.55. V . ¢ , . .39 ‘ . xx: 1 . .1. 2.. _ ‘ 25:. r. 1 ......v3. 4 \r n :3 t5. :5? . i . z 4 z. 3.... h 14... .9! :3 Vaniln PLACE IN RETURN Box to remove this chedF values from global ANOVA analysis of all accessions. ................................................................................................. 133 Supplementary Table 2. Pr>F values from ANOVA analysis of individual accessions. ................................................................................................. 134 viii LIST OF FIGURES CHAPTER I Figure 1. Overview of the relationships among Arabidopsis flowering pathways ' (Mouradov et al., 2002) .................................................................... 4 Figure 2. F LC delays flowering in a rheostat—like manner. The delay in flowering in a FRI—containing background is proportional to FLC copy number. Transgenic plants containing additional copies of FLC are biennials and do not flower without vernalization. (From Michaels and Amasino, 2000) ...... 7 CHAPTER II Figure 1. Levels for each lighting treatment from 29 Jan to 4 June 2001. The I2ight treatments were: 50% shading of ambient 2Iig1ht plus PPF of 100 pmol m 2-'s (L1); ambient light plus PPF of 20 pmol m'2 -'s (L2), ambient light plus PPF of pmol rn'2 s (L3); and ambient light plus PPF of 150 pmol rn’2 -'s (L4). ...................................................................................................................... 65 Figure 2. Quality ratings for Achillea (A), Gaura (B) and Lavandula (C) as described in Table 2 after 8, 6, and 7 weeks of forcing, respectively. Numerical ratings were determined following the first forcing period for each species and were used to assess quality in subsequent forcing periods. ..... 66 Figure 3. Quality ratings, lateral inflorescence number at anthesis, dry mass accumulation and stem length measured at visible bud (shaded squares) and anthesis (open squares) for Achillea ’Red Velvet', Gaura 'Siskiyou Pink' and Lavandula 'Hidcote Blue' as a function of daily light integral. Regression lines are shown for significant correlations only. Error bars represent 95% confidence intervals. Quality ratings and dry mass were determined for Achillea, Gaura and Lavandula after 8, 6, and 7 weeks of forcing, respectively. .................................................................................................. 67 Figure 4. Time from forcing to visible bud, from forcing to anthesis, and from visible bud to anthesis for Achillea millefolium ‘Red Velvet‘, Gaura Iindheimeri ‘Siskiyou Pink' and Lavandula angustifolia ‘Hidcote Blue' grown under varying average daily light integrals. Regression lines are shown for significant correlations only. Error bars represent 95% confidence intervals. ............... 68 CHAPTER III Figure 1. Days to visible bud (A), rate of progress to visible bud (B), and inflorescence number (C) for Achillea 'Moonshine’ plants following 2, 4, 6, and 8 weeks at -2.5 to 20°C. Error bars represent 95% confidence intervals. Control plants reached visible bud in 25.5 days with a rate to visible bud of ix 0.039 and formed an average of 1.8 inflorescences (solid lines in A, B, and C, respectively). ................................................................................................. 83 Figure 2. Number of leaves unfolded during (A) and after (B) treatment and their sum (C) for Achillea 'Moonshine' plants following 2, 4, 6, and 8 weeks of storage at -2.5 to 20°C. Error bars represent 95% confidence intervals. Control plants unfolded 10.9113 leaves after transfer to the greenhouse (solid line, B and C) ....................................................................................... 84 Figure 3. The rate of leaf unfolding for Achillea ’Moonshine' plants held at -2.5 to 20°C for 2, 4, 6, and 8 weeks. Data from two replications were pooled for analysis. ........................................................................................................ 85 Figure 4. Correlation between the number of leaves unfolded following 2, 4, 6, or 8 weeks treatment at -2.5 to 20°C and the number of days required to reach visible flower bud. Greenhouse temperatures were 2012°C. ....................... 86 Figure 5. The flowering physiology of Achillea ‘Moonshine’ stock plants and cuttings. Stock plants grown under a short day photoperiod for >15 weeks and transferred to an inductive long day photoperiod flower in approximately 24 weeks with few inflorescences per plant. Vegetative cuttings taken from these stock plants grown under short days, if rooted and grown under inductive long day photoperiods will flower in 6 weeks at 20°C. ................... 87 CHAPTER IV Figure 1. Survival percentage (A,E), flowering percentage (B,F), nodes formed after transplant below the inflorescence (C, G), and percent of flowering lateral shoots (D) or flower number per shoot (H) for Veronica spicata 'Red Fox‘ (A-D) and Isotoma axillaris (E-H) held at -2.5 to 20°C for 2 to 8 or 2.5 to 15 weeks, respectively. Node and flower number for Isotoma are reported for treatments with 100% survival. No control plants flowered. Error bars represent 95% confidence intervals. ........................................................... 109 Figure 2. Flowering percentage (A) and percent of lateral shoots that flowered (B) for Veronica spicata 'Red Fox’ held at -2.5, O, or 25°C for 12 to 32 days. Each symbol represents the mean of 16 plants. ......................................... 111 Figure 3. Number of nodes formed during temperature treatment (A, C) and the rate of node formation during treatment (B, D) for Veronica spicata 'Red Fox‘ (A, B) and Isotoma axillan‘s (C, D) held at -2.5 to 20°C for 2 to 8 or 2.5 to 15 weeks, respectively. Error bars represent 95% confidence intervals. ......... 112 Figure 4. Correlation between node number after transplant and days to flower for Isotoma axillan's plants stored at -2.5 to 20°C for 2.5 to 15 weeks. Treatments with less that 100% survival are presented with (+) inside symbol and were not included in regression analysis. ............................................ 113 CH Fig Figl. II.‘ I I n”.- I! Fig. CHAPTER V Figure 1. The number of leaves > 3mm formed during the 9-week vernalization treatment and the rate of leaf formation for Columbia, Bolsena, and ROdasen accessions grown at -2.5 to 15°C for 3, 6 and 9 weeks. Leaf formation rates were nonsignificant for duration of cooling and were pooled for each accession at each temperature. Error bars represent 95% confidence intervals. ..................................................................................................... 135 Figure 2. Number of leaves formed in the greenhouse below the inflorescence after vernalization and the number of days from placement in the greenhouse to visible inflorescence for all accessions following 3, 6, and 9 weeks vernalization. Error bars represent 95% confidence intervals. Visible inflorescences were recorded after 25 d for nonvernalized Columbia plants, and one nonvernalized Bolsena plant reached visible bud in 32 d. ............ 136 Figure 3. Total inflorescence number (A) and the number of flowers on the primary inflorescence (B) for each accession after 9 weeks of vernalization. Columbia control plants held at 20°C formed an average of 9.9 :t 2.4 inflorescences and 9.3 :I: 1.2 flowers. No ROdasen and one Bolsena control plant flowered. Error bars indicate 95% confidence intervals ..................... 137 xi CHAPTER I VERNALIZATION AND ITS ROLE IN THE FLOWERING PROCESS dorr dorr deve inille spit Wide Prue mat sam num the f Vernalization as defined by Chouard (1960) is the “acquisition or acceleration of the ability to flower by a chilling treatment.” Vernalization is an inductive process primarily found in temperate herbaceous biennial and perennial species that promotes flowering in imbibed seeds or whole plants once the requirement for low temperature is fulfilled (Thomas and Vince-Prue, 1984). Flower primordia initiate and develop only after plants are returned to warmer temperatures, not during the low temperature treatment. In addition to vernalization, low temperatures are involved in breaking dormancy of seeds and existing vegetative and reproductive buds. Chilling of dormant seeds and buds releases the growth inhibition and allows for continued development under favorable conditions. Low temperatures also promote flower initiation and development during chilling in several plants including brussel sprouts (Brassica oleracea gemmifera L.) (Thomas and Vince-Prue, 1984). Several reviews have been published characterizing vernalization of a wide range of plant species (Chouard, 1960; Lang, 1952, Thomas and Vince- Prue, 1984), most of which are decades old. Despite the breadth of subject matter, the role of vernalization in flowering has not been investigated to the same extent as environmental processes such as photoperiodism. However, numerous recent investigations at the genetic and molecular level have reopened the field of vernalization and provided new insight into its fundamental basis. Wl’.‘ Ulldi res; gene cont Dam: (wint (Sprii habit Genetic and Molecular Basis of Vernalization Vernalization is an important, heritable agronomic trait found in many crop species (Chouard, 1960; Table 1). Allelic variation of vernalization genes is the basis of worldwide adaptation of many crops to low temperature and enables the tailoring of varieties to sowing times for maximize yield potential (Snape et al., 2001). For example, in cultivated beet, breeders select for the vernalization trait to prevent premature bolting. The vernalization requirement of cultivated Beta vulgaris ssp. vulgaris is under the control of the bolting gene 8 where genotypes with the dominant allele (8) do not have a vernalization requirement and flower under long days (LD); those with recessive alleles (bb) have a quantitative response to cold exposure (Boudry et al., 2001). In wheat (Triticum Sp.) the VRN genes (VRN-A1, VRN-B1, VRN-D1, and VRN4) influence heading time by controlling the sensitivity of strains to low temperature (Kato et al., 2001). In particular, the VRN-A1 allele determines whether wheat strains have an obligate (winter habit) or facultative response to vernalization, or no response at all (spring habit) (Snape et al., 2001). The spring habit is dominant to the winter habit (all VRN loci are recessive) (Kato et al., 2001). Table 1. Major field crops requiring vernalization (Henderson et al., 2003) Species Common Name Species Common Allium cepa onion Lens culinan's lentil Avena sativa oat Lolium ryeg rass Beta vulgaris beet Papaver poppy Brassica rape Pisum sativum pea Brassica oilseed Raphanus radish Cicer chickpea Secale cereale rye Daucus carrot Spinacia spinach Hordeum barley Tn'ticum wheat Lactuca lettuce Vicia fabia faba bean Arabidopsis thaliana, a Model System. Arabidopisis thaliana (L.) Heynh. has been utilized extensively as a model plant in studying the molecular basis of vernalization and genetic control of flower timing. Arabidopsis ecoptypes are found worldwide in diverse geographic and climatic ranges. Arabidopsis is a facultative long day plant and flowers most quickly under LD. Flowering is strongly delayed or absent under LD in many winter annual ecotypes unless plants are exposed to a cold treatment, ie. these ecotypes require vernalization for flowering in the spring under favorable conditions. In contrast, early-flowering ecotypes such as Landsberg erecta[Ler1 and Columbia [Col] flower most rapidly under LB with or without exposure to cold (Sheldon et al., 2000b); however, vernalization does hasten flowering of these ecotypes when grown under SD (Lee and Amasino, 1995). The use of ecotypes and induced mutations have identified approximately 80 loci known to affect flowering time in arabidopsis, many of which are mapped to known flowering pathways (Levy and Dean, 1998). At least four signaling pathways interact to control flowering time in Arabidopsis: the photoperiod, V6 ,.l .l | St 90‘. iii FR Vernal SUfl alle al., SUm Clea for ll vernalization, autonomous (independent of external factors) and gibberellin pathways (Levy and Dean, 1998). These four distinct pathways converge on the same flowering time genes such as FLOWERING LOCUS T (FT) and SUPPRESSOR OF OVEREXPRESSION OF CONSTANS (SOC1) which coordinate flowering responses to multiple, simultaneously changing environmental signals (Mouradov et al., 2002) (Fig. 1). In particular, three primary genes are involved in the vernalization responses of