UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Flavonoids and taxonomy of the Limnanthaceae Parker, William Harrison 1975

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

Item Metadata

Download

Media
831-UBC_1975_A1 P37_7.pdf [ 11.46MB ]
Metadata
JSON: 831-1.0100112.json
JSON-LD: 831-1.0100112-ld.json
RDF/XML (Pretty): 831-1.0100112-rdf.xml
RDF/JSON: 831-1.0100112-rdf.json
Turtle: 831-1.0100112-turtle.txt
N-Triples: 831-1.0100112-rdf-ntriples.txt
Original Record: 831-1.0100112-source.json
Full Text
831-1.0100112-fulltext.txt
Citation
831-1.0100112.ris

Full Text

FLAVONOIDS AND TAXONOMY OF THE LIMNANTHACEAE by WILLIAM HARRISON PARKER " B.A. , Reed College, 1968 M.Sc, Univ. B r i t i s h Columbia, 1972 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of Botany We accept t h i s t hesis as conforming to the required standard. THE UNIVERSITY OF BRITISH COLUMBIA August, 1975 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shal make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my Department or by his representatives. It is understood that copying or publication of this thesis for financial gain shal not be alowed without my written permission. Depa rtment The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 ABSTRACT The Limnanthaceae i s a small family of North American annual herbs with uncertain relationships. I t i s generally recognized to consist of two genera, Limnanthes and Floerkea, together containing 10 species with 13 v a r i -eties and subspecies. To help c l a r i f y relationships within t h i s family, a com-parative flavonoid investigation was undertaken. A l l taxa were compared on the basis of flavonoids occurring i n a l l tissues, and most Limnanthes taxa were additionally compared on the basis of flavonoids occurring i n the petals and the u l t r a v i o l e t reflectance characteristics of t h e i r flowers. A t o t a l of hd flavonol glycosides were encountered. Of these, 35 were i d e n t i f i e d as deriva-tives of s i x flavonol aglycone types: syringetin, isorhamnetin, kaempferol, l a r i c y t r i n (myricetin 3'-methyl ether), quereetin and myricetin, a l l gly-cosylated with combinations of glucose and rhamnose. The flavonoid data were analyzed by three numerical taxonomic techniques: clustering by the weighted pair group method, P r i n c i p a l Components Analysis, and Varimax Factor Analysis with rotation. Duplicate comparisons were made according to type of tissue analyzed and concentrations of flavonoids considered. C l a s s i f i c a t i o n s based on petal flavonoids occurring i n r e l a t i v e l y higher concentrations were found to r e f l e c t most c l e a r l y natural relationships i n Limnanthes. The u l t r a v i o l e t reflectance characteristics of flowers were found to distinguish one supra-s p e c i f i c group of Limnanthes taxa from the remainder of the genus. The adap-t i v e significance of u l t r a v i o l e t patterning i s discussed, together with i t s implications concerning flavonoid compositions. The method and v a l i d i t y of using flavonoids as taxonomic characters i s also discussed. Flavonoid and u l t r a v i o l e t reflectance characteristics are integrated with a l l other taxonomi-c a l l y s i g n i f i c a n t information known for the Limnanthaceae, and proposals for taxonomic revision are made. A synopsis of the family i s presented which re-cognizes one genus, Floerkea, containing 15 species with 5 .varieties._ - i i i -TABLE OF CONTENTS Abstract i i L i s t of Tables v i L i s t of Figures v i i L i s t of Appendices i x Acknowledgement x i i i Introduction Family Description 1 Taxonomic Background A. Below the Family Level 3 B. Family A f f i n i t i e s 7 Previous Chemical Investigations 10 Agronomic Evaluation of Limnanthes 11 Thesis Objectives 12 Materials and Methods The Plants and Their Sources lh . Plant Growth 18 Plant Tissue S e l e c t i o n 20 F l o r a l Patterning i n U l t r a V i o l e t Light A. Introduction 22 B. Method 2h Flavonoid I s o l a t i o n and I d e n t i f i c a t i o n c A. Introduction 25 B. Plant Extractions 27 C. I d e n t i f i c a t i o n of the Flayonoids of L. douglasii. yar. d o u g l a s i i , OTU 1 i . Chromatography 28 i i . S p ectral Analyses 30 - i v -TABLE OF CONTENTS — Continued i i i . Hydrolyses 31 i v . Acetate Preparation 32 D. I d e n t i f i c a t i o n of the Remaining Limnanthes Flavonoids • 33 Expression of Flavonoids as Taxonomic Characters 33 Treatment of the Flavonoid Data A. Introduction 3h B. Method ' 35 Results F l o r a l Patterning i n U l t r a V i o l e t Light 39 Flavonoid Characterization and. I d e n t i f i c a t i o n I A. Data Presentation kk B. Explanation of the Data 51 Sig n i f i c a n c e of the Flavonoids of the Limnanthaceae A. Aglycones 60 B. Glycosides ' 62 The Contribution to V i s i b l e Flower Colour by Syringetin Derivatives 62 Environmental Modification of Flavonoid Composition 63 Comparative 'Flavonoid Data 65 Flavonoid Differences Between P e t a l and.Whole Plant Material 68 Numerical Taxonomic Results 68 Comparisons, of OTUs by Flavonoids A. Occurrence Tables 83 B. Cluster Analyses Qh C. 'Factor Analyses 88 D. Glycosylation Classes 89 TABLE OF CONTENTS — Continued Discussion Method and V a l i d i t y of Using Flavonoids as Taxonomic Characters A. Flavonoids as Taximetric Characters i . Introduction 92 i i . I nclusion of Negative Matches 92 i i i . Flavonoid Concentration 94 i v . Lack of V a r i a t i o n among Characters 95 B. Use of Dry versus Fresh Plant Material 96 C. P e t a l versus Whole Plant Results 97 D. Factor Analysis versus Conventional Cluster Analysis 98 E. Flavonoid Differences "between Duplicate Taxa. 99 Taxonomy of the Limnanthaceae . A. Introduction 101 B. Hypothetical Evolutionary History of the Limnanthaceae 104 C. Assumptions that 'Led to the Hypothetical History 107 D. Comparison of the Proposed C l a s s i f i c a t i o n to the' E x i s t i n g C l a s s i f i c a t i o n i . Above the Species Level' 111 i i . Species Level'and Below' 116 E. Summary of Proposals f o r Revision of the Limnanthaceae \^ . i , Introduction 128 i i . Genus Level 129 i i i . Section Level 129 i v . Species and V a r i e t a l Levels 129. F. Synopsis of the Family Limnanthaceae 130 . Conclusions 133 Li t e r a t u r e Cited 135 - v i -LIST OF TABLES TABLE I — Present Day C l a s s i f i c a t i o n of the Limnanthaceae TABLE II — I d e n t i f i c a t i o n and Sources of Operational Taxonomic Units of Limnanthes and Floerkea 1 TABLE I I I — I d e n t i f i c a t i o n s of the Flavonoids of the Limnanthaceae TABLE IV — Rfs and Colours' of Flavonoids of the Limnanthaceae TABLE V — Products Obtained upon T o t a l Hydrolysis with 1.0 N HC1 TABLE VI — Derivatives Obtained upon P a r t i a l Acid Hydrolysis and Hydrolysis with Emulsin TABLE VII — UV Absorption Maxima of Flavonoids of the Limnanthaceae TABLE VIII — Proton Chemical S h i f t Values of TMS Ethers of Six Flavonol y3-D-Rutinosides of the Limnanthaceae TABLE IX — Whole Plant Flavonoids of 30 OTUs of the Limnanthaceae TABLE X -.- Pe t a l Flavonoids of 1 8 OTUs of Limnanthes TABLE XI •— Synopsis of the Results of Four Cluster Analyses Based on Whole Plant Flavonoids of 30 OTUs TABLE XII — Synopsis of'•the Results of Four Cluster Analyses Based on Pe t a l Flavonoids of 1 8 OTUs - V l l -LIST OF FIGURES FIGURE I — D i s t r i b u t i o n of Limnanthes i n Oregon and C a l i f o r n i a FIGURE II — The Flowers of Six Selected Limnanthes taxa Taken in:" V i s i b l e and. U l t r a v i o l e t Light FIGURE III — Composite Two-Dimensional Map of 48 Flavonoid Glycosides of the Limnanthaceae FIGURE IV — Stepwise Elaboration of the B-Ring of Flavonols of the Limnanthaceae FIGURE V — TLC Map of Whole Plant Flavonoids of L. montana. Taken i n U l t r a v i o l e t Light FIGURE VI — TLC Map of Whole Plant Flavonoids o f L. montana Taken i n V i s i b l e Light FIGURE VIT -- Weighted Pair Group Clustering of 30 OTUs by Jaccard C o e f f i c i e n t : High Concentrations of Whole Plant 'Flavonoids • 'FIGURE VIII — Weighted Pair Group Clustering of 30 OTUs by Simple Match C o e f f i c i e n t : High Concentrations of Whole Plant Flavonoids FIGURE IX — Weighted Pair Group Clustering of 30 OTUs by Jaccard C o e f f i c i e n t : A l l Concentrations of Whole Plant Flavonoids FIGURE X — Weighted Pair Group Clustering of 30 OTUs by Simple Match C o e f f i c i e n t : A l l Concentrations of Whole Plant Flavonoids FIGURE XI — Weighted Pair Group Clustering of 18 OTUs by Jaccard C o e f f i c i e n t : High Concentrations of P e t a l Flavonoids • FIGURE XII — Weighted Pair Group Clustering of 18 OTUs by Simple Match C o e f f i c i e n t : HighOConcentrations of Petal Flavonoids FIGURE XIII — Weighted Pair Group Clustering of 18 OTUs by Jaccard C o e f f i c i e n t : A l l Concentrations of Pe t a l Flavonoids FIGURE XIV — Weighted Pair Group Clustering of 18 OTUs by Simple Match C o e f f i c i e n t : A l l Concentrations of Petal Flavonoids FIGURE XV - - P r i n c i p a l Components Analysis of 30 OTUs Based on Whole Plant 'Flavonoids FIGURE XVI — Varimax Factor Analysis of 30 OTUs Based on Whole Plant Flavonoids FIGURE XVII — P r i n c i p a l Components Analysis of 18 OTUs Based on P e t a l Flavonoids - V I 1 1 -LIST OF FIGURES — Continued FIGURE XVIII — Varimax Factor Analysis of 18 OTUs Based on Peta l Flavonoids FIGURE XIX — Hypothetical Phylogenetic Tree of the Limnanthaceae FIGURE XX — P r i n c i p a l Components Analysis of 9 OTUs of L. flo c c o s a Based on Whole Plant Flavonoids - i x -LIST OF APPENDICES APPENDIX I — Mason's. C l a s s i f i c a t i o n of Limnanthes R. Br. 139 APPENDIX II — Voucher Specimens of Limnanthes Taxa Grown from seed ikO APPENDIX III — TLC Map of Petal Flavonoids of L. do u g l a s i i var. d o u g l a s i i , OTU 1 l 4 l APPENDIX IV — TLC Map of Whole Plant Flavonoids of L. do u g l a s i i var. d o u g l a s i i , OTU 1 1 142 APPENDIX V — TLC Map of Pe t a l Flavonoids of L. do u g l a s i i var. d o u g l a s i i , OTU 2 143 APPENDIX VI — TLC Map of Whole Plant Flavonoids of L. do u g l a s i i var. d o u g l a s i i , OTU 2- ihk APPENDIX VII — TLC Map of'- Petal Flavonoids of L. do u g l a s i i var. nivea, OTU 3 1^5 APPENDIX VIII — TLC Map of Whole Plant Flavonoids of L. do u g l a s i i var. nivea, OTU 3 146 APPENDIX IX — TLC Map of Pe t a l Flavonoids of L. do u g l a s i i var. rosea, OTU h ikj APPENDIX X — TLC Map of Whole Plant Flavonoids of L. do u g l a s i i var. rosea, OTU h. 148 APPENDIX XI — TLC Map of Petal Flavonoids of L. do u g l a s i i var. sulphurea, OTU 5 149 APPENDIX XII -- TLC Map of Whole Plant Flavonoids of L. do u g l a s i i var. sulphurea, OTU 5 150 APPENDIX XIII — TLC Map of Whole Plant Flavonoids of L. vinculans 151 APPENDIX XIV -- TLC Map of Pe t a l Flavonoids of L. bakeri, OTU 7 152 APPENDIX- XV — TLC Map .of Whole Plant Flavonoids of L. bakeri, OTU 7 • 153 APPENDIX XVI — TLC Map of Pe t a l Flavonoids of L. s t r i a t a , OTU 8 154 APPENDIX XVII — TLC Map of Whole Plant Flavonoids of L. s t r i a t a , OTU 8 155 APPENDIX XVTII — TLC Map of Pe t a l Flavonoids of L. macounii, OTU 9 156 APPENDIX XIX — TLC Map of Whole Plant Flavonoids of L. macounii OTU'9-' 157 - X -LIST OF APPENDICES — Continued APPENDIX XX — TLC Map of Pet a l Flavonoids of L. macounii, OTU 1<0) APPENDIX XXI — TLC Map of Whole Plant Flavonoids of L. macounii, OTU 10 APPENDIX XXII — TLC Map of Petal Flavonoids of L. alba var. alba, OTU 11 APPENDIX XXIII — TLC .Map of Whole Plant Flavonoids of L. alba. var. alba, OTU 11 APPENDIX XXIV — TLC Map of Pet a l Flavonoids of L. alba var. alba, OTU 12 •APPENDIX XXV — TLC Map of Whole Plant Flavonoids of L. alba var. alba, OTU 12 APPENDIX XXVI — TLC Map of Petal Flavonoids of L. alba var. v e r s i c o l o r O T U 13 APPENDIX XXVII — TLC Map of Whole Plant Flavonoids of L. alba var. v e r s i c o l o r , OTU 13 APPENDIX XXVIIl — TLC Map of Pet a l Flavonoids of L. g r a c i l i s var. g r a c i l i s , OTU 'l.te APPENDIX XXIX — TLC Map of Whole Plant Flavonoids of L_. g r a c i l i s var. g r a c i l i s , OTU ik APPENDIX XXX — TLC Map of Pet a l Flavonoids of L. g r a c i l i s var. g r a c i l i s , OTU 15 APPENDIX XXXI — TLC Map of Whole Plant Flavonoids of L_. g r a c i l i s var. g r a c i l i s , OTU 15 APPENDIX XXXII — TLC Map of Petal Flavonoids of L. g r a c i l i s var. p a r i s h i i , OTU 16 APPENDIX XXXITI — TLC Map of Whole Plant Flavonoids of L_. g r a c i l i s var. p a r i s h i i , OTU 16 APPENDIX XXXIV — TLC Map of Pet a l Flavonoids of L. montana, OTU 17 APPENDIX XXXV — TLC Map of Whole Plant Flavonoids of L. montana, OTU 17 APPENDIX XXXVI1:— TLC Map of Petal Flavonoids of L. flo c c o s a ssp. b e l l i r i g e r i a n a , OTU 18 APPENDIX XXXVTI — TLC Map. of Whole Plant Flavonoids of L. floccosa' ssp. b e l l i r i g e r i a n a , OTU 18 - x i _ LIST OF APPENDICES — Continued APPENDIX XXXVIII — TLC Map of Whole Plant Flavonoids of L_. f l o c c o s a ssp. b e l l i n g e r i a n a , OTU 19 APPENDIX XXXIX — TLC Map of Pe t a l Flavonoids of L. flo c c o s a ssp. pumila, OTU 20 APPENDIX XL — TLC Map of Whole Plant Flavonoids of L. f l o c c o s a ssp. pumila., OTU 20 APPENDIX XLI — TLC Map of Whole Plant Flavonoids of L. f l o c c o s a ssp. pumila, OTU 21 APPENDIX XLII — TLC Map of Whole Plant Flavonoids of L_. f l o c c o s a ssp. g r a n d i f l o r a , OTU 22 APPENDIX XLI'II. — TLC. Map of Whole Plant Flavonoids of L. flo c c o s a ssp. f l o e cos a,;-OTU 23 APPENDIX XLIV — TLC Map' of Whole Plant Flavonoids of L. flo c c o s a ssp. f l o c c o s a , OTU 2k . APPENDIX XLV — TLC Map of Whole Plant Flavonoids of L. floccosa-ssp. f l o c c o s a , OTU 25 APPENDIX XLVI — TLC Map of Whole Plant Flavonoids of L. floccosa ssp. ' c a l i f ornica, OTU 26 APPENDIX XLVII — TLC Map. of Whole Plant -Flavonoids of F_. proserpinacoides,. OTU 27 • APPENDIX XLVITI — TLC Map of Whole Plant Flavonoids of F_. proserpinacoides ^ ©TU 28 APPENDIX XLIX — TLC Map of Whole Plant Flavonoids of F_. proserpinacoides, OTU 29 APPENDIX L — TLC Map of Whole Plant Flavonoids of F. proserpinacoides, OTU 30 APPENDIX LT — 100. mHz NMR Spectrum of TMS Ether of Kaempferol 3-0-y3-D-Rutinoside APPENDIX LIT — 100,mHz NMR Spectrum of TMS Ether of Quercetin 3-0-y3-D-Rut inos ide APPENDIX H I T r,-, 100. mHz NMR, Spectrum of - TMS Ether of Isorhamnetin 3-0-/3-D-Rut inos ide APPENDIX LTV — 100. mHz NMR Spectrum of TMS Ether of Myricetin 3-0-/3-D-Rut inos ide - x i i _ LIST OF APPENDICES — Continued APPENDIX LV — 100 mHz NMR Spectrum of TMS Ether of Laricytrin 3-0-y3-D-Rutinoside . 193 APPENDIX LVI — 100 mHz NMR Spectrum of TMS Ether'of Syringetin 3-0-y3-D-Rutinoside 19^ APPENDIX LVII — Matrix of Similarity Coefficients Calculated for 30 OTUs by Jaccard Coefficient: High Concentrations. 195 APPENDIX LVIIl — Matrix of Similarity Coefficients Calculated for 30 OTUs by Simple Match Coefficient: High Concentrations . 196 APPENDIX LIX — Matrix of Similarity Coefficients Calculated for 30 OTUs by Jaccard.Coefficient: A l l Concentrations 197 APPENDIX LX — Matrix of. Similarity Coefficients Calculated for 30 OTUs 'by'"; Simple Match Coefficient: A l l Concentrations 198 APPENDIX LXl — Matrix of Similarity Coefficients Calculated for 18 OTUs by Jaccard Coefficient: High Concentrations 199 APPENDIX LXII — Matrix of Similarity Coefficients Calculated for 18 OTUs by Simple Match Coefficient: High Concentrations 200 APPENDIX L X I I I — Matrix of Similarity Coefficients Calculated f 6 r r l 8 OTUs by Jaccard Coefficient: A l l Concentrations 201 APPENDIX LXIV — Matrix of Similarity Coefficients Calculated for 18 OTUs by Simple Match Coefficient!:;-All Concentrations 202 APPENDIX. LXV — Coordinates of 30 OTUs Plotted in Figure XV 203. APPENDIX LXVI — Coordinates of 30 OTUs Plotted in Figure XVI 20k APPENDIX LXVIT — Coordinates of 18 OTUs Plotted in Figure XVII 205 APPENDIX LXVIIT — Coordinates of 18 OTUs Plotted in Figure XVIII 206 7 - x i i i -ACKNOWLEDGEMENT I thank the following people who as s i s t e d i n the completion of t h i s t h e s i s : Dr. B.A. Bohm for his support and advice throughout the course of t h i s work; Drs. C.J. Marchant, J.R. Maze, W.B. S c h o f i e l d , I.E.P. Taylor and R.L. Taylor f o r t h e i r c r i t i c i s m s of the manuscript and h e l p f u l suggestions f o r i t s improvement; Dr. F;,-.W\ C o l l i n s f o r t e c h n i c a l advice; Mr. Steve Borden, I n s t i t u t e of Animal Resource Ecology, Uni v e r s i t y of B r i t i s h Columbia, for performing the computer analyses; Dr. Bob Bos.e, Environment Canada, Department of F i s h e r i e s , Vancouver, f o r determining the NMR spectra; Mr. G.A. White, U.S.D.A, ,' A g r i c u l t u r a l Research Service, B e l t s v i l l e , Maryland, for providing seed of Limnanthes; Dr. R. Ornduff, Dept. of Botany, Uni v e r s i t y of C a l i f o r n i a , Berkeley), f o r providing dried material of Limnanthes.; and my wife Jane C. Parker f o r her continued support and assistance. INTRODUCTION FAMILY DESCRIPTION The Limnanthaceae i s a small family of annual herbs containing two genera: Floerkea W i l l d . , which i s monotypic, and Limnanthes R. Br., which contains 18 recognized taxa. Floerkea proserpinacoides W i l l d . i s widely d i s t r i b u t e d across the United States and southern Canada. Limnanthes is. r e s t r i c t e d to southern Oregon, C a l i f o r n i a and the southern t i p of Vancou- • ver Island, B r i t i s h Columbia. Figure I i s a map showing the d i s t r i b u t i o n s of a l l Limnanthes taxa except L_. macounii T r e l . which occurs only i n the v i c i n i t y of V i c t o r i a , B.C. • A l l of the Limnanthaceae are early spring annuals, and most are associated with vernal pools or s o i l pockets kept moist by seepage. Limnanthes taxa vary i n t h e i r moisture requirements; some a c t u a l l y grow i n standing water, others at the edges of pools or streams, and others on moist h i l l s i d e s . With the exception of L_. montana Jepso'n and L. s t r i a t a . Jepson, both of which frequently occupy more shaded streamside habita t s , Limnanthes grows i n open spots. Taxa of Limnanthes occur over a range of elevations from nearly sea l e v e l to about lB'OO meters [Mason, 1952;. Gentry & M i l l e r , 1965; A r r o y o , 1973a; Munz & Keck, 1963). Floerkea grows i n the shade of shrubs or trees. A l l of these plants are ephemeral; they appear early i n the spring, then flower and set seed, and are soon overtopped by other plants and die. Although all.members of the Limnanthaceae are.self-compatible, t h e i r breeding systems vary from nearly t o t a l outcrossing to complete autogamy. Outcrossing members are adapted to insect p o l l i n a t i o n . These plants have, large numbers of a t t r a c t i v e , fragrant flowers, and tend to produce a carpet - 2 -FIGURE I D i s t r i b u t i o n of Limnanthes i n C a l i f o r n i a and Oregon L. doug.var.-L. s t r i a t a L_. baker i ^ - - . L. alba var. alba-L. f l o e . ssp.. f l o c c o s a -L_. vinculans - ' " L_. doug. var. sulphurea-- " L. doug. var. d o u g l a s i i " L. doug. var. n i v e a — L_. grac . var. p a r i s h i i - -- 3 -of blooms over a meadow. From t h i s character the common name for Limnanthes, "meadow foam", i s derived. A d d i t i o n a l t r a i t s associated with outcrossing species are protandry and the presence of nectar glands at the base of the filaments. Autogamous members of the family tend to have unscented, nectar-l e s s , and generally inconspicuous flowers. The extreme i s found i n L. macounii and F. proserpinacoides, which are t o f a l l y s e l f - p o l l i n a t i n g , have-small greenish flowers and.reduced numbers of f l o r a l parts.. These characters, coupled with small stature, make these plants very d i f f i c u l t to f i n d . TAXONOMIC BACKGROUND Past taxonomic treatments of the Limnanthaceae recognized that t h i s family comprises a d i s c r e t e group of c l o s e l y r e l a t e d plants. . However, these treatments have expressed c o n f l i c t i n g views regarding a f f i n i t i e s with others f a m i l i e s and re l a t i o n s h i p s within the family. To some extent, these c o n f l i c t s . remain unsettled today. A. Below the Family Level Robert Brown's o r i g i n a l circumscription of the family (Brown, 1833) delimits two genera, Limnanthes and Floerkea. Some.later authors have ques-tioned the v a l i d i t y of t h i s separation and united a l l members of the family i n the genus Floerkea. This subject i s reviewed by Mason i n his monograph of Limnanthes (1952). A c t u a l l y , the questrop of whether the family consists-of one genus or two has not yet been adequately resolved. The primary reason for t h i s i s that Mason excluded F. proserpinacoides from his treatment of Limnanthes. ;•' Since the modern c l a s s i f i c a t i o n of the Limnanthaceae i s based on Mason's treatment, t h i s .exclusion has r e s u l t e d i n a de facto separation of the family into two genera. Whether t h i s separation was j u s t i f i e d cannot be determined without applying the same taxonomic c r i t e r i a used by Mason to a l l - h -members of the family simultaneously, including Floerkea. A more recent study by Ornduff and Crovello (1968), based on a t a x i -metric analysis of morphological characters, concluded that the differences between Limnanthes and Floerkea are probably not great enough to d i s t i n g u i s h two genera on either phenetic or phylogenetic grounds. However, the two genera have not been formally reunited. Mason (1952) divided Limnanthes into 8 species and 1 1 v a r i e t i e s t o t a l i n g 1 5 taxa. His c l a s s i f i c a t i o n i s presented as Appendix I. Mason's taxonomic scheme i s based on the r e s u l t s of a r t i f i c i a l crosses and morphologi-c a l and c y t o l o g i c a l analyses. The r e s u l t s of Mason's c y t o l o g i c a l i n v e s t i g a t i o n show uniform pat-terns of 5 i d e n t i f i a b l e p a i r s of chromosomes, i n d i c a t i n g a close r e l a t i o n s h i p between a l l Limnanthes taxa (except L_. macounii which was thought to be ex-t i n c t at that time and f o r which Mason had no l i v i n g m a t e r i a l ) . This complete lack of karyotype v a r i a t i o n suggests either a comparatively recent divergence from ancestral stock of the taxa investigated, or a strong evolutionary conservancy of karyotype i n the family. Mason's c l a s s i f i c a t i o n of Limnanthes r e l i e s h e avily on the b i o l o g i c a l species concept. Thus, i n many instances experimentally i n t e r f e r t i l e taxa, formerly treated as separate species, are reduced to v a r i e t y rank. Taxa so reduced include two of the four v a r i e t i e s of L. douglasii.R. Br., the two v a r i e t i e s of L. alba Benth. , and the two v a r i e t i e s of L_. g r a c i l i s Howell. According to Mason, Limnanthes can be divided n a t u r a l l y into two sections, the Reflexae and the Inflexae (See Appendix I ) . Names of these two sections are derived from the p e t a l p o s i t i o n s assumed and maintained a f t e r p o l l i n a t i o n . Although; Mason did not detect other morphological differences coincident with t h i s d i v i s i o n , he concluded that the two sections represent two phylogenetically d i s t i n c t groups since no successful crosses could be made - 5 -across the sectional b a r r i e r . Ornduff and Crovello (1968) agreed with the separation of Limnanthes into two sections, although t h e i r presented results only weakly support t h i s separation.'- JJsing a taximetric analysis based on 35 morphological characters, a l l recognized taxa of the Limnanthaceae were clustered using the weighted pair group method and a va r i a t i o n of factor analysis. Two additional factor analyses were also performed, one on 18 f l o r a l characters, and the other on the remaining 17, or "vegetative" characters. The results of each clustering analysis are markedly different. Only the results of the vegetative analysis follow Mason's sectional alignment, producing a d i s t i n c t cluster of taxa corresponding to the section Reflexae and a f a i r l y diffuse grouping corres-ponding to the section Inflexae. I t i s notable that i n t h i s analysis F. proser- pinacoides also grouped with the Reflexae. The results of the analysis u t i -l i z i n g f l o r a l characters apparently contradict the presence of a sectional d i v i s i o n since no groups were separated. According to the authors, these results are.easily explained, i f not expected, since i n th i s Instance a c l a s s i f i c a t i o n based on f l o r a l characters i s i n effect closely linked to l e v e l of autogamy. Therefore, they hypothesize that the two sections are not separated by the f l o r a l analysis, since p a r a l l e l trends toward autogamy have apparently occurred simultaneously i n both sections. The analysis based on the combined data of the f l o r a l and vegetative analysis separated three clusters of taxa: the f i r s t contains L_. macounii and F. proserpinacoides, the second contains Mason's.three va r i e t i e s of L. floccosa Howell, and the t h i r d contains the remaining Limnanthes. taxa. Again the authors explained that the'observed phenetic a f f i n i t i e s are closely linked to l e v e l of autogamy and do not i n v a l i -date Mason's sectional d i v i s i o n , which they considered an accurate phylo-• genetic arrangement. In t h e i r 1968 study, Ornduff and.Crovello included two.undescribed - 6 -Limnanthes taxa with a f f i n i t i e s to members of the Reflexae. Ornduff has subsequently described one of these as a new species, Limnanthes vinculans Ornduff ( l 9 6 9 a - i - I t i s a narrow endemic growing sympatrically with L_. doug- l a s i i R. Br. var. nivea Mason which i t c l o s e l y resembles i n a l l but i t s l e a f l e t morphology. Arroyo (1973a) reexamined L_. f l o c c o s a , elevating Mason's three v a r i e t i e s to the rank of subspecies and creating two new subspecies. Level of autogamy, as demonstrated by a phenetic c l u s t e r analysis h eavily weighted with f l o r a l characters, was the sole c r i t e r i o n used to d i s t i n g u i s h the f i v e subspecies, although geographical and e c o l o g i c a l evidence was c i t e d to support t h i s separation. Arroyo concluded that the f i v e taxa of L_. f l o c c o s a repre-sent stages i n the trend toward complete autogamy i n the Inflexae and form a natural group recently derived from the predominantly outcrossing L. alba Benth. (1973a, 1973b). She d i d not explain why she chose to divide L_. f l o c - cosa into subspecies i n contrast to the v a r i e t a l subdivision used uniformly by Mason f o r the r e s t of the genus. In h i s most recent p u b l i c a t i o n concerning the Limnanthaceae, Ornduff C l 9 7 l ) reported that both L. macounii and F_.. proserpinacoides.' have chromosome numbers of n=5, the same as the rest of the family. Ornduff (1971) also pre-sented the r e s u l t s of extensive Limnanthes.interbreeding t r i a l s . In general •his r e s u l t s confirmed the work'1- of,';' Mas on, demonstrating low i n t e r s p e c i f i c c r o s s a b i l i t y and a high degree of chromosomal homology (inc l u d i n g L. macounii). However, Ornduff made three important observations regarding s t e r i l i t y bar-r i e r s i n Limnanthes-. l ) In general, the presence of s t e r i l i t y b a r r i e r s be-tween taxa does not correspond with the mophological differences d i s t i n g u i s h i n g these taxa. •2)'. Since the c y t o l o g i c a l differences between taxa are of a r e l a -t i v e l y minor nature, hybrid s t e r i l i t y , where present, probably has a genetic bas i s . 3) The presence of a s t e r i l i t y b a r r i e r between two taxa i s most l i k e l y - 7 -i f the two lack d i s t i n c t geographical d i f f e r e n c e s . In t h i s l a s t regard, i t i s of p a r t i c u l a r i n t e r e s t that L. macounii, which i s the most disjunct species, of Limnanthes, produced the; highest number of i n t e r s p e c i f i c hybrids including the only i n t e r s e c t i o n a l hybrid, a s t e r i l e o f f s p r i n g r e s u l t i n g from a cross with L_. montana. Table I shows the modern c l a s s i f i c a t i o n of the Limnari'thaceae. It • i s based on Mason's c l a s s i f i c a t i o n of Limnanthes, and i t includes the more recent taxonomic additions previously c i t e d . In a d d i t i o n , i t l i s t s Floerkea as a d i s t i n c t genus. In summary, r e l a t i o n s h i p s below the family l e v e l i n the Limnanthaceae remain unclear f o r two primary reasons; l ) i t i s probable that c e r t a i n simi-l a r i t i e s have ar i s e n by convergence; the-trend toward autogamy has most l i k e l y occurred independently i n the family at l e a s t three times and possibly more often. 2) the b i o l o g i c a l species concept employed by Mason i s probably of l i m i t e d value i n t h i s context,.since genetic, b a r r i e r s to interbreeding exist between c l o s e l y r e l a t e d sympatric taxa, but may not be present between more d i s t a n t l y r e l a t e d a l l o p a t r i c taxa. B. Family- A f f i n i t i e s -In h i s d e s c r i p t i o n of the Limnanthaceae, Robert Brown (1833) was uncertain regarding a f f i n i t i e s with other f a m i l i e s . He implied that his e a r l i e r examinations of Floerkea suggested an a s s o c i a t i o n with the perigynous f a m i l i e s . However, a f t e r the discovery of Limnanthes, he decided that the family was more properly a l l i e d with the hypogynous f a m i l i e s . Other nineteenth century authors generally agreed with Brown, either l i n k i n g t o , or including members of the Limnanthaceae i n , the Geraniaceae. Engler and P r a n t l (1896) disagreed including the family i n the Sapindales. Most modern authors, including Hutchinson (1926), Cronquist (1968), - 8 -TABLE I Present Day C l a s s i f i c a t i o n of the Limnanthaceae Floerkea W i l l d . l ) F_. proserpinacoides W i l l d . Limnanthes R. Br. Section Reflexae 2) L. bakeri J.T. Howell 3) L_. do u g l a s i i R. Br. var. do u g l a s i i " var.- nivea Mason " var. rosea (Benth.) Mason " var. sulphurea Mason 4) L_. macounii T r e l . 5) L_. s t r i a t a Jepson 6) L. vinculans. Ornduff Section Inflexae 7) L_. alba Benth. yar. . alba ". var. v e r s i c o l o r (Greene) Mason 8) L_. flo c c o s a Howell ssp. flo c c o s a . " ssp. b e l l i n g e r i a n a (M.E. Peck) Arroyo " ssp. c a l i f o r n i c a Arroyo " ssp. g r a n d i f l o r a Arroyo " ssp. pumila (Howell) Arroyo 9) L_. g r a c i l i s Howell var. g r a c i l i s " var. p a r i s h i i (Jepson) Mason 10) L_. montana Jepson - 9 -and Takhtajan (1969), agree that the Limnanthaceae constitutes a d i s t i n c t family belonging to the Geraniales. Maheshwari and J o h r i (1956) took excep-t i o n to t h i s viewpoint a f t e r comparing the Limnanthaceae to the Sapindales and the Geraniales. Their comparison was based on embryological and morpho-l o g i c a l features of Limnanthes discovered by Mason (1951) and Mathur (1956), together with t h e i r own findings on embryo development, f l o r a l morphology, and seed morphology of F. proserpinacoides. The authors showed that the Limnanthaceae has few morphological s i m i l a r i t i e s to either the Geraniales or Sapindales, with regard to embryogenesis, and seed, p o l l e n , and gynoecium morphology. On t h i s b a s i s , Maheswari and J o h r i concluded that assignment of the Limnanthaceae to either the Sapindales or Geraniales i s uns a t i s f a c t o r y , and that probably the family should be r a i s e d to o r d i n a l rank. Unfortunately, the authors do not speculate.on the d e r i v a t i o n of the Limnanthaceae, nor do they suggest with which orders, i f any, the Limnanthaceae shares the greatest a f f i n i t y . Warburg (1938) discussed i n c l u s i o n of the Limnanthaceae i n the Gerani-ales i n l i g h t of i t s chromosome cytology. He made chromosome counts, deter-mined karyotype morphology, and observed chromosome behaviour at meiosis f o r at l e a s t onfe-'representative of each family i n the Geraniales sensu Hutchinson (1926), i..e/, Geraniaceae, Oxalidaceae, Tropaeolaceae, Limnanthaceae, Linaceae, Balsaminaceae and Zygophyllaceae. The behaviour of chromosomes of Limnanthes at meiosis, t h e i r large s i z e , and small number caused Warburg to propose two alternate hypotheses regarding placement of the Limnanthaceae; either the family i s improperly placed i n the Geraniales and.more l i k e l y f i t s i n to the Sapindales as proposed by Engler a n d P r a n t l ; or the family i s a pr i m i t i v e r e l i c c l o s e l y r e l a t e d to ancestral taxa from which members of the Geraniales are derived. Neither of these hypotheses has been supported by more recent d i s c o v e r i e s , but the c y t o l o g i c a l observations on which they were - 10 -based remain valuable., .. PREVIOUS CHEMICAL INVESTIGATIONS Considering the small s i z e of the Limnanthaceae, a r e l a t i v e l y large amount of research' has been dir e c t e d toward characterizing t h i s family'.;s cherr.-risitry-^- This i s not s u r p r i s i n g i n view of the i n t e r e s t i n g r e s u l t s that have, been obtained. Members of the family contain glucosinolates, a character known to be shared by only about a dozen other f a m i l i e s . m-Methoxybenzyl i s o t h i o -cyanate was i s o l a t e d and characterized from seed of L_. d o u g l a s i i by E t t l i n g e r and Lundeen (.1956). A c t u a l l y , t h i s isothiocyanate i s not n a t u r a l l y occurring; instead, i t i s a d e r i v a t i v e of the ,glueoside present i n the seed, which i s formed by the enzymic release of glucose. Hence, the n a t u r a l l y occurring glucoside of L_. d o u g l a s i i i s m-methoxybenzyl glucosinolate, given the t r i v i a l name glucolimnanthin. This compound possesses an extremely rare s u b s t i -t u t i o n pattern not known outside the Limnanthaceae.' M i l l e r et a l . (1964) have detected an a d d i t i o n a l u n i d e n t i f i e d glucosinolate i n L. d o u g l a s i i var. nivea, and also reported the presence of glucosinolates i n a l l other Lim- nanthes taxa.(Gentry & M i l l e r , 1965). The United States Department of A g r i c u l t u r e (U.S.D.A.) has conducted extensive research on the seed o i l chemistry of Limnanthes (Gentry & M i l l e r , 1965; M i l l e r et a l . , 1964). Unlike common vegetable o i l s based on 16 or l8-carbon f a t t y acids', the seed o i l of Limnanthes i s based on 20 and 22-carbon f a t t y acids. Positions of unsaturation of the f a t t y acids of Limnanthes are'-also uncommon. Four such f a t t y acids comprise about 95% of the t o t a l seed o i l f r a c t i o n of Limnanthes; of these, three are unknown from other sources, while the" fourth i s erucic a c i d , also present i n rape and mustard o i l s . Before the present i n v e s t i g a t i o n was completed, l i t t l e was known of - l i -the phenolic substances of the Limnanthaceae. Bate-Smith (1962) reported that hydrolyzed extracts of leaves of L_. d o u g l a s i i contained c a f f e i c a c i d , kaempferol, quercetin, myricetin, leucocyanidin and leucodelphinidin. As part of the present i n v e s t i g a t i o n , Parker and Bohm (1975) characterized 18 f l a v o n o l glycosides from L_. d o u g l a s i i . The r e s u l t s and implications of that study are presented i n l a t e r chapters. AGRONOMIC EVALUATION OF LIMNANTHES The.fatty acids present i n the seed o i l of Limnanthes are w e l l suited to the production of high q u a l i t y wax products. Because of t h i s character, and since the o i l content of the nutlets i s 20-33% (Gentry & Miller , 1 9 6 5 ) , the U.S.D.A. has conducted extensive t r i a l s to determine the s u i t a b i l i t y of Limnanthes as a crop plant. T r i a l s were made.in Alaska, Oregon, C a l i f o r n i a and Maryland, and.consisted of p l a n t i n g , harvesting and c a l c u l a t i n g y i e l d s f o r most of the Limnanthes taxa (Higgins et_ a l . , 1971). In addition, various researchers have t r i e d to determine optimal conditions f o r germi-nation of Limnanthes. seed (Toy & Willingham, 1966; 1967; Cole, 197*0. The r e s u l t s of the f i e l d t r i a l s i n d i c a t e that most of the w i l d populations of Limnanthes produce abundant seed i n the range of 1000 kg/ hectare when c u l t i v a t e d . From an agronomic standpoint L_. alba i s considered the most desirable species due to i t s upright form and good seed r e t e n t i o n , both q u a l i t i e s being necessary f o r mechanical harvesting. Although p r o f i t s are'marginal at present, Higgin et al.