|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 IN VITRO ECOLOGY OF Calopogon tuberosus var. tuberosus : A NEW CONCEPT IN SPECIES CONSERVATION By PHILIP KAUTH A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009
2 2009 Philip Kauth
3 To Meredith, who else
4 ACKNOWLEDGMENTS I thank Dr. Michael Kane fo r his dedication, passion, enthusiasm, and encouragement, which in turn fueled my passion and excitement I thank my co-chair, Dr. Wagner Vendrame for his support and encouragement from a distance. I also thank Dr Carrie Reinhardt-Adams, Dr. Debbie Miller, and Dr. Thomas Sheehan for se rving on my supervisory committee and sharing their knowledge, ideas, and excitement. Over the last several years I had the opportunity to share a work environment with great people. I could not have shared space, ideas, an d knowledge over the years with better colleagues and friends than Tim Johnson, Daniela Dutra, Scott Stewart, Xiuli She n, and Carmen ValeroAracama. I especially thank Tim for the hours of brainstorming, talking, watching the fights, and trips to the Refuge. I thank Nancy Philman not only for her incredible friendship, but also for much of my development (even though she probabl y will deny her involvement). This research could not have been conducted without the assist ance from many that have either collected seed, issued permits, served as fiel d trip guides, or provided t houghtful insights including Larry Richardson (USFWS-FPNWR), Mary Bunch (South Carolina Heritage Preserves-SCDNR), Jim Fowler (Greenville, South Carolina), Howard Lorenz (Wisconsin), and Doug Goldman (Harvard University). I thank Kip Knudson not only for co llecting seed and lead ing me through Carney Fen numerous times, but also for influencing me more than me takes credit for. I also thank Dr. Hector Perez and Dr. Charles Guy for use of their incubators and growth chambers. I thank my parents for their endless suppor t, love, and encouragement throughout my LONG educational process. I thank my tw o beautiful dogs, Mya and Ladybug, for providing hours of laughter, and for showing me that life is a true blessi ng and that everyone deserves a second chance at life. Finally I thank the love of my life, Mered ith. I could not have done this without her love, sacrifice, and support and for that I am forever grateful.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........9 LIST OF FIGURES.......................................................................................................................10 ABSTRACT...................................................................................................................................13 CHAP TER 1 LITERATURE REVIEW.......................................................................................................15 Introduction................................................................................................................... ..........15 The Orchid Seed................................................................................................................ .....16 Orchid Seed Germination....................................................................................................... 18 History and Background.................................................................................................. 18 Symbiotic Orchid Seed Germination............................................................................... 19 Asymbiotic Orchid Seed Germination............................................................................ 20 Factors Effecting Orchid Seed Germination................................................................... 25 Photoperiod.............................................................................................................. 25 Temperature............................................................................................................. 28 Ecotypes..................................................................................................................................31 Introduction................................................................................................................... ..31 Importance of Ecotypes to Plant Conservation............................................................... 32 Ecotype Development and Differentiation...................................................................... 34 Common garden studies...........................................................................................34 Seed germination ecology........................................................................................ 37 Biomass allocation................................................................................................... 38 In vitro ecology ........................................................................................................40 Ecotypic Differentiation in Orchids................................................................................ 42 Seed source............................................................................................................... 42 Pollination ecotypes................................................................................................. 44 Plant of Study.........................................................................................................................45 Diversity within Calopogon tuberosus ............................................................................46 Floral Biology of Calopogon tuberosus ..........................................................................48 Seed Germination of Calopogon tuberosus ....................................................................49 Habitat Descriptions of Calopogon tuberosus ........................................................................51 Carney Fen.......................................................................................................................52 Ashmore Heritage Preserve............................................................................................. 54 Site C...............................................................................................................................55 Eva Chandler Heritage Preserve......................................................................................56 Goethe State Forest.......................................................................................................... 58
6 Florida Panther National Wildlife Refuge....................................................................... 60 2 EFFECTS OF PHOTOPERIOD AND GE RMINATION MEDIA ON IN VITRO SEED ECOLOGY OF Calopogon tuberosus ....................................................................................70 Introduction................................................................................................................... ..........70 Materials and Methods...........................................................................................................71 Seed Source.....................................................................................................................71 Seed Viability Test..........................................................................................................72 Media and Seed Preparation............................................................................................ 72 Photoperiod Effects on Asymbiotic Germin ation and Early S eedling Development.....73 Photoperiod Effects on Advanced I n Vitro Seedling Development................................ 73 Asymbiotic Germination Media Evaluation .................................................................... 74 Soil Analysis....................................................................................................................75 Statistical Analysis.......................................................................................................... 75 Results.....................................................................................................................................76 Seed Viability..................................................................................................................76 Photoperiod Effects on Germination and Early Development........................................ 76 Photoperiod effects on Advanced Seedling Development.............................................. 78 Media Effects on Germination and Early Development................................................. 79 Media Effects on Advanced Seedling Development.......................................................81 Soil Nutrient Analysis..................................................................................................... 81 Discussion...............................................................................................................................82 Seed Viability and Quality..............................................................................................82 Photoperiod.................................................................................................................... ..83 Media Screen and Soil Nutrient Availability.................................................................. 85 Conclusions.............................................................................................................................87 3 EFFECTS OF COLD STRATIFICATI ON AND DIUR NAL TEMPERATURES ON IN VITRO GERMINATION OF Calopogon tuberosus ............................................................101 Introduction................................................................................................................... ........101 Materials and Methods.........................................................................................................103 Seed Source...................................................................................................................103 Media and Seed Preparation.......................................................................................... 103 Cold-stratification Effects on Seed Germination.......................................................... 104 Diurnal Temperature Effects on Seed Germination...................................................... 104 Scanning Electron Microscopy......................................................................................105 Histological Sectioning.................................................................................................105 Statistical Analysis........................................................................................................ 106 Results...................................................................................................................................106 Effects of Cold-stratifica tion on Seed Germ ination......................................................106 Effects of Diurnal Temperatures on Seed Germ ination and Early Development......... 107 Scanning Electron Microscopy......................................................................................108 Histology.......................................................................................................................108 Discussion.............................................................................................................................109 Cold-stratification Effects on Seed Germination.......................................................... 109
7 Diurnal Temperature Effects on Seed Germination...................................................... 112 Conclusions...........................................................................................................................114 4 COMPARATIVE SEEDLING BIOMASS ALLOCATION AND CORM FORMATI ON AMONG WIDESPREAD Calopogon tuberosus POPULATIONS............ 121 Introduction................................................................................................................... ........121 Materials and Methods.........................................................................................................124 Seed Source...................................................................................................................124 Seedling Transfer and Data Collection..........................................................................125 Results...................................................................................................................................126 Corm Formation............................................................................................................126 Shoot Length.................................................................................................................127 Root Length and Number..............................................................................................128 Biomass Allocation....................................................................................................... 128 Discussion.............................................................................................................................130 Conclusions...........................................................................................................................134 5 EFFECTS OF CHILLING AND CUTTI NG C ORMS ON CORM DORMANCY AMONG WIDESPREAD POPULATIONS OF Calopogon tuberosus ...............................144 Introduction................................................................................................................... ........144 Materials and Methods.........................................................................................................147 Chilling Effects on Corm Dormancy and Shoot Regrowth...........................................147 Data Collection and Statistical Analysis....................................................................... 148 Effects of Cutting Corms on Corm Dormancy and Shoot Regrowth............................ 149 Data Collection and Statistical Analysis....................................................................... 150 Results...................................................................................................................................151 Chilling Effects on Corm Dormancy and Shoot Regrowth...........................................151 Effects of Cutting Corms on Corm Dormancy and Shoot Regrowth............................ 154 Discussion.............................................................................................................................156 The Role of Chilling on Dormancy............................................................................... 156 The Effect of Cutting Corms on Dormancy.................................................................. 158 Conclusions...........................................................................................................................159 6 SUMMARY AND CONCLUSIONS...................................................................................167 APPENDIX A FIELD TRANSPLANT OF Calopogon tuberosu s IN SOUTH FLORIDA......................... 170 Introduction................................................................................................................... ........170 Materials and Methods.........................................................................................................172 Study Site..................................................................................................................... ..172 Seed Source and Propagation........................................................................................172 Field Establishment.......................................................................................................173 Comparison of propagule type on field survival.................................................... 173 Seedling survival in a burned and unburned field plot ........................................... 174
8 Data Recording and Statistical Analysis....................................................................... 174 Results...................................................................................................................................174 Comparison of Propagule Type on Field Survival........................................................ 174 Seedling Survival in a Burned and Unburned Field Plot.............................................. 175 Discussion.............................................................................................................................175 Management Recommendations........................................................................................... 178 B MORPHOMETRIC ANALYSIS OF Calopogon tuberosu s POPULATIONS....................184 LIST OF REFERENCES.............................................................................................................192 BIOGRAPHICAL SKETCH.......................................................................................................215
9 LIST OF TABLES Table page 1-1 Summary of Calopogon tuberosus populations studied ....................................................62 2-1 Six stages of orchid seed development.............................................................................. 89 2-2 Comparative mineral salt content of orchid seed germ ination media................................90 2-3 Comparative soil nutrient analysis from the study sites.................................................... 91 2-4 Effect of scarification time on embryo viability of Calopogon tuberosus seeds from populations studied............................................................................................................ 92 4-1 Comparative changes in mean corm diameter of Calopogon tuberosus seedlings of different geographic sources during 20 weeks in vitro culture........................................ 135 4-2 Comparative change in mean shoot length of Calopogon tuberosus seedlings of different geographic source during 20 weeks in vitro culture ......................................... 136 4-3 Comparative changes in m ean root number of Calopogon tuberosus seedlings of different geographic source during 20 weeks in vitro culture......................................... 137 4-4 Comparative changes in m ean root length of Calopogon tuberosus seedlings of different geographic source during 20 weeks in vitro culture......................................... 138 4-5 ANOVA results for Calopogon tuberosus seedling biom ass allocation after 20 weeks in vitro culture..................................................................................................................139 4-6 Comparative biomass allocation to shoots, roots, and corm s of Calopogon tuberosus seedlings of different geographic source......................................................................... 140 A-1 Shoot lengths recorded for actively growing Calopogon tuberosus seedlin gs in February and April 2009.................................................................................................. 179
10 LIST OF FIGURES Figure page 1-1 Monthly temperatures at population locations studied ...................................................... 63 1-2 Carney Fen, Michigan...................................................................................................... ..64 1-3 Ashmore Heritage Preserve, South Carolina..................................................................... 65 1-4 Site C, South Carolina........................................................................................................66 1-5 Eva Chandler Heritage Preserve, South Carolina .............................................................. 67 1-6 Goethe State Forest, Florida.............................................................................................. 68 1-7 Florida Panther National W ildlife Reserve, Florida.......................................................... 69 2-1 Photoperiod effects on in vitro seed germ ination and development of Calopogon tuberosus from widespread populations............................................................................93 2-2 Photoperiodic effects on the developm ental index of Calopogon tuberosus seedlings from widespread populations............................................................................................. 94 2-3 Effects of photoperiod on in vitro seedling developm ent of Calopogon tuberosus from widespread populations............................................................................................. 95 2-4 Percent dry weight biomass allocation in Calopogon tuberosus seedlings ....................... 96 2-5 Comparative leaf number, shoot length, root number, root length, and corm diameter in Calopogon tuberosus seedlings from widespread populations...................................... 97 2-6 Effects of culture media on in vitro seed germ ination and subsequent development of Calopogon tuberosus from widespread populations.......................................................... 98 2-7 Effects of culture media on seedling developm ental index of Calopogon tuberosus from different populations................................................................................................. 99 2-8 Culture media effects on early seedling development of Calopogon tuberosus from widespread populations.................................................................................................... 100 3-1 Effects of chilli ng seeds at 10C in darkness on germ ination of Calopogon tuberosus seeds from distant populations......................................................................................... 116 3-2 Diurnal temperature effects on germina tion and d evelopment of unchilled seed and Calopogon tuberosus seeds from different populations.................................................. 117 3-3 Developmental index of unchilled Calopogon tuberosus seeds from widespread populations.......................................................................................................................118
11 3-4 Comparative scanning elect ron m icroscopy of seeds from widespread populations of Calopogon tuberosus .......................................................................................................119 3-5 Light micrograph cross sections of mature Calopogon tuberosus seeds from widespread populations.................................................................................................... 120 4-1 In vitro seedling developm ent of Calopogon tuberosus from widespread populations..141 4-2 Correlation of growing s eason length and percent co rm biomass allocation of Calopogon tuberosus seedlings from widespread populations........................................ 143 5-1 Outline of in vitro to ex vitro growth of Calopogon tuberosus .......................................160 5-2 Ex vitro growth com parison of representative Calopogon tuberosus plantlets............... 161 5-3 Effects of chilling corms at 10 C on shoot em ergence after 16 weeks of ex vitro growth of Calopogon tuberosus plantlets........................................................................162 5-4 Effects of chilling corms at 10C on growth and development of Calopogon tuberosus plantlets ...........................................................................................................163 5-5 Effects of cut and uncut unchilled corm s on regrowth of Calopogon tuberosus shoots.164 5-6 Percent shoot regrowth from cut and uncut corms of Calopogon tuberosus over 8 weeks in vitro culture .......................................................................................................165 5-7 Comparative regrowth from cut and whole corms of Calopogon tuberosus Data was collected after 8 weeks in vitro culture ............................................................................ 166 A-1 Field translocation study at the F lorida Panther Nationa l Wildlife Refuge. A) Map of the FPNWR......................................................................................................................180 A-2 Monthly temperatures recorded at Un it 23 in the Florida Pan ther National Wildlife Refuge..............................................................................................................................181 A-3 Survival of Calopogon tuberosus propagules at the Florid a Panther Na tional Wildlife Refuge..............................................................................................................................182 A-4 Survival of Calopogon tuberosus seedlings in a burned and unburned plot at the Florida Panther National W ildlife Refuge....................................................................... 183 B-1 Labeled parts of a Calopogon tuberosus flower m easured..............................................186 B-2 Labeled close up of a Calopogon tuberosus flower.........................................................187 B-3 Whole plant morpho m etrics analysis of Calopogon tuberosus .......................................188 B-4 Flower morphometrics of Calopogon tuberosus .............................................................189
12 B-5 Flower part morphometrics of Calopogon tuberosus ......................................................190 B-6 Labellum and column morphometrics of Calopogon tuberosus ......................................191
13 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy IN VITRO ECOLOGY OF Calopogon tuberosus var. tuberosus : A NEW CONCEPT In SPECIES CONSERVATION By Philip Kauth December 2009 Chair: Michael Kane Cochair: Wagner Vendrame Major: Horticultural Science The importance of ecotypic differentiation has recently been highlighted for plant conservation purposes, but use of local plant mate rial for orchid restor ation has been largely ignored. Using local plant material is necessa ry to maintain ecosystem and plant population stability, but little is known re garding the degree of ecotypi c differentiation necessary to maintain stability. Ecotypic differentiation am ong geographically dist ant populations of the orchid, Calopogon tuberosus was examined using in vitro ecology methods. Comparative effects of photoperiod, germination media, temperature, and chilling on asymbiotic in vitro seed germination and seedling development of C. tuberosus populations were examined. Seedling biomass allocation and corm dormancy were also examined. Regardless of germination treatment, Michigan seeds germinated and developed rapidly, while south Florida seeds germinated and developed slowly. This was likely due to the faster onset of winter conditions following seed dispersal in more northern e nvironments. Photoperiod did not significantly influence seed germination and development. Different germination media significantly influenced germination and development depe nding on source due to soil nutrient variation among populations. Higher temperatures promoted increased germination percentages in South
14 Carolina seeds, while Michigan and south Florida seed germination were higher at lower temperatures. Chilling seeds effectively increas ed germination in all populations, but seed germination in northern populations exceeded em bryo viability in longe r chilling treatments. This was reflected by the thicker testae require d to protect seeds of northern populations form harsher winter climates. During a detailed tim ecourse seedling development study, Michigan seedlings allocated more biomass to corms, and developed corms fa ster than all other populations. Higher corm biomass allocation in nor thern populations corre lated strongly with a shortened growing season. The rapid corm formation and biomass allocation in seedlings from more northern populations represented an adap tive response to a shorter growing season to increase survival. All populati ons required a chilling period to break corm dormancy, but longer chilling periods promoted faster and a higher percentage regrowth. Cutting corms also broke dormancy, but Michigan corms responded more rapidly compared to other populations. In conclusion, in vitro techniques were found to be appropr iate to differentiate ecotypes by detecting their unique ecologica l growth strategies. Ecotypic differentiation influenced by growing season length, temperature, and soil nutrient availability is occurring in C. tuberosus
15 CHAPTER 1 LITERATURE REVIEW Introduction The Orchidaceae is one of the largest and most diverse family of angiosperms with an estimated 17,000 to 35,000 species (Dressler, 1993) All orchids share at least six common features including zygomorphic flowers, a gynandrium (column), a rostellum, pollinia, a labellum, and microscopic seeds (Sheehan a nd Black, 2007). Seventy percent of all orchid species are epiphytes, but terrest rial, aquatic, lithophytic, and unde rground species exist as well (Dressler, 1993). The Orchidaceae has evolved highly specialized associations with pollinators and mycorrhizal fungi that allow species to occupy and survive various habitats (Otero et al., 2004; Jerskov et al., 2006). The highly diverse floral array is a direct result of the specific pollination mechanism of each species (Tremblay, 1992). Civilization has been fascinated with orchids since antiquity. Early references to orchids can be found in ancient Chinese literature dati ng back to 800 B.C., in which Confucius lauded the orchid for its fragrance (Ber liocchi, 2000). Orchids owe much of their appeal to the ancient Greeks who associated orchids wi th passion and fertility, and crea ted many myths concerning the origin of orchids. These myths often involved devious characters and death, thus setting the future for these highly desired plants. The name orchid is derived from the Greek word for testicles, orchis first used by Theophrastus in scien tifically describing orchid plants. Great scientists and writers have popularized orchids. Charles Darwin was captivated by their diverse pollination mechanisms. Angraecum sesquipedale an orchid with a 30 cm nectar spur, is often referred to as Da rwins orchid since he hypothesized that a moth was the potential pollinator. Although Darwin was criticized, 40 year s later a sphinx moth with a 30 cm proboscis was discovered in the same habitat as the orch id. Writers including Shakespeare, H.G. Wells,
16 and Thoreau included orchids in many of their writings (Berliocchi, 2000). In the James Bond novel and film, Moonraker, an orchid was the so urce of a deadly nerve toxin. The book Orchid Thief and the movie based on the book, Adaptati on, popularized the famous and sought after Dendrophylax lindenii the ghost orchid. As orchids gain fame and popularity an opportunity exists to educate the public a bout conservation of orchids. Although the Orchidaceae is diverse and widesp read, the family is critically imperiled. Over-collecting, pollinator decline, habitat c onversion and loss, and habitat mismanagement have caused a drastic reduction in wild orchid populations (Koopowitz et al. 2003). Many organizations and individuals are associated w ith restoring orchid populations to historic numbers by propagating orchids fr om seed. Large amounts of info rmation exist regarding the propagation science of orchids for conservation and reintroduction pur poses. However, large gaps in propagation science still exist since the fi rst orchid seeds were successfully germinated asymbiotically in vitro. A major omission in propagation sc ience of orchids is the use of in vitro conditions to study ecological factors that promot e seed germination and seedling development. This area of propagation science has the potential to assist in orchid conservation, as well as further the scientific kno wledge of orchids. The Orchid Seed Orchid seeds are exceedingly diverse in shape, size, and pattern. They are microscopic and range from 0.05 to 6 mm in length, 0.01 to 0.93 mm in width, and weigh 0.3 to 14 g (Arditti, 1967; Arditti and Ghani, 2000). Seed capsules may hold anywhere between 1,300 to 4 million seeds (Arditti, 1967). Shapes are also various including filiform, fusiform, clavate, and ellipsoidal seeds (Molvray and Kores, 1995). Orchid seeds share a common characteristic of a reduced embryo and the absence of endosperm or a cotyledon (Prutsch et al. 2000) with the exception of Sobralia and Bletilla seeds
17 that have a rudimentary cotyledon (Arditti, 196 7). Various surface depressions and patterns in the testa increase air resistance and allow seeds to remain airand water-borne for long periods (Arditti and Ghani, 2000; Prutsch et al ., 2000). The testa is normally derived from the outer integument, but as in the case of Paphiopedilum delenatii the testa is derived from both the inner and outer integument (Molvray and Kores, 1995; Lee et al. 2006). In most species the testa is usually one cell thick, but made up of 20 to 600 cells (Molvray and Kores, 1995; Prutsch et al ., 2000). The embryo is attached to the te sta by several cells, contains dense cytoplasm, and is made up of as few as ten cells (Stoutamire, 1964). At early globular stages, plastids with starch are visible, but soon disappear during the mature globular stage (Lee et al ., 2006). At the mature globular stage, starch is repla ced by lipid and protein bodies (Lee et al ., 2006). Cuticular substances appear in the surface wall cells of the embryo during the early globular stage, but are not found in the suspensor region (Lee et al ., 2006). The suspensor serves as a channel for free movement of nutrients and water as well as a food storage site for the embryo (Yeung et al., 1996). The two-cell thick inner integument dehydr ates and compresses around the embryo at full maturity (Lee et al., 2005). A layer inside th e inner integument becomes cutinized and a layer outside the inner integument b ecomes lignified at seed matur ity (Yamazaki and Miyoshi, 2006). The lignification and cutinization process strengthens the carapace, which restricts embryo growth through mechanical or chemical means (Yamazaki and Miyoshi, 2006). Orchid seeds contain an undifferentiated embryo lacking enzymes to metabolize polysaccharides (Manning and van Staden, 1987; Molvray and Kores, 1995). Sugars present in orchid embryos include sucrose, fructose, maltose rhamnose, and glucose, but these are either utilized fully prior to germina tion or are present in insufficient quantities to support and sustain
18 germination (Manning and van Staden, 1987). A lthough lipids are used as a major nutrient source, embryos lack enzymes to convert lipids to soluble sugars (Manning and van Staden, 1987). Given that orchid seeds can not metaboliz e polysaccharides and lipids, they utilize a mycorrhizal relationship with compatible fungi (i.e. mycobiont) during germination and early development (Rasmussen et al., 1990b). This mycorrhizal relati onship is critical during germination and seedling deve lopment (Zettler, 1997). Followi ng penetration, embryos digest the fungi providing water, carbohydrates, mi nerals, and vitamins (Rasmussen, 1992; Yoder et al. 2000). The digestion of the mycorrhizal fungi in turn stimulates glucose and enzyme production, reserve mobilization, and postgermination nutrient support (Manning and van Staden, 1987). Orchid Seed Germination History and Background Interest in orchid seed germination bega n in the 1800s. Early attempts to initiate germination involved sowing seed s onto organic substances such as sphagnum moss, bark, or leaf mold, but this often proved unsuccessful (Arditti, 1967). Growers also attempted to germinate seeds at the base of potted wild-collect ed mother plants. Bernard and Burgeff were the first to recognize the role of f ungi in orchid seed germination a nd co-cultured fungi with orchid seeds (Bernard, 1899; Burgeff, 1909). They experi mented with symbiotic seed germination, the co-culture of fungi with orchid seeds. Although seeds did not germinate readily, they concluded that orchid seeds could only germinate in vitro in the presence of an appropriate mycorrhizal fungus (Knudson, 1922). Based on initial experiments by Bernard and Burgeff, Lewis Knudson further examined orchid seed germination. Using hi s knowledge and interest in the effects of sugars on plants and enzyme production in fungi, Knudson recognized th at mycorrhizal fungi hydrolyzed starch and cellulose into usable simple sugars (Arditti, 1967). Using a nutrient so lution supplemented with
19 1% sucrose, Knudson (1922) successfully germinated Cattleya seeds. From these initial experiments Knudson demonstrated asymbiotic in vitro germination, and developed Knudson Solution B with 2% sucrose or glucose. Knudson eventually improved Solution B and developed Solution C, which is widely used as an asymbiotic germination medium (Knudson, 1946). Asymbiotic germination represents an ideal system for studying the growth and development of orchid seeds and seedlings. While asymbiotic germination is often the more popular in vitro technique, symbiotic seed germina tion has recently gained popularity for conservation and restoration projects. Factors su ch as photoperiod, temperature, culture media, and seed dormancy may influence both as ymbiotic and symbiotic germination. Symbiotic Orchid Seed Germination The Orchidaceae have evolved a unique relatio nship with mycorrhizal fungi (mycobionts). Bernard in 1899 first recognized the role of mycobionts in orchid seed germination, and attempted to co-culture orch id seeds with mycobionts (Had ley, 1982). Burgeff, who was a contemporary to Bernard, not only studied th e relationship between mycobionts and orchid seeds, but also isolated and identified ma ny fungi in culture (Hadley, 1982). Although early attempts at symbiotic germination were often not successful, Bernard and Burgeff created the foundation to further study orchid seed germination. The focus of symbiotic seed germination research changed throughout the 20th century. After Knudson (1922) discovered asymbiotic germination, little research on symbiotic germination continued until the problematic relati onship of mycobiont specificity was examined (Curtis, 1939), and a more efficient symbiotic germination technique wa s developed (Downie, 1940). The nutritional requirements of mycobionts a nd the movement of nut rients between seeds and mycobionts were studied thoroughly in th e 1960s and 1970s (Smith, 1966, 1967; Hijner and
20 Arditti, 1973; Hadley and Purves, 1974; Blakeman et al., 1976; Hadley and Ong, 1978). In the late 1980s, symbiotic techniques were again refined (Clements et al., 1985; Dixon, 1987). Until the 1990s, the environmental cues that affected symbiotic seed germination were not studied as intensively as asymbiotic germin ation. Photoperiod and temperature affects on symbiotic seed germination were studied in depth (Rasmussen et al. 1990a b; Rasmussen and Rasmussen, 1991; Rasmussen, 1992; Zettler and McInnis, 1992, 1994; Zettler and Hofer, 1997, 1998). Recently, the focus has on ce again turned to refining sy mbiotic techniques to increase seedling acclimatization survival fo r reintroduction pur poses (Brundrett et al. 2003; Batty et al. 2006a; Scade et al., 2006). The role of fungal specificity has once again been revisited (Otero et al. 2004, 2005; Stewart and Kane, 2006 b) as well as molecular identification and mycobiont diversity (Taylor et al. 2003; McCormick et al., 2004; Shefferson et al. 2005, 2007). The mutualistic/parasitic relationship between orchids and mycobionts has also been recently studied (Cameron et al., 2006, 2007, 2008 Shimura et al. 2007). An important study involved the propag ation of the federally endangered, Platanthera holochila from Hawaii (Zettler et al. 2005a). The mycobiont isolated from Hawaiian plants did not support germination; however, the mycobiont isolated from Spiranthes brevilabris from Florida supported germination ra ising concerns of the ecological importance of local adaptation. Introducing plants harboring the Florida my cobiont may adversely affect not only the P. holochila population, but also the entire ecosystem. Di fferent strains within a fungal species may damage isolated ecosystems or other rare and endangered plants (Zettler et al ., 2005a ). For this reason, considering differences in mycobionts not only among but also within species is crucial. Asymbiotic Orchid Seed Germination Since Knudson (1922) showed that or chid seeds could be germinated in vitro without mycobionts, much information has been publis hed on asymbiotic germination. Asymbiotic
21 germination is used more often than symbiotic techniques since asymbiotic methods are generally less complicated. Asymbiotic methods ar e ideal means to study physiological effects of photoperiod, temperature, and mineral nutrition on germination and subsequent development. Replicating the environmental conditions found in situ under in vitro conditions may lead to further insight into how environm ental cues trigger germination. A major difference between symbiotic and as ymbiotic techniques is the germination medium. Because mycobionts are not utilized to provide nutrients to developing embryos, asymbiotic germination media is often suppl emented with various sources and types of carbohydrates, nitrogen, vitamins, and undefined or ganic additives. Since Knudson demonstrated the feasibility of asymbiotic germination, the role of mineral nutrition ha s been researched extensively. Many different culture media have been developed since Knudsons original formula was published. Although many of these media have only minor differences in composition, growth and development of species may be significantly affected. More recently the role of individual media components have not been as extensively i nvestigated, but rather commercially prepared media are often used to conduct screens to obtain satisfactory germination. Such studies also focused on ch aracterizing the growth and development of embryos and seedlings to more precise ly examine growth and development. Nitrogen has long been considered an important component in the germination of orchid seeds. Recent reports have shown that differe nt media may better suppo rt initial germination compared to advanced seedling development, and vice versa. The higher seed germination percentages of Encyclia boothiana var. erythronioides and Calopogon tuberosus on Knudson C. were attributed to high ammonium content, which can be utilized by seeds during early germination and development (Stenberg and Kane, 1998; Kauth et al. 2006). Seedling fresh
22 weight of Cattleya and Cymbidium hybrids was greater when grown on a medium with a high ratio of ammonium to nitrate (Spoerl and Curtis, 1948). While Knudson C also promoted seedling development of E boothiana (Stenberg and Kane, 1998), C tuberosus seedlings developed to more adva nced stages on P723 Orchid Seed Sowing Medium ( Phyto Technology Laboratories, Inc., Shawnee Mission, KS) (Kauth et al. 2006). The limited development of C tuberosus on Knudson C was attributed to a high nitrate concentration and the inability of the embryos to utilize nitrates dur ing early growth and development (Raghavan and Torrey, 1964). Peptone, an organic nitrogen source present in P723, may have contributed to the increased seedli ng development by supplying auxin-like compounds or various amino acids (Curtis, 1947; Kauth et al. 2006). However peptone responses may be species specific. Seed ge rmination percentages of Paphiopedilum insigne and P. hirsutissimum were approximately 30% higher with peptone th an without peptone, and seedlings were more uniform with peptone (Curtis, 1947). Adversely, seed germination of Phaius grandiflorus and Platanthera clavellata was hindered in the presence of peptone. Although ammonium was found beneficial in asymbiotic germination of E boothiana var. erythronioides and C tuberosus, seed germination of other terre strial orchids may be inhibited by it. Germination and growth of Dactylorhiza incarnata seeds, a European terrestrial orchid, were reduced in the presence of ammonium (Dijk and Eck, 1995 b). As nitrogen concentration increased, embryo weight decreased in two species of Dactylorhiza (Dijk and Eck, 1995 a). Likewise, a high ratio of ammonium to nitrate reduced the germination of Vanda tricolor (Curtis and Spoerl, 1948). Amino acids have also been used as a s ubstitute nitrogen source. Raghavan (1964) reported that only certain amino acids increase seed germination of Cattleya Glycine, the
23 simplest amino acid, decreased overall germination of Cattleya seeds from 53% to 41%. However, germination in the presence of argini ne, proline, and glutamine was similar to that with ammonium nitrate (Raghavan, 1964). Spoerl and Curtis (1948) also reported that glycine significantly reduced germination of Cattleya seeds after 2 months when compared with other amino acids. However, after 5 months germinatio n in the presence of glycine increased from 22.5% to 64%. Amino acid enzyme systems within developing embryos change over time. Amino acids may not be available as initial ni trogen sources, but may be metabolized after a certain period of time (Spoerl and Curtis, 1948). Various orchid species respond differently to various amino acids during germination, and ther efore further investigat ion should be carried out. Since not all amino acids are beneficial for seed germination, combinations of amino acids may increase germination (Spoerl and Curtis, 1948). Edamin, a lactalbumin hydrolysate with pept ides and 18 amino acids, increased the germination of a Cattleya Laelia hybrid (Ziegler et al. 1967). Embryos became green faster and seedling dry weight was gr eater with Edamin than seed lings cultured without Edamin. Tissue analysis of seedlings cultured on Edamin yi elded showed increased levels of glutamine, asparagine, and gamma amino butyric acid. Although these amino acids were not found in Edamin, complex organic nitrogen sources, such as Edamin, may be important for the synthesis of amino acids (Ziegler et al. 1967). Nitrogen in the form of amino acids may be mo re readily available to germinating seeds or developing embryos than inorganic n itrogen (van Waes and Debergh, 1986 a; Malmgren, 1993; Anderson, 1996; Malmgren, 1996; Stewart and Kane, 2006 a). Majerowicz et al. (2000) reported increased growth of Catasetum fimbriatum seedlings in the presence of glutamine compared to media containing ammonium or n itrate. Stewart and Kane (2006 a) reported improved
24 germination and subsequent development of Habenaria macroceratitis on Malmgren Modified Terrestrial Orchid Medium, which contains glycine as the sole nitrogen source. When inorganic nitrogen, such as ammonium, is utilized by germinat ing seeds, the nitrogen is converted to amino acids (Majerowicz et al. 2000). Using amino acids as the sole nitrogen source may lead to more efficient nitrogen assimilation by avoidi ng certain nitrogen conversion steps. Since orchid seeds have minimal carbohydr ate reserves, an e xogenous source of carbohydrates is required for in vitro orchid seed germination. Two sources of carbohydrates are available to the germinating embryo during the fi rst stages of developm ent in nature: minimal carbohydrates in the embryo, and those obtaine d from mycobionts (Rasmussen, 1995). Some orchid seeds contain glucoprot eins that may release glucos e upon hydrolyzation, explaining why some orchid seeds germinate in water (Rasmussen, 1995). The role of carbohydrates in asymbiotic germin ation has often been contradictory. In an early study on the germination of Cattleya seeds using several carbohyd rate sources, d-mannose supported the highest germination percentage, whil e pentose sugars such as xylose supported no germination (Wynd, 1933). Conversely, Ernst (1967) reported that xylose proved effective in supporting germination and development of Phalaenopsis seeds. Both concl uded that fructose did allow for moderate to exceptional germinatio n. Several sugars are translocated to embryos from their mycobiont including trehalose, gluc ose, and mannitol (Smith, 1973). Trehalose was found to be suitable for several species, but to lesser extents than other sources (Ernst, 1967; Ernst et al. 1971; Smith, 1973). The role of carbohydrates is also crucial for seedling development. Phalaenopsis seedlings developed best on media with glucose, fructose and oligosaccharides containing glucose or fructose (Ernst et al. 1971). However, polysaccharides proved unsuccessful in germinating and
25 sustaining seedlings. The benefits of using glucos e as the lone carbohydrate source were recently reported (Wotavov-Novotn et al. 2007). Using lower concentra tions of sucrose promotes shoot development, while higher concentrations promote root gr owth (Yates and Curtis, 1949). Ernst and Arditti (1990) reported that Phalaenopsis seedlings developed in the presence of many carbohydrate sources including glucose, a si mple sugar, and maltoheptaose, a long chain sugar. Germination percentage and seedling de velopment was highest on glucose, with fewer seeds germinating on maltooligosaccharides. Although embryos did not develop further without sugar or at least a low concentration, endoge nous carbohydrates must have been present to support early germination and de velopment. After 6 months culture, seedlings cultured on glucose had higher fresh weights and surviv al than seedlings cultured on long-chain carbohydrates. The lower fresh weight of Phalaenopsis seedlings cultured with long-chain carbohydrates may be caused by insufficient enzy mes responsible for breaking bonds in these carbohydrates (Ernst and Arditti, 1990). Factors Effecting Orchid Seed Germination Photoperiod Although the effects of photoperiod have been widely researched, the results are often contradictory. Incubating terres trial orchid seeds in complete darkness is often recommended while light incubation is epiphytic seeds is recommended for epiphytic seeds. Several explanations have been offered regarding this relationship. Upon dehiscence, seeds of terrestrial orchids may not germinate until buried (Ras mussen and Rasmussen, 1991). Many terrestrial orchids also grow in more shaded environments than their epiphytic counterparts (Rasmussen, 1995), and light may not penetrate the canopy and reach soil as readily (Rasmussen and Rasmussen, 1991).
