Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Population variation in North American Menziesia (Ericaceae) Wells, Thomas Cameron 1992

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

Item Metadata

Download

Media
831-ubc_1992_spring_wells_thomas_cameron.pdf [ 3.95MB ]
Metadata
JSON: 831-1.0098874.json
JSON-LD: 831-1.0098874-ld.json
RDF/XML (Pretty): 831-1.0098874-rdf.xml
RDF/JSON: 831-1.0098874-rdf.json
Turtle: 831-1.0098874-turtle.txt
N-Triples: 831-1.0098874-rdf-ntriples.txt
Original Record: 831-1.0098874-source.json
Full Text
831-1.0098874-fulltext.txt
Citation
831-1.0098874.ris

Full Text

POPULATION VARIATION IN NORTH AMERICAN MENZIESIA (ERICACEAE)byTHOMAS CAMERON WELLSB.Sc., The University of Guelph, 1983M.Sc., The University of Western Ontario, 1986A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIES(Department of Botany)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAFebruary 1992© Thomas Cameron Wells, 1992In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.____________________________Department of BotanyThe University of British ColumbiaVancouver, CanadaDate April 29, 1992DE-6 (2/88)AbstractMenziesip, a widespread genus of shrubs, occurs intemperate montane regions of North Zmerica (2 species) andJapan (8 species). Although this disjunct distribution isshared by numerous vascular plant genera, few have beenexamined biosystematically within and among the floristicregions. In western North rnerica, N. ferrupinea is a highlyvariable species and apparently is related closely toAppalachian N. pilosa, based on flavonoid analyses. In thisstudy, univariate and multivariate analyses of 22morphological descriptors revealed discontinuous and clinalvariation along west-east and north-south gradients in N.ferrupinea, allowing recognition of two phases. Using thesame procedures, N. pilosa proved to be comparativelyuniform. Variation was partitioned largely within rather thanamong populations, with morphological patterns spatiallycorrelated on a regional scale. Past and present migrationalevents and, to a lesser degree, ecological variables havecontributed to clinal development. Analyses also were madeusing starch gel electrophoresis with 13 enzyme systems codingfor 19 loci. Isozyme variation resided predominantly withinrather than among populations, with lower than expected totalgenetic diversity. Levels of allozyme variation withinpopulations were comparable to other xenogamous, entomophilousspecies. Populations situated near distributional extremesiiH H-H-ItCtC)CDH-HD)t-(tQ.CD)IIC)hCDIIQfrHP)U)H-H-0HItC)U)CDCl)‘1)crHhCDH-N0)C)0ii CDU)kDDCDIjCiItIN))C)C)P)CDIH-HM-rr‘tjH-)CrU)U)H-0)jH-H-CDIP)U)CrU)CD-CDH-CD‘- 0‘-‘t•c.QU)CDCU1hItU)U)CUHCDCD0.oC)C0Q4)jIIH-CUU)CtCUCDItCt0HHU)CU0C)ItCDHCUCDU)00H‘frCl)U)CrCDCDItU)U)HCrIt<0JH-0CDHCDC)00H-CDtCU)CItCDH<0CUU)CU-1c5‘-<çtCrCUHCDH-CDI0-.CDCr0-.CUH-CrH-:iC)0C)0-.-•CDCUC)Z(•1CtZCrCtCDU)0CUCT)0-.C)IIH-çtU)0-.0..CtQ.I-H-CDU)CUU)CUU)•CrCrH-CD‘-0ZCDC)0H-U)çti-C)CrCD-CrCUU)t5G)(0çTiCDCDçtiCDCDCDçtçtCrH-H-0CDC)C)C)H-H-HCU0..0-.CDI-CDCDU)CDIi CtCrI-H-H-H-CDCrCrU)CU‘<<0-.U)H-t5I—-.HCDCDItI.<Cr<ItCD00-.CDHI—H-Cl)U)CrU):iCDCr00-.H-CrH(0CDrDH-U)‘-0U)WCDt(JJU)CDU)CD0-.ItCUCDCUC)0-.oH-CDI-CDCftU)‘1‘-0CDoH-t\)I—iHU)iCUoCr(0<1CDH-CDt-0-.C)‘1CDCU<ItcrHCD0H:—C)‘<H-CrI-(0H CDP’HHCto‘<H•00U) CU CU Crh‘-0CD0CUH-CDCDC)IU)çtCUQçtJI-CDcoHCDH-H-,H-II(0CDFCr:iItH-0H-CD0(0EoiU)ItJ—0c-rC)HC)H-oH-l))iSM-,H-Crl-t3CDH-U)IICtU)0CDCDCUCDot.U)U)CDCr0-.H-H-H-CD ‘CD(Q(Qr-h(0U) ItCUHCUftCD<U)00C)H-MItU)ftH-c-i-F—CDc-tCUCUC)HHCrCDH-0-.CUHI-’-I—hH-(-t<0C)H-U)H-CDC)çt0H-CU0-.H-ZHCrU)CUC)ctU)‘<CrCDCUCDCDC)H-0..1-i QIHCD0-.U)U)H-U)-rçrU)HC)•ItIt00CD0iU)IjHi-cU)CDHU)HCDU)CUH-0M0HH-NCUCD0<c-i-CDCUHCUC)CDU)0CDCrU)—0-.CUU)CDH-IttTIt<ItCDCDCUCri-cIiU)CDCDH-CDCrCtCD x H H Cr CD 0-. Cr CD H CD CU U) Ci (0 CD CD Cr H-C) CU H CU H- 0 Ii CD U) It H-Cr CD H (0 H CD CD H U) 0 I-hTable of ContentsPageAbstract.iiList of Tables viList of Figures ixAcknowledgements xiiChapter 1 - Introduction 11.1 Circumscription of Menziesia 11.2 Taxonomic Treatments of Menziesia 21.3 Phytogeographic Considerations 71.4 Objectives of the Thesis 10Chapter 2 - Patterns of Morphological Variationin Menziesia 142.1 Introduction 142.2 Materials and Methods 142.2.1 Herbarium Specimens and SpeciesDistribution Mapping 142.2.2 Selection of OTUs for Study 152.2.3 Selection of Descriptors 162.2.3.1 Leaf Characters 172.2.3.2 Fruit Characters 182.2.4 Univariate Analysis of Descriptors 182.2.5 Multivariate Analysis of Group Structure. .202.2.6 Tests for Association betweenMorphological Variation and Geography 212.2.7 Comparison of Morphological Variationwith Ecological Parameters 232.2.8 Analysis of Japanese Taxa and theCladistic Analysis of Menziesia 252.3 Results 272.3.1 Univariate Analyses of CharacterVariation 272.3.2 Cluster Analysis 282.3.3 Principal Components Analyses 302.3.3.1 Within-group PCA of Menziesiaferruainea, sensu lato 302.3.3.2 Within-group PCA of Menziesiapilosa 312.3.4 Discriminant Analysis 322.3.5 Association between MorphologicalVariation and Geography 342.3.6 Ecology and Patterns of MorphologicalVariation in the Western North AmericanMenziesia Sites 342.3.7 Ecology and Patterns of MorphologicalVariation in the Appalachian MenziesiaSites 37ivPage2.3.8 Comparisons between North American andJapanese Menziesia 392.4 Discussion 422.4.1 Taxonomic Considerations in NorthAmerican Menziesia 422.4.2 Key, Descriptions, and Nomenclatureof North American Menziesia 462.4.3 Factors Influencing MorphologicalVariation in North American Menziesiawith Reference to the Japanese Species 51Chapter3.13.23 - Isozyme Analyses of North American Menziesia. . .102Introduction 102Materials and Methods 1023.2.1 Selection of Populations and Material 1023.2.2 Extraction and Electrophoresis ofEnzymes 1043.2.3 Analysis of Genetic Variation 1073.3 Results 1093.3.1 Interpretation of Isozyme BandingPatterns 1093.3.2 Genetic Variation Within Populations 1113.3.3 Genetic Variation Among Populations 1143.3.4 Estimates of Gene Flow in Menziesia 1163.4 Discussion 1173.4.1 Genetic Variation Within Populations 1173.4.2 Genetic Variation Among Populations 1213.4.3 Factors Influencing Isozyme Variationin North American Menziesia 128Chapter4.14.24 - Concluding Remarks 152Objectives Revisited 152Further Avenues of Study 157References 161Appendix 1 - Summary of Voucher Specimens Examined 174Appendix 2 - Summary of Ecological Parameters for theNorth American Menziesia Field Sites 187AppendixA3.1A3 .2A3 .3A3 .43 - Growing Menziesia Seedlings Using SterileCulture 195Surface Sterilization and Preparation of Seeds... .195Preparation of Growth Media 196Growth Conditions 196Discussion 198vList of TablesTable Description Page2.1 Summary of Menziesia herbarium specimens 60used in mapping species distributions.2.2 OTUs utilized in morphometric and ecological 64analyses of Menziesia.2.3 Description of the twenty-two morphological 68characters used in the morphometric analysesof Menziesia.2.4 Ecological descriptors used in the canonical 70correlation analyses of North AmericanMenziesia.2.5 Characters used in the cladistic analysis of 71Menziesia.2.6 Character states of taxa used in the clad- 73istic analysis of Menziesia.2.7 Principal components analysis of North 85American Menziesia using 22 morphologicaldescriptors: loadings, eigenvalues, andvariance per component.2.8 Principal components analysis of . 87ferrupinea, sens. lat. using 12 morphological descriptors: loadings, eigenvalues,and variance per component.2.9 Principal components analysis of Appalachian 87. oilosa using 10 morphological descriptors:loadings, eigenvalues, and variance percomponent2.10 Canonical correlation analysis of the 93relationship between morphological andecological descriptors of OTUs from thewestern North American Menziesia field sites:loadings, intraset and interset communalities.2.11 Canonical correlation analysis of the 97relationship between morphological andecological descriptors of OTUs from theeastern North American Menziesia field sites:loadings, intraset and interset communalities.viList of TablesTable Description Page2.12 Principal components analysis of Japanese 100Menziesia using 20 morphological descriptors:loadings, eigenvalues, and variance percomponent.3.1 Collection sites and sample sizes of 34 134populations of North American Menziesiasampled for isozyme analyses.3.2 Electrode and gel buffers used to resolve 13516 enzyme systems in North American Menziesia.3.3 Summary of intrapopulational genetic 139variation in 34 populations of North AmericanMenziesia: mean number of alleles per locus,proportion of polymorphic loci, observed andexpected heterozygosities, mean fixationindices.3.4 Fixation indices for all polymorphic loci in 141populations of North American Menziesia.3.5 Allele frequencies summarized by geographic 143region for North American Menziesia.3.6 Nei’s genetic diversity statistics for North 145American Menziesia taxa for each polymorphiclocus and pooled over all loci: HT, H5, DST,GST.3.7 Mean genetic identities and mean genetic 147distances among the three North Americantaxa of Menziesia.3.8 Nei’s genetic identities and genetic 147distances depicting the relationship betweenthe three North American taxa of Menziesia.3.9 Gene flow (Nm) estimates from allele frequen- 150cies for the three taxa of North AmericanMenziesia. Estimates are calculated by threedifferent methods: Wright, Crow, and Slatkin.viiList of TablesTable Description Page3.10 Frequency of private alleles used to 150calculate gene flow among populations ofNorth American Menziesia using Slatkin’smethod.3.11 Levels of intrapopulational genetic variation 151in North American Menziesia compared toaverage values summarized from other studies:mean number of alleles per locus; percentageof polymorphic loci; mean expectedheterozygosity.Al.1 Summary of lending institutions from which 174Menziesia specimens were obtained.A1.2 Summary of voucher specimens examined in 175the study of North American Menziesia.Al.3 Summary of Japanese Menziesia voucher 184specimens.A2.l Species association matrix for western North 187American Menziesia sites.A2.2 Species association matrix for eastern North 189American Menziesia sites.A2.3 Physical features of North American Menziesia 191field sites.A2.4 Summary of soils found at the North American 192Menziesia field sites.A2.5 Summary of climatic factors at the North 194American Menziesia field sites.viiiList of FiguresFigure Description Page2.1 Distribution of Menziesia ferrupinea, sens. 61lat., in western North America.2.2 Distribution of Menziesia pilosa in eastern 62North America.2.3 Distribution of the common taxa of Japanese 63Menziesia based on a representative sampleof herbarium specimens.2.4 Sampling regions and collection sites of the 66western North American Menziesia specimensexamined.2.5 Sampling regions and collection sites of 67Menziesia specimens examined in easternNorth America.2.6 Box plots summarizing variation in 22 74morphological descriptors of North AmericanMenziesia.2.7 Contour plots of descriptors exhibiting 80clinal variation in j. ferruainea, sens.lat., in western North America, superimposedon sampling localities.2.8 Contour plots of descriptors exhibiting 82clinal variation in . pilosa in easternNorth America, superimposed on the samplinglocalities.2.9 Summary dendrogram of a UPGMA cluster 83analysis of North American Menziesia OTUsbased on morphological data.2.10 Distribution of 238 OTUs of North American 84Menziesia within the first two axes of aprincipal components analysis based on 22morphological descriptors.2.11 Disposition of 150 OTUs of . ferruainea, 86sens. lat., within the first two axes of aprincipal components analysis using 12morphological descriptors.ixList of FiguresFigure Description Page2.12 Disposition of 88 OTUs of M. pilosa within 88the first two axes of a principal componentsanalysis using 10 morphological descriptors.2.13 Distribution of 238 OTUs of North American 89Menziesia within the space of the firsttwo canonical variates of a discriminantanalysis based on 14 morphological descriptors.2.14 Gabriel plot defining intersite relatedness 90among the collection localities of .ferrupinea, sens. lat., OTUs in westernNorth America.2.15 Gabriel plot defining intersite relatedness 91among the collection localities of . pilosaOTUs in eastern North America.2.16 Dendrogram of western North American 92Menziesia field sites based on a UPGMAcluster analysis of community types asdefined by a Jaccard similarity matrix ofspecies association.2.17 Distribution of 101 OTUs of western North 95American M. ferrupinea, sens. lat., withinthe morphological and ecological domains ofa canonical correlation analysis.2.18 Dendrogram of eastern North American 96Menziesia field sites based on a UPGMAcluster analysis of community types asdefined by a Jaccard similarity matrix ofspecies association.2.19 Distribution of 61 OTUs of Appalachian . 98pilosa within the morphological and ecologicaldomains of a canonical correlation analysis.2.20 Distribution of 30 OTUs, representing the 99common species of Japanese Menziesia, withinthe first two axes of a principal componentsanalysis based on 20 morphological descriptors.xList of FiguresFigure Description Page2.21 Strict consensus tree obtained from 6 101maximally parsimonious cladograms of Japaneseand North American taxa of Menziesia.2.22 One of the 6 cladograms of Menziesia taxa, 101similar to the strict consensus solution,illustrating the changes in character statesneeded to obtain the tree.3.1 Representative illustrations of isozyme 136banding patterns observed in an electrophoretic study of North American Menziesia.3.2 Dendrogram of North American Menziesia popu- 148lations, grouped by geographic regions, basedon a UPGMA cluster analysis of Nei’s geneticidentities.3.3 Dendrogram of populations of North American 149Menziesia based on a UPGMA cluster analysisof Nei’s genetic identities.xiAcknowledgementsI would like to express my gratitude to my supervisor,Dr. Bruce Bohm, for giving me the opportunity and support topursue my interests in studying variation in woody plants. Iwould also like to thank the members of my advisory committee:Drs. Gary Bradfield, Fred Ganders, Jack Maze, and WilfSchofield for helpful input at various stages of the project.Post-graduate scholarships from the Natural Sciences andEngineering Research Council of Canada and the University ofBritish Columbia, as well as teaching assistantships in theDepartment of Botany are gratefully acknowledged.Drs. Fred Ganders and Carl Douglas provided lab space forthe electrophoretic and sterile culture work, respectively.Drs. Doug and Pam Soltis at Washington State University,Pullman also generously supplied lab space and helpful advicewhile Dr. Steve Novak deserves special thanks for teaching methe art of starch gel electrophoresis. A productive conversation with Dr. Randy Bayer at the University of Alberta,Edmonton led me to develop a protocol for the collection andfreezing of Menziesia field samples needed for the isozymework. Discussions with fellow graduate students, especiallyStewart Schultz and Brian Compton, and of ficemates Drs. BillCrins and Kadi Hauffe helped me to look at science andMenziesia in different and interesting ways.And of course, I cannot adequately express my appreciation to my wife, Muriel, for her love and encouragement; andto both of my families for their guidance and support.xii1Chapter 1Introduction1.1 Circumscription of MenziesiaMenziesia Smith is a genus of erect or spreadingdeciduous shrubs that belongs to tribe Rhodoreae D. Don(subfamily Rhododendroideae Endi.) of the Ericaceae (HeathFamily) (Stevens 1971). Its general appearance providesMenziesia with the local common names of “false azalea” or“fool’s huckleberry”, as it can be mistaken in some seasonsfor these other ericaceous shrubs. Nevertheless, it is adistinct and perhaps relatively primitive member of theRhodoreae according to a recent cladistic analysis of thetribe (Kron and Judd 1990). The genus is disjunctivelydistributed, with eight species in Japan (Hatta and Tashiro1986) and two species in North America (Gleason 1952;Hitchcock et al. 1959)Vegetatively, Menziesia is distinguished by itselliptical to ovate to obovate deciduous leaves with distinctsubulate or strigose hairs on the lower midrib. Theinflorescences are arranged in umbels or racemes withtetramerous or pentamerous urceolate to tubular-campanulateflowers that have superior ovaries and anthers dehiscing byshort slits. The fruits are dry septicidal capsules thatproduce numerous small seeds.Menziesia grows in a variety of habitats includingcoastal or mountainous temperate mesophytic forests, subalpine2forests, and heath balds (Braun 1950; Daubenmire 1978; Hattaand Tashiro 1986). In both Japan and North America, Menziesiatends to occur in areas of high rainfall or where persistentfog or mist ensures a surplus water supply over the growingseason. Occasionally, Menziesia is a major component of theshrubby understory community whereby it is an importantcontributor to the stabilization of slopes. Although it issometimes grown horticulturally as an interesting ornamentalshrub (Rehder 1940; Kruckeberg 1982), it is not generallyconsidered to be economically important. However, it is ofnote ethnobotanically because a fungus, Exobasidium sp. af fin.vaccinii (Fuck.) Woron., that attacks its leaves andinflorescences is eaten by various groups of coastal FirstNations people in British Columbia and Alaska (Compton 1991).1.2 Taxonomic Treatments of MenziesiaThe genus Menziesia was first described by J.E. Smith(1791) from material collected along the coast of westernNorth America. Today, two species are generally recognized inNorth America: Menziesia pilosa (Michx.) Juss. from theAppalachians of southern Pennsylvania to northern Georgia(Gleason 1952) and Menziesia ferruainea Smith from montanewestern North America (Hitchcock et al. 1959). The two NorthAmerican species are considered to be similar to one anotherbased on general observations (Radford et al. 1968; Bohm etal. 1984).3In western North America,. ferrupinea is highlyvariable both within and among populations. The recognitionof a less glandular and generally less pubescent phase,distinct from coastal j. ferruainea, led Asa Gray (1878) todescribe a new species, Menziesia alabella Gray, from theRocky Mountains. Peck (1941), in consideration ofintermediate specimens collected from the Cascades, reducedGray’s taxon to varietal rank, ferruainea var. alabella(Gray) Peck. According to this alignment, . ferruainea var.alabella was described in the Vascular Plants of the PacificNorthwest as having a range encompassing the Rockies fromBritish Columbia and Alberta south to Montana, Idaho, Wyoming,eastern Washington and Oregon and down the Columbia Rivervalley to Mt. Adams and Mt. Hood of the southern Cascades.Specimens from coastal Alaska and British Columbia south tonorthern California and inland to the northern Cascades werereferred to as . ferruainea var. ferruainea (Hitchcock et al.1959; Hitchcock and Cronquist 1973). In preparation for theFlora of the Queen Charlotte Islands these combinations werechanged to subspecific rank (Calder and Taylor 1956; 1968).In an attempt to clarify the taxonomy of western NorthAmerican Menziesia, Hickman and Johnson (1969) conducted amorphometric study and documented complex patterns of clinalvariation, expressed primarily as differences in severalpubescence characters and in leaf tip shape. The patternsobserved, however, could not be partitioned into discretetaxonomic categories. Rather, Hickman and Johnson4hypothesized that the dines observed reflected migrationalevents as influenced by glacial history, past and presentpatterns of gene flow, and the adaptation of populations totheir existing environments. Nevertheless, Martin (1973), inan unpublished thesis comparing two widely separatedpopulations from Mt. Seymour in the Coast Range of BritishColumbia and from Waterton Lakes National Park in southernAlberta, found significant differences between the populationsover several morphological features (pubescence, leaf tipshape, and stomatal density). Based on these data, Martinagain raised the possibility that taxonomic recognition waswarranted for the Rocky Mountain phase.A study that compared eastern and western North AmericanMenziesia utilized foliar flavonoids (Bohm et al. 1984)Menziesia ferrupinea was found to accumulate derivatives ofkaempferol, 7-O-methylkaempferol, quercetin, 7-0-methylquercetin, myricetin and gossypetin that occurredvariously as a complex mixture of mono-, di-, andtriglycosides, some of which were acylated derivatives. Incomparison, Appalachian N. pilosa had a simpler flavonoidprofile lacking 7-0-methylated flavonols, triglycosides, andgossypetin. In addition the two species were furtherdistinguished by the presence of an unidentified flavanone inN. ferrupinea and dihydromyricetin in N. pilosa. Flavonoidprofiles in N. ferrupinea were highly variable but notsufficiently distinct from the coast to the Rockies to allowrecognition of infraspecific taxa, thereby supporting the5conclusion of Hickman and Johnson (1969) that a single speciesexists in western North America. Based on extensive reviewsof the flavonoid literature (Bohin et al. 1984; Bohrn 1987), itappears that North American Menziesia exhibits one of thehighest levels of intrapopulational flavonoid variation everobserved in vascular plants and closely resembles thesituation described in Phlox oilosa L. (Levy 1983).Nevertheless, it is in Japan that Menziesia reaches itsgreatest diversity. The standard treatment of the genus isthat of Ohwi (1965) who, in his Flora of Japan, recognizedfour species and numerous taxa of intraspecific rank.Recently, two new species were described based onmorphological and ecological considerations (Tashiro and Hatta1986). Although a stable taxonomic treatment does not appearto exist, detailed work by Tashiro and Hatta supports therecognition of eight Japanese species of Menziesia (Hatta andTashiro 1986). The species are readily divisible into twogroups according to the number of floral parts. The firstgroup, which is pentamerous, is endemic to Japan and includes. ciliicalvx Maxim., . lasioohvlla Nakai, N. katsumatpe M.Tashiro et Hatta, M. rnultiflora Maxim., and N. oentandraMaxim. A second, tetramerous, group consists of N.povozanensis M. Kikuchi, M. ourourea Maxim., and N.vakushimensis M. Tashiro et Hatta. Menziesia oentandra, inparticular, is common and widely distributed, occurring fromthe southern island of Kyushu north to Sakhalin Island and thesouthern Kuriles (Ohwi 1965). In contrast, the tetramerous6species have more restricted distributions relative to theirpentamerous counterparts. They are morphologically distinctfrom North American Menziesia which are also tetramerous. Amajor difference between Tashiro and Hatta’s work, and that ofOhwi, is the inclusion of . multiflora and . lasioohvllawithin . ciliicalvx, sens. lat., by Ohwi (1965).Biologists generally acknowledge that variation is theraw material for evolution. Nevertheless, surprisingly fewwidely distributed plants have been studied throughout theirgeographic range with a view to examining comprehensively thedegree of variation present and the factors accounting for it.Some exemplary studies that employ a variety of investigativetools are represented in work done on Phlox L. (Levin 1966;Levin and Levy 1971; Levy 1983), the Chenooodium fremontiiWats. complex (Crawford 1976; Crawford and Wilson 1977;Crawford and Mabry 1978; LaDuke and Crawford 1979), and theClavtonia virainica L. complex (Doyle 1983, 1984 a, b; Doyleet al. 1984; Doyle and Doyle 1988). In each case,morphological, cytological and biochemical races werediscovered.It is not clear whether western North American Menziesiaform one or two distinct biological entities. A thoroughanalysis of the complex patterns within . ferruainea isrequired to determine if observed variation justifiespartitioning this taxon into discrete races or groups.Furthermore, despite their similarity, no genetic analysesexist to assess the relationship between the disjunctMi))C)HiC))0ljCrC))CrHrrZC:•F-CD0‘t50HMiCDI-COH-C‘.00CD•00C))C))CD0tIoCl)0tI0QCD0H-CDI-CDt-‘cj<<i-.t-çtQ.Cl)çtI-jCl)3CDCr(JiCrC))c-rC))CD0C))C))CDP.)CDH-0C))MiC))C))C))JC)Q.HCOI-I—’hCOH-COF-hCD0tJCDC))Q.0C))‘.0tICtC))CrC))0H-‘<0I-H-<H-COQ.CDC)o-JCO0HCDH-hII3COCOC1CDCDClCrCOHiZ1’)C‘C)ClC))<C))C))CDCD‘.—-0<CDIIH-0C))C))COCO<I-()I-iC(QCD-CoCDCDH-C)t3I-•CDI-C))COCrCrH-F-CD<CDH-‘tiCOCrocrI-<MiH-0—CDI-’HC)0CDC)0C))C)CDCOClCtMiMiCOt’iCX)C))COH-Q.Ct(C))C)CDCDC))C))C)H-HiH-HH-CrtICDUICDCO1-H-CO‘tiCDC)H‘.0C))CDCD‘.0C))tiH-ClCDCDC)ClCDI-H-0H-—1CD-.MiH-CrC))COCrI-jCDC)CDCrC))I-CtCDP.QC)CDClHCDCOCr3•H-C)COI-C))HC))H-Cl<c-r0‘-CrI-’-CDClCDCIt-CDCrH-H-HCOCD•ClH-H-CDCDH-CD1CDCI-j0i-CrU)CDC)COC))I-1)H-H-CC)iiH-CrCrhC))H0C)HClC))HCl)CO—CD1J0CC))0CD<CD<CrMiH00‘.0H-‘.0CrCr‘C0HCD0C))CtCDCOI-hU]CDH-H-0ç-tQ.C))‘C))cTZMiCD<CrCrwCrht’.)IC)NCH-WCDC))CrC))0HCrC))CDH-H-—0—CCO0I-’‘CDfrjH-HCiZI-0CDCOCOH-C))COCOCOC))0‘tIC))0-t-CDCrCDCC)MiI-QCOCDCrCDCt3‘tiI-JC)),QCDCD‘tiCOH-CoI-COMiH-CDC))H-COC))C))CtClCi-cCD0CtCrHCrCDC))H-H-COØCO0H-COCDH-iH-C))C))CDH-ri-ClCDH-CrCDCtCrCO1CD0MiC))H-IIF-’•H-tiC))COCDC))H-CD0H-COC)C))CrHC))H-0NCDCO<0C))H-‘.0H(H-H-0CDH-Q.C))ClH-CDHrr3H-H<CD-0C)a)0CD:i:i-cC)COH-C).I-CD0CrH‘.0CDCOtOMiI-i0COCr0H-C))C)I-hHC))i-cH---•MH-<3C)H-CDHiC))CCrCrrrC))H-CDtOtYCD<CrC))COCDH-NC))C))3CD0COC)CO—CDCOC)0CrCfCDC)H-0l-ClC))CDCt-CDtII-COC))CDClCOMiCDCDCr—0CDCC)CtH-IISCOClSH-C))MiC)C)CCOCDC))CrH-tiCDP.1C)CrCDCrCDCOHhHI-H-H0CrC)CDC))0COiCrCDCD<3CDC))CDC))COQCrH-CDCOC)CO‘tiCrCDCDH-COC)ClC)Hi-cCDi-cC)CDH-CrH-CDC)C)COCOCOCD0HCD0CDC))CDCDC)MiH-1iC))ZP.)CDC)CDCQ<1CI0CCO0C))<COI-0CDNC0CtC)C)CCl(QHI—htICDCri-cCDi-cCCOI-—Cti-H-H-tIC))MiCCr()CtCOCOCDCrCDc-rCDC)NCDH0Q.i-0CCDCDClI-CrCDCr‘.0rrIIH-CDCOC))H-H-IICrCr3ClCDCOCDCD0CDH-CO5Cr‘ijCrtiCDCDWNCO5CDC))CrC))CDC)COCrH-CDHI-HiCD-.H-H-CDCOCrC)CD0CrC))CD0CD0i-cCrCDNCrHH-CDCDi-‘—3C)HiH-C><H-0CD‘.05MiCDCr5CDH-CrH-CDC)ClC))CD-CDH-Hi5CDCl)C))C))C))MiJCOC))H-5COCrtO1C)H-HC))‘<COMiCDCO0CDH-H--.H-0COHCrC))CrCOC))CDC)Ht-Cr01’J0CDCDC))IIC))CDP.1C))C))i-hCDt--<‘.<ClCOCOCDCD8Similarly, little attention has been placed on disjunctionsbetween eastern and western North America, with a notableexception of Fernald’s comparison of the vascular borealfloras (Fernald 1925).The disjunction of the temperate floras of eastern Asiaand North America is thought to reflect largely changes inclimate and geography since the Tertiary, coupled withevolution within the floras themselves (Tiffney l985a). Fromthe Eocene to mid-Miocene (55-13 million years beforepresent), floras of eastern Asia and North America werecontinuously linked via the Bering and North Atlantic landbridges (Tiffney 1985 a,b) . It is likely that temperategenera including Menziesia became widely established ineastern Asia and North America at this time (Wolfe and Leopold1967; Wolfe 1969, 1972, 1981, 1985) . During the Oligocene toMiocene Epochs (38-5 my b.p.), the Rocky Mountains rose,creating an increasingly arid midcontinental region in NorthAmerica. A simultaneous cooling trend in the late Miocene (13my b.p.) led to the southward migration of the temperate florain North America (Wolfe and Leopold 1967; Wolfe 1987a,b). Asplit between west coast and eastern temperate florasresulted, which could account for the disjunct distribution ofMenziesia in North America. Later, during the Pliocene (5-2my b.p.), the Sierra-Cascade axis uplifted, drying the regionbetween the Cascades and Rockies thereby likely contributingto the separation of coastal and interior populations of i.ferrupinea.9Of equally dramatic proportions, however, were thePleistocene glaciation events (1.9-0.01 my b.p.), whererecurring advances of ice covered most of Canada and thenorthern parts of the contiguous United States (Prest 1984).The distribution of Menziesia in North America was undoubtedlyaltered during this period and, in the west, the genus mayhave been limited to refugia in the unglaciated regions ofcentral Alaska and below the glacial boundary of southernAlberta, Washington and Montana. The effects of glaciationmay have been less pronounced in the Appalachians where theice sheets advanced only as far as northern Pennsylvania(Mickelson et al. 1983). Temperate elements were thus able toretreat further south unimpeded in the Appalachians where theyremained at lower elevations during glacial maxima (Watts1983) . Menziesia tilosa is currently distributed south of themaximum glacial boundary. Although parts of Japan were alsoglaciated, there apparently was no period during which thetemperate flora was eliminated (Maekawa 1974). Despite north-south and altitudinal oscillations, a temperate flora wasprobably maintained over the greater part of Japan allowingcontinuous survival of similar floral elements that date fromthe Eocene to the present.Unfortunately, little biosystematic work has been done ongenera that exhibit North American and east Asiatic disjunctdistributions. The few cases to date represent a number ofscenarios. Some genera such as Bovkinia Nutt. (Gornall andBohm 1985), and Acer L. section Palmatum Ogata (Chang and10Giannasi 1991) exhibit close affinities between the Asian andNorth American species. In Tiarella L. (Soltis and Bohin 1984)and Elliotia Muhl. (=Cladotharnnus Bong.) (Bohxn et al. 1978),all the species are very distinct. In Aaastache Claytonsection Aaastache (Lint and Epling 1945; Vogelmann 1984;Vogelmann and Gastony 1987), Liauidambar L. (He and Santamour1983; Hoey and Parks 1991), and Liriodendron L. (He andSantamour 1983; Parks et al. 1983; Parks and Wendel 1990),alternative interpretations of the relationships between theAsian and North American taxa arise depending on whethermorphological or biochemical data are considered. Wood (1972)has stated that some genera, notably Gaultheria L., OsmorhizaRaf. and Stvrax L., have species occurring in eastern NorthAmerica that are more similar morphologically to their easternAsian counterparts than they are to their geographicallycloser western North American relatives. Such findings stressthe need for further examination of genera exibiting thispattern of disjunction.1.4 Objectives of the ThesisBecause of its biogeographical distribution andrelatively small number of species, Menziesia is a genus wellsuited for intensive studies that consider the evolution ofcontinental disjuncts and the partitioning of variation inshrubby plants. For logistical reasons this study focusedprimarily on the North American members of the genus.11The primary goal was to determine the degree ofintrapopulational and interpopulational variation present inAppalachian . oilosa and western N. ferruainea, sens. lat.This involved both detailed examination of morphologicalvariation (Chapter 2) and isozyme variation analyzed among 34populations of North American Menziesia (Chapter 3). Withthese approaches, several questions were addressed: 1. Arethere discrete patterns of variation in western North AmericanMenziesia that warrant taxonomic recognition? If so, whatcriteria apply and at what taxonomic level is recognitionvalid? 2. How similar are the western and eastern NorthAmerican species of Menziesia to one another morphologicallyand ecologically, and how much izozyme divergence has takenplace between these areas? Achievement of this objective willaid in understanding how Menziesia has evolved in NorthAmerica and the degree to which factors such as geographicisolation, glaciation and habitat preference have influencedthe shaping of past and present patterns of gene flow. ... Canclinal variation in morphology be correlated with currentenvironmental factors (Chapter 2)? . Can allele frequenciesbe used to estimate levels of gene flow among populations(Chapter 3)? This thesis then complements and expands on thecurrent body of knowledge of North American Menziesia ascircumscribed by the work of Hickman and Johnson (1969) andBohm et al. (1984)Despite the limited availability of Japanese material forthe current study, it was possible to compare the Japanese andH--r-rCl))2)f-HCl)-Cl)(QH-ZCf)CrS2)ZCl)Cl)Cf1CD-DCrCD‘10000CD05CT)0H0oC)CDCCD<0)0Mc-fCt)5I-iHNQ-I-CDCDQjH-0CiCrt-‘1‘t5CrCl)H-2)0Q.H.H-CDIHCl)CT)H-CDCDJ’5H)CDHXH-CDCDCl)52)‘‘1CDCr()Cl)0<CDCl)Cl)H-CDCl)02)CrH-0Cr2)CDHCtCD1•5(Q0$)H-CrH-C)H-<CD302)<C)HCD02)2)50H-02)Cl)Cr(QCD2)0C)M-,3Crcni-jH0C)0Hi-iiCrQ.CiCl)H-Cr2)CfQCD<(QH-MHCl)HH-I-’Cl)CD2)C)H-H-CfCl)HCl)CDC)0CD0<C)CDXCf2)CDC)f0‘t5‘<Cr0hCl)00-$)CrH-1$)Hrj2)c-iT)))0CDCDCDCDI-hCfCDCDCD3H-Cl)I-Q—J5Ct2)CrHCrHCDH-I-QC)oCf2)CD<CDC)Cl)H-HH<CrX2)k<Cl)H-I-h2)H-5-SZCDC)CL2)CDCDCD‘-QCrCiH-CfCDCD1HCr0)—.1CfICDH2)0)12)Cl)H-M-H-00)H-Cf1C)DH-CDNH-H-<CfCl)CT)C)><2)2)CLH-0-.0T))HH-Cl)CD5CLCrCr0CDQ2)H>LCl)0))CD02-)H-CrCDH-MI-jCl)Cl)0)CDCDCrCfrrI-NCl)Cr0CD2)Cl)HHC)2)CDc-rH--<CfCD:-(QH-CrCD1CDDH-2)—CD0Cl)H5CDCLCLH-T))5CDCL2)CDCfCr0c-fQ<XHH-‘0frCD-c-rCDI-l-jHQCD2)2)C)CDCD0))0c-f0)H-CDC)H-—CL2)i,C-i5H-NH-—CLS5LQ<.OC)Ci2)0C)2)0Cr2)HH-c-iH-H-I-iH-H-—H-CLH-CLI-hH-CfCfCl)CD3H-C)H-•C)CDCLH-2))Cl)<02)CDCL0C)C)CD2)CDCD3HH-1CDCr2)H-CDMI-CLHCfCl)0Cl)QCf<210CtCl)CD0I-Cl)CrCD-Cl)-3Crl-CDcniM-CDCl)CDCDCDCrCD2)0)Ci2)I-c0I-H-Cl)H-<CD2)iCl)I-TCDCDHCl)CDHCl)CL‘10C2)2)CLCDCL02)C)Cl)•C)0)WH-H-Cl)HC)2)H-Cl)Cl)H-CLH-H-CDCf-Q.CQ—0Cl)0Cl)0H<CDH-C)CDCrCDH-CDI-hLi.Cl)Cf2)Cf‘<CDC)CDH-0<Cl)Cl)‘tS-.otiCiCl)i<5H-3N0CLCLCD•2)-JCDH-CDciCl)CL00c-iCDCDCDIc-iCrtiDCDCfti0C)c-f<CDH-CLCrXti-0CD-.tiCl)H-titirCD2)‘10)Zci0CDH)CDCDC)Hc-fCl)50C)HCl)Cl)HC)O)0HCDtiCl)C)H-•2)M-c-i‘-<21Q.C)Crpic-iHCDCDI-CCl)H-0CDCfCLCDc-fI-CCDCLH--QCDCrH-Q2)IH-Cl)H-CfCDCrC)2)CLHC)Cf—CD2)CD<tJ<5C)Cl)H-CD0CD02)J0H0CD0CD5CDCDCDCl)5CfCtCDCf<CDMCDCDHIt-H-CfCUCDCDH-C)CDCDH-CDHCDC)H-3LCtC)2)I-jH-C)T))CDCL)t3CfCDtiCDH-CDH-H-C)MCD2)tiCL‘-Chht-ZCtCDCDCDCf2)03X<CDt”i(Q2)CDCrCl)H-C)CD0H2)0Z2)<2)C).—HCD‘12)CDCrCD5Cl)•00Cr2)H-13CDCD0CrCL13CDCDH-CDCUI—,,13HSI-QCDSci-CDNCDCL13ClCr2)CrCDtiCD0H-CDCLCDH-13CrI-CCD0Cl)CD2)5CDCL0Cl13H-H-H-Cr05CDti13H-CDti0Cl)CDCfNI-2)CUCl)2)0<H-H0CLCrCr0H13H-,13C)13CLCDJCrCDCDCl)HI-bc-’-Q.t:1<—HCO—COrt5CDICl)H-DiH‘-0CDHCDCDtoCDDitiI-tQCD0C)H-rtCD‘-0lCDCtP1H-H-J’IIH’çtCrCDI-cciDlCO-.H-•COCDH-Cr‘ZCrH•0—•H’CDH-CD<QCOCDtoCD0<i:DiCDCODiCDCT)I-0COCO5CtCOH’I—i5COI-h‘-<0CDDlCODiCO0CroCOCOS•CD05M-Cr(-r•00CD0•totjDlCDI-h—5C)-çH-5I-cI-CrCO0CtCrCDC1Ct0CODlH’CD‘tiH-I-h-0CDM-H--0CtDlDiHH’ciDiCDCl)‘COCOCl)MDlH-5H-ciCD0I-htoH’CDCl)HCt0c-ic-tCDCDCDCtDiDiI-CDCD0to5CttiDiDiCtCOC)JtotiH-DiH-CQ.H’<çtQH-IIi’Q‘CD‘<COCODlCDtoDi—Ct‘COH‘<I—’<toQoH-C)Q.