k[CYTOTAXONOMIC STUDY OF THE POLYPODIUM VUIGARE COMPLEX /IN NORTHWESTERN NORTH AMERICA by Frank Alexander Lang B. S. Oregon State U n i v e r s i t y , 1959 M. S. Un i v e r s i t y of Washington, I96I A t h e s i s submitted i n p a r t i a l f u l f i l l m e n t of the requirements f o r the degree of Doctor of Philosophy i n the Department of • Biology and Botany We accept t h i s t h e s i s as conforming to the required standard The U n i v e r s i t y of B r i t i s h Columbia J u l y 1965 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an advanced degree a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r a g r e e t h a t p e r -m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by t h e Head o f my Department o r by h i s r e p r e s e n t a t i v e s , , I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i -c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department o f B i o l o g y and B o t a n y The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada Date 10 September 196$ i x ABSTRACT: Supervisor, T. M. C. Taylor The circumboreal Polypodium vulgare complex consists of a series of closely related ferns of different ploidy levels. Two northwestern North American members of the complex, P. glycyrrhiza D. C. Eaton, and P. hesperium Maxon, have long been a source of taxonomic confusion since l i t t l e has been known of their relation-ships within the complex in the Northern Hemisphere. The present cytotaxonomic investigation of these taxa has shown that they are composed of three cytotypes, two diploid and one tetraploid. P. glycyrrhiza has proved to be uniformly diploid (n = 37) and morphologically, ecologically and geographically distinct from P. hesperium. Investigation of P. hesperium has shown that this taxon, as usually treated by North American taxonomists, is composed of at least two distinct entities, one tetraploid (n = 7h), the other diploid (n = 37). These two cytotypes are morphologically and ecologically as well as cytologically separable, and have indepen-dent geographical distributions. The type specimen of P. hesperium is morphologically comparable to tetraploid populations from the interior of British Columbia, and specimens from the type locality have proved to be tetraploid. It is recommended that P. hesperium be reserved for the tetraploid cytotype. The epithet montense is tentatively proposed for the diploid cytotype. Two morphologically distinct triploid hybrids were found in areas of sympatric occurrence of the three cytotypes. Morphologically i i i these hybrids appear to be P. hesperium x P. glycyrrhiza and P. hes-perium x P. montense. At meiosis both hybrids showed n II • n I, which is interpreted to mean that the montense genome and the glycyrrhiza genome are both present in P. hesperium. P. hesperium sensu stricto appears to be of alloploid rather than autoploid ori-gin since i t forms only bivalents at meiosis. P. hesperium is also intermediate in morphology and ecology between P. glycyrrhiza and P. montense. It is postulated that P. hesperium is an allotetraploid derived from a pre-Pleistocene hybridization between r . glycyrrhiza and P. montense or their immediate progenitors. The hypotnesis is also made .that P. hesperium originated largely because of climatic i changes in the interior of the continent imposed by pre-Pleistocene orogenic activity. Morphology, ecology and geographical distribution indicate three main lines-of differentiation among the diploid cytotypes. These diploids eventually gave rise to polyploid derivatives, pro-, bably in the late Tertiary before the advent of Pleistocene glacia-tion. The intergradation and morphological variability of these' taxa are attributed to alloploidy, hybridization and phenotypic plasticity. The morphology, biochemistry, ecology and geographical distribution of the three species is circumscribed and discussed and a pragmatic key provided. iv TABLE OF CONTENTS CHAPTER PAGE I. SCOPE AND OBJECTIVES 1 II. APPROACH, MATERIALS AND METHODS OF STUDY 5 III. RESULTS 13 Cytology 13 Morphology-anatomy 20 The Gametophyte i|2 Biochemistry U2 Ecology and Geographical Distribution US IV. TAXONOMY 59 Discussion of Nomenclature 59 A Pragmatic Key to the Species of Polypodium in Northwestern North America 67-68 Species Descriptions 69 Polypodium glycyrrhiza 69 ' ' Specimens Cited 70 Polypodium montens'e 76 Specimens Cited 77 Polypodium hesperium 80 Specimens Cited 81 Triploid Hybrids 83 Polypodium virginianum 8U V TABLE OF CONTENTS (continued) CHAPTER PAGE V. DISCUSSION OF RESULTS 85 Origin and Relationships of the Polypoclium vulgare Complex in Northwestern North America 85 Taxonomic Confusion and the Intergradation and Variability of Forms 101 V I . SUMMARY AND CONCLUSIONS 102; LITERATURE CITED 107 APPENDIX ' 113 vi LIST OF TABLES TABLE PAGE 1. Chromosome Numbers in Polypodium in Northwestern North America lb-15 2. Summary of Frond Measurements of Polypodium glycyrrhiza, P. hesperium A and P. hesperium B • 22 ,3. Phenology: Time of Annual Production of New Fronds and of Spores h9 k. Characters Distinguishing Polypodium glycyrrhiza, P. hesperium, P. montense and P. yirginianum ,... 62 5. Chromosome Counts of the Polypodium vulgare Complex in North America ....„, 90 v i i LIST OF MAPS MAP PAGE 1. Approximate world distribution of members of the Polypodium vulgare complex 2 2. Distribution of Polypodium (except P. scouleri) in Northwestern North America 53 V . v i i i LIST OF FIGURES FIGURE PAGE 1. Quantitative measurements made on Polypodium vulgare s . l . 11 2. Photographs of meiosis and mitosis i n Polypodium 17 3. Photographs of meiosis and mitosis i n Polypodium ........ 18 U. Camera lucida drawings of meiosis and mitosis i n Polypodium 19 $. Graphic summary of frond measurements i n Polypodium '23 6. Morphology of Polypodium glycyrrhiza 2h 7. Morphology of Polypodium hesperium A 2$ 8. Morphology of P. hesperium B and P. virginianum 26 9. Morphology of putative hybrids Polypodium hesperium A x glycyrrhiza and P. hesperium A x hesperium B 27 10. Correlation of frond segment length versus width 30 11. Correlation of blade width, frond segment length to width r a t i o and frond segment t i p shape 31 12. Distributions of the numerical index of frond segment shape 32 13. Epidermal patterns i n Polypodium 33 I i i . Scatter diagram showing correlations of stoma length to stoma frequency 37 lf>. Rhizome scales i n Polypodium 38 16. Spores and paraphyses of Polypodium 31 ( ix LIST OF FIGURES (continued) FIGURE PAGE 17. Correlation of mean spore length to mean stoma length ... k3 I'd. Habitats of Polypodium hesperium A, P. hesperium B and P. glycyrrhiza hi 19. Field and voucher collections of Polypodium $0 20. Comparison of mean stoma length vs. frequency for Polypodium 56 21. Graphic presentation of climatic data from selected weather stations from the ranges of Polypodium in British Columbia 58 22. Relationships in the P. vulgare complex 88 ( X ACKNOWLEDGEMENTS Sincere thanks i s expressed to Dr. T. M. C. Taylor f o r h i s help and c r i t i c i s m . The i n t e r e s t and assistance of Dr. K. I. Beamish, Dr. J . K u i j t , Dr. W. B. Sc h o f i e l d , Dr. G. E. Rouse, and Dr. J . R. S t e i n during the course of t h i s study i s also g r e a t l y appreciated. And I wish to thank my wife Suzanne f o r her encouragement and a s s i s -tance i n the completion of t h i s t h e s i s . CHAPTER I i SCOPE AND OBJECTIVES The Polypodium vulgare complex i s a somewhat nebulous aggre-gation of morphologically similar ferns characterized by a creeping rhizome with articulate, mostly free veined, glabrous fronds and exindusiate sori. As seen in Map I the taxa of the complex are dis-tributed around the Northern Hemisphere and occupy distinct although often overlapping ranges. In addition there are also reports i n the literature of P. vulgare i n the Southern Hemisphere (Christensen 1928; Sims 1915; Gilbert 1899). On the basis of these reports one must ten-tatively accept the fact that the P. vulgare complex is probably fur-ther complicated by a bipolar distribution. In view of the morphology and distribution of the complex, the question naturally arises, i s P. vulgare a single widely ranging species or is i t an aggregation of related similar species? In North America members of the P. vulgare complex have been variously treated from nearly complete synonomy with P. vulgare of Europe to recognition as distinct species. The only major taxonomic treatment of the North American P. vulgare complex i s that of Fernald (1922). On the basis of morphology and geographical distribution he accepts Linnaeus' recognition of P.• virginianum as distinct from P. vulgare. His view is basically the one subsequently adopted by most taxonomists. Since Fernald's To face page 2 Map 1. Approximate world distribution of members of the Poly- podium vulgare complex (after Martens 1950, Shivas 196lb, and Hulten 1962). P. californicum P.* glycyrrhiza P. hesperium s.l. P. virginianum ooooooo P. australe P. interjectum + +++t+ P. vulgare 2 3 classical investigation, however, new evidence and new methods have been discovered which justify a re-examination of the complex in North America. During the past two decades, in particular, many new techniques have been developed for taxonomic and evolutionary studies of pterido-phytes. Manton's (1950) perfection of the squash technique for the study of fern chromosomes has enabled the pteridologist to make use of cytotaxonomic and cytogenetic methods to aid in solving many perplexing taxonomic and evolutionary problems. Modern taxonomists utilize data from many sources, e.g., ecology, genetics, cytology, biochemistry, to supplement information from comparative morphology-anatomy. Inherent in this new approach is the study of living wild populations. Using Manton's methodology, Shivas (1961a and b) has been largely successful in clarifying the taxonomic and evolutionary situa-tion in the European members of the P. vulgare complex. The completion of her work s t i l l leaves two extensive geographical areas, Asia and North America, where very l i t t l e is known about the cytotaxonomy and probable evolution of the plexus. The present study, limited to a critical cytotaxonomic investi-gation of certain members of the Polypodium vulgare complex in North-western North America, is an attempt to help f i l l this gap in our knowledge. As defined in this st\idy, Northwestern North America com-prises the area from Alaska-south along the Pacific Coast to central California and east to the Rocky Mountains. The genus Polypodium in this geographic area revolves around two principal nodes; one is represented by P. glycyrrhiza D. C. Eaton, k the other by P. hesperium Maxon. At t h e i r morphological extremes these two types are d i s t i n c t , but they are so v a r i a b l e that t h e i r precise boundaries are often b l u r r e d and i t i s often d i f f i c u l t to d i s t i n g u i s h them. Several species, i n addition to P. g l y c y r r h i z a and P. hesperium, are also found i n the area, but are omitted from t h i s study. P. scou-l e r i Hook & Grev., d i s t i n c t i v e i n i t s morphology and d i s t r i b u t i o n , o f f e r s no problems and i s not dealt with here. In c e n t r a l C a l i f o r n i a P. californicum Kaulf. and P. g l y c y r r h i z a are apparently involved e v o l u t i o n a l l y with each other (Lloyd and Lang 196U, Lloyd 1962). Since t h i s s i t u a t i o n involves only g l y c y r r h i z a i n a small part of the present study area and does not g r e a t l y influence t h i s i n v e s t i g a t i o n , consideration of P. cal i f o r n i c u m and i t s accompanying problem i s deferred u n t i l another time. P. virginianum L., known from North-eastern B r i t i s n Columbia and the Yukon, i s considered here only, as i t r e l a t e s t o P. hesperium. The present study has four major o b j e c t i v e s : 1) t o inve s t i g a t e the v a r i a b i l i t y and intergradation among the species; 2) to examine the r e l a t i o n s h i p of these taxa with other members of the complex i n the Northern Hemisphere; 3) to re-examine the nomenclature and taxonomy of the complex i n Western North America i n l i g h t of the new evidence; ii) to c l a r i f y and elucidate the probable o r i g i n of the Western North American species of Polypodium. CHAPTER I I APPROACH, MATERIALS AND METHODS OF STUDY I n o r d e r t o s o l v e t h e t a x o n o m i c and e v o l u t i o n a r y p r o b l e m s p r e s e n t e d b y t h i s h i g h l y v a r i a b l e c o m p l e x i n N o r t h w e s t e r n N o r t h A m e r i c a a number o f p o p u l a t i o n s were e x a m i n e d s y s t e m a t i c a l l y f r o m v a r i o u s p o i n t s o f v i e w . F a m i l i a r i t y w i t h t h e s p e c i e s was g a i n e d t h r o u g h a s u r v e y o f t h e l i t e r a t u r e , b y h e r b a r i u m and l a b o r a t o r y s t u d i e s o f g r o s s and m i c r o s c o p i c m o r p h o l o g i c a l f e a t u r e s , b y o b s e r v a t i o n s o f l i v i n g p l a n t s , a n d b y m a k i n g a c y t o l o g i c a l e x a m i n a t i o n o f t h e number and b e h a v i o r o f t h e i r chromosomes . P o p u l a t i o n s , p r i m a r i l y f r o m B r i t i s h C o l u m b i a , w e r e s t u d i e d , mass c o l l e c t i o n s made, chromosomes c o u n t e d and l i v i n g m a t e r i a l b r o u g h t b a c k f o r g r e e n h o u s e c u l t u r e . T h e s o u r c e s o f l i v i n g p o p u l a t i o n s u s e d i n t h i s s t u d y a r e l i s t e d i n T a b l e I . The l o a n o f h e r b a r i u m m a t e r i a l s b y t h e c u r a t o r s o f t h e f o l l o w i n g h e r b a r i a i s g r a t e f u l l y a c k n o w l e d g e d : U C , U n i v e r s i t y o f C a l i f o r n i a , B e r k e l e y ; U S , U n i t e d S t a t e s N a t i o n a l H e r b a r i u m (Type o f P . h e s p e r i u m ) ; V , P r o v i n c i a l Museum, V i c t o r i a , B . C ; WS, W a s h i n g t o n S t a t e U n i v e r s i t y ( T y p e o f P . amorphum); WTU, U n i v e r s i t y o f W a s h i n g t o n ; Y U , Y a l e U n i v e r s i t y ( T y p e o f P . g l y c y r r h i z a ) . M e i o t i c chromosomes were o b s e r v e d i n s p o r e m o t h e r c e l l s (SMC) f r o m s p o r a n g i a l s q u a s h e s made f o l l o w i n g t h e t e c h n i q u e s o f M a n t o n (1950). T h e chromosomes w e r e s t a i n e d i n a c e t o c a r m i n e a f t e r f i x a t i o n i n C a r n o y ' s f i x a t i v e (12-2U h r s . ) . T h e m o r d a n t , a s a t u r a t e d s o l u t i o n 6 of iron acetate-acetic acid, was added to the fixative as part of the normal amount of acetic acid until the solution was a light straw colour. Meiotic figures were obtained from young fronds bearing small white sori: yellowing sori usually were too old, and the mature sporangia interfered with the squashing. Material collected in the forenoon seemed to have more divisions than that collected at other times of the day. Most of the counts were made on SMC at diakinesis. Root tips and small circinate fronds were used for mitotic counts. Excised root tips and croziers were pretreated in a saturated aqueous solution of para-dichlorobenzene for 1 to 3 hours. Before fixing in Carnoy's solution the material was washed in water. After fixation the material was washed in several changes of 70 per cent alcohol until a l l the fixative was removed. It was then bulk stained in Snow's hydrochloric-alcoholic carmine (Snow 1963) and stored in 70 per cent alcohol at 0-3*C. Squashes were made in U5 per cent acetic acid and the slides temporarily sealed with paraffin gum arabic. Some of the slides were made permanent by mounting in Hoyer's medium i (Benson 1963). Whenever possible, a photomicrograph was made of the analyzed cells. Photography was not always possible because of the large num-ber of chromosomes, their frequent clumping, and poor spreading of SMC. Some of these cells, nevertheless, could be accurately counted by using a camera lucida and phase contrast microscope. In normal species most of the spores are approximately the same size. Spores produced by hybrids, however, are usually irregular 1 and exhibit a range i n size from large to very small aborted spores, as can be seen in Figure 16, page I4I. Aborted spores produced by ferns are often used as an indication of hybridity when taken i n conjunction with other c r i t e r i a such as intermediate morphology (Manton 19 $ 0 ; Shivas 196lb anu 1962; Wagner and Chen 196$; Walker 1955). Since spore abortion may be caused by factors other than hybridity, such as high temperature, i t i s wise to use i t largely as confirmatory evidence. The original drying papers were retained with the specimens so that spores could be collected from them and mounted in Hoyer's solution for examination. In the case of herbarium specimens suspected of being hybrid, i t was often necessary to remove unopened sporangia,, break them open, and examine the spores for abortion. Two types of apomixis, a phenomenon