arabidopsis: FRIGIDA (FRI), FLOWERING LOCUS C (FLC), and VERNALIZA TION (VRN). VRN2 FRI H_|O_S1 5’ / FCA/FY Autonomous Vernalization _'| FLC I'— FVE/FPA Pathway SOC1 Figure 1. Overview of the \ /FT relationships among Arabidopsis flowering FLOWERING pathways (Mouradov et aL,2002) FRIGIDA (FRI). Differences in the early and late flowering habit of summer and winter annual arabidopsis ecotypes, respectively, is determined by allelic variation of two loci, FRI and FLC (Lee and Amasino, 1995; Koomneef et al., 1998; Johansen et al., 2000). Recessive FRI alleles confer early flowering in summer annual strains while dominant FRI alleles in combination with FLC create the vernalization requirement of winter annual strains. FRI is responsible for the delay in flowering (ie the vernalization requirement) in natural winter role aial neg and flora prei (Sin neg. Veri allhc mecf of la annt FLC annual ecotypes by increasing FLC mRNA levels. Vernalization acts to reduce FLC expression (Lee and Amasino, 1995). Interestingly, early flowering ecotypes such as Ler and Col evolved from late flowering ecotypes through one of two base pair deletions in the FRI allele resulting in a loss of FRI function (Johansen et al., 2000). These loss-of-function FRI mutations minimize the vernalization requirement and enable other flowering pathways to predominate. FLOWERING LOCUS C (FLC), a Floral Repressor. FLC plays a key role in the control of flower initiation and the response to vernalization in arabidopsis (Sheldon et al., 2000a). FLC encodes a MADS box protein that negatively regulates the transcription of genes that promote flowering (Michaels and Amasino, 1999; Sheldon et al., 2000a). Specifically, FLC represses the floral pathway integrator genes FT and SOC1 (also known as AGL20) thereby preventing the action of other pathways under otherwise inductive conditions (Simpson and Dean, 2002). FLC functions in part to repress flowering by negatively regulating SOC1 expression downstream of FLC (Rouse et al., 2002). Vernalization increases SOC1 expression presumably by decreasing FLC levels, although SOC1 expression can be mediated through other pathways. Dominant FLC alleles quantitatively represses flowering in a “rheostat” mechanism (Figure 2). Michaels and Amasino (2000) illustrated that the degree Of lateness in flowering is proportional to FLC copy number, and the winter annual habit of arabidopsis can be converted to a biennial habit by increasing FLC copy number. mil M can of la flow fedu betii M01; autc pair. IEpre FY. i mRN Figure 2. FLC delays flowering in 1a FLC/FLC a rheostat-like manner. The delay in flowering in a FRI- - containing background is proportional to FLC copy number. Additiona| Transgenic plants containing copies of additional copies of F LC are FLC biennials and do not flower Increasing FLC number without vernalization. (From Michaels and Amasino, 2000) Arabidopsis populations from high altitudes with an obligate vernalization requirement have high levels of FLC activity compared with equatorial populations that flower without cold (Dennis, 2001 ). In fact, FLC protein levels correlate with mRNA transcript levels and determine the vernalization component of late flowering (Rouse et al., 2002). When FLC protein levels are high, flowering is delayed unless FLC is suppressed by vernalization, and the extent of reduction in FLC mRNA is proportional to vernalization duration. The interactions of genes upstream and downstream of FLC as well as between different flowering pathways control flower timing (Rouse et al., 2002). Mouradov et al. (2002) emphasized the effects of the vernalization and autonomous pathway on the central role of FLC in flower timing (Figure 1). Both pathways act by related mechanisms and converge in the control of FLC by repressing FLC expression. The autonomous pathway genes (FCA, FVE, FPA, FY, FLD, LD) regulate FLC expression by altering transcript levels to limit FLC mRNA accumulation (Figure 1, Sheldon et al., 2000a). In particular, dominant alc. and l in mg 2000 IEQUII (Moo; FRI alleles Upregulate FLC expression while LD decrease FLC mRNA (Michaels and Amasino, 1999). In addition, vernalization suppresses the lateness of flowering of recessive autonomous mutants under long days and short days, suggesting both pathways act in parallel to promote flowering (Henderson et al., 2003). Reeves and Coupland (2001) hypothesized that the autonomous pathway alone has little promotive activity but indirectly promotes flowering through the photoperiod and GA pathway by reducing the inhibitory effect of FLC. Mutants impaired in the photoperiod and GA pathways flowered to a lesser degree than fully vernalized mutants impaired in the photoperiod, GA, and autonomous pathways (Reeves and Coupland, 2001). The vernalization pathway was more effective in promoting flowering than the autonomous pathway alone. Therefore, it is likely that vernalization promotes flowering by other mechanisms in addition to FLC repression that may involve an unidentified Signaling pathway (Reeves and Coupland, 2001). In addition, the activity of VRN1 provides additional evidence for FLC-independent mechanisms for flowering in response to vernalization. VERNALIZA TION (VRN1 and VRN2). The vernalization genes (VRN1 and VRN2) are required for maintenance of the vernalized state and are involved in mediating vernalization-induced down regulation of FLC (Sheldon et al., 2000a). VRN2 is required for maintenance of the vernalized state but is not required for the initial response to cold or the establishment of FLC repression (Mouradov, 2002). VRN2 likely functions to maintain transcriptional repression of ta' ma he tut eaI do“ act con Inez verr Ant nece wani appe wthi FLC FLC by affecting Chromatin organization. These findings suggest that FLC regulation is biphasic with separate phases for establishment and maintenance (Henderson et al., 2003). VRN1 shares similar characteristics with VRN2 and is involved in transcriptional repression of FLC. VRN1 and VRN2 likely act together to maintain FLC silencing. However, in contrast to VRN2, VRN1 has additional functions during vernalization and in flowering time control that are not mediated through FLC (Levy et al., 2002). For example, overexpression of VRN1 leads to earlier flowering with No change in FLC mRNA. However, expression of the downstream floral integrator genes FT and AGL20 change with expression VRN1 activity. VERNALIZA TION INSENSITIVE 3 (VIN3). VIN3 is the most upstream component of the vernalization pathway identified to date and is involved in measuring the length of the cold treatment as well as in establishing the vernalized state through FLC repression in root and shoot apices (Sung and Amasino, 2004). VIN3 expression occurs following sufficient cold exposure necessary for vernalization yet becomes undetectable 3 days after a return to warmer temperatures. Therefore, effective vernalization can be marked by the appearance of VIN3 expression, and the degree of VIN3 expression correlates with the duration of cold and degree of FLC repression. Once VIN3 is induced, FLC repression occurs through deacetylation and histone modification events. or: the Chi 20 he Epigenetic Characteristics of Vernalization. Flower induction in response to vernalization often occurs several weeks after plants are returned to warmer temperatures. Once perceived, how are plants able to “remember” the cold in order to flower? Vernalization involves a cellular memory where the vernalized state is maintained and acquired through a series of subsequent mitotic divisions (Metzger, 1996). Although the vernalized state is stable through vegetative development, the requirement for vernalization is reestablished during meiosis, and seedling progeny must also be vernalized in order to flower. Therefore, vernalization is an epigenetic, reversible phenomena that does Not involve changes in DNA nucleotide sequences but likely involves changes in chromatin structure and DNA methylation (Goodrich and Tweedie, 2002). Chromatin Remodeling. Because changes in DNA sequence are irreversible, the use of chromatin to record transient events in a plants life cycle, such as winter has the advantage of being mitotically stable yet can be reset after meiosis. All chromosomes are composed of chromatin which are repeating units of nucleosome particles consisting of DNA and associated materials such as histories (Goodrich and Tweedie, 2002). Chromatin remodeling involves the alteration of a chromatin region resulting from covalent modifications of histone tails due to methylation, acetylation, phosphorylation or ubiquitination (Goodrich and Tweedie, 2002). These modifications affect transcription; an “open” chromatin structure allows for transcription while a “compact” structure reduces 10 is e site epi the accessibility of DNA to transcription machinery and converts the chromatin region into heterochromatin. DNA Methylation. It is now hypothesized that regulation of gene expression by DNA cytosine methylation leading to chromatin remodeling could be the basis of vernalization (Michaels and Amasino, 2000; Sheldon et al., 2000a). DNA methylation acts by covalently modifying the cytosines of a DNA strand into 5-methylcytosines (Fransz and Jong, 2002). Heavily methylated DNA is similar to heterochromatin and has been associated with transcriptional silencing and gene repression. Once established, methylation provides an epigentic mark to remember Chromatin states through subsequent cell divisions. Vernalization reduces DNA methylation of arabidopsis and thus has been correlated in early flowering proportional to the length of the cold treatment and extent of demethylation (Finnegan et al., 1998). Cold-induced demethylation may disrupt maintenance methylation and decrease the activity of a cold- sensitive DNA methyltransferase (Finnegan et al., 1998). Demethylation by treatment with the compound 5-azacytidine or by the METHYL TRANSFERASE1 (ME T1 ) antisense construct promoted flowering in vernalization-sensitive arabidopsis strains (Sheldon et al., 2000b). Methylation may also block expression of a FLC repressor or bind a repressor to the FLC promoter (Sheldon et al., 2000b). The promotive effects of vernalization and demethylation on flowering are additive and may act upon separate flowering pathways, or demethylation may not fully activate the vernalization pathway (Finnegan et al., 1 998). 11 ups al.. inv: 20; mei usin acte plar.‘ Imps. Syne ides] the l: ‘i anait Low Temperature Sensing How do plants perceive and measure temperature fluctuations to determine when adequate cooling has been achieved for flowering? To date, the genes required for the establishment of FLC repression and those in the cold- sensing pathway remain uncharacterized (Henderson et al., 2003). Although cold acclimation and vernalization are difference processes, research investigating cold acclimation in arabidopsis provided insight into genes involved in low temperature sensing. One viable candidate is HOS1, a gene that acts upstream of FLC as well as in the cold acclimation genes CBF1 and COR (Lee et al., 2001). The cold acclimation COR genes and the protein CBF1 are not involved with VRN genes or associated with vernalization processes (Liu et al., 2002). Possible temperature sensing mechanisms in plants include changes in membrane fluidity and accompanying ion fluxes. Crude physiological models using alfalfa cell cultures suggested that changes in plasma membrane fluidity acted as biological thermometers although mechanistic explanations in higher plants are not available (Orvar et al., 2000). Ca2+ fluxes may also be an important step in temperature signaling. Studies with the cyanobacterium Synechocystis found that increased expression of fatty acid desaturase genes (desA, desB, desD) in response to temperature shifts from 34 to 22°C affected the lipid desaturation and membrane fluidity (Los et al, 1997). Mutational analysis of Synechocystis provided a two-temperature-sensor model involving 12 S’AI IA U 00: aCC histidine kinases located in the plasma membrane (Browse and Xin, 2001). Each sensor was proposed to regulate a unique set of genes and, together, the sensors regulate a third set of genes. Although homologues to histidine kinase genes have not been found in Arabidopsis and direct evidence is lacking, higher plants may employ a similar two-sensor mechanism for temperature sensing. The Physiology of Vernalization Molecular and genetic studies have increased our knowledge of the signaling pathways involved in