o l o g i c a l and evolu t ionary processes i n a spatial context. J . S i l v e r t o w n and J . A n t o n o v i c s (Eds.) B l a c k w e l l Sc ience . O x f o r d , U . K . pp. 319-339. B u r c z y k , J . 1991. T h e mat ing system i n a Scots p ine c l o n a l seed orchard i n P o l a n d . A n n a l e s des Sciences Forest ieres 48: 443-451 . B u r c z y k , J . , W . T . A d a m s , and J . Y . S h i m i z u . 1996. M a t i n g patterns and po l l en dispersal i n a natural knobcone p ine (Pinus attenuata L e m m o n . ) stand. Hered i ty 77: 251-260. B u r c z y k , J . , and W . C h a l u p k a . 1997. F l o w e r i n g and cone product ion va r i ab i l i ty and its effect on parental balance i n a Scots p ine c l o n a l seed orchard. A n n a l e s des Sc ience Forestieres 54: 129-144. B u r n s , R . M . , and B . H . H o n k a l a (Eds. ) . 1990. S i l v i c s o f N o r t h A m e r i c a , V o l u m e 1. Coni fe rs . U . S . D . A . Forest Se rv ice A g r i c u l t u r e H a n d b o o k 654. W a s h i n g t o n , D C . 675 pps. Butcher , P . A . , J . C . B e l l , and G . F . M o r a n . 1992. Patterns o f genetic d ivers i ty and nature o f the breeding sys tem i n Melaleuca alternifolia (Myr taceae) . A u s t r a l i a n Journa l o f B o t a n y 40: 365-375. B y e r s , D . L . , and D . M . W a l l e r . 1999. D o plant populat ions purge their genetic load? Effects o f popula t ion s ize and mat ing history on inbreeding depression. A n n u a l R e v i e w i n E c o l o g y and Systemat ics 30: 479-513 . 125 Casgra in , P . , and P . Legendre . 2001 . The R - P a c k a g e for mul t ivar ia te and spatial analysis , vers ion 4.0 d6. User ' s M a n u a l . Depar tement de sciences b io log iques , U n i v e r s i t y de M o n t r e a l , M o n t r e a l , Quebec . U R L : h t tp : / /www.fas .umontrea l .ca /BIOL/ legendre / . Cha i su r i s r i , K . , J . B . M i t t o n , and W . A . E l - K a s s a b y . 1994. V a r i a t i o n i n the mat ing system o f S i t k a spruce (Picea sitchensis): evidence for par t ia l assortative mat ing . A m e r i c a n Journal o f B o t a n y 81 : 1410-1415. Changt ragoon, S., and R . F i n k e l d e y . 1995. Patterns o f genetic var ia t ion and character izat ion o f the mat ing sys tem o f Pinus merkusii i n Tha i l and . Forest Genet ics 2: 87-97. Char leswor th , D . , and B . Char leswor th . 1987. Inbreeding depression and its evolu t ionary consequences. A n n u a l R e v i e w i n E c o l o g y and Systemat ics 18: 237-268. Char l e swor th , D . , and B . Char leswor th . 1995. Quant i ta t ive genetics o f plants: the effect o f the breeding sys tem on genetic var iab i l i ty . E v o l u t i o n 49 : 911-920. Char l e swor th , D . , M . T . M o r g a n , and B . Char leswor th . 1990. Inbreeding depression, genetic load , and the evo lu t ion o f outcross ing rates i n a mul t i l ocus sys tem w i t h no l inkage . E v o l u t i o n 44: 1469-1489. C h e l i a k , W . M . , J . A . P i t e l , and G . M u r r a y . 1985a. Popu la t ion structure and the mat ing sys tem o f whi te spruce. Canad ian Journal o f Forest Research 15: 301-308. C h e l i a k , W . M . , B . P . D a n c i k , K . M o r g a n , F . C . H . Y e h , and C . Strobeck. 1985b. T e m p o r a l var ia t ion o f the mat ing system i n a natural popula t ion o f j a c k pine . Genet ics 109: 569-584. C o l a n g e l i , A . M . , and J . N . O w e n s . 1990. The relat ionship between t ime o f po l l ina t ion , po l l i na t ion eff ic iency, and cone size i n western red cedar (Thuja plicata). Canad ian Journal o f B o t a n y 68: 439-443 . C o l l e v a t t i , R . G . , D . Grat tapagl ia , and J . D . H a y . 2001 . H i g h resolut ion microsate l l i te based analysis o f the mat ing sys tem a l lows the detection o f s ignif icant biparental inbreeding i n Caryocar brasiliense, an endangered t ropica l tree species. Hered i ty 86: 60-67. C o n k l e , M . T . 1987. Elec t rophore t ic analysis o f var ia t ion i n nat ive monterey cypress (Cupressus macrocarpa Ha r tw . ) . In Conse rva t ion and management o f rare and endangered plants. Proceedings o f the conference. T . S. E l i a s (Ed.) Sacramento, C A . pp. 249-256. Copes , D . L . 1981. I soenzyme un i fo rmi ty i n western red cedar seedlings f rom O r e g o n and W a s h i n g t o n . Canad ian Journa l o f Forest Research 11: 451-453 . Cornuet , J . M . , and G . L u i k a r t . 1996. Desc r ip t i on and power analysis o f two tests for detecting recent popula t ion bott lenecks f rom al lele frequency data. Gene t ics 144: 2001-2015 . Co t t r e l l , J . E . , and I. M . S. W h i t e . 1995. T h e use o f i s o z y m e genetic markers to estimate the rate o f outcross ing i n a S i t k a spruce (Picea sitchensis (Bong . ) Carr . ) seed orchard i n Scot land . N e w Forests 10: 111-122. C r i t c h f i e l d , W . B . 1984. Impact o f the pleistocene on the genetic structure o f N o r t h A m e r i c a n conifers . In Proceedings o f the 8th N o r t h A m e r i c a n Forest B i o l o g y W o r k s h o p . R . M . 126 L a n n e r (Ed . ) Depar tment o f Forest Resources , U t a h State U n i v e r s i t y , L o g a n , U t a h . pp. 70-117 . C r o o k , R . W . , and W . E . F r i e d m a n . 1992. Effects o f po l l en tube number and a rchegon ium number o f reproduct ion i n Douglas - f i r : s ignif icance for seed orchard management. Canad ian Journa l o f Forest Research 22: 1483-1488. D a v i d s o n , R . , and Y . A . E l - K a s s a b y . 1997. Genet ic d ivers i ty and gene conservat ion o f Pac i f i c s i lver fir (Abies amabilis) on V a n c o u v e r Is land, B r i t i s h C o l u m b i a . Forest Genet ics 4: 85-98. D e n t i , D . , and D . J . Schoen . 1988. Sel f - fer t i l iza t ion rates i n whi te spruce: effect o f po l l en and seed product ion . Journa l o f Hered i ty 79: 2 8 4 - 2 8 8 . Desponts , M . , and J . P . S i m o n . 1987. Gene t ic structure and var iab i l i ty i n populat ions o f b l ack spruce (Picea mariana) in the subarctic region o f northern Quebec . Canad ian Journa l o f Forest Research 17: 1006-1012. D i R i e n z o , A . , A . C . Peterson, J . C . G a r z a , A . M . V a l d e s , M . S la tk in , and N . B . F re imer . 1994. M u t a t i o n a l processes o f s imple-sequence repeat l o c i i n human populat ions . Proceedings o f the N a t i o n a l A c a d e m y o f Sciences , U S A 9 1 : 3166-3170. D o n g M e i , Z . , L . Y u e , S. X i H u a n , Z . C h u n X i a o , L . G u o F e n g , and L . Jun . 2000 . A p re l imina ry study on the mat ing system o f three different populat ions w i t h i n an improvement p rogramme for Pinus tabulaeformis Car r . Journa l o f B e i j i n g Fores t ry U n i v e r s i t y 22: 13-18. D o y l e J . J . , and J . L . D o y l e . 1987. A r ap id D N A iso la t ion procedure for s m a l l quantit ies o f fresh leaf tissue. P h y t o c h e m i c a l B u l l e t i n 19: 11-15. Duret t , R . , L . Bu t t e l , and R . Har r i son . 2000. Spa t ia l models for h y b r i d zones. Hered i ty 84: 9-19. Ech t , C . S., L . L . D e V e r n o , M . A n z i d e , and G . G . V e n d r a m i n . 1998. Ch lo rop la s t microsatel l i tes reveal popula t ion genetic d ivers i ty i n red pine, Pinus resinosa A i t . M o l e c u l a r E c o l o g y 7: 307-316. Ecker t , C . G . , and M . A l l e n . 1997. C r y p t i c se l f - incompat ib i l i ty i n t r is tylous Decodon verticillatus (Lythraceae) . A m e r i c a n Journa l o f B o t a n y 84: 1391-1397. E l - K a s s a b y , Y . A . 1999. Pheno typ ic plas t ic i ty i n western redcedar. Forest Genet ics 6: 235-240. E l - K a s s a b y , Y . A . , and B . Jaquish. 1996. Popu la t ion densi ty and mat ing sys tem pattern i n western larch . Journal o f Hered i ty 87: 438-443 . E l - K a s s a b y , Y . A . , M . D . Meaghe r , and R . D a v i d s o n . 1993. T e m p o r a l var ia t ion i n the outcross ing rate i n a stand o f western whi te pine. S i l v a e Gene t i ca 42: 131-135. E l - K a s s a b y , Y . A . , M . D . Meaghe r , J . Pa rk inson , and F . T . Por t lock . 1987. A l l o z y m e inheri tance heterozygosi ty and outcrossing rate among Pinus monticola near L a d y s m i t h , B r i t i s h C o l u m b i a . Hered i ty 58: 173-181. 127 E l - K a s s a b y , Y . A . , J . Pa rk inson , and W . J . B . Dev i t t . 1986. The effect o f c r o w n segment on the mat ing system i n a D o u g l a s - F i r {Pseudotsuga menziesii ( M i r b . ) Franco . ) Seed Orcha rd . S i l v a e G e n e t i c a 3 5 : 149-155. E l - K a s s a b y , Y . A . , and K . R i t l a n d . 1996. Genet ic var ia t ion i n l o w e levat ion Doug la s - f i r o f B r i t i s h C o l u m b i a and its re levance to gene conservat ion. B i o d i v e r s i t y and Conse rva t ion 5: 779-794. E l - K a s s a b y , Y . A . , K . R i t l a n d , A . M . K . Fashler , and W . J . B . Dev i t t . 1988. T h e role o f reproduct ive phenology upon the mat ing system o f a Doug la s - f i r seed orchard. S i l v a e Gene t i ca 37: 76-82 . E l - K a s s a b y , Y . A . , D . R u d i n , and R . Y a z d a n i . 1989. L e v e l s o f outcross ing and contaminat ion i n two Pinus sylvestris L . seed orchards i n northern Sweden . Scandanavian Journa l o f Forest Research 4: 41-49 . E l - K a s s a b y , Y . A . , J . R u s s e l l , and K . R i t l a n d , 1994. M i x e d mat ing i n an exper imenta l popula t ion o f western red cedar, Thuja plicata. The Journal o f Hered i ty 85: 227-231 . E l - K a s s a b y , Y . A . , and A . D . Y a n c h u k . 1994. Genet ic d ivers i ty , differentiat ion and inbreeding i n Pac i f c y e w f rom B r i t i s h C o l u m b i a . Journal o f Hered i ty 85: 112-117. E l - K a s s a b y , Y . A . , F . C . Y e h , and O . S z i k l a i . 1981. Es t ima t ion o f the outcross ing rate o f Doug la s - f i r {Pseudotsuga menziesii [ M i r b . ] F ranco) us ing a l l o z y m e p o l y m o r p h i s m s . S i l v a e Gene t i ca 30: 182-183. E l l e g r e n , H . 2000a . Mic rosa t e l l i t e mutations i n the germline; impl ica t ions for evolu t ionary inference. Trends i n Genet ics 16: 551-558. E l l e g r e n , H . 2000b. Heterogenous mutat ion processes i n human microsate l l i te D N A sequences. Nature Genet ics 24: 400-402 . E r i c k s o n , V . J . , and W . T . A d a m s . 1990. M a t i n g system var ia t ion among i n d i v i d u a l ramets i n a Doug la s - f i r seed orchard. Canad ian Journal o f Forest Research , 20: 1672-1675. E t t l , G . J . , and D . L . Peterson. 2001 . Genet ic var ia t ion o f subalpine fir {Abies lasiocarpa ( H o o k . ) Nut t . ) i n the O l y m p i c M o u n t a i n s , W A , U S A . S i l v a e Gene t i ca 50: 145-153. F a d y , B . , and R . D . Wes t f a l l . 