26 A small increase in light intensity fr om complete darkness to 1.2 mol m-2 s-1 reduced germination of European terrestrial orchids (van Waes and Debergh, 1986 b). Seed germination of Cypripedium acaule was lower under a16/8 h photoperiod (6.7% germination) compared to complete darkness (96.7%) (St-Arnaud et al., 1992). In addition, all embryos developed leaves in darkness, but only 60% of the embryos in the 16/8 h photoperiod developed leaves (St-Arnaud et al. 1992). Zettler and Hofer (1997) reported a si gnificant decrease in germination when S. odorata seeds were exposed to a brief period of illumination. Germination in complete darkness for 3 weeks was greater than germination of seeds exposed to either 7 days of an 8/16 h or 14/10 h photoperiod, and then placed in darkne ss for 2 weeks. Stewart and Kane (2006 a) reported that light inhibited asymbiotic germination and development of Habenaria macroceratitis Although embryos developed to a leaf-bearing stage in all photoperiod treatm ents, over 90% of the embryos developed leaves in comple te darkness (Stewart and Kane, 2006 a). Embryos cultured in complete darkness often produce more rhizoids than those in light (Stewart and Kane, 2006 a). Rhizoids, which are sites of f ungal infections, may not be produced until seeds/embryos are buried and likely to encounter fungal mycobionts (Rasmussen, 1995). Rhizoid inhibition under light conditions may prevent embryo death by preventing the mobilization of valuable energy reserves prior to encountering conditions of likely mycorrhizal infection (Stewart and Kane, 2006 a). Stoutamire (Stoutamire, 1974) suggested that bog-inhabiting North American terrestrial orchids that are adapted to an open ca nopy are less sensitive to light. Kauth et al. (2006) found evidence for this with seeds of Calopogon tuberosus var. tuberosus a North American terrestrial orchid. Calopogon tuberosus not only inhabits bogs, but also gr ows in areas of full sun such as open prairies and pine flatwoods. Although asymbiotic germination in complete darkness was
27 generally greater than germination in a 16/8 h photoperiod, seedling development was superior in a 16/8 h photoperiod. No embryos developed to an advance leaf-beari ng stage under complete darkness, but over 20% of the embryos on P723 cu lture medium developed to advanced leafbearing stages in the 16/8 h photoperiod (Kauth et al. 2006). Similar results were obtained with asymbiotic germination of Bletia purpurea a terrestrial orchid that grows in prairies and under open canopies in south Florida (Dutra et al., 2008). Germination and subsequent development under long day conditions may be an adaptation to shallow seed burial or germination above the substrate. Several researchers reported that germination increases with brief periods of illumination. Rasmussen et al. (1990 a ) reported 75% germination of Dactylorhiza majalis seeds when illuminated for 10 days prior to dark incubation. This was a significant increase from 45% germination under continual darkne ss. Zettler and McInnis (1994) reported similar results with symbiotic germination of Platanthera integrilabia Germination increased from 20% under complete darkness to 44% when seeds were exposed to 7 days under a 16/8 h photoperiod prior to dark incubation. While the exact function of light pretreatment is not understood, mycorrhizal fungi may benefit from brief periods of illumination (Zettler and McInnis, 1994). While photoperiod has been stud ied extensively in orchid s eed germination, light quality and quantity has been generally neglected. Fukai et al. (1997) examined the role of light quality on asymbiotic seed germination of the hybrid Calanthe Satsuma. After 4 months germination percentage was highest in complete darkness (57.7%) compared to 40.2% and 1.3% germination under red and blue light, respectiv ely. Germination was also low (12.4%) under a combination of red and blue light as well as fluorescent lights (13.2%). Blue light, although inhibitory to germination, promoted a high level of embryo development (Fukai et al. 1997). Blue light has
28 been shown to be important in photomorphogenesis as well as chlorophyll accumulation in nonorchid species (Kamiya et al., 1981). Likewise, red light proved beneficial for asymbiotic seed germination of Goodyera pubescens, while blue and far red inhi bited germination (McKinley and Camper, 1997). Approximately 33% germin ation was reported under red light and fluorescent light, while germination under blue light, UV light, and co mplete darkness was approximately 20%. Rasmussen and Rasmussen (1991) studied the effects of light quality and quantity on symbiotic germination of D majalis Under a low white light intensity of 13 W m-2 (ca. 60 mol m-2 s-1), germination decreased from 20% in co mplete darkness to less than 5% (8/16 h photoperiod) and 0% (16/8 h photop eriod). Green or red light illumination before white light decreased germination to less than 10%. Ho wever, red light followed by dark incubation increased germination to 17%. Red light, which is physiologically active, promotes germination; however, canopies absorb red light. Red light stimulation may be an adaptation for D majalis growing in open areas (Rasmussen and Rasmussen, 1991). The role of phyt ochrome and red/far-red light has not been fully investigated in orchid seeds. Experiments with non-orchid seeds may be useful as models for future orchid seed research regarding phytochrome and light quality. Although only a few published articles exist that examine light quality on orchid se ed germination, more research is required on more species in order to find a de finitive role of phytochr ome and light quality. Temperature The effects of temperature on orchid seed ge rmination and development have been largely ignored compared to the effects of photoperiod an d germination media. Temperatures are often selected with no justification or reference to t hose found in nature. Temperature is a major factor responsible for the onset and breaking of physiological seed dormancy (Baskin and Baskin,
29 2004). The lack of understanding regarding orchid seed germination and temperature may simply be due to many studies focusing on refining methods of existing germination protocols, as well as understanding the nutrient requirements of symbiotic a nd asymbiotic germination. Several studies that provide insight into the relationship between orchid seed germination and temperature are available. As with many ot her species, orchid seeds germinate within a range of temperatures, but maximum germinatio n is achieved only in a narrow range. Although Dactylorhiza majalis seeds germinated between 10 and 30 C, germination decreased below 15C and above 27C while the optimum ra nge was between 23 and 24.5C (Rasmussen et al., 1990a; Rasmussen and Rasmussen, 1991). At higher temp eratures, rhizoid form ation was inhibited, which may be due to the lack of mycorrhizal colonization. Since rhizoids ar e the primary site of mycorrhizal infection, the lack of rhizoids may cause reduced my corrhizal infections (Rasmussen et al. 1990b ). Several thermo-inductive treatments are effec tive at breaking dormancy in mature orchid seeds including cold-stra tification. The use of co ld-stratification to break dormancy in orchid seeds is often used for difficultto-germinate genera such as Cypripedium Epipactis and Dactylorhiza (Rasmussen, 1995). However, there is limited information on the exact mechanism by which cold-stratification promotes orchid seed germination. Cold temperatures may decrease enzymatic reactions, slow metabo lic processes, or change enzyme production and concentration, and thus promoting germination (Bewley and Black, 1994). However, variable results have been reported not only between species, but also within the same species. Ballard (1990) reported a maximum germination in Cypripedium calceolus of 16% after 4 months of cold-stratification at 5C, while Coke (1990) reported 50% germination after 5 months cold-stratification. After 160 days incubation, germination of C calceolus increased to
30 over 90% after cold-stratification at 5C for 8 weeks (Chu and Mudge, 1994). Pretreatment of C calceolus seeds at 6C for 8 weeks reduced germination to 0.8% when incubated (van Waes and Debergh, 1986b). De Pauw and Remphrey (1993) reported higher germination for C. candidum seeds after two months. Germination of Dactylorhiza lapponica increased over 80% after three months chill at 3-4C (ien et al. 2008). Different capsule ripeni ng conditions and seed age may have caused the different result s. van Waes and Debergh (1986 b) used fully mature seeds collected from dehisced capsules, while Chu and Mudge (1994) used non-dehisced mature seeds. Dehisced seeds may need a longer period of co ld-stratificatio n than van Waes and Debergh (1986b) provided. Since Chu and Mudge (1994) cultu red seeds in complete darkness while van Waes and Debergh (1986 b) cultured seeds under a 14/10 h photoperiod, differences in germination may be attributed to other culture conditions. The length of cold-stratification (chilling) is al so an important factor to consider, and may be species specific. Rasmussen (1992), Tomita and Tomita (1997), and Miyoshi and Mii (1998) reported higher germination percentages when seeds of Cypripedium macranthos C candidum and Epipactis palustris respectively, were cold-stratif ied for 8 to 12 weeks. Zettler et al. (2001) found that germination percentage of Platanthera leucophaea increased after two coldstratifications for 11 months as well as 107 da ys at 6C following 95 days at 23C. Sharma et al. (2003) reported a higher ge rmination percentage of Platanthera praeclara after 6 months of cold-stratification compared to 0 and 4 months. Shimura and Koda (2005) reported the importance of fungal inoculation corresponding to cold -stratification on symbiotic germination of C macranthos A higher germination percentage wa s reported when seed cultures were inoculated with fungi after a 12 week cold-stratification compared to inoculation before or
31 several weeks after the co ld-stratification. This might suggest that fungal infection in nature takes place after winter and prior to germina tion in early spring (Shimura and Koda, 2005). Cold-stratification in orchid seeds has severa l ecological functions a nd effects. If seed dispersal occurs in fall, seed germination may be delayed until the next growing season when conditions are more favorable for growth and de velopment. A low temperature requirement for long periods of time may prevent seeds from germinating immediately after dispersal (Rasmussen, 1995). The effects of chilling and th awing may cause degradation of the testa, which could lead to leaching of germinati on inhibitors, imbibiti on, and fungal infection (Rasmussen, 1995). Chilling also promotes the grow th of rhizoids, which are important for the uptake of water and nutrients as well as es tablishing the mycorrhizal fungal relationship (Rasmussen, 1992). Ecotypes Introduction The term ecotype was first defined by Turesson (1922 b) in describing alpine plant populations. Ecotypes were defined as ecological sub-units of a sp ecies resulting from genotypic responses to a particular habitat (Turesson, 1922 a). The term ecotype wa s preceded by several other terms that were established in debating the species concept. The coenospecies was considered the complete combinati on of genotypes within a complex, and ecospecies the genotypic sub-units of a co enospecies (Turesson, 1922 b). Gregor et al. (1936) stated that coenospecies and ecospecies were distinguished by morphological, physiologi cal, or cytological characters, while cytological co mponents were not characteristic of the ecotype. In addition, ecotypes could not interbreed with other ecotypes of the ecospecies (Gregor et al. 1936). Clausen et al. (1939) defined the coenospecies as a sp ecies-complex incapable of interbreeding. The ecospecies was considered a distinct species that may pr oduce hybrid generations, and the
32 ecotype a subspecies of the ecospecies that could pr oduce viable hybrids (Clausen et al. 1939). Gregor (1939) further defined co enospecies as populations that were unable to exchange genes and ecospecies as populations with low gene exchange with other populati ons of a coenospecies. Gregor (1939) also introduced the term cline as any gradation in measurable characters, and ecocline as a cline correlated with an ecological gradient. In shor t, an ecotype was defined as a particular range on an ecocline (Gregor, 1939). Daehler et al. (1999) defined ecotypes as subpopulations with genetic differences in morphology, physiology, and life history. However, this definition did not recognize the role of e nvironment in ecotypic differentiation. Hufford and Mazer (2003) stated that ecotypes are distinct genotypes or populations within a species, resulting from adaptation to loca l environmental conditions; capable of interbreeding with other ecotypes of the same species. This is a comp lete definition since th e role of genetics and environmental conditions are recognized. Importance of Ecotypes to Plant Conservation A major concern in conservation and reintr oduction programs is the source of plant material. Many species are adapted to local environmental conditions and selection pressures, thus the use of locally adapted material is essential. However, using local ecotypes for restorations is expensive and difficult to mainta in and verify because collecting plant material directly from native habitats is often unregulated and guidelines vary (Smith et al. 2007). Debate exists concerning collecting plant material from single or multiple sources for use in restoration projects. Sanders and McGraw (2005) found that populations of Hydrastis canadensis established more readily from single sources at the restoratio n site, but recommended using multiple sources over multiple sites to increase compatibility to source-sites. Regardless, introducing non-local plant material to naturally occurring popul ations might reduce fitness, fecundity, and ecosystem functions (Linha rt, 1995; Hufford and Mazer, 2003; McKay et al.,
33 2005). These factors may be affected by heterosi s, outbreeding, inbreeding, founder effects, and genetic swamping (Hufford and Mazer, 2003; McKay et al. 2005). Numerous studies have shown a home-site advantage for local populations possibly due to habitat-specific selection pressures influenced by genetic adaptation to environmental conditions (van Tienderen, 1992; Kindell et al. 1996; Nagy and Rice, 1997; Keller and Kollmann, 1999; Bischoff et al., 2006). Finding the correct balance be tween inbreeding and outbreedi ng depression of introduced populations is difficult yet essential (McKay et al. 2005). Inbreeding depression occurs when populations become relatively small increasing the loss of genetic diversity (McKay et al. 2005). Inbreeding depression may be the result of f ounder effects as reported for the Mauna Kea silversword. A restored population of the silv ersword was generated from two founders and a loss of genetic diversity resulted (Robichaux et al. 1997; Friar et al., 2000). The adverse consequences of inbreeding depression may be overcome by careful plant material collections (Williams, 2001). Outbreeding depression often results from non-locally adapted populations producing viable hybrids (McKay et al. 2005). These hybrids, in turn, ca n express lower fitness levels by increasing the numbers of unfit offspr ing (Hufford and Mazer, 2003; McKay et al. 2005). Conversely, heterosis or hybrid vigor may result from outbreedi ng depression. This adversely affects restorations since non-local vigorous hybrids may cause genetic swamping and loss of diversity in native populations (McKay et al. 2005). Ecotypic differentiation may also be a conse quence of primary productivity. Ecotypes from colder climates may not respond to increased temp eratures as efficiently as southern ecotypes when transplanted to warmer climates (Shaver et al. 1986). Leaf senescence of Eriophorum vaginatum ecotypes from northern latitudes occurre d earlier than southern ecotypes when
34 transplanted to warmer climates. These ecoty pes represented a genetic limitation on primary productivity, which could adversely a ffect restoration projects (Shaver et al. 1986; Fetcher and Shaver, 1990). Ecotype Development and Differentiation The evolution and selection of ecotypes occurs over time and space, but development is not limited to single biotic or abiotic pressure s (Linhart and Grant, 1996). Ecotypic development occurs from differences in many environmenta l characteristics such as photoperiod (Howe et al., 1995; Kurepin et al. 2007), temperature (Campbell and Sorensen, 1973; Downs and Bevington, 1981; Li et al. 2005), altitude or elevation (Turesson, 1922 a; Clausen et al. 1941; Li et al., 2005), water availability (Peas-Fronteras et al. 2009), and soil nutrient availability (Grze 2007; Macel et al., 2007; Sambatti and Rice, 2007). Although generally more difficult to detect, biotic pressures such as pollination vector s (Robertson and Wyatt, 1990; Johnson, 1997) and grazing or herbivory (Suzuki, 2008) also influence ecotypic differentiation. Many of these pressures can be selected for or differentiated experimentally. Common garden studies Local adaptation has been studi ed in numerous species incl uding both plants and animals (Nuismer and Gandon, 2008). Common garden and tran splant studies are often used to detect local adaptation. Common garden st udies test local adaptation a nd fitness of individuals from local or distant habitats in a common environm ent. These studies minimize the environmental impacts on fitness and may better identify the ro le of genetics. Transplant studies may better estimate the role of environmen tal variation on fitness since in dividuals are transplanted to foreign habitat (Nuismer and Gandon, 2008). Collecti ng plant material from natural populations and planting under controlled conditions often excludes maternal effects (Volis et al. 2002).
35 Several techniques exist for c onducting common garden and transplant studies. Controlled environments include greenhouses, growth chambe rs, field sites, and out door plots (Gallagher et al. 1988; Howe et al., 1995; Majerowicz et al. 2000; Suzuki, 2008). Mature or juvenile plant material including seeds can be collected and gr own. Also, using seeds from the F1 generation for transplant studies can reveal natural selection strategies (Volis et al. 2002). When using controlled environments, the na tural conditions such as photoperiod and temperature found in situ can be replicated. Photoperiod is essential for the induction of dormancy and flowering (Howe et al. 1995; Kurepin et al. 2007). The critical photoperiod that induces flowering, germination, or dormancy in ecotypes is often a main focus. Critical photoperiodic responses in plants ar e classified as long-day requir ing long days and short nights, short-day requiring short days and long nights, and day-neutral plants (Kurepin et al. 2007). Under growth chamber conditions, the critical photoperiod can be tested by controlling the photoperiod. Northern ecotypes of many tree specie s often require longer cr itical photoperiods to maintain growth (Downs and Bevingt on, 1981) and initiate bud set (Howe et al. 1995). The ability of plants to flower under sun or shad e can also be analyzed. Alpine ecotypes of Stellaria longipes, growing in full-sun, were clas sified as day-neutral while prairie ecotypes, growing in shaded habitats, required long days to flower (Kurepin et al. 2007). When using field or outdoor plots in a common garden study, environmental conditions can not be controlled as they are in greenhouses or growth chambers. Field plot experiments are useful to study population varia tion. Ecotypic responses to grazing or herbivory can be studied efficiently. Dwarf ecotypes of Persicaria longiseta and Spartina alterniflora remained in their dwarf form regardless of grazing or fertilizing nutrients, respectively (Daehler et al. 1999; Suzuki, 2008). The dwarf form of P. longiseta and S. alterniflora were found to be more
36 influenced by genetics than environmental c onditions. Soil type and nutrient availability can influence ecotypic differentiation (Grze 2007; Sambatti and Rice, 2007). Soil was collected from two California sites where Helianthus exilis ecotypes were found, and used in a common garden study (Sambatti and Rice, 2007). Although gene flow did occur between populations, local adaptation to soil was detected. Ecotypes from riparian area were not adapted to water deficits in serpentine soils, and extending gr owth for longer periods led to higher mortality (Sambatti and Rice, 2007). While common garden studies can detect ge netic components of ecotypes, reciprocal transplant studies may more efficiently detect environm ental components of ecotypic differentiation. Fang et al. (2006) conducted a reciprocal tr ansplant with dwarf and normalstature pitch pines from Long Is land, New York. Over a six year period, dwarf plants did not retain their dwarf stature when planted at nondwarf sites, and normal stature plants became dwarf when planted at dwarf sites. Plants in dwarf sites also reproduced slowly and had lower survival. Fang et al. (2006) recommended preserving the dw arf-site habitat since the dwarf ecotype could not be preserved ex situ While plants are generally tr ansplanted among sites, soils can also be transplanted to detect soil and climate adaptation. Macel et al. (2007) transplanted Holcus lanatus and Lotus corniculatus and local soils between three sites, and plants were grown in native or foreign soils. Over two years, H. lanatus exhibited home-site advantage with no adaptation to local soil c onditions indicating adaptati on to climate. Conversely, L. corniculatus showed slight adaptation to soil conditions and no adaptation to clim ate. These results show that local adaptation is likely influenced by comple x interactions among climat e, soil nutrients, and soil biota (Macel et al., 2007).
37 Seed germination ecology When mature plants are not available for co mmon garden or transpla nt studies, studying seed germination ecology can be used to differe ntiate ecotypes. Optimal germination conditions often differ among seed lots, populations, or sources, and sampling from one population or populations within the same vicinity may not sufficiently determine optimum conditions for germination (Quinn and Colosi, 1977). However, using seeds collected across a population may provide a representation of population adaptation. To determine whether optimum conditions for germination, seed from several popula tions should be collected (Nelson et al. 1970; Seneca, 1974). However, controversy exists whether seeds s hould be germinated directly from plants in nature or plants grown under uniform conditio ns for several generations (Quinn and Colosi, 1977). Using wild-collected seeds may not be a ppropriate for germination ecology studies because they may have environmental preconditioni ng that masks the effects of genetics (Nelson et al. 1970). Growing plants under uniform conditi ons and then germinating second generation seeds can successfully determine whether germination ecotypes exist (Nelson et al., 1970; Baskin and Baskin, 1973). However, limiting seed germination to non-wild grown seed ignores all adaptations and ecolo gical influences are diminished (Quinn and Colosi, 1977). Baskin and Baskin (1973) suggested growing plants under un iform conditions for several generations before genetic differences can be allo cated. Growing plants for several and not just one generation under uniform conditions may remove previous e nvironmental influences (Baskin and Baskin, 1973; Linhart, 1996). Growing plants for one ge neration under uniform conditions and then germinating seeds may be an appropriate measure (Nelson et al. 1970; Quinn and Colosi, 1977). Germination between first and second generation seeds should be comparatively studied (Baskin and Baskin, 1973; Quinn and Colosi, 1977).
38 Regardless of the argument concerning whether first or second seed generations should be used for germination ecology studies, these studie s are powerful ways to differentiate ecotypes. Three sea oats eco-regions were identified through ge rmination studies (Seneca, 1972). The first group included Virginia and North Carolina populations whose seeds required chilling and seedlings expressed intermediate vegetative gr owth. The second group included Atlantic Florida populations that did not require seed chilling and had low cap acity for vegetative growth. The third region was the Gulf coast whose seeds ha d a chilling response and high potential for vegetative growth (Seneca, 1972). Probert et al. (1985 b) found that northern European Dactylis glomerata ecotypes required both light and altern ating temperatures for germination, while Mediterranean ecotypes germinated in continua l darkness and constant temperatures. More northern ecotypes may have a deeper dormancy condition and therefore, have a more narrow range of germination conditions (Seneca, 1972; Probert et al. 1985a). Meyer (1992) and Mondoni et al. (2008) correlated habitat a nd local adaptation to ecotypes of Penstemom eatonii and Anemone nemerosa respectively. Penstemon eatonii ecotypes from sites with colder winters required longer chill pe riods before germination began, and were slower to germinate than seeds from sites with warm er winters (Meyer, 1992). The cold stratification requirement for A. nemorosa germination was expressed more in mountain ecotypes, and radicles emerged one month earlier at lower temperatures compared to lowland ecotypes (Mondoni et al. 2008). Biomass allocation Differences in biomass allocation and storage organ formation are strong indicators of ecotypic differentiation. The differential develo pment may be influenced by growing season length. In Sagittaria latifolia Rhode Island ecotypes formed corms two months before South Carolina ecotypes when grown in a common garden in Florida (Kane et al. 2000). The
39 differences in timing of corm formation was lik ely influenced by a shorter growing season in Rhode Island, and quicker corm formati on that favored winter survival (Kane et al., 2000). Biomass accumulation has also been correlate d with differences in reproductive strategy. While many instances of differences in bioma ss allocation were due to genetics, others are plastic responses to environmental cues (Abr ahamson, 1979). Larger allocation to underground storage organs promoted vegetative or clonal re production, while allocation to flower and seed organs promoted sexual repr oduction (Abrahamson, 1975, 1979; Sun et al., 2001; Thompson et al. 2001). In Scirpus mariqueter ecotypes, reproductive strategy switched from lower elevational marsh locations to higher locations (Sun et al., 2001). Higher biomass allocation to corms and rhizomes occurred in lower elevati on ecotypes, thus allowing for more efficient colonization. Also, higher allocation to corms also provided efficient strategies to survive under disturbances such as tidal current or flooding (Sun et al., 2001). Lowland ecotypes of Cyperus rotundus had larger tubers with higher non-struct ural carbohydrate contents (Peas-Fronteras et al. 2009). The allocation of C. rotundus followed similar patterns to S. mariqueter in that lower elevational ecotypes were better able to tolerate flooding and a void starvation and seedling death (Peas-Fronteras et al. 2009) Ecotypes of Rubus hispidus wildflower species, and Spartina anglica responded similarly (Abrahamson, 1975, 1979; Thompson et al. 1991). Higher biomass allocation to seed organs was observed in ecotypes from younger successiona l habitats, while those in mature habitats allocated more biomass to leaves and undergro und organs. This switch in reproductive strategy related to different rates of su ccession. In later successional ha bitats, woodlands species were found to be k-strategists with higher rates of clonal and vegetativ e growth. In earlier successional habitats, such as fields, species were reported to be r-strategists with higher rates of seed
40 production and sexual reproduction (Abrahamson, 1975, 1979; Thompson et al. 1991). Abrahamson (1975) also noted that pollinators were less abundant in mature habitats compared to the earlier successional habitats. In vitro ecology Common garden and transplant studies, while useful, are generally limited to non-rare species since obtaining permits to collect and transplant rare species, such as orchids, is often difficult. Alternatively, seeds can be used to e ither produce mature plan ts for common garden studies or study the germination ecology of ecot ypes. Unfortunately, collecting and germinating orchid seeds in situ is not an easy task. Orchid s often require four or more years to flower from initial seed germination (Stoutamire, 1964), and in situ orchid seed germination is difficult and time consuming since germination is often low (Brundrett et al. 2003; Zettler et al. 2005 b; Diez, 2007). Collecting native soils for germinat ion purposes under green house settings is not efficient either (Tim Johnson, unpublished data). Using in vitro techniques to study orchid seed germination and seedling development is more feasible since orchid seeds germinate in vitro more readily than in situ (Dijk and Eck, 1995b; Kauth et al., 2008). Many in vitro culture techniques can be grouped under the discipline of in vitro ecology. In vitro ecology has been previously defi ned to include exogenous factors (i.e., temperature, light, gas phase, cultu re media, photoautotrophy) that affect in vitro growth and development (Hughes, 1981; Williams, 2007). In vitro ecology can also be used to identify, pr opagate, evaluate, and select plant genotypes and ecotypes for ecological purposes Specifically, environmental and genetic variables that affect plan t growth and development in vitro with ecological factors affecting growth and development in situ can be studied. In vitro ecology could also be used to assess ecotypic differentiation for habi tat restoration and plant reintroduction programs by conducting
41 in vitro common garden studies or reciprocal transp lant studies under controlled environmental conditions. While genotypic and eco typic selection from tissue culture has been previously attempted, the validity of the system must be verified. Also, the ecological strategies of plants have not been thoroughly studied in vitro In vitro conditions are different than those conditions found in situ Conducting common garden studies in vitro may exclude possi ble environmental interactions that can influence gene expr ession. First generation seeds may also have environmental preconditioning that effect in vitro ecology studies (Seneca, 1974). Wetland and marsh plants have been regene rated through micropropaga tion for restoration purposes (Kane, 1996; Rogers, 2003). Li et al. (1995) successfully regene rated callus cultures of Spartina patens and subsequently studied its salt tolerance. Tissue culture generated plants have also been used to block invasive species (Wang et al., 2006). A more controversial technique is to engineer plants in vitro via somaclonal variation, and subs equently selecting genetically superior genotypes (Seliksar and Gallagher, 2000; Wang et al., 2007). While this provides a major breakthrough for producing plants, introducing bi oengineered plants to native habitats may lead to genetically altering the composition of other native plants due to hybridization. A major concern with developing plants th rough micropropagation is the difficulty of acclimatizing plants to ex vitro conditions. Difficult-to-acclima tize sea oats genotypes were found to utilize leaf carbohydrat e reserves less efficiently in vitro, and contained a lower photosynthetic capacity (Valero-Aracama et al., 2006). Adjusting the in vitro environment also provided successful protocols for acclim atizing sea oats (Valero-Aracama et al., 2007). By providing optimum growing conditions for all genotype s, restoration studies can be successful by incorporating numerous local genotypes.
42 Ecotypic Differentiation in Orchids Research on ecotypic differentiation in plan t species is abundant with the exception of orchids. Only a few articles exist on identifying pollinator ecotypes in two orchid species (Robertson and Wyatt, 1990; J ohnson, 1997). Numerous authors that have investigated morphological or genetic diversity in orchids have come short of stating whether ecotypes exist (Dijk and Eck, 1995b; Goldman et al., 2004a; Pillon et al. 2007; Swarts, 2007). Also the issue of seed source has been addressed, but not in de tail regarding ecotypic diffe rentiation. As interest in orchid conservation continues to grow, research on ecotypic differentiation in orchids should be of concern. However, many organizations involv ed with orchid conservation are not informed about ecotypes or are conten t to reintroduce plants wit hout knowing the source. Seed source One purpose of orchid seed germination is to provide plants for speci es-level conservation and reintroduction. However, popula tions of one species may inhabi t different habitats across a geographic range. Differences in habitats may alter the genotypic and/or phenotypic compositions producing distinct ecotypes adapted to local environmental conditions (Hufford and Mazer, 2003). Introducing inap propriate ecotypes into a part icular habitat could not only lead to the death of transplant ed individuals, but loss of geneti c diversity, as well as population degradation. With an increasing interest in orchid -species conservation, care must be taken to use local seed. Zettler and McInnis (1992) reported germination differences between seed sources of Platanthera integrilabia The highest germination percentage and seedling establishment was observed in seeds from th e largest population of P integrilabia, while smaller populations had lower seed germination and seedling establishment. Inbreeding depression in smaller populations could lead to differences in germinability, low viability, or reduced vigo r (Zettler and McInnis,
43 1992). Zettler and Hofer (1998) reported diffe rences in germination among populations of Platanthera clavellata Although seed originatin g from Georgia had lower germination than other sources, seedling development was superior with Georgia seeds. Since P clavellata is an auto-pollinated species, it may be likely that small differences in seed viability or genetic diversity would occur between populations (Zettler and Hofer, 1998). Although habitat conditions were not incorporate d, the size of the populations and apparent isolation may have caused genetic differences in seed germination. Recently the symbiotic germina tion between two populations of Epidendrum nocturnum was examined (Zettler et al. 2007). Seed germination from Fa kahatchee Strand State Preserve plants was 55.7%, but 12.7% from the plants lo cated at the Florida Pa nther National Wildlife Refuge (FPNWR). However, seeds from the FP NWR had a viability of 79.7% compared to a viability of 72.6% for Fakahatchee seeds. Althou gh seed handling and age may have contributed to these differences (Zettler et al., 2007), the self-pollinating breeding system may have also contributed to the germination and viability differences. Dijk and Eck (1995b) investigated the role of in vitro seedling mineral nutrition between coastal and inland populations of Dactylorhiza incarnata in the Netherlands. Major differences in seed germination responses to nitrogen type and population location were noted. Seedlings from coastal areas grew faster in vitro and were more tolerant of exogenous ammonium and nitrate, while the inland seedlings were more sensitive to both ammonium and nitrate. However, seedlings from both populations were more se nsitive to high concentrations of exogenous nitrogen. Since the coastal seedlings developed quickly, they were al so able to assimilate nitrate more efficiently. Both populations inhabit calc areous areas where high nutrient levels are found due to the introduction of fertilizers and poor dr ainage. These soil conditions have led to
44 decreased D incarnata plant numbers. Increased nitrogen mineralization inland may have caused increased nitrogen sensitivity of these plants. Although Dijk and Eck (1995 b) were uncertain whether habitat influenced developmenta l differences, habitat differences seem to have influenced the ecotype differentiation as s hown by the observed differences in seedling development. Pollination ecotypes The only reports of ecotypes in the Orch idaceae have identified pollinator ecotypes. Ecotypic development influenced by pollinators may be due to colonization of the species into new habitats where more effective or differe nt pollinators exist (Robertson and Wyatt, 1990; Johnson and Steiner, 1997). The s hort-spurred mountain ecotypes of Platanthera ciliaris from South Carolina were polli nated by the short-tongued Papilio troilus while the long-spurred coastal ecotype was polli nated by the long-tongued P. palamedes (Robertson and Wyatt, 1990). Johnson and Steiner (1997) reported that Disa draconis from South African mountain sites was pollinated by horseflies while plants in sandplai ns were pollinated by tanglewing flies. Johnson (1997) also reported that shor t-spurred coastal ecotypes of Satyrium hallackii were pollinated by carpenter bees while long-spurre d ecotypes from grasslands we re pollinated by hawkmoths. The differentiation of ecotypes was likely due to pollinator availability. Robertson and Wyatt (1990) reported that P. troilus was scarce in coastal areas, but P. palamedes was absent from mountain populations. Likewise carpenter bees were rare in South African grasslands and hawkmoths were scarce in coas tal habitats (Johnson, 1997). Johns on and Steiner (1997) noted that both shortand long-spurred plants in m ountainous regions in South Africa were pollinated by short-tongued horseflies. They concluded that climate did not in fluence spur length since both ecotypes inhabit similar montane environments. Selection favored long-spurred plants in the sandplains since longer spurs ensured pollinaria attachment. Artificially shortening long-spurred
45 ecotypes did not decrease pollinaria removal, but did decrease pollination levels (Johnson and Steiner, 1997). Plant of Study Calopogon tuberosus (L) Britton, Sterns, and Poggenberg is a widespread terrestrial orchid of eastern North America. Calopogon tuberosus is distributed from Newfoundland, Canada to Cuba, and west to Texas (Luer, 1972). Calopogon tuberosus is an unusual orchid species given that it occupies di verse habitats including alkaline prairies, pine flatwoods, mesic roadsides, fens, and sphagnum bogs (Luer, 1972). The generic name Calopogon, meaning beautiful beard, is derived from the Greek words kalos (beautiful) and pogon (beard). The species name, tuberosus is derived from the Latin word tube rosus, which inaccurately refers to the corms of the species (Luer, 1972). Nonresupinate pink flowers, corms, and grass-like leaves are characteristic of Calopogon species. Two varieties of Calopogon tuberosus are recognized: var. tuberosus and simpsonii (Small) Magrath. Variety tuberosus is 10-75 cm in height with 217 flowers per stem and 1 or 2 leaves (Luer, 1975). Individual flowers with yello w trichomes on the floral lip are 2-3.5 cm in diameter (Brown, 2002). Although pink is the mo st common color, the rarer color form, albiflorus PM Brown, is often scattered among popul ations. The flowering period of var. tuberosus is March in southern stat es to August in northern states and Canada (Luer, 1975). Variety tuberosus grows in full sun and commonly inhabi ts sphagnum bogs and swamps in the north, as well as wet meadows, pine flatwoods and sandy roadsides in the south (Brown, 2002; Luer, 1972). Plants readily form large populations with scattered plants, but occasionally large groups form. In a northern Michigan bog, Case (1987) observed 1,000 plants in less than one square meter.
46 Variety simpsonii is one of the rarest orchids in the United States being re stricted to rocky prairies and marls of 4 countie s in South Florida (Brown, 2002). Plants are larger than var. tuberosus with inflorescences up to 120 cm and 525 flowers (Brown, 2002). This variety also has 4-5 slender leaves, and pink to white trichomes on the white to light pink floral lip. Flowers are commonly pale pink, but a white flower form also exists (forma niveus P.M. Brown). The flowering period is earlier than var. tuberosus with plants initiating flowers in late December and continuing through June (Brown, 2002). Historically two other varie ties have been described, but recent work relegated these varieties to var. tuberosus Variety nanum was first described as Limodorum tuberosum var. nanum in 1913 as small plants with 2-3 purple flowers from Newfoundland, Canada (Nieuwland, 1913). However, this variety was never officially recognized. While var. tuberosus can be 75 cm in height, var. nanum was a maximum of 13 cm in height. St. John first described var. latifolius as Calopogon pulchellus forma latifolius in 1921 from Sable Island, Nova Scotia, Canada. Fernald (1946) elevated the form to varietal level, and Boivin (1967) published the variety as Calopogon tuberosus var. latifolius The plants were originally se parated from var. tuberosus as short plants with wide leaves and larger co rms. The morphological characteristics of var. latifolius are found in plants inhabiting coastal bogs of eastern Maine and the Canadian Maritime (Goldman et al., 2004a). Diversity within Calopogon tuberosus Recently the genetic and morphological diversity within Calopogon tuberosus has been examined. With a large range of distribution, mo rphological and genetic tr aits varied between C tuberosus var. tuberosus C. tuberosus var. simpsonii and C. tuberosus var. latifolius (Goldman et al. 2004a b; Trapnell et al., 2004). Although var. latifolius is currently not recognized as an official variety, Goldman et al. (2004 b) included plants expressing trai ts of this variety in his
47 publications. However, the authors did not state whether ecotypes existed within the C. tuberosus complex. In a morphometric treatment of Calopogon, Goldman et al (2004 a) sampled C. tuberosus from the following areas of the species ra nge: southwest, south east including var. simpsonii and north including var. latifolius The northern range extending into southern Canada was divided from the southern range according to the Wisconsin glaciation, and the eastern range was separated from the western range by th e Mississippi embayment. The morphological separation between southern plants and northern plants was evident by the larger size of southern plants. Based on three tests (Principle compone nt analysis-PCA, average Euclidean distance, average squared Mahalanobis distance) for dive rsity, southern plants were more diverse morphologically than northern plants, with var. simpsonii being more diverse than both the northern and southeastern plants. Although southwes t plants were larger than the rest of the species, morphologically they were similar to s outheast plants. Interestingly, southwest plants were found to be the most diverse in the average Euclidian distance test, but least diverse in the squared Mahalanobis distance test. The reason for this may be due to isolation and unsuitable habitat in the Mississippi embayment. Since var. latifolius did not form a distinct group and was similar to the northern var. tuberosus Goldman et al. (2004 a ) chose not to recognize var. latifolius In an earlier morphological study, Catling and Lucas (1987) relegated var. latifolius to var. tuberosus Plants from Sable Island (var. latifolius ) as well as eastern Ontario (var. tuberosus ) were cultivated under uniform conditions. Both va rieties expressed plasticity year to year. A plant referred to as var. latifolius with leaves 14 mm wide and 10 cm long became more tuberosus-like the second year of cultivation with leaves 9.5 mm wide and 10 cm long. Another
48 plant referred to as var. latifolius had leaves 13 mm wide and 9 cm long the first year and 10 mm wide and 14.5 cm long leaves the second year. Although these plants expressed phenotypic plasticity between years, several plants referred to as var. latifolius became even more extreme with 30 mm wide leaves. Catling and Lucas (1987) reported that var. latifolius was the end of a continuum of variation among the plants of th e northeast, and plants were not discrete individuals in nature. Likewise, si nce the characteristics of var. nanum were observed also throughout the northeastern range, the variety was not officially recognized. Using AFLP analysis, Goldman et al. (2004b) reported that var. latifolius was genetically similar to var. tuberosus and not recognized as a distinct variety. Also ba sed solely on genetic analysis, var. simpsonii formed a coherent group and was recognized as a distinct variety. Although Goldman et al. (2004a) reported high morphologica l diversity of var. simpsonii Trapnell et al. (2004) found that var. simpsonii had low genetic diversity. However, Trapnell et al. (2004) sampled only one population of var. simpsonii Goldman et al. (2004a) noted that var. simpsonii is often self-pollinated due to a short rostellum, possibly contributing to the low genetic diversity among the populati on sampled by Trapnell. Goldman et al. (2004a) sampled large plants of var. simpsonii from Florida and smaller plants of var. simpsonii from the Bahamas, contributing to th e diversity in morphology. Floral Biology of Calopogon tuberosus The biology and flower orientation of Calopogon tuberosus provide a unique pollination mechanism. Orchid flowers ar e normally non-resupinate, but Calopogon flowers are nonresupinate with the lip peta l the uppermost (Robertson, 1887). The gynandrium protrudes from the base of the labellum, and the stigma and anther is found at the apex of the column (Robertson, 1887). Since the flower s are nectarless, pollination occurs by deceit or mimicking other rewarding orchids in close proxim ity (Firmage and Cole, 1988). However, Calopogon
49 flowers may also mimic non-orchids such as Rhexia species (pers. obs.). Yellow trichomes resembling pollen are found at the peak of the labellum (Firmage and Cole, 1988). The flowers and trichomes reflect ultra-violet light, which attracts potential pollinators to the uppermost portion of the labellum (Thien and Marcks, 1972). Upon landing on the hinged labellum, the pollinator is dropped onto the column where the pollin ia attach to the top of the abdomen (Thien and Marcks, 1972; Proctor, 1998). Several bee species have b een reported to pollinate C. tuberosus including Bombus (Thien and Marcks, 1972), halictid bees (Firmage and Cole, 1988), and Xylocopa (Dressler, 1981). However, Thien and Marcks (1972) observed that queen bumblebees and carpenter bees were too large to effectively pollinate flowers becaus e they were strong enough to leave the flower before falling on the pollinia. Both carpe nter and honeybees were observed visiting C. tuberosus flowers at Goethe State Forest, Levy County, Florida; however, neither effectively removed the pollinia. Although the weight of the carpenter bee caused the labellum to collapse onto the column, the bee escaped. The honeybee was too small to collapse the labellum onto the column. Smaller worker bees are likely more efficien t pollinators than queen bumblebees (Thien and Marcks, 1972). Other visitors to Calopogon flowers include beetles, f lies, and small butterflies, but are not likely pollinators (Firma ge and Cole, 1988; Proctor, 1998). Seed Germination of Calopogon tuberosus Conditions for in vitro seed germination of native orchid s have long been debated, and one protocol may not be appropriate for ge rmination of another species (Arditti et al. 1985). While one study might recommend dark incubation, an other study may recommend light incubation. The optimum temperature for seed germination is also variable between studies. Although these parameters are essential for seed germination, germination comparisons between genotypes or populations of native orchids have not been studied. The following studies demonstrate
50 differences between seed germination for C. tuberosus Interestingly, only one study cited the source of seeds (Kauth et al. 2006). Whitlow (1996) outlined a commercial seed culture protocol for C. tuberosus No germination media was cited, although C. tuberosus germinated on various culture media (Whitlow, pers. comm.). Cultures were incu bated under cool-white fluorescent lights (no intensity or photoperiod cited). Af ter development for 5 months, seedlings were placed in cold storage in darkness for 3 months at 4oC, and subsequently placed under fluorescent lights. Although not all seedlings required chilling, s eedlings were programmed on the same developmental schedule after chilling. After 5 months development under fluorescent lights, seedlings were taken out of culture and placed under cold storage for an additional 3 months. Following the cold treatment seedlings were plan ted in a 1:1 mixture of sand and peat moss. Although no quantitative germination rates were reported, Whitlow stated that seeds of C. tuberosus germinated readily. Henrich et al. (1981) and Myers and Ascher (1982) followed similar protocols for C. tuberosus Both used a culture medium develope d by Norstog for barley embryos. Henrich et al. (1981) stored cultures in polyeth ylene bags in the dark at 25oC for 6 months. After 29 days culture, germination was 29%. Myers and Ascher ( 1982) also stored cultur es in the dark at 25oC. After embryos developed, embryos were transferred to MS medium and stored in the dark at 25oC. When leaf expansion occurred, seedlings were placed under fluorescent lights for 1-3 weeks. Seedlings were then transferred to gla ss jars with MS medium and placed under lights. When at least one leaf fully expanded and dark green roots formed, the plantlets were removed from flask and potted in sphagnum moss (Myers and Ascher, 1982).