vernalization and flower induction processes. We now know that FLC plays an essential role in vernalization but does not totally account for the vernalization response in arabidopsis. We have also learned from molecular and physiological studies that several common features exist between FLC expression and the vernalized state (Michaels and Amasino, 1999). First, meristematic regions are the sites of vernalization perception and FLC mRNA levels are highest in root and vegetative shoot tips of Arabidopsis Col following chilling (Michaels and Amasino, 1999). Secondly, plants maintain the vernalized state through subsequent mitotic divisions, and FLC transcript levels in vemalization-responsive arabidopsis ecotypes decrease and remain mitotically stable in root, leaf, and floral tissues following vernalization (Sheldon et al., 2000a). Thirdly, both the vernalization requirement and FLC transcript levels are reestablished in the progeny of vernalized plants (Michaels and Amasino, 1999; Sheldon et al., 20003). Despite the recent knowledge gained in the molecular arena, the mechanisms by which plants perceive low temperatures and the cold 13 .A S. b lES WE (0 the WC? (8a yer: wor Rus tern pret mor. vem Meir Vern curre IDOV signal transduction pathway remain largely unknown. Yet, the physiological responses of vernalization in many plant species have been well described. Studies investigating the effects of low temperature on flower induction were first described in the mid-19th century by Klippert in the United States (Chouard, 1960). He observed that winter wheat (Triticum aestivum L.) could behave as spring wheat if germinated in the fall and stored cool until planting in the spring. However, vernalization did not become widely known until Gassner’s work on the cold requirement of cereals and other plants published in 1918 (Salisbury and Ross, 1992). The Russian scientist Lysenko coined the term yarovizacij in 1928, meaning “to make spring,” and translated it to the English word vernalization (Thomas and Vince-Prue, 1984). Lysenko encouraged Russian farmers to expose winter wheat and rye (Secale cereale L.) seed to low temperatures before planting them in the spring (Wilsie, 1962). This chilling pretreatment promoted flowering in the same summer as planting and enabled more Northern areas in Russia to successfully grow winter strains. Classical vernalization works by Gregory and Purvis with ‘Petkus’ winter rye and Lang and Melchers with biennial henbane (Hyoscyamus niger L.) further characterized vernalization responses of cereals and biennials (Chouard, 1960). Much of our current knowledge of vernalization is based upon these works, and little new knowledge has been gained in this field. 14 Ve the an. rep VGI I08 IUSi ther SDEI VEIH VBQe Vernalization Perception The Apical Meristem. Most studies beginning in the 1930s suggested that the apical meristem is the site of vernalization perception for many species, including annuals, biennials and perennials (Chouard, 1960). Apical meristems function to initiate tissues and organs as well as communicate with other cells in the plant, and the apical meristem is indeed the site of vernalization perception and signal transduction (Metzger, 1988). Although the shoot apex is the plant structure typically associated with reproductive development, all dividing shoot cells are potential sites of vernalization (Thomas and Vince-Prue, 1984). Other cell types aside from apical meristems may also be thermoinduced and organized into meristems that give rise to reproductive organs. Wellensiek (1960) found that leaf cuttings of overwintered Lunaria biennis developed into plants capable of flowering without additional exposure to cold. Further work with Lunaria leaf cuttings led to the belief that cell division is necessary for vernalization perception (Wellensiek, 1964). However, Metzger (1988) showed that field pennycress (Th/aspi arvense L.) Shoots regenerated from fully expanded petiole cells with no present meristematic tissue displayed normal reproductive development when therrnoinduced. Therefore the requirement for cell division may be species specific, or DNA replication and not cell division per se may be required for vernalization perception (Michaels and Amasino, 2000). Juvenility. Plant age is an important factor that controls the shift from vegetative to reproductive growth in some species. A plant must ensure that the 15 energy demands required for flowering and seed set can be met before flowering occurs. As a result, young seedlings of herbaceous and woody species are often incapable of responding to environmental stimuli such as low temperatures that otherwise induce flowering in mature plants. This developmental phase is called the juvenile or incompetent phase, and plants with a juvenile phase can only be vernalized when fully mature. The switch from vegetative to floral development can be triggered when a critical stage of development is reached after the juvenile period. Plants in this Stage have been termed ripe-to-flower (Bemier et al., 1981). The length of this juvenile period varies and can last for days or years, depending on species. Competence to Flower. How can one tell if a plant is competent to perceive low temperatures and flower? For some species such as English ivy (Hedera helix L.), changes in morphological characteristics such as leaf shape can distinguish between reproductively competent and incompetent tissues (Thomas and Vince-Prue, 1984). However, for the majority of species, reproductive competence is often difficult to assess with absolute certainty as a lack of flowering does not necessarily indicate the juvenile phase (Thomas and Vince-Prue, 1984). Plant maturity is most often determined by plant size as indicated by node or leaf number, not chronological age. Although plant size and leaf number are convenient measurements, they are affected by growing conditions and cannot be considered physiological constants (Brewster, 1985; Wellensiek, 1964). 16 30 inc gr: V6" 6X8 afle oil- flow The Vain A species’ responsiveness to vernalizing treatments often varies with age and stage of development. In some instances, the cold requirement can be partially satisfied during seed development on the mother plant; examples include red beet (Beta vulgaris var. rubra L.), Chicory (Cichon’um intybus L.), lettuce (Lactuca sativa L.), winter wheat and winter rye (Heide et al., 1976; Krekule, 1987; Schwabe, 1973; Wiebe, 1989). Plants that do not experience a juvenile period, including arabidopsis, winter rye, chicory, lettuce, and spinach (Spinacia oleracea L.), may be vernalized during seed germination or as small seedlings (Chouard, 1960). In fact, arabidopsis is reportedly more readily vernalized as a seed compared with juvenile plant; yet its ability to be vernalized increases several weeks after germination to an optimum after 45 to 90 days of growth (Napp-Zinn, 1969; Chandler and Dean, 1994). Additionally, incomplete, variable flowering may occur if plants are vernalized while making the transition from the juvenile to adult phase. For example, young plants of Lunaria biennis (L.) are less sensitive and require more cold than older plants (Wellensiek, 1958). Complete flowering of lunaria resulted after 12 weeks at 5 °C provided plants were at least 9 weeks old while flowering of 7-week-old plants was more variable and incomplete; younger plants failed to flower. The Vernalization Signal The nature of the vernalization stimulus remains unclear. Does vernalization change the cell so that a transmissible hormone is capable of being 17 €08 tern likeli, VGili ism: produced or perceived? Does vernalization lead directly to the production of a transmissible signal? In photoperiodic species, the leaf is the site of day length perception and produces a floral stimulus (the elusive florigen) that moves in the phloem to the apex where floral induction occurs. Researchers have been unsuccessful in isolating and determining “florigen” or “vemalin”, the compound proposed to be generated in response to vernalization. Interestingly, grafting experiments have shown that a photoperiodic floral stimulus can induce flowering in nonvernalized plants of cold-requiring, nonphotoperiodic species. Metzger (1988) found that nonvernalized field pennycress shoot tips were receptive to the photoperiodic floral induction stimulus produced by Sinapsis aIba scions grown under LD inductive conditions. It was hypothesized that field pennycress may have a similar or identical floral stimulus as Sinapis, although the induction mechanisms differ in these species. Gibberellin. The application of the plant hormone gibberellin (GA) enables some cold-requiring plants to flower without exposure to low temperatures thereby suggesting a role for GA in the vernalization process. It is likely that GA is involved in stem elongation responses associated with vernalization but may not be directly involved in the vernalization response. GA is most effective in causing flowering in LDP that rosette without exposure to cold, although it is not effective for all species (Chouard, 1960; Thomas and Vince-Prue, 1984). However, differences in flowering characteristics in response to GA compared with vernalization have been reported. Lang (1965) reported that GA substituted for vernalization in some biennial plants although shoot 18 elongation preceded flowering while in vernalized plants, flower formation preceded shoot elongation. GA only partially substituted for vernalization in Arabidopsis mutants with the effect of vernalization and GA being additive suggesting they operate in independent pathways (Chandler and Dean, 1994). Characterization of Vernalization Vernalization is an integral component of the life cycle of numerous floriculture crops, including many herbaceous perennial species, and their cold requirements for flowering must be met at some point during production. Therefore, an understanding of the specific low temperature requirements and responses of individual species is especially important, especially as greenhouse production expands to warmer climates and shifts to year round production. For plants that require vernalization, inadequate cooling may result in delayed or a complete lack of flowering. Similarly exposing plants to warm temperatures immediately following vernalization may adversely affect or completely reverse flower initiation and development. Facultative versus Obligate Responses. Although not all plants require or respond to vernalization, those that do are characterized as having a facultative (quantitative) or an obligate (qualitative) response to low temperature. Herbaceous perennials can be classified according to their vernalization response (Table 2). For plants exhibiting a facultative response, exposure to low temperatures promotes flowering; low temperatures generally accelerate flower timing and improve flower quality and uniformity but are not absolutely required 19 If II I lrlFr.ulll.1 IGQ C00 OCC v.1 U00 for flower induction. Winter annuals such as arabidopsis (Johanson et al., 2000) and numerous herbaceous perennials exhibit facultative responses to cold. For example, Agastache ‘Blue Fortune’ and Rudbeckia fulgida Ait. ‘Goldsturrn’ flower more quickly with fewer nodes following exposure to 5 °C compare to noncooled plants (Fausey et al., 2002; Runkle et al., 1999). Additional examples of facultative responses are presented in Table 2. Plants with an obligate vernalization requirement only flower following a sufficiently long cold treatment; plants do not flower without exposure to cold and generally will not rebloom once flowering has ceased. Some temperate biennial and perennial herbaceous species such as Digitalis sp. and Phox subulata have an obligate cold requirement which ensures that growth and flowering occur under favorable conditions following a sufficiently long period of winter (Chouard, 1960; Fausey et al., 2003; Runkle et al, 2001; Table 2). Interestingly, the obligate requirement can be “leaky” for a number of species, and a proportion of insufficiently cooled plants may flower provided other environmental factors such as photoperiod and light intensity are favorable. It is important to note that without cold, flowering of these plants is generally not as rapid or profuse as that of cooled plants. For example, Oenothera fruticosa ‘Youngii-lapsley’ generally requires a minimum of 3 weeks at 5 °C for flowering (Clough et al., 2001), and complete flowering of mature Lavandula angustifolia Mill. ‘Munstead’ plants occurs following 10-15 weeks exposure to 5 °C (Whitman et al., 1996). However, we have observed flowering of these plants with little or no cold when grown under long days with high daily light integrals (Fausey, unpublished data). 