1997. M a t i n g system parameters i n a natural popula t ion o f Abies borisii regis Ma t t f e ld . A n n a l e s des Sciences Forestieres 54: 643-647. Fast , W . , B . P . D a n c i k , and R . C . B o w e r . 1986. M a t i n g system and p o l l e n contaminat ion i n a D o u g l a s - f i r c lone bank. Canad ian Journa l o f Forest Research 16: 1314-1319. Far r i s , M . A . , and J . B . M i t t o n , 1984. Popu la t ion density, outcross ing rate, and heterozygote superiori ty i n ponderosa pine . E v o l u t i o n 38: 1151-1154. Felsenste in , J . 1995. P H Y L I P (Phy logeny Inference Package) V e r s i o n 3.57c. Depar tment o f Genet ics , U n i v e r s i t y o f Wash ing ton , Seattle, W a s h i n g t o n Ferreyra , L . L . , A . L a t i n o , A . Ca lde ron , and C . N . Gardena l . 1996. A l l o z y m e p o l y m o r p h i s m i n Austrocedrus chilensis ( D . D o n ) F l o r i n and Boute l je f r om Patagonia , Argen t i na . S i l v a e Gene t i ca 45 : 61-64. 128 F o w l e r , D . P . 1965. Effects o f inbreeding i n red pine, Pinus resinosa A i t . II. P o l l i n a t i o n studies. S i l v a e Gene t i ca 14: 12-23. F o w l e r , D . P . , and R . W . M o r r i s . 1977. Genet ic d ivers i ty i n red pine: ev idence for l o w genie heterozygosi ty. Canad ian Journa l o f Forest Research 7: 343-347. F r a n k l i n , E . C . 1972. Gene t ic l oad i n l o b l o l l y pine. A m e r i c a n Naturat l is t 102: 262-265 . F r i e d m a n , S. T . , and W . T . A d a m s . 1985. L e v e l s o f outcross ing i n two l o b l o l l y pine seed orchard. S i l v a e Gene t i ca 34: 157-162. F r o g l e y , M . R . , and P . C . Tzedak i s . 1999. C l i m a t e var iab i l i ty i n northwest Greece dur ing the last in terg lac ia l . Sc ience 285: 1886-1889. Furn ie r , G . R . , and W . T . A d a m s . 1986. M a t i n g system i n natural populat ions o f Jeffrey pine. A m e r i c a n Journa l o f B o t a n y 73 : 1002-1008. G e , S., M . X . W a n g , and Y . W . C h e n . 1988. A n analysis o f popula t ion genetic structure o f M a s s o n p ine b y i soenzyme technique. Sc ien t i a S i l v a e S in icae 24: 399-409. G i a n n i n i , R . , M . Morgan te , and G . V e n d r a m i n . 1991. A l l o z y m e var ia t ion i n I tal ian popula t ions o f Picea abies ( L . ) Kars t . S i l v a e Gene t i ca 40 : 3-4. G i a n n i n i , R . , L . Pa rducc i , P . R o s s i , and F . V i l l a n i . 1994. Gene t ic structure and mat ing system o f s i lver f i r i n the C a m p o l i n o reserve (Nor th Apenn ines , I taly) . Journa l o f Genet ics & B r e e d i n g 48 : 335-337. • -G i b s o n , J . P . , and J . L . H a m r i c k . 1991a. Genet ic d ivers i ty and structure i n Pinus pungens (Table M o u n t a i n pine) populat ions. Canad ian Journal o f Forest Research 2 1 : 635-642 . G i b s o n , J . P . , and J . L . H a m r i c k . 1991b. Heterogenei ty in po l l en al lele frequencies among cones, whor l s , and trees o f Tab le M o u n t a i n p ine (Pinus pungens). A m e r i c a n Journa l o f B o t a n y 78: 1244-1251. G i l l , D . E . , L . C h a o , S. L . Perk ins , and J . B . W o l f . 1995. Gene t ic m o s a i c i s m i n plants and c lona l animals . A n n u a l R e v i e w o f E c o l o g y and Systematics 26: 423-444 . G l a u b i t z , J . , Y . A . E l - K a s s a b y , and J . E . C a r l s o n , 2000. N u c l e a r res t r ic t ion fragment length p o l y m o r p h i s m analysis o f genetic d ivers i ty i n western red cedar. Canad i an Journa l o f Forest Research 30: 379-389. G o d t , M . J . W . , J . L . H a m r i c k , M . A . E d w a r d s - B u r k e , and J . H . W i l l i a m s . 2001 . Compar i sons o f genetic d ivers i ty i n whi te spruce (Picea glauca) and j a c k p ine (Pinus banksiana) seed orchards w i t h natural populat ions. Canad ian Journa l o f Fores t Research 31 : 943-949. G o m o r y , D . , and L . Paule . 1992. Inferences on mat ing system and genetic compos i t i on o f a seed orchard c rop i n the European larch (Larix decidua M i l l . ) . Journa l o f Genet ics and B r e e d i n g 46: 309-313. Goncha renko , G . G . , V . E . Padutov, and A . E . S i l i n . 1993. A l l o z y m e var ia t ion i n natural popula t ions o f Euras ian pines II. Gene t ic var ia t ion, d ivers i ty , differentiat ion, and gene f l o w i n Pinus sibirica D u T o u r i n some l o w l a n d and mounta in populat ions . S i l v a e Gene t i ca 42 : 246-253. 129 Goncha renko , G . G . , I. V . Zade ika , and J . J . B i r g e l i s . 1995. Gene t ic structure, d ivers i ty and differentiat ion o f N o r w a y spruce (Picea abies ( L . ) Kars t . ) i n natural populat ions o f L a t v i a . Forest E c o l o g y & Managemen t 72: 31-38. Got tesfe ld , A . S., F . J . Swanson , and L . M . J . Got tesfeld . 1981. A pleistocene low-e leva t ion subalpine forest i n the western cascades, Oregon . Nor thwes t Sc ience 55: 157-167. G u o , S. W . , and E . A . T h o m p s o n . 1992. Pe r fo rming the exact test o f H a r d y - W e i n b e r g proport ions for mul t ip l e al leles. B i o m e t r i c s 48: 361-372 Gur ie s , R . P . , and F . T . L e d i g . 1982. Gene t ic d ivers i ty and popula t ion structure i n P i t c h P i n e (Pinus rigida M i l l . ) . E v o l u t i o n 36: 387-402. H a l l , P . , L . C . O r r e l l , and K . S. B a w a . 1994. Gene t ic d ivers i ty and mat ing system i n a t ropica l tree, Carapa guianensis (Mel iaceae) . A m e r i c a n j ou rna l o f B o t a n y 81 : 1104-1111. H a l l , P . , S. W a l k e r , and K . B a w a . 1996. Effect o f forest fragmentation on genetic d ivers i ty and mat ing sys tem i n a t ropica l tree, Pithecellobium elegans. Conse rva t ion B i o l o g y 10: 757-768 . H a m a n n , A . , Y . A . E l - K a s s a b y , M . P . K o s h y , and G . N a m k o o n g . 1998. M u l t i v a r i a t e analysis o f a l l o z y m i c and quantitative trait var ia t ion i n Alnus rubra: geographic patterns and evolu t ionary impl ica t ions . Canad ian Journal o f Forest Research 28: 1557-1565. H a m r i c k , J . L . , and M . J . W . God t . 1996. Effects o f l i fe his tory traits on genetic d ivers i ty i n plant species. P h i l o s o p h i c a l Transact ions o f the R o y a l Soc ie ty o f L o n d o n : Series B 351 : 1291-1298. H a m r i c k , J . L . , M . J . W . God t , and S. L . She rman-Broy le s . 1992. Factors in f luenc ing levels o f genetic d ivers i ty i n w o o d y plants species. N e w Forests 6: 95-124. Harder , L . D . , and S. C . H . Barrett . 1996. P o l l e n dispersal and mat ing patterns i n an ima l po l l ina ted plants. In F l o r a l b i o l o g y : studies on f lo ra l evo lu t ion i n a n i m a l po l l ina ted plants. D . G . L l o y d and S . C . H . Barrett . (Eds.) C h a p m a n & H a l l , N e w Y o r k . pp. 140-190. H e b d a , R . J . , and R . W . M a t h e w e s . 1984. H o l o c e n e history o f cedar and native ind ian cultures o f the N o r t h A m e r i c a n P a c i f i c Coast . Sc ience 225: 711-713 Hewi t t , G . M . 1993. Pos tg lac ia l d is t r ibut ion and species substructures: lessons f r o m po l l en , insects and h y b r i d zones. In E v o l u t i o n a r y patterns and processes. D . R . L e e s and D . E d w a r d s (Eds.) A c a d e m i c Press, L o n d o n , U K . pp. 97-123 . Hewi t t , G . 2000. T h e genetic legacy o f the Quaternary ice ages. Nature 405 : 907-913 . Ho l s inge r , K . E . 1991. M a s s act ion models for plant mat ing systems: the evolu t ionary stabi l i ty o f m i x e d mat ing systems. A m e r i c a n Natura l i s t 138: 606-622 . Ho l s inge r , K . E . 1996. P o l l i n a t i o n b i o l o g y and the evo lu t ion o f mat ing systems i n f l ower ing plants. E v o l u t i o n a r y B i o l o g y 29: 107-149. H u , X . - S., and R . A . Ennos . 1999. S c o r i n g the mat ing systems o f natural popula t ions o f three Larix taxa i n C h i n a : L. gmelinii (Rupr . ) Rupr . , L. olgensis H e n r y and L. principis-rupprechtii M a y r . Sc ien t i a S i l v a e S in icae 35: 21-31 . 130 H u s b a n d , B . C . , and D . W . Schemske . 1996. E v o l u t i o n o f the magni tude and t i m i n g o f inbreeding depression i n plants. E v o l u t i o n . 50: 54-70. Innes, D . J . , and G . G . R i n g i u s . 1990. M a t i n g sys tem and genetic structure o f two populat ions o f whi te spruce (Picea glauca) i n eastern N e w f o u n d l a n d . Canad ian Journa l o f B o t a n y 68: 1661-1666. Jaquish, B . , and Y . A . E l - K a s s a b y . 1998. Gene t ic var ia t ion o f western la rch i n B r i t i s h C o l u m b i a and its conservat ion. Journal o f Hered i ty 89: 248-253. Jarne, P . , and D . Char leswor th . 1993. T h e evo lu t ion o f self ing rate i n funct ional ly hermaphrodite plants and animals . A n n u a l R e v i e w i n E c o l o g y and Systematics 24: 441-466. Johnston, M . O . 1998. E v o l u t i o n o f intermediate self ing rates i n plants: po l l i na t ion eco logy versus deleterious mutations. Gene t i ca 102/103: 267-278. K a r h u , A . , J . - H . D ie t e r i ch , and O . Savo la inen . 2000. R a p i d expansion o f microsate l l i te sequences i n pines. M o l e c u l a r B i o l o g y and E v o l u t i o n 17: 259-265 . K a r k k a i n e n , K . , and O . Savo la inen . 1993. T h e degree o f early inbreeding depression determines the sel f ing rate at the seed stage: m o d e l and results for Pinus sylvestris (Scots pine) . Hered i ty 7 1 : 160-166. K i j a s J . M . H , and J . C . S .Fowle r .1994 . En r i chmen t o f microsatel l i tes f r o m the citrus genome us ing b io t iny la ted o l igonucleo t ide sequences bound to streptavidin-coated magnet ic part icles. B io techn iques 16: 657-662. K i m u r a , M . , and J . F . C r o w . 1964. T h e number o f al leles that can be main ta ined i n a f inite popula t ion . Genet ics 49 : 725-738. K i r k p a t r i c k , M . , and P . Jarne. 2000. The effects o f a bot t leneck on inbreed ing depression and the genetic load. The A m e r i c a n Natural is t 155: 154-167. Kjaer , E . D . , and V . Suangtho. 1995. Outc ross ing rate o f teak (Tectona grandis (L . ) ) . S i l v a e Gene t i ca 44: 175-177. K n o w l e s , P . , G . R . Furn ie r , M . A . A l e k s i u k , and D . J . Perry , 1987. S ign i f i can t levels o f self-fer t i l iza t ion i n natural populat ions o f tamarack. Canad ian Journa l o f B o t a n y 65: 1087-1091. Kormut ' ak , A . , and D . L i n d g r e n . 1996. M a t i n g system and empty seeds i n s i lver f i r (Abies alba M i l l . ) . Forest Genet ics 3: 231-235. K o r n , R . W . 2001 . A n a l y s i s o f shoot ap ica l organizat ion i n s ix species o f the Cupressaceae based on ch imer i c behavior . A m e r i c a n Journal o f B o t a n y 88: 1945-1952. K o r o l , L . , N . K a r a , K . Is ik, and G . Sch i l l e r . 