51 Of the studies published on seed culture of C. tuberosus Anderson (Anderson, 1990) directly compared seed germination under contin ual darkness and light incubation. Cultures were placed in both light and darkness at 22oC. Burgeff N3f, Knudson C, and modified Luckes medium were compared for germination. Leaf development was more efficient on Luckes medium with leaves reaching 5 cm in 8 weeks. Seedlings incubated under fluorescent lights were larger than seedlings developed in darkness, but no statistics were re ported. Within 2 months, light-grown seedlings produced small corms, 5 cm long leaves, and roots. Unlike Whitlow, who allowed the seedlings to end their growth cycle in vitro Anderson (1990) acclimatized 2 monthold seedlings. In a comparative study between culture medium and light treatment of seeds from Florida, Kauth et al. (2006) found that seeds germinated under both continual darkness and a 16 h photoperiod, but light incubation was superior for seedling development. Seed germination on Knudson C (Knudson, 1946) was higher compared to seeds germinated on Malmgren modified terrestrial orchid medium (Malmgren, 1996) and Phyto Technology orchid seed sowing medium ( Phyto Technology Laboratories, LLC, Shawnee Mission, KS). However, seedling development was superior on Phyto Technology orchid seed sowing medium. Habitat Descriptions of Calopogon tuberosus The populations studied were chosen on seve ral criteria including their widespread distribution, varying habitats, a nd proximity to local individuals who collected seed on my behalf. The populations were locate d at the following sites: 1) Fl orida Panther National Wildlife Refuge, Collier County, FL; 2) Goethe State Fo rest, Levy County, FL; 3) Ashmore Heritage Preserve, Greenville County, SC; 4) Site C: Greenville County, SC; 5) Eva Chandler Heritage Preserve, Greenville County, SC; and 6) Ca rney Fen, Menominee County, MI. The following
52 paragraphs describe each sites and C. tuberosus populations using pr eviously published documents as well as personal observations. Habitat and environmental conditions at each site are vastly differe nt (Table 1-1). The populations in South Carolina are found where th e Blue Ridge Escarpment begins about 300-500 m above sea level. The Michigan population is the only population from the glaciated area. Due to latitudinal differences, maximum and mini mum day lengths differ from almost 16 h photoperiods during Michigan summers to fewer than 14 h photoperiods during south Florida summers. Average yearly precipitation is great in Florida, about double that in Michigan (excluding snow levels). Mean monthly temperat ure averages (Figure 1-1) also differ greatly leading to differences in grow ing season lengths (Table 1-1). Carney Fen Carney Fen (designated as Mi chigan) is located on the upper Michigan peninsula (Figure 1-2E) in Menominee County and is part of th e Escanaba State Forest Management Unit. The 2,485 acre parcel of land is the newest addition to the Michigan State Natural Areas Program as of June 2009. Carney Fen is classified as a norther n fen (Figure 1-2D), which is characterized as a graminoid dominated wetland with neut ral to moderate alkaline soils (Kost et al., 2007). Northern fens occur in flat or depressed area s in glacial lakeplains, outwashes, or kettle depressions (Cohen and Kost, 2008). The poorly drained soils are com posed of peat or marl with a high nutrient availability due to groundwater rich in calcium and magnesium (Cohen and Kost, 2008). The groundwater provides a stable hydroperiod year-round, but water-levels normally do not exceed a few centimeters (Kost et al. 2007). Northern fens rich in nutrients support rich species diversity of both plants and animals, including many rare species (Cohen and Kost, 2008).
53 Carney Fen proper is divided into two dis tinct parcels separate d by County Road 374 and north of the Wiregrass Lake wetland complex. Th e fen is surrounded by a conifer swamp in the low areas and hardwood forests in the upland areas. The area nor th of CR 374 transitions into a tamarack savanna with a fen understory with open canopy that provides sufficient sunlight. The savanna transitions into a conifer swamp with a canopy of white cedar, tamarack, and black spruce, and in recent years the tree species have colonized the savannah area (personal observation). The area south of CR 374 occurs on sedge peat with a higher pH that supports more plant species. The C. tuberosus population at Carney Fen is dist ributed throughout the area including north and south areas. Where it is not growing in direct full-sun, it occupies canopy gaps where sunlight penetrates the fo rest floor. In all cases, C. tuberosus is found at the top of sphagnum hummocks often at the base of tr ees that can be colonized by more than 30 plants (Figure 1-2C). Calopogon tuberosus generally flowers from late June to ea rly July. Vegetative or juvenile plants have been observed in early June. Carney Fen is dominated by rushes, sedges, and grasses. While sphagnum mosses ( Sphagnaceae ) are present, brown mosses ( Amblystegiaceae ) are more common. Northern fens support a high diversity of plant species due to the rich minerotrophic conditions and microtopography. Peat mounds a nd hummocks are found scatte red throughout Carney Fen creating micro-environments c onsisting of more acidic-lovi ng sphagnum mosses and plants (Amon et al. 2002). Common graminoids found at Carney Fen include (Kost et al. 2007; Cohen and Kost, 2008): Calamagrostis canadensis (bluejoint grass), Calamagrostis stricta (reed grass), Deschampsia cespitosa (hair grass), Panicum lindheimeri (panic grass), Muhlenbergia
54 glomarata (marsh-wild timothy), Carex lasiocarpa (wiregrass sedge), Carex aquatilis (water sedge), Cares livida (livid sedge), Carex limosa (mud sedge), Carex sterilis (dioecious sedge), Scirpus cespitosa (tufted bulrush), and Eleocharis elliptica (golden-seeded spike rush). Herbs and forbs commonly found include (Kost et al. 2007; Cohen and Kost, 2008): Aster borealis (rush aster), Campanula aparinoides (marsh bellflower), Euthamia graminifolia (grass leaved goldenrod), Solidago uliginosa (bog goldenrod), Iris versicolor (blue-flag iris), Iris lacustris (dwarf lake iris), Lysimachia terrestris (swamp candles), Potentilla palustris (marsh cinquefoil), Menyanthes trifoliate (bog buckbean), Lobelia kalmii (Kalms lobelia), Drosera rotundifolia (round-leaved sundew), Sarracenia pupurea (purple pitcher plant), Utricularia intermedia (flat-leaved bladdwort), and Parnassia glauca (grass-of-Parnassus). Carney Fen also has approximately 26 species of terrestrial or chids many of which are state protected and threatened. A few common shrubs and trees include (Kost et al. 2007; Cohen and Kost, 2008): Potentilla futicosa (shrubby cinquefoil), Myrica gale (sweet gale), Salix pedicellaris (bog willow), Vaccinium sp. (blueberry), Vaccinium oxycoccos (cranberry), Adromeda glaucophylla (bog rosemary), Ledum groenlandicum (Labrad or tea), Kalmia polifolia (bog laurel), and Chamaedaphne calyculata (leather leaf), Larix larnicina (tamarack), Thuja occidentalis (white cedar), and Picea mariana (black spruce). Ashmore Heritage Preserve Ashmore Heritage Preserve (designated as South Carolina 1) is located in northern Greenville County, South Carolina, and is part of Caesars Head State Park and the Mountain Bridge Wilderness Area, and is pa rt of the South Carolina Herita ge Preserve program (Figure 12E). The preserve is a total of 1,125 acres, and is host to a variety of habitats (South Carolina Department of Natural Resources). A 1.6 km long hiking trail encompasses the preserve starting
55 along Persimmon Ridge Road. This particular area of the preserve is approximately 400 m in elevation, descending to 320-340 m. At the center of the preserve is the 53 acre Lake Wattacoo, which is fed by Wattacoo Creek (Figure 1-3D) that falls 45 m down a granitic cliff (South Carolina Department of Natural Resources). A bog-like area with a sphagnum moss floor is found on the north and east side of the lake where rare plants grow (Figur e 1-3C). This area is not a true bog because Wattacoo Creek provides a supply of flowing water, which is characteristic of a fen. Calopogon tuberosus is found along Lake Wattacoo growi ng in the fen areas. The area supports a rather large number of plants with approximately 50-100 in the flowering stage. Plants flower from mid-May to early June, but the majo rity flower toward the end of May. Plants not only grow directly into the spha gnum moss, but also in areas of bare soil. Because the area is inundated with water, several plan ts grow in standing water. The fen area is dominated by sphagnum mosses and several graminoid genera such as Andropogon, Rhynchospora and Fimbristylis (Nelson, 1986). Due to the lack of prescribed fire, the area is slowly being coloni zed by tag alder species. Other plants found in the fen area at Ashmore Preserve include Amianthium muscitoxicum (flypoison), Rhexia virgnica (common meadow beauty), Sarracenia jonesii (mountain sweet pitcher plant), Sarracenia rubra (sweet pitcher plant), Sarracenia purpurea (purple pitcher plant), Drosera rotundifolia (round-leaved sundew), Osmunda regalis (royal fern), Pogonia ophioglossoides (rose pogonia orchid), Aletris farinosa (white colicroot), and Lycopodium sp (clubmoss). Site C Site C (designated as South Carolina 2) is not recognized as a pres erve, but is in close proximity to Eva Chandler and Ashmore Heri tage Preserves in Gr eenville County, South Carolina (Figure 1-4E). The site has characteristics of both hi gh elevation seeps and granitic
56 flatrocks with a gradual slope (Figure 1-4B, C, D). Granitic flatrock communities do not have a year-round supply of water, which determines pl ant diversity (Porcher and Rayner, 2001). Large graminoid-dominated areas are found at the peri meter (Figure 1-4D), and small micro-islands habitats are found toward the base of the slope (Figure 1-4B, C). The soil depth within the microisland habitats at Site C is extremely sh allow and no more than 4 cm deep. The C. tuberosus population at Site C is larger than the other South Carolina populations, with as many as 100 flowering plants. Many plants are in the juvenile stage and do not flower, therefore more than 500 plants may be found in the area. The plants are found both in the grassy perimeter areas as well as the micro-islands, where they grow direc tly on top of granite surrounded by the shallow soil. The vegetation is dominated by gr aminoid species including many Juncus (rush), Andropogon (bluestem), and Calamagrostis (reed grass) spp. (N elson, 1986; Porcher and Rayner, 2001). Herbs and forbs found at Site C include: Krigia montana (dwarf mountain dandelion), Oenothera perennis (little evening primrose), Castilleja coccinea (Indian paintbrush), Helenium autumnale (common sneezeweed), Crotonopsis elliptica (outcrop rushfoil), Talinum teretifolium (Appalachian fameflower), Senecio tomentosus (woolly ragwort), Hypericum gentianoides (pineweed), and Tiarella cordifolia (heartleaf foamflower). Eva Chandler Heritage Preserve Eva Chandler Heritage Preserve (designated as South Carolina 3) is part of the South Carolina Heritage Preserve program. The preser ve is located in northern Greenville County, South Carolina, and is next to Caesars Head State Park (Figure 1-5E). The 253 acre preserve, which is approximately 500 m in elevati on, contains several communities including a pine/hardwood ecosystem and granite outcrop (South Carolina Department of Natural Resources). A small stream originating from a floodplain flows down the outcrop, and supplies a
57 constant flow of water (Figure 1-5D). The gr anite outcrop area has elements of a granitic dome and flatrock, high elevation seep, acidic cliff, and spray cliff. Spray cliffs differ from high elevational seeps by a constant s upply of flowing water, and vasc ular plants are restricted to edges of the water flow (Nelson, 1986). Acidic cl iffs differ from granitic domes and flatrocks by a more sloped rock surface and lack of a canopy and woody species (Nelson, 1986). Granitic domes have a sloping terrain compar ed to granitic flatrock and of ten have sliding water (Nelson, 1986). Regardless of the granite outcrop classificati on, Eva Chandler Preser ve contains several seepage communities called cataract bogs (Porch er and Rayner, 2001). Although the term bog is used, the communities are actually fens because their main water source is seepage and not rainwater (Porcher and Rayner, 2001). Micro-island habitats form along edges of streams where organic matter accumulates in cracks and fissures in the granite outcropping (Figure 1-5 B, C). While the soil is saturated th roughout the year, the availability of flowing water changes seasonally (Nelson, 1986). The underlying soil is shallow leading to slow succession, but woody species can colonize the deeper soils (Porcher and Rayner, 2001). Approximately eight cataract bogs are found at Eva Chandler Heritage Preserve varying in size from 30 cm to more than 2 m in diameter. The soil in the catar act bogs are a few cm to about 25 cm deep in the largest bogs, and a distinct layering ex ists of leaf litter, soil, and then granite. The C. tuberosus population is rather large with more than 200 plants among the eight bogs; however; more plants may exist since many ar e in the juvenile stag e. The plants are found at the periphery of the cataract bogs where woody species do not colonize. Due to the slightly higher elevation, the flowering period is a few days later than the other South Carolina
58 populations. The plants at this lo cation are diverse in color with most plants a paler shade of pink. The vegetation in the surrounding area is very diverse including xe ric species such as Opuntia compressa (eastern prickly pear cactus), gram inoids, herbs, and small trees (Nelson, 1986; Porcher and Rayner, 2001). Small trees that colonize larger cracks and fissures in the granite include Alnus serrulata (tag alder), Acer rubrum (red maple), and Pinus virginiana (Virginia pine). Carnivorous plants growing alongside C. tuberosus include Drosera rotundifolia (round-leaved sundew), Untricularia cornuta (horned bladderwort), and Sarracenia jonesii (mountain sweet pitcher plant). Other herbs and forbs growing with or near C. tuberosus include Spiranthes cernua (nodding ladies tresses orchid), Pogonia ophioglossoides (rose pogonia orchid), Platanthera integrilabia (white fringeless orchid), Parnassia grandifolia (large leaf grass-of-Parnassus), Achillea millefolium (thousand-leaved groundsel), Helenium autumnale (common sneezeweed), and Castilleja coccinea (Indian paintbrush). Graminoids found in the area include Carex prasina (drooping sedge), Carex biltmoreana (stiff sedge), Danthonia spicata (poverty wild-oat grass), Dicahnthelium acuminatum (tapered rosette grass), and Calamagrostis spp (reed grass). Goethe State Forest Goethe State Forest (designated as north central Florida) is located in Levy County, Florida, and is part of the stat e forest system of Florida (Figure 1-6E). The area encompasses more than 53,000 acres with approximately 15 different natural communities including flatwoods, dome and basin swamps sandhills, and ruderal sites in cluding mesic to wet roadways (Florida Division of Forestry, 2005) Goethe State Forest also holds one of the largest old-growth longleaf pine flatwoods in Florida (F lorida Division of Forestry, 2005).
59 Calopogon tuberosus is found along a contiguous 3 km stretch of County Road 336. The areas where C. tuberosus grows is a mesic roadside with sa ndy soils at the edge of a longleaf pine flatwoods (Figure 1-6C, D). The roadside is slightly sloped with a seasonally wet sphagnum moss slough. Plants are scattered across both side s of CR 336, and are found in both the higher road and slough areas. The population of C. tuberosus is large (> 1000) with a mix of blooming and juvenile plants. Plants flower from late April to mid June, with peak flowering in mid May. Of all the populations, this is the only location where C. tuberosus is found in large groups, possibly due to a high percentage of vegetative growth via corms (Figure 1-6B). The population is morphologically diverse containing numerous color forms includi ng the pale and alba form of C. tuberosus (Figure 1-6B). Hybridization may also be occurring with other Calopogon species including C. pallidus Compared to the other sites, the vegetation al ong the roadside is dominated more by forbs and herbs rather than graminoids. Se veral commonly found graminoids include Rhynchospora colorata (star rush white-top), Xyris spp. (yelloweyed grass), Juncus effusus (soft rush), Andropogon virginicus (broom sedge bluestem), and Schizachyrium scoparium (little bluestem). Herbs and forbs found are as follows: Coreopsis spp. (tickseed), Mimosa nuttallii (sensitive briar), Aletris lutea (yellow colic root), Aletris obovate (southern colic root), Rhexia spp. (meadowbeauty ), Asclepias tuberosa (butterfly weed), Helinathus spp. (sunflower), Eriocaulon decangulare (tanangle pipewort), Pogonia ophioglossoides (rose pogonia orchid), Spiranthes vernalis (grass-leaved ladies tresses orchid), Hypericum myrtifolium (myrtleleaf St. Johns wort), Sabatia campanulata (slender rose gentian), Erigeron sp. (fleabane), Baluduina angustifolia (coastalplain honeycomb head), Polygonella polygama (October flower), and Solidago odora var. chapmanii (Chapm ans goldenrod). Larger shrubs are scarce but include
60 Serenoa repens (saw palmetto) and Vaccinium sp. (blueberry). Several carnivorous plants are found including Drosera spp. (sundew) and Pinguicula spp. (butterwort). Florida Panther National Wildlife Refuge The Florida Panther National Wildlife Refuge (FPNWR; designated as south Florida) is located 20 miles east of Naples, FL along Interstate 75, and consis ts of 26,400 acres within the Big Cypress Basin (U.S. Fish and Wildlife Service, 2009). The FPNWR (Figure 1-7E) was established in 1989 in order to protect the Florid a panther and the mosaic of habitats located throughout. As a national wildlife refuge, the FPNWR actively manages the area with invasive plant removal, prescribed burning, native plan t propagation, and restor ation activities. The FPNWR is divided into 50 fire-management units. Calopogon tuberosus is currently found in units 23, 24, 25 where wet prairies are located (Figure 1-7C, D). Wet prairies are included under the flatwoods mars h habitat classification, and bot h are included under the larger freshwater marsh ecosystem (Kushlan, 1990). Wet prairies have the shortest hydroperiods of freshwater marshes of appr oximately 50-170 days (Duever et al. 1986; Kushlan, 1990; Rutchey et al. 2006). Wet prairies dry duri ng low rainfall and are only inundated during seasonal rains (Kushlan, 1990), explaining the dry conditions often encountered in units 23, 24, and 25 at the FPNWR. During the wet season water leve ls are normally below 20 cm (Duever et al. 1986). Sandy/clay mineral soils with low organic matter, high fire frequency of 2-4 years, and diverse vegetation dominated by graminoids are found are characteristic of wet prairies (Davis, 1943; Florida Natural Areas Inventor y, 1990; Kushlan, 1990; Rutchey et al., 2006). The prairies at the FPNWR are found between pine flatwoods dominated by Pinus elliotii and are dominated by a short grass and sedge (less than 1 m) flora (Davis, 1943; Duever et al. 1986). The population of C. tuberosus is morphologically diverse, because both var. tuberosus and simpsonii (Figure 1-7A, B) are found together. Possible interbreedin g between the two
61 varieties exists. The largest population of C. tuberosus is found in unit 23 where several hundred plants flower from March through May with peak flowering season in mid April. The vegetation is dominated by graminoids with a few forbs and herbs. Herbs and forbs include Crinum americanum (swamp lily), Helenium nuttallii (sneezeweed), Oxypolis filiformis (water carrot/cowbane), Hypericum fasciculatum (St. Johns wort), Dyscheriste oblongifolia (twinflower), and Polygala incarnata (procession flower). Serenoa repens (saw palmetto) and Sabal palmetto (cabbage palm) are found in very low numbers mainly due to colonization of nearby pine flatwoods. Grami noids commonly found in wet pr airies include (Davis, 1943; Duever et al. 1986): Aristida virgata (switch grass), Aristida affinis (poverty grass), Cladium jamaicensis (saw grass), Rhynchospora tracyi (Tracys beak-rush), Schoenus nigricans (black sedge), Muhlenbergia spp. (muhly grass), Schizachyrium scoparium (little bluestem), Panicum hemitomom (maidencane), Dichromena colorata (white-top sedge), and Hypoxis juncea (yellowstar grass).
62Table 1-1. Summary of Calopogon tuberosus populations studied. Population Coordinates Habitat Elevationa Growing Seasonb Day Lengthc Max (h) Min Precipitationb Michigan 45 34' 47" N 87 39' 38" W Northern fen 240 m 125 days 15.7 8.8 740 mm South Carolina 1 35 05' 12" N 82 34' 47" W Fen 340 m 210 days 14.5 9.8 800 mm South Carolina 2 35 05' 06" N 82 35' 43" W Seepage 450 m 210 days 14.5 9.8 800 mm South Carolina 3 35 04' 28" N 82 36' 19" W Cataract bog 500 m 210 days 14.5 9.8 800 mm North Central Florida 29 09 18" N 82 37 12" W Road 12 m 270 days 14.1 10.3 1400 mm South Florida 26 10' 06" N 81 21' 51" W Prairie 4 m 365 days 13.8 10.5 1450 mm aElevational data from Google Earth last retrieved 18 Aug 2009. bData from the National Climactic Data Center (2009). Growing season length is the number of days between the last sp ring frost (above 0C) and first fall frost (below 0C). cPhotoperiod from National Weather Service (http://www.weat her.gov/) last retrieved 18 August 2009.
63 Figure 1-1. Monthly temperatur es at population locations studi ed. Temperature data except South Florida were obtained from the closest city within 50 km to specific sites. A) Michigan: Escanaba, MI. B) South Caroli na: Greenville, SC. C) North Central Florida: Ocala, FL. D) South Florida te mperature data collected on the Florida Panther National Wildlife Refuge with a HOB O H8 Pro series we ather station. Data from National Weather Service (http://www. weather.gov/) last retrieved Feb 18, 2009.
64 Figure 1-2. Carney Fen, Michigan. A) Individual flower note th e strong apex on the labellum. B) Multiple plants growing in close pr oximity. C) Sphagnum hummock containing numerous C. tuberosus plants. D) Northern fen habitat. E) Location of Carney Fen in the upper Michigan peninsula.
65 Figure 1-3. Ashmore Heritage Preserve, South Carolina. A) Individual flower. B) C. tuberosus plant growing in sphagnum moss note the wide leaf. C) Area containing C. tuberosus population around the lake edge. D) Ha bitat at Ashmore Heritage Preserve. E) Location of South Carolina popu lations in the upstate.
66 Figure 1-4. Site C, South Carolina. A) Individual flower. B) Small micro-island habitat. C) Large micro-island habitat. D) Seepage area with flowing water. E) Location of South Carolina populations in the upstate.
67 Figure 1-5. Eva Chandler Heritage Preserve, South Carolina. A) I ndividual flower. B) Cataract bog habitat. C) Multiple cataract bogs. D) Flowing water on granite outcropping. E) Location of South Carolina populations in the upstate.
68 Figure 1-6. Goethe State Forest, Florida. A) Individual flower. B) Group shot with lighter pale flower. C) Roadside habitat. D) Roadside habitat. E) Location of Goethe State Forest in north central Florida.
69 Figure 1-7. Florida Panther National Wildlife Re serve, Florida. A) Individual flower with crested labellum characteristic of C. tuberosus var. tuberosus B) Flower with flat labellum and white trichomes characteristic of C. tuberosus var. simpsonii C) Prairie habitat. D) Wet prairie habitat. E) Lo cation of Florida Panther National Wildlife Refuge in south Florida.
70 CHAPTER 2 EFFECTS OF PHOTOPERIOD AND GERMINATION MEDIA ON IN VITRO SEED ECOLOGY OF Calopogon tuberosus Introduction Ecotypic differentiation has recen tly been recognized as an important issue in several plant sciences including conservation, restoration, and population genetics (Hufford and Mazer, 2003). Ecotypic differentiation enables sp ecies to survive dive rse habitats and envi ronmental conditions across their geographical range, bu t the specific functions they serve in ecosystems remain unclear (Seliksar et al. 2002). For this reason, using local pl ant material for restoration purposes or reintroductions may be necessary to mainta in ecosystem stability (Linhart, 1995). Introducing poorly adapted ecotypes into unsuitable habitats may lead to reduced plant population fitness (Hufford and Mazer, 2003; McKay et al. 2005). Common garden studies are often utilized to detect local ad aptation (Sanders and McGraw, 2005), but obtaining permits to collect and transp lant protected, rare, th reatened, or endangered species is often difficult. Alte rnatively, studying the ecology and physiology of seed germination and seedling development from widespread popul ations may provide insi ght into the range of ecotypic differentiation (Singh, 1973). Studies of orchid seed germination ecology are needed to support reintroduction programs that typically us e seed germination as a propagation tool. Studying the germination ecology of orchid seeds in situ is difficult and time consuming because orchid seeds are minute and germination is often low (Brundrett et al. 2003; Zettler et al. 2005b; Diez, 2007). Alternative to in situ germination, asymbiotic in vitro techniques were developed to germinate orchid seed (Kauth et al. 2008). In vitro conditions can also be manipulated to, at best, mimic in situ conditions such as photoperiod and temperature. Given that Calopogon tuberosus is a commonly recognized or chid in North America, information exists regarding ecology, pollinati on, and seed germination for this species.
71 However, information regarding seed germina tion is often conflicting. Different photoperiods were recommended for seed germination of C. tuberosus including complete darkness to light incubation (Stoutamire, 1974; Whitlow, 1996; Kauth et al. 2006). Likewise, different germination media were also recommended (Henrich et al. 1981; Arditti et al. 1985; Anderson, 1990; Kauth et al. 2006). Differences in germination and seedling de velopment might be the result of local adaptation to specific environmental conditi ons. Attributing ecotypic differentiation to germination differences is difficult sin ce seed source is rarely reported in C. tuberosus seed germination studies, and basing recommendations for seed germination of C. tuberosus on one population is tenuous. Evaluation of in vitro seed germination from diverse populations may clarify whether ecotypic differentiation occurs among C. tuberosus populations. In this paper, the effects of photoperiod and culture media on as ymbiotic seed germination and seedling development are compared among widespread populations of C. tuberosus The objectives of these experiments were: 1) Compare the effects of photoperiod on seed germination ecology among latitudinally widespread C. tuberosus populations; 2) Compare the effects of germination media on seed germin ation ecology among latitudinally widespread C. tuberosus populations; 3) Determine wh ether soil nutrients may contri bute to differences in seed germination among latitudinally widespread C. tuberosus populations; and 4) Validate the use of in vitro studies in differen tiating ecotypes of C. tuberosus Materials and Methods Seed Source Intact seed capsules (sli ghtly yellow in color) were collected before dehiscence approximately 2 months after peak flowering. Caps ules were collected from the Florida Panther National Wildlife Refuge (Collier County, Florida) Goethe State Forest (Levy County, Florida),
72 Ashmore Heritage Preserve (Greenville Count y, South Carolina), Eva Chandler Heritage Preserve (Greenville County, Sout h Carolina), Site C near Ev a Chandler Heritage Preserve (Greenville County, South Carolina), and Carn ey Fen (Menominee County, Michigan). Nondehisced capsules were collected to reduce the potential for surface contamination of seeds. Upon collecting and receiving capsules, seeds we re pooled according to source and stored at 23C over silica desiccant for 2 weeks. After 2 weeks, seeds were removed from the capsules and stored over silica desiccant at -11C until use. Seed Viability Test A viability test (Lakon, 1949) was performed on seed from all populations by staining embryos with 2, 3, 5 triphenyl tetrazolium chloride (TTC). Seeds were first scarified in an aqueous 5% CaOCl2 solution for 0, 0.5, 1, 2, or 3 h. Two replications of approximately 100 seeds each were used per treatment. After scarification, seeds were rinsed twice in distilled-deionized (dd) water and suspended in sterile water for 24 h in darkness at 23 2 C. Water was replaced with TTC and seeds were soaked for 24 h at 30 C in darkness. After the TTC soak, embryos were scored as viable if any de gree of red staining was observed. Media and Seed Preparation Media were prepared in 1000 mL batches, a nd the pH was adjusted to 5.7 with 0.1N KOH prior to autoclaving for 40 min at 117.7 kPa and 121C. Forty mL sterile medium was dispensed into square 100x15 mm Petri plates with a 36-cell bottom (Integrid Petri Dish, Becton Dickinson and Company, Franklin Lakes, NJ, USA). Seeds were surface sterilized in sterile scintillation vials for 3 min in a solution of 5 mL absolute ethanol, 5 mL 6% NaOCl, and 90 mL sterile dd water. Seeds were rinsed twice with sterile dd water after surface sterilization. Solutions were removed from the vials with steril e 5 mL glass Pasteur pipettes. Seeds were then placed on the surface of the germination media with a 10L sterile inoculating loop. The interior
73 16 cells of the Petri plates were used for subr eplications to avoid uneven media drying at the edges. Petri plates were sealed with one layer of Nescofilm (Karlan Research Products, Santa Rosa, CA, USA). Seed germination and seedli ng development (Table 2-1) were monitored weekly for 8 weeks according to the six deve lopmental stages described by Kauth (2005). Photoperiod Effects on Asymbiotic Germ ination and Early Seedling Development The first experiment was a 6x3 factorial was designed with six seed sources and three photoperiods including a short da y (SD; 8/16 h L/D), neutral da y (ND; 12/12 h L/D), long day (LD; 16/8 h L/D). These photope riods were chosen to represent commonly encountered photoperiods throughout the range of C. tuberosus Phyto Technology Orchid Seed Sowing Medium (#P723; Phyto Technology Laboratories, Shawnee Mi ssion, KS, USA) was used based on previous success with C. tuberosus seed germination and development (Kauth et al. 2006). Ten replicate Petri plates with five randomly sel ected subreplications were used per seed source and photoperiod treatment. An average of 48.5 s eeds was sown for each subreplication for an average of 242.5 seeds per replic ation and 2425 seeds per treatment. Each subreplication Culture vessels were placed under cool-whi te fluorescent lights (F96712, General Electric) at an average of 33.3 7 (12/12 photoperiod), 31.6 5 (8/16 photoperiod), and 31.6 6 (16/8 photoperiod) mol m-2 s-1 and incubated at 25 0.4C. Photoperiod Effects on Advanced In Vitro Seedling Development After 8 weeks, developing embryos and seedling s were transferred from Petri plates to 95x95x 100 mm Phyto Tech culture boxes ( PhytoTechnology Laboratories, Shawnee Mission, KS, USA) containing 100 mL P723 medium; seedlings were maintained in corresponding photoperiods. Ten seedlings were transferred to e ach culture vessel. After an additional 8 weeks and 16 weeks total, five cultu re vessels per treatment (50 total seedlings) were randomly selected. However, due to the lack of develope d embryos 30 seedlings from South Carolina were
74 used. Seedling percent biomass allocation was determined by dividing corm, root, and shoot weights by the total seedling weight. Additional da ta recorded were leaf number, shoot length, root number, root length, and corm diameter. Cu lture conditions were the same as previously described. Asymbiotic Germination Media Evaluation A 6x6 factorial design with six germination me dia (Table 2-2) and six seed sources was used. Five media commercially prepared by Phyto Technology Laboratories were used: BM-1 Terrestrial Orchid Medium (BM-1; #B141; van Waes and Debergh, 1986 a), Knudson C Orchid Medium (KC; #K400; Knudson, 1946), Malmgren M odified Terrestrial Or chid Medium (MM; #M482; Malmgren, 1996), Orchid Seed Sowing Medi um (#P723), and Vacin and Went Modified Orchid Medium (VW; #V895; Vacin and Went 1949). Murashige and Skoog Medium in half strength (MS; #M5524; Mura shige and Skoog, 1962) was comme rcially prepared by SigmaAldrich (St. Louis, MO, USA). BM-1 and VW were further supplemented with 0.1% charcoal, KC was further supplemented with 0.1% charcoal and 0.8% TC agar ( Phyto Technology Laboratories), MM was further supplemented with 0.8% TC agar, and MS was further supplemented with 0.1% charcoal, 0.8% TC agar, and organic supplements including peptone found in P723. All media contained 2% sucrose. Fi ve replicate Petri plat es with three randomly selected subreplications with an average of 62.6 seeds were used per treatment. An average of 187.8 seeds was sown per replicate plate for a tota l of 939 seeds per treatment. Germination and development were monitored biweekly for 8 weeks. Culture vessels were placed under ND conditions, cool-white fluores cent lights at 33.3 7 mol m-2 s-1 and 25 0.4C. Only one photoperiod was chosen as to not co nfound the experimental design.
75 Soil Analysis In order to determine if soil nutrient availa bility reflected germination media, a soil analysis from each site was conducted. Soil was co llected from each study site during the 2009 growing season. Soil cores were taken with a 5.5 x 6 cm tulip bulb planter and stored at approximately 6C until preparation. Because of th e fragile nature of the cataract bogs at Eva Chandler Preserve (South Carolina 3), soil was colle cted from the forested area approximately 5 m from the cataract bog (see Figure 1-5D). Soil at Site C (South Carolina 2) was not collected from the micro-island habitats, but ra ther from the larger areas where C. tuberosus grows (see Figure 1-4D). Soils were dried at 30C for 72 hou rs, ground with a mortar and pestle, and sieved through a No. 20 mesh screen. Samples were submitte d to the University of Florida/Institute of Food and Agricultural Sciences Extension Soil Tes ting Laboratory. All soil tests were performed according to Environmental Protec tion Agency (EPA) standards. Total Kjeldahl Nitrogen (TKN), which tests the sum of organic ni trogen, ammonia, and ammonium, was performed according to EPA 351.2. Howe ver, this test does not give the amount of available nitrogen to pl ants. Potassium, calcium, magnesium, phosphorus, and iron were extracted using the Mehlich 1 technique and analyzed using inductively coupled plasma spectrometry (ICP). ICP metals were tested according to EPA 200.7. Organic matter was tested using loss on ignition, and pH was tested acco rding to EPA 150.1. For detailed testing methods see Mylavarapu and Kennelley (2002). Statistical Analysis Germination percentages were calculated by di viding the number of germinated seeds by the total number of seeds with an embryo in each subreplication. The percentage of embryos and seedlings in a developmental stage (Table 2-1) was calculated by dividing the number of seeds in
76 a stage by the total number of s eeds. A developmental index of germinated seeds was modified from Otero et al. (2004): DI = (N 1 + N 2 *2 + N 3 *3 + N 4 *4 +N 5 *5 + N 6 *6) (N1 +N2 + N3 + N4 + N5 + N6) where N1 is the number of seeds in stage 1, etc. St age 0 ungerminated seeds were excluded since only germinated seeds were of concern. Data were arcsine transformed to normalize variation prior to analysis. Germination, seedling developm ent, and embryo viability data were analyzed using general linear model procedures (proc glm) and least-square means (lsmeans) using SAS v9.1 (SAS Institute, 2003). Results Seed Viability For all populations except south Florida (no di fference in pretreatment time), the highest percentage of viable embryos was observed after 3 h of calcium hypoc hlorite pretreatment (Table 2-4). Maximum embryo viability from seed collected in 2006 was as follows: 85.4% south Florida, 66.7% north central Florida, 25.0% South Carolina 1, 38.1% South Carolina 2, 42.1% South Carolina 3, and 50.3% Michigan. Em bryo viability was also tested for seeds collected in 2007, but seeds we re only scarified in CaOCl2 for 3 h. Embryo viabilities were as follows: 70.5% south Florida, 53.9% north centra l Florida, 25.0% South Carolina 1, 10.6% South Carolina 2, 30.3% South Carolina 3, 20.8% Michigan. Photoperiod Effects on Germination and Early Development ANOVA results indicate that pho toperiod alone did not signific antly influence germination (F2 = 0.58, p = 0.56), but the interacti on between photoperiod and source (F5 = 9.97, p < 0.0001) as well as source alone (F10 = 225.1, p < 0.0001) were highly si gnificant. However, photoperiod (F2= 8.43, p = 0.0002), source (F5 = 211.5, p < 0.0001), and their interaction (F10 = 6.31, p <
77 0.0001) were highly significant for early embryo and seedling development during weeks 1-8. Seeds from Michigan germinated and developed more quickly compared to other populations. Imbibition occurred one week after seeds we re inoculated onto the germination medium. However, total germination by week 8 was s till low compared to Florida populations. Germination of Michigan seeds was similar re gardless of photoperiod (Figure 2-1A). Embryos developed quickly with corms forming by week 6. Approximately 95% of the germinated seeds in all photoperiods developed to stage 6 (Figure 2-1A). The average developmental stage was between stage 5 and 6 with no differe nce among photoperiods (Figure 2-2A). Germination of South Carolina seeds was the lowest of all populations (Figure 2-1). Germination percentages in all South Carolin a populations did not exceed 4%. The highest germination was observed under long day conditions in South Carolina 3 seeds (Figure 2-1D). Although embryos did develop to stages 5 and 6 in all South Carolina popul ations (Figure 2-1B, C, D), development on average was low. South Carolina 1 (Figure 2-2B) and South Carolina 2 (Figure 2-2C) embryos developed slowly. The average developmental index for both South Carolina 1 and 2 was less than the stage 2. De velopment of South Carolina 3 (Figure 2-2D) embryos was more advanced than other South Carolina populations. U nder long day conditions, South Carolina 3 embryo development was to stage 4. Seeds from north central Florida germinated quickly and corms formed on seedlings after week 8 (Figure 2-1E). Greater than 16% of the protocorms in each photoperiod developed to an advanced leaf-bearing stage (Fi gure 2-1E). Total seed germina tion percentage was highest under short day conditions for both th e north central Florida (60.2 %) and south Florida (48.5%) populations (Figure 2-1E, F). P hotoperiod did not significantly influence embryo development in either Florida population. However, development wa s more advanced in the north central Florida
78 population with most embryos developing to stag e 4 by week 8 (Figure 2-2E). Development of north central Florida embryos was only exceed ed by the Michigan population. Although South Florida seed germination was relatively high, de velopment was low (Figure 2-2F). Fewer than 5% of the south Florida seeds under SD conditio ns developed past imbibition after 8 weeks culture (Figure 2-1F). Approximately 10% of the seeds under both ND and LD conditions developed past imbibition (Figure 2-1F). S outh Florida seeds germinated slowly and development was delayed compared to all other sources. Photoperiod effects on Advanced Seedling Development Advanced seedling development was monitore d from weeks 8-16. After 16 weeks culture, Michigan seedlings began to se nesce while South Carolina and Florida seedlings continued to grow (Figure 2-3). Corm formation was limited in south Florida seedlings, while Michigan, South Carolina, and north central Florida seed lings all formed corms (Figure 2-3). Biomass allocation was similar among photoperiods within each seed source (Figure 2-4). Maximum dry weight allocation to corms was observed in Mi chigan seedlings (Figur e 2-4A). Although north central Florida seedlings formed large corms, the percent dry biomass allocation was more evenly distributed among shoots, corms, and r oots than other populations (Figure 2-4C). The greatest seedling shoot biomass allocation was obse rved in South Carolina 3 and south Florida populations (Figure 2-4B, D). After 16 weeks in vitro culture, few differences in s eedling growth and development existed among photoperiod within each individual seedling source (Figure 2-5). Photoperiod had a significant influence on shoot length (F2 = 8.58, p = 0.0002) and root length (F2 = 4.24, p = 0.02), but not leaf number (F2 = 1.45, p = 0.23), root number (F2 = 1.92, p = 0.15), and corm diameter (F2 = 0.57, p = 0.57). However, major differences were observed among seedling source for leaf number (F3 = 2.92, p = 0.03), shoot length (F3 = 119.3, p < 0.0001), root number
79 (F3 = 46.7, p < 0.0001), root length (F3 = 73.3, p < 0.0001), and corm diameter (F3 = 20.8, p < 0.0001). Michigan seedlings generally had the lowest shoot length, root number, and root length (Figure 2-5). North Central Florida seedlings had the longest shoot length, highest root number, longest roots, and largest co rms. Within seed source, ne utral day photoperiods generally promoted growth of south Florida seedlings further than short and long days (Figure 2-5). Photoperiod did not influence seedling growth of Michigan and north central Florida seedlings with the exception of corm diameter. Neutra l day photoperiods promot ed the largest corm growth (Figure 2-5E), while Michigan corms we re the smallest in th e short day photoperiod (Figure 2-5E). Media Effects on Germination and Early Development Germination media (F5 = 2.25, p = 0.05), seed source (F5 = 375.2, p < 0.0001), and the interaction between source and medium (F25 = 4.95, p < 0.0001) significantly influence germination. Early embryo and seedling develo pment during weeks 1-8 were significantly influenced by source (F5 = 63.8, p < 0.0001), medium (F5 = 4.73, p < 0.0001), and their interaction (F25 = 3.11, p < 0.0001). Michigan seeds germin ated and embryos developed quickly on all media compared to other sources. The high est germination percenta ge of Michigan seeds occurred on P723 and BM-1 (Figure 2-6A). Seed germination was still lower compared to Florida populations with germination remaini ng below 40%. A large percentage of embryos developed to stage 6. However, embryo develo pment was lowest on KC and VW with the average stage of development remaining betw een 4 and 5, while embryos on all other media approached stage 6 (Figure 2-7A). Seed germination for South Carolina 1 wa s highest on VW, but germination was only 4.9% (Figure 2-6B). No germination occurred on P723 (Figure 2-6B). Clear differences in germination were not evident in South Carolin a 2 (Figure 2-6C), but germination on P723 was
80 significantly lower than all other media. Germin ation on KC and MS (no difference between MS and VW) was highest for South Carolina 3 seeds, while lowest germination occurred on P723 (Figure 2-6D). Although germination of South Carolina 1 and 2 seeds improved, germination was still lower than 5% on all media. Development among South Carolina populations was very different. No South Carolina 1 embryos developed to stage 6, and less than 1% developed to stage 5 (Figure 2-6B). Advanced stages of de velopment (stages 4-6) were observed for both South Carolina 2 and 3 (Figure 2-7C, D). The aver age stage of development was lower than 3 for South Carolina 1 embryos, which was lower than near stage 6 development for South Carolina 2 and 3. However, examining development of So uth Carolina populations was deceptive since germination was very low. Development of So uth Carolina 2 embryos approached stage 6 (Figure 2-7C), but germination was lower than 5%. Development of South Carolina 3 embryos was highest on MS, VW, P723, and BM-1 (Figure 2-7D). However, more embryos development on MS, KC, and VW since germination was very low on P723 and BM-1. In both Florida populations, few differences in total germination existed among media; at least four media provided high germination percentages (Figure 2-6E, F). Subsequent development differed greatly. For north central Fl orida, more advanced development (stage 4+) occurred on BM-1, MS, and P723 (Figure 2-7E). High germination percentages of north central Florida seeds were observed on MM, but the majority of seeds remained in stage 1 (Figure 26E). Only a small percentage of south Florid a embryos developed to stage 6 (Figure 2-6F). Development was highest on P723, VW, and BM-1 for south Florida embr yos with the average developmental stage of near 4 (Figure 2-7F). Germination percentages were high on KC and MM for south Florida seeds, but as with north central Florida seeds development was below an average stage of 3.