whereby sexual fusion i s circumvented, have been well described by Manton (1950X meiotic, and Evans (196U), ameiotic. Both types produce 32 spores per sporangium rather than the 6JU spores found in most normal sexually reproducing ferns. The number of spores per sporangium were counted to ascertain i f apomixis was present i n the complex. Morphological structures, such as sporangia, spores, paraphyses, and rhizome scales requiring microscopic examination, were mounted directly in Hoyer's mounting medium without fixation. Hoyer's medium seems to preserve the tissue or structures with l i t t l e , damage or ef-fect on size, and i t has the additional advantage of being water soluble. Paraphyses were easiest to find i n immature s o r i , especially 8 in material collected for cytological examination. As the sorus matures, the paraphyses become: senescent and are broken off, probably by the developing., enlarging sporangia. When examining herbarium specimens for the presence of paraphyses, i t is necessary to soak the sorus with detergent and scrape off pieces of the receptacle as well as the sporangia onto the slide. Epidermal features, the shape of upper and lower epidermal c e l l s , size and distribution of stomata, and morphology of the hyda-thodes, were studied using a cellulose acetate technique similar to that of Sax and Sax (1937). When a small drop of cellulose acetate i s spread on the upper and lower epidermis of the segments of mature fronds, the solution flows over the surface forming impressions of the c e l l walls, stomata, and hydathodes. When dry, after 10-1$ minutes, the cellulose acetate with i t s replica of the ^surface can easily be peeled from the frond. The peel, shiny side up, was fastened to a microscope slide with "Scotch" tape. From this slide camera lucida drawings and measurements were made with relative ease. Unfortunately, these preparations were usually not suitable for photography. The i great advantage of this technique i s that the epidermis of herbarium specimens can be examined without damage. Gametophytes for morphological and cytogenetical study were grown using the method outlined by Walker (1955). S o i l was chosen as the medium to avoid the abnormal growth reported for gametophytes on liquid or agar media (Steeves, Sussex and Partanen 19$$; Pickett and Thayer 1927). The manipulations for synthesizing hybrids are the same as those of Walker (1955) and Shivas (1961). Gametophytes for 9 comparative study were fixed and stored i n 60 per cent formalin-aceto-alcohol (Johansen 1°1|0) u n t i l they could be examined. The terminology used to describe the simple symmetrical plane shapes of frond outline, frond segments, and rhizome scales i s that of the language equivalents of shapes advocated by the Systematics Association Committee for Descriptive Biological Terminology (ly62). Spore and stomata lengths were measured and stomatal frequen-cies computed to see i f any differences existed between populations of known ploidy level. In order to reduce some of the sources of v a r i a b i l i t y i n c e l l size, only specimens with mature dehisced sporangia were used. Plants from a variety of ecological conditions were measured to provide norms that might be used to indicate the ploidy level of herbarium specimens. If differences showed only when plants were grown under uniform conditions, such measurements would not be neces-sary since ploidy level could be ascertained by actual chromosome determinations. The length of the f i r s t 25 normal mature spores encountered was measured using an ocular micrometer, and a mean spore length calculated. Stoma length and frequency were measured from cellulose ace-tate impressions (made as described earl i e r ) . Care was again taken to see that only mature fronds were used. The lengths of 25 stomata per plant were measured and means computed from three selected areas of a frond segment. The selected areas were near the margin of the segment, near the costa, and medially between the two. This was done ID to assure mean values that would represent the stomata from a l l areas of the frond. To measure the frequency of stomata per unit area, the 10X eyepieces and objectives were used. One eyepiece of the binocular microscope was blacked out and a piece of opaque paper with a square cut out of i t was placed into the other. The area on a slide v i s i b l e through this square was computed with a stage micrometer. Stomata lying completely within the square were counted from 20 different areas on each segment, and the mean number for the area was calculated. 2 In order to convert this figure to mean number of stoma per mm , i t was necessary to multiply by the factor . 1 8 . The matter of describing and comparing fronds quantitatively posed problems particularly as to the measurements that would best serve for this purpose. A preliminary t r i a l of frond measurements showed the following linear dimensions to be the most discriminative in describing frond v a r i a b i l i t y among the taxa: stipe length, blade length and width, and the length and width of the longest frond seg-ment. The original dimensions of the frond segments varied con-siderably from plant to plant, but the ratio of length to width was much more constant. For this reason the ratio of length to width of the longest frond segment from the middle of the frond was computed. A number of quantitative methods were tried to describe segment t i p shape, but they were later discarded i n favor of scoring i t on an arbitrary scale. An attenuated t i p was given a score of 1, the most rounded a score of 3, with intermediate values of 1.5, 2.0, and 2.5. Figure 1 shows the linear measurement and the position where i t To f a c e page 11 F i g u r e 1. Q u a n t i t a t i v e measurements made on P o l y p o d i u m v u l g a r e a . L i n e a r measurements; E L - b l a d e l e n g t h Bw - b l a d e w i d t h P I - p i n n a e l e n g t h Pw - p i n n a e w i d t h S l - s t i p e l e n g t h b . N u m e r i c a l i n d e x f o r e x p r e s s i n g t h e shape o f t h e f r o n d s egment s . 11 12 were made on the frond. This figure also shows the approximate shape of the segments and their scored values. The observed range, the mean, and the range between which 80 per cent of the observations f e l l were used in presenting the quanti-tative data. The 80 per cent range was chosen as a quick convenient way to describe the distribution of the observations so that the measurements would be more meaningful. The study of herbarium material was greatly aided by the use of data sheets (for an example see Appendix, page 113) • These provide a permanent record of the herbarium material examined, the information and observations being recorded quickly and efficiently. The localities of specimens identified by the writer were recorded on a Rand McNally road atlas of North America. Each species was given a symbol and the collector and his number written by the locality. In this way the locality could be cross-referenced to a particular specimen and a l l the accompanying data. The scale of the base map used for Map 2 made i t impractical to indicate a l l the stations of a species. Stations were, therefore, i selected that would display the general distribution of each species. Field observations and herbarium specimens showed that dif-ferent populations produced new fronds at different seasons. Fertile herbarium material was examined noting whether the sporangia were immature, f i l l e d with mature spores, or dehisced, and i f young cir -cinate fronds were present. By correlating these findings with the month the specimen was collected, i t is possible to make some state-ments as to the phenology of the species. CHAPTER III RESULTS • •' ' I. CYTOLOGY Wagner (I9$h), Britton (1953), Manton (1950) and Shivas (196la and b) offer considerable evidence that polyploidy in the pteridophytes is generally accompanied by slight morphological differences. It was thought, especially in view of Manton's and Shivas' findings in Europe, that the variation exhibited in Western North American members of the P. vulgare complex might be due to polyploidy. Accordingly, a cytological approach was taken for the entire study. ~^ For each taxon, the chromosomal configuration at diakinesis (meiotic association), the chromosome number of the sporophyte, the collection locality, the collector and the collection number of populations analyzed cytologically in this investigation are given in Table 1. 1 P. hesperium consists of two cytotypes (see:Table 1) , one a diploid with 37 II at meiosis, the other a tetraploid with 7I4. II. For convenience of reference the tetraploid is designated as P. hes-1 perium A, the diploid P. hesperium B. Of special note in Table 1 are the three triploid populations designated as P. hesperium AxY. These are undoubtedly hybrids with the tetraploid cytotype serving as one of the parents. The possible identity of the second, diploid, parent is discussed after the morphological results are presented. I l l TABLE 1. CHROMOSOME NUMBERS IN POLYPODIUM IN NORTHWESTERN NORTH AMERICA Species hesperium A Meiotic 2n Assoc. 7k II 7h II 7h II 7U II 7k II 7h II 7h II 7h II 7U II 7h II 7k II hesperium B 37 II .37, II 37 II 37 II Ui8 1U8 1U8 1U8 1U8 1U8 1U8 1U8 lii8 7h 7k 37 II Source of Material 11.5 miles N of Clearwater, Br i t i s h Columbia, Lang 70 . 10 miles S of Clearwater, Br i t i s h Columbia, Lang 71 . Mara Lake, B r i t i s h Columbia, Lang 72. Craigellachie, British Columbia, Lang 73. Wigwam, Columbia River, Br i t i s h Columbia, Lang 7k. Kootenay Bay, Kootenay Lake, Bri t i s h Columbia, Lang 102. ca. 10 miles S of Clearwater, British Columbia, near Lang 71, Lang 111. H4 miles S of Clearwater, British Columbia, Lang 112. lU.5 miles S of Clearwater, B r i t i s h Columbia, Lang l lU . Coyote Creek, Lake Chelan, Washington, Lang 130, 138. 3 miles N of Green River, Br i t i s h Columbia, Lang 210 A and B. E of Alta Lake, Br i t i s h Columbia, Lang 212. Marblehead, B r i t i s h Columbia, Taylor and Szczawinski 610. Cambie, B r i t i s h Columbia, Taylor and Szczawinski kk2. Stream draining into Goldie Lake, Mount Seymour, Bri t i s h Columbia, Lang 96 . 11 miles N of Cheekeye, Cheakamus River, British Columbia, Lang 117-C-2. 1 mile N of Yale, British Columbia, Lang 210, 118. Alexandra Bridge, Fraser River, Bri t i s h Columbia, Lang 122-6. McGuire, Cheakamus River, Br i t i s h Columbia, Lang 211. W of Daisy Lake, Garibaldi, British Columbia, Lang 21U. S face of Dinky Peak, Mount Seymour, Brit i s h Columbia, Lang 218. 15 TABLE 1 (continued) glycyrrhiza 37 I I 37 I I 37 I I 37 II 37 I I 37 I I 37 I I 37 I I 37 I I 37 II 37 I I 37 I I 37 I I 7k hesperium AxY 37 I I + 1 1 1 37 I 1 1 1 37 I I + 37 I Lewis and Clark State Park, Troutdale Oregon, Lang 188, 18°_. .5 miles E of Crown Point, Columbia River Gorge, Oregon, Lang 190, 192. 2 miles S of Masset, Queen Charlotte Islands, B r i t i s h Columbia, Lang 2k, U.6 miles E of Royal Canadian Navy radio s t a t i o n , Tow H i l l Road, Graham Island, Queen Charlotte Islands, Lang 2$. St. Mary's Spring, Graham Island, Queen Charlotte Islands, Lang 28. North face of Tow H i l l , Graham Island Queen Charlotte Islands, Lang 30. North end of Copper Bay, Moresby" Island, Queen Charlotte Islands, Lang 33. 8 miles E of Terrace Bridge, Skeena River, B r i t i s h Columbia, Lang 53. .75 miles N of Squamish, B r i t i s h Columbia, Lang 221. Murrin Lake, Howe Sound, B r i t i s h Columbia, Lang 222. P i t t River, B r i t i s h Columbia, Lang 216, 217. 1 mile NW of Cheekeye, B r i t i s h Columbia, Lang 106. West of Daisy Lake, Garibaldi Station B r i t i s h Columbia, Lang 213. 2 miles S of Furry Creek Bridge, Howe .Sound, B r i t i s h Columbia, Lang 223. Alexandra Bridge, Fraser River, B r i t i s h Columbia, Lang 125-2. 3 miles N of Green River, B r i t i s h Columbia, Lang 210-C. Alexandra Bridge, Fraser River, B r i t i s h Columbia, Lang 122-3. 16 These triploids have 37 II + 37 I at meiosis. P. glycyrrhiza is uniformly diploid with 37 bivalents. The 37 bivalent chromosomes in P. glycyrrhiza are shown in Figures 2e and l*c. Chromosomes from the sporophyte generation, 2n = 7h9 are shown in Figure 2f. There are 7h bivalent chromosomes at diakinesis (Figures 2a and lib) in P. hesperium A. The sporophyte of this cytotype has 11*8 chromosomes as seen in Figures 2b and he. P. hesperium B has 37 bivalent chromosomes at diakinesis, Figures 2c and Ua, and 7h chromosomes in the sporophyte, Figure 2d. ; Meiosis in these ferns follows the pattern already described by Manton (19!?0) for the majority of the pteridophytes studied. Chromosome pairing is regular in the diploids, and in the tetraploids only bivalents are formed. On the basis of spore number there is no evidence of apomixis. Metaphase I of -meiosis in the triploid hybrids shows 37 b i -valents plus 37 univalent chromosomes (Figure 3a, explanatory diagram lid; Figure 3b, explanatory diagram Ue). Mitosis in P. hesperium AxY is shown in Figure 3f (explanatory diagram Iih) and Figure hg with 2n - 111. The effect of hybridity on the later stages of meiosis can be compared with similar stages in normal species in Figure 3c, d, e, and g. Scattered univalent chromosomes can be seen between the nuclei in telophase II of meiosis in P. hesperium AxY (Figure 3c). They are absent from the same stage in P. hesperium A (Figure 3d). Metaphase I in these triploid hybrids is shown in Figure 3e. Here the univalent chromosomes are seen scattered in the cytoplasm To f a c e page 17 Figure 2. Photographs of meiosis and mitosis i n Polypodium. a. Diakinesis in P. hesperium A. n = 7h. Arrow indicates nucleolus. FAL 111-C. b. Mitosis i n P. hesperium A. 2n = IJ48. Taylor and Szczawinski 610. c. Diakinesis i n P. hesperium B. n = 3 7 . FAL, 120-10. dl. Mitosis i n P. hesperium B. 2n - 7k. FAL 120-2. e. Diakinesis i n P. glycyrrhiza. n = 3 7 . FAL 223. f. Mitosis i n P. glycyrrhiza. 2n - 7k. FAL U9. 1 To face page 18 Figure 3. Photographs of meiosis and mitosis in Polypodium1. a. Metaphase I in P. hesperium AxY. 37 II * 37 I« Explanatory diagram Figure Ud, page 19 . FAL 122-3. b. Metaphase I in P. hesperium A x Y . 37 II + 37 !• Explanatory diagram Figure he, page 19 . FAL 125-2:. c. Telophase II in P. hesperium AxY. Note scattered univalent chromosomes in the cytoplasm between the nuclei. FAL 122-3. d. Telophase II in P; hesperium A. Note lack of univalent ' chromosomes. F A L l l l - C . e. Lateral and polar views of metaphase I in P. hesperium A x Y . Note univalent chromosomes off the metaphase plate. FAL 125-2. f. Mitosis in P^ hesperium A x X. Photographs taken at different levels of focus in the same c e l l . 2n - 111. Explanatory diagram Figure l\h, page ±9 • F A L 125-2. g. Lateral view of metaphase"I in P. hesperium B. FAL 12Q-9. \ \ To face page 19 Figure k. Camera lucida drawings of meiosis and mitosis in Polypodium, a. Diakinesis in P. hesperium B. n = 37. FAL 117-C2. b. Diakinesis in P. hesperium A from the type locality, n - 7k. FAL 130-U. c. Diakinesis in P. glycyrrhiza. n - 37. FAL 189. d. Explanatory diagram for Figure 3a, page 18. P. hesperium AxY. FAL 122-3. e. Explanatory diagram for Figure 3b, page 18. P. hesperium AxY., FAL 125-2. f. Mitosis in P. hesperium A. 2n = lU8. FAL 210-A. g. Mitosis in P. hesperium AxY. 2n * i l l . FAL 210-C. h. Explanatory diagram for Figure 3f, page 18. P. hesperium AxY. 2n = 111. FAL 125-2. 