1997. Gene t ic differentiat ion among and w i t h i n natural and planted Cupressus sempervirens L . eastern Medi ter raneaen populat ions . S i l v a e Gene t i ca 46: 151-155. K r a k o w s k i , J . 2001 . Conserva t ion genetics o f whi tebark p ine (Pinus albicaulis E n g e l m . ) i n B r i t i s h C o l u m b i a . M S c , 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 . 131 K u i t t i n e n , H . , O . M u o n a , K . K a r k a i n e n , and Z . B o r z a n . 1991. Serbian spruce, a nar row endemic , contains m u c h genetic var ia t ion. Canad ian Journa l o f Fores t Research 21 : 363-367. K u i t t i n e n , H . , and O . Savo la inen . 1992. Picea omorika is a self-fertile but outcross ing conifer . Hered i ty 68: 183-187. K u s e r , J . E . , T . R . Meaghe r , D . L . Sheely , and A . W h i t e . 1997. A l l o z y m e frequencies i n N e w Jersey and N o r t h C a r o l i n a populat ions o f A t l a n t i c white-cedar, Chamaecyparis thyoides (Cupressaceae). A m e r i c a n Journal o f B o t a n y 84: 1536-1541. Lagercrantz , TJ., and N . R y m a n . 1990. Gene t ic structure o f N o r w a y spruce (Picea abies): concordance o f m o r p h o l o g i c a l and a l l o z y m i c var ia t ion. E v o l u t i o n 44: 38-53. L a i , H . - L . , and M . - X . W a n g . 1997. S tudy on mat ing sys tem i n man-made populat ions o f M a s s o n pine. Sc ien t i a S i l vae S in icae 33: 219-224. L a m y , S., A . B o u c h a r d , and J . - P . S i m o n . 1999. Genet ic structure, var iab i l i ty , and mat ing sys tem i n eastern whi te cedar (Thuja occidentalis) popula t ions o f recent o r ig in i n an agr icul tura l landscape i n southern Quebec . Canad ian Journal o f Forest Research 29: 1383-1392. L a n d e , R . , and D . W . Schemske . 1985. The evo lu t ion o f self-fer t i l izat ion and inbreeding depression i n plants. I. Genet ic models . E v o l u t i o n 39: 24-40. L a n d e , R . , D . W . Schemske , and S. T . Schu l t z . 1994. H i g h inbreeding depression, selective interference among l o c i , and the threshold self ing rate for pu rg ing recessive lethal mutations. E v o l u t i o n 48 : 965-978. La t ta , R . G . 1995. The effects o f embryo compet i t ion w i t h m i x e d mat ing on the genetic load i n plants. Hered i ty 75 : 637-643. La t ta , R . G . , and J . B . M i t t o n . 1999. H i s t o r i c a l separation and present gene f l o w through a zone o f secondary contact i n ponderosa pine . E v o l u t i o n 53: 769-776 . L e d i g , F . T . , B . B e r m e j o - V e l a z q u e z , P . D . H o d g s k i s s , D . R . Johnson, C . F l o r e s - L o p e z , and V . Jacob-Cervantes . 2000. T h e mat ing system and genie d ivers i ty i n M a r t i n e z spruce, an ext remely rare endemic o f M e x i c o ' s S ier ra M a d r e Or ien ta l : an example o f facultat ive self ing and su rv iva l i n in terg lac ia l refugia. Canad ian Journa l o f Fores t Research 30: 1156-1164. L e d i g , F . T . , M . A . Capo-Ar t eaga , P . D . H o d g s k i s s , H . Sbay , S. F l o r e s - L o p e z , M . T . C o n k l e , and B . B e r m e j o - V e l a z q u e z . 2001 . Gene t ic d ivers i ty and the mat ing sys tem o f a rare M e x i c a n p inon , Pinus pinceana, and a compar i son w i t h Pinus maximartinezii (Pinaceae). A m e r i c a n Journal o f B o t a n y 88: 1977-1987. L e d i g , F . T . , and M . T . C o n k l e . 1983. Gene d ivers i ty and genetic structure i n a nar row endemic , Tor rey p ine (Pinus torreyana Parry ex Carr . ) . E v o l u t i o n 37: 79-85 . L e d i g , F . T . , M . T . C o n k l e , B . B e r m e j o - V e l a z q u e z , T . E g u i l u z - P i e d r a , P . D . H o d g s k i s s , D . R . Johnson, and W . S. D v o r a k . 1999. E v i d e n c e for an extreme bot t leneck i n a rare M e x i c a n 132 p i n y o n : genetic d ivers i ty , d i s equ i l i b r ium, and the mat ing sys tem i n Pinus maximartinezii. E v o l u t i o n 53: 91-99. L e d i g , F . T . , V . Jacob-Cervantes , P . D . Hodgsk i s s , and T . E g u i l u z - P i e d r a . 1997. Recen t evo lu t ion and divergence among populat ions o f a rare m e x i c a n endemic , C h i h u a h u a spruce, f o l l o w i n g holocene c l ima t i c w a r m i n g . E v o l u t i o n 5 1 : 1815-1827. L e v i n s o n , G . , and G . A . G u t m a n . 1987. Sl ipped-s t rand mispar ing : a major m e c h a n i s m for D N A sequence evo lu t ion . M o l e c u l a r B i o l o g y and E v o l u t i o n 4: 203-211 . L e w a n d o w s k i , A . , and J . B u r c z y k . 2000. M a t i n g system and genetic d ivers i ty i n natural popula t ions o f European la rch {Larix decidua) and stone pine {Pinus cembrd) located at higher elevations. S i l vae Gene t i ca 49: 158-161. L e w a n d o w s k i , A . , J . B u r c z y k , and L . M e j n a r t o w i c z . 1991. Gene t ic structure and the mat ing sys tem i n an o l d stand o f P o l i s h la rch . S i l v a e Gene t i ca 40 : 75-79 . L i , P . , and W . T . A d a m s . 1989. Ra nge - w i de patterns o f a l l o z y m e var ia t ion i n Doug la s - f i r {Pseudostuga menziesii). Canad ian Journal o f Fores t Research 19: 149-161. L i a n , C , M . M i w a , and T . Hegetsu . 2001 . Outcross ing and paternity analysis o f Pinus densiflora (Japanese red pine) b y microsate l l i te p o l y m o r p h i s m . Hered i ty 87: 88-98. L i n , T . - P . , T . - Y . L e e , L . - F . Y a n g , Y . - L . C h u n g , and J . - C . Y a n g . 1994. C o m p a r i s o n o f the a l l o z y m e divers i ty i n several populat ions o f Chamaecyparis formosensis and Chamaecyparis taiwanensis. Canad ian Journal o f Forest Research 24: 2128-2134. L i n , T . - P . , C . - T . W a n g , and J . - C . Y a n g . 1998. C o m p a r i s o n o f genetic d ivers i ty between Cunninghamia konishii and C. lanceolata. Journa l o f Hered i ty 89: 370-373 . Longaue r , R . , P . Z h e l e v , L . Paule , and D . G o m o r y . 1992. T h e mat ing sys tem outcross ing rate and genetic differentiat ion i n natural Scots p ine {Pinus sylvestris L . ) popula t ions f rom B u l g a r i a . B i o l o g i a 47: 539-547. L o v e l e s s , M . D . , J . L . H a m r i c k , and R . B . Foster . 1998. Popu la t i on structure and mat ing system i n Tachigali versicolor, a monocarp ic neot ropica l tree. Hered i ty 81 : 134-143. L u i k a r t , G . , F . W . A l l e n d o r f , J . - M . Cornuet , and W . B . S h e r w i n . 1998. D i s t o r t i o n o f a l le le frequency distr ibut ions provides a test for recent popula t ion bott lenecks. The Journa l o f Hered i ty 89: 238-247. M a l v o t i , M . E . , S. F i n e s c h i , M . Morgan te , and G . G . V e n d r a m i n . 1995. M a t i n g system i n natura l ized Juglans regia L . popula t ion i n Italy. In Popu la t i on genetics and genetic conservat ion o f forest trees. P . Baradat , W . T . A d a m s , and G . M u l l e r - S t a r c k (Eds.) Papers presented at an internat ional s y m p o s i u m organized by I U F R O , he ld 24-28 A u g u s t 1992 at C a r c a n - M a u b u i s s o n , France . S P B A c a d e m i c P u b l i s h i n g . A m s t e r d a m , Netherlands, pp. 305-308. M a n d a l , A . K . , and R . A . Ennos . 1995. M a t i n g sys tem analysis i n a natural popula t ion o f Acacia milotica subspecies kraussiana. Forest E c o l o g y and Managemen t 79: 235-240. 133 M a r s h a l l , H . D . , C . N e w t o n , and K . R i t l a n d . 2002. Ch lo rop las t phylogeography and evo lu t ion o f h i g h l y p o l y m o r p h i c microsatel l i tes i n lodgepole p ine (Pinus contorta). Theore t ica l and A p p l i e d Genet ics 104: 367-378. M a r u y a m a , T . , and P . A . Fuerst . 1984. Popu la t ion bott lenecks and n o n e q u i l i b r i u m mode ls i n popula t ion genetics. I. A l l e l e numbers when populat ions evo lve f rom zero var iab i l i ty . Genet ics 108: 745-763 . M a r u y a m a , T . , and P . A . Fuerst . 1985. Popu la t ion bott lenecks and n o n e q u i l i b r i u m mode ls i n popula t ion genetics. II. N u m b e r o f alleles i n a s m a l l popula t ion that was fo rmed by a recent bott leneck. Genet ics 111: 675-689. M a t h e s o n , A . C . , J . C . B e l l , and R . D . Barnes . 1989. B r e e d i n g systems and genetics i n some Cen t ra l A m e r i c a n pine populat ions. S i l v a e Gene t i ca 38: 107-113. Matthes-Sears , U . , S. C . Stewart, and D . W . L a r s o n . 1991. Sources o f a l l o z y m i c var ia t ion i n Thuja occidentalis i n southern Ontar io , Canada . S i l v a e Gene t i ca 40 : 100-105. M c G r a n a h a n , M . , J . C . B e l l , G . F . M o r a n , and M . Slee . 1997. H i g h genetic d ivergence between geographic regions i n the h igh ly outcrossing species Acacia aulacocarpa ( C u n n . ex Benth . ) . Forest Genet ics 4: 1-13. M e h r i n g e r , P . J . Jr. 1985. Late-Quaternary po l l en records f rom the inter ior P a c i f i c Nor thwes t and northern Great B a s i n o f the U n i t e d States. In P o l l e n Reco rds o f Late-Quar tenary N o r t h A m e r i c a n Sediments . J . V . B r y a n t and R . G . H o l l o w a y (Eds.) A m e r i c a n A s s o c i a t i o n o f Strat igraphic Pa lyno log i s t s Founda t ion . D a l l a s , Texas , pp. 167-189. M e j n a r t o w i c z , L . , and A . L e w a n d o w s k i . 1994. A l l o z y m e p o l y m o r p h i s m i n seeds co l lec ted f rom a I U F R O - 6 8 Doug la s - f i r test-plantation. S i l v a e Gene t i ca 43 : 181-186. M e n o t t i - R a y m o n d , M . , and S. J . O ' B r i e n . 1993. D a t i n g the genetic bot t leneck o f the A f r i c a n cheetah. Proceedings o f the N a t i o n a l A c a d e m y o f Sc ience 90: 3172-3176. M e r z e a u , D . , B . C o m p s , B . Thiebaut , and J . L e t o u z e y . 1994. E s t i m a t i o n o f Fagus sylvatica L . ma t ing system parameters i n natural populat ions . A n n a l e s des Sciences Forestieres 5 1 : 163-173. M i l l a r , C . I., and K . A . M a r s h a l l . 1991. A l l o z y m e var ia t ion o f Por t -Or fo rd -Cedar (Chamaecyparis lawsoniana): impl ica t ions for genetic conservat ion. Forest Sc ience 37: 1060-1077. M i l l a r , M . A . , M . B y r n e , D . J . Coates , M . J . C . S tuke ly , and J . A . M c C o m b . 2000. M a t i n g sys tem studies i n jarrah, Eucalyptus marginata (Myrtacaeae) . A u s t r a l i a n Journal o f B o t a n y 48: 475-479 . M i n o r e , D . 1983. Wes te rn redcedar: A literature rev iew. Pac i f i c Nor thwes t Fores t and Range E x p e r i m e n t Stat ion. U S D A Fores t Se rv ice Gene ra l T e c h n i c a l Repor t P N W - 1 5 0 . Por t land , Oregon . 70 pps. 134 M i n o r e , D . 1990. Wes te rn redcedar (Thuja plicata D o n n ex D . D o n ) . In S i l v i c s o f N o r t h A m e r i c a , V o l . 1, Coni fe r s . R . M . B u m s and B . H . H o n k a l a (Eds.) U S D A Forest Se rv ice A g r i c u l t u r e H a n d b o o k 654 Wash ing ton , D C . pp. 590-600. M i t t o n , J . B . 