•DiDi‘-00DiCDIII-H-C)0CtCriiH’H<C)CrCI)H-CtH’CO•CDDiCt0C)C)H0Dli—cI-H-ociH’H’HDiCDCDI-H-CODl‘-0CD‘-0COCOCODiC)toDiCr(DCDC)Ct0H-H0CD—1X013oi-I-hC)5—ciDi-DiH’CDoH-H‘tiCDH50C)ciCC)-0Di,-t•H’H-DiH-CDMCD-cH-Diti‘0H-Cr1C_IlH-C)<tiI—’CDCDNCOCDHCl)0DiCOCDDl‘-0(DciCDCDCl)0Ct5-0<HCDCl)H-DiDiCD-CtCOCrC)H-—‘1CDC)-•0—0Cl)H-CD1<0CDM-,i0H-CtDlH-ID)ciH-,<DiH-NCl)CrDiI-IC)Di5Di<“<0H-COiCDIC)Ctci0i-cCD5to‘CI0H-CDH-H-,H-COi-cCDCDH-0COçtC)‘CiCl)C)CDCDH-CtCDCDH-Cl)CDDiI—iiH-I--cIC)CrCrCrH’CD0H’5IHi-cDlH-<H-H’CtCDiiCIciiCDciC)Dl0JICD0ciCODlHDiCtCDICOciSi-iCDH’H-HCtH-0CDCOH<DiH0‘<CD<DiDiciCtHCl)5H’CrHCFCOCt•CDWCD014Chapter 2Patterns of Morphological Variation in Menziesia2.1 IntroductionThis chapter examines intrapopulational andinterpopulational morphological variation in North AmericanMenziesia. Since Menziesia is generally well represented inherbarium collections, detailed study of its morphologythroughout its range was possible. Furthermore, informationfrom herbarium labels was used to generate distributional mapsof the genus in North America. Most of the material examined,however, came from population sampling carried out duringfield seasons of 1977-1990. This work permitted anexamination of variation within and among populations andprovided ecological data to be used to test the relationshipbetween morphological variation patterns and environmental andgeographical factors. A cladistic analysis, which exploresthe relationships between the North American and Japanesespecies of Menziesia based on morphology, is also presented.2.2 Materials and Methods2.2.1 Herbarium Specimens and Species Distribution MappingHerbarium specimens of Menziesia were obtained from 29major and regional herbaria located in North America as wellas from three Japanese herbaria (Appendix 1.1). Over 1800North American specimens and 164 specimens of JapaneseMenziesia were examined in the course of the study (TableP)H-b..)D)(1H•HJCDu-I0‘1ci 5— 0HCo•tQHH-C)5:-ci00CoçrCDrrCtHCDCoCDCo0CtNH-MiH-CtCDCDCDCDCDCti-Cl)CoCo‘1CDçtH-CDcuCDCQC))(QI-•CDci‘-CCDC)CDo•0C)MiCDCoC)MiC))00IICtCoC))Q:-H-CDCo5CtNC)H-(-h)C))HCtC))(y-H0ciciciH-MiCoH-H-HH-Cl)C))I-QC)H-HHciciS ociHZCooCDC))0HC)JCtCo‘-CtC))C))CrCDC))CoCD<oC))--oC))1I-iJ5I-]ClH-H-CDC))CociCD‘-Q‘—sCrCoH-HH-U)k<•C)CDjC))0t’)-CtC))Hi:-HaCD[-JCDr H-C))C))C) CDHCDHcici0CtCrC)C))C))CtCtCoCD ciH ci H H-‘—3ciCl)CD(DC)) H(DC))C)‘t5CoCtHH-1.1.C))C)0CtCo—cu’0SH)Ct CDHO‘-CH-I-S CD‘-QaciH-,C))0H-I•CtoCl)‘-cCDCo‘-CCDQci1<Ct H-H-o:iC))CrHHf-3CoC)) >‘CooCtociS H- oH-CoSC))H- Cr 0 ‘-35CoC))c:‘tiMiMiH-H C)CQH-CDCoCrH-H ci‘tiCD‘-C(-t0C))tQHIIHC))CDSci H C))‘-CDCrHciC)o0H-0CoHoQçtI—’(QH-‘—CCDI--CCrH-C)C))ct‘ticiHCDCrQH-H-C)C)00ciSC))C))00c1MI-hC))Coci H-CD C))C))C)‘—3I’Ct‘tiCDCD‘tiCoCoI--C‘tiCo0MiCDHX0C)CrH-I-CH-CDSCDC))CDCoHCtC))CDC)tiMiH‘-<0CociS‘-0H-CDC))U]CrCrçtO’,OCDCDHI--C00Mi5MiMH-0Ct5H-C)-’ CoCD(QCC))CoNCrCo‘tiCDCDCDCrCDciC)CDH-CDHCrS‘--CCI)0CDCoCr HSCoC))ctC)jciciC))CDçtciC))CDci(tCOCoLQC)-’•Cl)5C)k<-CDtiWHCDH-CoCUCoC))H‘-CI-CçtCDH-CDCDC))C)H5H-CoCoCoC))‘-C)C)‘tiCDCDCD—CoCD‘-CMiCDC)C)5CD0C))-•0‘tiH-C))I-5C)CDCt0ci‘C)CoCDtCrCDI-CCtHCl-C))C))H-C)-’CD‘tiHCD(QI-CC)-)H-‘tiCoI-CCDCDHC)C))CD0HH‘tiC))5CI)LOCoHI-C<5C)CDCo(I)CDC))0‘—00Cl)CoH-H-C)CDHCU•H-C))0CoCUctç-tCDHCDCDCDciCD<C)CDH-Co<CDH-CoCDCr‘-CCDNCtHh0CoH-‘-CCoCDH-CtH‘-Q0COtiCD0CoH-I-hH-I-C-çtCDC4H-C))HoCU—0Ctcti‘1H-HCtC))H-0U]c_QSCDCDCoC))C))QCo•‘tiHI-CCDHCDCoH-s)HCDS•Mi.0‘1H-CDH0CDC)CUCrI-CDH(DC))ci-Cr‘1ciCDC))C))CoH-CD5H-tOCtH(Q0•CDC)CDtoI-jCo—CDC)CtjMiH-Z00C))0C))F-3I-CHi•CoC))Cl-QI 0CD C))LiiCo‘-iJk<I:-’ CT]‘--C>CD Cl-Q‘--CC))H-çtCDC))C))C))HCo CDC))—ci< CDCo‘--C0Co‘--CHCt0H-CtOoC)MiHWo0C)‘--CC))HHC))H- IIH H CD ‘—-C C)) ‘--C H Ii S H C)) CD H H Mi 0 ‘--C S C)) Ct H 0 C)) Co CD Cl CD I--C CD ci H Ct 0 C))H I—n16Appalachians. Populations situated at or near the rangeextremities (eg. HUN, TET, WBR, LWS) were of special interest.Since most sites contained large numbers of Menziesia bushes,OTUs were selected with the aid of a random numbers tablealong transects, usually near trails or logging roads.Herbarium specimens in good condition were chosen to representareas not covered by the field sampling. Of the 238individuals used for the morphometric analyses of NorthAmerican Menziesia, 160 were randomly sampled (99 west/61east) and 78 were herbarium specimens (57 west/27 east).To facilitate the statistical analyses of variation, OTUswere assigned to regions defined by physiographic, climatic,and ecological factors (Figs. 2.4-2.5; Table 2.2). Theseregions were similar to those used in earlier studies ofMenziesia (Hickman and Johnson 1969; Bohm et al. 1984), butdiffered by ignoring political boundaries.Material of Japanese Menziesia came entirely fromherbarium specimens. Unfortunately, only 30 specimens of thecommon taxa were sufficiently mature to allow their inclusionin the morphometric analyses (Table 2.2).2.2.3 Selection of DescriptorsTwenty-two characters were chosen to describemorphological variation in Menziesia (Table 2.3). Becauseleaf expansion in Menziesia is usually not complete untilflowering has ended, for consistency, measurements were madeonly from mature fruiting specimens. This posed no major—H-C)U)c1çt—CUU)CrcJHU)H-‘ZCUHHU)CDCDC)H-I—’-I-CDCD•H-I-CDH-‘DCDCDCl)IICD))H-C)NC)CUCUI-hc-r0I-FØCUQC)<U)C)0CDU)Q.I—hCDCD•I-CDH-OH-I-H-H-IIU)CDU)HC)HU)HLiH-CrI-H-<CUC)U)CUC)•C)C)MCDCUCDCr<<CDC))CDCDc-rCUCDHCD0CfCD‘-IIH-C))HU)-I-QQCUHH-H-CDU)IQCDH-C))U)Q.CD0H-<C)I-IU)C)HCf‘—atQH-HU)CDCDCUtYCDCDHCrCDC)CDCrU)H-H0QCrCUU)H00CDCDCD-.CDH-H-C))0U)CU<0CrCU<-CUCDHNU)C)c-fU)I-CrQZCUC)CrCDC)H-CDCUH-I-hCr0Q.I-1QQ.CrCUCDC)CDU)CDCDC))CDH-Cf0-ICDCDH-H-CDU)C)—CDC)U)U)XCDU)0U)t‘I‘dU)c-rCDU)HI-H-Cr(1CDCDCDCDHH-HCDCUC))IH-CUhCD0Q0H-H0NCDI-i)CDU)CrH-0HCrtiCrCDCU1>3Q.C))cv<U)tiCDCDCrC))C))rti0Q0-C)rrbCDH-CDCDC))IjU)H-C-IC))IC))CDCrCrCDCDMI-hH<C)crF-I.QCDH-CtHQU)—0Cl)I-I-U)I—.CrCDH-H-HCUCDCDHC)CDICDI-I0)0U)HCDC))C))C))DCUCUQHQU)0-CrU)tiIIH-<C))MH-I-hMCDCUH-0tQCUCDI-IHHU)CUCDWC))tQHH-0I-CDHU)Q.0—CU<CDc-tCDCDU)Ct‘-<I-i0CDC)cv‘-jlHc-vHCUU)U)QC))0C)HClC))H-CD00CDU)CDCUCD<CrCDZcvH1<CDH-,Cl)CrCDC)U)CtI-CCDClU)U)CC)<C)CDC))C)I-’-1—hCDb’CDCD‘-10U)U)C)CD‘-Cl1-CCDCUCDI-CCD0CtI-CCr0U)‘<H-1<U)C)0I-CH-CrH-H-CDH-U)-CUC)CtMCUCD1-CH-hH-,C)I-C0‘tIlCDCDCDo‘c-viCD0HC))H-CUC)QU)cv-j‘ci)HH-U)C)HH-CD‘ciCD0Cl1-CHC)CDC))0U)C))<‘-CIH-CDCrH-,CDU)tI‘<CDH-CUC))1-CHHH-0U)CDCtCDU)lCD(Q0I-CU)CDU)C)H-ilCDCDHCDNMH-,QI-CICflC)U)H-Cr1iClCl‘<CUCDCrCt-CUIH-CDClH-CDCDC)U)‘<‘ciINH-CD‘-CIC))CDCDCr‘-CCD‘ciCU-Ct‘CiU)CDCtCD0CDrrCDHU)C)CD0CDCilCDCDU)C))Crt3CUHCDC)H-,CDi)CD‘-CCUCDCC)HCDCDiC))‘CiCDCUU)—-‘Ci0H‘<CDU)ClCDCD<HClC)IC))C))0CDHCII-CU)-ClHCDCUI-CU)I.QHCDCDClCD‘-3Cl‘-CCDU)CDClU)crc-v•CDU)U)CDC•0HC))CDH-‘<CDH-HCfU)CC)IILYCD0CD0<Cl‘CiC)‘<tQCDH-0CDCDC))0Hc-I-C))I-hCD‘CiCDCfH-,CUCr‘-CcrC)CQ1-h-31-CCDc-v1-hU)CfHH‘CiC)C)H-CD‘-CHCr0CD0CD0U)0-U)U)0IIU)-CDCUU)I-CC))H-,I-0CUCDCD‘-CC))‘-CClU)CDH‘-C0CUI-iCDCiC))‘-CCDC))CDCDCDH-U)CDCD‘-CiCUU)CDClCDH0‘CiC)H-iNC)HCfClC)I-CI-CC)H1-CCDH-CrCDCDC))CrCDH-C))H-H-c-I-CDCDC))CD‘-CCUC)CDClU)CrCDCDClU)CD0ClCl(QWI-0p)xt H-HC)0)(1CDHU)H ‘-<J_I-H-HC)c:u H-CDU)H‘-<H0—J0-0H,CDCDQ H-C)I.-]U)H-CDQU)MCDU)U)H-C)U)0I-iU)H It’1t’0CrCDH,0 ‘1H,00U)U)CDU)CD‘-Q<CDU)< U)U)H-HH-0U))U)(-r CDCt CDU) HU)IDClHCDCDHU)U)C)C)0‘-I-H,H-HCtCtCD0ClCD‘•lU)U)U)Ct<HU)CDZçtI-(QH-U)U)(Q0CtCD‘-H-U)ClC)CDU) HCDCD‘1ClU)CDU)C)HU)Hk<HCtU)C)H-CDU)U)HCtU)U)CtCl-CDCDClClHCD0U)l-C) 11CDHU)‘tiocr 0oC-)It,cClW LI)-U) CDC)Cl H-Cr0H0CDH-,U)U)CD‘-‘ICDU)Cl-(QC)CDCD1iU))-U)‘CD‘t’CDCr-U)CDIt’C)çrHH-U)CDU)CDHU)U)U)l-C)CDHC)CDtQC)CDCt‘tiC)CDCDU)U)CDU)C-tC)ClHClU)CDCDCDU)C)U)‘tiU)CDCDU)H-IIU)0t’•CrCDC)CDH-‘t’H-U)CDC)CDU)C)U)11CtCD11CDU)<It’ClC)C)CDHCDC)ClU)‘ti—CDCl‘-•ICl-ClCDU)H-•HCrCrC)Cl)CDCDH-HIriHCr‘-<U)CDU)C) U)CDC) 0H0U)CDCrc-fCDH Cl‘tictCDU)0U)t’•U)U)CDU)CDU)ClH-U)frU)CDt’CDClCrHI-ClCDCDCDHQI-U)H,IiCrC)‘-IU)H-0CD0CrU)CDCDHCl0CD-H,U) C)U)to•CDL’.)•i-h U)0•C)I-I-IL’.)CDH-U)CDCrexjl-<CDC)U)U)-I.’.U)(1HClCDU)(U)C)C)H,0Cr0)•CDI•1‘-‘I0)CDU)( CrCl-CDCDF-ClU)aCtH0CDU)to •(_.)U) CDU)U)U) H-CDH z C) HCD I-Cl CD0ClH,U)U) H Cl)CDI-iCD0 H,to-U)t’HCDCD‘-CDCD‘tSU)U)ClU)HU)1JClH,H-$iHC)I-ICDCDt’U)CDH-c-rCttOU)0CDCDH-U)UCrU)crCrU)•CrU)IICDCDCDiJ00H-CDi-IXCDC)CDCDCt(1)Cl<H-Cr‘tiCDU)U)CrCrlHCD—U)CDH,CDCD0C)ItsU)H-Cl)CtI-i-CDCl)0CDU)H,H‘1Wt’CT)tCDHHU)CrrCrH,U)HCDClCrCD<U)U)U)U)t’U)hIDH-HH-CDClClCDIU)Cl-CDr1U)HCThCDH-CDH,QCDCtU)H-0HU)CrCDH,U)U)U)-CrHCtH-0<U)MH,U)CDCt)HH-CDHU)‘tiU)CDH CDClHt’U)U)CDCDCDU)C)MU)U)0U)H,C)U)H-CDCDU.CDCrU)CClCl‘<ci-C)ClCCDQU)CDU)Cl)CrQU)1<CrCrCD0CDS CD U) U) C I-s CD S CD U) CD I-s CD I-h 0 C Cl Ct 0 CD U) 0 U) I—c H U) H CD CD CD 0 Cr t3 CD U) U) S CDI.’.0) 0) Cr CD 0) I-I 1< I.J•0) 0 1h CD 0) () I-I.Cr 0 0,U) H- N CD U) U) It, CD U) ClIt, C CD U) C) CD C) CD Cl CD U) C) II H Ct 0 ‘-.5 U) I-Il 0 II C) 0 U) H- U) Cr CD C) Cr CDC CD I—c U) 0 H, U) H I-c U) C) 0 C Cr CD Cl H Cr H U) Cl H- U) U) CD C) Cr H U) C) 0 It, CD H,H-CD H Cl U)H cx19graphics package SYGRAPI-{ (Wilkinson 1990b). Variation wasexamined both within and among populations as well as byregion and taxonomic group.Descriptor dispersion was assessed within and amongregions by fitting the observed data to a standard normalcurve and checking for departures using the Kolmogorov-Smirnovgoodness of fit test (Zar 1984) . Lilliefors correctedprobabilities, needed for the proper use of the test(Lilliefors 1967; Sokal and Rohlf 1981), were obtained usingSYSTAT program NPAR (Wilkinson 1990a).For normally distributed descriptors, differences amongtaxonomic groups were tested with one-way ANOVAs using theSYSTAT routine STATS (Wilkinson 1990a). Bartlett’s test wasemployed to check for the equality of group variances. Insome cases, variables were transformed using eitherlogarithmic or square-root transformations to correct fordepartures from the test conditions (Zar 1984). Differencesin variable means among taxa were determined oosterioriusing Tukey’s multiple comparison procedure (Zar 1984). Fornon-normally distributed variables, where the assumptions of aone-way ANOVA could not be met, the Kruskal-Wallis nonparametric ANOVA was used instead, followed by Dunn’s multiplecomparisons tests (Dunn 1964; Zar 1984). The lattercalculations were performed using output from SYSTAT routinesNPAR and TABLES.Clinal variation in the data was explored by creatingcontour plots of descriptor observations versus locality using20the inverse smoothing option of SYGRAPH routine PLOT(Wilkinson 1990b)2.2.5 Multivariate Analyses of Group StructureA variety of multivariate analyses were employed toexamine group structure in the morphometric data. An initialexploration of the entire data set was accomplished usingunweighted pair-group method (UPGMA) cluster analysis (Sneathand Sokal 1973). A Euclidean distance matrix of OTUs, withvariables standardized to zero mean and unit variance, wasused. All calculations were done with the NTSYS-pc routinesSTAND (data standardization), SIMINT (calculation of thedistance matrix) and SAHN (clustering) (Rohif 1988). By theiralgorithmic nature, cluster analyses will partition the datainto groups, whether such groups are meaningful or not (Orlóci1978; Pilou 1984) . Therefore, a principal components analysis(PCA) based on a correlation matrix of the original data wasperformed to check the validity of the groups defined in thecluster analysis. This was carried out using the SYSTATroutine FACTOR (Wilkinson l990a).Group structure within Appalachian and western NorthImerican Menziesia were examined separately using subsets ofvariables to perform principal components analyses. Thevariables were chosen based on their near normal distributionwithin regions and on their ability to explain uniquecomponents of variation within the first few principalcomponents. Descriptors were removed if they were found to be21highly correlated with other retained variables. This wastested by comparing the Pearson product-moment correlationsbetween all descriptors with their partial correlations(SYSTAT routines CORR and MGLH; Wilkinson 1990a). If theproduct moment correlation (rhi) of two descriptors is greaterthan the partial correlation (rhi.), then their relationshipis largely explainable by correlations with other descriptors(Rohif 1977). Such redundancy adds little to the overallstructure of the data.Groups defined by cluster analysis and PCA5 were alsosubjected to discriminant analysis to test for distinctness ofthe priori chosen groups and to identify the descriptorsthat maximize those differences. To meet the conditions ofdiscriminant analysis, a subset of descriptors was chosen thatmost nearly approximated normality among the groups.Descriptors with little or no variation in one or more of theselected groups were discarded prior to the analysis.2.2.6 Tests for Association between Morphological Variationand GeographyThe association between morphological variation andgeography was examined by the Mantel test which statisticallyassesses the degree of similarity between two independentlyderived data matrices (Mantel 1967; Sokal 1979). Twosymmetric distance matrices are required; in this case, X, aEuclidean distance matrix derived from the morphological data,and Y, a matrix of geographic distances between OTUs. The22test computes the sum of products of the corresponding matrixelements according to:Z = I.. *where x and y are off diagonal elementsof matrices X and Y, respectively.The null hypothesis of no association between the matrices istested by iteratively generating random permutations of theelements of one matrix, and then computing the inner productbetween these permutations and the elements of the othermatrix, to obtain Z’ . The two matrices are considered tocorrespond significantly when Z exceeds Z’. The Mantel testwas performed using the NTSYS-pc subroutine MXCOMP whichautomatically calculates Z and Z’ and derives a t-statistic todetermine the difference between them (Rohlf 1988). Twohundred fifty iterations were performed to generate Z’Two forms of geographic distance matrices were used toconduct the tests. The first test employed a binaryconnectivity matrix which joins only localities considered tobe near neighbours (Sokal 1979). Related localities, linkedaccording to Gabriel networks (Gabriel and Sokal 1969), wereassigned a value of one; all other elements were set to zero.This type of matrix assesses whether geographic variation is afirst order process; that is, where only neighbouringlocalities or individuals are capable of influencing oneanother (Sokal 1979). A second, regional test of association,used a matrix of Euclidean distances between localitiescalculated from their latitude and longitude coordinates. In23this case, all localities are related to one another, inproportion to the continuous measure of geographic distancebetween them.2.2.7 Comparison of Morphological Variation with EcologicalParametersField sites were compared by constructing speciesassociation matrices to analyze their community structure.Presence/absence species matrices were used to optimize theefficiency of data collection (Mueller—Dombois and Ellenberg1974; Gauch 1982). Only perennial plants, primarily from thetree or shrub layer, were included because they formed themost stable visible element of vegetation within a site. Thisminimizes differences in site evaluation owing to seasonalvariation. Community relationships among populations weresummarized by unweighted pair-group method (UPGMA) clusteringof Jaccard similarity matrices of the data using NTSYS-pcprograms SIMEQUAL and SAHN (Rohif 1988). The vegetationgroups defined were then cross-referenced to the vegetationprovinces of Daubenmire (1978) and Braun (1950) to define thehabitats in which North American Menziesia grows.In addition, several ecological factors were compiled foreach site, including four physical variables, two soildescriptors, and four climatological parameters (Table 2.4).The physical and soil variables were measured at the site atseveral points along the collecting transect. Elevations wereobtained from a Thommen TX-22 altimeter corrected to sea level24while slope and aspect were measured with a Silva 15T compassand clinometer. Exposure was estimated using canopy photostaken over a sample plant with a 35mm camera mounted on alevelled tripod (Chan et al. 1986; Mason 1990) . A 28 mm wideangle lens with a 750 field of view was employed. The amountof cover was calculated from the ratio of dark to light areason the high contrast black and white film (Ilford XP400).Climatological data were obtained from meteorological records(Baldwin 1968; Canada Dept. of Energy, Mines, and Resources1974; Farley 1979) . A summary of these site parameters isfound in Appendix 2.Relationships between the patterns of morphologicalvariation observed in Menziesip and ecological factorsoccurring at the field sites were assessed using canonicalcorrelation analysis (CCA). This analysis compares twoindependently derived morphological and ecological datamatrices and projects a solution onto a limited number of axesthat maximize the correlation between the two data sets(Gittins 1979). Consequently, canonical axes differ fromprincipal components in that they do not simultaneouslyextract the maximum amount of variation present within eachdata domain.To meet the test conditions, individuals from 20 westernNorth American populations and from 14 populations in theAppalachians were compared separately, to minimize groupheterogeneity. In each area different subsets ofmorphological and ecological descriptors were used to simplify25interpretation and to reduce potential problems caused by non-normally distributed descriptors or co-linearity in the data(Gittins 1979). Calculations were performed using SYSTATsubroutine MGLH (Wilkinson 1990a). The results weresummarized by plotting the first two canonical variates anddescriptor loadings for each data domain. In addition,correlations between the descriptors and the canonicalvariates of each domain determined to be significant were usedto calculate intraset and interset descriptor communalities(SYSTAT routine CORR; Wilkinson 1990a). These communalitiesrepresent the amount of variation summarized by eachdescriptor within and between the two data domains. Theamount of variance and redundancy within and between thecanonical variates of each data set, respectively, were alsocalculated as outlined by Gittins (1979)2.2.8 Analysis of Japanese Taxa and the Cladistic Analysisof MenziesiaA full set of morphometric data was collected for 30 OTUsof the common species of Japanese Menziesia: . ciliicalvx, .multiflora, . tentandrp, and jy. purourea (Table 2.2). Thedescriptors chosen were the same as those used in the NorthAmerican study (Table 2.3), except for BPUBS and CPUBP whichwere invariant in the Japanese taxa. Group structure wasexamined using principal components analysis.Comparison of the North American and Japanese taxa wasaccomplished by cladistic analysis. Seventeen binary26characters were chosen which could be polarized amongst thetaxa (Table 2.5). Some characters used in the main study wereincluded because they exhibited discontinuous variation amongthe taxa. The remaining characters chosen were mostly floraldescriptors. The genus Cladothamnus Bong. was used as anoutgroup for character polarization as it is considered to beone of the most primitive members of the Rhododendroideae(Kron and Judd 1990). However, some characters could not bepolarized against the outgroup because both states were foundin Cladothamnus (Bohm et. al., 1978). In this case,polarization was assigned on the assumption that common equalsprimitive.Included in the analysis were the eight Japanese speciesrecognized by Hatta and Tashiro (1986), as well as . oilosaand . ferruQinea (both coastal and Rocky Mountain phases)from North America. Character states for each species werescored by examining a large number of herbarium specimens. Inaddition, published descriptions by Ohwi (1965) and Tashiroand Hatta (1986) were used to score the newly described orless common Japanese taxa, for which herbarium material wasnot seen. The resultant data set (Table 2.6) was analyzedusing Wagner parsimony with the cladistics package PAUP(version 2.1; Swof ford 1985). To ensure that all maximallyparsimonious trees were found, the branch and bound algorithmwas employed (Hendy and Penny 1982).272.3 Results2.3.1 Univariate Analyses of Character VariationOf the 22 variables selected for use in the multivariateanalyses of morphological variation, 14 were found to benormally distributed over most of the regions and populationsexamined. Nine of these were continuous variables and fivewere meristic: ANGTIP, LABOVE, LBELOW, LWIDTH, and PETIOLEdescribe leaf shape; CILIA, SPUBS, and SUBLATE describe leafpubescence; CALPUB, CALWID, CAPLEN, and SEGWID describe fruitsize, shape and calyx pubescence; and PEDLEN and PEDPUBdescribe pedicel length and pubescence, respectively. Onlycapsule pubescence (CPUBP, CPUBG) and leaf venation (NTJMVEIN)were non-normally distributed in all of the regions.Box plots of all 22 descriptors indicate differencesbetween three broad groups of North American Menziesia (Fig.2.6). For example, several leaf descriptors (ANGBASE, BPUBG,LABOVE, LBELOW, NTJMVEIN, SPUBG, SUBLATE) and fruit descriptors(CALWID, PEDLEN, PEDPUB) illustrate differences betweenAppalachian.pilosa and western M. ferrupinea, sens. lat.,although the coastal (ferrupinea) and Rocky Mountain(alabella) phases are not significantly different from oneanother (Fig. 2.6). All three regions are distinct from oneanother as described by seven other variables (CAPLEN, CILIA,CPUBG, FRTNUM, PETIOLE, SEGWID, SPUBS). Coastal . ferrupineaand N. pilosa are similar to one another in calyx width andpubescence as well as leaf width (CALPUB, CALWID, LWIDTH), andboth have capsules lacking puberulent hairs (CPUBP). However,28the Rocky Mountain phase N. ferruainea “alabella” and N.oilosa have generally broader leaf tips and densely puberulentleaf undersides (ANGTIP and BPUBP).In western North American Menziesia, clinal variation wasobserved in several descriptors (Fig. 2.7). Leaf tip angle(ANGTIP), the density of puberulent hairs on the leafunderside (BPUBP) and calyx pubescence (CALPUB) increasedinland from the coast to the Rockies, while capsule length(CAPLEN) gradually decreased inland from the coast and furtherdeclined southward in the Rockies. Other characters, varyingalong north-south trends, included the density of puberulenthairs on the upper leaf surface (SPUBS) and the number ofsubulate hairs on the midvein (SUBLATE), which both increasedsouthward along the coast from Alaska to California.Conversely, the density of glandular hairs on lower leafsurfaces (BPUBG) was highest in Alaska and along the B.C.north coast and decreased southward toward Oregon andCalifornia. Although clinal variation was less apparent in N.pilosa (Fig. 2.8), the density of puberulent hairs on leafundersides (BPUBP) and glandular hairs on capsules (CPUBG), aswell as capsule length and segment width (CAPLEN, SEGWID) alldeclined toward the southern part of the range from southernVirginia to northern Georgia.2.3.2 Cluster AnalysisA cluster analysis of the entire morphological data setpartitioned individuals into two major groups corresponding to29an Appalachian cluster and a western North American cluster(Fig. 2.9).Within the Appalachian group, chaining is apparent withintermixing of individuals from the five physiographicregions. One fairly coherent group consists largely of OTUsfrom the southern Appalachians (region A5). Other individualsappear to cluster because they come from open sub-alpinepopulations (DCL, W, MIT).In western North America, OTUs are clearly divided intotwo discrete clusters consisting of either coastal andCascades OTUs (ferrupinea) or Rocky Mountains OTUs (alabella).The intrapopulational cohesiveness of OTU5 is not particularlystrong in any geographic region. For example, while a clusterof individuals from the southern coast of Oregon and northernCalifornia (region W3) is evident within the coastal“ferruainea” subgroup, they are combined with a few northcoastal (region W2) and Cascades (region W5) OTUs. Alsopresent in the “ferruainea” cluster are all of the northernCascades OTUs (region W5), plus one individual (WAO2) fromFerry Co., WA in the Columbia Plateau (region W7), and oneindividual (CR01) from Government Camp, OR near Mt. Hood inthe Southern Cascades (region W6). The interior “alabella”cluster includes all of the OTUs from the Rocky Mountains plusall of the southern Cascades OTUs except ORO1 from GovernmentCamp. Despite a fair degree of interpopulational variation inthe interior, there appears to be some segregation of30individuals from the westernmost ranges of the Rockies (regionW7) relative to the main cordilleran range (region W8)2.3.3 Principal Components AnalysesThe results obtained from a PCA of the entire datasetcompare well with observations made from the cluster analysis(Fig 2.10). Again, two major groups of OTUs corresponding toM. pilosa from the Appalachians and . ferruainea, sens. lat.,from the west are evident. There is some segregation ofindividuals from the coastal or Cascades populations(ferrucinea) relative to OTUs from the Rocky Mountains(cxlabella), but there is considerable overlap. This points toclinal variation in western North American Menziesia.Together, the first two axes account for 41% of the totalvariation, with most of the descriptors loaded highly on twoor more of the first three principal components (Fig. 2.10;Table 2.7).2.3.3.1 within-group PCA of Menziesia ferrucginea, sensu latoEleven descriptors were chosen to examine variation inwestern North American Menziesia, including six continuousdescriptors (five lengths and one angle) plus six meristic(pubescence) descriptors (Table 2.8). Another ten descriptorswere dropped because they were non-normally distributed insome regions, or because they added little to the overall datastructure independent of the retained descriptors. A PCA ofthese data (Fig. 2.11) was very similar to one obtained using31all characters (not shown), with 54% of the total variation inthe data set described by the first two axes. Most strikingis a gradation of OTUs along a west to east dine.Consequently, coastal (regions Wl-W4) and Rocky Mountains(regions W7-W8) OTUs are largely distinct from one another,with individuals from the Cascades (regions W5-W6) lyingbetween them. Separation of the two phases is primarilyinfluenced by characters such as capsule length (CAPLEN) andcalyx pubescence (CALPUB), leaf surface pubescence (SPUBS) andleaf tip shape (ANGTIP). The overlap of some OTUs from thedifferent physiographic regions indicates a high level ofintrapopulational variation relative to interpopulationalvariation. Also apparent is a north-south dine among thecoastal populations with individuals from southern Oregon andnorthern California (region W3) being most distinct. Asimilar, but less obvious, trend is also seen in the Rockiesalong a west (W7) to east (W8) dine. Leaf size variables(LABOVE, LBELOW, LWIDTH, PETIOLE) and glandular pubescencedensity (BPUBG) dominate the first component, while other leafpubescence descriptors (SPUBS, SUBLATE) and calyx pubescence(CALPUB) are highly loaded on the second axis.2.3.3.2 Within-Group PCA of Menziesia DilosaTen descriptors were chosen for the analysis of . oilosaOTUs (Table 2.9). All ten were normally distributed among thephysiographic regions and include five length descriptors,leaf tip angle (ANGTIP), and four meristic pubescence32descriptors. Other descriptors were dropped from the analysisbecause they were either non-normal, or contributed little tothe data structure, or were highly correlated with othervariables retained in the analysis. The resultant PCA (Fig.2.12) was very similar to one obtained using all 22 variables(not shown), thereby validating the ability of the reduceddata set to explain most of the morphological variation in .pilosa. Of the total variation present in the data set, 54%was explained by the first two principal components (Table2.9).In M. pilosa, intrapopulational variation is fairly high,with individuals from a given population or region tending togroup more closely to OTUs from other widely separatedregions, than to one another. Variation among populations isnot very pronounced. However, southern Appalachian OTUs(region A5) and Pennsylvanian OTUs (region Al) form somewhatdistinct groups separated mainly along the second axis. Thevariables most responsible for this are capsule size (CAPLEN,SEGWID) and pedicel pubescence (PEDPUB). On the firstcomponent, leaf surface subulate pubescence (SPUBS) andpetiole length (PETIOLE) are most highly loaded (Fig. 2.12;Table 2.9).2.3.4 Discriminant AnalysisAll of the previous analyses indicate the presence ofthree broadly defined groups of North American Menziesia.Their coherence was further tested by discriminant analysis.‘Dc1)‘tiH-IC’)—c5tQrrHCDI-iQ<COrtbH,CQO\0Q(Q0><IH00COCl)tCI)(1)C)H-H.00—HCDQIH,:CI)ICQCI)crtH-rJCI)HHf-t0CO•iCl)CI)IH-IICI),QCDH-CI)JCDr‘CDCOH-kDHCDC)COIHHH-CICDCr0CDC5CDCOCI)H-CtJU)COCtCDCOtQCI)CO-Cl)-thHCI)CtCO(Q)HC)I-CDCI)Ct-.H-0H-CI0‘<COCDCDCDCOCtH-COIICl—H-CDCOCI)CDCtCD=-0CD-CDH-0CO•CO0IH,Cl)ZC)ZCI)Cl)C)H-bClC)C)H,CI)CQCD0lCD0H—COlCDt-CtHH,CI)0AH-0C)I—i<‘JCOIt-iH-‘t‘d0IS))C)CDH‘<CtCt00‘tCrC)IhCDCI)CDCI)0-JC)(QCDS))CI)CO<:CI)L1I<CDCOCI)‘jHIH,‘—H,HCICD•CtI-iCD‘1HCDClC)COCO‘1CtCI)ICDH-ClCDI-CDt’lH-<0I-iCtC)CDCDCDCDC)COCDCDlCD‘H,COCOCDCI)H-I-—CI)I-xjCOCr‘-I-iClCtCDCt0CD•CDH-H-0CICI)CI)H-0CI)‘t3CI)CDCDCDCI)H(QCOCO‘1COCDtCI)1jH-HH-fl-ICI)CDCDC)H-I-5H,CI)CI)COHCD•Ct0C)CDH-H(QCI)CD‘1X‘i0CrCQC)1-50CI)CI)CI)CtCDH-CI)CtH-CDI-C-tCOH,ClCOC)C)çtCI)0COCI)H-COHCl)0H-H-•0CDHCDHCOCDH-Ct-HCDHf-tCOCDCO<H1CCOCOCDCl-H(QCD1COCD-CD(-‘-)HHClCOCOQCtCI)0—CI)ICDH-CO0CDH-H,CO:CrCDCOI-IHH-H,CtCOHCt0‘<NCD1-5l-’J1-5‘CD•H)CI)ClIIHCDHH-IIQLflH,CDCDCDCD0H-S))CI)CDH<C)CD—CX)CCH,Z1-5HCDCtACOCrCDCDZCI)Cl)Ct-•U]CDIQH-Ct0CI)•CDHHH-ClHCOZCDl—5‘-5HCtCS))HCtCOC)I-i‘<L,-H-CD=0CrH,‘1CDClClCD•CD-5CDCDCOClICCCDCDClCDCI)CD•C)—J-C<1-5CtCDHIHH,CDC)CDC)ClCOH,HCC)CDCtCI)H--5COICICICtCOC)C•WHC)‘Z31-5COCDH-C)C)I53c-tC)CI)CI)CI)CDCrCDCOH-Ct1-51-5UilCDH,Cl‘‘-5‘tICOCO‘tSCO•Cl•CDCI)H-CDiH-—H-lF-‘-sCt’CDHH-CDCtCDc-tH-lC)Cr‘CDIHCCOCI)‘CIHCDCDCOCI)HC)CDCrECO1CIIS))C)C)Cr=CI)CClCO1-5CtCIHCOCD‘CICClCCD‘1CDCH-I-5HCDC)3‘CIH-CI)HCDCO1-51-5—CridCl(QCOCDti‘CIuCDCOCrC)0HCrCClCCDCOCrCDC)C)ZS)C)HCDCrCOCDCI)H-It-jCDH-H-1-5CDIHCrCI)CCtCOlCDCH-CD1-5CI)Cr‘CII-iC)CDCOCl)H-.5CDClIHCDCrC)CtCCCDCDCH-COCOCDCOC‘CIIHCtCtClCZCt3CDH-QCOC)COtQCD‘-tQ.H,CI)HIS)-)CO0CDH-CDH,0COCDCI)C)3aClCDH-1-5CD=CD1-5COCI)CD0CI)CI)•C)C)COC)f-tçtHCO<CDCO:iCDIC)H,CrCrH,CI)01-5CDHH-H,H-Cl0Clo-CH-H-CO-H-1-5I‘1CDCONCOCC)U]CI)CO3CrC)C)ClC)‘CIH-CD-CtHCDt3CDCr—‘COCDCD(H-S))S))CtC)CDH,CO0CDCD0—•Qc-tQ‘CIHClCI)0CI)H-H,COH-CI)CrVCt‘<CD><‘—5‘-5Cr><COH-HCOCOCtClCDCOCDCOCtCDCl-CDw C))34discriminant analysis. Only two OTUs from ferruainea wereassigned to “tilosa” while another two OTU5 were placed with“plabell”; all four OTUs were from the Cascades. A total offive OTUs of “alabella” were likewise transferred to“ferruQinea2.3.5 Association between Morphological Variation andGeographyMantel test comparisons between a distance matrix ofwestern . ferrupinea, using all 22 morphological descriptors,and a binary connectivity matrix of geographical association,based on a Gabriel network (Fig. 2.14), resulted in a nonsignificant correlation between the two data sets (r = -0.343;p(Z > Z’) < 0.01). The same test applied to the Appalachiandata set and its associated binary connectivity network (Fig.2.15), was also not significant (r = -0.131; p(Z > Z’) <0.01) . However, when regional variation was tested usinggeographic Euclidean distance matrices, highly significantcorrelations with the morphological data sets were found, bothin the west (r = 0.164, p(Z > Z’) > 0.001) and in theAppalachians (r = 0.179; p(Z > Z’) > 0.001).2.3.6 Ecology and Patterns of Morphological Variationin the Western North American Menziesia sitesFour main community types were identified in the westernMenziesia field sites according to a cluster analysis (Fig.2.16) of the vegetation found at those sites (Appendix 2.1).35Most of the sites in the Rocky Mountains, and some from theCascades, are in subalpine forest dominated by Engelmannspruce and subalpine fir (the Picea enaelmanii Parry Province)(Daubenmire 1978). The remaining populations are found in theTemperate Mesophytic Forest Region of Daubenmire (1978). Ofthese, most belong to the Tsuaa heteroohvlla (Raf.) Sarg.Province (central section) . However, Moscow Mountain, ID(MOS) is an ecotone site between the Tsuaa heteroohvllaProvince and the drier Pseudotsuaa menziesii (Mirbel) FrancoProvince (eastern section). Although Humboldt County, CA(HUM) is in the southern section of Daubenmire’s Tsuaaheteroohvlla Province, it is quite distinct because it isdominated by a canopy of redwood (Sequoia semoervirens (D.Don) Endl.). Here Menziesia attains a unique lifestyle,growing in organic debris on nurse logs and as an epiphyte inthe lower branches of redwood trees.Morphological and ecological variation in the westernNorth Anierican Menziesia populations were compared usingcanonical correlations analysis (CCA). The morphologicaldescriptors used included the same 12 characters used in thewithin-group PCA of western Menziesia, plus shrub height(HEIGHT) (Table 2.10) . Nine ecological factors (Table 2.10)were used in the analysis; only FROSFREE, the number of frostfree days, was dropped because it was highly correlated withthe mean January temperature (JANTEMP), and thereforecontributed little unique information to the data structure.C)HC)C)0U)PU)C)I-cUiC)cvH-cvCDNQ—)H-0T))H-h(1)CDi0CDC)0H-IQWCDiHicv05CDQI-I—iH-5C)CD0i—hIi--i—ii-H-tQ0c-vcvH-cv<5Q.cvQUiti‘-I—CDM-Ic))Hk<:tQHiMH-CDT))H-c))cv0C)CDCDH-0CDI5a”H-(I)CDI—’-0I—cQi-U)H-CT)—H-I-cD)‘-Q<h1(1)-U)C)cvi-cCDCDU)Qc))U)cvH-CDCDII—’H-t’JCDU)c))T))c))tCDCl)00CDcvH-CDC)I-cIHC)5CQHcvi-ncvH-Cl)C)cicvc-vI-c))U)cv))pC)H-HhH-I-hCDT))ctQH-CD‘<0I-c(QpHH-=i0H—JCDCD<U)CD0C)c)-)C)cvH-CDC)U)CliCli—CD‘tipcvC)cvHcvI-CDHcvU)CDcvcic-v—U)l-Hl-H-cv0H-,H-H-CDCDU)CliCDHH-CDcvH-l-CDcv0C)H-C)05Qc)i-ClCD0U)0CDicvH-,cvCl)Cl)0I-c))cvU)0tQtiC)CD(QH0HCDcvHCDH-l-cvCDCDMCDHH-Qc)U)I-H-H-C)ciCDciHCDl-H-U)I-Cl)tQcvH-cvU)cv0U)cvC)CD5cv•HClH-‘tiCDCT)cvcvCDHC)CDH-cvCDHcvCD0CDcv5cvU)H-ci0Z‘-Q<U)H-CDU)cvCDCl)CDU)I-cU)cv0tiU)CD‘-CT)cvci0I-5C)C)Mc)•-i—cCDCD(QC)CDcvI-U)Cl)CD‘ti0i-H0CD‘-CD—ICl)5CDJCDHHcvU)CDClQ‘tiCDMCl)CDHCl)IH-U)00CDH-CDH-i-iU)Cl)U)<cvH0ctU)C)l-C)H5<CDH-C)Cl)cvH-U)CDCD0c).CD0CX>Jc)-IU)‘tiCDlCDci0CDhcvCl)H-CDNHIcvQCDciCDC)H-I-cHHC)U)CDU)CQH-5•ciCD1-i0CDC)‘tiU)U)l-C)U)CDI-hCDCDU)U)<HCDCDCl)c)00CDCD0U)CDH0i-ctU)I—nCD0cvU)U)cv00hU)0IHM-,0H-cvCn00l-0ciI.QCl)CDHI-hcvNH-C)tU)cvM-U)0H-CDCDt-ci0H-c))U)HCDticv0-SH-C)CDC)‘<(QU)CDCDU)UiH-I‘I]‘<‘ClU)zcvcvH-ZU)hCl—0ClHQ.H-cv0CD0HI—’-cvC)Cl)cviH-Cl)Ui<C)CDi-c0CltQH-‘CSH0cvcv<0Clc)ciCDClCD-iLI]I-cU)CDcv•C)HciCDcvHH-i-jQCQCDC)CDCDCl)HU)CD0CDU)I-i-HH-UiCl)CDI-I-’)NCl)SC)0<CDCD00H-C)U)U)•0cvcv0•ClcvCl)SCl)CD(QCl)UiH-C)CD0HH-H-,H-H-5Hl-><0C)(QU)C)LIIZCDCl)H-5—1CD0I—’-li-ncv00c))‘<H-U)‘Cl:—CDi-ct-QU)--•Cl)cvt-Cl)H0CDI-cCD0-0cvH-Cl)H-Q.H-H-cvcv0U)ClU)c))U)C)CD‘-H-CDcvCl)C)NH-‘Cli-hIU)Ui-I-iU)c))H-Cl)IcvCl)CDCD0CDi-icic))l-H-(I)Cl)-0iH-tCDCl)HCD0—ciClCl>Hi--iH-icv•ClU)5U)—(QCDC‘ClCD‘ClcvC)cvi)i-H-‘Cl0Cl)cv-H-HUi‘Clci‘ii—iCDCDCDc)CDI-‘Cl-‘-XI‘Cl—-CD0IO0CDI-ii-c‘ClU)cv‘-U)CDcvH-Mi-cC)LI]ClU)l-•HCDH-U)H-cvClCr)5I-cCDi-c><C)i-cH0U)H0-‘iiHI-CDCDCDC)CD—‘-UHU)SCDC’QCll-i-C)cvH-cvcvU)LI]I-5CDhH-Cl)05—H-ycvCDCl)U)CDC)CD‘ClCl)Cc-iHcvCDCliC)ClCDH-,QU)i--ccvU)H-CD0i-c50Cl)Hci0H--c))CDCD0—CDCDcv-‘ClH‘ClI-i0‘-i]i-<CDH-Cl)H-Cl)Cl)H-LI]HcvH-5H-CDH-Cl)SCQU)cvU)H—CDHCl)H-U)IQClCl):—0H-CD-H-CDcvCD<cvZCD0H-ciCDi-U)CDcv37relationship