19 20 off the metaphase plate. This i s contrasted by a latera l view of metaphase I i n P. hesperium B. (Figure 3g) where there i s no evidence of univalent chromosomes. D i f f i c u l t y was encountered in finding cells where chromosome counts could be determined accurately because the bivalents clumped together and would not spread out when squashed; furthermore some of the smaller bivalents approached the size of the larger univalent chromosomes. The pairing i n the triploids always approached the i "Drosera" type (Wood 1955) which in this case i s 37 II + 37 I. Other representative counts were 3k II + 1)0 I and 3 6 II + 38 I. These departures from the Drosera type pairing are probably caused by con-fusion of bivalent with univalent chromosomes rather than actual pairing differences. Neither Shivas (196la and b) nor Manton (1950) report other than the expected n II + n I i n similar t r i p l o i d hybrids which they have examined. As mentioned above t r i p l o i d hybridity also shows an effect on spore formation resulting i n abortion. Aborted spores are shown in Figure 16c, page i l l . Figures 16a, b, and d show normal spores from P. hesperium A, P. glycyrrhiza, and P. hesperium B. II. MORPHOLOGY-ANATOMY The two cytotypes of P. hesperium, P. glycyrrhiza, and the t r i p l o i d hybrids were carefully compared morphologically to discover what similarities and differences existed 6 among them. The results of this aspect of the study are presented next. The frond i s the most conspicuous part of the sporophyte, and 2 1 its shape, features of its segments, and epidermal morphology provide many useful taxonomic criteria for delimiting taxa and perhaps indi-cating relationships. Other parts of the sporophyte, the rhizome and reproductive structures, are also of taxonomic value, but the incon-spicuous gametophyte is morphologically uniform among" species and thus contributes nothing to the recognition of our species. Photographs of these cytotypes and drawings of some of their morphological features are shown in Figures 6-9, 13, 1$, 16. The taxonomic characters of frond size and segment shape are frequently used to circumscribe the taxa of the Polypodium vulgare complex. The value of such quantitative characters in the complex was tested with the results summarized in Table 2, page 22, and shown graphically in Figure $, page 23. The relationships of several measurements of taxonomic interest are demonstrated in Figures 10-12. Because of the small sample size, the triploids are not included in the measurements. Frond shape differs slightly among the cytotypes as shown in Figures 6-9 and 19. In P. glycyrrhiza the frond varies from oblong, i (Figure 6f), to ovate (Figure 19a). There is no difference'between the P. hesperium cytotypes (compare Figures 7 and 8). It is apparent that there are two different triploid hybrids, one shown in Figure 9a, b, c, the other in Figure 9d. The fronds in Figure 9a, b, c, are more or less narrowly ovate while the frond in Figure 9d is oblong. As seen in Figure 5 the three cytotypes are difficult to separate using the quantitative characters of blade length, blade width, and stipe length primarily because of the overlap in size range. One item of 1 TABLE 2. SUMMARY OF FRONDMEASUREMENTS OF POLYPODIUM GLYCYRRHIZA, P. HESPERIUM A AND P. HESPERIUM B Measurements in mm Blade Blade Stipe Segment Segment Ratio Segment t i p length width length length width Seg. 1/w index -P. g l y c y r r h i z a N. 325 354 375 339 287 349 210-> 0. R.— 20-470 14-156 11-340 7-85 3-11 1.8-12.00 1.0-3.0 Mean 172.6 60.30 93.56 33.28 6.65 5.6 1.65 80%- obs. 69-280 35-92 36-143 18-53 5^8 3.3-7.4 P.hesperium A N 199 171 223 218 . 215 284 206 0. R. 30-265 12-83 12-155 9-45 5-11 1.3-3.8 1.5-3.0 Mean 114.09 33.29 56.75 17.64 7.25 2.44 2.72 80% obs. 63-189 22-48 28-100 11-25 6-9 1.9-3.1 P. hesperium B N 216 224 213 221 220 315 217 0. R. 18-190 11-45 8-142 5-25 3-12 1.2-3.6 2.0-3.0 Mean 80.58 23.58 57.94 12.92 / 5.60 2.35 2.99 80% obs. 46-122 17-30 28-100 9-17 ' 4-7 1.8-3.0 £i See Figure 1 page 11. b Observed range. £ The two extreme values between which 80 per cent of the observations f e l l . To face page 23 Figure $. Graphic summary of frond measurements in Polypodium based on Table 2, page 22. 23 Blade 5 0 0 r 'ength 1 R glycyrrhiza 250 OL FROND MEASUREMENTS (mm) Segment width E4=*—O.R. 1 2 10 Stipe 1 length 2 R hesperiumA 3 R hesperium B 1 Blade 1 width 1 3-0, 2-o; 1.0' Segment tip index 1 2 3 100 Segment length 1 Segment l/w 1 3 I 1 T 50 0 L 1O0 5.0 o-o1-To face page 2k Figure 6. Morphology of Polypodium glycyrrhiza. a. Type specimen of P. glycyrrhiza.. A. V. Kautz s.n. b. Diploid P. glycyrrhi za growing near e. FAL I89. c. Portion of blade of type specimen of P. glycyrrhiza. Note round median sori. d. Diploid P. glycyrrhiza growing on old Alnus. Note lobing on segment margina. FAL 26. e. Small depauperate diploid P. glycyrrhiza growing near b. FAL 188. f. Diploid P. glycyrrhiza. Note rounded segment tips and slightly oval sori. FAL 190. 1 To face page 2$ Figure 7. Morphology of Polypodium hesperium A. a. Type specimen of P. hesperium A. M. W. Gorman 6h2. b. Portion of type specimen showing large median oval sori. c. Frond of tetraploid P. hesperium A. Note large median oval sori. FAL 111. d. Type specimen of P. vulgare var. columbianum. A photo-graph of a photograph of the original specimen. e. Portion of blade of tetraploid P. hesperium A. Arrow indicates elongate receptacle of the sorus. FAL 111. f. Tetraploid P. hesperium. FAL 210-A. To face page 26 Figure 8. Morphology of Polypodium hesperium B and P. virginianum. a. Diploid P. hesperium B with round sori and blunt frond segments. FAL 211-B, b. P. hesperium B (probably diploid). Schofield s.n. c. Portion of blade of diploid P. hesperium B with round marginal sori. FAL 122-6. f d. Diploid P. hesperium B from more or less exposed habitat. FAL 2l£. e. P. virginianum from Northeastern British Columbia (ploidy level not known but probably diploid). Raup and Correll 10119. f. Diploid P. virginianum from He Perrot, Quebec."• FAL 1|6. g. >i II n II II II II II •• h. P. hesperium B. Note position of sori. Krajina 29 Aug. 1950. i To f a c e page 2 ? F i g u r e 9. Morphology of p u t a t i v e h y b r i d s Polypodium hesperium A x hesperium B and P. hesperium A x g l y c y r r h i z a . a - c . T r i p l o i d P . hesperium A x g l y c y r r h i z a showing f e a t u r e s of the f r o n d , c . D e t a i l s o f s o r a l shape and p o s i t i o n . FAL 2 1 0 - C . d . T r i p l o i d P . hesperium A x hesperium E . FAL 1 2 2 - 3 . 28 particular interest is the intermediate position of P. hesperium A between P. glycyrrhiza and P. hesperium B. In P. glycyrrhiz a the frond segment margins are serrate to serrulate with spreading or appressed teeth (Figure 6). The margin in P. hesperium A may be crenate (Figure 7c, e, g) or entire (Figure 8), although there are'often small notches in the margins at the terminal vein endings and some segments have some fine serrations at the tip (Figure 8a). One triploid (Figure 9d) has more or less entire margins while the other hybrid has serrate margins similar to those of P. glycyrrhiza. Frond segments with lobes are found sporadically in a l l three taxa (Figure 6d, 7d). Such lobing is of no practical taxonomic value. Particular attention has been paid to the use of segment length, width and shape because these characters are.frequently used to separate P. hesperium from P. glycyrrhiza. Although segment length is sometimes useful for separating • P. glycyrrhiza from P. hesperium, the ratio of segment length to width seems to be more reliable. As can be seen in Figure $ the 80 per cent range of segment lengths in a l l three species overlaps. On the other hand, the 80 per cent ranges of the ratios of segment length : width between P. glycyrrhiza and the other two cytotypes do not overlap. However, there is s t i l l considerable overlap in the observed ranges of the ratio. The relationship of segment length to width is shown in Figure 10. Here the similarity of expression of these dimensions is particularly evident between P. hesperium A and P. hesperium B. The relationship of blade width and the ratio of segment length : i 29 width (Figure 11) exhibits a similar picture. Taxonomically, frond segment shape is perhaps the most useful of the characters treated quantitatively (shape is actually qualitative). For the method used in scoring tip shape, see page 10. The distribution of the segment shape index in the three taxa expressed as a percentage of the total number of observations is shown on the bar diagram on page 32 (Figure 12). P. glycyrrhiza is the most variable with two tip shapes particularly prevalent, 33 per cent with attenuate tips (1.0) and hO per cent with acute tips (2.0). P. hes-perium B is the least variable with most of the plants having obtuse segments (3.0). P. hesperium A more closely resembles P. hesperium B with more than 60 per cent of the plants having obtuse segments. There is , however, considerable overlap with P. glycyrrhiza. These simi-larities and differences are further expressed by the "means" in Figure 5 for the segment shape index. The mean for P. hesperium Mis 2,7 and P. hesperium B .3.0, while the mean for P. glycyrrhiza is 1.6. A general trend, evident in Figure 11, partially explains the overlap in segment shape between P. hesperium and P. glycyrrhiza. There is a tendency towards more sharply pointed segment tips with an increase in frond size (expressed in Figure 11 as increased blade width). Smaller specimens of P. glycyrrhiza have blunter segments, while larger specimens of P. hesperium have more pointed, ones. The morphology of surface epidermal cells has recently proven to be valuable in taxonomic and evolutionary studies of the pterido-phytes (Bladsell 1963; Wagner 195U; Evans and Wagner 1961j). Although epidermal morphology does not provide any clear diagnostic characters To face page 30 Figure 10. Correlation of frond segment length versus width in a number of individuals of the complex from different populations. Notice the overlap of P. hesperium A and B. and the tendency of P. hesperium A to be intermediate between P. hesperium E and P. glycyrrhiza. The numeral 2 indicates the number of individuals at a particular point. i \ 60 O O 50 • O R g lycy r rh iza • R hesperium A A R hesper iumB £ £ 401- O o 30-o • 10-o 8 A A • ft r2 -2 A 5 6 7 8 9 10 FROND SEGMENT WIDTH (mm) To face page 31 Figure 11. Correlation of blade width, frond segment length to width ratio, and frond segment tip shape in a number of individuals of the complex from different populations. Notice that the larger specimens of each cytotype tend to have more sharply pointed frond segments. 31 90 80 70 E i 60 50 LU < _ l CD 40 30 20 A Segment tip index o 1-0 € 1-5 . e 2-0 ® 2-5 3.0 OS Ll © 0 A A A i l A A J L 9 O O © P. g l y c y r r h i z a 0 R h e s p e r i u m A A R h e s p e r i u m B o o o i _ 7-5 1 5 2 0 25 3 0 3-5 4-0 F R O N D S E G M E N T 4-5 50 5-5 60 L E N G T H : W I D T H 6-5 7-0 To face page 3 2 Figure 1 2 . Distributions of the numerical index of frond segment shape expressed as per cent of total in Polypodium glycyrrhiza, P. hesperium A, and P. hesperium B« 32 lOOr 80 60 c Ci) c 40 30 20 10 Numerical index of frond segment shape in Polypodium n= 210 R glycyrrhiza 1.5^2.0 2-5 3-0 R hesperium A LO 1.5 2-0 2-5 3-0 . R hesperium B To.face page 33 Figure 13. Epidermal patterns in Polypodium. a-f. Lower (left) and upper (right) epidermis. g-1. Hydathodes of upper epidermis. a. Tetraploid P. hesperium A. lower epidermis. FAL 111, upper epidermis. FAL 7U. b. Triploid P. hesperium A x glycyrrhiza. I.e. FAL 125-2. u.e. FAL 210-C. c. Diploid P. glycyrrhiza. I.e. FAL 217. u.e. FAL 33. d. Triploid P. hesperium A x hesperium B. 1. and u.e. FAL 122-3. e. Diploid P. hesperium B. I.e. FAL 218. u.e. FAL 211-G. f. Diploid P. virginianum. 1. and u.e. FAL U6. ... g. P. hesperium A x glycyrrhiza. FAL 210-C. h. P. hesperium A. FAL 71. i . P. glycyrrhiza. FAL 2k» j , P. hesperium A x hesperium B. FAL 122-3. k-1. P. hesperium B. FAL 215. ." 3k for distinguishing members of the P. vulgare complex, these cells do exhibit patterns which are useful in assessing relationships and for use as corroborative evidence in indicating the possible parentage of hybrids. The shape of epidermal cells differs in the lower and, parti-cularly, the upper epidermis among these taxa. The cells in P. gly- cyrrhiza are narrow and elongate (Figure 13c), while in P. hesperium B (Figure 13e) they are broadly oblong in outline. The epidermis of P. hesperium A (Figure 13a) is more or less intermediate between the two diploids. As. seen in Figure 9 there are two morphologically different triploids. A comparison of Figure 9 with Figures 6, 8 shows that one of these triploids, Figure 9a-c, is intermediate in gross morphology between P. hesperium A and P. glycyrrhiza, while the other fall s between P. hesperium A and P. hesperium B . The intermediate state of the triploids is also apparent in their epidermal features. Figure 13d (from the plant shown in Figure 9d), with its blockier cells is intermediate between P . hesperium B (Figure 13e) and P . hesperium A (Figure 13a). The other triploid, Figure 13b, with its longer cells, is intermediate between P. hesperium A and P . glycyrrhiza. In addition to cel l shape there are also differences in cell size and stoma frequency evident in Figure 13. These differences are apparently related to polyploidy. The cells of the polyploid taxa, shown in Figure 13a, b, d, are larger than those of the diploids in Figure 13c, e, f, and the stomata are few in number. In the two diploids, the difference in stomata size (Figure 13b and d) suggests 35 the additional action of genetic factors in determining cell size. Differences in cell size and stomatal frequency are best expressed in quantitative terms. Means of stomal' length and fre-quency taken from individual plants of each cytotype from different geographical areas are plotted together in Figure lh» Here i t is apparent that the stomata of the polyploids average more than 1*2 yi - long, while those of the diploids are less. The most distinctive epidermal feature of the upper frond sur-face is the hydathodes, which (is? located at the vein endings. Outlines of their cells are shown in Figure 13g-l. The hydathodes of P. hes- perium B generally tend to be smaller and composed of fewer cells than those of the other cytotypes. The rhizome of P. hesperium B is slender and compact, while in comparably sized plants of P. glycyrrhiza and P. hesperium A i t is thicker and more fleshy. Furthermore, the rhizome of P. hesperium B is sometimes pruinose, while those of the other cytotypes are not.. It is difficult to get comparable measurements of the rhizome because ,of size differences imposed by age, amount of available moisture, and the size of the plant. A small plant of P. glycyrrhiza might have a thinner rhizome than a normal sized plant of P. hesperium B, but i f compared to a plant of P. hesperium B of the same size, i t would be thicker. The rhizome scales of P. glycyrrhiza (Figure l5d) are uni-formly colored, and tapering with a relatively smooth margin. When compared with similarly sized scales from the other cytotypes, those of P. glycyrrhiza are composed of smaller more numerous cells. The 36 scales of P. hesperium B (Figure I5g) often have a median strip of darkly stained cells, and the margins are coarsely toothed. In P. hesperium A (Figure l$b) the scales are morphologically intermediate between P. glycyrrhiza (Figure l$d) and P. hesperium B (Figure l|?g). The rhizome scales of the two triploids, as shown also in their epidermal patterns, are intermediate between the tetraploid P. hesperium A and P. glycyrrhiza on the one hand and P. hesperium B on the other. Fernald (1922) used "scales peltate" versus "scales hot pel-tate" to distinguish P. virginianum from P. vulgare s . l . Subsequent investigation-(Lloyd & Lang 1961*) has shown that the scales in P. glycyrrhiza (considered a variety of P. vulgare by Fernald) are not peltately attached. A sinus with overlapping lobes"' leading from the point of attachment to the outside gives the scales of both taxa the appearance of being peltately attached (Figure 15). It should be noted that the rhizome scales at the stipe bases are essentially truncate and not cordate in a l l cytotypes. Bobrov (1961i) uses characters of the rhizome scales, parti-i cularly the distribution of marginal teeth and the presence of a sclerenchyma strand, together with spore characters to separate the Russian species of Polypodium. The present investigation shows these .1 characters to be quite variable and they do not appear to be useful. There has been no opportunity, however, to check the validity of Bobrov's findings. Reproductive structures have long been used in the classifi-cation of pteridophytes. Sorus shape and location, presence and To face page 37 Figure Iii. Scatter diagram showing correlations of stoma length to stoma frequency in a number of individuals of the complex from different populations. Notice how the symbols tend to segregate into two groups, one representing the diploid cytotypes, the other, polyploid cytotypes. \ 37 10Cf 9C4-80r CM E £ LU o LU _J 60r 00 50r-IX 40r © O © 1 A A CP A A A A ( © R glycyrrhiza E R hesperium A A R hesperium 8 S • 30 A G3 1 • • Q 20*-30 35 40 45 xSTOMA , LENGTH (u) . 