- (1971) concluded that L_. alba prob-Yiably contains s u f f i c i e n t genetic v a r i a b i l i t y such that an a r t i f i c i a l s e l e c t i o n program could s u b s t a n t i a l l y increase seed y i e l d s . Since Limnanthes seed contains 15-25% p r o t e i n , the authors suggested that the extracted meal could be used as stock fodder, i f r e s i d u a l l e v e l s of t o x i c glucosinolates are not too high.' - 12 -THESIS OBJECTIVES The Limnanthaceae i s p a r t i c u l a r l y w ell suited to the a p p l i c a t i o n of chemical taxonomic techniques. Due to the uncertainty of r e l a t i o n s h i p s , both within the family and with other f a m i l i e s , the discovery and a p p l i c a t i o n of new information with taxonomic s i g n i f i c a n c e i s d e s i r a b l e . Unless new information can be a p p l i e d , . i t i s probably impossible to f u l l y elucidate family r e l a t i o n s h i p s , and therefore, impossible to t e s t the v a l i d i t y of the c l a s s i f i c a t i o n shown i n Table I. The.chemical characters of the family are an untapped source of p o t e n t i a l l y valuable taxonomic information. Secondary chemical constituents have served as useful taxonomic• characters f o r many groups of plants, both large and small, and i t i s possible that the systematic u t i l i s a t i o n of these characters might be u s e f u l i n the context of the Limnanthaceae. The i n f o r -mation a v a i l a b l e on seed fats suggests that these constituents possess l i t t l e taxonomic s i g n i f i c a n c e ( M i l l e r . e t a l . , 196k). The possible taxonomic s i g n i -ficance of other classes of secondary compounds, including the g l u c o s i n o l a t e s , cannot be estimated u n t i l they are more completely elucidated. The flavonoids of the Limnanthaceae also f a l l i nto t h i s l a t t e r category. The primary intentions of the present study are: l ) to determine the flavonoid composition of each,member of the Limnanthaceae; 2 ) . t o produce an a r t i f i c i a l c l a s s i f i c a t i o n ( s ) of the family based s o l e l y on flavonoid compo-s i t i o n s ; 3) to compare the chemical c l a s s i f i c a t i o n with the e x i s t i n g pheno-typ i c treatments; and k) to discuss any implications .regarding evolution within the family or regarding a f f i n i t i e s to other f a m i l i e s uncovered i n the course .of t h i s work. l h Y There are three secondary objectives 'o§ t h i s study: l ) to assess the value of the contribution flavonoid character taxonomy can make to the syste-matics of the Limnanthaceae; . 2) to discuss b r i e f l y the method of a p p l i c a t i o n - 13 -of flavonoids as characters i n plant taxonomy; and 3) to assess the taxo-nomic s i g n i f i c a n c e of flower patterning i n the family. - 14 -MATERIALS AND METHODS THE PLANTS AND THEIR SOURCES Whenever possible f r e s h l y grown plant material'was used f o r flavonoid comparisons. This p r a c t i c e tends to eliminate several p o t e n t i a l sources of chemical v a r i a t i o n f o r the following reasons: the use of fresh material eliminates most ambiguities a r i s i n g either from compounds being present i n low concentration or from i n d i v i d u a l v a r i a t i o n , since a large amount of plant material i s us u a l l y a v a i l a b l e and many plants can be analyzed simul-taneously; environmentally induced v a r i a t i o n i n flavonoid composition, i f any e x i s t s , also i s eliminated, since a l l plants can be grown under nearly uniform conditions; and although flavonoid breakdown i s apparently i n s i g n i f i -cant i n pressed specimens of Limnanthes and Floerkea, the use of fr e s h material f o r chemical comparisons eliminates t h i s possible source of v a r i a t i o n . Approximately three quarters of the Limnanthes taxa were grown from seed provided by G.A. White, U.S.D.A., A g r i c u l t u r a l Research Service, B e l t s v i l l e , Maryland. This seed was a v a i l a b l e as a r e s u l t of the agronomic evaluation being conducted by the U.S.D.A. The taxa f o r which no seed was ava i l a b l e were those described a f t e r the U.S.D.A. work was underway, including L_. vinculans and the various subspecies of L_. floc c o s a. Unfortunately, seed of Floerkea was unobtainable from any source. Since i t i s nearly e s s e n t i a l to include a l l members i n a comparative study, herbarium specimens were used f or analysis of those taxa of the Limnan-thaceae f o r which no seed could be obtained. Dried material of L_. vinculans and L_. flo c c o s a was provided by Dr.. R. Ornduff, Department of Botany, Univer-s i t y of C a l i f o r n i a , Berkeley, while several c o l l e c t i o n s of Floerkea were provided by the U n i v e r s i t y of B r i t i s h Columbia Herbarium. The p a r t i a l use of d r i e d material causes c e r t a i n i n t e r p r e t i v e prob-lems a r i s i n g from the analysis of non-parallel material. The e f f e c t s of t h i s p r a c t i c e on the r e s u l t s of t h i s study -will be discussed i n the l a s t chapter. In the course of t h i s work, i t was r e a l i z e d that Dr. W.B. S c h o f i e l d , Department of Botany, University of B r i t i s h Columbia, had discovered a popu-l a t i o n of the extremely rare L_. macounii at .William Head, Vancouver Island about 25 miles from the only previously known population at V i c t o r i a , B.C. (Mr. A. Ceska, Department of Biology, U n i v e r s i t y of V i c t o r i a , has informed me that he and his wife have discovered a d d i t i o n a l populations since that time.) Material was grown from seed taken from the S c h o f i e l d c o l l e c t i o n , and addi-t i o n a l plants were c o l l e c t e d from t h i s new population. Table II i d e n t i f i e s the sources of a l l plants used i n t h i s study. The c o l l e c t o r s and t h e i r c o l l e c t i o n numbers are given f o r the herbarium materi; The remaining plants are 'Identified by the o r i g i n a l U.S.D.A. plant accession numbers. These numbers are the same as those r e f e r r e d to by Higgins et a l . (1971) i n t h e i r agronomic evaluation study. Each entry i n Table II was treated as a d i s t i n c t taxon and analyzed separately, and each has been assigned an Operational Taxonomic Unit (OTU) number. The assignment of these numbers roughly follows the accepted c l a s s i -f i c a t i o n of the family, but t h i s assignment was made only to i d e n t i f y OTUs. Certain v a r i e t i e s , subspecies or species are represented by more than one OTU i n Table I I . These r e p e t i t i v e OTUs represent d i f f e r e n t c o l l e c t i o n s and almost" c e r t a i n l y d i f f e r e n t populations as w e l l . The decision to analyze separately d i f f e r e n t c o l l e c t i o n s of a taxon does not r e f l e c t any preconceived i taxonomic judgements. A primary reason f o r separateutreatment i s that some of the taxa were i d e n t i f i e d according to d i f f e r e n t taxonomic c r i t e r i a and • conventions. 'For instance, i t was considered unwise to group fresh material - 16 -TABLE II I d e n t i f i c a t i o n and Sources of Operational Taxonomic Units Of Limnanthes and Floerkea OTU Taxon Name U.S.D.A. Accession Plant C o l l e c t o r and No. C o l l e c t i o n No. 1 L. do u g l a s i i var. d o u g l a s i i 2T81T0 2 L. d o u g l a s i i var. d o u g l a s i i 283T08 3 L. do u g l a s i i var. nivea 283T13 k L. do u g l a s i i var. rosea 283T15 5 L. do u g l a s i i var. sulphurea 283T18 6 L. vinculans Rubtzoff 5699 7 L. bakeri 283T06 8 L. s t r i a t a 283T2T 9 L. macounii ( V i c t o r i a ) 315048 10 L. macounii (William Head) Scho f i e l d 11 L. alba var. alba 283T01 12 L. alba var. alba B55689 13 L. alba var. v e r s i c o l o r 283705 14 L. g r a c i l i s var. g r a c i l i s 283722 15 L. g r a c i l i s var. g r a c i l i s 283723 16 L. g r a c i l i s , var. p a r i s h i i 283724 IT L. montana 283725 18 L. floccosa ssp. b e l l i n g e r i a n a . 283720 19 • L. fl o c c o s a ssp. b e l l i n g e r i a n a K a l i n (Arroyo) 7031 20 L. fl o c c o s a ssp. pumila 283721 21 L. flo c c o s a ssp. pumila K a l i n 7033 22 L. f l o c c o s a ssp. grandiflora. K a l i n 7028 23 L. flo c c o s a ssp. floccosa K a l i n 7026 24. L. flo c c o s a ssp. fl o c c o s a - K a l i n 7022 25 L. flo c c o s a ssp. fl o c c o s a K a l i n 6917 26. L.' floccosa s s p . ' c a l i f o r n i c a Niehaus 371 & Ornduff 6885 2T F.- proserpinacoides (Dutchess Co. , N.Y. ) Ahles 76368 28 F. proserpinacoides (Kettle F a l l s , Wash.) Beamish 60352 29 F. proserpinacoides (Montreal, Quebec) Rouleau 4001 30' F. proserpinacoides (Shushan, N.Y. .) ' Dobbin 1758 - 17 -of L_. floccosa var. b e l l i n g e r i a n a , i d e n t i f i e d on the basis of a three variety-subdivision of L_. f l o c c o s a , with pressed material of L_. flo c c o s a ssp. b e l l i n - geriana^ i d e n t i f i e d according to a f i v e subspecies d i v i s i o n . Therefore, OTUs 18 and 19 were analyzed separately. Three taxa obtained from the U.S.D.A. were not i d e n t i f i e d according to Mason's c l a s s i f i c a t i o n . These taxa are: L. douglasii., OTU 1; L_. g r a c i l i s , OTU 14; and L_. alba, OTU 11. A f t e r growth and examination, each of these taxa was i d e n t i f i e d as the respective t y p i c a l variety,, and they are designated i n this, fashion i n Table I I . , I t i s of i n t e r e s t that the seed of OTU 1 was o r i g i n a l l y from Europe, and probably t h i s taxon i s descended from the plants o r i g i n a l l y c o l l e c t e d by David Douglas i n the early 1830s from which Brown described the family (Gentry & M i l l e r , 1965). In t h i s case an a r t i f i c i a l i s o l a t i o n has probably been maintained f o r nearly 150 years, which allows an estimation of the e f f e c t s of genetic' d r i f t and inbreeding over t h i s period. Floerkea pros.erpinacoides and L. macounii are subdivided i n t o four and two OTUs r e s p e c t i v e l y , as shown i n Table I I . Both are t o t a l l y auto-gamous, and populations from d i f f e r e n t locations are e f f e c t i v e l y g e n e t i c a l l y i s o l a t e d . For t h i s reason, separate analyses of these populations may reveal differences i n flavonoid composition a t t r i b u t a b l e to genetic d r i f t or d i f f e r e n t s e l e c t i o n pressures. In the case.of L. macounii, the two populations are only a few miles apart, but i n the case.of F. proserpinacoides, the most widely d i s t r i b u t e d element of the family, the populations are from the ex-tremes of i t s range. Three d i f f e r e n t c o l l e c t i o n s of L_. floccosa ssp. f l o c c o s a were a v a i l a b l e from various l o c a t i o n s . Each of these c o l l e c t i o n s was made by Arroyo and was used i n her systematic studies. These c o l l e c t i o n s are treated.here as d i s t i n c t OTUs to assess the v a r i a t i o n within t h i s taxon and to t e s t the validity-.of'Arroyo's recent taxonomic treatment of L. flo c c o s a (Arroyo, 1973a). - 18 -In the course of growing L. d o u g l a s i i var. sulphurea, OTU 5, U.S.D.A. Accession #283718, i t was observed that these plants had p e t a l patterning d i f f e r i n g from Mason's de s c r i p t i o n of t h i s v a r i e t y . Mason (1952) described i t as having completely yellow p e t a l s , while the U.S.D.A. plants have a small amount of white at the t i p s . However, they have a smaller amount than the petals of L. d o u g l a s i i var.. d o u g l a s i i . Otherwise,' these plants correspond to Mason's d e s c r i p t i o n of v a r i e t y sulphurea, including l e a f l e t shape, a char-acter which distinguishes t h i s taxon from other v a r i e t i e s of L_. d o u g l a s i i . Mason described F^ hybrids of v a r i e t i e s sulphurea and.douglasii as having p e t a l t i p s with an intermediate amount of white. Whether the abnormal colouring of the U.S.D.A. v a r i e t y r e s u l t s from such crossing i s not known, but the i d e n t i f i c a t i o n of OTU 5 as L_. doug l a s i i var. . sulphurea was t e n t a t i v e l y accepted f o r t h i s study. Voucher specimens of a l l plants grown for t h i s i n v e s t i g a t i o n are deposited i n the U.B.C. Herbarium. Appendix II l i s t s these, vouchers and t h e i r c o l l e c t i o n numbers. The four c o l l e c t i o n s of Floerkea are.also on f i l e • i n the U.B.C. Herbarium. Voucher specimens of L. flo c c o s a and. L. vinculans, that were provided by Ornduff, are on f i l e at the Herbarium of the Un i v e r s i t y of C a l i f o r n i a , Berkeley. PLAWT GROWTH Most plants were grown i n a greenhou®^' under nearly uniform condi-t i o n s . The d i f f e r e n t taxa were grown i n random groups, two to four at a time, from the end of 1972 to the beginning of 1975- One taxon,-L.. d o u g l a s i i var. douglasii,. OTU 1, was also grown outdoors on a much la r g e r s c a l e - i n the summer of 1973 to provide a quantity.of fresh plant material so that the major-flavonoids could be i s o l a t e d . Mason (1952)'. reported that Limnanthes seed w i l l not germinate, unless - 19 -the s o i l temperature i s kept under 15°. Preliminary germination t e s t s con-firmed t h i s observation and also showed that s t r a t i f i c a t i o n of the seed at about 5-10° f o r 3-5 days enhanced germination. Therefore, . a l l seed was placed i n a r e f r i g e r a t o r f o r a few days before planting i n pots. The pots were placed i n c o l d frames on top of the Bio-Sciences B u i l d i n g , U n i v e r s i t y of B r i t i s h Columbia and kept there u n t i l germination was complete. This p r a c t i c e was f a i r l y s a t i s f a c t o r y except i n the middle of winter or the middle of sum-mer, when temperatures i n the cold frame exceeded the range i n which Limnanthes w i l l germinate. A f t e r germination, pots of young Limnanthes plants were tr a n s f e r r e d to a greenhouse and allowed a short growth period. A f t e r development of the second or t h i r d permanent l e a f , the young seedlings were transplanted 20 to a f l a t , . u s i n g a mixture of about 90%. s t e r i l i z e d garden s o i l and 10%. sand. The - s u r v i v a l rate of transplanted seedlings was-close to 100%,, i f they, were not overwatered. Since the moisture requirements of the d i f f e r e n t taxa v a r i e d , each taxon was watered an appropriate amount a r r i v e d at by t r i a l and error; i f any signs of damping-ioff appeared, the amount of water.was reduced. Plants used f o r flavonoid comparisons were grown on the same bench under the same bank.of ordinary fluorescent l i g h t s . Like many early spring annuals, taxa of Limnanthes require long daylight periods to.flower (Mason, 1952). I t was found that a l l taxa r a p i d l y flowered under.l6'hour daylength, so t h i s l i g h t duration was used co n s i s t e n t l y . Also, i t was discovered early i n the course of t h i s work that should daylength be shortened d r a s t i c a l l y (i.e_. , 16 hour to an 8 hour daylength) after-the plants have.begun flowering, they r a p i d l y cease flowering and die. A complete generation of Limnanthes, from seed germination, to seed ripening, occurs i n 3-h months'under the'con-di t i o n s outlined above. • The procedure f o r growing L. d o u g l a s i i var. d o u g l a s i i outdoors was - 20 -s i m i l a r to the above, except that seedlings were transplanted d i r e c t l y into the ground instead of f l a t s . Plants were spaced every 15 cm i n a s i n g l e row about 50 m long. Seed f o r t h i s planting was germinated i n l a t e June, and plants were harvested i n l a t e September, j u s t p r i o r to flowering. Outdoor-grown plants produced much coarser, more lush vegetation than the corresponding greenhouse grown plants. Whether t h i s d i f f e r e n c e r e s u l t s from the n a t u r a l l y shortening dayleng'th or some a d d i t i o n a l factors associated with the outdoor planting i s not known. • There are two problems associated with the growth of Limnanthes plants which deserve comment. Their succulent vegetation i s p a r t i c u l a r l y a t t r a c t i v e to aphids and w h i t e f l i e s , both common greenhouse pests. These insects s e l e c t i v e l y i n f e s t Limnanthes plants, often occurring i n great?numbers on them while adjacent plants are.unaffected. Therefore, i n order to obtain s a t i s f a c t o r y y i e l d s , i t was necessary to spray plants p e r i o d i c a l l y with "Raid" i n s e c t i c i d e (.S.C. Johnson & Son, Ltd . ) . I n t e r e s t i n g l y , the f i e l d grown plants.of L . douglasii- were not attacked by ins e c t s . The second problem associated with Limnanthes culture regards i ; t s root c o l l a r . . In order f o r a plant to reach maturity, i t s root c o l l a r must remain buried. Since t h i s root c o l l a r extends only a short distance below the ground (.1-2 cm), care must be taken when watering and weeding not to disturb the s o i l immediately around the plant. I f the root c o l l a r i s uncovered, the plant w i l l l i v e only a day or so. PLANT.. TISSUE. SELECTION The type of t i s s u e used f o r flavonoid analysis v a r i e d with the source of plant material.. Since.only small.amounts of pressed specimens were a v a i l -able, a l l the material at hand was used. However, for- plants grown from seed, choices had to be made regarding .1) whether a l l tissues should be - 21 -included i n the flavonoid comparison t e s t s , and 2) whether any t i s s u e s should be analyzed separately. A l l members of the Limnanthaceae have a basal rosette of succulent, but f i n e l y divided, compound leaves and.caulescent flowering stems, which ar e . l a r g e l y indeterminate.and may e a s i l y produce over 25 blooms i n most species. The plants have short root c o l l a r s with a d i f f u s e network of f i n e l y dissected, deeply penetrating roots, a c h a r a c t e r i s t i c which makes root c o l -l e c t i o n d i f f i c u l t and.tedious. E a r l y i n the course of t h i s i n v e s t i g a t i o n , the flavonoid contents of the roots of L_. d o u g l a s i i var. nivea, OTU 3, were compared to the contents.of the'above ground parts to determine whether root recovery was necessary. For t h i s experiment the roots were e f f i c i e n t l y r e -covered using a stream of water to wash away the surrounding s o i l . This pre-liminary, t e s t showed r e l a t i v e l y weak concentrations and a more s i m p l i f i e d pattern of flavonoids i n the roots.than i n l e a f and stem material. For t h i s reason, no further'attempts were made, to rec'over quantitative, amounts of root material from other-taxa. However, some root material was included with the plants used f o r comparative study since they were harvested by p u l l i n g them from the s o i l a f t e r flowering had begun. Although t h i s procedure y i e l d s only small" amounts.af-root t i s s u e , the'flavonoids present i n t h i s material were considered to comprise the t o t a l flavonoid composition of any given taxon, and t h i s composition i s hereafter r e f e r r e d to as the "whole plant flavonoid composition'' No' root material was c o l l e c t e d when L_. d o u g l a s i i var. d o u g l a s i i was grown i n the open.field. To f a c i l i t a t e . h a r v e s t i n g , these plants were simply cut o f f at ground l e v e l . However, f o r comparative a n a l y s i s , t h i s taxon was regrown -under, the same conditions as the other taxa. The' flowers of many Limnanthes taxa are.quite s t r i k i n g , some taxa having flowers with.various shades of yellow patterning on otherwise white - 22 -'.to ivory coloured petals. Since there i s s t r i k i n g v a r i a t i o n between flowers of some taxa, and preliminary t e s t s showed high flavonoid concentrations i n the p e t a l s , the decision was made to analyze separately the petals of each taxon grown from seed. In t h i s manner the flavonoid compositions of these OTUs could be compared on the basis of pe t a l composition as w e l l as whole plant composition. The petals from 10-20 i n d i v i d u a l s , c o l l e c t e d over a 2-3 week period, y i e l d e d quantities of flavonoids adequate for comparative a n a l y s i s . FLORAL PATTERNING IN ULTRA VIOLET LIGHT  A. Introduction I n s e c t - p o l l i n a t e d , outcrossing plants frequently possess s p e c i a l adaptations, such as nectar glands and patterned flowers that are associated with t h i s type of breeding system. Contrasting petal colouring i s common i n such plants and often r e s u l t s i n a "bullseye" e f f e c t . This type of pattern tends to draw an observer's eye toward the flower center. Since the goal of the insect p o l l i n a t o r i s the nectar glands rather than the reproductive parts of a flower, bullseye patterns have.been interpreted as adaptations which a t t r a c t v i s i t i n g insects to the center of the flower and.have thus been termed "nectar guides". Only by looking at a flower as a bee sees i t , i s i t possible to f u l l y appreciate colour adaptations that? f a c i l i t a t e p o l l i n a t i o n by t h i s vector. Although t h i s cannot be done d i r e c t l y , i t i s possible to examine flowers at a l l wavelengths to which bees respond. Von Frisch' s work on honeybee v i s i o n (1967) has revealed that bees, unlike humans, possess l i g h t receptors sensi-t i v e to near u l t r a violet^MY.1} l i g h t . By photography using a f i l t e r trans-parent only to UV l i g h t and f i l m s e n s i t i v e to t h i s colour range, various investigators have discovered that c e r t a i n flowers have nectar guides that are v i s i b l e only i n UV l i g h t . Flowers with such nectar guides have contrasting - 23 -bright and dark regions, corresponding to areas of reflectance and absorption of UV l i g h t . To date, two classes of flavonoids, flavonols and.chalcones, both with absorption maxima i n the near UV range, have been i d e n t i f i e d as the pigments responsible f o r the UV absorbing portion of the nectar guides of Rudbeckia and Oenothera v i s i b l e under UV l i g h t (Thompson et_ al_. , 1972; Dement & Raven, 19lh). Comparison of the f l o r a l patterns of c l o s e l y r e l a t e d plants as the bee sees them may reveal information with p o t e n t i a l systematic importance. Daumer (1958) has found that flowers with patterning v i s i b l e i n UV are more a t t r a c t i v e to p o l l i n a t i n g bees than unpatterned flowers. This phenomenon has le d Ornduff and Mosquin (1970) to conclude that plants which have evolved a system of nectar guides v i s i b l e i n UV l i g h t are more highly adapted to an out-crossing habit than r e l a t e d plants without such a system. Discrimination by p o l l i n a t i n g agents on the basis of f l o r a l appear- '• ance can play an important r o l e i n the divergence of populations and the creation of new taxonomic e n t i t i e s . Two c l o s e l y r e l a t e d populations with no genetic b a r r i e r s to crossing, d i f f e r i n g only i n t h e i r appearance to bees, may become reproductively i s o l a t e d on e t h o l o g i c a l grounds (Mac i o r , 1971)• I f two populations become i s o l a t e d by p o l l i n a t o r s e l e c t i o n , they may eventually become d i s t i n c t as a r e s u l t of genetic d r i f t , and possibly d i f f e r e n t i a t e into new species i n response to changing s e l e c t i o n pressures. Differences i n flower colouring revealed by UV photography have. proven u s e f u l f o r d i f f e r e n t i a t i o n of c l o s e l y r e l a t e d taxa. Horovitz and Cohen Q-972).have shown that the UV reflectance c h a r a c t e r i s t i c s of Sinapis  alba and S\ arvensis help d i s t i n g u i s h these two c l o s e l y r e l a t e d species. S i m i l a r l y , Ornduff and Mosquin (l970),have shown that c l o s e l y r e l a t e d , but geographically i s o l a t e d , elements.of the Nymphoides indiea complex are d i s -tinguished i n the same manner. - 24 -. Several taxa i n the Limnanthaceae are almost t o t a l l y outcrossing; these plants often have v i s i b l y patterned flowers, nectar glands, and are bee-pollinated. The v a r i e t i e s of L. d o u g l a s i i follow t h i s pattern, and i t i s i n t e r e s t i n g that the most s t r i k i n g differences between these plants r e s u l t from v a r i a t i o n i n f l o r a l colours and patterning. Since ,>fTower petals of Limnanthes were found to contain high concentrations of flavonoids which might produce UV v i s i b l e nectar guides, and because information of p o t e n t i a l systematic importance might be discovered, the flowers of a l l Limnanthes taxa grown from seed were examined by UV photography. I t was hoped that any f l o r a l differences revealed by t h i s technique would complement chemical differences and would be u s e f u l i n reevaluating the systematics of the family. B. Method Flowers of OTUs grown from seed were photographed on colour f i l m , f i r s t with no f i l t e r , and second with a f i l t e r which transmitted UV l i g h t only. A l l photographs were taken with an Exa 35 mm SLR, F 3.5 lens, and extension r i n g s adjusted so that the r e s u l t i n g flower image on the colour transparency was approximately l i f e s i z e . High Speed Ektachrome (Kodak) f i l m , ASA-l60 was used throughout. Photography was done i n the l a b , but only on bright sunny days when a strong natural window l i g h t was present. Preliminary t r i a l s showed that ah exposure time of 1/25.second produced s a t i s f a c t o r y pictures i n e x i s t i n g daylight under these conditions. A f t e r photography' i n natural l i g h t , each flower was rephotographed through a Wjratten No, l8A (Kodak) 2 inch glass f i l t e r under the same conditions as above, but with.an a d d i t i o n a l hand-held UV l i g h t source (366 my*). The 18A f i l t e r i s transparent only to l i g h t with wavelengths l e s s than 400 ny*. Since a regular camera lens w i l l transmit l i g h t over 350 nyi, with the f i l t e r i n place the frequency range of l i g h t reaching the f i l m i s 350-400 ny* - 25 -(Kodak Data Book M-27, 1968). The photographs made by t h i s technique theo-r e t i c a l l y represent the UV component of a bee's v i s i o n . An exposure time of 30 seconds for UV photographs was selected on the basis of preliminary t r i a l s . The films were developed by regular com-mercial processing. For presentation i n t h i s t h e s i s , selected transparencies were copied and printed i n black and white. Before the technique of UV flower photography can produce s a t i s -factory r e s u l t s , some tes t photographs must be made. This i s necessary to determine exposure and focusing d e t a i l s and choice of background. Since the l i g h t r e f r a c t i o n by an ordinary camera lens i s d i f f e r e n t f o r UV than for v i s i -ble, l i g h t , the lens must be refocused before a UV exposure i s made. However, the UV image i s i n v i s i b l e through the camera viewfinder. So, unless a f l u -orescent viewing screen i s a v a i l a b l e , the dgree of lens movement must be pre-determined by t r i a l exposures. For t h i s work a uniform decrease of extension of about 3 mm produced UV exposures that were reasonably w e l l focused. Choice of flower background must also be determined by t r i a l and error. This i s because the UV refl e c t a n c e and absorptive characters of the subject flower must be known before a contrasting background can be chosen. For the UV photography of Limnanthes flowers, a white background gave the best) r e s u l t s , although a black background was also t r i e d . FLAVONOID ISOLATION AND IDENTIFICATION . A. Introduction .A comparative survey of the flavonoids of any group of plants con-s i s t s of two parts: l ) i s o l a t i n g and i d e n t i f y i n g a l l the flavonoids present i n t o t a l , and 2) assessing which compounds occur i n each taxon under study. In p r a c t i c e , however, t h i s technique i s modified s l i g h t l y ; each.taxon i s ana-lyzed i n sequence, rather than a l l taxa at.once. In c l o s e l y r e l a t e d groups - 26 -of plants, each with s i m i l a r flavonoid complements, the i s o l a t i o n and iden-t i f i c a t i o n of the flavonoids of. one taxon y i e l d s a set of standard compounds. By d i r e c t comparison to these standards, i t i s usually possible to i d e n t i f y most of the compounds present i n the remaining taxa. Therefore, only a few ad d i t i o n a l unknown flavonoids need he; i s o l a t e d and i d e n t i f i e d from each new taxon. Once these compounds have been i d e n t i f i e d , they i n turn serve as standards f o r comparison. The family Limnanthaceae consists of r e l a t i v e l y c l o s e l y r e l a t e d taxa. Preliminary observations indicated that the flavonoid complements of these taxa were very s i m i l a r . Therefore, L_. doug l a s i i var. d o u g l a s i i , OTU 1, with i t s large yellow and white patterned flowers was selected f o r i n i t i a l f l avonoid i d e n t i f i c a t i o n s based on the amount of seed a v a i l a b l e and the s i z e and.colour of i t s flowers, rather than the uniqueness of i t s flavonoid pattern. Plants of OTU 1 were grown outside to provide enough fresh material f o r flavonoid i d e n t i f i c a t i o n , and four f l a t s of plants were grown i n the green-house to provide s u f f i c i e n t numbers of flowers for separate analysis of the petals. Since most of the flavonoids present i n remaining taxa were i d e n t i -f i e d by comparison to the compounds of OTU 1, l e s s fresh material was r e -quired f o r each taxon grown subsequently, and the material from a sin g l e f l a t of plants was usually adequate. However, i n c e r t a i n cases i t was necessary to regrow a taxon to i s o l a t e enough of a previously unknown compound(s) for i d e n t i f i c a t i o n . • Unfortunately, a few compounds, always occurred i n trace amounts or were present i n taxa. f o r which no seed was a v a i l a b l e (e_.g_. , Floerkea). I d e n t i f i c a t i o n of these compounds was impossible without reference standards f o r comparison, and.these, were not a v a i l a b l e . However, a f t e r the flavonoids of the f i r s t several taxa were characterized, the flavonoid patterns of the - 27 -remaining taxa often could be completely determined on the basis of two chromatograms, one of the petals and one of whole plant material. In the course of t h i s work, s l i g h t l y d i f f e r e n t procedures were used to i s o l a t e flavonoids appropriate to the type and''amount of material a v a i l -able and the purpose f o r which the material was used. These d i f f e r e n t pro-cedures w i l l be described i n the following sections. B. Plant Extractions Fresh leaf-stem material of f i e l d grown L_. d o u g l a s i i var. d o u g l a s i i , OTU 1, weighing 3.5 kg was repeatedly extracted with hot 90% MeOH. This extract was concentrated i n vacuo at 30°, followed by t r i t u r a t i o n with C e l i t e A n a l y t i c a l F i l t e r A i d to remove chlorophylls. The presence of large quantities S.f carbohydrates and other components necessitated a step to remove these com-ponents before- flavonoid separation was possible. This was accomplished by passing the extract dissolved i n water through a short column of polyamide (SC - 6 ). S u f f i c i e n t adsorbent was used- to hold a l l flavonoids while carbo-hydrates and other non-adsorbing compounds were eluted with water. Subse-quently, the flavonoids were removed from the column by repeated e l u t i o n with 50%, aqueous MeOH u n t i l no traces could be detected under UV l i g h t . The f l a -vonoid containing eluent was then concentrated i n vacuo. The preliminary,cleaning of MeOH extracts by means of a short poly-amide column as described above proved invaluable for separation of flavonoids of Limnanthes, and.this procedure was used as a preliminary step whenever any fresh material was extracted. However, caution had to be exercised while e l u t i n g with water, since highly glycosylated flavonoids, as are-found i n a l l Limnanthes taxa, have Rfs approaching 1.0 under these circumstances. There-fore, to avoid flavonoid lo s s or contamination, c a r e f u l monitoring under UV l i g h t was.required during e l u t i o n . - 28 -Approximately 10,000 petals were c o l l e c t e d from k f l a t s of OTU 1 over a period of flowering of 2-3 weeks. The flavonoids were i s o l a t e d from t h i s material following the technique used f o r leaf-stem m a t e r i a l , except that the t r i t u r a t i o n step with C e l i t e was eliminated. Although the presence of carbohydrates was not as troublesome i n t h i s and other p e t a l extracts, these extracts were eluted from a short polyamide column with 50% MeOH, leaving behind carotenoid pigments which were sometimes present i n high con-centration. Following the work on OTU 1, the i s o l a t i o n procedure was changed s l i g h t l y ; rather than extraction with hot MeOH, fresh plant t i s s u e s were repeatedly extracted with MeOH at room temperature f o r a duration of 2-5 days. Hopefully, t h i s prevented any possible breakdown of highly glycosylated f l a v o -noids on exposure to heat. Although le s s e f f i c i e n t than extraction with hot solvent, acceptable amounts of flavonoids were extracted under the milder conditions. The procedure used to i s o l a t e flavonoids from pressed plant material was simpler than those already outlined. The dried material was extracted with MeOH at room temperature for 2-5 days and then concentrated. No pre-liminary clean-up steps were required, although such steps might have been necessary had larger amounts of material been a v a i l a b l e . In every case where a taxon was analyzed from pressed m a t e r i a l , only about 0.1 g dry weight of plant.was a v a i l a b l e . . C. I d e n t i f i c a t i o n of the Flavonoids of L. dou g l a s i i var. d o u g l a s i i , OTU 1  i . Chromatography Af t e r preliminary cleaning, p e t a l and whole plant extracts were d i s -solved i n small volumes of water and.chromatographed on columns of SC-6 poly-amide using a l i n e a r gradient 0-50% MeOH i n HO. This p r a c t i c e separated - 29 -flavonoids roughly according to g l y c o s y l a t i o n class — biosides from t r i o -sides, etc. Column f r a c t i o n s , containing compounds with a common gl y c o s y l a -t i o n pattern, were consolidated and concentrated in_ vacuo. These f r a c t i o n s were taken up i n a small volume of Q0% MeOH and streaked on TLC plates (20 x 20 cm) using DC 6.6 polyamide (MacNarey-Nagel) as the adsorbent. These plates were run repeatedly i n CHCl^MeOH-butanone-HgO (55:22:20:3). This organic solvent was developed f o r t h i s purpose and tends to separate f l a v o -noids according to t h e i r degree of s u b s t i t u t i o n on the flavonoid nucleus, rather than type of sugar moiety. The r e s u l t i n g bands of pure compounds were marked under UV l i g h t and scraped, together with the polyamide, o f f the glass plates. The compounds were eluted with 80% aqueous MeOH with a vacuum a s p i -rator. The r e s u l t i n g eluents were concentrated, and again taken up i n small volumes of 80% MeOH. Some groups of flavonoids with d i f f e r e n t g l y c o s y l a t i o n patterns, could not be separated completely by aqueous column chromatography. Separation was achieved using an aqueous TLC system — DC 6.6 polyamide with H^O-n-BuOH-acetone (8:1:1). The chloroform-based TLC system was then used to resolve these classes of compounds into i n d i v i d u a l components. The above procedure used to i s o l a t e pure flavonoids was successful f o r most types of flavonoids present i n Limnanthes.. However, compounds with sugars attached at the 3 and 7 positions of the flavonoid nucleus were not separated by TLC i n the above organic solvent system. Unfortunately, e f f o r t s to solve the problem were only p a r t i a l l y s u ccessful; an a c i d i c system using DC 6.6 polyamide and CHCl^-isopropanol-butanone-HOAc (10:3^3|,0~w a s .developed which w i l l resolve the 3 , 7-diglycosides, but only i f they are applied to plates i n small amounts. However, t h i s system made possible the i d e n t i -f i c a t i o n of these compounds, even though they were not i s o l a t e d as i n d i v i d u a l pure compounds. - 30 -The e f f i c i e n c y of compound recovery from the TLC polyamide was v a r i -able. Apparently, e f f i c i e n c y of e l u t i o n i s inversely correlated with degree of hydroxylation of the B-ring of the .flavonoid nucleus — i - i l - > the more hydroxyl groups present on a compound, the lower the e f f i c i e n c y of recovery of that compound. Recovery of myricetin derivatives with t h e i r three B-ring hydroxyl groups was probably l e s s than 50%, and.the recovery of compounds with two hydroxyl groups' on the B-ring, such as quercetin d e r i v a t i v e s , was only s l i g h t l y better. Compounds with only one B-ring hydroxyl group !were recovered more e f f i c i e n t l y . Attempts were made to f i n d a better method for .eluting compounds from TLC polyamide, but no better e l u t i n g solvent than 80% MeOH was discovered. I t i s also worth noting that recovery of flavonoids from polyamide columns was by no means complete. i i . S p ectral Analyses U l t r a v i o l e t absorption spectra were determined f o r a l l compounds i s o l a t e d i n s u f f i c i e n t quantities (ca. l.mg). Standard procedures outlined-by Mabry et a l . (1970) were followed. The technique consists of s p e c t r a l de-termination i n MeOH, and then i n MeOH with various s h i f t reagents added. The data obtained are u s e f u l f o r s t r u c t u r a l determination of flavonoids. UV-spec-t r a of Limnanthes compounds were determined on a Pye-Unicam SP 1800 spectro-photometer. Nuclear magnetic resonance (NMR) spectroscopy also may be u s e f u l f o r i d e n t i f i c a t i o n of flavonoid compounds. However, s a t i s f a c t o r y r e s u l t s using t h i s technique require the i s o l a t i o n of comparatively large amounts of compound (10-20 mg). Six compounds were i s o l a t e d from field-grown material of OTU 1 i n quantities ranging from 7-30 mgs, and.their NMR spectra were determined. To obtain s a t i s f a c t o r y NMR spectra, the s i x Limnanthes flavonoids were c r y s t a l i z e d from 80% MeOH a f t e r i s o l a t i o n by TLC. T r i m e t h y l s i l y l (TMS) - 31 -ethers of the s i x compounds were prepared using " T r i s i l K i t s " (Pierce Chemical Company). NMR spectra of the TMS ethers dissolved i n CCl^ were determined on a Varian HA-100 instrument (100 mHz) using tetramethylsilane as i n t e r n a l standard. TMS ethers were hydrolyzed to y i e l d the o r i g i n a l flavonoids by addition of aqueous MeOH to the CCl^ solutions. The r e s u l t i n g solutions of l i b e r a t e d flavonoid compounds were then concentrated i n vacuo. i i i . Hydrolyses The hydrolysis of flavonoid glycosides was accomplished by several d i f f e r e n t procedures. A l l hydrolyses were performed to determine either the number and type of sugar moieties attached to a flavonoid nucleus, or, when a compound was glycosylated with more.than one sugar, the p o s i t i o n and order of attachment of the : sugars. T o t a l hydrolyses were accomplished with 1.0 N HC1 i n water f or 30 minutes at 95-100°. Liberated sugars were i s o l a t e d , neu-t r a l i z e d and i d e n t i f i e d by TLC following a procedure s i m i l a r to that described by Mabry et al.