81 Media Effects on Advanced Seedling Development Corm development was more advanced afte r 8 weeks on BM-1, MS, and P723 compared to development on MS, KC, and VW (Figure 28). Seedling development of Michigan, South Carolina 3, and north central Florida seedlings wa s superior to other populations (Figure 2-8). However, development of Michigan, South Caro lina 3, and north central Florida seedlings differed markedly. Corm formation was more pr onounced in seedlings from northern latitudes. Thus by week 8, no corm formation was observed in Florida seedlings while early and advanced corm formation was observed in South Caroli na and Michigan seedlings, respectively. Soil Nutrient Analysis Soil analysis revealed that nutrient availabil ity differed among habitats. Overall, soil from Michigan was the richest in nutri ents and poorest in south Florida with the exception of nitrogen and calcium. Northern fens generally have high le vels of calcium and magnesium as a result of the mineral rich groundwater percolating to the surface (Amon et al., 2002; Bedford and Godwin, 2003; Cohen and Kost, 2008). This was observed in Carney Fen (Michigan), which contained the highest amount of calcium and magnesium (Table 2-3). In addition, nitrogen, phosphorus, and potassium levels we re also highest at Carney Fe n. Soil from south Florida was also high in calcium and total nitrogen, but lowest in phosphoru s and iron. Iron levels were highest in soils from South Caro lina 1 and 3 as well as north central Florida. Total nitrogen levels were similar in soils from South Carolina 1 and 3 as well as north central Florida, but much higher in South Carolina 2. Soil pH levels were alkaline in Michigan and south Florida, and acidic South Carolina and north central Fl orida. Total organic matter was exceedingly high (78.5%) in the Michigan soil, followed by South Carolina 2 and 3, south Florida, South Carolina 1, and north central Florida.
82 Discussion Seed Viability and Quality Differences in seed germination responses are often attributed to seed viability and quality. Comparisons of orchid seed germination among populations of the same species have been reported, but C. tuberosus has not been examined. Symbio tic germination and mycorrhizal specificity among populations rather than ecotypic differentiation were examined in these studies (Zettler and McInnis, 1992; Ze ttler and Hofer, 1998; Sharma et al. 2003). However, differences in seed germination and viability among populations were described which might be accounted for by ecotypic differentiation. Population size and inbreeding de pression may influence low seed germination of several C. tuberosus populations as well as differences in seed viability. Lower germination percentages in small populations of Platanthera integrilabia compared to larger popul ations, were attributed to lower seed viability (Zettler and McInnis, 1992). Similarly, Platanthera clavellata seed germination differences were attributed to inbreeding depression (Zettler and Hofer, 1998). Reduction in pollinator numbers at different sites may lead to seed viability differences in C. tuberosus as reported for Platanthera leucophaea and P. praeclara (Bowles et al., 2002; Sharma et al. 2003). Another plausible explanation regarding diffe rences in seed viability may be selfpollination. Calopogon tuberosus is a non-rewarding/out-crosser pollinated by Bombus Xylocopa and Megachile bees through deception (van der Pijl and Dodson, 1966; Thien and Marcks, 1972; Dressler, 1981). Self-pollination in C. tuberosus may be common as Firmage and Cole (1988) reported in Maine populations. Self-pollination in Calypso bulbosa and likely C. tuberosus was mediated by bumblebees since a mechanism for autogamy does not exist (Alexandersson and Agren, 2000). While fruit set is generally not affected by self-pollination,
83 reduced seed viability or embryo production can be reduced (Tremblay et al., 2005). Low seed viability and germinability in certain C. tuberosus populations may be cause d by higher levels of self-pollination; however, further investigation is warranted. The correlation between TTC determined seed viability and the corresponding observed percent germination is often variable and species specific (van Waes and Debergh, 1986 a; StArnaud et al. 1992; Shoushtari et al. 1994). van Waes and Debergh (1986 b) reported various optimal pretreatment times from 45 min to 16 h in calcium hypochlorite in 31 species of terrestrial orchids, thus differences in C. tuberosus viability are not surprising. Tetrazolium testing can overestimate viability because this test does not detect inactive enzymes that may become active during germination (Lauzer et al. 2007). For this reason, fluorescein diacetate (FDA) is used with results of ten correlating with germina tion (Pritchard, 1985; Vendrame et al., 2007). Lower germination percentages compared to viability may reflect non-optimal temperatures with seeds from northern climates requiring cooler temperatures in vitro or stratification to germinate; these concerns are cu rrently being addressed in separate experiments. In addition, seeds that do not germinate in vitro may have an intrinsic dormancy mechanism. Embryo damage during surface steri lization is also a likely scen ario that may have reduced germination. Photoperiod No other published research ex ists that compare photoperiodic effects on North American orchid seed germination spanning several popu lations of the same species from distant geographic sources. For non-orchid ecotypes photoperiod was reported to be an important factor on germination (Singh, 1973; Probert et al. 1985 b; Seneca, 1974). Due to latitudinal differences in location, C. tuberosus populations experience different se asonal variations in photoperiod, temperature regimes, and growing season duration. Calopogon tuberosus flowers in early June to
84 mid-July in the north and mid-May to early June in the south (Luer, 1972). In Florida, seed capsules dehisce and seeds are disbursed in Ju ly when photoperiods are approximately 13-14 h. Short day conditions promoted the highest germin ation percentage for both Florida populations. At both Florida locations, the shortest na tural photoperiods do not approach 8 h, but approximately 10 h. Whether Florida seeds are somewhat light sensi tive during germination remains unclear without also conducting in situ germination studies. Development of embryos in all three photoperiod s for north central Florida was very rapid compared to south Florida embryos. A large perc entage of south Florida seeds germinated only to the imbibition stage by week 8, perhaps due to the longer growing season in south Florida. In north central Florida, lower daily surface temper atures in winter can drop below the freezing point, while low daily winter temp eratures in south Florida rare ly drop below 5C. The warmer climate in south Florida may allow slower em bryo development withou t increased mortality, assuming that temperatures within the soil wher e embryos reside are also different. After 16 weeks culture, south Florida seedlings were sma ll and did not form corms, while north central Florida seedlings were larger and had readily formed corms. Although total germination was low in Michig an seeds, seedling development and corm formation was more rapid than those from the southern populations. Imbi bition occurred after 1 week, and corm initiation began by week 6. Regardless of photoperiod, after 16 weeks culture the large seedling corm:shoot:root ratios generate d in seedlings from th e Michigan population suggest a high percentage of car bohydrates are allocated to co rms. Rapid seed germination, seedling development, and corm formation in northern populations may indicate that seedlings do not respond to photoperiod en suring rapid corm development prior to the shorter growing
85 seasons. Kane et al. (2000) similarly reported rapid corm development in northern ecotypes of the wetland non-orchid species Sagittaria latifolia. Media Screen and Soil Nutrient Availability P723 proved to be an adequate medium for ge rminating Florida and Michigan seeds, but discrepancies between germination and viabi lity were perhaps caused by using a non-optimal medium for other seed sources. Ab undant literature ex ists on the influenc e of media mineral nutrition for orchid seed germination and subseque nt seedling development (Curtis, 1947; Spoerl and Curtis, 1948; Raghavan, 1964; van Waes and Debergh, 1986a; Kauth et al., 2006). However, site-specific differences in soil nutrient availability could explain differences in germination and development as found in Dactylorhiza incarnata by Dijk and Eck (1995 b). Seedlings from coastal areas grew faster in vitro and were more tolerant of exogenous ammonium and nitrate compared to seedlings from inland populations. Coastal populations inha bited calcareous areas where high nutrient levels were found due to th e introduction of fertilizers and poor drainage (Dijk and Eck, 1995b). The wide range in pH levels among the populat ions is not surprising. Sheviak (1983) found C. tuberosus in soils with a wide pH tolerance from 4.5 to 7.5, but few plants were found in soils with a moderate pH of 6.0-7.0. Sheviak (1983) noti ced that nutrient availability was different with varying pH levels. Phosphorus was higher in soils with a pH from 6.0-7.0, iron was higher in more acidic soils, and potassium, calcium, and magnesium were higher in neutral to alkaline soils. The same trends were observed in the C. tuberosus soil samples. Michigan soil, with a pH of 7.0, contained higher phosphorus le vels than acidic South Carolin a soils and highly alkaline south Florida soils. The more acidic South Carolina and north cen tral Florida soils contained more iron compared to the alkaline soils of Mich igan and south Florida. Calcium levels were
86 higher in the alkaline soils of Michigan and s outh Florida, and magnesium and potassium levels were higher in Michigan soils. Although soil from Michigan contained the highest amount of nitrogen and phosphorus, available nitrogen and phosphorus levels were reported in low le vels in northern fens (Amon et al. 2002; Bedford and Godwin, 2003). However, highe r nitrogen levels can be present due to agricultural and septic tank drainage, and atmospheric nitroge n deposition (Bedford and Godwin, 2003). Carney Fen (Michigan) is in close proximity to agricultura l areas (personal observation), which may be an influence on the higher nitr ogen and phosphorus levels. The Kjedahl total nitrogen test includes organic nitrogen, ammonium, and nitrate, which may also influence the high nitrogen levels in Carney Fen. The low phosphorus in south Florida soil was surprising since increa sed urbanization and agriculture in south Florida has in creased phosphorus levels (Bruland et al. 2006). Bruland et al. (2006) reported higher nutrient levels in south Fl orida soils near the Ever glades compared to the soil from the Florida Panther National Wildlife Re fuge (south Florida). This may be influenced by the higher nutrient load from Lake Okeechob ee flowing through southeast Florida. Lower nutrient levels in soil on the Fl orida Panther National Wildlife Re fuge might be explained by the lower development in southwest compared to southeast Florida, and is relatively isolated from the Lake Okeechobee flowage. Porcher and Rayner (2001) reported that soils in cataract bog areas in South Carolina were high in magnesium and calcium, but Michigan soil was higher in calcium and magnesium while south Florida soil wa s higher in calcium. Because soil analysis for reference habitats in South Carolina were not found, it is difficult to assess what Porcher and Rayner (2001) meant by high levels.
87 Assessing the nutrients and the appropriate conc entrations from soil analysis to promote seed germination is difficult because nutrient leve ls in the soils and germination media are not equivalent. However, general conclusions can be made by comparing the two. Seeds from South Carolina 3 germinated better on media with high er mineral salt concentrations, calcium, and magnesium. High germination of Michigan s eeds was observed on P723 containing peptone, a source of organic nitrogen. While MS was also supplemented with pept one, it also contained higher levels of inorganic nitrogen. Due to the higher organic matter in the Michigan soil, the higher organic nitrogen in P723 may have increased germination of Michigan seeds. Florida soils were generally nutrient poor with the excepti on of calcium. Because few differences in germination were observed among media within each Florida population, this may indicate that Florida seeds will germinate well on any media with a necessary supply of nutrients. Further investigation is needed to addre ss soil nutrient availability in order to develop germination media based on soil analysis. Conclusions The results presented provide insight into phy siological and developmental aspects that are important aspects of ecotypic differentiation. Based on in vitro seed germination studies, ecotypic differentiation is likely occurring within C. tuberosus evident by rapid germination and subsequent seedling development, as well as im mediate corm formation in northern populations. Rapid corm development in northern plants ma y be a consequence of the relatively shorter growing season experienced by these populations. Conversely, southern plants display greater shoot biomass allocation and a slower tendency to form corms. Results indicate that photoperi od is not a strong selection pressure for seed germination and subsequent development, but germination media containing different nutrients is a strong selection pressure. The differences in germination media reflect the variation in soil type and
88 nutrient availability among C. tuberosus habitats, and the species ability to adapt to wide-ranging conditions. The ability of Calopogon tuberosus to adapt to different soil types is remarkable. These soils include the alkaline, peat-based, and nutr ient rich soils of northern fens to the acidic, nutrient poor, sand-based soils in north central Florida.
89 Table 2-1. Six stages of orchid seed development. Stage Description 1 Imbibed seed, swollen and turning green 2 Enlarged embryo without testa 3 Presence of shoot apex 4 Elongated shoot apex and rhizoids 5 Emerging leaf and developing roots 6 Seedling with fully developed leaves and roots
90 Table 2-2. Comparative mineral sa lt content of orchid seed germ ination media. BM-1 Terrestrial Orchid Medium (BM-1), Knudson C (KC), Ma lmgren Modified Terrestrial Orchid Medium (MM), Murashige and Skoog (MS), PhytoTechnology Orchid Seed Sowing Medium (P723), Vacin and Went (VW). MM BM-1 P723 VW KC MS Macronutrients (mM) Ammonium 5.15 7.57 13.82 10.31 Calcium 0.73 0.75 1.93 2.12 1.50 Chlorine 0.0021 1.50 3.35 3.10 Magnesium 0.81 0.83 0.62 1.01 1.01 0.75 Nitrate 9.85 5.19 10.49 19.70 Potassium 0.55 2.20 5.62 7.03 5.19 10.89 Phosphate 1.03 2.20 0.31 3.77 1.84 0.63 Sulfate 0.92 1.10 0.71 8.71 8.69 0.86 Sodium 0.20 0.20 0.10 0.20 0.10 Micronutrients ( M) Boron 161.7 26.7 50 Cobalt 0.105 0.026 0.053 Copper 0.10 0.025 0.5 Iron 100 100.2 50 100 90 50 Iodine 1.25 2.50 Manganese 10 147.9 25 30 30 50 Molybdenum 1.03 0.26 0.52 Zinc 34.8 9.22 14.95 Organics (mg l-1) Biotin 0.05 0.05 Casein hybrolysate 400 500 Folic acid 0.5 0.5 L-Glutamine 100 Glycine 2.0 2.0 myo-Inositol 100 100 100 100 Nicotinic acid 5.0 1.0 1.0 Peptone 2000 2000 Pyridoxine HCl 0.5 1.0 1.0 Thiamine HCl 0.5 10 10 Total mineral salt 4.35 6.98 24.72 35.54 46.72 48.01 Total inorganic N n/a n/a 15.00 12.76 24.31 30.01
91 Table 2-3. Comparative soil nutr ient analysis from the study sites. All units are in mg kg-1 with the exception of organic matter (OM) and pH. Michigan South Carolina 1 South Carolina 2 South Carolina 3 North Central Florida South Florida Total N 12453 1603 5150 1639 1705 5729 P 21.7 3.58 2.89 3.30 4.28 0.08 K 226.6 45.5 93.5 124.1 20.1 23.2 Ca 20780 99.3 876.2 605.0 670.2 6515 Mg 4194 20.3 316.7 128.5 38.4 50.4 Fe 9.86 86.5 46.3 276.8 81.4 0.10 OM (%) 78.5 7.45 19.3 11.4 6.74 8.91 pH 7.00 4.90 5.50 5.00 5.00 8.00
92Table 2-4. Effect of scarification time on embryo viability of Calopogon tuberosus seeds from populations studied. Seeds were scarified for 0, 0.5, 1, 2, and 3 h in a 5% aqueous CaClO2 solution prior to staining in tetrazolium. Values represent the mean of two replications with 100 seeds each. Means with th e same letter within populations are not significantly different at =0.05. All values are percent viability. Treatment Length South Florida North Central Florida South Carolina 1 South Carolina 2 South Carolina 3 Michigan Control 78.3a 41.1b 4.40b 5.40b 14.9b 24.1b 0.5 hour 85.4a 50.0b 5.85b 11.6b 20.4b 30.7b 1 hour 79.0a 66.1a 10.6b 6.30b 39.1a 32.5b 2 hours 84.9a 63.8a 15.0b 11.5b 41.1a 33.3b 3 hours 84.3a 66.7a 25.0a 38.1a 42.1a 50.3a
93 Figure 2-1. Photoperiod effects on in vitro seed germination and development of Calopogon tuberosus from widespread populations. Data wa s collected after 8 weeks culture on P723 medium. A) Michigan population fr om Carney Fen. B) South Carolina population from Ashmore Heritage Preserve. C) South Carolina population from Site C. D) South Carolina population from Eva Chandler Heritage Preserve. E) North central Florida population from Goethe St ate Forest. F) South Florida population from the Florida Panther National Wildlif e Refuge. Histobars (mean response of 10 replications with 5 subreplica tions each) within each seed source with the same letter are not significantly different ( =0.05). See Table 2-1 for stages of development.
94 Figure 2-2. Photoperiodic effects on the developmental index of Calopogon tuberosus seedlings from widespread populations. Data was collected after 8 w eeks culture on P723 medium. See Equation 3-1 for developmental index. A) Michigan population from Carney Fen. B) South Carolina population from Ashmore Heritage Preserve. C) South Carolina population from Site C. D) South Carolina population from Eva Chandler Heritage Preserve. E) North cen tral Florida population from Goethe State Forest. F) South Florida population from the Florida Panther National Wildlife Refuge. Histobars (mean response of 10 re plications with 5 subreplications each within each seed source with the same letter are not signi ficantly different ( =0.05). Developmental index was calcu lated as follows: DI = (N1 + N2*2 + N3*3 + N4*4 +N5*5 + N6*6)/ (N1 +N2 + N3 + N4 + N5 + N6) where N1 is the number of seeds in stage 1, etc.
95 Figure 2-3. Effects of photoperiod on in vitro seedling development of Calopogon tuberosus from widespread populations. Seedlings repres ent average size afte r 16 weeks culture. A, B, C, D) Seedlings cultured under an 8/16 h L/D photoperiod. E, F, G, H) Seedlings cultured under a 12/12 h L/D photope riod. I, J, K, L) Seedlings cultured under a 16/8 h L/D photoperiod. A, E, I) Sout h Florida seedlings. B, F, J) North Central Florida seedlings. C, G, K) South Carolina seedlings. D, H, L) Michigan seedlings. Scale bars = 1 cm.
96 Figure 2-4. Percent dry wei ght biomass allocation in Calopogon tuberosus seedlings. Data was collected after 16 weeks in vitro culture. A) Michigan population. B) South Carolina population from Eva Chandler Heritage Preserve. C) North central Florida population. D) South Florida population. Histobars represen t the mean response of 50 seedlings S.E.
97 Figure 2-5. Comparative leaf numbe r, shoot length, root number, root length, and corm diameter in Calopogon tuberosus seedlings from widespread populations. Data was collected after 16 weeks in vitro culture under short day (8/16 h light/dark), neutral day (12/12 h light/dark), and long day (16/8 h light/dar k) photoperiods. Histobars represent the mean response of 50 seedlings. Histobars within with the same letter are not significantly different ( =0.05).
98 Figure 2-6. Effects of culture media on in vitro seed germination and subsequent development of Calopogon tuberosus from widespread populations. Data was collected after 8 weeks culture under a 12/12 h L/D photoperiod. A) Michigan population from Carney Fen. B) South Carolina population from Ashmore Heritage Preserve. C) South Carolina population from Site C. D) South Carolin a population from Eva Chandler Heritage Preserve. E) North central Florida populat ion from Goethe State Forest. F) South Florida population from the Florida Pant her National Wildlife Refuge. Histobars (mean response of 5 replications with 3 s ubreplications each) within each seed source with the same letter are not significantly different ( =0.05). See Table 2-1 for stages of development. For media abbreviations and formulae see Table 2-2.
99 Figure 2-7. Effects of culture medi a on seedling developmental index of Calopogon tuberosus from different populations. Data was colle cted after 8 weeks culture under a 12/12 h L/D photoperiod. See equation 3-1 for devel opmental index. A) Michigan population from Carney Fen. B) South Carolina populat ion from Ashmore Heritage Preserve. C) South Carolina population from Site C. D) South Carolina population from Eva Chandler Heritage Preserve. E) North cen tral Florida population from Goethe State Forest. F) South Florida population from the Florida Panther National Wildlife Refuge. Histobars (mean response of 5 re plications with 3 subreplications each) within each seed source with the same letter are not signi ficantly different ( =0.05).
100 Figure 2-8. Culture media effects on early seedling development of Calopogon tuberosus from widespread populations. Seedli ngs representative average size after 8 weeks culture. A, B, C) North central Florida seedlings. D, E, F) South Carolina seedlings from Eva Chandler Heritage Preserve. G, H, I) Michig an seedlings. A, D, G) Seedlings cultured on BM-1 Terrestrial Orchid Medium. B, E, H) Seedlings cultured on P723 Orchid Seed Sowing Medium. C, F, I) Seedlings cultured on MS. Scale bars = 1 cm.
101 CHAPTER 3 EFFECTS OF COLD STRATIFICATI ON AND DIURNAL TEMPERATURES ON IN VITRO GERMINATION OF Calopogon tuberosus Introduction Differences in photoperiod (Florida populat ions) and nutrient availability among germination media contributed to differences in seed germination and development among Calopogon tuberosus populations as found in Chapter 2. Th is may be caused by abiotic selection pressures such as soil nutrient availability and soil type as well as photoperiod influenced by latitude. Although different germination media increased germination of Michigan and South Carolina populations, germination was still relatively low compar ed to Florida populations. The low germination percentages may be influen ced by a non-optimum temperature used in the previous experiments. Given that C. tuberosus populations are found in areas with different maximum and minimum temperatures, this may im pose a selection pressure influencing ecotypic differentiation. Chilling length requirements may also be contributing to ecotypic differentiation due to the different winter conditions at each site. Studying seed germination ecology has been repo rted as an effective method in detecting ecotypic differentiation in widespread plan t populations (Seneca, 1972; Singh, 1973; Seneca, 1974; Probert et al. 1985b). Understanding germination ecology is essential to further our knowledge about the timing of germination, seed maturation, seed dispersal, dormancy, and environmental cues that promote germin ation (Baskin and Baskin, 2001; Donohue, 2005). Studying the germination ecology of orchid seeds in situ is difficult and time consuming because orchid seeds are minute and germination is often low (Brundrett et al. 2003; Zettler et al. 2005b; Diez, 2007). Asymbiotic in vitro techniques were developed to successfully germination orchid seeds (Kauth et al. 2008). In vitro conditions can also be ma nipulated to, at best, mimic in situ conditions such as photoperi od and temperature.
102 For many plant species temperature is a major factor responsible for the onset and breaking of seed dormancy (Baskin and Baskin, 2004). Unfort unately, the effects of temperature on orchid seed germination have largely been ignored compared to photoperiod and media nutrition. Furthermore, constant temperatur es were used in orchid seed germination more often than fluctuating diurnal temperatur es (Harvais, 1973; Rasmussen et al., 1990a ; Rasmussen and Rasmussen, 1991; Rasmussen, 1992; Mweetwa et al. 2008). Using fluctuati ng temperatures was recommended over constant temperatures because seeds are not commonly exposed to constant temperatures in nature (Baskin et al. 2006). However, constant temperatures improved in vitro germination of several orchids over alterna ting temperatures (van Waes and Debergh, 1986 a). Regardless, selecting temperat ures similar to those found in situ may lead to better insight into ecotypic differentiation by conducting in vitro common garden or transplant studies. The role of cold-stratif ication in orchid seed s has also been examin ed, but its physiological role is not well-understood. In non-orchids cold-stratification wa s reported to decrease enzymatic reactions, slow metabolic processes that inhib it germination, or change enzyme production and concentration (Bewley and Black, 1994). However, reports of co ld-stratification influence on orchid seed germination have been variable due to differences in the chilling length (Rasmussen, 1992; Tomita and Tomita, 1997; Miyoshi and Mii, 1998) and seed age (van Waes and Debergh, 1986a; De Pauw and Remphrey, 1993; Chu and Mudge 1994). In addition, the effects of coldstratification among distant populatio ns of the same orchid species have not been reported. Examining the effects of chilling period on seed germination and embryo development may also provide further insight into ecotypic differentiation. Seeds from northern populations may require longer chill periods to increase germin ation since these seeds are exposed to longer winters.
103 The objectives were to: 1) Compare seed germination among latitudinally widespread C. tuberosus populations after chilling seeds; 2) Examin e the role of diurnal temperatures on seed germination of latitudinally widespread C. tuberosus populations; 3) Examine differences in anatomy and morphology of C. tuberosus seeds from widespread populations; and 4) Validate the use of in vitro seed germination techniques to differentiate ecotypes. Materials and Methods Seed Source Intact seed capsules (sli ghtly yellow in color) were collected before dehiscence approximately 2 months after peak flowering. Caps ules were collected from the Florida Panther National Wildlife Refuge (Collier County, Florida) Goethe State Forest (Levy County, Florida), Ashmore Heritage Preserve (Greenville Count y, South Carolina), Eva Chandler Heritage Preserve (Greenville County, S outh Carolina), Site C near Ev a Chandler Heritage Preserve (Greenville County, South Carolina), and Carn ey Fen (Menominee County, Michigan). Nondehisced capsules were collected to reduce the potential for surf ace contamination of individual seeds. Upon collecting and receiving capsules, seeds were pool ed according to source and stored at 23C over silica desiccant for 2 weeks. After 2 weeks, seeds were removed from the capsules and stored over silica desiccant at -11C until use. Media and Seed Preparation BM-1 Terrestrial Orchid germ ination medium (chosen based on results from Chapter 2) supplemented with 0.1% activated charcoal was us ed as the germination medium for both the cold-stratification and temperature studies. The medium was prepared in 1000 mL batches, and the pH was adjusted to 5.7 with 0.1N KOH prior to autoclaving for 40 min at 117.7 kPa and 121C. Forty mL sterile medium was dispensed in to square 100x15 mm Petri plates with a 36cell bottom (Integrid Petri Dish, Becton Dickin son and Company, Franklin Lakes, NJ, USA).
104 Mature seeds were surface sterili zed in sterile scintillation vials for 3 min in a solution of 5 mL absolute ethanol, 5 mL 6% NaOCl, and 90 mL sterile dd water. Seeds were rinsed twice with sterile dd water after surface steri lization. Solutions were removed from the vials with sterile 5 mL glass Pasteur pipettes. Seeds were then placed on the surface of the germination medium with a 10L sterile inoculating l oop. The interior 16 cells of the Petri plates were used for subreplications to avoid areas of uneven medium drying at the edges. Petri plates were sealed with one layer of Nescofilm (Karlan Res earch Products, Santa Rosa, CA, USA). Seed germination and seedling development (Table 2-1) were monitored weekly for 8 weeks according to the six developmental stages described by Kauth (2005). Cold-stratification Effects on Seed Germination Upon seed inoculation, cultures were wrapped in aluminum foil and placed in continuous darkness at 9.9 0.3C for 2, 4, 6, or 8 weeks. A no chill control of 24.2 0.2C was also used. Five replications with three subreplications each were used for each seed source/chilling treatment combination. Approximately 57 seeds we re sown onto each replicate plate (mean seeds per subreplication = 19) for an average of 285 s eeds per treatment. After each chilling period, vessels were removed from cold-storage, unwrapp ed, and placed in a 12/12 h L/D photoperiod at 24.2 0.2C under cool-white fluorescent light s with a light level of 40 mol m-2 s-1. Culture vessels were initially scored upon removal fr om the dark-chill period. Germination and development were examined bi-weekly for 10 w eeks after culture vessels were removed from cold storage. Diurnal Temperature Effects on Seed Germination Diurnal temperatures were chosen based on av erage seasonal day and night temperatures in Gainesville, FL. The following day/night temperat ure treatments were used: 33/24C (summer); 29/19C (spring); 27/15C (fall); 22/11C (winter). A constant temperature of 25C was also
105 used. Five replications with three subreplications each were used for each seed source/chilling treatment combination. An average of 73.5 seed s were sown onto each plate (mean seeds per subreplication = 24.5) for an average of 367.5 seed s per treatment. BM-1 Terrestrial Orchid germination medium supplemented w ith 0.1% activated charcoal (B141; Phyto Technology Laboratories, Shawnee Mission, KS, USA) was used. Cultures were placed under a 12/12 L/D photoperiod and cool white fluorescent lights with a light level of 40 mol m-2 s-1. Cultures under fluctuating temperatures were incubated in Percival growth chambers (Model 130VL, Percival Scientific, Perry, IA, USA), while cultur es in constant temperatures were placed in Percival growth chamber model 136LL. Culture s were incubated in light during higher temperature periods and incubated in darkness during the cooler temperature periods. Seed germination and seedling development we re monitored bi-weekly for 8 weeks. Scanning Electron Microscopy Seeds from all six populations were obser ved by scanning electron microscopy (SEM). Seeds were not stored in fixative solutions, but prepared as dry se ed. Seeds were stored at -11C over desiccant until use as previously described. Seeds were gold sputter coated for 40 s at 45 mA before images were taken. Digital imag es were captured with a Hitachi S-4000 SEM (Hitachi Scientific Instrume nts, Danbury, CT) at 10.0 kV. Histological Sectioning Mature seeds stored at -11C over desiccant we re submitted for histological sectioning to the Interdisciplinary Center for Biotechnology Research Electron Microscopy and Bioimaging Laboratory at the University of Florida. Seeds were hydrated in distilled water for 24 h and fixed in formalin acetic acid alcohol (FAA) consisting of 50% ethanol (95%), 5% glacial acetic acid, 10% formalin (37% formaldehyde), and 35% disti lled water. Seeds were then dehydrated in an ascending series of 95% and 100% ethanol followed by dehydration in 100% acetone. The ends
106 of the testae were cut to faci litate resin embedding. Dehydrated seeds were then embedded in Spurrs resin (Spurr, 1969) and cured at 60C for 2 days. Polymerized blocks were trimmed and semi-thick sections (500 nm) were cur with a Leica Ultracut ultramicrotome R (Leica Microscopy and Scientific Instruments, Deerfield, IL). Sections were collected on glass slides and stained with toluidine blue. Sections we re viewed under an Olympus BH2 brightfield microscope (Olympus America Inc., Center Va lley, PA, USA). Images were captured with a Retiga-2000R Fast 1394 digital camera (QImagi ng, Surrey, British Columbia, Canada) and QCapture Pro 6 software (QImaging). Statistical Analysis Germination percentages were calculated by di viding the number of germinated seeds by the total number of seeds with an embryo in each subreplication. The percentage of embryos and seedlings in a developmental stage (Table 2-1) was calculated by dividing the number of seeds in a stage by the total number of s eeds. A developmental index of germinated seeds was modified from Otero et al. (2004): DI = (N 1 + N 2 *2 + N 3 *3 + N 4 *4 +N 5 *5 + N 6 *6) (N1 +N2 + N3 + N4 + N5 + N6) where N1 is the number of seeds in stage 1, etc. St age 0 ungerminated seeds were excluded since only germinated seeds were of concern. Data were arcsine transformed to normalize variation. Germination, seedling development, and embryo vi ability data were analyzed using general linear model procedures and least square means using SAS v9.1 (SAS Institute, 2003). Results Effects of Cold-stratification on Seed Germination Source (F4 = 44.2, p < 0.0001), chilling period (F4 = 264.4, p < 0.0001), and their interaction (F16 = 22.4, p < 0.0001) were hi ghly significant for germination. Longer chilling
107 periods resulted in significantly higher germination percentages after 10 weeks culture with the exception of north central Florida (Figure 3-1). Lower germination was observed in unchilled seeds for all populations. In fact, no germination was observed in the control for South Carolina 2. In Michigan seeds, chill peri ods over 4 weeks promoted the highest seed germination. In both South Carolina populations chilling seeds for 6 or 8 weeks promoted the highest germination. No differences among chilling treatments were obser ved in north central Florida seeds, but germination of chilled seeds was significantly higher than unchilled seeds. Higher germination in south Florida seeds was observed following chi lling periods of 4 weeks or longer. In south Florida (86.8%) and South Caro lina 1 (86.4%) and 2 (88.1%), maximum germination was higher compared to 72.3% and 64.6% germination in Mi chigan and north central Florida seeds, respectively (Figure 3-1). Effects of Diurnal Temperatures on Seed Germination and Early Development Source (F5 = 410.4, p < 0.0001), temperature (F4 = 10.6, p < 0.0001), and their interaction (F20 = 5.08, p < 0.0001) were highly si gnificant for germination. In all populations, cultures incubated at a constant 25C reduced germin ation compared to at least one alternating temperature regime (Figure 3-2). Higher germin ation (>20%) in Michigan seeds occurred at temperatures below 33/24C with the excepti on of 25C (Figure 3-2A). Higher fluctuating temperatures promoted germination in all S outh Carolina populations. However, germination (<10%) was still low in South Carolina 1 and 2 populations regardless of temperature compared to all other seed sources (Figure 3-2B, C). Higher fluctuating temperatures (29/19C and 33/24C) promoted germination in north central Florida seeds (Figure 3-2E). Surprisingly, south Florida seed germination was highest at bot h 22/11C and 33/24C (Figure 3-2F). Among all populations, germination was highest in south Flor ida seeds, followed by north central Florida, South Carolina 3, Michigan, and South Carolina 1 and 2.
108 Source (F5 = 79.6, p < 0.0001), temperature (F4 = 9.83, p < 0.0001), and their interaction (F20 = 4.38, p < 0.0001) were highly significant for early development. Among seed sources, early development from weeks 1-8 was more a dvanced in Michigan (Figure 3-3A) and South Carolina 3 (Figure 3-3D) seedlings with the majo rity near or above stage 5. However, a large portion of embryos developed to stage 6 (Fi gure 3-3A, D). Above 22/11C development was more advanced in Michigan, South Carolina 3, and north Central Florida (Figure 3-3E). Although South Carolina 1 and 2 embryos develope d to stages 5 and 6 (Figure 3-3B, C), the average developmental index was less than 3 (Figure 3-3B, C). In fact, very few embryos among all populations developed to stages 5 and 6 (Fig ure 3-3). South Florida seedlings were more developed at temperatures below 33/24C (Figure 3-3F), but the majority of embryos remained in stage 1 (Figure 3-3F). Scanning Electron Microscopy Overall Michigan and south Florida seeds were the smallest (Figure 3-4). The testae on all South Carolina seeds had cracks a nd holes, but this was not seen in seeds from Florida and only a few from Michigan. On average South Carolina seeds were the largest, followed by north central Florida, Michigan, and south Florida. So uth Florida seeds were co nsistently rounder than other seeds, while other seeds we re more linear with tapered ends Testae on Michigan and South Carolina seeds were more collapsed than Florida seeds, which may have been caused by the gold sputtering process or storing seed below freezing temperatures. Histology Seed cross sections revealed several interesting anatomical features. Seeds from Michigan and South Carolina were more oval shaped comp ared to the rounder seeds from Florida. The embryo was not evident in the Michigan seed pos sibly due to a poor sample (Figure 3-5A). The embryo contained few cells and was disorganized in the South Carolina 1 seed (Figure 3-5B).
109 Embryos in South Carolina 2 (Figure 3-5C), nor th central Florida (Figure 3-5D), and south Florida (Figure 3-5E) were well organized. Star ch grains were eviden t in the north central Florida embryo (Figure 3-5D), wh ile dehydration was noticeable in the south Florida and South Carolina 2 embryos. Testae surrounding the Florid a seeds were generally one cell thick and thinner than other populations. The testa on the Michigan seed was approximately 15-30 m thick, 5-20 m on South Carolina 1 seed, 10-50 m on South Carolina 2 seed, and 5-10 m on both Florida populations. Discussion Beyond morphological variati on reported (Goldman et al. 2004 a), differences in seed germination and seedling development in response to diurnal temperatures and cold-stratification among C tuberosus populations provide further ev idence for ecotypic differentiation. Morphological variation in C. tuberosus correlating to geographic location was likely caused by different selection pressures and abiotic factors (Goldman et al. 2004a), but these selective pressures were not specifically explored with respect to ecotypic differentiation. Although photoperiod and nutrient availability were often concerns for succe ssful orchid seed germination (for review see Kauth et al. 2008), temperature should be include d as well. The present results along with results from Chapter 2 indicate that photoperiod may not be as influential as temperature for C. tuberosus seed germination and developm ent, especially for northern ecotypes. Cold-stratification Effects on Seed Germination Cold-stratification increased germination signi ficantly compared to the control among all seed populations. This is not su rprising since cold-stratification was reported to increase germination in numerous orchid species (De Pauw and Remphrey, 1993; Chu and Mudge, 1994; Tomita and Tomita, 1997; Miyoshi and Mii, 1998; Shimura and Koda, 2005; ien et al. 2008).