between the two data domains adequately. Theamount of variation described by these three sets of canonicalvariates are roughly similar for both the morphological andecological data domains, though a greater amount of variationin the ecological domain is explained over the first two axes(Table 2.10) . The morphological canonical variates (Ul-U3)account for more of the variation in their own domain thanthey do in the ecological domain (Vl-V3), and vice versa, buteach set does account for a reasonable portion of the other asseen in the redundancies. Ecological variables such as ELEV,JANTEMP, PRECIP, and SOILORG have high intraset and intersetconimunalities and therefore are the most importantcontributors to the analysis. Leaf size, shape and pubescencedescriptors (LABOVE, LEELOW, LWIDTH, ANGTIP, SUBLATE), as wellas capsule size (CAPLEN) are similarly important in describingvariation in the morphological domain.2.3.7 Ecology and Patterns of Morphological Variation inthe Appalachian Menziesia SitesIn general, Appalachian Menziesia field sites arecharacterized by a conspicuous deciduous tree component andhigh species diversity (Appendix 2.2). A cluster analysis ofthe sites based on the presence or absence of 41 perennialplant species resulted in the recognition of two majorcommunity types (Fig. 2.18). These are representative ofeither Picea rubens Sarg. subalpine forests or temperatemesophytic forest habitats largely dominated by oak38(Daubenrnire 1978; Braun 1950). Of the latter, most arereferrable to either the Ridge and Valley or Northern BlueRidge sections of Braun’s Oak-Chesnut Forest Type, except forsites from the southern Appalachians (LWS, WSM, PIS, BB) whichare particularly species-rich.In the canonical correlation analysis, eightmorphological descriptors were chosen, including five of the10 characters used in the within-group PCA of the Appalachiandata (2.3.3.2). A preliminary CCA indicated that ANGTIP,CILIA, CAPLEN, SEGWID, and PEDPUB were not significantcontributors to the analysis; hence they were dropped.However, despite their correlation with LABOVE, the leafdescriptors LBELOW and LWIDTH were shown to contribute to theanalysis and were included on that basis. Of the ecologicalcharacters, FROSFREE and SOILDEP were found to be highlycolinear with JANTEMP and AUGTEMP, respectively, and were notretained. In addition, ASPECT, SLOPE and SOILORG werediscarded as they added little to the overall data structure.Only the first three canonical variates were necessary tosummarize the interactions between the two data domains (Table2.11). The ecological canonical variates (V1-V3), inparticular, describe a greater amount of variation than do themorphological canonical variates (Ul-U3). As well, each setof canonical variates accounts for more variation within theircorresponding domain (total variance) relative to theiropposite domain (total redundancy). Nevertheless, each setdoes account for a reasonable portion of the other. All of39the ecological variables included in the analysis have highintraset cornmunalities and two, JANTEMP and AUGTEMP, have highinterset communalities as well. Similarly, leaf size andpubescence characters (LABOVE, LBELOW, LWIDTH, SPUBS, BPUBP)and HEIGHT are important contributors to the morphologicaldomain and also have reasonable interset conimunalities.An illustration of the two data domains (Fig. 2.19)reveals a correspondence between morphological variation andecological factors. With an increase in elevation (lowersummer and winter temperatures and a shorter frost-freeseason), leaves tend to become smaller (LABOVE, LBELOW,LWIDTH) while leaf surface pubescence (SPUBS) increases.Moreover, certain types of leaf pubescence (SPUBS, BPUBP)appear to become more dense when subjected to the higherexposures experienced on the open heath balds in the subalpinezone. With increased shade in the oak forest sites, leavesbecome larger and plant height increases. Consequently, thereis some relation between the disposition of OTUs in the datadomains with the ecological community types defined in thecluster analysis of sites (Fig. 2.18). To a lesser degree,southern Appalachian OTUs are also distinguishable from OTUsof the more northerly populations.2.3.8 Comparisons between North American and JapaneseMenz ies iaAn initial examination of morphological variation inJapanese Menziesia involved the use of PCA to explore group40structure within and among samples of the common taxa (Fig.2.20). Two distinct groups are apparent. One clusterconsists of a mix of . multiflora and . ciliicalvx OTU5,along with an individual of tetramerous . ourourea. Thesecond cluster comprises individuals of j. pentandra, whichappear to be quite distinct from the other common five-merousJapanese taxa. The first two components of the PCA togetheraccount for over 46% of the total variation present, with mostof the descriptors being loaded highly on one or more of thefirst three components (Fig. 2.20; Table 2.12). In general,pubescence characters are most highly loaded along the firstaxis while variables of size or shape dominate the secondaxis.In a cladistic comparison of the North American andJapanese species of Menziesia, based on morphology, sixmaximally parsimonious trees were obtained using the branchand bound option of PAUP (Swof ford 1985). All had a length of32 steps with a consistency index of 0.531. A strictconsensus tree (Fig. 2.21) illustrates the elements common tothe different solutions. Most notably, all of the NorthAmerican taxa, with the coastal (FERR) and Rocky Mountain(GLAB) phases of . ferruainea being most similar, are linkedin a cohesive group with . oentandra from Japan. As seen inthe PCA of Japanese Menziesia, j. oentpndra is distinct fromthe other five-merous taxa based on a number of characters(Fig. 2.22; Table 2.6). Like all of the North American taxa,N. oentandra has small flowers, glandular hairs on the upperriIQ0II-hI‘tjIIc-I-Ic-iH‘jHHCl)H-CDIHH-IIHH-HQ.CDc-I-lI-IQc-I-QC)0c-I-CDk-I-C)c-I-CDDlCDbCDCDH,H-HNH-‘IIH-CDH-<D)c-IIIH-IIIIH-0CDII0C1Ii00IICDH-iII0CDC)<CDH-H,‘tSHHIH-Q.H-c-I-lCDH0Dic-I-0H,III-hCDIICDHCDCDIIIc-ICDc-IHCDCD<CD0DlH-0-HQ.Dlc-IID)CDCDH-IICDHCDc-I-II<CDII<c-I-CD(DHH,c-I-kCDJc-I-CDDiI0CDCDCDCDDic-I-I-DiIIH-H-CDDl--c-I3‘t5H-C)oC)IICDCDZHH-DiCQQ.CDCDCDCDDl0CDIHCDDiCDDiIIH0IICDc-I-0ID)IIc-tQ-IH0c-I-I-DiICDH0LQH-c-I-kD-hH-J-DlH-MDi0c-IIH-H-0CDCDCDC)c-ICDCDc-I-CDCDIC)<CD1<c-I-Ø‘ZIQ.Ic0H-IIDl1C5CDCDQHID)H-C)0CDHCDCDCDCDH,CDCDCDCDHCDH-c-VhDlIICD<IIH,Hc-I-H,0c-I-<H-H-DlII-cDl0—CDIHcICDDlZCDH-DiDl0‘1CDCD1ID)0C)II<H-HDlQH-DiDlDlCDDiHCDCQCD0CDDiH-DlCDMCDHCDDlLQ(QDlCDH-CDCDCDCD-IICD(Q<DiH-<CDC)-ooDlCDCDHCDH-CDCDH-IIDl3CDCDDiH-c-I-DlI-hDlCDc-I-0HCDDiHc-I-CD0CDc-I-c-I-c-I-Q.•0Qc-I-CDCDC)c-VDlHCD•CDI-hCDZc-I-cH-CDDiH-H-c-I-CD1)IIoCDCDICDCDCDQ.“<CD00DiDlc-I-DiHi-ci-c•Dlc-I-CDc-I-CDCD3CDlCDc-I-CD0c-I-CDc-I-0H-c-I-0CDCQCD1(1)c-V0H,0Dl1<CDl-’3<CDC)0i3I-HID)ItC)CD‘tiHc-I-0c-I-DlINDiCl)CDQII-i-i•Hc-I-H-H0H,If-’-Dlc-I-I0CDC)k<i-cCDCDCDH-kDCDCDCDtihc00I5IIDlCDCL)HtiHHCD(QIH-0DiHH-IH-Ic-I-‘TjCD‘<c-ICDH-CDID)i-c0DlC)CDCDICDQ-DititiIICDCDQ.c-ID)•H-lCDCDH-HCDC)c-I-DlC-DiItsCDCDCDH-C)lCDrCDCDDitiCD11c-Ik,CDCiIICDCDDlCD0H-DiCDHCDlCDC-4c-IKJ0DlC)•CDC)l-CDti0CDQiiDiDlc-I-‘—9CDIH-c-I-H-CDCD0QIDiHIc-I-‘tiH-c-IDlDllCDCDDlMH•iDlIDIDic-ICDH><CDc-ICD00c-I-Hc-I-CDc-IIIHCiHI-cCD0CDc-IIICDI‘<HDiH-HtsCDz0CD•c-I-DlDlCD><CDIZCD•Ic-I-CD0HDl‘-<HDlc-I-i-cHI-hDl0CDH-CDIiCD‘<CDCD0‘DlDli-hH-DiHc-I-H-CDb0H,0DiC)DlCDCDIH-I-iDlHC)-9H,-<c-I-CDDi‘IDiDlCDIF-’CD‘Q-CDH,c-I-C)bCDts•DlC)JHCDH-QCDH-CDCDCD3H-lCDCDc-Ic-I-DlIICD•0H-CDID)CDCDDlIIH•H,CDHCDc-I-CDI0Zi-tiHC)F-çtCrpiçt)CoF-CDHQ.CtQ.Lsi301CUCDCH0CDCUi0Q1(1)QCl)H-i-H-0COHCD0CD(QC)iCDCOCDCtH-I—iCflCDiCrCoHH‘tiHCDCDK’iCDCoCrCrCoH-I-tQI-CDC)CDC)CDH‘tiIH-CDH-H—-.çrCCUQ‘-000HCDCr1(1)HC)HC)H‘<HCrCoHDCUUCUHI-iH5k<tiCUi5CUC)CDCr))))COCo0CDH---H-CUCWCD‘t50CDCr1CrQQCUCUCUQ‘HHCr0)0I-Co5H,çtClCl-—H-CDiH,I-jH-CI)CoH-CC)0CHC)C)CoQICiH-HC)H-00)C)H)0ctHCDC)CUCI)CrCo0IH-0CUCo0)C)-’CDCCUC))CUCUCOC))HHCDHCIH-CoH-tYN0I-i-CoH-CoCDh<CDCO‘‘-CD‘-Cr10CDCUHH-Co0CDF-Cr-CDI-iCo><<COH,1WCUCDCDCri-’.C))I-COCD(QCC))CoC))IC))COCOC)CDI-HCCUC)CDHHH‘IZC)fl-QH-ClCDH0HCDH-C))CDCH-0‘-QCDCUH-‘tiC))‘<(NCDQC))0C))CoC))frHCohCoCoCD0—0H-QCtI-CDc-tCrCU05I-CC)CDCDCrHH-C))CDH-CC))COH-C))HfiC))H-‘ti0Cl0)COCD0C)CrCO<HIIC)0QI-H-CDHCOC)CoCDI-H-COCrH-C))CDCDCCD5HCI-1f-tCUCDCoC))CDCrCDC))f1ClI-5HC)CUI-H-C))CD•H0H-hCCOC)C))CDC))H-I-j(I-hIIH-“<CI-CDCOCUC))5CDI-I-I-CDHC))C))5H0)çtCOCoCrCtHCDC))CDH-•COF-C))C))C)COJCDCr(I)CCDCoHHH-C)COCDtCDCDc-rQ1.1.H-CCDH-‘10XCDC))0H5C))HCoCD01COH,IH-I-I-hCDC)0C—CO5CD5C)CDC))CrH-I-CDl--MII‘1CCDHCO000)CDI-CtC)0CDCDCoCC))C)CDCoCDI-CC)‘<CDCDH-55C))C))CDCo-i0c-r‘CrHC))CoI-Cr00CoC)COCDH-IC))CtH-CDCoH---‘-‘10ClCD0C)CDI-h‘tiI-H-ClCo0CDC)C)ClCrCDl-05-CD0CCoCUH-CtHZCoCrC))C))()H-CDCrH0tiI-iCDHCoCD00CD-ClHC))CoCDClQCDHCtCU5CDHCr(QICDHH-CoC)hCD0CDC))CDH-CoCO<H-C)H-CFClCUCo<HC)0II<CUXCUClCrC)IiCDC):‘H-CDC))COCD•-CDCDC)C).)CoCC)),çlCtC)CDC))C)Co(QQC)hCtQCr0HC)CH-CDH-0HHI-ICD0CtHH0H-ClH-CoC<ClH-C))C))C)0CC))C))H-C))CDCr—CCDClCrC))tQC)CtCOCrCrCçrCDCt0Ct‘C)CDC)H-H-CrtiCUCDC))i-aH-CDHCCDH-C))0CDHCDCoH-0‘CIIHC)CDCDCDCOtHH05CD‘1CC))CCoJHCD<0)CMH-C))0H-C)CDCtH-CrCC))CUCDCOCtCDXJ(QJ<<C))CoH,C)HCoCDHH-IIICiCDH-C))00CDC))ICD0CHH-CDCoH-CDI-H-COCDCrC)C)CoIICrI-ICDC))0CrCoCoHHCOCH-CoC))COCOHH,H,C))CDCo•CD00hH-VH-HC))•1DH-COCUI-I-0CUCOI-hN0CDCDCDC))Ct0CUCUC?CDCDI-i<C))I—’.C)CoHCoC)H-H‘rJClH-C))CDZCDC)H-C)]0HCDC))0H-‘tiCrCrCoH,0CC)Cti-HI-hI-i.I-•CUCDtQ0IICDCDCD0CrCDCo0H-SClCDSCtM43over relatively small distances, individuals from the coast ornorthern Cascades are readily distinguished from individualscollected in the Rockies. Hovever, accurate identificationmay be confounded in the Cascades of southern Washington andnorthern Oregon, where intergradation between the two phasesis apparent.In their study of flavonoids, Bohm et al. (1984) observedsimilar trends in North American Menziesia. Both . uilosaand N. ferrupinea had similar arrays of flavonoids, butdiffered in key respects. In particular, N. oilosa had asimpler array of flavonoids since it does not accumulate 7-0-methylated flavonols, triglycosides and gossypetin. It does,however, accumulate dihydromyricetin, which . ferrupinea doesnot, while M. ferrupinea accumulates an unidentified flavanonenot seen in N. oilosa. Less variation in flavonoids wasobserved within and among populations of N. pilosa compared toN. ferruainea, in which complex and highly variable profilesprevented the recognition of coastal and interior phases.It was Peck (1941) who realized that distinct coastal andRocky Mountains phases of Menziesia tended to intergrade inthe Cascades. Accordingly, he recognized a single species, N.ferrupinea, with two varieties, an arrangement maintained inthe Vascular Flora of the Pacific Northwest where the zone ofintergradation was identified as centering around Mt. Adamsand Mt. Hood (Hitchcock et al. 1959). Hickman and Johnson(1969), focusing primarily on pubescence characters and leaftip shape, documented complex patterns of clinal variation,CDHH)CD0H-<H-)C)PCI)SiH-0CDCD)Cl‘OCDI-I-CO(1)COfrCOtiM0COC)f-tC)<C)COH-iLflCflrrCD1C)ZrtH-C)FlCDH-C)CDHCD0CO‘<‘DCDH-I-jH-CD0CS)SiC)0ICOSiH-Fl(QC)C)—CflCrCDCDH-ZCrt-hClf-rSiH-C)ClH0CDCrZ00CD•H-HH-JC)H-CDH-H-CDCOCI)H-COH-ZI-<CD0HFlH-(I)çr0COZ<COCD:iZCDCrColJZH-ZCDCDCDCDC)51C).QCrCOHH-H-CDH-FlH-M-lCDCDCOCDHCOHCDCDCD0ClZCDZFlZSiCDCrH-HCDMFlCOZCDC)H-CDtQ01C1CrCOZCDCDCnI-hZ‘-<51FlCDCOI-CDCrCOHH-CDCOFlH-FlCOCDCDCDC)00CDH-0IH-CDCO0CDFlHC)SiHM51I-hCrICZCOH-CDH-CO0FlCD-rCDIICD0HCD<C)H-ZFlCrFlH-CDH<CDH-HCD01-50CDCrFlC)CDCrCrCDZ(CDZi-3H-HCDH,CDFlFlCDHFl<H-ZCI)CD<MCOjI-CO51COHf-r0rrCDH-CDZCrHFlSiCDClH-0CI)H-CDc-rCDFlCOCD)CDZSiC)<CDCDMCDH-CDSiMCOFlZCDCDClFlCrCD0COCDC)COMHCrCrCrH-•QCDCOHH-CS)CS)SiCtCOClH-CDCDH-CDCDSi0C)<0ZCOCOSiC)H-ZCDC)C)Z0CDCDHCrFlCDHM0Z51Si51CDCO0C)H-Cl0ZFlCl0<CDFlc-rCDClFlCO<51CrH-CDCD0ZH-COCrCOH-H-CCDH-Cl)H-Ht10COH-(QCDC)CtC)CrCO0CONCOZCDH-CrI-iCrCDZC)FlFl0D)CDCDCr-CDCD51C)H‘1CD0ZCDH-ZCDHH-Si(II5))‘<<COSi-DCrCOZFlZCDCOFlCDCDH-CDCrCDCDCDC)UIH-CrCD0CDCQCO0CDCDH-CO515))5))COCOCtCOCDCOo\°ZCDCFCDC)ZSiCrZCD51<COSiCDH-Cr51SiFlZZ5))51H-CDCrCQCDCrCD0H-‘-Q0CDCD00ZH-NCrH-COFl—MCOCDZCr1-5FlCOMSiCOH-CrCDZHH-ZH-H0)H-CDCDCOZCOCDCi)CO0CD.DFlCD51ZCDCrZCrQH-ç-rCDCOH-FlCDç-t005))CO3<CDZH-ZCDCrH0DCDCD51CDHH-M(-FCDC)C)CDi)COCDCDCDZZCDFl—51FlhZCDCD510HFlCDH—Cl51CDCr51CO0ClC)ClCDNZCD0H-CO<FlCDSiCDH-M5))0H-COCO1-35))COCrH-çrCDCOH-ClCrCDCOCtJH-ZC)ZC)CrH-C)H-I-j51<-5<H-COHCDCO0CDC)H0<FlCDCD0ClFl05))CDH-SiI-hCDCOSic-rCDSiCDH-CrCDSiZI-ZCDH-COCOCOFlCD00CDCO5))<SIH-ZCDCDCO51ZCD0C)CrH-<HCrM•CrHH-c-rH-COCrH-Crl-5CiCD‘-3CiCDCDC)CD0CDCOC)iHCDc-r511-55))1-5Cr1-551HCD(Q0Cr‘<H-CO0CDCDH-ZCDZCOCDI-hFlC)0ZCDCT)Z0CrCDCOCDCDCOH-1-5•I-c,5I0ClCOCrCOtSFlZZClCDC)Z00CDC)CrCrH-COCrFlH-Z0CDCDCDCFFlClFlCO<H-H-FlCDZ-3F-ZCOFlClH-H-C)0CDH-tQZ5))IC‘-<NCSCD‘tS<CrCDFlCDClZZCOCOH-I-’-ZHCrH-JHH-CrSiCDCrHH-FlH-Cr51051CDCDCDbZCDC)00COCDCOFlSiCrCDH-HHCD5))CiH-,H-COCD0HZCOCDH‘CIHI-j•lCOCOJH-C)5(QC))=H-C)CrI—irtCI-iCOH-COC)ZC))I-iC))C)‘1I-’50‘.0CDC))C3‘H0<C))0hhC))CDI-I0QCDC))O0C)C)C)M-CDCDCD‘CDU)‘IH-CU)CD-c‘C))HhC)C))CrCD0C))tQCD(1)CrQ.H—‘tiCDC))H-IIC))U)COCOCDCr0COC)hH-00COC)C))-cU)CDHC)CDtiU)C)‘<H<H-C)rrC)CDc-r(I)‘tiU)<0‘‘HI-C))QCDCD0HC)CrCrH-CDU)C)CDS5CrCD‘1C)H-HC))tqH-0CtCDCDQ.‘1CD0C)H-C))‘CDHCD0H-C)QCO‘<rrhCDHCDCDU)CD).QH-C))I-j0C))M0CD-Q.“<HH-Q.hH-CCrQ.H-‘tit-CrH-)—COC))ç-tH-I-hCDC)H-H-CtH-CDC))QH-H-CrC)Cr0I-‘.0CrU)HCDC)CrC)U)QC)j<U)H-0C))CDC)o0CC))C)CDHH-CDCrCD-tCD<I-C)i0CrCtH-0CrH-CrHH-C))CHQ0Q—Q.0Q.I-h—C))H-‘-<‘-CD‘-<CDU)•H-I-’U)U)H-CDCtHI-h><CDC)—r0C)C))H-QH—tiCOØ.•H-I-‘.0I-I-C))U)CrCrU)C))CC))‘<CDCrH-COCO0C))aC)CDCrN‘tiC))H-CtCC))C)(I)H-CrHCO<COIH-I-0U)H-CO1HH-ZH-CC))Cl)CDCD—tiCl)H-H-ti•CDC))I-i<0CDCD0ciCDC)c-rCCr‘-QCDC)-tU)CDU)CO0CD0I-cI-cI-hU)U)-tCDCr00H-CC)HC)ICH-CDIIM,0COI—s—U)QCOH-U)H-H-IH-C))COMiçtC)WCDCOCr‘tiCCOCDCDI-h0CDIH—NC))CrH-0CDCOtriiC)C0CDCOCC)ciH-5CrCOIC0CrU)CII‘H-CDH-C)CD(QC))HC)0IICOCr1CDCciCDCDCrC)CD‘-QU)H-tQU)Cr‘-IIC))CDCDH-H-C)C))0COII‘CICD0CDC)c-r<CDC))CDciQ.U)CrI-IH-cT-Q0CDU)MiCOC))CDCDC))MiI-Mi<I-I0CDC))tiC)CDI-hiI-CC)CrQI-I-ç-rCDi-cCoIICDI-hCrCD><HH-‘CIH-CD‘CIH-I-CD(CHH-IIC))CC))C))‘CIC)•C)C)CC)i-cN‘CICOCDCDC))CrU)I-CDC))C))frCO5i-ciCOC))H-•C)CDC))I-U)i-c•C))<CrCCrCCDH<‘CIMiCDCrI-jMiCICD=CDCDCDH-CDC)ç-tHSCrI-CDU)C))C))CDHU)H-CUH0H-H-Q.3<tiCDI--‘ICDCD-rI-’-CC)CrCr--CDCDC))0U)I-CrCrCDH-COC))H-0CDH-CtI-hCD00CrH-CDH0MiCO0C))CI•I-j1<QMCrC)CDiH-I-hCriCCD‘CICD‘<:CrCrH-I-0C))CDciH0CD<0COH-CrC)‘CICO•CDCDCOCOCrC))CDCUCDMiCDCUCC)CO0CUH-CrCDciCDC)S•COXIIHCCrCQH5H-H-C))0C))HCDCDU)<‘CI‘IU)C))H-H-Q.I-‘.-J-C))H-HiIIH-HCC))CDHCU)CDCDCCDH-C)HH-0CDCD5CIHCIl-C)0t3Qt-tc-i-CrCUC))I-I-II-COH-H-C))ICOH-H-I-IU)Ct)‘<CD•CrCH-H-CDC))CDCDNCrCDU)CO‘CICDMiC))ti‘-3ciH-U)0CUH0•CD0CDH-CDCDf-rH-CDC))i-IMiCDC-?CD<COSciCU)(QC)H-C)0C)“<t5‘CICDH-CO0CDCD•CDU)i-H-CDMiH-HCD‘1C)0CD00H‘-‘—3c-rCD0CDCDU)-rCD0c-rCDCI-hI-h0-C))HC)0MiCQU)C))CrU)IU)tQCDCDC))H0C))=>‘JCDCCrH-CDCtQCrC))C)I-I,CrCDC))CDH--CD5H-CDCDC)CDZCr•CrC))CD—HiC))NI-cC)0C))U)CDCr-CCOciC)CDiO‘-0CICD—C))Cr01CDC))C0UiCrHciCH-CCDCDH-C))COHCDI-o‘-<CrCOU)CrI-‘-<U)HciCDci-CDH-CDCDCDCOCDCDI-iI—cI--cU,46rationale of Calder and Taylor for recognizing subspecies in. ferruQinea is sound and more in keeping with moderntaxonomic practices of recognizing infraspecific variation(Stace 1986)Because this is the first study to examine morphologicalvariation in detail throughout the range of Menziesia in NorthAmerica, a synoptic treatment is presented below.2.4.2 Key, Descriptions, and Nomenclature of North AmericanMenz ies iaKEY TO NORTH AMERICAN MENZIESIA Smith (ERICACEAE)Erect or straggling shrubs to 4 m tall, young branches finelypuberulent and glandular-pilose, becoming less pubescent orglabrous with age, with older bark shredding in longitudinalstrips. Leaves alternate, often in compact whorl-likeclusters; thin and deciduous, glaucous or pale green below;pilose or glandular-pilose above, glandular-pubescent orpilose to densely puberulent below, with subulate hairs alongthe midrib; margins ciliate, minutely crenulate-serrulate;ovate to elliptic to oblanceolate or obovate, 3-8 cm long,1.0-3.0 cm wide, tips acute and apiculate to rounded andslightly mucronate, short petioles to 1 cm, leaf bases acuteto cuneate. Inflorescence a 2-8 flowered terminal corymb orumbel (ours) or abbreviated raceme, nodding, on shoots of theprevious year, appearing with the leaves; pedicels glandularpubescent to pilose or puberulent, 10-40 nim long, subtended bydeciduous membranous bracts; perianth 4 (5) merous, calyxF-CDr1CDU))frh()C)U)QCl)I-QQHHf-I0H•0CDH-c1Cl)CDCoZC)ctQr-CDClCDC)J‘tSci-H-ci-HHCDci-I-CD5CtHClCDF-Cl)CDCl)ci-L-CD—-H-hh•U)Cl)H-Cl)HCDrtHU)iH50C?I-ICD<C)-U)Cl0<CD(1)0CDt-•HO(Q‘ClCDClU)CflCDCDClH’U)5•CD<HsiCl)CD‘-IC)U)U)COtiHci-CDJ-c-rci•Cl)Cl)tiI-ICl)HClCDtiCDc-tCl)H-HCDti•Cri-iCDClH(QCDCDH-I-ICDHC)U)H0H•H-ClU)CD1-10r-rHC)ClC)ICltYHci•CDC)CQCl)0ICl)H-Cl)H-C)I-CDCl)LQU)CD•U)HCDI-II-I‘titQCDCl0cCLH-f-I()Cl)CDci-CClU)SCltiI-iCDCDClQClMH•0(CI-ICrCDH-50H-HCD0ClCDCDH,H•Cl)HCrCDH5HC!)HH-I-I-•ClCDQ•U)Cl)ZCl)ClCDCl)t’JC)HU)U)H•CrCl)Cr5C)I-ICrI-ICr0CDCl)C)COCDc-tU)•Cl)I-IH-CD0Cl)CDCDH-H-IH-CD<0CDHrtf-ICl-Cr•H0I-i<C)hi5ti0c-rCt0Cl)C)H•U)CDH-CtCDCl)H05CDti•U)f-IHCrC)-COCl0<H<(QCD0H-CDC)•VCDCl)CD0ClI-ICl-0CDHCrhCt.ci-WCDU)C)tiCO‘-ICrCrt-Cl)Cl)Ci)•Cr0XICl)0Cl)C)Cr0ClCDCDCr•U)•0tiStiCDCDClI-ICrCDtiCl)CD•(CCrCDI-IH-CDCOCD5CDCl)-5H-Cr•HI-ICDC)•CriI-CrCDHCOH-Cl)—CrCDCl)••Cl)CDCD•H-HI-HCDtiCOi00H•HH•SCDHCl)CD0H-HCDCrCl)CtCl)IC)CltiCDCl)•CDU)CDf-I‘ti‘<-I‘<HCtf-I—-.Cl)I-I0—•H-Cl)Cr•U)ClHU)U)H-H-HHCrtiCD•-CrC)-0‘ti0CD‘tiCDH-Cl00•Cl0Cl)HI-I0-CDI-I0CDC)f-IU)Cr•CD<tiH-tiCltC!)f-IH0C!)‘-It-jhiCl)ICDCDCDH-•ClCDH0-CDHCDH-H-H0U)I-IU)tiH-••C!)-IH0—JH-f-iCDU)Cl)HC)0C!)•CDH-0Cl)WCDC)CrHCl)Cl)IHCDCDC))CtIH-Cl—Cl)HCD(QCl)H-ClCDH•HCl)CD.DCDUiCDU)hCiICl)CDU)I-Cl)—Cl)H-Q—f-ClC)5CQHCDU)C)Cl)HIl-i,tici-0CDCl)0I-ICl)MCD0I-ItiHCr<C-jCrCD‘tiCrCl-tiCDCDQ.C’)f-ICrClClI-ICDCl)CDCl)HH‘ti‘t3CrCl)0C)H-U)f-Iti-.C)Hf-IU)CiMCl)H-HCl)CD•f-ICl)CQH-0Cl)•C)Cl)‘-IClS‘tiCDtiSCl)C)Hsif-If-hCl)0Cl)CD0HCDHHU)Cl)•tiCDCrClH-H0H-(QCrHtiQ><H•HH0WCDU)H0f-ICDH•JCDU)-H-CrClClClIIU))5çtU)C)h-i,C)<Cl)ICDC!)CDCl)CDCDH-lCDCDHQtiCl)CrCD(HCQCl)FCr f-ICDHH-•CtU)Clc-rCDHCr-•CDH-Hf-I0U)HCl)H-0ClU)Cl)CDClCrCrHCr0HU)tYHU)U)I-h5CDCl)U)i0ClU)Cl)f-I<CrH-Cl)CrH••tI-ICDU)Hf-Ic’-)CD0H•Hci-CDCDHHU)Cl)IIHSiCDH0CDf-I0Cl)CrClU)5HCl)Sif-I—CDCl)<o-.•II•wjII- CD C)H,C)rt 0H‘10 ZCoH1Qf-r—cliHtQ•tQCoIs)HCoIH-‘tSr’-) U]H,CoCDC)I-[1H-HQc-vDOgCD çt H-0 HI:—’CDCDCocli <tOCDICooCD H H—.H It’ftftH-HItiC)Cl)ftcli0C)0fttDCD0-cli ft CDH 0 H II H- Co 0 0 I-1 w 0 H- H H H wCl)HC)IHt5cli01-hN-0H-CDCDtO<CD0D0CDIIH-tQHCDCo$iQHclicli<HHhcli11<•.CocliCl)CDH-CDcliCoQQIt’1<ftt’Q.cliCoCD<H-CDCD—I‘-<CDIl-hft•CDClcliftt’H()cliH•ICokD•H0Cl•I-IC)Col-CDHl-<•.It-cli<Cl)•CDC)cliCl)H-I-00CDIiCDIhC)IICDCoHCDCoCo—C)NH,‘t5tit’CD•cli0CDCoft•CDH-cliIH-ItiCoCDClH•ftHtI0l-CDH-0CDH1icliHCD<•CD3CDo-rIicl)-.Co•H--C)‘<CoHH-cl)HCoColIDHH0-Cot’‘-QcliH-CDidcliCDCo1t’-•CDC)Qt’i-tQl-cliC)Co-CDH-•l-I-jcliHCDCDHftH-0Cl)HH•clitidlHCDcliH-CoI-’H--<0•-HCo0tiH-cli0H-Cocli0CD-.QCo(I)fti<ft1t’CD•ftcliC)H-dl0Cl)cli•H-ftCDHHHC)••II•0QH-CocliCDHcli0H-H-C)CDCDCD1cliI-t-CoCoCD0II(QCoftdlC)ICDt’cliH•CocliH••-HJCDQ.1t’•N•—IIHcli•H-0dl1t’-cliCoCD•ftCodiHftHt’0HHClCDCl•‘-<CDtCk<I-cliCDCDC)CDft••CoCl)H-ICoclicli•-ttQHICoIC)ClI-j0CoHcli•Cl)C)ICo(QA1t’CDCDcliHCDclit’ticliC)0HCoQ.H,HiH-‘I-h•CoditO•yftQ.H0lCDC)ftCDZCD-ft•0CD0ClQhIhIMcli0ftQWC)CoH•C)(QHCoIhIll)H-)<cliC)Cocli0-HcliCD0IliHCDHCoCDHCldlIHIH--CoI-icliftCQcli••ClC!)cli<hi0CDCl3C)Coft•dl-CoWCDdlftC)H-H-I-hCoCo1t’CDcliftI0‘ttiCoiHftCoC)-Ht5ClCoU]•cliHCDCoclit’H--ClI<CDt’HG’i•hH(1-H-Cocli•ftIICl)HH-Co0Cl1t’CDCoC)CoCDH-Coft•ftIt’‘t’HcliCD-•CDH-ft-Coft(QCDCDCDHHt’IIHICDC)H0HC!)0H••cliI<IQclicliftH-CDCoQCDcliHIHtiCDdiC)HCDHIt’IjH-H(11111t’Cl•CocliH-CD<ft‘H-hHk3cli0IIHcliHt-IH1t’CoclilIDIIC)QCo(0t’loIHCDiH-•0Cot’0ClIHftidIIIdftCDCok<•49usually prominantly apiculate; glandular-pilose above;glandular-pubescent, occasionally glabrous below, with few tomany (0-25) fine subulate hairs on midrib; dark green or blue-green above, paler below. Flowers 3-7, on pedicels withconspicuously long glandular hairs greater than 3x the pilosehairs, 15-40 mm long, calyx disciform, glandular-ciliate;corolla greenish-white to cream to salmon pink, 6-8 rrun, 4-lobed, urceolate to urceolate-campanulate. Capsules ovoid,(5) 6-8 (9) mm long, glabrous to glandular-pubescent. Moistmesic woods, open montane or subalpine forests. Alaska southalong the coast to N. California; Cascades of BC to southernWA. Fl.: April-Aug; Frt.: July-Nov.. ferruainea ssp. alabella (Gray) Calder and Taylor, Canad.J. Bot. 43: 1387—1400. (1956).Synonyms: . alabella Gray, Syn. Fl. N. Am. 2(1):39 (1878).. ferruainea var. alabella (Gray) Peck, Manualof Higher Plants of Oregon: 542 (1941).Short compact shrub to 1.5 m. Leaves obovate to oblanceolate,3-8 cm long, 1.5-3.0 cm wide, petioles 3-9 mm long; tips morerounded than acute, only slightly mucronate; glandularpubescent and only slightly pilose above; glandular pubescentand typically densely puberulent below, though only sparselyso in the southern Cascades, with few (0-10) fine subulatehairs along midrib; yellow-green above, pale to glaucousbelow. Pedicels slightly to moderately glandular and lessthan 2-3x the length of pilose hairs; calyx disciform,50glandular-ciliate and often puberulent. Capsules small, 3.5-7mm long, glandular-pubescent and slightly to denselypuberulent. Open montane to subalpine woods; Rockies ofsoutheast BC and southern Alberta, south to eastern WA and OR,ID, MT, and northern WY, as well as Cascades of WA (SkarnaniaCo.) and OR, where intergradation with ssp. ferrupinea isapparent. Fl.: late May-July; Frt.: Aug.—Oct.M. pilosa (M±chx.) Juss., Ann. Mus. Paris 1: 56 (1802).Common Names: minnie-bush.Synonoyms: N. Smithii Michx., Fl. Bor. Am. 1: 235 (1803).N. alobularis Salisb., Parad. Lond. t. 44 (1806).Shrub to 2 (3) m tall. Leaves elliptic to oblanceolate, 2.0-5.5 cm long, 1.0-2.5 cm wide, petioles 2-9 mm long; tips acuteto rounded, prominantly mucronate; densely pilose above andeglandular to rarely slightly glandular; densely puberulentand usually eglandular below, with 10-25 coarse glandularsubulate hairs on the midrib. Flowers 3-7, on glandularpilose pedicels 10-25 mm long; calyx disciform, ciliate;corolla greenish-white suffused with pink near the tips,urceolate 6-10 mm long, 4 lobed; capsules ellipsoid to ovoid,(3) 4-6 (7) mm long; densely glandular, never puberulent.Moist, open montane oak woods to high elevation heath balds;southern PA and western MD, south through WV, VA, and NC toeastern TN and northern GA. Fl. May-July; Frt. Aug-Oct.512.4.3 Factors Influencing Morphological Variation in NorthAmerican Menziesia with Reference to the Japanese SpeciesHickman and Johnson (1969), hypothesized that patterns ofmorphological variation in western North Zmerican Menziesiawere explainable in terms of migrational history, past andpresent patterns of gene flow, and adaptations to existingenvironments. Although they did not examine these factors indetail, they discussed the effects of glaciation on westernMenziesia. In both western and eastern North Zmerica,morphological variation is greater within populations thanamong populations. This pattern shows significant correlationwith geography on a regional but not on a local scale, asindicated by the Mantel test results. The apportionment ofvariation is consistent with a sexual diploid xenogamousbreeding strategy (Grant 1981). Chromosome counts forMenziesia are in fact diploid with 2n = 26 (Taylor andBrockman 1966; Packer 1983).Along the coast, N. ferruainea occurs in a variety ofhabitats, including sea level Picea sitchensis (Bong.) Carr.forests from Alaska to Oregon, and montane woods with Thuiaplicata Donn. and Tsuaa heterophvlla, or Tsuaa mertensiana(Bong.) Carr. at higher elevations. Unique among thesehabitats are the coastal redwood (Sequoia semoervirens)forests of northern California, where Menziesia grows eitheron fallen logs or as an epiphyte in the lower branches oftrees. In the Cascades and Rocky Mountains, Menziesiatypically grows in open subalpine forests, usually along theCQHP1‘iCDCDCDOP1(t<::-NCrCDCD0H3ftCDC)Cr‘-<Cl)H-CDP1P1HCl)hP1CDHP1P1H1C)CDHCDP10P1oci5Cr5CroCD0()Cl)l:3øi0CD CDCrH-<P1‘10CD<‘1CDCD(CQ.Cl)Q-P1CDCDHCrHCDH-Cl)<P1CDCDCDCrH-HCrCl)0P1CDtQHClIHJHCrHOH-Cl)CrCD(Q‘ZiCD‘1P1‘1P1CrP1() C)(I)P1H$HH-P1H-5CDH<P1H-LQHP1?J’rfP1CrHP1CrH-CI)CrCDoCDCDHHH-CDP1CrCQCDt-CDCD(QSçtICD•CDH-CDCrP1Cr5P11:-I0CrCDHiCDXk<p1 ‘-r-rI-5CQCDH-P1CDP1<0 I-hCrCl)CDH-CDSMiHP1P1Ci)0<HiCl)<HC)H-CDCDCrCl)I-I-<zP105(Q00P1IIH-HCDCr0CDZ•P1‘tICrCl)Q CDCDH-‘t3P1CD(Q‘tC)Cl)Q.CDCDIIQCrCDCDHCD‘tiCDH-HI-hC)P1CDP1C)HC)P1CrCl)CDCrCl)Cl)H-O0P1Hi0 H-,CDMi<H‘1CD0CDHP15CrH-<CrCQCDCDZ.CDH-IICr •SQ-P1P1•P1CDP1HCrHiCD<Cl)C)0I-CDCl)CrP1P10‘t3C)Cl)OCrCDCl)Q.H-CDP1CtcLCDCDP1Cr P1OP1 CrCrCDCDC/) OCl)OCD‘‘t5CDHO<HH-0O (0CDH-fr’C)5P1CDCrHCrCD<P1P1H‘-QH-H-,0P1P1 C)H-H-CrH0(QH-I-CtCl)CD CD P1OCl)CDMiI-QCD:iHi0P1N H-CDQ.Cl) H-CrP1P10<•CDP1p1<C)CrCDHP1H-H-C)CDCDCDMi0I-H-,H-CDHC)Cl)CrOHCrCI)‘CDCDCDHCD<C)CDC)CDCDP1P1<H5H5P1CDCDCD0CDCr•HCl)CD50H-5‘-03P1I-Cl)Ot3HCDCDCDH5Cl)CDI-‘‘CiCDCDMiH-HP1HP1CrC)C)0Cl)DCDCDCrP1CDCDCDCl)CD ><(Q0Z‘CiCD‘CiCDCDCl)HI-P1CrCDH-5CDP1CrCDCDCl)<CDCDSC)C)P1CDCiCD‘CiHQCDCD‘CiHIIC)CDCDH-P1CDCDI-‘1P1CLCl)Cl)P1Cr‘CiH-H-Crfrj5IlCLCDP1CDH-MiCrC)HP15CDH-CrP1C)00HI-Cl)‘1k<HCl)CDP1Cl)HCDP1C)‘CiCrXH-0H-0‘CiCLS0C)‘CiCDCrCl)h5P1CDCD0C)NCDClP1I-Cr0CDCDC)CrCDCD00CL(QHHCLP1P1HCl)CDH-3HH-I-iQCLCLCD—z-C)HHHCD5HCDP1CD0H-P1hP1Q<CDCD0CDCrCDCl)‘CiCiP1tlP111CDCDHP1CDH-H-0CrCLP1HCDCQCLP1‘-QCl)CL0QiCDCrCl)C)Cl)CDHCLH-•H-I-CL•Cl)0C)P10CDMiH-CrCDIhCrC)CDHCl)HCrhi<0CDCDH-CrIH<00P1CDP1‘-C)Cl)CDH-Crhi‘<o,QCrH-IF-’-5H0kCl)‘CiCD0CDCDP100Cl)P1h--C)CQCrCr•H-C)Cl)HI’iCi’Cl)P1HP1‘<CD0Cl)5Cl)IIR3H-3H-CrkD0CrCD13HCDF-CrKD0P1H-CDH-IP1‘CiHNC)CDI‘CiHP1H-P1P1<P1C)IH-CLHCLCiCDH-lCD0P1H-H-H-0(QlCDSC)CrCrH-P1Cl)CDlI-hH-H-H-CrCDCDIIIP1P1Cl)CD>CDCrlC)1CrlCDCDMiP1H-CDlCDCLl0HICi’HI-•CDH-CiCDH-CrCl)<0ICiCLHC)CDP1—0H-00H-Cr‘tjQ.,H0H-CCD0H-P1HH-0IICl)CrII-‘Cl)P1P1CDCDtY’HII-—0P1CD‘-<C)H<C)I‘ti‘-QP1CD0CD•00H><H-CrCl)0P10‘CiICi‘(Q0IH-•CDCDC)CDH-Cl)IH0C)CD10Cl)HiIP1CllCDCrH0Cr‘CiHlP1P10H:iCDI—iCrCDCLCLCrCrCl)P1CDP1CrP1UiP1CDCrQjDlCoH-DiQjCDU)1H-I-jrtDlDlH-I-IçtC)H-CDDiCDU)HCDU)C))F-DiH-DiU)DlU)0CDDiOHHDiHH-CDH,DiU)C)friCDDiDiI-IU)DiHHH-CrCrDiCDH-I-ICDU)DiHDiH0‘1CDH-DiCDHU)HH<DlU)ftftH0CDH-DiCDCD<iCDCDQ.HU)CrHC)‘<0CD0 HC)0CDrrHHH-CD0:ç1DlC)U)H-CDiCrCDU)CDHQI-Ia-D)0Ui0U)><—I-IH-•ç1H,DlHlCr0H-CDc-rU)U)00CrzDl0DlU)I-iHCr01t3H,Dl-rCDHCD0QCrIcsCri—ii-ilCDDlDlIDlijH•ICrCDCDDlI-Itcsi-i0DlDl0I-I;sI-IU)CrU)DlU)U)DlH-H-DlU)()DlCDU)DlU) CrDlCL)CD0‘-1H-U)I-iH-U)CDU)U).oCDU)U)H-CrCDH-cscic4‘tiH-H-DliDlCDU)csCDI-IU)çtDl‘CDH-CDCrIC)DiCiHçrZCD‘<ciDlrrDi0zHci0‘<‘-10I-ICl)NrrH-ciDlH-ciC)CDCDCrCrU)H-U)tcsCDCD0H‘i‘-IZU)CDH-QDlC)CDHciDlI0U)CrH,H,CDCr0I-1I-I‘-10CDCDS0I-ICDCD 0 H,CrH,H,CDDlçrC)I-iI-Ci--Dl 5CrCDHt-DlZoçtCD DlU)çr-J•DlCDCr<ciDiI‘.<I-I CDDiId H‘csH-CDU)0Di-J-U)11DiCtC)Dioc-rI-i CDcsU)Cl)td CDCD,QC) H-H-CDCrU)C)CD0ciU)H-r)’CDU)HHCrCr‘-<H-0‘—1C)CDCr(QI- DlHiCrI-ICrCD0ciSCDciCrCDDiU)CDI-IdCDH Cr CD Cr CDC)Cr5050CrDiU)50CDJ0H-0H-0H-CD5CD‘-IU)i-i<I-IdNCrCr‘dCDDiCDH-DiH-H-0C)DiU)C)‘-ICDZ0I-IH-0C)ZH-Di0H-U)H,H<U)HCr0CDZ(CU)H-HO0<HCD0DiU)(QCDCDCDCrCrCD5C)H-Crfri<CI)DiDiC)U)CrCDI-XCD0C)U)HDiH-H,HCDDilH,CDCrCDHtQCDU)Di-HH0U)EDiCDCrU)d<H-HCD0I-jH-F-CDDiH,<U)H,(QDiCrHC)CDDifriH-U)Di1HU)U)CD5H-C)H,:5CDc-rDiCDCDXCDDiDi0lHIciCiDiHc-rI-ICDCDC)5•CD0H,5<H-CrC)DiCQDiCrH-C)Cr<H-U)CDI-IC)0HfriH-‘d0‘<CD><DlCrCD0CDCDI-IH-HCr50DiciDi(Q0ç-rDiCrDiI-II-I•H-C)DiU)‘dDidU)I-Ii0H5Cr0CriiH-CrH-CDHSCiCD5CD(QDi00‘-1U)U)DiCDI-ICI)HDiCrDiI-iDitiCD‘<C)50HDi‘tiCDDiHtQC)HDi5H,C)CrCrDitQI-I0Cri-iCDH-friH-CDDiCiCDDifri‘-IH,CDfriC)0CDci<DlH-0tIiCDDiSU)CDU)HCDC)C)CrH,—Cr0‘<HiCr0CDCD01CDH-(Q-DlU)055DiDiH,H-DiCrC)H-CrfriCrtiQCrC)0H-0U)0CDDi•CDDiDlCr0dI-I‘lCD0i-iHU)DiDlCD—DlCDCDHCrU)N)—C)fritfrI0•CrH-i-OCr‘<CDfriCDHCiU)DiH,CrCDCDCDCDU)—CDCDH-DiU)HH-friU)‘-QC)CDCi-CrDiH-ciU)-CI)CrCDH-0CiU)CDCrcn0DiCDfrifri<DifriCDCDCDCrCDDiDiU)Cr0CiCiCDC)JHCiU)Cr0CD0H-CDCi0CDCi‘tiDlHI0H,5(CD‘ti11H-H-CD0HlCDtiCrNi0DiDiCDCDCDU)CrH,C)CrDlfrCiCrH-H-U)I-I•0DiH,HtQUi54Menziesia ferrupinea and N. oentandra often have beentermed vicariant species in phytogeographic comparisons of theeastern Asian and North American floras (Mizushima 1972;Kruckeberg 1983). Based on the results of the cladisticanalysis, this supposition seems reasonable. Theirrelationship suggests that an ancestral taxon, most like N.IDentandra, migrated to North America from Japan via Beringia.The morphological similarity between N. ferruainea and N.pilosa, and the fact that both species are tetramerous,suggests that they were derived from a common ancestor whichmade its way across North America. The transition from apentamerous ancestor to a widespread tetramerous taxon mayhave occurred in Japan, but more likely evolved in Beringia ornorthwestern North America since N. oentandra appears to bethe only closely related Japanese species. These eventsprobably took place during the Eocene or early to mid-Miocenewhen the Beringia link existed and the floras of eastern Asiaand North America were contiguous (Tiffney 1985a).Uplifting of the Rockies and the increasing aridity ofthe midcontinental plains, which started in the Oligocene andbecame pronounced in the Miocene, divided the temperatemesophytic flora of North America into eastern and westerncomponents (Daubenmire 1978). In this regard, it is worthnoting that there have been reports of Menziesia fromMinnesota, several hundred kilometres distant from the nearestpopulations in either the Rockies or the Appalachians (Gleason1952; Rosendahl 1963). The material, from the Duluth area,D))C)CD‘-QCD‘tii-ZP1Cl)ICDCDCDIIHHHI-0F-HCD0H-H-CDCDCDHH-CD0C)HCD0C))C))CDI-CDQCl)C)Cl)F-C/)CDC)‘tiC5C)Cr‘H-U)Cl)C))QI-C)H-CDCi)0H-NCDCDCrCD1H-CD0CtHU)H-0H-U)H-U)HH-I-iC/)CDHQI-I-CrC))‘-OHC))H-Cr-tC))C))CDH-CDCrU)C))CDCrCL)U)CDODf-rHH-HCQ0HHI-iCl)CDItCDCrCDC!)CDC))W0CDIC)<CDC))H-H-H-CDMI-C))H-CrCrCD0I-CDCDC))CD-3CrCDc-r0CDC))Hc-i-0C))H-0QCrCDc-rQ.CriCDaH-Q.l-I-’H<H-,U)CrIDCD<H-CDI-<H-C))CDH-H-CrCDCDCDCDU)CDHC))0CD0IX0Cr<><C))H-H-HI—h0H-‘<‘H•1hCDCD‘<U)‘<CDC)CrCDCrU)C)C)CDQ.F-CrX0H-Cr-CDCr0CDCrCl)0U)CDCr113CrCD0NH-U)CDl-NCDI-CrH,CrCrH-DC!)CrtiH-CDH-00H-01CDCrU)CDCD0IiCDc-tHF-CtQCDI-i,CDQ0C))ICrCDCDCD•C))C!)H,CrCD0C))C))Cl)CrIIH-H-HCrCDC)NH-<C))CrCD-U)<CD‘1U)CrQ.I-hCDCrH-C))CDII0CrCrCDCrc-rCrNCDti‘-<Q0CD0QCrH-I-iCDH-hH-H-I-Z‘130CDI-CrH-0MH-IXCDCD-0Cl)C)C))0C))H,C!)CDIIC)H-hH,CD‘ii13H,C))0CD‘-‘-C))c-rC!)H-CD0H-0H-C)1iC))H-QQCDC!)CrC))CrCrCrC!)C))CrC))C)0CDIIWCDCDCrc-tCrC)C!)QCD<U)CDhC)C/)IQCDCDH-CDU)0C!)CDCDHCDC))CDc-tF-H-H-0CDCl)I-0C)Cr‘<H-CDt-13U)tQl-HCt0U)‘13C)HHCDCr13CDF-CDHCDH-U)H,CDCDC))‘-0Q.H-CDCDC))HCDC))Q‘-0HU)CrH-C)‘1310U)CDC)CD0HC))CDC))I-HCDC)CDCl(C)H-C-tH-H,C)tH-,H<U)CrH-C))(-tH-tCDCDHCrH-CDC)C))—C))H-H-CD0CDC))‘-<—CDC)CDCl)C))C)Cl<CD0CrCDCri-H-c-t•IIC)C))C)H-CDCD•0CDCDX-U)13C))HCDC))CDHCD13H-HU)C))U)H-CDCrCDCDC))I-jCrQHU)HC))U)U)U)HC))U)CDC))HCDc-r‘13HC)CrCDCrCt0CrClCDCrC))0130l-C)JC))H-0b•H-H-,C))I-iCDCrHCr(U0CDCDI-<H-H-0H-H-CDHI-<ClCDC)H-CDC))H-I-U)CDCDC))CDCDU)U)IXH-CDICDU)H-CrQCC))Cl‘13•C))Q.H-(C)H,I-C!)iU)(C)ClCDC))C))ClCrC)0(C)HCDHCDH-0H-HH-Q.I-C)H-0I-hIIClC))C))0Cl0C)CDU)0U)CrH-C))CrQ.H-c-I-I-ZCD‘13CrU)XC)‘1-CDI-I-hCDCDC)H-13H-CDU)hH-H-H-CDC))l-13ClH-0‘-<C))CC)0H-IIC)CDC))ZCDH-•Cr00CrlCDCD:ii-(C)I-hU)HCrHç-t—I-U)CDCnCDI-CrCD‘13I-c-i-0CD‘13CDH-•U)H-CrC!)U)CD(I)ClCrCDCDH-CC)IIH0U)hiClC)HCrH-U)HCDCrCDCDU)IH-H-H-H-U)0IH-CD0H-,lCr—H-C!)CDCDCrCDC)00U)CrjC)U)CDH‘—CDC)‘-UCrC!)0Cr•CDCDI-LC)H-C))lCD‘13HCrCD0HU)CD0CD0Cl‘-HCDlCDl-ClU)‘13‘-D[-U)H-H-CrHCDCDCDH-c-r(3-,CrCrU)Crc-r00CDCDZCDCDHWCDI-ICDH-00IICl•ClZCDC))1<C))I-<MCrIXCrCrCrHCDCDCrH-CD•U)CD-CrH-Cr0 I-hCDU,H-CrHH,I-C))CrCD3COQHtQC)0‘CDC0HCC))0HC))(1)C))H-C))CDhQHhCrClCDH-II1C))I-CC))<C))CO<COCr‘tSHCr<H-t-HCONCrC)H-CtC)Cl)‘iCtCDH-CDC)ClCCCrH-C)H-C))c-t-tH-HCDCDCDHClCDC))C))HCDICDC))C))CDC))C)C‘C)0Q3C)(I)HH-CrCOC))ClCOHCOri-COCrC)CrC)ciC)0Cl<—C)C))C)H-CdCOH,H-CDCrCrç-tH0CDCrC)(I)C)<H-H-C)t-CDCI-cCDCDCDClHC)COC)CrC))CrClCDC)Cr0CDciH,HCDClCClciCOCDCrCDHC))ClI-C)CClHJCr-C))COCDCOC))HHH-CDI-cH-H-C)COCrC))C))H-C)CrCOt-HtJHCC)C))CCDCOCOHC))CO<H-CCDI-jHOC)HI-hCrC))CDC))I-hCrCOCDCr•C))CciCOCQH-ClHC))CDC))CD1CDH-CC))t-<ClCDC))CrIIciCDIQt.QCDClC))CtC))COI-NHCDH-COI-hI-hC)HH-C))CDIHI-CDQC))HI-C))H-•C))ClCHciCrC)CDCCOIC))CDCOIICCDCHCOCDCtQCDCriiH,HbIkI-3H-HCrC))I-HClCDC)-ClCDH,CCDC)C))C))C)H-LCOC)<C))C))C))C))CDCDIH0H,H-C))0c-r0CDciCDC))COCDI-IHhNHC)CrHC)HC)C)C))—COC)CCtCrIC))H-CCOCD<C)CDH-CrCHCCC))‘z3C))H-I-CD0CrI-H•C))COCD‘-C))I-ClciCOCDClH-COCDciCrC))C))COCciCDHH-CCQCrH-CDCOCO><COI-jCOCrc-rCOXI-hCOC))(1C)CDH-CQCrciH-H-C))HC))COCDciCCttQCDCDCOljCOCJI.QCCO<HC))CCrHC)CDcF-C))H-CDIII-iCrCDMCtCDCOC))frNCrCrCCOH-I-hCCCl)CD-.CDCDI-CDHClc-rI-jCC)C))3COClHC))<H-CDHCDCDCDCOCDCrCDCtCDCCOCCrCC))CDCrI-hoCOI-frCOClI-CDHClI-cI-c—CDCOH-I-CDC))ciCDCDHCCCCciI-CrCrCO-COCrCQCD(iJC)t5ciH,I-hClC)C)ClCDC)CDCDC))CrCOCDC))CrCciCC)H-bC))C))C))CI-CrCD•I-jC))I-C)I-ciCOH-ClHC)CrHCOCrClI-I-CD—.ciI-iCOCDCrI-I-CrC))H-CCCOciCCD><C)ClH-H-ChiH-H-I-H,C))CO(QCDCDI-H-C))(Q‘dCD3MNIH-CClH-H-I-hC))CCOCDH-I-HH-HCDC)COHHC))ChH-IC)CCC)IICrC3C)CDCDXC)-CrClC))I-CrC))CrCDlCDI-ciHC))C))HCOCQH-COCDCDC)CrCrCOIC))CCrHH-C))1HDCOC))COC)ClI-H-COCDCClCDH-H,-HH-CH-C))COC))CDCrCrC-tC))C))I-C))—C))C•c-t<b)H-CDlClC)CrC(Q1COH-CrCDC))—HCDC)CDciCDH,CDC))HCCDCOCCDi-i•H-C).C)CO1ClClCDC))CDCOC))•HIIH-tHCDH-CDtQC)NCr-CrHCDC))H-C))•CC))CDH-CDH-H-CQCDH,CrC)CrCrCI-hQC)CDCDHC))HH-(1)HCDCOC))H-CDCDCDH,<Cl‘1HH‘tiCO-DCrH-C))CDCO(QCI-ClLII0HiQCOQI-IIH-CDC)C)H-Cr<COIIH-CC)ci0tiCDCDH-C)H-CI-COC))C))C))H-H-ciCDCOIICCDCCCOCr—COCDC))C))CDCH-H-I-COClH-(QC)ciCDI-hH-ciciCrHCOH,3CDC)CDCrCCHHC))ClCrciCrHC)-00)‘-0ClCrC)ClC))H•C))CQ<COC)tYCOCD0)•CrHH-COCDHH-CDCDCOIIHC))CCH-CD•Ik<CDI-hI-ha,H-0F-1tiCOM,Mcv0ClcvClHHti-Dh00<0M000HD00CDCl)CDCl)H-H-H-CD0I-I-CDHCDI-HC)<cvC)HCl)MCl)cvC)ICDNHC)0cvHClCDCDCD0HtQ—Cl)CDH)0frHQ.00CDI-ClQ.Cl.HtCDCOCDH-CO0C)tH0C)if-vHM•CDQ.Cl)H-CDCDSH-II0‘<I—hH-H-<H-H-Cl)f-fI-CDCQCDH-HCl)f-vI-H-cttQ.C0ZcvcvIICl)cvCOCQcvf-vCDCl)hf-fCD0iCDM-iH-CD0IINH-CDI-iCD0cvCDCDv‘1CDCDI—h0HCOCDHCDCDH-0MCDiicvtYQ.0C)xCDCDH)Cl)0Cl)HCQCDi0tiCD<COHC)0cvC)cvLQ-HC)C)0HCl)cvcviCDCl)QH-0H-Cl)HH-HCDCl)CO00C)hCDi-cvCOlCD<0cvCl)Cl)0Cl)COCDH-CDf-I-cvCDCDctC)C)HCDCl)COCDCl)cv<H-cvCOC)IICDI-iH-HH-cvCDCl)Hhf-vCDf-vCl)H-0IIH-cvH-COCD0cv00Cl)0‘-C)CDCl)H-<cvCDC)Cl)CD<CD0H-ti0Cl)Hcrk<0CD0h’ct0Cl)CDQ.‘1H-Cl)‘ti0CDI-iCl)H-C)I-00f-vC)CO-H-COI‘-CDH-Cl)f-vHH-H-HM-,Cl)cvCl)CD0ctCO•0cvZ0CDZcvCl)0H01C)COcvClHCOCDCOCD(QIIClcv‘tiMltic-I-CD0ItiHCD<-CDCl)cv‘-<i-jcvH-CDHH-HCOH-0H-cvQCl)I-cvH-Cl)Cl)H-0C)CDf-vCDCOfrHH-Cl)cvCDI-CDcvCDC)CD00H-COcv0f-vI-CCCDcrI-CDCDCD0(QCl)f-vCDCDCl)QMl(QcvH-CO<HHI-0CO5COtQI-COHCDCOICl)Q.hCl)NCDXCOf-v0Q.Cl)H-HcvH-50IICOCDH-0C)H-Cl)H-H-f-tt<Q.Cl)CDCl)t1)00CO5H-CDf-v5cvCDcvC)C)HH0COH-Cl)COH-H-CDQ.