- (1970), Relative amounts of the i d e n t i f i e d sugars were also estimated using t h i s technique. The l i b e r a t e d aglycones were i d e n t i f i e d by co-chromatographyC^^ . When a glycoside contains more than one su g a r , . p a r t i a l hydrolysis may a s s i s t i n it s - i d e n t i f i c a t i o n by' comparing the r e s u l t i n g d e rivatives with standard.compounds. By t h i s process both the order and p o s i t i o n of sugar attachment may sometimes be determined. Limnanthes. compounds were p a r t i a l l y hydrolyzed by two methods, depending on the mode of g l y c o s y l a t i o n : hydrolysis for short i n t e r v a l s at 95-100°. with either 0.1 N HC1 or 20% (v/v) acetic a c i d . Enzymic hydrolysis with emulsin proved a us e f u l technique f o r the char a c t e r i z a t i o n of many of-the Limnanthes compounds. These hydrolyses were accomplished by mixing a small amount of enzymic mixture ( N u t r i t i o n a l B i o r chemical Corp.) into. 1 . 0 N acetate.buffer (pH 5.1) containing dissolved - 32 -compound and allowing the suspension to stand overnight at room temperature (Mabry et a l . , 1970). In t h i s manner glucose was s e l e c t i v e l y removed by the p-glucosidase contained i n the emulsin, when glucose was the only s u b s t i -tuent at a given s i t e . (Glucose i s also removed i f i t i s the terminal sub-s t i t u e n t of a glycoside moiety; however, enzymic hydrolyses were not performed on compounds with t h i s mode of g l y c o s y l a t i o n in the present i n v e s t i g a t i o n . ) i v . Acetate Preparation Two previously unreported compounds, each based on a rare aglycone type, were i s o l a t e d i n comparatively large quantities from L_. d o u g l a s i i var. d o u g l a s i i . To characterize these compounds more completely, t h e i r acetate derivatives were prepared, and t h e i r melting points determined. P r i o r to d e r i v a t i z a t i o n of these two compounds, an a d d i t i o n a l acetate d e r i v a t i v e of a t h i r d Limnanthes compound possessing the same g l y c o s y l a t i o n pattern was pre-pared to te s t the planned procedure and to compare.the melting point of t h i s t e s t d e r i v a t i v e with the published value. The j|iKo:'.3iiedure used to prepare acetate derivatives was as follows: 1 ml of ac e t i c anhydride was added to 5 mg of compound i n a 25 ml round bottom f l a s k ; a few drops of triethylamine were added, and the s o l u t i o n was l e f t overnight at room temperature; t h i s s o l u t i o n was then concentrated, taken up i n chloroform and the d e r i v a t i v e p r e c i p i t a t e d by the addition of EtOH. The p r e c i p i t a t e d d e r i v a t i v e was removed from s o l u t i o n by f i l t r a t i o n , dried and c r y s t a l i z e d from a mixture of benzene and cyclohexane. By following t h i s process, the amount of c r y s t a l i n e d e r i v a t i v e recovered was JO-80% i n each case. These acetate derivatives were not substituted at a l l free hydroxyl groups of the flavonoid nucleus since the 5-hydroxyl group remains unacetylated due to hydrogen bonding. - 33 -D. Ide n t i f i c a t i o n - of the Remaining Limnanthes Flavonoids When compounds were encountered which did not occur i n L_. d o u g l a s i i , they were i s o l a t e d and i d e n t i f i e d by the processes already described, i f they were present i n s u f f i c i e n t concentration.. Compounds c o n s i s t e n t l y occurring i n trace amounts remain u n i d e n t i f i e d . However, the presence of these compounds was s t i l l recorded, and they were included with the known compounds f o r ana l y s i s . The flavonoid complement of each taxon, as i t was determined for comparative purposes, was taken to be a l l the flavonoids that could, be resolved by two-dimensional TLC of whole plant and p e t a l extracts. The TLC system used was polyamide DC 6.6, 1 s t dimension — H^O-n-BuOH-acetone-HOAc -(16:2:1:1), 2nd dimension — CHCl^-isopropanol-butanone-HOAc - (10:3:3:4). Some plates were run twice i n the second dimension. A f t e r drying thoroughly:;; the two-dimensional maps were sprayed with a boronate reagent. This reagent i s a modification of that described by' Somaroo et_ a l . (1973) and gr e a t l y increases the v i s i b i l i t y of flavonoids on polyamide under UV l i g h t . I t was prepared by d i s s o l v i n g a small amount of diphenyl b o r i c a c i d ethanolamine complex (Aldrich) ifr MeOH, and then d i l u t i n g with water to 0.1% (w/v). The compounds resolved by t h i s process were i d e n t i f i e d whenever possible by com-parison to standard compounds using the technique of co-chromatography, i f necessary. ' EXPRESSION OF FLAVONOIDS AS TAXONOMIC CHARACTERS Flavonoids are.often used as characters i n taxonomic-studies by simply comparing presence or absence i n the taxa under study. In p r a c t i c e , however, the r e l a t i v e concentrations of occurrence are.often also'considered. In the present i n v e s t i g a t i o n , taxa were compared on the basis of 4 6 flavonoid characters.- Two d i f f e r e n t l e v e l s of compound concentration on comparative two-dimensional chromatograms were recognized and recorded: l ) present i n r e l a t i v e l y high concentration and v i s i b l e before spraying with boronate r e -agent , (several orders of magnitude of concentration are represented i n t h i s f i r s t c l a s s i f i c a t i o n ) and 2) present i n trace amount and v i s i b l e only a f t e r spraying. By t h i s process each OTU was described on the basis of flavonoids present i n whole plant material and also on the basis of p e t a l material when i t was a v a i l a b l e . One advantage of using a technique which takes into account concen-t r a t i o n s of flavonoid /occurrences, i s that a better approximation can be made of the r e s u l t s as they a c t u a l l y occur on a chromatogram. Since the r e l a t i v e concentrations of flavonoids seem to be consistent f o r a given taxon, a pro-cedure which considers concentration data probably conserves more taxonomically s i g n i f i c a n t information than one that does not. TREATMENT OF THE FLAVONOID DATA  A. Introduction Small groups of taxa can be compared onfcthe basis of t h e i r flavonoids by constructing a matrix of OTUs versus flavonoid characters. The order of OTUs and/or characters can be v i s u a l l y arranged so that taxa with the greatest character s i m i l a r i t i e s c l u s t e r together i n the matrix. As the number of OTUs and/or characters increases, the technique of manually rearranging OTUs into c l u s t e r s becomes more d i f f i c u l t and.subjective. This i s because the eye can only consider so much information at once, and when confronted with excessive data, tends to concentrate on a small portion of the data. This process leads to r e s u l t s prejudiced i n favor of the portion of the data considered. For t h i s reason, phenetic comparisons of large numbers of OTUs on the basis of many characters are best made using the techniques of numerical taxonomy. T h e o r e t i c a l l y , the' use of these techniques eliminates the above prejudice. - 35 -A most important t o o l of the numerical taxonomist i s c l u s t e r a n a l y s i s , a technique recently made p r a c t i c a b l e by the development of the computer. There are many va r i a t i o n s of computerized c l u s t e r a n a l y s i s , but e s s e n t i a l l y a l l of these techniques are designed to consider character s i m i l a r i t i e s between a l l taxa simultaneously and produce groupings or clu s t e r s of taxa based on r e l a t i v e s i m i l a r i t i e s which provide the best " f i t " to the data. It was desirable to use numberical taxonomic techniques f o r analysis of the flavonoid data of the Limnanthaceae for two reasons: l ) a data matrix containing 30 OTUs described by 46 characters exceeds the s i z e f or which an accurate assessment of s i m i l a r i t y can be made by eye; and 2) c l u s t e r analy-s i s based on.flavonoid characters allows a d i r e c t comparison to the e a r l i e r numerical taxonomic analysis of the Limnanthaceae by Ornduff and Crovello based on morphological data (1968). B, Method To c l u s t e r taxa by conventional methods, a matrix of s i m i l a r i t y coef-f i c i e n t s must be calculated between a l l taxa. Since there are many d i f f e r e n t ways to ca l c u l a t e s i m i l a r i t y c o e f f i c i e n t s , . t h e r e s u l t s of c l u s t e r analysis may vary with the. type - of c o e f f i c i e n t used. Whether or not two taxa.are made more s i m i l a r due to the'mutual absence of a given flavonoid character (nega-t i v e matches' are considered),•is a question that has not been s a t i s f a c t o r i l y answered. • To help answer t h i s question, c l u s t e r analysed of Limnanthes and Floerkea were performed using two types of s i m i l a r i t y c o e f f i c i e n t s : l ) the simple matching c o e f f i c i e n t (Sokal & Sneath, 1963, p. 133) which considers ' negative matches'; and 2) the c o e f f i c i e n t of Jaccard (Sokal & Sneath, 1963, p. 133)' which calculates s i m i l a r i t y only on the basis of mutual occurrence. Flayonoid data were compiled into two Basic Data.Matrices l i s t i n g , i n one case, a l l OTUs (30) as rows versus the flavonoid characters (46) found - 36 -found i n whole plant m a t e r i a l , and i n the other case the OTUs grown fresh ( l8) versus the flavonoid characters (46) found i n the f r e s h p e t a l s . From these two matrices the simple matching and Jaccard s i m i l a r i t y c o e f f i c i e n t s were calculated by computer between a l l taxa. These c o e f f i c i e n t s were ar-ranged i n h a l f matrices of OTUs, either 30 x 30 (whole plant data) or 18 x 18 (petal data). In each instance duplicate matrices were c a l c u l a t e d handling the data.in.two ways: l ) s i m i l a r i t i e s were determined considering only those compounds occurring i n r e l a t i v e l y high concentrations; 2) s i m i l a r i t i e s were determined on the basis of any.occurrence regardless of concentration. By t h i s process a t o t a l of eight matrices of s i m i l a r i t y c o e f f i c i e n t s were computed. From t h i s data, eight d i f f e r e n t c l u s t e r analyses were computed' using the weighted p a i r group c l u s t e r i n g method, and the r e s u l t s were expressed i n eight dendrograms (phenograms). In t h e i r taximetric study of the Limnanthaceae, Ornduff and Crovello (.1968) used two techniques to cluster' OTUs. The f i r s t was a conventional method, the weighted p a i r group method, that producec&a dendrogram expressing o v e r a l l cophenetic s i m i l a r i t i e s between OTUs. The second method was a form of fa c t o r analysis which produced a three-dimensional p l o t of OTUs i n "reduced character' space". Distances between OTUs pl o t t e d by the second technique are proportional to major sources of v a r i a t i o n between taxa. Although the com-puter program used by Ornduff and Crovello was not a v a i l a b l e f o r the present i n v e s t i g a t i o n , two other standard types of factor analysis w e r f e . P e r f o r m e < i °n flavonoid data. The two techniques, P r i n c i p a l Components Analysis and Varimax Factor Analysis (Both.programs are from the S c i e n t i f i c Subroutine Package of IBM.), w i l l be b r i e f l y described here since the method d i f f e r e d from standard c l u s -t e r i n g programs and may not neces s a r i l y express o v e r a l l cophenetic s i m i l a r i t i e s . Instead these techniques are based on key components) of the t o t a l v a r i a t i o n - 37 -present between taxa. The f i r s t step i n performing any factor analysis i s to set up a character-by-character c o r r e l a t i o n c o e f f i c i e n t matrix. Then, o v e r a l l c l u s -t e r i n g of characters (not OTUs as by conventional c l u s t e r i n g programs) i s computed. From the n dimensions of hyperspace, corresponding to the number of characters considered, the r e l a t i v e amount of v a r i a b i l i t y expressed by each dimension i s factored. The three ( i n t h i s case) factors or vectors which together account for the greatest amount of v a r i a t i o n between taxa are. selected. These factors are not correlated. The " f a c t o r loadings" f o r each character are calculated'^ i..e_. , the. r e l a t i v e c ontribution that each character makes to the v a r i a t i o n expressed by each of the three selected vectors. Then, the character complement of each.OTU i s m u l t i p l i e d by the respective factor loadings, producing three coordinates f o r each OTU. These coordinates are.used to p l o t OTUs i n three dimensional space using a standard type of p l o t t i n g program which p i c t o r i a l i z e s the three-dimensional plot i n two dimensions. The type of analysis described above i s f a c t o r analysis by p r i n -c i p a l components. The other technique used i n t h i s study was Varimax Factor Analysis with r o t a t i o n . The preliminary steps are the same f o r t h i s technique. A f t e r the factors are extracted, they are transformed by r o t a t i o n t o d i f f e r e n t coordinates which t h e o r e t i c a l l y show i n t e r r e l a t i o n s h i p s between OTUs i n their--simplest form. From these adjusted f a c t o r s , new factor' loadings are calcu-l a t e d , and the OTUs are p l o t t e d i n three'dimensions as aboye. Since preliminary results, i n d i c a t e d that the c l u s t e r analyses.were more taxonomically meaningful i f only the' compounds occurring i n r e l a t i v e l y , heavy concentration were used as characters, the two types of f a c t o r analysis were performed only on the basis of flavonoids v i s i b l e before spraying the chromatograms. 'Eighteen OTUs were analyzed on the basis of 36 v a r i a b l e - 38 -flavonoids occurring i n the petals (the remaining 10 flavonoids did not vary under these conditions), while 30 OTUs were analyzed on the basis of 31 v a r i -able flavonoids present i n whole plant material. - 39 -RESULTS FLORAL PATTERNING IN ULTRA VIOLET LIGHT Figure II i s a cbmposicb.e plate made up of selected photograph pa i r s of Limnanthes flowers taken i n v i s i b l e and UV l i g h t . The taxa represented are the v a r i e t i e s of L_. d o u g l a s i i , OTUs 2,3,4 and 5, L.. macounii, OTU 10, and L. g r a c i l i s var. g r a c i l i s , OTU 14. Of these s i x taxa, the f i r s t f i v e exhibit some form of UV patterning while the l a s t (Figure H E ) i s included i n the plate to i l l u s t r a t e the pattern of uniform absorption observed f o r the remaining twelve unpictured taxa. The photograph p a i r of L_. d o u g l a s i i var. rosea (Figure IIB) shows the most s t r i k i n g r e s u l t revealed by the flower photography. Variety rosea has white petals with rose coloured veins and anthers. Under UV l i g h t the outer portions of the' petals of t h i s v a r i e t y r e f l e c t the l i g h t i n sharp con-t r a s t to the strongly absorbing inner p e t a l zones, thus producing a bullseye e f f e c t . This nectar guide v i s i b l e only under UV l i g h t i s analogous to the familiar^, yellow and white pattern of v a r i e t y d o u g l a s i i (Figure IIC). The photographs of L. d o u g l a s i i var. nivea (Figure IID) show some s i m i l a r i t i e s to those of v a r i e t y rosea. Variety nivea, which also has white flowers under v i s i b l e . l i g h t , also shows the bullseye e f f e c t under UV l i g h t , but with le s s contrast than v a r i e t y rosea. Also contributing to the UV pat-terning e f f e c t of v a r i e t y nivea are the l i g h t coloured regions at the p e t a l bases',and the l i g h t coloured anthers, both contrasting with the dark flower center. The pale appearance of the p e t a l bases i s due i n part.to the r e f l e c -tance of the basal h a i r s . A s i m i l a r r e f l e c t a n c e i s also evident for v a r i e t y rosea, and somewhat evident f o r v a r i e t y sulphurea (Figure IIA). R e f l e c t i n g • 40a -FIGURE II ~- The 'Flowers of Six Selected Limnanthes taxa Taken i n V i s i b l e and U l t r a V i o l e t Light. The upper pictures of the photograph pa i r s are taken i n v i s i b l e l i g h t . The lower pictures are taken through the l8A f i l t e r . A) L. d o u g l a s i i var. sulphurea, OTU 5; B) L. d o u g l a s i i var. rosea, OTU 4; C) L. d o u g l a s i i var. d o u g l a s i i , OTU 2; D.), L. d o u g l a s i i var. nivea, OTU 3; E) L. g r a c i l i s var. g r a c i l i s , OTU 14; and F) L. macounii, OTU 10. ! - 41 -hairs at the p e t a l hases were occasionally observed for other unpictured taxa as w e l l , including L_. alba var. v e r s i c o l o r . Both L. d o u g l a s i i vars. sulphurea (Figure HA) and d o u g l a s i i (Figure IIC) have some yellow flower pigmentation i n v i s i b l e l i g h t . When photographed under UV l i g h t , neither.^'bfi.'ffi'ese v a r i e t i e s has sharply contrasting zones, but each exhibits a c e r t a i n amount of patterning. Both have petals divided into regions of d i f f e r i n g degrees of UV absorption, with the darker zones toward the flower centers. The v a r i e t y sulphurea also exhibits c e r t a i n features i n addition to the p l a i n pattern of v a r i e t y d o u g l a s i i ; the p e t a l veins of v a r i -ety sulphurea appear under UV l i g h t as contrasting r a d i a l lines,, absorbing les s UV l i g h t than the inner zone but more than the outer, l i g h t e r p e t a l parts.. The appearance under UV l i g h t of the small p l a i n flowers of L_. macounii (Figure ITF) was s u r p r i s i n g . Its p e t a l bases r e f l e c t to some extent i n con-t r a s t to the outer p e t a l regions , the p e t a l veins, and the heavily absorbing sepals. Although there are.no hairs a,t the p e t a l bases and no l i g h t e r coloured zones at the p e t a l t i p s of L_. macounii, the c e n t r a l zone of UV r e -flectance and the contrasting veins show s i m i l a r i t i e s to the patterns of L. d o u g l a s i i vars. nivea and rosea. Flowers appearing white i n the v i s i b l e spectrum, .but which absorb UV Tight,- are-common and include those of most taxa of Limnanthes. The net e f f e c t to' a bee looking at t h i s type of flower i s a colour that contrasts with the background vegetation. Presumably, t h i s condition has evolved i n many l i n e s ' o f plants r e s u l t i n g from s e l e c t i o n pressures to increase the proba-b i l i t y , of a v i s i t by a p o l l i n a t o r and.thus enhance cross f e r t i l i z a t i o n . The elaboration of nectar guides, which function to draw the insect to. the center of a flower, i s a further adaptation to f a c i l i t a t e cross p o l l i n a t i o n brought about by.the same s e l e c t i o n pressures. - 42--The presence of UV-visible nectar guides i n the v a r i e t i e s of L_. d o u g l a s i i suggests that these taxa are more highly evolved outcrossers than are other Limnanthes taxa. Of these v a r i e t i e s , rosea, with i t s sharply contrasting zones, i s the most highly evolved of the four. Extrapolating backwards, i t i s probable that the outcrossing Limananthes taxa with p l a i n unpatterned flowers most c l o s e l y resemble the a n c e s t r a l form of t h i s group, and probably the genus as w e l l . The observed differences i n f l o r a l patterning of the four v a r i e t i e s of L. d o u g l a s i i suggest an important implication regarding distance of r e l a -t i o n s h i p i n t h i s species. Various sympatric taxa of Limnanthes have evolved genetic b a r r i e r s to h y b r i d i z a t i o n which were used by Mason (1952) to d i s t i n -guish species. Since Mason found the four v a r i e t i e s of L_. d o u g l a s i i experi-mentally i n t e r f e r t i l e , he grouped them together as v a r i e t i e s of one poly-morphic species i n s p i t e of the fact that e a r l i e r authors had described some of them as d i s t i n c t species (Bentham, 1848; Loudon, 1855; Greene, 1891; Abrams, 1941). Although Mason stated that these f o u r v a r i e t i e s r e a d i l y produce f e r -t i l e natural hybrids, the only s u b s t a n t i a l evidence he c i t e d i n support of t h i s statement wlasibthe^intermingling of characters between two of the v a r i e t i e s , rosea and nivea, i n an area of sympatry. However, the mixing of characters between these two v a r i e t i e s appears to be the extent of intermingling'within the species. The other v a r i e t i e s p e r s i s t as d i s c r e t e taxonomic e n t i t i e s , even i n zones of sympatry. I t i s possible that the general lack of i n t e r -mingling among the four v a r i e t i e s i s a t t r i b u t a b l e to p o l l i n a t o r s p e c i f i c i t y r e s u l t i n g from t h e i r d i f f e r e n t l y appearing flowers. This s p e c i f i c i t y may create an e c o l o g i c a l b a r r i e r to genetic exchange, causing the i s o l a t i o n of these taxa, even though no b a r r i e r s to h y b r i d i z a t i o n with a genetic basis have evolved. Therefore, the absence of a genetic b a r r i e r does not neces-s a r i l y i n d i c a t e that a l l . o f these four taxa are v a r i e t i e s of one species.-- k3 -Instead, some of them may be as d i s t a n t l y r e l a t e d as other,.isolated, recog-nized species of Limnanthes,. - The presence of clearcut morphological d i f f e r -ences between them tends to support t h i s p o s s i b i l i t y . The presence of UV-visible flower patterns i n only some taxa of Limnanthes has other implications concerning evolution i n ' t h i s genus.. The common occurrence of such patterning • i n L_. macounii and the four v a r i e t i e s of L_. d o u g l a s i i suggests that these f i v e taxa have evolved from a common ancestor with s i m i l a r flower patterning. I f t h i s i s true, these taxa c o n s t i -tute a natural supraspecific group distinguished from the rest of the genus by t h i s t r a i t . Ornduff (1969% Ornduff & Crovello, 1968) has indicated that autogamy i s a derived condition i n Limnanthes.• The presence of a UV-visible f l o r a l pattern i n L. macounii with no obvious s e l e c t i v e advantage to t h i s s e l f -p o l l i n a t i n g species' supports t h i s conclusion. This pattern i s probably a remnant of an adaptation valuable only to an outcrossing plant. There have been several changes of Slower morphology accompanying autogamy in- L. macounii, such as reduction of s i z e and number of parts. However, the presence of UV-visible nectar guides i n t h i s taxon suggests that not a l l t r a i t s associated with the outcrossing habit have been l o s t . The retention of such t r a i t s f m a y j , h e explained by the p r i o r establishment of linkage groups caused by s e l e c t i o n pressures to preserve outcrossing. Like L_. macounii, some of the subspecies of L. f l o c c o s a are t o t a l l y autogamous ( K a l i n , 1973a, 1973b). However, trends to autogamy have occurred independently i n the two species, since each belongs to a d i f f e r e n t section of the genus. The flowers of L_. f l o c c o s a , a member of the section Inflexae, exhibit UV absorption c h a r a c t e r i s t i c s i d e n t i c a l to outcrossing members of the section; i / e / j flowers with uniformly high UV absorption as shown by the photograph of L. g r a c i l i s (Figure H E ) . Although nectar guides v i s i b l e i n UV - hk -l i g h t have not evolved i n t h i s s e c t i o n , the observed flower characters s t i l l undoubtedly represent an adaptation to the outcrossing habit. Therefore, a conservation of flower pigmentation has operated i n L. f l o c c o s a p a r a l l e l to that described above for L. macounii. R e l a t i v e l y high concentrations of flavonoids with.absorption maxima i n the near UV were extracted from petals•of a l l Limnanthes taxa. This pre-sence suggests .that the absorption recorded by UV photography r e s u l t s d i r e c t l y from these compounds. I f correct, t h i s contention agrees with the conclu-sions of other recent investigations designed to determine the chemical nature of UV-visible nectar guides (Thompson et a l . , 1972; Dement & Raven, 197*0. Assumming that the flower flavonoids do cause the observed absorption exhibited by flowers of Limnanthes taxa, then these flavonoids are adaptations which confer a . s e l e c t i v e advantage to outcrossing plants. Since the UV absorption c h a r a c t e r i s t i c s of autogamous members of Limnanthes do not d i f f e r s i g n i f i -cantly from the outcrossing members, an evolutionary conservation has been working on the flower flavonoids of a l l tested members of t h i s genus. The' argument that the flower flavonoids of Limnanthes are adapta-tions for cross f e r t i l i z a t i o n , and.that these compounds w i l l tend to be con-served by evolution, has profound implications regarding the contribution flavonoid d i s t r i b u t i o n may make to the taxonomy"of t h i s group of plants. These implications w i l l be discussed f u l l y i n the following chapter. FLAVONOID CHARACTERIZATION AND IDENTIFICATION  A. Data Presentation Flavonoid i d e n t i f i c a t i o n s depend on the synthesis of many pieces of information gathered from a v a r i e t y of techniques. In the f i r s t part of t h i s section the data gathered for Limnanthes and Floerkea. flavonoids w i l l be tabu-l a t e d . In the second part of the section,- an explanation w i l l be presented - 45 -of how the compounds were i d e n t i f i e d . A t o t a l of 48 flavonoid glycosides were found i n taxa of Limnanthes and Floerkea. Table III l i s t s notations assigned to these flavonoids, t h e i r i d e n t i f i c a t i o n s , and the abbreviations used i n following tab l e s . Where an i d e n t i f i c a t i o n was not p o s s i b l e , the most reasonable guess i s provided. Flavonoid aglycones were present occasionally i n some taxa, including those analyzed from pressed material. However, the appearance of aglycones was e r r a t i c , and when present they occurred only i n low concentration. These observations suggest that the appearance of aglycones was caused by a low l e v e l of h y d r o l y t i c a c t i v i t y , either i n extracts, or i n pressed material. For t h i s reason, i t was assummed that, flavonoid aglycones do not n a t u r a l l y occur i n the Limnanthaceae. Figure I I I i s a composite drawing showing the approximate r e l a t i v e Rfs (20.x 20. cm TLC glass plates of Polyamide DC 6.6; 1st dimension — H^O-n-BuOH-acetone-HOAc - l6:2:1:1, 2nd dimension — CHCl^-isopropanol-butanone-HOAc - 10:3:3:4) of a l l flavonoid glycosides occurring i n the Lim-nanthaceae. Although t h i s drawing i s a f u l l s i z e d reproduction of a 20 x 20 cm two-dimensional map, i t i s not meant to represent the r e s u l t s of any actual chromatogram. Each of the flavonoids i n the drawing i s labeled with the number or l e t t e r randomly assigned- during the course of t h i s work. Table IV l i s t s Rfs of the 48 .flavonoids as they appear i n Figure I I I . These Rfs are approximations only, since flavonoid Rfs on polyamide. are v a r i -able, being c o n t r o l l e d by many f a c t o r s ; however, the r e l a t i v e positions of these compounds are consistent. Table IV also l i s t s colour c h a r a c t e r i s t i c s of most flavonoids as they appear on polyamide chromatograms under UV l i g h t (366 mji) , with and without NH^ vapour, and also a f t e r spraying with boronate reagent. Colour data are not l i s t e d f o r compounds that occurred only i n trace amounts, since no judgement could be made beyond presence or absence. - he -TABLE I I I I d e n t i f i c a t i o n s of the Flavonoids of the Limnanthaceae Symbol Compound Abbreviation H Unknown kaempferol monoside; perhaps acylated D Probably the 7-0-glucoside of kaempferol E Probably the 7-0-glucoside of quercetin 0 Syringetin-3-0-glucoside Sg-•3-•0--G P . Isorhamnetin-3-0-glucoside I r -•3-•0--G Q Kaempferol-3-0-glucoside• Kp-•3-•0--G R Laricytrin-3-0-glucoside Lc-•3-•0--G S Quercetin-3-0-glucoside Qu-•3-•0--G T Myricetin-3-0-glucoside My-•3-•0--G C Unknown.3-0-monoside of quercetin A Probably the 7-0-rutinoside of kaempferol B Probably the 7-0-rutinoside of quercetin 19 Unknown 3-0-monoside or bio s i d e of s y r i n g e t i n 20, . Unknown 3-0-monoside or bio s i d e of isorhamnetin 21 Unknown' 3-0-monoside or bioside of kaempferol 22. Unknown 3-0-monoside or bioside of l a r i c y t r i n 23 Unknown 3-0-monoside or bioside of. quercetin U Syringetin-3-0-rutinoside Sg-•3-•0--Rut V Isorhamnetin-3-0-rutinosIde ' I r -•3-•0--Rut W Kaempferol-3-0-rutinoside Kp-•3-•0--Rut X La r i c y t r i n - 3 - 0 - r u t i n o s i d e Lc-•3-•0--Rut Y Quercetin-3-0-rutinoside Qu-•3-• 0-Rut Z Myricetin-3-0-rutinoside My-•3--0--Rut I • Syringetin-3-0-diglucoside Sg-•3--0--GG - kl -TABLE III — Continued Symbol Compound Abbreviation J Isorhamnetin - 3 - 0-diglucoside Ir-3-0-GG K Kaempferol - 3 - 0-diglucoside Kp-3-0-GG L L a r i c y t r i h - 3 - 0 - d i g l u c o s i d e Lc-3-O-GG M Quercetin - 3 - 0-diglucoside • Qu-3-O-GG N Myricetin - 3 - 0-diglucoside My-3-0-GG F A bioside of kaempferol; probably kaempferol -3 -•0-glucoside ,7 -0--g lucoside G A bioside of quercetin; probably quercetin 3 - 0 -•glucoside , 7 - 0-glucoside 1 Syringetin - 3 - 0-rhamnosylrutinoside Sg-3-0-GRR 2 Isorhamnetin -3 - 0-rhamnosylrutinoside Ir-3-0-GRR 3 Kaempf erol - 3 - 0-rhamnosylrutinoside •' Kp-3-0-GRR k Laricytrin - 3 - 0-rhamnosylrutinoside Lc-3-O-GRR 5 Quercetin -3 - 0-rhamnosylrutinoside Qu-3^0-GRR 6 Myric etin -3 - 0-rhamnosylrut inos ide My-3-0-GRR 7 • S y r i n g e t i n - 3 - 0 - r u t inos ide,7 - 0-glucos ide • Sg -3 - 0-Rut ,7 --0--G 8 Isorhamnetin - 3 - 0-rutinoside,7 - 0-glucoside I r - 3 - 0 - R u t , 7 - -0--G 9 • Kaempferol -3 - 0-rut inos ide,7 - 0-glucos ide Kp -3 - 0-Rut ,7 --0--G 10 . L a r i c y t r i n - 3 - 0 - r u t i n o s i d e , ' 7 - 0 - g l u c o s i d e Lc - 3 - 0-Rut , 7 --0--G 11 Quercetin - 3 - 0-rutinoside , 7 - 0-glucoside Qu-3-0-Rut,7- •0--G 12 Myricetin - 3 - 0-rutinoside , 7 - 0-glucoside My-3-0-Rut,7- •0--G -1 3 Syr ingetin - 3-O-rut inos ide, 7-0^-rut inos ide Sg -3 - 0-Rut ,7 -•0--Rut Ik Isorhamnetin -3 -0-ru'tinoside , 7 - 0-rutinoside Ir - 3 - 0-Rut , 7 -•0--Rut 15 . Kaempferol - 3 - 0-rutinoside , 7 - 0-rutinoside Kp-3-0-Rut,7-•0--Rut 16 L a r i c y t r i n - 3 - 0 - r u t i n o s i d e , 7 - 0 - r u t i n o s i d e Lc - 3 - 0-Rut ,7 -•0--Rut . 17 Quercetin - 3 - 0-rutinoside , 7 - 0-rutinoside Qu-3-0-Rut,7- •0--Rut - 48 -FIGURE III Composite Two-Dimensional Map^of 48 2 A Flavonoid Glycosides- of the Limnanthaceae © © '20) © •22s' ( x \ — • R > A „ ' '23 ©1-°© 12 1 — 1st dimension (horizontal) — . HgO-n-BuOH-acetone-HOAc ( l 6 : 2 : l : l ) 2nd dimension (v e r t i c a l ) — CHCl^'Vbutanone-isopropanol-HOAc (10:3:3:4) 2 — S o l i d . c i r c l e s denote i d e n t i f i e d compounds; broken c i r c l e s denote compounds present i n quantities i n s u f f i c i e n t f o r complete i d e n t i f i c a t i o n . - 49 -TABLE IV 1 2 Rfs and Colours of Flavonoids of the Limnanthaceae Compound Rf - 1 Rf - 2 UV UV/NH^ UV'/Spray-3 H 0.08 0.10 P P G D 0.14 0.15 - Id E 0.15 0.07 - Id 0 Sg-3-0-G 0.34 0. 54 P PY YG P Ir -3 -0-G 0.24 0.43 . P G G Q Kp-3-0-G 0.20 0.31 P P G R Lc-3-O-G 0.24 0.18 Br Y Ro S Qu-3-0-G 0.18 0.14 Br Y Or T My-3-0-G 0.18 0.06 Br Y Ro C 0.22 0.11 - - Id A 0.31 0.18 Y .' Y G B 0.30 0.06 Y Y Or • 19 0.46 0. 54 - - Id 20 . o.4i 0.44 - - • Id 21 0.36 0.34 • - - Id 22 0.41 • 0.23 ' - - Id 23 0.38 0.18 - - Id U Sg-3-0-Rut ' 0.54 0. 54 P PY YG V Ir-3-0-Rut 0.50 0.44 P G G W Kp-3-0-Rut 0.43 0.34 . P P G X Lc-3-0-Rut 0.50 . 0.24 Br Y Ro Y Qu-3-0-Rut . ' 0.46 ' 0.18 • Br Y Or Z • My-3-0-Rut 0.46 0.08 Br Y Ro I Sg-3-0-GC- 0.55 0.38 P PY YG J . Ir-3-O-GG 0.55 0.31 P G G K Kp-3-0-GG 0. 50 0.28 P P G L Lc-3-0-GG 0.55 0.19 - - Id M Qu-3-0-GG 0.52 0.14 Br Y Or N My-3-0-GG 0.51 0.'Q6 • - - Id. F 0.58 • 0.23 • - - Id G O.58 0.10 - - Id 1 Sg-3-0-GRR O.69 o.4o P PY YG 2 Ir-3-0-GRR .0. 70 . 0.31 . P • G G - 50 -TABLE IV — Continued Compound Rf - 1 Rf - 2 UV UV/NH^ UV/Spray 3 Kp-3-0-GRR 0.62 0.23 P P G 4 Lc-3-O-GRR 0.68 0.18 P Y Ro 5 Qu-3-0-GRR 0.63 0.13 P Y Or 6 My-3-0-GRR 0 .65 0.08 P Y Ro 7 Sg-3-0-Rut,7-0-G 0.83 0.38 P PY YG 8 Ir-3-0-Rut ,7-0-G 0.80 0.32 . P Y G 9 Kp-3-0-Rut,7-0-G 0.77 0.24 P Y G 10 . Lc-3-0-Rut ,7-0-G 0.82 0.14 Br Y Rd 11 Qu-3-0-Rut,7-0-G 0,. 77 ' 0.11 Br Y Or 12 . My-3-0-Rut,7-0-G 0.78- 0.03 Br Y Rd 13 Sg-3-0-Rut,7-0-Rut 0.90 0.38 P PY YG 14 Ir-3-0-Rut ,7-0-Rut 0.87 0.31 P Y G 15 Kp-3-0-Rut,7-0-Rut 0.84 0.24 P Y G 16 Lc - 3-0-Rut., 7- 0-Rut 0.88 0.15 Br Y Rd 17 Qu-3-0-Rut,7-0-Rut 0.86 0.10 Br Y Or 1 — Rf - 1 =. 