20 In di- 5: Ch l Ch 5 I m“ s de VG pc- EI‘ L V6 res Up Hc flo l l m. PO. to C ab: Quantitative Nature. Vernalization is a quantitative process where the degree of flower induction is a function of the temperature and duration of the Chilling treatment. The vernalization response becomes increasingly “stable” as the duration of exposure to effective chilling temperatures increases to a maximal, saturating level (Thomas and Vince-Prue, 1984). Once the vernalized state is achieved it is stable throughout subsequent mitotic divisions. If the chilling requirement for flowering is not completely fulfilled, flowering may be delayed or absent. Most vernalization studies use an empirical approach to determine the effectiveness of a particular vernalization treatment. Quantitative responses to vernalization in a population of plants include an increase in reproductive percentage, a decrease in leaf number below the inflorescence, intemode elongation, and a general decrease in time to anthesis with increasing degree of vernalization. Published methods to determine saturation of the vernalization response in wheat include obtaining the lowest total leaf number at anthesis Upon transfer from vernalizing to inductive conditions (Brooking, 1996). However, in our studies on vernalization of herbaceous perennials, rapid, uniform flowering of a population is often assessed by other factors in addition to leaf number. Saturating treatments generally result in complete flowering of a population (100%), minimum leaf number below the inflorescence, minimal time to anthesis, and high flower bud count compared to less effective treatments. Effective Vernalization Treatments. From an evolutionary perspective, a broad range of temperatures would be effective in vernalizing plants to 21 I III I I p I I l (L cc maximize the likelihood of flower induction in environments with fluctuating temperatures (Metzger, 1996). If flower induction only occurred after exposure to a narrow range of temperatures, reproduction and species success would be compromised. Effective vernalizing temperatures vary for a wide range of temperate species and are generally considered to be 1 to 5 or 6 °C although a broad range of -6 to 14 °C has been documented (Chouard, 1960). While vernalization is generally considered to be a feature of temperate plants, subtropical orchid species such as Zygopetalum and Phalaenopsis are also “vernalized” by exposure to temperatures of 14 to 17 °C and <29°C, respectively (Lopez et al., 2003; Robinson, 2002). The effective duration of exposure to vernalizing temperature varies considerably with Species and chilling temperature. Generally, plants with long cold requirements would be protected from blooming during the fall and early winter until favorable conditions for flowering return in the Spring and early summer. Most cold-requiring species reportedly require 4 or more weeks of vernalization for maximum flower induction (Metzger, 1996). In our experience, few cold-requiring herbaceous perennial species have required greater than 15 weeks of cold at 5 °C for flowering, and many require 6 or fewer weeks of vernalization at 5 °C (Table 2). In contrast, species such as Raphanus sativus L. Chinese Radish Jumbo Scarlet can be vernalized at 6 °C in as little as 8 days (Erwin et al., 2002). Classical studies with biennial strains of Hyoscyamus niger illustrated the quantitative nature of vernalization, a process whose intensity at effective 22 600 temperatures gradually increases with the length of exposure until saturation occurs (Lang, 1965; Vince-Prue, 1975). The effectiveness of vernalization is highly dependent on temperature following short durations of cold, yet all effective temperatures eventually promote similar flowering responses when the duration is sufficiently long (Lang, 1952). For example, vernalization of oxeye daisy (Leucanthemum vulgare Lam.) occurred in 6 weeks at the optimum temperature of 6 °C while greater durations of exposure up to 12 weeks were required for complete flowering at 9 °C (Heide, 1995). Varying vernalization responses among cultivars and ecotypes of a particular species have been extensively documented in arabidopsis (Sheldon et.al, 2000a), beet (Heide, 1973), carrot (Daucus carota L.) (Hiller and Kelly, 1979), celery (Ramin and Atherton, 1991), spring rape (Brassica napus L. var. annua) (Dahanayake and Galwey, 1998) and wheat (Brooking and Jamieson, 2002). Growth and flowering phenologies of ecotypes adapt to local environmental conditions such as photoperiod and temperature. For example, Northern Belgian populations of sugar beet (Beta vulgaris ssp. maritime) have a longer vernalization requirement compared with populations from more southern Sites which experience warmer, shorter winter seasons (Boudry, et al., 2002). Similarly, strong selection of arabidopsis ecotypes for a reduced vernalization requirement (early-flowering) occurs in some environments (Johansen et al., 2000). This was demonstrated in a comparison of 40 arabidopsis ecotypes showing a north-south distribution in vernalization response with the majority of ecotypes from Northern latitudes having a late-flowering phenotype 23 (vernalization-responsive) while the majority from central and eastern Europe have early-flowering phenotypes (Johansen et al., 2000). Stability of the Vemalized State. How permanent is the vernalized state? While few studies have investigated the longevity of the vernalized state, fully vernalized biennial henbane plants remained vegetative when held in noninductive short day (SD) conditions for up to 300 days and flowered if transferred to LB without a loss in the vernalized condition (Lang, 1965, cited by Chouard, 1960). Although the vernalized state is generally considered to be highly stable and irreversible once fully established, incomplete vernalization followed by adverse environmental conditions such as high temperature or short days can result in devernalization, a complete or partial reversal of flower induction (Purvis and Gregory, 1952; Lang, 1965, cited by Chouard, 1960). The degree of devernalization decreases with increasing degree of vernalization (Gregory and Purvis, 1952). The ranges of inductive and devernalizing temperatures are often close and found to be separated by a narrow range of neutral temperatures. Neutral temperatures (commonly 12-15 °C) act to stabilize the vernalized state before plants are exposed to higher, potentially devernalizing conditions (Thomas and Vince-Prue, 1984). Exposure to moderate temperatures (15-20 °C) for several days “fixes” the vernalized state and renders further devernalization ineffective (Chouard, 1960). Gregory and Purvis (1952) found reduced devernalization of winter rye by high temperature when plants were held at 15 °C for 2 or 4 days 24 teri following vernalization. For some plants, light during the chilling period appears to have a stabilizing effect on vernalization (Thomas and Vince-Prue, 1984). As with vernalization, effective devernalization conditions vary with Species, temperature, duration, and timing in relation to the vernalization treatment. Devernalization effects by high temperature have been reported in numerous crops including winter rye (Gregory and Purvis, 1952), carrot (Hiller and Kelly, 1979), celery (Ramin and Atherton, 1991), spring rape (Dahanayake and Galwey, 1998), and radish (Sheen, 2000), among others. Generally, temperatures from 18 to 40 °C, depending on species, are effective with the degree of devernalization increasing with increasing temperature (Chouard, 1960; Gregory and Purvis, 1952). Similarly, high day temperatures of 20 or 30 °C were effective in preventing vernalization of radish plants grown with cool night temperatures of 5 °C (Sheen, 2000). Schwabe (1955) reported complete devernalization of Chrysanthemum ‘Sunbeam’ by exposure to short day photoperiods with low light levels (25 footcandles) for four weeks following vemalization. In all reported cases, devemalized plants are fully revemalized by subsequent exposure to low temperatures. There is limited evidence that exposure to high temperatures before vernalization (antivemalization) can delay or nullify the effect of a vernalizing low temperature treatment. Spring rape genotypes exposed to 30 °C before a vernalization treatment produced more leaf nodes at flowering with increased time to flowering (Dahanayake and Galwey, 1998). Exposure to high temperatures before vernalization may decrease the vernalization response by 25 de ar qu ho siz Hc res an: delaying growth before the onset of effective vernalization. Antivemalization effects are overcome by prolonged exposure to vernalizing conditions. Light Effects on Vernalization Light Quantity. Light transmission in a greenhouse is often reduced by 40 to 60% due to several factors including glazing material, overhead structures, dust and aging, as well as weather patterns and time of year (Both and Faust, 2003). For example, in the northern United States, including Michigan (lat. 42° to let. 45°N), ambient greenhouse daily light integral (DLI) ranges from 2.5-10 mol-m’z-d‘1 from November to February and greater than 25 mol-m’z-d'1 in July and August (Korczynski et al., 2002). Vernalization studies have shown that light quantity prior to, during, and following vernalization can affect flowering; however, plant responses have been variable. lrradiance before chilling can affect flower initiation by influencing plant Size as measured by leaf number and carbohydrate status (Pierik, 1966). However, it is difficult to determine whether a size requirement for flowering results from the carbohydrate status of the plant or from physiological maturity and juvenility issues. As previously discussed, many cold-requiring herbaceous perennial species must reach a minimum size before they will flower or flower well in response to low temperature. Some examples include Hosta sp. (Fausey et al, 1999) and Aquilegia (Finical, 1998). Exposure to high daily light integrals (DLI) before vernalization may be expected to load plants with carbohydrates and enhance flowering. However, 26 Ii *.-=—. exposure to DLI of 14 mol-m‘Z-d‘1 for 5 weeks before vernalization reduced flowering of Aquilegia xhybn'da Sims ‘Remembrance’ and Lavandula ‘Hidcote Blue’ seedlings compared to those grown with a DLI of 4 mol-m'z-d'1 (Niu et al., 2002). In contrast, vernalization of onion (Allium cepa L.) positively correlated with irradiance (600 versus 200 umol-m'z-s") prior to chilling presumably due to increased soluble carbohydrate levels under the higher irradiance conditions (Brewster, 1985). Light during vernalization is also beneficial for herbaceous perennials that maintain foliage during the cold treatment (Elzroth and Link, 1970). In particular, maintaining light at 500 footcandles (~ 100 pmol-m’z-s'I) increased flowering percentage and reduced flower timing of Ajuga reptans and Dianthus caryophyllus compared to plants grown in the dark. Light during vernalization may play a regulatory role and Not act entirely through photosynthesis (Ramin and Atherton, 1994). Darkness during chilling completely prevented the vernalization response of celery cv. New Dwarf. However, very low irradiance of 0.2 W-m'2 (~1 umol-m'Z-s'I) promoted slight vernalization (33% flowering) while further 6 to 85 W-m‘2 (~ 30 to 425 pmoI-m'z-s") resulted in 100% flowering (Ramin and Atherton, 1994). In some species, substantial light levels may eliminate the vernalization requirement. In Graminae high irradiance levels have successfully replaced the obligate vernalization requirement of unchilled winter wheat and Festuca arundinacea grown at warm temperatures (Krekule, 1987). Other cold-requiring herbaceous perennials have been shown to flower completely under high light 27 conditions with no chilling, one example is Digitalis purpurea ‘Foxy’ (Fausey et al., 2003; Table 2). Light Quality. It is well-known that phytochrome is the principal photoreceptor for daylength responses in plant species including Arabidopsis and that flowering in long-day plants is promoted by far-red (FR: 700 to 800 nm) compared to red (R: 600 to 700 nm) light. The relative proportions of R and FR light (RzFR ratio) have also been shown to interact with vernalization to affect flowering. Bagnall (1993) found that