‘Z0hf-fH-CDC)jC)k<cvcvcvCOH-HI-’-0cv(Q0CDCl)Cl)H-Cl)<CDcv00cvCl)i-h<QC)CDI-i<I-CO<CDf-vCl)0Cl.ClIH-ClCOC)0cvtic-I-CDQH-ctCDH-Zcvcv0cv00Cl)-COCl)50H--Qcv0C)Cl)MlMlCOcvH-C)HCl)f-f00HCDICSCSCScvCDQCOCDcvCl)cvClC)cvCS•COhH-000IIhH-f-tQCl)CDCOCScvCDCOf-vcvtiHC)Q.I-cvH-H-HCOCDC)CDf-I-H-CDH-)Cl)0IICl)CDH-IC)CD<CDC)‘Cl)Cl)Q-‘CSCl)CDH-0Ht5IICOCl)CD‘<‘d00CDCDCl)C)5Q.COHH-ClHf-vCl‘tivCDf-vCl)CDCDCl)COCO‘CSHCDCOCO—COCl)COtQlHcvHCOCOCO<COCOcv‘CI••H-k<Cl)H-H-—H-CDCl)tQCDCOCl)COH-HCDCOiH-Cl)0Cl)Cl)Cl)-CDHiH-0Cl)<<CO0<CDCDCC<IcviH-,HCl)<cvC)HCOH-HcvcvCOcv—CD0Q.H5CDC)Cl)H-0CO<I-HHCOCD•ClCDI‘CI-D0i-h<5f-vCl)H-ClCDH-Il•H-CDf-ICDH-CD0Cl)tiHCOCl)0H-CQ.l-LI)Cl)COCl)CDH‘DH-COCOcv0IlHHCQH-CDf-tHcv5IIH-CDCl)cvCl)<II0Cl)CDCD0•Cl)Cl)COCl)CDCDCl)LI)H<Cl)‘CS‘CSCDcvcvfrjH‘<cvCD-.Cl)0CDH0CD‘CSIIcvCD•IIcvCl)CDcvH-H-HI—hCOCOCDCl)-CO0II•CO<‘-<CDCOCl)I—hCl)lICDcvCDCl)C)H-H-CO00QH-ctCl)HCDHcvH-CDIIi-hCl)Cl‘-<cvcvCDHCOCDSCDCOClCDCOIILfl—5QC)-]H-F-H-))CD)H-CflIIHQIiIiCrCDCH-C)QCD)t-CCCD5HCDINH-ICrCCD‘1CDCDCDCCDHC)C)CrHL-H-CUCDCDH-IH-f-tIc))5I—iH-HCDC))HCrH-H-C))IiCC))CDC)CHd‘<ClCDCDHCUH-HCrH-kD)lCDCDCCDF-H-HCHCDH-CDCrC)C))H-C))t-lcnf-tHC))C))C))CDCrCDCrHC))C))H-C))fr<IICDIH-CDI—cCDCDCD•5çt3CtCDII‘10CDC))IC))-CrCDCr-C))H-CDI-Q.Cr—QCrH-CDC))C))5CD0CDCCDC)•CUciCDCDH--t-ri-C))CtHCDQCr5I-CrCCD5HI1ZcihH-CtC))CDCC)0CDC))C)h5<CDH,CtH,CDNCrC))CDCDC))HC))CDCCDCDCrCDJNH-3Cr0HCDH-CDC))CrCDCDH-CCl0lCDHCDtQ<C)C))CrH-(QC))CDHF-H,Cr0CD0ciFC))I-iICDH-H-ICDIC)ciI-5I-hCDCDH-Crw0H-ciH-CDCrCr<0CDH-CCDCDCDCD<3CDCDCrCCtCD<—CDC))CD•ci11CDH<0CDCDCDH-CDCDCDI-jIH-ClI-CD(CC))CrHCDH-C)C))HH-5-IC))CDCDF-CrCDCciCDCU5H-CDH,Q<ICUH-CDQ<H-5CrCrC)uCDCi—cI--cC)IH-I-CrClCDCtiH-CCDLi.CDC)H,CCUtJCDC•C00CDCDCDClCDCDC<CC)05ClCDCUHciHC)CDI-Cl)3ClCI-CUcii-iH-CDCDlCDH-Cl55H-H-C))CDCUF-It-tQCrCDCDI3CDci-CDCCDH-CrCDCUII<CDCDCDCHk-tC)LHCDCDCDH-CDXCrCDCrHCDIçtHC)C)H-c-tCrCD0lHClCtH,H-CrCDC)CiCDCIH-0HCDCH,H,C))CCDCDH-CDCrid>IIMCCrQH,CI)C))H5H,CCDH-lii‘tiIHH-CDCDIIH-S04H-5H-CrCci0IC))Hk)CDH-CDH-CDC)0I-0tQH-MClC))CrCrIXII-CDHCtlCDHC))HhdtQCDH,•CDC))I-CDçtH-C))IC))C))Cr<ciH-I3C))CQCDH,C))hCtCC))C))CD‘3HH-CrClCDCrC))II3CrIIPCDI-C))CUCl)I-’dCDC)Ii-’-H-CrCH-CrH-5—IH-I-’dCDCD30(QCDClCrlCDCDH,0CDCDCDCrIHCDCD•CDH-‘<ClH-CrCDH3ClC)0IX0H,Cr0C)CUH-CrCUH,CDIXH-CDLDCD0C))CDciCDC)CC))CD•HCDC)CDC))CDC\)3CCDCl)I-QC))C)CDCr•C))HIXClC)C)ClCUtiCr-CrCtCDCDl-dH5HCD:—ciIHCD•ciH-CC))CDL!ZCDCDC))C))H-CDItikvC)CrIiCDCUMHlCD0IiCDCrCUC)<3CI’-’‘H-H-H-C)‘<CD-JINCr(QH-CrCDH-HCI)lo‘-0Ic<HH-H-CrCDCD‘U1IH-ZCDU)CDC))CDH-CDkouikjc-fCCDCDHCD‘-0L3t30CtC))IHHCDH-CCl)H-CD—lCDCDCt<CrCtIC))CDHH-hC))CDCDHlCD‘H-ciCl‘tiCDCDC))CD‘<C)CrCUIX‘tiH-‘C))H,IH-C))CDH-CDQ.<IH,lH-CD0ICUCD<3H-H,H-H-03CDHHCDCDC)C))I-cCDCr<CClC))HC))C))C))CDH-H,H-‘tiH-1-iC))CDHCDCrH5><<CDClCtH-C)I-’dCrU)5l—HH-CCCC))-CDH-CDCDCrCUCH-CDCDCC)C‘-I-cHiCDCrl-dCDCDH-CrI-H-CD3H-CtCDU)XH-CrCCUC)CCUCD‘-<CCrH.C)<HC1ci‘<CCUciCrHCDCQH,:iCCDHCH,ClI-hU]coD)QIIHi0IYCD0M“IU)Di(-r‘I-CDI-iCDCD0CD0CDI-H0U)CD0CDçrCl)C)U)U)Di0H‘-jQCDDl<H-DlCl)CrtiN0C)ZCr(<U)H-I-H-Cl)IICDCrH-tiCD1Cri.QJCDDl17jCDt—U)H-C)CD‘—JCrU)P1H-(I)Cl)I-HU)Cl)CDICDti0C)Cr0U)‘-<U)CrH-I-CrH-0H-Cl)CDQCD‘1-rH-0H-HCD0U)CDCl)CDI-CC)CDCD<0DlCDI-hQ.CrCCrDlMH-DiDiI-H-QCDCr‘-<CrCrDlI-jCrH-HCrCDDiH0H-CDCrCtCrIICD0CDCD0DiDiH-U)HH-Q.<CDN0C)CrDiCDDi-DiU)HooCDHQ.H-DiI-CDCDC)Cl)CrU)IIDi<CDU)—U)0DiDlQCD0DiL0H-Cl)I-jHCr00-NtQDiH-H-.H,Di0H-C)C)I-<I-iDlCDCDti0CrQ(QH-HH-ZCtItiLQU)J0DlH‘titQ0Q.H-CD0CDCDCDC)-iCDCl)H-U)H-CDH-DlU)DiCtI-DiHH-0Cl)CDU)-rHI-iHCl)<Cl)-I-H-CrHC)<HHCrH--oCDH-CDU)CDQ.H-H-H-hDl0ti0<I-0oCCDI-hH,H-HCDI-(QohCDH-(flH-II-DiU)CrH-HDi(I)HC)H-H-H-CrCDCDI-U)-.U)•0CDti00H0DiH-DiDlC)Q.CDHC)HDiCDI-hI-H-,H-I-HC)DiCDHCDI—’CrU)DiItiC)DiC)HCDCDHHCl)IiQ.DiCDDlICrCDH-Cl)0U)00XHQ.CDHH-HCrQ-QCQ-h‘1H-U)DiC)I-hH-U)H0CDCQU)riU)<CDHDiCDCDU)Di><IU)QDlCDH-IliCDHU)i-.DiCl)0hH-HU)‘<QC)C)0C)ClCDoCt-C)Cr0H,lCD‘<‘tiH-CDCD0HC)Cl-3CDIDiCDCl)Cl)H0CDDiCl)CrDlCri-CrC)DiH-CDH,U)CrZ<DiJCDH-CDH-CDiH0CDH-U)DiCD0DiClCDU)CDClCDI-<DiH-i-CrU).CrCl)C)C)CrCD‘tiI-i.I-iC)Cr-30IXH-HHIIIDi0C-fDiIX0H-Di0Dic-rDiHCDCD-CriiDiU)H-ClHI-CDDiCDCrDiCfCrCD‘tiCDH-C)I-hH-HClfrç-rU)k<DiCDDiClH-I-H-CHU)CDCDU)CDClCDClDiCDClDiI-iI-i0HHII0CrCDU)ç-rCDDlHI-‘CD‘ti0‘DH-CriiHDiClCDU)XHI-‘-0‘-<iCl‘tiClCDU)OH-DiC)ClC)‘DiDiCDCDDlU)H-DiIIDiUiC)CDClc-fH-CDCrCrC-‘CIU)I-Ct‘CI‘—DiI—DiDl0CDIIC)-rCtII—H-DirrCD0DlClCD•HI-IX0U)IH-DiCDCDCl)IItil-CDDi•CDCDCDCrH-CfCfU)XU)HH-HU)HH-CIHDiH-0Cl‘<Cr‘-oCfDiU)0‘-<CD‘CICrU)H-H-H-U)rrCDDlH-H,Cri-c-rCrc-rU)Cl)Cl)H-HHClDi0a0CrCDU)H-0H-DiCDJCrU)‘oH-U)H-0C)H-H-0HDiC)•U)IIDi0CfH-HCrH-C)H,U]I-hHH‘CI0IU)(Qrr‘-DCrDiC)H—CD‘-<DiCDI-hCDU)•I-H-CDU)CrI-U)IIC)0<1O0Cl•I-’H-CDH-H-H-DiCD—(-TiU)H,CD‘CIH-CrU)U)60Table 2.1. Summary of Menziesia herbarium specimens used inmapping species distributions.Taxon/Territory No. Specimens TotalNorth American TaxaN. ferrupinea, sens. lat. 1309Alaska 198Alberta 97British Columbia 325California 38Idaho 196Montana 169Oregon 101Washington 162Wyoming 23N. oilosa 503Georgia 11Maryland 24North Carolina 169Pennsylvania 26Tennessee 34Virginia 147West Virginia 92Japanese TaxaN. ciliicalvx 55Honshu 53Shikoku 2N. povozanensisHonshu 1N. lasioohvlla 3Honshu 3N. multiflora 54Honshu 54N. entandra 48Hokkaido 8Honshu 38Shikoku 2N. ourourea 3Kyushu 3Fig.2.1.DistributionofMenziesiaferruginea,sens.lat.,inwesternNorthAmerica.13S°UC’z86°NDISTRIBUTIONOFN.PILOSAINTHSTERNNORTHAMERICA81°NFig.2.2.DistributionofMenziesiapilosaineasternNorthAmerica.76°N63250km 38°N,-7 j44E1. —‘ •boo0 M. ciliicalyx M. multiflorap * M. goyozanensis • M. pentandra0 M. lasiophylla M. purpureaçOFig. 2.3. Distribution of the conunon taxa of JapaneseMenziesia based on a representative sample of herbariumspecimens.64Table 2.2. OTUs utilized in morphometric and ecologicalanalyses of Menziesia. Arranged within regions by state orprovince with field sites highlighted in bold. Collectioninformation for all OTU5 is given in Appendices 1.2 and 1.3.All accession numbers are T.C. Wells collections, except stateor provincially coded OTUs which represent herbarium specimens.Taxon / Region No. OTUs Individuals ExaminedNorth American Taxa. ferrupinea, sens. lat.Wi Alaska 5 AKO1-AKO2, AKO5, AKO8-AKO9W2 Alaska/B.C 12 AKO3-AKO4, AKO6-AKO7, AK1O; BCO1-BCO7CoastW3 U.S. Coast 13 WAO1, WAO3; OR: 08W (1009), BEV (623-624) , PER (926, 939) , CC (948) ; CAO1—CAO2, HUN (628, 637, 639)W4 Inner Coast 33 BC: BWF (1725-1729, 1733, 1735, 1739—of B.C. and 1741, 1744-1745, 1747—1752, 1754),Washington STL (707, 711, 717, 726, 729, 733, 738,740-741), PL5 (662, 1010, 1017),SEY (BCO8); WAO5W5 North Cascades 20 BCO9—BC1O, ROT (745, 750, 754, 759—760,762, 766), SKY (799, 808);WA: STV (1677—1678, 1680-1686)W6 South Cascades 4 WAO4; GVC (ORO1), 0R02-ORO3W7 Western Ranges 31 B.C.: SPA (810, 812, 814), TRO (821),of Rockies MUR (873), GNP (1121—1123, 1126, 1129,1132, 1134, 1137, 1140, 1143, 1145)WAO2; 1D02—1D04, FRE (1D05), MOS (1715-1722) ; MTO8—MTO9W8 Main Cordillera 32 ABO1, LL (1151, 1169, 1177, 1184),WAT (1190, 1219); IDOl; MTO1—MTO7,MT1O—MT12, ALV (1228—1230, 1232, 1235,1237—1238, 1240, 1247—1248, 1252);WYO1-WYO2, TET (WYO3). pilosaAl Central Penn. 5 PAO1-PAO5A2 North Allegheny 9 PAO6-PAO7; MDO1-MDO3, WBR (1273-1276)Table 2.2 continued next page65Table 2.2 continued. OTUs utilized in morphometric andecological analyses of Menziesia.Taxon / Region No. OTUs Individuals ExaminedA3 VA/WVA Ridge & 24 WVO1-WV03, DOL (1280, 1285, 1289, 1290,Valley 1292, 1309), CAP (1334, 1341, 1358,1360), MIN (1374, 1384, 1390, 1399,1401) ; VAO5, MTL (1601—1602, 1607,1611, 1623)A4 N Blue Ridge 12 VAO1-VA04, VAO6-VAO7, JEN (1312, 1316,1325, 1327, 1332), RKN (1402)A5 S Blue Ridge 38 VA: WTM (1406, 1414, 1419, 1422, 1431,1434-1436); NCO1-NCO5, MIT (1440, 1453,1458), PIS (1470, 1478, 1482, 1502),WSM (1508, 1514, 1519, 1530) ; TNO1—TNO2, LEC (1572—1573, 1580); GA: BB(1557-1558, 1561, 1566, 1568), LWS(1538, 1541, 1544, 1550)Japanese Taxa. ciliicalvxHonshu 6 CI01-C106. multifloraHonshu 13 MUO1-M[J13. oentandraHokkaido 3 PEO4, PEO6-PEO7Honshu 7 PEO1-PEO3, PEO5, PEO8-PE1O. turoureaKyushu 1 PUO1/1/2 /7O’,4 0I .500kmo •3I T( H• 115w’\ 3844Fig. 2.4. Sampling regions and collection sites of the westernNorth American Menziesia specimens examined. Open circlesdenote samples collected from field sites; black circlesindicate that samples came from herbarium collections.Further details are given in Table 2.2 and Appendix 1.2.Fig.2.5.Samp1incregionsandcollectionsitesofMenziesiaspecimensexaminedineasternNorthAmerica.Opencirclesdenotesamplescollectedfromfieldsites;blackcirclesindicatethatsamplescamefromherbariumcollections.DetailedinformationonallsamplesisoutlinedinTable2.2andAppendix1.2.286w----.\\II125km68Table 2.3. Description of the twenty-two morphologicalcharacters used in the morphometric analyses of Menziesia.Leaf CharactersANGBASE Angle between tangent to leaf base and midvein withvertex at petiole insertion point (degrees)ANGTIP Angle of leaf apex between midvein and widest pointof lamina (degrees)BPUBG Density of glandular hirs on lower leaf surface(number of hairs/lO nim surface area)BPUBP Density of puberulent hairs on lower leaf surface(number of hairs/lO mm2 surface area)CILIA Number of ciliate hairs along leaf margin (measuredalong 1 cm length near leaf apex)LABOVE Leaf length above widest point (mm)LBELOW Leaf length below widest point (mm)LWIDTH Leaf width at widest point (mm)NUNVEIN Number of secondary veins along one side of midveinPETIOLE Petiole length (mm)SPUBG Density of glandular hirs on upper leaf surface(number of hairs/iC rnmh surface area)SPUBS Density of subulate hairs on upper leaf surface(number of hairs/iC mm2 surface area)SUBLATE Number of subulate hairs along lower midveinFruit CharactersCALPUB Number of glandular-ciliate hairs on margin ofcalyx discCALWID Calyx disc width (mm)CAPLEN Length of capsule (mm)Table 2.3 continued next pageIC?)1inflCD13LiiLxiLxijIIC)0Ccii0ZWilF-I-IHLiiC“d0110(Dzii IIH-taIIIC).IIIIHHJZQeiQllIH-IH-frCDCDCDCDCD(DlllQIIJC)ctH’COH-i’jCi)CI)CflCI)II0IctH’C)CDCDH’CDH-ilICI)H-rtCDi—c-tI—ctilH-act0CD<<‘<<Ii‘<HIM-tQ0IiI0F-M00ilCDIC)CDMCDH,MiICl)CDID)CDH‘cjctQCQI-CDCDiI0.ICl)H-F-CtCl)Cl)F-HIH-C)D)H’C)tC)D)Hlf-CDCrCDCDCDiijlCDI-Q.-Cl)QiICDCDIrrctctIIiCl)ICl)-I—i-.I—l—IINC)lCDCDD)CDCDD)IIH-I-ji(QIICDHçtIICJ)’TJI(ClH-Y0H-Cl)Cl)IiH-ctIQ.(DctrtD)iiD)0iCt$1IMD))D)H’IiICI)F-Cl)MCtH-ctIiC/)CD0CDCDCflhiI—.-M3C)Cl)Cl)Cl)Ii0F-Ct——0IiCl)CD0IiCDCl)iictH-p)C)C)iiIJIictCDC)D)iiH’ICDC/)(DIIC)‘tiCl)HCDU)H(DOD)i-ii0CDIIIH’jCDHC)rrCl)—flCDCD(Qil0LQi-IlID)IICDtailCtIIIOCQI-OilH’I0liC)I(QCCIIICl)iiICtIICtOh0Ii0-,IIij‘iJC(r:01’l(nLillhhOIXJClCD000>tIIOOIIi00ZrtHHI0LII‘<UIC)(r‘-](D1:-ILi0LII‘-d<)llIHiLIILII00tJC!)C)LIII-’-IIs1(DCDIILiiCli-(IlCDHIr’i00‘-oIIflPILiIILIIIIP)C1•I0IIIciI—-C)C’C’I-’.JOCl)CD—.rtQnct(Dcn10<<<()H-f-i0(1H-HfrHCCIiICDCDCDCDP(QQH-CDCDCDt-0PJCI)(DIIOP)LIII1fr1IP)HII-CD<III-hC)IP))P)P)OQ.0CDCnn)III(Q(QtQI1<H-QiCDCDrr(DctIIHHICDCDCDCDP0CDH-H-OH-IICD’<OIII0c-ru0rtQPIICl)QIP)Q.Qi)110c-I-—CD0IIEj)CDHIC))C))(QQCDH-CDIICl)OI0H-H-CDP)li0H-CtrtIII-iC))cHHrrc-r—01-i,—C))CDIICDOHI3CD<<CDH-CDCl)C))IIçtI-OEnc-I-CDflhICDQiICOrtC))CflHH-II3CDI00C))Cl)H00IIc-I-CEOII-hI-hEQ00HC))‘Z3C))IIIH-M0QCDc-I-IIc-tIc-I-I—hCl)C))HH-IIIIHI0IIc-I-11CflCO‘-ICDCOIIIc-I-0‘-<—•0cTEQCD0CDIIX>c-IIC))COc-I-H-000IIIHc-I-CDc-I-CD011CDII(D11ICDQ-><IIQi11IIIEl)—CDQ.CDC))IIHIIIIICDc-t11CDCDCDQ.HIIICDCD11CDPJ0C5IIQiIIC))CnI0CDC))I-jctO—•H-IICDIH-c-I-C))CDCDM00IIICnh<(.J0IIH-IIIH-C))11—0I-(IIHIc-I-<CDIIc-I-Cl)CDCDc-tIIIC))Cl)CD0c-I-QCDC)ICOIINIci--C)C))0IIH-ClIH-‘C)—c-I-0IICDI0CD°C)H-CD<HIIEOCDIIICl)00IIHICDC))0IIC))OI<IIEQII•C))ICDCDC))CIIIIiI11C))H0--c-I-IIQIII-Q.IIICl)ClQCDCDIICDHICD—0CDEQII(DOIC))QjH-EQIICOIIC))IIIC))H11CDH-IIc-I-HIk<COCDCDc-I-IICDICOCDCOCDIII—c-I-CO—IIftI0—IITable2.5.CharactersusedinthecladisticanalysisofMenziesia.Seetextfordetailsoncharacterpolarization.StateCharacterPlesiomorphicApomorphic1.numberoffloralparts5merous4merous2.numberofstamenstwiceasmanyaspetalssamenumberaspetals3.sepalprominencecalyxlobesverydistinctcalyxdisciform4.corollashapecampanulateorurceolatetubular-campanulate5.corollasizelarge(10-17mmlong)small(mostly<10mmlong)6.corollacolourgreenish-whitetoyellowdeeprosetopurpletosalmon-pink7.pedicellengthshort(typically<20mmlong(typically>25mmlongbutupto30mm)longbutupto40mm)8.pedicelpubescencemostly<10hairs/cmtypically>13hairs/cm(uptol3)(upto30)9.ratiolengthofglandularglandular>>twicelengthglandularatmosttwicetopuberulenthairsofpuberulenthairslengthofpuberulenthairs10.densityofpuberulentglabroustoslightlymoderatelytodenselyhairsoncapsulespuberulentpuberulent11.densityofglandularglandularhairsesentiallyslightlytomoderatelyhairsoncapsulesabsentwhenmatureglandularpubescentTable2.5continuednextpageIHI—iI—aI—iIC)II—.1wIiiI.•itIIII—’IC)P0Q0Q.0Q.oc1IwIICD ICI—uiCDCDICDi(DIC)iiIF—0IrrII0it—toI—tflCflCflICDIII(Q(DOH-OH-‘tiP-‘tIH-IIIUit-’çc-rCt‘rJCr‘tJCrIIIII—iCD<CD<(D<CD<III1000‘1‘1‘1IIIIIC) IM,M,00C)01II0ICDF-H,f—iH,I—aH,I—aH,IIIIf—it-(OCDCDCDCDIIICrlCD(t)CQI)’tJP)(QP)toIIIH’ IP)f-I-t,F-I-hM,I—H,IIIIH(It3IIIItoCO(D(0C0IIICDIM,)CrhhthHIClCDH,HH,HH,I—aH,CrIIIlCDH-(l))P)(DP)P))CDlIIIQ3C)fr’C)C)C)IHC)ItoP)Cl)CDrtCDCDIiiIH-H-)IIIP)H-IIIiiIH,toH-(1)H-1-jIIICt)Ct)H-II(1)1IIC)IC)tnI-ItoIICrlCD(0IIICDIIIIIIIIIIIItoIf—ic-rI-H,CDCrCDIIICrlCDH• 1<C)CDto‘<toIF—IICt)IP)S1’tIto‘z5toICDIICt1<H-CDH-CDItoIICDICT)H-C)C)IH-HCl)Ito1’JP)t\iCtç-tP)rtioIIIF-k<H-HH-IsIII‘CiF—atoCiCt-’F—il))I0II(I)Il))<H-Hk<F—IIICDIIIC)H-i-I‘CiIICllCDAH,T))Ct)‘<I3IIIPJ1—IH-IIHI(QI-C)fJ)toCt)IC)iiIC)CD<CDCi-iII1(T)to(01IICrlCDCtC)CDCrCDIII I:iCt)IlIICDIH-CDU1CrCr1IIltj1-Ct)I(1)11C)lCDCl)Ic-rnII-Ct)ICt)IICt)10H-ICr11ClI1-ICDIIHICl)III(I)IIiiCrIIIIHIIIIC)IIIIIIIICt)IHCri—c-ri—CrIkVIJ’IIlCDH-’<C)’<<D’<PtoC)‘CIIICt)ICt)QJ’tI‘Ci‘CiH-CiL.iIOIII<H-H-H-11Ct)IIIIlCDH-C)C)C)toF-LflIOIItoIto(JCt)t’3Ct)1’J)F-JC)II-IIIHIH-Hl—H‘<I‘CiIItoI(QF-toI—Cl)F-C)totoIII‘<i<si<VIH-II0Ct)II-jI-jIC)IIHIVMVhhVU)H,IIIIC)Ct)Ct)OICt)HIII10I-C)F—C)HC..-)C)Ct)III01(DC)(DC)toC)CDCIIIItoICDCDIIINIC...)Ct)CDCt)II-CQ)IIIH-IC)IU]II@H,HhIIICDCD(DC)WCt)CDP)III If—iCt)1JCt)C)C)Ct)H-IIIH-10)Ct)(DCI1-jIIICt)IH-H-(01IIII-jCt))—IIIIto(I)H-I-Ct)IIII-.‘iCDIII(1)Ct)IIIIIIIIIITable2.6.CharacterstatesoftaxausedinthecladisticanalysisofMenziesia.(0plesiomorphic;1apomorphic;U=unpolarized).SeetextandTable2.5forcharacterdescriptionsanddetailsofscoring.TaxonNorthAmericantaxa.ferruainea“ferruginea”ferrupinea“glabella”j.pilosaJapanesetaxaCharacterCode1234567891011121314151617OutgroupM.ciliicalyxjvj.aovozanensisjj.katsumataeM.lasioohvllayj.multifloraI.oentandra.ourourea.vakushimensisFERR100GLAB100PILO100CILI000GOYO100KATS000LASI000MULT001PENT010PURP101YAKU10101001001010010000000100101000110011001111010000000000100001000000000100010001ill11111010001000011011011100100001011101101100010001000100010001001000010001CladothamnusCLAD00U00UUO00OUO000074Fig. 2.6. Box plots summarizing variation in 22 morphologicaldescriptors (Table 2.3) of North American Menziesia grouped by:F ferruginea; G glabella; P pilosa. The central horizontalline represents the median, while box edges define the upperand lower quartiles, respectively. Whiskers show the range ofvalues falling within l.5x the interquartile range. Outliersare plotted as asterisks or as open circles when beyond 3x theinterquartile range. For a given character, groups with thesame letter do not differ significantly from one another(p < 0.05). a) leaf base and b) leaf tip angles; c) glandularor d) puberulent hair densities on leaf undersides.a ANGBASE b ANGTIP50— 80 I* a b bGRPCODE GflPCOOEC BPUBG d BPUBP50 I 800 I40- J 0 600E30 :GRP000E GRPCODE75Fig. 2.6 continued. Box plots summarizing variation inmorphological descriptors of North American Menziesia.e) calyx pubescence; f) calyx width; g) capsule length;h) ciliate hairs/cm on leaf edge.e CALPUB f CALWID20 515(I) 4(‘30 2F G PGRPCODE GRP000Eg CAPLEN h CILIA10 40a b c930GRPCODE GRPCODEab b a1*F G Pb C76Fig. 2.6 continued. Box plots summarizing variation inmorphological descriptors of North American Menziesia.i) glandular or j) puberulent hair densities on fruitcapsules; k) no. of fruits/infructescence; 1) leaf lengthabove widest point._3 _3> >(I) (1)cc2) U)_o10CU)00D(I)____aC354CPUBG I54CPUBP000a0—1k8F G PGRPCODEFRTNUMb aF G P. GRPCODELABOVE0—1403020100t I I2F 0 P F 0 PGRP000E GRPCODE77LB ELOWp50Fig. 2.6 continued. Box plots summarizing variation inmorphological descriptors of North American Menziesia.in) leaf length below widest point; n) leaf width; o) numberof side veins per leaf; p) pedicel length.LWIDTHn353025201510m50 -40 -30201008743a a b*GRPCODENUMVEINF G PGRPCODEPEDLENEE40302010F G P F G PGRPCODEGRPCODE78Fig. 2.6 continued. Box plots summarizing variation inmorphological descriptors of North American Nenziesia.q) pedicel pubescence; r) petiole length; s) capsule segmentwidth; t) glandular hair density on leaf upper surfaces.q PEDPUB r PETIOLE25 - -E2O(I)1Ct’41035F G PGRPCODE GRPCODES SEGWID t SPUBG4 I 30a b c20 -*c’JI E* E0ioGRPCODE GRP000E10987*a a b*2F G P**Iaab79Fig. 2.6 continued. Box plots summarizing variation inmorphological descriptors of North American Menziesia.u) subulate hair density on leaf upper surfaces; v) numberof subulate hairs on the leaf lower surface midrib.£21-oE___,10SPUBS*V SUBLATE30U4030E20(t 100-1020** II I I***aabCF0-10 -G PGRPCODEF G PGRPCODE80Fig. 2.7. Contour plots of descriptors exhibiting clinalvariation in M. ferruginea, sens. lat., in western NorthAmerica, superimposed on the sampling localities (Fig. 2.4).a) leaf tip angle; b) glandular or c) puberulent hairdensities on lower leaf surfaces; d) calyx pubescence.a ANGTIP b BPUBGdegrees hairs/lOmm265 I I 65 I I25 150 4N 60605020-25-J 45G50 )7 50 10-150 ( 6051045 \ 55\40 I I 40160 150 140 130 120 110 160 150 140 130 120 110L0NG LONG°C BPUBP d CALPUBhairs/lOmm2 no. ciliate hairs65 I I 65 II6010010.155 - 860 -8-J504550-12212:I8I\50\ 10•/1 \ 40 I160 150 140 130 120 110 160 150 140 130 120 110LONG° LONG81Fig. 2.7 continued. Contour plots of descriptors exhibitingclinal variation in . ferruginea, sens. lat., in westernNorth America. e) capsule length; f) subulate hair density onupper leaf surfaces; g) number of subulate hairs on the leaflower surface midrib.e CAPLENmm6560F—-Jf656055504540160. 150SPUBShairs/lOmm2F—4:-Jg656055504540140 130 120 110 160SUB LATEhairs/midrib150 140 130 120 110-J55504540160 150 140 130 120LONG°110LONG° LONG82Fig. 2.8. Contour plots of descriptors exhibiting clinalvariation in pilosa in eastern North America, superimposedon the sampling localities (Fig. 2.5). a) puberulent hairdensity on lower leaf surfaces; b) capsule length; C) capsuleglandular pubescence density; d) capsule segment width.a BPUBP b CAPLENhairs/lOmm2 mm4—1 I I I .4—1 I I(:) 5.0 5.240 200 588O385.2 -5.438 250-300350.®?-J3737 2503636 150-200 .200 4 6 4.810035 . -34 I I / I I 34 I I I I84 83 82 81 80 79 78 77 76 84 83 82 81 80 79 78 77 76LONG° LONGC CPUBG d SEGWIDdensity level mm41 41 I40 40 2.0\\•c739.. N3.24:: 1.9-2.138 -J-J 37 3.0 3724 35-1.7-2.0.8-2.2 II )l1362.63484 83 82 81 80 79 78 77 76 84 83 82 81 80 79 78 77 76LONGS LONGII1.6____11.4__________12I1__I-;;;1.04W2(OSW)12W330W43W62WA19BWF4Al2BEV8STLBA24W725W72PER2PL50.1A217A32GNP2SPA1CCBCO81D051TRO2CA3?.31.A47A4MTO91MtJR3HUN8W54A3(MTS)2VA9GNP2BC(DOL)15AS4JEN22W88MOS2W42ROT2A4NCO11RKNABO13ID1BWF4STy0.61A41MIT4LLMTO8WAO5(JEN)5AS4PIS10A52WAT1W6NCO23WSM4WTMIDOl10W812W5ORO16ASTNO1TNO23NC3MT7MT5ROT4WTM1BB2LEC1BB9ALV2ALV2SKY1W7A42MIT2LWS3BB2LWS2WYWYO3SSTVWAO2W20.4ASVAO2SAlW7W4w2BCO64WSMLECPAO1SPASTLAKO4PL5PILOGLABFERRFig.2.9.SummaryderidrogramofaUPGMAclusteranalysisofNorthAmericanMenziesiaOTUs•basedonmorphologicaldata.OTUsareidentifiedbyregion(Al,W2,etc.).orsitecodesasoutlinedinTable2.2.Scalerepresentsfusionlevels.A•ferruginea00c0•••b.•—Fig.2.10.Distributionof238OTUsofNorthAmericanMeriziesiawithinthefirsttwoaxesofaprincipalcomponentsanalysisbasedon22xnorpholoicaldescriptors.Thefirstandsecondaxesaccountfor24.5and16.7%ofthetotalvariance,respectively.Descriptorloadinsonthefirsttwoaxesareillustratedbythevectordiagram.TheanalysisissummarizedinTable2.7.0oglabellaCpilosaDC)LWIDTH3 2 1 0—1-2-3--3LBELOWLABOVECCANGT1P0CPUBGC-2-10pci1SPUBS2SEGWIDCILIABPUBG85Table 2.7. Principal components analysis of North AmericanMenziesia using 22 morphological descriptors. Componentloadings, eigenvalues, and the absolute and cumulative varianceaccounted for by the first 6 components are presented.Descriptor abbreviations as in Table 2.3.Descriptor Component Loadings1 2 3 4 5 6ANGBASE —0.410 —0.090 0.137 —0.066 —0.665 —0.434ANGTIP —0.305 0.472 0.500 —0.046 —0.351 —0.191BPUBG 0.532 —0.444 0.242 —0.096 —0.119 —0.066BPUBP —0.600 0.155 0.425 0.384 0.095 0.164CALPUB 0.141 0.198 0.632 0.363 0.040 0.292CALWID —0.084 —0.076 —0.220 0.572 —0.503 0.396CAPLEN 0.727 —0.251 —0.418 0.105 —0.201 0.025CILIA 0.276 —0.403 0.487 0.392 0.122 —0.154CPUBG —0.778 0.293 0.112 0.258 —0.032 0.067CPUBP 0.312 0.310 0.757 0.059 0.031 0.080FRTNUM 0.544 —0.058 —0.226 0.318 0.105 0.035LABOVE 0.438 0.721 —0.367 0.074 0.024 —0.106LBELOW 0.440 0.806 —0.204 0.043 0.101 0.068LWIDTH 0.247 0.867 —0.022 0.057 —0.229 —0.214NTJMVEIN 0.270 0.183 —0.356 0.228 0.124 —0.095PEDLEN 0.753 0.257 0.013 0.118 —0.062 0.030PEDPUB 0.252 0.001 0.162 0.478 0.276 —0.644PETIOLE —0.231 0.656 —0.112 —0.109 0.018 0.098SEGWID 0.603 —0.333 —0.211 0.290 —0.335 0.028SPUBG 0.579 —0.133 0.470 —0.113 0.040 0.007SPUBS —0.657 —0.397 —0.307 0.273 0.113 —0.122SUBLATE -0.716 0.141 -0.420 0.322 0.117 -0.107Bigenvalues 5.381 3.664 2.883 1.550 1.210 1.057% Variance 24.46 16.66 13.10 7.05 5.50 4.81% Cum. Var. 24.46 41.12 54.22 61.27 66.77 71.583dODDo2 1 0 1(.) 0.00.ICALPUBCILIAANGTIP0 0I 10.0PETIOLELWIDTH..SPUBSBPUBGWRegiono1,2,4•32A617.3I08-3—2—101234pciFig.2.11.Dispositionof150OTUsof.ferruginea,sens.lat.,withinthefirsttwoaxesofaprincipalcomponentsanalysisusing12morphologicaldescriptors.Thefirstandsecondaxesaccountfor31.8and22.1%ofthetotalvariance,respectively.OTUsareidentifiedaccordingtotheirsamplingregionsasdefinedinTable2.2andFig.2.4.Descriptorloadingsonthefirsttwoaxesareillustratedbythevectordiagram.TheanalysisissummarizedinTable2.8.LABOVESUBLATECAPLEN87Table 2.8. Principal components analysis of . ferruainea,sens. lat., using 12 morphological descriptors. Componentloadings, eigenvalues, and the absolute and cumulativevariance accounted for by the first 3 components arepresented. Descriptor abbreviations as in Table 2.3.Descriptor Component Loadings1 2 3ANGTIP 0.318 0.705 0.014BPUBG —0.663 0.207 0.092CALPUB 0.120 0.670 0.232CAPLEN —0.157 -0.781 -0.085CILIA —0.495 0.441 0.491LABOVE 0.828 -0.359 0.075LBELOW 0.920 -0.157 0.012LWIDTH 0.898 0.112 0.113PEDPUB 0.150 0.143 0.743PETIOLE 0.607 0.181 0.032SPUBS —0.460 -0.577 0.401SUBLATE 0.230 -0.558 0.596E±genvalues 3.818 2.656 1.398% Variance 31.82 22.13 11.65% Cum. Var. 31.82 53,95 65.60Table 2.9. Principal components analysis of Appalachian. icilosa using 10 morphological descriptors. Componentloadings, eigenvalues, and the absolute and cumulativevariance accounted for by the first 3 components arepresented. Descriptor abbreviations as in Table 2.3.Descriptor Component Loadings1 2 3ANGTIP 0.359 0.073 0.184BPUBP -0.580 0.237 0.463CAPLEN 0.302 0.799 —0.325CILIA -0.671 0.274 0.390LABOVE 0.781 0.292 0.328PEDLEN 0.570 0.349 0.467PEDPUB —0.381 0.638 0.166PETIOLE 0.729 —0.139 0.230SEGWID 0.055 0.836 —0.369SPUBS -0.814 0.091 0.054Eigenvalues 3.284 2.117 1.051% Variance 32.84 21.17 10.51% Cum. Var. 32.84 54.01 64.523SEGWIDA2QA0Ai!—ARegionAolPETIDLE-2A3B—3I-3—2—l0123pciFig.2.12.Dispositionof88OTUsof.pilosawithinthefirsttwoaxesofaprincipalcomponentsanalysisusing10morphologicaldescriptors.Thefirstandsecondaxesaccountfor32.8and21.2%ofthetotalvariance,respectively.OTUsareidentifiedaccordingtotheirsamplingregionsasdefinedinTable2.2andFig.2.5.Descriptorloadingsonthefirsttwoaxesareillustratedbythevectordiagram.DetailedsummaryisinTable2.9.CILIACAPLENPEDLENLABOVEUIFig.2.13.Distributionof238OTUsofNorthAmericanMenziesiawithinthespaceofthefirsttwocanonicalvariatesofadiscriminantanalysisbasedon14morphologicaldescriptors.Descriptorloadingsonthefirsttwocanonicalvariatesareillustratedbythevectordiagram.•ferrugineaOglabellaCpilosa0 000006 4 2C40—2-4 -6-4CALPUBDC0C•ILWI DTHANGTIPPETOLE0 •.••••S UBLATECAPLEN—2SPUBS0DA1246co/ K/1ç11500kmI II T\Fig. 2.14. Gabriel plot defining intersite relatedness amongthe collection localities of . ferruginea, sens. lat.., OTUsin western North America.125kmFig.2.15.Gabrielplotdefiningintersiterelatednessamongthecollectionlocalitiesof.pilosaOTU5ineasternNorthAmerica.H\I--.0.1>0)--rC)(I)t00)-,,r>cm01<CZ-1mom-lr-u0C<--<0-H-C)330<‘]j<rC)>(I)-0.2-0.3-0.4•0.5SUBALPINETEMPERATEMESOPHYTICFig.2.16.DendrograrnofwesternNorthAmericanMenziesiafieldsitesbasedonaUPGMAclusteranalysisofcommunitytypesasdefinedbyaJaccardsimilaritymatrixofspeciesassociation.Scalerepresentsfusionlevels.11Table2.10.CanonicalcorrelationanalysisoftherelationshipbetweenmorphologicalandecologicaldescriptorsofOPUsfromthewesternNorthAmericanMenziesiafieldsites.Correlationsbetweenmorphologicalandecologicaldescriptorsandthefirstthreecanonicalvariatesofeachdomain(U1,U2,U3;V,V21V3)1aswellasintraset(H1)andinterset(H2)descriptorcoromunalitiesforthefirstthreecanonicalvariatesissummarized.Seetextfordetails.MorphologicalDataDomain:U1U2U3H1V1V2V3H2ANGTIP—0.7740.118—0.2870.695—0.7290.0960.2060.583BPUBG0.129—0.2790.4310.2800.122—0.227—0.3090.162CALPUB—0.653—0.0910.3670.569—0.616—0.074—0.2630.454CAPLEN0.823—0.103—0.1500.7100.776—0.0830.1080.621CILIA—0.234—0.1100.6440.482—0.221—0.089—0.4620.270HEIGHT0.5100.433—0.1280.4640.4810.3520.0920.364LABOVE—0.0580.584—0.4740.569—0.0540.4750.3400.344LBELOW—0.2800.594—0.5070.688—0.2640.4840.3630.436LWIDTH—0.4360.523—0.6660.907—0.4110.4260.4780.579PEDPUB—0.2300.5880.2020.439—0.2160.479—0.1450.297PETIOLE—0.3520.254—0.2250.239—0.3320.2070.1620.179SPUBS0.6070.0750.2830.4540.5730.061-0.2030.373SUBLATE0.3240.7760.1070.7190.3050.632—0.0770.498%Variance22.7417.6615.1155.51%Redundancy20.2211.707.7739.69Table2.10continuednextpageTable2.10continued.CanonicalcorrelationanalysisoftherelationshipbetweenmorphologicalandecologicaldescriptorsofOTUsfromwesternNorthAmericanMenziesiafieldsites.Correlationsbetweenmorphologicalandecologicaldescriptorsandthefirstthreecanonicalvariatesofeachdomain(U1,U2,U3;V1,V21V3)1aswellasintraset(H1)andinterset(H2)descriptorcorrtrnunalitiesforthefirstthreecanonicalvariatesissunimarized.Seetextfordetails.EcologicalDataDomain:V1V2V3H1U1U2U3H2ASPECT0.440—0.327—0.0750.3060.415—0.2660.0540.246AUGTEMP0.1000.7300.0460.5450.0950.594—0.0330.363ELEV-0.751—0.4900.3680.940—0.707—0.399—0.2640.729EXPOSURE—0.1730.4310.2880.299—0.1630.350—0.2060.192JANTEMP0.7310.4480.1420.7550.6890.365—0.1020.618PRECIP0.881—0.165—0.1070.8150.830—0.1350.0770.713SLOPE0.2070.1150.5200.3260.1950.093—0.3720.185SOILDEP—0.3250.5000.3100.452-0.3070.407—0.2220.309SOILORG0.843—0.197—0.0120.7500.795—0.1600.0080.658%Variance32.9717.736.9457.64%Redundancy29.2811.743.5644.58UiviFig.2.17.Distributionof101OTUsofwesternNorthAmerican.ferruginea,sens.lat.,withinthemorphological(Ui,U2)andecological(Vi,V2)domainsofacanonicalcorrelationanalysis.IndividualsarecodedbygeographicregionasdefinedinTable2.2andFig.2.4.Vectorsillustratethecontributionofdescriptorstotheirrespectivedatadomains.SeeTable2.10forfurtherdetails.WRegion:•304z5A617D83 2.••JANTEMP•I1ELEV-3-2-101-223-2-10123U,96•0.0I 0.1___0.2•0.30.40.50.6c j o r-— Ill- > Cl) 0)C) 1 Z j Z Z !-u (/)It-.SUBALPINE TEMPERATE MESOPHYTICFig. 2.18. Dendrogram of eastern North American Menziesiafield sites based on a UPGMA cluster analysis of communitytypes as defined by a Jaccard similarity matrix of speciesassociation. Scale represents fusion levels.Table2.11.CanonicalcorrelationanalysisoftherelationshipbetweenmorphologicalandecologicaldescriptorsofOTUsfromtheeasternNorthAmericanMenziesiafieldsites.Correlationsbetweenmorphologicalandecologicaldescriptorsandthefirstthreecanonicalvariatesofeachdomain(U1,U2,U3;V,V21V3)1aswellasintraset(H1)andinterset(H2)descriptorcommunalitiesforthefirstthreecanonicalvariatesissunmar±zed.Seetextfordetails.MorphologicalDataDomain:U1U2U3H1V1V2V3H2BPUBP—0.516—0.5580.2370.634—0.457—0.4400.1350.421HEIGHT0.8070.3150.2690.8230.7150.2490.1540.597LABOVE0.722—0.457—0.1590.7550.639—0.361—0.0910.547LBELOW0.757—0.404—0.4380.9280.671—0.319—0.2500.615LWIDTH0.695—0.422—0.4280.8440.616—0.333—0.2450.550PEDLEN0.482—0.526—0.0800.5150.427—0.415—0.0460.357PETIOLE0.5730.146—0.3200.4520.5080.115—0.1830.305SPUBS—0.7870.116—0.0440.635—0.6970.091—0.0250.495%Variance45.9415.907.9969.83%Redundancy36.059.902.6148.56EcologicalDataDomain:V1V2V3H1U1U2U3H2AUGTEMP0.882—0.4080.0250.9450.781—0.3220.0140.714JANTEMP0.940—0.1680.1610.9380.833—0.1320.0920.720ELEV—0.1160.8980.2990.909—0.1030.7090.1710.543EXPOSURE0.795—0.018—0.5420.9260.704—0.014—0.3100.592PRECIP0.4610.3630.7260.8710.4080.2860.4150.420%Variance50.3922.6618.7491.79%Redundancy39.5314.126.1359.78ARegion:•2A34•533ELEVI22 1I A AALBELOW—1•HLABOVE-2—2—3I—3I-3-2—10123-3-2—f0123UiviFig.2.19.Distributionof61OTUsofAppalachian.pilosawithinthemorphological(Ui,U2)andecological(Vi,V2)domainsofacanonicalcorrelationanalysis.IndividualsarecodedbygeographicregionasdefinedinTable2.2andFig.2.5.Vectorsillustratethecontributionofdescriptorstotheirrespectivedatadomains.SeeTable2.11forfurtherdetails.1I••.•..SPUBSHEIGHTIPETIOLEI—1PRECIPBPUBPPEDLENAIAUGTEMP2.0•AA A0PC1SEGWIDCALWIDoM.ciliicalyxM. multiflora•M.pentandraIM.purpureaFig.2.20.Distributionof30OTUs,representingthecommonspeciesofJapaneseMenziesia,withinthefirsttwoaxesofaprincipalcomponentsanalysisbasedon20morphologicaldescriptors.Thefirstandsecondaxesaccountfor26.8and19.6%ofthetotalvariance,respectively.Descriptorloadingsonthefirsttwoaxesareillustratedbythevectordiagram;seeTable2.12forfurtherdetails.•LBELOW• .o•0•1L000a.—1—2 -2-1PETIOLENUMVEtN•FRTNUM•CALPUBSPUBGBPUBG2ANGBASE3100Table 2.12. Principal components analysis of Japanese Menziesiausing 20 morphological descriptors. Component loadings, eigenvalues, and the absolute and cumulative variance accounted forby the first 6 components are presented. Descriptor abbreviations as in Table 2.3.Descriptor Component Loadings1 2 3 4 5 6ANGBASE 0.372 —0.504 0.072 —0.564 0.137 0.204ANGTIP —0.708 —0.216 —0.323 —0.342 0.097 0.187BPUBG 0.467 —0.142 —0.472 0.029 0.403 —0.117CALPUB —0.475 —0.305 0.195 0.438 —0.085 0.420CALWID -0.798 —0.247 0.182 —0.034 0.311 —0.135CAPLEN —0.043 0.435 0.360 0.281 0.395 —0.429CILIA 0.698 0.280 —0.171 0.226 0.074 0.308CPUBG 0.205 0.235 0.702 0.345 0.248 —0.089FRTNUM —0.471 0.162 —0.316 0.543 —0.094 0.469LABOVE —0.153 0.866 0.003 —0.173 0.046 0.009LBELOW —0.543 0.780 —0.008 —0.011 0.004 0.020LWIDTH —0.577 0.609 —0.129 —0.323 0.253 0.132NUNVEIN 0.203 0.524 0.038 —0.468 0.342 0.352PEDLEN —0.331 0.611 —0.265 0.357 —0.163 —0.108PEDPUB 0.182 0.147 0.719 —0.334 —0.333 0.080PETIOLE 0.130 0.678 —0.216 —0.319 —0.320 —0.142SEGWID —0.692 —0.170 0.271 0.014 0.483 0.186SPUBG 0.703 —0.111 —0.505 0.135 0.321 —0.083SPUBS 0.707 0.254 0.367 0.097 0.042 0.398SUBLATE 0.750 0.372 0.011 0.154 0.159 0.133Eigenvalues 5.361 3.925 2.278 1.919 1.313 1.189% Variance 26.81 19.63 11.39 9.60 6.57 5.95% Cum. Var. 26.81 46.44 57.83 67.43 74.00 79.95101FERRGLAB_______I PILOPENTLASIMULTCI LIGO YOPuRPYAKUKATSCLADFig. 2.21. Strict consensus tree obtained from 6 maximallyparsimonious cladograms of Japanese and North American taxa ofMenziesia. Identification codes as in Table 2.6. Cladothamnus(CLAD) is the outgroup.FERR GLAB PILO PENT LASI MULT CILI GOYO PURP YAKU KATS/ / / \i34/ /14 \./“ g 8 \./ 3 / 6 // 13 /17 / / 12 \ / /16/ )C10 / / \o //2 / / y3 /61481161312411 7417CLADFig. 2.22. One of the 6 cladograms of I4enziesia taxa, similarto the strict consensus solution (Fig. 2.21), illustrating thechanges in character states (Table 2.6) needed to obtain thetree. Identification codes as in Table 2.6.102Chapter 3Isozyme Analyses of North American Menziesip3.1 IntroductionThis chapter contains the results of an electrophoreticsurvey of isozymes obtained from populations of North AmericanMenziesia. The data are used to examine the levels of geneticvariation in North American Menziesia and how this variationis partitioned within and among populations and taxa.Estimates of gene flow are also derived. Using the data, twospecific questions are addressed:.Does the Rocky Mountainsphase of j. ferruainea differ isozymically from coastal orCascades populations? .. How far has Appalachian. pilosadiverged from western North American Menziesia?3.2 Materials and Methods3.2.1 Selection of Populations and MaterialThe 34 populations chosen for the electrophoretic studyof isozyrnes in Menziesia (Table 3.1) include all populationsused in the morphometric analyses described in Chapter 2, withthe exception of Glacier National Park, BC where suitablematerial was unavailable. In addition, two populations wereadded, namely: Yew Lake, BC (YEW) and Yoho National Park, BC(YNP) . Together the sites encompass most of the range ofMenziesia in North America except for the north coast ofBritish Columbia and Alaska.CDHQcv))0HH0H<cv‘Zj()CDCT)C)ØcvHHCl)o(J)Qt5CDLCD-vCl)HCDH-CDH-H-F-Hcv0Q.ZC)))cvCDZU)C))C))•Cl;cvc_vtj•ZCDC))H,C!)CDcvht-Cl)ICU)0oCDCDHC))N0<cvCD•‘-CDC))I‘tSWZHC))HCl)C))U)ICDC)‘-QCi)C))H01)U)•CDCDI0HCDHC)HQ.<C))0ClC))cvHU)C))0CHH,0C)C)‘1CD‘U)ZCDcvCDC)CDcvU)0cvCD0C))—CD-C)CDNfrCDC))ClCDCD‘1Cl)C)I-QH-i0H,HC))CiIIC))C))C))ZH0cvc-vHU)ClCD U)IiH-H-U)0Mcv5CDCDC)C))0CI)0c-tCD(-vQC)J•QH-CD‘1-CDH,H-H-F—HH-CD0CDCDCl)Cl<cvCDcv-<00c-vCDZCDHCDZU)CDzCDH-CDcvcvcvQCT)c-vU)-CDçcvCD0CDCDCDZ‘1CD0oCDH-Cl)t-cvZCDH-‘cvCDU)IICl)‘10C))CDH,U) CDctHC))CDCDHZC’)C)) T:5-0Ht-CDCDcvCl)HC!)cvC))HZU)<CDc-vCDZH--C)H-HZC))C))c-vC)L0CDU)Cl-HH-Cl)C))ZZU)ClrvC))H-HC)C))Ct3 CD‘1HCl)HHCDCDIClC))C))b•C)C)<CD cvCC))C))‘tiH-C‘HHHH-Cl)C))icrC)(QH-CD0ClCDZ •H ZCD Hi-U)0Z‘CDC))CDH H‘rJCDC))Cl)WcvHCDZC))HC))C)IC))hH-U)C)c-ICC))U)CDC)H-H-cvHH-Cl)<ZH-‘<ZcvCDC)CDcvU)HHC))cv‘-<‘OI-cvCD1iC)—1CDC))<C))I-h0(QU’‘-QC)Z0ZI-C))U)QCl)H-U)0H0CDHc-vHHI-0•H-CDCl)CD‘1CZ‘‘1H<H-cvC)(QhCI)cvZC)C))Hc-vCDU)0t5ZCDcvCl)cvcvHHClCDZoHCT)cv‘-0CDCDJ0HCC))C)H-H,H-CDHMcvC))U)C))QH,H-CDZClC))H,‘1CDClClC))HoHC))cvCDNClC))‘t3H-CDHCD-0Cl)Z0II‘tihcCl)Z‘<ZCDcvC)cvCC)‘tIH-CDHCD‘tiCDCD0CHHcvC))I-ZClC))H-U)U)C))CD‘tiC)CDZC))cvU)C))H-cvQQC)ZcvC))‘C:,C)H(QIIH-HCD0Z•C))ZCDHCl)HcvZCDU)U)HcvCl)Cl0CDZCDCDU)0C)CDC)H-CDC)cvcvocvZI-ZU)(UCl)ZH-•CDClClcvZcvC)CH-ZH,ClCDI-Cl)CDC)ClI-k<cvCDZ‘ti0CU)CD0cvCD‘ti-‘CDClHZC)‘t3ClHCl)cvC))HC))CD-C)ZcvC))C)‘-CDH-CDCDU)oU)cvHCDH-C))cvCl)cvZH-HU)HCDCD ‘1 CDClH-CD‘ti‘1‘tiCDbCDCDClcvH,otiCDCDC)U)CDZcvi-H-0CD‘ticvH-Z•CDZCD‘ti(Qcv-H C))C)Cl)ti•0C))•HCD H-C) H CD U) Cl)0 H,C)) U)cv ZC)H-CU)HH-HcvCDZC))C)HZcvZ<CD ClCl)H-C));iMC)ClMH-0< H-0ClC))‘tiC‘tiCC))‘HHC))U)0cv>H‘tiH-0CDZI-cC))-cv‘tiCD0HZ•‘ti‘<CD i-cHWCDC))C)<cvCDH-i-c0CDZH-0H-ZU)ClU)‘tiH-Hl-<lJ•0H-H<ClCDCDCClC)) HHcvU)CD-•Cl)CDH,C)cvZZH-C))CDU)ZU)C)CCCDCl)CDCl)CD0H-CrU)C)ti—1CrQ9)tQrjF—.0H,CflCD0000nH-9)CDtiCDH0IICD9)CrH-CD-tiU)IIH-Cr‘1CD9)Q.I-’9)9)C)Q•IH,CDU)Q.CDCrU)9)CDU)CD0C)0U)H-I-H0‘-HH-C)-Cr•H-CD0HH-“jHH-09)H,Cr9)H-H0H-H-Q9)CDCDZCDHCrH-CHti9)C)ZCr9)(1(L)CDLHU)CDCr(QIHH-HI-i<H0tt1CrH-QCrU)XHCrU)-CD0H-QIIHI<CDH-H-0-iCr00CrH,0M-IH-ç-tQCrH-CD9)H,Cr1Cr-HI-9)09)H-ICrCrCr0Ø9)o0C)-C)IF-CL)09)CD1CDI-0U)CDH-QCDCDZCr.DCrICI-cCCDt-H-<CrCDCDIIC)H-CDC)CrCrH,CDH-U)U)CD9)H0()tiU)-QC)U)tiP‘II-c0I-Qfr’1jH1Cr•HH-tiCDtiCrCD0C:I-H9)00H9)00U)CD9)CDC:CDCDI-chcI-CDtl9)tC)NH-(DC)C:9)U)U)HI-I-CDCDC:r-rCDII-CrCrU)CDCr9)H-QH-H-CrCD9)CrH,H-i(QH-H-H-(Q9)CD9)U)H-Crç-rH-HCrCDH,CHCrC)I-r-rQ.QH-><CD9)00-ci(D‘-D9)9)CD0C:H-ciCrof-i-9)C:IIH-—]I-’LHI-I-hU)0CrwC)H-I—hU)9)H-0CDI—’Ui0C)-Cr9)hIl)9)U)—C:00CrCDI-hC)CD)QCD0HCDCrCDU)-HCDciCDCD0•CDtiI-Cl)9)CDCrCrh9)c-i-cCDU)rr09)C)CDH9)9)9)9)C:9)0‘1CDII<H,QH0ZCDCOi-cU)Hci9)N009)cif-rU)‘l(DH-HH-CDci-C)CD<C:<I-hCDCrciH-HHH-CT)i-cCOCl)H-9)<H-H-9)9)H-9)H-Ci-U)9)ciCDrtiCDCDH0CrCD0H-CtHIIciCrQCDCDH9)tQ9)C)ciCOCDI-jtQH-CD(9)<CrC)CDSk<i-Q.9)CDH--9)0(QIti00ci0U)HCrI-hM02Ci-HiCDiH-CrCrHiHCO-9)Cr0I-’-I-QU)SU)i-c0CD0C:H-H-CrI-9)‘-1HCt020CDU)05CrciCDU)HHCDHH-U)CD9)i-cC)H-Ci-U)CDC:l.Q9)H-<I-CDCO0CD9)5Cr0I-ICr9)NCDHCrHH-9)H,COH-.CrH‘ti9)C:9)CDCD05U)Hc-i-5H-C)I-C:0CrH-HHHCi-5H-l-HH-Ci-LQCDCDLIMiCDiH-H-ciH-I-MiCrHCDCD00CDCl)CDU)Cr0CDH-CDC))HCDMiCr•0U)NCrU)tQ‘-<U)CDCD•ciCr(QCDCUC)U)H-CDHH-CDICDClI-ciCD5Cr9)CD-009)0ciI-cC)Z0•‘-3tiH-CDCDH,I—hU)H,CrN09)CDHU)H0)5“<I-9)9)‘<HtiCDCrCD9)CDU)9)CDCDII-cU)CrH9)5HI-U)-C:C)CrC)lCDCD(DH-HCD9)CDHCDCDHCrCrCD5CD0C:NIU)CD<CDH-ciC)H-CDCri-ct’3I-5U)<INC:ciCD<Cr<<CrCD9)ci9)9)U)‘itiCD5IH-HCDH-9)9)9)CDiCrHCO0“<Cri-w9)9)ciCDlCDCr09)CrhCrH-U)CD0IHI-U)COCOH-,CrH-Cr9)Cll-N.CDH-CrH-•CD9)9)HH-CQtJH-CDI-jC9)9)COCr(QCDCOCOCDHiciCDHCDH-CD9)CDCDU)I-LHi—c9)I—cCr0I-h1-cCDCDN0CriCrCD<CDCD‘tiCrCD0I-CDciciU)COCDCDciH C)105tissue was extracted by grinding only a few samples (10-12) ata time, after they had been placed, still frozen, into coldbuffer. In practice, 7-8 drops of grinding buffer were neededto make a good homogenate from the frozen tissues. Thegrinding buffer itself, was made up in bulk (125 ml) and keptrefrigerated at 4°C for up to three weeks, and contained: 125ml of 0.1M Tris[hydroxymethyl]aminomethane-HC1, pH 7.5; 0.5gof EDTA (ethylenediaminetetraacetic acid, tetrasodium salt);0.95 g of KC1; and 0.25 g of MgC12 6H20. Before use, 1.5g ofpolyvinylpyrrolidone (PVP; MW 40000) was added to 12.5 ml ofbuffer and allowed to dissolve. In addition, 7-8 drops(approx. 