50 To face page 38 Figure 15. Rhizome scales in Polypodium. a. Triploid P. hesperium A x glycyrrhiza. FAL 210-C. b. Tetraploid P. hesperium A. FAL 212-C. c. Triploid P. hesperium A x hesperium B. FAL 122-3• d. Diploid P. glycyrrhiza. FAL 213» e. Diploid P. virginianum. FAL U6. f. P. virginianum. Raup and Correll 10119. g. Diploid P. hesperium B. FAL 211-B. 38 39 morphology of paraphyses, and details of the anatomy of the sporangia have been found by other investigators to be useful taxonomic char-acters in the Polypodium vulgare complex. The sori of P. glycyrrhiza are circular (Figure 6), or at most, slightly oval, as seen in Figure 6f. In P. hesperium the sori are of two shapes: P. hesperium A has oval sori (Figure 7), while P. hes-perium B has circular sori (Figure 8). The sori of both triploids (Figure 9) are oval. It should be noted that sorus shape is often obscured in mature specimens by opened sporangia. In contrast, the sorus shape is usually quite clear in specimens with immature spo-rangia. Shivas (1961a ana b) found the number of annular cells a use-ful taxonomic character. She utilized plants of known ploidy level grown under uniform conditions. Rothmaler and Schneider (1962) counted the number of cells per annulus using wild material. They found con-siderably more overlap in numbers, particularly between P. vulgare and P. inter jecturn, and so minimized the value of this character. Counts were made of the number of annular cells of the North i American species from a variety of natural and cultural conditions. Although a considerable range in numbers occurred, a definite pattern could not be distinguished among the species, a l l having an average number of 12-11* annular cells. This character does not appear to be useful in separating the Western North American cytotypes. The paraphyses of the Filicineae were recently discussed by Wagner (1961*). He considers paraphyses in the ferns to be any sterile organ in the sorus intermixed with the sporangia. Martens and Pirard hQ> (19U3) and Martens (19$0a and b) have reported in detail on the mor-phology, ontogeny, and distribution of paraphyses in the P. vulgare complex in the Northern Hemisphere. They found two morphologically distinct types. One type, composed of branching glandular hairs, is typical of the European diploid, P. australe. The other type, first found in P. virginianum, is shown in Figure 16. The paraphyses pictured in these figures show the changes encountered in morphology from young (Figures l6e, h).! to old paraphyses (Figures I6f, g). A typical 'virginianum' paraphysis consists of a stalk and globular head bearing bicellular glandular hairs (Figure l6e). The head frequently consists of only one or two cells, but there may be more; the bicellular hairs range in number from one to ten or more, or are entirely absent. Martens (1950a) discusses most of these variations. Occasional sporangia are also encountered bearing these bicellular hairs. In Western North America typical 'virginianum' paraphyses are very common and numerous in P. hesperium B. Paraphyses (Figure l6h) are also found in the hybrid pictured in Figure 9d, although they are much fewer in number than in P. hesperium B (Figure l6e, f ) . Para-physes have not been observed in P. glycyrrhiza or in the other t r i -ploid hybrid, Figure 9a, b, c. The paraphyses are usually lacking in P. hesperium A, although an occasional specimen will have them (Figure l6g). When present,1 there are only one or two per sorus, and in many sori on the same specimen there are none at a l l . To face page hJi Figure 16. Spores and paraphysis of Polypodium. a. Normal well formed spores of tetraploid P. hesperium A. FAL 210-A. b. Spores of diploid P. glycyrrhi za. FAL 217. c. Aborted spores of triploid hybrid P. hesperium A x glycyrrhiza. FAL 210-C. d. Spores of diploid P. hesperium B. FAL 211-C. e. Young paraphyses of P. hesperium B. FAL 120-l8a» f. Mature paraphyses of P. hesperium B. FAL 211-G. g. Mature paraphyses of tetraploid P. hesperium A. Very rare. FAL 210-A. h. Young paraphyses of triploid P. hesperium A x hesperium B. FAL 122-3. i \ u2 III. THE GAMETOPHYTE The spores of a l l species are bilateral, monolete, and uni-form with the spore wall covered with papillose protuberances. When dehisced in quantity, they are a bright yellow. The major spore difference among the three cytotypes is that of size, and the spore photographs in Figure 16 show the size differences. The spores of the tetraploid, P. hesperium A (Figure 16a), are largest. The smaller diploid spores may be seen in Figure 16b, P. glycyrrhiza, and l6d, P. hesperium B. When measurements of mean stomat-a length and mean spore length are plotted together, as in Figure 17, the effect of polyploidy is clearly shown by the grouping of the diploid and tetra-ploid taxon symbols. The ordinate in Figure 17 shows the x spore lengths taken from a number of individuals from different populations of the three cytotypes. The spores of the tetraploid tend to be larger than the two diploids, although there is some overlapping. Gametophytes of p. hesperium A and B, P. glycyrrhiza, and P. virginianum were grown for cytogenetic studies. Some were also collected at different developmental stages for mutual comparison. It was found that there was no significant difference in morphology, and the development of the gametophytes agrees with that described in detail by Pickett and Thayer (1927) for P. glycyrrhiza. v IV. BIOCHEMISTRY In Europe, from earliest times, the sweet rhizome of P. vul- gare s . l . has been collected for medicinal use. Because of the phar-macological interest in the species, the chemical composition of the Ta face page U3 Figure 17. Correlation of mean spore length to mean stoma length in a number of individuals of the complex from different populations. Notice how the symbols segregate into two groups, each representing different ploidy levels. \ h3 65 E 60 © © R g l y c y r r h i z a • R he s p e r i u m A A R h e s p e r i u m B • O z u _J UJ Cr: o Q_ oO IX 55-50-© A 45 A _] 35 40 . X S T O M A 45 L E N G T H (ju) 50 hk plants is well known. Hegnauer (1962) reviews the chemotaxonomic aspects of the numerous chemical investigations of members of the complex. The sweet, licorice flavor of the rhizome of P. glycyrrhiza was mentioned by Kellogg as early as l&$h and by Eaton (1856). In his description of P. hesperium, Maxon (1900) noted that i t , too, had a "hard licorice-like root stock". He compared i t to the Eastern 'vulgare' (P. virginianum) which was "...not...only quite acrid but more or less unsavory in taste." Fernald (1922) used the same char-acter to separate P. vulgare from P. virginianum. Fischer and Lynn (1933) investigated P. glycyrrhiza as a possible source of licorice and found that no glycyrrhizin (a com-pound found in true licorice Glycyrrhiza glabra that is £0 times sweeter than sugar) was present. They concluded that the licorice-like taste was due to a mixture of a glucdside, which they called polypodin, some unidentified substances, and sucrose. In the present investigation the rather unsophisticated method of chewing the rhizome or frond was used to determine the chemical composition of the plant being tested. In British Columbia, both P. glycyrrhiza and P. hesperium A have sweet, licorice-like rhizomes. P. hesperium B , on the other hand, had an acrid rhizome similar to that of P. virginianum. The triploid hybrid in Figure 9a, b, e tasted of licorice. Unfortunately, only dried material of the triploid in Figure 9d was available for tasting and was sweetish, but less so than that of dried P. hesperium A. 1*5 V. ECOLOGY AND GEOGRAPHICAL DISTRIBUTION Polypodium glycyrrhiza and P. hesperium A and B show differ-ences in habitat, ecology, phenology and geographical distribution which aid in separating the taxa and perhaps suggest their evolu-tionary relationship. It can be presumed that the characteristically different geographical distributions of these cytotypes are based largely on their ecological differences. The only species of the complex found a s an epiphyte is P. glycyrrhiza (Figure l8e, page li7). It commonly grows on the trunks and horizontal branches of Acer macrophyllum throughout Western Oregon, Washington, and Southern British Columbia. In regions of high r a i n f a l l (100 in. per annum) such as the western Olympic Peninsula;, the west coast of Vancouver Island, and the Queen Charlotte Islands where the growth of epiphytic bryophytes is abundant, P. glycyrrhiza is found growing with them on Alnus rubra, Picea sitchensis, Salix sitchensis, S. scouleriana, and Sambucus racemosa. In addition, how-ever, i t occurs almost as frequently on the mossy surface of rocky outcrops, talus slopes, stumps, logs, and soil. It has even been collected growing on sand dunes. Figure 18b shows P. glycyrrhiza growing in shallow soil on the surface of a rock exposure. It grows in shaded as well as exposed sites, usually below 2000 f t . in elevation. In the interior of British Columbia P. hesperium A appears to range from 1500-6000 f t . , usually in the vicinity of water. It grows U 6 on rock surfaces (Figure l8a), in rock crevices (Figure l8d), or on soil (Figure 18c), commonly in shade (Figure l8d), but often doing as well in exposed sites (Figure l8c):. P. hesperium B is typically a plant of mountainous areas, growing in the Northwest between 1$00 and 6000 f t . elevation. It has been collected at sea level at Bella Coola, British Columbia. When collected at lower elevations, i t is usually found growing in montane valley bottoms. This cytotype characteristically grows in rock cre-vices, either shaded or exposed. Its rhizome is typically at the back of a deep crevice with the stipe extending to the surface, so exposing the blade to light. Figures l8f and g show the species growing in deep shade. As mentioned above, neither P. hesperium A nor B has ever been observed growing as an epiphyte. The growth habit of the rhizome is different in the three cytotypes. In P. hesperium A and P. glycyrrhiza the rhizome tends to spread out and grow over the surface of the substrate, while the rhizome of P. hesperium B grows in rock crevices and rarely extends to the adjacent surface. P. hesperium A, on the other hand, will i start growth in a crevice, but the rhizomes are likely to spread out along the edges of the cleft over the surface of the rock. A l l three species seem to thrive equally well under shaded or exposed conditions. However, plants from sites of varying degrees of exposure show morphological differences. Specimens from exposed sites are smaller, with fronds coriaceous in texture rather than membranous or herbaceous, and have more reflexed frond segments than plants from shaded sites. In deep shade the growth form P. hesperium A and B To face page k7 Figure 18. Habitats of Polypodium hesperium A, P. hesperium B, and P. glycyrrhiza. a. P. hesperium A (tetraploid) growing over the surface of rock, North Thompson River, British Columbia. FAL Hq. c. P. hesperium A (tetraploid) i n exposed situation above the North Thompson River, B r i t i s h Columbia. FAL 111. d. P. hesperium A (tetraploid) i n shaded crevice near a. FAL 112. b. P. glycyrrhi^a on rock, Horseshoe Bay, B r i t i s h Columbia, f-g. P. hesperium B (diploid) growing i n rock crevices i n deep shade. McGuire, Br i t i s h Columbia. FAL 211. k7 HQ show a superficial resemblance to one another (Figure l8d and f ) . There are two obvious alternatives to explain these morpholo-gical changes. The changes could be caused either by phenotypic plasticity or by ecotypic variation under genetic control. This pro-blem can only be solved by cloning and transplant experiments. In the present investigation plants brought in from the wild have shown marked morphological change under greenhouse conditions. Figure 19 illustrates fronds of a l l three cytotypes collected in the wild and later from the same plants after they have been in culture. Most of the difference noted in P. glycyrrhiza is in size. Figure 19j shows fronds formed under exposed greenhouse conditions, which are similar in morphology to plants of P. hesperium B in natural, exposed sites. The original fronds from the same rhizome were like those in Figures l8f and g. Figure 19g shows two fronds of P. hesperium from the same rhizome. The plant was collected in deep shade on the south wall of a deep, narrow canyon. The frond on the left was formed in nature the previous spring, the one on the right was formed after the rhizome had been in cultivation for several i months. In a l l three cytotypes the old fronds are shed during or shortly after the annual initiation of new fronds. These cytotypes produce new fronds at different times of the year and, as a conse-quence, meiosis occurs and the spores mature at different times. These observations are summarized in Table 3. k3> TABLE 3. PHENOLOGY: TIME OF ANNUAL PRODUCTION OF NEW FRONDS AND OF SPORES BY P. GLYCYRRHIZA, P. HESPERIUM A, AND P. HESPERIUM B Species Month J F M A M J J A S 0 N D P. glycyrrhiza s s s s s p. hesperium A s s s s s P. hesperium B s s s s s s s months new fronds produced s mature spores observed It w i l l be noted that young fronds are produced by P. glycyr- rhiza from June onward. Although i t is not apparent from Table 3, most specimens seem to produce new fronds in late summer with meiosis occurring in the f a l l . Mature spores can be found throughout the f a l l and winter months. P. hesperium A produces new fronds throughout the summer with the first being initiated in May. Plants can be found in a l l stages of development during most of the summer. New fronds are usually produced by P. hesperium B in April or May with the spores maturing in June, July, and August. Hybrids between P. hesperium A and the other two cytotypes also show seasonal differences. The t r i -ploid in Figure 9d was undergoing meiosis in the spring when i t was collected. The other triploid hybrid in the Fraser Canyon, similar to the one in Figure 9a, b, c, was in i t i a l l y collected because a plant was noticed putting up fronds unseasonably early for P. glycyrrhiza To face page 5Q> Figure 19• Field collections and voucher specimens of Polypodium to show comparison of fronds collected in the wild and fronds produced in greenhouse culture. a-f, h, i . P. glycyrrhiza. g. P. hesperium B. j . P. hesperium A. a. Field collection and d. voucher specimens of FAL 23. b. Field collection and c. voucher of FAL 25_. e. Field collection and h. voucher of FAL 28. i . Field collection and f. voucher of FAL 33. g. Tetraploid P.. hesperium A. Frond on left formed in the wild, on right formed under cultivation. FAL 130-1;. j . Diploid P. hesperium B. Small depauperate fronds formed under cultivation. Original fronds similar to Figures I8f and g. FAL 218. 5u 5i in that area. In any given region the majority of the plants of a particular cytotype produce new fronds at about the same time. In most localities where the cytotypes are found together, there is usually a considerable interval between the initiation of fronds in each case. In the Fraser Canyon, for example, P. hesperium B produces new fronds several months before P. glycyrrhiza initiates them. Variation in the time of initiation of new fronds can be related to different climatic conditions in different parts of the cytotypes' ranges or to annual fluctuations in local climatic condi-tions. Plants of P. glycyrrhiza from the north coast of British Columbia tend to put up their fronds earlier than those on the south coast. This could be accounted for by differences in climatic conditions in the north. An example of an annual fluctuation of local climate affecting initiation of new fronds was noted in a population of P. hesperium B from Mount Seymour, British Columbia. Cytological material for jineiosis was collected late in May 1962 . Next spring material in the same condition could not be collected until almost a month later because the onset of the growing season was delayed by the late melting of the unusually heavy snowfall of the preceding winter. These ferns are perennial from a creeping rhizome and often form rather extensive mats of intertwining branched rhizomes. The older parts of these die off so causing the plants to divide—a form of vegetative reproduction. It is often impossible to determine i f the individuals in a mat are clonal or were derived from sexual 52 reproduction. During the present investigation gametophytes were rarely found in nature and then only in disturbed sites. The geographical distributions of P. glycyrrhiza, P. hesperium A, P.. hesperium B and the triploid hybrids are shown in Map 2. The coastal P. glycyrrhiza extends from Kamtchatka along the Aleutian Islands to Alaska, (see Map 1), through British Columbia, Washington, Oregon and as far south as Monterey County in Central California (Munz 19$9). In Oregon, Washington, and Southern British Columbia the species extends as far inland as the Cascade Mountains, deepest penetration occuring in major river valleys, as far as Usk on the Skeena River, Boston Bar on the Fraser River, and Hood River on the Columbia River. The most easterly population is at Satus Pass in Washington, although there are several collections of interest from elsewhere in the interior (see page 100). In Map 2 the cytotypes of P. hesperium s . l . are more or less geographically distinct with P. hesperium A present mainly in the interior and P. hesperium B limited to the coast and Cascade Mountains. P. hesperium A is found in the eastern part of the southern l coast mountains in British Columbia and is separated from the remainder of its distributional area in the southeastern part of the province by the Okanagan-Kamloops-Lillooet-Thompson River dry belt. It occurs south through Eastern Washington, although i t is absent from the Columbia Plateau, Northern Idaho, and Northwestern Montana, as far as the Wallowa Mountains and the Mount Hood region of Oregon. P. hesperium B is found in the coast mountains of British Columbia from Hazelton south through the Cascade Mountains to the To face page 53 Map 2. Distribution of Polypodium (except P. scouleri) in North-western North America (based on selected localities of specimens examined). Sise difference of symbols not signifi-cant. 53 Sk Mount Hood region of Oregon, where i t also occurs at Saddle Mountain in the Oregon Coast range. In Washington the cytotype is also found