vernalization and photoinduction have similar effects on flowering in arabidopsis and can substitute for one another. In particular, vernalization and growth of nonvernalized arabidopsis ecotypes Eifel, Innsbruck, and Pitztal under a R (660 nm) to FR (730 nm) ratio of 1.0 reduced time to flower and leaf number at flowering compared to plants grown with a R:F R of 5.8. Lee and Amasino (1995) also found that when compared to early flowering ecotypes Ler and Col, F R-enriched light accelerated flowering in nonvernalized Arabidopsis Ler and Col lines introgressed with FRI (Lee and Amasino, 1995). Vernalization at 4°C lessened the promotive effects of FR light on flowering in the introgressed lines after 30 to 40 days and completely eliminated responsiveness to light quality by 80 days. However, it was noted that in all cases, increased petiole length due to F R—enriched light was not affected by vernalization, and vernalization likely acts downstream of F R-light perception. Photoperiod. Many cold-requiring plants also have a photoperiodic requirement for flowering. Generally, the cold treatment precedes exposure to inductive photoperiods and may render plants more sensitive to day length 28 following vernalization. The most common requirement is for LD following the cold treatment, which naturally leads to flowering in late spring or early summer (Thomas and Vince-Prue, 1984). Photoperiod can partially or fully substitute for a vernalizing treatment in some plants. SD followed by LD successfully replace or partially replace the low temperature requirement for a number of LDP (Thomas and Vince-Prue, 1997). Some of these plants are termed short-long-day plants (Roberts et al., 1988). SD partially replace the chilling requirement for flowering of oxeye daisy (Heide, 1995) and enhance or fully substitute for vernalization in most temperate cereals including some winter wheat (Krekule, 1987) and barley (Hordeum vulgare L.) strains (Roberts et al., 1988). In a comparison often barley genotypes, ear initiation of a single strain ‘Arabi Abiad’ was advanced by ~5 leaves with SD or ~9 leaves with low temperature vernalization accompanied by marked decreases in time to flower compared to nonvernalized plants (Roberts et al., 1988). SD vernalization of barley seed is ineffective once the seed is adequately vernalized by low temperature. Early work with Campanula medium also found that nonvernalized plants exposed to SD followed by LD at warm temperatures flowered similarly to fully vernalized plants (Wellensiek, 1960). Long days can also substitute for vernalization in species such as Easter lily (Lilium Iongiflorum Thumb.) (Rees, 1987). Easter lilies require 6 weeks vernalization at 1.5 to 7 °C and generally receive some exposure to low temperatures while in the field before the bulbs are lifted. Although LD are not required for flowering, forcing bulbs under LD in the greenhouse can partially 29 substitute for incomplete cooling thus ensuring rapid growth and uniform flowering. For example, a minimum of 3 weeks of cold followed by 3 weeks of LD resulted in 100% flowering with a Slight increase in days from emergence to visible bud compared to fully vernalized plants (Dole and Wilkins, 1994). Concluding Remarks. Vernalization responses are extremely diverse, and we have only begun to understand the basis of vernalization in herbaceous perennial species. While molecular and genetic studies have increased our knowledge of the signaling pathways involved in flower induction of Arabidopsis thaliana, the extent of the overlap in ornamental Species remains to be seen. 30 Literature Cited Bagnall, DJ. 1993. Light quality and vernalization interact in controlling late flowering in Arabidopsis ecotypes and mutants. Ann. Of Botany 71:75-83. Bemier, G., J. Kinet, and RM. Sachs. 1981. The Physiology of Flowering. Volume I: The Initiation of Flowers. CRC Press, Boca Raton, FL. Both, A.J. and J. Faust. 2003. Light transmission in greenhouse design and coverings. Greenhouse Grower: 21(14). Boudry, P., H. Mccombie, and H.Van Dijk. 2002. Vernalization requirement of wild beet Beta vulgaris ssp. maritime: among population variation and its adaptive Significance. J. of Ecology 90:693-703. Brewster, J. L. 1985. The influence of seedling size and carbohydrate status and of photon flux density during vernalization on inflorescence initiation in onion (Allium cepa L.). Ann. of Botany 55:403-414. Brooking, I. R. 1996. Temperature response of vernalization in wheat: A developmental analysis. Ann. of Botany 78:507-512. Brooking, IR, and PD. Jamieson. 2002. Temperature and photoperiod response of vernalization in near-isogenic lines of wheat. Field Crops Research 79:21-38. Browse, J. and Z. Xin. 2001. Temperature sensing and cold acclimation. Current Opinion in Plant Biology 4:241-246. Chandler, J. and C. Dean. 1994. Factors influencing the vernalization response and flowring time of late flowering mutants of Arabidopsis thaliana (L.) Heynh. J. of Experimental Botany 45(278):1279-1288. Chouard, P. 1960. Vernalization and its relation to dormancy. Annual Review Plant Phys. 11:191-238. Clough, E., A.C. Cameron, R.D. Heins, and WC. Carlson. 2001. Growth and development of OeNothera fruticosa is influenced by vernalization duration, photoperiod, forcing temperatures and plant growth regulators. J. Amer. Soc. Hort. Sci. 126(3):269-274. Dahanayake, S. R. and NW. Galwey. 1998. Effects of interactions between low and high temperature treatments on flowering of spring rape (Brassica napus var. annua). Ann. of Botany 81:609-617. 31 Dennis, L. 2001. Genes for Flowering. Combined Proceedings International Plant Propagator’s Society 51:136-138. Elzroth, DE. and CB. Link. 1970. The influence of light during vernalization on the flowering response of Ajuga and Dianthus. J. Amer. Soc. Hort. Sci. 95(1):95-98. EnIvin, J. E., R.M. Warner, and A. G. Smith. 2002. Vernalization, photoperiod, and GA3 interact to affect flowering of Japanese radish (Raphanus sativus Chinese Radish Jumbo Scarlet). Physiologia Plantarum 115:298-302. Fausey, B. 1999. The influence of cold treatment, photoperiod, and forcing temperature on the dormancy, growth, and flowering of Hosta. MS. Thesis, Michigan State University, East Lansing, Michigan. Fausey, B., E. Runkle, R. Heins, A.Cameron, and W. Carlson. 2002. Herbaceous Perennials: Agastache. Greenhouse Grower 20(9):74-82. Fausey, B. A. Cameron, R. Heins, and E. Runkle. 2003. Herbaceous Perennials: Digitalis. Greenhouse Grower 21 (1 0):29-42. Finnegan, E.J., R.K. Genger, K. Kovac, W.J. Peacock, and ES. Dennis. 1998. DNA methylation and the promotion of flowering by vernalization. Proc. Natl. Acad. Sci. 95:5824-5829. F inical, L.M. 1998. The effects of plant age, photoperiod, vernalization, and temperature on the growth and development of Aquilegia and Gaura. MS. Thesis, Michigan State University, East Lansing, MI. Goodrich, J. and S. Tweedie. 2002. Rembrance of things past: chromatin remodeling in plant development. Annual Review Cell Div. Biology. 18:707-746. Henderson, l.R., C. Shindo, and C. Dean. 2003. The need for winter in the switch to flowering. Annual Rev. Genet. 37:371-392. Heide, OM. 1973. Environmental control of bolting and flowering in red garden beets. Scientific Reports of the Agricultural University of NonIvay. 52(15): 1-17. Heide, OM. 1995. Dual induction control of flowering in Leucanthemum vulgare. Physiologia Plantarum 95:159-165. Heide, O.M., O. Junttila, and RT. Samuelsen. 1976. Seed germination and bolting in red beet as affected by parent plant environment. Physiologia Plantarum 36:343-349. 32 Li Le Le Hiller, L.K. and WC. Kelly. 1979. The effect of post-vernalization temperature on seedstalk elongation and flowering in carrots. J. Amer. Soc. Hort. Sci. 104(2):253-257. Johanson, U., J. West, C. Lister, S. Michaels, R. Amasino, and C. Dean. 2000. Moloecular analysis of FRIGIDA, a major determinant of natural variation in Arabidopsis flowering time. Science 290:344-347. Kato, H., S. Taketa, T. Ban, N. lriki, and K. Mural. 2001. The influence of a spring habit gene, VRN-D1, on heading time in wheat. Plant Breeding 120:115—120. Koomneef, M., C. Alonso-Blanca, A. J.M. Peeters, and W. Soppe. 1998. Genetic control of flowering time in Arabidopsis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 492345-70. Korczynski, P.C., J. Logan, and J.E. Faust. 2002. Mapping monthly distribution of daily light integrals across the contiguous United States. HortTech. 12(1):12-16. Krekule, J. 1987. Vernalization in wheat, p. 159-169. In: J.G. Atherton (ed.). Manipulation of Flowering, ButtenNorths, London. Lang, A. 1952. Physiology of Flowering. Ann. Rev. Plant Physiol. 3:265-306. Lee, and R. Amasino. 1995. Effect of vernalization, photoperiod, and light quality on the flowering phenotype of Arabidopsis plants containing the FRIGIDA gene. Plant Physiol. 108:157-162. Lee, H., L. Xiong, Z. Gong, M. lshitani, B. Stevenson, and J. Zhu. 2001. The Arabidopsis HOS1 gene negatively regulates cold signal transduction and encodes a RING finger protein that displays a cold-regulated nucleo- cytoplasmic partitioning. Genes and Development 15:912-924. Levy, Y. Y. and C. Dean. 1998. The Transition to Flowering. The Plant Cell 10:1973—1989. Liu, J., S.J. Gilmour, M.F. Thomashow, and S. vanNocker. 2002. Cold signaling associated with vernalization in Arabidopsis thaliana does not involve CBF1 or abscisic acid. Physiol. Plant. 114:124-135. Lopez, R.G., E.S. Runkle, R.D. Heins, and CM. Whitman. 2003. Temperature and photoperiodic effects on growth and flowering of Zygopetalum Redvale ‘Fire Kiss’ orchids. Acta. Hort. 624:155-162. 33 R; Los, D.A., M.K. Ray, and N. Murata. 1997. Differences in the control of the temperature-dependent expression of four genes for desaturases in Synechocysteis sp. PCC 6803. Mol. Microbiology 25:1167-1175. Metzger, JD. 1988. Localization of the site of perception of therrnoinductive temperatures in Thlaspi arvense L. Plant Physiol. 88:424-428. Metzger, JD. 1996. A physiological comparison of vernalization and dormancy chilling requirement. In: Lang GA (ed) Plant Dorrnancy: Physiology, Biochemistry and Molecular Biology. CAB International, Wallingford UK, pp. 147-156. Michaels, SD. and RM. Amasino. 1999. FLOWERING LOCUS C encodes a Novel MADS domain protein that acts as a repressor of flowering. Plant Cell 1 1 :949-956. Michaels, SD. and RM. Amasino. 2000. Memories of winter: vernalization and the competence to flower. Plant, Cell and Environment 23:1145-1153. Mouradov, A., F. Cremer, and G. Coupland. 2002. Control of Flowering Time: Interacting pathways as a basis for diversity. The Plant Cell SI 11-S130. Niu, G., R. Heins, A. Cameron, and W. Carlson. 2002. Prevernalization daily light integral and vernalization temperature influences flowering of herbaceous perennials. HortScience 37(7):1028—1031. Orvar, B.L., V. Sangwan, F. Omann, R. Dhindsa. 2000. Early steps in cold sensing by plant cells: the role of actin cytoskeleton and membrane fluidity. Plant Journal 23:785-794. Pierik, R.L.M. 1966. Effect of light and temperature on flowering in Cardamine pratensis L. Z. Pflanzenphysiol. Bd. 56(141-152). Purvis, ON. and F.G. Gregory. 1952. Studies in Vernalization. XII. The reversibility by high temperature of the vernalized condition in Petkus winter rye. Ann. of Botany 16(1):1-21. Ramin, A. A. and J. G. Atherton. 1991. Manipulation of bolting and flowering in celery (Apium graveolens L. var. dulce) I. Effects of chilling during germination and seed development. J. of Hort. Sci.66(4):435-441. Ramin, AA. and JG. Atherton. 1994. Manipulation of bolting and flowering in celery (Apium graveolens L. var. dulce). lll. Effects of photoperiod and irradiance. J. of Hort. Sci. 69(5):861-868. 34 Rees, AR. 1987. Environmental and genetic regulation of photoperiodism - a review, p. 187-202. In: J.G. Atherton (ed.). Manipulation of Flowering, Buttenrvorths, London. Reeves, RH. and G. Coupland. 2001. Analysis of flowering time control in Arabidopsis by comparison of double and triple mutants. Plant Physiol. 126:1085-1091. Roberts, E.H., R.J. Summerfield, J.P. Cooper, and RH. Ellis. 1988. Environmental control of flowering in barley (Hordeum vulgare L.) l. Photoperiod limits to long-day responses, photoperiod-insensitive phases and effects of low-temperature and short-day vernalization. Ann. of Botany 62:127-144. Robinson, K. A. 2002. Effects of temperature on flower development rate and morphology of Phalaenopsis. MS. Thesis, Michigan State University. Rouse, D.T., C.C. Sheldon, D.J. Bagnall, W. J. Peacock, and ES. Dennis. 