1992. The d y n a m i c mat ing systems o f conifers . N e w Forests 6: 197-216. M i t t o n , J . B . , Y . B . L inha r t , M . L . D a v i s , and K . B . Sturgeon. 1981. E s t i m a t i o n o f outcross ing i n ponderosa pine, Pinus ponderosa L a w s , f r o m patterns o f segregation o f protein p o l y m o r p h i s m s and f rom frequencies o f a lb ino seedlings. S i l v a e Gene t i ca 30: 117-121. M i t t o n , J . B . , R . G . La t ta , and G . E . Rehfeldt . 1997. The pattern o f inbreed ing i n W a s h o e pine and su rv iva l o f inbred progeny under op t ima l env i ronmenta l condi t ions . S i l v a e Gene t i ca 4 6 : 2 1 5 - 2 1 9 . M i t t o n , J . B . , Y . B . L inhar t , K . B . Sturgeon, and J . L . H a m r i c k . 1979. A l l o z y m e p o l y m o r p h i s m s detected i n mature needle tissue o f ponderosa pine . Journa l o f Hered i ty 70 : 86-89. M o r a n , G . F . , and W . T . A d a m s . 1989. M i c r o g e o g r a p h i c a l patterns o f a l l o z y m e differentiat ion i n Doug la s - f i r f r om southwest Oregon . Forest Sc ience 35: 3-15. M o r a n , G . F . , J . C . B e l l , and A . C . Ma theson . 1980. T h e genetic structure and levels o f inbreeding i n a Pinus radiata D . D o n seed orchard. S i l v a e Gene t i ca 29: 190-193. M o r a n , G . F . , J . C . B e l l , and K . G . E l d r i d g e . 1988. T h e genetic structure and the conservat ion o f the f ive natural populat ions o f Pinus radiata. Canad ian Journa l o f Forest Research 18: 506-514. M o r a n , G . F . , P . M u o n a , and J . C . B e l l . 1989. B r e e d i n g systems and genetic d ivers i ty i n Acacia auriculiformis and A. crassicarpa. B i o t r o p i c a 2 1 : 250-256. M o r g a n , M . T . , D . J . Schoen and T . M . B a t a i l l o n . 1997. T h e evo lu t ion o f sel f - fer t i l izat ion i n perennials . The A m e r i c a n Natura l i s t 150: 618-638. M o r g a n , M . T . 2001 . Consequences o f l i fe his tory for inbreeding depression and mat ing system evo lu t ion i n plants. Proceedings o f the R o y a l Soc ie ty o f L o n d o n : Series B 1817-1824. Morgan te , M . , P . R o s s i , G . G . V e n d r a m i n , and G . B o s c h e r i n i . 1994. L o w levels o f outcrossing i n Pinus leucodermis: further evidence i n ar t i f ic ia l stands. Canad i an Journa l o f B o t a n y 72 : 1289-1293. Morgan te , M . , G . G . V e n d r a m i n , and A . M . O l i v i e r i . 1991b. M a t i n g system analysis i n Pinus leucodermis A n t . : detection o f self-fer t i l izat ion i n natural populat ions . Hered i ty 67: 197-203. Morgan te , M . , G . G . V e n d r a m i n , and P . R o s s i . 1991a. Effects o f stand densi ty on outcrossing rate i n two N o r w a y spruce (Picea abies) populat ions. Canad i an Journa l o f B o t a n y 69: 2704-2708. Morgan te , M . , G . G . V e n d r a m i n , P . R o s s i , and A . M . O l i v i e r i . 1993. Se lec t ion against inbreds i n early l i f e -cyc le phases i n Pinus leucodermis A n t . Hered i ty 70 : 622-627. 135 M o s s e l e r , A . , D . J . Innes, and B . A . Roberts . 1991. L a c k o f a l l o z y m e var ia t ion i n disjunct N e w f o u n d l a n d populat ions o f red pine (Pinus resinosa). Canad ian Journa l o f Fores t Research 21 : 525-528. M u o n a , O . 1988. Popu la t ion genetics i n forest tree improvement . In P lant popula t ion genetics, breeding and genetic resources. A . D . H . B r o w n , M . T . C l e g g , A . L . K a h l e r , and B . S. W e i r (Eds.) S inauer Assoc ia tes Inc., Sunder land, Massachuset ts , pp. 282-298. M u o n a , O . , and A . H a r i u . 1989. Ef fec t ive popula t ion sizes, genetic va r iab i l i ty , and mat ing sys tem i n natural stands and seed orchards o f Pinus sylvestris. S i l v a e Gene t i ca 38: 2 2 1 -228. M u r a w s k i , D . A . , B . Dayanandan , and K . S. B a w a . 1994a. Outc ross ing rates o f two endemic Shorea species f rom S r i L a n k a n t ropica l ra in forests. B i o t r o p i c a 26: 23-29. M u r a w s k i , D . A . , I. A . U . N . Guna t i l l eke , and K . S. B a w a . 1994b. T h e effects o f selective l o g g i n g on inbreeding i n Shorea megistophylla (Dipterocarpaceae) f rom S r i L a n k a . Conse rva t ion B i o l o g y 8: 997-1002. M u r a w s k i , D . A . , J . L . H a m r i c k , S. P . H u b b e l l , and R . B . Foster . 1990. M a t i n g systems o f two B o m b a c e o u s trees o f a neotropical mois t forest. O e c o l o g i a 82: 501-506. M u r a w s k i , D . A . , and J . L . H a m r i c k . 1991. T h e effect o f the density o f f l ower ing i nd iv idua l s on the mat ing systems o f nine t ropica l tree species. Hered i ty 67: 167-174. Nagami t su , T . , S. I c h i k a w a , M . O z a w a , R . S h i m a m u r a , N . K a c h i , Y . T s u m u r a , and N . M u h a m m a d . 2001 . Mic rosa t e l l i t e analysis o f the breeding system and seed dispersal i n Shorea leprosula (Dipterocarpaceae). International Journa l o f P lan t Sc ience 162: 155-159. N a k a m u r a , R . R . , and N . C . Whee le r . 1992. Sel f - fer t i l i ty var ia t ion and paternal success through outcross ing i n D o u g l a s f i r . Theore t ica l and A p p l i e d Genet ics 83: 851-854. N a m k o o n g , G . , and J . B i s h i r . 1987. T h e frequency o f lethal al leles i n forest tree populat ions . E v o l u t i o n 4 1 : 1123-1126. N e a l e , D . B . , and W . T . A d a m s . 1985. A l l o z y m e and mat ing-sys tem var ia t ion i n ba l sam f i r (Abies balsamea) across a cont inuous e levat ional transect. Canad i an Journa l o f B o t a n y 63: 2448-2453 . N e i , M . 1972. Gene t ic distance between populat ions . A m e r i c a n Natura l i s t 106: 283-292. N e i , M . , T . M a r u y a m a , and R . Chakrabor ty . 1975. The bot t leneck effect and genetic va r iab i l i ty i n populat ions . E v o l u t i o n 29: 1-10. N e w c o m b e , R . G . 1998. T w o - s i d e d confidence intervals for the s ingle propor t ion: compar i son o f seven methods. Statistics i n M e d i c i n e 17: 857-872. O ' C o n n e l l , L . M . , and C . E . R i t l a n d . 2000. Charac ter iza t ion o f microsate l l i te l o c i i n western redcedar (Thuja plicata). M o l e c u l a r E c o l o g y 9: 1920-1922. 136 O ' C o n n e l l , L . M . , F . V i a r d , J . R u s s e l l , and K . R i t l a n d . 2001 . T h e mat ing system i n natural populat ions o f western redcedar (Thuja plicata). Canad i an Journa l o f B o t a n y 79: 753-756. Oet t ing , W . S., H . K . L e e , D . J . F landers , G . L . Wiesne r , T . A . Sel lers , and R . A . K i n g . 1995. L i n k a g e analysis w i t h mu l t i p l exed short tandem repeat p o l y m o r p h i s m s us ing infrared f luorescence and M 1 3 ta i led pr imers . G e n o m i c s 30: 450-458 . Ohta , T . , and M . K i m u r a . 1973. A m o d e l o f mutat ion appropriate to estimate the number o f e lect rophoret ical ly detectable al leles i n a f inite popula t ion . Gene t ic Research C a m b r i d g e 22: 201-204. O ' M a l l e y , D . M . , and K . S. B a w a . 1987. M a t i n g system o f a t ropica l ra in forest tree species. A m e r i c a n Journa l o f B o t a n y 74: 1143-1149. O ' M a l l e y , D . M . , D . P . B u c k l e y , G . T . Prance, and K . S. B a w a . 1988. Gene t ics o f B r a z i l nut (Bertholletia excelsa H u m b . & B o n p l . : Lecyth idaceae) , 2. M a t i n g system. Theore t ica l and A p p l i e d Genet ics 76: 929-932 . O m i , S . K . , and W . T . A d a m s , 1986. V a r i a t i o n i n seed set and proport ions o f outcrossed progeny w i t h c lones , c r o w n pos i t ion , and top pruning i n a Doug la s - f i r seed orchard. Canad ian Journa l o f Fores t Research 16: 502-507. O r i v e , M . E . 2001 . Somat i c mutations i n organisms w i t h c o m p l e x l i fe histories. Theore t ica l Popu la t i on B i o l o g y 59: 235-249. Otto , S. P . , and I. M . Has t ings . 1998. M u t a t i o n and select ion w i t h i n the i n d i v i d u a l . Gene t i ca 102/103: 507-524. O w e n s , J . N . , and M . M o l d e r . 1980. S e x u a l reproduct ion i n western red cedar (Thuja plicata). Canad ian Journa l o f B o t a n y 58: 1376-1393. O w e n s , J . N . , A . M . C o l a n g e l i , and S. J . M o r r i s . 1990. T h e effect o f self-, cross-, and no po l l i na t ion on ovu le , embryo , seed, and cone development i n western red cedar (Thuja plicata). Canad ian Journa l o f Forest Research 20: 66-75. Page, R . D . M . 1998. Tree V i e w for M a c i n t o s h . V e r s i o n 1.5. U n i v e r s i t y o f G l a s g o w , U K . P a i v a , J . R . D . , P . Y . K a g e y a m a , and R . V e n c o v s k y . 1993. Outc ross ing rates and inbreeding coefficients i n rubber trees (Hevea brasiliensis ( W i l l d . ex A d r . de Juss.) M u e l l e r . A r g . ) . R e v i s t a B r a s i l i e r a de Gene t i ca 16: 1003-1011. Pa rducc i , L . , A . E . S z m i d t , F . V i l l a n i , X . - R . W a n g , and M . C h e r u b i n i . 1996. Gene t ic var ia t ion o f Abies alba i n Italy. Heredi tas (Landskrona) 125: 11-18. Park , W . S., and D . P . F o w l e r . 1984. Inbreeding i n b l ack spruce (Picea mariana ( M i l l . ) B . S . P . ) : self-fert i l i ty, genetic load , and performance. Canad ian Journa l o f Fores t ry Research 14: 17-21. Parker , T . , and F . D . Johnson, 1988. Seed and vegetative regeneration o f western redcedar i n the northern R o c k y M o u n t a i n s . In Wes te rn red ceda r -does it have a future? N . J . S m i t h (Ed.) 137 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 , F a c u l t y o f Forestry, V a n c o u v e r , B r i t i s h C o l u m b i a , pp. 136-144. Paule , L . , D . L i n d g r e n , and R . Y a z d a n i . 1993. A l l o z y m e frequencies, outcross ing rate and po l l en contaminat ion i n Picea abies orchards. Scand inav ian Journa l o f Forest Research 8: 8-17. Perez-Nasser , N . , L . E . Eguiar te , and D . P inero . 1993. M a t i n g sys tem and genetic structure o f the d is ty lous t ropica l tree Psychotriafaxlucens (Rubiaceae) . A m e r i c a n Journa l o f B o t a n y 80: 45-52 . Perry , D . J . , and B . P . D a n c i k . 1986. M a t i n g system dynamics o f L o d g e p o l e P i n e i n A lbe r t a , Canada . S i l v a e Gene t i ca 35: 190-195. Per ry , D . J . , and P . K n o w l e s , 1990. E v i d e n c e o f h igh self-fer t i l izat ion i n natural populat ions o f eastern whi te cedar (Thuja occidentalis). Canad ian Journa l o f B o t a n y 68: 663-668. Per ry , D . J . , P . K n o w l e s , and F . C . Y e h , 1990. A l l o z y m e var ia t ion o f Thuja occidentalis L . i n Nor thwes te rn Ontar io . B i o c h e m i c a l Systematics and E c o l o g y 18: 111-115. P o l i t o v , D . V . , and K . V . K r u t o v s k i i . 