110 These reports were concerned with propagation and increasing germination, and not with the ecological aspect of chilling se eds with the exception of ien et al. (2008). However, the present results first report the effects of chilling length on several pop ulations of the same species. Although the exact mechanism of chilling in orchid seeds is not well-understood, coldstratification likely breaks physiological dorma ncy under favorable environmental conditions (Baskin and Baskin, 2001). Breaking dormancy a nd germination requirements are factors in timing of in situ germination (Baskin and Baskin, 2001) Although a constant temperature was used after the chill peri od, previous research has indicated th at cold-stratification may remove a requirement for fluctuating temperatures (van Assche et al. 2003). Calopogon tuberosus seeds exhibited some type of dormancy since chilling seeds increased germination significantly, and the de gree of dormancy may differ among populations due to differences in testae thickness. Embryos may not be completely dormant, but rather the testa may be inducing physiological dormancy (Lauzer et al. 2007). Once removed from the testa, embryos may germinate readily as reported for Aplectrum hyemale (Lauzer et al. 2007). Delayed or non-existent germination may be caused by a high concentration of phenolic compounds in the testa, which reduces th e permeability of the testa (Thompson et al., 2001). Chilling seeds may release harmful phenolics by increasing the permeability of the testa (Thompson et al. 2001). South Carolina seeds required longer th an 6 weeks chilling to obtain maximum germination, while Michigan and south Florid a seeds required 4 weeks or longer, and north central Florida seeds required 2 weeks or l onger. The longer chilling requirement in South Carolina seeds may be the influence of the micr o-niche habitats that are mesic in spring and summer and xeric in fall and winter. Seeds likel y require a longer ch illing period to break
111 prolonged dormancy and germinate when the mesic environment and warmer temperatures return in spring. The degree of dormancy is also likely diffe rent among populations since Michigan seeds did not germinate to the maxi mum percentages observed in other populations. Seeds from northern populations may require longer chilling pe riods that delay germination until late winter or spring and break dormancy (Meyer et al. 1995; Allen and Meyer, 1998). Although nonchilled C. tuberosus seeds from Michigan did germinate, these seeds likely germinated quickly to form corms immediately in order to survive winter. South Florida seeds also required coldstratification to promote maximum germination. Southern populations may require coldstratification since fluctuating winter temperat ures may lead to early seedling emergence and subsequent death from colder temperatures (Fowler and Dwight, 1694; Schtz and Milberg, 1997). The required chilling period may ensure that seeds germinate only upon experiencing temperatures (above 22/11C) that do harm em erged seedlings. Nondormant seeds from south Florida germinate slowly with little embryo de velopment indicating that autumn emergence is unlikely (Meyer et al. 1995). South Florida seeds likely are not exposed to a long chilling period in situ but long chilling periods lead to a breakdo wn of chilling cue resistance so seeds can germinate (Meyer, 1992). Germination of chilled seeds exceeded the tested embryo viability in all populations, but more so in the northern C. tuberosus populations. Seeds from thes e populations may require a longer treatment in CaOCl2 than the 3 h period to further degr ade the testa. The thicker testae observed on seeds from northern populations may be a result of harsher climates during winter months providing extra protection for the embryo. Longer chilling periods may aid in weakening
112 the thicker testae of northern popul ations leading to increased ge rmination. The well-organized embryos of Florida seeds may also lead to high er germination percentages in unchilled seeds. Diurnal Temperature Effects on Seed Germination The highest germination percentages were ach ieved under fluctuating diurnal rather than the constant temperatures. Since consta nt temperatures are not normally found in situ (Baskin et al. 2006), the decrease in germination observed at 25C is not surprising. Previous research with non-orchids indicated that fluctu ating temperatures promoted ge rmination more than constant temperatures (Thompson et al., 1977; Thompson and Grime, 1983; Probert et al. 1986). Although seed germination of C. tuberosus still occurred at 25C, the presence of light may have removed the requirement for fluctuating temp erature (Thompson and Grime, 1983). Given that temperature fluctuations occur more often at the soil surface, increased germination under fluctuating temperatures may represent a mechan ism to prevent germination of deeply buried seeds (Thompson and Grime, 1983). Constant temperatures increased germination in Dactylorhiza majalis which was attributed to germinati on and mycorrhizal fungi infection at deeper soil depths (Rasmussen and Rasmussen, 1991). However, D. majalis seeds exposed to fluctuating temperatures where not exposed to light, which may have enhanced germination (Toole et al. 1955). Seeds near the soil surface are expos ed to fluctuating temperatures, light, and higher nitrate concentrations than seeds buried deeper potentially leading to higher germination (Roberts and Benjamin, 1979). In Michigan, C. tuberosus flowers into July with seed dehiscence occurring into September. Higher in vitro germination at 22/11C and 29/19C corresponds to late or early summer and early fall germination in situ Seedling development was more advanced at temperatures over 22/11C, which corresponds to summer development. Germination of Michigan seeds was much faster than other populations and corm development occurred within
113 6-8 weeks from initiation of germination. The presen t results also reaffirm that Michigan seeds germinate and develop to advanced stages more quickly than other seed sources. The onset of lower temperatures, which happens rapidly du ring September in Michigan, may cue seed dormancy or seedling leaf senescence. Seeds that do not germinate and de velop storage organs may require cold-treatment to germinate the following spring. Once seed s germinate in spring, seedling development may be enhanced at hi gher temperatures. Determining when seeds germinate in nature may be elucidated by conducting in situ germination studies. All South Carolina populations germinated best at higher temperatures. However, germination was still low in S outh Carolina 1 and 2 populations, which may have been caused by physiological dormancy. Populations from South Carolina 1 and 2 are located in micro-island habitats with shallow soil overlaying granite in close proximity to seasonal flowing water (Porcher and Rayner, 2001). Seeds may likely only ge rminate at higher temperatures because of the micro-niche habitat in these areas. Since the so il is shallow, colder temperatures may impact soil conditions more than areas with deeper soils. Also, since gr anite is found directly below the soil, the cataract bogs may be exposed to larger temperature fluctuations. Seeds in north central Florida likely germinate in late summer or spring as indicated by the maximum in vitro germination at 29/19C a nd 33/24C. Seedling development was more advanced at 27/15C and 29/19C corresponding to spring and fall developm ent. Since seeds in north central Florida are subjec t to a longer growing season than Michigan and South Carolina, thus fall development is likely. Seeds from north central Florida that germinate in vitro developed relatively fast with corms forming af ter 12 weeks culture, ensuri ng that seedlings have adequate storage organs to surv ive winter. Viable seeds that do not germinate in fall likely germinate and develop the following spring.
114 Germination of south Florida seeds wa s highest at 22/11C and 33/24C, while development was lowest at 33/24C. This indicated that most germination may take place in the cooler winter months or warm er summer months, and that seeds may have coldor warmstratification requirements. Although germination at 33/24C was equal 22/11C, development was suppressed at 33/24C. Warmer temperatures may place physiological stress on smaller embryos and seedlings, leading to embryo and seedling development under cooler temperatures in late winter. In South Florida, C. tuberosus seedlings from the previous growing season reemerge in mid-to-late February when temperatures are not cold enough to damage seedlings (see Appendix A). Higher germination at 25C compared to other popul ations may be a consequence of burial. Seeds may not germinate at the soil surface under higher temperatures, but rather germinate slowly underneath th e soil and emerge under cool er temperatures (Probert et al. 1985b). Conclusions These results provide insight into the nature of orch id ecology and physiology, and specifically germination requirements within a widespread species. Along with our previous results, these results provide further evidence of ecotypic differentiation for temperature within the geographic range of C. tuberosus The results of both experiments indicates that C. tuberosus seeds may germinate before winter months and spring, and those seeds that germinate in spring require some period of chilling. However, ch illing period and optimum temperatures differ depending on seed source, and local environmen tal cues promote germination when conditions are optimal for seedling survival (Bischoff et al. 2006). Given that cold-s tratification increased germination significantly, chilling seeds is an ab solute requirement to break seed dormancy in Michigan and South Carolina seeds. Once dor mancy is broken, seeds may become more
115 responsive to wider environmental conditions such as photoperiod; thus, the results from Chapter 2 may be different with chilled seeds. When interpreting in vit ro data, it must be scrutinized closely since in vitro common garden studies like other controlled common gard en studies can oversimplify the interaction among environmental factors. Under controlled environments, environmental conditions are controlled for to simplify complex interactions that would otherwise confound the experiments. However, environmental factors do not act alone, but photoperiod, soil nutrients, and temperature all interact and form a comple x network directing eco typic differentiation. In vitro seed germination is only one tech nique that can be utilized to differentiate ecotypes, although it is a reliable method. Combining in vitro results with in situ field germination data may provide more understanding into ecotypic di fferentiation since conditions e xperienced in the field differ from those in vitro .
116 Figure 3-1. Effects of chilling seeds at 10C in darkness on germination of Calopogon tuberosus seeds from distant populations. Data was collected after 10 weeks culture. Cultures were incubated under a 12 h photoperiod at 24C following the chilling treatment. Histobars (mean response of 5 replications w ith 3 subreplications each) with the same letter are not significantly different ( =0.05).
117 Figure 3-2. Diurnal temperature effects on germ ination and development of unchilled seed and Calopogon tuberosus seeds from different populations Data was collected after 8 weeks culture under a 12 h photoperiod. A) Michigan population from Carney Fen. B) South Carolina population from Ashmore Heritage Preserve. C) South Carolina population from Site C. D) South Carolin a population from Eva Chandler Heritage Preserve. E) North central Florida populat ion from Goethe State Forest. F) South Florida population from the Florida Pant her National Wildlife Refuge. Histobars (mean response of 5 replications with 3 s ubreplications each) within each seed source with the same letter are not significantly different ( =0.05). See Table 2-1 for stages of development.
118 Figure 3-3. Developmenta l index of unchilled Calopogon tuberosus seeds from widespread populations. Seeds were incubated under di urnal temperatures for 8 weeks under a 12 h photoperiod. A) Michigan population fr om Carney Fen. B) South Carolina population from Ashmore Heritage Preserve. C) South Carolina population from Site C. D) South Carolina population from Eva Chandler Heritage Preserve. E) North central Florida population from Goethe St ate Forest. F) South Florida population from the Florida Panther National Wildlif e Refuge. Histobars (mean response of 5 replications with 3 subreplica tions each) within each seed source with the same letter are not significantly different ( =0.05). Developmental index was calculated as follows: DI = (N1 + N2*2 + N3*3 + N4*4 +N5*5 + N6*6)/ (N1 +N2 + N3 + N4 + N5 + N6) where N1 is the number of s eeds in stage 1, etc.
119 Figure 3-4. Comparative scanning electron microscopy of seeds from widespread populations of Calopogon tuberosus A) Michigan population from Carney Fen. Scale bar = 150 m at 200x. B) South Carolina population from Ashmore Heritage Preserve. Scale bar = 120 m at 250x. C) South Carolina population from Site C. Scale bar = 200 m at 150x. D) South Carolina population from Eva Chandler Heritage Preserve. Scale bar = 200 m at 1500x. E) North central Florida population from Goethe State Forest. Scale bar = 167 m at 180x. F) South Florid a population from the Florida Panther National Wildlife Refuge. Scale bar = 167 m at 180x.
120 Figure 3-5. Light micrograph cross sections of mature Calopogon tuberosus seeds from widespread populations. A) Michigan seed at 40x. The lack of embryo is possibly due to a poor sample. B) South Carolina 1 seed at 40x. C) South Carolina 2 seed at 20x. D) North Central Florida at 40x. E) South Florida at 40x. T = testa; Em = embryo; SG = starch grains. Scale bars = 50 m.
121 CHAPTER 4 COMPARATIVE SEEDLING BIOMASS AL LOCATION AND CORM FORMATION AMONG WIDESPREAD Calopogon tuberosus POPULATIONS Introduction A similar trend was observed throughout the ex periments presented in Chapters 2 and 3. Regardless of seed germination treatment ( photoperiod, germination media, temperature), Michigan seeds germinated quick ly and seedlings developed corm s before all other populations followed by South Carolina, north central Florida, and south Florida. Th e growing conditions at each population are markedly different with a shor t growing season and long winter in Michigan and a long growing season and short winter in south Florida. Thus timing of corm formation and differences in biomass allocation may be influen ced by length of growing season, which in turn may be responsible for ecotypic differentiation. Widely distributed plant species evolved th e ability to survive wide environmental conditions leading to local adaptation to biot ic and abiotic conditions (Linhart, 1995; Joshi et al. 2001; Sanders and McGraw, 2005). Local adaptation in plants was first examined using common garden studies by Turesson (1922 a), who first used the term ecotype, and by Clausen et al. (1941) using reciprocal transplant studies. Th ese studies showed that local adaptation to environmental conditions, such as altitude and temperature, were infl uencing differences in growth and development of plan t species. In recent years, th e use of locally adapted plant material for restoration purpos es was highlighted for the purpo se of maintaining ecosystem function and stability because non-locally adap ted ecotypes can reduce plant population fitness (Linhart and Grant, 1996; Hufford and Mazer, 2003; McKay et al. 2005). Local adaptation has been st udied in numerous species through common garden and reciprocal transplant experiments (Nuismer and Gandon, 2008). Common garden studies test for adaptation and fitness of individu als from local or distant hab itats in a common environment,
122 while transplant studies examine the role of non-local conditions on adaptation and fitness. Common garden studies may more efficiently test the genetic contribution to fitness while minimizing environmental impacts on fitness (Nuismer and Gandon, 2008). Transplant studies may better estimate environmental variation since i ndividuals are transplanted to habitats with environmental conditions not experienced in the natural habitat (Nuismer and Gandon, 2008). Local adaptation can be studied by examini ng performance of ecotypes under different photoperiods (Howe et al. 1995; Kurepin et al. 2007), temperatures (Seneca, 1972; Probert et al. 1985b), and soil regimes (Grze 2007; Sambatti and Rice, 2007). Differences in biomass allocation were propos ed as an important aspect of ecotypic differentiation. Northern ecotypes of Spartina alterniflora allocated more biomass to underground organs including roots and rhizom es (Gallagher, 1983; Gallagher and Howarth, 1987; Gross et al. 1991). Greater biomass allocation to underground organs in northern ecotypes of several species was due to a shor ter growing season (Potvin, 1986; Sawada et al. 1994; Kane et al. 2000; Liancourt and Tielbrger, 2009) and a higher allocation of carbohydrate reserves to overwintering structures (Mooney and Billings, 1960). Biomass allocation was also correlated with various reproductive strategi es in ecotypes. Ecotypes found in fields or areas of younger succession allocated more biomass to reproductive organs than those in wooded habitats that allocated more biomass to vegetative struct ures (Abrahamson, 1975, 1979). Marsh plants that occupied areas of greater dist urbance allocated more biomass and carbohydrate reserves to underground storage organs (Sun et al. 2001; Peas-Fronteras et al., 2009) Common garden and transplant studies can be performed in greenhouses, growth chambers, natural habitats, and outdoor plots (Gallagher et al. 1988; Howe et al., 1995; Majerowicz et al. 2000; Suzuki, 2008), but obtaining permits to collect and transplant protected,
123 rare, threatened, or endangered species, as many or chids are, is difficult. Seeds can be used to produce mature plants for common garden and transplant studies. While this may be an effective method for quick-growing species, orchids often require four or more years to flower from initial seed germination (Stout amire, 1964). Additionally, in situ orchid seed germination is difficult and time consuming since germination is often low (Brundrett et al. 2003; Zettler et al. 2005b; Diez, 2007). Alternatively, in vitro techniques can be used to study environmental requirements for orchid seed germination (Kauth et al. 2008) as well as seedling growth and development (Dijk and Eck, 1995b). Many in vitro culture techniques can be gr ouped under the discipline of in vitro ecology. In vitro ecology has been previously defined to in clude environmental and exogenous factors (i.e., temperature, light, gas phase, culture media) that affect in vitro growth and development (Hughes, 1981; Williams, 2007). Here, we further define in vitro ecology to include the evaluation and use of in vitro culture techniques to identify, pr opagate, evaluate, and select plant genotypes and ecotypes for ecol ogical purposes. Specifically, in vitro ecology studies can be used to correlate environmental and genetic vari ables that affect plant growth and development in vitro with ecological factors aff ecting growth and development in situ In vitro ecology could also be used to assess ecotypic differentiation for habitat restoration and plant reintroduction programs by conducting in vitro common garden studies under controlled environmental conditions. Since this use of in vitro ecology is a new area of rese arch its validity must be verified. Based on morphological variation, Goldman et al. (2004a) defined three distinct geographic areas of Calopogon tuberosus : northern plants in gl aciated areas, southwest populations west of the Missi ssippi Embayment, and southe ast populations east of the
124 Mississippi River and sout h of the glaciated zone. However, Goldman et al. (2004a) did not classify C. tuberosus ecotypes, but stated that variation in C. tuberosus could be caused by environmental conditions. Further ecotypic differen tiation has not been previously explored in C. tuberosus Although morphological and gene tic variation exists in C. tuberosus all plants throughout its range form corms. Differences in biomass allocation among C. tuberosus populations were reported in chapter 2. Howe ver, a detailed timecourse comparison for C. tuberosus seedling development has not been reporte d, and little information exists regarding this area of research for ecotypes in general. Evaluation of in vitro seedling development with regard s to biomass allocation and timing of corm formation from several C. tuberosus populations from divers e geographic sources might clarify the extent of ecotypic di fferentiation across its range. Seedling development in relation to the effects of photoperiod, germination media, a nd temperature was examined in Chapters 2 and 3. However, a detailed comparison of seedling bi omass allocation and corm formation was not examined over several weeks. The objectiv es were to: 1) Validate the use of in vitro common garden studies to differentiate ecotypes; 2) Examine differences in the timing of corm formation among widespread C. tuberosus populations; and 3) Compare seedling biomass allocation among widespread C. tuberosus populations. Materials and Methods Seed Source Seeds were collected from the following locations: Carney Fen (Menominee County, Michigan, USA), Eva Chandler Heritage Preserve (Greenville County, South Carolina, USA), Goethe State Forest (Levy County, Florida, US A), Florida Panther National Wildlife Refuge (Collier County, Florida, USA). Seed lots fr om Ohio, Oklahoma, and Central Florida were obtained from cultivated plants that were hand pollinated under greenhouse conditions (D.
125 Goldman). Seeds from greenhouse plants were used to determine if germination and development would still be different after years of ex situ cultivation. Seed capsules from all populations were collected before complete dehi scence and were stored at 23C over silica gel for 2 weeks. Seeds were then removed from capsu les, pooled by geographic source, and stored in complete darkness at -11C until used. Seed Germination and Medium Preparation Seeds were surface disinfected in sterile scintilla tion vials for 3 minutes in a solution of 5 mL absolute ethanol, 5 mL 6% NaOCl, and 90 mL ster ile distilled-deionized (dd) water. Seeds were rinsed with sterile dd water afte r surface sterilizati on, and solutions were removed with sterile Pasteur pipettes. Seeds were tr ansferred with a sterile inocul ating loop to BM-1 Terrestrial Orchid Medium ( Phyto Technology Laboratories, Shawnee Mission, KS, USA) in 100 x 15 mm Petri plates (Fisher Scientific Pittsburgh, PA, USA). The medi um was supplemented with 1% activated charcoal. Medium pH was adjusted to 5.7 with 0.1N KOH prior to autoclaving for 40 minutes at 117.7 kPa and 121C. Ten replicate Petr i plates with 30 mL medium each were used for each seed source with approximately 100 seed s per plate. Cultures were placed in an environmental growth chamber (# I-35LL; Percival Scientific, Pe rry, IA, USA) under cool-white fluorescent lights in a 12/12 hr photoperiod at 24.2 0.2C and a light level of 40 mol m-2 s-1. Seedling Transfer and Data Collection After 6 weeks culture seedlings were transferred from Petri plates to PhytoTech Culture Boxes ( PhytoTechnology Laboratories) containing 100 mL of BM-1 medium. Medium was prepared as described previousl y. Uniform-sized seedlings with developing leaves were then transferred to individu al culture boxes. Three PhytoTech Culture Boxes with nine seedlings each were prepared per seed s ource per week. A total of 21 Phyto Tech Culture Boxes were prepared
126 per seed source. Cultures were completely ra ndomized within the growth chamber under the same conditions previously described. Data were collected bi-weekly on three replicate Phyto Tech Culture Boxes containing nine seedlings each per seed source. Data were taken on 27 seedlings per seed source each week. Data for week 10 South Carolina seedlings were collect ed on two replications due to contamination of one replicate. Also, data was collected on two replications of Oklahoma and Ohio from weeks 12-20 due to the limited number of seeds that germinated and developed into seedlings.The following data were collected: shoot length, root number, root length, co rm diameter, and dry weight. Shoot, root, and corm dry weights were measured after ti ssues were dried for 24 hours at 60C. Seedling percent biomass allocation was determined by dividing corm, root, and shoot weights by the total seedling weig ht. Shoot length, root number, root length, corm diameter, and biomass data were statistically analyzed using general linear procedures, ANOVA, and Tukeys HSD test at =0.05 in SAS v9.1 (SAS Institute, 2003). Regression and Pearsons correlation analyses were performed on corm biomass allo cation and growing season length reported in Table 1-1. Corm biomass allocation data were arcsine transformed prior to regression analysis. Results Corm Formation Timing of corm formation di ffered significantly by source (F6 = 48.2, p < 0.0001), week (F6 = 80.9, p < 0.0001), as well as source by week (F36 = 8.97, p < 0.0001). Corm formation on Michigan seedlings was evident by week 8, w eek 10 on South Carolina, Ohio, and Oklahoma seedlings, week 14 on north central Florida seedlings, week 16 on central Florida seedlings, and week 18 on south Florida seedlings (Table 4-1). Initial mean corm diameter on Michigan and South Carolina seedlings was similar until week 16. Mean corm diameter was similar in Michigan and south Florida seedlings, but south Fl orida seedlings continue d to grow after week
127 20 (personal observation). Mean corm diameter was largest on South Caro lina, north central and central Florida seedlings at week 20. Ohio seedli ng corms were only smaller than north central Florida seedling corms, while co rms on Oklahoma seedlings were only larger than Michigan and south Florida corms by week 20 (Table 4-1). Shoot Length Shoot lengths were significan tly different among sources (F6 = 266.7, p < 0.0001), week (F6 = 256.9, p < 0.0001), and source by week (F36 = 45.4, p < 0.0001). Initial shoot lengths on Michigan and South Carolina seedlings were larger than both Florida populations (Table 4-2). After week 12, mean shoot length on Michigan seedlings were the shortest of all seedlings. Shoot growth on Michigan seedlings did not sign ificantly increase from week 8 to 16, but did decrease significantly there afte r. Similarly, shoot growth did not increase significantly on South Carolina seedlings from week 12 to 20. Shoots on north central and central Florida seedlings were the largest by week 14, and were the largest by week 20 as well. Shoot growth on Ohio and Oklahoma seedlings was similar to South Carolina Shoot growth on south Florida seedlings was initially small, and only north central Florida seedlings exceeded mean shoot length of south Florida seedlings at week 20 (Tab le 4-2). Although leaf width m easurements were not collected, leaves on Oklahoma seedlings were very thin compared to all other seedlings (personal observation). Shoot senescence, characterized by yellowing and browning of leaves, began on Michigan seedlings after 16 weeks culture and by week 20 almost 100% of shoots senesced (Figure 41M). Shoot senescence was obser ved at week 20 on Ohio seedlings. Shoot senescence was delayed in southern populations, with no sene scence occurring by week 20 on any other seedling source.
128 Root Length and Number Source (F6 = 160.9, p < 0.0001), week (F6 = 105.8, p < 0.0001), and the interaction between source and week (F36 = 19.4, p < 0.0001) all significantly influenced root number. By week 18, north central and central Florida seedlings had the most r oots and this was seen at week 20 as well (Table 4-3). Michigan seedlings ha d the lowest number of roots throughout the experiment, and root number declined after week 16 due to root die-back. Root development was similar on Ohio and Oklahoma seedlings through out and similar to South Carolina and south Florida seedlings by week 18 and 14, respectively (Table 4-3). Root length was significantly influenced by source (F6 = 161.1, p < 0.0001), week (F6 = 209.7, p < 0.0001), and source by week (F36 = 16.8, p < 0.0001). At week 20, the longest roots were observed on north central and central Florida seedlings, and th e shortest roots on Michigan seedlings (Table 4-4). By week 14, mean root length was longest on north central and central Florida seedlings. Root elongation was similar on Ohio, Oklahoma, South Carolina, and south Florida seedlings after week 16 and continued until week 20. Biomass Allocation ANOVA results revealed that percent bioma ss allocation to shoots, corms, and roots differed significantly among seed sources (Table 4-5). Corm biomass allocation was inversely related to latitude with the highest allo cation being observed on Michigan seedlings. Approximately 97% biomass was allocated to co rms in Michigan seedlings by week 20, which was significantly higher than the 77% in South Carolina seedlings, 74% in Ohio and Oklahoma seedlings, 53% in north central Florida seedlings 35% in central Florida seedlings, and 7% to corms in south Florida seedlings (Table 4-6). By week 8, grea ter corm biomass allocation was evident on Michigan seedlings compared to all other populations. Higher corm biomass allocation compared to shoot a nd root biomass allocation was evident on Michigan seedling by
129 week 10. Corm biomass allocation on South Carolina, Ohio, and Oklahoma seedlings was significantly greater than all Florida populations after 20 weeks culture (Table 4-6). Allocation to corms was lowest in south Florida seedlings followed by central and north central Florida seedlings. Ohio, Oklahoma, and central Florida donor plants were under cultivation for more than 10 years, but still main tained genetic identity. Percent shoot biomass allocation declined as greater biomass was allocated to corms regardless of population (Table 4-6). However, s hoot biomass allocation was significantly higher on south Florida seedlings than all other seedling sources. Shoot biomass allocation of Ohio and Oklahoma seedlings was generally greater than Michigan, similar to South Carolina, and less than all Florida populations. Michigan seedlings had the lowest shoot biomass allocation. South Florida seedlings allocated more biomass to shoot s compared to roots an d corms over the entire 20 week period. Root biomass allocation was si gnificantly higher on central Florida seedlings compared to all other populations, while root biomass was lowest on Michigan seedlings. Root biomass allocation on Ohio, Oklahoma, and South Carolina seedlings was similar throughout the experiment and was higher than Michigan seedli ngs, but lower than all Florida seedlings (Table 4-6). Correlation analysis revealed a strong negative correlation between corm biomass allocation and growing season le ngth: higher corm biomass a llocation was correlated with shorter growing season lengths. Gr owing season was considered the number of days between the first spring and last fall frost. Pearsons correl ation coefficients (all p values < 0.0001) were as follows: -0.73 (all weeks), -0.67 (week 8); 0.81 (week 10); -0.87 (week 12); -0.93 (week 14); 0.96 (week 16); -0.91 (week 18); -0.95 (week 20). Regr ession analysis also revealed a negative trend for all weeks (Figure 4-2). With the exception of week 8 a nd 10, regression models
130 accounted for much of the data variance with strong r2 values over 0.75 (Figure 4-2). Due to the lack of corm formation in week 8 and 10 data, r2 values were not as st rong. When weekly data were combined the r2 was 0.54, but the model was significant. Discussion This represents the first application of in vitro ecology to assess the extent of ecotypic differentiation of a latitudinall y widespread orchid species through studying biomass allocation. Although information relating timing of biomass a llocation to ecotypic differentiation is scarce (Gallagher, 1983; Gallagher and Howarth, 1987; Gross et al., 1991; Seliksar et al. 2002; Yoshie, 2007), timing of corm formation is an importa nt factor in the ecotypic development of C. tuberosus Few reports exist that utilize in vitro techniques to correlate ec otypic life history traits with in vitro growth strategies of orchids (Dijk and Eck, 1995 a). The present results also show the potential use of in vitro common garden studies to detect unique growth strategies. In particular biomass allocation in C. tuberosus ecotypes is most likely influenced by growing season length. Biomass allocation dynamics and storage organ function have been previously described in situ for single orchid populations (Whigham, 1984; Snow and Whigha m, 1989; Zimmerman and Whigham, 1992; Tissue et al. 1995; ien and Pederson, 2003, 2005). However, biomass allocation in orchids has not been explored with respect to ecot ypic differentiation. In the present study, C. tuberosus biomass allocation to corms ranged from 7% to 97%, depending on seed source. Whigham (1984) reported near ly 80% of biomass in a single Tipularia discolor population was allocated to underground storag e organs. Zimmerman and Whigham (1992) reported that 61% and 66% of the total non-stru ctural carbohydrates we re allocated to the youngest corms in vegetative and do rmant plants, respectively. In a detailed analysis of biomass allocation in T. discolor 66% of the total biomass was allocated to corms during fruit maturation
131 and 80% during leaf senescence (Tissue et al., 1995). These data are comparable to C. tuberosus since more biomass was allocated to corms just prior to and during leaf senescence. Although carbohydrate analysis of C. tuberosus was not investigated, rea llocation of carbohydrates from leaves to corms might explain incr eased corm biomass allocation in C. tuberosus as was similarly reported for Dactylorhiza lapponica tubers (ien and Pederson, 2005). Regardless of orchid species storage organs such as corms represent ecological adaptations to ensure survival during unfavorable growing conditions. In T. discolor corms are vital to support growth and reproducti on (Zimmerman and Whigham, 1992; Tissue et al. 1995), and serve as sinks for nutrient reserves (Whi gham, 1984). Corms may also aid in long term survival by protecting the shoot meristem duri ng periods of stress (Whigham, 1984). Greater and faster biomass allocatio n to underground organs in northern C. tuberosus ecotypes followed a similar trend to ecotypes of Spartina alterniflora (Gallagher, 1983; Gallagher and Howarth, 1987; Gross et al., 1991) and Sagittaria latifolia (Kane et al. 2000; Kane et al., 2003). The faster biomass allocation to corms in C. tuberosus is likely a selection pressure favored by a shorter growing season in more northern latitudes as reported with ecotypes of S. alterniflora (Seliksar et al. 2002), Plantago asiatica (Sawada et al., 1994), grass species (Potvin, 1986; Liancour t and Tielbrger, 2009), and Eriophorum vaginatum (Fetcher and Shaver, 1990). Northern ecotypes of C. tuberosus may allocate larger carbohydrate reserves in storage organs to survive winter conditions, and subsequently r eallocate those carbohydrates to rapid growth the following spring (Seliksar et al. 2002). Greater corm biomass in northern C. tuberosus ecotypes could be influenced by faster reallocation of carbohydrates from shoots to corms leading to faster shoot senescence compared to southern ecotypes (Mooney and Billings, 1960). Further investigation may also determine wh ether northern ecotypes are more tolerant to
132 freezing temperatures due to higher corm carbohydrate reserves (Hofgaard et al., 2003; Shahba et al. 2003), and thus are better able to survive in colder climates. A short life cycle from initi al shoot production to shoot senescence as well as low temperature tolerance is an adaptation to north ern environments where the growing season is short (Potvin, 1986). Even under the same environmental conditions in vitro, northern C. tuberosus ecotypes expressed a shorter growth cycle and faster corm biomass allocation. Since seeds were collected directly from wild popul ations, pre-conditioned environmental carry-over effects may have explained this adaptation. It is interesting to note that seedlings from Ohio, Oklahoma, and central Florida were morphologically different even after the seed donor plants had been in greenhouse culture for more than 10 years (D. Goldman, personal communication). A long-term genetic adaptation to shorter growin g seasons may also explain the differences in development (Shaver et al. 1986), and plants from northern latitudes may always express the characteristics of a shorter life cycle and greater corm biomass allocation regardless of environmental conditions. The adaptation may also be a consequence of primary productivity, and plants from northern latitudes may not be able to take advantage of increased temperatures or constant growing conditions (Fetcher and Shaver, 1990). Greater biomass to corms may also represen t a successful survival strategy under periods of environmental stress such as flooding. The populations used in the present study from Michigan and South Carolina have long periods of water availabili ty in the form of ground water (Nelson, 1986; Cohen and Kost, 2008), while popul ations in Florida ex perience distinct dry seasons (Davis, 1943). Ecotypes in areas prone to flooding allocated more biomass and carbohydrates to corms and tubers indicating a vegetative growth strategy (Li et al. 2001; Sun et al. 2001; Peas-Fronteras et al., 2009). Higher biomass to underground storage organs may be a
133 response to prolonged flooding when plants wo uld need a readily available source of carbohydrates (Peas-Fronteras et al. 2009). Growth differences may be related to reproductive strategy as well. Florida populations in the pres ent study produce more flowers and seed capsules than the plants in Michigan and South Caroli na, which may lead to higher seed production (Peas-Fronteras et al. 2009). Higher seed production may be necessary in order to colonize areas of earlier succession such as prairies and non-wooded areas in so uth Florida (Abrahamson, 1975, 1979). Differences in root number, length, and biomass of C. tuberosus ecotypes may be related to soil nutrient and water availa bility. Biomass allocation to root s was greater in several annual plant species and Populus davidiana ecotypes under low nutrient and water stressed soils (McConnaughay and Coleman, 1999; Zhang et al. 2005). Massachusetts ecotypes of S. alterniflora were found to have shorter roots due to th e shallow, organic soils compared to the deeper sand-based soils in Georgia (Seliksar et al. 2002). Longer or deeper roots on southern C. tuberosus ecotypes may be an adaptation to water-st ressed environments where the upper soil layers are less hydrated (Kondo et al. 2003). Shoot biomass as well as shoot length on C. tuberosus was highest in Florida populations that experience higher growing temperat ures. The larger shoots on Florida C. tuberosus seedlings may be a selection pressure to maximize phot osynthesis to outcompete vegetation during a longer growing season (Gallagher and Howart h, 1987). A higher shoot biomass may be a requirement to reach reproductive size to set s eed before adverse environmental conditions are experienced (Rice et al. 1992). Faster shoot growth in Michig an seedlings may be due to earlier carbohydrate allocation.
134 Common garden studies are useful tools to detect local adap tation influenced by genetics, but often disregard the relative im pact of environmental conditions in situ depending on source (Nuismer and Gandon, 2008). Transplant and recipr ocal transplant studi es better indicate environmental effects on local adaptation (N uismer and Gandon, 2008). Cultural conditions in vitro can be controlle d to represent in situ conditions by controlling environmental conditions experienced across a species di stribution such as photoperiod, te mperature, and humidity. Using a captive generation of seeds may be necessary to further investigate the ro le of environment and genetics on local adaptation. Growing C. tuberosus seedlings in vitro under different temperatures or photoperiods may lead to differe nt results. However, biomass allocation in the South Carolina and Florida populations were una ffected by different photoperiods after 16 weeks culture (Figure 2-4). Greater bi omass allocation to shoots occurre d in Michigan seedlings under short days, while neutral and long days promoted higher corm biomass a llocation (Figure 2-4). Conclusions Screening for ecotypic differentiation is an exciting application for in vitro culture, and in vitro common garden studies can be effective at de tecting different growth strategies. Biomass allocation and corm formation results along with data from Chapters 2 and 3 provide strong evidence that, based on the differences in biom ass allocation and timing of corm formation, C. tuberosus ecotypes exist. The ecological adaptations presented are likely influenced by differences in growing season length influen ced by both photoperiod and temperature. Due to shorter growing seasons prior to winter dormanc y, seedlings from northern populations displayed an adaptation for accelerated corm initiation and development under in vitro conditions. Conversely, seedlings from southern populations de monstrate a delay in corm formation possibly an adaptation to a longer growing season.
135 Table 4-1. Comparative changes in mean corm diameter of Calopogon tuberosus seedlings of different geographic sources during 20 weeks in vitro culture. Means with the same letter among populations by week are not signi ficantly different according to Tukeys HSD test at =0.5. All units in mm. Michigan Ohio Oklahoma South Carolina North Central Florida Central Florida South Florida Week 8 1.42 a 0 b 0 b 0 b 0 b 0 b 0 b Week 10 1.73 a 2.09 a 1.85 a 1.78 a 0 b 0 b 0 b Week 12 2.11 c 3.20 a 2.69 b 2.18 c 0 d 0 d 0 d Week 14 2.37 ab 3.23 a 2.40 ab 2.75 ab 2.05 b 0 c 0 c Week 16 2.60 b 3.37 ab 3.29 ab 3.90 a 3.59 a 2.63 b 0 c Week 18 2.72 cd 4.04 a 3.17 bc 4.18 a 4.49 a 3.73 ab 2.34 d Week 20 2.77 d 4.60 b 3.62 c 4.76 ab 5.48 a 4.88 ab 2.59 d
136 Table 4-2. Comparative change in mean shoot length of Calopogon tuberosus seedlings of different geographic source during 20 weeks in vitro culture. Means with the same letter among populations by week are not signi ficantly different according to Tukeys HSD test at =0.5. All units in mm. Michigan Ohio OklahomaSouth Carolina North Central Florida Central Florida South Florida Week 8 13.4 abc 14.8 a 11.9 bc 13.8 ab 10.6 cd 8.15 d 5.04 e Week 10 14.4 de 31.9 b 44.9 a 29.0 b 22.4 c 18.9 cd 9.56 e Week 12 14.7 d 37.4 c 63.3 a 50.2 b 40.0 b 35.8 c 22.8 d Week 14 14.9 d 37.9 c 56.4 b 52.2 b 74.4 a 62.6 ab 29.3 c Week 16 11.7 c 38.1 b 46.6 b 50.0 b 111.1 a 93.4 a 41.8 b Week 18 6.00 c 41.7 b 46.7 b 49.9 b 137.2 a 128.4 a 52.4 b Week 20 2.19 d 47.9 c 49.6 c 48.0 c 131.9 a 142.4 a 90.7 b
137 Table 4-3. Comparative change s in mean root number of Calopogon tuberosus seedlings of different geographic source during 20 weeks in vitro culture. Means with the same letter among populations by week are not signi ficantly different according to Tukeys HSD test at =0.5. Michigan Ohio OklahomaSouth Carolina North Central Florida Central Florida South Florida Week 8 1.22 ab 1.14 ab 1.35 a 1.33 ab 1.04 ab 1.00 b 0.33 c Week 10 0.96 d 2.33 b 2.19 b 2.83 a 2.33 b 1.59 c 0.56 d Week 12 1.33 b 2.56 a 2.72 a 2.52 a 2.56 a 1.78 b 1.37 b Week 14 1.07 d 2.00 c 1.78 c 2.74 ab 3.00 a 2.78 a 2.15 bc Week 16 1.04 e 2.18 cd 2.06 d 2.89 bc 3.85 a 3.63 ab 2.22 cd Week 18 0.82 c 2.44 b 2.17 b 2.85 b 4.85 a 4.33 a 2.63 b Week 20 0.33 c 2.22 b 2.00 b 2.67 b 4.11 a 4.67 a 2.59 b
138 Table 4-4. Comparative change s in mean root length of Calopogon tuberosus seedlings of different geographic source during 20 weeks in vitro culture. Means with the same letter among populations by week are not signi ficantly different according to Tukeys HSD test at =0.5.All units in mm. Michigan Ohio OklahomaSouth Carolina North Central Florida Central Florida South Florida Week 8 10.0 ab 7.29 bc 7.35 bc 10.7 a 10.0 ab 6.22 c 1.26 d Week 10 9.56 c 27.5 a 23.4 ab 19.2 b 23.3 ab 17.5 b 5.37 c Week 12 16.8 d 36.9 a 32.6 ab 27.1 bc 29.1 bc 31.1 ab 27.7 cd Week 14 14.9 d 37.9 c 56.4 b 52.2 b 74.4 a 62.6 ab 29.3 c Week 16 13.1 c 29.3 b 36.6 b 38.6 b 52.4 a 64.4 a 31.0 b Week 18 17.3 d 40.1 c 29.9 cd 39.7 c 58.5 b 73.9 a 30.9 cd Week 20 7.22 c 47.7 b 34.1 b 41.1 b 65.5 a 80.3 a 43.9 b
139 Table 4-5. ANOVA results for Calopogon tuberosus seedling biomass allocation after 20 weeks in vitro culture. Shoot Root Corm Variation df F p df F p df F p Source 6 272.5 < 0.0001 6 167.1 < 0.00016 412.3 < 0.0001 Week 6 266.9 < 0.0001 6 65.4 < 0.00016 260.1 < 0.0001 S*W 36 6.68 < 0.0001 36 14.3 < 0.000136 10.5 < 0.0001
140 Table 4-6. Comparative biomass allocation to shoots, roots, and corms of Calopogon tuberosus seedlings of different geographic sour ce. Means with the same letter among populations by week (horizontal) are not significantly different according to Tukeys HSD test at =0.5. All units in percent. Michigan Ohio Oklahoma South Carolina North Central Florida Central Florida South Florida Shoot Week 8 37.9 e 57.9 cd 53.1 d 64.6 c 75.5 b 52.2 d 90.4 a Week 10 27.2 d 37.2 cd 38.5 bc 47.8 bc 50.4 b 50.0 b 93.7 a Week 12 17.9 e 22.9 de 32.5 cd 37.6 bc 55.8 a 47.8 ab 53.5 a Week 14 10.1 d 26.1 c 26.2 c 30.7 c 49.0 b 40.1 b 60.6 a Week 16 14.1 c 15.9 c 16.6 c 20.7 c 37.7 b 42.5 ab 50.2 a Week 18 7.78 d 12.1 cd 16.8 c 14.9 cd 32.3 b 29.1 b 45.4 a Week 20 4.24 d 10.1 c 12.7 c 11.3 c 25.3 b 23.8 b 54.2 a Root Week 8 35.7 bc 42.1 ab 46.9 a 5.4 bc 24.5 d 47.8 a 28.8 cd Week 10 17.8 d 44.5 ab 37.8 bc 44.9 ab 49.6 a 50.0 a 29.5 c Week 12 14.9 e 34.9 d 33.6 cd 40.6 bcd 44.2 abc 52.2 a 50.2 ab Week 14 10.5 e 41.8 b 25.7 d 32.2 cd 42.4 b 59.3 a 39.4 bc Week 16 10.2 d 22.3 c 19.2 c 21.9 c 36.2 b 52.7 a 49.8 a Week 18 10.5 d 20.1 cd 21.0 c 16.7 cd 29.2 b 45.5 a 35.8 b Week 20 9.25 d 17.9 bc 13.5 bc 12.6 c 22.0 b 41.0 a 38.2 b Corm Week 8 32.4 a 0 b 0 b 0 b 0b 0 b 0 b Week 10 59.3 a 29.4 c 39.9 b 18.7 d 0 e 0 e 0 e Week 12 67.2 a 46.1 b 33.9 c 21.9 d 0 e 0 e 0 e Week 14 76.0 a 32.2 c 48.1 b 37.2 c 14.6 d 0 e 0 e Week 16 80.5 a 66.7 b 61.2 bc 57.3 c 26.2 d 12.9 e 0 f Week 18 89.4 a 67.8 b 62.2 b 68.4 b 38.5 c 35.4 c 33.7 c Week 20 97.3 a 73.8 b 73.8 b 77.3 b 52.7 c 35.2 d 20.5 e
141 Figure 4-1. In vitro seedling development of Calopogon tuberosus from widespread populations. Note the progressively long delay in corm formation in southern populations. Shoot die-back was characterized by yellowing and browning of leaves. A-D, Q-S) Seedlings after 8 weeks culture. E-H, T-V) Seedlings after 12 weeks culture. I-L, WX) Seedlings after 16 weeks culture. M-P, Z-BB) Seedlings after 20 weeks culture. A, E, I, M) Michigan seedlings. B, F, J, N) Ohio seedlings. C, G, K, O) Oklahoma seedlings. D, H, L, P) South Carolina seedlings. Q, T, W, Z) North Central Florida seedlings. R, U, X, AA) Central Florida seedlings. S, V, Y, BB) South Florida seedlings. Scale bars = 1 cm.