2002. F LC, a repressor of flowering, is regulated by genes in different inductive pathways. The Plant Journal. 29(2):183-191. Runkle, E.S., R.D. Heins, A.C. Cameron, and W.H. Carlson. 1999. Photoperiod and cold treatment regulate flowering of Rudbeckia fulgida ‘Goldsturrn’. HortScience 34(1):55-58. Runkle, E., R. Heins, A. Cameron, and W. Carlson. 2001. Herbaceous perennials: Phlox subulata. Greenhouse Grower 19(9):80-86. Salisbury, EB. and CW. Ross. 1992. Plant Physiology. 4th Edition. Wadsworth Publishing Co., Belmont, CA. Schwabe, W.W. 1955. Factors controlling flowering in the Chrysanthemum V. De—vernalizatin in relation to high temperature and low light intensity treatments. J. of Experimental Botany 6(18):435—450. Sheen, T. 2000. Effect of day and night temperature variation and of high temperature in radish. Acta Hort. 514:157-162. Sheldon, C. C., D.T. Rouse, E. J. Finnegan, W.J. Peacock, and ES. Dennis. 2000a. The molecular basis of vernalization: The central role of FLOWERING LOCUS C(FLC). Proc. Natl. Acad. Sci. 97(6):3753-3758. Sheldon, C. C., E. J. Finnegan, D.T. Rouse, M. Tadege, D. J. Bagnall, C, Helliwell, W.J. Peacock, and ES. Dennis. 2000b. The control of flowering by vernalization. Current Opinion in Plant Biology 3:418—422. 35 Simpson, G. G., and C. Dean. 2002. Arabidopsis, the Rosetta Stone of flowering time? Science 296:285—289. Snape, J.W., K. Butterworth, E. Whitechurch, and A.J. Worland. 2001. Waiting for fine times: genetics of flowering time in wheat. Euphytica 1 19: 1 85-1 90. Sung, S. and RM. Amasino. 2004. Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3. Nature 427:159-164. Thomas, B. and D. Vince-Prue. 1984. Juvenility. photoperiodism, and vernalization. In: Advanced Plant Physiology, M. B. Wilkins (ed.). Pp. 408- 439. Pitman Publishing, London. Thomas, B. and D. Vince-Prue. 1997. Photoperiodism in Plants. Academic Press, San Diego. Wellensiek, SJ. 1958. Vernalization and age in Lunaria biennis. Mededelingen van de Landbouwhogeschool te Wageningen, Nederland. 61:561-571. Wellensiek, SJ. 1960. Flower formation in Campanula medium. Mededelingen van de Landbouwhogeschool te Wageningen, Nederland. 60(7):1-18. Wellensiek, SJ. 1964. Dividing cells as the prerequisite for vernalization. Plant Physiology 39:832-835. Whitman, C.M., R.D. Heins, A.C. Cameron, and W.H. Carlson. 1996. Cold treatments, photoperiod, and forcing temperature influence flowering of Lavandula angustifolia. HortScience 31(7):1150-1153. Wiebe, H.J. 1989. Effects of low temperature during seed development on the mother plant on subsequent bolting of chicory, lettuce and spinach. Scientia Horticulturae 38:223-229. Wilsie, CF. 1962. Crop Adaptation and Distribution. W.H. Freeman and Company, San Francisco. 36 Table 2. Vernalization, photoperiod, and light quantity requirements for flowering of numerous herbaceous perennial species as well as general production timing and garden rebloom potential. filipendulina Plate', ‘Cloth of millefolium , ‘Terra Cotta', 0 ptarmica 15" or less tomentosa 15" or less 15" or less rupestris x reptans mollis 9 15* or less depressus 1 5" or less azurea 1 0 hupehensis 6 flabellata flabel/ata x hybrida LD required LD beneficial LD beneficial LD beneficial LD beneficial LD beneficial LD beneficial Day neutral LD beneficial LD beneficial Day neutral Day neutral Day neutral Day neutral Day neutral LD required Day neutral Day neutral Day neutral Day neutral 37 Yes Not tested Yes Not tested Not tested Not tested No Yes Yes Yes Yes Not tested NO Not tested Yes Yes Not tested Not tested Not tested Not tested Yes Table 2. Vernalization, photoperiod, and light quantity requirements for flowering of numerous herbaceous perennial species as well as general production timing and garden rebloom potential. maritima 10 or less Day neutral Yes No "meme 0 LD required Yes Yes 5 LD beneficial Not tested Not tested dumosus 15* or less LD-SD Yes Yes chinensis pumila 9 LD beneficial Yes No x hybn'da 10+ Day neutral Yes No 15" or less Day neutral Not tested Not tested 6-12 Day neutral Yes No macrophylla 5 LD required No 15* or less Day neutral No Yes 12 LD required Not tested Not tested carpatica 0 LD required Yes Yes garganica 6-9 LD beneficial Yes Not tested glomerata 15* or less LD beneficial Not tested Not tested Pe’SICIfOIIa 12 LD required Not tested Not tested 0 Day neutral Yes Yes 6-9 LD beneficial Yes Not tested Puncmta LD Yes Yes x Day neutral No LD required Not tested 38 Table 2. Vernalization, photoperiod, and light quantity requirements for flowering of numerous herbaceous perennial species as well as general production timing and garden rebloom potential. Ceratostigma Intermediate plumbaginoides 0 day plant’ 7 Not tested No 15* or less Che/one glabra beneficial LD SD 10 Yes Not tested Chrysogonum , 0 Day neutral 34 Yes Yes vrgrnranum Pierre C’micm’ga ”cam“ 15* or less LD 6-7 Not tested Not tested Clematis montana 15' or less - ‘John Paul II' beneficial LD reqUIred 14 Not tested Not tested gethra alnifolia 15" or less Day neutral 15 Not tested Not tested osea Coreopsis . _ 'Limerock Rub , 0 LD reqmred 6 7 Yes Yes Coreopsis auriculala 0 Day neutral 34 Yes Yes nana ,C°’°°ps’s Q’af’d’flma 0 LD required 7 Yes Yes Early Sunrise Coreopsis grandiflora _ LD required 'Baby Sun' and ‘Sunray' 10 15 LD beneficial 7 Yes Yes Coreopsis rosea . _ ‘Sweet Dreams' 0 LD reqUIred 7 8 Yes Yes Coreopsis verticillata . ‘Moonbeam' ‘Za re , 0 LD reqUIred 8 Yes Yes Corydalis 15* or less - ‘BlackbeLry Wine’ beneficial LD benefic'a' 4 Yes Yes Corydalis flexuosa ,, ,, “China Blue’ - 15 or less Day neutral 5-6 Yes Not tested Corydalis lutea bengficial LD beneficial 6 Yes Yes 15* or less Delaspemia cooperi beneficial Day neutral 9 Yes Yes Delphinium _ _. ‘Volke rfrie den’ 0 Day neutral 7 11 No Yes If cool Pelphinium °’?‘"{" 15' or less Day neutral 7-8 Yes Not tested Magic Fountains .Delphinium _ . grandiflorum ‘Summer b 5fi . l Day neutral 7-9 Yes Yiisgafdceilo' Blues’ ‘Blue Mirror' ene cra 39 Table 2. Vernalization, photoperiod, and light quantity requirements for flowering of numerous herbaceous perennial species as well as general production timing and garden rebloom potential. deltoides 5-15 Rose' beneficial Day neutral N0t tested Day neutral Yes 15" or less Day neutral Not tested grandiflora 0 Day neutral Yes obscure 5 Day neutral Yes Limited 5 Pu’Puma . LD beneficial Limited beneficral thapsi 15 LB beneficial o SD-LD LD“ beneficial 0 LD beneficial cheiri 8 Dust’ beneficial LD required 15* or less LD required 5 , Day neutral requrred epithymoides 15* or less Day neutral 10* or less Day neutral Not tested 0 Day neutral Yes xgrandiflora 9-12 ’ . 'BabY Cole' beneficial LD beneficial Yes Iindheimeri 0' LD beneficial Yes 40 Table 2. Vernalization, photoperiod, and light quantity requirements for flowering of numerous herbaceous perennial species as well as general Whirling Butterflies', production timing and garden rebloom potential. ‘Siskyou Pin ’ Iggrzaa’m': 0 Day neutral 5-6 Yes Yes i——.— . Ge’an’u’" da’mat’cu’" 6 LD beneficial 7 Yes No Geranium sanguineum 5 LD before, DN ‘New Hampshire Purple’ beneficial after cold 5 Yes Yes Geranium . xcantabn‘giense 1: orfilesls LD beneficial 4 Yes Not tested ‘Biokovo' ene Cla Geum coccineum . . ‘Borisii', “Mrs. Bradshaw 6-15 Day neutral 6-8 Yes letted Gypsophila paniculata 15" or less . 'Double Snowflake' beneficial LD benefiCIal 1,2 Yes N°t tested :Gypsophila paniculata 15' or less . ‘Happy F estiv al' beneficial LD beneficral 8 Yes Not tested Helenium 'Mardi Gras' 0 Day neutral 4-6 Yes Yes Helenium autumna/e - , , 0 LD reqUIred 10-14 Yes Yes Bruno , and others W’amus, o LD-SD 12-14 Yes Yes Low Down Hemerocallis 8-10 'Stella D'Oro’ varies Day neutral 8 Yes Yes Heuchera cvs. 10 Day neutral 5-8 Yes Limited Heucherella cvs. 9-12 Day neutral 5-6 Yes Yes Hibiscus moscheutos . 'Disco Belle' and others 0 LD requrred 9 10 Yes Yes I-Iosta 'Francee', 3 LD required 9 Yes No Hyacrnthlna Hosta ‘Golden Tiara’ . and ‘Golden Sce te r’ 3 LD requrred 12 Yes No Hosta 'Lanclfolia' and . 'Undulata Van'e ata' 0 LD requrred 9 Yes No Hosta p/antaginea - and 'Ro al Stan dar d' 0 LD requrred 14+ Yes No 41 Table 2. Vernalization, photoperiod, and light quantity requirements for flowering of numerous herbaceous perennial species as well as general production timing and garden rebloom potential. . ,y . ,.,;, -:: rant 42 Hosta - 'Tokudama' 6 LD requrred 11 Yes No lben's sempervirens 'Snowflake' and others 10 Day neutral 3 Not tested No Lamiu’" maculatum 0 Day neutral 4-6 Yes Yes icultlvars Isotoma axillan's 5 LD required 10 Yes Yes Lavandula angustifolia - 'Hidcote', ‘Munstead’ 10-15 LD benefiCIal 7 Yes Yes ,La‘fiamu’a 5mm”, 0 LD beneficial 55 Yes Yes Chlca RoseI Purple Lavandula stoechas - ‘Coco Blue &White’ 0 LD beneficral 8 9 Yes Yes Leucanthemum x - su rbum “Beck , 0 LD requrred 8-9 Yes Yes Leucanthemum x 6 _ . superbum 'Snow Cap’, beneficial LD beneficial 6-7 Yes Limited 'lce Star’ Lewisia cotyledon 15* or less Day neutral 11 Yes Limited ””5“? Pu’Pu’ea 15' or less Day neutral 6-7 Yes Not tested ,L’"“”’ 9959"” 10 Day neutral 5 Yes Limited Sapphire Lobelia xspeciosa 6 - ‘Compliment Scarlet’ beneficial LD benefiCIal 9 Yes NOt tested Lychnis flos-cucu/i 15. or less Day neutral 4 Yes Yes ‘Jenny' after cold Lysimachia ciliata 0 LD required 7-8 Yes Yes Firecracker .Lysimachia p un ctata 0 LD required 8 Yes Yes ‘Alexander’ Malva alcea 0 LD required 8 Yes Not tested Mondara didyma 15' or less LD required 9 Yes Yes 'Cambridge Scarlet’ beneficial Nepeta faassenii ‘Blue Wonder’, 0 LD required 4-6 Yes Yes Walkers Low’ Table 2. Vernalization, photoperiod, and light quantity requirements for flowering of numerous herbaceous perennial species as well as general production timing and garden rebloom potential. Nepeta faassenii ‘Snowflake' 0 Day neutral 34 Yes Yes Nepeta subsessilis 0 LD beneficial 4-5 Yes Yes Oenothera . - ‘Cold Crick‘ 15 or less LD reqUIred 5 Not tested Not tested Oenothera fruticosa ‘Fireworks', 'Summer - Solstice’, ‘Highlight', 3+ LD requrred 3 Yes No 'Youngii-Lapsley' °°"°"’°.'° . 1045. LD beneficial 7 Not tested Not tested mrssounensrs beneficral Oenothera speciosa 0 LD 6 Yes Yes SISKIYOU beneficial Osteospermum 15' or less ‘Purple Mountain‘ beneficial Day neutral 4 Yes Yes Oxalis crassipes 15' or less . ‘Rosea' beneficial LD reqUIred 5-6 Yes Yes iPenstemon . ‘Violet Dusk’ 0 LD benefiCIal 5.5 Yes Yes IEeHIi'I's’temon barbatus 15* or less LD required 5-6 Yes Yes Penstemon . , 15 °' "5.55 LD beneficial 12 Yes Not tested campanu/atus Garnet benefiCIal Penstemon digitalis 3 - _ ‘Husker Re d’ beneficial LD beneficral 9 10 Yes Not tested iPenstemon heterophyllus 10 LD beneficial 10 Yes Not tested ‘Electric Blue’ Penstemon mexicale . ‘Pike's Peak' 0 LD beneficral 5-6 Yes Yes Perovskia atn‘plicifolia _ ‘Fili ran', ‘Little S ire’ 0 Day neutral 10 12 Yes No Phlox divan'cata 'Laphammi’, ‘London 15* or less Day neutral 3 No No Grove’ PM” Pa”’°“’°‘° 0 LD required 10 Not tested Not tested ‘Eva Collum’l ‘Tremor’ 43 Table 2. Vernalization, photoperiod, and light quantity requirements for flowering of numerous herbaceous perennial species as well as general at“ Phlox subulata numerous cultivars production timing and gard Day neutral No Phygelius aequalis 'Yellow Trum et’ 0 Day neutral 8-10 Yes Yes Physostegia virginiana 6 LD required 10 Not tested Not tested Rosea , and others Platycodon gran diflorus 0* Day neutral 10-12 Not tested Not tested Sentimental Blue Polemonium . 4-6 ‘Brise D'Anjou', others 0 LD varies Yes No Egg? on um affine beneficial LD required 13 Not tested Not tested Potentilla atrosanguinea . ‘Mrs Willmott' 15 or less Day neutral 6 Yes Yes Primula veris 10 ‘Pacific Giants' beneficial Day neutral 5 Not tested Not tested Pulmonan'a 6 Day neutral 2-3 No No man cultivars Rodgersia aesculifolia 15" or less LD required 6-8 Not tested Not tested Rudbeckia fulgida - _ ‘Goldsturm‘ 0 LD reqUIred 11 12 Yes Yes Salvia guaranitica ‘Black and Blue’ 0 Day neutral 5-6 Yes Yes Salvia nemorosa 0 ‘Caradonna’ beneficial Day neutral 4 Yes Limited ‘May Night’, ‘Marcus’ wllow light S l ' 0 , a we superb a beneficial LD beneficial 5.5 Yes Yes Blue Queen . wllow light saX’I’aga . 9+ Day neutral 4-5 Not tested No s ecres and cultivars Scabiosa caucasica 1 0'1 5 ‘Butterfly Blue‘, others beneficial Day neutral 5 Yes Yes ,Sedu’" 'M°'?"