1990. Gene t i c var ia t ion i n Pinus sibirica du Tou r . V . A n a l y s i s o f mat ing system. Gene t ika 26: 2002-2011 . Po tenko , V . V . , and A . V . V e l i k o v . 1998. Gene t ic d ivers i ty and differentiat ion o f natural popula t ions o f Pinus koraiensis ( S I E B . et Z u c c ) i n R u s s i a . S i l v a e G e n e t i c a 47 : 202-208. Po tenko , V . V . , and A . V . V e l i k o v . 2001 . A l l o z y m e var ia t ion and mat ing sys tem o f coastal popula t ions o f Pinus koraiensis S ieb . et Z u c c . i n Russ i a . S i l v a e Gene t i ca 50: 3-4. P r i m m e r , C . R . , H . E l l e g r e n , N . Sa ino , and A . P . M 0 l l e r . 1996. D i r e c t i o n a l evo lu t ion i n germl ine microsate l l i te mutations. Nature Genet ics 391-393. P u g l i s i , S., R . L o v r e g l i o , and M . A t t o l i c o . 1999. Subpopula t ion differentiat ion a long e levat ional transects w i t h i n two Ital ian populat ions o f Scots p ine (Pinus sylvestris L . ) . Forest Gene t ics 6: 247-256. R a d d i , S., and S. Sumer . 1999. Gene t ic d ivers i ty i n natural Cupressus sempervirens L . populat ions in T u r k e y . B i o c h e m i c a l and Systemat ic E c o l o g y 27: 799-814. Rajora , O . P . , A . M o s s e l e r , and J . E . M a j o r . 2000. Indicators o f popula t ion v i ab i l i t y i n red spruce, Picea rubens. II. Gene t ic d ivers i ty , popula t ion structure, and mat ing behavior . Canad ian Journa l o f Bo tany . 78 : 941-956. R a y m o n d , M . , and F . Rousset . 1995. G E N E P O P (vers ion 1.2) popula t ion genetics software for exact tests and ecumen ic i sm . Journal o f Hered i ty 86: 248-249. Rehfeldt , G . E . , 1994. Gene t ic structure o f western red cedar popula t ions i n the inter ior west. Canad i an Journal o f Forest Research 24: 670-680. R e u s c h , T . B . H . 2000. P o l l i n a t i o n i n the mar ine rea lm: microsate l l i tes reveal h i g h outcross ing rates and mul t ip le paternity i n eelgrass Zostera marina. Hered i ty 85: 459-464 . R e u s c h , T . B . H . 2001 . Fi tness-consequences o f gei tonogamous self ing i n a c l o n a l marine ang iosperm (Zostera marina). Journal o f E v o l u t i o n a r y B i o l o g y 14: 129-138. R i c e , W . R . 1989. A n a l y z i n g tables o f statistical tests. E v o l u t i o n 43 : 223-225 . 138 Richa rds , A . J . 1986. P lan t B r e e d i n g Systems. George A l l e n & U n w i n . L o n d o n , U . K . 529 p . R icha rdson , B . A . , S. J . B runs f e ld , and N . B . Klopfens te in . 2002 . D N A f rom bird-dispersed seed and wind-d i ssemina ted po l l en provides insights in to pos tg lac ia l co lon i za t i on and popula t ion genetic structure o f whi tebark pine (Pinus albicaulis). M o l e c u l a r E c o l o g y 11: 215-227. R i t l a n d , K . 1983. E s t i m a t i o n o f mat ing systems. In I sozyme i n plant genetics and breeding, Part A . S . D . T a n k s l e y and T J . Or ton (Eds. ) . E l s e v i e r Sc ien t i f i c Pub l ica t ions , A m s t e r d a m , pp. 289-302 . R i t l a n d , K . 1989. Cor re la ted matings i n the part ia l selfer, Mimulus guttatus. E v o l u t i o n , 43 : 848-859. . • , • . • R i t l a n d , K . 1990a. A series o f F O R T R A N computer programs for es t imat ing plant mat ing systems. Journal o f Hered i ty 81: 235-237. R i t l a n d , K . 1990b. Inferences about inbreeding depression based on changes o f the inbreeding coefficient . E v o l u t i o n 44: 1230-1241. R i t l a n d , K . 2002 . E s t i m a t i o n o f gene frequency and heterozygosi ty f rom p o o l e d samples. M o l e c u l a r E c o l o g y Notes 2: 370-372. R i t l a n d , K . , and Y . A . E l - K a s s a b y . 1985. The nature o f inbreeding i n a seed orchard o f D o u g l a s f i r as shown b y an efficient mul t i locus m o d e l . Theore t i ca l and A p p l i e d Genet ics 7 1 : 275-384. R i t l a n d , C , T . Pape, and K . R i t l a n d . 2001 . Gene t ic structure o f y e l l o w cedar (Chamaecyparis nootkatensis). Canad ian Journa l o f B o t a n y 79: 822-828. R o s s i , P . , G . G . V e n d r a m i n , and R . G i a n n i n i . 1996. E s t i m a t i o n o f mat ing sys tem parameters i n two Ital ian natural populat ions o f Fagus sylvatica. Canad i an Journa l o f Forest Research 26: 1187-1192. Ri i te r , B . , J . L . H a m r i c k , and B . W . W o o d . 2000. Outc ross ing rates and relatedness estimates i n pecan (Carya illinoinensis) populat ions . The Journa l o f Hered i ty 9 1 : 72-75 . Sampson , J . F . , S. D . Hopper , and S. H . James. 1989. T h e mat ing sys tem and popula t ion genetic structure i n a b i rd -po l l ina ted mal lee , Eucalyptus rhodantha. Hered i ty 63 : 383-393. S A S Institute. 1997. J M P user's guide, vers ion 3.2.1. S A S Institude Inc. C a r y , N o r t h C a r o l i n a . Savo la inen , O . K . , K a r k k a i n e n and H . Ku i t t i nen . 1992. Es t ima t ing numbers o f embryon ic lethals i n conifers . Hered i ty 69: 308-314. Schlotterer, C , R . Ri t ter , B . Har r , and G . B r e m . 1998. H i g h mutat ion rate o f a long microsate l l i te a l le le i n Drosophila melanogaster p rovides evidence for a l le le-speci f ic mutat ion rates. M o l e c u l a r B i o l o g y and E v o l u t i o n 15: 1269-1274. Schneider , S., D . R o e s s l i , and L . Exco f f i e r . 2000. A r l e q u i n ver 2 .000: A software for popula t ion genetics data analysis . Genet ics and B i o m e t r y Labora to ry , Dep t . o f A n t h r o p o l o g y and E c o l o g y , Geneva , Swi tze r l and . U R L : http:/ /anthro.unige.ch/arlequin. 139 Schroeder , S. 1989. Outc ross ing rates and seed characterist ics i n damaged natural populat ions o f Abies alba M i l l . S i l vae Gene t i ca 38: 185-189. Seavey, S. R . , and K . S. B a w a . 1986. La te -ac t ing se l f - incompat ib i l i ty i n angiosperms. T h e B o t a n i c a l R e v i e w 52: 195-219. Shaw, D . V . , and R . W . A l l a r d . 1981. A n a l y s i s o f mat ing system parameters and popula t ion structure in Doug la s - f i r us ing s ingle- locus and mul t i locus methods. In Proceedings o f s y m p o s i u m on i sozymes o f N o r t h A m e r i c a n forest trees and insects, J u l y 27, 1979. P a c i f i c Southwest Forest and Range E x p . Stn. , Forest Serv ice , U . S . Depar tment o f A g r i c u l t u r e , B e r k e l e y , Ca l i f i f o rn i a . pp. 18-22. Shaw, D . V . , and R . W . A l l a r d . 1982. Es t ima t ion o f outcross ing rates i n Doug la s - f i r us ing i s o z y m e markers . Theore t i ca l and A p p l i e d Genet ics 62: 113-120. Shea, K . L . 1987. Effects o f popula t ion structure and cone produc t ion on outcross ing rates i n E n g e l m a n n spruce and subalpine fir. E v o l u t i o n 4 1 : 124-136. Shea, K . L . 1990. Genet ic var ia t ion between and w i t h i n popula t ions o f E n g e l m a n n spruce and subalpine fir. G e n o m e 33: 1-8. S i eg i smund , H . R . , E . D . Kjaer , and U . B . N i e l s e n . 1996. M a t i n g sys tem estimates and effective popula t ion numbers for an isolated noble fir (Abies procera) c l o n a l seed orchard i n D e n m a r k . Canad ian Journal o f Forest Research 26: 1135-1141. S i eg i smund , H . R . , and E . D . Kjaer . 1997. Outc ross ing rates i n two stands o f noble fir (Abies procera Rehd . ) in D e n m a r k . S i l v a e G e n e t i c a l 4 4 - 1 4 6 . S i m o n , J . P . , Y . Berge ron , and D . G a g n o n . 1986. I sozyme un i fo rmi ty i n popula t ions o f red pine (Pinus resinosa) i n the A b i t i b i R e g i o n , Quebec . Canad i an Journa l o f Forest Research 16: 1133-1135. Snyder , T . P . , D . A . Stewart, and A . F . S t r ick ler . 1985. T e m p o r a l analysis o f breeding structure i n j a c k pine (Pinus banksiana L a m b . ) . Canad ian Journa l o f Fores t Research 15: 1159-1166. So l t i s D . E . , M . A . Gi tzendanner , D . D . S t rengem and P . S. So l t i s . 1997. Ch lo rop las t D N A intraspecif ic phylogeography o f plants f rom the P a c i f i c Nor thwes t o f N o r t h A m e r i c a . P lan t Systemat ic and E v o l u t i o n 206: 353-373. Sorensen, F . 1969. E m b r y o n i c genetic l o a d i n coastal Doug la s - f i r Pseudotsuga menziesii var. menziesii. T h e A m e r i c a n Natura l i s t 103: 389-398. Sorensen, F . C . 1982. The roles o f p o l y e m b r y o n y and e mbr yo v i a b i l i t y i n the genetic system o f conifers . E v o l u t i o n 36: 725-733 . Sorensen, F . C . 1999. Re la t ionsh ip between self-fert i l i ty, a l loca t ion o f g rowth , and inbreeding depression i n three coniferous species. E v o l u t i o n 53: 417-425 . Sproule , A . T . , and B . P . D a n c i k . 1996. T h e mat ing system o f b l ack spruce i n north-central A l b e r t a , Canada . S i l v a e Gene t i ca 45 : 159-164. 140 Stacy, E . A . , J . L . H a m r i c k , J . D . N a s o n , S. P . H u b b e l l , R . B . Foster, and R . Cond i t . 1996. P o l l e n dispersal i n low-dens i ty populat ions o f three neot ropica l tree species. T h e A m e r i c a n Natura l i s t 148: 275-298. Stauffer, A . , and W . T . A d a m s . 1993. A l l o z y m e var ia t ion and mat ing sys tem o f three Doug la s - f i r stands i n Swi t ze r l and . S i l v a e Gene t i ca 42 : 254-258. Steinhoff , R . J . , D . G . Joyce , and L . F i n s . 1983. I sozyme var ia t ion i n Pinus monticola. Canad ian Journa l o f Forest Research 13: 1122-1132. Stern, K . , and L . R o c h e . 1974. Genet ics o f Forest Ecosys tems . B e r l i n : Sp r inge r -Ve r l ag . Stebbins, G . L . 1950. V a r i a t i o n and evo lu t ion i n plants. C o l u m b i a U n i v e r s i t y Press, N e w Y o r k . 643 p. Stoehr, M . U . , and Y . A . E l - K a s s a b y . 1997. L e v e l s o f genetic d ivers i ty at different stages o f the domest ica t ion cyc l e on inter ior spruce i n B r i t i s h C o l u m b i a . Theore t i ca l and A p p l i e d Gene t ics 94: 83-90. Strauss, S. H . , and M . T . C o n k l e . 1986. Segregat ion, l inkage , and d ivers i ty o f a l l ozymes i n knobcone pine. Theore t ica l and A p p l i e d Genet ics 72: 483-493 . T r e m b l a y , M . , and J . P . S i m o n . 1989. Genet ic structure o f marg ina l popula t ions o f whi te spruce (Picea glauca) at its northern l i m i t o f d is t r ibut ion i n Nouveau -Quebec . Canad ian Journa l o f Forest Research 19: 1371-1379. Truesdale , H . D . , and L . R . M c C l e n a g h a n . 1998. Popu la t ion genetic structure o f tecate cypress (Cupressaceae). Southwestern Natural is t 43 : 363-373. Turgeon , J . , and L . Bernatchez . 2001 . C l i n a l var ia t ion at microsate l l i te l o c i reveals h i s to r ica l secondary intergradation between g l ac i a l races o f Coregonus artedi (Teleostei : coregoninae) . E v o l u t i o n 11: 2274-2286. T u s k a n , G . A . , K . E . Franc is , S. L . Russ , W . H . R o m m e , and M . G . Turner . 