142 Figure 4-1. Continued.
143 Figure 4-2. Correlation of grow ing season length and percent corm biomass allocation of Calopogon tuberosus seedlings from widespread populations. Percent biomass allocation is represented as mg of dry we ight per total dry weight. A) Combined biomass allocation. B) Bioma ss allocation after 8 weeks in vitro culture. C) Biomass allocation after 10 weeks in vitro culture. D) Biomass allocation after 12 weeks in vitro culture. E) Biomass a llocation after 14 weeks in vitro culture. F) Biomass allocation after 16 weeks in vitro culture. G) Biomass allocation after 18 weeks in vitro culture. H) Biomass a llocation after 20 weeks in vitro culture. Each point represents the mean response of three replications with nine seedlings each. Corm biomass percentages were arcsine tran sformed prior to regression analysis.
144 CHAPTER 5 EFFECTS OF CHILLING AND CUTTING CORMS ON CORM DORMANCY AMONG WIDESPREAD POPULATIONS OF Calopogon tuberosus Introduction The results of Chapter 4 indicated that diffe rences in biomass allocation and timing of corm formation were influenced by growing seas on length, but the actual dormancy of the corms produced was not investigated. Because Michigan plants are located in a climate with a longer and colder winter, they may exhibit longer corm dormancy that may only be broken by long chilling periods compared to more southern plants. The next sequential step is an examination of corm dormancy in widespread populations of Calopogon tuberosus The issue of dormancy in plants has been debated because the term often has conflicting meanings based on the plant, plant part, whole plant or cellular level, a nd seasonality (Rohde and Bhalerao, 2007). Amen (1968) defined dor mancy as endogenously controlled, but environmentally imposed temporary suspension of growth. However, complete suspension of growth is difficult to assess (Rees, 1981). Th ree types of dormancy ha ve been previously identified including ecodormancy, paradormancy, and endodormancy (Lang, 1987). Ecodormancy is influenced by unfavorable envi ronmental conditions, paradormancy occurs due to an inhibition from another part of the plant, and endodormancy is found in the dormant structure (Lang et al., 1987). Seed dormancy is defined as the inability of a viable seed to germinate under favorable conditions (Bewley, 1 997). When considering the whole plant level dormancy may be defined as the inability to initiate growth from a meristem under favorable conditions (Rohde and Bhalerao, 2007). Many temperate plant species form overwinte ring structures su ch as buds, tubers, rhizomes, and corms (Garbisch et al. 1995; Rohde and Bhalerao, 2007). These structures are formed during the growing season before unfa vorable growth conditions are encountered
145 (Garbisch et al. 1995). The dormant structures continue to remain dormant until favorable growth conditions are en countered the following growing season (Garbisch et al. 1995). In order to break dormancy, chilling is required (Rohde and Bhalerao, 2007). Longer chilling periods are often required to break dormancy in tubers and corms of temperate species, but extended periods often inhibit growth and deve lopment (Clark, 1995; Yaez et al. 2005; Fukai et al. 2006). However, chilling period requirement may be di fferent according to plant provenance in that southern species may require shorter ch ill periods (Perry and Wang, 1960; Garbisch et al. 1995). Chilling has been explored in ecotypic differe ntiation of tree species (Perry and Wang, 1960; Kriebel and Wang, 1962), aquatic species (Garbisch et al. 1996), and forage grasses (Silsbury, 1961; Cooper, 1964; Eagles, 1967 a, b; MacColl and Cooper, 1967). Acer rubrum ecotypes from Florida required no chilling to break dormancy, but longer chill periods were required to break dormancy in north ern ecotypes (Perry and Wang, 1960). Acer saccharum ecotypes Georgia and Tennessee required shorter ch ill periods to break dormancy than ecotypes in Michigan and Ohio (Kriebel and Wang, 1962). In several species of forage grasses, relative growth rate of Mediterranean populations was highe r at cooler temperatures compared to north European populations that had a higher growth rate at warmer temperatures (Cooper, 1964). Prolonged chilling decreased both survival and s hoot growth of aquatic plant ecotypes from Florida (Garbisch et al. 1996). Several hormones have been shown to be i nvolved in storage organ dormancy. However, with the majority of research has been conducte d with potato tubers. Low levels of indole-3acetic acid (IAA) were found in dormant tubers while higher levels were present once shoot regrowth occurred (Hemberg, 1949). Abscisic acid (ABA) levels were relatively high in dormant tubers (Coleman and King, 1984), inhibited tuber sprouting (Su ttle and Hultstrand, 1994), and
146 subsequently declined during storag e (Suttle, 1995). Gibberellic acid (GA3) broke tuber dormancy by increasing shoot regrowth, and cyt okinins increased the act ivity of mersitematic areas (Suttle, 2004). GA has been shown to stimul ate cell division, increase glucose levels near buds, and increase synthesis of DNA and R NA (Burton, 1989; Taiz and Ziegler, 1998; Alexopoulos et al., 2007). Ethylene has been shown to increase as tuber dormancy decreased, and subsequently promoted shoot regrowth compared to control treatments (Rylski et al. 1974). However, tuber response to ethylene may be cultivar dependent (Alam et al., 1994). In addition to hormonal control, cutting t ubers increased bud sprouting due to a possible wound response or removal of endogenous inhi bitory hormones (Alexopoulos et al. 2008). Dormancy in corm-forming species, such as ma ny terrestrial orchids, has not been fully examined. Calopogon tuberosus is a model orchid species to study the extent of corm dormancy since it is a widespread species. Studying corm dormancy, the chilling peri od length necessary to break dormancy, and shoot emergence may provide insight into C. tuberosus ecotypic differentiation. Southern ecotypes may require a shorter chilling period to break dormancy and initiate shoot regrowth compared to northern ecotypes that are exposed to longer winters. The effects of cutting corms on dormancy of widespread C. tuberosus populations were also examined. This research also further validates the use of in vitro ecology to screen for ecotypic differentiation. The objectives were to: 1) Determin e the degree of corm dormancy in C. tuberosus ecotypes; 2) Verify the length of chilling that is appropriate to break corm dormancy in C. tuberosus ecotypes; and 3) Determine whether cutting C. tuberosus corms can effectively relieve dormancy. Given that southern ec otypes do not experience long wint ers, shorter chilling periods may be sufficient to break corm dormancy.
147 Materials and Methods Chilling Effects on Corm Dormancy and Shoot Regrowth Seeds were used from the following locatio ns: Carney Fen (Menominee County, Michigan, USA), Eva Chandler Heritage Pr eserve (Greenville County, South Carolina, USA), Goethe State Forest (Levy County, Florida, USA), Florida Pa nther National Wildlife Refuge (Collier County, Florida, USA). Seed capsules from all population s were collected before complete dehiscence and were stored at 23C over silica gel for 2 we eks. Seeds were then removed from capsules, pooled by geographic source, and stored dry in the dark at -11C until used. Seeds were surface disinfected in sterile scinti llation vials for 3 minutes in a solution of 5 mL absolute ethanol, 5 mL 6% NaOCl, and 90 mL sterile dd water. Seeds were rinsed with sterile dd water after surface steri lization. Solutions were removed with sterile Pasteur pipettes. Seeds were transferred with a 10L sterile inoculating loop onto BM-1 Terrestrial Orchid Medium contained in 100 x 15 mm Petri plat es. The medium was supplemented with 1% activated charcoal. Medium pH was adjusted to 5.7 with 0.1N KOH prior to autoclaving for 40 minutes at 117.7 kPa and 121C. Ten replicate Petr i plates with 30 mL medium each were used for each seed source with approximately 100 seeds per plate (Figure 5-1A). Cultures were placed in an environmental growth incubator (#I-35LL; Percival Scientific, Perry, IA, USA) under coolwhite fluorescent lights in a 12 h phot operiod at 24.2 0.2C and 40 mol m-2 s-1. After 8 weeks culture, seedlings (Figure 5-1B) were transferred to larger culture vessels for further growth and development. Nine s eedlings were transferred to individual Phyto Tech Culture Boxes containing 100 mL of BM-1 Terrestrial Orchid Medium. Fi ve replicate vessels were prepared for each treatment and seed source combination for a total of 45 seedlings per treatment. A total of 25 vessels with a total of 225 seedlings were prepared for each seed source.
148 A total of 900 seedlings were transferred. Seedlings grew in vitro for another 12 weeks, for a total of 20 weeks culture. Envir onmental conditions were the sa me as described previously. After the 20 weeks (Figure 5-1C ), shoots and roots on seedlings were removed so that only corms remained (Figure 5-1D). The nine corms in each PhytoTech box were transferred to Sigma Phytatrays I (#P1552, Sigma-Aldrich, St. Louis, MO) containing 100 mL of moist, sterilized vermiculite (Figur e 5-1E). Five Phytatrays I (114 mm x 86 mm x 63.5 mm) were prepared for each treatment. Cultures containing the corms were subsequently stored at 10 0.3C for 2, 4, 6, and 8 weeks in complete darkness; a control of no cold storage was also used. Five culture vessels per seed source were al located to each chilling period treatment. After the each chilling period, the five culture boxes for each time source*time treatment were removed. Corms were subsequently planted in a 9-cell pack containing Fafard 2 (Conrad Fafard, Inc., Agawam, MA, USA). Corms were pl anted in a randomized complete block design with block designated as the chill treatment so that block 1 was the control, etc. Each seed source was allocated to each block, and blocks were replicated five times. Corms were buried approximately 1 cm below the soil line. Trays we re placed in a walk-in growth chamber (Figure 5-1F) under a 16/8 h L/D photoperiod at 27 2.2 C and an average relative humidity of 85%. Four 400-watt metal halide bulbs (Sylvania, Da nvers, MA, USA) provided a light level of 167 mole m-2 s-1. Corms were watered as needed and as frequently as daily. Data Collection and Statistical Analysis Shoot emergence date was recorded by the pr esence of the new shoot breaking the soil surface. Every 2 weeks, starting upon emergence and continuing until week 16, shoot length was measured from the soil surface to the shoot apex. At the final data collect ion, leaf number, leaf width, shoot height, root number, root length, corm diam eter, and dropper formation were recorded. Droppers (Figure 5-2) are axillary shoots growing from storage organs to form new
149 storage organs (Dixon and Pate, 1978; Hollick et al. 2001). Percent shoot emergence and survival, noted by the presence of a corm beneat h the soil surface, were recorded. Logistic regression was used to assess th e affect of chilling treatmen t and source on percent shoot emergence, percent survival, and percent dropper formation using the generalized linear mixed model procedure (proc g limmix macro) in SAS v9.1. Least-square means (lsmeans) were used to assess mean separation. Endpoint measurement da ta were analyzed using the general linear procedure (proc glm), ANOVA, and least-square means in SAS v9.1. Effects of Cutting Corms on Corm Dormancy and Shoot Regrowth Seeds were used from the following locatio ns: Carney Fen (Menominee County, Michigan, USA), Eva Chandler Heritage Pr eserve (Greenville County, South Carolina, USA), Goethe State Forest (Levy County, Florida, USA), Florida Pa nther National Wildlife Refuge (Collier County, Florida, USA). In addition, seeds from Oklahoma, Ohio, and Central Florida were obtained from cultivated plants that were hand pollinated under greenhouse conditions (D. Goldman). Seed capsules from all populations were collected before complete dehiscence and were stored at 23C over silica gel for 2 weeks. Seeds were th en removed from capsules, pooled by geographic source, and stored in the dark at -11C until used. Seeds were surface disinfected in sterile scinti llation vials for 3 minutes in a solution of 5 mL absolute ethanol, 5 mL 6% NaOCl, and 90 mL sterile dd water. Seeds were rinsed with sterile dd water after surface steri lization. Solutions were removed with sterile Pasteur pipettes. Seeds were transferred with a 10L sterile inoculating loop onto BM-1 Terrestrial Orchid Medium contained in 100 x 15 mm Petri plat es. The medium was supplemented with 1% activated charcoal. Medium pH was adjusted to 5.7 with 0.1N KOH prior to autoclaving for 40 minutes at 117.7 kPa and 121C. Ten replicate Petr i plates with 30 mL medium each were used for each seed source with approximately 100 seed s per plate. Cultures were placed in an
150 environmental growth incubator (#I-35LL; Perciv al Scientific, Perry, IA, USA) under cool-white fluorescent lights in a 12 h photoperiod at 24.2 0.2C and light levels of 40 mol m-2 s-1. After 8 weeks culture, seedlings were transf erred to Sigma Phytatrays II (#P5929, SigmaAldrich, St. Louis, MO) with 100 mL BM-1 Terre strial Orchid Medium. Ten Phytatrays (114mm x 86 mm x 102 mm) with nine seedlings were pr epared per seed source. After 20 weeks culture shoots and roots were removed from seedlings, and corms were transferred to fresh BM-1 Terrestrial Orchid Medium in P hytatrays II. Corms were either left whole or cut in half longitudinally. Previous experime nts using the north central Flor ida population indicated that no difference was found between shoot regrowth on th e bottom or top corm half (unpublished data). Both the top and bottom halves were randomly distributed throughout the culture vessels. For each seed source five replicate vessels with nine propagules each were pr epared for each corm treatment. Cut corms were transferred cut-side down. Environmental conditions were the same as previously mentioned. Data Collection and Statistical Analysis Shoot regrowth was monitored bi-weekly and percent shoot regrowth was recorded. After 8 weeks final data was collected including percent shoot regrowth, number of shoots per explant, shoot length, root length and number, and corm formation on new shoots. Final endpoint measurement data as well as the days to shoot emergence were analyzed in SAS v9.1 using ANOVA, general linear model proce dure (proc glm), and least-square means. Logistic regression was used to assess the affect of cutting corms on percent shoot regrowth using generalized linear mixed model procedure (proc glimmix macro) in SAS v9.1. Least-square means (lsmeans) were used to assess mean separation.
151 Results Chilling Effects on Corm Dormancy and Shoot Regrowth Chilling treatment (F4 = 61.2, p < 0.0001), source (F3 = 11.1, p < 0.0001), and the interaction between source and chilling treatment (F12 = 4.29, p < 0.0001) all significantly influenced the number of days to shoot emergence. The average number of days to shoot emergence was less under the longer chill periods of 6 and 8 weeks, regardless of source (Figure 5-3A). Corms subjected to no chilling and the 2 week chilling period exhibited the slowest shoot emergence. South Carolina and south Florida corms chilled longer than 6 weeks exhibited the quickest shoot emergence. Michigan corms re quired 4 weeks or longer for quickest shoot emergence, while shoots emerged faster when nor th central Florida corms were chilled for 8 weeks (Figure 5-3A). Percent shoot emergence was highly influenced by chilling treatment (F4 = 56.0, p < 0.0001), but source (F3 = 0.65, p = 0.58) and the interaction between chilling treatment and source (F12 = 1.66, p = 0.07) were not significant. Lowe r percent shoot emergence was observed when corms were chilled for shorter periods (F igure 5-3B). Less than 20% shoot emergence was observed in unchilled corms and following the 2 week chilling period among all populations. In fact, only one shoot from South Carolina emerge d in the control and only one shoot from north Central Florida emerged in the 2 week chilling treatment. Chilling periods longer than 6 weeks provided the highest percent shoot emergence in Michigan corms (Figure 5-3B), while 8 weeks chilling provided the highest shoot emergence for all other populations. Approximately 90% shoot emergence was observed for South Carolina, north central Florida, and south Florida corms, compared to 78% for Michigan corms (Figure 5-3B). After 16 weeks, corm survival was high rega rdless of source or ch illing length. Survival was measured by the presence of a viable corm below the soil surf ace, and not the presence the
152 emerged shoot. The original corm often remained viable, but did not form a shoot by the end of the experiment. Major differences in survival we re not clearly evident according to source (F3 = 0.00, p = 1.00), chilling treatment (F4 = 0.03, p = 0.99), or their interaction (F12 = 0.37, p = 0.97). Differences in survival were only observed wi thin Michigan and south Florida populations. Survival of south Florida propagul es was highest with 4-8 weeks chilling, while Michigan corm survival was highest after 2 week s chilling (Figure 5-3C). Near 100% survival was observed in South Carolina and North Central Florida corm s regardless of treatment (Figure 5-3C). Only source (F3= 3.89, p = 0.009) significantly influe nced dropper formation, while chilling treatment (F4 = 0.12, p = 0.97) and their interaction (F12 = 0.93, p = 0.94) were not significant. Droppers were considered formed by the presence of a shoot connecting the original corm to the new shoot (Figure 2A, B). Dropper fo rmation was prevalent in Michigan and South Carolina populations after 16 weeks (Figure 5-3D ). Dropper formation was higher on corms that were chilled for 6 or 8 weeks compared to shorter chilling periods in Michigan and South Carolina seedlings (Figure 5-3D). No droppers fo rmed on north central Florida and south Florida corms in the control, 2 weeks, and 4 weeks chil ling periods (Figure 5-3D). Dropper formation in both Florida populations was highest when corm s were chilled for 6 weeks, while 8 weeks chilling suppressed formation. The interaction between source and chill trea tment was significant (F = 2.45, p = 0.006) as well as source (F = 12.1, p < 0.0001) and chil ling treatment (F = 5.26, p = 0.004) for shoot length. No differences in shoot length were obs erved among chilling periods in the Michigan and South Florida populations (Figure 5-4A). Few di fferences were observed in South Carolina and north central Florida populations. Michigan and south Florida shoots were generally the shortest of all populations and north central Fl orida the highest (Figure 5-4A).
153 The interaction between sour ce and chilling treatment (F11 =1.19, p = 0.29) and chilling treatment (F4 = 1.52, p = 0.19) did not significantly infl uence leaf number. However, source significantly influenced leaf development (F3 = 6.87, p = 0.002). Michigan had the highest leaf number after 16 weeks followed by South Carolina and both Florida populat ions (Figure 5-4B). On average, less than two leaves were present on all seedlings regardless of source. The interaction between source and chill tr eatment was significant for leaf width (F11 = 2.74, p = 0.002) as well as source (F3 = 33.7, p < 0.0001), but not chilling treatment (F4 = 0.40, p = 0.81). Average leaf width was lowest on south Fl orida plantlets (Figure 5-4C) while the widest leaves were observed on north cen tral Florida and South Carolina populations (Figure 5-4C). No differences among treatments were observed on Sout h Carolina and south Florida. Wider leaves were observed in chilling periods longer than 4 weeks on Michigan plantlets. Widest leaves were observed on north central Florida pl antlets from the control, 2 w eeks, and 8 weeks chill period. The interaction between chil ling treatment and source (F11 = 2.08, p = 0.02), source (F3 = 12.5, p <0.0001) and chilling treatment (F4 = 4.46, p = 0.002) significan tly influenced root development. No differences were observed in root number for both South Carolina and Michigan populations, but more roots were observe d in shorter chilling periods for both Florida populations (Figure 5-4D). The hi ghest root number was observed in the control treatment in north central Florida plantlets, but South Carolina seedlings had the greatest root number on average. Chilling treatment (F4 = 4.27, p = 0.002), source (F3 = 8.72, p < 0.0001), and their interaction (F11 = 2.72, p = 0.002) significantly influenced root length. Longer chilling periods promoted the longest roots on Michigan plantlet s (Figure 5-4E). No difference was observed on South Carolina plantlets. Chilling periods less than 8 weeks promoted the longest roots on north
154 central Florida plantlets. South Florida plantlet s from the no chill treatment did not form roots, while few differences were observed among chilling treatments (Figure 5-4E). Source (F4 = 3.18, p = 0.03) significantly influenced new corm development while chilling treatment (F4 = 1.73, p = 0.15) and th eir interaction (F4 = 1.63, p = 0.09) were not significant. Few differences were observed among chilling tr eatments within each source with the exception of South Carolina and north central Florida where the largest corms were observed in the control and control/2 weeks chilling treatment, respectively (Figure 5-4F). Regardless of treatment, smallest corms were observed on south Florid a seedlings, and new corms did not form on seedlings in the control treatment. The largest corms were observed on So uth Carolina and north central Florida seedlings in the control and 2 weeks chilling treatment, while Michigan and South Carolina seedlings had the largest corm s in the 4, 6, 8 week chilling treatments. Effects of Cutting Corms on Corm Dormancy and Shoot Regrowth Cutting treatment (F1 = 109.3, p < 0.0001) had a highly signif icant effect on percent shoot growth, while source (F6 = 0.44, p = 0.86) and the interaction between source and treatment (F6 = 1.18, p = 0.32) did not have a significant effect. Cutting corms had a highly significant effect on shoot regrowth compared to whole corms for all sources (Figure 5-5). Less than 11% of uncut corms formed shoots among all populations. Percen t shoot regrowth on cut corms was well over 50% for all populations except nort h central and south Florida, but this was only significantly different than corms from Michigan. Michigan had the highest shoot regrowth with 66.7% of propagules forming shoots, but this was not diffe rent than Ohio (58.3%), Oklahoma (62.2%), and South Carolina (53.3%). Northern populations generally expressed a higher percent shoot regrowth sooner than southern populations (Figure 5-6). At week 2, Florida populations had less than 20% shoot regrowth while other all ot her populations had over 20% shoot regrowth.
155 However, after week 4 shoot regrowth genera lly did not differ significantly within each population (Figure 5-6). Cutting treatment (F1 = 23.1, p < 0.0001) and the interaction between treatment and source (F6 = 3.28, p = 0.005) significantly influe nced shoot length, but source (F6 = 0.82, p = 0.56) did not have a significant effect. Shoots produced on cut corms from Oklahoma exhibited the greatest mean shoot length (140 mm), as well as the smallest shoot length (10.5 mm) in the whole corm treatment (Figure 57A). However, shoot length on uncut Oklahoma corms was only significantly different than shoot leng th on uncut central Florida corms. Within several weeks of shoot initiation, corm s began to form at the base of the newly formed shoots. The interaction betw een source and cutting treatment (F6 = 4.69, p = 0.001) significantly influenced new corm diameter, but source (F6 = 1.98, p = 0.07) and cutting treatment (F1 = 2.04, p = 0.16) did not. Few differences were observed among treatments and sources, but the smallest corms were observed on South Florida (cut treatment), Ohio (uncut treatment), and Michigan (uncut treatment) shoots. No corms formed in the uncut treatment on south Florida and Oklahoma shoots (Figure 5-7B). Mean root number was signifi cantly influenced by source (F6 = 4.52, p = 0.0003) and cutting treatment (F6 = 12.2, p = 0.0006), but not their interaction (F6 = 1.46, p = 0.19). Root number on Ohio, Oklahoma, and north central Fl orida plantlets differe d between treatments (Figure 5-7C). The highest number of roots was observed in the cut treatment, regardless of source, as well as South Carolina and central Fl orida whole corm treatments. Root length was also significantly in fluenced by source (F6 = 4.50, p = 0.0003) and treatment (F6 = 15.5, p = 0.0001), but not their interaction (F6 = 1.24, p = 0.29). Few differe nces were observed among
156 treatments and sources with the exception that the longest mean roots were observed on central Florida in the cut corm tr eatment (Figure 5-7D). Discussion This is the first study examining the role of chilling in orchid ecotypes, and one of a few that investigated the role of chilling on storage organs among plant ecotypes. Previous research on chilling of storage organs has focused on eff ects of temperature and chilling length on corm and tuber sprouting and flowering for horticultural purposes (Clark, 1995; Kim et al., 1996; Gonzlez et al. 1998; Yaez et al. 2005; Fukai et al. 2006), but little information exists focusing on local adaptation to ch illing (Cooper, 1964; Eagles, 1967 a, b; MacColl and Cooper, 1967). While few differences in chilling were found among populations, local adaptation to different habitats and environmen tal conditions may, in part, expl ain the chilling requirement. The Role of Chilling on Dormancy Comparative influence of chilling storage organs such as tubers and corms resulting from ecotypic differentiation has received little a ttention. Bud chilling of tree ecotypes has been investigated extensively (Perry and Wang, 1960; Kriebel and Wang, 1962; Myking and Heide, 1995; Li et al. 2003; Li et al., 2005), but correlating chilling response of buds with underground storage organs may be difficult due to location of plant parts. Regardless, chilling requirements should be considered in a restor ation context if plants are mo ved from their home-site since southern ecotypes may not be cold-hardy (Garbisch et al., 1996). Clearly, further research is required on the ecological implications of chilling requirements on ecotypes. Chilling of C. tuberosus corms followed a similar pattern to bud chilling in birch species ( Betula pendula and B. pubescens ). Longer chilling treatments reduced days to bud break regardless of source latitude, a nd the number of days to bud break was more pronounced for southern ecotypes (Myking and Heide, 1995). Longer chilling periods increased shoot
157 emergence and reduced the number of days to s hoot emergence regardless of chilling length in ecotypes of C. tuberosus Longer chilling treatments had a more pronounced influence on emergence days of southern C. tuberosus ecotypes. In addition, northern ecotypes of both C. tuberosus and Betula sp. generally broke dormancy ear lier than southern ecotypes. The requirement for a chill peri od longer than 6 weeks for the C. tuberosus Michigan population is not surprising. The long winters and relatively constant temperatures below freezing require plants to maintain dormancy until environmental conditions are appropriate (Garbisch et al. 1996). No difference in shoot emerge nce was observed between the 6 and 8 week chilling period for Michig an plants, and shoot emergence was lower than all other populations in the 8 week chilling treatment. Bei ng subjected to longer winters, Michigan plants may require a chill period longer than 8 weeks for maximum shoot regrowth. The required chilling period for southern C. tuberosus ecotypes may be explained by differences in temperatures. Winter temper atures in the south often exceed 17C, but temperatures may drop suddenly in subsequent da ys. Even in south Florida temperatures can drop below 0C, albeit for shorter periods than northern climates. Longe r chilling periods would ensure that plants in Florida do not initiate regr owth until the threat of freezing temperatures is surpassed (Garbisch et al. 1996). An interesting morphological feature obser ved in Michigan and South Carolina populations was the formation of droppers. Droppe rs, which are not exclusive to orchids, are axillary shoots that grow downward to form replacement storage organs for the next seasons growth (Dixon and Pate, 1978; Hollick et al., 2001). Most species that form droppers are tuber forming plants rather than corm fo rming. The axillary shoot that forms on C. tuberosus corms is not an actually dropper since th e shoot grows upward toward the surface. Also, this may be an ex
158 situ phenomenon because axillary shoots have not been seen on wild plants from any location (personal observation). Michig an and South Carolina seedlings were more prone to form droppers, and more droppers were formed with longer chill periods regardless of seedlings source. Longer chilling treatment s may have influenced northern C. tuberosus ecotypes to form axillary shoots in order to position the new seas ons growth to the soil surface where warmer temperatures exist. Further investigation into the dynamics of axillary shoot formation is warranted. Clear differences in growth and development of plantlets after corm chilling were not evident. However, north central Florida and Sout h Carolina plantlets generally were the largest after 16 weeks. This was also observed throughout numerous experiments (personal observation). Specifically, north cen tral Florida plantlets from th e control had the longest shoots and roots, largest number of root s, and largest corms. Several f actors may have influenced the inconsistent seedling growth and development resu lts. Plantlet numbers were rather low in the control and shorter chill treatments thus error was much larger creating fewer significant results. The few plantlets that did emerge and develop in the shorter chilling trea tments had a longer time to develop compared to those in the longer chilling treatments. The Effect of Cutting Corms on Dormancy Exogenous application of ethylene and GA ha s been shown to break potato tuber dormancy (Rylski et al. 1974; Alam et al., 1994; Alexopoulos et al., 2007, 2008). Free abscisic acid (ABA) decreased during dormancy release in freesia corms, and exogenous application of the cytokinin benzlyadenine (BA) broke dormancy (Uyemura and Imanishi, 1987). In addition, ethylene production increased simultaneously as ABA decreased prior to freesia corm dormancy release, and the increase in ethylene was attri buted to dormancy release (Uyemura and Imanishi, 1983).
159 Cutting corms may trigger a w ound response leading to an in crease in respiration and metabolic activity that could influe nce faster shoot regrowth (Passam et al., 1977; Burton, 1989). Alexopoulos et al. (2008) found that cutting potato tubers prior to treatment with GA increased shoot production compared to whole tubers and cu tting tubers prior to incubation in water. Also, an increase uptake of GA may have influenced shoot regrowth. Cutting C. tuberosus corms may lead to increased concentrations of ethylene or GA causing enhanced shoot regrowth, or decrease the amount of inhibitory hormones (ABA) blocking regrowth. Due to sugars and nutrients present in the medium, cutting C. tuberosus corms may lead to enhanced uptake of nutrients leading to increased sh oot regrowth (Alexopoulos et al., 2008). Initial diff erences in shoot regrowth among C. tuberous ecotypes could be explained by differential rates of ethylene evolution or GA accumulation upon cutting corms, and further investigation is warranted. Conclusions The results presented here are not entirely conclu sive concerning C. tuberosus ecotypic differentiation with respect to chilling and co rm dormancy, but the experiments provide an excellent foundation for further investigation. Areas of research to examine include the effects of temperature on corm chilling, longer chill periods photoperiodic effects on regrowth of chilled corms, and exogenous hormones, such as GA, ABA, and cytokinins, on shoot regrowth. Given that the chilling temperature used (10C) is moderately cool and this temperature is experienced throughout the distribution of C. tuberosus chilling corms at colder temperatures may provide better insight into corm dormancy. Also, investig ating low temperature tolerance of propagules may reveal whether southern ecotypes can survive long-term under colder conditions.
160 Figure 5-1. Outline of in vitro to ex vitro growth of Calopogon tuberosus A) Germinating embryos in a Petri dish. Scale bar = 0.5 mm. B) In vitro seedling after 8 weeks culture and subsequently transferred to larger culture vessels. Scale bar = 0.5 cm. C) Seedlings after 20 weeks in vitro culture. D) Corm isolated from in vitro seedlings and placed in chilling conditions. Scale bar = 0.5 cm. E) Culture vessel containing sterilized vermiculite used to chill corms. F) Plantlets under ex vitro conditions in a walk-in growth chamber.
161 Figure 5-2. Ex vitro growth comparison of representative Calopogon tuberosus plantlets. Plantlets represent average si ze after 16 weeks growth in a walk-in growth chamber. Plantlets were generated after ch illed corms were planted under ex vitro conditions. A) Michigan plantlet with dropper formed between the original and new corm. B) South Carolina plantlet with dropper formed between the original and new corm. C) North Central Florida plantlet. D) South Florida plantlet. Sc ale bars = 1 cm.
162 Figure 5-3. Effects of chilling corms at 10C on shoot emergence after 16 weeks of ex vitro growth of Calopogon tuberosus plantlets. A) Number of days to shoot emergence. B) Percent shoot emergence recorded by the pr esence of a shoot emer ged from the soil. C) Percent survival measured by the presence of a corm beneath the soil. D) Percent dropper formation. Each histobar represents the mean response of five replications with nine plantlets for a to tal of 45 plantlets per treatment*source. Means with the same letter are not significantly different at = 0.05.
163 Figure 5-4. Effects of chilling corms at 10C on growth and development of Calopogon tuberosus plantlets. Data was co llected after 16 weeks ex vitro growth. A) Shoot length measured from the soil surface to the tip of the longest leaf. B) Leaf number. C) Leaf width measured at th e widest point of the widest leaf. D) Root number. E) Root length of the longest root. F) Corm diameter of the new corm measured horizontally at the widest poi nt. Each histobar represents the mean response of five replications with nine plantle ts for a total of 45 plantlets per treatment*source. Means with the same letter are no t significantly different at = 0.05.
164 Figure 5-5. Effects of cut and uncut unchilled corms on regrowth of Calopogon tuberosus shoots. Data was collected after 8 weeks in vitro culture. Histobars represent the mean response of five replications with nine propagules for 45 total measurements per treatment*source. Histobars with the sa me letter are not significantly different = 0.05.
165 Figure 5-6. Percent shoot regrowth from cut and uncut corms of Calopogon tuberosus over 8 weeks in vitro culture. A) Michigan. B) Ohio. C) Oklahoma. D) South Carolina. E) North Central Florida. F) Central Flor ida G) South Florida. Each data point represents the mean response S.E. of five replications with nine propagules. Points with the same letter are no t significantly different at =0.05.
166 Figure 5-7. Comparative regrowth from cut and whole corms of Calopogon tuberosus Data was collected after 8 weeks in vitro culture. A) Shoot height m easured from the above the corm to the tip of the longest leaf. B) New corm diameter measured horizontally from the widest point. C) Root number. D) Root length of the longest root. Histobars represent the mean response of five replications with nine propagules for 45 total measurements per treatment*source. Histobars with the same letter are not significantly different = 0.05.
167 CHAPTER 6 SUMMARY AND CONCLUSIONS The Orchidaceae is a high profile plant family that is imperiled worldwide due to habitat loss and illegal collecting. Because orchids have captured the attention of many individuals and organizations, their conservation is at the fore front of many concerns. While many organizations are involved in conserving and purchasing land to save orchids, others focused their attention on the restoration of rare and th reatened species. In order to restore population numbers, the propagation of orchids has received much attention. In vitro methods were developed beginning in the early 1900s to germinate orchid seed. In vitro germination methods are popular tools for germinating orchid seed to produce plants for population restoration. However, the source of plant material for orchid population restoration is often not a concern. For this reason, differences in local adaptation of Calopogon tuberosus to local environmental conditions were studied using in vitro methods. By controlling conditions such as photoperiod and temperature, in vitro methods were shown to be efficient and effective at differentiating ecotypes of Calopogon tuberosus based on seed germination, seedling development, biomass allocation, and corm formation. This area of research known as in vitro ecology utilizes methods to study the unique grow th and development of plant species based on geographic source. The effects of photoperiod and germination media on orchid seed germination were researched extensively over the past 100 years. However, my results indicate that photoperiod is not crucial for the germination of C. tuberosus seeds. Although seed germination percentages from Florida were higher under a short day photo period, differences in ph otoperiodic response of northern populations were not significant. In vitro responses to germinati on media contributed to differences in germination and development. Th ese differences reflected the varying habitats,
168 soil types, and soil nutrient availability at each individual site. Different temperature treatments were a strong influence on seed germinati on compared to photoperiod. In addition, coldstratification significantly increased germination by removing physiological dormancy. Germination results indicate that C. tuberosus ecotypes are influenced more by soil type and nutrients as well as temperature compared to photoperiod. However, photoperiod can not be ruled out as a major selection pressure influenc ing the growth and develo pment of adult plants. Once dormancy is removed from seeds, they may be able to germinate under a wider range of conditions leading to different results reported in Chapters 2 and 3. A different approach would be to stratify seeds on several germination media followed by a photoperiod and temperature screen. Throughout the germination experiments, Mi chigan seeds germinated and developed corms quicker than all other populations, while south Florida seeds germinated slowly. A detailed timecourse biomass allocation study show ed that Michigan seedlings allocated more biomass to corms faster than all other populat ions, while corm formation was delayed in southern plants. The differential corm biomass a llocation was positively correlated with a shorter growing season. Thus, the more rapid seed germination, corm formation, and corm biomass allocation in northern plants reflects adaptive responses to more severe winter conditions. Although all C. tuberosus populations required a chilling period to break corm dormancy, the ecological strategy for a chilling requirement is different. Southern plan ts require a chilling period to protect from sudden temperature fluc tuations during winter months, while northern plants require a chilling period to re main dormant during a long winter. The results from this research indicate that Calopogon tuberosus ecotypes do exist, but further investigation is still necessary to determine whether these ecotypes demonstrate
169 phenotypic plasticity and the ability to adapt to non-local environmental conditions. Data indicate that even under in vitro conditions, C. tuberosus ecotypes maintain their unique characteristics. For this reason, using local seed for population restoration projects is recommended. Local adaptation of C. tuberosus is, at least, based on temperature, growing season length, habitat type, and soil type and nutrient availability. These selection pressures do not interact as single entities, but rather form a complex web of se lection pressures that influence the development of ecotypes. Biotic selection pr essures, such as polli nators, likely influence differences in morphology, seed viability, and capsule set that coul d influence ecotypic differentiation In vitro ecology methods are effective to di fferentiation ecotypes, and should be applied worldwide as a component of sound cons ervation management plans to recover at-risk species.