°"'v 0 LD for all 12 Yes Yes Autumn Jo , others Sida/cea 15* or less . l 1 s Y 5 ‘Party Girls' beneficial Day neutra 0 Ye e 44 Table 2. Vernalization, photoperiod, and light quantity requirements for flowering of numerous herbaceous perennial species as well as general production timing and garden rebloom potential. ‘Sisyn‘nchium LD before, DN . . angustifolium 'Lucerne' 0 after cold 5 Yes L'm'ted . . . . LD before, DN . . SIsynncthm beIIum 0 after cold 5 Yes Limited Sisyrinchium tinctorium SD or LD - but _ . . ‘Puerto Yellow' 0 Not 16 h 7 8 N° L‘m'ted Solidago _ _ 'Fireworks' 0 LD SD 12 15 Yes Yes ,S°”dag° Sp”°°,°’°t° 9 SD beneficial 13 Not tested Not tested Golden Fleece ,S‘afhys F°°°’”°a 0 LD beneficial 5-6 Yes Yes Chinook Stachys macrantha 15' or less LD beneficial 8-9 Yes No Hummelo Stokesia laevis . . - . . Faculatatlve Klaus Jellrto , Purple 6-9 int errn e di at e‘ 9-11 Yes No Parasols , many others ,Tanacetu’" ”New" 15- LD 5 Yes Not tested Jackgt’ Thalictrum 15" or less . aquilegifolium beneficial LD benefiCIal 8 Not tested Not tested Thalictrum kuisianum 9 Day neutral 6 Yes Not tested Thymus serpyllum 6 ‘Pink Chintz' and others I beneficial LD 8 Yes Yes Tiarella species and cultivars 0* Day neutral 3-5 Yes Yes Tn'cyn‘is hirta 15* or less . ‘Miyazaki‘ beneficial LD reqUIred 15—1 6 Yes No Verbascum 15' or less . . ‘Southern Charm’ beneficial Day neutral 6 Yes L'm'ted Veronica ‘Sunny Border Blue’ 6 Day neutral 9 Yes Yes Veronica longifo/ia 'lcicle' 6 Day neutral 7-9 Yes Yes 45 Table 2. Vernalization, photoperiod, and light quantity requirements for flowering of numerous herbaceous perennial species as well as general production timing and garden rebloom potential. Veronica spicata ‘Red Fox', 'Goodness 5-6 Day neutral Grows’ Number of weeks at 5°C(41°F) for complete vernalization; ' indicates the only duration tested 2Short day (SD), long day (LD) or either (DN) are required or beneficial for flowering 3High light levels may improve flowering characteristics: percentage, uniformity, or flower number ‘The ability of a plant to rebloom once initial flush of flowering has ceased. 46 CHAPTER II: DAILY LIGHT INTEGRAL AFFECTS FLOWERING OF GREENHOUSE-GROWN ACHILLEA, GAURA, AND LAVANDULA Fausey, B.A., R.D. Heins, and AC. Cameron. 2005. Daily light integral affects flowering of greenhouse-grown Achillea, Gaura and Lavandula. HortScience 40(1): 114-118. 47 Daily light integral impacts flowering and quality of greenhouse-grown Achillea, Gaura, and Lavandula Beth A. Fausey, Royal D. Heins, and Arthur C. Cameron Department of Horticulture, Michigan State University, East Lansing, MI 48824- 1325 Additional index words. herbaceous perennial, light quantity, flower induction Abstract The growth and development of Achillea x millefolium L. ‘Red Velvet’, Gaura Iindheimeri Engelm. & Gray ‘Siskiyou Pink’ and Lavandula angustifolia Mill. ‘Hidcote Blue’ were evaluated under average daily light integrals (DLI) of 5-20 mol-m’z-d". Plants were grown in a 20 °C glass greenhouse with a 16-h photoperiod under four light environments: 50% shading of ambient light plus PPF of100 umol-m°-s"(L1); ambient light plus PPF of 20 pmoI-m‘z-s" (L2); ambient light plus PPF of 100 umol-m‘z-s'1 (L3); and ambient light plus PPF of 150 pmol-m’z-s“ (L4). Between 5-20 mol-m'z-d'I, DLI did not limit flowering and had little effect on timing in these studies. Hence, the minimum DLI required for flowering of Achillea, Gaura and Lavandula must be <5 mol-m'Z-d", the lowest light level tested. However, all species exhibited prostrate growth with weakened stems when grown at DLI,1O mol-m'z-d“. Visual quality and shoot dry mass of Achillea, Gaura and Lavandula linearly increased as DLI increased from 5-20 48 moi-ii declir result prodc lntroc drama for a v day le quality of son 700 nr flowerl to lat, 2 Noven~ and no require Perenn Ihese re 2003) IEVBIS a mol-m'z-d'1 and there was no evidence that the response to light was beginning to decline. While 10 mol-m'z-d'1 has been suggested as an adequate DLI, these results suggest that 15-20 mol-m'Z-d'1 should be considered a minimum for production of these herbaceous perennials. Introduction Greenhouse production of flowering herbaceous perennials has increased dramatically in the past decade. Though forcing schedules have been developed for a wide range of perennials (Heins et al., 1997), seasonal variations in light, day length and temperature provide challenges for consistent production of high quality crops. Northern growers have reported incomplete or delayed flowering of some perennials during winter production. DLI, the cumulative number of photosynthetically active photons (400 to 700 nm) received in a 24-h period, is a useful variable related to plant growth and flowering responses. In the northern United States, including Michigan (lat. 42° to lat. 45°N), ambient greenhouse DLI ranges from 2.5-10 mol-m'z-d'1 from November to February (Korczynski et al., 2002). At higher latitudes in the UK and northern Europe, DLI during this same period could be lower. While DLI requirements vary considerably with species, it has been generalized that most perennials require 10-16 mol-m’Z-d'1 for minimum acceptable quality, though these recommendations are subjective and not based on specific data (Faust, 2003). The consequences of growing perennials at DLI below recommended levels are not well documented. 49 cause i instanc geranil Heins, respe. Descl Warn and ii ‘l. H re(lull lespo eXiStir It is well known that photosynthesis increases asymptotically with instantaneous light and the response is species specific. In whole-plant studies, the relationship between DLI and dry mass (DM) is generally linear up to 20-30 mol-m'Z-d‘1 or higher for a diverse group of plants such as Chrysanthemum, corn, and cucumber (Warrington and Norton, 1991). Depending on species, increasing DLI also improves numerous plant quality Characteristics such as lateral branching and flower number (Faust, 2003). The optimum DLI for sun species can be as high as 250 mol-m'z-d", far exceeding levels that can be achieved in greenhouses (Warrington and Norton, 1991). Minimum DLI requirements for flowering can be very low, and low DLI can cause incomplete or delayed flowering of some potted flowering crops. For instance, Chrysanthemum “Bright Golden Anne’ (Karlsson et al., 1989b), geranium ‘Red Elite’ (White and Warrington, 1988), and African violet (Faust and Heins, 1994) required minimum DLls of only 1.8, 3.3 and 2.0 mol-m'z-d'I, respectively, to initiate flowering. In field studies, however, the grass Deschampsia flexuosa (L.) Trin. did not flower at DLls < 9 mol-m‘Z-d'1 (Foggo and Warrington, 1989). Warner and Erwin (2003) studied several Hibiscus species and found that most flowered even at the lowest light levels tested (< 5 mol-m'z-d' 1). However, the herbaceous perennial Hibiscus moscheutos ‘Disco Belle’ required at least 14 mol-m‘z-d'1 to consistently flower (Warner and Erwin, 2003). The objective of this study was to characterize growth and flowering responses of select herbaceous perennial species to a range of DLI using existing greenhouse production protocols. We selected three common species 50 known Gaura Specif expect sun pla Materi Velvet nodes and 05 500 pp Ore.) a come Mich.) media Plants mEdia. Commf were 8 aVelaqi 5°Cc known to be sun-loving plants in the garden: Achillea millefolium ‘Red Velvet’, Gaura Iindheimeri ‘Siskiyou Pink’, and Lavandula angustifolia ‘Hidcote Blue’. Specifically, we quantified plant responses to DLI between 5-20 mol-m'z-d", expected levels under greenhouse conditions but well below expected optima for sun plants. Materials and Methods Plant material and culture. Vegetative cuttings of Achillea millefolium ‘Red Velvet’ with 5 to 8 nodes and Gaura Iindheimeri ‘Whirling Butterflies’ with 3 to 4 nodes were taken from stock plants on 08 Jan. (Achillea only), 05 Feb., 19 Feb., and 05 Mar. 2001. Cuttings were dipped into 1000 ppm indoIe-3-butyric acid and 500 ppm naphthalene acetic acid rooting hormone (Dip ‘N Grow, Clackamas, Ore.) and placed in 72-cell plug trays (50-mL cell volume) filled with a commercial peat-perlite media (Sure-Mix, Michigan Grower Products, Galesburg, Mich.). Cuttings were rooted for 3 weeks in a propagation house with air and media temperatures of 23 and 26 °C, respectively. Once rooted, 40 uniform plants were transplanted to 13-cm (1 .1-L) square containers filled with the above media. Lavandula angustifolia ‘Hidcote Blue’ seedlings were received from a commercial plug producer on 22 Oct., 19 Nov., and 17 Dec. 2000. Seedlings were subsequently grown under natural photoperiods for 10 weeks until they averaged 11 to 14 Nodes. To induce flowering, seedlings were then placed in a 5 °C cooler for 9 weeks with 10 LImOI-m'z's'1 provided by cool white fluorescent 51 lamps ’7 descnb I I 1| dieint i 1andi 125K-l lamps for 9 h-d". Following cold treatment, plants were transplanted as described for Achillea and Gaura. Plants were fertilized at every irrigation with well water (EC of 0.70 mS-cm' 1 and 105 mg Ca, 35 mg Mg, and 85 mg S-L“) acidified with sto. to a titratabie alkalinity of 130 mg CaC03-L'1 and water soluble fertilizer providing 125N—12P— 125K—13Ca mg-L'1 (30% ammoniacal N) plus 1.0Fe—0.5Mn—0.52n—0.SCu—0.1B- 0.1Mo mg-L'1 (MSU Special; Greencare Fertilizers, Chicago, Ill.). Environmental control. Plants were grown in a 20,-: 4 °C glass greenhouse with a 16-h photoperiod provided by 400-watt high pressure sodium (HPS) lamps from 0600 to 0800 HR and 1700 to 2200 HR with varying PPF. Ten plants of each species were placed under each of four light environments: 50% shading of ambient light plus PPF of 100 pmol-m'z-s'1 (L1); ambient light plus PPF of 20 umoI-m'Z-s'1 (L2); ambient light plus PPF of 100 umoI-m'z-s'1 (L3); and ambient light plus PPF of 150 pmol-m‘z-s'1 (L4) (Fig.1). Additional supplemental lighting from HPS lamps provided a PPF of 100 pmol-m'z-s'1 for L1, L3 and L4 Or 20 timol-m'z-s'1 for L2 when the ambient greenhouse PPF dropped below 200 pmol-m‘z-S'1 from 0800 to 1700 HR; supplemental lighting ceased when the PPF exceeded 400 pmoI-m'z-s'I. When combined with the multiple forcing dates, these light treatments created 16 average DLls for Achillea and 12 average DLls for Gaura and Lavandula. Instantaneous light in each treatment was measured at plant height with two LI-COR quantum sensors (model Ll-189; Ll-COR, Lincoln, Nebr.) connected to a CR10 datalogger (Campbell Scientific, Logan, Utah). Greenhouse air 52 gauge dalalci and it from i . hmmgl each s anthes anthes hstfle calcula numbe. treating Achille, onetOi Followil harvest measur I PROCI regress, temperatures were controlled by a climate-control computer (Priva, Model CD750, De Lier, The Netherlands) and were monitored on each bench with 36- gauge (0.127-mm diameter) type-E thermocouples connected to a CR10 datalogger. Temperature and light measurements were collected every 10 s, and the hourly average was recorded. The average daily temperature and DLI from forcing to visible bud, to anthesis, from visible bud to anthesis, and from forcing to final plant quality and DM assessment (see below) were calculated for each species. Data collection and analysis. The date of visible flower bud, date of anthesis, stem length at visible bud and anthesis, lateral inflorescence number at anthesis, and final node number below first open flower were measured. Time to visible bud, time to anthesis, and time from visible bud to anthesis were calculated and used in analyses. Flowering percentage was calculated as the number of flowering plants divided by the total number of plants in each treatment. Plant quality was assessed 8, 6 and 7 weeks after transplant for Achillea, Gaura and Lavandula, respectively, using a numerical rating scale from one to five where one equaled poor quality and five equaled high quality. Following quality assessment, the above-ground biomass of each plant was harvested and dried in a 60 °C forced-air oven for 6 d after which shoot DM was measured for each species. Linear regression analysis was performed on treatment means using PROC REG procedures in SAS version 8.0 (SAS Institute, Cary, NC). Linear regression lines are presented only when the correlation was statistically 53 snnmc _.,l wube usedir confide Resuh send; find-Wk fig.h expenr underl Lavanfi ihdcot Propag Unionr prOSIra "iol-m‘2 three SI f0rAch significant. Calculated DLI levels from forcing to visible bud, to anthesis, from visible bud to anthesis, or to final plant quality assessment and harvest were used in regression analyses, depending on variable. Error bars represent 95% confidence intervals. Results Light levels. Average DLI ranged from 5—20 mol-m'Z-d'1 depending on the start date and treatment (Table 1). DLI generally increased from late