in the Olympic Mountains, Wenatchee Mountains and on Mount Constitu-tion, Orcas Island, in the San Juan Islands. It is apparently absent from the Oregon Cascades. The status of the two California specimens of P. hesperium B shown in Map 2 is in doubt, and further work in this area is necessary. They seem to be closest to the B, cytotype. As seen on Map 1, P. hesperium, in its broadest sense, is widely distributed in the Western United States, being found in the mountains of Colorado, Southeastern Wyoming, Utah, Arizona, New Mexico, Southern California, and Baja California. Although no information based on living material and mass collections was available for inves-tigation from this part of the taxon's range, some interesting observa-tions were made on a limited number of herbarium specimens. Until adequate material can be critically examined, the following remarks must be regarded as tentative. From what is known of P. hesperium s . l . in the Pacific North-;• west, herbarium material from the remainder of its range can be placed i in two groups, one centering about P. hesperium B, the other about P. hesperium A. It is often difficult to place some specimens because of inadequate material. Specimens from Southwestern Colorado, Eastern Utah, and Arizona, in Oak Creek Canyon, along the Mologon Rim, and in the isolated high mountains in the south were much like P. hesperium A. Some specimens resembled Clute's (1910) picture and description of P. prolongilobum. Specimens from Southern California and Baja California could not be 55 positively identified because of the limited amount and condition of the material available. The collections centering about P. hesperium B are from the Laramie Hills of Southeastern Wyoming and the mountains of North-eastern Colorado with a few isolated specimens from the high mountains of Southern Arizona. In most aspects of their morphology the speci-mens seem to be similar to P. hesperium B. In Figure 20 the enclosed areas represent the extreme boun-daries of mean stomal frequency versus stomatal length of cytotypes of different ploidy level shown in Figure 1U. When these measurements were computed and plbtted in Figure 20, i t was discovered that they f e l l in the area represented by tetraploid individuals in the Pacific Northwest. This strongly suggests that plants from Wyoming, Colorado and Arizona might be polyploid instead of diploid as P. hesperium B is in the Northwest. The two triploids (Figure 9) were found in areas where a l l the cytotypes occur sympatrically or at least where a l l may poten-ti a l l y be found together. Throughout most of their ranges P. glycyr-rhiza, P. hesperium A, and P. hesperium B are spatially and geogra-phically isolated from each other. P. hesperium B, a plant of mountainous areas usually found at high elevations (1500-6000 f t . ) , descends into the bottoms of mountain river valleys and antecedent streams, such as the Fraser River and the Columbia River which pass through the major mountain ranges. The lowland species P. glycyrrhiza, usually found below 2000 f t . , grows some distance up these same valleys. It is here that P. glycyrrhiza and P. hesperium B are found growing To face page 56 Figure 20. Comparison of mean stomatal length versus mean stomatal frequency for Polypodium from NE Colorado and Arizona with plants of known ploidy level. The closed points represent the area covered by points in Figure LU. On these enclosed areas have been superimposed the measurements of stomatal length and frequency from herbarium material of unknown ploidy level from Arizona and Colorado. 56 A • i • i i i 30 35 40 45 . 50 X S T O M A ; LENGTH (JJ) 5? near each other. Certain areas, such as the low Cheakamus River-Pemberton-Lillooet divide, and the antecedent Fraser and Columbia Rivers, provide convenient routes for interior and coastal floras to intermix. As one proceeds upstream, there is a more or less gradual transition in these deep river valleys from a humid coastal climate to a dry interior climate. Changes in the vegetation generally reflect this climatic transition. At Hope on the Fraser River the climate and vegetation are coastal. Midway through the Fraser Canyon near Boston Bar and Hells-gate, Figure 21, the climate is intermediate and elements of both the coastal and interior floras are present. At Lytton the climate and flora are typical of the dry interior. Detling (1958) reports a similar picture of floristic and climatic change in the Columbia River Gorge. Al l three cytotypes are found in the Cheakamus-Lillooet region (Map 2, inset 1). P. glycyrrhiz a and P. hesperium B are both in the Fraser Canyon and, because of the presence of the triploid hybrids, i t is probable that P. hesperium A is also there (Map 2, inset 2), although i t was not found. Similarly, a search of the Columbia River Gorge yielded only P. hesperium B and P. glycyrrhiza; no evidence of \ P. hesperium A was found. \ To face page $8 Figure 21. Graphic presentation of climatic data from selected weather stations from the ranges of Polypodium in British Columbia (after Krajina 1959) • / J F M A M J J A S O N D Pachena Point max. temp. x monthly temp. min-temp. 'y-A i . 10-?. 20-10 1 A . -TP 111-36' Nanaimo T.f? 37-9V Hollyburn Mount Seymour 58 J F M A M J T A S O N D 1 ° F Revelstoke Salmon Arm T.R 19.68" Hells Gate -•80 ••60 -•40 --20 -•20 40 •100 T80 60 40 - •20 TP 46-33 CHAPTER IV TAXONOMY I. DISCUSSION OF NOMENCLATURE It is how possible to examine the taxonomy of the Polypodium vulgare complex in light of the results of this study. A detailed assessment of the nomenclature would be premature at the present, how-ever, since several taxa are s t i l l incompletely known through parts of their ranges and a number of type specimens s t i l l must be located and examined. Taxonomic treatments of aggregations, such as the Polypodium vulgare complex, made up of a number of similar but genetically dis-tinct taxa of different ploidy levels must of necessity be very sub-jective. The f i r s t question is the rank to which the various taxa should be assigned; should they be considered species, subspecies, or placed in some other category? Admittedly a subspecific ranking has the obvious advantage of expressing the relationships of the taxa b y i tying them to a common epithet. On the other hand, the choice of a subspecific category may have the disadvantage of concealing the existence of genetic discontinuities. Another alternative would be the use of the 'species aggregate' as a formal category as has been advocated by Manton (l°f?8) and Lovis (196U). This would probably be the simplest way to handle the problem and has much to recommend i t . However, until the time such a rank appears in the taxonomic hierarchy existing categories must be used. 60 Since Linnaeus first described Polypodium vulgare, the inves-tigations of Manton (1950) and Shivas (1961a and b) have shown that there are three ploidy levels in the European Polypodium species. The question i s : should the epithet vulgare be applied to the diploid, tetraploid, or hexaploid cytotype? This can be determined in several ways. The type specimen may be compared to other material of known ploidy level or topotype material may be examined cytologically. Since the type specimen of P. vulgare is not extant, Shivas has essentially used the latter approach in limiting P. vulgare to the tetraploid cytotype. Because the tetraploid is the prevalent form in Sweden, and is probably the one Linnaeus had in mind when he described the taxon (it occurs abundantly in his garden), Shivas proposes to attach the name P. vulgare L. sensu stricto to this cytotype.- The diploid and hexaploid cytotypes are given other specific names. The example of Manton and Shivas in considering the cytotypes of the P. vulgare complex as species has been followed in the present investigation. As in the European ones the Northwestern North Americas cytotypes display the fundamental characteristics of distinct natural i species. They are reproductively isolated, possess distinctive geo-graphical and ecological distributions, and can be distinguished morphologically, although admittedly not always with ease. Before the taxonomy of the Western North American cytotypes can be considered, i t must be established whether or not P. vulgare L. (Shivas) occurs in our area. If one accepts Shivas' decision as to the application of P. vulgare the two Western North American diploids P. hesperium B and Sk P. glycyrrhiza must be eliminated; fi r s t because of the difference in ploidy level, and second because of morphological and ecological differences. P. hesperium B is diploid, has short blunt frond seg-ments, paraphyses, marginal sori, and an acrid rhizome, a l l characters not possessed by P. vulgare s.s. P. glycyrrhiza is similar to P. vul-gare s.s. but differs from the latter in its acute to attenuate frond segments with their serrate margins, its distinctive rhizome scales, having its meiotic season in the f a l l , and in its being diploid. Although P. hesperium A and P. vulgare s.s. are both tetraploid, i t is clear that they are not the same taxon. P. hesperium A has short rounded frond segments with oval sori as opposed to the longer, more acute segments of P. vulgare s.s. with circular sori. On the basis of the foregoing the present author concludes that P. vulgare s.s. is absent from this area. This leaves the problem of naming the three Northwestern North American cytotypes. There is no doubt that the correct name of the Coastal diploid species is P. gly- cyrrhiza D. C. Eaton since this epithet has priority over the others. ( Two varieties of P. glycyrrhiza might be recognized on the basis of frond segment shape, one with attenuate frond segments, the other with acute to subobtuse segments. However, formal recognition of the varieties will have to be delayed until the precise relation-ship of the two forms has been intensively studied. The differences between the two cytotypes included in P. hes- perium are summarized in Table U . It is felt that the differences are such to warrant their recognition as two distinct species. The oldest available name which must be used for one or the other of the 62 TABLE i i . CHARACTERS DISTINGUISHING POLYPODIUM GLYCYRRHIZA, P. HESPERIUM A, P. HESPERIUM B AND P. VIRGINIANUM, Species ' Frond shape Segment shape Index (mean) Sorus shape Sorus location Paraphyses Rhizome taste Scale strip New fronds Geographical distribution P. glycyrrhiza P. hesperium A P. hesperium B P. virginianum oblong oblong to ovate 1.65 acuminate to acute circular median absent licorice absent autumn coastal 2.72 acute to obtuse oval median very rare licorice absent summer interior oblong 2.99 obtuse circular marginal common acrid + -spring western mountains oblong-triangular acute circular marginal common 'acrid present summer eastern N. Amer. 63 two cytotypes is P. hesperium Maxon. It thus becomes necessary to establish to which cytotype the epithet hesperium properly belongs, the diploid or the tetraploid. The type specimen of P. hesperium was compared morphologically with a range of both diploid and tetraploid specimens from throughout the Northwest and chromosome determinations were made on topotype material. This comparison makes i t clear that the tetraploid (P. hes- perium A) agrees very closely with the type specimen of P. hesperium Maxon. Among other features in common, both have oval median sori and a sweet licorice-like rhizome, as mentioned by Maxon (1900) in his original description. The diploid, P. hesperium B, on the other hand, has circular marginal sori and an acrid rhizome. The geographical distributions give further evidence that the type of P. hesperium is tetraploid since, as far as is known, the diploid cytotype does not occur in the same area as the tetraploid cytotype. A collecting trip to the type locality of P. hesperium in Coyote Canyon, Lake Chelan, Washington, yielded several isolated colonies of Polypodium. They were essentially similar in morphology and a l l plants on which chromosome i counts were made proved to be tetraploid. Figure 3b shows a spore . mother c e l l from one of these plants at diakinesis with 7k bivalents. On the basis of the evidence there is no doubt that the tetraploid should bear the name P. hesperium Maxon. This leaves the diploid cytotype, P. hesperium B, without a name. There are, however, several possible names in the literature which must be considered. One epithet that might apply is P. amorphum Suksdorf, the type specimen of which was collected at the base of a shady c l i f f in 61* Dog Creek Canyon, Skamania County, Washington, in the region of the Columbia River Gorge. On the basis of a photograph in Frye (193U)> examination of the type specimen and Suksdorf's description, this species has some features in common with P. hesperium B, viz., a thin slender pruinose rhizome with typical scales and circular marginal sori with paraphyses. In the morphology of the frond P. amorphum is very variable, with mostly semicircular frond segments and bifurcate frond tips. P. hesperium B does not agree with i t in these features. Several attempts to re-collect P. amorphum at the type loca-l i t y have failed (Slater 196U). In making up specimens for d i s t r i -bution Suksdorf returned to the type locality several times; in doing so he probably collected most of the existing plants, any survivors having subsequently died out. The writer accepts the view that P. amorphum is based on a monstrosity and even i f P. hesperium B is conspecific with i t P. amorphum is regarded as illegitimate under Article 71 of the 1961 edition of the International Rules of Botanical Nomenclature. Recommendation 60a of the International Rules requires that i several varietal names must be considered. P. vulgare var.. colum-bianum Gilbert is a synonym of P. hesperium s.s., and in any case the new combination P. columbianum for the diploid would be illegitimate since i t would be a later homonym of P. columbianum Baker. Clute (1910) described Polypodium vulgare var. perpusilium from Mt. Lemmon, Arizona, and gave the following brief description: "Fronds one to four inches long, one half to three quarters of an inch wide diminishing below, pinnules oblong, obtuse, about eight pairs; sori 65 medium size, numerous, near margin than midrib." This description more or less f i t s the diploid cytotype, especially the sori being nearer the margin than the midrib. Unfortunately, Clute's type specimen has not been located nor has topotype material been available for examination. The description, with its emphasis on unreliable quantitative char-acters, is not precise enough to state with certainty that i t is the same as the diploid cytotype of P. hesperium s . l , viz. P. hesperium B of this study. Specimens from elsewhere in Arizona do f i t Clute's description and are similar to P. hesperium B. Certain of these herbarium specimens have been examined to determine their stoma length and frequency, with the results shown in Figure 20. In this figure the frequency and lengths of stomata of plants of unknown ploidy level are compared to the areas covered by the diploid and tetraploid cytotypes in Figure lU. The specimens from Arizona and Colorado examined f a l l much closer to the tetraploids than to the diploids and i t l i s almost certain in fact that they are tetraploid. It is recognized that, genetic factors may affect cell size in addition to the ploidy level. A good example of i • • this is seen in the large spores of the diploid P. australe which are greater than 7U U long compared to those of P. vulgare s.s. which are less than 70 u long (Shivas l°6la and b). Nevertheless the stoma size and frequency relationship seems important evidence favoring tetraploidy. In view of these uncertainties, the writer does not feel justi-fied in taking up Clute's varietal epithet of perpusillum for the diploid cytotype and so the new name Polypodium montense is proposed. 66 The triploid hybrids found in this study are discussed in some detail under P. hesperium s.s., as i t seems fairly certain that this is one of the putative parents. As only a few plants of each has been found there is no evidence as to the uniformity of these crosses and so no adequate grounds for naming them. They are referred to simply by a hybrid formula which also has the advantage of indica-ting their probable parentage. \ 67 II. A PRAGMATIC KEY TO THE SPECIES OF POLYPODIUM IN NORTHWESTERN NORTH AMERICA 1. Paraphyses present 2. Sori circular, marginal; rhizome scale often with a median stripe, margin coarsely toothed, rhizome acrid. 3 . Frond segments narrowly ovate, tips mostly acute P. virginianum 3. Frond segments oblong to ob-ovate, tips mostly obtuse .... 2. P. montense 2. Sori oval, median; rhizome scale uniformly colored, margin not coarsely toothed; rhi-zome sweet. ii. Spores normal (not aborted) . . 3 * P» hesperium U. Spores aborted 3." P. hesperium x montense Paraphyses absent 5. Sori circular 6. Sori median 7. Frond segments usually more than 30 mm long, ratio of length to width usually more than 3 .3 (mean $ . 