1996. R A P D markers reveal d ivers i ty w i t h i n and among c l o n a l and seedl ing stands o f aspen i n Y e l l o w s t o n e N a t i o n a l Park , U S A . Canad ian Journal o f Fores t Research 26: 2088-2098. U c h i d a , K . , N . T o m a r u , C . T o m a r u , C . Y a m a m o t o , and K . O h b a . 1997. A l l o z y m e var ia t ion i n natural populat ions o f H i n o k i , Chamaecyparis obtusa (Sieb. E t Z u c c . ) E n d l and its compar i son w i t h the plus-tree selected f rom ar t i f ic ia l stands. B r e e d i n g Sc ience 47 : 7-14. U d u p a , S. M . , and M . B a u m . 2001 . H i g h mutat ion rate and muta t ional bias at ( T A A ) n microsate l l i te l o c i i n ch i ckpea (Cicer arietum L . ) . M o l e c u l a r Genet ics and G e n o m i c s 265: 1097-1103. U y e n o y a m a , M . K . , K . E . Ho l s inge r , and D . M . W a l l e r . 1993. E c o l o g i c a l and genetic factors d i rec t ing the evo lu t ion o f self-fer t i l izat ion. O x f o r d Surveys o f E v o l u t i o n a r y B i o l o g y 9: 3 2 7 - 3 8 1 . V a z q u e z , J . F . , T . Perez , J . A l b o r n o z , and A . Domfnguez . 2000. E s t i m a t i o n o f microsate l l i te mutat ion rates i n Drosophila melanogaster Gene t i ca l Research 76: 323-326. 141 V i d a k o v i c , M . 1991. Con i fe r s : morpho logy and var ia t ion ( M . So l jan , Translator) . G r a f i c k i z a v o d Hrva t ske , Zagreb . 756 p. V i g o u r o u x , Y . , J . S. Jaqueth, Y . M a t s u o k a , O . S. S m i t h , W . D . B e a v i s , J . S. C . S m i t h , and J . D o e b l e y . 2002. Rate and pattern o f mutat ion at microsate l l i te l o c i i n maize . M o l e c u l a r B i o l o g y and E v o l u t i o n 19: 1251-1260. vonRud lo f f , E . , and M . S. L a p p . 1979. Popu la t iona l va i ra t ion i n leaf o i l terpene compos i t i on o f western red cedar, Thuja plicata. Canad ian Journal o f B o t a n y 57: 476-479 . vonRud lo f f , E . , M . S. L a p p , and F . C . Y e h . 1988. Chemosys temat ic study o f Thuja plicata: mult ivar ia te analysis o f leaf o i l terpene compos i t i on . B i o c h e m i c a l Systematics and E c o l o g y 16: 119-125. W a n g , C . - T . , and T . - P . L i n . 1998. Inheritance and l inkage relat ionships o f a l lozymes and est imat ion o f outcross ing rates i n a seed orchard o f Cunninghamia konishii H a y . S i l v a e Gene t i ca 47 : 33-37. Warner , B . G . , and R . W . M a t h e w e s . 1982. Ice-free condi t ions on the Queen Charlot te Islands, B r i t i s h C o l u m b i a , at the height o f late W i s c o n s i n g lac ia t ion . Sc ience 218: 667. W e i r , B . S. 1990. Gene t ic D a t a A n a l y s i s . Sinauer. Sunder land , U . S . A W e i r , B . S., and C . C . C o c k e r h a m . 1984. Es t ima t ing F-stat ist ics for the analysis o f popula t ion structure. E v o l u t i o n , 38: 1358-1370 . Whee le r , N . C , and R . P . Gur i e s . 1982. Popu la t ion structure, genie d ivers i ty , and m o r p h o l o g i c a l var ia t ion i n Pinus contorta D o u g l . Canad ian Journa l o f Forest Research 12: 595-606. W h i t h a m , T . G . , and C . N . S lobodch iko f f . 1981. E v o l u t i o n b y ind iv idua l s , plant-herbivore interactions, and mosa ics o f genetic va r iab i l i ty : T h e adaptive s igni f icance o f somatic mutat ion i n plants. O e c o l o g i a 49: 287-292. W i l l i a m s , C . G . , Y . Z h o u , and S. E . H a l l . 2001 . A c h r o m o s o m a l reg ion p romot ing outcrossing i n a conifer . Genet ics 159: 1283-1289. W i l s o n , E . B . 1927. Probab le inference, the l aw o f succession, and statistical inference. Journa l o f the A m e r i c a n Stat is t ical A s s o c i a t i o n 22: 209-212 . X i e , C . Y . , B . P . D a n c i k , and F . C . Y e h . 1991. T h e mat ing sys tem i n natural popula t ions o f Thuja orientalis. Canad ian Journal o f Forest Research 21 : 333-339. X i e , C . Y . , B . P . D a n c i k , and F . C . Y e h . 1992. Gene t ic Structure o f Thuja orientalis. B i o c h e m i c a l Systematics and E c o l o g y 20: 433-441 . X i e , C . Y . , and P . K n o w l e s . 1994. M a t i n g system and effective po l l en i m m i g r a t i o n i n a N o r w a y spruce (Picea abies ( L . ) Kars t ) plantat ion. S i l v a e Gene t i ca 43 : 48-52 . Y e h , F . C . 1988. I sozyme var ia t ion o f Thuja plicata (Cupressaceae) i n B r i t i s h C o l u m b i a . B i o c h e m i c a l Systematics and E c o l o g y 16: 373-377. Y e h , F . C , and Y . A . E l - K a s s a b y . 1980. E n z y m e var ia t ion i n natural popula t ions o f S i t k a spmce (Picea sitchensis). 1. Genet ic var ia t ion patterns among trees f rom 1 0 I U F R O provenances. Canad ian Journal o f Forest Research 10: 415-422 . 142 Y e h , F . C , and D . O ' M a l l e y . 1980. E n z y m e variat ions i n natural popula t ions o f Douglas - f i r , Pseudotsuga menziesii ( M i r b . ) F ranco , f r om B r i t i s h C o l u m b i a . 1. Gene t i c var ia t ion patterns i n coastal populat ions. S i l v a e Gene t i ca 29: 83-92. Z h e n g , Y . Q . , and R . Ennos . 1997. Changes i n the mat ing systems o f popula t ions o f Pinus caribaea M o r e l e t var. caribaea under domest ica t ion. Forest Genet ics 4: 209-215. Z h e n g , Y . Q . , and R . A . E n n o s . 1999. Gene t ic var iab i l i ty and structure o f natural and domest icated populat ions o f Car ibbean pine (Pinus caribaea M o r e l e t ) . Theore t i ca l and A p p l i e d Genet ics 98: 765-771 . 143 Appendix I Description of genetic diversity parameters PPL: Percentage o f p o l y m o r p h i c l o c i , ca lcula ted as the number o f l o c i w i t h more than one al lele d i v i d e d b y the total number o f l o c i assayed. A/L: N u m b e r o f alleles per locus averaged. Includes both p o l y m o r p h i c and m o n o m o r p h i c l o c i . Hes: T o t a l genetic d ivers i ty for a species was obtained b y averaging genetic d ivers i ty over a l l l o c i ( po lymorph i c and monomorph ic ) . where p . is the mean frequency o f i t h a l le le across a l l popula t ions ana lyzed for a species. Hep: T o t a l genetic d ivers i ty w i t h i n a popula t ion and was obtained b y averaging genetic d ivers i ty over a l l l o c i i n a popula t ion ( inc lud ing p o l y m o r p h i c and m o n o m o r p h i c ) . where p , is the frequency o f i ' t h a l le le i n a popula t ion , and Hepis the mean over a l l populat ions . HT: T o t a l genetic d ivers i ty for a species ca lcula ted us ing on ly p o l y m o r p h i c l o c i Hs: M e a n d ivers i ty w i t h i n a popula t ion ca lcula ted us ing on ly p o l y m o r p h i c l o c i Gsl: T h e propor t ion o f the genetic d ivers i ty res id ing among popula t ions o f a species. G s t is the average over a l l p o l y m o r p h i c l o c i . XJ a < a o cu cS C J o cS PH 3 u CU 4 3 X I C J C J e B o I CM O a .5 >-C J 4 3 C J ' o C J a en oo cj 5 •a C J > JD B O 3 PH O PH C J 4 3 -t—» -4—» CS o -35 o 60 N o cj -4-1 C J 4 3 X I 3 C J PH X cj B c S C J s C J ' l-l C J > cS C J 3 cs a C J B O XJ C J >N C J > s a •3 £ o cj 3 PH O O m PH J> XJ '<-> 5 » a o o - C J C J XJ 3 c o C J 4= 4-* 1-a a _o J3 "3 PH o PH _a >> 'S3 C J > X) CJ a CS "3 P H o OH CJ o 4 3 & O o PH C J 60 .2 a C J C J OH sf 0, 3 o O o PH JD tt, = 3 O «» t VH S 2 CO CO 2 3 a I O cu 60 4-1 o 5t CO cu es cu cu '> X ) o X CS H cu B < " CS B 4 3 -4—* *CO O 60 >> N O I i 0J. C U 4 3 X ) C U -fcH C J C U PH X C U § C U a C J 4 3 3 B 1 X ! C J XJ 13 6 U o o '3 cj B 60 C J C J fc > ia cs x) I&: a c/3 on O N 1 CU O N O N >> 4 3 W O N O N ON 2 o o C N DH ~ -s « CO *. • CO « • N^ I 4 S .2 * a-J3 i f o O o O N O N o o O N O N 04 O N ON 43 op •a J3 o O N O N « J3 CU o c55 p-O N O N m o iJ5 T- "5 a 8 TC C3 a a « s a 3 C3 3 5^ Cj C3 0 0 R 0 0 c cu 3 5 0 0 c a CU Cj 0 0 O N W W . U a CJ 3 .a 0 0 cs oo C N C N in en r- Tt" I T ) o o o o o O o — C5 c > d d d d d d C O — o 0 0 m r - N O i> C N N O o in r - O N C N N O > n N O C O N O r - C N C O N O N O 0 0 r ~ , , Tf C N m N O C O r - N O N O O N 0 0 C N 0 0 c n > n — C O C N " C N C N C N C N Tf 0 0 c n oo in m O N m C N O N oo 1/1 O N 0 0 o o —— C N o o o d d d d d d C O T(" 0 0 C O 0 0 N O in ro C N C N C O C O O N .—( C N O N 0 0 0 0 N O — H O N m o .—1 0 0 —— — — — — O - H C N C N - H C O o d d d d d d d d d d d d o 0 0 O N O N m O N o r-C N O N TT - N O m >n 0 0 O N O C N TI- c~ Tt 1 > C N C N C N C N C S C N C N V •« "c3 "c3 "ca "ca "c3 "ca "c3 S3 l a | 3 ] 3 | 3 % 4 3 o 03 ca ca ca ca ca t a ca l a 1- l a l a C B B B B B B B B B B o B B n 0 00 cn CO CN OS so so •>* "* •<* t -CO r- O SD CN so OO SO 06 m CN CN t-; OS co co co r~ SO OS SO SO r-» SO m CN m CN •* CN •ct m 00 CO Os CN CN CN CN CN Os OS CN ro m >n CN r-00 0 CO SO m CN so .2274 1795 O 0 O 0 O 0 0 0 0 O 0 0 SO P-m .2465 in co - CO >n r~ m CN r~ 00 0 CN 00 CN m m CN •* CN 00 CN SO LZZ CO 00 O 0 0 O 0 O •, 0 O 0 0 0 0 , O 0 O O 0 0 0 CN >n CN 0 O Os in CN CN in 00 CN in 00 00 co V~i CN CN CN Os CN r~ CO 00 m SO SO CN O CN SO CN 13 13 13 % cd cd cd B B B CJ CJ u CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ cd cd cd cd cd C3 cd cd cd cd cd cs cd cd cd cd cd cd cd CJ CJ cj CJ CJ CJ CJ CJ CJ cj cj CJ cj u CJ CJ CJ CJ CJ 0 0 0 0 cj O 0 0 0 CJ u 0 0 0 0 CJ 0 0 CJ cd cd cd cd cd CO cd cd cd cd cd cd cd cd cd cd cd cd cd fin Pin Pin Pin Pin Pin Pin Pin Pin Pin Pin Pin Pin Pin Pin Pin Pin Pin Pin S T3 CD a a o a CD Cu a t/3 o o CN > o "cj > <*) o J«i c Si O CU OS Os ON <3 C CJ •a LL. <& B O O 60 t-oo Os oo Os Os CN 2 is o 60 CJ O CJ - J u •S « -s CO o o CN 60 •3 J3 o 60 •3 CJ —1 =3 O o S3 J3 O B O o Os 00 Os X ca B o 3 s OS 00 Os "S K ca B o 3 OS Os Os 60 s ° <=> ?J 00 00 OS OS C i— — 2 >, B CJ CJ 03 b to 3 K S| S 1 3 K SI £ to 3 C £1 o c o £ to 3 C SI a fi-ts u Q O to 3 C S| 3 C s: 3 ft. to 3 S s: "3 3 K SI i : ^ «o oj 3 3 S s: 3 S s; 3 B s: a §P >> c a ? ts o "a 3 cu 3 S s; 3 C SI S o o CN m m CO CN SO CO CN O •* o o o o o o O o m CO in -* SO CO oo en m O CO CN SO SO •<* m SO c~ sb m SO O in oo SO 00 SO oo CO t-; so t-; Ov CN oo OS t-; so Os OS ~* CN " CN aaverage. aaverage. so CN CN CN 00 in o 00 00 SO CN -* CN 00 OS o SO so CN 00 CN 00 SO in O o o o © ' o © O o d o o o o las SO r-SO CN CN CN •<* 00 m o 00 00 00 so r-CN "* CN 00 OS o SO "3- SO CO CO 00 CN m SO so CN OS CN 00 m © © © ' © ' © ' o o © ' © ' © ' o O o o O o o o o POP CO - - oo CN m 00 O CN m CN 00 CO CN CN SO CN CO CO SO •* CN SO CN CO o CN CO CO CN oo CN CN o "ca 13 13 13 In 3 13 13 13 13 13 13 13 13 % 13 % 13 1 CJ o 13 13 13 13 ca s ca B ca B fa B B ca B ca B ca B ca B ca B B ca B ca B ca B ca B ca B ca B ca c CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ cj CJ CJ CJ CJ CJ CJ o CJ cj cj CJ CJ CJ CJ cj CJ CJ cj cj CJ CJ o CJ C3 ca ca ca ca ca ca ca CO ca ca ca ca ca ca ca ca ca ca B s c c B c B B B E B B B B B B B B B 0- CU CM CL, OH OH OH OH OH OH OH CL, CU Cu Cu CU CU 0. OH <3 00 3 to •8 3 cu tu oS ON ON ON OO OO OO ON ON ON ^ i tO to co T 3 T3 T 3 < < < =3 <=3 =8 5 0J3 C3 tU >• a l»5 OH o CH s a) fa -a cu 3 a a o o T3 a cu CM CM < a, C/3 O 3 00 3 3 NO o\ ON •a I 2 o o o CN NO ON ON CO C CU T t ON ON 13 as =3 S3 o o CN t-ON ON f -ON ON oo ON ON aa U cj ON ON 0O ON CS 2 •o JJ „ fi 5 O CJ o o CU CU tu tu tu CN CN m oo O d d d d d d d d d d d d in NO in r-o r> CN 3 oo o r-o 00 o CO m T t CN O CN CN in o oo CN CN 0 0 oo CN T t CN ON CN d d d d d d d d d d d d d d d d d d m T i -NO co T T CN ON T t CO CN ON CN r» CO NO T t CN CN CN NO to CN o CN O CN o CN m ON CN o T t o r-CN 00 CN CN CN CO o CN 13 3 13 b 13 b 13 b 13 b 1 b 13 b 13 b 13 b nati 3 M B -3 CS B - J 13 B ca B •a ca B ea B B ca B ca B CX 3 u S3 C/3 5! cu an « u cu > es cx iccT PH HH u id CD =S . a a o o T3 a o> OH < ex C/J 2 c :s cj © S ON ON o ON ON CN ON ON CO cs C „i, CJ X ffl CO CO o NO CN os x; T3 S3 cx 3 u u u D . 