0.125 ml) of 2-mercaptoethanol were added by Pasteurpipette to the buffer just prior to use. A teflon-tippedgrinder was used to macerate the tissue in ceramic spot wells,that were placed on a tray of compacted crushed ice.Supernatant solutions of the plant extracts were then taken upwith Whatman 3MW filter paper wicks (3 x 10 mm), which werethen loaded onto 12.5% starch gels.Sixteen enzyme systems were examined in all of thepopulations: aspartate amino transferase (AAT), E.C. 2.6.1.1;aldolase (ALD), E.C. 4.1.2.13; glucose-6-phosphatedehydrogenase (G6PDH), E.C. 1.1.1.49; glutamate dehydrogenase(GDH), E.C. 1.4.1.2; glyceraldehyde-3-phosphate dehydrogenase(G3PDH), E.C. 1.2.1.12; hexokinase (HK), E.C. 2.7.1.1;isocitrate dehydrogenase (IDH), E.C. 1.1.1.42; leucine aminopeptidase (LAP), E.C. 3.4.11.1; malate dehydrogenase (MDH),E.C. 1.1.1.37; malic enzyme (ME), E.C. 1.1.1.40;106phosphoglucoisomerase (PGI) = glucose-6-phosphate isomerase(GPI), E.C. 5.3.1.9; phosphoglucomutase (PGM), E.C. 5.4.2.2;6-phospogluconate dehydrogenase (6PGD), E.C. 1.1.1.44;shikimate dehydrogenase (SkDH), E.C. 1.1.1.25; superoxidedismutase (SOD), E.C. 1.15.1.1; and triose phosphate isomerase(TPI), E.C. 5.3.1.1.The gel and electrode buffers used for electrophoreticseparation are presented in Table 3.2. Typically, enzymeswere resolved on systems described by Soltis et al. (1983):system 1 for G6PDH, G3PDH, IDH, and SkDH; system 6 for ALD,AAT, GDH, and ME; and system 9 for PGM and occasionally 6PGDor MDH. A modification of system 8 gel and electrode buffers(Hauffler 1985) was used for HK, LAP, PGI, SOD, and TPI. Amorpholine-citrate electrode buffer (system M), modified fromWendel and Weeden (1989), was used routinely for resolving MDHand 6PGD.The enzymes were resolved using staining recipesdescribed in Soltis et al. (1983), with the exception of GDHwhere the amounts of MTT (3-[4,5-dimethylthiazol-2]yl-2,5-diphenyltetrazolium bromide) and PMS (phenazine methosulfate)were doubled to increase staining intensity. For LAP, therecipes of either Soltis and Rieseberg (1986) or Cheliak andPitel (1984) were followed. Agarose overlays were used forALD and TPI, from which SOD was resolved as an achromatic bandon the gel. Occasionally, enzymes were resolved on otherbuffer systems to help interpret complex banding patterns andto test for hidden heterogeneity (Kephart 1990). This was107done particularly when comparing co-migrating bands betweenthe western North American and Appalachian samples. Bandingpatterns were scored by numbering sequentially from thefastest anodally-migrating isozyme; similarly, allozymes of agiven locus were labelled alphabetically starting with thefastest anodally-migrating form.3.2.3 Analysis of Genetic VariationFive measures of genetic variation within populationswere applied to the elect.rophoretic data: the mean number ofalleles per locus, including monomorphic loci (A); theproportion of loci observed to be polymorphic, where therarest allele occurred with a frequency of 0.01 or greater(P); observed heterozygosity (Hobs); expected heterozygosity(Hexp) where Hexp = 1 - Zp12, and p is the frequency of thejth allele; and the mean fixation index (F) whereF = [1- (Hobs / Hexp)]. Chi-square tests were performed todetermine whether the observed heterozygosity of a populationdeviated significantly from expected heterozygosity based onHardy-Weinberg equilibrium expectations (Ayala 1982).Averages of the genetic variation statistics for the threeNorth American taxa also were compared with one-way ANOVA andTukey multiple comparison tests (Zar 1984).Genetic diversity among populations was analyzed usingNei’s (1972, 1973) gene diversity statistics; where HT is thetotal gene diversity within a taxon, Hs is the gene diversitywithin populations of a taxon, DST is the gene diversity108between populations within a taxon, and GST is the coefficientof gene differentiation. Note that HT = H5 + DST and thatGST DST / HT. Standard genetic distances and identities,unbiased by sample size, among populations were alsodetermined (Nei 1972, 1973). These calculations wereperformed using the GENESTAT-PC program (version 2.1 by P.Lewis and R. Whitkus; Whitkus 1988). Populations were groupedbased on genetic identities with unweighted pair-group (UPGMA)cluster analysis using the NTSYS-pc subroutine SAHN (Rohif1988)Estimates of gene flow (Nm) among populations usingallele frequencies were also calculated by three differentmethods. The first method employs Wright’s FST statistic,where FST = 1 / (4Nm + 1) (Wright 1951; Futuyma 1979). Thisapproach depends largely on the distribution of common allelesbecause of the calculation of FST V / p (1 - p)j; where (V)is the variance in gene frequency among populations and (p) isthe mean gene frequency. The value of FST is calculated oneach allele at a locus. A slightly different approachsubstitutes the values of GST for FST in the formula, with acorrection factor for the population sample size (Crow 1986).Another, more recent method of estimating gene flow (Slatkin1985), uses the frequency of private alleles, that is, allelesthat appear in only one population. It is estimated by theformula:ln(p(1)) = -0.505 (ln Nm) - 2.44;109where: p(1) = frequency of private alleles;N = population size;m = migration rate.All of the methods of estimating gene flow are dependant oncalculating the average number of migrants exchanged betweenpopulations, since N (the population size) is rarely knownwith any degree of certainty.3.3 Results3.3.1 Interpretation of Isozyme Banding PatternsOf the 16 enzyme systems examined, 13 were interpretablefor all of the 34 populations of North American Menziesiastudied (Fig. 3.1). They presumably coded for 19 loci: AAT-l(a-b); AAT-2 (a); ALD-l (a); ALD-2 (a); G3PDH-l (a); HK-2 (a);IDH-1 (a-c); LAP-l (a-d); MDH-3 (a-d); MDH-4 (a-c); PGI-i (a);PGI-2 (a-g); PGM-l (a-c); PGM-2 (a-b); 6PGD-l (a-d); 6PGD-2(a-d); SkDH-l (a-c); SOD-i (a); and TPI-i (a-b). A second,poorly resolved, locus of G3PDH was seen but could not beinterpreted. Although not always visible, faint fast anodalbands, HK-l, LAP-2, MDH-l, and MDH-2, were apparentlymonomorphic. In MDH, PGM, and TPI, extra, usually faint,bands were occasionally observed. In PGM these turned out tobe ghost bands, artifacts likely resulting from the extractionprocedure, and appeared most frequently when frozen or agedtissue was used. Similarly, one or two faint bands were seenfrequently in TPI but only TPI-l was strongly stained andconsistently scorable. These faint bands appeared mostcommonly in the western North American populations but were110occasionally present in Appalachian populations as well.However in MDH, the extra bands were observed only in someindividuals and appeared regardless of whether duplicatetissues were run on systems 9 or Morpholine. Similaradditional bands in MDH have been reported in diploid membersof Vaccinium section Cvanococcus, for which the genetic basisis not known (van Heemstra et al. 1991).Three enzyme systems were not retained in the analysesbecause of difficulties in interpreting the banding patternsobserved. These include G6PDH (1-2) and ME-l which wereapparently monomorphic loci but were sometimes poorly resolvedwhen frozen tissue was used. As well, GDH was not retainedbecause it was only faintly stained in some populations. Whenresolved, multiple banding patterns, possibly coded by onelocus, were observed.Of the 19 loci scored, seven were monomorphic: AAT-2,ALD-l, ALD-2, G3PDH-l, HK-2, PGI-1, and SOD-l (Fig. 3.1). Oneenzyme system locus (AAT-l) was invariant within populationsand was fixed for AAT-la in the west and for AAT-lb in theAppalachians. The other 11 enzyme loci examined werepolymorphic in at least some of the populations. Bandingpatterns of IDH, MDH, PGI, 6PGD, and TPI were either onebanded or three-banded at each locus, consistent with adimeric enzyme structure (Fig. 3.1). Likewise, LAP, PGM, andSkDH had one-banded or two-banded patterns at each locus,indicative of monomeric enzymes (Fig. 3.1). These results arein agreement with quaternary structures commonly reported for111these systems in other vascular plants (Weeden and Wendel1989)Seedling enzyme banding patterns, when compared to knownmaternal patterns, supported the genetic interpretation ofthese enzyme systems. Extra bands appearing in MDH-3 in somematernal plants were similarly inherited by a portion of theirprogeny, though reliable ratios were not obtained owing to thesmall number of seedlings examined. All of the enzyme locihad banding patterns consistent with expectations for diploidindividuals. For several enzyme systems, many of the commonallozymes observed in Appalachian i. oilosa were scored asbeing identical to allozymes seen in western North AmericanMenziesia because they would co—migrate in different buffersystems.3.3.2 Genetic Variation Within PopulationsStatistics measuring levels of genetic variation withinpopulations (Table 3.3) reveal a high degree of similarityamong the western populations of . ferrupinea, sens. lat.The mean number of alleles per population (A) ranged from 1.21at Prairie Creek in Humboldt Co., CA (HUM) to 1.74 at YewLake, BC (YEW). There was no significant difference betweenthe overall ranges of (A) between coastal and Cascadespopulations of . ferrupinea ssp. ferructinea, and the interiorpopulations of . ferruginea ssp. alabella. The same trendwas observed for the number of polymorphic loci per population(P), which varied from 0.158 (HUM) to 0.474 (OSW). Although112the average number of observed heterozygotes was slightlylower in coastal . ferrupinea (0.061) than in populationsfrom the Rockies (0,070), the difference was not significant.Observed mean heterozygosity was highest (0.104) at YohoNational Park, BC (YNP) and lowest (0.022) at Prairie Creek,CA (HUM). In most of the populations examined, the observedheterozygosity deviated significantly from Hardy-Weinbergexpectations (Hexp). This is reflected in the averagefixation index (F) for western Menziesia populations whichranged from 0.102 in Manning Park, BC (ROT) to 0.576 at MurtleLake, BC (MUR). The positive values of (F) indicate that afair degree of inbreeding exists in each population. Thelowest degree of genetic variation, as measured by theseparameters, occurred at Prairie Creek, CA (HUM) on the coastand at String Lake, WY (TET) in the Rockies. Thesepopulations are at the southernmost limits of distribution ofMenziesia in western North America.Levels of genetic variation in Appalachian . o±losa wereoften lower, in most instances significantly so, than thoseobserved in western North American Menziesia (Table 3.3). Themean number of alleles per population (A) in . pilosa rangedfrom 1.16 at Capon Springs, WV (CAP) to 1.58 at WhitesidesMountain, NC (WSM). The proportion of polymorphic loci (P)varied from 0.158 (CAP, WSM, LWS) to 0.474 (JEN). While theobserved heterozygosity for . oilosa averaged 0.051, this wasnot significantly lower than the mean observed value incoastal N. ferruinea (0.061), but was significantly lower113than the mean observed in the interior phase, N. ferruaineassp. alabella (0.070) . Interestingly, the observedheterozygosity of most Appalachian populations, while slightlylower than expected, did not deviate significantly from levelscalculated using the Hardy-Weinberg model. Although there wasa tendency for inbreeding in the Appalachian populationssurveyed, it was lower than observed in the west, withfixation indices (F) varying from -0.170 at White TopMountain, VA (WTM) to 0.481 at Mt. Pisgah, NC (PIS). Thenegative value of (F) at WTM indicates a high degree of randommating in this population.When examined over all populations, some loci (PGI-2,6PGD-l, TPI-l), had heterozygote frequencies that did notdeviate much from Hardy-Weinberg expectations (Table 3.4).Indeed, PGI-2 showed an excess of heterozygotes in severalpopulations, particularly in . pilosa. Conversely, LAP-i,MDH-4, PGM-1, 6PGD-2, and SkDH had significantly fewerobserved heterozygotes than expected in many populations of .ferruainea, sens. lat., as reflected in the fixation indices(F). The remaining loci exhibited variable deviations fromrandom mating.Clinal patterns in allele frequencies were also observedin several of the enzyme systems (Table 3.5). In the west,most of these trends varied along a west-east gradient. Forexample, PGI-2a and PGI-2d were commonly observed only incoastal and Cascades populations and occurred infrequently inWells Gray Park, BC (TRO and MUR). In contrast, PGI-2e was114seen only in a single individual at Moscow Mountain, ID (MOS)while PGI-2g was locally common in the northern Rockies (MUR,LL, YNP, WAT). While 6PGD-l was fixed for 6PGD-lc in coastaland northern Cascades populations of . ferruainea, it waspolymorphic in many of the interior populations, and highly soin the Rockies. A similar trend of increasing polymorphismwas observed in 6PGD-2, and IDH-l. In general, the number ofobserved alleles and their frequency of occurrence becamelower towards the southern part of the range of . ferrupinea,both along the coast and in the Rockies.In the Appalachians, PGM—2b occurred only in thenorthernmost populations (WBR, DOL, JEN, MIN) while PGI-2a wasfound exclusively in the southern Appalachians. As well,6PGD-l was usually highly polymorphic while 6PGD-2 was not asvariable as observed in . ferrupinea. Polymorphism wascommon in TPI-l in j. pilosa, whereas it was fixed in M.ferrupinea, sens. lat. Both IDH and SkDH were invariant inthe Appalachians. The number of observed alleles and degreeof polymorphism was usually lower in populations located inhigh altitude heath balds relative to cove woods populations,except at Capon Springs, WV (CAP).3.3.3 Genetic Variation Among PopulationsGenetic variation among populations was assessed usingNel’s genetic diversity statistics (Table 3.6). For most lociexcept MDH-3, total gene diversity (HT) was highest in eithercoastal or interior populations of . ferrupinea relative toZiC5QC)COC)CrtQC)IQIM-‘H)jH-I—iH-H’00CD(1)H-(0H-•MiCDIF-IkDokD‘H-H-jJQCrC)I-(Q‘-C)CDCrF-P1iCDIP)IhZh-Ij—crCrtQI-CDC)Cr•a’(QCDIIyIt-iIhloF-CDCX)P1aiCDlCDçjIIcoH-CrCDP1CDC)iHpH-pI—p0JH-CCDh))Cr0HIP)13H-CtCDP1CDZCrCrF-H-C)b-CD5C)Ii-h-a-CrICfl000I-hCD0MiCoCoP1H’o(i)CrCr(1)XCDH’(,QH-H 0CO-C)X‘t3CnCrCDCOCD-h-P1b-tiCO•r—-CD0H-P1CDIICDCDlCDCDCDI—’(QH-P10CDIP1)t-It-’I-0I-H’CO—frF-P1P1CII-MiCDQh)IhCl)‘t3COCDCDHCrIICrCDHH-H-P1•(I)CDWH’CrJCOCDCrCDH-CDI.Q•P1IH-ttiH-C‘IQCrF-CrC)0C)H-IIj-J-<P1<Q.0ItiCDP1•CtHI-H’C)H’ZoI-CDCO•C)SCrC)CoWCDCDJ,)‘H)Cl)iiP1H-tQCDH-H-CrCoIt-iCtØ.H’WCrCD‘CDH-CDC)P1-0H-CD0COp1H-ItCI)H-—--CiP1HH’i-b-ID—-Cl)CDI(i<C)—P1C)P1CDIfrH’H-H-I—tiCOCl)CO‘rJP1CrIQCD•Cl)CrCOCDHH0P1‘<P1CDP1CDIH-C)F-H—P1P1CrCDCOP15IQP1H’—‘1H-1i•<C)XCrCrkt)C)•QP1CDH-C)0IH-MP1Co-3H-ClHIP)C)P1H-5COIiH-HP1oQ.H’H-CrICr)CrH-I-I-<IIQH-CrH-‘<-I-hI-)))<1<)I’Jft-iC)H-H-C)P1H0(QQH-CDP1COP1-IP1CO0P1I-hICDHH-HiCI)‘.QCD‘tiH-SPiCD0HI-HCDCrH’Cl0IH-0Cr-H-COCr0CDCP1CDP1COCOC)0MiCoH-CDCDLQhjC)H-IS)QCDCl)‘(3‘<HH5C)MiP1b—MCD-ISCtICrCr1(3CI)‘1CDCDWI-ICI)H’frCr‘-C)-‘(3CD0H-H(3IH-I-CDCDCl•C)H’0CDCDH-IS))h5F-)<C)CI)CDIF-A—]•P15(QClI-hCrP1(3I-H-I-pçICrbrrCD—CDi—CrPIH-H-0p1IIHH’I-’-CCI)C)P1•IS)C)I-hCDH-HI-hIHCOhCDCl0CDCDCI)•oCrQc-r0CDH0HCiI-CDI-C)5çC)0C)CoCDCD:iCDP1P1CO01COCDH-—1H’5CD(3CoCO<Ci00Cr•P)‘(3CICo1(3°C)Cl0CrI-ClCDC)HiH-,H-CrCOiP1CD-C)CrHhJH0CrH-H-°ClCMiC)H-C)P1—CDSP1H-MiClC)HMi‘Hi<0-0CI(.QP1CrCD1(flCD0P1lCDC)CDCI)CD—CDiCOH’-IP1IIS)CrCrc-r(I)II-HhCo00Cr0ClHCD-‘••CrH-II-CiCOCr0t3CD<H-C)I0H’—CD0CD<IH-CI)MiH’I-CDHCtClCO•1)‘(3C)5Hi-rCDIQMiCrH<CDt-Cl)CDCOMi—CDP1‘(3H’P1I-CD00CDItsi-Mi•IP1CI)WCII-Cri-•CDtsIICl)C3CDICrlCDP10C)COHMiP1P1COIC))I5iCiH’COH•CDtsH-0P1HCoiH•H’IS)CI)CCDH’•C)CDCrI-IH-0<C)CDwI-5ClH-P1(Q—HiCrH•05—)Cr0H-COCOCrH-tjlC)HCD)QCrCrCDH-HbC)CDC!)CO(QH,CoP1iHCrCrCOH-Cts—(Q0CDCDIC))H-h‘(300CD<CDCDCrP1CD•MiI-CDP10CDH-HI-tiC)CrCrH0-•c-n116yields similar results, with two major groups, correspondingto N. pilosa and N, ferruinea, sens. lat., being mostapparent. In pooled calculations, N. pilosa is marginallymore similar genetically to coastal N. ferrupinea ssp.ferrupinea (I 0.933) than it is to interior N. ferruaineassp. alabella (I = 0.924) (Table 3.8) . A high copheneticcorrelation coefficient (r = 0.924) indicates that thedendrogram accurately portrays the relationships defined bythe genetic identity matrix. Most of the variation observedamong populations occurs in western N. ferruainea, andalthough some partitioning of the coastal and interiorpopulations is evident, there is considerable intermixing ofthe two groups. Populations from the Cascades (GVC, ROT, SKY,STV) are scattered over several branches of the N. ferrupineacluster. Consequently, the genetic identity between the twowestern subspecies is very high (0.990) (Table 3.8). The mostdistinct populations in the west are located along the maincordillera of the B.C. Rockies (LL, YNP, WAT). In contrast,N. pilosa appears to be a largely homogenous group with a muchnarrower range of genetic identities (Fig. 3.3; Table 3.7).There appears to be complete intermixing of populations fromthe northern and southern Blue Ridge populations and betweensubalpine and temperate mesophytic sites.3.3.4 Estimates of Gene Flow in MenziesiaEstimates of gene flow (Nm) in Menziesia based onWright’s (1951) FST statistics range from 1.90 for N.‘tSH-‘tiH-WWHU)CDOU)bZCDCU)‘M00iCDU)0H-f-a-••DCDH-U)H-•U)))0H-•U)U)1(1)tICIi<U)CrCDHCDCrCQCDCIU)CQH-tIfr-iCF-CrCDP)CDCr•C_flP)IIH-H-CD•Cl)Cr.IHQ.CDiH0-.CrF-’-HCD—CrIH-(QIHH,P)tQCDI-iCr(7)I-•H-C)C))C))U)H-CDU)CD000CD•CDCDH-)CrCD‘tiH-CDU)<CD0a)Ci)<)Iiç1ç1f-Q(7)IHI-jU)Cl)IiIIiilCD0tIi-iHCCrCDHHU)CDCDC))U)U)C))IiClCr(Da)Cr<IU)CDlCDcnIihiIc’U)C)CrC)H-H-rra)H-CrCDCr0QH-CrC3IHHH-C))CrCDt’JH-CDIt—C)iC)Cr0C)H-U)0IH,CrCr00H-CrCDU)I-hH,U)C))U)Ii<ZCD‘-0CrH-H,H-C))IC))C))CCrC))H-C))<H0lCDtiIH-YH-0tiI-IøClC’)Cl-CDI-jIHCDkDIiC)0QI-jH,CH-I-ClH-I—’Q.,CQ0-HJH-INC)HC)H-HU)H0HH-CDiHHC)CDH-CrC)WCDC))CDC))U)CD0HIH-H-C‘<0H-rr0IIC))U)QCDCr—‘H-CDIt3IC’)H-CDClH‘tiC))H-0CDCDCl)CDCrCD.DU)ICDIF—lCD‘-IIU)HI-’-0CD‘-l-HL—HlCDC))HHCDH-•C))IF-IH-CQH-H-DCDHCrIC))U)CDC)‘<H0H-’C))CDiC_flCDIiWIC))F-’I-I-.CDC))H-CDC)CDCD0CrCDH-H-H-U)(1C)H-CDSIiU)—0<CrCltiCDCDC))CQCrH-0C))0Ii0CDC))ICC))—CD‘(5I-’-IiH-CDJCDCl(5CS•C))CrH-h’-DH-,H-F--t00—I-ifCDHCrHCt5W0CDC)U)0U)CDHU)CDU)U)C)CDIi‘(5C))CrCCDClF-CDCDH-C)•CrClH-C(5C‘5H,H-’CD—0Øi0CDC)ZI—’0H-’C))C))U)H,CZC))W<CDC))IiH,CDC)‘dH-0CtCrHU)5‘1•CDC))rrIiH,(7)U)H-CDCrC)CDH-0CDCDHClCl)0CQH-CC))HClH-I-’HQ,5ClIICDU)HC)I—CCD<C)U)H-00)••-t---—CCrCDC)CDCl-iH5CD<CrWH--5‘(5CDC)CDCDC)‘-0CQU)1’-D0C))CDCD50CDCl0CrU)Cr00CDHU)r’C))CDH-CQIiC))H-CrU)I0U)H-‘-DCQH-C<C)HCrCDQjQ-C’)C)H-‘(5Cl<U)a)—CDCDCDIHC))CrCD00CrC))U)•‘(5QU)wi-ClCDf-tIC))HC))<U)5CtC)0H-•C))CrC)I(5-C)CHC))CDC)HHI-CCD0‘(5C)0C-.)c-QC)CrHCDCDI-CCDH-H,CDC)HCrU)0LH-I-CHH-H-5F-CC))H-,CDCD‘<Cl0HH‘-<U)lC))CDU)C))-OCDC)H-ZlIt0iCkIC)çrC))Crc-f—C))F-Cr<5C))H,••‘(5•(QC))Il-hH-CfC))CD•CDH-CDIi0CDCt--kD0CCQClCrClCr0H-CDIiC)-)I(5IH,U)HMC)II-jU)IH-kDCrC))Ci)IiCIiiI-CCDCD0CDIC))HC)-C’)H,5IHhiH-CfH-CHICU)H,U)lCDHH-0I-CCrIClI-CSH-5C)IQH--03Ii.H,H—C))0U)H-IC))ICC))0C)CrIHCI-I-CI-CINCDC))Cl)‘-CCrlCDIC))IQCCDHCD0‘oH-iU)(5lCD‘-0‘-QC))CrU)0CDCIH-HI-HCrCDU)•ZCDrrCDCr-zU)‘<H-‘-0I—’U)5H,H-•‘<H-‘<HIIH H118Levels of within-population allozyme variation were verysimilar in the two subspecies of N. ferruainea (Table 3.11).Although these levels are low, compared to mean valuesreported for widely distributed species, they are very closeto average values for regionally distributed species (Table3.11; Hamrick and Godt 1989). The mean levels of withinpopulation allozyme variation in Appalachian N. oilosa aremost similar to values observed in endemic species (Table3.11; Hamrick and Godt 1989). In general, the levels ofwithin-population genetic variation are somewhat lower thanpredicted on the basis of geographic distribution, especiallythe levels of expected heterozygosity (Hexp) . Lower geneticvariability in N. oilosa might reflect its smallerdistribution relative to N. ferrupinea. The results arecertainly consistent with the more homogenous morphology(Chapter 2) and simpler flavonoid profiles (Bohm et al. 1984)seen in N. pilosa.With regard to life form, the values of within-populationallozyme variation observed in North American Menziesia areagain lower than expected for long-lived woody perennials(Table 3.11). However, they are comparable to levels observedin short-lived woody perennials (Table 3.11; Hamrick and Godt1989) . It should be noted that most of the studies of longlived woody perennials upon which these statistics are basedare wind-pollinated tree species, particularly gymnosperms.Few studies have focused on short-lived woody species, andconsequently, the data base may not be entirely reliable.CDCrrCDO)C)QHcrC)H5CCoZCl)>CH,H-CWCCD‘-.0“<CH-‘-.00)b’C0COCl)‘ZCl)CDtQCDHCLCl)‘.0tIçtCX)C-tCoHCDhSCCfCl)II‘-<NCDCDH-‘.0H-CDHC)CfCfH-HCrCCr<CQ5H-0C-f—C)Cr3I-iH-H-3Cl)COCCDH-CCD<C)H-Cl)CDCQ<CDZCDCl)CrCDi—hHCDtII-IICoCl)CC)H0XCDCl)COCDc-fH-l‘.0frCCoC3H-CtC)CDH,)CLCr-i-H-C<‘‘.0CDHS—Cl)CfC)<H,CDHCDCDCDCDCCDCl)HCO<H-H-)Cts-)C)H-H-Ql-l-CO3CoQ.l-Cr5lJC)Cl)Hl-C)I-iCrCL5Cl)H-Cf-Cf•H-CH‘.0H-CCl)CDCl).—.hC)HCH-5l-HHCD‘.0—Cl)5CLIICfCDHIc-fCl)CDCri--hCDCDC)Cl)CDH-CF-’WC-ftIH-H-CLI•Cl)H-HCOCrCrH-CC)CfIIH-0)tIt-JC—tC)CDCH,‘tICDHH-H-CCl.H--CDCHGCICHCOH,H-CDI-Cl)CC)CDCDCCoCrHH-CDHCDCOC)ChH-C-tCLH-H,HHCl)H,0)H-NCDCoHCDCoCLCH,CDH-CL-F—NCDc-fCl)‘<H-CHCCDI-iCrCfCOI-CCo-HH-I.QH-HH,NF-’CDIIHCl)C‘1Cl)Cl)H<CDCDCDCHCCfCoCDHCDCrCl)C-fI—’C!)‘-<CoCDCr55tIC)COCDH-CH-CLCDCl)-.CrH-CDH-l-CDCD•Cr<CDCoCrCfH-1QII0)CrH-H-CfCDCDC)C-CDCrCrNtIH-CDCH-CD<CoCOCrH--HH-CDH-H-F-’CO0)COC)C)CtCLCrH‘-<Cl)CXCoCl)C)IICCQ‘SSCDCCDH-COCDCOCDH—Cl)HCOH-CDChI5hIC)CDCl)<CrHtC-CDCr‘-CLC)IIH-Cr‘Cl)CCDCDhICDCDCo‘<H-HH,Cl)hICl)H-CDC)Cl)hINCoH,CDH5c-fCDF-’hIH-CDH-CrCDCl)hIH-CDCoCoCl)tC‘.0CDC)—C-tCoCOc-fCDCl)CQ—hIH-Cf-CrH-WCD‘.00)CLCDH-Cl)C!)ChICH-(Q•H‘C)Cl)H-hI5Cl)H-c-fCl)CDH,CoH-HCD—CDC<<H-CLhIH,F-’CDCo‘Cl.F-’Cr-tCCLHH-CoCl)HCDCl)H-CoHH-CDCo<H-H--—H-HCCCL‘Cl)C)<C-fCD<H-Cl)C)C)C1•C)Cl)H-CohII-CHI-CCC‘-<CDCD(QhICoCfCoCDCl)‘CDCLCDWCl)I-jH-hIH<CoCl)Cl)F-’CoCfCtc-fhIH-•-Cl)H-Cl)5Cl))HCl)CDCDCOCDL)CC)Cl)HHH,CrCl)hIhICLhICLhI•HhICr(C—COHH-‘<CDCL5CDH-CD—<H<‘.0CHCDCC)CCl)H-WCDHCl)(X)Co1<H-ChIH,Cl)Cl)‘CrI<CoI-hI‘CL-•H‘-.0CoCDCOCl)Cl)hIH-CCH-—CDCo<•CfCr—<CoCL><CCJH,C)F-’Cl)CD•CLH-‘CDCDHCiCDCDH-tICCDhICl)CoCrCI-iCl)HCD<CrCLH,CHH-hICl)5CD5Cl)<H0)I-hIhICHCl)Cl)I-Co<CDCDCrH0)CDC0)HCr‘H-CLC)H-CDCl)CLCHCLCL•CCoCrH-H-Co5C)CC)CDHCf0)CLCrCCD5H-F-’0)CrCl)CrNCDHCShI—C)‘.0hICoCDCDH-CrH-hICl)CrHCnCO‘CL0iCH-Cl)H-CoH-hICCl)CDCDH‘.0CCDCrCDCLo5Coi5CoCfCoHCLCL‘<‘-C>H,CLhIC)CCrCLCl)CoCoCH-CDHCl)CfCoCDSHCD5Cl.H-—H-H-H-C0H-I><CCDCD3•CLCCo5CrC><H,F-’H-HCDCLCDCl)I-’CoCOCrCLH‘.0120loci (LAP-i, MDH-4, PGM-l, 6PGD-2, and SkDH-1) than wasexpected from the allele frequencies. This is suggestive ofeither significant levels of inbreeding in the populations orof selection against certain enzyme phenotypes at these loci.In contrast, other loci (MDH-3, PGI-2, TPI-1), generallyexhibited a slight excess of heterozygotes. Populations thatdeviated the least from Hardy-Weinberg expectations were foundin the Appalachians and at the distributional limits of bothj. ferrupinea, sens. lat., (HUM, TET), and . pilosa (CAP,LWS). These areas also exhibited the lowest levels of geneticvariability seen in North American Menziesia. Neitherpopulations of Vaccinium (Breuderle et al. 1991) norpopulations of Leioohvllum (Strand and Wyatt 1991) exhibitedsignificant deviations from the Hardy-Weinberg equilibrium.Reduced levels of genetic variation within populations ator near the range extremities could be a result of therelative isolation of these areas resulting in bottleneckeffects as Menziesia migrated northward following glacialretreat. Areas to the south may have become too warm or toodry to support Menziesia, except in isolated pockets.Menziesia in Humboldt Co., CA (HUM), for example, aregeographically isolated from Oregon coastal populations by anarea of drier habitat roughly from Coos Bay, OR south to theOregon-California border. Similarly, the southern limit forMenziesia in the Rockies occurs in the Grand Tetons of Wyomingat the bottom of cold air drainage valleys, isolated frompopulations in southern Montana and adjacent Idaho. In the121Appalachians, Capon Springs, WV (CAP) and Lake Winfield Scott,GA (LWS) are comparatively less isolated, and within-population genetic variability is not much lower than observedin many . pilosa populations. Other examples of probablepost-glacial isolation of populations and reduced levels ofgenetic variability have been observed in Phlox (Levin 1984)and Leioohvllum (Strand and Wyatt 1991).3.4.2 Genetic Variation Among PopulationsIn North American Menziesia, levels of gene diversityamong populations (mean DST = 0.007 to 0.023) are quite smallrelative to gene diversity within populations (mean Hs = 0.067to 0.113). The partitioning of genetic variation withinrather than among populations correlates well with thepartitioning of morphological variation observed in Chapter 2,as well as in the profiles of leaf flavonoids (Bohm et al.1984) . The low levels of total genetic diversity distributedamong populations are reflected in the small values of GSTthat range from 0.092 in . oilosa, to 0.166 in . ferrupineassp. alabella. These levels are much lower than the averagereported for animal-outcrossed species (GST = 0.197 ± 0.017;mean ± standard error), and in fact, are comparable to levelsobserved in wind-pollinated (GST = 0.099 ± 0.012) or latesuccessional species (GST = 0.101 ± 0.013) (Hamrick and Godt1989). Similarly, the levels of GST are low relative to meanvalues reported for widespread species (GST = 0.210 ± 0.025),but are slightly higher than levels observed in either short-122lived (GST = 0.088 ± 0.024), or long-lived (GST = 0.076 ±0.01) woody perennials (Hamrick and Godt 1989).However, the mean levels of total genetic diversity (HT)in North American Menziesia, ranging from 0.074 in . ilosato 0.136 in . ferruainea ssp. alabella, are much lower onaverage than values reported in most studies, where HT = 0.310± 0.007 (Hamrick and Godt 1989). This appears to reflect lowlevels of total diversity among populations in a number ofpolymorphic loci, with the notable exception of PGI-2.Consequently, levels of genetic diversity within populations(Hs) are correspondingly low, ranging from 0.067 in . Dilosato 0.113 in . ferruainea ssp. alabella, relative to mostplant taxa, where the mean H5 = 0.230 ± 0.007 (Hamrick andGodt 1989)Other ericaceous genera also exhibit lower than expectedtotal levels of allozyme variation. For example, Leioohvllumbuxifolium had levels of HT = 0.118 ± 0.022 and H5 = 0.108 ±0.062. However, like Menziesia, Leioohvllum had low levels ofgenetic diversity among populations relative to the totalgenetic diversity (GPT = 0.109 ± 0.22) (Strand and Wyatt1991) . In diploid species of Vaccinium section Cvanococcus,HT ranged from 0.151 in . elliotii Chapm. to 0.212 in .tenellum Alt., with H similarly varying from 0.132 to 0.185for the same species. The levels of GST in these Vacciniumspecies were also low, ranging from 0.126 to 0.133 (Breuderleet al. 1991) . Breuderle and his associates (1991) attributedC)i-h-C)Q.IQi-hCl)WC-tH-P)CDl-CDl-U)(Q‘ZH—CD00H-IHH-CDHXCD><CDCD)‘OW00CD‘10U)IC))IWCl)HC))H‘Z3QCDZrfcx0iii-rrctIrJl-H-NI-CDC))CDC)CDrrUi$CDCDrtrr0l-lCDH•HH-CDHH-C1C)H-H-‘1CDC))U)CD‘1HH-HC))CDCDCDiCD(1CDC))hQl-CDH-H-HHH-l-C))CDQ.Cl)CDQCDCDU)H0CDLCCD-<C))H-QU)H-H-H-Q.U)HCD0C)CDCDCDctCDCti-h-t-ZC))•çt<H-C-tI-hCDCDiH-,C))IIH-—C))<U)tCt0HM-,C))HCDoHCD0HLQCtCD<CtCDc-r,—t‘1C))-O‘ZU)i-H-C)(QH-CtH<I-CDU)Ct0CtH-CD‘1Co0U)IIU)CDCDH-><CD0CDrrI-’CtH-i0CDC)CDHI-•‘CtC))MC)Cl)H-HCDC))I—iU)C)ICt<-.CtC’•CtCDCDU)Cl)Cl)CDH-HCDC))0Q.oCDCl)•CDCDI-C))CtCDHl-l-Ct•i-HHHf-t<C)QDHCDU)CD‘-IH-CD0CD0H-CDfrWf-tCDC)NHH-<MH-U)3‘-<0HC)QctCDQIiC)M-,H-WI-CtH-0‘.QCDCl)H-‘.DC))H-H-IIH-0U)—I-H-H-H-C)HCt<CD(XC)<0Cl)Cl)MII-C))H-ClH-CtCl)0CD-tHC)—]CDCDZHCt00H‘CtCt(QCl)H--CtU)C)l-0t-C))CtH-I-jI-jC)Cl)H-C))CDH-0C))CtC))HHCl)I<CDQ.0<HI-0CtCDI-hHH-CDc-r‘<C))U)H-“<tCCDH-CDC)CDH-‘tQU)<U)CtHU)HCl)HC)C))C)Cl)HCt0CtClI-QC))0CD•CtQCl)‘<0H-H-C)CtC)H-‘tC)CDCtH0Cl)‘.0CD0—H-VCDCDH-CDC))I-C))HH-C)C)H-,H-HC))H-0U)H0H-CD0Cl)H-$iI—(C)H—H-Cl)Cl)iC)C))ICtC))Ct0C))U)HC))•i-hU)IICQ•CtCD0i—FH-•C))i-hH-‘<H-0Hi-iH-H-Cl)ICt‘.0H-H-H-C)CtH‘<H-CDC)CtC)•‘—Ct0HCl)0LQCDC)0CtH-I-•CD0toH-CDiClHClCD‘‘zH-,H-00HH-‘.0Id—H-Cl)HH-H-HCDH‘dCD0C))toIF-’--•CDCDC))<H-,CD0CtC))CDCt0‘dIiH-IHH-HHC)çtCDH-,CtC))CtCtH-—IC)tQC))C))‘-<CtCDH-QCDCthCDf1CDHCt(QIC))CtH-HH-U)U)IC)CD‘1—Cl)U)Cl)0CD0C))<0IC))H-CFHH-00H-NCDCl)CDC))Ct(QI-Cl0U)CrH-Q.,Ii--F<Ctf-FU)H-H-H-C))CDCD‘-<CtH-HU)0Cl)C))00Cl)<I‘-QCl)ICtC)<H-H-CF0‘dCDU)c-VCD•Q.,I-•(Q0C))H-Cl)CD<Cl)c-FCDH-Cl)Cl)C))CD-0CDC))H‘dC)IICDC))C))CDC)tIU)CD‘dU)MCDK)H-‘-CDCl)ClH-H-0IICD-c-FCDC)H-t-ri-ic-FCDH-Cl)C))HU)CDU)ot-u0HCDU)C))HC))CFHCl)U)H-CDH-CDIICFI-hH0U)c-FCDI-0fr<H-C))ClC))•H-HCDCI)QH-CDCDCl)H-I-hIi-H-U)I-I-jcQC))HC)(CCtCD0CDtQCl)IF-CDH-Cl)IIH-CDWCDCFCDCD0i-IU)CDC)IClHooHC))ClCl)IiI-CFi—hiIiH-NC)HU)H-CDICD—)H-0CtCDbCFJ‘10CDC))ICl)f-V1<.IiU)C)•CF-.0C)i-hC))CDC))hIC))‘0HLi.H-C))H-HC))ct1JC))HCD‘.0•CDCDU)C)‘XIH-i0tYU)HC))0U)H-iC)i-h0CDC)‘C3tQtQH-C))C)HCt••H—Cl)toCl124variation. This is perhaps not unusual, as high levels ofgenetic identity are commonly reported for infraspecific taxarecognized on the basis of morphology (Crawford 1990)Although populations from the Cascades and coast arealmost indistinguishable based on isozyme analyses, they dodiffer from the Rocky Mountains populations. This appears toreflect clinal variation in the allele frequencies of PGI-2,6PGD-l, 6PGD-2, and IDH-l, which occurred along a west-eastgradient. This is comparable to patterns of morphologicalvariation in . ferrupinea (Chapter 2), but not with patternsof variation in flavonoids (Bohm et al. 1984). In . pilosa,populations from the northern and southern Blue Ridge werevery similar genetically, illustrating the high degree ofuniformity in the Appalachian species. Nevertheless, north-south clinal variation in the allele frequencies of PGI-2 andPGM-1 were observed, again mirroring morphological trends.The relative homogeneity of . oilosa compared to j.ferruginea is perhaps best explained by their geographicdistributions. While . ferrupinea, sens. lat., is widelydistributed in western North America in a number of habitats,M. pilosa occurs over a much smaller range and is restrictedto higher elevations in the Appalachians.As discussed in the previous chapter, there is strongmorphological evidence that . oilosa and . ferruainea arederived from a comnmon widespread ancestor. Although there issome genetic divergence, the electrophoretic data confirm ahigh degree of similarity between western . ferrupinea, sens.1Li(QIt1Hi0—Ir0C)QIMiHCOHI-HCOCl)IlCDMi0iIQP)CDIQ00CI)CD0CD<ç-t0CD‘IH-ICI)CI)COCOP)CDC)CD‘1c-rCOCrIICDH-CDhCrCOICJ)COCDCrCDICOIl0C)CDH-IlCrC)rrCrII0CrICtH--CDC)(QCI)HH-Cl)CDF-CD0Ic))CI)<HICIQ.CDCDXCD0CrI-CI)CDCrC)CI<H•C)C)CD<C)KH-COH-0C)HCICDCDCOCI)00I’CO0Cr0—COC)Q.H-CD(QCDCDCDHH-lCDC)MiCl)CDIQØH-COCDQ.COCl)H-C)COH-Cr•CI)‘DCLII(QCD<1HCI)QCDCDII0rrC)CDCDCDCrIlIICDC)IIC)CD0Cr)CDCDCI)CD0-tCr3CI)11H-CI‘tSCDC)HiCrCLçtIlHCD‘13CrCICDCDiHHCOH-CDH-H-H<CD0(Q•H-CLCDCI)CrCDtCrCD0<CDCrH-Mi<0C)HCI)COHCO<I-COCDIICOCD(-tH-CI)tQCDCDCrCI)MiCI)CDCI)COCrH-0CI)HCDC)CDCDHCOHC)CL0CO0HCrIIH-CDCI)COCD0CI)COCDCDCI)MiCrCOI-h•CQCI)CrQIlCDCI)(QCO3IICO<HH-H-f-tH-MiIlCDIICr00CI)CDQtCLH-HH-CDCr0CI)—‘CI)HI-hCO0ZIH-CrCr)CDCOQ.HCrHCrH-CLI-h0CLI-j(QIICI)Il0H-CDN<IItC(QH-II(QCDH-lCDH-H-CDCrIlH-NCOCDC)CD0H-CDH-CDCDC)tQCLH0H-C)QC)‘<—i-iiCr—ECDIIC)COC)HCDC)Q.00••CI)H-Il0CDCDCOQCDCI)•CDCD01C5HCI)CDCDC)0IICI)COCrHCrQCrCOQCI)H‘-oCDWH-C)CI)‘rjCDCOCrI-hQ.H—)H-IIH‘DCDCLCOIltJCOCLCrH-CrCDH-iCD<CI)—C)MiCI)CL0C)1CI)CI)0C)COCDI-iCD0I-hCrH-CO—0COCrCrMiCOIlH-IIH-H--QCrCDCI)•COCD0CI)HCl)CDCOCDC)H-CrH-H0(COCI)Cr-i•0CDCrH-HC)CD0X0<H-C)CDIlC)H-CI)H0CO•CLHCI)CrCDC)C)CDCI)CDCDCDHC)COCI)0I-jCI)CrtQCOCri-30<COCO0tIlCDCI)III—cCr0roCDCrH•H(IlHCDCrCDCOHH-H-COHCrHiCOHCDH-CDCrCD0CDC)H-CI)0LIHH-Cr3CDCLCDCDC)tiCI)IbCrCDtQH-(0‘-<CO0rrC)CLtC)HHC)‘IHCDHCLC)H-Mi-CrCD0CrCOCDCO00CI)OCDCI)DH-C)CrH-CDIIC)HCrIICrCDCDCI)IICOCrCOCDIICDC)0CD0i-C)‘tiH-CDCH-CDCOHCDCDJC)H-LIH-Cr•‘tiC)CDCDCI)0CrIICDH-H•CrHCDC)I)CrCIH-C)H•‘-DH-CrHCtCfH•HCOCLti<IICrC)CI)C)Ui0CDH--.rrC)CDIIH0CDCI)CDCI)CQCI)3I-jCOCDCD-CDHHHH-CO3CD0Ct0IHCDCDC)CDI-C)CLC)CO0CI)CLLIH-Ct•COCL—CrI.Q3LI3LICI)CDCrCOHCDCDCOLtiCD<H-IICrCI)CI)CrHCDH-COOH-LI“<ILIC5CD0C)0CI)C)C)CI)tiCr<CLCI)CI)0CI)HHCI)lCDC)CrtQCI)HMi00HiCDCOCDiQC)COCI)H-CI)IQC)H-U-CDH00H-CrCI)HiLIJCOCDICH-CDCrHHCOCLLI(CLICOLILICI)CD•CI)H-CI)CrLIICflCDCOH-H-CLCD0ClCDCDCO0ZCrHi<CI)CDICI)COCDCI)-0COCDLICLC)H0CI)C)0CO3I—hLi.HCDHH0CDtiCD0LICr0Hi‘)LI<CrCD0CI)CI)0HiH-3CDH-CDCrI-h0COH•CDCI)CrC)UiC)COUi0CDI-iH-LiCOCI)LICr—CDLICrCI)•0CDCLCO•0CD-H r’)UiH-H-H-‘tj<0tItID)H-QH,t500tYC))U)0CflØ.HCl)0C))H,C))CDH-HCl)QIfrC))H,H-CDti0CDCDH-0U)t-I-iCrCl)H-C!)CDC))0H-Cl)0H-(-tQCDCrHil-QNCOH-C)CrçtQQi-JCDC)C)C)C))C))çtH-C)‘<H-C)0H-CDC)C)C)CrIC))‘1CDJCDC)H-HC-tH-C)CDCrC)CDiCDH-H-H,Cr0kt<LiCDCDCl)CDCrC))CDHH-Cr$iC)C))CrIIH,b-a-CD<CDC)COCDQ<CDCD0H-i—iCl)CDC)çt<0CDk-nCDCrCL—CLC))ICDCDClkiC)CC)hHU)C))QCDI-lCl)0(1)CDIC))H-tiC))H0C)tIC)CrH-H-HH-H-H-H,CrC)COCDHCDH0<CD0C))<Cr‘-<lH,QCDCDt-Cr0C)Cl)ClCDC)CDC))<H-CrIIZHC!)CrCDHCDC))•‘tiH-C))1j-1CD00CrH0H-V0CDU)CDCrCrCDCDHCDU)IHCDH1tiCDCrHCrH-H-<C))<CDLi.H-HCD‘Z3H-1jIHCl)ClC)U)H-CDHC)CDC)I-C!)CrCDU)C))C)Cl)CDCOC)CDIIClrCDU)ClCD0IC)C)t-CDIC)CDLiH-HCr0:)<C))CrH-C)H-00CDC))CDCDH-CTCDtiC))0CDCDCr1jC))CDCrtiQ(QC))0<0JH-HHCOU)U)CD‘tiCDC)H,CDC))C))C))CC)Cl)C!)CDQIC))CDCDCDfrICrCTH-CD0H-00111H-hCDCDC))H-H-H-Cl)(QCrCrCOClCr0H,CDCDPC))0Cl)iCDC)JQIjC!)H-Q.CrCrCOCD0QHH-IH-C))CrH-Cl)<XC)0CDCDCD0CrC))CrH,‘-<CDHCDI0H-IIk’CD0Cl)CDCl‘1ICCDCDC)CDIcyHC)CrICrI—cC)H-HCrCDMHCr-cIi‘tiC))ICUC))0oCr0CrC)•Cr0ICD•CDCDCDC)CDti-QINH-C))f-tH-1<C))3H-GDH-(0C))C))C))I1CDH-IH-HI-•llCDHlit,‘OCDC)hCOU)kCDCDC))CD0Il))CDCDCDII•tEiH-I—aH-C)HICIIHC)knCLCDOCDC)H,CDIH-tiC))Crt-H-HCDC))IC))C))H-CDC!)C))H,I-C)0C))H-H-oCDCOClII0CDCl)HICD11CDCrH-Q.CrC)CDI-C))0I-iCDi-cIC))HH-H-C))c-rCr‘<C))H-‘-0CTCDHCrC))0C))kC))(C))—i-cCDC!)C)Cl)Cl)—QH-CDiHCl)HCrJC)CDi-c•-—CDCr(-t(Q—Cr010CDC)CtHCDC))•I-iCDC)C)<H-(0•CDi-cIC))CD‘ci0CD-VIIC))Cl)H-CDHciClCr0C!)HHCDClCDCrH-HH,C)010ClHC))C))C)Cl)Cl),—rQC))ClI-c‘ciIciiCD0CDCOH,CDCDCIHH-C))•i-jH,QI-(1)HIC))CrHI-hH-C!)