1 development, Polyamide DC6.6, H20--n-BuOH-•acetone-HOAc ( l 6 : 2 : l : l ) Rf - 2 = Xldevelopment, Polyamide (10:3:3:4) DC6.6, CHCl^-butanone-isopropanol-HOAc 2 — Colour explanation — - P8= purple; Br = brown; Y = yellow; PY = pale yellow; G = green;' Or = oran£ ;e; Ro = rose; Rd. = red; YG = yellow green; Id = limit;.,of detection - Spray = 0.1% (w/v) diphenyl boric a c i d ethanolamine complex i n HcO & MeOH - 51 -Table V l i s t s the products obtained upon t o t a l hydrolysis with HC1, including flavonoid aglycone, sugar(s), and r e l a t i v e concentration of each sugar where appropriate. Table VI l i s t s the derivatives of p a r t i a l a c i d hy-dr o l y s i s and the derivatives of enzymic hydrolysis with emulsin (^-glucosidase) where these procedures were employed. Table VII l i s t s UV absorption maxima of the flavonoid glycosides that were i s o l a t e d i n s u f f i c i e n t quantities f o r determination. Table VII l i s t s maxima i n MeOH and also i n MeOH following the addition of various s h i f t reagents. Table VIII i s a tabulat i o n of the r e s u l t s of NMR spectroscopy of si x flavonoids of Limnanthes. It presents the proton s h i f t s of TMS.-ethers'.,af s i x ^-D-rutinosides r e l a t i v e to tetramethylsilane. Reproductions of the NMR spectra of thse s i x derivatives are presented as Appendices LI-LVI. B. Explanation of the Data The' flavonoids of the Limnanthaceae follow a d e f i n i t e pattern. A series of s i x f l a v o n o l aglycone types repeats many times, each time com-p r i s i n g a class of compounds possessing uniform g l y c o s y l a t i o n . In t o t a l 12 gly c o s y l a t i o n patterns were found i n the Limnanthaceae. A l l of these i s o l a t e d i n quantities s u f f i c i e n t f o r hydrolysis proved to be based on two sugars, glucose and rhamnose. Using aqueous solvents, flavonoids were resolved into groups, each with uniform g l y c o s y l a t i o n . The a c i d i c chloroform solvent resolved these groups into s i x components, each based on a p a r t i c u l a r aglycone type. Thus, the array of compounds diagrammed i n Figure III i s divided along the h o r i -zontal axis into g l y c o s y l a t i o n classes ( , U-Z, 1 - 6 , ... etc.) that were separated by development i n the aqueous solvent. These groups i n turn are separated along the v e r t i c a l axis into i n d i v i d u a l compounds based on each - 52 -TABLE V Products Obtained upon T o t a l Hydrolysis with 1.0 N HC1 Compound . Aglycone Sugar (s ) . 0 Sg-3-0-G Syringetin Glucose P Ir-3-0-G Isorhamnetin Q Kp-3-0-G Kaempferol R Lc-3-O-G L a r i c y t r i n S Qu-3-O-G Quercetin T My-3-0-G Myricetin U Sg-3-0-Rut Syringetin Glucose, Rhamnose V Ir-3-0-Rut Isorhamnetin I! W Kp-3-0-Rut Kaempferol 1! X Lc-3-0-Rut L a r i c y t r i n 11 Y Qu£:3-0-Rut . Quercetin II Z My-3-0-Rut Myricetin II I Sg-3-0-GG• Syringetin Glucose K Kp-3-0-GG Kaempferol i i M Qu-3-0-GG" Quercetin II 1 Sg-3-0-GRR Syringetin 1 Glucose, 2 Rhamnose 2 , Ir-3-0-GRR ' Isorhamnetin 3 Kp-3-0-GRR' Kaempferol ' " k Lc-3-O-GRR L a r i c y t r i n " 5 Qu-3-O-GRR Quercetin 6 My-3-0-GRR Myricetin 7 Sg-3-0r-Rut,7-0-G' Syringetin 2 Glucose, 1 Rhamnose 8 Ir-3-0-Rut,7-0-G Isorhamnetin 9 ' Kp-3-0-Rut,7-0-G Kaempferol 10 Lc-3-0-Rut,7-0-G L a r i c y t r i n 11 Qu-3-0-Rut,7-0-G Quercetin' 12 My-3-0-Rut,7-0-G Myricetin 13 Sg-3-0-Rut,7-0-Rut Syringetin G.lucose, Rhamnose Ik Ir_3_0-Rut,7-0-Rut Isorhamnetin II 15 Kp-3-0-Rut,7-0-Rut Kaempferol n 16 Lc-3-0-Rut,7-0-Rut L a r i c y t r i n i i 17 • Qu-S-O-Rut^-O-Rut Quercetin II - 53 -TABLE VI-Derivatives Obtained upon P a r t i a l Acid Hydrolysis and Hydrolysis with Emulsin Compound Hydrolysis Derivatives 1 Sg-3-0-GRR 20% HOAc , Sg, Sg-3-O-G, Sg-3-O-Rut 2 Ir-3-0-GRR TI I r , I r_3_0-G, Ir-3-0-Rut 3 Kp-3-0-GRR it Kp, Kp-3-O-G, Kp-3-O-Rut k Lc-3-0-GRR tt Lc, Lc-3-0-G, Lc-3-O-Rut 5 Qu-3-O-GRR tt Qu, Qu-3-O-G, Qu-3-O-Rut 6 My-3-0-GRR. it My, My-3-O-G, My-3-0-Rut •7 Sg-3-0-Rut ,7-0-G 0.1 N HC1 Sg, Sg-7-O-G, Glucose, Rhamnose 8 Ir-3-0-Rut ,7-0-G tt I r , Ir-7-O-G,.Glucose, Rhamnose 9 Kp-3-0-Rut ,7-0-G tt Kp, Kp-7-O-G, Glucose, Rhamnose 10 . Lc-3-0-Rut ,7-0-G tt . Lc, Lc-7-O-G, Glucose, Rhamnose 11 Qu-3-0-Rut ,7-0-G it Qu, Qu-7-0-G, Glucose, Rhamnose 12 My-3-0-Rut ,7-0-G tt My, Hy-7-0-G, Glucose, Rhamnose 7 Sg-3-0-Rut ,7-0-G' Emulsin Sg- 3-0-Rut, Glucose 8 Ir-3-0-Rut ,7-0-G tt I r - 3-0-Rut, Glucose 9 Kp-3-0-Rut ,7-0-G tt • Kp- 3-0-Rut, Glucose 10 Lc-3-0-Rut ,7-0-G tt. Lc- 3-0-Rut, Glucose. 11 Qu-3-0-Rut ,7-0-G' tt Qu- 3-0-Rut, Glucose 12 My-3-0-Rut ,7-0-G tt My- 3-0-Rut, Glucose 13 Sg-3-0-Rut ,7-0-Rut' 0.1 N.-HCl Sg, Sg-7-0-G, Sg-7-0-Rut Ik Ir-3-0-Rut ,7-0-Rut tt.. I r , Ir-7-0-G,.Ir-7-0-Rut 15 Kp-3-0-Rut ,7-0-Rut it Kp, Kp-7-0-G, Kp-7-0-Rut 16 Lc-3-O-Rut ,7-0-Rut tt Lc, Lc-7-O-G, Lc-7-0-Rut 17 Qu-3-0-Rut ,7-0-Rut tt Qu, Qu-7-O-G, Qu-7-0-Rut 13 ' Sg-3-0-Rut ,7-0-Rut Emulsin . No A l t e r a t i o n Ik Ir-3-0-Rut ,7-0-Rut tt tt 15 ' Kp-3-0-Rut ,7-0-Rut tt tt 16 Lc-3-0-Rut ,7-0-Rut it ti 17 Qu-3-0-Rut ,7-0-Rut tt a. it TABLE VII UV*kb:sorp!biph'-Maxiii|, of; E l a v o n o i d s o f - t h e L imnanthaceae .MeOH ?• Kef KeOR h MeOH & MeOH ft-MeOH T Ta0':e Aica A 1 C 1 , & HC1 NaOAo TaO.a r v V •3 BO COMPOUND •M I I I 1 I I I -'• 1 I I I M 3 . ? L c-3 - O - G 255,266s ; sos- 36-1 268 327 1*19 272 301s 433 271* 309 371,.40 6 272 325 1.11* 262 302s 383 Sg-3-0-G 252,262s , sops 359 266 329 It 27 273 310 3 7 0 s , 1*06 273 310 366,1*05 270 325 1(22 251,261* 306s 361* My-3-0-Rut 261 306 362 266 322 1*09 273 318 s 1*23 271* 320s .3743,1*08 269 328 1(09 261 306s 383 Qu-3-0-P.ut 258,268s . 299 359 266 322 Ul8 - 276 30os 435 271 299s 366,1(04 271* 328 1(09 261*' 295s 382 Lc-3-0-Rut 255,266s 301 361 266,286S 329s 1*31 • 27!* 306s 1*39 273 302s 367,407 270 322 1*23 '267 302s 381* Km-3-0-Rut 266 : ; 350 266 32!* 1*05 275 306s 35,*,397 275 306s 350,394 275 306s 330 355 Ir-3-0-Rut 255,268s ' ;300 356 270 332 1*11 269 ' 293s •373s, 1*07 269 203 3oU,Uo6 271* 323 4 09 255,268s 361 Sg-3-0-Rut 254,268s \ 300 361 266 330 1*26 272 307 ' 3 8 0 s . 1*10 271* 307 360,1(07 265 323 427 266s 308 365 Qu-3-O-GG 257,268s ' 298 357 270 322s 1*09 ' 277 302s i»3l* 271 2Q8s 365,1*06 271 32l( 395- . 262 " 302s 371*" Km-3-0-GG 268 - 3I+9 270 322 1*01 275 302s 352,1+00 275 302s 349,396 275 302s 375 269 302s 351 S g - 3 - O - G G 251»,2,65s 303 362 266,288s 358 1*29 272 310 3 8 l s , . l * i a 27l* 310s 370,1*0-3 266 322 1*30 26.IJS 302 365 My-3-0-GRR 255,266s 297 357 266 ,285S 326s 1*12 271* 310s 1*26 273 30ks 370,1(07. 270 318s 390 263s 366 Qu-3-O-GRR 257,267s 305s 356 270 319 hO.P- 271* 306s 1*37 - 270 301s 368,1*05 271 326 383 262 310s 374 Lc-3 -0 -GRR 255,266s 300 359 266 324 1H7 271* •30ps 1(1*1 275 309s 367,1(08 270- 32n 1*12 260 375 Km-3-0-GRR . 267 297s 3^9 27 u 324 395 275 305 355 ,403 276 303 348 ,399 271* 313 333 267 352 255,268s 302s 355 266 326 1*09 270 301s 371 s , 1*09 26o 301s 363.1*01+ 273 323 !*0l* 255,26" S 358 Sg-3-0-GRR 254,268s 303 359 266 328 1*18 271* 306s 368s, 1*11 275 30'6s 371 ,407 . 272 316 388 263s 360 Lc-3- 0 - R u t , 7-0-.G 258,266 291* 358 269 1*03 273 206s 3 6 2 s , 1*32 273 296s 368s, 1*03 265 2°6s 1*03 261 2C2s 373 1). Expressed 3s TABLE VIII Proton Chemical S h i f t Values of TMS Ethers of Six Flavonol J3-D-Rut inos ides of the Limnanthaceae A-Rin£ B-Rihg Carbohydrate Compound 6 8 2* 6' 3' 5' Glc-l-H Rhm-l-H Rhm-Me Other Kp-3-0-Rut 6.08 d,J2.5 5.36 d,J2.5 (7.68) 2H,d,J8.5 (6.78) 2H,d,j8 .5 5.80 d,J7 h.l9k d , J l 0.71 3.26-3.83 Qu-3-0-Rut •6.08 '6.36 ' 7-30 7.34 d,J2 q,J2.,9.5 6.78 d,J9-5 5.75 4.20 0.81 3.29-3.80 I r_3_0-Rut 6.08 6.38 . 7-45 7.27 d,J2 q.,J2,9 3 .84 3H,s^ 6.77 d,J9 5.84 4.20 0.72 3.25-3.74 My-3-0-Rut 6.06 6.32 (7.08) 2H,s - - 5.81 4.22 1.00 3.29-3.85 Lc-3-0-Rut . 6.09 6.37 7.20 • 6.90 d,J2 d,J2 - 3.85 s 3H 3s ? 5-89 4.23 0.81 3.28-3.76 Sg-3-0-Rut 6.10 6.41 (7.12) 2H,s (3 . 84) 6H,S 5 5-93 4.22 0.73 3.25-3.75 1 — Expressed as (Tin ppm r e l a t i v e to tetramethylsilarie (0.00) i n CCl^. 2 — Consistent value;s f o r H-6 & H-8. 3 — Consistent values f o r glucose-l-H. 4 — Consistent values for rhamnose-l-H. 5 — 0-methyl groups - 56 -aglycone type • (e_. g_. , U,V,W, ... etc.) that were resolved a f t e r development i n the organic solvent. Only three of the many, g l y c o s y l a t i o n classes occurring i n the Limnan-thaceae are present i n r e l a t i v e l y great amounts. The most prominent group among these, one that i s present i n a l l taxa, consist's^of the s i x compounds designated U,V,W,X,Y and.Z. The i d e n t i f i c a t i o n of these s i x flavonoids was c r u c i a l to the enti r e i n v e s t i g a t i o n as they were often obtained as hydrolysis products from other flavonoid glycosides, and from these the si x aglycone types were i d e n t i f i e d . Rfs i n aqueous solvents of the compounds U-Z suggested that they were diglycosid'es';'" Upon hydrolysis each y i e l d e d a d i f f e r e n t aglycone, and the sugars, glucose and rhamnose (Table V). Four of the s i x aglycones were kaemp-f e r o l , quercetin, isorhamnetin and myricetin. UV and NMR sp e c t r a l data of the corresponding glycosides (Tables VII & VIII) agreed completely with pub-l i s h e d data for 3 - 0 - jJ-D-rutinosides. The remaining two rutinosides were new. One (U) had Rf and UV behaviour s i m i l a r to isorhamnetin 3 - 0-rutinoside (V) which suggested the presence of an O-methylated B-ring. The NMR spectrum of the TMS derivative;, indeed showed the presence of two equivalent O-methyl groups, 3.84c*, and. a s i n g l e t at 7 • 12 S which integrated f o r two protons. This s i n g l e t indicates equivalent protons on the B-rihg. These data indicate that the B-ring i s a syringyl-typevv Signals at 6.1C& and.6.l4cT showed the expected meta s p l i t t i n g of the 6-.K and 8-H... The aglycone i s , therefore, myricetin 3',5'-dimethyl ether, or s y r i n g e t i n . The second new rut inos idi,e.(x) was s i m i l a r to r u t i n (Y) i n i t s Rf and UV behaviour (Figure I I I , Table VII). The NMR spectrum of the TMS de r i v a t i v e showed four aromatic protons, as i n the case.of s y r i n g e t i n r u t i n o s i d e , but only one O-methyl group (Table VIII':);. The B-ring protons appeared as doublets at 7-20S and 6.90S" and displayed meta coupling. These r e s u l t s allow the - 57 -assignment of the structure myricetin 3'-methyl ether to the aglycone, or l a r i c y t r i n as i f has recently been named by Tyukavkin et_ al_. (1974). The acetate derivatives were prepared of s y r i n g e t i n 3-0-rutinoside, l a r i c y t r i n 3-0-rutinoside and quercetin 3-0-rutinoside, and t h e i r melting points determined. The melting points of the three acetate d e r i v a t i v e s are: quercetin rutinoside — Il6-ll8°, l a r i c y t r i n r u tinoside — 124 -126° , and s y r i n g e t i n rutinoside — 130-132°. The second major group of glycosides investigated, termed 1,2,3,4,5 and 6, moved further i n aqueous solvents than d i d the rutinosides (Figure I I I ) , suggesting a-greater degree of g l y c o s y l a t i o n . Their UV spectra showed them to be 3-0-glycosides (Table VII). T o t a l hydrolyses gave the .same s i x aglycones already described and, i n each case, two equivalents of rhamnose and one of glucose (Table V). P a r t i a l a c i d hydrolysis of these t r i g l y c o s i d e s gave, i n each case, the corresponding 3-0-rutinoside and 3-0-glucoside (Table VI). The point of s u b s t i t u t i o n of the terminal rhamnose has not been determined, but the s i x t r i g l y c o s i d e s are c l e a r l y 3-0-rhamnosyl rutinosides. The t h i r d major group of glycosides, c o n s i s t i n g of the s i x components 7,8,9,10,11 and 12, had higher Rfs i n aqueous solvents than the two preceding groups [Figure I I I ) . A l l attempts to f r a c t i o n a t e these glycosides into s i x components by preparative TLC, as was done for the rutinosides and rhamnosyl-ru t i n o s i d e s , were unsuccessful. However, one of these s i x compounds was i s o l a t e d from the r e s t by c r y s t a l i z a t i o n from MeOH, and i t s UV spectrum cor-responded to a 3,7-diglycoide (Compound 10,' Table VII). Since separation could not be achieved, hydrolysis experiments were performed on a l l s i x gly-cosides simultaneously. T o t a l hydrolysis gave the s i x aglycones present i n the other groups, two equivalents of glucose and one equivalent of rhamnose (Table V). P a r t i a l a c i d hydrolysis .yielded f l a v o n o l derivatives with Rfs and UV s p e c t r a l c h a r a c t e r i s t i c s t y p i c a l of 7-0-monosides. Treatment with ^-gluco-- 58 -sidase- (emulsin) produced the si x 3 - 0 -rutinosides and glucose (Table VI). Therefore, these' compounds must be the fla v o n o l 3-0-^-D-rutinoside^7-0-^-D-glucosides. The 18 f l a v o n o l glycosides described above are ubiquitous i n the Limnanthaceae, occurring i n high concentrations i n nearly every taxon inves-tigated. The other groups of glycosides encountered i n t h i s work occur s p o r a d i c a l l y throughout the family, but i n l e s s e r concentrations than the three common groups. The group of glycosides c o n s i s t i n g of compounds I,J,K,L,M and N had Rfs i n aqueous solvents i n d i c a t i n g that they were biosides. Rfs i n the organic solvent and colours a f t e r spraying indicated that these s i x glyco-sides were based on the si x aglycones already encountered. Of the s i x , only I,K and M were i s o l a t e d i n quantities s u f f i c i e n t f o r s p e c t r a l and hy d r o l y t i c analysis. These three had UV spectra corresponding to 3 - 0-glycosides (Table VTl) , and t o t a l a c i d hydrolysis gave the aglycones s y r i n g e t i n , kaempferol and quercetin, and the sugar glucose (Table V). This information allows the conclusion that the three are f l a v o n o l 3 - 0-digiucosides. Although the com-pounds <T,L and.M occur only i n trace amounts, and.were not i s o l a t e d i n quantity^ they are almost c e r t a i n l y the 3-0-d'iglucoside analogues of i s o -rhamnetin, l a r i c y t r i n and.myricetin, r e s p e c t i v e l y . The g l y c o s y l a t i o n series containing the six compounds 0,P,Q,R,S and T had Rfs i n aqueous solvents c h a r a c t e r i s t i c of fl a v o n o l monosides (Figure III).'. Their UV spectra were those of fla v o n o l 3 - 0-glycosides .(Table VII). T o t a l acid hydrolysis y i e l d e d each of the s i x aglycones and only glucose- (Table V). Therefore, these compounds are.flavonol 3 -0-^-D-glucosides. A series of f i v e compounds,' 13 , l h , 1 5 , 1 6 and 17, had Rfs i n aqueous solvents even higher than those of the 3,7-branched t r i o s i d e s (Figure I I I ) , suggesting a higher degree of gl y c o s y l a t i o n . Attempts to resolve t h i s group - 59 -into i t s f i v e components Py preparative TLC were unsuccessful. Therefore, i t was necessary to hydrolyze these compounds as a group. . T o t a l hydrolysis y i e l d e d the aglycones, s y r i n g e t i n , •isprhamnetin, kaempferol, l a r i c y t r i n and quercetin, and two sugars, glucose and rhamnose,. both i n apparently equal proportions (Table V). Attempts to hydrolyze these compounds with emulsin were unsuccessful, i n d i c a t i n g that glucose does not occur i n a terminal p o s i t i o n on these compounds. P a r t i a l a c i d hydrolysis gave three series of f i v e com-pounds each, one series consisting of the f i v e aglycones, and the other two with Rfs and colours c h a r a c t e r i s t i c of 7-0-monosides and 7-0-biosides. Insuf-f i c i e n t material and time prevented further experiments on these f i v e glyco-sides, but a l l of the information gathered i s consistent with the assignments of f l a v o n o l 3-0-rutinoslde,7-0-rutinosides. A l l other flavonoid compounds present i n the Limnanthaceae with the exceptions of H,A and B occurred i n concentrations too low for i s o l a t i o n and i d e n t i f i c a t i o n . Compound H was found"; only i n Floerkea, of which only small amounts of material were a v a i l a b l e f o r study. I t s colour c h a r a c t e r i s t i c s and low Rf i n the organic solvent suggest that t h i s i s a kaempferol 3-0-monoside (Figure I I I & Table IV). Its extremely low m o b i l i t y i n the aqueous d i r e c t i o n suggests that t h i s i s an acylated d e r i v a t i v e . Compounds A and B have colour c h a r a c t e r i s t i c s and RFs which suggest that they are 7-0-biosides of kaempferol and quercetin. Problems i n obtaining s u f f i c i e n t amounts of these two com-pounds, which w i l l be discussed i n a l a t e r section, have.not allowed complete i d e n t i f i c a t i o n ' , but they are probably f l a v o n o l 7-0-rutinosides. I d e n t i f i c a t i o n of the three remaining g l y c o s y l a t i o n series can only be speculated upon based on Rfs. Compounds D and.E have Rfs that correspond to 7-0-monosides and are probably the 7-0-glucosides of kaempferol and quer-c e t i n . Compounds F and G also have Rfs i n the organic solvent i n d i c a t i v e of g l y c o s y l a t i o n at the 7-position of the f l a v o n o l nucleus. This fact-, taken - 60 -together with t h e i r m o b i l i t y i n the aqueous d i r e c t i o n , suggests that they are 3,7-dimonosides; i f t h i s i s true, they are probably the 3 - 0 -glucoside , 7 - 0 -glucosides of kaempferol and-quercetin. Compounds\ 19,20,21,22 and 23 remain unknown. Although c l e a r l y d e r ivatives of s y r i n g e t i n , isorhamnetin, kaemp-f e r o l , l a r i c y t r i n and quercetin, t h e i r Rfs i n the aqueous solvent, midway between the 3 - 0-glucosides and 3 - 0-rutinosides (Figure I I I ) , makes the extent of g l y c o s y l a t i o n unclear. These compounds are either 3-0-monosides or biosides. 3-0-Dirhamnosides would probably have Rfs s i m i l a r to these compounds, but f u r -ther information i s required before t h e i r i d e n t i t y can be established. SIGNIFICANCE OF THE FLAVONOIDS OF THE LIMNANTHACEAE  A. Aglycones Although anthocyanins occur i n trace amounts i n some taxa, flavonols were the major type-of flavonoid encountered i n the present i n v e s t i g a t i o n . Six f l a v o n o l aglycone types are made and accummulated by every member of the Limnanthaceae.. These include the common fla v o n o l types: kaempferol, quer-c e t i n , isorhamnetin and myricetin, the r e l a t i v e l y rare s y r i n g e t i n , and the extremely rare laricytrin,, or •3':-methyl myricetin. These aglycones d i f f e r only i n t h e i r B-rings, and together they comprise a series of derivatives made by successive hydroxylation and methoxylation additions to t h i s r i n g . This stepwise elaboration of the B-ring, from kaempferol to s y r i n g e t i n , i s shown i n Figure IV. Syringetin has r a r e l y been reported previously. Besides Limnanthes barker & Bohm, 1975) and Floerkea, t h i s aglycone has been reported from Lathyrus (Harborne, 1965), L a r i x ;(Niemann, 1972; 1973; Tyukavkin et a l . , 1974), Soymida (Parkhasaradhi & Sidhu, 1972), and.Philydrum. (Bohm & C o l l i n s , 1975). L a r i c y t r i n - has previously been reported only from L a r i x . I t s occur-rence was suspected by -Niemann(1972, 1973) and was v e r i f i e d by Tyukavkin and FIGURE IV Stepwise Elaboration of the B-Ring of Flavonols of the Limnanthaceae - 62 -coworkers (197*0. However, i t i s probable .that l a r i c y t r i n has a much wider d i s t r i b u t i o n than these data i n d i c a t e , s i n c e t h i s compound i s not e a s i l y sep-arated from quercetin. I f l a r i c y t r i n i s not separated from quercetin, i t s presence w i l l go) undetected since these compounds possess very s i m i l a r phys-i c a l properties. TLC chromatography on polyamide DC 6.6 with organic solvents appears to be the only way presently known to separate these two aglycone types. When t h i s technique i s more widely adopted, seyerrali, a d d i t i o n a l r e -ports of l a r i c y t r i n w i l l probably be made. The recent discovery of 'lar i c y -t r i n i n Heuchera (C..K. Wilkins, personal communication) tends to support t h i s observation. B. Glycosides A t o t a l of twelve g l y c o s y l a t i o n patterns was>,discovered i n the course of t h i s i n v e s t i g a t i o n . Although monosides do occur i n various members of the Limnanthac eae, the' majority of the flavonoids of Limnanthes and Floerkea are.glycosylated with two or more sugars. A l l of the i d e n t i f i e d flavonoid g l y c o s y l a t i o n patterns are based on d i f f e r e n t combinations of glucose and rhamnose. R e l a t i v e l y few t r i o s i d e s and t e t r a s i d e s have been reported, with the exception of 3 - 0-rutinoside , 7 - 0-glucosides. To the best of my knowledge, there have been no previous reports of 3-0-rhamnosyl r u t i n o s i d e s , nor have. there been any reports of 3 - 0-rutinoside , 7 - 0-rutinosides. THE CONTRIBUTION TO VISIBLE FLOWER COLOUR BY SYRINGETIN DERIVATIVES Harborne (T967) has reported that s y r i n g e t i n acts as a yellow pigment i n the flowers of Lathyrus. Although certain.Limnanthes taxa have flowers that contain bright yellow pigments, in c l u d i n g two v a r i e t i e s of L_. d o u g l a s i i , s y r i n g e t i n apparently does not contribute to t h i s phenomenon. This was - 63 -concluded since high concentrations of s y r i n g e t i n glycosides occur i n the petals of most Limnanthes taxa, the majority of which have white flowers. Furthermore, the yellow pigment present i n Limnanthes flowers i s probably a carotenoid(s) since i t i s r e a d i l y extracted with chloroform. A l l the white flowered forms of Limnanthes have petals that are tinged with a pale yellow colour, p a r t i c u l a r l y toward t h e i r bases. This e f f e c t i s accentuated by drying specimens. I t i s quite probable that s y r i n g e t i n g l y -cosides, as w e l l as other f l a v o n o l glycosides, contribute to t h i s yellowing. A l l s y r i n g e t i n glycosides i s o l a t e d during t h i s investigation' share an obvious property; they are a l l extremely s e n s i t i v e to base. When fumed with NH^ these compounds i n s t a n t l y fluoresce an intense pale yellow colour under UV l i g h t . In f a c t , o ccasionally s y r i n g e t i n derivatives were detected only by t h i s technique, being i n v i s i b l e under UV following spraying but p r i o r to fuming. A f t e r fuming, these compounds appear yellow i n v i s i b l e l i g h t and remain t h i s way f o r a short period (normally they appear a very pale y e l -low when present i n high concentration, and-are i n v i s i b l e when present i n lower concentration). This observation indicates that s l i g h t l y basic conditions i n a flower might e a s i l y cause s y r i n g e t i n glycosides to serve as yellow p i g -merits i n v i s i b l e l i g h t . ENVIRONMENTAL MODIFICATION OF FLAVONOID COMPOSITION Taxonomic studies using flavonoids as characters r o u t i n e l y assume that flavonoid compositions do not vary with changes i n the environment. (This statement i s not meant to include the small amounts of flavonoid hydros l y s i s that occur as plants age.) However, the presence of two flavonoid compounds of Limnanthes, A'and B (probable T-O-rutinosides of kaempferol and. quercetin) apparently depends upon environmental f a c t o r s . Germination of a l l Limnanthes seed drops to a low l e v e l as s o i l - 64 -temperatures, increase from'spring to summer. Since seed was germinated out-side, t h i s phenomenon caused an acute problem i n some instances. On the f i r s t attempt, germination of OTUs 12 and 15, L. alba var. alba and L_. g r a c i l i s var. g r a c i l i s , was 1-2%, producing only two plants of each taxon. The-pots con-t a i n i n g these plants were not tr a n s f e r r e d to the greenhouse following the standard p r a c t i c e , but were l e f t i n cold frames pending a d d i t i o n a l germination. No further germination occurred, and eventually the two plants of each OTU were harvested, and the flavonoid compositions of petals and whole plant material were determined. Both of these taxa contained A and B i n reasonably high concentration. However, i n s u f f i c i e n t plant material made it-' impossible to i d e n t i f y these flavonoids. An a d d i t i o n a l f l a t of L_. alba var. alba was grown at a cooler time of year, and germination was much improved. However, at t h i s l a t e r date the plants were tr a n s f e r r e d to the greenhouse following the usual p r a c t i c e . Neither the flowers nor whole plant material contained compounds A and B, although otherwise they were chemically i d e n t i c a l to the plants grown i n the cold frames. The presence of A and B was not detected i n any other members of the Limnanthaceae including those taxa represented by drie d material. Presence of A and B i n plant material grown i n the cold frame i s assumed to be environmentally induced rather than a genetic p e c u l i a r i t y of a few plants, since the phenomenon was observed f o r two d i f f e r e n t taxa. The nature of the inducing f a c t o r i s not known; however, i t seems u n l i k e l y that the compounds A and B are breakdown products since neither L_. alba nor L_. g r a c i l i s contains any 3 , T-dirut inos ides, presumably the only class of compounds encountered i n any taxa. of Limnanthes which could possibly-breab--. down to y i e l d A or-B. At any rate these two compounds have been eliminated from the comparative a n a l y s i s s i n c e t h e i r presence i s v a r i a b l e and.they are, therefore, unsuitable f o r taxonomic comparisons within the Limnanthaceae. - 6 5 -COMPARATIVE FLAVONOID' DATA . Flavonoid complements of a l l taxa were determined from two-dimensional "thin layer chromatograms. Figures V and VI are colour photographs of an actual two-dimensional map of the whole plant flavonoids of L. montana, OTU 17, taken a f t e r spraying and marking. Figure V shows the chromatogram under UV l i g h t . Exposure d e t a i l s are: 2 minutes _at.F 3.5, using a Wratten 2E f i l t e r (Kodak) and High Speed Ektachrome f i l m . Figure VI shows the same chromatogram taken i n daylight at 1/25 second at F 3.5- Two points are noteworthy about the ,T picture made with UV l i g h t , l ) , ,The compounds are c o r r e c t l y . shown as f l u -orescent spots a f t e r spraying, but the colours are not accurately represented. Actual differences i n colours between compounds are greater than they appear i n t h i s p i c t u r e . 2) The photographic f i l m i s more s e n s i t i v e than the human eye. Compounds which appeared at the l i m i t s of detection are c l e a r l y v i s i b l e i n t h i s p i c t u r e , and c e r t a i n features of the photograph were i n v i s i b l e to the naked eye. The pictures of the L_. montana chromatogram exemplify the experimental r e s u l t s upon which the conclusions of t h i s t h e s i s are based. Photographs of other chromatograms have not been included; however, tracings of labeled and marked chromatograms'of p e t a l and whole plant material are included as appen-dices (Appendices I I I - L ) . In both the photographs and the t r a c i n g s , com-pounds v i s i b l e before spraying are.marked by s o l i d l i n e s , while compounds v i s i b l e only a f t e r spraying are marked by broken l i n e s . Table IX l i s t s the flavonoids present i n whole plant.material of a l l 30 OTUs included i n t h i s study. Compounds present i n r e l a t i v e l y great con-centration are indicated by'"Xs", while' compounds that became v i s i b l e only a f t e r spraying are indicated by " t s " . The compounds are grouped according to g l y c o s y l a t i o n type with l i n e s drawn between these groups. The groups are distinguished i n t h i s fashion so that occurrence of a g l y c o s y l a t i o n class can - 66 -FIGURE V TLC Map of Whole Plant Flavonoids of L. montana Taken i n Ul t r a v i o l e t Light * 4 • FIGURE VT TLC Map of Whole Plant Flavonoids of 1, montana Taken i n V i s i b l e Light TABLE IX WHOLE PLANT FLAVONOIDS OF 30 OTUS OF THE LIMNANTHACEAE OTU CCMFOUKD: j-O-Rut U V W X 1 3-0-2 3 SSI! It 5 6 3-o-7 8 Rut g ,7-10 o -s 11 12 3-o-Ruty-o-Rut '13 I1* 15 16 17 0 p 3-0-G Q R s T 19 20 21 22 23 F G c ' I 3-0-GG J K L M t X X X X X X X X X X X X X X X X X X X X X X X X .X X X X X t X X X X X X X X X X t t X X t t X X X X X X X X X X X X X t t t t X X X X X X X X X X X X X X t t t X X t t X X X X X X X X X X X X X t t X X X X X X X X X X X X X t t t X X X X X X X X X X X X X t X t X t t X t X X X X X X X X t X X X . X X X X X X X X X X X X X X X t t X X X X t X X t t X X X t t X t X X X t X X X t X X X X t t X X X X X t X X X X X X X X t X X X X X t X X t X X X X X X X X X t X t X X X x' X X X X t t X t X X X X X X X X X X X X t t X X X t X X. X X X X t t X X X X X X X X X X t X X X X t t X X X X t t X X X X X X X X X t t X X X X X X X t X X X X t t X X X X X X X X X X t t X X X X X X X X X t X X X t X t X X X X X X X X X X X X X X t t X X X X X X t X X X t X X t t X t t X X X X X X t t t X X X X X X t t X t X X X X r> E A B s t r i a t a 2 5 3 h 6 10 1 Q doug. doug. dcug. sulph. doug. nivea  dcav. rosea  vjr.culans macounii dog;, doug. rcicounii 21 floe, pusdla 25 floe. floe. 19 flee, hell. floe. floe. floe, boll, grac. parish, alba versic. 2k 18 16 13 17 11 alba alba 12 alba alba 15 l i t 22 20 26 grac. grac. grac. grac. floe, grand. floe, purlla floe, cali f . 23 floe, floe. 29 Floerkea 28 Flcerkea 27 Floerkea -0 Floerkea X X X X. X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X t t t t t I X t X t - .68 -be r e a d i l y assessed'"!" or. each. OTU. Table X l i s t s the flavonoids present i n fresh petals of 18 OTUs of Limnanthes. The format i s the same as that described for Table IX. Taxa have been arranged manually i n Tables IX and X to place OTUs with the greatest flavonoid s i m i l a r i t i e s next to each other. Although the r e s u l t i n g arrangements i n both tables are somewhat subjective, i t i s possible to see a certain.amount of c l u s t e r i n g between OTUs. FLAVONOID DIFFERENCES BETWEEN PETAL AND WHOLE PLANT MATERIAL Relative concentrations of c e r t a i n flavonoids often"' d i f f e r e d markedly between p e t a l and whole plant material. In the v a r i e t i e s of L_. d o u g l a s i i , L_. f l o c c o s a and-L. g r a c i l i s , the 3-0-rutinoside,7-0-glucosides occur i n high concentration i n l e a f m a t e r i a l , but are nearly absent from the flowers. The most extreme examples are found i n L_.. d o u g l a s i i vars. rosea and nivea which • possess no detectable amounts of these compounds i n t h e i r petals. Conversely, the 3-0-rhamnosyl rutinosides appear i n higher concentrations i n petals than whole plant m a t e r i a l , although the differences are not as extreme. A comparison of Tables IX and X reveals that occasionally compounds occur i n p e t a l material but not i n whole plant material of some OTUs. Theo-r e t i c a l l y , the p e t a l flavonoids should always be a subset of the whole plant flavonoids since whole plant material c o n s i s t e n t l y contained flowers. However, t h i s s i t u a t i o n has hot been r e a l i z e d , presumably because c e r t a i n p e t a l f l a v o -noids are present i n concentrations too low to detect once the petals become a minor component of the t o t a l t i s s u e analyzed. NUMERICAL TAXONOMIC RESULTS-The data contained i n Tables IX and X were analyzed by computer using several taximetric.techniques described i n the previous chapter. Figures PE^AL FLAVONOIDS OF 18 OTUS OF LIMNANTHES COMPOUND: 'J 3-0-V w Rut X Y z 1 3-0-2 3 GRR It •5 6 3-o-7 8 Rut,7-9 10 0-G 11 12 3-O-Rut7-0-Rut 13 ll+ 15 16 17 0 p 3-0 Q -G R s T 19 20 21 22 23 F G c I 3-0 J K -GG L M X X X X X X X X X X X t X X X X X X X X X X X . t X X t t t X X X X X X X t X X X X X X X X X X X X X X t t X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X t t X X X X X t X X X X X X X x X X X X X X X x X X X X X X X X X X X X X X X X X X X X X X X X t X X X t t • t t t t X X X X X X X X X X X X X t t t t t t X X X X X X X t X X t X X X X X t X X X X X X X t t X X X t t X X t t t t X X X X X X X X t X X t X t t - X t t t t X X X X X t t t t t. t t X X X X t t t X X X X X X t X X t X X X X t t X X X X X X t X X t X X X X X t t X X X X X X t t t t X X X X X X t X X X X X X X t t X X X X X X X X X t t X X t • t t X X t D E A B 3 2 3 1* 10 1 o 13 lr-13 17 11 1U 20 l a t a e c u ; , d o-.'.g. dcug. s u l p h . c a i ^ . r.lvea c^u.;. r o s e a a a j c u r . i l  f l e e , b e l l . : : r a c . r a r i s h . aV.-a a l b a ;i!!'a al'.a  ;:vac • i':ae . .;:-ao. ..-'.-ac • r'.imila VII to X are the'results of c l u s t e r analysis by the weighted'pair group method based on whole plant flavonoids. The figures are dendrograms showing degrees of phenetic s i m i l a r i t y between a l l 30 OTUs based on h6 flavonoids. Figures VII and VIII show phenetic s i m i l a r i t i e s based only on compounds occurring i n higher concentration (the Xs i n Table IX), while Figures IX and X show, a f f i n i t i e s based on flavonoids present i n a l l concentrations (the Xs and ts of Table IX). The f i r s t of each p a i r of dendrograms (Figures VII and IX) i s calculated using the c o e f f i c i e n t of Jaccard which does not consider negative matches, while the second of each p a i r (Figures VIII and.X) i s based on the simple matching c o e f f i c i e n t which does consider negative matches. Figures XI' to XIV are dendrograms expressing phenetic s i m i l a r i t i e s between 18 OTUs based on the pe t a l flavonoid data contained i n Table X. As fo r the whole plant flavonoid c l u s t e r analyses, two of the dendrograms con-sider a l l flavonoid occurrences (Figures XIII and XIV) , while two only con-sider compounds present i n r e l a t i v e l y higher concentration (Figure XI and XII). Also, two' are calculated using the c o e f f i c i e n t of Jaccard (Figures XI and X I I I ) , while two are based on the simple matching c o e f f i c i e n t (Figures XII and XIV). The matrices of s i m i l a r i t y c o e f f i c i e n t s used to produce the eight dendrograms have been attached' as Appendices LVII" to LXIV. These matrices are arranged i n the same order as t h e i r corresponding dendrograms. Figures XV and XVI are representations of three-dimensional plots of 30 OTUs using P r i n c i p a l Components Analysis and Varimax Factor Analysis of 31 whole plant flavonoids. Figures XVII and XVIII are representations of three-dimensional plots of 18 OTUs using 36 p e t a l flavonoids. In these four figures the positions of OTUs i n space are represented by the locations of the end sections of cones. These cones have been drawn, either with points downward, to indicate a p o s i t i o n above the plane determined by the two h o r i z o n t a l vectors, or with points upward to indic a t e a p o s i t i o n below t h i s plane. When the FIGURE VII Weighted P a i r Group Clustering of 30 OTUs by Jaccard Coefficient:.High Concentrations of Whole Plant Flavonoids FIGURE VIII Weighted P a i r Group Clustering of 30 OTUs by Simple Match C o e f f i c i e n t : High Concentrations of Whole Plant Flavonoids FIGURE IX Weighted Pair Group Clustering of 30 OTUs by Jaccard C o e f f i c i e n t : M'- Concentrations of Whole Plant Flavonoids Ri PJ LU H- 01 •r-LO FIGURE X . Weighted Pair Group Clustering of 30 OTUs hy Simple Match C o e f f i c i e n t : A l l Concentrations of Whole Plant Flavonoids FIGURE XI Weighted Pair Group Clustering of 18 OTUs by Jaccard C o e f f i c i e n t : High Concentrations of Petal Flavonoids i H- H' H-N U Ui 03 a ui in oi F I G U R E X I I Weighted P a i r Simple Match C o e f f i c i e n t Group Clustering of 18 OTUs by-High Concentrations of Petal Flavonoids ru o K CO O FIGURE XIII Weighted P a i r Group Clu s t e r i n g of 18 OTUs by Jaccard C o e f f i c i e n t : A l l Concentrations of Petal Flavonoids —3 —] I 3 CD FIGURE XIV Weighted Pair Group Clustering of 18 OTUs by Simple. Match C o e f f i c i e n t : A l l Concentrations of Petal Flavonoids ... • FIGURE - XV-N ' • P r i n c i p a l Components Analysis of 30 OTUs Based on Whole Plant Flavonoids FIGURE XVI . Varimax Factor Analysis of 30 OTUs Based on Whole Plant Flavonoids 7 7 Co O FIGURE XVII P r i n c i p a l Components Analysis of 18 OTUs Based on Petal Flavonoids FIGURE XVIII Varimax Factor Analysis of 18 OTUs Based on Petal Flavonoids - 83 -p o s i t i o n of an OTU.was obscured by the presence of another i n front of i t , t h e i r positions i n the drawings were separated s l i g h t l y f o r clarity.