6 ) ; tips acute to attenuate 1. P. glycyrrhiza 7. Frond segments usually less than 30 mm long, ratio of length to width usually less than 3 . 8 (mean 2.ii); tips obtuse to acute ^ This species and P. scouleri are included for completeness but are not dealt with directly in this study. 68 8. Rhizome scale often with a median stripe, margin coarsely toothed; rhizome acrid, pruinose 2. 8. Rhizome scale uniformly colored, margin not coarsely toothed; rhizome sweet, not pruinose 3« Sori oval 9. Veins anastomosing 9. Veins free 10. Spores normal (not aborted) 11. Frond segments usually more than 30 mm long, ratio of length to width usually more than 3.3 (mean $.6), tips acute to attenuate 1. 11. Frond segments usually less than 30 mm long, ratio of length to width usually less than 3.8 (mean 2.J4), tips obtuse to acute 3. 10. Spores aborted •12. Segment margins entire to \ crenate, tips obtuse; ratio of segment length to width less than 2.0 3* P. montense 6. Sori marginal 2. P, P. hesperium P. montense P. scouleri P. glycyrrhiza P. hesperium P. hesperium x montense 12. Segment margins serrate, tips acute; ratio of seg-ment length to width more than 2.0 3. P." hesperium x glycyrrhiza 63 III. SPECIES DESCRIPTIONS 1. Polypodium glycyrrhiza D. C. Eaton, Am. Journ. Sci. II. 22:138 moT. -Polypodium vulgare L. var. Occidentale Hook., Fl. Bor. Am. 2:258 (181*0). Polypodium faleaturn Kellogg, proc. Cal. Acad. 1:20 (185U). Non-L. f i l . Suppl. 1*1*6 (1781). Polypodium vulgare var. commune Milde, F i l . Eu. Atla. 18 ( 1 B 6 7 T : Polypodium vulgare var. faleaturn (Kellogg) Christ, Beitr. Krypt. Schweiz. 2:51 (1900). Polypodium occidentale (Hook.) Maxon, Fern Bull. 12:102 . (1901*7: ~ Polypodium vulgare subsp. occidentale (Hook.), Hulten, F l . Alaska, J. Yukon, I. 1*1* (191*1). Polypodium aleuticum A. Bobr. Botanicheskii Zhurnal 1*9:5U2 Rhizome creeping, sweet, with licorice-like taste, 3-6 mm in diameter and paleaceous; rhizome scales narrowly ovate to ovate, up to 9-10 mm long, tan to custaneus, uniformly colored, more or less entire, sometimes with prolonged capillary tip, cells small, about 1*0 in num-i ber across scale just above point of attachment; frond averaging 260 mm long, max. ca. 600 mm long; stipe stout, *11) (36-91*-11*3) (31*0 mm long; blade coriaceous on exposed rock, herbaceous or membraneous in shade; oblong to ovate, sometimes deltoid, 20) (69-173-280) (1*70 mm long, ll*) Quantitative measurements are presented in this manner to best display their variability. Using stipe length as an example the following should suffice as an explanation. The shortest stipe measured was 11 mm long) (80$ of a l l stipes measured were from 36 to 11*3 ram long, the average stipe length being 9h mm long) (the longest stipe was 31*0 mm long. For further information see page 12, Table 2, and Figure $. 7a (35-60-92) (156 mm widej blade segments narrowly oblong-attenuate, more or less falcate, tips acute to acuminate, 7) (18-33-53) (85 mro long, 3) (5-7-8) (11 mm wide, ratio of length to width 1.8) (3.3-5.6-7.U) (12.Oj hydathodes large, oval, many celled, veins free, forking 2-k timesj sori usually circular, sometimes slightly oval, located midway between costa and margin of segments; paraphyses absent, chromosome number n => 37> 2n » 7k; type: Port Orford, Kuntz s.n. (YU) 1 . New fronds produced from August to February; found growing at lower elevations on rocks and trees along the Aleutian Islands to Alaska south through coastal British Columbia, Washington, Oregon and California to Monterrey County; east to the Cascade Mountains in the southern part of the range and extending up inlets and river valleys in Coast Mountains of British Columbia. Specimens cited ALASKA: M. W. Gorman 172 (UC)*; Old Harbour, Kodiak Is. Eyerdam 6kk (UC); Raspberry Is. Port Vita, Eyerdam 3213 (UC); Yakutat Bay i Funston 13 (UC); Larson's Bay, Kodiak Island, D. E. Grether h$9$ (UC); Sitka Heller LU927 (UC); Dutch Harbor, Unalaska, H. L. Mason 6037 (UC); Windfall Harbor, Admiralty Island, Stephens 155 (UC); Red Bluff Bay, Baranoff Island, Stephens 159 (UC); Tongas Village, Mr. and Mrs. E. P. Walker 888 (UC); Sitka and vicinity, W. G. Wright 1611 (UC). Herbarium designations are those of Lanjouw and Stafleu (I96U). 71 BRITISH COLUMBIA: Texada Island, Allen s.n. (UBC); Islands South, Victoria, Anderson s.n. (UBC); Deer Lake Ashlee s.n. (UBC); Ganges, Salt Spring Island, Ashlee s.n. (UBC); Salt Spring Island, Mountain t r a i l N. W. Isabella Pt., Ashlee s.n. (UBC); Pender Island, Ashlee s.n. (UBC); Capilano Canyon, Baggs s.n. (UBC); N. of Mile 126, Hope-Princeton Hgwy., Beamish, Vrugt-man, and Stone 7609 (UBC); Islands South, Wellington, W. Bent- ham s.n. (UBC); Capilano River, Buckland s.n. (UBC); Burrard Inlet, Bucklana s.n. (UBC); Bodega H i l l , Galiano Island, Calder and MacKay 28859 (UBC); Junction of Upper Campbell Lake and Buttle Lake, Calder and MacKay 30586 (UBC); Lulu Island, Fraser River, Davidson s.n. (UBC); Caulfields, Eastham s.n. (UBC); Lasqueti Island, Eastham s.n. (UBC); Alberni, Vancouver Island, Eastham s.n. (UBC); H i l l Meadows, Evans s.n. (UBC); Ganges, Salt Spring Island, Harrington G671 (UBC); Thetis Island, Harrington G672 (UBC); Tynehead, Surrey, H i l l s.n. (UBC); Fraser River at Hope, Hitchcock and, Martin 7362 (UC); Skeena R. west of Terrace, King s.n. (UBC); Cultus Lake, Krajina s.n. (UBC); Tofino, Krajina s.n. (UBC); Vancouver near UBC, Krajina 218 (UBC); Cheam Lake, Krajina 592 (UBC); Agassiz-Harrison Mills, Krajina 7U8 (UBC); UBC campus, Krajina 7^ 6 (UBC) Saanich, Victoria, Vancouver Island, Krajina and Spilsbury s.n. (UBC); Mt. Douglas Park near Victoria, Krajina and Spilsbury 3816 (UBC); Anderson H i l l , Krajina and Spilsbury 3867 (UBC); Near 1st Nanaimo Lake, Krajina, Spilsbury and Szczawinski hh9h (UBC); Othello, Coquihalla River, Lang s.n. (UBC); Murrin Lake, 72 Howe Sound, Lang 20 (UBC); T l e l l , Queen Charlotte Islands, Lang 23 (UBC)j 2 miles S of Masset, Queen Charlotte Islands, Lang 2h (UBC); Rd. to Tow H i l l , Queen Charlotte Islands, Lang 2$ (UBC); S Port Clements, Queen Charlotte Islands, Lang 26 (UBC); Rd. to Juskatla, Queen Charlotte Islands, Lang 27 (UBC); St. Mary's Spring, Queen Charlotte Islands, Lang 28 (UBC); Tow H i l l , Queen Charlotte Islands, Lang 30 (UBC); Tow H i l l Sound, Queen Charlotte Islands, Lang 31 (UBC); 19 miles E Terrace Bridge, Skeena River, Lang h9 (UBC); 8 miles E Terrace Bridge, Skeena River, Lang $3 (UBC); within 8 miles of Terrace Bridge, Skeena River, Lang $1* (UBC); Long Beach Provincial Park, Vancouver Island, Lang $6, $7, $8 (UBC); Rainbow Lake, Prince Rupert, Lang 100 (UBC); lh miles W of Seagram Creek, Skeena River, Lang 101 (UBC); First c l i f f from Cheekeye Bridge on Squamish River Road, Lang 103, 10$, 106 (UBC); B. C. Hydro Powerhouse, Squamish River, Lang 108 (UBC); 2 miles S of B. C. Hydro Powerhouse, Squamish River, Lang 109 (UBC); 6 miles S of B." C. Hydro Power-house, Squamish River, Lang 110 (UBC); I; miles N of Cheekeye Bridge, Squamish River, Lang 11$ (UBC); 1 mile N of Cheekeye Bridge near junction of Squamish and Cheakamus Rivers, Lang 116 (UBC); 11 miles N of Cheekeye, Cheakamus River, Lang 117-B, D, F (UBC); NW of Mount Douglas, Victoria, Lang 119 (UBC); Alexandra Bridge, Fraser River, Lang 121 (UBC); W side of Cheakamus River Valley, N of Garibaldi Station and B. C. Hydro dam, Lang 206 (UBC); W Daisy Lake, N of Garibaldi Station, Lang 213 (UBC); above Fox Reach, Pitt River, Lang 216, 217 (UBC); 73 2 miles S Furry Creek, Howe Sound, Lang 223 (UBC); Calvert Is-land, McCabe 35 (UC); Yale, McCabe 7kk (UC); Alexandra Bridge, Fraser River, McCabe 76? (UC); Stuart Island, McCabe I6lt7 (UC); Kelsey Bay, McCabe I676 (UC);.Christy Pass, McCabe 169IA, 1695 (UC); Ann Island, McCabe 1802 (UC); Boston Bar, McCabe 1851 (UC); 5-10 miles E of Rosedale, McCabe 1852 (UC); Alexandra Bridge, Fraser Canyon, McCabe 1853 (UC); between Alexandra Bridge and Yale, McCabe 2512 (UC); Whytecliffe, McCabe 2555 (UC); Koeye River, Fitzhugh Sound, McCabe 3120 (UC); rocky i s -let one-half mile from Lama Pass and Fisher Channel, McCabe 3130 (UC); McKenny Islands, NW of Aristazabal Island, McCabe 3^ 31 (UC); Vedder Mountain, McCabe 3797a, b (UC); Calvert Island, McCabe U077 (UC); Table Island, McCabe 1^ 116 (UC); Zero Rock, McCabe kh09 (UC); 12 miles N of Victoria, McCabe 5606 (UC); Sooke, Vancouver Island, McCabe 5687 (UC); 2 miles N of Yale, McCabe 5790 (UC,WTU); Reginald Island, Mulbank Sound, McCabe 6800 (UC); Fraser River at Hunber's Creek, McCabe 7059 (UC); Hurst Island, McCabe 7115 (UC); Hecate Island, McCabe 7130 (UC); Neekis River, Don Peninsula, McCabe 7UJ9 (UC); Calvert Island, McCabe 7179 (UC); Larson Harbor, Banks Island, McCabe 7315 (UC); Murder Cove, Macauley'Island (sic), McCabe 7316 (UC); Porcher Island, Freeman Pass, McCabe 7357 (UC); Neekis River, Don Peninsula, McCabe 7390 (UC); Saanichton, Vancouver Island, Mounce s.n. (UBC); Victoria, Pineo s.n. (UBC); New Westminster, Pirie s.n. (UBC); McKay Creek, Sumas Mountain, K. Racey 36OI (UC); Dollarton, K. Racey, s.n. (UBC); Abbots-Ik ford, Ridewood 26 (UBC); North Vancouver, Roller s.n. (UBC); District of Renfrew, Rosendahl & Brand 98 (UC); Third Beach, Vancouver, Sanderson 67 (UBC); Powell River, Szy s.n. (UBC); Markale, Kyuquot, Vancouver Island, Taylor 21$ (UBC); Salvus, Skeena River, Taylor 606 (UBC); Saanichton, Vancouver Island, Taylor 1007 (UBC); Prince Rupert, Taylor and Lewis 726 (UBC); Markale, Kyuquot, Vancouver Island, Taylor and Szczawinski 2U0, 2U7 (UBC); Malcolm Island, Zarelli s.n. (UBC). CALIFORNIA: Alameda Co: Wildcat Canon, Berkeley, M. A. Howe s.n. (UC); Humboldt Co.: Bull Creek region, S Fork Eel River, Constance ' 7U8 (JEPS); Wilder Ranch near Carlotta, Hall 23k (UC); Carlotta, Hawver s.n. (UC); Willow Creek near Weaverville, M. E. Jones 2917$ (UC); Hupa Valley, Manning s.n. (UC); Lawrence Creek, H. E. Parks 3233 (UC); Samoa, opp. Eureka, J. P. Tracy 986 (UC); Bald Hills road near Redwood Creek, Tracy 3113 (UC); Hydesville, Tracy U023 (UC); N fork of Eel River, Tracy kk?0 (UC); McKee's Mill, Eel River from S fork to Scotia, Tracy 6602 (UC); Knee-land Prairie, Tracy 66U4 (UC); Trinity River Valley near the S fork, Tracy 7326 (UC, JEPS); Blue Slide on Van Duzen River, Tracy 929li (UC); Miranda, S fork of Eel River, Tracy lkllS (UC); Lake Co.: Clear Lake, McLean s.n. (UC); Marin Co.: Big Carson, Howell s.n. (UC); Cataract Gulch, Mt. Tamalpais, Howell 230i;6 (UC); Redwood Canon, J. B. Davy 12$k (UC); Mt. Tamalpais E. L. Greene s.n. (UC); Napa Co.: Yountville, F. L. Clarke s.n. (UC); Mendocino Co.: Sherwood Valley, Davy and Blasdale $170 (UC); Mendocino, M. E. Jones 29173 (UC); S fork of Eel 75 at "Bridges Creek," Tracy 137UO (UC); Sonoma Co.: Austin Creek, J. B. Davy 1662 (UC)j Siskiyou Co.: Klamath River U miles E of Seiad, Constance and Rollins 2908 (UC); Walker Creek, a tributary of the Klamath River, E. Lee 103U (UC). OREGON: Benton Co.: Sulfur Spring Rd. N of Corvallis, Lang s.n. (UBC),- Coos Co.: Florence, W. S. Cooper 33-169 (UC); Coos Bay, Lemmon s.n. (UC); Curry Co.: Port Orford, A. V. Kautz s.n. (YU); Wyeth between Bonneville and Hood River, Lang 199 (UBC); Jackson Co.: Prospect Alexander s.n. (UC); Lane Co.: Spencer Butte, South of Eugene, Constance s.n. (UC); Lincoln Co.: . S of Newport, Lawrence lU8i| (UC); Multhomah Co.: Oneonta Gorge, Cronqulst 6098 (UBC); Clarnie Station, M. Eaton 1 (UC); Lewis and Clark State Park, Troutdale, Lang 188, 189 (UBC); Crown Point, Columbia River Gorge, Lang 190, 192 (UBC); Oneonta Gorge, Lang 198 (UBC). WASHINGTON: Yakima Region, Brandegee 330 (UC); Clallam Co.: Lake Crescent, H. E. and S. T. Parks 0325 (UC); Clark Co.: at intersection of Skamania, Clark Co. line and highway U. S. t 830, Lang 203 (UBC); Island Co., Camano Is., N. S. Gardner s.n. (UC); Jefferson Co., Brinnon, R. K. Beattie 3001 (UC); along bank of Hoh River, B. I. Brown & W. C. Muenscher 117 (UC); Constance Ridge, G. N. Jones 5779 (WTU); Port Discovery, St. John 31*33 (UC); Klickitat Co.: Columbia Highway, Eyerdam 153U (WTU); King Co.: Seattle, Pelton s.n. (UC); Snoqualime, Rodin 6786 (UC); Lewis Co.: 2.75 miles S of Elbe, Slater s.n. (UBC); Mason Co.: Hoodsport, Hood Canal, Smith I83 (UC); San 1 76 Juan Co.: Brown's Island, near Friday Harbor, Blanchard 27 (UC); Stuart Island, Lawrence 203 (UC); Brown's Island, J. J. Martin 270 (UC)j Skamania Co.: Rocky Canyon of Dog Creek, near Cook, G. N. Jones 63LU (UC, WTU)j 1.5 miles E of Steveson, Lang 200 (UBC)j Beacon Rock State Park, Lang 201 (UBC); Thurs-ton Co.: W of Bay, Olympia, Townsend s.n. (UC); Yakima Co.: Satus Pas3, Hitchcock and Marsh 335U (UC, WTU). 2. Polypodium montense F. A. Lang, sp. nov. (ined.) Polypodium amorphum Suksdorf, Werdenda 1:16 (1927). (Type: Dog Creek Canyon, near Cooks, Skamania Co., Washington, Suksdorf 11667 (WS) l) Polypodium vulgare L. var. perpusilium Clute, Fern Bull. 18:98 (1910). Rhizome creeping, acrid, 3-5 mm in diameter, often pruinose, paleaceous; rhizome scales dark brown to castaneus, often with a cen-tral strip of darkly colored cells, narrowly ovate to ovate, often con-stricted above point of attachment, to 5 mm long, usually with a capillary tip, margin coarsely toothed, cells large, ca". 25 in number across scale just above point of attachment; frond averaging 130 mm 1 long; max. ca. 300 mm long; stipe slender, 8) (28-58-100) (llj2 mm long; blades coriaceous to membraneous, oblong l8Xii6-8l-122) (190 mm long, 11) (17-2U-30) (U5 mm wide; segments oblong to obovate, tips obtuse to rarely acute, margin entire to crenulate, 5) (9-13-17) (25 mm long, 3) (lj-6-7) (12 mm wide, ratio of length to width 1.2) (1.8-2.3-3.0) (3*6; hydathodes small, round, few-celled; veins free, forking 1-2 times; sorus circular, nearer the margin than the costa; paraphyses 77 many) chromosome number n =- 37, 2n = 7k', (Type: McGuire, Cheakamus River, British Columbia, Lang 211 (UBC). New fronds produced from late April to June; found growing in rock crevices in mountains from central Coast Range in British Columbia south through the Cascade Mountains in Washington to Oregon and the Sierra Nevada Range in California, in the Olympic Mountains and Wenatchee Mountains of Washington and the Northern Coast Range of Oregon, usually at high elevations but descending to bottoms of river valleys. Also found in mountains of Northeast Colorado and the Laramie Hills of Southeast Wyoming and some of the high mountains of Arizona. Specimens cited ARIZONA: Pima, Spud Rock, Blumer 3U39 (UC). BRITISH COLUMBIA: Mt. Prevost, Vancouver Island, Anderson lt$a (V); Cheam, Anderson and Fletcher U5 (V); Pemberton, Baggs s.n., (UBC); Yale, Calder and Savile 8 1 t 3 1 (UBC); Cruickshank Canyon, Forbidden Plateau, Vancouver Island, Clark s.n. (V); Alpine Lodge, Garibaldi, Eastham s.n. (UBC); Pemberton, Eastham s.n. (UBC); Fraser River, Hitchcock and Martin 7363 (UC, WTU); Barrier, Garibaldi Park, Jackson s.n. (UBC); Yale, Krajina s.n. (UBC); Lytton-Boston Bar, Krajina s.n. (UBC); Lytton-Stein River Valley, Krajina H1I6 (UBC); Bulkley River, Hazel-ton, Lang $1 (UBC); Goldie Lake, Mt. Seymour, Lang 96-(UBC); 11 miles N of Cheekeye, Cheakamus River, Lang 117A, C, E, (UBC); Yale, Lang 118A, B, C, 120 (UBC); Fishways, Fraser River, Lang 161 (UBC); McGuire, Cheakamus River,. Lang 207, 209, 211 (UBC); 78 Daisy Lake, Garibaldi Station, Lang 2lU, 21$ (UBC); Dinky Peak, Mt. Seymour, Lang 218 (UBC); Bella Coola, McCabe 1, 130 (UC); Alexandra Bridge, Fraser River, McCabe 781 (UC); Stuie, hO miles E of Bella Coola, McCabe 1$11 (UC); Hell's Gate, New- combe s.n. (V); Dinky Peak, Mt. Seymour, E. B. Peterson s.n. (UBC); Othello, Coquihall River, T. M. C. Taylor 1287 (UBC); Yale, Taylor and Lewis-67, 71 (UBC); Dam Mt., W. Taylor s.n. (UBC);'. Castlecrag Mtn. Underhill h91 (V). CALIFORNIA: Trinity Co*:: Devil's Canyon Mts., White's Creek, Tracy 1U621 (UC); Tuolumne Co.: Strawberry Lake, Pinecrest, Rodin 6313 (UC). COLORADO: Boulder Co.: Boulder Canyon, Clokey 3993 (UC); Green Mt. below twin springs, Ewan 121U8 (WTU); Twin Springs, Green Mt., Weber 397b (WTU); South Vrain Canyon, 9 miles S of Lyons, Wherry s.n. (WTU); Larimer Co.; N of the Thompson, Qsterhout 5708 (UC); Summit Co.: Ten mile Canyon near Blue River, Brandegee s.n. (UC). OREGON: Clatsop Co.: Saddle Mt., Detling 709b (UBC); Hood River Co.: 1 Lost Lake, Mt. Hood Nat. For., J. W. Thompson 3619 (WTU); Cascade Locks, J. W. Thompson U876 (WTU); Multnomah Co.: Oneonta Gorge, Foster s.n. (WTU); 1/2 mile E of Crown Point, Lang 191 (UBC); Shepard's Dell, Lang 19$ (UBC); Wahkeen Falls, Lang 196 (UBC); Oneonta Gorge, Lang 197 (UBC). WASHINGTON: Yakima Region, Brandegee 323 (UC); Olympic Mts. Flett s.n. (WTU); Chelan Co.: Mt. Stuart, Wenatchee Mts., J. W. Thompson $872, 7722 (WTU); Clallam Co.: Halfway Rock, Mt. 79 Angeles Trail, Hitchcock and Martin 3532, (UC); Mt. Angeles, 0. N. Jones 330j (WTU); Hurricane Ridge, G. N. Jones U019 (WTU); near summit of Mt. Angeles, J. V/. Thompson 5581, IhPh (WTU); Clark Co.: intersection Skamania-Clark County line and Highway U. S. 830, Lang 20U, 205 (UBC); W of Camus, J. W. Thompson 8118 (UC); Jefferson Co. : Mt. Olympus, J. B. Flett 3090 (WTU); Humes Glacier, Queets River, Frye s.n. (WTU); King Co.: Stampede Tunnel, L. F. Henderson s.n. (WTU); Gaye Peak, J. W. Thompson 9682 (WTU); Kittitas Co.: Upper N Fork Tean-away River, Wenatchee Mts., Kruckeberg 5795 (UBC); Head of Beverly Creek, Wenatchee Mts., J. W. Thompson 9598 (WTU); Mason Co.: Mt. Ellinor, Eyerdam 128U (UC); Pierce Co.: Switch-back Trail, Mt. Ranier, J. B. Flett 192U (WTU); Indian Henry's Trail, Mt. Ranier, J. B. Flett 213U (WTU); Glacier Basin, Mt. Ranier, Grant s.n. (WTU); Mt. Ranier, McCalla s.n. (UBC); Mt. Ranier, Piper 297 (WTU); Mt. Rainier, Shumway s.n. (WTU); San Juan Co.: Mt. Constitution, Orcas Island, T. C. Frye s.n. (WTU); Mt. Constitution, A. S. Pope s.n. (WTU); Skamania Co.: Beacon Rock State Park, Lang 202 (UBC); Cape Horn, Suksdorf 2336 (UC); Dog Creek Canyon, near Cooks, Columbia River, Suksdorf II667 (WS); Snohomish Co.: Twin Lakes, Broadbent s.n. (WTU); Mt. Dickerman, J. W. Thompson 8776 (WTU); Whatcom Co.: Devil's Elbow, Skagit River Canyon, Otis 1071 (WTU); Mt. Hermann, near Mt. Baker Lodge, J. W. Thompson 5357 (WTU). WYOMING: Albany Co.: Crow Creek, Nelson 8902 (UC). 