•H-C))COC))—C))0II0300C!)CDCDU)0C))‘-0CrCrCLCC)i-cI-(l)i-c(QU)Cr0CI(QH-0C!)CrSCDH-CrCrH-H-CD0Cri-•CDhH-H-IC!)-H-•CDH-::3C)C)0H,CDCD-‘ciCDC!)H,CD•C))HC)C))l-C)•CrI-0I-HC))HC))i‘-‘.)C!’H><I-(QSCDI-’-)00“<ZCl)CrCDH-H•C))CD—HCOCOCLCQ6NCLI))—CLCD<‘ciCrH-0CtCD1CDI5’IH-‘CICDICDi-cC))CD0Crl-Qi-cC))CDIC))WCOCDCDH-U)H-Crii—CrH-QC)H-CD<H-Li.C)CD•CrC)HQC)H-CrCDC)H-CDHCDCOCrlCD‘CIH-l-jC)]H-C))C))C)JCQCDC))CDCl0iH-[iCDCDHCrH-ICH•HU)CrC)C)U)s)H-H•HH-CrCDC))<IC))127aambelli Nutt. (Schnabel and Hamrick 1990) . There is evidencethat these oaks have had the opportunity to hybridize withinthe past two or three million years, which could account fortheir high genetic identity of I = 0.915 (Schnabel and Hamrick1990)It is possible that some allozyme variants betweenpilosa and yj. ferruainea were not detected electrophoretically, which would have resulted in higher genetic identityestimates. However, underestimation of genetic variation is apotential problem common to all isozyme studies (Crawford1990). An examination of mutations in plastid DNA in NorthAmerican Menziesia would provide an alternative source of datathat could be used to assess the time of divergence between .ferrupinea and N. Dilosa. In addition, it would be useful tocompare the Japanese species of Menziesia, especially N.pentandra, with the North American species, using bothisozymes and DNA data, in order to obtain a clearer picture ofthe species relationships in the genus. At present, there isa paucity of information available on isozyme or DNAvariability in disjunct shrubby genera that could be comparedto Menziesia. The examination of genera, such as RhododendronL., Vaccinium, and Cladothamnus (= Elliotia), which havevicariant species pairs in eastern Asia and North Americawould, therefore, be beneficial. At present, the high degreeof isozyme similarity between the North American species ofMenziesia remains an interesting, if not somewhat puzzling,problem which must await further study.1283.4.3 Factors Influencing Isozyme Variation in North AmericanMenziesipEstimates of gene flow (Nm) in North American Menziesia,calculated by three different methods, ranged over all speciesfrom 1.45 to 3.30. These levels are higher than averageestimates for animal-pollinated species (Hamrick 1989). Thisindicates that gene flow among populations of Menziesia inboth western North America and the Appalachians may be highenough to counteract influences promoting differentiationwithin populations. Estimates of gene flow in Leioohvllum,using Wright’s method substituted by GST, were 3.08 for geneflow among regions and 6.69 for gene flow among populations(Strand and Wyatt 1991). These values are high compared toMenziesia and may be the result of using only a few highlypolymorphic loci in the calculation (Strand and Wyatt 1991).Direct measures of gene flow, such as following pollenmovement and seed dispersal, have not been carried out inMenziesia or Leioohvllum. Therefore, the high levels of geneflow estimated indirectly by isozyme data must be viewed withcaution.In general, there is a high degree of concordance betweenpatterns of morphological, isozyme and flavonoid data inMenziesia. Variation within all three data sets are higherwithin than among populations, consistent with xenogamousspecies. The relatively high levels of gene flow amongpopulations are also consistent with this breeding strategyand could, in part, explain why . pilosa is a fairly129homogenous species. There are factors, however, that promotedifferentiation within populations. Menziesia is an insect-pollinated genus and, in this study, the predominantpollinators observed visiting Menziesia were bumblebees(Bombus spp.) and other small bees. Since these pollinatorstend to forage between flowers of the same or neighbouringplants they promote inbreeding within populations. However,some longer distance flights are probable so that gene flow isnot entirely localized (Waddington 1983).Menziesia typically has one major flush of floweringearly in the growing season, with an extended period when someflowers are available to pollinators. The length of theflowering period depends largely on the temperature regime.For example, flowering is extended in warmer areas such asCalifornia and this may increase the chances of long-distancepollination between unrelated plants (Loveless and Hamrick1984) . Conversely in cold subalpine areas, the growing seasonis short and flowering must take place within the span of fourweeks to ensure enough time for seeds to mature. This couldlower the potential for frequent long-distance pollinationevents. In examining allele frequency variation of the highlypolymorphic locus, PGI-2, along transects within the westcoast populations, partitioning of individuals intosubpopulations was often apparent. For example, at Yew Lake,B.C. (YEW), PGI-2a was common along one portion of thetransect, becoming largely replaced by PGI-2d along anotherpart of the transect only 100-200 m away. This patchiness inU)CtCç-tM-,QU)ci)H(QH’Cl-Cl’0U)I-H-U)CD0H’H’I-H’0H’ci)CDCDCDCDH’‘-00CDCDU)CDQ.C)U)Cl-ci)U)CrU)iC)CDci)CrU)<HOQciCtCDU)H-CD0HCl-H-H,ci)dl-‘tS1C)H5ci)H’C)CD‘tiH’U)CtCDt-<ci)U)HH-H’CDhIci)C)Q.Q.ci)CDCDH’ci)t-CDci)Q)H’C)CDU)CDLQCDCDci)CDXt3HiH,CttQHCrCDci)U)ci)U)U)Cl-CDQHciHCrCDH-U)CDNCDU)HCDci)tiCrC)U)ciCDçtC)t-ci)Q.U)0H’QH‘HH’ti-CtC)CDH’00CDU)CDH-CrCrH’U)H’Q.Crci)055t-)U)U)ciH’‘ti5CrIQ0CDU)CDCDtiIICDti‘tiH’C)SU)tiCDCD0CDci)CrC)CDCDH<CDci)Cl-0ciC)QU)H,H-,j0IItI0Q0H-I-U)0ci)5C)ci)ciCrU)Ctci)H,I—’CDHCDCDH’U)HI-<H’‘tiCD1ci)U)tI0QC)CDCT)H’Q.C)CrHH-CD5H<UH-CrCDCDtiH’H-ci)Cl-C)<(Q5SCDII0U)CDci)CD5cici,QU)IICDci)U)U)CrCrCDCtci)0ci)I-U)I-Cl-HI0Cl-—I—’U)U)CrCl-ci.0H-CDI-CrU)CDci)C)3’—U)0CDHH-tClU)U)Hti0H-,U)C)tiHHci)‘lciCDCrI-DCDhjci)ci)H-)Ctci)Cl)H-U)U)0Crci0tiCDCl0tiH’ci)U)55HC)H’C)U)ci0ç-r-t-,U)0ci)‘-0CDtici)H0Cl-C)‘<CDci)U)Hci)ci-ci)U)H—HH’CD‘ti0‘<ci)tici)-C)ciH’HCD<0HCDci)ciClCDCDCDtici)CDCl)05I-jH-5H‘tiI.QU)ciCD‘ç-tU)H’ICDCD0‘Qci)U)ZCDHH00ciQ‘1Clci)ticiCDCrC)5ci)CDH’Hci)ci)H’tici)ti0U)HHci)Ctci)0C-tHt-C)CDHici)H0CrCrCl‘tlci)ciiCDci)CDci)CtC)0‘-O<U)HCtCl-C-tci)H0ciCD<U)CtCt)CliCDci—1CT)H’0U)‘-<CrciiC)C)•Ct-0U)H’ClCDti‘-0ClC)H’0tiCDCrCDH-H-Cl)C)U)U)U)H’0U)(Qci)ci‘<ClU)CDCDCrciCDciCDSF-”.ci<CrCD5Cl-U)F-’-H,HCl‘CDtQCDHU)ci)CD-C)0CDClci)Cl-CD0HU)U)ClCDCrQCr0‘tiCD0—H-J‘10ciU)‘tiLiiCDCL)ci)U)‘‘<ciU)U)5ci)HHCDH<tici)CL)ci)CDCl0CDCDU)CDMiCDci)tiCDci)ci)H’0C)CDHCDH-H’CL)ci)<‘tiCT)CDHHCDU)‘1NiClNC)U)0ClCrU)CrClClCDH-ECl0CDZH’0Cl-CD<0CtH’HH’CDH’Ct)I-ClCDCDU)Cl-II<Cl-CrU)CL)ciH’CDH’0frjH’XSU)‘ICD0ClC)0C-tt-ii<Cl-H’H0ClH-CDIIH’-CD0CDU)Cl-CDCDCL)U)ci)0-05U)H-ci)H,U)-<HH’I-C)U)Ctci)0ci)MCDCDCDU)HMCDF-”Cl-H,çtH0S—H’•CtHCD‘tiH-U)CDTY’ClCD0<0HCDci)Ct0tiCl)U)0ClCDciC)5U)tiH’<CL)‘1Cl0‘ti05CD‘iU)tiU)H-ci)U)C)H’C)0‘tiU)‘tiLii<CDU)H’CDHci)CT)Cl-Cl-0CDCl-t-ci)CDCDci<C)0ClH’U)CDI—’Cl3’•ci)ciCl)0Cl-C)i0HCD0U)U)C)ClCL)CrCDC)CDClCl-ciCDU)CD0HciHci)tiHci)ci)C)HU)H,C)H-Ctti<tiCl-5C)Cl-’‘tiCL)tiCrHH-U)>cU)H’H’CDCDCrci)ciH’CDU)H’0H’0CDCD0H’ci)tiHCDH-Cl-CDScici)H,U)CL)WCDNHtiL),<ciH<H,U)CL)CD0CrciiH-,<CDHC)CDH’CDH’ti•Hci)HCDCDCl<ClU)Ctt-H-,H-CDci)0‘<CDCL)IICl‘-bI-hctC)rtC)0H-MiQCl)ZQ.CDH0CDI-C))H-C))H,CDC))H-C))(D0C)IIH-H-HCr00CDCQCDHHhICD5<IIC)C))<CrH,CDCDH-QN5C))H-C))IIc-tI-(1)Cr0<CDNCDC)CrCrHH-Cl)I-hC))ZC0l-H-I-CH-I-H-H-H-C))CC))CDCDHF-I-HOciH-C)C)CrtQ(1)Q.CD0CDl-H-H-H‘CDCI-<CDI5CDC)CDciH-CC))HCDCrCr0C5IiH-CDCD-CDc-tHC)0CDI-H-C))0CDl-CrC)CDCDCDI-hciC)0CrCrC))0H,<CrC))CrC))H-Cl-0H-CDh0CrC))H-CDC)CDC)CDCD0ci5H-CrI-iCDCC).ICDCDCDCrCDF-C))HH-5C)J0HCDCD05H-hXCDH-CDciCDC)C))C))(DCrCDC))‘MHC))C))H-0ciCDCD0I-HZCrCD0CDCDCDH-CDCD0C)C)I-55ciH-H-CDCDCDH-C))H,C)C)C5CDCDCrCrH,CrHCtH-H-C))CrCDCDCDCD00I-H-0C))H--H-H-CciC)H-CDCDHciCrCDCCrC)Cri5tQCDCDciCDC))HC))-CDCc-rCDl-CHCDCDC))H(QCDCDH-C))HHCrD.—<CDHQ0CD-CDh<<0C)<HC)0CrCDCDC)H-CDCDH-HH-CD0(Qci<H-DC))CrHCDCr<CD—Cr<1tCDCrHci—5CDCDciH-CDCDCDCCrCD0C))CDI-H-HciCrCDH-ciCD0CDCtC))CDCDC)-t0CDH-tQH-I-L0ciCDciCDJH-C))CCDC-rCDIC))C)CDH,H-CDH-H-HciCDH-C))C)<H-H-Crl-CrH,QCD•ct(Q‘rJ0C))I.QC)(QC))CDIQ5H-IIC))CDC))QH-C))5Crc-t0CD‘1H-H-CrHC)hjCDI-jCr0CD0C))H-IHCtCCDCD‘0-tC))I-XCrCDCDI-’H-IICDCDC))CDC)ci0CrCII—)0l-CDC)5CDCl)Cr<IIHCDCr0‘DIICDC))CD0CrNC))rrC))C))0H-C))H-CtCtciCDH-.H-CDHCDI-cC•H-HCHHI-iCC)I-HC))H-CD<ciCDCrC)ci<CD(QI-cciCrCrCDH-H-Ct50H,H-5CDciCDCDH-3<CDCtCDCSi-iCDCDCr0C))ci(Di-c<C)H-H-CrCDC)C)Cr<QCDH-0H<CrH-I.QC)CrCDCDCQCIICDl-<CDl-5CDCDC))CDCnC)0iCrI-ciH-C))0C))(CC)CrH5Ct0C))H-0Cl)H-C))C))CDCD0C))0iCrCDCt-H-0CDCI-ctYjiC5CtQi-cC)H,H,CDCDHCl)H,CDCDCDH-CDH-CDC))MC)H-CDCrCDH-CDiC))CDCDCrt3CDCl)CDCrH-HH-CDCDCD0H-H0CCI-HCD-ciCDCI-c5C)CD0CDCD5CtH-C))1C)CDCrH-C)C))Cr<CDCrhCtCr><CCDCrCDCDciC))<H-CDH-CDIC)C)HH-CDCrCDH-‘<H-CDI-iIHCDCDci0ciC)C)H-HH-Hi-i-cC)CDCrI-I-’-H,0I-Cr3i-cCDC)00CrCDHC))CDC))•5H-C))CDH-C00CD0H-CHH-CDi-c5CrC))CtCDH-I-ci-tH-C))0HHCDH<CDCDCDHCDiCDC)CrH-5CDI-hciH-(QH-H<C))CDH•(QH,H,C))CD<CrH-(QCDH0(DCD•H-00H-HH-ciCDHC))HCrCl)H-C))I-l-Cri-cI-H-CrC)CDCDC))Cr0H-0CDCDICSCDCDi-hCDiiNhH-C)HI-hCi)l-IH-C))CDCDCDC))0CDC))H-CDNH-C0C))CDI-QIF-CrCrI-CrciHQH-CDiH<Cl)0ICCDCDCD-CD-O•CDCDHH-CDS(QlCDtiCrC))H-ciCDCDH-<C))NHCDIC))CDJ00I-C))0H-C)CDCDH-C)C))H,H,H-•-••ciciCrH-H-CrCDC))0CDCrH-H,H-CD<ciH CJ-)HHCDCrC)CrH-C)CO[1QQ.HCQz00CDH-HF-0‘-DHCD0tI0tICr)CDCDH-H-CrCOI-l))CX)C)I-iHtICDH-CtQ.CO!-C)HCrc-tH-0C)CDCrWC)NCDH-CrI))HI-HH-COH-CDJCl)C)H-I-H-—H-H-H-JF-)CL)H-CrC)CDC))(QbC))Cr0—C))CDI-iCrC))CO0H-C))I-C))Q0JHHCrC)CrHC))iHCDH-C)HCrI-ICO0COCDiH-<C))CDç-tCO(DCl)CrH-C)ZCDH-H-CrCOCD<CDCOH0C))CD(QCDHC))—tQhCD-C))H0hCD0‘-C)ç-t(L0CDC))CrC)<HCrCrH-H-HCDQC))CrhC))C))0H-H-CD0HC)COC))<><C))NiH-CDH-CrCrC))tçtC)-Q.CrHH-H-ØCflC-rH-rr1HCDC))C)CO‘<C))-CrC))CD(1)C))CDCDC))<COCDf-rhI0hIH-hIHICOCDhIC)CCHCrCDCQC))hI‘<C)c-r‘<‘QCDI-CD•H-H-CDECrCflHHC))Q.CDCDhIHC))H-t5CDH-Cr0CD<‘OH-<C))H-C))hIH-C))CDtIZC))CrK)Cr0C))0CDQC)-)CDC))CrhçhçCrH-C))tQIH-CDtIC)tICrHCOZCDH-CD0CDCOCOCDCDIHCDQhIH-Q.H-Cr—H-C)Q.COCOl-jCrçtCrCO0K)C))HQ.IhI0C)CrCD00CrC))JCt0HK)C))C))HCDCOCOCOhICD0C)C’JQCDIC))COCrCOC))QH-Cr0c-r—l-jH-CD)-flC))-COHH-<K)CrhINtIC)C))HC)i-jICrH-H-0IH-H-HH-H-CDH-0$iH-CrCDI“<0CDCDWCDHCl)CIIHC)k<I-hCrCOhICrI-CQr-t,C,QCO-10CrCOCOhICDCDCr(QCrCrCOCD(C))H-H-CDC))K)HH-tQH-C))K)C))tiC))CDhIC))C))hItiIH-H-H-CH-C))H-QtiCDCrhIIH-H-COCtHCrQCDIHC))C)HCDCOCDhICDIHCl)CDH-CDZK)H-C)H-HhIrrCD100CDCDCrpH-IICrC))K))(31H-C))H-QC)Q0H-K)<H-HHH-0COIC))hIC))COCr00H-hIQ.IC)’CDtiCDC)hIH-COQH-CrH-H-CtCOCDhIH-C))CrHH-f-rH-C0CDHCrCDhICrQCOCtH-hI0CQ<COH-Hto0C))C))0H-H-0CDc-rH-<hIWCl)CQH-C)-CDCDNII‘<hIMCDCD<HQCDCDCDClCrCOCDC))tiCDWC))ClCDHCDC))H-H-CD<IiH-Q.CDhI<C))HhIC)CDhICrCrCr0C))Cl)C)CDClC))CDCOCOC))hIClIC)C))C))CDCOCOH-0C))H-iC))Cr0Qi-:iH-COC)c-rClCOH-0C)HCrhIHNCl)C))ç-rQC))Q.C)CD0H-QCrCrC))Z‘-<H-H-CrCOlICDC))H-H-<CDCQH-HC))QC))COQtoCDhI0CDHMhI0CrCDCrHCDCDCrH-C))0CDHCrH-H-CDCO00-UiCOhICD‘<0C))CO3’HHCDHCDC))CO0CDI-h‘tiCDhIH-ClCrCrhI‘ti‘tiH-HC))hICDClHCrCDCDCQ0COH-CDCD00‘tIH0CrC))I-hCrCLCDJ’C)ClhICr1-hI-hItiC))H-H-C))ICO1-H-><CDH-H-C))H-tCHCrH-<H•0CL0CDt-CrCDiCrC)COHC)‘<iCDCt0C))CDCDCO(QHC))CDC))C)HH-K)CrC)COI-b<CD•CDCDCrH0Cr0CrH-QC))COCDHCOCOH-CDhIH-H-M-QIH-‘-CDCr‘-.DH-Cr0M-,0C)Ct0C))HC/)CL100C))HhI—)<CDCOC))CrC))CDK)MC)C))CD0CDCDhICrC)-H-COCOC)0COIC))CDhIhICt0-H-IICriCQC)C).Cr3CrCDCDHCDClH w toçr)CDU)o0<f-IftCDH-CDCD0Q.H<‘1‘1CDF-)P)CD(QCDZI-QC)I-P)-U)CDp)IiCDrt•c-iCDU)0çrU)HF-?JCDCDCDF-‘-3CC)Cl)Cl)t3)•U)CDU)H-‘CDCl)Cl)CDCD(QZC)F-CD<I-CDH-CDH-CDCl)CDC)Cl)‘1H-CDU)f-i-cr0I-H1-h00C)0I-h‘CDClC)0H-c-i-CDCDCD‘<U)CDCDU)ci-NCl-hH-çl-pHH(QCDC)CDC)CDClU)CDH-ClC)H-Cl)Cl)Cl)CDClci-CDCDCl)ci-C))0Cl-l)ci-U)I-iIIHCDci-CDPClCtCD•0I-0Hfi-t-tU)U)0It,<IIF-0CDCD‘iCl)IHMC)ci-c-i-Ii—CDK)I-Cl)HClKfl0Cl)<IP)Fci-$))))I)H-H(Q0F-•ClCl)HCl)U)C)U)Cl)Cl)H-1Cl)CDCl)CD CDCDH-, 0Hw134Table 3.1. Collection sites and sample sizes (n) of 34populations of North American Menziesia sampled for isozymeanalyses. Unless otherwise noted, collection numbers ofvouchers are T.C. Wells accessions deposited at UBC; W/H =T.C. Wells & M.E. Hiebert (UBC); S/C = W.L. Stern & K.L.Chambers (OSC); B = S. Brunsfeld (WS). Complete collectioninformation is found in Appendix 1.2.Pop. Voucher No. n LocationN. ferruainea ssp. ferrupineaBWF 1725 34 Brandywine Falls Park, BCSTL 707 36 Stump Lake, Alice Lake Park, BCPL5 1010 36 Parking Lot 5, Cypress Park, BCYEW 1042 38 Yew Lake, Cypress Park, BCSEY W/H 1762 36 Goldie Lake, Mt. Seymour Pk., BCOSW 1009 35 Oswald West Park,Tilamook Co., ORPER 926 35 Cape Perpetua, Lane Co., ORHUN 628 36 Prairie Cr., Humboldt Co., CAROT 745 30 Rein Orchid Trail, Manning Park, BCSKY 799 36 Skyline Trail, Manning Park, BCSTV 1677 33 Stevens Pass, Chelan Co.,WAM. ferrupinep ssp. label1aGVC S/C 33 34 Government Camp, Clackamas Co., ORTRO 821 36 Trophy Mtn., near Clearwater, BCMUR 873 37 Murtle Lake, Wells Gray Park, BCYNP 889 36 Kicking Horse R., Yoho N.P., BCLL 1151 36 Lake Louise, Banff N.P., ABWAT 1190 38 Akimina Pass, Waterton Lks. Pk., ABMOS 1715 33 Moscow Mtn., Latah Co., IDFRE B s.n. 17 Freezeout Saddle, Shoshone Co., IDALV 1228 38 Alva Lake, Missoula Co., MTTET 1265 36 String Lake, Teton Co., WYN. pilosaWBR 1273 18 Walnut Bottom Rd., Garrett Co., MIDDOL 1280 36 Dolly Sods, Grant Co., WVCAP 1341 36 Capon Springs, Hampshire Co., WVJEN 1316 36 Jenkins Gap, Warren Co., VAMIN 1374 28 Minnehaha Spr., Pocahontas Co., WVMTL 1601 32 Mountain Lake, Giles Co., VA1406 38 White Top Mtn., Grayson Co., VAMIT 1440 35 Mt. Mitchell, Yancey Co., NCPIS 1470 36 Mt. Pisgah, Haywood Co., NCWSM 1508 34 Whitesides Mtn., Macon Co., NCLEC 1572 26 Mt. LeConte, Sevier Co., TNSB 1558 36 Brasstown Bald Mtn., Union Co., GALWS 1538 34 Lake Winfield Scott, Union Co., GA135Table 3.2. Electrode and gel buffers used to resolve 16enzyme systems in North American Menziesia. Systems 1, 6, and9 from Soltis et al. (1983); System 8 from Hauffler (1985);System M modified from Wendel and Weeden (1989). Enzymesoccasionally resolved on alternate buffer systems areindicated in parentheses.Electrode Buffer Gel Buffer Enzymes1. 0.40 M Citric Acid, 0.020 M Histidine.HC1; G6PDH G3PDHtrisodium salt; 1.0 M NaOH to pH 7.0 IDH SkDH1.0 M HC1 to pH 7.06. 0.100 M NaOH; 0.015 M Tris; 0.004 M ALD AAT GDH0.30 M Boric Acid, Citric Acid, pH 7.8 (LAP) MEpH 8.6 (PGI) SOD8. 0.039 M LiOH; 0.033 M Tris; 0.005 M (AAT) HK LAP0.263 M Boric Acid, Citric Acid; 0.004 M PGI SOD TPIpH 8.0 LiOH; 0.030 M BoricAcid; 1.0 M HC1 topH 7.69. 0.065 M L-Histidine; 0.009 M L-Histidine; MDH PGM 6PGD0.015 M approx. 0.002 M Citric Acid (SkDH)Citric Acid, topH 5.7M. 0.04 M Citric Acid; 1 part electrode MDH (PGM)0.068 M approx. buffer: 25 parts H20 6PGD (SkDH)N- (3-aminopropyl) -Morpholine to pH 6.2RmAATALDG3PDHHKIDHLAP0.8——a——b——C—-al———d0.6a2—a—b10.4_al——a2——a—al*0.2—a2—a2—b1—_C0Fig.3.1.RepresentativeillustrationsofisozymebandingpatternsobservedinanelectrophoreticstudyofNorthAmericanMenziesia.Numberingandletteringreferstolocusandalleledesignations,respectively.FaintbandsoccasionallyobservedinMDH,HPGM,andTPIaredrawnasdashedlines.Scaleindicatestherelativemobilityofanodallymigratingisozyxnes.RmMDHPGI0.6_al—a2ai0.4a-—————Ca———————d0.2__b.,——e2—.J———-——————---a—__b40——=————C———QFig.3.1continued.RepresentativeillustrationsofisozyxnebandingpatternsobservedinanelectrophoreticstudyofNorthAmericanNenziesia.(AJRmPGM6PGDSkDHSODTPI0.8—a—al0.6————Ca0.4aa————b—bi——Ca0.2————bia——C——_b2—=——C——d0Fig.3.1continued.RepresentativeillustrationsofisozymebandingpatternsobservedinanelectrophoreticstudyofNorthAmericanMenziesia.139Table 3.3. Surranary of intrapopulational genetic variation in34 populations of North American Menziesia. Population codesare given in Table 3.1. Included are: mean number of allelesper locus (A); proportion of polymorphic loci, with thefrequency of the rarest allele > 0.01 (P); mean observedheterozygosity (Hobs); mean expected heterozygosity (Hexp);and the mean fixation index (F). Deviations from randommating are indicated at the following levels of significance:* p<O.O5, ** p<0.01, p<0.00l. For a given parameter,group means with the same letter do not differ significantlyfrom one another at the p < 0.05 level.Population A ± S.E. P Hobs Hexp FN. ferrupinea ssp. ferrupineaAverages± S.E.1.57 ± 0.04a 0.354a 0.O6lab 0.lO2a±0.024 ±0.005 ±0.0120 .356a±0.041BWF 1.53 ± 0.16 0.368 0.063 0.089 0.292*STL 1.53 ± 0.14 0.368 0.067 0.092 0.272*PL5 1.68 ± 0.17 0.421 0.058 0.125 0.536***YEW 1.74 ± 0.17 0.421 0.064 0.133 0.519***SEY 1.68 ± 0.15 0.421 0.088 0.157 0.439***OSW 1.74 ± 0.17 0.474 0.077 0.126 0.389**PER 1.53 ± 0.15 0.316 0.051 0.082 0.378**HUM 1.21 ± 0.09 0.158 0.022 0.026 0.154SKY 1.53 ± 0.17 0.316 0.066 0.101 0.347**ROT 1.53 ± 0.18 0.316 0.044 0.049 0.102STV 1.58 ± 0.18 0.316 0.073 0.142 0.486***N. ferrupinea ssp. alabellaGVC 1.53 ± 0.17 0.316 0.054 0.079 0.316*TRO 1.63 ± 0.15 0.421 0.054 0.110 0.509***MUR 1.47 ± 0.15 0.316 0.053 0.125 0.576***YNP 1.58 ± 0.14 0.421 0.104 0.121 0.140LL 1.53 ± 0.14 0.368 0.086 0.139 0.319***WAT 1.63 ± 0.12 0.473 0.090 0.128 0.297**MOS 1.47 ± 0.12 0.368 0.062 0.121 0.488***FRE 1.53 ± 0.18 0.368 0.076 0.105 0.410*ALV 1.47 ± 0.12 0.316 0.076 0.104 0.269*TET 1.32 ± 0.10 0.263 0.054 0.073 0.260Averages 1.52 ± 0.03a 0.363a 0.070b 0.llla 0.358a± S.E. ±0.019 ±0.006 ±0.006 ±0.040Table 3.3 continued next page140Averages± S.E.Grand1.35 ± 0.03b 0.292a 0.051a±0.023 ±0.0050 . 067b±0 .0070. 222a±0.046Averages 1.47 ± 0.03± S.E.0.091 0.305±0.006 ±0.027Table 3.3 continued. Sunimary of intrapopulational geneticvariation in 34 populations of North American Menziesia.Population A ± S.E. P Hobs Hexp F. pilosaWBR 1.42 ± 0.14 0.368 0.076 0.103 0.262DOL 1.32 ± 0.08 0.316 0.041 0.053 0.226CAP 1.16 ± 0.06 0.158 0.023 0.034 0.324JEN 1.42 ± 0.08 0.474 0.042 0.060 0.300MIN 1.37 ± 0.11 0.316 0.066 0.080 0.175MTL 1.42 ± 0.12 0.316 0.087 0.104 0.163WTM 1.21 ± 0.09 0.158 0.062 0.053 —0.170MIT 1.47 ± 0.14 0.316 0.065 0.072 0.097PIS 1.37 ± 0.10 0.316 0.042 0.081 0.481***WSM 1.58 ± 0.13 0.421 0.051 0.095 0.463***LEC 1.32 ± 0.13 0.211 0.038 0.051 0.255BB 1.32 ± 0.10 0.263 0.035 0.048 0.271LWS 1.21 ± 0.09 0.158 0.029 0.030 0.0330.333±0 .0150 .060±0 . 003Table3.4.Fixationindices(F)forallpolymorphiclociinpopulationsofNorth?mericanMenziesia.PositivevaluesindicatethatobservedheterozygotesarefewerthanexpectedfromHardy-Weinbergexpectations;negativevaluesindicateanexcessofheterozygotes.Deviationsfromrandommatingareshownatthefollowinglevelsofsignificance:ap<0.05;bp<0.01;cp<0.001.Blankentriesrepresentmonomorphicloci.LocusPop.IDH-lLAP-iMDH-3MDH-4PGI-2PGM-1PGM-26PGD-l6PGD-2SkDH-1TPI-1ji.ferrupineassp.ferruaineaBWF..0843a—0.054—0.114—0.1100•697c0.4680.522.STL...0.3650•915c0.045—0.0820.301—0.0380.650.PL5...0704c—0.0380•727c0.0270756c0.3521.0001000aYEW1.0000•731c0664a0485a0.0880•775c0.527...1.000SEY1.0000•722c0.0460457a0.1780673c0.270...0•753bOSW0.7800•718a—0.0180.1690.1010•712c0.4720.3510780aPER...0.780...-0.062-0.0220•618b0766c-0.036HUN......—0.0370.208......—0.038.SKYO.663—0.0470.0610.0810•912c..0.206.ROT0.653—0.6250.0000.022......—0.0641.000STV1000c..0.181—0.0720•861c..0.4700•778cyj.ferruQineassp.alabellaGVC...1.0000•783c..—0.0641.000...—0.0730.634.TRO..0.752—0.0370789a0.2840•833c..0.0000•480c—0.099MUR..0943c•..l.OOO0.1231•000c..—0.0190.297.YNP-0.1210.041...1•000b-0.086......0.3570.263LL—0.1210•446a—0.0380.789—0.028...0.7810•527c0759cWAT—0.0400•402a—0.0390•629a—0.2200•729c—0.0390.1020660c.MOS1000c1•000c..0.0960.0431000a..0.0000.389..FRE..0•892c..—0.070—0.0611•000c—0.054...—0.0541.000ALV-0.1290.116...0660b-0.157...0769c..0388cTET1.0000.080...0.4770.166.........0.159Table3.4continuednextpageTable3.4continued.Fixationindices(F)forallpolymorphiclociinpopulationsofNorthAmericanMenziesia.LocusPop.IDH-1LAP-iMDH-3MDH-4PGI-2PGM-1PGM-26PGD-i6PGD-2SkDH-1TPI-1j.pilosaWBR0.0650.619—0.037—0.004...1•000c0.112...—0.093DOL1.0000.466—0.037—0.072...—0.038—0.038....CAP0.472......0.200......1.000...JEN1.000—0.047—0.037—0.0651•000a—0.0371.0000.6500.650MIN0.4160.432—0.020—0.161...—0.059......0.183MTL0710b0475a-0.046......-0.2160.000-0.022WTM.........—0.114......—0.134...—0.264MIT...0•744b1.000_0•465c-0.036...0.292...-0.1470•706c0733b0.1941000a..-0.038...-0.037WSM0•720a0.6590.2670.0231000a..0.6400.6160.524LEC0.6510•782a..0.052......0.000.....BB1.0000.270...—0.1061•000a..—0.086.....LWS...0.000......—0.094........0.655143Table 3.5. Allele frequencies summarized by geographicregion for North American Menziesia. Refer to Fig. 3.1 forlocus and allele designations. Population codes are given inTable 3.1.Locus- RegionaAllele CST CAS RN NBR SBRAAT—1A 1.0000 1.0000 1.0000 0 0B 0 0 0 1.0000 1.0000AAT—2A 1.0000 1.0000 1.0000 1.0000 1.0000ALD—1A 1.0000 1.0000 1.0000 1.0000 1.0000ALD—2A 1.0000 1.0000 1.0000 1.0000 1.0000G3PDH—1A 1.0000 1.0000 1.0000 1.0000 1.0000HK—2A 1.0000 1.0000 1.0000 1.0000 1.0000IDH—lA 0 0 0.0340 0 0B 0.9843 1.0000 0.9347 1.0000 1.0000C 0.0157 0 0.0313 0 0LAP—lA 0.2517 0.1953 0.3617 0.1212 0.0315B 0 0 0.0033 0 0C 0.7483 0.7895 0.6221 0.8788 0.9663D 0 0.0152 0.0130 0 0.0021MDH—3A 0.0175 0.0340 0 0 0B 0.0752 0.0563 0.0098 0.1397 0.1109C 0 0 0 0.0054 0D 0.9073 0.9097 0.9902 0.8548 0.8891MDH—4A 0.1241 0.0828 0.1011 0 0.0398B 0.0507 0.0940 0.0570 0.0135 0.0313C 0.8253 0.8231 0.8419 0.9865 0.9289PGI—1A 1.0000 1.0000 1.0000 1.0000 1.0000PGI—2A 0.0577 0.0825 0.0017 0 0.0334B 0.2813 0.1431 0.2964 0.3685 0.2656C 0.5175 0.4658 0.5471 0.5642 0.6631D 0.0491 0.1542 0.0016 0 0B 0 0 0.0016 0 0F 0.0943 0.1544 0.0880 0.0673 0.0380G 0 0 0.0635 0 0Table 3.5 continued next page144TPI-1AB0 0 0 0.0673 0.07131.0000 1.0000 1.0000 0.9327 0.9287aGeographic regions include the following populations:Western North AmericaCST CoastCAS Cascades:RN Rockies:Eastern North AmericaNBR N. Blue Ridge:BWF, STL, PL5, YEW, SEY, OSW,PER, HUNSKY, ROT, STy, GVCTRO, MIJR, LL, YNP, WAT, MOS,FRE, ALV, TETWBR, DOL, CAP, JEN, MIN, MTLTable 3.5 continued. Allele frequencies summarized bygeographic region for North American Menziesia.CSTVCASRegionaRN NBR SBR0.0122 0.0074 0 0 0.00210.6729 0.8611 0.7996 0.9839 0.96650.3148 0.1315 0.2004 0.0161 0.0335Locus-AllelePGM-1ABCPGM-2AB6PGD-1ABCD6PGD-2ABCDSkDH-1ABCSOD-lA0.9494 1.0000 0.9610 0.9622 1.00000.0506 0 0.0390 0.0378 00 0 0.0229 0.0054 0.00630 0 0.0700 0.0108 0.00411.0000 0.9811 0.9039 0.9273 0.90380 0.0189 0.0033 0.0565 0.08580 0 0 0.0081 00.9159 0.9210 0.5243 0.9891 0.97500 0 0 0.0028 0.00630.0841 0.0790 0.4757 0 0.01880 0 0.0033 0 00.9582 0.9211 0.9870 1.0000 1.00000.0418 0.0789 0.0097 0 01.0000 1.0000 1.0000 1.0000 1.0000SBR S. Blue Ridge: WThI, MIT, PIS, WSM, LEC, BB, LWS145Table 3.6. Nei’s genetic diversity statistics for North AmericanMenziesia taxa. Gene diversities are presented for each polymorphic locus as well as pooled values over all loci. HT = totalgene diversity within a taxon; Hs = gene diversity within populations of a taxon; DT = gene diversity between populations withina taxon; GST = coefficient of gene differentiation.Locus TaxonAAT-1 M. ferrupinea. ferrupinea. pilosaAllssp. ferrupineassp. alabellaHT0000.472DST GST0000.4720001.0000000IDH-l . ferruainea ssp. ferruainea 0.023 0.022 0.001 0.029jy. ferruginea ssp. alabella 0.108 0.103 0.006 0.051pilosa 0 0 0 0All 0.040 0.037 0.003 0.067LAP-i . ferrualnea ssp. ferruainea 0.38i 0.296 0.086 0.225M. ferruainea ssp. alabella 0.470 0.374 0.095 0.203M. pilosa 0.158 0.138 0.020 0.127All 0.346 0.259 0.087 0.252MDH-3 M. ferruainea ssp. ferruainea 0.143 0.131 0.012 0.083. ferruainea ssp. alabella 0.065 0.057 0.007 0.114M. pilosa 0.226 0.207 0.019 0.083All 0.155 0.138 0.016 0.106MDH-4 . ferruainea ssp. ferrupinea 0.325 0.295 0.030 0.093. ferruainea ssp. alabella 0.264 0.238 0.026 0.099M. pilosa 0.084 0.071 0.013 0.157All 0.221 0.193 0.029 0.129PGI-2 ferruainea ssp. ferruainea 0.647 0.580 0.067 0.103. ferrupinea ssp. alabella 0.630 0.582 0.049 0.077M. pilosa 0.525 0.482 0.043 0.083All 0.602 0.543 0.059 0.098PGM-1 Iyi. ferruainea ssp. ferruainea 0.412 0.337 0.074 0.180M. ferruainea ssp. alabella 0.296 0.184 0.112 0.378M. Dilosa 0.047 0.045 0.002 0.043All 0.261 0.181 0.080 0.308PGM-2 . ferruainep ssp. ferruainea 0.071 0.069 0.002 0.023j. ferruainea ssp. alabella 0.066 0.063 0.003 0.045yj. pilosa 0.044 0.040 0.004 0.081All 0.059 0.056 0.003 0.047Table 3.6 continued next page146Table 3.6 continued. Nei’s genetic diversity statistics forNorth American Menziesia taxa. Gene diversities are presentedfor each polymorphic locus as well as pooled values over allloci. HT = total gene diversity within a taxon; Hs = genediversity within populations of a taxon; DST = gene diversitybetween populations within a taxon; GST = coefficient of genedifferentiation.Locus Taxon HT Hs DST GST6PGD-1 N. ferrupinea ssp. ferrupinea 0 0 0 0N. ferruainea ssp. alabella 0.166 0.135 0.031 0.186M. pilosa 0.161 0.145 0.016 0.100All 0.114 0.095 0.019 0.1656PGD-2 ferrupinea ssp. ferruainea 0.152 0.142 0.011 0.070j. ferrupinea ssp. alabella 0.484 0.386 0.098 0.203M. pilosa 0.036 0.033 0.002 0.066All 0.262 0.172 0.090 0.343SkDH-l N. ssp. ferrupinea 0.112 0.099 0.013 0.114. ssp. alabella 0.028 0.027 0.001 0.050N. 0 0 0 0All 0.046 0.040 0.006 0.127TPI-1 M. ssp. ferruainea 0 0 0 0M. ssp. alabella 0 0 0 0N. 0.127 0.117 0.010 0.082All 0.051 0.045 0.006 0.121ferruainea 0.119 0.104 0.016 0.130alabella 0.136 0.113 0.023 0.1660.074 0.067 0.007 0.0920.139 0.093 0.046 0.331ferrupineaferruaineapilosaf erruaineaf errupineapilosaAlllociM.M.M.Allferruainea ssp.ferruainea ssp.pilosaIICDICDlIIIt313IICDICDIIII)i-3I.CD.CD.III(DCDP)I.CD.CD.IP)llIdII’IddXiiOblICj•I-hIICD-’lid••MI0II3HIIH-IIICDcfCDH-CDIII(QCDIll-’IQl-hhIII(DH-HIQHhIIiHIIi3C)WI0IH’-CDIIIcrwIDP11jIII•lCDIS))CIIIIP)lIIIctQIS))ItJI-SIII(DICDH-H-IIIJH-•ICDH-H-IIIIIII(DC’)II-3lIIrtIHH-CDIIIrrIIl—CDH-fl)IiiIP1IIIrtPiZIIP)P)iIIIICDIIIJICDIIIi(D(DIP1IIiC)H-IIIICC’S))IIII(DCD-IIIIIII(DCDCJ)IIiiZIlCDIIIlllQ(QIlCDIIZ—cQIC)CDC)CDIICDIldIIOHCDl•••IPillçtII.IIIO‘0‘..OIllCDIC)C)IIICtCDID-COIIICtI•IM-II(1)f-tIF.\.)OIHIIt’iH-ICDCDIIII-5H-IIIIC)10)HIIC)I—-—I(DlICDI‘.0CDIIIICtICDCDCDIIIIIHIIIICDH-I•••ICtIIIIII-‘5H51I‘.0‘.0‘.0IH-IIC)CDIIH-IIH-P)CDI--(51IçtIIP13II3IlII‘.00)OIk<iiII•IIP1(CtIIIIIIHIIIIHH-IHCDCDI-IIçtf-tIIMDIICDrtI.••ll-’5ilP)HIIIIt—H-IC)‘.0‘-0IP1ulXCDIIl-’5IIP1CDlCD‘-.0‘.0lIPit’)II1-5IiX51CDICD0)-0IIIIltDIIP1CDl——ICDlIOS))IltDii—I1—lIII‘dH-II0H-CIIIi51I1lII-to’t3IIIlCDCDIIICt’dIIIII.•IQIIH-CDIllCDICD‘-0IHII3-‘5ICDC)C)ICDIINP1I—)‘.DIc))II3(.•.IPIIH-ZI‘-0CDIItJIINCtICDCDCDIIIIIICDIIH-CCI-5lCDL\iHIIIIIIl-IIH-ICO‘.0-0IIIH-CDIIII—IICDcI)IIH—IlIlb))IIH-3I-ICI)Il•CDIIIIl-jl.QICDCDIc-CiiIIII•CDl—’I••pIllH—IIIIHCDICD0lIIC)IIIIP1—I—CDCDlollIIIiiCrIC)HCDII51II•IIH-SDIIIIIIIHI0CDIII03ICDCDCDI-IICl)I••IIdilI•••l’5IICtI‘-.0‘.0IIH-IICDICDCDCDIS))IISDILi-)IIHIIIJC)]iIlIIU)I10IIH-IH0)‘-.0IIIC)HIICDIItii-.——ICDIICDIIISDIlII-IlCD148Fig. 3.2. Dendrogram of North American Menziesia populations,grouped by geographic regions, based on a UPGMA clusteranalysis of Nei’s genetic identities. Population membershipin these regions is defined in Table 3.5.GENETIC IDENTITY0.92 0.94 0.96 0.98 1.00COASTL CASCADESROCKIESN. BLUE RIDGEL S.BLUE RIDGE149GENETIC IDENTITY0.904 0.920 0.936 0.952 0.968 0.984 1.000r—BWF_fi— STLGVCoswSKY_PL5EEWATWBREDOL—MI NBBCAPLWSLEC 3[ JENWSM—LILMITV______MTLPISWTMFig. 3.3. Dendrogram of populations of North AmericanMenziesia based on a UPGMA cluster analysis of Nei’s geneticidentities. Population codes as in Table 3.1. The copheneticcorrelation coefficient is 0.924.I)tilWHC)3II+III‘-3IIIjGLiJ’.loll•kDCIF—ilclIICDOIIC)IIOjF1bIJ)F-XIILiLillIIciiifrtiICll)H1IciIl-tiIt-tbIIC))HIIF-1I1IH-IMF-1ICOIIC))IH-CtCDIt’lCtIH-IIIH•C)Ct(DIt\)1-s)C)LiIHH1’JIIItYkDCDI•C))IF—IIF—i0)ILi1010)11F-l)WIX10IIICflI—(DWIII—aIICD--(Q•I0)III•C)IIC))IHIIDCDI-IIC))III-I(0IIWi0-IIIF-‘IIHII•CCD•IIIC))frI(01(011l—CJ)I(0IIIrtCDIC))III•H-I-hIC))IIICDCD0IIIII-jIIIIQCl)IWIII(QQICO(01IIrtWIIIt’JCDICl)(01IIt3C))CDIIII(Ott)I‘dIIII•lOllrtF---•III0)1-1,IIIIH-C))ZCDIIItH‘‘1’1IQCIIiftII-hiII00ICI)LiHW(!)tiQtiHIHilC))-C)IIF-IIIIfrH-IL’iCt)L1C))IC))IIctH-‘<IIC))‘1IIICDIlcTllCDC))I1t3I-iIIICDZ—ICDIHIIc°-ioICDIII0II—l0II(001--tiII-IIIIHIII01IH’1-IIIIC))IIIh(Q’tlIC))iIIIr-hIIIICDfrjIIIIH,CDIIII(QtJF-’-IIIICDCOIIIICD0<IIIIIIII0C))IIIICDCDHIIIICD1Cc-tIIIIIIt1jII•F-’CDIIIIçtH-C))IlIIMiC))IH-IIIOCtIICDIIF-’CtC))I<‘tJI.)HF-)II-LIIC))CDIIlI0JH-H’1•••IH-IICDCl)ICDC)C)C)C)ICII00H’IF)WQ,‘0O’I(QIICtI.ICDIICDC)‘OHC)C’IIIII1C)C)C)C)C)C)C)IIICCOH’IICtII0lCDt-IiF-’UF-1UiI-’UIC)III—’CDIIIIQ.I30IF)F)O01‘OUI‘OI‘<IIC))0Cl)IILiiIICl)INIIIIçtt—tIICl)II—.II—’—IItIIIH-—IICfIIICDC))I—IIC))OZ’Z11+IH11IIH1II0-IC)F-1F-1F-’I-’ICThIICD1HHII—IICDMF-11•••IMC))IICD)CDIIIIC)Cr—IaoC)WI0CtII•I—’IIIIH,0—I01iF)‘..DICI)II-tCD0Q,II(011CDCCDCIIII<LIIMiCDU)IIOIlf-tWQ.CDCDIIMillCtCDIIIIt—hH-H-1-H11+lD2II0CCDC)CtIF)WIIl-II1-C))CDC))QI•••IIC))IICti-3irIL.-)a)a)IICrIICDC)C))H-C)IC)C)F1IIIICDH1tJ<C))IIH-IICDF-CDHIIII(I)CDI151Table 3.11. Levels of intrapopulational genetic variation inNorth American Menziesia compared to average values summarizedfrom other studies. Included are: the mean number of alleles!locus (A); the percentage of polymorphic loci (P); and meanexpected heterozygosity (Hexp).Category A P Hexp(S.E) (S.E) (S.E)MenziesiaM. ferrupinea ssp. ferrupinea 1.57 35.4 0.102(0.04) (2.4) (0.012)M. ferrupinea ssp. alabella 1.52 36.3 0.111(0.03) (1.9) (0.006)j. oilosa 1.35 29.2 0.067(0.03) (2.3) (0.007)widely distributed taxa a 1.72 43.0 0.159(0.07) (3.3) (0.013)regionally distributed taxa a 1.55 36.4 0.118(0.04) (2.0) (0.007)endemic taxa a 1.39 26.3 0.063(0.03) (2.1) (0.006)long-lived woody taxa a 1.79 50.0 0.149(0.012) (2.5) (0.009)short-lived woody taxa a 1.55 31.3 0.094(0.12) (6.7) (0.021)sexual reproducers a 1.53 34.9 0.114(0.03) (1.3) (0.005)outcrossed-animal pollinated a 1.54 35.9 0.124taxa (0.03) (1.8) (0.008)mixed-animal pollinated a 1.43 29.2 0.090taxa (0.04) (2.5) (0.010)Vaccinium sect. Cvanococcus b 2.82 48.2 0.148(0.17) (2.6) (0.038)Leioohvllum buxifolium c 1.89 64.6 0.108(0.120) (6.3) (0.018)Sources: a Hamrick and Godt (1989); b Breuderle et al. (1991);c Strand and Wyatt (1991)152Chapter 4Concluding Remarks4.1 Objectives RevisitedOne of the primary objectives of this study was todescribe the patterns and quantify the degree of morphologicaland isozyme variation present in North American Menziesia.Morphological, flavonoid and isozyme data are now available toassess the interrelationships between the eastern and westernNorth American species. There is a high degree of congruenceamong the data sets. In all cases, the amount of variationobserved within populations greatly exceeds the amount ofvariation among populations. This pattern is typical ofspecies that are sexual diploid outcrossers. The levels ofisozyme variation within populations and the degree ofvariation among populations also agrees well with reportedfigures for outcrossed animal-pollinated taxa (Hamrick andGodt 1989)Clinal variation is common in the North American species,particularly . ferruainea. This was most apparent in themorphological analyses, though similar variation trends wereobserved in the allele frequencies of certain polymorphicloci, namely PGI-2, 6PGD-l, IDH-l, and PGM-2. While somemorphological variation can be related to ecological andenvironmental factors, there is sufficient discontinuousvariation in western North American Menziesia to allow therecognition of two subspecies of . ferrupinea. The twosubspecies have a very high genetic identity (0.99) and<ti‘tSH-tYC)QH-Q.ftflftIItQftC))lCDCDIH-H-lCDH-Q(DHH-U)H-CD))‘10‘11i‘1CDft<tiU)CDft0U)0U)))P)(1)C)I-CD‘1CDQC))IH-H-H-IIlH-0H-IN(1)QCDC)Cl)ftNftft0H-C))CIIIQftCDCl)H-k<H-ft(QC))ftCE)H-ftICE)C)U)kD<oH-CDHH-CD0ZftCD0H-knH-CFH-lCnCDfti<C)CDC)Cl)Cit-t-H-H-I-F-H-LQoC))CDft-iiCE)H-ftftC1ftN0C)U)IC))HQ0‘<IC))‘<(E)ftI-HC))tOH-ftQti-U)Ct)M-C))fr1-(I)HH-C))CD-Hçt0C))CDHHCl)C))QC))C))C)H-HH-HI1(1)hC))0H-C))U)C)C))CDH-,C)C)ftC))ftC)ftftC))<t.-CT)H-CDHC))H-CDCl)ftHH-CT)tY0<H-H-I-jCDfrC)CDH-H-0C)U)0Q.H-C))C))H-H-M-U)QI-XH-Q0H-U)H-CDH0hC)IIC))0H-0C)C)H-U)CDHftU)H-H-H-C))C))tNM()ftC))(CH-C))ftftQC))IU)‘<H-,CDftC)‘1U)><CE)<‘ftCl)NC))0ftCl)0F-HIQhh‘-i-0‘-tiF-Cl)CD0CDt-0tQ<1‘1ftCD‘<IHC)‘CDC1‘C)0CDCDCDH-CDCD‘<C))-0i00C)H-,C)H-ftCD00CDHSC)IS0ftIIHlCDftIHH-H-0<H-0CDH-01000C))C))ftHU)ftIj—CDC))‘—ak1CDC)5CDC)i-tC)5CDU)1MI-jCDtiIC))U)‘‘1ft‘1ft)-H-,CDr-’CD-0H-0C))CDCD0‘<CDH0CDCD‘Q5CDU)0C)ClIIHU)H-XXS‘-O5CD‘ftftIH-CDftC))0C))C)fttiQCF00‘-IC))-DCDCDU)CrHtQH-U)CDCX)0fthI00H-fi-CD0C)’IH-CDJ’C))ft‘)ft0U)0H-IIftU)0U)hIH-QH-lCD--CDftfthICD—CDS‘(5C)ftC))tQCl)CDt-C)hIW•00C))CDHt3CDH-CD-C))CD0tH-‘(5H-,F-ClC))U)C))HI—--CDU)CD0hI0CDCDC)HftH0ftHCDftftCi)HC))C))CDH-C))tQU)0C)H-5ftU)H-CD‘(5-hL)I—i‘<ftftC-jH0H-‘(5iCl)kD(-.30H-HHH-H-Cl)ftU)C))‘iC)‘(5ftftH-CDlICDhIZ’CDCDH-IQCDC))HH-C))CDCDC)Cl)Cl)U)H-F-H-ECDU)IH-hIU)<U)C—’CD0CDC))C)0Cl(5HU)I-hIi--H-ft1iU)H-ftU)CDhI0ftClCDftH-lCD0U)c‘(5t300U)CDC)CDIC))(5H-hIHlCDH-0HCD(QCl)H-ftCDCDH-C))IC))•U)5H-CD‘-0IC))ftH<ClC)H-Cl)CDCD(QClCDCD‘(5hU)CXH-‘(5JU)CDH-C))U)I-lCDH-hIhIH0CDiftHCDHH50U)Cl)hI<ft<CDhI00)C))hIHU)(QCDH-ftCDCDCDHCDhI5ft<U)H<C))C)3CtU)ItSfthICDftClClCl)U)C)CDH-,ftt-ft<CDHC)CDCD0HCD‘(I0H0U)C))U)CD<Cl‘(5-H-QHJCD(50CDiU)Cl)ftCDH-0H-trU)hIH-H-—C)(QC))C))H-C)ftC)U)(<C)HClC))C)CDCDH-CDU)C))U)CDftCF(QC))HCDH-—Cl)<C))HCD‘<ClHCD0CDCDC))U)U)Cl)(5CDLIH-ftHftCDCDU)C)H--Cl00CD•CDC)0CD•HH-tIH-,H (-.3c1c3H-çtIl-hI-h)C)HI-C))H-0G-H-Ctwi-00C))0aH-H-H-C))CDSaui0zCDHCrC))IlIIaQU)I-çtC)C))H-0C)HH-çtl-j0U)(QCtCr(Ql-CDCD0CDU)CtaQC)H-0H-CDIIH-0IIIIIlHIIC)iaitJH-H-CDiC))CDHCDU)CDIlCtCDCrC))0H-CDCrCrH-‘-QQ.XU)SIlC))Q.H-CD0IICrI-jU)C))CrH-C)-CD‘IQCl)ciH-CDC5frhtCrdM)ciCDç-tH-1Cl)IlCDCDCD00CrCDH-0H5H-IlH-aCDCtCDU)H-l-’5CD•Q-tIlCDCD0C)’0U)c-tCDt5aHC)c-r0IlH-CDU)H5CtCDH-‘<CD0H-C)aHCtC))CrH-C))Ilci0ciCD5U)IIC)C))CD(QaLQCD3C))C)C))aaH-CDciC)tiCDCtU)CDZ0C))Cl)CDHIIC))C))CD5C/)CDCDCrH-CDCrH-0QH-0Cr•CDCl)CDI-iHHU)Ill.QU)><H-CtC))C))HCrciCD)<HH-Cl)C))CDCDCDH-5H-C))0H-Cr0ciCDC))5H-Cl)C)(QCtIaU)5000C))5CrCDQCDtiI—hCDc-tCtC/)a-CDt5H-C!)CDCrCDH-CDaIIM0C))ClC’)CDH0U)C!)tQHCD0iici0tiHaMH-Cl)<0‘1IH-CDC))0C)ci0CrHHCtHC)H-iCl)-C)i-iCDCr0HI0CtM,0C))CD00C))H-0cCtU)H-<IICDQCrTJCr‘ICrCr01(1aCDH-Cr0C))U)C))CrCl)CDH-H-C))IH-0Jci0U)CD11HaC!)HCrC))U)H-HIlIICDH0Ctl-QI-(0H-0Cl‘<ICCDCDH-CDaiiH-IIC))><CrQF-c-tCnC!)i-a0CDHH-CrtCDU’IC)çtaCDU)HtiC)0CDCrC!)0IHU)CtNc-I-CrCD‘<CDCDM011CttiH5HU)H-)0CD‘-CCrCrH-CDIl0CDC))C!)C))C))CrC5H-IciCl)QHCQ‘ICD5CrH<Ct5HH-U)0C))Il-hH0IIHC)CDCDa3ci0JC))CDIC))U)CDH-H-I-’-I-CC))CrCDCtH‘IHH-C)aU)C))5CDci0CDH-CDIlH-U)CDHCD0CDCDH‘-QCDU)HciIIHCrCD0CDCDh5HU)HCraC))<I-j00IIiClU)U)aHC))CDH-CDC)0C))C!)HMCDaC))-Ct0CDH-C)l-jiCDi-c-l-C)lHtYU)0C))H-0U)C!)tCHU)U)IC))CD‘QtEC)HCDI-h‘Qa0U)H-C))HtiHCDCDCl<C!)C))CDCrCr0tIQCrtiI-hCtC))I--hC!)(U)C))5C))t-Cl)U))0C!)I.QH--<HHCDCrC))CDIiCD•CrC))HH-CD--aCD0tiCDU)C))0IIH-aCDH-HdCDC)C!)CDi-ca0CraI—50HH-CrCDH-C)‘CDCDl-CC))CrCDI)C)H-Hi)C))t-3CDi‘<rtC!)CDCtC))51JU)<U)IlC)I-C)0P0‘-<U)I-CCDCDH-HaH-‘1C)U)CDU)HU)HIC))CD0aC))H-H-C)HCrQCDH•H-H-IlC)H-C!)CCD<II<ClC))H-H-C))Qd<IIC))0U)CDC!)<CDHCrU)U)H-IIC))CDiZCDC))CC))CrCD0H-U)t3I-hCDH-tiU)Cr‘-0ciC))aIlC))C))Il00HH-MU)CrC))CrCDH-tYH-H-HCrH-U)0H-CDCDCD‘-Ci-CCDCrH-CDC))HC))C!)