- The p o s i t i o n of L_. bakeri, OTU 7', has been moved i n somewhat closer to the other OTUs than was determined by the computer analysis. This was done so that a l l of the taxa would be included i n the drawings. The actual coordinates of the OTUs determined by computer which were used to make the four drawings are presented i n Appendices LXV-LXVIII. COMPARISONS OF OTUS BY FLAVONOIDS ' A. Occurrence Tables The majority of the flavonoids discovered i n the Limnanthaceae are unusual, and many have not been previously reported. A perusal of Tables IX and X, which l i s t flavonoid occurrences by OTUs, reveals that a l l OTUs of Limnanthes and Floerkea possess rare derivatives of syringetin. and l a r i c y t r i n . Also, rhamnosyl rutinosides are found i n a l l OTUs of Limnanthes and i n some of the OTUs of Floerkea. No other plants are known which share these charac-t e r i s t i c s . ' This indicates two things: l ) the OTUs of Floerkea and Limnanthes constitute a natural grouping, and.2) no a f f i n i t i e s to other f a m i l i e s are indicated on the basis of flavonoid s i m i l a r i t i e s . However, the flavonoid compositions of f a m i l i e s with suspected a f f i n i t i e s to the Limnanthaceae are poorly known. The future discovery of Limnanthes-type flavonoids i n any of these f a m i l i e s would be s i g n i f i c a n t and might help c l a r i f y r e l a t i o n s h i p s above the family l e v e l . The most s t r i k i n g feature of Tables IX and X i s the o v e r a l l simi-l a r i t i e s i n flavonoid composition shown by a l l the taxa investigated. These s i m i l a r i t i e s are not s u r p r i s i n g i n view of e x i s t i n g information, a l l of which indicates- that members of the Limnanthaceae are c l o s e l y r e l a t e d . The r e s u l t s presented i n Table IX indicate that Floerkea.is d i s t i n -- 81i -guished from Limnanthes by two di f f e r e n c e s : T) three of the four OTUs of Floerkea contain compound H (possibly an acylated monoside of kaempferol) , which i s not found i n any OTUs of Limnanthes, and 2) the flavonoid patterns of the four OTUs of Floerkea are generally simpler than those of most Limnan- thes . , However, L_. flo c c o s a ssp. pumila, OTU 21, has a pattern of comparable simplicity.. Also the rare aglycone types are.not present as derivatives i n most glycoside s e r i e s , or i f present, occur i n trace amounts. It i s s i g n i f i c a n t that OTUs of Limnanthes are.separated i n Tables IX and X by c e r t a i n d i f f e r e n c e s , corresponding to the s e c t i o n a l d i v i s i o n proposed by Mason. OTUs described as members of the section Reflexae usually possess more'derivatives of s y t i n g e t i n , isorhamnetin, l a r i c y t r i n and.myricetin i n ce r t a i n g l y c o s y l a t i o n s e r i e s . This observation i s p a r t i c u l a r l y , t r u e f o r the pe t a l data. OTUs belonging to the same species have us u a l l y f a l l e n i n adjacent rows i n Tables IX and X, although there are notable exceptions: L. do u g l a s i i var. d o u g l a s i i , OTU 1,' from Europe, f a l l s between the two OTUs of L_. macounii; • Ii.' g r a c i l i s var. p a r i s h i i , OTU l 6 , more c l o s e l y resembles L_. alba var. v e r s i - color and L_. flo c c o s a than OTUs of i t s s i b l i n g v a r i e t y ; and subspecies of L_. fl o c c o s a do not l i e together, but instead form subgroupings. B. Cluster.Analyses The dendrograms are c l a s s i f i c a t i o n s of the'Limnanthaceae based purely on s i m i l a r i t i e s i n flavonoid composition. A l l flavonoid characters were given•equal weight, and c l u s t e r development was not prejudiced by previous taxonomic assessments. Computation of the dendrograms var i e d according to the way simi-l a r i t i e s , were determined, what concentrations of flavonoids were considered, and whether whole plant.or p e t a l material was analyzed. The r e s u l t i n g d i a -grams r e f l e c t these differences. - 8 5 . -Although the c l a s s i f i c a t i o n s of OTUs presented i n the dendrograms are d i f f e r e n t , a l l eight have c e r t a i n gross features i n common. The features of these dendrograms roughly coincide with the recognized taxonomic scheme of the family. In three of the four dendrograms based on whole p l a n t . f l a v o -noids, Floerkea. i s separated from Limnanthes. However, L_. flo c c o s a ssp. pumila, OTU 21, i s also' d i s t i n c t from the remainder of Limnanthes taxa and cl u s t e r s with Floerkea i n two of these three dendrograms. • OTUs belonging to the same section and.species generally have.clustered together i n both p e t a l and whole plant'analyses. However, as was observed f o r the manual, arrange-ment of OTUs i n the occurrence t a b l e s , c e r t a i n OTUs f a l l out of place. Tables XI and XII are attempts to summarize the information contained i n the eight dendrograms^ Table XI i s a synopsis of the r e s u l t s of the four c l u s t e r analyses based on whole plant flavonoids of 3Q OTUs, and Table XII i s a synopsis of the r e s u l t s of the four c l u s t e r analyses based on p e t a l f l a v o -noids of 18 OTUs. The OTUs analyzed from dried material have been underscored with dotted l i n e s i n Table XI. OTUs are grouped together i n the l e f t hand columns of Tables XI and XIr i f they clustered together i n each of the four appropriate c l u s t e r analy-ses. OTUs that did not c l u s t e r c o n s i s t e n t l y with other taxa, but had v a r i -able a f f i n i t i e s dependent on the conditions of a n a l y s i s , a r e . l i s t e d i n the r i g h t hand columns of the tables. The a f f i n i t i e s of these " f l o a t i n g " OTUs are indicated by dotted l i n e s drawn to the clu s t e r s at the l e f t or to other f l o a t i n g taxa,.where appropriate. I t i s important to note that no set simi-l a r i t y c o e f f i c i e n t value was used to separate f l o a t e r s from consistent c l u s t e r s , or to separate the c l u s t e r s themselves. Therefore, the consistent c l u s t e r s shown i n these tables are somewhat a r b i t r a r y , and they are not d i r e c t l y comparable. • ' In both Tables XI and XII the compositions of consistent c l u s t e r s - 86 -TABLE XI Synopsis of the Results of Four Cluster Analyses Based on Whole Plant.Flavonoids of 30 OTUs 1 Taxa with Consistent A f f i n i t i e s Cluster OTU Taxa with Variable A f f i n i t i e s OTU 1 T bakeri 2 8 s t r i a t a 18 ::floc. ssp. b e l l i n g 14 grac. var. g r a c i l i s 23. f l o e , ssp. fl o c c o s a 9 macounii 10 macounii 4 doug. var. rosea 5 doug. var. sulph. .6.. vinculans. 1 d o u g l a s i i var. d o u g l a s i i ^ —20 f l o e . ssp. pumila 2 doug. var. doug 3 doug. var. nivea 15 grac. var. g r a c i l i s 22. f l o e . var. g r a n d i f l o r a 26 f l o e . var. c a l i f o r n i c s 8 11 alba var. 12 alba var. IT montana 9 27 Floerkea 30 .Floerkea 10. 28 Floerkea .22 . Floerkea — — ~ l 6 grac. var. p a r i s h i i 13 alba var. v e r s i c o l o r 1 f l o e . ssp b e l l i n g e r i a n a __ ^2.4. f l o c c o s a ssp. fl o c c o s a f^2^ f l o c c o s a ssp. floccosa — 21. f l o c c o s a ssp. pumila 1 — The numbers of OTUs analyzed from dri e d material are underscored with dotted l i n e s . 2 — Variable a f f i n i t i e s are.indicated with broken l i n e s drawn between columns and OTUs where appropriate. - 87 -TABLE XII Synopsis of the Results of Four Cluster Analyses Based on Pet a l Flavonoids of 18 OTUs Taxa with Consistent A f f i n i t i e s Cluster OTU . . Taxa with Variable A f f i n i t i e s OTU 11 alba var. alba 12 alba var. alba 14 grac. var. g r a c i l i s 20 f l o e ssp. pumila-13 alba var. v e r s i c o l o r 15 grac. var. g r a c i l i s 17 -T.or.tar.a . ' 3 doug. var. nivea k doug. var. rosea , v l 8 f l o e , ssp. b e l l i n g e r i a n a 1 v s. 1 \ l d o u g l a s i i var. d o u g l a s i i \ ^16 g r a c i l i s var.- p a r i s h i i 2• doug. var. d o u g l a s i i 5 doug. var. sulphurea 9 macounii 7 • bakeri 8 s t r i a t a — "10 macounii 1 -— Variable a f f i n i t i e s are indicated with broken l i n e s drawn between columns and OTUs where appropriate. -•88 -approximately follow-the, recognized family taxonomy: Floerkea i s d i s t i n c t ; OTUs of the Inflexae are not grouped with OTUs of the Reflexae, with the ex-ception that L. s t r i a t a and L_. flo c c o s a ssp. b e l l i n g e r i a n a , OTU 18 , group together i n the whole plant flavonoid analyses; . and us u a l l y OTUs of the same species f a l l i nto the same group. But there are numerous exceptions to t h i s l a s t observation. OTUs of L_. flo c c o s a tend to. group with OTUs of L . g r a c i l i s , rather than together, and the converse i s also true. Limnanthes alba var. v e r s i c o l o r , OTU 13, groups with L_. montana and L_. g r a c i l i s rather than i t s s i b l i n g v a r i e t i e s . There are eight f l o a t i n g OTUs i n Table XI and four f l o a t i n g OTUs i n Table XII. I t i s notable that L_. d o u g l a s i i var. d o u g l a s i i , OTU 1, and L_. g r a c i l i s var. p a r i s h i i , OTU l 6 , which have v a r i a b l e a f f i n i t i e s on the basis of either whole plant or p e t a l flavonoids, were also singled out as anomalies i n the preceding section a f t e r v i s u a l arrangement of.OTUs i n Tables IX-and.X. C. Factor. Analyses, The arrangement of OTUs i n Figures XV and XVI, the drawings of three-dimensional plots based on whole plant flavonoids, generally agrees with the re s u l t s of c l u s t e r analysis.• However, the f a c t o r analyses provide a much better representation of the magnitude and d i r e c t i o n of v a r i a t i o n between OTUs, and between c l u s t e r s of OTUs. The s t r i k i n g feature i n these diagrams i s that the Limnanthaceae i s unexpectedly divided into three d i s t i n c t parts: Floerkea, OTUs 27-30, L. bakeri, OTU 7, and the remaining Limnanthes taxa. Three sub-groups of taxa can be distinguished within the main c l u s t e r of Limnanthes OTUs. These subgroups do not follow the recognized taxonomy of the genus, but gener-a l l y c o r r e l a t e with the-type of breeding system; the most highly evolved outcrossers form a group at one extreme, while.many of the autogamous taxa f a l l into the group at the other extreme. The features o f Figures XV and XVI d i f f e r by the r o t a t i o n of the array - 89 -of OTUs i n space. The differences between these figures i s comparable to the dif f e r e n c e that would be detected i f the array of OTUs were viewed from two d i f f e r e n t points. Figures XVII and XVIII p i c t o r i a l i z e the r e s u l t s of the- f a c t o r analysis of p e t a l flavonoids.- As above, the two diagrams d i f f e r mainly i n perspective. Again, L_. bakeri i s sharply separated from the remainder of the OTUs of Lim- nanthes (flower material of Floerkea was unavailable f o r t h i s a n a l y s i s ) . The remaining OTUs of.Limnanthes form a long continuous grouping. There are several important features of t h i s grouping: l ) L. s t r i a t a l i e s at one extreme, and shows a greater a f f i n i t y , to L_. bakeri than any shown by the other OTUs; 2) a l l the v a r i e t i e s of L_. d o u g l a s i i (OTUs 1-5) and L_. macounii- (OTUs 9 and 10) form- a reasonably t i g h t c l u s t e r near L. s t r i a t a , but below i t ; and 3) the OTUs belonging to the 'section'.Inflexae are a l l d i s t r i b u t e d i n a long c l u s t e r l i n k e d to L_. s t r i a t a , by L_. g r a c i l i s var. g r a c i l i s , OTU 14, and to a l e s s e r extent, by v a r i e t y p a r i s h i i , OTU l 6 . The drawings of f a c t o r analyses of p e t a l flavonoids show the section Inflexae. and a subgroup of section Reflexae to be noh-overlapping groups. However, these groups are joined by a f f i n i t i e s to L_. s t r i a t a and, to a smaller extent, L. g r a c i l i s . For.these features to become apparent, i t i s necessary to v i s u a l i z e ' these two figures i n a l l three dimensions, since the l i n k s to L_. s t r i a t a depend on the a f f i n i t i e s shown by the dimension of v a r i a t i o n per-pendicular to the plane formed by the two h o r i z o n t a l vectors. D. Glycosylation Classes. -Differences i n g l y c o s y l a t i o n pattern between taxa are probably more taxonomically s i g n i f i c a n t than simple differences i n aglycone types within a s i n g l e g l y c o s y l a t i o n c l a s s , providing that the missing aglycones are pre-sent as derivatives of other classes. This i s because the presence or.i.ab-- 90 -sence of the synthetic machinery necessary to attach an a d d i t i o n a l sugar at a new l o c a t i o n may r e f l e c t greater differences between taxa than does the presence or absence of aglycone types within a g l y c o s y l a t i o n s e r i e s . Since a l l OTUs of the Limnanthaceae contain the s i x f l a v o n o l aglycone types, d i f f e r -ences i n g l y c o s y l a t i o n between these taxa may deserve s p e c i a l consideration. Compounds l i s t e d i n Tables IX and X have been grouped so that pre-sence or absence of each g l y c o s y l a t i o n type i s e a s i l y assessed v i s u a l l y f o r each taxon. In t h i s • f a s h i o n OTUs are compared on the basis of 12 characters, or the number of gl y c o s y l a t i o n classes. Although there are differences between OTUs according to g l y c o s y l a t i o n , these differences are s u r p r i s i n g l y few,-and there i s l i t t l e v a r i a t i o n between taxa.- In f a c t , an attempt was made to cl u s t e r OTUs using the 12 g l y c o s y l a t i o n classes as characters. This attempt was unsuccessful, due to lack of v a r i a t i o n . Certain OTUs i n both Tables IX and X are set apart.on the basis of possessing an infrequently occurring g l y c o s y l a t i o n type. As previously i n d i -cated, three of the four OTUs of Floerkea f a l l into t h i s category, since they each possess compound H. Limnanthes s t r i a t a , L. bak e r i , and L_. d o u g l a s i i var. . nivea are a l l l i n k e d by the common occurrence of two g l y c o s y l a t i o n types, c o n s i s t i n g of F and G, and 19,20,-21,22 and 23. Limnanthes s t r i a t a , L. bak e r i , L.. d o u g l a s i i vars. d o u g l a s i i (OTU 2, but not l ) , sulphurea- and rosea, and L. vinculans and L. macounii .(OTU 10', but not 9) a l l • contain. at l e a s t one of the 3 , 7-dirutino-sides. The g l y c o s y l a t i o n class c o n s i s t i n g of the sing l e compound C i s found only i n whole plant material of three OTUs of the section Inflexae; .i.e.. , L. g r a c i l i s var. p a r i s h i i , OTU 16, L./alba var. v e r s i c o l o r , OTU 13, and L. alba var. alba, OTU 11, but not 12. The only other r e s t r i c t e d glycoside c l a s s , composed of compounds D and E, l i n k s L. macounii, OTU 10, with L. alba Var. - 91 -alba, OTUs 11 and 12, L. floccosa ssp. g r a n d i f l o r a and.L_. g r a c i l i s var. g r a c i l i s , OTU 14. , Unfortunately, most compounds belonging to the g l y c o s y l a t i o n classes with r e s t r i c t e d d i s t r i b u t i o n s occurred i n .trace amounts. Consistent presence i n trace amounts introduces a problem of accurately determining presence or absence, a problem that i s d i r e c t l y r e l a t e d to the i n a b i l i t y to completely i d e n t i f y these compounds. Furthermore, compounds c o n s i s t e n t l y appearing i n trace amounts may be overlooked i n some taxa, i f they occur i n concentra-tions too low f o r detection. - 92 -DISCUSSION METHOD AND VALIDITY OF USING FLAVONOIDS AS TAXONOMIC CHARACTERS  A. Flavonoids as Taximetric Characters i . Introduction Both the a p p l i c a t i o n of taximetric techniques and the u t i l i z a t i o n of flavonoids as characters are r e l a t i v e l y new to plant taxonomy. The mating of these two techniques i s more recent and has l e f t unresolved c e r t a i n t e c h n i c a l and p h i l o s o p h i c a l questions associated with t h i s process. To help resolve these questions, the flavonoid data gat ered i n t h i s i n v e s t i g a t i o n -were;; analysed using a v a r i e t y of methods to assess s i m i l a r i t i e s between taxa. i i . I nclusion of Negative Matches The question of whether the mutual absence of a compound contributes to s i m i l a r i t y between taxa i s one that has not been discussed i n reference to flavonoid taximetrics. However, t h i s question has been treated by various authors i n r e l a t i o n to other types of characters. Davis and Heywood (1963) say that mutual absence does not constitute s i m i l a r i t y , while Sokal and Sneath (1963) say that i t does i n some cases. The negative argument says that a l l organisms possess an i n f i n i t e number of mutually absent characters, and that s i m i l a r i t i e s based on such absences are i r r e l e v a n t and r e f l e c t no taxonomic r e l a t i o n s h i p s . The p o s i t i v e argument says that, as long as one of the taxa under consideration exhibits a p a r t i c u l a r character, the other taxa, not pos-sessing t h i s character, are made more s i m i l a r by the common absence. In-at least.two instances, the consideration of negative matches i n comparisons, based on flavonoid characters may r e s u l t i n the misplacement of - 93 -emphasis, and thus lead to unwarranted taxonomic conclusions. I f the absence of a flavonoid compound always r e s u l t s i n the presence of another, or the con-verse i s true, then the,A-presence or absence of these compounds should not be considered independently i n taxonomic comparisons. S i m i l a r i t i e s based on the mutual occurrence of such compounds should not be increased further by the absence of the alternate compounds, since t h i s p r a c t i c e would introduce r e -dundant information into the comparison. Instead, presence or absence of both compounds should be treated as a single character state. I t i s possible that two flavonoids might occur i n t h i s l i n k e d , either-or fashion, i f they each provide an i d e n t i c a l function i n d i f f e r e n t plants. However, whether t h i s s i t u a t i o n a c t u a l l y exists f o r flavonoids cannot- be determined on an a p r i o r i b a s i s , since the functions of most of these compounds are unknown. Unnatural emphasis i s also placed on mutual absence, i f a compound occurs i n none of the taxa being compared. This i s the objection of Davis and Heywood, and i t i s nearly axiomatic, since there i s no taxonomic information content i n such characters, just as there i s none for characters that are c o n s i s t e n t l y present within a group of taxa. However, providing that a f l a v o -noid does occur-.in at l e a s t one of the taxa under consideration, and that the taxa are of p a r a l l e l rank and are being compared simultaneously, the mutual absence of that flavonoid does indi c a t e a s i m i l a r i t y with possible taxonomic s i g n i f i c a n c e . Since r e s o l u t i o n of the above arguments i s not possible without experi-mental evidence, t.axa;-were£clustered using both Jaccard's s i m i l a r i t y coef-f i c i e n t , which does not consider negative matches, and the simple matching c o e f f i c i e n t , which does. I t was suspected that the r e s u l t s of the simple matching c o e f f i c i e n t might possess the greater taxonomic s i g n i f i c a n c e f o r two reasons: l ) c e r t a i n OTUs of the Limnanthaceae such as L. bakeri contain nearly every flavonoid encountered,«which eliminates the p o s s i b i l i t y of any of . - 9h-these compounds occurring on an either-or b a s i s ; and 2) c e r t a i n OTUs have much reduced flavonoid complements that are very s i m i l a r ; thus, s i m i l a r i t i e s between these taxa are higher i f calculated considering negative matches. I n t e r e s t i n g l y , the dendrogram p a i r s , d i f f e r i n g i n type of s i m i l a r i t y c o e f f i c i e n t , are quite s i m i l a r i n each of the four cases. It i s possible to select two dendrograms, one.from the four based on whole plant material and one from the four based on p e t a l material, which best agree with the accepted taxonomic scheme of the Limnanthaceae. For whole plant material the choice i s Figure VII calculated by the Jaccard c o e f f i c i e n t , and for p e t a l the choice i s Figure XII calculated with the simple matching c o e f f i c i e n t . These obser-vations indicate, that the r e s u l t s obtained using either c o e f f i c i e n t are roughly comparable, and i n t h i s case, the choice of which to use i s a r b i t r a r y . i i i . Flavonoid Concentration In comparisons of taxa based on flavonoid compositions, i t i s natural to make.comparisons using as many compounds as possible to contribute the maximum information input to the process. However, at the r i s k of l o s i n g some information, perhaps some compounds are best excluded from consideration i f they occur i n amounts which make detection and/or i d e n t i f i c a t i o n uncertain.,--•the' reason being that the p a r t i a l l o s s of information i s preferable to the i n c l u s i o n of misleading or i n c o r r e c t information. Various flavono'ids occurring, i n the Limnanthaceae c o n s i s t e n t l y were found i n trace amounts. Although detection of these compounds was enhanced by the use of a s e n s i t i v e spray reagent, the determination of compounds made v i s i b l e by t h i s spray i s d i f f i c u l t . Because of t h i s d i f f i c u l t y and because of the p o s s i b i l i t y that 'compounds c o n s i s t e n t l y occurring i n trace amounts might be overlooked i n some taxa, the flavonoid data was analyzed i n two ways: l ) considering a l l occurrences i n a taxon regardless of concentration, - 95 -and,-2) considering only the compounds occurring i n s u f f i c i e n t concentration to he v i s i b l e under UV l i g h t before spraying. The r e s u l t s of c l u s t e r analysis are v a r i a b l e depending on which con-centration of flavonoids was used. Dendrograms based on flavonoids present i n higher concentration more c l o s e l y approximate the accepted taxonomy of the family. Providing that t h i s taxonomy i s b a s i c a l l y c o r r e c t , t h i s r e s u l t indicates that the data based on higher concentrations have greater taxonomic s i g n i f i c a n c e than do the data based on a l l concentrations. However, there i s . no reason to suspect that compounds occurring i n trace amounts are inherently les s s i g n i f i c a n t ; Therefore, the r e s u l t s themselves probably account for t h i s phenomenon, presumably because incorrect or incomplete information was introduced into the analysis by including the trace-amount data. Since the flavonoid data based on higher concentrations have greater taxonomic s i g n i f i c a n c e to the Limnanthaceae, the f a c t o r analyses were performed only on these data. The amount of time involved i n the computation of factor analyses also influenced the decision to omit further analyses including trace-amount data. i v . Lack of V a r i a t i o n among Characters-Each c l u s t e r analysis by the weighted p a i r group method was based on 46 flavonoid characters. However, i n each of these analyses, the number of v a r i a b l e characters was l e s s than 46 depending upon the conditions of a n a l y s i s . Therefore, a c e r t a i n number of nonvarying characters (either p o s i t i v e or nega-t i v e ) were included i n each analysis which increased s i m i l a r i t i e s between taxa. Although the r e s u l t i n g increases are uniform and the c l u s t e r composition does not vary, t h i s procedure i s not completely acceptable from a systematic point of view. The j u s t i f i c a t i o n for t h i s procedure i s that the disadvantage rea-l i z e d by uniformly increasing s i m i l a r i t y c o e f f i c i e n t s was outweighed by the - 96 -convenience provided to the computer operator. Invariant characters cannot be used i n fa c t o r analyses. Therefore, the whole plant f a c t o r analyses were performed on 31 v a r i a b l e flavonoids, and the p e t a l f a c t o r analyses on 36 v a r i a b l e compounds. B. Use of Dry versus Fresh Plant Material Whenever possible OTUs of the Limnanthaceae were analyzed using fresh material. Unfortunately, such material was not a v a i l a b l e f o r a l l taxa, and the remainder were ne c e s s a r i l y analyzed from dr i e d m a t e r i a l . This procedure introduced some problems of i n t e r p r e t i n g comparative r e s u l t s . The f i r s t i n -volved breakdown i n d r i e d material. Fortunately, the f l a v o n o l glycosides pre-sent i n Limnanthes and Floerkea are comparatively stable types, so breakdown eff e c t s were probably n e g l i g i b l e . A more serious problem a r i s i n g from the lack of fresh material was the i n a b i l i t y to complete the studies of p e t a l flavonoids and UV flower photo-graphy. However, the 18 OTUs grown from seed are a good representation of the species and sections of Limnanthes. Fresh material of Floerkea. would have been desirable to balance the i n v e s t i g a t i o n s . The t h i r d problem r e l a t e d to lack of fresh material was the p o s s i b i l i t y that i n d i v i d u a l s of a given taxon may vary i n t h e i r flavonoid compositions. The flavonoid patterns l i s t e d f o r OTUs grown fresh represent the compounds present i n many plants, whereas the flavonoids determined from dr i e d material represent the composition of only one or a few a v a i l a b l e plants. I f i n d i -v i d u a l v a r i a t i o n i s extensive, r e s u l t s based on s i n g l e plants could be taxo-nomically misleading. It i s not possible oh the.basis of the a v a i l a b l e data to completely evaluate the e f f e c t s of comparing i n d i v i d u a l s with groups i n some instances. . However, no discrepancies''-;' were noted f o r the OTUs analyzed from a s i n g l e , or few, d r i e d plants. Apparently the e f f e c t s are minimal, since these OTUs c l u s t e r c o n s i s t e n t l y i n roughly the same frequencies as the other OTUs (Table XI) and according to the accepted taxonomic pattern. C. P e t a l versus Whole Plant Results In an e a r l i e r chapter the suggestion was made that flavonoids are d i r e c t l y responsible for both the nectar guides of L_. douglasii. made v i s i b l e under UV l i g h t and the uniformly high absorption c h a r a c t e r i s t i c s of the flowers of other species. This assumption implies that flower flavonoids serve a u s e f u l function i n outcrossing plants; i-e_. , a t t r a c t i n g and guiding p o l -l i n a t o r s . Therefore, i n outcrossing taxa of Limnanthes these flower pigments w i l l be selected f o r , and eventually, linkage groups determining these and other outcrossing t r a i t s may be formed. I f such linkage groups' are formed, c e r t a i n t r a i t s associated with outcrossing w i l l l i k e l y p e r s i s t f or a time a f t e r the outcrossing habit has been l o s t . The presence i n the flowers of autogamous Limnanthes taxa of pigments with UV absorption c h a r a c t e r i s t i c s comparable to those observed i n the out-crossing taxa indicates that the flower flavonoids are conserved i n the genus. This . conservation of the p e t a l flavonoids has taxonomic s i g n i f i c a n c e which applies d i r e c t l y to the r e s u l t s of t h i s study. I f the p e t a l flavonoids have been s e l e c t i v e l y conserved by evolution, i t i s probable that these compounds w i l l be better i n d i c a t o r s of taxonomic re l a t i o n s h i p s i n Limnanthes than w i l l the whole plant flavonoids unless these compounds also have been conserved f o r some reason. Stebbins (1974) has sug-sested that flavonoids help discourage herbivores and.insects by making vege-t a t i o n unpalatable. However, there i s l i t t l e evidence to support t h i s sugges-t i o n i n the context of Limnanthes. Cluster and f a c t o r analyses based on the p e t a l data more c l o s e l y - 98 -agree with many points of the e x i s t i n g taxonomy of Limnanthes than do the analyses based on the whole plant flavonoids. Providing that the e x i s t i n g c l a s s i f i c a t i o n i s b a s i c a l l y sound, t h i s observation supports the hypothesis that the p e t a l flavonoids are indeed better taxonomic i n d i c a t o r s i n t h i s genus. D. Factor Analysis versus - Conventional Cluster Analysis Because i t should be possible to produce an accurate and natural taxonomic scheme for a group of plants-fusing d i f f e r e n t methods and d i f f e r e n t types of s i g n i f i c a n t information, the recurring features of the taximetric analyses of the Limnanthaceae possess s p e c i a l s i g n i f i c a n c e . A f f i n i t i e s that are co n s i s t e n t l y demonstrated between taxa by these methods l i k e l y r e f l e c t natural r e l a t i o n s h i p s . The consistent c l u s t e r s of OTUs (Tables XI and XII) produced by the weighted p a i r group method demonstrate close s i m i l a r i t i e s i n flavonoid compo-s i t i o n s . These s i m i l a r i t i e s undoubtedly r e f l e c t natural a f f i n i t i e s between OTUs i n these c l u s t e r s . However, many OTUs do not f a l l i n t o consistent c l u s -t e r s . In the whole-plant flavonoid analyses, L_. bakeri never c l u s t e r s with other taxa and has been treated as a separate group. Other nonclustering OTUs f l o a t between consistent c l u s t e r s depending on the method of data analy-s i s . What i s the s i g n i f i c a n c e of t h i s f l o a t i n g phenomenon? There are at l e a s t two possible explanations f o r the f l o a t i n g pheno-menon: l ) the evidence used to c l u s t e r taxa i s i n v a l i d or i n s i g n i f i c a n t ; 2) the flavonoid compositions of the f l o a t i n g OTUs are intermediate, and these OTUs serve as l i n k s between c l u s t e r s . Thus, as the conditions of analy-s i s change, OTUs with intermediate compositions show v a r i a b l e a f f i n i t i e s . On the basis of the dendrograms alone, i t i s impossible to determine which of these explanations i s correct. - 99 -Because the f a c t o r analyses ordinate taxa by arranging them i n three-dimensional space ( i n t h i s study) at distances that are proportional to v a r i a t i o n between them, these techniques make a t possible to estimate super-imposed, branched r e l a t i o n s h i p s between OTUs. Such multi-branched configura-tions can be explained i f groups have repeatedly diverged from an ancestral l i n e . This pattern of evolution has apparently taken place i n the Limnan-thaceae, and i t i s f o r t h i s reason that some taxa were consistent f l o a t e r s . They are indeed l i n k s between divergent groups. Although Sokal and Sneath (1963) reported that there i s generally good agreement between the r e s u l t s of f a c t o r analysis and c l u s t e r i n g by the weighted p a i r group method, a comparison of the r e s u l t s obtained by these two techniques i n the present i n v e s t i g a t i o n indicates that t h i s consistency depends on how the v a r i a t i o n between taxa.is expressed. I f there i s more than one uncor-r e l a t e d d i r e c t i o n of v a r i a t i o n expressed between taxa, a one-dimensional c l u s -t e r i n g cannot adequately convey the r e l a t i o n s h i p s between them. Therefore, since most of the •••variation present i n the Limnanthaceae cannot be accounted fo r i n one dimension, r e l a t i o n s h i p s cannot be properly assessed within t h i s group unless a multi-dimensional c l u s t e r i n g technique i s used. E. Flavonoid Differences between Duplicate Taxa In many instances two or more OTUs were analyzed that were i d e n t i f i e d as the same species, subspecies or v a r i e t y . By doing so, i t was possible to estimate the approximate flavonoid v a r i a t i o n between populations or recognized taxonomic e n t i t i e s of Limnanthes and Floerkea. The two OTUs of L. macounii group together as would be predicted. The same was true f o r the two OTUs of g. laTbajgvar. alba. OTUs of L. floccosa genera-a l l y grouped together, i f somewhat d i f f u s e l y , i n the factor analyses but not i n the c l u s t e r analyses. Although the s i m i l a r i t i e s were not as great between-- 100 -the four OTUs of F. proserpinacoides, these taxa also grouped together i n three out of four c l u s t e r analyses and the fac t o r analyses. Since these four OTUs of Floerkea were chosen to represent diverse geographic elements of the species, elements that have probably been i s o l a t e d f o r some time, a c e r t a i n degree of v a r i a t i o n was expected a t t r i b u t a b l e to divergence caused by genetic d r i f t or d i f f e r e n t s e l e c t i o n pressures. Differences i n flavonoid patterns of the duplicate OTUs of L. doug- l a s i i and L. g r a c i l i s are both s i g n i f i c a n t . Limnanthes d o u g l a s i i i s unusual i n two respects. F i r s t , OTU 1 only shows a f f i n i t i e s to it's r e p l i c a t e , OTU 2, i n the p e t a l f a c t o r analyses. In the whole-plant c l u s t e r analyses and whole-plant factor analyses i t often c l u s t e r s with more or l e s s d i s t a n t l y r e l a t e d autogamous d e r i v a t i v e s . This si n g l e observation provides further evidence that\) the p e t a l flavonoids are-conserved to a greater degree than the whole plant flavonoids. I t i s possible to p a r t i a l l y explain the strange a f f i n i t i e s of OTU 1. These plants may be descendants of plants taken to Europe by David Douglas i n the early l830's. and have since been maintained as a h o r t i c u l t u r a l v a r i e t y . Therefore, over the i n t e r v a l of about 140 years (probably somewhat fewer gener-a t i o n s ) , the flavonoid composition of t h i s OTU has d i f f e r e n t i a t e d from other members of t h i s v a r i e t y that n a t u r a l l y occur i n the western United States, even though the two remain.very s i m i l a r morphologically. This d i f f e r e n t i a t i o n may have r e s u l t e d from the processes of inbreeding, genetic s e l e c t i o n , or d r i f t . Because of the magnitude of the differences between OTU 1 and the na t u r a l l y occurring OTUs of L_. d o u g l a s i i , which c l u s t e r c o n s i s t e n t l y with each other, i t i s probable that inbreeding or some unknown form of a r t i f i c i a l s e l e c t i o n has played a greater r o l e than random d r i f t i n causing t h i s pheno-menon. Because OTU 1 clu s t e r s with autogamous plants i n the whole plant taximetric analyses, i t i s possible that inbreeding may lead to dire c t e d - 101 -changes, i n whole plant flavonoid composition. Why there should he such a d i r e c t i o n i n flavonoid change associated with autogamy or inbreeding i s not apparent. The r e s u l t s of the c l u s t e r analyses and factor analyses i n d i c a t e that there i s great v a r i a t i o n i n the flavonoid patterns within both L. g r a c i l i s and L. d o u g l a s i i . The observed v a r i a t i o n i n both of these species i s much greater than that observed within other species of Limnanthes, with the po s s i -ble exception of L. floc c o s a . Providing that the divergence i n flavonoid compositions between segments of a species (or any r e l a t e d group of plants) r e s u l t s mainly from the random process, of genetic d r i f t or from comparable s e l e c t i o n pressures, the extent of the divergence between elements of the species d o u g l a s i i , g r a c i l i s and flo c c o s a has implications with regard to both the r e l a t i v e ages of these species, and to ranks of the components of these species. TAXONOMY' OF THE LIMNANTHACEAE  A. Introduction The currently accepted taxonomic scheme of the Limnanthaceae i s a s l i g h t l y expanded form of the system proposed by Mason i n 1952 (Table I ) . There are three primary features of t h i s scheme, l ) Floerkea i s d i s t i n -guished from Limnanthes by numbers of f l o r a l parts and by having hypogeal • cotyledons. 2) The genus Limnanthes i s s p l i t into two sections, Reflexae and Inflexae, based on p e t a l p o s i t i o n a f t e r p o l l i n a t i o n . 