80 3. Polypodium hesperium Maxon, Proc. Biol. Soc. Wash. 13:200 (1900). Polypodium vulgare L. var. columbianum Gilbert, last N. Am. Pterid. 19, 38 (1901). Polypodium vulgare L. var. hesperium (Maxon) Nelson & Mac-bride, Bot. Gaz. U.:30 (1916). Polypodium prolongilobum Clute, Fern Bull. 18;97 (1910). Rhizome creeping, sweet, licorice-like taste and 3-6 mm in diameter, paleaceous; rhizome scales tan to castaneus, ovate to $ mm long, margin somewhat crenate, cells large, 25-30 in number across scale just above point of attachment; frond averaging 170 mm long, max. ca. 375 mm long; stipe stout, 12) (28-57-100)(155 mm long; blade coriaceous or herbaceous to membranous in shade, oblong 30) (63-lli*-189) (265 mm long, 12) (22-33-1*8 (83 mm wide; segment oblong, tip obtuse to acute, margin entire to serrate, 9) (11-18-26) (1*5 mm long, 5) (6-7-9) (11 mm wide, ratio of length to width 1.3) (1.9-2.1*-3.1) (3.8; hydathodes large, oval, many celled, veins free, forking 1-3 times, sorus oval, located midway between costa and segment margin, paraphyses very rare; chromosome number n = 7l*, 2n « 11*8; (Type: Coyote Canyon, Lake Chelan, Washington, Gorman 61*2 (US) I) 1 New fronds produced from June to September; found growing on rock and in rock crevices from South Central British Columbia south to New Mexico, Arizona, and Northern Mexico, east to the Rocky Mountains, west nearly to the coast in Cheakamus and Fraser River valleys of British Columbia, south along the east slope of the Cascade Mountains in Washington to Mount Hood, Oregon, absent from the Columbia Plateau in Washington, Central Oregon, and the Great Basin region of Utah and Nevada, Western limit of eastern southerly extension of range; 1 61 Eastern Washington, Northwestern Oregon, Idaho, Eastern Utah west to Zion Park in southern part of the state* Specimens cited ARIZONA: Coconino Co.: W fork, Oak Creek Canyon, Goddard $98 (UC); Upper Oak Creek, Goodding 188-1*7 (UC); W of Troutdale, Oak Creek, Goodding 197-U7 (UC); Kehl Canyon, Mogollon Rim Road, Phillips 29736 (UC); Graham Co. Marihild Canyon, Pinaleno Mts., Darrow, Pultz, and Phillips 21*83, 2U8U (UC); Pinecrest, . Pinaleno Mts. Maguire and Maguire 12015 (UC); Pima Co.: Soldier Camp, Catalina Mts., Phillips 2309 (UC); Sabino Canyon, Catalina Mts., Phillips 2376 (UC). BRITISH COLUMBIA: Okanagan, Okanagan Lake, J. R. Anderson s.n. (UBC); Edgewood, Arrow Lake, Beamish and Gilmartin 7272, 7316 (UBC); Watshan River, Beamish and Gilmartin 731*3 (UBC); Burton, Bell s.n. (UBC); Naksup, Bell s.n. (UBC); Silverton, Bell s.n. (UBC); Kootenay Bay, Kootenay Lake, Bosorowich s.n. (UBC); Gray Creek Kootenay Lake, Eastham s.n. (UBC); Marysville, Fodor 212 (UBC); Balfour, G. Foster s.n. (UBC); Victor Lake, W of Revelstoke, Hitchcock and Martin 7579 (UBC); 11.5 miles N of Clearwater, Lang 70 (UBC); 10 miles S of Clearwater, Lang 71, 111 (UBC); Mara Lake, Lang 72(UBC); Craigellachie, Lang 73 (UBC); Wigwam, Lang 71* (UBC); ll * miles S of Clearwater, Lang 112 (UBC); ll*.5 miles S of Clearwater, Lang 111* (UBC); Green River, Pemberton, Lang 210A, B (UBC); E of Alta Lake, Lang 212 (UBC); Ainsworth, McCabe 5991, 60Ui, 601*8, 61*58 (UC); between Fauquier 82 and Borton, McCabe 6620 (UC); Green Mt., Penticton, M. Stonor s.n. (UBC); Indian Reserve Summerland, M. Stonor s.n. (UBC); Cambie, Taylor and Szczawinski hh2 (UBC); Marblehead, Taylor and Szczawinski 610 (UBC); near St. Leon, Upper Arrow Lakes, F. Tusko s.n. (UBC); Trout Lake, Columbia R. Basin, E. Wilson 630 (UBC); Armstrong Okanagan, E. Wilson s.n. (UBC)^ COLORADO: Ovray Co.: Bear Creek Falls Canyon, San Ovray U. S. 550 Maguire, Pirarian, Richards 12761 (UC). IDAHO: Idaho Co.; Selway River below Moose Cr. Kirkwood 1722 (UC); Selway River above Selway Falls, M. Qwnbey and G. H. Ward 3130 (WTU); Kootenai Co.: Coeur D'Alene, M. Milburge 1075 (WTU); Shoshone Co.: Elk Creek, Abrams 762 (UC). MONTANA: Glacier Co.: Sun Camp, Glacier Nat. Park, W. T. McLaughlin 2961 (WTU); Lincoln Co.: 70 miles W of Kalispell, C. L. Hitch-cock 17768 (UC, WTU). OREGON: Clackamas Co.: Mirror Lake, Mt. Hood Nat. For., J. W. Thompson. 1633, 163b (WTU); Umatilla Co.: Bingham Springs, Umatilla River, Cusick 3287 (UC, WTU); Wallowa River, G. N. Jones 6597 (UC). i UTAH: Salt Lake Co.: "Little Cottonwood Canon, J. Clemens s.n. (UC); above Lake Blanche, Wasatch Mts., Holmgren, Holmgren, Holmgren 7117 (UC, WTU); Alta, Wasatch Mts., M. E. Jones s.n. (UC); Washington Co.: Zion Nat. Pk., Boyle 361 (UC). WASHINGTON: Brandegee 1208 (UC); Chelan Co.: Coyote Creek, Lake Chelan, Lang 130, 130-138, 138 (UBC); Coyote Canyon, Lake Chelan, Gorman 6b2 (US); Preston Falls, Entiat Valley, Morrill 255 (WTU); near Leavenworth, J. W. Thompson 6k0h (WTU); Lewis Co.: 83 Ohanapecosh Hot Springs, Mt. Rainier, J. W. Thompson 12689 (WTU); Pierce Co.: Tipsoo Lake, Warren 15$8 (UW); Stevens Co. Kettle Falls, Beattie and Chapman 2 2 2 5 (UC); Kettle Falls, Piper s.n. (UC); Whitman Co.: Kamiak Butte H. St. John 7 6 1 5 (UC). MEXICO: San Pedro Martin, Brandegee s.n. (UC); La Encantada, Sierra San Pedro Martir, Wigging and Demaree 5 0 0 0 (UC). In order to accurately interpret the evolutionary implications of meiotic chromosome behavior in the triploid hybrids found in North-western North America during this study, their putative parents must be established. By comparing the specimens in Figure 9, i t is clear that there are two morphologically distinct triploid hybrids. Because they have 1 1 1 sporophytic chromosomes they must have been derived from hybridization between a tetraploid with 7 U gametic chromosomes and a diploid with 3 7 gametic chromosomes. The fact that the tetraploid P. hesperium and the two diploids, P. glycyrrhiza and P. montense, are the only species known from the region where the triploids were found (see Map 2 ) is sufficient basis to suspect them as being the parental species. Each triploid probably shared P. hesperium as a common parent since i t is the only available tetraploid. This hypo-thesis is also borne out by their morphology. A;s often indicated in Chapter II, Results, the two triploids are more or less intermediate between P. hesperium and P. glycyrrhiza in one case (Figure 9a, b, c), and P. hesperium and p. montense in the other (Figure 9d). On the basis of the available evidence i t seems that these taxa are the 81* parental species. The parentage of the triploids is indicated by the following hybrid formulas: P. hesperium x glycyrrhiza and P. hesperium x montense. Because of the very small number of collections (see Specimens Cited), no attempt is made to formally characterize these taxa. Specimens Cited P. hesperium x glycyrrhiza BRITISH COLUMBIA: Kaske Creek, 59 miles E of Prince Rupert, Lang 99 (UBC)j Alexandra Bridge, Fraser River, Lang 125 (UBC)j, Green River, Pemberton, Lang 210-0 (UBC). P. hesperium x montense BRITISH COLUMBIA: Alexandra Bridge, Fraser River, Lang 122-3 (UBC). Several specimens of P. virginianum from Northeastern British Columbia and the Yukon were examined because of. their close proximity to the study area. Their distributions are shown in Map 2, and one of the collections is pictured in Figure 8e and l5f. Specimens Cited P. virginianum BRITISH COLUMBIA: Pouce Coupe Hills, Calverley 117 (V); vicinity of Beatton R. app. Lat. 57°5' N, Long., 122°35' W, H. M. Raup and D. S. Correll 10119 (UBC). YUKON: Sawmill Gulch, Yukon River, Berton U07 (UBC). CHAPTER V DISCUSSION OF RESULTS The evidence accumulated in this and previous investigations indicates the genus Polypodium in Western North America, as in Europe, presents an intricate evolutionary picture. It is now possible to speculate on the probable origin and relationship of members of the complex throughout its range and particularly in Northwestern North America. The basis of the general taxonomic confusion which has surrounded the complex will also be discussed. I. ORIGIN AND RELATIONSHIPS OF THE POLYPODIUM VULGARE COMPLEX IN NORTHWESTERN NORTH AMERICA When more than two identical genomes are found in a plant, as in autopolyploids, multivalent chromosome associations are formed at meiosis. Instead of only two homologous chromosomes there will be three or four or more homologues, a l l of which may undergo synapsis to 'form multivalents. The presence of multivalents is one of the char-acteristic features of autopolyploidy. The plants are often partially or completely sterile and are usually derived from a single ancestral diploid. The sterility may have genetic basis or may be caused by segregation of the multivalent associations. Autoploids usually bear a strong morphological resemblance to their diploid progenitor because of the presence of identical genomes. There are a few changes such as 56 increase in cel l size and often some physiological ones as well. Alloploids, on the other hand, contain two distinct genomes, each derived from two diploid progenitors. The f i r s t generation hybrid is usually sterile since meiosis is often upset by the lack of pairing between the nonhomologous genomes or chromosomes. This sterility "bottleneck" can be overcome by a doubling of its chromo-some complement so that each genome is duplicated to produce an indi-vidual with four genomes, two pairs of which are homologous. These homologous genomes can pair and so meiosis can proceed normally, thus overcoming the sterility barrier. An allopolyploid.is usually inter-mediate between its diploid progenitors in most characteristics and as a result often have different ecological preferences. By combining the attributes of the diploid species i t is often possible for the polyploid to occupy habitats not available to either of the diploids. One of the basic assumptions of meiosis is that homologous chromosomes pair in a gene for gene or segment for segment fashion during synapsis. A further assumption is often made that the degree of homology between chromosomes from different plants combined in i hybrids reflects the relationship of the sources of the chromosomes: thus the greater the degree of homology the closer the relationship. If two genomes are identical or very closely related, then, at prophase of meiosis there should be perfect bivalent pairing of homologous chromosomes. If the two genomes are not closely*related, then there should be l i t t l e or no pairing of the chromosomes. This comparative method of determining genome homologies and origins by observing meiosis in hybrids is not always reliable, since . 87 various factors such as translocations between chromosomes of non-homologous genomes or pairing between members of different genomes may obscure their origins. Wagner (1963) does not believe that chromo-some pairing necessarily shows homologies between chromosomes of entire genomes. He suggests that chromosome pairing in "aggregate species", such as the Polypodium vulgare complex, might be determined by simple genetic factors rather than a multitude of differences in genetic homology, but offers only circumstantial evidence (Mehra 1961; Verma and Loyal I960) to support this view. There is certainly no evidence of this in Polypodium vulgare s. l . The concept of homology, as re-flected in pairing at meiosis, s t i l l provides the best cytological criterion of related ancestry. It is only when the cytological picture is complete, however, that the concept of chromosome homology can provide a crit i c a l basis for outlining the mode and probable course of evolution of a particular plant group (Swanson 1957). The presumed relationship of the members of the complex is summarized in Figure 22. Although most of the information in the figure is of a cytogenetic nature, i t should be remembered that such evidence correlates with and is augmented by the morphology, ecology and geographical distribution of the cytotypes. Because of the com-plexity of Figure 22, some explanation of i t is given. The different cytotypes of the complex are aligned horizon-tally by ploidy level. The taxa in bold-lined boxes represent spe-cies, those in narrow-lined boxes hybrids. Within each box the name of the cytotype and its chromosomal configuration at meiosis is indicated. Shivas (196la) assigned a different letter to each To face page 88 Figure 22. Relationships in the Polypodium vulgare complex. \ \ 6x 5x 4 x 3x 2x R interjectu-n m n A A B B C C Rvulgare x interjectum 74 • • 371 AA B B C Raustrate x interjectum 37 D • 741 C C A B Rvulgare x australe 111 I A B C Z R australe 37 n C C Pvirgintanum x interjectum 3711.111 I AA B C D Rgiycyrmiza x vukjare 37 D•371 B B A R glycyrrhiza 37 n Rvirginianum x interjectum 37 n. 74 ] AA B C ZZJZXZN Rvirginianum x vulgare 37D-37J I virginianum 37 n Rvirgnianum 74 n AA DD Rvirginianum x virginianum 37 n. 37J R montense 37 U D O 3x Pglycyrrhiza x nesperium 37 n . 371 BB D R montense x hesperium 37 n 371 DD B 4x Ploidy level R hesperium 74 n 6x 5x R californicum 74 • B B E E 4x R « o u l « n x californicum 37 n • 37 I 3x Pcalifornicum 37D F?scouleri 37 n 2x Relationships in the Polypodium vulgare complex based on: 3x Tax o n Meiotic association Genome cytoiogical evidence MMHi synthesized hybrids — r — w i l d hybrids morpholog ica l evidence 4x cr CO 89 nonhomologous genome she discovered. For example, A represents a gen-ome of P. virginianum; B, a genome of P. glycyrrhiza, etc. The new genomes in this investigation were arbitrarily given new letters following Shivas' designations. In the hybrids these letters are used to indicate the source of the paired (II) or unpaired (I) chromosomes. In the triploid P. virginianum x vulgare, the 37 II are presumed to be the pairing of two A genomes from P. virginianum, while the 37 I are probably the P. glycyrrhiza genome B. This is represented in Figure 22 by AA B. The hybrids are connected to their known or putative parents by thick lines i f i t is a synthesized hybridization or by thin lines i f a natural hybrid. Detailed information on the hybridizations in Figure 22 con-nected by broad lines may be found in Shivas (196la). The results of the present investigation are shown below the diploid ploidy level in the figure. The remainder of the figure was compiled from the refer-ences in Table $. The following discussion is based largely on the cytogenetic study of the P. vulgare complex done by Shivas (1961a and b). It is intended to show how members of the complex are known to act cyto-logically so that the results of the present investigation may be intelligently assessed. Ideally such a cytogenetic study should have been carried out in conjunction with the present investigation, but limitations of time and the slow growth of the experimental material would not permit its execution. Fortunately, Shivas (196la and b) has done a considerable amount of work on the cytogenetics of the European members of the 9© TABLE 5. CHROMOSOME COUNTS OF THE POLYPODIUM VULGARE COMPLEX IN NORTH AMERICA, COMPILED FROM THE LITERATURE Species n or 2n Location Reference P. californicum 7k Portola, California Manton 1951 Searsville Lake, California 7k Mendocino Co., California Lloyd 1963 7k Solano Co., California Evans 1963 37 Monterey Co., California Lloyd 1963 P. glycyrrhiza 37 Mill Valley, California Manton 19$1 37 Duncan, British Columbia Manton 1950 37 Canadian Rockies Manton 1950 occidentale) 37 Humboldt Co., California Evans 1963 p. hesperium s . l . 7k Santa Catalina Mts., Arizona Knobloch 1962 7k Oak Creek Canyon, Arizona Lloyd 1963 7k Columbia River, Oregon Manton 1950 37 Snohomish Co., Washington Evans 1963 p. scouleri 37 Marin Co., California Evans 1963 37 Pt. Reyes, California Manton 1951 37 Sooke, British Columbia Taylor & Lang 1963 p. virginianum s . l . 