‘tiC)t3U)‘IICrH-0H-CtC/)CrHC))CD3CDHCDCDCDCr•ClaI-h0C))H-CDH-<aC))CD<HaCDCrHCr30C))HH‘<CrCrH-CU)U)C))U)H-CDCrIIa0CDU)CrCrU)CDI.<‘-<IIH U,H(QCD‘jH-Cl)MIH-CDI-C)0tY0Cn0P10HQ.0Cl)0H-CDCDP1XP1CDP1LCCDH,0CDCDC)iIIP1tQH1U)CDtiU)CDU)H,ctHCDC)Q.CDC)CDtiCDHI-cvI-H-NHHZcvU)CDMU)P1U)H-U)CD‘1H-HCDP1H-P1H0H-Z•U)C)U)0P1><00c-tCDcvH-Cl)cvC)P1I-c-FC)H-P1cvH0P1tYP1H-CDIH-Cl)CDQC)CHHH-C0CIP1H,P1cv3QC)•0H-H-HH-U)0P1HCDt5cvI-P1Z3CDCDZ-HcvP1HCDHHH-CDHU)H,•CQC)MCQHP1HCDCDcvCDCDU)P1-C)CDcvH,‘-<cvP1I--0I-P1hP1c-vH-CDCDP1cfCDH-P)H<0CDNcvHU)U)cvQi-<H0C)<1jcvP1i-tU)0CD(ICTP1H-P1P1•CDCDCDc-FH-<C)CD-cCDH‘tic-FU)P1cvCDU)I-iP1CDU)CDt.P1CD—U)c-FCDCflH-CDcvHc1H-0CDCDHiH-C)•H-P1MJCDI-JH-cvP1P1Hc-v‘-CDP10cvCDU)CDHCD<CDsi0CDI-cvc-vC)P1P1P1H01p1CDCDU)cvP1U)LQCDU)c-FP1Q.CDP1HC)HU)U)HU)H-I-Cl)CD1jH-H-P1H-U)QcvP1‘-iiC)c-vCDU)‘<H3H-CDP1P1Qcv<c-vCD-c-vHP1U)H,‘<H-Cl)P1C)P1CDH-CDP1-H-CDP1cv0CDP1I-HCDCl)cv0P1<Z0H-MU)00c-FHQH,It‘1(QH-Cl)QI-jHU)c-FH,‘z3HCDCDH0•HCDU)CDH-CDctI-CDQ(QHCD-c-FI-jCD0C)P1P100U)‘1CD—ai-tYCDQP1CDCl)U)K)U)0P1U)H-CDc-FC)tiCDU)‘—0P1Q.CDHU)U)cvIH-U)Hc-F><Q.CD•0P1CDc-v—H-P1cvCDCD0P1IHCD0c-FCDcvcv•Q<<U)U)‘CDH-‘tiH-IC‘Z3P1HCDU)CDH-P10H-P1CD•c-F3H-i-cK)l-CQ<H-CDp1HCDi-C)HU)H-0CD0Cl)IS))0P1CDCDU)(QU)H-HQQ(Q3I-c-vH-H-U)U)Hc-vH<00Dl0C/)H-0P1CDc-vc-FH-P1P10CDP1P)30H-U)‘<cvJP13CDCD<CDH-<C)Q.<l-jcvCñH-QcvcICD0CDU)HtiQCD0CDH-Q.cvCDH-CDH-HU)C)0CDc-vC)I-cU)CDU)<P1H-HS)1H-CDc-FI-c-vCDCDCDDl0I-(QCDcvc-vCDU)c-FC)1QH-c-F0H-U)Q><0H-CDI-CH-H-H-C)c-F<I-jC)DlCD•C)H-C)QcvCD‘ti<100<00P1CDHCDU)C)3CD0cvCDP1H,U)U)H-0cvH,P1c-I-Q.CDC)cv‘ti-IIH-HNH-CCD1-51l-CDH-CDCDCl)H-0QP1QH-U)<CDi-0(QQi-c<3Q.C)H3<CDH-tCD‘tiU)CDCDc-F1-5P1I-SCDCDCDP1tIc-vCD0cvc-vU)I-1-5Q1-5U)1-5(QCDU)H-l-5151cvP1U)35i<CDCDCD‘tIH-H-H-‘-5H-0•H-CH-H-U)cvH-1-551U)Sic1tiP1P1CD<H-H,U)c-F<0cv0U)cv0HDlP1SiCD3cv0cvcvCDH,P1H,ICDHDlI-50ctH-NI—5H-5H-5U)U)H-‘ti51cvcvHCTMH-c-FH-ti•cvCDHH-CDU)CDIP130CDH-CDH-cv0‘-<CDCDJ0ty3H-Dl<•P1tiH-00P1CDU)1-5CDC)SiCDc-vi-CDCD5ySiH,H3H-P1P1HH-cvC)‘-<CD‘-5SiH-HU)P10513H-•P1H-cvP1cvC)CQH,H(QCDtCC)U)0HHP1cvC)H-P11-5H-0CD‘-S3•cvH-CD0cv0U)•-P1HcvH-U)0SiP10I-sHcvCDHcv‘tiU)51U)‘<H0c-nI-f,c-fl[(0 5C)COC-r1C)IHiLQHQ<JC)Hii-3MiF-SCDrl)-)SC)0C)‘11iCDhH-IlH-H0C)l)C-tIlC)CDN(1QH-FtI-IH-HCDCOCDC)H-HtirtH--)CDCrIIH-CII—S))IC1)H-)Q.‘1CrH-H-C0Ci)CDCDHP)CD‘)0)JlCD)0CDCDjJC)CI)CT)CDCDSSCD0CDIICI)HCDCrIC)IICI)II1<IlCl)(I)CDCDH-IICIcCo‘tiH-FlCIH-H-COCDCDCI)HHtiCDIIIICl)COH-Cr0111HCDHCl)-rCrCl)CI)CI)00F>JCDrrH-C)CDH-HCI)Cl)HCCrIIIICOC-C)0-COç-t(5CDC-r0H-CrCDtiCOCOCI)CI)CI)H-ti<CI)C)H-CDiCI)0C)Cl)CDti‘tI0Cl)CrQC)CDIICr<COCr-QC)CI)CI)CIlC)H-F-IICI)C)CrH-CDCr-H-0IIH-Cl)CD0Cl)H-0HCDCDiiH-(DC)I-iCI)CrF-1H-CrCO<CrCDCDHCI)‘<HI‘10CrCOCDCrCI)HCI)Cl)hi,I-H-H-Q.COf-tC)Cl)CrCDCl)HCDI-IHH-00CDC)Cl)H-hi,ICrQCDH-çI-051-’-0I-Cr<H[CD CI)QIl‘HC)dCl)CDCO0Cl)CDCDHH-CI)HHCrHCl)Cl)CDHCrH-U)COHCD05CI)CI)H-C)H-H-‘-<Cl)CrCrH-CrCl)HH-Cr01(1)II‘10ClCl)0rrH-CDC)H-5H-ltlCl)H-çrIIH-Cr—CICDCl)çrr-i,‘ICDhi,CDH-hliiICrç-rC)Cl)CI)QCI)H-CD0CDCDCrQHI-iI-iC)ID)0CI)Cr(Q(CD0C)St-IICl)oo0CrI3C)0H-CDIlCI)0CI)HH-CDCDCr5-IClHC)CDCDCD01CIHCl)H-H-Cl)lIlCrCD0CDCDCrQD)ClCI)CrCrCl)I-•CDlID)COCII-4CCI)I-5H-l-’H-IIH-COH-0‘tI(0C•0•H-xH-H-CI)CDC)HCl)‘0Cl)‘tiH(0Cr5CI)C)C)CI)H--0Cr5(QH(Q5CICI)H-CI)CI)H-ç-r<HiCr1’iIIH-00CDC)o‘tiCDHCI)•H‘<D)CD-.CrC)IICrCl)HCl)Cl)CI)HiCI)0CrCDCOICrHCDIlQH-5H-H-Cr3CDHCrC)D)CI)I-iCI)HCl)•CI)‘-<CDHH-COI-iCrCI)H-H-CCrC)‘tiH-CDCDCrCI)H-CrCON5Cl)H0CDIIpCI)101MiCI)0IICD0C)H-ffIICDH0l<Cl0CltiU)ZlIlLItiC5IIHiCr00CDCOhç‘<(Q•CDCO<CDLti0JD)0CI)CrCrC))COIIIICDH-Cl)H-II0CI)CI)CDS0IIH-ICI)CD‘<QH-C)H-C)1(1D)CrCrC)CDCDCDCrH-,kDH-H-5CI)Cl)D)IH-HHC)SCDCDJCDCI)C)C)INCOLI(0H0CO)H-<CrCOH-H-1iCl01i•HF-1CDCDU)IICDH-<lCI)IH-G)CDHQH0<LICDCDCI)H-1(0ClCD‘ICr‘<—3COCr<H-COCl)CrCrCl)CDLI0IH-CriCrCOtJLi.H-0H-CDHClLI-CDH-Cr01CDHCrkD0(00CODH-‘—3CDrH-CI)COHiLI‘tiIC))CDCDCr-•U)Cl)HCLICrC)CD00CI)LI0HLICDCl)Cl)CDC‘<LILI5CDHCr5’‘tiLOCl)55C)‘tiD)CrD)COCDCDCI)5rrCDCI)<Cl)LIH-II-r)‘50-ç-rtiCI)CI)LIClCDCDCOCD.0HCDLi.COCrCI)CDHiHLIçtCDCl)CI)Cl)LICI)0ClCOLI0C)CrCDCOCI)CDCDCrCD•H-‘IsCDCDZCDICOXhi,CI)COC)CDCI)LICO0H-ClCtCrCO•CDCrLICtCDCI)COCDLIC)•Cl)CDLICI)0CDCDCDCDCrH-H-5ClLIt30lC)Cr(0CDHi•HCI)H-H-CDU] a.,HCi):):‘0O9)2)U)-‘Cx)f-I-<fI-13cf1H-CD9)CDH,I-0-I-I9)H9)C)0HU)1QCDQco9)9)H-C)CDC)co‘iCD9)t02)HCDf-to< 9)c-I-9)ItiI- H-2)2)CD‘-<2)<(I-fI-CD0‘tioH-I(0crQCD—oHHH-DCD9)‘otQHCD2)H-H-U)C)orr2)H-U)CDU)9)H-CDI-icrtQ9)H-JCDH0c-IU)U)c-I-0U)c-I-CDH-•CDb<CDCDc-IZCD02)c-II-U)‘CDH-2)9)9)U)C)HCi)cIH0CD0H-9)HCDc-tc-I-H-9)I0H-(QHçI-U)I-j9)•CDCD(-I2)F-’c-I-Hc-I-2)0c-I-c-I-9):iCDCDHfr-c U)‘ti‘titio09)2)‘tSU)c-If-I-cIfJHCD•2)I--cf-I-H-2)H-U)U)0c-r 009)I--cH,Ii.-H10C)IHC)0HH,HIH-I—’KU09)9)19)I--cH0H,‘tiH<CDH09)9)‘1I--cNU)H-H-H-f-I-U)9)CDIf-I-fII-flCjH-H-H-ZC)0H-13I--c—U)CDC)cI-tQ9)H-IiI--cQ0I-flH-13I--c13CD•CDU)CD0H9)•H13HR’IH-KUCDC)[::I-flC:I•kçr0113-CDI--cH-lCDI--cfI13I-c139)i2)13C-I-çtI0CDIH-I--cüIC)c-tHI--cC)c-rH-H-1Y13013.H(QH-0132)HHf-I-c-I-HCDHI--cH-CDH-13CDtQU)a9)13.0‘-QCt’c-I-CDc-I-CDCDHC)0H0Q013CD0c-I-C13CD‘1Ct)CDC)1313CD13I-flCD2)Q.f-I-fI-I-cI-flCDH-H-13U0H-2)hH1313CDH-I-fl(QCDHCD2)1313H,frj13(QCDHCD9)CD‘<9)<132)U)CDCDQH-H-cI-CD1313H,H-0CDC)cI-c-I-I-cCD0H130’0CDH,H-c-I-I-flC)<13-‘<13HCDCDI-flCDI—I-9)I--cc-I13IIU)U)CDf-I-CDH-H-1313I--cH,fI-2)9)I-flH-<HrtCDU)CDHHC)cI-0’H-C)c-t-13H-,ci0’0<c-I-U)f-I-CD13-13fI•CDI-(02)U)ciI-cH-0IIt50H13CDtJCDI-fl‘-<Hci130F’d(QXH-,c-I-CD0I-flh?JiCD0CD00H-13‘<<C)H-CD9)0hcHH,-U)9)13CDH13130HdNclCDH-<H‘-ijI--c‘H,ciCDF’H0H-U)H,H,I--cH-0HH,F’13•9)13CD9)13CD013(0ciU)c-I-CD13H-I13130cif-I-(0I-hI-fl9)HHCDCD2)c-I-I-fl9)U)H-CD‘ti‘ti0‘tiCDc-I-CDC)13-jI)H-CDI-CDHI-I)U)I-flF’H-9)I-flf-I-U)0H-2)013I--cH-CDU)i-i-CDI—’CD9)‘9)C)2)I--cCDc-I-I--cI-H-I--cc-I-13-jCD(1)02)U)CDF’ciU)9)I-flF’c-If-I-13’H-09)13U)CDH-F’c-I-F’H‘-<13’I--c‘-<H-H-CDH-13U)U)‘tiC)0ciI-i--cU)HH-F’‘tifI-çI-1313’ciCDH-><00’0’f-I-•HH-0H-(09)13CDH0’C)09)9)U)H13-’ CDII rrf-I-CD CDH-m0 CD0H0)0Ct-H-I-flpf-I-1<I--c H ci c-I- CD 0 c-I- 9) >‘ 9) c-I 1:-’ 9) c-I- CD x ID-’ H H c-I-c5’9)CDIIf-Ic-I-f-I-H-CDC)I—ci F’ci9)13II0HCD‘-<I--c U)I-’-c-I139) 13 00HI--c f-I-(0ID’0c-I 0 ID’ CD C) ci I I--c CD 13 c-I- H H H 13 H f-I- CD 0’ 13 ci CD I--c 0 I-h C) 0 13 f-I- H 13 CD ID 9) H 0’ H- I-fl ci 13 C) c-I f-I 2) x 9)I-fl CD C) H- CD U) I CD 13 c-I 2) 13 0’ I--c 9) H-U) f-I 13-’CD 13 0 U) c-I 0’ H- U) c-I- H 13 C) c-IF’ —1C)F-hOS))WQ0ctCU)1tQ0CCDI-F-I,H0H-CDMpCDCDH-CD0CCDH,CDFl0CDCt3c-fFl0CDU)CMI-U)0çtCtFlCDLH-LQ<NH-F-bCD))ç-tCD<1C))CDH-Fl•CDFlC))H-Cl(QCDC))CDC)NDFlC))H-Cr“jC)U)CrC))FlCDCD0FlFl-3CDFlFlCDCDCDC)FlCDNCC))C))C))H-‘-<H-U)<0ZU)CrFlCDCDC))FlHHF-Cl)H-Fl•Cr3YC))HCDFlCDCDCDC))CDCtCDClH-0H-ctC))—çtQQC)C))()FO-C)U)HH,C)H-0C))CHHrrC)CDFl-C))C))HFlHH-CDCDC))0H-FlFlU)flç-t0CDFlC))FlJFlU)U)CfCrC)CDCDC)CD0CDCr(I)ClC))CDC)CD0Z‘-QC))1’JHCDU)HH-E0ClCtC))C))C))3C))CrCDC)Cl0F-I—i0H-Fl0PU)H-FltiH-HFlCtFlH-QU)‘t3<(I)CfC))CtCtHCDHClU)0CDCD(QCDC)CI)U)U)FlH-H-HCDCDCtCl00C)CD00C)FlJPCDti0C)U)FlU)CrFl0hClFl0CD0H-00FlHC)ClH-.CDCl0CtU)Q.C)H--j,<CnC))CDH0FlC))HCtI-hH-CDCt0CDH0I-h0CCtC)C)C)FlC)(C)CrCt<ClClH-C)CnCrCtC)CDU)CD<C))H-CD0FlC))CDrtiF-CDH-C))CtCtU)ClCDHC)H-FlHU)‘<CtFlFlIC)0CD0ClI-hCCfClH-0H-H-H-U)00IF-AFlHH-HClCDI-bCrCF-<H-CDC))ItJU)CDCDCl0<HH-0U)HH-0ftCDH-U)C)>ClIH-C))Flti0ClFlCDHH-CDClCfHC))IM,C))C))CtCtC)H-C)FlCD0‘-Q<0tiC))C)IF-ACDFl<10CDU)CrU)c-fH-IC<CDNCtH-I-FlH-HIC)C)CDCDH-CfJ‘tiC))CCD—CDU)H-o0H-H-IFlCDI-hC))FlCDCDFl0CDClI-hC))ZICU)t3<103CDU)CDCDItU)‘t3CC‘<CtC)ICDCDCDFlC)CDFlICU)HFlC))C)H-HH<CDCDC))HCDClC]IC)lCDCDC)oCDCt0CD3CtU)CtCrFlClC)C)HFl[-tFl<C))Cl:3’<0ClFlU)CtCtH-‘tiH-C))CDI<C))H-CDC))00C))CDCDCDC)C))U)C))<CDlCDCD‘tiClCDH-0FlCrH0Fl0C))çtFl00IC))Cl‘tiCDU)3’I—I,H-.C)‘ti3CCtC))HH,CDCDC))C))FlCD0CtHCDFl00U)IciH-C))C))CftiCtCtCDH-FlFlClF-hFlC))CfH-H-CDlCD3FlU)0H-HC))0CtH3CtCtCtQIC)FltiH-CfCD0H-CC))CDCtFl0C))0CDCDU)H-CfCrc-fCDCDU)C)U)CtQH-U)Cr‘tiFl0IIH-0HC)CDCCDH-tiC))H,1QHFlFlU)1iCDH-H-U)U)C))H-ClH-(C)CDCDFlU)CI)0J’IC))U)C)C))CDI-fCD‘t5C)C)c-rC)<C)HH,FltCD0U)CU)C))0FlU)0CDCtFlC)CC)QH-CIC))H-HClC0ClCDCDH-Fl‘C33CDHb’0CD0‘<QCDCDHHFlCDH-FlFlCrC)H-U)ClCDCfC)H,CDU)U)ClC)CCrC))CfH-3’C))U)U)H-3’HCttSCDCtICY’C))U)C)CC)Ct03CCDClC))CDCD‘HH-HC))CtHH-I-hC))H-H-FlFlCDFlU)FlFlH‘<HC))‘•P0c-fCrU)0‘Ci0FlC))CDCtFlCfClU)NCtCD<FlU)C(C)CrFl:3’‘-C)C)ICD—CtH-0H-CDC)H-‘CiU)CDCtU)H-CDH-)<ICCCrCDClH-0ClCDH-CDCrU)HC))lCDCl‘<U)CDFlCDC)FlC))ICrH-H-U)CtH-(C)33H-ICDCfH0C))CDU)ClU)C))lCDU)3’(-nI-hU)C))‘<IC))•CDCD159the group. In addition, relating the Japanese species to theNorth American taxa of Menziesia using DNA, electrophoretic,and flavonoid sources of data would aid in resolvingevolutionary trends within the genus as well as among theeastern Asian and North American floras as a whole.It is important to carry out these further studiesconsidering the high genetic identities observed between thewestern North American and Appalachian species of Menziesia.This contrasts with the low genetic identities observed amongother disjunct taxa in Liriodendron, Liauidambar, Apastache,and Datisca as discussed in Chapter 3.4.2. Examining DNAdivergence in Menziesia ferruinea, . pilosa, and M.oentandra would permit one to estimate the possible times ofdivergence among these species which could then be compared tothe isozyme data obtained in this study. Since Menziesia isthe first eastern Asian and North American disjunct genus ofshrubs to be examined in biosystematic detail, it is importantto examine other shrubby genera exhibiting the same pattern ofdistribution to form a comparative database. Vicariantspecies pairs in Rhododendron, Cladothamnus, or Vaccinium, toname but a few, could serve as good examples worthy ofdetailed study.Finally, deviations from random mating in Menziesia,despite relatively high levels of gene flow, differ from theresults found in other ericaceous shrubs, notably: Vaccinium(Breuderle et al. 1991) and Leioohvllum (Strand and Wyatt1991) . A closer examination of the breeding biology ofMenziesia would help determine which factors contribute toreduced total genetic diversity and the trend towardsinbreeding within populations.160WWWb1WII000000P)CDCl)CDCL)F-CDp_I--‘1IIC)<<IU1H-G)HI-H-f-HCflCDCI)CDti2CrIH1c1-I—ICr“CD•<•QCCD<C)H-tiC)0CDH-ICDH--I—P)HHCD•I-’f-CD.<JH-I-iIC)iC)CD(QP_I•(C5’-.•OCDCH-H-CCDCDC1CrIF—P)CDiP’CDPPQjOP_IWIorr(DWCDW(DP)iH-O030C)•HCrCPC11<CDCroCr—HOH--uiiZ•o•PJCDH-.IH----H-.---H-.CD•.H-I—(flC)OCDP)CD(QOCD.•ICDIx’0I—flCDIH-(r•.H—I-i•Cr.0HP5Ii-•H-H-.p’jt.-)CTF—-10WH-H0F-<F—sHCr0IDI())Ii-Ct:Icn••CrPH-H-F-NDOH-IØH-QPD.Q-f-tH-•r-0o—1••OI—aCl)I-c-r(DCDCrC)ICDH-CWH-I•H-))O•.Cr_II-H--Cr•H-•O{OWCrI•I•I-.C)OI3CDZP_IH-CI)I-nGCD0H-cOP_I‘IiH-CThnHIP_Ip_Iui0H-—]HCOCDi•l—Q5F—•CDH-‘OIIP)•.H-F-P_Ici.CDCDCDrCDC)P)CD•CrCr—•Crt’JICr<•c-rP)P)OHHQnpF—•ctF—QH-h(J1CDIH-CDCCDCDCrt-rJCC3•P)H-H-C)CDCriPi)CDk<CDF-CDIP_IHCDF—CD•CIPiCDH-P.)CD3iI•.HH-CiCQ_I5—DiCDC.CDCrCrCrH-P)’-jflP_I•p_IpJ—CDF--rr(DCDQ.•CD‘IP)CDCDc-tI-Cr(JPiCD••CDP).DIIOCrC)Q.CD‘-CD••H-g-r<(1)CDWQ-QP)OC5P)fr.H-C-iH-QP)p_IH-0WH-H0iQ..C).OH-CDCr(r0CDCDWli10(1)(1)Cr(QP.CDQP_IH-QIH-CrP_IHCrCD-CD(DO•CDrpCDCDQ.C)C)P_ICr<0P)0H-L’JP_ICrCrF-CIH-OLJ.13—JOF—CI)LflCrCr•.CrF--ICDWF—F-CDCDCDOiHCDOP)(-)OP_IHCD<NI)H-QiCI)t’JOCDIDCDH--(DF-tIF-H-H-.Q.OQ0CDH-’H-CD‘CI—.OIOWCD—-JPiPi.‘-OW.OWiIXQ.fl-Y1CDWH-F-CDP)1I-hCrP_ICr•.H--1H-cj)k1JNOP)P)P_IH-iCDH-CDIXjCj)(DOOCD0(71CDQCDCDCr0iF-aoHIQ-.CDCrH-IH•CDCDP)JThI0H-O’-<H-CrCr10O(X)H-OQ.OQCDH-H(DW•H-C)Hc1-I-)cr1Q.C)ICDI’CI’CI(DC-JlfrhCDC)<•xJP)H-OOCDH-H-ICDP_IHCD-L’iCD-CX)•P)(QflCDOH•(QfrCD—iP_ICDP_IH-OIiF—H-HIQctOCrODWIICD1.C4CD(1P_IO)—OCDCD(QWCDG-H-CrH(DHH-C)•CrIIO1PIH-CDCDCDIo•CrH-’D(DOOiH-—CDH-CDIjCrCrCD13<‘<ICDJQCX)H-C)jP_IQ.CrP_IP_ICD•-CDCD-(1P_IF-aCrH-CDtf—Q.CrQCl)-W•_ICDC)P_IH-iH--fr(()CCDJI)CD•0H•..CD<00(l)0I-P)H-tI(/)•3HCD0Q.CDP_IOHP)•CDC)C)P_ICrflH-CDH-CDCrCDMCD-ClCDQC)•CDCrP••CDCrHF—P_I—0CD)P)CI)P_ICrCDHHCr•H-P_IH•CDCrCDH-C)P_I’tIO—0(I)P_ICDDW•.Cl)0‘CICDCD-hHCDNlNC)CItY———-.J0Q.•I-•IQ-•IF—<OP_ICfl0Crk<Cr•CrCDI.<t-P_I.•CDHflWCDCrCDCD0I-I—i0CDCDCrI-h•Lii162Breuderle, L.P., N. Vorsa, and J.R. Ballington. 1991.Population genetic structure in diploid blueberryVaccinium section Cvanococcus (Ericaceae) . Amer. J. Bot.78: 230—237.Brown, A.II.D. 1979. Enzyme polymorphism in plantpopulations. Theor. Pop. Biol. 15: 1-42.Calder, J.A. and R.L. Taylor. 1956. New taxa andnomenclatural changes with respect ot the flora of theQueen Charlotte Islands, British Columbia. Canad. J.Bot. 43: 1387—1400.Calder, J.A. and R.L. Taylor. 1968. Flora of the QueenCharlotte Islands. Part 1. Systematics of the VascularPlants. Ottawa: Queen’s Printer.Canada Department of Energy, Mines, and Resources. 1974. TheNational Atlas of Canada, 4th revised edition. Ottawa:Macmillan Co.Chan, S.S., R.W. McCreight, J.D. Walstad, and T.A. Spies.1986. Evaluating forest vegetative cover withcomputerized analysis of fisheye photographs. ForestSci. 32: 1085—1091.Chang, C. and D.E. Giannasi. 1991. Foliar flavonoids of Acersect. Palmata series Palmata. Syst. Bot. 16: 225-241.Cheliak, W.M. and J.A. Pitel. 1984. Techniques for starchgel electrophoresis of enzymes from forest tree species.Agric. Canada Forestry Service, Petawawa Nat. For. Inst.Information Report PI-X-42.Compton, B.D. 1991. “Ghost ears” (Exobasidium sp. af fin.vaccinii) and fool’s huckleberries (Menziesiaferruciinea): a unique report of mycophagy on the centralcoast of British Columbia. unpublished m.s.Computer Associates International. 1988. DISSPLA User’sManual for version 10.5. San Diego: Computer AssociatesInternational (formerly ISSCO).Crawford, D.J. 1976. Variation in the seed protein profilesof Chenopodium fremonii. Biochem. Syst. Ecol. 4: 169-172.Crawford, D.J. 1985. Electrophoretic data and plantspeciation. Syst. Bot. 10: 405-416.Crawford, D.J. 1990. Plant Molecular systematics:macromolecular approaches. New York: John Wiley and SonsInc.163Crawford, D.J. and T.J. Mabry. 1978. Flavonoid chemistry ofChenopodium fremontii. Infraspecific variation andsystematic implications at the interspecific level.Biochem. Syst. Ecol. 6: 189-192.Crawford, D.J. and H.D. Wilson. 1977. Allozyme variation inChenopodium fremontii. Syst. Bot. 2: 180-190.Crow, J.F. 1986. Basic concepts in population, quantitative,and evolutionary genetics. New York: W.H. Freeman andCo.Daubenmire, R. 1978. Plant geography with special referenceto North America. New York: Academic Press, Inc.Davis, M.B. 1983. Quaternary history of deciduous forests ofeastern North America and Europe. Ann. Missouri Bot.Gard. 70: 550-563.Dickinson, T.A. and J.B. Phipps. 1984. Studies in CrataeausL. (Rosaceae: Maloideae) IX. Short shoot leafheteroblasty in Crataeaus crus-aalli L. sensu lato.Canad. J. Bot. 62: 1775-1780.Doyle, J.J. 1983. Flavonoid races of Clavtonia virainica(Portulacaceae) . Amer. J. Bot. 70: 1085-1091.Doyle, J.J. 1984a. Karyotypic variation of eastern NorthAmerican Clavtonia chemical races. Amer. J. Bot. 71:970—978.Doyle, J.J. l984b. Leaf morphology of Clavtonia virainica:racial and clinal variation. Canad. J. Bot. 62: 1469-1473.Doyle, J.J. and J.L. Doyle. 1988. Natural interspecifichybridization in eastern North American Clavtonia. Amer.J. Bot. 75: 1238—1246.Doyle, J.J., R.J. Beachy, and H.W. Lewis. 1984. Evolution ofrDNA in Claytonia polyploid complexes. Pp. 321-341, inPlant Biosystematics, W.F. Grant, ed. Orlando: AcademicPress Inc.Dunn, O.J. 1964. Multiple contrasts using rank sums.Technometrics 6: 241-252.Farley, A. 1979. Atlas of British Columbia. Vancouver:University of British Columbia Press.Fernald, M.L. 1925. Persistence of plants in unglaciatedareas of boreal America. Mem. Amer. Acad. Arts. Sci. 15:239—242.C)C)C)C)C)C)C)G)C)C)C)C)MMM10000I—aH)c-rrri-CDrt‘<<c-rçtc-C0--c-CIIHCD())C!)H-H-‘-<Zfl—C)CDMflQIH-Hc-CHC!)CHC)CTht—tiOCD-F-P)(DCfl(D••)0MP)CD-P)f—”OCDMi(DF-))F—rtOCfl0F—H-.f—<)HCfl0MCDGCDODCDF—MO-H•0-C•.-tILJcC).c-COci-•HLirttF--H.M-<—C!)tIDCD1-Q.•M•H.QH-0JCDQJH-CD0’-<CC)MQ7OH-CDHb-hMSCDt3•C!)F-C!).LQ.U1P).CD•t-ODIOC!)ODH-H•f-fliH-)c-C.C!)QC-•-OJCDH•H-‘<OQJMC—1iH-CDU1Dc-tHrtH-••-h.H-MOD)•I“ODLnMZH-’OCDQOCDH•WrtM-.cZ.F—H-00---1MMCi)Cl)H-P)Qç1(QH.o-)c-CJH•Hi•D)MOC)C!)t-.j—.](DODcQC!)H<CDOH-f-Cc-C.<••HCDOH-QciCDD1I1I))M•QCDCDCl)MO<H3O(DH-•(51CPQ...Cfl<D)CD0Mi)Cfl’j‘--(DJciH-l-CDH.DCCC!)MHCI)x’0HH-HCnc-t.•0•o•.CMc-C(DC!)))Mici(DOM•OOC-tiJi(--‘MCDHCtTIH-c-C-5c-tOMH-iCCDHCCQI)O•C0.c-t-.).oO1ciMCDf-C<c-rc-r-I-hMCHCnQQ.CDçCHCDIC(QHWH-(Dc-Cm<H-MH-0•(DO)QQ.U1H(Dc-CrrJj)HOMrCC!)OH-HC!)CDCZP)CD•()QH-C0))iP)H-‘-C!)P)CnCD•CDDi(DC!)Lfl(Qc<H-HHHc-taMCDH-•c-CDiOHC)MCtit-H•OD)ciCD(DODiH-C<c-CQCcn••Di.CDCDJ.oHoctMi(DCC!)CI)H-‘ijOMiHHc-CMMO-i<M•.H-c))•H-<.•HDiH?C-hC!)D)Di’t3CO00(DIeHOOH••D)c))CCH-ci•c-C•D)rCCODOC(QCD-IDiI—MWHMc-CI--MC!)ZrJH-QJojOH—)iDCMHD)CDCDH-HDiDiCJOC0-.)I-hCDCn.(DOCDPlHQOD••H-c-tO’.‘MiiIH-•.M••(iiH-c-CCDOH’Dic-CaiPlO(Dc-CPlF-O•(D.c-Cc-CH-H‘<MiMc1OH-I-c-Cc-C-CDO-H-H-MD)0.HCDC!)c-CH-c))—.)H-ZH-tDi<0H-HHCZ•‘CDCDH(Dc-C••0CC’CJ)(1)c-toDic1’<CMO0MHCDHciO)DiH-.Mc-CCn)H•MCDM01MHCDc-C•ortOH-H-CDH-Cc-CMCDMCD1<JC!)MOCDDiQ..CC!)p1jCDH-H‘jc))H-.H-ciC-.4I—hertDiHCDODiHCDCDo’Q(DtYOH-CCDbJHC!)H-•D)c-CCDIIc-CC)..CD•CDCnPP1(C)‘-<D)MDiH-C!)H-M1Hc-CHMCC(D00<H-Cl)CDP)CH-H-CD5CflH0C!)—())MDiCMi(DOC!)c-C(I)Mc-tCDH-C!)CDO)H-OCMC•P)ciHCDQ..CflH-C1)0DiciDiOC!)c-C0MCHc-CMiC)DlH-CDH-CDDiH-H-C)..Di•OH-OMCDC)H-COPIHMCI.c-Ce0.C!)CD--h’MC!)MC)..MiQMiMic-CQ..HP1C‘<i‘<c-CMH-H-H-CD<0Cc-CMc))U-c-CH-C!)H-H<000DiOH’•CDMc-CP)L3jDiC!)H-.H•H-‘uc-Cc-CODi0P)c-CMiP1HHCDCD(QOHOc-C<iOM’<.c))HCDQ.P)HC!)i—Dl(TIC!)C!)DiOCDMCD.DiCDHC)..••CDc-C(J1c-CMiOP)P)HC!)NMCDC)..C)CD(-C..CDCDH-Hc-Cc-CCl)CC3CD1<C)..MH-HPit5CDC.-•C)..C!)H-H-H-CDCDCDC))C))C))çt0-rtC)C)U)çtU)I-IIU)<U)H-C)(1)H-H-H0Oti(DQC))c-(DU1(tr-1o.ICDHHr’1CJC))j(QU)I0HCflO000C)<DtI--o0H-I-tic<CDOC)J-—(DtCDHQctI-jo0-Q-H-H--•.I-iC))QjH-Q.C))CDF—”0I-C))-I--rr‘tiNtQ<C-tc-t1.H-Z•F---•CDC))•HrrrjiC))H-I-jC4CDE(DOt’.)CDJH-300.rrrtrtC-.(DrtC--0CrrlQ.C)).I-itQj•—Jrr.CICDc10—CDC)U)3tflI-jH-.frj-F-.C))Cr.(DOH-CDIOU)C-]CDH-.H-U)li-j..00-1C))U)t1li-1U).li.C))H-J.H-U).II-jC))liH-•0liCDCDCI05li•C))U).CIfrjCrtQliH-CrI-jrtMliQC)U)U)JU)li.(1)CD0uJC))‘.OH-CII-C)00r—Q-jo0<—c))liHctH<DliH-C))’<CD.0HliI-jH-’-<•I-jH-iIJH-H-HC))CDC))U)JHQ.0LjCDD•CDJlilU)‘deli•0liCD(I•CDCDCIH-Cli’-OU)liI-jt-hH-liU)•OC))‘.0C))N-(Q—.J•f-rCflUiOU)U)OH(DCICr00H-<(-i)WODliH-C))W’<.H-•..H-lU)0’ti’.0C))frjQ.OC)).CtH-H-.H-H-CeCU](DC))U)—)CD•0H.rrCDWI-jJH-CI-j(D.flfrj(fl•.H-CD(DU)QI-dCiCC)cl.<OH-.li0H-liliCDCDC)H-H-li’-c)<HCrH-H-I.Q5•Il-liH(DC))(DC))liCDliF—’<lililiU)0CD—C))hjH0•HliliH-ctH(DOI-jH-U)clCCDliiCDC))”<•(DI-jiJ.1)iCIc-rU)DliC))0Q.CDU)I-j..I-j(DH-0U)CItQH.tjH-liC))C)).F—YfrLii(DC))40CDCI’0C•—0li0-liH-CDNH-liW(DQQQI-jCDH-Q.(flU)ONcticlliCrCDH-’-<‘<crQ.CDHCCDc-tI-jIc-tU)C))(D(QC))<HU)C-4—liCDH•‘jC))HCrI--•0i.H-liCDH’H-CCrC’.DCflI-jfrjCDH-DC))0•CDH-ittHCD<H-‘.0.H-F—’OH-•c-fliCr)OH-OM,C))U)I-j-(D.C))(fl(D(fl(ODC))•I—IMiHClit’.)(D<U)0iH-C)rjc-r(DU)C))Q‘-U)hiiCrI-jcflia)CI(DU)C))jc-f.frjU)I-jC)).t-tiH‘<C))QCliH-’-0U).H-C)>U).H-.I-j(DU).CflC))H-H-HH-I-ICOH-U)OHliCX)•H-<0U)HU)CThF—C0’.oCIH-0I-t,CrF—H-C.H-(DH-‘.0U)QCrCD)c-t’.0..(D-fQ)C))C))C))-C•CDU)-C))I-j’-i.CU)‘-CCDQi•C))t-jCWOH-I-jrftlc-f-fJ.ç-f‘.0U)I-j’.D‘-H-1C))Cr.CDH-U)C))CDOCDf—tQ0•CD‘-CrC—)H-CDt’J.CliU)•H-C)—COI-jcflDCDhU)C)0liH-CDM)rrC)frjH-’ICr0CDH-Cr0Crct-li<HI-I-<t••ZCrci-CCr0H—CrNCDHCDH-CICDC))HC))HC(D(D.I-jI-IC))H-U)•‘<liI-jC))U)HCr0CrHCr0C))liC).’CU)F-hjI-jQ‘<(DC)-’I-IC)H-U)CD—1H-DCrlitiH-NliC))U)ijI-t,U)iCr•tOCDQ.CDC))U)U)U)0WOC))C0.liQ’<WC))U)WQU)QliU)CrH-c-toIU)liH-CL3ititOliCr’.Ot-h.bN‘<CIH-H-I-h.(I)I>J<C)).-CDIU)lit-i.C)):C))I—1C))’-<U)OH-U)(DC))O.I.QC))H-HH-CrtOCrCoI-jCrCDU)0U)C)I-jCr(DCDCIU)C)U)CrCr0CD’<liCDU]liCD•C•C))i-RH-.H-CDHCDQ.liH-.U]00H-hiI-jU)ZU)CDC))liC))CIOCrI-joC))frj’.oH-U)<-Clit-hU)C))H-H-CD-CrQj—1li<W(DHt-hH-C))lic-rH---.)j.(1)U)H-,OU)C))liQ.CDCU)CH-U)’.0H-I-jOHCrC))‘t<H-C.C))(XlI-j—tQliCl)0CrU)OH-liC00liI-HIH-.0I-j.0H-•H-CrCDCDCDCOI-hHHc-f—)I-hHi00H0IXCrH-IHliU)F-’U)H-1t’JH-CH-H0’I-hP’H-H-U)‘<U)IC!>CDU)liIC)Ch‘-00C)—Ctli•DCr.C))..(D-f(DU)H-•C))..••l(DC))C)).H-INH0(DU)OCtCr.I-jCrliliI0H-H-I—’ICfH-H-‘.0H•CDCIII-hC)).CI)•Ctoto-H-CDI-jU)IHoliI-1CjCDCDCDCD)‘1CD000H<00CDftH-H-H-H-C)C)U)C)33)H-H-CD-IiQC),--3U)P)CDC)CDOWH-’-HCD‘iHCD11JF-C)CDH-Ii-QjP)H-Ocorr••OPcni—-H-P-oH•ZjU)jfrJ><CDQ.CD(DC)CDft.cr0-.OC-t-•U).U)H-U1C)clQ.U-.•.0.CC)C-43U)Hhftt3‘NU)CDCDEfti-3H-•CD(QU)M,(H-.‘<Cs)cyr.’U)Q1HCD•H-CU.-OCD•••0•CDC)H--CD—:is...ft.C1i-C)C)H-U)OU)’<CUftC)CU—ICD•CUCU(QCDC)H-H-CUftU)0CDI—H-‘C5ft•.U)..Q.fti)<.HU)F—f—“0CD0’Z•••CD0-bCIC)Q.ItJ’-o0oCDOHC1-OCUCDi-t,iF-i)U)U).fs)Q13CX)CDo-(-.)H-13•C)cOCUF-CDaDNH-flCtC)Ii—f—C)OC)•ftCDX.L.J‘-D(Qr-t,.Wii).••C•0.H-.CD•—.]C-4Y’Cs)C)’-oH-I-C))I><F-I1iH(DO’—JQC)CUO•CDOCU•)ftWCULCDDDF-CU,TP)(DG)0ftF-)HOHflaDU)aDft0C4tJCDH-tjU)LQC)O<CDftH-H-CU•L..>C-tMtç.C).I)QC))•‘ijIH-<11U)OCDoftC))0CDQMHH-CD-hcHOIJf-.<CD‘C33C1H-M,CUQCUrtUCUftCDftflCUZOH-OftQCDU)dH-Or-hH-CD’ftC))CDU)NHY’U)CUQ(Jlftf-H-CUr-SJ’U)-3(1)0IH-tiH-H-‘‘<HC)tiU)HHU)tiU)CD(DC))lCD—<C)H-CUCDQr’U)HDftQ.CU00CUL-lQC1]f-CDC))U)C))U)H----H-<HU)><OH-’ftQ.ItiD3H-C)CDH0ftQC))t-h’tStiCDrtCDlH-)CUtQNH-QC))CDti’-D•3tiO<ICIU)H-li—i.C)C)CDH-CUQ.CUH-’.r.H‘tJH-CDCDIH-ftHCUCDH-CCDOHCUft(-JCU..‘1tLiIF-H-)ft..C)ft.F-—CU.F-LU)CDftU)M)CD—1OCDLQ(flCD‘<CDH-0.CDCDOCD•.H-MiH-lU)‘-C)ftC‘<CUCIU)HtO®.CI’-CU0><C)0WU)ftlCUU)•I—hU)CDH-CDHCD0t’iWftCDCDHH-H-ft—’-<U)C)CU’-<F-CUU)CUtSICItSMI(Q00U)<00<ft.F-ftIH-rtC))C))tSOH-Q.WC))MCDtSOftCDC)CUt-SCDOOtiHCDft1-5CDQ.NH-CDH-ECUitSUltSH-CDlitS’tiC))IiH-(Plh0IC)liOftH-Ii1<QMQW(DCU’CI‘ICUCUQU)U)tSOftCDrJlU)U).U)’<M•HtSHti•..H-CUfttSrtCU”CD•CDCUC)CUU1QM-HCDC))ftft-CD(DOIICi.h5ftC))H-•H-HCX)<C)C)—Ift‘<H-U)ftC)NH-U)•‘T’QftZtS—I<0tSHftC)Y’<HCIH-ZCDH-HftCDH-OCD“CDOCDU1CDC)tSU)‘CIHSIlU)<(DliOCDiiHOftU)liH-,•.()H-H-3CDHC•OtS’CD..ftC-1Hr-cH-,liCDCDU1Z‘CIH-liU)U)H-Iiftr(D(QCDtSMOçiHC)oOHUH-CD.CDO..U)li‘U)‘<0HCUCU—JH-HCDOCUP)0‘CICDtSCDCC))C)‘CIHU)Cr)ftU)IC)C))U)H-,ftIi—)CDH-U)tSCDHOiH-ftti’CIcritS’tS’ftc’-ft.OH-H•C)C)HP)CUMtSCUCDCU‘-.0H-U)COOH-‘CIU)J’00•.H-CDCD-J’CUftH-3•(QrtHrt3•QQHCD-c0OH-ftCDCUP)•CU(QOC))HU)H‘-0U)CI’U)U(DOU)CUU)ftCI’C)ftH-U)U)•CDWCDH-CUeCUtSCUCDH-H-0CDtS’CDftCU••r-hI-iCOOftftliU)CDQQ.HI-htSrttSH-l-‘-50lhh‘<ftO—C))H-Hft’CI.H-Ml-5oCDtSftC)CDU)U)U)HH-CDQ.Q1<——1—:tI:-’I:-’t-4I:—’P)P)P)0H-H-H-H-H-CDCl)IICD0Cl)CD0çrCDc-tCl)c1H-CDDl0H00-HCtH---QflIJCDP)M(Q—’CDM,i2.CD(-,JCDCD(!)i(DWP).HjjhhOF—I(DjC11—’MCDQ..HP)P)ZOCI)HiOODH-5ri-OM-,0rtF—YIiH-DlhF-Q•0-ctCDH•ri-l1OS(D•P)frO1—Q.•CD(D•WOIN<iDl-•OI-iCDQCD-CDCD•OH-I-IhiH-Il(rIl-DlIjCDIIOIl.H-I--a•P1H-DlCDOIJCDP)H-Yri-t(Dl-jOH-Oj•I-i00-OP1-CflH--Mp1F——I-OHri-H-DlCD.CDHOLQ.H.‘<13iOH--HQHCDIlQ.rrH--OCD’-OIOD‘-DH-)IH-H-H-JH-<itJtO0t’H•H-H-ri-ri-uCD.IP)ri-P1IlCD•N‘dF-tQP)•..ctlQ.Oct.Q-fl0’<.CDt’-)(.)c0(I)H‘<OU)OH5OIlH-H--h’<H.DlIl.1c3•Q(n)Dl.--JOHtxjc-i-p)U1O1HH-CDOQL’J—J‘-iiQH-(y-ci--C)0CDMOP)I(f)H-H-IHOWIl-QIl—O(0Q.U1HICDCDCI>‘çtjQzj(DH-I—’ZiH-•(D•‘ICD•C-’-)IlOHOcl-H-iH-Dlf-CDi-—OHICDF-IlZri-OCDQtQICD••M-C-i()1P)CDOOZCDri-H-Cl)‘-DOIP)P)CDP1tY-DlCD•0-‘1X0•Hcl-CflOHCDiiO5a-rtIlH-cl-HbH0CD-CIQt’tODlCDH-.QCDI-QOCD-—1H-—0ri-J’<H-H-CJ)CDCD•CD5O-H-‘-<CD-•QCnOP1.H-CU.CD‘‘-<H.clP)CD(DP)OP)i--hri-(DCI)LQ‘JIlCflH-H-CDCDHP1r00H--H-‘<H-H-OP)ri-0CDCDIl—tTt-jH-ccH-HOb005—5Il0OQ.IlO•ri-CULJQ(D(QCDHH-PlO(DO0<l-1,HP)<QOCDO•.çtrtH-Sp)-(Dci--crIIOHP)CDOIlci-DlHH-HOIlc-t:H-IlOIlIlIl•00PlOt5PlOU]F-’-OQPlOOP)(.‘-)LYl-t00MCDCD0OH-O0QP1--(DOc-i-I-j(DW.o-<P1HH-CUH-Mci-OQ5CDCDri-Q,Q.,I—•-Pl0CUP)c-irtMOCU0C)CD-MHCUM-Oo-Il-QO0--Ilci-OHH-’<Cl<IlIlP1‘100U1Ci0HOH-’-]MH-H-00CD‘<OQP)PlCDCU]O10O•P1•(-‘->P)O‘-30CU0H-O‘octHOCii0rtCr.—J‘DHMCD•P10M-Ilci-’.oOplP150.U]DCD.—O.‘CD‘-005CUCD—CUflQ.,t-t,H-H----)•M-,03<•frH•M<(DOOP1ci-ri-Q0U1c-i-CDOIIiCUtljOLiCiH-,QCDHP1O(Q-ClH-CC)lDlIlHOHC)CD(DO-c1-0O(I)•flH-5IIHH-In-H-CH-•H-MIlCD(ñCD00P1CDH-H-P1II--ClP15rtCU0•OCZ0On-CflMt’JflH-CDOH-CDZ-<5CD3H-Hctcr<PlQ•OP)0ci-Oct‘00<-Mi•Cl(DOri-CDDl-ci-0(l)ClH-Cl)HOIP1Cl)CD0Ct)Mci-•CDO0M0•-H-MO•0CDP)F-IlI-h•CDCflCDH•--5CD•IlO—(OHC-ci-H•H-CDO<0<H-DliiO0ci-tQiCDo•CUoIlCDOHCDbrtC)OWOHrtOH-Jici-M-MCUHDlQlCI)H-I-hCU‘Ort00lioC-40I-’--H00F-’-ri-O5CDCD-CD0(QOP)H-Ot5OClPl.<HO0ri-H-3-C)P)c-i-C-Dl‘CiHCD•ri-H-.c-i-OP15H-OCDII-5MCI)Dlci-OH--CDDlrt(DCD•-Q.<CDH-Cl)SH-C‘—O’H-I—’U.050.flHCD<DlO0-H-H--H-I-hP)0P10—G)OCD0H<Il-ct0HI-h(QH•ci-(Dc-I-aCDCDhc-i-CI)MHCI)•IlCD•M(QC-H-.•O••OCOIl0CD‘ii(0C)CUCDI-h000ri-CDM-0(0H-Dl‘-oP)0——C-’-)ci-0.HOci-IlI)IlCD•COH-IICDM1<.W0H-a’O--ItOp)•CDDl168Michaux, A. 1803. Menziesia Smithii. Fl. Bor. Am. 1: 235.Mickelson, D.M., L. Clayton, D.S. Fullerton, and H.W. Borns,Jr. 1983. The Late Wisconsin glacial record of theLaurentide ice sheet in the United States. Pp. 3-37, inLate Ouaternarv environments of the United States, volume1. The Late Pleistocene, ed. S.C. Porter. Minneapolis:University of Minnesota Press.Mizushima, M. 1972. Taxonomic comparison of vascular plantsfound in western North America and Japan. Pp. 83-91, inFloristics and Paleofloristics of Asia and Eastern NorthAmerica, ed. A. Graham. Amsterdam: Elsevier PublishingCo.Mueller-Dombois, D. and H. Ellenberg. 1974. Aims and methodsof vegetation ecology. New York: Wiley.Murashige, T. and F. Skoog. 1962. A revised medium for rapidgrowth and bioassays with tobacco tissue cultures.Physiol. Plant. 15: 473-497.Nei, M. 1972. Genetic distance between populations. Amer.Naturalist 106: 283-292.Nei, M. 1973. Analysis of gene diversity in subdividedpopulations. Proc. Nati. Acad. Sci. U.S.A. 70: 3321-3323.Nei, M. 1987. Molecular evolutionary genetics. New York:Columbia University Press.Nikias, K.J. and D.E. Giannasi. 1978. Angiospermpaleobiochemistry of the Succor Creek Flora (Miocene)Oregon, USA. Amer. J. Bot. 65: 943-952.Ohwi, J. 1965. Flora of Japan (in English). Washington:Smithsonian Institution.Orlóci, L. 1978. Multivariate analysis in vegetationresearch, 2nd. ed. The Hague: Dr. W. Junk B.V.Packer, J.G. 1983. Flora of Alberta, second edition.Toronto: University of Toronto Press.Parks, C.R. and J.F. Wendel. 1990. Molecular divergencebetween Asian and North American species of Liriodendron(Magnol±aceae) with implications for interpretation offossil floras. Amer. J. Bot. 77: 1243-1256.Parks, C.R., N.G. Miller, J.F. Wendel, and K.M. McDougal.1983. Genetic divergence within the genus Liriodendron(Magnoliaceae). Ann. Missouri Bot. Gard. 70: 658-666.169Peck, M.E. 1941. A manual of the higher plants of Oregon.Portland: Binfords and Mort.Pilou, E.C. 1984. The interpretation of ecological data: aprimer on classification and ordination. New York: JohnWiley and Sons.Porter, S.C., K.L. Pierce, and T.D. Hamilton. 1983. LateWisconsin mountain glaciation in the western UnitedStates. Pp. 71-114, in Late Ouaternarv Environments ofthe United States, Volume 1. The Late Pleistocene, ed.S.C. Porter. Minneapolis: The University of MinnesotaPress.Powell, C.L. 1982. The effects of the ericoid mycorrhizalfungus Pezizella ericae (Read) on the growth andnutrition of seedlings of blueberry (Vaccinium corvmbosumL.) . J. Amer. Soc. Hort. Sci. 107: 1012-1015.Prest, V.K. 1984. The Late Wisconsinan glacier complex. Pp.22-36, in Ouaternarv stratiaraohv of Canada - A Canadiancontribution to IGCP Proiect 24, ed. R.J. Fulton.Geological Survey of Canada paper 84-10. Ottawa: Supplyand Services Canada.Radford, A.E., H.E. Ahies, and C.R. Bell. 1968. Manual ofthe vacular flora of the Carolinas. Chapel Hill:University of North Carolina Press.Rehder, A. 1940. Manual of cultivated trees and shrubs hardyin North America, 2nd. ed. New York: Macmillan.Rohlf, F.J. 1977. A note on the measurement of redundancy.Vegetatio 34: 63-64.Rohlf, F.J. 1988. NTSYS-pc: numerical taxonomy andmultivariate analysis system, version 1.50. Setauket NY:Exeter Publishing, Ltd.Rosendahl, O.C. 1963. Trees and shrubs of the Upper Midwest.Minneapolis: University of Minnesota Press.Salisbury, R.A. 1806. Menziesia alobularis and jvj.urceolaris. Parad. Lond. tab. 44.Schnabel, A. and J.L. Hamrick. 1990. Comparative analysis ofpopulation genetic structure in Ouercus macrocarpa and Q.crambelii (Fagaceae) . Syst. Bot. 15: 240-251.Slatkin, M. 1985. Rare alleles as indicators of gene flow.Evolution 39: 53-65.Smith, J.E. 1791. Plantae Icones med. p1. 56.Cl)Cl)Cl)Cl)Ci)Cl)Cl)Cl)Cl)Cl)Cl)Cl)Cl)çr000000000CDC))H)—HHHHCDC))C)Cl-rtctctftC))C))C))CDCDH-H-H-H-H-H-HHcCWCDCfl‘tjIcIHC)-CDCl)flcnjC)flC0Cn0F-fl(D-Cn—C)CnCl)—tQ-Cl)t’JH--IFJ<-CflIHHIOWCD<i3-rtHO-HO—CJ)J-C))CT)C))-_)C)t5CDIC))hhC))C)IHrtZHOrrHC))CDHC))CDH-Cfl00I-hC))I’(-I<H-N0H00jH-(QL’J.tQ•C)’-<•(,JI-.C))IJI)COC”C))•H-h.$‘t5.•Cl-H--H-H-.iJJFC)CnL11u1.•flCflH-.IC)Cl)(DOL’JctOOL’iH-H-t’iFl(DLI1-hCl)t1Fl•C)).FlJC))H-jI-hI(--cr.Cn1t1-H-F--j.QC).ct.C))C)).CDCDH-tiH.CDH-IJCDHH-0<H-tQQjCl)C))IjC)C))C))NQ.H-(DC)C)HC))CflC))H-C))(IC))H-cr(Q’<C))C)H-HC)•CDC))C)CDQH-DCflF-U)’jCnjC))flQCDC)-)H-C)OH(DCflF-•I-t)cr—CCIcl.0c-r.Fl:iQjC)OCItn—iC’J)C))•F-’-•-OH-H-OQI-hCi))I-’-ICl)C)CDNC)’1<OC)iOCl)C)—JUI0.Q—Qi.(D3tJCflH-C)UI0C)).OCI(fl-h.HC)’(Q.CDZH-(Dc’-•.CIICl)CI•‘—3(Q(DLiCj)—‘CJ)F-t)Cl)—D•H•0-3Fl-CD0•-H-0I—F-t-iCDC))•H-.•CD-C))-3CflM,’<F—c-CW(Dct<I-i<H(D(DHH-•Ic-Cl))•c’C)‘tIFl‘-ivCl)CDC)UIH-C))ctGC))ctCflFl1-h••iCIW0H-Cl)t••CflclH-I-iH-c-C(DH-I--C)FlH-LflrZFlH-•CflCDr-i-H-Cl)H-QOc-rC)ctDC))(IC)<-H-’(IC))-’0CñH-(flcICDH-cfl-HF-AQCD<Cflj•’HZI--”Cl)(DCDHHi—frl(000C))-IlCiCi)-CD(DCD(DCD(flLJ-‘IIM,•CflC))CDH-CIG)I—Cl-GC)C-tC)FlFl(Drtç.I-i.j(QI-c-CCD•D2Cl)Cl)•1s)H-i-3U1ZH-C))CDcr-•CD(I•CDC))I-ICl)‘(lCDH-H0•a-CD0CDF-’I-Ir’CDC)CDc-to(DHH-HH--•t”I—iOCiCH-l-WctF—Hl-c-t.C)-DCAM-,H-C))c-tH•C))•C))’(l.•C)Q0oC))D(DC))C))OctF—0O•H-Zk)’DCX)H•C))W.c-tCflac-r:i•.F-.H-HC)CDCl-C))CDII—’CX)I.‘iCIO-WH-•‘(HH-HCDIC))Cfl‘-lIC))O‘—0C))H-QjCD‘-I0CD-C)-Cl)D)C)))-0UiH-cl-•(I•F•IC))C).FlCDCDCJCl-CDF—L0C)Ctb-.)‘(IH-‘-<M-IiF—’CflC))•H-’-O“H-H-Q‘-DCDCDHt’I00H-,C,II—’-1-J(DC))CiCDH-FlCDUIC)IHOC))iFl‘-oi•C))Q0H-hlHCDCC)C!)Fl—]JOCnC))CThH-Cf)C))C)U90oH-clCD•F—---c-CCtCD•c-C0(Jc-t.1100‘<H0ctQ(l)O•••CWICJiCIH-CflM0•.CflHQ<•oUiC))U’(IOhhhi03CDC))‘tI•—JOH-r--IH-0CDc-COT)Ci•ti•C)0‘-OCDc-tI(DctF-’WQ,L.JH,F--’Cl-•H•c-CIIC)H-—CDCDr-tC)F—1iCiI—3OI-(QCDH,JH-p-C)C))Cl)H(D0c-C—--tQ‘<NhiCiC)flH-CD0CI—110UIC))QO‘-3hiCIIQ•<.OH-HC)UIH-)WT)<c-C.Xhi(X)iCiL-)‘(IOF-hC)HOWCIhi•.lC))C))-H-H-’-DMM-,CnC)HCDOI(DC))Cl)I-’H-C))PDC))I—hc-rhiH-C))I-I•I—hLQC))—JI-h.-•IF-HiCD•C)HCICi•H-HCT)HCflH--3hiH--•CDC-•N’<IHØ,C)C)XC)).C)cf‘tiQ-ci0C))jhi‘-<‘tiIC))01-J3c-CH-hiOQ0C))H-H-CD—)t3c-tH(DtQCflCl)I-’H-,0)0C))C1H-C)<<C)—ICflH-CflC))-C)O(DOHQ..(DXQC))••(t)(DC))H-i50H-Cl)00OC))H-C))0HCTI-hC)I-Ic-CI-hH.CD-CDMC))CflCICD1iQH-H-WhiHCDCT)CDH-0iC))C))C-t<0QL-JM-0000I..H-CIC)0)C))CDCIH-OH-—1C-fUic-C1-)—F—fliH-0c-tI•<••C))’<H-H-C)c-C•c-C-CI0F—”HCIOCD0‘-3‘-3‘-3‘-3‘-3(1)rJ)000H-H-I)‘<<II‘QH-H-,<COCtCt0COCDCDCDHiI-hHCl)Cl)I—hHHCD<()H-HiW<3CDJ’E‘F-HCDQHQ.1’I-OOP)CDZHCDHCnfrcI_’P)rP)E’rJOP)-DOliCDCDO—JHili<OF—’-<HO0<P)H---H--H-P)I-••-zCDNliP)iI0liOcD—cn1C-iQO-cli-OliiOI--tCnQH-.i<liCI)liH-CDCnHI-rtH-.•Ctc-rhWCDNf-P)H-lI’rJHctD)—JP)•HCt-.hrrOH-.P-ICOH-WH-LII•H-U1H-CD.OH—irrP)CDPJ•H-H-I-sQOCtO•(Dct.Cn..tP)C‘-O0CitiOH-hCc1-C-OC-liftP)H-P)CDliCOrrtl(D•H-ct•WP)CD.‘-<O•I0••H-P)•J•P)-0.3Cl)I•H-.Cl)H-P)OCli•F-fliOliL’ICDLIIF—OH-CDOctP)lilictH-P)CtP)(n’-<(X•tjiliCD.••H-OrHCDCDDI-ICDliH-eliD)H-liHf-P)OCOHICD•ECDCfl—1Ct’tIHH-COH(-JHQliP)Q.‘-‘o•liCt‘<P)HHU)P)’z3f-----1liODOH-MOH-I1C1Q-P)•OCOliWf-OP)•CnP)PiCDCDI-I-i.••NOt3IH-3iCDCDU[-hH-QH-’-OH-liliCt‘-<(-tUiH-UiC).0H-I.t5O•‘-<•(QO—-1Cl)COD<H-H-Ct•HP)ttUiliIH-Cnli•Cl)(DCI)ICI0OCDliOLIIli-P).li•P)-o(QIli••Hi-••IQO-J•DI--.<(QliCDX1110CI-C))C1Ct4IliOdCDP)I--hC4CDHH-’-rj•‘zCDCD•H-QLIiWCtILIKDCJP)P)(lrtP)WICI)-li•OHCtH‘-3CtdH-Il‘CC))a\ZIlOColiC)C:H-liO)ICtZCD0H-H-X<CDP)OP).DH-1JCDli0QUiIC)JOQH)>OIICOliCDIilioli‘r)IlP)(l)C))CDLQIlrtH-,liCn••Il—P)(DC))H-WIliCtC/)<<IlCDflHCtli-3L’JOCD’‘tIeCDQHO<H•li•DliCtCOOIlCD••P)Ct0CDCDOHCtCDHP)Cl)(D•liO•0rr‘<OQ-COli1IlIlOQOUIctliHCDCO’liOMi(D(Ct)CDCtCDCDctli0IlCtH-II(I)<H-‘I•0•H-liCDH-CtCtH-H-HC/)CtH110OCD•L’IQrtH(l)HHC))(t)Illi<U)•OC))’-DOH-coCtP)CtIlOCnCliCD•C))IIbCD•HCtli03NOODP)CDC•H-CtCtCtC<OCDCI)-OX3H-C))Ui-<liUililiOHHO‘1liCD•0lIP)a-C))OO3C))CDIC))co•M-CtP)(DCDlIP)IlCDCOOOG’(-QOliCDHH.“CtCDCDIQlicXSCDli‘CSH-C))liCtCtCtliP)•.C))OliH-CDIC))—1HI(DCDQ.CtC:OIlQliCDliliHCDHOICI)C))•t’-)1QIlli•liIlC))liIlCtC))-3(DlI—Q-C))G)H[-rli•.•H-HliliQflOC))liC))CL)HP)CtIC))C1ICIrJ.QctQ<HOCto-CD•JtQCO’-<CtCOliH-liIC)LII(.)ICtcnOH-CDCDCI)QQ.F—aII‘-<Cl)‘CHliP)HP)Ili(DH0IC))OH<Q.OH-CDPi•.COOC-iCO’CSCDCDO.CDCtOHICDP)CDWIC)O’-<OCtHCl)H-Il•CDliotQOH-CDCOOIIli•Il‘(!)O•IlliCt—JC))H-0‘-00CtH-(DH-C)(flU)—CtCtWICDCCI)CDCt’-QCSH-W(QWCI)‘-OH-CDliLIICDHCDF—(DIl0iI-P)C))H-OOIP)H-00“(ClP)Cfl(D<Cfl<C/)Cl)P)IlO0,CI)O’CI•HCtOI-‘-oliliCtCl)liCtCC))CtIlCX)CSli•CDIlliliCDC))CDCD’-<H•CD‘)P)OH-HOIl‘-<0)H-0rI-H-rtCflHC))CtCtP)•0coOHOC)OH-CtCl)C))OCt•OH(1)QOliCDI--hlilI--hkU)CtP)CDHCtCflIlH-IlOCIDH-S(QQ.C))t-çiWCOH-’C3CtP)H-CD0(f)H-’-DH-HiliCD(OCt•.C’CtH-HOOHHOHCDP)CtOCOH0IIQo-OCS’Ctli•liCl)CDI--hliCDOC))(DCC)—liH-H-O•IlQIl(DCL)HCD(I)•XXliCt0•NCl)••IlH-I—iIlCtLJ-ON‘-<P)Qli0CI)-QH-JcooIIH•••-JCQW(1)liCl)“I--hC))CDH(DCD0liH000H-H-CDCDCDCD(I)P)HF-HHHHHCDc-I-U)I-hI-hHHc-I-CDCDCDCDH-H-CDU)CDU)MHZ-OM-0<<3Q.H-H-H-HH(t)-LIJH.I-<Z(DO5M-,F-CDCOCDD)(1)0O-c3CDMic-I-OIQMMCIU)OCCMOMOc-tMMc-I-•.C)’-H-tI-Ø•H-CD(1)QQMc-Ic-tc-.cI-M.c-I-.U)U)(1)c-I-.c-I-.cr.F—5.0H-JCDP)UH--H--•C)HHMH-flH-tYC)d’CDCD••P)CDMH-Cfl.CD•000CII—1ZCl)H.•MP)MU)•c-I-•5‘-0(1)3H-Cf)lP)U)c-I-H(1)••j’ZIQclCDic-I-P)cfl•JiH-CDP)MOH-P)C)I7c•H5HP)HUiU]•.ctP)iflHc-I-MU]Hio•-oC)o.•c-I-’-oQOCDDk<-Q)J(J)IJCD—)‘—]H-OCDHCDHH-CX>IjH•Ci-Q.W—HSI0C4HCDCX>‘I1(1)CD0•5uiL’J•I-Ol-tOHDH0H-OH-•IiW‘-0HH-.OCDLIICDWWIQ5H-.H•H-.‘0‘-0CDtQ•OHOZO•SM.c-I-H,H-iCD.••IQ-HU)CnC)I-hCDI—hCDU)OU)’OCDHEU)<IM<OH-H•CDU)CO0(1)CDQC)(1).H->tIU)HCDc-tH-c-I-HICD.i0DCl<HM•M•c-I-CD0O•C)P)OCnH-OMO<c>jQCDOH-IOCDHOMMCD-MOSIC)OCDIP)dMWH-M5CDMI’iMOOCDQ.H-(1)0(I-NCDCDH(Q—Ic-I-MCDCflCD(DM(-)H-c-ttQ0Cl0(1)(nc-I-H-Q-CDCDCD<-‘-dHCD5CDCDIIP)H-U)F-U)(-CD(1)’-<(f<H-MCDQ.HCD••5(1)•OS.U)CDCD-c-I-I-’••C)U)HOSIP)-c-t.0•(,)ZOH-U)cQ.MCD0CDPHCD•H-1jCCflOCDCDCDCDCDCD(1)C)CD0ctU)c-I-CDc-I-CDc-I-cn•MMLYMHMU)U)Q.c-I-HI-’P)QU)CDH-‘-DCDMH-H-CDHjc-I-Mi[IJL’J>-)U)c-rMCJ•H-H(1)..P)H•D0•CD(1)C)MH<<..HP)<CD.H-M0<M‘iiO<H-tQWtQc’“ICD0(1•P)CDc-I-CDU)L’ICnPCDc•I-SI0M(1)I-hCDP)‘-0H-OU)flHc1-Mc1-c-tH-MU).OHCD)M5••j.CD.P)F—U)c-I-c-I-O0P)P)H-CncI-U)c-tU)Cn•N00.CDHU)C)MCD0•OCDMC1SIU)c-I-Jc-I-CDc-I-H-(&<(X)H-H-(1)U).ct.wCDt-liMH-H-Mc-I-O(D0H-OO5oSIc-rH-0H-H-0H-U)OOH-tSU)HCD.CD’Ic-I-Oc-I-CDP)CnCDOc-I-OCl)><..U)c-I-U)P)HOCDU)0(Qc-t00H<HU)0H-•CDP)MiMOWCDMOO0(1)SIc-IP)-00(1)tCObc-I-F—hht-hCl)<—<(1)OH-c-I-c-I-H-.Mc-I•CD5U)M-,t-J.•c-t••CDCDMH-.0C)MOHCDOH-H-c-I-CDCD5•00P)’-dU)Oc-I-H-HOM-0o->CThSI(1)5(1)SL1I‘CI0c-I-H-SIOCn•oi-i‘-<‘-<MiSI•P)LIIH-(D<CDM-U)HCDM••CDCDSIP)COCD(1)frh(1)0(1)1’]HH-SICDU)0.c-toU)(-RQ>-O-MHC1)‘tjUiH-OU)‘<-I-hc-I-MO.0lCDM•INC)-H-H-çI-hOiOM(11(1)Hc-t-3U)CD(1)(1)HMCDVoCDc-rCDP)0HHP)-‘-Q-c-I-H-3.CJic-I-CDI-hCOCDQ.HQC).IOMCDC)c-tM(1)H-..-H-5c-rU)U)P)CD‘CDM0P)OF3•H-HP)Hc-tC)Cl)0.IF:-1H-c-IU).(1)W5DI-hOMiCD0i>CIH-(1)OH-iH-•c-I-‘HU)H-U)CDH0P)HMiC)OU)HiCl)ICDCD0CDc-tc-I-U)P)H•5Qc-I-(1).H-•c-I-M’c-I-(1)I-<c-IM(•H-H-CDH-H•MH-HC)H-CDMH-IHU)ICl)LIICOH-HMCDSIH-P)Cl)C),QOCnKnSIc-nd)H’t.0U)H-iU)CQc-Ic-I-C1)M.U).pJ.c-I-hN.•5U)0c-t<H--CDMIN<O•HCDH-Cl)1<LII05LiCDU)c-i-05ISC)OMI-CDH(1)H-0‘<lCD0HiP)c-I-Cl)c-I-c-IH5H-.CDCD173Wolfe, J.A. 1981. Paleoclimatic significance of theOligocene and Neogene floras of the northwestern UnitedStates. Pp. 79-101, in Paleobotanv, paleoecolocw, andevolution, ed. K.J. Niklas. New York: Praeger.Wolfe, J.A. 1985. Distribution of major vegetational typesduring the Tertiary. Pp. 357-375, in The carbon cycleand atmospheric C02: natural variations Archean toPresent, eds. E.T.Sundquist and W.S. Broecker. AmericanGeophysical Union Geophysical Monograph 32.Wolfe, J.A. 1987a. An overview of the origins of the modernvegetation and flora of the northern Rocky Mountains.Ann. Missouri Bot. Gard. 74: 785-803.Wolfe, J.A. 1987b. Late Cretaceous-Cenozoic history ofdeciduousness and the terminal Cretaceous event.Paleobiology 13: 215-226.Wolfe, J.A. and E.B. Leopold. 