3) Various Limnanthes taxa, including some formerly recognized as d i s t i n c t species with clearcut morphological and geographical d i f f e r e n c e s , are reduced to v a r i e t a l or sub-s p e c i f i c rank l a r g e l y on the basis of i n t e r ' f e r t i l l t y i n a r t i f i c i a l crosses. Plant taxonomists have been reluctant.to revise c l a s s i f i c a t i o n schemes arr i v e d at by biosystematic methods when c o n f l i c t i n g evidence i s produced by - 102 -phenetic analyses. The type of evidence on which the phenetic analysis i s based, whether morphological or chemical, does not seem to matter; the f e e l -ing i s that biosystematic studies more c l e a r l y indicate phylogenetic r e l a -t i o n s h i p s . Mason's treatment of Limnanthes, based on morphological and c y t o l o g i -c a l comparisons, and the r e s u l t s of interbreeding t r i a l s , i s generally con-sidered to be an accurate phylogenetic arrangement of the genus. However, not a l l of the a v a i l a b l e evidence supports t h i s conclusion. Of the three type of evidence considered by Mason, his conclusions depend most heavily on the r e s u l t s of a r t i f i c i a l crosses. His other r e s u l t s do not n e c e s s a r i l y support his arrangement of the genus. Since there was no v a r i a t i o n i n karyotype mor-phology of the genus, Mason's r e v i s i o n of Limnanthes was not inf-luenced by his c y t o l o g i c a l r e s u l t s . Other morphological treatments and studies of the genus c o n f l i c t with Mason's conclusions. Previous authors (Howell, I89T; Rydberg, • 1910; Jepson, 1936) considered several of Mason's v a r i e t i e s to be d i s t i n c t species based on morphological differences. In a d d i t i o n , the phenetic a n a l y s i of Ornduff and Crovello (1968), based on morphological characters, generally does not support a l l the d e t a i l s of Mason's taxonomic scheme. Therefore, the net e f f e c t has been to attach s p e c i a l s i g n i f i c a n c e to the r e s u l t s of h y b r i d i -zation t r i a l s i n Limnanthes. Three a d d i t i o n a l observations suggest that the phyl'ogenetic s i g n i f i -cance of experimental crossing t r i a l s i n Limnanthes may need reevaluation. l ) Ornduff (197T) has reported that genetic i s o l a t i n g mechanisms apparently have ar i s e n between Limnanthes taxa,' not on the basis of distance of r e l a -t i o n s h i p , but rather on a purely s p a t i a l b a s i s ; the more geographically d i s t i n c t two taxa are, the more l i k e l y that no genetic b a r r i e r to crossing w i l l have developed between them. 2) UV flower photography demonstrates that an i s o l a t i n g mechanism of an e c o l o g i c a l nature may exist between v a r i e t i e s - 103 -of L. d o u g l a s i i , and that a r t i f i c i a l i n t e r f e r t i l i t y does not preclude the existence of such a mechanism. 3) The flavonoid data c o n s i s t e n t l y disagrees with c e r t a i n key features of the e x i s t i n g c l a s s i f i c a t i o n . There i s evidence i n d i c a t i n g that the taxonomic status of Floerkea requires reevaluation. l ) Since the o r i g i n a l d e s c r i p t i o n of the Limnanthaceae, there has "been disagreement regarding whether the morphological differences i n the family are.great enough or discontinuous enough to warrant recognition of two genera. 2) The phenetic analysis of Ornduff and Crovello (1968) does not support the maintenance of separate genera. 3) The c y t o l o g i c a l characters of Floerkea do not d i s t i n g u i s h i t from Limnanthes (Ornduff, 1971)- 4)' The flavonoid data does not support the separation of Floerkea and Limnanthes. The phenetic s i m i l a r i t i e s between OTUs based on flavonoids provide new evidence that may be applied to the systematics of the Limnanthaceae. Because the flavonoid evidence stands independently of the r e s u l t s of other taxonomic treatments, t h i s evidence provides the basis for a new c l a s s i f i -c ation of the family. But a key question must be answered regarding the s i g -n i f i c a n c e of the flavonoid evidence: do the r e s u l t s of the flavonoid analysis warrant any new conelus'ions"."? or require a r e i n t e r p r e t a t i o n of r e l a t i o n s h i p s within the family? The demonstration of flavonoid a f f i n i t i e s does j u s t i f y the r e v i s i o n of a c l a s s i f i c a t i o n providing that two conditions are met: l ) other a v a i l -able evidence supports such a r e v i s i o n , and 2) the flavonoid r e s u l t s are un-equivocal. Both of these conditions are f u l f i l l e d i n the context of the Limnanthaceae. A comparison of the flavonoid evidence with the relevant evidence from a l l other sources has l e d to a reconstruction of a hypothetical evolutionary h i s t o r y of the Limnanthaceae and a new proposed c l a s s i f i c a t i o n of the family. In the following sections, the .proposed h i s t o r y and c l a s s i f i c a t i o n w i l l be - 104 -presented f i r s t , followed by a discussion of the evidence f o r and against these proposals. F i n a l l y , the need for taxonomic r e v i s i o n w i l l be discussed, and a synopsis of the family w i l l be presented including the necessary nomen-c l a t u r a l changes. B. Hypothetical Evolutionary History of the Limnanthaceae ' The information presented i n t h i s t h e s i s , taken together with the a v a i l a b l e information regarding d i s t r i b u t i o n s and e c o l o g i c a l requirements of members of the Limnanthaceae, has many implications concerning the family's h i s t o r y . These implications have made i t possible to reconstruct a hypotheti-c a l h i s t o r y which i s presented below. However, t h i s h i s t o r y i s purely specu-l a t i v e since i t s evaluation requires nonexistent f o s s i l information. In the early T e r t i a r y , Limnanthes probably c l o s e l y resembled some of i t s modern descendents. I t was a predominantly outcrossing spring annual, with a haploid number of f i v e , and had medium sized white flowers and compound leaves with s l i g h t l y divided l e a f l e t s . Since the grassland s i t e s occupied by most modern counterparts had not yet appeared, Limnanthes was probably a streamside herb of the woodlands, occurring at moderate elevations, and pro-ducing a determinate number of r e l a t i v e l y large n u t l e t s . I t s distribution'.-' was probably reasonably extensive i n western North America, and perhaps i t extended across the continent. With the u p l i f t of the coastal mountain ranges, the population of Limnanthes probably was s p l i t i n two. I t i s l i k e l y that occasional trends to autogamy were ongoing i n early elements of Limnanthes, as they are today. One such element, g e n e t i c a l l y i s o l a t e d by autogamy, and better adapted to the cooling trend of the T e r t i a r y , was probably able to p e r s i s t further north than the parent population. This autogamous^ Limnanthes r a p i d l y evolved reduced flowers with accompanying t r a i t s such as cleistogamy. I t s dissemination was probably somewhat improved, and - 105 -i t may have become s l i g h t l y weedy. Three modern members of the Limnanthaceae most c l o s e l y resemble the three postulated ancestral elements. The autogamous d e r i v a t i v e i s represented by F. proserpinacoides which s t i l l occupies the deciduous woodland habitat of i t s ancestor. Since the s e l e c t i o n pressures associated with t h i s habitat have probably stayed f a i r l y constant, t h i s species probably c l o s e l y resembles the ancestor s p l i t o f f m i l l i o n s of years ago. The secondmajor element, i s o -l a t e d to the east of the coast range, i s represented today by_L_. s t r i a t a , a species which may s t i l l occupy moderate elevation streamside s i t e s . The western element.'.persisting from the s p l i t i s represented today only by L_. bakeri, a narrow endemic of C a l i f o r n i a associated with moderately high elevation vernal pools. With the gradual cooling that occurred during the Miocene Epoch, each of the three ancestral elements of Limnanthes was pushed southward. In i t s southward migration, the population represented today by L_. s t r i a t a was divided by the c e n t r a l v a l l e y of C a l i f o r n i a , one part proceeding, down the west side of the newly u p l i f t e d S i e r r a Nevadas, and the other down the east side of the coast range. Elements of t h i s migration apparently p e r s i s t today. Although the main range of L. striata:, i s east of the c e n t r a l v a l l e y i n the S i e r r a Neva-das, Ornduff ( O r n d u f f . C r o v e l l o , 1968; 1971) has recently discovered a v a r i -ant western element of t h i s species i n the northern C a l i f o r n i a coastal moun- . t a i n s , which he at f i r s t thought was a new species and r e f e r r e d to as " t r i n i t y " (from T r i n i t y County). I t i s l i k e l y that Limnanthes and.Floerkea p e r s i s t e d v i r t u a l l y unchanged as woodland plants through the T e r t i a r y u n t i l the comparatively recent develop-ment of grasslands i n the Pleiocene Epoch. However, with the a v a i l a b i l i t y of these new s i t e s , i t i s conceivable that elements of L_. s t r i a t a invaded and r a p i d l y evolved adaptations appropriate to the new l i f e s t y l e . In c o a s t a l - io6 -areas, the grassland invaders were probably the forerunners of the Reflexae, characterized by the L." d o u g l a s i i type, while the eastern invaders were proba-b l y the ancestors of the Inflexae, characterized by the L. g r a c i l i s type. The oldest western elements occupying the range west of the coast (mountains were apparently unable to compete s u c c e s s f u l l y in.the new s i t e s and have i a r g e l y disappeared with the exception of L_. bakeri. In the comparatively modern Pleistocene, three repeating processes may have taken place which account, f o r both the d i s t r i b u t i o n s and d i v e r s i t y of extant forms of Limnanthes. l ) A l l populations, including Floerkea, prob-•gably have advanced and retreated north and south with vegetational changes accompanying periods of warming and cooling associated with g l a c i a t i o n . During the cool periods, continuous populations may have been s p l i t into many groups. 2) Geographically i s o l a t e d populations may have become g e n e t i c a l l y d i s t i n c t due to ra p i d adaptations to the d i f f e r i n g niches, made a v a i l a b l e by the new grassland habitats. In t h i s fashion populations of the sections Reflexae and Inflexae may have diverged on an e c o l o g i c a l basis and according to a l t i t u d i n a l and l a t i t u d i n a l zonation. This process could have r e s u l t e d i n various new species, subspecies or v a r i e t i e s . 3) The recurring trend to autogamy i n Limnanthes, coupled with the generally better tolerance of these derivatives of s l i g h t l y cooler conditions, probably has l e d to the present day persistence of these plants i n s i t e s marginal for most outcrossing taxa of Limnanthes. I f so, these plants can be interpreted as r e l i c s of e a r l i e r northward migrations.• I t i s l i k e l y that rapidly, occurring adaptations to the modern grass-land habitats i n the •'Reflexae. hare l e d to the creation of L. vinculans and the four v a r i e t i e s of L_. d o u g l a s i i . S i m i l a r d i f f e r e n t i a t i o n i n the Inflexae may have l e d to the creation of L_. montana,. the v a r i e t i e s of L_. g r a c i l i s , the v a r i e t i e s of L. alba and the v a r i e t i e s of L. floc c o s a . The persistent auto-- 107 -gamous r e l i c s of the northward migrations are.L. macounii, probably derived from an ancestor resembling- L. douglasii,:'arid the subspecies of L_. f l o c c o s a , derived from an ancestral form of t h i s species. Figure XIX, the hypothetical phylogenetic tree of the Limnanthaceae, summarizes the h i s t o r y of the family. No attempt has been made to, i n d i c a t e which of the: •species, L. bakeri or F. proserpinacoides, f i r s t diverged from the main l i n e of Limnanthes. The amount of divergence from t h i s main l i n e i s roughly comparable for both, and i t i s probable that the s p l i t happened at approximately the same time. Sectional divsions have.been made i n Figure XIX, and these w i l l be discussed below. C. Assumptions that Led to the Hypothetical History Processes of evolution i n Limnanthes have made i t d i f f i c u l t to under-stand r e l a t i o n s h i p s i n t h i s group. Certain features have been strongly con-served such as karyotype, f r u i t type, and basic f l o r a l structure. However, other features may have changed r a p i d l y including number of f l o r a l p arts, flower and plant s i z e , breeding system, and l e a f l e t morphology. These changes r e s u l t either from adaptations to new niches, or from the loss of the out-crossing habit. Because of evolutionary convergence, taxa possessing common t r a i t s of the sort e a s i l y modified by evolution may not necessar i l y have natural a f f i n i t i e s . The use of flavonoid characters f i l l s an obvious gap i n evaluating the taxonomy of the Limnanthaceae. C o n f l i c t s i n previous taxonomic conclu-sions r e s u l t from d i f f e r e n t i n t e r p r e t a t i o n s of incomplete or i n s u b s t a n t i a l information. The flavonoid data presented i n t h i s t h e s i s i s taxonomically useful because i t b a s i c a l l y supports the taxonomy of the group, as f a r as i t i s presently known, and simultaneously, provides a d d i t i o n a l information by which the' present system can reevaluated and re f i n e d . FIGURE XIX Hypothetical Phylogenetic. Tree of the Limnanthaceae Section Line Taxon Section Hypothetical Ancestor bakeri s t r i a t a . macounii doug. douglasii doug. sulphurea vinculans doug. rosea  doug. nivea grac. p a r i s h i i alba alba  alba v e r s i c o l o r  montana  grac. g r a c i l i s f l o e . b e l l i n g e r i a n a  f l o c . floccosa j — f l o e . g r a n d i f l o r a 1—floe, c a l i f o r n i c a f l o e . pumila proserpinacoides Bakera Limnanthes Reflexae Inflexae Floerkea - 109 -The taxonomic implications of the flavonoid data are a l l r e l a t e d to amount of divergence i n flavonoid patterns between taxa. Several assumptions must be made regarding t h i s divergence. The f i r s t i s that divergence i n flavonoid pattern from that of a common ancestor i s a random process that arises from genetic d r i f t ; or-, i f the changes are caused by s e l e c t i o n pres-sures, these pressures w i l l be much the same for a l l taxa. This assumption probably cannot be made, for h o r t i c u l t u r a l v a r i e t i e s which are the products of a r t i f i c i a l s e l e c t i o n . The second assumption i s that the rate of divergence of flavonoid com-p o s i t i o n i s approximately uniform f o r c l o s e l y r e l a t e d outcrossing annuals, such as most taxa of Limnanthes. A f t e r conversion to autogamy, i t i s probable that the rate of divergence w i l l slow down with the accompanying loss of heterozygosity. However, sub s t a n t i a l amounts of v a r i a t i o n were detected i n a l l four completely autogamous members of the family. This v a r i a t i o n i n d i -cates that divergence does occur i n autogamous plants, and that the conver-sion to autogamy does not preclude the retention of some v a r i a b i l i t y . However, whether v a r i a t i o n continues to increase i n d e f i n i t e l y at the same rate i n these autogamous taxa i s impossible to determine on the basis of the a v a i l a b l e data. The t h i r d assumption i s c l o s e l y r e l a t e d to the second. Divergence i n flavonoid composition proceeds at a slower rate i n the petals than i n the rest of the plant. This conservation i n the petals r e s u l t s d i r e c t l y from the a b i l i t y of these compounds to a t t r a c t . p o l l i n a t o r s . Once'the above three assumptions are accepted, four conclusions can be drawn which suggest trends i n the evolution of Limnanthes. The f i r s t i s that the amount of divergence between two taxa i s roughly proportional to the length of time that has passed since these taxa were derived from a common ancestor. On t h i s b a s i s , F_. proserpinacoides and L. bakeri each separated from Limnanthes long before the creation of any of the remaining taxa, with - 110 -the possible exception of L_. s t r i a t a . The second conclusion i s that taxa with much i n t e r n a l flavonoid v a r i -ation are more ancient than those with l i t t l e v a r i a t i o n . For instance, the amount of v a r i a t i o n i n L. g r a c i l i s i ndicates that t h i s taxon i s r e l a t i v e l y more ancient than L_. alba. Likewise, the roughly comparable t o t a l v a r i a t i o n present i n members of the sections Inflexae and Reflexae- (excluding the species s t r i a t a and bakeri) indicates that these supraspecific taxa are of approxi-mately the same age. The t h i r d conclusion i s that divergence i n autogamous derivatives of the same magnitude as that present i n outcrossers probably indicates a more ancient o r i g i n of the autogamous plants. For t h i s reason the v a r i a t i o n i n the populations of Floerkea, an amount nearly as great as that observed for a l l taxa of Limnanthes (excluding Ly b a k e r i ) , indicates that F. proserpinacoides has been a d i s t i n c t taxon since well before the evolution of most other family members. The fourth conclusion i s that the implications of p e t a l flavonoids are more taxonomically s i g n i f i c a n t than those of the whole plant flavonoids^ and thus suggest a more natural c l a s s i f i c a t i o n system. This conclusion i s supported by' the separation of the section Inflexae from the Reflexae (with-out L_. s t r i a t a and L_. bakeri) by the flavonoid analyses of petals but not of whole plant material. Although the whole plant flavonoid analyses do not separate the two sections recognized by Mason, they c l e a r l y d i s t i n g u i s h Floerkea and the s i n g l e Limnanthes species bakeri from the main family l i n e . Like the flavonoid evidence, the present day d i s t r i b u t i o n s of a l l members of the Limnanthaceae support the hypothetical h i s t o r y of the group and were u s e f u l i n formulating t h i s theory. I t i s true that d i s t r i b u t i o n s may change r a p i d l y and thus suggest misleading evolutionary conclusions. How-ever, the close agreement between flavonoid and d i s t r i b u t i o n a l evidence - I l l -suggests that present day locations are s i g n i f i c a n t clues u s e f u l i n t r a c i n g the family h i s t o r y . Great emphasis has been placed on the d r i v i n g force for evolutionary change created by r e l o c a t i o n i n new grassland niches. Most modern taxa of Limnanthes can be regarded as recent derivatives moulded by t h i s force. Con-ver s e l y , certain.extant family members have retained the more ancient stream-side or woodland habitat and exist today probably l i t t l e changed f o r m i l l i o n s of years. Here, the forces of natural s e l e c t i o n have probably stayed r e l a -t i v e l y constant, and.changes are l a r g e l y the r e s u l t of genetic d r i f t . Several pieces of evidence suggest that L_. s t r i a t a probably most c l o s e l y resembles the ancestral Limnanthes type: l ) the whole-plant f l a v o -noid f a c t o r analysis shows t h i s s p e c i e s . c e n t r a l l y located between Floerkea and L_. baker 1; 2) the p e t a l flavonoid f a c t o r analysis shows L_. s t r i a t a forming a common l i n k between the section Inflexae and a reduced version of the section Reflexae, while simultaneously showing a greater a f f i n i t y to L_. bakeri than any other taxon; 3) L. s t r i a t a i s the only family member, except F_. proser- piriacoides and L_. montana,, which may occupy a streamside habitat rather than the newer grassland.sites. As discussed above the streamside habitat pro-bably represents the ancestral condition. The continued presence of a species i n t h i s type of s i t e might tend to minimize evolutionary change, k) The d i s -covery of a f a r removed, morphologically varied disjunct of t h i s species sug-gests that i t indeed has a comparatively ancient o r i g i n . D. Comparison of the Proposed C l a s s i f i c a t i o n to the.Existing C l a s s i f i c a t i o n  i . Above the' Species Level' The c l a s s i f i c a t i o n presented i n Figure XIX i s l a r g e l y the product of factor; and c l u s t e r analyses'of flavonoid data. However, t h i s scheme was de-v i s e d to provide the'best f i t u t i l i z i n g a l l the a v a i l a b l e evidence, including . - 112 -that presented by Mason, Ornduff and Arroyo. Since a l l of the a v a i l a b l e . information was used, and a l l family members were considered simultaneously, the new c l a s s i f i c a t i o n probably more c l o s e l y approximates actual r e l a t i o n -ships i n the Limnanthaceae than d i d the old scheme. The hypothetical h i s t o r y and c l a s s i f i c a t i o n of the Limnanthaceae depart from the accepted taxonomy of the family i n c e r t a i n key points. How-ever, both systems share many features, and the new scheme i s a c t u a l l y a r e -finement of the old. The ,major differences a r i s e because of d i f f e r i n g i n t e r -pretations 0% group ranks. This presents a problem, since i n l i g h t of new information^ taxa.with comparable a f f i n i t i e s should be adjusted so that they are'of p a r a l l e l rank. The proposed c l a s s i f i c a t i o n of the Limnanthaceae has been drawn.with f i v e s e c t i o n a l d i v i s i o n s . An a l t e r n a t e , but le s s d e s i r a b l e , choice would have been d i v i s i o n of the family into three sections or three genera con s i s t i n g of Floerkea, Limnanthes b a k e r i , and the remaining taxa of Limnanthes. From the flavonoid evidence there i s l i t t l e doubt that both L. bakeri and F. proserpinacoides have diverged from the remaining family members to a comparatively great extent. This p a r a l l e l divergence leads to a necessary conclusion; i f Floerkea i s recognized as a genus apart from Limnanthes, then L. bakeri should also be so recognized. However, the morphological differences between the three groups are r e l a t i v e l y t r i v i a l , and L_. bakeri has been demon-strated to, be i n t e r f e r t i l e with various other members of Limnanthes. This i n t e r f e r t i l i t y makes a three generic s p l i t of the family untenable. Therefore, the family consists of only one genus. Although the accepted c l a s s i f i c a t i o n of the Limnanthaceae recognizes two genera, i t i s u n l i k e l y that many taxonomists w i l l object to the i n c l u s i o n of Floerkea i n the same genus with Limnanthes. A l l the hard.data suggests that they are not separate genera. The morphological d i f f e r e n c e s , including - 113 -loss of flower parts, are e a s i l y accounted for by the conversion to autogamy, p a r t i c u l a r l y since p a r a l l e l changes have taken place i n L_. macounii. Only one good morphological t r a i t separates.Floerkea from Limnanthes; Floerkea i s -hypogeous, while Limnanthes i s epigeous. Ornduff and Crovello (1968) have suggested that t h i s t r a i t alone does not constitute s u f f i c i e n t grounds f o r the maintenance of two genera, and I agree. Although extensive interbreeding t r i a l s have been made between a l l Limnanthes taxa, apparently no attempts have.been made to cross Floerkea with Limnanthes. I f such a cross- produced f e r -t i l e hybrids, t h i s would further support the j o i n i n g of these two taxa into a s i n g l e genus. Unfortunately, the t i n y cleistogamous flowers of F_. proser- pinacoides would make a r t i f i c i a l cross p o l l i n a t i o n s extremely d i f f i c u l t . Since the family Limnanthaceae has probably evolved from three d i s -t i n c t p h y l e t i c l i n e s of approximately equal rank, the.decision to recognize f i v e sections instead ofj three was based on convenience and t r a d i t i o n . The two sections, Inflexae and Reflexae, recognized by Mason are most probably natural groupings pending the removal of L_. s t r i a t a and L_. bakeri from the Reflexae. Since Mason's two groups have been w e l l accepted, and contain a l l but three of the species i n the family, i t i s inappropriate to merge these two sections: a merger that would be necessary, i f the family were to be divided into three groups of p a r a l l e l evolutionary rank. Having made the decision to preserve Mason's se c t i o n a l d i v i s i o n to the maximum extent allowed by the new evidence, i t i s necessary to erect a section i n addi t i o n to the Inflexae and Reflexae c o n s i s t i n g only of L. s t r i a t a . This section i s named Limnanthes, since i t i s probable that L_. s t r i a t a most c l o s e l y resembles the ancestral family type. This d i v i s i o n into three sections keeps the Reflexae, Inflexae and Limnanthes p a r a l l e l and i s a necessary step since these three probably evolved from a common ancestor some time a f t e r the separation of sections Floerkea (containing F. proserpinacoides) and Bakera - 114 -(containing L_. b a k e r i ) . Since the f i v e groups diverged at d i f f e r e n t times, they are not completely p a r a l l e l i n rank. However, i n view of the circum-stances, a d i v i s i o n of the family into f i v e sections i s the most desirable among the a v a i l a b l e a l t e r n a t i v e s . The old , i f;two-sectional d i v i s i o n of Limnanthes i s supported, by two pieces of evidence: l ) p e t a l p o s i t i o n a f t e r p o l l i n a t i o n , and 2) the presence of an i n t e r s e c t i o n a l b a r r i e r to h y b r i d i z a t i o n . This evidence can equally w e l l be interpreted to support the proposed f i v e - s e c t i o n d i v i s i o n . P e t a l p o s i t i o n a f t e r p o l l i n a t i o n i s by i t s e l f a t r i v i a l character, . . probably c o n t r o l l e d by one or a few a l l e l e s . I t i s possible that i n f l e x i o n of petals might have r e s u l t e d from a s i n g l e mutation of an ancestor with the r e f l e x e d p e t a l type. Whether such a mutation might have.happened.more than once i s hard to determine, but the a v a i l a b l e evidence indicates that i t proba-b l y has not. The following i s a hypothesis attempting to explain why there may be a s e l e c t i o n pressure for maintenance of the i n f l e x e d flower type i n grassland-inhabiting Limnanthes taxa. I t i s possible that the p e t a l i n f l e x i o n , which tends to r e t a i n nutlets i n the flower, was selected f o r and maintained i n the grassland invading species of Limnanthes since t h i s t r a i t might a i d dissemination of nutlets by h e r b i -vores or other animals. I f t h i s hypothesis i s c o r r e c t , L_. floccosa. i s the most highly evolved member of the section Inflexae i n t h i s regard since t h i s spe-cies has also evolved i n f l e x i n g sepals, and.most subspecies possess an ab-s c i s s i o n layer below the calyx which tends to cause d i s p e r s a l of the flower as a u n i t . Dispersal of nutlets by animals, either by ingestion or by the flowers c l i n g i n g to f u r , may overcome one problem of dissemination i n Limnanthes. The comparatively large nutlets of t h i s genus are of a s i z e normally associated with woodland plants (Salisbury, 1942; Stebbins, 1974). . Although apparently - 115 -necessary for competition i n shady s i t e s , grassland.invaders would he put at a disadvantage by t h i s t r a i t , and thus compensating, mechanisms-./might;, evolve. Besides i n f l e x i o n of:;flower parts, natural s e l e c t i o n probably has favored increased production of seed by grassland inhabitants to provide the p o t e n t i a l f o r rapid c o l o n i z a t i o n of these newer s i t e s . In t h i s regard, i t i s s i g n i f i c a n t that Higgins et .al. ( l 9 T l ) found that L_. s t r i a t a and L_. bakeri were more determinate i n t h e i r flowering than other Limnanthes taxa. This shared character implies that both of these plants are more p r i m i t i v e i n t h i s respect than other members of the genus. How^well do the r e s u l t s of a r t i f i c i a l crosses support d i v i s i o n of the Limnanthaceae into the f i v e sections, Floerkea,. Bakera, Limnanthes, Reflexae and.Inflexae? ' Because of b a r r i e r s to crossing, the Inflexae i s d i s t i n c t . Since no interbreeding t r i a l s have been conducted using Floerkea as a parent, no conclusions can be drawn.regarding t h i s section. The sections Bakera, Reflexae and Limnanthes are p a r t i a l l y i ' n t e r f e r t i l e . However, the successful crosses between L_. macounii, L_. vinculans and the v a r i e t i e s of L. d o u g l a s i i generally produce more f e r t i l e hybids than those between these taxa and L. s t r i - ata. (Ornduff, 1971). Also, hybrids between L. bakeri and the previous taxa showed the' lowest f e r t i l i t y . On the basis of flavonoid evidence a strong case can be made.for • t r e a t i n g the' sections Limnanthes and Bakera as groups d i s t i n c t from the Reflexae. In a d d i t i o n , other evidence does not contradict t h i s separation. One ad-d i t i o n a l piece of evidence, presented e a r l i e r i n t h i s t h e s i s , also supports such a separation. Limnanthes macounii and the v a r i e t i e s of L_. d o u g l a s i i have each evolved UV-visible f l o r a l patterning. This character holds these f i v e taxa together, just ,as i t ' separates L. bakeri and L — s t r i a t a from t h i s group. - 116 -i i . Species Level and Below Three of the f i v e newly proposed sections, Bakera, Limnanthes and Floerkea ,'r are monotypic , containing the species L. bakeri, L. s t r i a t a and F. proserpinacoides, r e s p e c t i v e l y . Each of these three taxa has t r a d i t i o n -a l l y been regarded as a d i s t i n c t species. Certain elements of doubt, either have been, or w i l l now be expressed concerning the accepted status and rank of the remaining Limnanthes taxa comprising the sections Inflexae and Reflexae. In the intervening years since Mason published h i s scheme, one of his- conclusions that has been questioned in v o l v e s . r e l a t i o n s h i p s between the three species L_. montana, L_. alba and L_. g r a c i l i s . Mason hypothesized that the taxa, L_. g r a c i l i s vars. g r a c i l i s and p a r i s h i i , and L_. montana, once were part of a continuous population of a s i n g l e species which became discontinu-ous and diverged into the three d i s t i n c t taxa. Mason accorded L_. montana spe-c i f i c status since i t was not i n t e r f e r t i l e with the other two taxa which he recognized as v a r i e t i e s of L. g r a c i l i s . Gentry and M i l l e r (1965) thought that t h i s was an i n c o s i s t e n t treatment and.suggested that the v a r i e t i e s of L_. g r a c i l i s should also be designated as separate species. Ornduff and Cro-v e l l o (1968) concluded that-the three taxa should a l l be recognized as either v a r i e t i e s or species, since either treatment would be consistent with Mason's hypothesis which they considered correct. I n t e r e s t i n g l y , the r e s u l t s of Orn-duff and Crovello support a d i f f e r e n t combination: that L. alba var. v e r s i - color , L_. - g r a c i l i s var. p a r i s h i i and L. montana, are a l l v a r i e t i e s of one spe-i e s , and that L_. alba var. alba and L. g r a c i l i s var. g r a c i l i s should be r a i s e d to the rank of species. However, they chose not to follow these r e s u l t s which they interpreted to be taxonomically i n s i g n i f i c a n t . A f t e r consideration of the-evidence presented by the .above authors, and analysis.of the f i v e taxa, the best s o l u t i o n i s probably to recognize each of the two v a r i e t i e s of L_. g r a c i l i s , the two v a r i e t i e s of L. alba and L. montana - H 7 -at the same rank. At any rate, L. montana should notObe maintained at a higher rank.than the other four taxa, since the flavonoid evidence indicates a very recent divergence of t h i s taxon from the v a r i e t y versicolor.. The a v a i l a b l e data support the hypothesis that the putative ancestor of L_. montana. and the v a r i e t i e s g r a c i l i s , parishi-i,- alba and, v e r s i c o l o r occu-pied a continuous range east of the c o a s t a l mountains from southern C a l i -f o r n i a to c e n t r a l Oregon. Since elements of L_. g r a c i l i s have the most..diverse flavonoid patterns and share a f f i n i t i e s with L_. s t r i a t a and. the section Re- flexae , i t i s probable that the putative ancestor of the f i v e taxa resembled L_. g r a c i l i s . At some comparatively recent time t h i s continuous ancestral population was s p l i t into three parts. Variety g r a c i l i s was derived from the northern population and v a r i e t y p a r i s h i i from the southern population. But the c e n t r a l population r a p i d l y diverged i n a more r a d i c a l fashion, creating three d i s t i n c t taxa, L. montana and.the v a r i e t i e s alba and v e r s i c o l o r , f o l -lowing t h e i r successful invasions into d i f f e r i n g habitats,. This hypothesis suggests that the f i v e taxa should be divided into three species and three v a r i e t i e s ; i _ . e_. , the v a r i e t i e s g r a c i l i s , p a r i s h i i and alba .become species, and the hypothetical new species of JJ. alba consists of the three v a r i e t i e s alba, v e r s i c o l o r , and montana. However, t h i s i s an impractical s o l u t i o n , since the morphological differences d i s t i n g u i s h i n g the v a r i e t i e s would be of a greater magnitude than those separating, the species'. WJhen a l l the factors are considered, there i s l i t t l e but confusion to be gained by recognizing any- of the above f i v e taxa. at d i f f e r e n t ranks). By t h i s c r i t e r i o n , the five-might be considered as v a r i e t i e s of one geo-g r a p h i c a l l y diverse, polymorphic species. However, i n view of the pronounejed e c o l o g i c a l , geographical, and.morphological differences between these taxa, and the lack of natural h y b r i d i z a t i o n even where b a r r i e r s to crossing a r e . i n -complete, treatment as f i v e v a r i e t i e s i s .unsatisfactory. Therefore, the most - 118 -reasonable way to t r e a t t h i s group i s to elevate each of the v a r i e t i e s of L_. alba and L_. g r a c i l i s to s p e c i f i c status along with L. montana, thus cre-ating f i v e d i s t i n c t species. Past taxonomic treatments of the L_. f l o c c o s a group have drawn con-f l i c t i n g conclusions. Mason (1952) recognized three v a r i e t i e s of L. f l o c c o s a , including f l o c c o s a , and the rare endemics pumila and b e l l i n g e r i a n a . His recognition of these taxa.at v a r i e t a l status was s l i g h t l y inconsistent, since he was unable to demonstrate i n t e r f e r t i l i t y between the three, and each was formerly recognized as a d i s t i n c t species. Mason's judgement i n t h i s case rested on the morphological s i m i l a r i t i t e s and.distributions of the three. Since Mason's work, Ornduff (1971) has produced hybrids between f l o c c o s a and b e l l i n g e r i a n a . Arroyo (1973a) also has indicated that a c e r t a i n amount of i n t e r f e r t i l i t y e xists between some u n i d e n t i f i e d elements of the L_. floccosa group. Recently, Arroyo (l973a) has changed the treatment of L_. floccosa,' r a i s i n g each of Mason's three v a r i e t i e s to subspecies and designating two ad-d i t i o n a l subspecies, g r a n d i f l o r a a n d . c a l i f o r n i c a . This further d i v i s i o n was based on the r e s u l t s of a taximetric analysis of plants grown under uniform conditions, considering various f l o r a l and\vegetative characters. Although Arroyo divided L_. floccosa' into f i v e taxa, her r e s u l t s , as • she presents them, indicate that a two-way s p l i t would have been more approp-r i a t e . By her a n a l y s i s , the subspecies g r a n d i f l o r a , c a l i f o r n i c a and pumila form a discrete.grouping d i s t i n c t from a second group con s i s t i n g of the sub-species' flj3ccosa_ and b e l l i n g e r i a n a . Arroyo explains that the two groups are separated since each represents one of two discontinuous l e v e l s of autogamy i n the species. One of Arroyo's major premises i s that L. f l o c c o s a recently evolved as an autogamous de r i v a t i v e of L. alba (1973a, 1973b). I t i s upon t h i s pre-- 119 -mise that she bases, at l e a s t i n p a r t a l l of her conclusions about evolu-t i o n i n L_. fl o c c o s a. This premise i s an i n t e r e s t i n g one and worthy of con-s i d e r a t i o n here. Although Arroyo does not present her r a t i o n a l e , presumably she sees the highly autogamous forms of L_. floccosa' l i n k e d to L. alba by the subspecies grandiflora:-, c a l i f o r n i c a and pumila, which are more or l e s s intermediate