170 APPENDIX A FIELD TRANSPLANT OF Calopogon tuberosus IN SOUTH FLORIDA Introduction Reintroducing and augmenting plant populations is a necessary step to maintain and reinstate ecosystem diversity (Maunder, 1992). Reintroducing and establishing new populations of charismatic species, such as terrestrial orchids, se rve as conservation sy mbols and to divert attention from a vulnerable population (Maunder, 1992). The worldwide loss of orchid taxa has led to an abundance of research focused on thei r conservation, ecology, an d field transplantation (Ramsay and Dixon, 2003). Unfortunately, few reports exist that detail management methods for both orchid populations and their habitat (Stewart, 2007). Successful establishment of plants into their habitats to augment or replace extent population is often the culmination and goal of conservation research (Batty et al. 2006 a). Field establishment of or chids is challenging since complex ecological requirements of individual taxa are not well-understood (Scade et al. 2006). Field establishment terrestrial orchids has b een previously attempted, but only for a few species (McKendrick, 1995; Rams ay and Stewart, 1998; Stewart et al. 2003; Batty et al. 2006b; Scade et al. 2006; Yamato and Iwase, 2008), and few invol ve successful field establishment of North American species (Stewart, 2007). Long-term survival of field translocated orchids is often very low because efficient methods fo r establishing orchids are lacking (Batty et al. 2006a). The influence of abiotic and biotic factor s on successful field establishment of orchids has not been studied in detail (Scade et al. 2006). However, field esta blishment of orchids could be an important tool for not only conserving orchid s, but also to further our knowledge of orchid ecology (McKendrick, 1995). A major obstacle to field establishment is initial survival of propagules. Only a few articles discuss techniques for increasing su rvival of orchid seedlings under in situ conditions (Batty et
171 al. 2006b; Scade et al. 2006; Smith et al., 2009). Batty et al. (2006 b) reported higher survival of several Australian orchid species when dormant tubers were used rather than seedlings. However, observations that Thelymitra manginiorum seedlings established more readily than tubers indicates that field perf ormance of different propagule t ype is species specific. Smith et al. (2009) found that actively growi ng plants established readily compared to dormant tubers. Competition may also be an important factor to consider for successful establishment (McKendrick, 1995). Dense coverage by native sp ecies and less weed coverage increased survival of field transplanted orch id species (McKendrick, 1995; Scade et al. 2006; Yamato and Iwase, 2008), but areas of highe st vegetation coverage impeded total survival (McKendrick, 1995). Calopogon tuberosus var. tuberosus (referred to as Calopogon tuberosus ) is a corm forming species found throughout eastern North Am erica including southwest Florida. In south Florida the flowering season begins in April and continues through th e end of May. Typical south Florida habitat includes mesic-alkaline pr airies surrounded by pine flatwoods growing in full sun. The ease of asymbiotic seed culture and rapid corm formation in vitro makes this orchid species an excellent candidate for development of a model system to test field establishment methods. Two separate experiments were conducted to: 1) Establish C. tuberosus seedlings at the Florida Panther National Wildlife Refuge (FPNWR ); 2) Compare survival of seedlings and corms of C. tuberosus ; 3) Compare survival and growth of seedlings in burned and unburned areas; and 4) Recommend manage ment practices for establishi ng terrestrial orchids at the FPNWR. In addition, the methods used can be transferred to other te rrestrial orchids worldwide.
172 Materials and Methods Study Site The Florida Panther National Wildlife Refuge is located in southwest Florida in Collier County (Figure A-1). The field plot s were established in the marl prairie in Unit 23. This area contains the largest population of C. tuberosus on the FPNWR. Seed Source and Propagation Seed was collected from the FPNWR as pr eviously described in chapter 3. Only propagules derived from the FPNWR were used to avoid any potential cross-pollination effects by introducing no-local ecotypes. For the 2008 experiment, seeds were germinated beginning March 2007 on P723 medium ( Phyto Technology Laboratories, Shawnee Mission, KS) in square Petri dishes as part of experiments in Chapter 2. After 8 weeks cultu re (May 2007), seedlings were transferred to Phyto Tech Culture Boxes (Phyto Technology Laboratories, Shawnee Mission, KS) containing 100 mL P723 medium. After an additional 20 weeks culture, corms were chilled from October 2007 to January 2008. This was ac complished by removing the shoots and roots from the seedlings, and transferring corm s to fresh P723 medium contained in Phyto Tech Culture Boxes. After the chilling period at 10C, corms were transferred to fresh P723 medium contained in Phyto Tech Culture Boxes for an additional 12 weeks. Seedlings were subsequently moved to greenhouse conditions April 2008. Seedlings were planted in 9-cell pack trays (Model #IKN0809, Hummert International, Earth City, MO) containing Fafard 2 soilless potting mix (Conrad Fafard, Inc., Agawam, MA). Seedlings we re covered with clear vinyl humidity domes to prevent desiccation and placed under 50% shade cloth and a natural photoperiod. Average light levels were 300 mol m-2 s-1 measured at 12 noon, and average temperatures ranged from 21.6 2C to 29.3 3C. After 1 week humidity domes were removed, and seedlings watered as needed.
173 Seedlings from the 2009 experiment were in itially started January 2008 as part of experiments from Chapter 3. Seeds were sown on BM-1 Terrestrial Orchid Medium ( Phyto Technology Laboratories, Shawnee Mission, KS) in square Petri dishes. After 8 weeks culture, seedlings were transferred March 2008 to Phyto Tech Culture Boxes containing 100 mL BM-1 medium. After 30 weeks culture, corms were transferred to new Phyto Tech Culture Boxes containing 100 mL BM-1 medium and chille d at 10C from October 2008 to December 2008. Seedlings were moved to greenhouse condi tions December 2008 until ready for field establishment February 2009. Greenhouse transfer pr ocedures were similar to those previously described. Average light levels were 253 mol m-2 s-1 measured at 12 noon, and average temperatures ranged from 20.8 2.3C to 28.8 2.8C. Field Establishment Planting occurred in successive years in April 2008 and February 2009. For the 2008 planting, differences in survival of field transplanted seedlings and corms were examined. For the 2009 planting, the response of planting seedli ngs in a burned and unburned area was studied. In all experiments square quadrats 30 cm x 30 cm were constructed from PVC piping (1.5 cm diameter). Each quadrat was divided into 16 sections approximately 7.5 cm x 7.5 cm by using 14 gauge coated electrical coppe r wire. A HOBO H8 Pro weat her station (www.microdaq.com, Ltd., Contoocook, NH) was placed at the site to reco rd daily temperatures and relative humidity (Figure A-2). Comparison of propagule type on field survival Three 10 m transects were establish in unit 23 on April 23, 2008. Each transect contained four quadrats that were 2.5 m apart. A randomiz ed block design was used to plant propagules. Corms and seedlings were assigned randomly to a quadrat and quadrat section. Sixteen propagules were used in each quadrat (8 seed lings and 8 corms per quadrat). A total of 192
174 propagules were planted. Propagules were irrigated with di stilled water upon initial planting. Data was collected on May 20th 2008, July 9th 2008, February 27th 2009, and April 23rd 2009. Seedling survival in a burned and unburned field plot In January 2009, unit 23 was burned except the area where C. tuberosus was previously established in 2008. This presented a unique oppo rtunity to compare the effects of planting seedlings in the burned and unbur ned areas in unit 23. Two 10 m tr ansects were established in the burned area and unburned area. Three quadrat s were allocated to each transect. Sixteen seedlings were planted in each quadrat for a tota l of 48 seedlings per transect and 192 seedlings for the experiment. Data Recording and Statistical Analysis Survival of all seedlings wa s recorded. Two different categories were classified in determining propagule survival. Percentage of actively growing shoots was recorded for those seedlings with an actively growing green shoot. Percentage of emergent shoots was recorded when shoots were present, but not necessar ily actively growing s hoots. Seedling leaf measurements were recorded before the Fe bruary 2009 experiment, and again in April 2009. Shoot emergence data were analyzed using proc glimmix, logistic regr ession, and least-square means in SAS v9.1. Results Comparison of Propagule Type on Field Survival Propagule type (F = 0.50, p = 0.48) did not infl uence survival, but da te of planting was significant (F = 20.4, p < 0.0001). At the initial data collection in May 2008, a higher proportion of seedlings (43.8%) had actively growing shoot s compared to corms (32.3%) (Figure A-3). After 1 month of field establis hment, less than 50% of all pr opagules had actively growing shoots regardless of treatment. In July 2008, all shoots had either senesced or were present but
175 not actively growing. A higher proportion of em ergent shoots were observed on seedlings (22.9%) compared to corms (12.5%). Data collect ed during February 2009 occurred during the early growing season in south Florida. Activ ely growing shoots were higher on corms (12.5%) compared to shoots (10.4%), but was not sign ificantly different. In April 2009 no significant difference was observed between the survival of corms (6.25%) and seedlings (8.33%), and the presence of shoots further declined. At this time, one seedling in the early flowering stage established from a corm propagule was observed No shoots were observed at the data recording in June 2009. Combined survivorship percen tages were as follow s: 38.0% (May 2008), 18.9% (July 2008), 11.4% (February 2009) and 7.3% (April 2009). Seedling Survival in a Burned and Unburned Field Plot Burning significantly influenced percent of emerged shoots (F = 48.7, p < 0.0001), while not burning influenced the percentage of ac tively growing shoots (F = 4.32, p = 0.04). Two months after field establishment, the number of actively growing shoots declined in both plots. Approximately 3% and 11% of actively grow ing shoots was observed in the burned and unburned areas, respectively (Figure A-4A). Howeve r, senesced shoots were visible on seedlings in the burned plot, but none in the unburned pl ot (Figure A-4B). Total survivorship was 7.3% when combining all data. Of the actively growing shoots, shoot lengths we re recorded in April 2009 (Table A-1). Shoot lengths on all seedlings with act ively growing shoots in the burned plot increased, while three of the elev en recorded leaf measurements in the unburned plot decreased (Table A-1). Discussion This is the first documentation of a field establishment study involving Calopogon tuberosus and one of the only scientifically documen ted orchid field establishment studies in North America (Stewart et al. 2003; Zettler et al. 2007). However, conclusive results were not
176 obtained due to the short-term nature of the st udy, but they are likely after several years of monitoring. Absence of an actively growing shoo t did not indicate propagule death since corms may have been present beneath the soil surface, but their presence beneath the soil was not confirmed in order to minimize soil disturbance. In addition, shoot s on field established seedlings may have senesced naturally because senescence naturally occurs in late May through early June. Field establishment of orchids depends may de pend on propagule type su ch as seedlings or storage organs. Dormant storage organs are often considered more likely to survive initial field establishment (Debeljak et al., 2002; Batty et al. 2006 b). Dormant storage organs may be able to survive drought conditions be tter than seedlings (Batty et al. 2006b); however, results are species specific. Caladenia arenicola and Diuris magnifica established more readily in the field when dormant tubers were plan ted rather than seedlings, but Thelymitra manginiorum established more readily from seedlings (Batty et al. 2006a). However, no C. arenicola propagules survived into the thir d growing season and only 10% of D. magnifica tubers. Approximately 70% and 35% of T. manginiorum seedlings and tubers, respectively, survived into the third growing season (Batty et al. 2006 a). Likewise, Smith et al. (2009) found that 2-3 year old plants (35%) established more readily in the field compared to tubers (11%) after 4 years. In the present study, no differences in shoot emergence were observed between corms (6.25%) and seedlings (8.33%) after the first year The low rates of shoot emergence may have been caused by propagule death or dormancy of co rms. Terrestrial orchids can remain dormant for several years (Kery and Gregg, 2004) so l ong-term monitoring is necessary to observe propagule survival. Throughout 2008-2009, south Florida experienced drought conditions which
177 may have promoted propagule death. The juvenile state of the propagules (1 year old) may have resulted in poor field establishment as well. Tuber size influenced th e survival of several Australian orchids with larger tubers increasi ng survival compared to smaller tubers (Batty et al. 2006a; Smith et al. 2009). Likewise, larger C. tuberosus corms or more mature plants may have likely increased survival in the present study by increasing storage reserves that tubers can utilize to sustain drought conditions and initiate growth (Batty et al. 2006a). The influence of competition, shading, and weed coverage influences the establishment of orchids in the field (M cKendrick, 1995; Scade et al. 2006). The effects of establishing orchids in burned plots have not been repor ted previously, but the influence of competition has been examined. In the present study, more actively growing shoots on C. tuberosus were observed on seedlings in the unburned area during April 2009 Shoots on the seedlings in the burned area were brown and senesced with the exception of three plants. The surroun ding native grasses in the unburned area likely shaded the seedlings provi ding increased survivorsh ip. Seedlings in the unburned area did not receive a ny level of shading and likel y caused seedling desiccation. Shading led to increased survival of severa l terrestrial orchids (M cKendrick, 1995, 1996; Scade et al. 2006; Yamato and Iwase, 2008), but areas of the densest shade and competition led to a decrease in orchid seedling survivorship (McKendrick, 1995; Yamato and Iwase, 2008). Fire is a necessary natural disturbance in many ecosystems (Duncan et al. 2008) including those found in south Florida. Competition with weeds and invasive species during field establishment often reduces the successful field establishment of seedlings (Moyes et al. 2005). Native perennials were established readily in a burned grassland and dolomite glade areas, reduced weedy species, and preven ted forest succession (Moyes et al. 2005; Duncan et al. 2008).
178 While the results of the study are mostly inconc lusive due to the shor t-term monitoring of the plots, the techniques employed can be applie d to other orchid spec ies worldwide. More definitive results may be observed after another growing season when seedlings in the burned area may re-emerge. Due to the drought conditions the past 2 years in south Florida, additional irrigation may have improved propagule survival. In addition, using symbiotically grown seedlings or inoculating soil with mycorrhizal fungi may have improved seedling survival as well (Batty et al. 2006a ; Scade et al. 2006; Smith et al. 2009). Although C. tuberosus is not a threatened or endangered orchid, the species is still threatened by habitat loss, fragmentation, and illegal collecting. Augmenting cu rrent or creating new populations may be necessary for the overall success of the species. Management Recommendations In order to successfully establish C. tuberosus seedlings in the field the following management techniques are recommended: 1) Pr opagules approximately 23 years old should be used instead of using young seedlings or corms th at do not have sufficient reserves to survive initial field establishment. 2) When planting do rmant corms, larger corms should be planted. 3) Propagules should be planted at the beginning of the growing season, preferably mid to late February. 4) Due to frequent drought conditions in south Florid a, supplemental irrigation should be applied when needed during pr opagule establishment. 5) Plots should be monitored for at least an additional year to observe successful field establishment. 6) If further field establishment is necessary, examine several different planting in tervals after a burn (i.e. 6 months, 1 year).
179 Table A-1. Shoot lengths r ecorded for actively growing Calopogon tuberosus seedlings in February and April 2009. All measurements are in mm. Seedlings were measured in February under greenhouse conditions prio r to transplant, and the April data collection was on seedlings after field tr ansplant on the Florida Panther National Wildlife Refuge. Treatment Transect Quadrat # Seedling # Height (Feb 2009) Height (April 2009) Unburned 1 1 14 85 28 1 2 3 25 66 1 3 3 100 75 1 3 6 60 90 1 3 13 90 108 2 1 14 70 92 2 2 4 62 220 2 2 9 76 35 2 2 10 63 72 2 3 5 85 125 2 3 12 26 102 Burned 1 1 13 52 91 1 2 14 10 165 2 1 3 95 140
180 Figure A-1. Field translocation study at the Florida Panther Na tional Wildlife Refuge. A) Map of the FPNWR. The red dot indicates the loca tion of the prairie habitat in Unit 23. B) Prairie habitat where tran slocation study was conducted. C) Burned (left) and unburned (right) areas in Fe bruary 2009. D) Burned (background) and unburned (foreground) areas. Yellow flags mark one of the transects. E) Transects and quadrats in the burned area. F) Transect and qu adrats in the unburned area in April 2008. G) Close-up of a quadrat.
181 Figure A-2. Monthly temp eratures recorded at Unit 23 in the Florida Panther National Wildlife Refuge. Average temperatures represent the mean daily high or low over the entire month. Data was collected with a HOB O H8 Pro series weather station.
182 Figure A-3. Survival of Calopogon tuberosus propagules at the Florida Panther National Wildlife Refuge. Histobars are the average of three separate tr ansects with four quadrats containing 16 propagules. A tota l of 96 propagules were planted per treatment for a total of 192 propagules.
183 Figure A-4. Survival of Calopogon tuberosus seedlings in a burned and unburned plot at the Florida Panther National Wildlife Refuge. A) Percentage of plants with actively growing shoots marked by the presence of a growing green shoot. B) Percentage of plants with either have actively growing shoots or previously emerged shoots that senesced. Histobars represent the mean of two transects with three quadrats containing 16 seedlings. Ninety-six seedlin gs were planted in each treatment for a total of 192 to tal seedlings.
184 APPENDIX B MORPHOMETRIC ANALYSIS OF Calopogon tuberosus POPULATIONS Calopogon tuberosus is a morphologically diverse terr estrial orchid of eastern North America. This diversity can be explained in part due to population isolations caused by glaciation, limited suitable transitional habitats between the coastal plai ns and glaciated areas, and possible differences in bee pollinators (Proctor, 1998; Goldman et al., 2004a). In a detailed morphometric treatment of C. tuberosus Goldman et al. (2004 a) distinguished plants from the northern, southeastern, and southw estern range. While many charac teristics were not different, northern plants tended to be smaller in stature with shorter leaves, smaller flowers, and narrower labella. In addition, the shape of the labellum lobe (Figure B-1) differed. Northern plants had a more rounded and strongly mucronate (with an ap ex) labellum lobe, while southern plants had wider, broader, and flatter labellum lobe that was weakly mucronate. A morphometric treatme nt based on Goldman et al. (2004a) was conducted on the populations from Michigan, South Carolina, and Florida. Ten plan ts from each population except Michigan were measured. Only two flowering plants were used for the Michigan population due to issues with encountering flowering plants. All data was collected on wild-growing plants during peak flowering season. The following data was collected: leaf numb er, leaf width, leaf height, inflorescence height, fl ower number, number of open fl owers, flower width, flower height, lateral sepal length and width, dorsal sepal length and width, lateral petal length and width, labellum length and width, labellum lobe width, column length and width, and column lobe width (see Figures B-1, 2). Data was analyzed using Tukeys HSD test at =0.05. Leaf number was consistent throughout the range, although plants in south Florida generally had more leaves. However, this was not significantly different from other sources (Figure B-3A). Leaf width was significantly di fferent among sources w ith Michigan and south
185 Florida populations having the na rrowest leaves (Figure B-3B). Leaf and inflorescence height were both shortest on Michigan plants and larg est in Florida populations (Figure B-3C, D). Flower number was highest on plants from nor th central Florida and lowest on Michigan plants, but no significant diffe rences were observed between Michigan and South Carolina populations (Figure B-4A). While plants from north central Florida had approximately 10 flowers, only about two were open simultaneous ly (Figure B-4B). South Carolina plants generally had the highest number of open flowers while Michigan plants had the lowest number at one open flower. However, no significant diff erences were detected between Michigan and Florida populations. No differences were observed in lateral se pal length (Figure B5A) and dorsal width (Figure B-5C), and few differences existed in lateral sepal widt h (Figure B-5B) and dorsal sepal length (Figure B-5C). However, plants from north central Florida had longer dorsal sepals than Michigan and South Carolina popula tions and wider lateral sepals than the Michigan population. North central Florida plants also had the longest and widest latera l petals compared to all other populations (Figure B-5E, F). Most of the characteristics measured we re similar to those reported by Goldman et al. (2004a). Wider labella were observed in Florida populations compared to Michigan and South Carolina populations. Northern plan ts were also smaller in statur e compared to southern plants. However, overall flower size was not signifi cant different perhaps due to sample size differences. Differences in indi vidual flower parts such may be due to different pollinators throughout the range of C. tuberosus (Thien and Marcks, 1972; Dressler, 1981; Firmage and Cole, 1988).
186 Figure B-1. Labeled parts of a Calopogon tuberosus flower measured.
187 Figure B-2. Labeled close up of a Calopogon tuberosus flower. A) Labellum dimensions measured. B) Column dimensions taken.
188 Figure B-3. Whole plant mo rphometrics analysis of Calopogon tuberosus. A) Leaf number. B) Leaf width measured at the widest point. C) Leaf height measured from soil to leaf apex. D) Inflorescence height measured fr om soil to inflorescence apex. Histobars with the same letter are no t significantly different at =0.05.
189 Figure B-4. Flower morphometrics of Calopogon tuberosus. A) Number of total flowers including open and closed. B) Number of flowers opened simultaneously. C) Flower width measured on a horizontal plane at the widest point. D) Flow er height measured on a vertical plane from the widest point. Histobars with the same letter are not significantly different at =0.05.
190 Figure B-5. Flower pa rt morphometrics of Calopogon tuberosus A) Lateral sepal length from point of attachment to apex. B) Lateral se pal width at the wide st point. C) Dorsal sepal length from point of attachment to ap ex. D) Dorsal sepal width at the widest point. E) Lateral petal length from point of attachment to apex. F) Lateral petal width at the widest point. Histobar s with the same letter are not significantly different at =0.05. For floral part lo cations see Figure B-1.
191 Figure B-6. Labellum and column morphometrics of Calopogon tuberosus A) Labellum length from point of attachment to apex. B) Labe llum width at the widest point. C) Labellum lobe width. D) Column length from point of attachment to apex. E) Column width at the widest point. F) Column lobe width at the widest point. Hist obars with the same letter are not signifi cantly different at =0.05. For floral part measurement locations see Figure B-2.
192 LIST OF REFERENCES Abrahamson WG. 1975. Reproduction of Rubus hispidus L. in different habitats. The American Midland Naturalist 95: 471-478. Abrahamson WG. 1979. Patterns of resource allocation in w ildflower populations of fields and woods. American Journal of Botany 66: 71-79. Alam S, Murr D, Kristof L. 1994. The effect of ethylene and of inhibitors of protein and nucleic acid syntheses on dormancy br eak and subsequent sprout growth. Potato Research 37 : 25-33. Alexandersson R, Agren J. 2000. Genetic structure in the no nrewarding, bumblebee-pollinated orchid Calypso bulbosa. Heredity 85 : 401-409. Alexopoulos AA, Akoumianakis KA, Vemmos SN, Passam HC. 2007. The effect of postharvest application of gibberellic aci d and benzyl adenine on the duration of dormancy of potatoes produced by plants grown from TPS. Postharvest Biology and Technology 46: 54-62. Alexopoulos AA, Aivalakis G, Akoumianakis KA, Passam HC. 2008. Effect of gibberellic acid on the duration of dormancy of potato tubers produced by plants derived from true potato seed. Postharvest Biology and Technology 49: 424-430. Allen PS, Meyer SE. 1998. Ecological aspects of seed dormancy loss. Seed Science Research 8: 183-191. Amen R. 1968. A model of seed dormancy. The Botanical Review 34: 1-31. Amon JP, Thompson CA, Carpenter QJ, Miner J. 2002. Temperate zone fens of the glaciated midwestern USA. Wetlands 22: 301-317. Anderson A 1990. Asymbiotic germination of seeds of some North American orchids. In: Sawyers C, ed. North American native terrestrial orchids: propagation and production Chadds Ford, Pennsylvania: Br andywine Conservancy, 75-80. Anderson A 1996. The reintroduction of Platanthera ciliaris in Canada. In: Allen C, ed. North American native terrestrial or chids: propagation and production. Germantown, Maryland: North American Native Te rres trial Orchid Conference, 63-71. Arditti J. 1967. Factors affecting the germ ination of orchid seed. The Botanical Review 33: 197. Arditti J, Oliva AP, Michaud JD. 1985. Practical germination of North American and related orchids-3Calopogon tuberosus Calypso bulbosa Cypripedium species and hybrids, Piperia elegans var. elata Piperia maritima, Platanthera hyperborea and Platanthera saccata American Orchid Society Bulletin 54: 859-872.
193 Arditti J, Ghani AKA. 2000. Tansley review No. 110 Numeri cal and physical properties of orchid seeds and their biological implications. New Phytologist 145: 367-421. Ballard WW 1990. Further notes on Cypripedium germination. In: Sawyers CE, ed. North American native terrestrial or chids: propagation and production Chadds Ford, Pennsylvania: Brandywine Conservancy, 87-89. Baskin CC, Baskin JM. 2001. Seeds: Ecology, biogeography, and evolution of dormancy and germination San Diego, California: Academic Press. Baskin CC, Baskin JM 2004. Determining dormancy-breaking and germination requirements from the fewest seeds. In: Guerra nt EO, Havens K, Maunder M, eds. Ex situ plant conservation: Supporting species survival in the wild. Washington DC: Island Press, 162179. Baskin CC, Thompson K, Baskin JM. 2006. Mistakes in germination ecology and how to avoid them. Seed Science Research 16: 165-168. Baskin JM, Baskin CC. 1973. Plant population differences in dormancy and germination characteristics of seeds: heredity or environment? The American Midland Naturalist 90 : 493-498. Batty AL, Brundrett MC, Dixon KW, Sivasithamparam K. 2006 a. New methods to improve symbiotic propagation of temperate terrestrial orchid seedlings from axenic culture to soil. Australian Journal of Botany 54 : 367-374. Batty AL, Brundrett MC, Dixon KW, Sivasithamparam K. 2006 b. In situ seed germination and propagation of terrestrial orchid seed lings for establishment at field sites. Australian Journal of Botany 54: 375-381. Bedford BL, Godwin KS. 2003. Fens of the United States: di stribution, characteristics, and scientific connection ve rsus legal isolation. Wetlands 23: 608-629. Berliocchi L. 2000. The orchid in lore and legend Portland, Oregon: Timber Press. Bernard N. 1899. Sur la germination du Neottia nidus avis. Comptes Rendus de lAcadmie des Sciences Paris 128: 1253-1255. Bewley JD, Black M. 1994. Seeds: Physiology of development and germination New York: Plenum Press. Bewley JD. 1997. Seed germ ination and dormancy. Plant Cell 9: 1055-1066. Bischoff A, Vonlanthen B, St einger T, Mller-Schrer H. 2006. Seed provenance matters -effects on germination of four plant sp ecies used for ecological restoration. Basic and Applied Ecology 7: 347-359.
194 Blakeman JP, Mokahel MA, Hadley G. 1976. Effect of mycorrhizal infection on respiration and activity of some oxidase en zymes of orchid protocorms. New Phytologist 77: 697704. Boivin B. 1967. Enumeration des plantes du Canada, Monopsides. Naturaliste Canadien 94: 131-157. Bowles ML, Jacobs KA, Zettler LW, Delaney TW. 2002. Crossing effects on seed viability and experimental germination of the federal threatened Platanthera leucophaea (Orchidaceae). Rhodora 104: 14-30. Brown PM. 2002. Wild orchids of Florida with refere nces to the Atlan tic and Gulf coastal plains Gainesville, Florida: Univ ersity Press of Florida. Bruland GL, Grunwald S, Osborne TZ, Ramesh Reddy K, Newman S. 2006. Spatial distribution of soil properties in water conservation area 3 of the Everglades. Soil Science Society of America Journal 70: 1662-1676. Brundrett MC, Scade A, Batty AL, Dixon KW, Sivasithamparam K. 2003. Development of in situ and ex situ seed baiting techniques to detect mycorrhizal fungi from terrestrial orchid habitats. Mycological Research 107: 1210-1220. Burgeff H. 1909. Die Wurzelpilze der Orchideen, ihre Kultur und ihre Leben in der Pflanze. G. Fischer Verlag : 202. Burton WG. 1989. The Potato Essex, England: Longman Sc ientific and Technical. Cameron D, Leake JR, Read DJ. 2006. Mutualistic mycorrhiza in orchids: evidence from plant-fungus carbon and nitrogen transfers in the green-leaved terrestrial orchid Goodyera repens New Phytologist 171 : 405-416. Cameron D, Johnson I, Leake JR, Read DJ. 2007. Mycorrhizal acquisition of inorganic phosphorus by the green-leav ed terrestrial orchid Goodyera pubescens Annals of Botany 99: 831-834. Cameron D, Johnson I, Read DJ, Leake JR. 2008. Giving and receiving: measuring the carbon cost of mycorrhizas in the green orchid, Goodyera repens New Phytologist 180 : 176-184. Campbell RK, Sorensen FC. 1973. Cold-acclimation in seedling douglas-fir related to phenology and provenance. Ecology 54: 1148-1151. Case FW. 1987. Orchids of the western great lakes region 2nd edn. Bloomfield Hills, Michigan: Cranbook Institute of Science, Bulletin 48. Catling PM, Lucas Z. 1987. The status of Calopogon tuberosus var. latifolius with comments on the application of varietal rank. Rhodora 89: 401-412.
195 Chu CC, Mudge KW. 1994. Effects of prechilling and liq uid suspension culture on seed germination of the yellow lady's slipper orchid ( Cypripedium calceolus var. pubescens ). Lindleyana 9: 153-159. Clark GE. 1995. Effects of storage temperatur e and duration on the dormancy of Sandersonia aurantiaca tubers. New Zealand Journal of Crop and Horticultural Science 23: 455-460. Clausen J, Keck DD, Hiesey WM. 1939. The concept of species based on experiment. American Journal of Botany 26: 103-106. Clausen J, Keck WM, Hiesey WM. 1941. Regional differentiation in plant species. American Naturalist 75 : 231-250. Clements MA, Muir H, Cribb PJ. 1985. A preliminary report on the symbiotic germination of European terrestrial orchids. Kew Bulletin 41: 437-445. Cohen JG, Kost MA. 2008. Natural community abstract for northern fen. Michigan Natural Features Inventory, Lansing, MI. 18 p. Coke JL 1990. Aseptic germination and growth of some terrestrial orchids. In: Sawyers CE, ed. North American native terrestrial orchids: propagation and production Chadds Ford, Pennsylvania: Brandywine Conservancy, 90-91. Coleman W, King R. 1984. Changes in endogenous abscisic acid, soluble sugars and proline levels during tuber dormancy in Solanum tuberosum L. American Journal of Potato Research 61 : 437-449. Cooper JP. 1964. Climatic variation in fora ge grasses. I. leaf devel opment in climatic races of Lolium and Dactylis Journal of Applied Ecology 1: 45-61. Curtis JT. 1939. The relation of specificity of orchid mycorrhizal fungi to the problem of symbiosis. American Journal of Botany 26: 390-399. Curtis JT. 1947. Studies on the nitrogen nutrition of orchid embryos. I. Complex nitrogen sources. American Orchid Society Bulletin 16: 654-660. Daehler CC, Anttila CK, Ayres DR, Strong DR, Bailey JP. 1999. Evolution of a new ecotype of Spartina alterniflora (Poaceae) in San Francisco Bay, California, USA. American Journal of B otany 86: 543-546. Davis JH. 1943. The natural features of southern Florida Tallahassee, FL: The Florida Geological Survey. De Pauw MA, Remphrey WR. 1993. In vitro germination of three Cypripedium species in relation to time of seed collection, media, and cold treatment. Canadian Journal of Botany 71: 879-885.
196 Debeljak N, Regvar M, Dixon KW, Sivasithamparam K. 2002. Induction of tuberisation in vitro with jasmonic acid and sucrose in an Australian terrestrial orchid, Pterostylis sanguinea Plant Growth Regulation 36: 253-260. Diez JM. 2007. Hierarchical patterns of symbiotic orchid germination linked to adult proximity and environmental gradients. Journal of Ecology 95: 159-170. Dijk E, Eck N. 1995 a. Axenic in vitro nitrogen and phosphorus responses of some Dutch marsh orchids. New Phytologist 131: 353-359. Dijk E, Eck N. 1995 b. Ammonium toxicity and nitrate response of axenically grown Dactylorhiza incarnata seedlings. New Phytologist 131: 361-367. Dixon KW, Pate JS. 1978. Phenology, morphology, and reproduc tive biology of tuberous sundew, Drosera erthryorhiza Lindl. Australian Journal of Botany 26: 441-454. Dixon KW 1987. Raising terrestrial orchids from seed. In: Harris W, ed. Modern orchid growing for pleasure and profit Adelaide, Australia: Orchid Club of South Australia, 47100. Donohue K. 2005. Seeds and seasons: interpreting germination timing in the field. Seed Science Research 15 : 175-187. Downie DG. 1940. On the germination and growth of Goodyera repens Transactions and Proceedings of the Botanical Society of Edinburgh 33: 36-51. Downs RJ, Bevington JM. 1981. Effect of temperature and photoperiod on growth and dormancy of Betula papyrifera American Journal of Botany 68 : 795-800. Dressler RL. 1981. The orchids: natural history and classification Cambridge, Massachusetts: Harvard University Press. Dressler RL. 1993. Phylogeny and classificatio n of the orchid family Portland, Oregon: Dioscorides Press. Duever MJ, Carlson JE, Meeder JF et al. 1986. The big cypress national preserve. National Audobon Society, New York. Duncan RS, Anderson CB, Sellers HN, Robbins EE. 2008. The effect of fire reintroduction on endem ic and rare plants of a southeastern glade ecosystem. Restoration Ecology 16: 3949. Dutra D, Johnson TR, Kauth PJ, Stewart SL, Kane ME, Richardson L. 2008. Asymbiotic seed germination, in vitro seedling developm ent, and greenhouse acc limatization of the threatened terrestrial orchid Bletia purpurea. Plant Cell Tissue and Organ Culture 94: 1121.
197 Eagles CF. 1967a. The effect of temperature on vegeta tive growth in climatic races of Dactylis glomerata in controlled environments. Annals of Botany 31: 31-39. Eagles CF. 1967b. Variation in the soluble carbohydrat e content of clim atic races of Dactylis glomerata (Cocksfoot) at diffe rent temperatures. Annals of Botany 31: 645-651. Ernst R. 1967. Effect of carbohydrate selection on the growth rate of freshly germinated Phalaenopsis and Dendrobium seed. Bulletin of the American Orchid Society 36: 10681073. Ernst R, Arditti J, Healey PL. 1971. Carbohydrate physiology of orchid seedlings. II. hydrolysis and effects of oligosaccharides. American Journal of Botany 58 : 827-835. Ernst R, Arditti J. 1990. Carbohydrate physiology of orchid seedlings. III. Hydrolysis of maltooligosaccharides by Phalaenopsis seedlings. American Journal of Botany 77: 188195. Fang W, Taub DR, Fox GA, Landis RM, Natali S, Gurevitch J. 2006. Sources of variation in growth, form, and survival in dwar f and normal-stature pitch pines ( Pinus rigida Pinaceae) in long-term transplant experiments. American Journal of Botany 93: 11251133. Fernald ML. 1946. Contributions from the Gray Herbarium of Harvard University: Identifications and reidentificati ons of North American plants. Rhodora 48: 185-200. Fetcher N, Shaver GR. 1990. Environmental sensitivity of ecotypes as a potential influence on primary productivity. American Naturalist 136: 126-131. Firmage DH, Cole FR. 1988. Reproductive success a nd inflorescence size of Calopogon tuberosus (Orchidaceae). American Journal of Botany 75: 1371-1377. Florida Division of Forestry. 2005. Goethe State Forest http://www.fldof.com /state_forests/goethe.html (9 July 2009). Florida Natural Areas Inventory. 1990. Guide to the natural communities of Florida http://www.fnai.org/PDF/Natural_Communities_Guide.pdf (9 July 2009). Fow ler DP, Dwight TW. 1694. Provenance differences in the stratification requirements of white pine. Canadian Journal of Botany 42: 669-675. Friar EA, Ladoux T, Roalson EH, Robichaux RH. 2000. Microsatellite analysis of a population crash and bottleneck in the Mauna Kea silversword, Argyroxiphium sandwicense ssp sandwicense (Asteraceae), and its implications for reintroduction. Molecular Ecology 9: 2027-2034. Fukai S, Fujiwara K, Okamata O, Hasegawa A, Goi M. 1997. Effects of red and blue light on germination and protocorm growth of Calanthe Satsuma. Lindleyana 12 : 169-171.
198 Fukai S, Kanechika R, Hasegawa A. 2006. Effect of low temperature on breaking dormancy and flowering of Arisaema sikokianum (Araceae). Scientia Horticulturae 111: 97-100. Gallagher JL. 1983. Seasonal patterns in recove rable underground reserves in Spartina alterniflora Loisel. American Journal of Botany 70: 212-215. Gallagher JL, Howarth RW. 1987. Seasonal differences in Spartina recoverable underground reserves in the Great Sippewissett marsh in Massachusetts. Estuarine, Coastal and Shelf Science 25 : 313-319. Gallagher JL, Somers GF, Grant DM, Seliksar DM. 1988. Persistent differences in two forms of Spartina alterniflora : a common garden experiment. Ecology 69: 1005-1008. Garbisch EW, McIninch SM, Swartz HJ, Salvaggio GJ. 1995. Chilling responses for some herbaceous wetland plants. Wetland Journal 7: 16-20. Garbisch EW, McIninch SM, Swartz HJ, Salvaggio GJ. 1996. The effects of controlled chilling on five wetland herbaceous plant species. Wetland Journal 8: 20-29. Goldman DH, van den Berg C, Griffith MP. 2004 a. Morphometric circumscription of species and infraspecific taxa in Calopogon R. Br. (Orchidaceae). Plant Systematics and Evolution 247 : 37-60. Goldman DH, Jansen RK, van den Berg C, Leitch IJ, Fay MF, Chase MW. 2004 b. Molecular and cytological examination of Calopogon (Orchidaceae, Epidendroideae): circumscription, phylogeny, polyploidy, and possible hybrid speciation. American Journal of Botany 91: 707-723. Gonzlez A, Ban S, Fernndez JA Franco JA, Casas JL, Ochoa J. 1998. Flowering responses of Gladiolus tristis (L.) after exposing corm s to cold treatment. Scientia Horticulturae 74: 279-284. Gregor JW, Davey VM, Lang JMS. 1936. Experimental taxonomy. I. experimental garden technique in relation to the recognition of the small taxonomic units. New Phytologist 35: 323-350. Gregor JW. 1939. Experim ental taxomony. IV. population differentiation in North American and European sea plantains allied to Plantago martima L. New Phytologist 38: 19-322. Gross M, Hardisky M, Wolf P, Klemas V. 1991. Relationship between aboveground and belowground biomass of Spartina alterniflora (Smooth Cordgrass). Estuaries and Coasts 14: 180-191. Grze IM. 2007. Does rare Gentianella germanica (Wild.) Borner originating from calamine spoils differ in selected morphologica l traits from reference populations? Plant Species Biology 22: 49-52.