37 Nova Scotia Manton 1950 37 Quebec Manton 1951 37 Smuggler's Notch, Vermont Manton 1957 1 7k Ontario Manton 1953 7k Amherstj Massachusetts Manton 1957 7k Britton 1953 \ 91 complex. She has analyzed the meiotic pairing behavior of chromosomes in synthesized hybrids between a l l the European and some American cyto-types. On the basis of her work we have some idea of how chromosomes act at meiosis in members of the complex. No F 1 hybrids between the diploid cytotypes of P. vulgare s; 1. have been found in nature nor have any been artificially synthesized. From this one must conclude that there is very l i t t l e homology between their genomes. This lack of homology is also reflected in the pairing behavior of chromosomes in polyploid hybrids. When Shivas crossed P. vulgare (n = 7k) with P. glycyrrhiza (n = 37) , she found that the resultant triploid had 37 bivalents and 37 univalents at meiosis (Figure 22). This precise pairing could be explained by autosyndesis or allosyndesis. In autosyndesis the 37 pairs in the triploid would be derived from the pairing of the 7k chromosomes from the gamete of P.' vulgare with the unpaired chromo-somes representing the 37 chromosomes from the P. glycyrrhiza gamete. In the other alternative, allosyndesis, the 37 bivalents would be iderived from the pairing of the 37 chromosomes from the P. glycyrrhiza gamete with 37 homologous chromosomes from the P. vulgare gamete, the 37 univalents being the remaining P. vulgare chromosomes. If auto-syndesis is the rule in this complex a l l triploids should result in the formation of n II + n I at meiosis. However, the cross between P. australe x P. vulgare yielded a triploid with 3n I, which could not occur in autosyndesis. In addition, autosyndesis would require an extra mechanism to prevent multivalent pairing in the tetraploid. On this basis one must conclude that pairing in the hybrids of the P. vulgare I i 92 complex is the result of allosyndesis. If pairing in the triploid does indicate that the tetraploid and diploid have a genome in common, then i t is assumed that the di-ploid or its ancestor gave rise, in conjunction with another diploid, to the tetraploid. With this brief introduction the possible parentage of the Western North American tetraploid Polypodium hesperium may be examined. P. hesperium s.s. is probably an allotetraploid. It forms only bivalents at meiosis, is not sterile, and is more or less intermediate morphologically between P. glycyrrhiza and P. montense. As soon as i t was realized that there was a tetraploid cyto-type in Western North America, a search was undertaken for triploid hybrids with the hope that their cytological analysis would give some clue to the origin of the tetraploid. As the distribution patterns of the two diploids and the tetraploid became known, i t was obvious that the best places to look for hybrids would be where the species were sympatric, e.g. the Fraser and Columbia River valleys. As shown on Map 2, page 5#, and Table 1, pages lb-15, triploid hybrids were found at Alexandra Bridge on the Fraser River and at Green River near Pemberton, B. C. The probable parentage of these triploid hybrids has already been established on the basis of their morphology, viz., P. hesperium x glycyrrhiza and P. hesperium x montense. Chromosome pairing in these hybrids is particularly significant since the diploid parents of the two triploids are also suspected, on the basis of morphology, 93 to be the progenitors of the tetraploid. The interpretation of these findings is shown in Figure 22. Since both morphologically-distinct triploids show 37 II + 37 I at meiosis, one must assume that one genome of P. glycyrrhiza and one of P. montense are present in p. hesperium. Meiosis in P. hesperium x glycyrrhiza is possibly open to question. The population examined morphologically in this study was from Green River, British Columbia, and meiosis was not seen. However, this pairing may be inferred from meiosis observed in a similar hybrid at Alexandra Bridge (Table 1 ) . It seems then that P. hesperium is an allotetraploid derived from hybridization between P. glycyrrhiza and p. montense or their immediate ancestors. This conclusion, however, must be considered tentative until i t can be confirmed by experimental hybridization using parents of known origin. With the information at hand, i t seems to be the most likely hypothesis. Although l i t t l e is known about the genetics of P. vulgare s . l . some observations have been made on the transmission of characters at the genome level. Each genome produces a phenotype characteristic of i its species. When these are combined with other genomes, the expression of certain characters may be suppressed or may dominate. For example, the paraphyses of P. virginianum are apparently recessive whem com-bined with European taxa of higher ploidy level (Shivas 1961); the ' paraphyses of P. montense are expressed in the triploid hybrid and virtually absent in the allotetraploid parent P. hesperium. Shivas (1961) also found that the oval shape of the sorus of P. australe is s t i l l expressed when combined with P. vulgare, with a circular sorus, 9k to produce P. interjectum. The presence of the oval sorus in P. hesperium is difficult to explain since neither putative parent commonly has oval sori. There seem to be two possibilities: f i r s t , the oval sorus could have risen anew in the tetraploid; or second, an ancestral diploid had an oval sorus which its modern descendent has lost. The only contemporary diploid that might qualify is P. australe, and this is not a likely parent on the basis of Shivas' hybridizations. Now that the probable parentage of P. hesperium has been esta-blished, one can theorize about the actual sequence of events which lead to the formation of P. hesperium and the other polyploid taxa of the complex. The long creeping scaled rhizome and relatively thick-textured articulated frond of P. vulgare s . l . are a l l features usually found in tropical epiphytic ferns. Although some members of the P. vulgare complex may be found as epiphytes, they also frequently grow on rocks, and soil. According to Christensen (1928) there are two possibilities that might explain the resemblance of P. vulgare to tropical epiphytic ferns. The complex may be a recent immigrant from the tropics, or i t could be a survivor from Tertiary times when subtropical vegetation was found over much of its present range. Since the present wide distribution argues against a recent immigration, the.complex probably has persisted in the Northern Hemisphere since the Tertiary (or before). Christensen (1928) contrived a series of morphologically similar species, distributed along both sides of the North Pacific, 9$ which show gradation from temperate species (free veined) to tropical ones with a 37), and P. hesperium Maxon (n = 7h). 2. P. hesperium is presumed to be an alloploid since i t forms only bivalents at meiosis and is more or less morphologically intermediate between the two diploid species. 3. Meiotic chromosome pairing in two morphologically dis-tinct triploid hybrids, P. hesperium->x glycyrrhiza and P. hesperium x 'montense both showed n II + n I. This pairing indicates that the glycyrrhiza genome and the montense genome are both present in the tetraploid. It. It is suggested that P. hesperium is an allotetraploid derived from hybridization between P. glycyrrhiza and P. montense or their immediate progenitors. The i n i t i a l hybridization may have occurred in the late Pliocene or early Pleistocene when mountain building activities caused the climate of the interior of the con-105 tinent to become more arid, providing new habitats for the tetraploid to colonize. ; 5. At the diploid level three main lines of divergence are recognized in the complex. The first line is characterized in North America by P. glycyrrhiza, the second by P. virginianum, and the third by P. californicum. 6 . The morphological intergradation of the taxa is due to phenotypic plasticity, alloploidy, hybridization, and the use of unreliable quantitative taxonomic characters. Much of the morphological intergradation is removed by using more qualitative characters to cir-cumscribe the taxa. 7. Further studies are s t i l l needed for a complete under-standing of evolution in the complex. Future investigations in the complex of the type done here are planned to include populations from throughout the range of the complex in Western North America and Eastern Asia. The study will be enlarged to include cytogenetical. analysis of hybrids, particularly between the diploid cytotypes. Knowledge of the complex has reached the place where chromatographic studies of the species would be most interesting. LITERATURE CITED 107 Abrams, L. 1°U0. Illustrated flora of the Pacific States. VI. Stanford University Press, Palo Alto, California. $38 p. Axelrod, D. I. 1962. Post-Pliocene uplift of the Sierra Nevada, California. Geol. Soc. Amer. Bull. 73:183-198. Benson, L. 1962. Plant taxonomy. Ronald Press Co., New York. 1*91* p. Blasdell, R. F. 1963. A monographic study of the fern genus Cystopteris. Mem. Torr. Botan. Club. 21(1*):1-102. Bobrov, A. E. 1961*. A comparative morphological and taxonomical study of the species of Polypodium L. of the flora of the USSR. Botanicheskii Zhurnal i*9:53 li-5U8. Britton, D. M. 1953. Chromosome studies on ferns. Am. Journ. Bot. h0:575-583. Chaney, R. 191*0. Tertiary forests and continental history. Bull. Geol. Soc. Am. 5:1*69-1*88. 19U7. Tertiary centers and migration routes. Ecol. Monogr. 17:139-1U8. 19h8. The bearing of the living Metasequoia on problems of Tertiary paleobotany. Proc. Natl. Acad. Sci. Wash. 3b:$03-5l$. Christensen, C. 1928. On the systematic position of Polypodium . vulgare. Dansk. Bot. Arkiv. 5:1-10. Clute, W. N. 1910. Two new polypodies from Arizona. Fern Bull. 18:96-98. Copeland, E. B. 19l*7. Genera Filicum. Ronald Press Co., New York. 21*7 p. Detling, L. 1958. Peculiarities of the Columbia River Gorge flora. Madrono. Il*:l60-172. Eaton, D. C. 1856. On three new ferns from California and Oregon. Am. Journ. Sci. and Arts. ser. 2. 22:138. Evans, A. M. 1963. New chromosome observations in the Polypodiaceae and Grammitidaceae. Caryologia 16:671-677. 1961*. Ameiotic alternation of generations: a new l i f e cycle in the ferns. Science lli3:261-263. 1 108 and VJ. H. Wagner, Jr. 1°6U. Dryopteris goldiana x inter-media—a natural woodfern cross of noteworthy morphology. Rhodora 66:255-266. Fernald, M. L. 1922. Polypodium virginianum and P. vulgare. Rhodora. 2U:125-1U2. Fischer, L. and E. V. Lynn. 1933. Licorice fern and wild licorice as substitutes for licorice. J. Am. Pharm. Assoc. 22:(12) 1225. Frye, T. C. 193b. Ferns of the Northwest. Metropolitan Press, Portland, Ore. 177 p. Gilbert, B. D. 1899. Two new Polypodia from New Zealand. Bull. Torr. Botan. Club. 26:316-317. Hegnauer, R. 1962. Chemotaxonomie der Pflanzen. Bd. 1. Thallo-phyten, Bryophyten, Pteridophyten, und Gymnospermen. Birk-hauser verlag Basel und Stuttgart. 517 p. Holttum, R. E. 19b7. A revised classification of leptosporangiate ferns. J. Linn. Soc. (Bot.) 53:123-158. 195b. Flora of Malaya. II. Ferns of Malaya. Gov. Printing Office, Singapore. 6b3 p. Hulten, E. I960. The circumpolar plants. I. K. V. A. Handl. 8:176-177, 253. Johansen, D. A. 19b0. Plant microtechnique. McGraw-Hill, New York and London. 523 p. Kellogg, A. 185U. Description of P. faleaturn. Proc. Calif. Acd. Nat. Sci. VI:19. Sec. Ed. published by the Acd. Dec. 1873. • Knobloch, I. W. 1962. Tetraploid Polypodium vulgare var. colum-bianum from Arizona. Am. Fern Journ. 52:65-68. Krajina, V. J. 1959. Bioclimatic zones in British Columbia. Univ. British Col. Botan. Series No. 1. 1*7 p. Lanjouw, J. and F. A. Stafleu. 1961. Index Herbariorum. Part I. 5th ed. Regnum Veg. 31:1-251. Lloyd, R. M. 1963. New chromosome numbers in Polypodium L. Am. Fern J. 53:28-123. and F. A. Lang. 196b. The Polypodium vulgare complex in ' North America. British Fern. Gaz. 9:i6d-177. 10? Lovis, J. D. 196b. The taxonomy of Asplenium trichomanes in Europe. British Fern Gaz. 9:lb7-160. Manton, I. 19$0. Problems of cytology and evolution in the pteri-dophytes. University Press, Cambridge. 316 p. 1901. The cytology of Polypodium in America. Nature, Lond. 167:37. 1957. The problem of Polypodium virginianum. Am. Fern Journ. U7:129-13b. 1958. The concept of the aggregate species in systematics of today. Uppsala Universitets Arsskrift. 6;10b-112. and M. G. Shivas. 1953. Two cytological forms of Poly-podium virginianum in eastern North America. Nature, Lond. 172:uio: Manum, S. 1962. Studies in the Tertiary flora of Spitzbergen, with notes on Tertiary floras of Ellesmere Island, Greenland, and Iceland. Norks. Polanist. Skrifter. 125:1-127. Martens, P. 1950a. Les organes glanduleux de Polypodium virginianum. III. Nouvelles donnees geographiques, systematiques et histologiques. La Cellule. 53:187-212. 1950b. Les paraphyses de Polypodium vulgare et la sous-espece serratum. Bull. Soc. R. Bot. Belgique. 52:225. and N. Pirard. 19b3» Les organes glanduleux de Polypodium virginianum. II. Structure, origine et signification. La" Cellule. £9:383. Mathews, W. H. and G. E. Rouse. I963. Late Tertiary volcanic rocks 1 and plant-bearing deposits in British Columbia. Geol. Soc. Amer. Bull. 7U:55-60. Maxon, W. R. 1900. Polypodium hesperium, a new fern from Western North America. Proc. Biol. Soc. Wash. 13:200. Mehra, p. N. 1961. Cytological evolution of ferns with particular reference to Himalayan forms. U8th Indian Science Congress, Part II: Presidential address, pp. l-2b. Moore, R. C. 1958. Introduction to historical geology. Second ed. McGraw-Hill, New York. 656 p. 110 Morton, C. V. and C. Neidorf. 195k. Polypodium vulgare var. v i r -ginianum. Am. Fern. J. l * l * : l l l - l l i * . Munz, P. A. 1959. A California f l o r a . U. of California Press, Berkeley and Los Angeles. 1681 p. Peck, M. E. 1961. A manual of the higher plants of Oregon. Bin-fords and Mort, Portland,. Ore. 936 p. Pickett, F. L. 1931. Notes on xerophytic ferns. Am. Fern J. 21:1*9-57. and L. A. Thayer. 1927. The gametophytic development of certain ferns: Polypodium vulgare var. occidentale and Pellaea densa. Bull. Torr. Botan. Club. 51*:2i*9-255. Piper, C. V. 1906. Flora of the state of Washington. Cont. U. S. Nat. Herb. 11:1-637. Province of Br i t i s h Columbia. Climate of B r i t i s h Columbia. Report for 1963. Queen's Printer, Victoria, B. C. 35 P« Rothmaler, W. and U. Schneider. 1962. Die gattung Polypodium in Europa. Kulturpf lanze. Beiheft. 3:23l*-2i*8. Sax, K. and H. J. Sax. 1937. Stomata size and distribution i n d i -ploid and polyploid plants. J. Arnold Arb. l8:10l*-172. Shivas, M. G. 196la. Polypodium in Europe and America. I. Cytology. J. Linn. Soc. (Bot.) $8s13-26. 196lb. Polypodium i n Europe and America. II. 'Taxonomy. J. Linn. Soc. (Bot.) 58:27-38. j 1962. The Polypodium vulgare complex. British Fern Gaz. 9:65-70. Sim, T. R. 1915. The ferns of South Africa. Second ed. Cambridge University Press. 381* p. 186 p l . Snow, R. 1963. Alcoholic hydrochloric acid-carmine as a stain for chromosomes in squash preparations. Stain Tech. 38:9-13. Slater, J. R. 1961*. Fern distribution i n Washington State. Occ. Papers Dept. of Bio. Univ. Puget Sound. 27:21*3-256. Stebbins, G. L. Jr. 1950. Variation and evolution in plants. Columbia University Press, New York. 61*3 p. I l l Steeves, T. A., I. M. Sussex, and C. R. Partanen. 195$. Invitro studies on abnormal growth of prothallia of the bracken fern. Am. Journ. Botan. 1*2:232-21;$. Swanson, I. 19$7. Cytology and cytogenetics. Prentice Hall, Inc., Inglewood Cliffs, N. J. $96 p. Systematics Association Committee for Descriptive Biological Termi-nology. 1962. II. Terminology of simple symmetrical plane shapes (chart 1). Taxon ll:ll*$-l$6. Taylor, T. M. C. 1963. The ferns and fern-allies of British Columbia. B. C. Provincial Museum, Victoria, B. C. Hand-book No. 12. 172 p. and F. A. Lang. 1963. Chromosome counts in some British Columbia ferns. Am. Fern. J. $3:123-126. Verma, S. C. and D. S. Loyal. I960. Colchi-autotetraploidy in Adiantum capillus-veneris. Nature. 188:1210-1211. Wagner, W. H . Jr. 19$1*. Reticulate evolution in Appalachian aspleniums. Evolution 8:103-118. 1963. Biosystematics and taxonomic categories in lower vascular plants. Regnum Veg. 27:63-71. 1961*. Paraphyses: Filicineae. Taxon. 13:56-61*. _ and L. L. Chen. 1965. Abortion of spores and'sporangia as a tool in the detection of Dryopteris hybrids. Am. Fern J. $5:9-29. ' Walker, S. 195$. Cytogenetic studies in the Dryopteris spinulosa t complex. I. Watsonia. 3(10:193-209. Whittaker, R. H. 1961. Vegetation history of the Pacific Coast states and the "central" significance of the Kalmath region. Madrono. 16:$-23. Wood, C. E., Jr. 19$$. Evidence for the hybrid origin of Drosera anglica. Rhodora. $7:10$-130. APPENDIX Polypodium Collector Herbarium No. Acc. No._ Date Locality. Can. U.S.; State or Prov. County Local, notes Ecological, notes: Growing on. rock ? s o i l ? tree Spores: aborted normal ^/sporangium Rhizome: dia. mra; trichomes,shape color stripe Frond: measurments SOL Bl. B/S Bw Bl/w li p Ibp l/b Pl. Pw El/w W5, SC A Sinus: W N Margin:. Serrate sp. ap. Entire Epidermal cells: Hydathodes 1 s Stoma 1. 2. mixed Reproductive structures.:. Sorus; 0 R mar. med. Sporangia # arm. cells gland, hairs Paraphyses 1. 2. 3. U . Age of sporangia* Y. M. D Ploidy level. 2n 3n Lai PRESUMED Basis: sp.s. st,s.;other