i n flower s i z e , degree of pubescence, and supposedly, l e v e l of autogamy. This hypthesis i s not consistent with the other a v a i l a b l e information. Ornduff and Crovello (1968) found L. floccosa d i s t i n c t from a l l other taxa of Limnanthes i n three out of four of t h e i r phenetic- analyses. This con-sistency was shared by only one other group c o n s i s t i n g of Floerkea and L_. macounii. In the'fourth a n a l y s i s , only one subspecies, pumila, showed any a f f i n i t y to L_. alba, while the subspecies floccosa and bellingeriana.formed a d i s t i n c t , separate c l u s t e r . (The subspecies g r a n d i f l o r a and c a l i f ornica,- ;:had; riot ;,y et., been described.) There are few morphological characters which possess much taxonomic u t i l i t y i n d i s t i n g u i s h i n g members of the Limnanthaceae.. Generally, the char-acters that are s t r o n g l y conserved by evolution are a l l uniform between taxa, while the p l a s t i c characters have evolved r a p i d l y many times i n d i f f e r e n t an-c e s t r a l l i n e s , often creating s i m i l a r i t i e s by convergence. I n t e r e s t i n g l y enough, the L_. floccosa group i s w e l l distinguished from other taxa of Lim- nanthes by what appear to be excellent taxonomic characters. These d i s t i n -guishing characters are: the .sepals have a c h a r a c t e r i s t i c apiculate t i p , the sepals become v a l v e - l i k e , i n f o l d i n g as the nutlets mature, and an a b s c i s -sion layer i s formed to some degree between receptacle and p e d i c e l , promoting d i s p e r s a l of the nutlet-containing-flower as a u n i t . The presence of these c h a r a c t e r i s t i c s i n L_. f l o c c o s a suggests that t h i s species diverged from other Limnanthes taxa i n comparatively ancient times. C e r t a i n l y , these morpho-- 120 -l o g i c a l differences do not support a close r e l a t i o n s h i p with any other members of the section Inflexae, including L_. alba. Mason (.1952) placed L_. floccosa, i n the Inflexae on the basis o f i t s p e t a l p o s i t i o n a f t e r p o l l i n a t i o n . The flavonoid data c l e a r l y support.;:; t h i s placement since i n the pe t a l factor analyses the subspecies b e l l i n g e r i a n a and pumila occur within the grouping formed by other members of the section (Figures XVII & XVIII). Most of the subspecies of L. flo c c o s a were analyzed from dri e d m a t e r i a l , which included several of Arroyo's c o l l e c t i o n s . Since i n s u f f i c i e n t p e t a l ma-t e r i a l was av a i l a b l e from these c o l l e c t i o n s , i t was, therefore, only possible to perform whole plant flavonoid analyses on the entire L_. floccosa. group. These analyses have several taxonomically s i g n i f i c a n t implications. Figure XX i s an expanded view of the P r i n c i p a l Components Analysis of whole plant flavonoids of the OTUs of L. fl o c c o s a . The three OTUs of sub-species floccosa (23,24,25) form a diffuse- grouping. Both of the subspecies pumila (i2Q" "&.4:'2lj..jaM^iellingeriana (18 & 1 9 ) , analyzed on the basis of fresh and dried material, form two d i s t i n c t c l u s t e r s separated cleanly by the approxi-mate t o t a l v a r i a t i o n present within the e n t i r e species. However, the sub-species of c a l i f o r n i c a (22) and.grandiflora (26) f a l l very close to the OTUs of subspecies b e l l i n g e r i a n a i n d i c a t i n g that these four OTUs have.very s i m i l a r flavonoid compositions. The flavonoid data support the following hypothesis. Since the d i -vergence i n the subspecies f l o c c o s a i s equivalent i n magnitude to the t o t a l divergence within the whole species complex, and a l l other subspecific taxa f a l l within the l i m i t s of t h i s v a r i a b i l i t y , subspecies f l o c c o s a most c l o s e l y approximates the ancestral type. The large discrepancy between pumila and the c l u s t e r formed by the remaining subspecies g r a n d i f l o r a , c a l i f o r n i c a and b e l l i n g e r i a n a , suggests that these two groups represent separate l i n e s of FIGURE XX P r i n c i p a l Components Analysis of 9 OTUs of L. floccosa Based on Whole.Plant Flavonoids i ro H - 122 -divergence from the ancestral form of L . floccosa. Because of the presence of two morphological characters, i t . i s probable that* subspecies pumila was the f i r s t to diverge from the L_. floccosa. group. These characters are: l ) l i k e other taxa of Limnanthes, the stigmatic branches of subspecies pumila are s p l i t only part way down t h e - s t y l e , while they are s p l i t a l l the way down i n the.other four subspecies; 2) the abscission layer that forms beneath the receptacle does not develop to as great .an extent i n pumila as i t does i n the remaining subspecies. Therefore, the flowers of pumila do n o t 1 r e a d i l y d i s a r t i c u l a t e as a unit (Arroyo, 1973a). Before a decision can be made regarding whether the fl o c c o s a group requires taxonomic r e v i s i o n , i t i s necessary to consider the v a l i d i t y of the. subspecific erections of Arroyo. Although her taximetric treatment was based on 42 characters, by her own admission the emphasis i s l a r g e l y on f l o r a l char-ac t e r s , thus emphasizing the taxonomic s i g n i f i c a n c e of autogamy. Since auto-gamy has evolved many times i n d i f f e r e n t l i n e s of the Limnanthaceae, taxo-nomic conclusions based on t r a i t s associated with s e l f i n g probably w i l l be erroneous, or at best misleading. Besides the weighting of characters l i n k e d to autogamy, Arroyo's analysis i s further prejudiced on the basis of two t r a i t s , pubescence and stature. Of the 42 characters determined, 7 of these concern pubescence of plant parts. The character used by Mason to d i s t i n g u i s h two of the v a r i e t i e s of L_. f l o c c o s a , rows of hairs at the pe t a l bases, i s included i n her analysis as an eighth character. Also, of the 42 characters, l6 or almost hQ%, r e l a t e s t r i c t l y to ph y s i c a l stature, being measurments of lengths and widths of various plant parts. Besides these redundancies, 1 a d d i t i o n a l character over-lap i s present.. For instance, " f r u i t i n g calyx abscission zone" and "mode of nutlet dehiscence" are presented as separate characters, when i n fact.they are 100%. correlated. - 123 -Because of the u n j u s t i f i e d weighting of c e r t a i n characters i n her a n a l y s i s , Arroyo's taxonomic conclusions are probably of l i t t l e value. Even her i n t e r p r e t a t i o n of her own r e s u l t s appears inconsistent with her data. Instead of her two p r i n c i p a l groups being separated by discontinuous l e v e l s of autogamy, i t i s probable that these two groups were separated simply on the basis of s i z e ; one group has l a r g e r component parts than the other. A f t e r consideration of the a v a i l a b l e information, i t i s c l e a r that c e r t a i n changes should be made i n the c l a s s i f i c a t i o n of the L_. f l o c c o s a group. Since the evidence indicates that the subspecies pumila diverged from L_. f l o c - cosa at an e a r l i e r time than the other taxa, that i t i s well d i f f e r e n t i a t e d morphologically and chemically from the r e l a t e d taxa, and that i t i s apparently not c r o s s f e r t i l e with the other subspecies, t h i s taxon should be r e i n s t a t e d as a . separate., species. The subspecies bellingeriana-, g r a n d i f l o r a and c a l i f o r n i c a a l l appear to be c l o s e l y r e l a t e d and probably can be considered as v a r i e t i e s or L_. f l o c - cosa, or a species d i s t i n c t from L_. fl o c c o s a. Of the three^'subspecies, b e l - l i n g e r i a n a .is morphologically d i s t i n c t from g r a n d i f l o r a and c a l i f o r n i c a , which are very close i n t h e i r morphological and e c o l o g i c a l c h a r a c t e r i s t i c s . However, these l a t t e r two subspecies are distinguished by a wide d i s c o n t i n u i t y i n t h e i r ranges, and each l i e s at a d i f f e r e n t extremity of the range of subspecies f l o c - cosa. Since Arroyo has reported that there i s a c e r t a i n amount of natural genetic exchange within the L_. floccosa- group, i t i s probably best to include the taxa b e l l i n g e r i a n a , g r a n d i f l o r a , c a l i f o r n i c a and f l o c c o s a into one species. The differences between grandiflora. and c a l i f o r n i c a are probably not great enough to preserve these as separate e n t i t i e s , and they should be merged into one taxon. I t i s suggested that members of the species L_. f l o c c o s a should be r e -instated at v a r i e t a l rank, rather than at the subspecies l e v e l chosen by Arroyo. - 124 -This i s because: l ) a c e r t a i n amount of genetic exchange occurs between these taxa, as demonstrated by intergrading of characters i n some areas; and.2) Mason's choice of rank i s adequate, and there was no apparent reason f o r her decision to change the. taxa to subspecies from v a r i e t i e s . The Reflexae has been regarded as a natural grouping of species con-t a i n i n g L. bakeri, L. s t r i a t a , L_. macounii, L. d o u g l a s i i , L. vinculans, and perhaps Floerkea proserpinacoides (Ornduff & Crovello, 1968). For the rea-sons already presented, L. bakeri and L_. s t r i a t a . should be deleted from t h i s section, and become the monotypic members of the sections Bakera and Limnanthes. Limnanthes macounii, L. vinculans, and the four v a r i e t i e s of L_. doug- l a s i i together form a group encompassing a great range of morphological diver-s i t y . In fact these taxa are held together l a r g e l y on the basis of r e f l e x e d p e t a l p o s i t i o n a f t e r p o l l i n a t i o n and demonstrated i n t e r f e r t i l i t y . However,-a l l of these taxa clustered t i g h t l y together i n one c l u s t e r analysis based on "vegetative characters" (Ornduff & Crovello, 1968). Mason did not have fresh material of L. macounii when he revised the taxonomy of Limnanthes. Therefore, on the basis of d r i e d material he deter-mined that t h i s taxon was a separate species, and i f i t belonged to the genus Limnanthes (instead of Floerkea)-,' i t s reflexed petals placed i t i n the section Reflexae. With the rediscovery of t h i s taxon i n nature, both of these judge-ments have been upheld. Af t e r having grown fresh material of L. macounii from two populations, i t i s evident that t h i s species i s more c l o s e l y l i n k e d to the Reflexae than i s apparent i f i t i s c o l l e c t e d from i t s natural habitat. Although a l l Lim- nanthes taxa are phenotypically p l a s t i c , L. macounii i s p a r t i c u l a r l y so. As i t i s found i n nature, t h i s species -is very inconspicuous, being t o t a l l y pro-cumbent, or i f upright, no t a l l e r than the surrounding herb elements, and u s u a l l y no larger than a few centimetres. When grown i n the greenhouse under - 125 -normal conditions, t h i s species a t t a i n s an unnaturally large s i z e approxi-mating that of the other Limnanthes taxa. I t s f o l i a g e resembles that .'of L. d o u g l a s i i vars. sulphurea or d o u g l a s i i , and i t not infrequently produces t o -t a l l y , or p a r t i a l l y , 5-merous flowers. The s i z e of i t s flowers, however, r e -mains small. The evolutionary consequences of autogamy have been extreme i n L. macounii, including l o s s of f l o r a l parts and.reduction i n flower s i z e . These changes notwithstanding, i t i s apparent that the o v e r a l l diminutive s i z e of t h i s plant can be a t t r i b u t e d l a r g e l y to the marginal conditions under which i t grows n a t u r a l l y . I t i s almost c e r t a i n that t h i s plant p e r s i s t s only as a r e l i c i n the warmest areas of Vancouver Island, cut o f f by the S t r a i t of Georgia from some receding population of d o u g l a s i i - l i k e plants pushed south-ward by a period of cooling. This hypothesis i s supported both by the absence of L_. macounii i n more southern regions and the present-day d i s j u n c t i v e d i s -t r i b u t i o n of L_. d o u g l a s i i along the west coast of the United States (Figure- I ) . It i s l i k e l y that the a b i l i t y of L. macounii to p e r s i s t f a r to the north of the remaining Limnanthes taxa i s made possible l a r g e l y by i t s conversion to autogamy.(Mosquin, 1966). I f so, t h i s s i t u a t i o n p a r a l l e l s the a b i l i t y of the t o t a l l y autogamous Floerkea to grow/in more northern areas. Although L. macounii appears to be i n l i t t l e danger of immediate.extinction, several un-u s u a l l y c o o l , or l a t e springs might decimate what remains of t h i s taxon. The p e t a l flavonoid data c l e a r l y links L_. macounii with the v a r i e t i e s of L_. d o u g l a s i i . Furthermore, the two species are l i n k e d by s i m i l a r i t i e s i n UV-visible flower patterning. The two OTUs of L_. macounii,• which c l u s t e r t o -gether i n both whole'plant.and p e t a l analyses, exhibit a f a i r l y small.amount of divergence i n flavonoid composition. This r e l a t i v e l y small amount of d i -vergence supports the hypothesis that L_. macounii has only recently diverged from the main Reflexae l i n e providing that there has been the retention of - 126 -some v a r i a b i l i t y by t h i s species. Although two were formerly considered separate species, Mason merged the taxa sulphurea, nivea, rosea and d o u g l a s i i as v a r i e t i e s of the species L_. d o u g l a s i i . However, these v a r i e t i e s have morphological differences of the same magnitude as those separating other species of Limnanthes. With the-exception of p a r t i a l intergrading of v a r i e t i e s rosea, and nivea i n an area of overlap, natural hybrids apparently are not formed between these four v a r i -e£Ip's>> However, a l l four are experimentally inter fertile,- p a r t i c u l a r l y when a l l o p a t r i c populations serve as parents. Therefore, Mason recognized the four as v a r i e t i e s since the b i o l o g i c a l species concept was uppermost i n Mason's taxonomic treatment. With the evidence made av a i l a b l e here, t h i s treatment requires reconsideration. A l l the v a r i e t i e s of L_. d o u g l a s i i are predominantly bee-pollinated outcrossers. UV f l o r a l photography discloses differences. between the v a r i -eties that may lead to i s o l a t i o n r e s u l t i n g from e t h o l o g i c a l , rather than genetic f a c t o r s . • ' The flavonoid f a c t o r analyses indicate-that most v a r i e t i e s of L_. doug- l a s i i are.as d i s t i n c t from one another as most of the recognized species i n the family. The exceptions are.the v a r i e t i e s rosea and.nivea which f a l l c l o s e l y together i n a l l the f a c t o r analyses. I f the taxa rosea-, nivea, sulphurea and d o u g l a s i i are to b;e' v a r i e t i e s of one species, i t i s apparent that equally distant.taxa have been assigned d i f f e r e n t ranks. The divergence i n flavonoid composition of these four v a r i -eties i s no le s s than the divergence of L";: macounii, a d i s t i n c t species, from these varieties-. A s i m i l a r s i t u a t i o n also exists f or the recently described species L. vinculans. When Ornduff described L.. vinculans (1969a), he c i t e d - l e a f l e t charac-' t e r i s t i c s that he interpreted to show that t h i s species was intermediate - 127 -"between L_. bakeri and L. d o u g l a s i i . However, the s i m i l a r i t i e s between the l e a f l e t s of L_. bakeri and L_. vinculans probably r e s u l t from d i f f e r e n t evo-l u t i o n a r y processes. Although the e f f e c t on the morphology i s s i m i l a r , i t i s probable that the l e a f l e t s of L. vinculans have.become e n t i r e by l o s s of l a t e r a l l y divided segments, while the l e a f l e t s of L.- bakeri remain mostly undivided. This hypothesis i s supported by the l e a f morphology of the seed-l i n g s of L_. vinculans, since at t h i s stage i t i s the only species of the Lim-nanthaceae having non-compound l i n e a r leaves. S i m i l a r i t i e s i n l e a f l e t margins between L_. bakeri and L_. vinculans probably r e s u l t from analogous adaptations caused by comparable s e l e c t i o n pres-sures c h a r a c t e r i s t i c of the c o a s t a l habitats of these species. In other characters L_. vinculans c l o s e l y resembles the L_. d o u g l a s i i group, and probably shares i t s c l o s e s t natural a f f i n i t y with v a r i e t y nivea. This a f f i n i t y i s indicated by the s i m i l a r i t i e s i n f l o r a l characters and by high i n t e r f e r t i l i t y between these two taxa (Ornduff, 1971), although hybrids are only formed be-tween a l l o p a t r i c populations. Although f r e s h material of L_. vinculans was not a v a i l a b l e f o r t h i s i n v e s t i g a t i o n , i t s whole plant flavonoids were determined from d r i e d material and compared to the other taxa. In the'cluster and f a c t o r analyses, " L . v i n - culans . clustered with OTUs of L_. d o u g l a s i i i n d i c a t i n g a close r e l a t i o n s h i p . In a d d i t i o n , the distance separating L_. vinculans from the v a r i e t i e s of L_. d o u g l a s i i i s no greater than the distance separating the v a r i e t i e s them-selves. On the'basis of the a v a i l a b l e information, i t i s possible to hypo-thesize how evolution proceeded i n the Reflexae. In a. process p a r a l l e l to that hypothesized for the Inflexae l i n e , elements of a douglasii-type ancestral population became i s o l a t e d geographically and/or e c o l o g i c a l l y and diverged into d i s t i n c t e n t i t i e s as a.result of d i f f e r e n t s e l e c t i o n pressures i n the - 128 -various new habitats. Six of these divergent e n t i t i e s e x i s t today, 'the four v a r i e t i e s of L_. d o u g l a s i i , and.the species L_. macounii and L. vinculans. Probably a l l s i x diverged comparatively recently. However, among these s i x taxa are some of'the most highly adapted forms i n the family. There i s l i t t l e j u s t i f i c a t i o n f o r maintaining the s i x taxa of the section Reflexae at d i f f e r e n t ranks, with the exceptions of the v a r i e t i e s nivea and rosea. Since these taxa have been found to intergrade i n a region of sympatry, perhaps they should be recognized as v a r i e t i e s of one species. Since L. macounii- and L. vinculans have been recognized and maintained as sepa-rate species, and there are good d i s t i n g u i s h i n g morphological, d i s t r i b u t i o n a l and e c o l o g i c a l characters separating the v a r i e t i e s d o u g l a s i i , sulphurea and rosea, i t would probably be most consistent to elevate these l a t t e r three taxa.to the rank of species. The proposed species rosea would consist of ' the two v a r i e t i e s nivea and rosea. E. Summary of Proposals f o r Revision of the Limnanthaceae  i . Introduction Af t e r consideration of the a v a i l a b l e evidence regarding r e l a t i o n s h i p s i n the Limnanthaceae, i t i s apparent that the family requires i n t e r n a l taxo-nomic r e v i s i o n at every l e v e l . Some researchers w i l l i n s i s t that a family should not be revised s o l e l y on the basis of chemical,characters; I agree, and t h i s i s not what i s being proposed. I t i s true that the flavonoids pro-vide a set of characters of great taxonomic u t i l i t y i n the Limnanthaceae., and. therefore were depended on h e a v i l y , i n formulating a new c l a s s i f i c a t i o n . How-ever, i n no instance does the other a v a i l a b l e evidence c o n f l i c t with the flavonoid information. Although interpretations of data may vary, i t i s improbable that d i f f e r e n t i n t e r p r e t a t i o n s would greatly a l t e r the proposed family c l a s s i f i c a t i o n , at l e a s t u n t i l new information from other sources i s - 129 -brought to bear on the problem. i i . Genus Level Floerkea and Limnanthes should be merged into one genus. The d i s -s i m i l a r i t i e s Let-ween these two genera are simply not great enough, or discon-tinuous enough, to preserve them as separate genera. On t h i s issue, the evi-. dence i s conclusive. With regard to the merit of maintaining Floerkea as a genus d i s t i n c t from Limnanthes on t r a d i t i o n a l grounds, the following passage from Jepson (1951, P- 14) has a p p l i c a t i o n : ... It i s , however, necessary that the l i m i t s of genera should, with increase of knowledge of t h e i r structure, probable phylogeny, geo-graphic h i s t o r y and ecology, be subject to r e v i s i o n and modification. No genus has any vested r i g h t s on account of long usage or approval by the great masters. Continued research, increased knowledge and an enlarged viewpoint must c o n t i n u a l l y f i n d vent i n new generic expressions. The generic name should become Floerkea since t h i s genus was described 32 years before Limnanthes, and.therefore takes precedence. i i i . Section Level The previous c l a s s i f i c a t i o n , which recognized Floerkea and two sections of Limnanthes should be amended to a si n g l e genus containing f i v e sections: Bakera, Floerkea, Inflexae, Reflexae and Limnanthes. The primary reason f o r t h i s a l t e r a t i o n is. to provide i n t e r n a l consistency of rank. i v . Species and V a r i e t a l Levels With the object of providing uniformity i n ranking p a r a l l e l taxa, i t i s necessary to rev i s e designations of many taxa i n the Limnanthaceae at the species l e v e l and.below. I t i s also necessary to rev i s e the concept of species differences as i t i s applied i n t h i s context. The b i o l o g i c a l species concept - 130 -has been found wanting as applied to the Limnanthaceae. Since t h i s c r i t e r i o n was previously assumed to hold the utmost i n taxonomic significance, other sorts of evidence were deemphasized i n e a r l i e r treatments. ° The Limnanthaceae should be revised to consist of 15 species, the species rosea containing two v a r i e t i e s and the species floccosa containing three v a r i e t i e s . With one exception, a l l of the taxa designated by previous authors would remain d i s t i n c t , but the ranks of many of these taxa would be changed. I t i s recommended that the subspecies of L. floccosa, grandiflora and c a l i f o r n i c a , as described by Arroyo, be merged into the single taxon floccosa var. grandiflora. F. Synopsis of the Family Limnanthaceae A single genus designated Floerkea W i l l d . , Neue S c r i f t , Geselschaft Nat. 3:449-1801. [Limnanthes R. Br., London and Edin. Philos. Mag. IXI,2 :70.. 1833] Section Limnanthes 1. F. s t r i a t a (JepVbh^Earker, comb. nov. [L. striata.Jepson, F l . C a l i f . 2:1*11. 1936.] Section Bakera 2/ F_. bakeri (J.T. Howell) Parker, comb. nov. [L_. bakeri J.T. Howell, Plantae Occi'dentales. I I I . L e a f l . West. Bot. 3:206. 1943.] . Section Floerkea 3. F. proserpinacoides W i l l d . , Neue S c h r i f t . Ges. Nat. 3:449. 1801. '[F. occidentalis. Rydberg, Mem. N.Y. Bot, Gard. 1:268. 1900.] Section Inflexae 4. F. alba Greene, F l . Fran. 100. 1891. [L. alba Bentham, PI. Hartw. 301. 1848.] I L . alba var. aetonsa Jepson, F l . C a l i f . 2:411. 1936..] ,[L. alba var. alba (Benth.) CT, Mason, Univ. C a l i f . Publ. Bot. 25:455. 1952."] - 131 -Section Inflexae — continued 5. F. fl o c c o s a (Howell) var. fl o c c o s a Parker, comb. nov. [L. floccosa Howell, F l . NW Amer. 1:108. I897.] [L. f l o c c o s a var. floccosa (Howell) C.T. Mason, Univ. C a l i f . Publ. Bot. 25:455- 1952.] [L. f l o c c o s a ssp. fl o c c o s a (Howell) Arroyo, B r i t t o n i a 25:177. 1973.] var. b e l l i n g e r i a n a (M.E. Peck) Parker, comb. nov. [L- b e l l i n g e r i a n a M.E. Peck, Proc. B i o l . Soc. Wash. 50:93. 1937.] [L. f l o c c o s a var. b e l l i n g e r i a n a (M.E. Peck) C.T. Mason, Univ. C a l i f . Publ. Bot. 25:455- 1952.] [L. f l o c c o s a ssp. b e l l i n g e r i a n a (M.E. Peck) Arroyo, B r i t t o n i a 25:177- 1973.] ........... var. g r a n d i f l o r a (Arroyo) Parker, comb. nov. [L. f l o c c o s a ssp. g r a n d i f l o r a Arroyo, B r i t t o n i a 25:177. 1973.] [L_. f l o c c o s a ssp. c a l i f o r n i c a Arroyo, B r i t t o n i a 25:177. 1973.] 6. F. g r a c i l i s (Howell) P a r k ercomb. nov. [L. g r a c i l i s Howell, F l . WW Amer. 1:108. I897.] I L . g r a c i l i s var. g r a c i l i s (Howell) C.T. Mason, Univ. C a l i f . Publ. Bot. 25:455- 1952.] 7-F_. montana (Jepson) Parker, comb. nov. I L . montana Jepson, F l . C a l i f . 2 :4l2. 1936.']" 8. F. p a r i s h i i (Jepson) Parker, comb. nov. [L. v e r s i c o l o r var. p a r i s h i i Jepson, F l . C a l i f . 2 : 4 l l . 1936.] [L. g r a c i l i s var. p a r i s h i i (Jepson) C.T. Mason, Univ. C a l i f . Publ. Bot. 25:455- 1952.] 9. F. pumila (Howell) Parker, comb. nov. [L. pumila Howell, F l . NW Amer. 1:108. I897.] [L. floccosa var. pumila (Howell) C.T. Mason, Univ. C a l i f . Publ. Bot. 25:455- 1952.] [L.- f l o c c o s a ssp. pumila'(Howell) Arroyo, B r i t t o n i a 25:177. 1973.] 10. F. v e r s i c o l o r Greene, Erythea 3:62. 1895. J L . v e r s i c o l o r (Greene) Rydberg, N. Amer. F l . 25:99. 1910.] J L . alba var. versicolor.(Greene) C.T. Mason, Univ. C a l i f . Publ.. Bot. 25:455- 1952.] Section Reflexae 11. F. d o u g l a s i i B a i l l o n , Adansoniay10:362..I87S. [L. d o u g l a s i i R. Br., London Edin. Ph i l o s . Mag. 111,2:70.. 1833.] ,[L_. sulphurea odorata Loud. , Enc. PI. 1543. 1855.] ["L. sulphurea (Loud.)" Rydberg, N. Amer. F l . 25:98. 1910.] [L_. howelliana Abrams, Madrono 6:27.' 194l .] [L. d o u g l a s i i var. d o u g l a s i i (R. Br.) C.T. Mason, Univ. C a l i f . P u b l . Bot, _ 25:455. 1952.] 12. F. macounii (Trelease) A. Gray, Syn. F l . 1:363. 1897. [F. proserpinacoides Macoun, Cat. Canad. PI. 1:90. 1883.] [L. d o u g l a s i i Macoun, Cat. Canad. PI. 3:502. 1884.] [L. macounii Trelease, Mem. Boston Soc. Nat. H i s t . 4:85. I887.] - 132 -Section Reflexae — continued 13. F. rosea Greene var. rosea, F l . Fran. 100. 1891. [L. rosea Bentham, PI.- Hartw. 302. 1848.] [L. p u l c h e l l a Hartweg, Jour. Hort. Soc. London 3:220.'1848.] .[L. rosea var. Candida Jepson, F l . C a l i f . 2:411. 1936.] [L. d o u g l a s i i var. rosea (Benth.) C.T. Mason, Univ. C a l i f . Publ. Bot. 25:455- 1952.] var. nivea (Mason) Parker, comb. nov. IL. d o u g l a s i i var. nivea C.T. Mason, Univ. C a l i f . Publ. Bot. 25:455- 1952.] 14. F. sulphurea (Mason) Parker, comb. nov. [L. d o u g l a s i i var. sulphurea C.T. Mason, Univ. C a l i f . Publ. Bot. 25:455. 1952] 15. F. vinculans (Ornduff) Parker, comb. nov. [L. vinculans Orncluff, B r i t t o n i a 21:11. I969.] - 133 -CONCLUSIONS 1) The flavonoids found i n the Limnanthaceae are fl a v o n o l glycosides. They are unusual with regard to type of aglycone, including various deriva-t i v e s of s y r i n g e t i n and l a r i c y t r i n . These flavonoids are also unusual r e -garding extent of g l y c o s y l a t i o n ; d i - r u t i n o s i d e s and rhamnosyl-rutinosides are found n a t u r a l l y occurring for the f i r s t time. 2) On the basis of presently known flavonoid s i m i l a r i t i e s , the Limnan-thaceae i s d i s t i n c t ; no fa m i l i e s appear to share any special, a f f i n i t i e s with i t . 3) Because of s i m i l a r i t i e s i n flavonoid composition between a l l members of the L i m n a n t h a c e a e i t i s apparent that they are a l l c l o s e l y r e l a t e d . k) • The a p p l i c a t i o n of taximetric techniques to flavonoid taxonomy of the Limnanthaceae indicates that the flavonoids are u s e f u l f o r the c l a r i f i -c ation of r e l a t i o n s h i p s -in t h i s family. 5) " 'Flavonoids occurring i n r e l a t i v e l y greater-concentrations are better indicators of r e l a t i o n s h i p s between members of the Limnanthaceae than those occurring i n trace amounts. 6) When flavonoid characters are analyzed by conventional c l u s t e r i n g techniques Ci.e_. , the weighted p a i r group method), i t makes l i t t l e d i f f e r e n c e whether or not mutual absences are-considered. - 134 -7) No taxonomically s i g n i f i c a n t differences were detected due to the use of d r i e d plant material when fresh material was unavailable. 8) C l a s s i f i c a t i o n s based on p e t a l flavonoids more c l e a r l y i n d i c a t e natural r e l a t i o n s h i p s between taxa of the Limnanthaceae than do c l a s s i f i c a t i o n s based on whole plant (including flowers) flavonoids. 9) ' The techniques of f a c t o r analysis provide c l e a r e r i n d i c a t i o n s of r e l a t i o n s h i p s between taxa i n the Limnanthaceae than do the standard c l u s -t e r i n g techniques. However, the r e s u l t s derived from the two techniques do not c o n f l i c t . . Apparently, ordination of taxa i s required to understand the multiple branching of evolutionary l i n e s and the d i f f e r e n t ongoing rates of evolution i n the-family. 10) . The flavono.fd pattern of the h o r t i c u l t u r a l v a r i e t y of L. d o u g l a s i i cannot be d i r e c t l y compared with other taxa of Limnanthes i n chemosystematic studies .-because of the e f f e c t s of a r t i f i c i a l s e l e c t i o n and/or inbreeding. . . 11) Certain members 'of the Limnanthaceae have f l o r a l patterning v i s i b l e only under UV l i g h t . This character i s taxonomically s i g n i f i c a n t and helps to d i s t i n g u i s h one supraspecific group from the rest of the family. 12) ' A new c l a s s i f i c a t i o n of the Limnanthaceae i s proposed a f t e r i n t e g r a t i o n of the flavonoid data and other a v a i l a b l e information. 13) I t i s apparent from discussion of the taxonomically s i g n i f i c a n t i n -formation known f o r the Limnanthaceae, that t h i s family requires taxonomic r e v i s i o n at a l l l e v e l s . - 135 -LITERATURE CITED: 1. Abrams, L. 1941. A new Limnanthes from Oregon. Madrono 6:29. 2. Arroyo, M.T.K. 1973a. A taximetric study of i n f r a s p e c i f i c v a r i a t i o n i n autogamous Limnanthes fl o c c o s a (Limnanthaceae). B r i t t o n i a 25:177-191-3 1973b. Chiasma frequency evidence on the evolution of autogamy i n Limnanthes floccosa- (Limnanthaceae). Evolution 27:679-688. 4. B a i l l o n , H. 1871. Notes sur l e s Geraniacees et l e s Linacees. Adansonia 10:360-371. 5. Bate-Smith, E.C. 1962. The phenolic constituents of plants and t h e i r taxonomic s i g n i f i c a n c e . I. Dicotyledons. J . Linn. Soc. (Bot.) 58:95-173. 6. Bentham, G. 1848. Plantae Hartwegianae [1839-1852]. London. 7. Bohm, B.A. & F.W. C o l l i n s . 1975. Flavonoids of Philydrum lanuginosum. Phytochemistry ..14:31'5-3l6. 8. Brown, R. 1833. Characters and descriptions of Limnanthes. London & Edin. Philos.. Mag.. 111,2:70. -9. Cole, D.F. 197k. E f f e c t s of temperature and l i g h t on germination of two accessions of Limnanthes alba seed. Economic Botany 28:155-159-10. Cronquist, A. I968. The Evolution and C l a s s i f i c a t i o n of Flowering Plants. Houghton M i f f l i n , Boston. 11. Daumer, K. T958. Blumenfarben, wie s i e die Bienen sehen. Zeitschr. Vergl. P h y s i o l . 41:49-110./ 12. Davis, P.H. & V.H. Heywood. 1963. P r i n c i p l e s of• Angiosperm Taxonomy. Van Nostrand, Princeton, ,N.J'. 13. ' Dement, W.A. & P.H. Raven. 1974. Pigments responsible f or u l t r a v i o l e t patterns, i n flowers, of Oenothera (Onagraceae). Nature 252:705-706. 14. Engler, A. & K. P r a n t l . 1896. Die naturlichen Pflanzenfamilien, T e i l 3 (Ant. 5):136-137. L e i p z i g . 15. E t t l i n g e r , M.G.. & A.J. Lundeen. 1956. The mustard o i l of Limnanthes do u g l a s i i seed, m-methoxybenzyl isothiocyanate. J.A.C.S. 78:1952-1954. 16. Gentry, H.S. & R.W. M i l l e r . 1965. The search for new i n d u s t r i a l crops XV. prospects of Limnanthes- Economic Botany 19:25-32. • 17. - Gray,A. I897. Synoptical F l o r a of North/America 1 ( p a r t . l , fasc. 2) New York. 18. Greene, E.L. I89I. F l o r a Franciscana. San Francisco. 19 -1895. Novitates occidentales. XII. Erythea 3:62. - 136 -20. Harborne, J.B. 1965. Plant polyphenols XV. Flavonols as yellow flower pigments. Phytochemistry 4:647-657. 21 . . . 1967. Comparative Biochemistry of the Flavonoids. Academic Press, London. 22. Hartweg, T. 1848. Journal of a mission to C a l i f o r n i a . Jour. Hort. Soc London 3:220. 23. Higgins,' J<I, W. Calhoun, B.C. Willingham, .D.H. Dinkel, W.L. R a i s l e r & G.A. White. 1971. Agronomic evaluation of prospective crop species I I . The American Limnanthes. Economic Botany 25:44-54. 24. Horovitz, A. & Y. Cohen. 1972. U l t r a v i o l e t r e f l e c t a n c e c h a r a c t e r i s t i c s i n flowers of c r u c i f e r s . Amer. J . Botany '59:706-713. 25. Howell, J.T. 1943. Plantae occidentales I I I . L e a f l . West. Bot. 3:206. 26. Howell, T. 1897. F l o r a of Northwest America, Vol. 1 (Phanerograms). Portland, Ore. 27. Hutchinson, J . 1926. The Families of Flowering Plants. Macmillan & Co., Ltd., London. 28. Jepson, W.L. 1936. F l o r a of C a l i f o r n i a 2:410-412. San Francisco. 29. 1951. A Manual of the- Flowering Plants of C a l i f o r n i a . Univ. of C a l i f . Press, Berkeley. 30. - Kodak Data.Book M-27. I968. U l t r a v i o l e t and fluorescence photography. Eastman Kodak Co., Rochester. 31. Loudon, J.C. T855. Encyclopedia of Plants. London. 32. Mabry, T.J., K.R. Markham & M.B. Thomas..1970. The Systematic I d e n t i f i c a t i o n of Flavonoids. Springer-Verlag, New York. 33. ' Macior, L.W. 1971-• Co-evolution ofvplants and.animals — - s y s t e m a t i c i n s i g h t s from plant-insect i n t e r a c t i o n s . Taxon 20:17-28. 34. Macoun, J . 1883-1886. Catalogue of Canadian plants, v o l s . 1,3. Montreal. 35. Maheshwari, P. & B.M. J o h r i . 1956. The morphology and embryology of Floerkea proserpinacoides W i l l d . with a discussion on the systematic p o s i t i o n of the family Limnanthaceae..Bot. Mag. Tokyo 69:410-423. 36. Mason, C.T. , JR. 1951. Development of the embryo-sac i n the genus Limnanthes. Amer. J . Bot. 38:17-:22. 37. 1952. A systematic study of the genus Limnanthes R. Br. Univ. C a l i f . Publ. Bot. 25:455-512. 38. Mathur, N. 1956. The embryology of Limnanthes..Phytomorphology .6:41-51. 39. M i l l e r , R.W., M.E. Daxenbichler, F.R. Earle & H.S. Gentry. 1964. Search f o r new i n d u s t r i a l o i l s . VIII. The genus Limnanthes. J.A.O.C.S. 41:167-169. - 137 -40. Mosquin, T. 1966. Reproductive s p e c i a l i z a t i o n as a f a c t o r : i n the evolution of the Canadian f l o r a . In: Taylor, R.L. & R.A. LudVig (Eds.). The Evolution of Canada's F l o r a . Univ. of Toronto Press. 41. Munz, P.A. & D.D. Keck. 1963. A C a l i f o r n i a F l o r a . Univ. of C a l i f . Press, Berkeley. 42. Niemann, G.J. -1972. Phenolics from Larix needles IV. Constituents of L. l a r i c i n a . Acta Bot. Neerl. 21:549-552. 43 1973. Flavonoids from needles of La r i x l e p t o l e p i s . Phyto-chemistry 12:2056. 44. Ornduff, R. 1969a. Limnanthes vinculans, a:'new C a l i f o r n i a endemic. B r i t t o n i a 21:11-14. 45 1969h. Reproductive biology i n - r e l a t i o n to systematics. Taxon 18:121-244. 46 1971. Systematic studies of Limnanthaceae. Madrono 21:103-111. 47. " Ornduff, R. & T.J. Crovello. 1968. Numerical taxonomy of the Limnanthaceae. Amer. J . Bot. 55:173-182. 48. Ornduff,. R. & T. Mosquin. 1970...'Variation i n the s p e c t r a l q u a l i t i e s of flowers i n the Nymphoides i n d i c a complex (Menyanthaceae) and i t s possible adaptive s i g n i f i c a n c e . Canad. J . Bot. 48:603-6o6. 49. Parker, W.H. & B.A. Bohm. 1975- Flavonol glycosides of Limnanthes d o u g l a s i i . Phytochemistry 14:553-556. 50. Pardhasaradhi, M. & G.S. Sidhu. 1972. O b t u s i f o l i o l , s y r i n g e t i n and d i -hydrosyringetin from Soymida f e b r i f u g a . Phytochemistry 11:1520-1521. 51. Peck, M.E. 1937. New plants from Oregon. Proc. B i o l . Soc. Wash. 50:93-94. 52. Rydberg, P.A. 1900. Catalogue of the f l o r a of Montana and the Yellowstone Park. Memoirs of the N.Y. Bot. Garden 1:1-492. 53. ' . . .' .... 1910.. Limnanthaceae. .North American F l o r a 25:97-100. 54. ' Salisbury, E.L. 1942. The Reproductive Capacity of Plants. Bell,'London. 55- Sokal, R.S. & P.A. Sneath. 