199 Hadley G, Purves S. 1974. Movement of C-14 from host to fungus in orchid mycorrhiza. New Phytologist 73: 475-482. Hadley G, Ong SH. 1978. Nutritional requirements of orchid endophytes. New Phytologist 81 : 561-569. Hadley G 1982. Orchid mycorrhiza. In: Arditti J, ed. Orchid biology: reviews and perspectives Ithaca, New York: Cornell University Press. Harvais G. 1973. Growth requirements and development of Cypripedium reginae in axenic culture. Canadian Journal of Botany 51: 327-332. Hemberg T. 1949. Significance of growth-inhibiting substanc es and auxins for the rest-period of the potato tuber. Physiologia Plantarum 2: 24-36. Henrich JE, Stimart DP, Ascher PD. 1981. Terrestrial orchid seed germination in vitro on a defined medium. Journal of the American Society for Horticultural Science 106: 193196. Hijner JA, Arditti J. 1973. Orchid mycorrhiza vitami n production and requirements by symbionts. American Journal of Botany 60: 829-835. Hofgaard IS, Vollsnes AV, Marum P, Larsen A, Tronsmo AM. 2003. Variation in resistance to different winter stress factors within a full-sib family of perennial ryegrass. Euphytica 134: 61-75. Hollick P, Senaratna T, McC omb J, Bunn E, Dixon KW. 2001. Response to paclobutrazol of symbiotic mycorrhizal fungi and dropper (tuber stalk) formation of host orchid seedlings. Plant Growth Regulation 00: 1-9. Howe GT, Hackett WP, Furnier GR, Klevorn RE. 1995. Photoperiodic responses of a northern and southern ecot ype of black cottonwood. Physiologia Plantarum 93: 695-708. Hufford KM, Mazer SJ. 2003. Plant ecotypes: genetic different iation in the age of ecological restoration. TRENDS in Ecology and Evolution 18 : 147-155. Hughes KW. 1981. In vitro ecology Exogenous factors affecting growth and morphogenesis in plant culture systems. Environmental and Experimental Botany 21: 281-288. Jerskov J, Johnson SD, Kindlmann P. 2006. Mechanisms and evolution of deceptive pollin ation in orchids. Biological Reviews 81: 219-235. Johnson SD. 1997. Pollination ecotypes of Satyrium hallackii (Orchidaceae) in South Africa. Botanical Journal of the Linnean Society 123 : 225-235. Johnson SD, Steiner KE. 1997. Long-tongued fly pollination and e volution of floral spur length in the Disa draconis complex (Orchidaceae). Evolution 51 : 45-53.
200 Joshi J, Schmid B, Caldeira MC et al. 2001. Local adaptation enhances performance of common plant species. Ecology Letters 4: 536-544. Kamiya A, Ikegami I, Hase E. 1981. Effects of light on chlorophyll formation in cultured tobacco cells I. chlorophyll accumulation and phototransformation of protochlorophyll(ide) in callus cells under blue and red light. Plant Cell and Physiology 22: 1385-1396. Kane ME. 1996. Wetland plant micropropagation: issues and opportunities. Aquatics 18: 4-11. Kane ME, Gillis MR, Philman N, Campbell S. 2000. Seasonal differences in ex vitro growth and corm formation between two micropropagated Sagittaria latifolia ecotypes. Acta Horticulturae 520: 229-237. Kane ME, Philman NL, Emshousen C. 2003. Wetland plant propagation: comparative growth and reproduction of micropropagated Sagittaria latifolia ecotypes. Combined Proceedings of the 2002 International Plant Propagators' Society 52 : 453-459. Kauth PJ. 2005. In vitro seed germination and seedling development of Calopogon tuberosus and Sacoila lanceolata var. lanceolata : two Florida native terrestrial orchids. University of Florida, Gainesville, FL, USA. Kauth PJ, Vendrame WA, Kane ME. 2006. In vitro seed culture and seed lings development of Calopogon tuberosus Plant Cell, Tissue and Organ Culture 85 : 91-102. Kauth PJ, Dutra D, Johnson TR, Stewart SL, Kane ME, Vendrame WA 2008. Techniques and applications of in vitro orchid seed germination. In: Teixeira da Silva JA, ed. Floriculture, ornamental and plant biot echnology: advances and topical issues Isleworth, UK: Global Science Books, 375-391. Keller M, Kollmann J. 1999. Effects of seed provenance on germination of herbs for agricultural compensation sites. Agriculture, Ecosystems & Environment 72: 87-99. Kery M, Gregg K. 2004. Demographic analysis of dormancy and survival in the terrestrial orchid Cypripedium reginae Journal of Ecology 92: 686-695. Kim HH, Ohkawa K, Sakaguchi K. 1996. Effects of storage temperature and duration on flower bud development, emergence and flowering of Zephyra elegans D. Don. Scientia Horticultura e 67: 55-63. Kindell CE, Winn AA, Miller TE. 1996. The effects of surrounding vegetation and transplant age on the detection of local ad aptation in the perennial grass Aristida stricta Journal of Ecology 84: 745-754. Knudson L. 1922. Nonsymbiotic germina tion of orchid seed. Botanical Gazette 73: 1-25. Knudson L. 1946. A new nutrient solution for the germination of orchid seed. American Orchid Society Bulletin 15 : 214-217.
201 Kondo M, Pablico PP, Aragones DV et al. 2003. Genotypic and environmental variations in root morphology in rice genotypes under upland field conditions. Plant and Soil 255: 189-200. Koopowitz H, Lavarack PS, Dixon KW 2003. The nature of threats to orchid conservation. In: Dixon KW, Kell SP, Barrett RL, Cribb PJ, eds. Orchid Conservation Borneo: Natural History Publications, 25-42. Kost MA, Albert DA, Cohen JG et al. 2007. Northern fen. Natural communities of Michigan: classification and description. Michigan Natural Features Inventory, Report Number 2007-21, Lansing, MI. 314 pp. Kriebel HB, Wang CW. 1962. The interaction between proven ance and degree of chilling in bud-break of sugar maple. Silvae Genetica 11 : 125-130. Kurepin LV, Walton LJ, Reid DM, Chinnappa CC, Pharis RP. 2007. Photoperiod, light quality, and irradiance effect s on flowering in the alpine and prairie ecotypes of Stellaria longipes. Canadian Journal of Botany-Revue Canadienne De Botanique 85 : 538-544. Kushlan JA. 1990. Freshwater Marshes. In: Myers RL, Ewel JJ, eds. Ecosystems of Florida Orlando, FL: University of Ce ntral Florida Press, 324-363. Lakon G. 1949. The topographical tetrazolium method fo r determining the germinating capacity of seeds. Plant Physiology 24: 389-394. Lang GA. 1987. Dormancy: a new universal terminology. Hortscience 22 : 817-820. Lang GA, Early JD, Martin GC, Darnell RL. 1987. Endo-, para-, and ecodormancy: physiological terminology and classification for dormancy research. Hortscience 22 : 371377. Lauzer D, Renaut S, St-Arnaud M, Barabe D. 2007. In vitro asymbiotic germination, protocorm development, and plantlet acclimatization of Aplectrum hyemale (Muhl. ex Willd.) Torr. (Orchidaceae). Journal of the Torre y Botanical Society 134: 344-348. Lee YI, Lee N, Yeung EC, Chung MC. 2005. Embryo development of Cypripedium formosanum in relation to seed germination in vitro Journal of the American Society for Horticultural Science 130 : 747-753. Lee YI, Yeung EC, Lee N, Chung MC. 2006. Em bryo development in the ladys slipper orchid, Paphiopedilum delenatii with emphasis on the ultrastr ucture of the suspensor. Annals of Botany 98: 1311-1319. Li B, Tomoko S, Yasuhiro Y, Toshihiko H, Masayuki Y. 2001. Interclonal differences, plasticity and trade-offs of life history traits of Cyperus esculentus in relation to water availability. Plant Species Biology 16 : 193-207.
202 Li C, Junttila O, Heino P, Palva ET. 2003. Different responses of northern and southern ecotypes of Betula pendula to exogenous ABA application. Tree Physiology 23: 481-487. Li C, Wu N, Liu S. 2005. Development of freezing tolerance in different altitudinal ecotypes of Salix paraplesia Biologia Plantarum 49: 65-71. Li XG, Seliskar DM, Moga JA, Gallagher JL. 1995. Plant regeneration from callus cultures of salt marsh hay, Spartina patens and its cellular based salt tolerance. Aquatic Botany 51: 103-113. Liancourt P, Tielbrger K. 2009. Competition and a short grow ing season lead to ecotypic differentiation at the two extr emes of the ecological range. Functional Ecology 23: 397404. Linhart YB. 1995. Restoration, revegetation, and the impor tance of genetic and evolutionary perspectives. US Department of Agricultur e, Forest Service, Intermountain Research Station, General Technical Report Nu mber INT-GTR-315, Odgen, Utah. pp. 271-287. Linhart YB, Grant MC. 1996. Evolutionary significance of local genetic differentiation in plants. Annual Review of Ecology and Systematics 27: 237-277. Luer CA. 1972. The native orchids of Florida New York: The New York Botanical Garden. Luer CA. 1975. The native orchids of the United States and Canada excluding Florida New York: The New York Botanical Garden. MacColl D, Cooper JP. 1967. Climatic variation in forage gr asses. III. seasonal changes in growth and assimilation in climatic races of Lolium Dactylis and Festuca Journal of Applied Ecology 4: 113-127. Macel M, Lawson CS, Mortimer SR et al. 2007. Climate vs soil factors in local adaptation of two common plant species. Ecology 88: 424-433. Majerowicz N, Kerbauy GB, Nievola CC, Suzuki RM. 2000. Growth and nitrogen metabolism of Catasetum fimbriatum (Orchidaceae) grown with different nitrogen sources. Environmental and Experimental Botany 44: 195-206. Malmgren S. 1993. Large-scale asymbiotic propagation of Cypripedium calceolus -plant physiology from a surgeons point of view. Botanic Gardens Micropropagation 1: 59-63. Malmgren S 1996. Orchid propagation: theory a nd practice. In: Allen C, ed. North American native terrestrial orchid s: propagation and production Germantown, Maryland: North American Native Terrestrial Orchid Conference, 63-71. Manning JC, van Staden J. 1987. The development and mobilizati on of seed reserves in some African orchids. Australian Journal of Botany 35: 343-353. Maunder M. 1992. Plant reintroductions: an overview. Biodiversity and Conservation 1: 51-61.
203 McConnaughay KDM, Coleman JS. 1999. Biomass allocation in plants: ontogeny or optimality? A test along three resource gradients. Ecology 80: 2581-2593. McCormick MK, Whigham DF, O'Neill J. 2004. Mycorrhizal diversity in photosynthetic terrestrial orchids. New Phytologist 163: 425-438. McKay JK, Christian CE, Harrison S, Rice KJ. 2005. "How local is lo cal?"a review of practical and conceptual issues in the genetics of restoration. Restoration Ecology 13: 432-440. McKendrick SL. 1995. The effects of herbivory and vegetation on laboratory-raised Dactylorhiza praetermissa (Orchidaceae) planted into grassland in Southern England. Biological Conservation 73: 215-220. McKendrick SL. 1996. The effects of shade on seedlings of Orchis morio and Dactylorhiza fuchsii in chalk and clay soil. New Phytologist 134 : 343-352. McKinley T, Camper ND. 1997. Action spectra of in vitro asymbiotic germination of Goodyera repens var ophioides. Lindleyana 12 : 30-33. Meyer SE. 1992. Habitat correlated variati on in firecracker penstemon ( Penstemon eatonii Gray: Scrophulariaceae) seed germination response. Bulletin of the Torrey Botanical Club 119 : 268-279. Meyer SE, Kitchen SG, Carlson SL. 1995. Seed germination timing patterns in intermountain Penstemon (Scrophulariaceae). American Journal of Botany 82 : 377-389. Miyoshi K, Mii M. 1998. Stimulatory effects of sodium a nd calcium hypochlor ite, pre-chilling and cytokinins on the germination of Cypripedium macranthos seed in vitro. Physiologia Plantarum 102: 481-486. Molvray M, Kores PJ. 1995. Character analysis of seed coat in Spiranthoideae and Orchidoideae, with special reference to the Diurideae (Orchidaceae). American Journal of Botany 82 : 1443-1454. Mondoni A, Rossi G, Hay F, Bonomi C. 2008. Habitat-correlated seed germ ination behaviour in populations of wood anemone (Anemone nemorsa L.) from northern Italy. Seed Science Research 18: 213-222. Mooney HA, Billings WD. 1960. The annual carbohydrate cycle of al pine plants as related to growth. American Journal of Botany 47: 594-598. Moyes AB, Witter MS, Gamon JA. 2005. Restoration of native perennials in a California annual grassland after prescribed spring burning and solarization. Restoration Ecology 13: 659-666.
204 Mweetwa AM, Welbaum GE, Tay D. 2008. Effects of development, temperature, and calcium hypochlorite treatment on in vitro germinability of Phalaenopsis seeds. Scientia Horticulturae 117: 257-262. Myers PJ, Ascher PD. 1982. Culture of North American orchids from seed. Hortscience 17: 550. Myking T, Heide OM. 1995. Dormancy release and chilling requirement of buds of latitudinal ecotypes of Betula pendula and B. pubescens. Tree Physiology 15: 697-704. Mylavarapu RS, Kennelley ED. 2002. UF/IFAS extension soil testing laboratory (ESTL) analytical procedures and training manual http://edis.ifas.ufl.edu/SS312 (21 Aug 2009). Nagy ES, Rice KJ. 1997. Local adaptation in tw o subspecies of an annual plant: implications for migration and gene flow. Evolution 51 : 1079-1089. National Climactic Data Center. 2009. Climatological data annual summary 2007 http://www.ncdc.noaa.gov/oa/c lim ate/stationlocator.html (18 Aug 2009). Nelson JB. 1986. The natural communities of South Carolina http://cricket.biol .sc.edu/acm oore/public ations/JBNDoc.pdf (10 July 2009). Nelson JR, Harris GA, Goebel CJ. 1970. Genetic vs. environmenta lly induced variation in medusahead (Taeniatherum asperum [Simonkai] Nevski). Ecology 51: 526-529. Nieuwland JA. 1913. New plants from various places. II. American Midland Naturalist 3 : 129133. Nuismer SL, Gandon S. 2008. Moving beyond common garden and transplant designs: Insight into the causes of local adapta tion in species interactions. The American Naturalist 171 : 658-668. ien DI, Pederson P. 2003. Seasonal pattern of dry matter allocation in Dactylorhiza lapponica (Orchidaceae) and the relation be tween tuber size and flowering. Nordic Journal of Botany 23: 441-451. ien DI, Pederson P. 2005. Seasonal pattern of dry matter allocation in Dactylorhiza lapponica (Orchidaceae) and the relation be tween tuber size and flowering. Nordic Journal of Botany 23: 441-451. ien DI, O'Neill JP, Whigham DF, McCormick MK. 2008. Germination ecology of the boreal-alpine terrestrial orchid Dactylorhiza lapponica (Orchidaceae). Annales Botanici Fennici 45 : 161-172. Otero JT, Ackerman JD, Bayman P. 2004. Differences in mycorrhizal preferences between two tropical orchids. Molecular Ecology 13: 2393-2404.
205 Otero JT, Bayman P, Ackerman JD. 2005. Variation in mycorrhizal performance in the epiphytic orchid Tolumnia variegata in vi tro: the potential for natural selection. Evolutionary Ecology 19 : 29-43. Passam HC, Hughes PA, Read SJ. 1977. Wound repair in yam tubers : the effect of inhibitors of respiration and protein s ynthesis on repair physiology. New Phytologist 79. Peas-Fronteras JT, Villalobos MC, Baltaz ar AM, Merca FE, Ismail AM, Johnson D. 2009. Adaptation to flooding in upla nd and lowland ecotypes of Cyperus rotundus, a troublesome sedge weed of rice: tube r morphology and carbohydrate metabolism. Annals of Botany 103 : 295-302. Perry TO, Wang CW. 1960. Genetic variation in the winter chilling requirement for date of dormacy break for Acer rubrum Ecology 41: 790-794. Pillon Y, Qamaruz-Zaman F, Fay MF, Hendoux F, Piquot Y. 2007. Genetic diversity and ecological differentiation in the endangered fen orchid ( Liparis loeselii ). Conservation Genetics 8 : 177-184. Porcher RD, Rayner DA. 2001. A guide to the wildflowers of South Carolina Colombia, South Carolina: University of South Carolina Press. Potvin C. 1986. Biomass allocation and phenological di fferences among southern and northern populations of the C4 grass Echinochloa crusgalli Journal of Ecology 74 : 915-923. Pritchard HW. 1985. Determination of orchid seed vi ability using fluorescein diacetate. 8 : 727730. Probert RJ, Smith RD, Birch P. 1985 a. Germination reponses to light and alternating temperatures in European populations of Dactylis glomerata L. II. the genetic and environmental components of germination. New Phytologist 99: 317-322. Probert RJ, Smith RD, Birch P. 1985 b. Germination responses to light and alternating temperatures in European populations of Dactylis glomerata L. I. variability in relation to origin. New Phytologist 99: 305-316. Probert RJ, Smith RD, Birch P. 1986. Germination responses to light and alternating temperatures in European populations of Dactylis glomerata L .5. The principle com ponents of the alternati ng temperature requirement. New Phytologist 102: 133-142. Proctor HC. 1998. Effect of pollen age on fruit set, frui t weight, and seed set in three orchid species. Canadian Journal of Botany-R evue Canadienne De Botanique 76 : 420-427. Prutsch J, Schardt A, Schill R. 2000. Adaptations of an orchid seed to water uptake and storage. Plant Systematics and Evolution 220: 69-75. Quinn JA, Colosi JC. 1977. Separating genotype from environment in germination ecology studies. The American Midland Naturalist 97: 484-489.
206 Raghavan V. 1964. Effects of certain organic nitrogen compounds on growth in vitro of seedlings of Cattleya Botanical Gazette 125 : 260-267. Raghavan V, Torrey JG. 1964. Inorganic nitrogen nutrition of the seedlings of the orchid, Cattleya American Journal of Botany 51: 264-274. Ramsay MM, Stewart J. 1998. Re-establishment of the lady's slipper orchid ( Cypripedium calceolus L.) in Britain. Botanical Journal of the Linnean Society : 173-181. Ramsay MM, Dixon KW 2003. Propagation science, recovery, a nd translocation or terrestrial orchids. In: Dixon KW, Kell SP, Barrett RL, Cribb PJ, eds. Orchid Conservation Kota Kinabalu, Sabah, Malaysia: Natural History Publications (Borneo). Rasmussen HN, Andersen TF, Johansen B. 1990 a. Temperature sensitivity of in vitro germination and seedling development of Dactylorhiza majalis (Orchidaceae) with and without a mycorrhizal fungus. Plant Cell and Environment 13 : 171-177. Rasmussen HN, Andersen TF, Johansen B. 1990 b. Light stimulation and darkness requirement for the symbiotic germination of Dactylorhiza majalis (Orchidaceae) in vitro Physiologia Plantarum 79: 226-230. Rasmussen HN, Rasmussen FN. 1991. Climactic and seasonal regulation of seed plant establishment in Dactylorhiza majalis inferred from symbiotic experiments in vitro. Lindleyana 5: 221-227. Rasmussen HN. 1992. Seed dormancy patterns in Epipactis palustris (Orchidaceae): Requirements for germination and establishment of mycorrhiza. Physiologia Plantarum 86: 161-167. Rasmussen HN. 1995. Terrestrial orchids from seed to mycotropic plant. Cambridge, Great Britain: Cambridge University Press. Rees AR. 1981. Concepts of dormancy as illust rated by the tulip and other bulbs. Annals of Applied Biology 98: 544-548. Rice KJ, Black RA, Radamaker G, Evans RD. 1992. Photosynthesis, growth, and biom ass allocation in habitat ecotypes of cheatgrass ( Bromus tectorum ). Functional Ecology 6: 32-40. Roberts EH, Benjamin SK. 1979. The interaction of light, nitr ate and alternating temperature on the germination of Chenopodium album Capsella bursa pastoris and Poa annua before and after chilling. Seed Science and Technology 7: 379-392. Robertson C. 1887. Fertilization of Calopogon parviflorus Botanical Gazette 12: 288-291. Robertson JL, Wyatt R. 1990. Evidence for pollination ecotypes in the yellow-fringed orchid, Platanthera ciliaris Evolution 44 : 121-133.
207 Robichaux RH, Friar EA, Mount DW. 1997. Molecular genetic consequences of a population bottleneck associated with reintroduc tion of the Mauna Kea silversword ( Argyroxiphium sandwicense ssp. sandwicense [Asteraceae]). Conservation Biology 11: 1140-1146. Rogers SMD. 2003. Tissue culture and wetland establishment of the freshwater monocots Carex, Juncus, Scirpus, and Typha. In Vitro Cellular and Developmental Biology-Plant 39: 1-5. Rohde A, Bhalerao RP. 2007. Plant dormancy in the perennial context. Trends in Plant Science 12: 217-223. Rutchey K, Schall T, Doren RF et al. 2006. Vegetation classification system for south Florida natural areas. United States Geological Survey, Open File Report 2006-1240, St. Petersburg, FL. 142 p. Rylski I, Rappaport L, Pratt HK. 1974. Dual effects of ethylene on potato dormancy and sprout growth. Plant Physiology 53 : 658-662. Sambatti JBM, Rice KJ. 2007. Functional ecology of ecotypic di fferentiation in the Californian serpentine sunflower ( Helianthus exilis ). New Phytologist 175 : 107-119. Sanders S, McGraw JB. 2005. Population differentiation of a th reatened plant: variation in response to local environment and implications for restoration. Journal of the Torrey Botanical Society 132: 561-572. SAS Institute. 2003. SAS for Windows, version 9.1. Cary, NC. Sawada S, Nakajima Y, Tsukuda M et al. 1994. Ecotypic differentiation of dry matter production processes in relation to survivorship and reproduc tive potential in Plantago asiatica populations along climatic gradients. Functional Ecology 8: 400-409. Scade A, Brundrett MC, Batty AL, Dixon KW, Sivasithamparam K. 2006. Survival of transplanted terrestrial orchid seedlings in urban bushland habitats with high or low weed cover. Australian Journal of Botany 54: 383-389. Schtz W, Milberg P. 1997. Seed dormancy in Carex canescens : Regional differences and ecological consequences. Oikos 78: 420-428. Seliksar DM, Gallagher JL, Burdick DM, Mutz LA. 2002. The regulation of ecosystem functions by ecotypic variati on in the dominant plant: a Spartina alterniflora salt-m arsh case study. Journal of Ecology 90: 1-11. Seliksar DM, Gallagher JL. 2000. Exploiting wild population diversity and somaclonal variation in the salt marsh grass Distichlis spicata (Poaceae) for marsh creation and restoration. American Journal of Botany 87: 141-146. Seneca ED. 1972. Germination and seedling response of Atlantic and Gulf coast populations of Uniola paniculata American Journal of Botany 59: 290-296.
208 Seneca ED. 1974. Germination and seedling re sponse of Atlantic and Gulf coasts populations of Spartina alterniflora American Journal of Botany 61: 947-956. Shahba MA, Qian YL, Hughes HG, Christensen D, Koski AJ. 2003. Cold hardiness of saltgrass accessions. Crop Science 43 : 2142-2147. Sharma J, Zettler LW, van Sambeek JW, Ellersieck MR, Starbuck CJ. 2003. Symbiotic seed germination and mycorrhizae of federally threatened Platanthera praeclara (Orchidaceae). American Midland Naturalist 149 : 104-120. Shaver GR, Fetcher N, Chapin FS. 1986. Growth and flowering in Eriophorum vaginatum : annual and latitudinal variation. Ecology 67: 1524-1535. Sheehan TJ, Black RJ. 2007. Orchids to know and grow Gainesville, Florida: University of Florida press. Shefferson RP, Wei M, Kull T, Taylor DL. 2005. High specificity generally characterizes mycorrhizal association in rare ladys slipper orchids, genus Cypripedium 14: 613-626. Shefferson RP, Taylor DL, Wei M et al. 2007. The evolutionary hi story of mycorrhizal specificity among lady's slipper orchids. Evolution 61: 1380-1390. Sheviak CJ 1983. United States terrestrial orchids: patt erns and problems. In: Plaxton EH, ed. North American Terrestrial Orchids Southfield, Michigan: Mi chigan Orchid Society, 4960. Shimura H, Koda Y. 2005. Enhanced symbiotic seed germination of Cypripedium macranthos var. rebunense following inoculation after cold treatment. Physiologia Plantarum 123: 281-287. Shimura H, Matsuura M, Takada N, Koda Y. 2007. An antifungal compound involved in symbiotic germination of Cypripedium macranthos var. rebunense (Orchidaceae). Phytochemistry 68: 1442-1447. Shoushtari BD, Heydari R, Johnson GL, Arditti J. 1994. Germination and viability staining of orchid seeds following prolonged storage. Lindleyana 9: 77-84. Silsbury JH. 1961. A study of dormancy, survival and other characteristics in Lolium perenne L. at Adelaide, S.A. Australian Journal of Agricultural Research 12: 1-9. Singh KP. 1973. Effect of tem perature and light on seed germination of two ecotypes of Portulaca oleracea L. New Phytologist 72: 289-295. Smith SE. 1966. Physiology and ecology of orchid mycorrh izal fungi with re ference to seedling nutrition. New Phytologist 65: 488-&. Smith SE. 1967. Carbohydrate translocation in orchid mycorrhizas. New Phytologist 66: 371-&.
209 Smith SE. 1973. Asymbiotic germination of orchid seeds on carbohydrates of fungal origin. New Phytologist 72: 497-499. Smith ZF, James EA, McDonnell MJ, McLean CB. 2009. Planting conditions improve translocation success of the e ndangered terrestrial orchid Diuris fragrantissima (Orchidaceae). Australian Journal of Botany 57: 200-209. Snow AA, Whigham DF. 1989. Costs of flower and fruit production in Tipularia discolor (Orchidaceae). Ecology 70: 1286-1293. South Carolina Department of Natural Resources. Eva Russell Chandler Heritage Preserve http://www.dnr.sc.gov/managed/heri tage/chandler/description.htm l (10 July 2009). Spoerl E, Curtis JT. 1948. Studies on the nitrogen nutrition of orchid embryos. II. comparative utilization of nitrate and ammonium nitrogen. American Orchid Society Bulletin 17: 111114. Spurr AR. 1969. A low-viscosity epoxy resin embeddi ng medium for electron microscopy. Journal of Ultrastructure Research 26 : 31-43. St-Arnaud M, Lauzer D, Barabe D. 1992. In vitro germination and early growth of seedlings of Cypripedium acaule (Orchidaceae). Lindleyana 7: 22-27. Stenberg M, Kane M. 1998. In vitro seed germination and greenhouse cultivation of Encyclia boothiana var. erythronioides, an endangered Florida orchid. Lindleyana 13: 101-112. Stewart SL, Zettler LW, Minso J, Brown PM. 2003. Symbiotic germination and reintroduction of Spiranthes brevilabris Lindley, an endangered orch id native to Florida. Selbyana 24 : 64-70. Stewart SL, Kane ME. 2006 a. Asymbiotic seed germination and in vitro seedling development of Habenaria macroceratitis (Orchidaceae), a rare Flor ida terrestrial orchid. Plant Cell, Tissue and Organ Culture 86: 147-158. Stewart SL, Kane ME. 2006 b. Symbiotic seed germination of Habenaria macroceratitis (Orchidaceae), a rare Florida terrestrial orchid. Plant Cell, Tissue and Organ Culture 86 : 159-167. Stewart SL. 2007. Integrated conservation of Fl orida Orchidaceae in the genera Habenaria and Spiranthes : model orchid conservation systems for the Americas. University of Florida, Gainesville, FL, USA. Stoutamire WP. 1964. Seeds and seedlings of native orchids. Michigan Botanist 3 : 107-119. Stoutamire WP 1974. Terrestrial orchid seedlings. In: Withner CL, ed. The orchids: scientific studies New York: John Wiley and Sons, 101-128.
210 Sun S-C, Cai Y-L, Liu H. 2001. Biomass allocation of Scirpus mariqueter along an elevational gradient in a salt marsh of the Yangtse River estuary. Acta Botanica Sinica 43: 178-185. Suttle JC, Hultstrand JF. 1994. Role of endogenous abscisic acid in potato microtuber dormancy. Plant Physiology 105: 891-896. Suttle JC. 1995. Postharvest changes in endogenous ABA levels and ABA metabolism in relation to dormancy in potato tubers. Physiologia Plantarum 95: 233-240. Suttle JC. 2004. Physiological regulation of potato tuber dormancy. American Journal of Potato Research 81 : 253-262. Suzuki RO. 2008. Dwarf morphology of the annual plant Persicaria longiseta as a local adaptation to a grazed habitat, Nara Park, Japan. Plant Species Biology 23 : 174-182. Swarts N. 2007. Integrated conservation of the rare and endangered terrestrial orchid Caladenia huegelii H.G. Reichb. University of Wester n Australia, Perth, Australia. Taiz L, Ziegler E. 1998. Plant Physiology Sunderland, MA, USA: Sinauer Associates, Inc. Taylor DL, Bruns TD, Szaro TM, Hodges SA. 2003. Divergence in mycorrhizal specialization within Hexalectris spicata (Orchidaceae), a nonphotosynthetic desert orchid. American Journal of Botany 90: 1168-1179. Thien LB, Marcks BG. 1972. The floral biology of Arethusa bulbosa Calopogon tuberosus and Pogonia ophioglossiodes (Orchidaceae). Canadian Journal of Botany 50: 2319-2325. Thompson DI, Edwards TJ, van Staden J. 2001. In vitro germination of several South African summer rainfall Disa (Orchidaceae) species: Is seed test a structure a func tion of habitat and a determinant of germinability? Systematics and Geography of Plants 71: 597-606. Thompson K, Grime JP, Mason G. 1977. Seed germination in response to diurnal fluctuations of temperature. Nature 267 : 147-149. Thompson K, Grime JP. 1983. A comparative study of germin ation responses to diurnally fluctuating temperatures. Journal of Applied Ecology 20: 141-156. Tissue DT, Skillman JB, McDonald EP, Strain BR. 1995. Photosynthesis and carbon alloc ation in Tipularia discolor (Orchidaceae), a wintergreen understory herb. American Journal of Botany 82: 1249-1256. Tomita M, Tomita M. 1997. Effects of culture media and cold treatment on germination in asymbiotic culture of Cypripedium macranthos and Cypripedium japonica Lindleyana 12: 208-213. Toole EH, Toole VK, Borthwick HA, Hendricks SB. 1955. Interaction of temperature and light in germination of seeds. Plant Physiology 30: 473-478.
211 Trapnell DW, Hamrick JL, Giannasi DE. 2004. Genetic variation and species boundaries in Calopogon (Orchidaceae). Systematic Botany 29: 308-315. Tremblay RL. 1992. Trends in pollination biology of the Orchidaceae: evolution and systematics. Canadian Journal of Botany 70: 642-650. Tremblay RL, Ackerman JD, Zimmerman JK, Calvo RN. 2005. Variation in sexual reproduction in orchids and its evolutionary consequences: a spasmodic journey to diversification. Biological Journal of the Linnean Society 84 : 1-54. Turesson G. 1922 a. The genotypical response of the plant species to the habitat. Hereditas 3: 211-350. Turesson G. 1922 b. The species and the variety as ecological units. Hereditas 3: 100-113. U.S. Fish and Wildlife Service. 2009. Florida panther national w ildlife refuge fact sheet. http://www.fws.gov/southeast/pubs/facts/flpcon.pdf (8 July 2009). Uyemura S, Imanishi H. 1983. Effe cts of gaseous compounds in smoke on dormancy release in freesia corms. Scientia Horticulturae 20: 91-99. Uyemura S, Imanishi H. 1987. Changes in abscisic acid le vels during dormancy release in freesia corms. Plant Growth Regulation 5: 97-103. Valero-Aracama C, Kane ME, Wilson SB Vu JC, Anderson J, Philman NL. 2006. Photosynthetic and carbohydrate status of eas yand difficult-to-acclimatize sea oats ( Uniola paniculata L.) genotypes during in vitro culture and ex vitro acclimatization. In Vitro Cellular & Developmental Biology-Plant 42 : 572-583. Valero-Aracama C, Wilson SB, Kane ME, Philman NL. 2007. Influence of in vitro growth conditions on in vitro and ex vitro phot osynthetic rates of easyand difficult-toacclimatize sea oats ( Uniola paniculata L.) genotypes. In Vitro Cellular & Developmental Biology-Plant 43: 237-246. van Assche JA, Debucquoy KLA, Rommens WAF. 2003. Seasonal cycles in the germination capacity of buried seeds of some Leguminosae (Fabaceae). New Phytologist 158: 315323. van der Pijl L, Dodson CH. 1966. Orchid flowers: their pollination and evolution Coral Gables, Florida: University of Miami press. van Tienderen PH. 1992. Variation in a population of Plantago lanceolata along a topographical gradient. Oikos 64: 560-572. van Waes JM, Debergh PC. 1986 a. In vitro germination of some Western European orchids. Physiologia Plantarum 67: 253-261.
212 van Waes JM, Debergh PC. 1986 b. Adaptation of the tetrazolium method for testing the seed viability, and scanning electron microscopy st udy of some Western European orchids. Physiologia Plantarum 66: 435-442. Vendrame WA, Carvalho VS, Dias JMM. 2007. In vitro germination and seedling development of cryopreserved Dendrobium hybrid mature seeds. Scientia Horticulturae 114: 188-193. Volis S, Mendlinger S, Ward D. 2002. Differentiation in populations of Hordeum spontaneum along a gradient of environmen tal productivity and predictab ility: life history and local adaptation. Biological Journal of the Linnean Society 77: 479-490. Wang J, Seliskar DM, Gallagher JL, League MT. 2006. Blocking Phragmites australis reinvasion of restored marshes using plants selected from wild populations and tissue culture. Wetlands Ecology and Management 14: 539-547. Wang J, Seliskar DM, Gallagher JL. 2007. Growth of tissue culture-regenerated salt marsh monocots, in a simulated marsh field pl ot: Implication for wetland creation and restoration. Ecological Engineering 29: 8-16. Whigham DF. 1984. Biomass and nutrient allocation of Tipularia discolor (Orchidaceae). Oikos 42: 303-313. Whitlow CE 1996. Mass production of Calopogon tuberosus. In: Allen C, ed. North American native terrestrial orchid s: propagation and production Germantown, Maryland: North American Native Terrestrial Orchid Conference, 5-10. Williams RR. 2007. Environmental factors affec ting plant tissue cultures. Combined Proceedings International Plant Propagators' Society 56: 91-95. Williams SL. 2001. Reduced genetic diversity in eelgrass transplantations affects both population growth and individual fitness. Ecological Applications 11: 1472-1488. Wotavov-Novotn K, Vejsadov H, Kindlmann P. 2007. Effects of sugars and growth regulators on in vitro growth of Dactylorhiza species. Biologia Plantarum 51: 198-200. Wynd FL. 1933. Sources of carbohydrate for germ inati on and growth of orchid seedlings. Annals of the Missouri Botanical Garden 20: 569-581. Yamato M, Iwase K. 2008. Introduction of asymbiotical ly propagated seedlings of Cephalanthera falcata (Orchidaceae) into natural habita t and investigation of colonized mycorrhizal fungi. Ecological Research 23: 329-337. Yamazaki J, Miyoshi K. 2006. In vitro asymbiotic germination of immature seed and formation of protocorm by Cephalanthera falcata (Orchidaceae). Annals of Botany 98: 1197-1206. Yaez P, Ohno H, Ohkawa K. 2005. Temperature effects on corm dormancy and growth of Zephyra elegans D.Don. Scientia Horticulturae 105 : 127-138.
213 Yates RC, Curtis JT. 1949. The effect of sucrose and other factors on the shoot root ratio of orchid seedlings. American Journal of Botany 36: 390-396. Yeung EC, Zee SY, Ye XL. 1996. Embryology of Cymbidium sinense : embryo development. Annals of Botany 78: 105-110. Yoder JA, Zettler LW, Stewart SL. 2000. Water requirements of te rrestrial and epiphytic orchid seeds and seedlings, and eviden ce for water uptake by means of mycotrophy. Plant Science 156: 145-150. Yoshie F. 2007. Length of the pre-reproductive period of Plantago asiatica L. from different latitudes. Plant Species Biology 22: 135-139. Zettler LW, McInnis TM. 1992. Propagation of Platanthera integrilabia (Correll) Luer, and endangered terrestrial orchid, through symbiotic seed germination. Lindleyana 7: 154161. Zettler LW, McInnis TM. 1994. Light enhancement of symbiotic seed germination and development of an endangered terrestrial orchid ( Platanthera integrilabia ). Plant Science 102: 133-138. Zettler LW. 1997. Orchid-fungal symbiosis and its value in conservation. McIlvainea 13: 40-45. Zettler LW, Hofer CJ. 1997. Sensitivity of Spiranthes odorata seeds to light during in vitro symbiotic seed germination. Lindleyana 12: 26-29. Zettler LW, Hofer CJ. 1998. Propagation of the little club-spur orchid ( Platanthera clavellata ) by symbiotic seed germination a nd its ecological implications. Environmental and Experimental Botany 39: 189-195. Zettler LW, Stewart SL, Bowles ML, Jacobs KA. 2001. Mycorrhizal fungi and cold-assisted symbiotic germination of the federally threatened eastern prairie fringed orchid, Platanthera leucophaea (Nuttall) Lindley. American Midland Naturalist 145 : 168-175. Zettler LW, Pearlman S, Dennis DJ, Hopkins SE, Poulter SB. 2005a. Symbiotic germination of a federally endangered Hawaiian endemic, Platanthera holochila (Orchidaceae), using a m ycobiont from Florida: a conservation dilemma. Selbyana 26: 269-276. Zettler LW, Piskin KA, Stewart SL, Hartsock JJ, Bowles ML, Bell TJ. 2005 b. Protocorm mycobionts of the federally threatened eastern prairie fringed orchid, Platanthera leucophaea (Nutt.) Lindley, and a technique to prompt leaf elongation in seedlings. Studies in Mycology 53 : 163-171. Zettler LW, Poulter SB, McDonald KI, Stewart SL. 2007. Conservation-driven propagation of an epiphytic orchid ( Epidendrum nocturnum ) with a mycorrh izal fungus. Hortscience 42: 135-139.
214 Zhang X, Wu N, Li C. 2005. Physiological and growth responses of Populus davidiana ecotypes to different soil water contents. Journal of Arid Environments 60: 567-579. Ziegler AR, Sheehan TJ, Poole RT. 1967. Influence of various media and photoperiod on growth and amino acid content of orchid seedlings. American Orchid Society Bulletin 36: 195-202. Zimmerman JK, Whigham DF. 1992. Ecological functions of carbohydr ates stored in corms of Tipularia discolor (Orchidaceae). Functional Ecology 6: 575-581.
215 BIOGRAPHICAL SKETCH Philip was born and raised in Manitowoc, WI, and attended the University of WisconsinStevens Point (UWSP). His research career be gan at UWSP where he worked on screening potato clones for disease resistance He received several awards fo r his potato research allowing him to attend local, regional, and national m eetings. Although he l oved researching potato genetics, he soon developed a passion for Wiscons ins native orchids. Ph ilip graduated from UWSP in 2003 with a major in biology and minor in chemistry. He joined the Plant Restoration, Conservation, and Propa gation Biotechnology Program in the Environmental Horticulture Department at the University of Florida in August 2003, and earned a Masters degree in August 2005. His thesis was titled In vitro Seed Germination and Seedling Development of Calopogon tuberosus and Sacoila lanceolata var. lanceolata : Two Florida Native Terrestrial Orchids. Fortunately, Calopogon seeds were much easier to work with than Sacoila seeds. Philip then transitioned into a PhD program in August 2005, and continued to study Calopogon tuberosus In his spare time Philip volunteers for a pet re scue organization, laughs at his dogs, enjoys photography, studies martial arts, and continues to write biographies in the third person, although this is probably the last one.