3 u s :§ cj cj o s s s o a cS a, e a, 3-ON o CO CO t~ t- CN NO CN CO lO NO ON CN oo in 00 CN ON ON in o CN o o o -* ON CO ON NO Tf o o o O CN o o o o O O NO NO © in >• > < < a ft x) S3 X! p X) CD CD co -O "c3 3 -4—* rt z > CO CJ CD CM CO CN en xi B rt a o o o to CD • -H o CD CM CO CN OO a CD CD 1- 1 - 1 oo B B CD CM 00 3 CO o l-H CJ -4—> B O x) CD CD > 3 PH O PH CJ B B CLT O OH Xl" CJ CJ > l-H B CO ' o O 3 o 3 O 4-1 O cj CD 1 3 B O hJ PH •a s 1 C8 I* en Tt as ON 00 ON O ON ON S3 Si s I ^ 3 ^ B < 5 g 3 2 "O tu ~* •C^ "3 CO > w o) C3 CU j3 Q Z M t— ON ON ZLJ CJ °> O O o CN M >> N NO ON ON NO ON ON •a B 3 E a CO o, =3 £> o a o a CN CN 00 o -a •s o 73 .3 ON ON ON w =3 3 BC 2 « Tt ON ON =3 =3 -S o 3 13 S3 es a a B 13 13 73 13 JB a o 73 t o 13 •3 73 S3 J3 o a. TC -CS TC s C3 -a TC a 8 CJ TC 8 a. TC l Cj •3 CJ 5 TC 3 3 3 3 3 s cj CJ o •s CJ o 3 W X CN CO r-oo U~t ON ON C- CO NO ON m NO 00 ON CN NO CN 00 ON 00 d d d 00 OO P - Tt ON d d d d d d o NO ON m ON ON Tt CN CO m NO ON Tt co Tt Tf oo r~ 00 ON m Tt r~ NO CN ON o ON CO ON d oo d d d 00 ON NO OO r» 00 00 d 00 O d d d d d d d d d NO o ON ON Tt oo CN CO UO CO oo CN CO CO Tt o OO ON ON ON tr> Tt o o o ON Tt ON O © O d 00 ON NO CN ON ON d 00 d 00 I OO >A 4 d d d -1 d d d d NO o r-; NO o 1 • • • 1 < ^ I D ON Tt r- NO CO Tt NO oo ro C- o CN d NO 00 ON 00 d O d d d ON ' oo Tt CN CN ON u-> CN CN CN o CN CO oo CN CO ON 00 t> >r> r- NO 73 S3 o CJ cu tu CJ tu CU CJ CU CJ CJ CJ CU CJ CU tu CJ CCS cs cs cs cs cs CS C3 ctf cS cS cs CS CS CS CS CJ CU tu CJ CU CU tu CU CJ CJ CJ CU tu tu CU CJ CJ CJ CJ CJ CJ CJ CJ o CJ o o CJ CJ o o CJ cs cs cs cs cs cs cs CO cs cS CS ICS tc! ICS cS cS B s B B B B B B B B B B B B B B ON Tt E? o 3 cd OH i M l ~ 00 cu O N T3 CJ CJ j-J on N O r -o o o O N O N 3 'So c 2 N O O N O N B CJ o JS Q « C O c^ j Tg1 °^ — 73 c P J u C L C O o s ">> o 03 00 O N 03 CN O N O N CN O N O N •a O N O N O O O 2? •3 J3 o o o \£3 w 5-" .5 o 3 3 a u o o CN o •a N O O N O N >N N O 3 03 I 7j JS U •a >> n co C O m O C N m o O N C N m in m C O C O O N O N N O C N O N oo oo oo m © m O N r - o O N d © O o © © o © 3 N O m m N O O N O N d © N O in CO oo CO *d- r - o in CO O N CN 00 m O N CO r- CN co CO oo o O N CN oo CN O N 00 o o 00 o © O N © O N 00 m © in O N O N P- o © o O o © © © © o © © OH o BH > N O O O CO CO N O 00 m >n O N m O N O N CO oo O N d O d d O N d d d d i m 00 o in O N o N O en O N o d d d © CN CN CN CN CN CN CN N O CN N O r - in r - C O 00 00 O N oo N O d d d d O N O N O N 00 C O d d N O CN N O m m O N d O N N O • * d O N o O N 3 d ^H N O CN oo 00 CO O N 00 O N d d d d CN CN d - CN N O o o 1 JS _ — T3 cd cd 5 I-H Ii cd 3 3 J= cd cd 3 3 •a S3 1 J 3 — -a 3 i i B O O 73 b 73 b 73 b 73 b B cd B cd B ca B cd B B B C CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ cd cd cd cd cd cd cd cd cd cd cd cd cd cd C3 cd cd cd cd cd CJ CJ CJ CJ CJ cj CJ CJ CJ CJ CJ CJ cj CJ CJ CJ CJ CJ CJ CJ o o CJ y CJ o CJ cj o CJ CJ CJ cj o CJ CJ CJ CJ CJ CJ cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd cd B n B B B C a B B B B B C B B B B C a S E E £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ £ T3 CD =3 a -i—i a o o •a c o 4 o B o o 0. ON ON T t ON ON o ON ON O N ON O N — ON Q ON ON a o o oo co ON oo ON © 2 CN * 2 op •3 •3 ca JB o op -3 .2 oo o T t no m O N —-, N O m uo T t T t O N ro CN N O co OO N O ON O N oo O N o O N oo oo 00 N O OO © T t d d d d d d d d d d ON d s If! Oa Z | O N 00 00 r - T t m m 00 CN N O N O no m T t t > N O o OO N O o T t O N ro CN N O o oo oo N O O N CN O N O N O N d O N 00 O N d ON 00 00 oo d N O 00 d T t 0.98 00 0.86 d d d d d d d d d d d d 0.98 d 0.86 CN ON d i ON ON CO ON d T 3 S3 •ss S P 894-1 0.808 0.983 0.94 0.971 -0.97 0.986 -0.84 o .926-00 00 0.68 0.751-0.765 o d 0.751-CO CO "JO T t CN in NO o CN NO CN T t 73 b 73 b 73 b 73 b 73 b 73 b 73 nati CO E nati nati £3 CO E natu nati — NO CN - H m m OO NO NO NO oo ON d O ON d f~- 1 t — o ON 00 d d d i»o —-1 NO T t NO NO CO - H _I 73 73 b 73 b 73 b 73 b c3 c3 ca CO z Z z Z CC) t-H CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca ca CJ CJ CJ CJ CJ CJ CJ CJ cu cu CJ CJ CJ cu CJ CJ CJ CJ CJ CJ o o CJ CJ o o CJ CJ o o CJ o CJ CJ CJ CJ CJ CJ CJ CJ ca ca ca ca ca ca ca ca ca ca CB ca ca ca ca ca ca ca ca ca B B B B B B B E B E E E E E E E E E E E £ E E E E £ £ £ £ £ £ £ £ £ £ £ E £ £ E T3 CU 3 _C a o o T3 a cu OH CM < "8 cu ft a a s R Cj cj «D 3 R 3 R a. 3 R C3 C-5 3 R -C3 •a o CS s 3 3 3 R a. 3 R a. 3 R a. C3 ft. a c <3 CO CJ .R S. 3 R -8 s Cj a. 3 R CN o as Os o 00 Os OS O •2 o 5 _*e CN 00 Os oo •3 J3 O CO ^ 2 o «8 o OH OS 00 as 3 2 2 x •a CN OS OS 00 a o OS Os 13 U CO m Os OS =8 o o o CN 2 00 c o Q o o o CN 2 00 a o Q Os Os CN 00 oo OS OS I < =8 cd J3 CO O CN m CN SO CN o Os OS m in CN OS SO 00 d d oo 0.96 Os Os 00 00 d 0.96 d d d d 3 so Os d oo d o A oo Os d Os d PH , PH CN — i —' CN - H 1 o _ i3 Cd -o S3 J3 •a S3 J3 •a S3 J3 __ __ T3 ca ca b 3 3 I tS 1 CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ ca ca ca ca ca ca ca ca ca ca ca CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ o CJ o CJ CJ o ca ca ca ca ca ca ca ca ca ca ca B B B c p b a S3 a a c OH a E a E c £ c £ c E c E c E T3 CO a -4—1 C o cj a CD a a. < •8 R O ft, do R a ft. 3^ C3 "3 i; i a co 3 R at a R a R a. to 3 R £1 to 3 R a R OH 3 R 3 -ft a a R 3 -ft 3 R 3 R 0, ft co I to 3 R a 00 3 to •§ 3 1 < C8 J3 CO m CN Os 00 in Os *^ 00 SO 3 00 CN o o CN 00 Os OS in u-> Os Os 00 Os m OS SO CN o CO 00 d d oo OS d d OS d OS oo Os OS OO Os Os d d d d d d d d d d d d SO oo CN CN CN CO SO CN CO SO Os OS 00 CO r~ SO SO O SO Os OS in oo OS Os d OS OS Os oo Os 00 d d r-d d d d SO d d d 1 m d i SO d 1 * *d- oo CO r- Os . 00 r-- in OS OS CO OS m Os m oo d OS d d d CO O CO -d- m 73 b b 13 b "c3 b p cs B p cd B p cd B p cd B R cu 5 CJ Oo 3 to •§ 3 R cu S C3 00 3 to •§ 3 •ct OS OS in CU 00 OS 2 a cu _L. •S <*! co ca c ^ -a S 2 so oo OS •c o -a CJ —1 O OS OS o £ OS m d Os OS OS •ct OS Os •a o o CN o U OS OS a =8 bo o Os d •o PH O PH > z m m so m CN t - Os •ct OS o OS d •* m in OS d © d d © d o •cl- CN o o O so m 00 m m 00 o o CO CN o i n CN CN SO so r - CO oo CN o Os OS OS 00 OS OS OS 00 d so CN CO in OS d d d d d d d d © d d d d d d o so CN OS so CN m m m CN r- CN o r- CN So p SD OS CO CO m o Os Os OS d •*Cf Os OS Os d CO d CN OS d d . d ^1 © d d d d d • m 1 i CN oo r~ o CO CO m o r-OS OS m CN d d d d d -ct Os CO CN — 1 SO so o oo CN Os ' r - CO OS CN V V •a -o •a H S3 S3 S3 "ca c3 "ca ~a "ca "ca "ca J3 J3 J3 J3 j= •3 | 3 | 3 ca Si a o O u< ca a ca ca ca ca ca c O o O O c o c s s B o c CN o OS d CJ CJ CJ CJ CJ CJ CJ CJ ca ca ca ca CO ca ca ca CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ CJ o CJ o CJ ca ca ca ca ca ca ca ca <3 (3 C3 (3 (3 n a c a. 3 u u u u u -a o T3 CD =3 B -*-» B O o T3 B CD CH DH < Co to CO to to to to cu cu cu cu cu cu cu KJ KJ KJ KJ S S s s c K c cu cu cu cu u cu cu s s 5 s s s s a <3 a a a a 00 00 00 00 00 00 s 3 3 3 3 3 3 co to to .to .to to o O o o O 0 o •a •C3 -a •a •« •« a 3 3 3 3 3 3 cu cu cu cu cu cu cu to to to to to to to a. a. a. a. a. a, a. <3 00 3 cm ca IH cu > < o 3" g 2 s cj Cj 0 .« 3" g s cj Cj « 5-a. C3 3-g a. C3 ca ca CA c« cu u a. a u cu s i ca IH CU > < s « S a •« 00 •5 5 s o •»! ca H cu SA ca IH CU Si T t cu c IU es S 04 c s PH o PH ea, fa T3 CJ .g •I—* a o cj T3 a CJ Cu CM < CS J2 ca T t T t •* ON ON ON ON ON ON -M -M -M •^J C3 C3 , >. CJ o CJ CO CO 5 5 CO CO CO VO VO p *ct o p CO vo VO p -ct o p CO VO vo o "3" o p 0 0 so CM O O CA > 0.59-0.67 1-0.92 0.93 0.986 0.951 0.921 0.998 0.95 0.642 0.967 1.077 •1.025 1.031 •1.304 1.006 0.59-0.67 d 0.97-0.986-1.008-1.03-0.989-cs - cs cs - cs - -•cf •* m o cs •* CO m CO o cs m Ov 13 b a b 73 b 13 b "co "co b "co b 13 b "co b 13 b 13 b a 13 b nati nati nati nati natx nati nati nati nati nati nati nati nati a CJ CJ CJ CO CO CO CJ CJ CJ CJ CJ CJ CO CO CO e t ts 2 2 2 CJ CJ CJ CJ CO CO CO CO CJ CJ CJ CJ o o o CJ CO CO CO CO J 3 X I J 3 J D CO CO CO CO LL, LL, LL, LL, •e o •a 3 LO S 2 J 3 3 C< T3 S3 o CO J3 O . CO •a CU 3 o CJ -a a a OH < u s •§ Cj •5 cj 3 S 5 •3 -S O ~ cj « -S SJ a J3 ,t-> 2 _g cu £ •3 -5 o K a 00 cu 5 a <3 <3 •SP -5 CJ K H <3 3 .3 -« 4 5 •3 .« .3 3 -B 5 o 5 0 e o •S S 156 A p p e n d i x IV Western red cedar foliage DNA CTAB extraction protocol (for 12 samples). 1. C h i l l mortar and pestle i n - 2 0 ° C freezer overnight 2. Preheat 2 x C T A B buffer at 6 0 ° C (or more 6 5 - 7 0 ° C ) 2 x C T A B buffer 0.1 M T r i s H C L p H 8.0 1.4 M N A C L 0.020 M E D T A 2 % C T A B d H 2 0 2% P-mercaptoethanol for 200 m l : 20 m l l M T r i s 56 m l 5 M N A C L 8 m l 0.5 M E D T A 40 m l 10% C T A B 76 m l d H 2 0 4 m l p - m e r c a p 3. G r i n d 0 .25g o f fol iage us ing l i q u i d ni t rogen and place i n 50 m l tubes 4. A d d 15 m l o f C T A B buffer per sample, incubate 30-45 minutes , s w i r l i n g every 10 mins 5. S p i n on table top centrifuge at 4000 r p m at 0 ° C for 10 mins to r emove most leaf debris 6. Transfer supernatant to new tube. A d d an equal v o l u m e o f c h l o r o f o r m : i s o a m y l a l coho l (24: T) to the extract. M i x 15 mins i n rotating machine . 7. Transfer top part to new tub and spin 10 0 0 0 r p m at 4 ° C for 10 mins (or more 30 mins) 8. R e m o v e top phase and transfer to new tube. A d d 2/3 v o l u m e c o l d i sopropanol . M i x gently and spin d o w n D N A Pel le t . P o u r out i sopropanol and air dry . 9. W a s h i n 20 -25 m l wash buffer o f for 120 m l : 7 5 % ethanol 0.01 M a m o n i u m acetate H 2 0 10. S p i n , pour out, dry 11. R e d i s o l v e D N A i n 500 |xl T E at 4 ° C 12. A d d R N A s e at a concentrat ion o f 10n l /ml Incubate at 3 7 ° C for 30 m i n 13. A d d Proteinase K to a final concentrat ion o f 10^1/ml Incubate at 3 7 ° C for 30 m i n 90 m l 100% ethanol 0.24 m l 5 M a m o n i u m acetate 30 m l H 2 0 ( R N A s e 10 m g / m l stock) 14. A d d equal v o l u m e o f phenol and ch lo ro fo rm (1:1), rotate for 15 m i n . S p i n d o w n at 8000 r p m at 0 ° C for 10 m i n Transfer supernatant to new tube 15. A d d s l ight ly less than equal v o l u m e o f ch lo ro fo rm: i soamyla l coho l (24:1). M i x for 10 S p i n d o w n at 10 0 0 0 r p m at 0 ° C for 10 m i n Transfer supernatant to new tube 16. A d d N A C L ( 5 M ) o f a f ina l concentrat ion o f 0.2 M . M i x w e l l . Q u i c k spin Precipi tate w i t h 2 vo lumes c o l d ethanol . M i x gent ly Store i n freezer for 30 mins or overnight . 