1963. P r i n c i p l e s of Numerical Taxonomy. W.H. Freeman & Co., San Francisco. 56.- Somaroo, B.H.,'M.L. Thakur & W.F. 'Grant. 1973. A u s e f u l spray reagent to d i f f e r e n t i a t e common phenolic compounds on t h i n - l a y e r plates and paper chromatograms. J . Chromatography 87:290-293'. 57- Stebbins, G.L. 1974. Flowering Plants: Evolution above the Species Level. Belknap (Harvard Univ. Press), Cambridge. 58. Takhtajan, A. 1969. Flowering -Plants: O r i g i n and Disper s a l . O l i v e r & Boyd, .Edinburgh. - 138 -59. Thompson, W.R., J . Meinwald, D. Anashansley & T. Eisner. 1972. Flavonols: pigments responsible f or u l t r a v i o l e t absorption i n nectar guide of flower. Science 77:528-530. 60. Toy, S.J. & B.G. Willingham..1966. E f f e c t s of temperature on seed germin-ation of ten species and v a r i e t i e s of Limnanthes. Economic Botany . 20:71-75. 6 l . . . 1967. Some studies on secondary dormancy i n Limnanthes. seed. Economic Botany 21:363-366. 62. Trelease, W. 1887. North American Geraniaceae. Mem. Boston Soc. Nat. H i s t . 4:84-85. ... 63. Tyukavkin, N.A., S.A, 'Medvedev & S.Z. Ivanov. 1974. New f l a v o n o l glycosides from La r i x s i b e r i c a . Khim. P r i r . Soedin. -197M2) , 157-160. 64. Von F r i s c h , K. T967. The Dance, Language and Orientation of Bees. Harvard-Uni v e r s i t y Press, Cambridge. 65. Warburg, E.F. -1938. Taxonomy and r e l a t i o n s h i p i n the Geraniales i n the l i g h t of t h e i r .cytology. New Phytologist 37:130-159,189-209-66. Willdenow.- 1801. Neue S c r i f t . Geselschaft Nat. 3:449-- 139 -APPENDIX I Mason's C l a s s i f i c a t i o n of Limnanthes R. B r . 1 Section Inflexae 1. L_. f l o c c o s a Howell var. pumila (Howell) Mason " var. h e l l i n g e r i a n a (M.E. Peck) Mason " var. flo c c o s a 2. L_. alba Benth. var.- alba " var. v e r s i c o l o r (Greene) Mason 3. L_. g r a c i l i s Howell var. g r a c i l i s " v a r . . p a r i s h i i (Jepson) Mason k. L. montana Jepson Section Reflexae 5. L_. do u g l a s i i R. Br. var. .douglasii " var. nivea Mason " var. sulphurea Mason " . var. rosea (Benth.) Mason 6. L; s t r i a t a Jepson 7. L. bakeri J.T. Howell 8. L. macounii Trelease 1 — Mason, C T . 1952. A systematic study of the genus Limnanthes R. Univ e r s i t y of C a l i f o r n i a Publications i n Botany 25:455-512. - 140 -APPENDIX II Voucher Specimens of Limnanthes Taxa Grown from Seed' OTU Taxon Name U.S.D.A. Plant Accession No. W-.H. Parker C o l l e c t i o n No. 1 L. do u g l a s i i var. d o u g l a s i i 278170 114,116,138 L. doug l a s i i var. doug l a s i i 283708 105,165 3 L. do u g l a s i i var. nivea 283713 100,166,117 1+ L. d o u g l a s i i var. rosea 283715 104,169,118 5 L. d o u g l a s i i var. sulphurea . 283718 101,170 7 L. bakeri 283706 102,120 8 L. s t r i a t a 283727 106,119 9 L. macounii 315048 112 10 L. macounii 128,167 11 L. alba var. alba 283701 111,160 12 L. alba var. alba B55689 113 13 L. alba var. v e r s i c o l o r 283705. 108,157 14 L. g r a c i l i s var. g r a c i l i s 283722 164,168 15 L. g r a c i l i s , var. g r a c i l i s - 283723 107. 16 L. gracill's var. p a r i s h i i 283724 103,159 17 L. montana 283725 109,158,161 18 L. fl o c c o s a :ssp.-. b e l l i n g e r i a n a . 283720 115,162 20 L. flo c c o s a ssp. pumila 283721 110,163 1 Voucher specimens are deposited i n the Herbarium of the Univ e r s i t y of B r i t i s h Columbia. - 141 --p a > H O C Q APPENDIX III TLC Map1 of Petal 7Flavonoids of L. d o u g l a s i i var. d o u g l a s i i , OTU 1 fl ai hO r H o -p fl CL) H o H 0 ) > Q fl O J O , -R ,-. Sr- , CD O r i g i n 1st Development, Aqueous Solvent — l ) See Materials and Methods for d e t a i l s of extractions and developments - 142 -APPENDIX IV TLC Map of Whole Plant Flavonoids of L_. d o u g l a s i i var. d o u g l a s i i , OTU 1 - 143 -APPENDIX V TLC Map of P e t a l Flavonoids of . d o u g l a s i i var. d o u g l a s i i , OTU 2 - 144 -APPENDIX VI TLC Map of Whole Plant Flavonoids of L_. d o u g l a s i i var. d o u g l a s i i , OTU 2 - 145 -APPENDIX VII TLC Map of Petal Flavonoids of L. douglasii var. nivea, OTU 3 - 146 -APPENDIX VIII TLC Map of Whole Plant Flavonoids of L_. do u g l a s i i var. nivea, OTU 3 - 147 -APPENDIX IX TLC Map of P e t a l Flavonoids of L. d o u g l a s i i var. rosea, OTU 4 - 148 -APPENDIX X TLC Map of Whole PlantsFlavonoids of L. douglasii var. rosea, OTU k APPENDIX .XI TLC Map of P e t a l Flavonoids of L. d o u g l a s i i var. sulphurea, OTU 5 - 150 -APPENDIX XII TLC Map of Whole Plant: Flavonoids of ' L_. d o u g l a s i i var. sulphurea, OTU 5 - 151 -APPENDIX XIII TLC Map of Whole Plant Flavonoids of L. vinculans, OTU 6 4 - 152 -APPENDIX XIV TLC Map of P e t a l Flavonoids of L. bakeri, OTU 7 - 153 -APPENDIX XV TLC Map of Whole Plant Flavonoids of L. bakeri, OTU 7 - 154. -APPENDIX XVI TLC Map of P e t a l Flavonoids of L. s t r i a t a , OTU 8 - 155 -APPENDIX XVII TLC Map of Whole Plant Flavonoids of L. s t r i a t a , OTU 8 - 1 5 6 -APPENDIX XVIII TLC Map of Petal Flavonoids of L. macounii, OTU 9 - 1 5 7 -APPENDIX XIX TLC Map of Whole Plant Flavonoids of L. macounii, OTU 9 - 158 -APPENDIX XX TLC Map of Petal Flavonoids of L. macounii, OTU 10 - 159 -APPENDIX XXI TLC Map of Whole Plant Flavonoids of L_. macounii, OTU 10 - i6o -APPENDIX XXII TLC Map of P e t a l Flavonoids of L- alba var. alba, OTU 11 - 161 -APPENDIX XXIII TLC Map of Whole Plant.Flavonoids of L. alba var. alba, OTU 11 - 162 -APPENDIX XXIV TLC Map of P e t a l Flavonoids of L. alba var. alba, OTU 12 - 1 6 3 -APPENDIX XXV TLC Map of Whole Plant Flavonoids of L. alba var, alba, OTU 12 4 - 164 -APPENDIX XXVI TLC Map of Pet a l Flavonoids of L_. alba var. v e r s i c o l o r , OTU 13 - 165 -APPENDIX XXVII TLC Map of Whole Plant.Flavonoids of L. alba var. v e r s i c o l o r , OTU 13 - 166 -APPENDIX XXVIII TLC Map of P e t a l Flavonoids of L. g r a c i l i s var.- g r a c i l i s , OTU ih - 16? -APPENDIX XXIX TLC Map of Whole Plant Flavonoids of L. g r a c i l i s var. g r a c i l i s , OTU 14 © (Ti? 1 0 - 168 -APPENDIX XXX TLC Map of P e t a l Flavonoids of L_. g r a c i l i s var. g r a c i l i s , OTU 15 - .169 -APPENDIX XXXI TLC Map of Whole Plant Flavonoids of L_. g r a c i l i s var. g r a c i l i s ,-• OTU 15 - 170 -APPENDIX XXXII TLC Map of P e t a l Flavonoids of L. g r a c i l i s var. p a r i s h i i , OTU 1 6 - 171 -APPENDIX XXXIII TLC Map of Whole Plant Flavonoids of L. g r a c i l i s var.' p a r i s h i i , OTU 16 - 172 -APPENDIX XXXIV TLC Map of Pet a l Flavonoids of L. montana, OTU 17 •V,' - 173 -APPENDIX XXXV TLC Map of Whole Plant Flavonoids of L. montana, OTU 17 - 174 -APPENDIX XXXVI TLC Map of Petal Flavonoids of L_. f l o c c o s a ssp. b e l l i n g e r i a n a , OTU 18 - 1 7 5 -APPENDIX XXXVII TLC Map of Whole Plant Flavonoids of L_. f l o c c o s a ssp. b e l l i n g e r i a n a , OTU 18 - 176 -APPENDIX XXXVIII TLC Map of Whole Plant.Flavonoids pf L_. f l o c c o s a ssp. b e l l i n g e r i a n a , OTU 19 - 177 -APPENDIX XXXIX TLC Map of Petal Flavonoids of L_. f l o c c o s a ssp. pumila, OTU 20 - 178 -APPENDIX XL TLC Map of Whole Plant Flavonoids < L_. f l o c c o s a ssp. pumila, OTU 20 - 179 -APPENDIX XL,! TLC Map of Whole Plant Flavonoids of L_. f l o c c o s a ssp. pumila, OTU 21 . Z - l8o -APPENDIX XLII TLC Map of Whole Plant Flavonoids of L. f l o c c o s a ssp. g r a n d i f l o r a , OTU 22 - 181 -APPENDIX XLIII TLC Map of Whole Plant Flavonoids of L. floccosa ssp. -floccosa, OTU 23 - 182 -APPENDIX XLIV TLC Map of Whole Plant Flavonoids of L_. f l o c c o s a ssp. fl o c c o s a , OTU 24 - 183 -APPENDIX XLV TLC Map of Whole Plant Flavonoids of L. floccosa ssp. f l o c c o s a , OTU 25 - 184 -APPENDIX XLVI TLC Map of. Whole Plant Flavonoids of L_. f l o c c o s a ssp. c a l i f o r n i c a , OTU 26 - 185 -APPENDIX XLVII .TLC Map of Whole Plant Flavonoids of F. proserpinacoides, OTU 27 - 186 -APPENDIX XLVIII TLC Map of Whole .Plant ..Flavonoids of F. proserpinacoides, OTU 28 - 187- -APPENDIX XLIX TLC Map of Whole Plant Flavonoids of F. proserpinacoides, OTU 29 - 188 -APPENDIX L • TLC Map of Whole Plant.Flavonoids of F. proserpinacoides, OTU 30 APPENDIX LI J 1 1 1 1 1 U-1 1 i 1 i ! 1 1 1 ! l _ J 1 1 i 1 : 1 i _ 4 I u—I ; i • • I • . . . I . . . . I • . . . I . . . . I • . . . I • ... I • . . . I • ... I I . . . . I APPENDIX L I I • • - i . . . . t . . . . i 1 . . . . i . . . . I . . . . i . . . . I . . . . i . . . . I . . . . i . . . . I . . . . t . APPENDIX LIII •100 mHz | NMR Spectrum of TMS Ether of Isorhamnetin 3-0-yft-D-Rutinoside i • ; ; • i • i • • i • 1 ; i i i 1 • i • 1 • i • i • • i • i • i • i • • • . . . I . . . . I . . . . I . . . . I . . . . I . . . . I . . . . I . . . . I . . . . I . . . . I . . . . I , . . . 1 . . . . t . . . . t . . . . t . APPENDIX LIV • 100 mHz1 NMR Spectrum of TMS Ether of Myricetin 3-0-/3 -D-Rut inos ide i 4 0 0 3 0 0 2 0 0 1 0 0 0 Hz . APPENDIX LVI 100 mHz NMR Spectrum of TMS Ether of Syringetin S-O-^-D-Rutinoside - 195 -APPENDIX LVIT. • ' . Matrix of S i m i l a r i t y C o e f f i c i e n t s Calculated f o r 30 OTUs by Jaccard C o e f f i c i e n t : High Concentrations "10 11 13 14 -T5-16 17 20 27 0. 789474 0. 714236 0. 714236 0. 714236 ~orsoooocr 0. 650000 0. 611111 0. 573947" 0. 565717 — O - 6 5 0 0 0 t r 0. 857143 0. 739474.. -0.--739130-0. 312500 0. 453333 — 0 T £ S 4 7 1 T " 0. 727273 0. 590909 • Or 7826 r09" 0. 650000 0. 772727 ""OTTJUOOOO-• 0. 652174 p. 68421. 1 ""07652174" 0. 750000 0. 545455 0. 590909 0. 73634 2 07 6 2 5 0 0 0 -0. 777778 0. 705832 0. 750000 0. 904767 0. 739130 0. 826087 0. 739130 0. 326037 0. 750000 0. 333333 -077533' O. 761905 0. 650000 0. 7 00000" 0. 739130 0. 857143 0. 666667 -07714234.-0. 750000 0. 909091 "07-555556-0. 772727 0. 666667 0." 636364' 0. 615335 0. 760000 -07-6T53T3TJ (.1. 5 / 1 4 2 V 0. 772727 0. 666667 "07 636364-O. 615335 0. 703333 0. 608696 -07652174-0. 692.303 0. 560000 0. 590909 0. 600000 "0.-565217 075714 29" 0. 4 48276 0. 772727 -07"933333-0. 521 739 0. 565217 "O. 7ST90T7 u. 6V5652 OTSaOOOO OrSSOOOCT—O7"6"10O0O~ 0. 739130 0. 600000 0. 600000 0. 630000 0. 809524 -07 830000 .0. 650000 0. 523310 "07"7 T 4 2 8 6 -0. 842105 0. 772727 0: -652174-0. 761905 O. 394737 ~ur>ppnrTA UTTWZBB—077454-51 5-0. 481431 0. 608696 O. 500000 -07-500000 0. 545455 0. 666667 0. 650000 -07"E52T74-0. 894737 0. 86:3636 -0.-625000' 0. 695652 0. 850000 i66667 0.666667 0.608696 0.590909 0.600000 0.647059 0.611111 0. 700000 "07:313132" 0. 736312 0. 573947 0. 705882 0. 77T^778_ '"a" 657174 0. 823529 0. 470533 07 590909 0. 714236 "07 7 50000" 0. 634211 0. 631579 0. 68471 1 0. 761905 "~0.~71'4?.86'~ 0. 650000 0. 777773 O. 809524 0: 947368 " O. 761905 0. 619048 "0. 8095i'4 0. 565217 0. 727273 "07-800000-0. 619043 0. 684211 "07-66666 7" 0. 736842 0. 695652 "0".""3'500'0'0'' 07"4-6"r5'T8 07-590909—07"4T6T9O—0775233T0~ 0.576923 0.30957.4 0.619048 0.666667 0. 500000 0. 636364 0. 523310 -0752"r739 0. 617059 0. 560000 "07684? IT 07-65-2T74— O. 590909 O. 520000 •O7"75OO0Cr-0. 545155 0. 631579 ~07"342103" 0. 666667 "07700000~ 0. 631579 0. 764706 ~0. /22772 O. 476190 0. 571429 0. 523310 0."7?58r3:3.T" 0. 777778 0. 500000 orsooooo"" 0. 533235 0. 761706 0. 555556 ~Q. 6OO0OO" 0. 550000 0_565217 0. 59~0?"0"9~" 0. 722222 0. 523310 "I~"84"2~l"0?~ 0r>J2"6-3T?T 0. 687500 0. 5714 29 ~07'7Z22Z7~ 0.590909 0.521739 0.5217; T~~750~~Cr TX"72 0. 608696 0.521739 0.600000 CrzrpVtTtS 07"777773" 07"E3"i 0. 555556 0. 526316 0. 476190 0. 600000 0. 368421 "07391304" 0. 562500 0. 312500 "CT2837T4~ 0. 333333 0. 470533' 0. 409091 " 0. 714286 0. 650000™ 0. 650000 0. 800000 ""67 7 50000" 0. 944441 0. 576316 0. 739130 "07"9"0d'00"0'*" 0. 736342 0. 714236. 0.600000 0^600000 0.680000 0. 481481 0.6818)3 "07"571'4?9 0. 800000 0:~6"956"5'2 07947368 07565217" 0. 500000 '0"7'7T4236" ~57"8"0750W" 0. 515155 0. 600000 0. 555556 0. 555556 0. 428571 "07450000" 0. 450000 0. 315.789 0. 619048 0. 705382 "0'~394737"' 0. 823529 0. 4761_90_ ~0. 526316 0. 500000 0. 391304 0. 428571"" 0. 333389 0. 235714 0. 565217 0. 565217 0. 652174 0. 705332 0. 61 9013 0 /2227.7 0. 800000 0. 347S26 0.634211 0.750000 0.650000 0.777778 J65217 0. 736342 0. 526316 0. 578947 0. 63421 1 0. 347826 0._666667_ "0. 681 818 0^ 520000 0. 812105" 0. 590909 0. 777778 0. 550000 0. 411765 0. 526316 0. 230000 "0731578:'" 0. 428571 0. 227273 0. 333333 0. 333333 0. 526316 "0. 47363 V 0. 555556 0. 31578? 0. 428571 0. 476190 07526516 0. 500000 0. 333333 0. 315789 0. 317876 07 4117 65 0. 526316 0. 538235 0. 562500 0. 280000 "07473631 0. 444141 0. 227273 "0. 332941"" 0. 400000 0. 317826 "0. 500000" 0. 4 70588 0. 375000 .0. 291667 0. 47365 ~0". 500000" 0. 533235 0. 307692 ' 0."409091" 0. 473681 0. 260870 0. 0. 888839 0. 263153 0. 750000 _0. 600000 PT "764706" 0. 315789 J5556 0. 421053 0. 583735 0. 533333 0. 300000 0. 352941 0. 375000 " 07500000" 0. 588235 0. 230769 07450000 0. 750000 0. 227273 ' 0^ 380952 0. 3333:33 0. 190476_ ' 0 7 473684 0. 23S093_ "0'411765 0. 600000 0. 291667 "07 555556' 0. 666667 0. 333333 0. 2S0000 0. 235294 0. 35294"! 0 7 T M 5 3 8 ~ 0. 352941 0. 416667 0. 400000 07350000" 0. 615383 0. 333333 07500000'" 0. 451545 0. 3888S9 07437500" - 196 -APPENDIX LVIIl Matrix of S i m i l a r i t y C o e f f i c i e n t s Calculated f o r 30 OTUs by Simple Match C o e f f i c i e n t : High Concentrations 0. 913013 0. 869063 0. 069:365 0. &69S65 0. 782609 ~0. 739130 0. 936322 0. 869365 0. 369365 0. 369565 0. 739130 0. 913043 0. 913013 0. 913013 0. 0. 913013 8 0. 317326 9 0. 847826 TCT 0. 826037 1 1 0. 782609 ~6~78^609 67739130 0. 391304 0. 934783 0. 891304 0. 347826 0. 817326 0. 847826 ' 07 869365 07 369365 ""0. 826037" 0. 369365 0. 869565 0. 732609 0. 869565 0. 732609 0. 391304 0. 847826 0/826037' 0. 782609 0. 7 3 9 1 3 0 0. 847826 0. 7 6 0 8 7 0 0. 80 ' !348 0 . 8 0 4 3 4 8 0 . 8 2 6 0 3 / " o r 8 2 6 0 3 7 - " 0. " 7 8 2 6 0 9 6 7 8 0 1 3 4 8 " 07 973261 0. 826087 0. 652174 0. 391304 0. 760870 0. 782609 a 391304 0 . 3 4 7 8 2 6 0. 7 6 0 3 7 0 ~ 0. 760S70 0 . 3 0 4 3 4 8 0 . 6 7 3 9 1 3 0 . 8 6 9 3 6 5 0 . 7 3 9 1 3 0 0 . 7 6 0 8 7 0 782609 _0. 326087 0. 6 9 5 6 5 2 p. 804348 0. 760870 0. 782609 847826 0 . 8 0 1 3 4 8 0 . 8 0 4 3 1 8 0 . 8 2 6 0 3 7 0 . 8 6 9 3 6 3 0 . 8 1 7 8 2 6 12 0. 347826 0. 931733 13 0.913043 0. 913043__0._86936S_ 0.782609 0. 369565 •'"07934783 "'"' 14 0. 934733 0. 347826 0. 847326 0. 847826 0. 717391 0. 782609 ~T~ 0. 869565 0. 869365 16 0. 804313 "6." 391304 17 0. 347826 0. 891304 0. 347826 0. 826087 0. 817826 "0." 304343 " 0. 760870 0. 931783 804348 869565 0. 826037" 0. 826087 18 ] 19 O. 86'; -'. o7s26"087 20 0. 913013 0. 732609 67369- .6S . 8 2 6 0 8 7 0 . 7 3 9 1 3 0 0. 9 3 1 7 3 3 0. 936522 0. 817826 0. 391304 0. 934733 0. 847876 826087" " 0". 801343" '67326037"' 0. 391301 0. 817826 0. 760370 0. 936522 0. 934783 0. 826087 "6. 8 6 9 3 6 5 ~ 0 7 9 1 3 0 4 3 0. "869565 0 . 8 6 9 3 6 5 0.' 8 6 9 3 6 5 " 0 . 8 0 4 3 4 3 0 . 7 8 2 6 0 9 0 . 8 9 1 3 0 1 0 . 3 2 6 0 8 / 0 . 8 9 1 3 0 4 0. 869565 _0._86936 0~8913d4~""' 0. 913043 0 0. 8 6 9 5 6 5 0. S26087 0. 782609 0. 7 8 2 6 0 9 0. 7 8 2 6 0 9 0 . 3 4 7 3 2 6 0 . 3 6 9 5 6 3 0 . 9 3 4 7 8 3 0 . 9 1 3 0 4 3 0 . .304348 "07739130 0. 782609 0. 695652 0. 804343 0. 760370 0. 782609 01_8/j782& O. 760870 0. 913043 0. 826087 0 :_847326 804348 0.931783 0.891301 0. 869565 0.717391 0.32608/ 0.782609 • J S 2 6 0 9 _0,_7S2609_. ._JL.S26037_ ." 891304 0.936522 0.847826 0. 826087 0. 804343 _0. 739130 0. 891304 0. 782609 0. 84 7826 _a_81782& Q, 760870 0. 732609 0. 69563 0. 847826 0. 913043 0. 847826 0. 913013 0. 826087 22 0. 391304 O. 891301 0. 804348 0. 913013 .0. 391304 0. 391304 0. 847826 0. 760370 0. 760870 0. 369365 0. 391301 0. 869363 0. 891304 0^391304 0. L.1782.6 0 76 0870 0^760870. 0. 369365 0. 934783 0. 869365 0. 934733 0. 760370 0. 0. 8 6 9 3 6 5 0. _ a .801 .34 8 0=. 0. 826087 0. 7 1 7 3 9 1 9 1 3 0 1 3 826087 804318 :2609_. 782609 891304 7826019_ 913043 0. 801318 0. 934733 0. 801318 0. 891 304 .Q,_801318.. 0. 891301 0. 847326 0. 804343 0. 760370 0. 760370 0. 804343 0. 760370 0. 826087 0. 326087 0. 804318. _ ______ _ . _ . — - _ - __ .-. .i y-. _•_ .—. n -t t \ I i~i 1 •* • r~\ / l ft O'. ' / ! / C ' ? 1 2 4 . 0. 8 0 1 3 4 3 0. 913043 0._8J.9565.. 0. 9 1 3 0 4 3 0. 891304 Jj_..826CJE___ 891304 0. 913043 0. 891304 0.826037 0.913013 0.817826 0.891304 0.9347330. 869365 0. 913013 _0. 913043_ 0^86956^_j3._7S2c^__0 ;_7i_3.609_^a 0. 973261 0. 956522 0. 304.313 0. 913043 0. 891304 0. 891301 •O. 826087 "P. S69S6S O. 782609 0. 7S2609 0_S5l6£____2 QJZS2&.Q2 (UtSlS-OA 0. 847826 0. 978261 0^847826__0^608a0___J? 1.782609 . 0. 782609 0. 869565 0. 326087 O. S47S26 0. 326037 0. 391304 0. 847826 O. 391301 0. 826087 0 .913013 _ 0. 869365 „0^7S26.09.___.0. ,_782.609-.__0._.826037 .0.773.9.1.3.0. a 891 S O l " 0 . 9 3 6 5 2 2 0 . 8 9 1 3 0 1 0 .95652.2 0 . 8 4 7 8 2 6 0 . 9 3 4 7 3 3 0 973261 0. 934783 0. 913043 0. 869565 0, 7AO«70 0. 673913 0. 673913 .0_-6:/.3i_13. ft 0. 869565 0. 891304 0. 847826 0. 913043 0. 913043 0. 913013 0. 302..343 0._8Q13.4.& Q_S260.SZ_ 0. 826087 0. 936522 0. 913013 0. 760870 0. 3 6 9 3 6 5 | J28 0. 7 3 9 l 3 0 _ "o7"693652 0. 847326 29 0. 760870 0. 804348 0. 826087 0. 326037 0. 739130 630435 0. 760870 0. 760870 0. 717391 0. 801348 0. 782609 0. 817326 0. 782609 O 826087 0. 804343 0. 817826 0. 817826 0. 693652 _ 0. 608696. 0. J.03696 ._0.._60_8696_. 6.739130 0.717391 0. 7S2609 0.717391 0 760870 0. 739130 0. 782609 0. 782609 0.673913 0.63043.J 0.63043!. 0. 630T33. 0. 8 2 6 0 8 7 0. 782609 0. 760370 0 6936"-' O. 717391 0. 847826 0. 817326 _0._S65217 O J 17391 Q_jS__J04_S_ A-&XXZ&-6. 7 6 0 8 7 0 0. 695652 0. 782609 0. 782609 0. 9 3 4 7 8 3 0 . 6 3 0 4 3 5 0 . 7 3 9 1 3 0 0 . 7 3 9 1 3 0 0 . 7 1 7 3 9 1 3 0 0. 673913 0. 782609 0. 804348 0. 717391 0. 913013 0 739130 0.717391 0. 78260V 0.760370 0.693632 0.739130 0.739130 0.826037 0.717391 0.804313 0.760870 0.913043 0 804318 0. 760870 0. 673913 0. 673913 0. 673913 .0,,630435. 07 7 8 2 6 0 9 0 . ' 8 0 1 3 4 S " 0. 782609" 0. 801313 0. 782609 0. 826087 0 826087 0. 826037 0. 801348 0. 304343 0. 847826 0. 913043 0. 760870 0. 760S70 0. 847826 0. 817826 0. 739130 P,7/3?J30.__i?J60870... 0. 717391 0. 801318 0. 801343 0. 891301 0. 869365 APPENDIX LIXV Matrix of S i m i l a r i t y Coefficients Calculated for 30 OTUs. by Jaccard Coefficient: A l l Concentrations - 197 " 0...863636. O. 714286 O. 708333 0. 8 4 6 1 5 4 0_900000_ o7607143 0. 769231.. 0,.777.778 . 6. 64.6667 0. 375000 _0__703_10A__J?_-J77_777J3._ 0. "75900 0. 800000 _Q, 720000 0.,730769.. 0. 615383 0. 730000 0. 800000_.. 0.. 7 77.77.S_ " 0.7,52 J 74 O. 6133S5 _0_--.<_ZL_.. 0. 826087 19 O. 782609. 0___i„ 31 0. 81.___54_ O. 714286 0. 730769 0. 875000 O. 789474 0. 625000 0. 576923 O. 518319 "" 0. 5S3333 0. 652174 0. 6666.67 O 600000 O. 615380, ' 0. 553556 6T"560000 0. 625000" 0. 652174 0. 633346 0 66666'/ " 0. 666667" 0. 615333 " C. 750000 0. 750000 . 0. 826.0-7.__0. .730769-0. 750000 O. 560000 0. 730769 0. 777778 0. 727273 0. 730000 0... 6.521.74 0...576923.. .0. 722722 O. 590V09 __0._ 625000 0. 615385 0. 777778 "O. 7*14286 0. 355556 0. 607143. 0. 750000 0. 837143 0...714286 0...6.29_:.30. CL.S.1.8.1.8.2 .0._63.6361. _0. 7.08333— .0.._..66_..&7 O._6.6.66.6.'„_0__3_O0.0.0.Q .0. _-7_6._2.___ .0. .830000.....0..7-684 2... _.0_'7142.._. 0. 692303 .0. .730769_ 0. 863636 20.-777.7_/8_ 0. 333333 -0..708333.. O. 640000 _0._629630.. O. 760000 0. 576923 .. 0. 631318 .0. 603696, 0. 608696 O. 739130 0-._63636__. 0. 782609 ..0.-.736842... 0._703S32 0. 789474 0. 681818 536 O. 520000 0. 700000 0. 500000 "O. 727273 0. 632174 0. 666667 O. 727273 _0._/2727_-_ J__.7_?i_3_l3_ .0. 7 00000. _O_590909_ 0. 6*31379 1 1 1 21 12 13 4 14 24 15 6 16 26 7 17 27 8 18 28 9 I V 29 10 20 . 2 r""~ 3 0. 791667J 0. 730769 0. 783714 '"0." 826037" ~o: 74074* f" 0.' "689653"*' 5 0, 730769 0. 703714 0. 733333 0. 881613 6 0. 666667 0. 724138 0. 793103 0. 814315 0. 837143 ' "7 07640000" 0 586207 ' 07 313337 "a 646667 ' " a 600000*' " a _.b'bbbo"" S ' 0. 850000 0. 680000 0. 692308 _0. 78260V 0. 692308 0. 629630 0. 600000 9 0. 736812 0. 0. 538462 0. 603696 0. 538167 0. 338162 0. 636361 0. 600000 10 0. 727273 0. 653316 0. 607143 7 ™ 615383 0. 607143 0. 607143 0, 610000 0. 608696 0. 736842 1 1 0. 666667 0. 666667 0. 7 33333 0. 580645 0. 623000 0. 6774 1V 0. 451513 0. 629630 0. 4814S1 0. 666667 _12_ 0. 680000 "6."8S4615 _0. 678571. 0. 750000 0. 336207 0. 633333 0. 689655 0. 451613 0. 703333 0. 480000 0. 615335 i 13 0. 326087 0. 807692 0. 311815 0. 703701 0. 750000 0. 750000 0. 531774 0. 708333 0. 608696 0. 680000 0. 314315 0. 769231 _ J 4 . -..P.-666667 0. 500000 _ 0 0 538162 500000 0, .613385. " 6. 625000 _<?. 675000 „P. _53.5556_ ..0 ,615385.. _..0 _5_83333_ JX_6.190.48 __,-_ 573?47_ _<_• 590909 i 15 0 625000 0 629630 0 703701 0 33:3714 0 33___Q___ 0 612837 0 148_7.6_ __C-__520.0_OQ .0. _4.7_826J_ o__-._o.ab_o 1 0. 703704 0 720000- 0. 720000 0 714286 O. 607143 o7_10000" 0. 464286. 0. 703333 0. 511.667.__0._500.QOO_ 6. 583333 6. 782609 _0._63717 4_ " 6. 619018 0. 636361 0. 533714 0. 650000 ; 24 O. 576923 0. 592393 0. 538462. 0. 666667 0. 800000 .38167 0. 592593 333333 0. 761903 592593 560000 0. 500000 0. 625000 0. 521739 0. 636364 0. 350000 0. 365217 0. 695652 0. 684211 0. 750000 0. 693652 _a_..07,__ 0. 769_31 0 0. 750000 0. 6 6 6 6 O. 8316IE 71666'/ 0. 769231_ **0. 693652 _qj314S 1.5 ' 6. 72000b' "67652174* 0 0. 607143 0 0. 666667 653316 680000 0. 727273 " 0. 730769 0. 750000 0. 714286 "0. 615333 0. 666667 0. 750000 0. 695652 0. 826037 0^81481.5_ 6. 875000 "6. 730769 0. 703333 _0._607_143 0^732609 0. 608696 0." 800000 0. 826037 0 0. 630000 ,6572 0.637174 0. 576923 0. 610000 O. 681818 0. 727273 07571429 0. 632174 0. 863636 0. 700000 0 652174 0. 653316 0. 730769 O. 666667 0. 680000 0. 750000 ' O. 750000 O. 900000 0. 800000 27 0. 619043_C^5600JX\. • | _ . . : 0. 464236 0. 72.772 "6. 500000 0. 423571 0._666667 0. 400000 0. 296296. 0. 170503 0. 666667 _ "6. 300000 O. 161533 0. 700000 '0.46 1338 0. 480000 ____.5Zi.429_ 0. 320000 0. 333333 0 100000 0_. 31 S 3 ) ' ? , " a " 533333 0 600000 0. 555556 0. 730000 0. 750000 P. 520000 0. 631579 0. 583333 0. 607) 43 0. 950000 0. 727273 O. 666667 O. 640000 0. 464286 0. 782609 0. 511667 0. 583333 0. 500000 0. 5S3333 0. 782609 0. 619048 0.318319 0. 464286 0.423077 0. 500000 0. 450000 Oj?.' 0. 666667 0. 545155 0. 608696 O. 700000 a 511*667 O. 315155 0.608696 _0_5154__i_ 0. 578917 6 0 0 0 0 0 0. 428371 0. 433077 0. 42837 l' 0. 379310 0. 0 480000 0. 500000 0. 545455 0. 300000 ' 0. _9, 476190_ _o_ 4 8 0 0 0 0 . , _ _ • _5000po_ _P: _37_1129, _o. 0. 296296 0. 333333 0. 296296 0. 296296 0. a 333333 0. 4 70588 0. 330932 0. 317826 . 0. .0. • . 1 4 1 4 1 0. 333333 b. 100000 0. 400000 " 0. _ 0. "•.'•.',•.56 0 360000 0 535556 0. 500000 0. ' 0 625000' 0 68 '• 21 1 " 0. 71 4 286 0. 652171 0. 0 . , 5 0 0 0 0 0 6-5000 0 590909 0. 750000 0. 300000 703882 291667 317826 500000 461338 937300 0. 573!. 10 0. 0. 431783 0. 0. 421053 0. 0. 400000.. 0. 0. 612837 0 513153.. 0. 0. 590909 0. 0. 761706 0. 500000  352941 :47S26 652174 4703SSS 0. 43S.83 0. 526316 0. 333333 0. 300000 0. 590909 0. 631579 APPENDIX LX —. -M a t r i x — o f - S i m i l a r i t y - C o e f f i c i e n t s Calculated f o r 30 OTUs by Simple Match C o e f f i c i e n t : A l l Concentrations - * s .*.+* » 3 0 <?TKs, Uk»le A _ * , S;~r.&_ AvicU - 198 -1 11 21 2 12. 22 13 4 14 24 16 26 7 17 27 3 18 28 29 10 20 _2 0. 891304 '3 0. 84 7826 0. 869565 4 0. 913013 '" 6. 817826'"' 0. 804343 ' 5 O. 847826 0. S69S6g_0. 826087 0. 934783 6 0. 804343 0. 826087 0. 869365 0. 391304 0. 913043 ~ 7'° :'S°4348 " ' "~07~73?^ "''0:7391.30'"" 07739130 _ 8 _ 0 . 934783 0.826087 0. 8260S7_,0. 391304 0.826037 0.782609 0. 782609 9 0.891.301 0.782609 .0.739130 0.804348 0.739130 0.739130 0.826087 0.82608/ jTT--b:m956S---re04348--o; 760870-b:'782609" ' 0. "760870 "~07760870 '767 801318"" a'804348 ^ " ^ T ^ l " I I 0.804348 Q.,732609 0.826037 0.717391 O. 7 3 9 1 _ o v ^ c 0 , - / 8 2 & 0 9 0 . - ^ - - ^ f g ^ ^ l ^ ^ - ^ ' ^ ° ^ ° S 7 0 - ^ - ^ - 0 - 4 7 8 2 6 0.71739. 0.847826 0.8043,8 C,^^ f 4 "0-71 739 ? " o : 8 0 4 3 4 S - - " - 8 - , S - - ° ^ ' 1 3 0 - 0 - 7 ^ ° * - 0...78>609_ .0. 826037__a 826087 .. 0..804 3,8_ U 1^. ..0,934783 _ 0,369565 p. 869365 rt 8 O J 3 0 4 0 •=•/-•=.".,< 5 j 0. 826087 6. 804343 0. 931783"" 0.'326dS7""'"o.7S26oV"' .. 913.043__a_826087 _p._S4787.6, ^ - - O - ^ - ^ O I — 0 _ ^ . 9 3 6 5 _ C X - 9 . 1 3 0 4 3 _ 0 . - 3 4 - / 8 2 6 _ 0 . 8 6 9 3 6 5 — O ^ S 6 0.326037 0.347326 ^ ^ ^ ^ ^ 1 ^ ^ ^ 6087 0--84j'-32_ " 15:5 * s » ' - ^ 2 3 °7 ®^!T ° 7 i : ° S 7 0 0 7 ^ 3 7 0 0. 72:9130 0. 760370 0. 760870 0. 717391 0. 760870 0. 804348 0 7S"60* 0 8478-6 0*1304- ° S - : 6 ° : 3 V a V 3 / ! 7 8 3 0 891301 0. 760870 0.801318 0. 826087 0. 847826 o! 869363 24 0.913013 0.891301 0.934783 0.869565 0.891304 0.891301 0.760870 0.89130.1 0. 801348 0 826037 o.jtltH o.ii^jl „ | j g ? a 8 , 7 8 3 6 ft8478ie6 a * 3 * 7 8 3 ft8M*" 0 9 1 3 0 , 1 3 0. 826037 0^804343 0. 847826 0. 78260? 0. 801343 0. 760870"" 6. 826037 0.""869565"0. 891301"'~07'847876" 0. 847826 0. 869563 0. 869565 0. 913043 a 847826 _0. 760870 6.' 347826' "oTS01348" a 817326 "0. "869365" 0± 30434S 0. 82608"/ 67 934783 0. 869365 26 0. 826087" 0. 80131S 0. 891304 27 0. 826087 0^  673913 6.891301" 0. S0434S 0. S47S26 0. S260S7 0. 869565 0. 956522 0. 9130'.3 0. 739730 0. 760870 0. 891301 0. 978261 0. 869365 0. 369365 0. 801318 0. 673913 0. 760870 0. 760870 0. 782609 0.801318 0.891304 0.782609 0.891301 0. 826037 0. 760870 0. 717391 0. 739130 0. 717391 0. 6956S2 a 782609 _0. 8478?6_0. 817826 _ 0. S69363 6.S26CS7" 0~7S2609 67732609 0. 673913 0. 673913 O, 804318 0. 760370 .0, 760870 a' 632174" 0. 86V565 0.695652 0. 7173*1" 0. 804318 •29 30 0. 739130 0. 386937 0. 801348 0. 847826 .0 693657 0. 918013 0. 430435 0. 652174 0. 739130. 0. 73760? 0, 7173'1 0. SV1304 _0. 652174 67 71739l'~ 0. 760S70 0. 5 6 9 5 7 0. 652174 .0,732609. 0 73»i;:0 0 80 * 7 <8 "0 3478:6" .0, 673913 0. 732609 0. 717391 0.. 652 f? 4 0. 801318 0. 652174 0. 760370 .0. 369365 0. 801318 .0. 652174 0. 782609 0. 760870 0. 586937 0. 717391 0. 739130 0. 739130 0. 369363 0. 801313 0. 869365 , 0. 608696 _ 0. 739130" 0. 801318 0. 586957 0. 673913 0,739130. 0. 693657 0. 826087 0. 891301 0. 760870 0. 782609 0. 760870 0. 804348 0. 782609 0. 826037 .0. 608696_.._0,782609, 0. 739130 0. 717391" 0. 891304 .p..693602._ 0, 71 7391 0. 739130 0. S01318 0. 630133 0. 760870 0. 673913 0. 739130 .0. 826087 . 0. 891304 0. 695632 0. 782609 0.. 782609__0. 801318. 0. 978:61 0. VI3013 0. 760870 0. 693652 0. 673913 0. 826087 0. 782609 0. .0. 626037_0. 0. 80131S 804318 817826 - 199 -APPENDIX LXI Matrix of S i m i l a r i t y C o e f f i c i e n t s Calculated f o r 18 OTUs by Jaccard C o e f f i c i e n t : High Concentrations ( 16 13 7 17 2 1 1 12 4 )5 10 1 4 ] 9 ! 8 >-— 0. 430000 ^ 7 0. 444444 0. 65517? 0. 666667 0. 560000 0. 5133)9 5 0. 736312 0. 555556 0. 51721) 0. 750000 4 0. 357143 0. 400000 0. 423077 0. 647059 0. 631579 10 0. 63157? 0. 600000 0. 615383 0. 736317 0. 714286 0. 6111)1 1 0. 323529 0. 428571 0. 400000 0. 571479 0. 636361 0. 7058.37 0. 545155 0. 526316 0. 570000 0. 600000 0. 631579 0. 619018 0. 500000 0. 777778 0. 451315 IS 0. 466667 0. 291667 0. 269231 0. 4.1 1765 0. 3.50000 0. 428571 0. 388889 0. 170388 0. 352941 16 0. 714286 0. 434783 0. 400000 0. 625000 p. 5263.1 6 0. 692308 0. 5832.35 0. !. '83783 0. 4 70388 0. 636361 13 0. 0. 266667 363636 0. 2 i7391 0. 200000 0. 3)2500 0. 263)53 0. 307692 0. 294118 0. 222277 0. 250000 0. 414 411 17 0. 0. 200000 272727 0. 0. 1739)3 800000 0. 160000 0. 250000 0. 210326 0. 230769 0. 235294 0. 166667 0. 187300 0. 333333 1 1 0. 0. 444444 571479 0. 0. 521739 416667 0. 0. 430000 0. 473684 0. 40909) 0. 500000 0. 450000 0. 380952 0. 35CCC0 0. 47837j 12 • 0. 0. 500000 642837 0. 0. 5652)7 384 615 0. 0. 520000 307692 0. 0. 576316 923077 0. 454543 •0. 4 70588 0. 300000 0. 128571 0. 400000 0. 500000 15 0. 0. 4 28571 600000 0. 0. 260870 571429 0. J L 240000 47837! 0. 0. 375000 500000_ 0. 0. 3)5789 461533 0. 500000 0. 332941 0. 352941 0. 235294 0. 555556 14 0. 0. 363636 444141 0. 0. 695652 312500 0. 0. 610000 250000 0. 0. 523810 750000 0. 0. 453333 705882 0. 0. 400000 375000 0. 571429 0. 320000 0. 4 76190 0. 333333 20 0. 0. 428571 0. 0. 434783 250000 0. 0. 400000 272727 0. 0. 444444 692303 0. 0. 330952 612857 0. 0. 375000 0. 0. 421 OSS 675000 0. 285714 0. 315739 0. 2357)1 - 200 -APPENDIX LXII ; . Matrix of S i m i l a r i t y C o e f f i c i e n t s Calculated f o r ' j 18 OTUs by Simple Match C o e f f i c i e n t : High Concentrations ! r 3 . s 7 2 5 4 10 1 9 18 16 53 17 1 1 17 13 14 \ s 0. 717391 r 7 0. 673913 0. 782609 •7 0. 369565 0. 760370 0. 717391 0. 391304 0. 739130 0. 693657 0. 891.304 4 0. 956522 0. 673913 0. 673913 0. 869565 0. 847826 10 0. 347326 0. 782609 0. 787609 0. 891304 0. 869363 0. 84 7826 1 0. 934733 0. 652174 0. 603696- 0. 304348 0. 826087 0. 891304 0. 732609 9 0. 804348 0. 739130 0. 782609 0. 847876 0. 826087 0. 80434 8 0. 913043 0. 739130 "IS 0. 826037""" "6Sf0".3_- 0758-6937 ""07 ".77391 0. 827-ro"87 " .377'67.870 0. 80434 8 0. 760870 16 0. 913013 0. 717391 0. 673913 0. 869363 0. 80134(-. 0. 913043 0. 847876 0. 34 732.6 0. 301318 0. 913013 13 0. 760370 0. 608696 0. 565217 0. 760870 0. 693637 0. 804343 0. 739130 0. 695652 0. 739130 0. 891301 0. "847826" 17 0. 739130 0. 586957 0. 513478 0. 739130 0. 673913 0. 782609 0. 71739) 0. 673913 0. 717391 0. 869363 0. 826087 0. 97877-rr 11 0. 782609 0. 760370 0. 717391 0. 787609 0. 717391. 0. 826037 0. 760870 0. 717391 0. 717391 0. 826087 0. 869365 0. 847826 67 '826087 12 0. 804348 0. 782609 0. 739130 0. 801348 0. 739130 .0. 304348 0. 782609 0. 789180 0. 739130 0. .34 787.6 0. 391304 0. 876037 0. 801318 0. 978261 15 0. 826087 0. 630435 0. 586937 0. 782609 0. 717391 0. 869565 0. 760870 0. 760870 0. 717391 0. 913043 0. 913043 0. 931783' 0. 913013 07869365"" 0. 817826 14 0. 695652 0. 847876 0. 804343 0. 782609 0. 717391 0. 739130 0. 804318 .0. 630435 0. 760870 0. 789130 0. 782609 0. 760370 0. 739180 0. 913013 0. 891301 0. 787609 70 0. 739130 0. 717391 0. 673913 0. 787609 0. 717391 0. 787609 760870 0. 673913 0. 7)7391 0. -0. 826087 0. 801348 0. 826037 0. 913043 0. 891301 0. 326087 869365 - 201 -APPENDIX LXIII Matrix of S i m i l a r i t y C o e f f i c i e n t s Calculated f o r 18 OTUs by Jaccard C o e f f i c i e n t : A l l Concentrations 7 8 0. 481431 7 0. 451613 0. 727273 -3 0.- 6666.1,7- ~0.~500000— 0.-468750-5 O. 636364 0. 548387 0. 538824 0. 652174 j 4 °" 837143 0. 407407 0. 387097 0. 617C59 0. 545435 10—0...4 6 VSaa^^UUO^ ^ 0 603696 0.580613 0.312857 0.625000 0.607143 0.321739 0.366667 9 0.634211. 0.371429 0.380643 0.789474 0.818)82 0.666667' 0.680000 0.576923 : ' 16 0.750000 0.588462 0.45,613 0.666667 0.5652,7 0.733333 0.383333 0.608696 0.68121,0 6 6 6 6 6 . - _ q . l - 0 - ft 4 6 , 9 3 8 °*«™\ o-^oooo 0.-00000 0 . 5 0 0 0 0 0 0 . 5 2 1 7 3 9 0 . 1 2 3 3 7 , 0 . 6 1 7 0 5 9 1 7 I S ° ' : ^ ° ^ 5 0 0 0.438333 0.47826, 0.880957 0.388233 n-^:.550S0.J::7038-.. l ^ ^ ^ ^ ° <>• ^ ^ S ^ ^ S ^ ^ X Z™™ 0,^0000.0.^00000 0.625000 0.652,74 0.37,179 0.777227 ', 0. 617039 l - - : - - o „ _ ? : ^ ^ ^ ^ ^ ^ ^ ^ ™ ° < » * ^ Q Q Q a 0.482759 0 . ^ 0 0 0 0 . 5 0 0 0 0 0 - 202 -APPENDIX LXIV . . ' Matrix of S i m i l a r i t y C o e f f i c i e n t s Calculated for 18 OTUs by Simple Match C o e f f i c i e n t : A l l Concentrations 3 0 7 7 5 A 10 16 13 17 11 12 ' I S 11 O. 693652 7 0. 630135 0. 301318 -_—0-369565-—Or-6 93652—0r-63043S-0.826087 0.695657 0. 673913 0.826087 4 0.936522 0.652174 0.586957 0.869365 0.782609 10—0,-69565-2—Ox-*S260S»—0.-7-1-739-1—0^-739-1^0—0^-7-3-9-1^:0—0t-A*S*S2-1 0.804348 0.7173?) 0.652174 0.304348 0.760870 0.760870 0.717391 9 0.369365 0.739130 0.717391 0.9)3013 0.913013 0.869563 0.87.6087 0.760870 „L8 0. 8260S7-....-0-695652. 0-5369.37 0.-7.32.609 0.-652O.4—0.-S7.6037-—0.7394 30 0.-84-7826—0.-739130 -16 0. 913043 0. 739)30 0. 630433 0. 869363 0. 787609 0. 91.3013 0. 782609 0. 804318 0. 869365 0. 869365 13 0.782609 0.69365? 0.536937 0.782609 0.637171 0.826087 0.739130 0.760870 0.739)30 0.869563 17 0.760870 0.673913 0.5657)7 0.760870 0.630433 0.801348 0.717391 0.739130 0.7)73?) C. 817826 0. 847826 C L _ — 2 & 1 : 11 0.717.39) 0.71739) 0.608696 0.7)7391 0.630433 0.760870 0.760870 0.787609 0.7)73?) 0.817826 0. 80.43.48. -0...89130-4—.0.-869.365 : ; : I 12 0 . 8 0 4 3 4 8 0 . 7 6 0 8 7 0 0 . 6 5 7 ) 7 4 0 .804318 0 . 7 ) 7 3 9 ) 0 . 8 0 4 3 4 8 0 . 8 0 1 3 4 8 0 . 8 2 6 0 8 7 0 .801348 0 . 8 9 ) 8 0 1 j fl 391.304 n 93478.3 O 913043 O 91304.3 \ • 15 0 . 7 3 2 6 0 9 0 . 6 9 5 6 5 2 0 .586957 0 .787609 0 . 6 3 2 1 7 1 - 0 . 8 7 6 0 8 7 0 . 7 3 9 1 3 0 0 . 7 6 0 8 7 0 0 .739130 0 369365 0. 369365 —-000000 0. . .973261—0„.8?.1.30.4—0_i>3.4.233 : : .. • 14 0.652174 0.826087 0.630133 0.693632 0.632)74 0.693652 0.782609 0.6739)3 0.739)30 0.739)30 0. 7?9) ,:o 0 826Q37 n 804348 _ , iy—. 1.3——34732.6 CU-82-60S7 20 0.695652 0.787609 0.536937 0.695632 0.603696 0.739)30 0.739)30 0.717391 0.693652 0.737609 I-•JO"—0.-S9.1304—0^-39-1304—0.-869363—0^-$'4S04S . - 203 -APPENDIX LXV Coordinates of 30 OTUs Plotted i n Figure XV FULL EVJDI-NCF: 85 OHARAO'TFR 30 TAXON IJHOL.F PLANTS FACTOR ANALYSIS 3 5. 1 0 4 7 2 - 1 . 3 5 0 4 8 3 . 7 4 8 7 5 8 2. 3 8 1 9 0 1. 3 4 9 4 9 3. 6 3 3 5 2 12. 5 6 7 6 3 16. 1 8 7 3 3 -7. 0 8 1 3 8 ( 2 2. 6 8 3 2 7 - 1 . 5 5 8 2 2 0. 3 6 5 2 5 5' 6. 4 9 2 5 6 0. 5 0 3 7 4 2 3 7 7 7 8 6 7. 9 6 0 7 5 -1. 9 0 3 7 4 5. 3 6 7 2 7 4 5. 8 6 2 9 0 - w. 8 7 7 7 9 4. 0 2 5 0 3 10 -3. 9 2 4 3 3 - 2 . 9 1 8 4 8 - 0 . 3.8532 1 - J . 6 1 4 0 3 - 1 . 7 2 0 0 9 - 2 . 5 6 1 7 9 9 - 2 . 7 7 4 4 5 - 3 . 3 2 7 3 4 - 0 . 513 3 8 21 - 3 . 4 1 6 7 6 0. 0 3 8 6 4 - 3. 6855:": 2 5 --0. 8 9 3 9 4 - 0 . 978.1 4 - 0 . 4 7 1 6 8 19 0. 6 4 4 3 2 - 0 . 8 8 1 8 9 - 0 . 9 6 8 3 7 2 4 2. 4 4 7 8 1 -1. 2 2 5 4 J 0. 0 7 2 3 8 18 1. 6 4 4 7 8 - 1 . 8 7 5 1 4 0. 8 8 2 3 2 . 16 4. 1 2 7 2 0 -.1. 0 5 2 1 4 2. 2 3 2 4 2 13 2. 3 4 4 6 5 - ] . 33 3 0 0 - 0 . 5 8 1 4 3 17 i . 4 7 4 9 2 - 1 . 0 9 4 4 4 - 0 . 4 9 0 2 2 11 5. 1 4 3 2 9 -•). 3 2 0 0 5 3. 2 5 8 6 5 12 i . 6 2 8 8 8 - 1 . 0 3 1 6 5 0. 2 6 3 7 3 15 0. 5 5 2 8 3 - 0 . 9 9 3 2 6 - 0 . 9 5 4 6 7 14 2 7 0 6 1 -.1. 9 8 1 5 1 - 2 . 6 3 4 9 2 2 2 3 9 7 2 1 -3. 23 4 2 4 -3. 2 7 7 7 3 2 0 - 3 . H i 7 3 - 1 . 8 1 0 8 1 - 2 . 4 0 2 8 3 2 6 0. 3 2 8 9 4 - 1 . 174 36 -3 . 8 3 7 6 8 2 3 - 2 . 3 9 2 9 6 - 1 . 8 0 7 8 9 - 2 . 2 8 9 7 6 2 9 - 1 6 . 1 3 7 6 4 6. 3 3 9 9 3 2. 1 3 8 7 4 2 8 - a 2 2 5 2 9 2. 5 0 8 0 4 - 0 . 9 0 2 3 4 2 7 - 1 0 . 9 9 7 6 6 5. 3 2 9 0 8 3. 8 6 7 J 3 3 0 - 1 1 . 0 2 9 2 1 4. 8 0 7 8 8 3. 03 3 53 C<=-rcL\w\iedr o f OlX^i - 204 -APPENDIX LXVI Coordinates of 30. OTUs Plot t e d i n Figure XVI' f H J L L K-:VIDFMI::F£ 31 CHARACTER 30 I AXON WHO. E PLAN. 8 H ALT. IR ANAI 3 3. 21928 0.62447 1.49291 8 1. 50211 -0.57834 2.753 7... 7 2. 52230 21. 45496 1. 86687 2 2.55308 -0. 11239 1.76760 2. 75269 2. 56790 5. 82146 3. 68256 0. 30923 8. 94659 4 2. 87802 - 0 , 23809 6. 76683 10 0.50069 -3.21428 -3.29830 I 1721282 -1 . 27731 = ! T ^ 6 3 T 9 0. 3 "279 -3. 827-.4 -2. 06476 21 -1.40398 -0.95383 -3.41115 25 0.21078 -1.053 47 -0.93 03 4 19 3.40729 -0.3 03 09 -0.37409 24 2. 25668 0. 08534 3. 54 88 3 18 i . 69776 -1. 0912"/ ' 1. 71099" 16 2. 18008 0. 26560 4. 26759 13 2: 56977 0. 5.8476 0. 96630 17 1. 83000 -0. 05950 0. 51094" 11 2.35664 0.33 730 5.7 3 596 12 1. 51780 -0. 19365 3. 20201 15 1.40422 -0.23906 -0.41786 14 0. 94048 -1 . 72693 -3. 45868 22 2. 24218 0. 0747.1 -0. 3 6 3.85 20 0.20707 -2.09082 -3. 784 00~ 26 1.79377 -0. 19609 -5.23886 23 0.61740 -1.79867 -3.25904 29 -15.08429 -3.04708 -8. 13 638 28 -6.33614 -1.38548 -5.71783 27 -11. 3.3691 -3. 27503 -5. 23 33 7 30 -10. 43195 -1. 42650 -5. 90945 - 205 -APPENDIX LXVII Coordinates of 18 OTUs Plotted i n Figure XVII 7 FULL - T K f l l — 0. 56728 -2. 94 474 -5. 7 3 675 8 10. 82163 -5. 20182 6. 18045 7 15. 55555. 6. 72893 —2. 47233 2 3. 47656 -2. 56263 -5. 20709 6. 08708 -3. 973 78 - J . 85070 (4 -0. 8339.. -2. 5 5 3.05. -1 . 39448 10~ 5. 03554 -3. 88063 -1 . 24277 1 3. 21599 -4. 38083 -3. 774 43 9 3. 05944 0. 23383 -3. 34 4 5.2 18 -7. 5.5.308 0. 727 5 1 -0. 5 3 984 16 -3. 97997 -0. 89735 -0. 94380 13 - 5.0. 335 32 3. 0879.5 5. 035 06 17 - 1.1. 52882 -v 79703 1. .32 J. 6 3 11 -2. 26424 0. 55 5 95. 5. J 8 5.9 5 12 - i . 59037 0. 22936 0. 854 29 1.5 -8. 85.080 1. 64442 0. 80772 14 3. 02657 J. 4 5 857 2. 08492 20 -4. 39254 3. 57670 5. 5 5384 - 206 -• APPENDIX LXVIII Coordinates of 18 OTUs Plotted i n Figure XVIII f FULL (f-F-A^ V 2. 80535 39534 -1. 45724 • 8 6. 46086 2. 86507 10. 38435 7 7. 53850 14. 4 2979 5. 31503 2_. 4. 46101 0. 30339 0. 34355 7. 40208 0. 841 45 0. 86824 1. 21 04 1 -1. 64758 - 1 . 75707 10 5. 1 4429 1. 68-5 3 4 1. 06514 1 5. 60514 -1. 05267 -0. 89649 9 2. 44199 2. 28316 0. 2 1 2 9 8 " ' . ~ 18 --5. 45416 -3. 24 556 -3. 29506 16 --5L 95039 ' —2. 531 21 -2. 70632 13 -9. 58709 -3. 54636 8. 584 07 17 -10. 97436 -3. 74321 •.*>. 82487 11 --2. 324 62 - 1 . 19221 0. 0554 9 12 - 1 . 51074 - 0. 96795 0. 04995 15 -7. 37026 -3. 59780 -3. 62824 14 0. 601 15 ^ / 05435 27592 20 -4. 46916 - 1 . 54248 -0. 871 29 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0100112/manifest

Comment

Related Items