1967. Neogene and earlyQuaternary vegetation of northwestern North America andnortheastern Asia. Pp. 193-206, in The Bering LandBridae, ed. D.M. Hopkins. Stanford: Stanford UniversityPress.Wood, C.E. 1972. Morphology and phytogeography: theclassical approach to the study of dijunctions. Ann.Missouri Bot. Gard. 59: 107-124.Wright, S. 1951. The genetical structure of populations.Ann. Eugenics 15: 323-354.Zar, J.H. 1984. Biostatistical analysis, second edition.Englewood Cliffs NJ: Prentice-Hall, Inc.Zhengyi, Wu. 1983. On the significance of Pacificintercontinental discontinuity. Ann. Missouri Bot. Gard.70: 577—590.174Appendix 1Summary of Voucher Specimens ExaminedTable A1.1. Summary of Lending Institutions from whichMenziesia specimens were obtained.North American HerbariaALA University of Alaska, Fairbanks AKALTA University of Alberta, Edmonton ABCAS California Academy of Sciences, San Francisco CADS Dudley Herbarium of Stanford University (at CAS)GA University of Georgia, Athens GAGH Gray Herbarium, Harvard University, Cambridge MAHSC Humboldt State University, Arcata CAID University of Idaho, Moscow IDIDS Idaho State University, Pocatello IDJEPS/UC University of California/Jepson Herbaria, Berkeley CAMONT Montana State University, Bozeman MTMONTU University of Montana, Missoula MTNY New York Botanical Garden, Bronx, New York NYORE University of Oregon, Eugene OROSC Oregon State University, Corvallis ORPAC Pennsylvania State University, University Park PAPENN/PH Academy of Natural Sciences of Philadelphia PAR1I Rocky Mountain Herbarium, U. of Wyoming, Laramie WYTENN University of Tennessee, Knoxville TNUAC University of Calgary, Calgary ABUBC University of British Columbia, Vancouver BCUNCC University of North Carolina, Charlotte NCUS United States National Herbarium, SmithsonianInstitution, Washington, DCUSFS United States Forest Service Herbarium (at RN)UVIC University of Victoria, Victoria BCVPI Virginia Polytechnic Institute, Blacksburg VAWS Washington State University, Pullman WAWTU University of Washington, Seattle WAWVA West Virginia University, Morgantown WVJapanese HerbariaKYO Kyoto University, Kyoto, JapanSHIN Shinshu University, Matsumoto, JapanTI University of Tokyo, Tokyo, Japan175Table A1.2. Summary of voucher specimens examined in thestudy of North American Menziesia. Field sites are highlighted in bold. Specimens marked with an asterisk were notused in morphornetric analyses, but are representative ofpopulations analyzed electrophoretically.Vouchers of Menziesia ferruginea, sens. lat.AlaskaAKO1 Talkeetna Quad, Curry. A. Nelson & R.A. Nelson 4152(RN 185258)AKO2 Kenai Peninsula, Palmer Creek Road, south of Hope.J.H. Langenheim 4256 (WTU 174238)AKO3 Ketchikan, Yes Bay. M.W. Gorman 72 (ORE 67742).AKO4 Sitka, above North Sandy Cove, alt. 610 m. D.B. Butts153 (DS 491616)AKO5 Talkeetna. J.P. Anderson 7614 (PH 809744).AKO6 Juneau Quad, Mt. Roberts. A. Nelson & R.A. Nelson 4407(ALA 1490)AKO7 Wrangell Quad, Kuiu Island, Washington Bay.W.J. Eyerdam 5383 (WTU 291382).AKO8 Talkeetna Quad, Little Susitna River, alt. 185 m.J. Kelly s.n. (ID 42320).AKO9 Cordova Quad, Chugach Mtns. G.M. Frohne 49-427(RN 219258)AK1O Ketchikan, Hyder. K. Whited 1234 (CAS 130669).AlbertaABO1 Jasper National Park, Sunwapta Falls. E.H. Moss 9476(ALTA 13943)LL Banff National Park, Lake Louise, alt. 1800 m.T.C. Wells 1151, 1169, 1177, 1184 (UBC)WAT Waterton Lakes National Park, near Akimina Pass, alt.1680 m. T.C. Wells 1190, 1219 (UBC)Table A1.2 continued next page176Table Al.2 continued. Summary of North American Menziesiavoucher specimens.British ColumbiaBCO1 Vancouver Island, near Coombs. T.R. Ashlee s.n.(UVIC 22926)BCO2 Blunden Harbour, Indian village site (abandoned).W.B. Schofield 85739 (UBC Vl9l374)BCO3 Forward Harbour, Douglas Bay. W.B. Schofield 85561(UBC V19l370)BCO4 Queen Charlotte Islands, between Sandspit and CopperBay. J.A. Calder, D.B.O. Savile & R.L. Taylor 23201(UBC 124575)BCO5 Haines Road, mile 47. A.F. Szczawinski s.n.(UBC 109904)BCO6 10 miles SSW of Hells Bells Cr., between Terrace and NewHazelton. J.A. Calder, D.B.O. Savile & J.M. Ferguson14825 (WS 234762)BCO7 N bank of Iskut River, 15 km NW of junction with ForrestKerr River, alt. 320-415 m. W. Gorman 1469(UBC V1769l4)SEY (BCO8) Mt. Seymour Provincial Park, alt. 1130 m.F. Szy s.n. (UBC 83481).SEY* Mt. Seymour Provincial Park, trail to Goldie Lake andFlower Lake, alt. 960 m. T.C. Wells & M.E. Hiebert 1762(UBC).BCO9 Cheam Lake, alt 20 m. V.J. Krajina 602 (UBC 88069).BC1O Stein River Valley near Lytton, alt. 245-365 m.V.J. Krajina 1662 (UBC 75935).BWF Brandywine Falls Provincial Park, Daisy Lake trail, alt.785 m. T.C. Wells 1725-1729, 1733, 1735, 1739-1741,1744—1745, 1747—1752, 1754 (UBC)STL Alice Lake Provincial Park, Stump Lake trail, alt 210 m.T.C. Wells 707, 711, 717, 726, 729, 733, 738, 740—741(UBC).Table A1.2 continued next page177Table A1.2 continued. Summary of North American Menziesiavoucher specimens.PL5 Cypress Provincial Park, trails near Parking Lot 5, alt.760 m. T.C. Wells 662, 1010, 1017 (UBC)YEW* Cypress Provincial Park, trail to Yew Lake from CypressBowl, alt. 970 m. T.C. Wells 1042 (UBC)ROT Manning Provincial Park, Rein Orchid Trail, alt 1200 m.T.C. Wells 745, 750, 754, 759—760, 762, 766 (UBC)SKY Manning Provincial Park, Skyline Trail from StrawberryFlats, alt. 1450-1600 m. T.C. Wells 799, 808 (UBC)SPA Wells Gray Provincial Park, Spahats Creek, alt. 570 m.T.C. Wells 810, 812, 814 (UBC)TRO Wells Gray Recreation Area, Trophy Mountain Access Road,km 6.7, alt 1350 m. T.C. Wells 821 (UBC)MUR Wells Gray Provincial Park, Murtle Lake Road, 13 km W ofBlue River, alt. 1000 m. T.C. Wells 873 (UBC).GNP Glacier National Park, Mountain Creek campsite, alt.810 m. T.C. Wells 1121—1123, 1126, 1129, 1132, 1134,1137, 1140, 1143, 1145 (UBC)YNP* Yoho National Park, roadside turnout past bridge, E overthe Kicking Horse River on the way to Takkakaw Falls,alt. 1290 m. T.C. Wells 889 (UBC).CaliforniaCAO1 Humboldt Co., Arcata, Community Forest main access road,170 m. L.P. Janesway 237 (HSC 77388).CAO2 Del Norte Co., Wilson Creek. C.B. Wolf 825 (DS 159386).HUM Humboldt Co., Humboldt State Redwood Forest, PrairieCreek Trail, alt 125 m. T.C. Wells 628, 637, 639 (UBC)IdahoIDOl Lernhi Co., Gibbonsville Pass. A. Cronquist 6810(ORE 67693)1D02 Shoshone Co., road from Bearskull to Roundtop.C.B. Wilson 506 (IDS)Table A1.2 continued next page178Table A1.2 continued. Summary of North American Menziesiavoucher specimens.1D03 Idaho Co., SE of Harpster. R.F. Daubeninire 47143(WS 176978)1D04 Idaho Co., Seven Devils Mountains, Dry Diggings Camp.A. Nelson & R.A. Nelson 2971 (RN 178443)FRE (1D05) Shoshone Co., along road 2 miles W of FreezeoutSaddle. P. Sullivan 22 (ID 087766)FRE* Shoshone Co., Crater Peak, along road just east ofFreezeout Saddle, alt. 1750 m. S. Brunsfeld s.n. (WS).MOS Latah Co., Moscow Mountain, Randle Flats Road, 11 kmfrom jct. with ID Hwy. 8, alt. 1520 m. T.C. Wells 1715-1722 (UBC)MontanaMTO1 Powell Co., Ovando. J.C. Kirkwood 1239a (UC 352086).MTO2 Missoula Co., Pattee Canyon, alt. 1550 m. M. Keller 374(UBC 167142)MTO3 Park Co., Yellowstone Valley. P.H. Hawkins s.n.(MONT 35513)MTO4 Lake Co., south fork Lost Creek, Swan Valley, alt.1080 m. J.A. Antos 215 (MONTU 80026)MTO5 Glacier Co., Summit, Glacier National Park.J.W. Blankinship s.n. (MONT 4229).MTO6 Flathead Co., Swan Range, Little Creek of AdditionCreek, alt. 1980 m. E. Chadwick s.n. (MONTU 71159).MTO7 Ravalli Co., Big Creek Lake, alt. 1788 m. B. Ranz 70(MONTU 81714)MTO8 Sanders Co., Thompson Falls. Booth s.n. (MONT 40495).MTO9 Lincoln Co., Leigh Lake, Cabinet Mtns. Wilderness Area,alt. 1676 m. D.W. Woodland 847 (MONTU 60494).MT1O Flathead Co., Bigfork, 915 m. M.E. Jones 8796(DS 156665)Table A1.2 continued next page179Table A1.2 continued. Summary of North American Menziesiavoucher specimens.MT11 Ravalli Co., Lost Horse Pass, alt. 1800 m. J.B. Leiberg2981 (ORE 67692)MT12 Powell Co., Monture Canyon near Ovando, alt. 1525 m.J.E. Kirkwood 2162 (MONTU 16378).ALV Missoula Co., Alva Lake campground, Lob NationalForest, alt. 1350 m. T.C. Wells 1228—1230, 1232, 1235,1237—1238, 1240, 1247—1248, 1252 (UBC)OregonGVC (ORO1) Clackamas Co., Government Camp, jct. with Hwy.26, alt. 1185 m. W.L. Stern & K.L. Chambers 33(OSC 109785)0R02 Hood River Co., Lost Lake. L.F. Henderson s.n.(ORE 67719)0R03 Hood River Co., Lost Lake. J.W. Thompson 11191(WTU 12463)OSW Tilamook Co., Oswald West State Park, trail to CapeFalcon, alt. 125 m. T.C. Wells 1009 (UBC)BEV Lincoln Co., Beverly Beach State Park, alt 40 m.T.C. Wells 623-624 (UBC)PER Lane Co., Cape Perpetua, Giant Spruce Trail, alt. 225 m.T.C. Wells 926, 939 (UBC).CC Lane Co., Cape Creek, alt. 100 m. T.C. Wells 948 (UBC).WashingtonWAO1 Wahkiakum Co., Cathlamet. A.S. Foster s.n. (WS 65591).WAO2 Ferry Co., Twin Lakes, alt 1220 m. H. St. John 8920(WS 138301)WAO3 Thurston Co., Black Hills, alt. 730 m. F.G. Meyer 1637(WS 123508)WAO4 Skamania Co., Chiquash Mountains. W.N. Suksdorf s.n.(WS 137812)Table Al.2 continued next page180Table A1.2 continued. Summary of North American Menziesiavoucher specimens.WAO5 Whatcom Co., Wiser Lake. W.N. Suksdorf s.n.(WS 138301)STV Chelan Co., Stevens Pass, Sno-Pac Road near Hwy 2, alt.1235 m. T.C. Wells 1677—1678, 1680—1686 (UBC)WyomingWYO1 Teton Co., N slope of Signal Mtn. near Jackson Hole,alt. 2100 m. J.F. & M.S. Reed 2743 (RN 214762)WYO2 Teton Co., Paint Brush Canyon, SW of Leigh Lake, alt.2285-2380 m. F,W. Pennell & R.L. Schaeffer 24294(PH 853589)TET (WYO3) Teton Co., Jenny Lake, alt. 1980 m. R. Lichvar838 (RN 316235)TET* Teton Co., along eastern shore of String Lake, alt.2100 m. T.C. Wells 1265 (UBC)Vouchers of Menziesia DilosaGeorgiaBB Union Co., Brasstown Bald, trail from SE corner ofparking lot, alt. 1375 m. T.C. Wells 1557-1558, 1561,1566, 1568 (UBC)LWS Union Co., Lake Winfield Scott, Slaughter Gap Trail,1.6-2.5 km from camp trailhead, alt. 950 m. T.C. Wells1538, 1541, 1544, 1550 (UBC)MarylandMDO1 Allegany Co., Dans Rock, 2 km ESE of Midland.R.M. Downs 8770 (UNCC 348227).MDO2 Garrett Co., Jennings. W. Stone 3704 (GH).MDO3 Garrett Co., Sampson Rock, alt. 885 m. S.R. Hill 12402(GH)WBR Garrett Co., Walnut Bottom Road, 200 m E of intersectionwith Hwy. 135, alt 750 m. T.C. Wells 1273-1276 (UBC)Table Al.2 continued next page181Table A1.2 continued. Summary of North American Menziesiavoucher specimens.North CarolinaNCO1 Mitchell Co., E of Carver’s Gap, Roan Mtn. E.W. Wood &D.E. Boufford 2203 (UNCC 467046).NCO2 Macon Co., Chesnut Ridge, E of Highlands. R.K. Godfrey,J.E. O’Connell & H. Wright 51760 (GA 57531).NCO3 Ashe Co., 1.5 mi. E of Brownwood. A.E. Radford 34310(GA 64957)NCO4 Watauga Co., 3 mi. E of Blowing Rock, N of US Rte. 221.W.B. Fox & R.K. Godfrey 3396 (NY).NCO5 Watauga Co., 5.3 mi. NNW of Laxon. A.E. Ahies &J.A. Duke 47677 (UNCC 167913).MIT Yancey Co., Mt. Mitchell, Deep Gap Trail to Mt. Craig,alt. 1950 m. T.C. Wells 1440, 1453, 1458 (UBC)PIS Haywood Co., Mt. Pisgah summit trail, alt. 1565 m.T.C. Wells 1470, 1478, 1482, 1502 (UBC)WSM Macon Co., Whitesides Mountain, summit trail, alt.1365 m. T.C. Wells 1508, 1514, 1519, 1530 (UBC)Pen.nsylvaniaPAO1 Lebanon Co., Cold Spring. E. Diffenbauch s.n.(PH 002995)PAO2 Schuylkill Co., 1.5 mi. S of Valley View, along RauschCr. below Bear Mtn. P.R. Wagner 1677 (PENN).PAO3 Dauphin Co., 2 mi. NE of Wiconisco, alt. 370 m.D. Berkheimer 14764 (PAC 59123).PAO4 Dauphin Co., 2.5 mi. ESE of Millersburg, alt. 130 m.D. Berkheimer 13683 (PENN).PAO5 Dauphin Co., along Rattling Cr., 0.5 mi. S of Lykens.H. Wilkens 8262 (PH 002997)PAO6 Somerset Co., 1.25 mi. W of Pleasant Union on 160 nearLaurel Run. E.T. Wherry s.n. (PENN).PAO7 Bedford Co., 0.4 mi. ESE of Martin Hill Fire Tower,alt. 760 m. D. Berkheimer 6399 (PENN)Table Al.2 continued next page182Table A1.2 continued. Summary of North American Menziesiavoucher specimens.TennesseeTNO1 Carter Co., Roan Mountain, alt. 1920 m. H.M. Jennisons.n. (TENN).TNO2 Sevier Co., between Porter’s Flat and the Sawteeth.T.S. Patrick & E. Rothberger 4445 (GA 163926).LEC Sevier Co., above Alum Cave Bluffs to Mt. LeContesunirnit, alt. 2020 m. T.C. Wells 1572-1573, 1580 (UBC).VirginiaVAO1 Rockbridge Co., near Marble Springs. K. Castro 814(NY).VAO2 Bedford Co., Sharptop, Peaks of Otter. W.W. Eggleston18633 (US 1533407)VAO3 Floyd-Patrick Co. line, Rocky Knob. F.R. Fosberg 33565(VPI 59255)VAO4 Floyd-Patrick Co. line, Rocky Knob. C.K. Dale s.n(VPI 29065)VAO5 Giles Co., Angels Rest Mtn., SW of Pearisburg, alt.1065 m. J.M. Fogg 11511 (PENN).VAO6 Roanoke Co., Little Brushy Mtn., ca. 2 mi. WNW of SalemP.O. C.E. Wood 2405 (PENN).VAO7 Page Co., Stony Man Mtn., SE of Luray. E.T. Wherry &F.W. Pennell 13336 (PH 728222)JEN Warren Co., Jenkins Gap, Shenandoah Parkway, alt. 760 m.T.C. Wells 1312, 1316, 1325, 1327, 1332 (UBC)RKN Floyd-Patrick Co. line, Blueridge Parkway, 10 km S ofRocky Knob, roadside turnout, alt. 880 m. T.C. Wells1402 (UBC)MTIJ Giles Co., Road 613, 7 km from Mountain Lake, alt.1060 m. T.C. Wells 1601—1602, 1607, 1611, 1623 (UBC)WTM Grayson Co., White Top Mtn., summit area along theAppalachian Trail, alt. 1535 m. T.C. Wells 1406, 1414,1419, 1422, 1431, 1434—1436 (UBC)Table A1.2 continued next page183Table A1.2 continued. Summary of North American Menziesiavoucher specimens.West VirginaWVO1 Grant Co., Greenland Gap. H.A. Davis 8606 (WTU 136863).WVO2 Pocahontas Co., Cranberry Glades. J.L. Sheldon 3839(WVA).WVO3 Pocahontas Co., W of Durbin on the Staunton Pike.A. Rehder s.n. (GH).DOL Grant Co., Dolly Sods, alt 1125 m. T.C. Wells 1280,1285, 1289, 1290, 1292, 1309 (UBC)CAP Hampshire Co., Capon Springs, alt. 520 m. T.C. Wells1334, 1341, 1358, 1360 (USC)MIN Pocahontas Co., forestry road near Minnehaha Springs,alt. 820 m. T.C. Wells 1374, 1384, 1390, 1399, 1401(UBC).184Table A1.3. Summary of Japanese Menziesia voucher specimens.i. ciliicalvxCIOl Honshu, Shiga Prefecture, Otu-shi, Mt. Hiei, NE ofKyoto. T. Shimizu 7153 (SHIN).C102 Honshu, Kyoto Prefecture, Kyoto-shi, en route fromsummit of Mt. Hiei-zan to Seiryu—ji, Sakyo-ku, alt.500 m. N. Fukuoka 7691 (SHIN).CIO3 Honshu, Gifu Prefecture, Shirotori-cho, Gujo-gun,vicinity of Nagataki, alt. 600 m. H. Takahashi,H. Takano & E. Nishio 344 (SHIN).C104 Honshu, Nagano Prefecture, Okuwa-mura, Kiso-gun, betweenNoziri and Mt. Aterayama, alt 500-800m. G. Murata &H. Nishimura 129 (KYO 324-07).C105 Honshu, Nagano Prefecture, Tatsuno-cho, Kamiina-gun,Otaki-sawa, alt. 1700 m. T. Umaba 103055 (SHIN 70352).C106 Honshu, Nagano Prefecture, Hase-mura, Kamiina-gun,Kokuken-tani, alt 1470-1490 m. T. Umaba 100303 (SHIN).M. multifloraMUO1 Honshu, Nagano Prefecture, Sakae-mura, Shimominochi-gun,around Nonomi Pond, alt. 1100 m. T. Shimizu 18963(SHIN)MUO2 Honshu, Nagano Prefecture, Sakae-mura, Shimominochi-gun,Hirataki to Nonomi Pond, 800-1100 m. T. Shimizu 18910(SHIN)MtJO3 Honshu, Akita Prefecture, Kisagata-cho, Yuri-gun,Shishigahama, N slope of Mt. Chokai-san, alt 400-500 m.N. Fukuoka 12320 (KYO 324-21).MUO4 Honshu, Yamagata Prefecture, Kaminoyama-shi, Mt.Zizo-yama in the Zaoo Mtns, alt. 1400-1700 m. G. Murata& H. Koyama 41229 (KYO 324-20).MtJQ5 Honshu, Aomori Prefecture, Aomori-shi, Mt. Hakkoda, nearSugayu. S. Okamnoto s.n. (KYO 324-26).MrJO6 Honshu, Gifu Prefecture, Itadori-mura, Mugi-gun,vicinity of Kakode, alt. 400 m. N. Fukuoka &K. Yamashita 256 (SHIN).Table A1.3 continued next page185Table A1.3 continued. Summary of Japanese Menziesia voucherspecimens.MUO7 Honshu, Nagano Prefecture, Matsumoto-shi, Chausu-yama,alt. 1900-2000 m. T. Shimizu 279 (SHIN)MUO8 Honshu, Nagano Prefecture, Yamanouti-machi, Shimo-takaigun, Mt. Iwasuge to Mt. Terakoya in the Shiga Heights,alt. 1700—2000m. T. Shimizu 17703 (SHIN)MTJO9 Honshu, Nagano Prefecture, Yamanouti-machi, Shimo-takaigun, Mt. Iwasuge to Mt. Terakoya in the Shiga Heights,alt. 1700-2000m. T. Shimizu 17699 (SHIN)MUlO Honshu, Nagano Prefecture, Fujimi-cho, Suwa-gun,Shirozare, between Mt. Kamanashi to Mt. Shiraiwa, alt.2100 m. T. Shimizu 26573 (SHIN)Mull Honshu, Nagano Prefecture, Kawakami-mura, Minamisakugun, near Jumoji Pass, alt. 1800 m. T. Shimizu 13802(SHIN)MU12 Honshu, Niigata Prefecture, Yuno-tani-mura, Kitauonumagun, Mt. Ogura-yama, Shiori-toga to Komanoyu, alt.1000 m. S. Kitamura & G. Murata 2866 (KYO 324-15).MtJ13 Honshu, Ishikawa Prefecture, Siramine-mura, Ishikawagun, Mt. Sunagozen-yama, Haku-san Mtns, alt. 800-1200 m. N. Kurosaki 12882 (KYO 324-14). oentandraPEO1 Honshu, Nagano Prefecture, Oomachi-shi, on the ascent ofMt. Gaki, SW of Qomachi, alt. 1180-1400 m. T. Shimizu18113 (SHIN)PEO2 Honshu, Nagano Prefecture, Hase-mura, Kamiina-gun, Mt.Shiraiwa, alt. 1500-1990 m. T. Shimizu 22972 (SHIN)PEO3 Honshu, Nagano Prefecture, Kiyo-sato-mura, Minamisakugun, Nembagahara, alt. 1300-2000 m. T. Shimizu 7765(SHIN)PEO4 Hokkaido, Nemuro-sityo, Nemuro-shi, Ottyshimisaki(Ochiishi-saki) . F. Yamazaki 2560 (TI H87—312)PEO5 Honshu, Shizuoka Prefecture, Shizuoka-shi, Mt. Akaishi.K. Asano 18677 (TI H87-3(4)).Table Al.3 continued next page186Table A1.3 continued. Summary of Japanese Menziesia voucherspecimens.PEO6 Hokkaido, Kamikawa-sityo, Tokiwa-mura, Nakagawa-gun,vicinity of Otoineppu, Nakagawa Enshurin, alt. 100 m.G. Murata, H. Koyama & T. Yahara 38440 (KYO 324-38).PEO7 Hokkaido, Rumoi-sityo, Horonobe-cho, Teshio-gun,Nupuromapporo, Hokkaido-daigaku, Teshio Enshurin, alt.200 m. G. Murata, H. Koyama & T. Yahara 38258(KYQ 324-39)PEO8 Honshu, Tochigi Prefecture, Nasu-machi, Nasu-gun,between Santogoya-onsen and Mt. Asahi, alt. 1500 m.G. Murata 18244 (KYO 324-34)PEO9 Honshu, Nagano Prefecture, Kami-mura, Shimoina-gun,Kitamata-sawa, alt. 1100-1200 m. K. Oso, K. Takumi &I. Fujii 126134 (SHIN)PE1O Honshu, Tochigi Prefecture, Nikko-shi, Nikko. T. Makino101431 (KYO 324-32). ouroureaPUO1 Kyushu, Miyazaki Prefecture, Takatiho-cho, Nisiusukigun, between Gokasyo and the summit of Mt. Sobo, alt.1600 m. S. Hatusima, S. Sako & Kawnabe 22690(KYO 324—42)Appendix2SummaryofEcologicalParametersfortheNorthAmericanMenziesiaFieldSitesTableA2.1.SpeciesassociationmatrixforwesternNorthAmericanMenziesiasites.1=presence;0=absence.NomenclaturefollowsHitchcockandCronquist(1973).TaxonTreesPopulat0a1234567891011121314151617181920AbiesamabalisAbieslasiocarpaAcercircinatumAlnusrubraLarixoccidentalisPiceaenqelmaniiPiceasitchensisPinuscontortaPooulustrichocarpaPooulustremuloidesPseudotsuaamenziesiiSequoiasemoervirensSorbussitchensisTaxusbrevifoliaThuipolicataTsuaaheteroohvlla00000000011000000011000000001000010001100010000000000000000000110110010100000000000010001000000001000000000000011010000000000001000000000000010000000000001110000001011111100000000000001000000100011111011000000000000000010000000110010111011000000000000001011000000000000000011100000000000000000000100000001100000101100100101000000000000000Tsuaamertensiana00apopulat ionCodes:1)BWF;2)STL;3)PL5;4)05W;5)BEV;6)PER;7)HUM;8)ROT;9)SKY;10)STV;11)GVC;12)SPA;13)TRO;14)MUR;15)GNP;16)LL;17)WAT;18)MOS;19)ALV;20)TETHTableA2.1continuednextpageTableA2.1continued.SpeciesassociationmatrixforwesternNorthAmericanMenziesiaTaxonShrubsGautheriashallonHolodiscusdiscolorLedumc1andu1osumLedumaroenlandicumLonicerainvolucrataRhododendronalbiflorumRubusDarviflorusRubusDedatusVacciniummernbranaceumVacciniumovalifoliumVacciniumparvifoliumVacciniumscopariumOtherCladinaspp.OxalisorepanaPolystichumInunitumXerpphyllumtenaxapopulat ionCodes:1)BWF;2)STL;3)PL5;4)OSW;5)BEV;6)PER;7)HUN;8)ROT;9)SKY;10)STy;11)GVC;12)SPA;13)TRO;14)MUR;15)GNP;16)LL;17)WAT;18)MOS;19)ALV;20)TET000000000000000000000000ODsites.1=presence;0=absencePopulationa123456789101112131415161718192001011000000000000000000000000000000100000001000000000000000000000000010000000000000000100100000000100001000000000010000000100100000000000000100000100000010000111001000010000000000011001010010000000000000000000000010110000000000000000000000000100000000000000100001000000000000000000000000000001010TableA2.2.SpeciesassociationmatrixforeasternNorthAmericanMenziesiasites.1=presence;0=absence.NomenclaturefollowsGleason(1952).TaxonPopulationWBRDOLCAPJENMINRKNMTLWTMMITPISWSMLECBBLWSTrees_______oo000000100100_________o0100000010000______11000010011011___________o0000000011000______oo000000000010______oo100110000000_________oo000001000000_________11001000001000___o0000000001000o0000000010000____________o0000000000001________o0000000001000________10011000000001_________o0001000000000_________o0000000011000_____oi000001100100____10001000000000________10000000000000__o0010010000010________10100010000000_____oo000010011000o0111010000001_____oaaaa01a011010_____o0001100000000______10000000000001________10011000001001_______oio00001100100________oo000000000001AbiesfraseriAceroennsvlvanicaAcerrubrumBetulaallegheniensisCarvaalabraCornusfloridaCrataegusounctataHamamelisvirainianahexdeciduahexmontanaLiriodendrontulipiferaMaQnoliafraseriNvssasylvaticaOstrvavirainignaOxvdendronarboreumPicearubensPinusstrobusPrunusserotinaOuercusalbaQuercuscoccineaQuercusmuhlenberaiiQuercusorinusOuercusrubraOuercusvelutinaRobiniaoseudoacaciaSassifrasalbidumSorbusamericanaTsuacanadensisI1TableA2.2continuednextpageTableA2.2continued.SpeciesassociationmatrixforeasternNorthAmericanMenziesiasites.1=presence;0=absence.TaxonPopulationWBRDCLCAPJENMINRKNMTLWThIMITPISWSMLECBBLWSShrubsKalmialatifolip11011110000100Prunusoennsylvanica00000000000100Rhododendroncalendulaceum00000100001001Rhododendroncatawbinense00000100111111Rhododendronmaximum10010100010100Rhododendronrpseum10000000000000Rosaspp.00000001000000Rubusspp.10000001000000Vacciniumcorvrnbosum00000000100000Vacciniumpallidum00011000000000Viburnumalnifolia00000000011011OtherCladinaarbuscula01000000000000Fragariavirginiana00000001000000191Table A2.3. Physical features of North American Menziesia fieldsites. Population abbreviations as in Appendix 1.2.Long. Slope Aspect Exposure(deg.) (% shade)Western North American sitesPop. No.OTUsLat. Elev.(in)BWF 19 50°01’N 123°07’W 785 10—15 W 30—75STL 9 49°46’N 123°07’W 210—230 0—15 N,E,SE,S,W 30—80PL5 3 49°24’N 123°l0’W 760 10—30 W 30—60OSW 1 45°46’N 123°58’W 125 10 Sw 203EV 2 44°43’N 124°04’W 40 10 S 70PER 3 44°16’N 124°06’W 225-240 5 N,S 75HUM 3 41°19’N 124°01’W 125 10—15 W 60—75ROT 7 49°04’N 120°48’W 1200 0—5 N,S 80SKY 2 49°03’N 120°52’W 1230—1400 15—20 N,E 50—705TV 9 47°45’N 121°04’W 1235 15 N 55—75GVC 1 45°16’N 121°45’W 1185 10 S 70SPA 3 51°43’N 120°01’w 570 2.5 w 65TRO 1 51°45’N 119°55’W 1350 5 W 50MUR 1 52°04’N 119°22’W 1000 25 E 70GNP 11 51°27’N 117°28’W 810 1-2 SE 40—75LL 4 51°25N 116°13’W 1800 0—25 E,WNW,NW 25—60WAT 2 49°01’N 114°OPW 1680 10 S 35-60MOS 8 46°48’N 116°54’W 1520 25 E 70ALV 11 47°19N 113°34’W 1350 0 SE 35—80TET 1 43°46’N ll0°43’W 2100 2.5 W 50Appalachian sitesWBR 4 39°28’N 79°12’W 750 5-8 W 85—90DOL 6 38°58’N 79°19’W 1125 0—2 NW 0—30CAP 4 39°07’N 78°30’W 520 15 ESE 50—70JEN 5 38°46N 78°12’W 760 0—10 WNW 60—80MIN 5 38°07’N 79°57’W 820 30 N 75—90RKN 1 36°46’N 80°24’W 880 30 N 75MTL 5 37°24’N 80°33’W 1060 10—25 SE 60—80WTM 8 36°39’N 81°34’W 1535 25 E 0—30MIT 3 35°45’N 82°33’W 1935—2020 5—40 ENE,E,W 20—25PIS 4 35°27’N 82°46’W 1530—1600 20—30 NW 70—85WSM 4 35°06’N 83°09’W 1330—1385 20-45 N,NE 75—95LEC 3 35°34’N 83°22’W 1880—2020 0—40 N,NW 60—80BB 5 34°49’N 83°51’W 1350-1380 20—25 NW 70—95LWS 4 34°42’N 83°58’W 940—970 5—30 ESE,WNW,NW 70—85192Table A2.4. Summary of soils found at the North AmericanMenziesia field sites. Population codes as in Appendix 1.2.Pop. Soil Organic Relatvejpea Colour’ Content2 DepthWestern North American sitesBWF Gi dark high shallowSTL Gi dark medium mediumPL5 G2 grey-brown low-medium mediumOSW 1-14 dark low-medium deep3EV 1-14 dark medium deepPER 1-14 dark high deepHUM U2 dark high shallow-med.ROT G2 dark medium mediumSKY G2 grey-brown medium mediumSW G2 dark medium mediumGVC G2 dark medium mediumSPA G2 dark medium mediumTRO K-G2 light-yellow low shallow-rockyMUR Gi dark high mediumGNP K-G2 grey-brown low medium-rockyLL K-G2 grey-brown low shallow-rockyWAT K-G2 grey-brown low shallow-rockyMOS A grey-brown low mediumALV M light-reddish low mediumTET M light yellow low mediumAppalachian sitesWBR 18 black high shallow-stonyDOL 18 black high shallow-stonyCAP 18 yellow low deepJEN U5-4 brown medium mediumMIN 18 yellow low deepRKN U5-6 brown medium mediumMTL 18 yellow-brown medium mediumU5-6 brown high shallow-stonyMIT U5-6 dark with mica high shallowPIS 1J5-4 dark high mediumWSM U5-6 brown medium medium-deepLEC U5-13 dark high shallow-stonyBB U5-6 brown high mediumLWS U5-6 red-brown medium deep2-Colour: predominant colour of A and or B horizons below thehumic layer2Organic Content: high = accumulated organic litter with asurface of undecomposed duff; medium = accumulated litterwhich has undergone modification; low = predominantly mineralsoilsTable A2.4 continued next page193Table A2.4 continued. Summary of soils found at the NorthAmerican Menziesia field sites.3Relative Depth: shallow = thin, often rocky soils, < 0.3 mto the C horizon; medium = deeper soils up to 0.7 m to the Chorizon; deep = soils at least 1 m to the C horizon.aSOl Type Codes (adapted from Farley 1979; Canada Dept. ofEnergy, Mines and Resources 1974; U.S. Dept. of the InteriorGeological Survey 1970):A grey-brown podzols: well-drained acidic soils with a thinor brownish layer of clay accumulation. Moist, coolenvironments.Gi ferro-humic podzols: well drained, infertile, acidic soilswith Al and Fe accumulated in the subsurface horizons.Surface horizons have accumulated humic material. Develops incool and moist environments.G2 humo-ferric podzols: well drained, acidic soils with Feand Al as the major accumulated components of the subsurfacehorizons. Less accumulated organic material than Gi.Develops in cool and moist environments.18 yellow podzols to grey-brown podzols: varies from shallowlithic (rocky) soils with surface accumulated humic layers todeeper, low to moderately humic, soils. All are leached,acidic, moist soils of cool environments.1-14 podzols: well drained acidic soils with crystallineclay minerals, and thick dark-coloured horizons. Alteredsubsurface horizons have lost mineral materials due toleaching. Occurs in moist, temperate environments.K shallow lithic (rocky) soils: of cold mountainousenvironments with moderate to steep slopes. In our studysites, humo-ferric podzols (G2) predominated.M podzols: well drained acidic soils with subsurface clayaccumulation. Typically nutrient-poor. Occurs in cold tocool, moist environments.U2 brown podzols: well drained, acidic soils with surfacehorizons high in organic matter. Subsurface horizons withappreciable weatherable minerals and a thin layer of clayaccumulation. Occurs in moist, temperate environments.U5 gray-brown to red-yellow podzols: have medium to highsurface organic layers and a subsurface horizon with clayaccumulation or weatherable minerals or both. Some soils (U5-6 dystrochrepts) are relativly infertile; others may lackmineral accumulation (U5-l3 paleudults). All are leached,acidic, moist soils of temperate to cool environments.194Table A2.5. Summary of climatic factors at the North rnericanMenziesia field sites. Population codes as in Table A1.2.Sources: Baldwin 1968; Canada Dept. of Energy Mines andResources 1974; Farley 1979.Pop.Frost-freeDaysAverage TemperaturesJanuary August±2.5°C ±1.0°CAveragePrecipitationcmWestern SitesBWF 100—140 -2.5 15.0 150—250STL 140—180 -2.5 15.0 150—250PL5 140—180 —2.5 17.0 150—250OSW 260—280 4.4 16.0 205—255BEV 260—280 4.4 16.0 205—255PER 260—280 4.4 16.0 205—255HUN 270—300 4.4 16.0 160—205ROT 60—100 —7.5 15.0 150-250SKY 60—100 —7.5 15.0 150—250STV 110—130 -4.0 15.0 80—120GVC 80—100 -3.0 15.0 75—100SPA 60—100 —5.0 17.0 80—120TRO 60-100 -7.5 15.0 75—100MUR 60—100 —12.5 15.0 100—150GNP <60 —12.5 15.0 150—250LL <60 —17.5 12.0 75—100WAT <60 —17.5 12.0 100—150MOS 120—160 —6.7 16.0 40—60ALV <90 -6.7 16.0 50-70TET <90 -12.5 13.0 95—115Appalachian SitesWBR 110—130 —1.0 18.0 120—135DOL 110—130 —2.0 16.0 120—135CAP 140—160 2.0 18.0 90—110JEN 125—145 -1.0 19.0 95—120MIN 140—160 2.0 21.0 90—110RKN 140—160 2.0 21.0 95—120MTL 140—160 2.0 21.0 95-120WTM 110—130 —1.0 16.0 125—145MIT 90-110 -2.0 15.0 125-145PIS 140—160 2.0 21.0 125—145WSM 170—190 4.0 21.0 180—205LEC 110—130 —1.5 15.0 110—130BB 170-190 4.0 21.0 150-165LWS 170—190 4.0 24.0 150-165195Appendix 3Growing Menziesia Seedlings Using Sterile CultureA3.1 Surface Sterilization and Preparation of SeedsTechniques for growing Menziesia from seed using sterileculture were adapted from guidelines employed in coinmercialRhododendron propagation (Macdonald 1986). Since Menziesiaseeds are quite fine, they require careful manipulation toensure proper surface sterilization, to minimize fungal orbacterial contamination of the agar growth media. In earlytrials, a major source of contamination came from theinclusion of small capsule fragments in the sown seeds.Consequently, in subsequent trials, 50-100 seeds per plantwere routinely sorted from capsule fragments using a camel-hair brush and were placed in 20 ml screw-cap vials. Surfacesterilization was achieved by rinsing the seeds in 95% EtOHfor 30 seconds followed by a 1:1 rinse of bleach (5.25%sodium hypochiorite) and water for 6 minutes. The seeds werethen washed three times with sterilized distilled water toremove all traces of bleach or alcohol. All work was doneunder a laminar air-flow hood using a Pipetman suctionapparatus and small aperture disposable pipette tips tochange solutions. After washing, the vials were looselycapped and the seeds were allowed to dry under low heat at30°C for 24 hours.196A3.2 Preparation of Growth MediaSeveral growth media were tried to optimize seedgermination and seedling growth. Amongst these were fullstrength, half strength and quarter strength solutions ofMurashige and Scoog nutrient media (Murashige and Scoog 1962)and Anderson’s Rhododendron medium (Anderson 1984), alongwith a control with no nutrient additives. In some trials, astandard mixture of plant vitamins (Sigma G-2519) was added,while growth regulators such as NAA (a-napthyleneacetic acid0.5 mg/i) were used to optimize the root development of theseedlings. Three different media pH levels (5.6, 5.0 and4.5) were tested. All media were agar based (0.6% DifcoBacto Agar) with 30 g/1 of sucrose added as a carbon source.Higher agar concentrations resulted in harder media thatMenziesia roots had trouble penetrating, while lowerconcentrations were too watery. All media were sterilized inan autoclave at 170°C for 15-20 minutes and then poured into250 ml sterilized glass jars under a laminar flow hood.A3.3 Growth ConditionsAfter the media had cooled and solidified, surfacesterilized seeds were sprinkled into the jars, again under alaminar air-flow hood, after which the jars were sealed andplaced on shelves in plant growth chambers. Two separatetemperature regimes were tested, either 25°C day/night or 19°Cday/l2°C night, with a daylength of 14-16 hours. Differentlight levels were achieved by placing replicate jars on197shelves 0.5m to 2.Om away from a high intensity fluorescentgrow-light source.High levels of germination (60-80%) were achieved usingsterile culture techniques about 10-12 days after sowing onagar, regardless of the nutrient media conditions. Althoughthe highest levels of germination occurred under high light(2000 lux), best growth after about two weeks was achievedunder shaded conditions where the jars were placed furtheraway from the light source. Cooler temperatures (19°C day!11°C night) gave the best results. However, after one or twomonths, plants growing in the M-S media started to die,regardless of the nutrient concentration levels. Thedegeneration of tissue began at the roots which were poorlydeveloped and advanced to the shoot. It appears that theseedlings suffered from excessively high nutrientconcentrations. Lowered concentrations of N and K, found inAndersonas medium, worked much better and were used routinelyafter the trials to grow Menziesia. The lower pH levels of5.0 or 4.5 gave best results. Nevertheless, growth of theseedlings would become arrested after the plants reached aheight of 1-1.5 cm and would remain static for up to 6months. Root development was usually poor, even with NAAtreatment.In an attempt to overcome this stasis, plantlets weretransferred to sterilized soil (1 part loam; 2 parts peat; 1part perlite) with or without the addition of nativeunsterilized soil gathered from field sites. All of the198plants were grown in the cool growth chamber, wateredregularly with water and fertilized every two weeks with aquarter strength solution of 20:20:20 fertilizer. Within twoweeks, many plants showed signs of renewed growth, especiallyin soil which had been inoculated with native soil. Thissuggested that Menziesia seedlings grew best in the presenceof a mycorrhizal fungus. The importance of mycorrhizae tothe growth of ericaceous seedlings has been documentedpreviously, especially in blueberry (Vaccinium corvmbosum L.)(Powell 1982). Interestingly, Godo Stoyke at the Universityof Alberta has shown that an endomycorrhizal fungus(Phialoceohala fortinii Wang et Wilcox) is associated withmany subalpine ericads, and is capable of infecting the rootsof Menziesia ferruginea (Stoyke and Currah 1991; personalcommunication). However, in culture the fungus often becomesparasitic resulting in seedling death.A3.4 DiscussionDespite all efforts, seedling mortality was generallyhigh (up to 90%) and survivors showed very slow growth,attaining a height of 5-7 cm in the first year. Seedlingswhich did grow were used in electrophoretic work. However,the number of usable seedlings per population was too smallfor routine populational screening of isozymes. Nevertheless,seedlings provided a useful way of comparing the enzymephenotypes of progeny with those of the maternal plants fromwhich the seeds were collected.199It appears that Menziesia seeds germinate and grow bestunder low to moderate temperatures (19°C day/ll°C night),shaded conditions, and in low nutrient acidic media or soils.Sustained growth may rely on the presence of mycorrhizalfungi. However, even under “optimal conditions”, growth isslow and a high degree of mortality is seen. In fact, fieldobservations reveal that seedling recruitment is low in manypopulations. Some plants 10-15 cm high are three or moreyears old based on counting growth scars. Considering that atypical Menziesia plant at maturity can produce thousands ofseeds per year, few offspring appear to survive. This isconsistent with a slow-growing conservative K-selection plantwhich grows in stabilized conditions. In tact, Menziesia inNorth America at least, does not appear to colonize recentlydisturbed sites. In the Clearwater Valley, B.C., Menziesiais absent from areas that have experienced extensive forestfires (T. Goward, personal communication). Trevor Goward isa noted naturalist in the Wells Gray-Clearwater Valley regionand co-author of Nature Wells Gray (Goward and Hickson 1989).

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0098874/manifest

Comment

Related Items