17. S p i n d o w n at 10 OOOrpm for 10 m i n 18. R e m o v e supernatant. W a s h w i t h c o l d 7 0 % ethanol and air dry . 19. R e d i s o l v e i n 500 ^1 H 2 0 158 A p p e n d i x V T w e l v e western redcedar D N A sequences for w h i c h p r imer pairs where des igned and that ampl i f i ed scorable and var iable microsate l l i te l o c i . T h e pos i t ion o f each p r imer is under l ined. A l l the sequences have been deposi ted i n Genbank . N , H , W , M or V , base not k n o w n . Locus Tpl Genbank accession no. AF245205 C C C A A G A G T T A G T A T A T C T C T T T G T C A T G T G Y B N S A T T T C C C T T G G T T T T T C C T T G C T T G G G G C T T C T C A T A G T G G A A A T A T C T C T T G T G G G T G C A T G T A T G C T C C A T T A A A T C C A T T T A T C C C A T T A A G G C A T T T G C A A T C A A T T T A T T A C G C G G G A A T A C A C T C T T C T C C T T T T T T C A C A T T T T G C G C G C A C G T A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A A T G A T T T A A T A C A T T T T C T T C T G T G G A T G G A T A A G G A G A C A A T G C T T T C G T T G G A C T T A T C T T T A C A T T G G C A T T C C T T G T G C A T C C T T A A T C A A A T T C C T T A A C A A A G A C A A T G C T T T C G T T A A A G A C C G A T C C T C C T T T T C T A G A T A C C T T T G G A A A C T A T G T T T T C C T T A G G A C C A G T C G C T T T T T T T G G A T C C C T T A A A T A A G A C A A T G C T T C C C T T A A G G A C T A Locus Tp2 Genbank accession no. AF245206 G G T C T T C T A A G T G C C T T A W A G T G T A C T W G G A C A T G T A C T A W G A Y A T G T A A T A T T T T A A C A T C T A G C T T T G T A C A T G T H A A T A A C A T A A A A A A A T A C A T A A C T Y T H A A A C A T A T A C A T G T W C A T G A T T G T G T G T G T G T G T G T G T G T G T G T G T G T G T K T G T G T G T G T G T G T G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A G A N A T T C C T A A T A A A T T T A C T A A C T T T A A A A G A A A A T A T T T T T A A T T T C T A A C A T T T C T A C T A N G G T T T T C T A A C A T T T T C T A A C A T A C T C A G T T A C A C T N T A T T A T A A C A T T T T G C T A C A T T C T A A T T A C A G T T T T C T A A T G T T T T C T A A T G T G T A C T A G G T T A T A A T T T T C T A Locus Tp3 Genbank accession no. AF245207 C C A A A T G C G T A T A G T T A G T T W A A A G A C C A A A T W A A N R A T A G G G A T C A G T C T C A A G A T C C A C T A A A A A A T A T T A T A A A T C T G T G T G T G T G T G T G T G T G D G T G T G T G T G T G T G T G T G T T G T T T T C G T A A N T C C A A A A N N A T N T T N T T N T T C A A A T C A A N A A C C C C A C C C T T T T N N C C T N C A N C C C T T A C T A A N A T A A T T G C A T T T C C A A G T T A T T T T T A T A G G A A A T A A A C C T A T G C A A M A G T A C A T T T A T T A A C T T A A C A A A A T T A T G A T A C A T T A A T T A A A C T A A T T T A C C T A A A A G A C A A A A T T A T C T G C G T C T T A C A A A A T C G T A T G A T C A A A G T C T T T T T T A A C T C A C T G T A A C A T A G G T A T T G A G G G A T G T T G G A A T G A G T T T C C A G G A C A C A G G A A A A A A A A 159 Locus Tp4 Genbank accession no. AF245208 C C C A T C T T G C C A C T T A T T G T A A C T T G T A T A A C C A C T A C A T C A T T G G A A A A A C A A C T G A T A A T T T T T T C C C A T G C T A G C A T T C C T A T C A A A T T A A C T A A C A T C T T G G C T C A A T A C A T A T T T T A A A T A T C G C A T T C T T A T G T G A G T T G A T T A T C G C T C A C T G A T T A T A T T G G A G T G G A G T T G T G T G T G T G T G T G T G T G T G T G T G T G T G T G T G T G T A T T A G T G T T G T T A A T A T T A T A T T G G A G T G A C C C T G A C A A A T C T C C A G G A T C T G A T G G C T T C C A A G C C T T C T T T T T A C A A A A A T G T T G G G A T N N N C A T T G G G G C T G A A T T A T G T G C A A G C A A T G C G Locus Tp5 Genbank accession no. AF245209 C C A G A A A A T A T T G A T G T T T A T A G C T T C G A T C A G A T R A A A T G G T C A T A A T A A T A T G A G A A A A T A G A T A A A T A A T A T G T G T G T G T G T G T G T G T G T G T G T G T G T G T G A A T A A T G T C A T T A C T A T T A T A T A T A T T N A T A T A C A T G T T T A A C A T T A T C C C T T T A T A A T A T T A A A C T V Y A T A A A C A T C A M T G A A T T G A T A A T A T Y A A A T A A T A T T A T T C T A C A T T A A T A T A T A T A G T A A T A A T G A T A T T A T T A T A T A T A T A T C A A T A C A C A T C T A C A A T G T A T A A T G T A T A T T A T A T T T G A T T A A A T A A T A T T T A A T A A T A C T A T G C G A T A C C G A T T A G T A T C G G T C A A T A C A T T G A T A T T G A T T T A G T T A A A C A G Locus Tp6 Genbank accession no. AF245210 A C T T A A T A W A T G T T G T A A C A T T A T G A C A A C C T A C T A A G A A C A T C C Y A A T C A T A A A A A T T G A A C C A A C A C A C A A A A C C G C C C A T G T A A A A G G A G G A G G A A A A C Y A G G A G A T C A C A A G C A A T G A A T G A A T A A G C A A G A A C A T A T A C G C G C G C G C G C G T G T G T G T G T G T G T G T G T G T G T G T G T G A T A T G T A T A T G T A T A T G T A T A T G T A T A T G T A T A T G T A T A T G T A T A T G T G T G T G T A T A T A T A C A T A T A T A T A C A C A T A T A T A C A C A C A T A C A C A C A C A C G T G T G T G T G T G T G T G T G T G T G T G T G T G T G T G T G T G T G T A C A C A T A C G T G T G T A T A G C C C T A G T C C G Locus Tp7 Genbank accession no. AF245211 G T C A T A T A C A G R G T T G C C A A T T T G G G T A A M A T C A A A C C T T A T C A A C C T C C C A T G G T G T G G G G G T G G T C C A A C A C A T G A C T A G G A C C T A A C T M A G G A C T T T T G A M A T A T A A C T T C T A A C T A A A C A C A C A A A A C A A C A A C T A A A A A G G T A T A A C C A T G G C G T C A A G A A C C T A G G K T T A T A A C T T T A C C T T A G G T C C T T C G G C T T A C A T C A A A G A C C C C A A C A C A C A A A C A C A C A C A A G G G T G A T C T A A T G G A T G C A C C T T A T T A A A A A C T C A A M A G R C A G C A A A A T T A G G R G T T T A G G R G T T T T A G G T G G K G T C C T G G C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A T A T A G A T C C C G C C T N G G A C C C C T T N G G A C A C A T T A G G G T N T C T T A A G A C C C C T T A G A A C A T A C T A A T C A N C G T N G G A T A C A C T A A T G G G T N T T A A T G G G T T C T A A G T G G T C C T A G G G A C C C T C A A G A C C C A T T A C G A C C C T T A A G A C C T A T T A G G A C A C A T T A G G A C 160 Locus Tp8 Genbank accession no. AF245212 A T A T T A G G T T A A T A G C T A G T T G A W A T T T T T A C G T H A A G T T T A C T A A T C T T G C A C A C A C A C A C A C A C A C A C G C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A T A T A T A T G T T C T N A C A A A T T T T T G A T A T T A G G T T G A T A G C T A G T T G A A G T T T T C A C G T C A A G T T G A C T A A C C C T T A T T T N N T A T T T T T T A T T G C C A C T T A N N N G C A T A G T T T G G Locus Tp9 Genbank accession no. AF245213 C C A G G D C A G G T T A C A T A G T T T C T A G A A A C C C A A T A C A G A C A C C A G T A A A C A A T A A T T T C C A G A T C T T C T A T A T T T Y A A G A T A A G T G T T T C A T T T C A T C C T A A T C T T T C A G A T C T A T T G T T T T A C T T T A T A A T T A A G A A A T T A G A C C C C C T T C C C C T C T T T A T C C A T A A C C A A T T G A T T C T A T T T T T M A C T G A T C A T C T C C T T T T G G T T A A T C A A T T V A V N T C T C C T T G T C T T G G A T T T G G A G A T A A T A T A T A T A T A C A A C A T T T T C C A T A C A T A G A A A A C A A A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A T C T T A T A G A G T T T A A T A A C A T A A A A C T A A G T A T A T A A A T A A T G T A A T C C A A A G G A G T A A T N C G T A A G A T T G T A A A G T A T G A T A T G T A C T C C A T T G A A A A A A A A A T T C T A T G T T T T C T A G T G A T A A T G A G A C T A C T J T C C G T A A N A A N N C C N N N N A A A A A G T T T G A A C A A C A T G A C C A C T A C A A G G Locus TplO Genbank accession no. AF245214 G G T C T T G T T T A T A G T T G T G T C C A T T C A G G C A T A A A C A T A T A T G T G T G T G T G C G T G T G T G T G T G T G T G T G T G T G T G T A A T G A A A A G A T A T A G T T G A G T G T C G A A A T A A A T A A A G A G A A G A G A T A T A G A A G A G C A G T T T A A A G A T G T T A G T A T G A G A T A G A G C C C T A A A A G A A G A J A A G A G C A T A A G T G C A T A A T A A A A A C T G A T C A G A C A T T T G T G T C A T A T A T A G T A G T T G A C T T T T G G A T A T A G T G C A T A T C T A T T C T C A G G A G G T T A T A G T C T T C T T T C T T A T T G A G C A G T G A G C T C T T G G A C A G T G A G T C C T C A C C C C T A A C A A S G C T G T T T G T A A A A G A C T C C T A A T A G G G T C A G G Locus Tpll Genbank accession no. AF245215 G A A T T C G G A C T A C C T G A T C C G C T T T G A T G G G T T G A A T A G C A A C T A T G T A G C T T G C T A G A A C T T C C A A T G C C A T A T T C T T T T G C T A C A C T T T T N C C T T G C A T G A N T N A T A G T A A G G N A T C T C C T T A C T A G G C T C T C T C T C T C T C T C T C T C T C T C T C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A C A T T T A T C T A T C T N T C T C G A G T G A T G C C T C T T A T C T T G G T T T T G G A T T T G A T G A A G A C T A G T C C G A A T T C 161 Locus Tpl2 Genbank accession no. AF245216 T A T G T A G C T A G T T T G A C C G A A G T T T G A C C C G C C A T A T C T C A G A C G T A T G G A C T C C T C T T T C A C C C G T C C A A G C T G C A T T G A A T C C A T T T T C T C A A G A A T T T T A A C T T G G T G T T A A C T T A T G G A G T A T T G G A G A T A C T T C A A T T T T A A G T A A A A A T A T C C C C C G G A T C A T T A A G G G C T C T A T C T C A T T C T G A T A T C T T A T A C T G C T C T A C N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N T G T G T G T G T G T G T G T G T T G T G T G T G T G T G T G T G T G T G T G T G T G T G T G T G T T T A T G C C T G A A T A G A T G C G A C N A A A A A C A A G A C C C15 vo 00 CN oo C N C N C N ro co O m C N N O C O O N C N en oo Tt O N N O C O N O O N C N O N 00 C N r--00 CQ © d d d d d d d d d d C14 R 3 m 00 C N Tt 00 C N O N C N N O C N O N -t N O O N C N C N r-C N O N OO CQ © d d d d d d d d d d ej O in C N o n o un © un O N © ro © O N r-» © © Tt r-» © d d d d d d d d CN NO o oo NO o Tt r-o in O N © NO O N © © CN 00 © OO oo © d d d d d d d d r--o O N o ro r-o CN O O N 00 © in O N © s ro d d d d d d d d ro o in R ro ro CN ro O N ro o ro o ro d d d d d d d d © Tt © ro t^ © in NO © NO © t--wo © o