January to mid-March and then was relatively stable until whitewash was applied in late April (Fig. 1). Average daily temperature rose by 1-2 °C during the course of the experiment (Table 1). Flowering percentage. All but one Achillea and all Gaura plants flowered under DLI ranging from 5-20 mol-m'z-d‘1 (Table 1). Flowering percentage of Lavandula varied from 60 to 100% but did Not correlate with DLI. Lavandula ‘Hidcote Blue’ is an open-pollinated cultivar and inherent variability of the seed- propagated starting material may have contributed to the lack of flowering uniformity. Growth and development responses. At low DLI, each Species exhibited prostrate growth with weakened stems (Fig. 2A-C). As DLI increased above 10 mol-m‘Z-d", stem strength, habit, lateral branching, and general appearance of all three species improved dramatically. Achillea developed pink flowers that failed to deepen in color under low DLI. Basal shoot development (number and length) for Achillea was also reduced under low DLI. Lavandula grew asymmetrically 54 Ge DL trip per mol- linea fold l molr receil “th r weeks inclea. git/ant increa for La v. and failed to develop gray foliage when grown with low DLI. A qualitative rating scale was developed based on these observed changes in growth and form (Table 2). Quality ratings of Achillea, Gaura, and Lavandula were linearly correlated with DLI from 5-20 mol-m'ztd", though there was substantial variation (Figs. 3A-C). Lateral inflorescence number increased linearly with DLI for Achillea and Gaura, while there was no significant correlation for Lavandula (Figs. 3D-F). As DLI increased from 5-20 mol-m'Z-d“, inflorescence number of Gaura nearly tripled from 7 to 19, and more than doubled for Achillea. Gaura flower number per inflorescence nearly doubled from ~25 to 2 40 as DLI increased from 5-20 mol-m‘z-d'1 (data not shown). Shoot DM accumulation of Achillea, Gaura, and Lavandula increased linearly with increasing DLI (Figs. 3G-l). DM increased 3-fold for Achillea, 3.4 fold for Gaura, and 2.6 fold for Lavandula as DLI increased from 5—20 mol-m'z-d". Achillea plants accumulated the most shoot biomass per unit of light received (0.96 g-plant‘I-DLI'1 in 8 weeks; Fig 3G) which infers that for each mol of light received above 5 moI-m'Z-d", Achillea gained nearly 1 g DM per plant in 8 weeks. Gaura had a relatively small rate of increase in DM accumulation with increasing DLI (0.15 g-plant'I-DLI'1 in 6 weeks). Lavandula accumulated 0.28 g-plant'I-DLI'1 in 7 weeks. When calculated over each forcing duration, DM increased 17.9 mg-plant'l-DLI'1 for Achillea (56 d), 2.9 for Gaura (42 d) and 5.7 for Lavandula (49 d) for plants grown with > 5 moI-m'Z-d". 55 II'ICIe lorA unafi vein anhe bud,f absol fidme howev d with Dhcm 0My8 Wiih a memj: I°West Plants of all species had slightly shorter stems at visible bud with increasing DLI (Figs. 3J-L). Stem length at visible bud decreased 24% (4.8 cm) for Achillea, 35% (2.4 cm) for Gaura, and 35% (5.4 cm) for Lavandula as DLI increased from 5-20 mol-m'z-d". Stem length at first flower was not correlated with DLI for any of the three species (Fig. 3J-L), though actual plant height was reduced at low DLI since stems were prostrate and could not support their weight (Fig. 2A-C). Flower Timing. Flower timing of each species was either reduced or unaffected by increasing DLI. For Achillea, DLI reduced the time from forcing to visible bud and from forcing to anthesis, but not significantly from visible bud to anthesis (Fig. 4A, D, G). For Gaura, DLI reduced the time from forcing to visible bud, forcing from visible bud to anthesis, and from forcing to anthesis although absolute differences were only 3 to 5 d (Fig. 48, E, H). Time from forcing to visible bud and anthesis did not vary significantly with DLI for Lavandula; however, time from visible bud to anthesis decreased, although minimally, by 1.5 d with increasing DLI (Fig. 40, F, l). Discussion Gaura had the highest DLI requirement of the three species tested with only 8 of 120 plants attaining the highest quality rating. Gaura had to be grown with a DLI, 18 moI-m'z-d'1 to consistently achieve a quality rating of 3 or greater, the minimum rating deemed acceptable for consumer sales. Gaura also had the lowest DM accumulation though inflorescence and flower number increased 56 subsl State July 1 Lava Laval conn) ophni challe receiv essen | photo: | satura I found I menu accurri range CIirysa I hand. i992). l i aCCUm, substantially with increasing DLI. Gaura is native to the southwestern United States where ambient light levels may be 55-60 mol-m'z-d'1 from May through July (Korczynski et al., 2002). To consistently achieve quality ratings of 4 or greater, Achillea and Lavandula plants required DLI 220 mol-m'z-d’1 (Figs 3A, C). Achillea and Lavandula are both native to Mediterranean regions with high light levels compared to the northern United States. Achieving DLI of > 20 mol-m'z-d’1 for optimal production of these species in northern commercial greenhouses is challenging, particularly during the winter months. Achillea plants accumulated the most shoot biomass per unit of light received (18 mg-plant'I-d'I-DLI") and the response of all three species was essentially linear between 5-20 mol-m'z-d‘I. There was no evidence that the photosynthetic response had begun to level off at 20 mol-m'Z-d", implying that saturation occurs at even higher light levels. Warrington and Norton (1991) found that DM of corn (Zea mays L.) increased linearly by approximately 18.4 mg-plant‘1-d'1-DLI'1 (29 d) with increasing DLI from 11.3-73.5 mol-m'Z-d" while DM accumulation of cucumber, Chrysanthemum, and radish were also linear in the range of DLIS (5-20 mol-m'z-d") investigated in this study. DM of Chrysanthemum ‘Bright Golden Anne’ grown at 20 °C increased linearly and more than doubled as DLI increased from 58-22 mol-m'2~d'1 (Karlsson and Heins, 1992). With increasing DLI, Lavandula had a relatively low shoot DM accumulation (Fig. 3|) and no Significant increase in inflorescence number (Fig. 57 3F), yet a marked increase in plant quality (Fig. 3C). The increase in quality was based more on stem strength, habit, form and foliage color than on DM or inflorescence number. DLI did not limit flowering and had little effect on timing in these studies. Hence, the minimum DLI required for flowering of Achillea, Gaura and Lavandula must be <5 mol-m'z-d'I, the lowest light level tested. AS DLI increased from 5-20 mol-m'z-d", time to flower was accelerated only by about 5 d for Achillea, 7 d for Gaura and not significantly for Lavandula. Node number below the inflorescence correlated with DLI for Achillea (p<0.0169; y=-0.185x +17.79) but not for Gaura or Lavandula. This indicates that Achillea plants may have flowered earlier with increasing DLI due to irradiance effects on stage of development. It is likely that timing could be more affected as DLI drops below 5 mol-m'z-d“. For instance, decreasing DLl from 5.8 to 1.8 mol-m"2-d'1 increased time to flower by 20 d for Chrysanthemum ‘Bright Golden Anne’ grown at 20 °C (Karlsson et al., 1989a). Pearson et al. (1993) also noted a sharp decline in Chrysanthemum floral development rate when grown at light levels below PAR radiation integral of 1 MJ-m'Z-d" (~5 moI-m'Z-d“). It is possible that the small decrease in flower timing observed in these studies may have resulted from associated increases in plant temperature with increasing DLI. Supplemental lighting with HPS lamps provides photosynthetic and nonphotosynthetic radiation energy that increases plant temperature and developmental rate thereby decreasing production time (Faust and Heins, 1997). Supplemental lighting at 50, 75, or 100 umol-m‘z-s‘1 from HPS lamps increased 58 vinca (Catharanthus roseus L.) plant temperature by 1.2, 1.5 and 1.7 °C, respectively, above air temperature (Faust and Heins, 1997). While plant temperature was not measured in this study, average differences in air temperature varied by ~2 °C between the L2 lighting treatment compared with other lighting treatments utilizing supplemental lighting (Table 1). While we presume plant meristem temperature may be greater under high DLI and with HPS lamps, time from visible bud to anthesis of Achillea was not affected by DLI and the time from visible bud to anthesis of Gaura and Lavandula was only affected slightly by 3 and 1.5 d, respectively (Fig. 2D-F). Linear increases in plant quality, DM accumulation, and inflorescence number with increasing DLI from 5-20 mol-m'z-d'1 illustrate the impact of DLI on growth and flowering of herbaceous perennial species. Prostrate habit with reduced stem strength and lateral branching are indicative of plants receiving too little light. While 10-16 moI-m'z-d'1 has been suggested for production of acceptable herbaceous perennials such as Campanula carpatica ‘Blue Clips’ (Faust, 2003, Niu et al., 2001), 20 mol-m'Z-d'1 is more appropriate for producing high-quality perennials from small starting material for these three selections. On a practical level, growers in northern climates should consider methods that optimize production and greenhouse light levels. 59 Literature Cited Faust, J.E. 2003. Light. p. 71-84. In: D. Hamrick(ed.). Ball Redbook, 17th edition, Volume 2: Crop Production. Ball Publishing, USA. Faust, J.E. and RD. Heins. 1994. Modeling inflorescence development of the African violet (Saintpaulia ionantha Wendi.) J. Amer. Soc. Hort. Sci. 1 19(4):?27-734. Faust, J.E. and RD. Heins. 1997. Quantifying the influence of high-pressure sodium lighting on shoot-tip temperature. Acta Hort. 418285-91. Foggo, MN. and I.J. Warrington. 1989. The influence of photosynthetically active radiation and vernalization on flowering of Deschampsia flexuosa (L.) Trin. (Poaceae). Functional Ecol. 3(5):561-567. Heins, R.D., A.C. Cameron, W.H. Carlson, E. Runkle, C. Whitman, M. Yuan, C. Hamaker, B. Engle, and P. Koreman. 1997. Controlled flowering of herbaceous perennials. p. 15-31. In: E. Goto et al. (eds.). Plant production in Closed ecosystems. Kluwer Academic Publishers, The Netherlands. Karlsson MG. and RD. Heins. 1992. Chrysanthemum dry matter partitioning patterns along irradiance and temperature gradients. Can. J. Plant Sci. 72:307-316. Karlsson, M.G., R.D. Heins, J.E. Enivin, R.D. Berghage, W.H. Carlson, and J.A. Biernbaum. 1989a. lrradiance and temperature effects on time of development and flower size in Chrysanthemum. Scientia Hort. 39:257-267. Karlsson, M.G., RD. l-leins, J.E. EnIvin, R.D. Berghage, W.H. Carlson, and J.A. Biernbaum. 1989b. Temperature and photosynthetic photon flux influence Chrysanthemum shoot development and flower initiation under short-day conditions. J. Amer. Soc. Hort. Sci. 114(1):158-163. Korczynski, P.C., J. Logan, and J.E. Faust. 2002. Mapping monthly distribution of daily light integrals across the contiguous United States. HortTech. 12(1):12-16. Niu, G., R.D. Heins, A.C. Cameron, and W.H. Carlson. 2001. Day and night temperatures, daily light integral, and 002 enrichment affect growth and flower development of Campanula carpatica ‘Blue Clips’. Scientia Hort. 87:93-105. Pearson, S., P. Hadley, and A.E. Wheldon. 1993. A reanalysis of the effects of temperature and irradiance on time to flowering in Chrysanthemum (Dendranthema grandiflora). J. Hort. Sci. 68(1):89—97. 60 Warner, RM. and J.E. Erwin. 2003. Effect of photoperiod and daily light integral on flowering of five Hibiscus Sp. Scientia Hort. 97:341-351. Warrington, l.J. and RA. Norton. 1991. An evaluation of plant growth and development under various daily quantum integrals. J. Amer. Soc. Hort. Sc. 1 16(3):544-551. White, J.W. and l.J. Warrington. 1988. Temperature and light integral effects on growth and flowering of hybrid geraniums. J. Amer. Soc. Hort. Sci. 1 13(3):354-359. 61 Table 1. Start date, flower percent, and average DLI and air temperature from forcing to anthesis for Achillea millefolium ‘Red Velvet’, Gaura Iindheimeri ‘Siskiyou Pink, and Lavandula angustifolia ‘Hidcote Blue’ grown under DLI treatments: 50% shading of ambient light plus PPF of 100 umol-m'°-s"(L1); ambient light plus PPF of 20 jimol-m’z-s'1 (L2); ambient light plus PPF of 100 pmol-m'Z-s" (L3); and ambient light plus PPF of 150 umol-m' s" (L4). 62 Light Start Date Flower Avg. DLI Avg. Air Temp Species Treament % (moltm'z-d") (°C) Achillea L1 1129/01 90 4.9 22.0 2119/01 100 5.6 21.7 ‘Red Velvet” 3105101 100 6.2 22.9 3/27/01 100 6.5 23.6 L2 1129/01 100 8.5 21.2 2119/01 100 10.2 21.7 3105101 100 11.4 22.1 3127/01 100 12.7 22.8 L3 1129/01 100 15.2 21.8 211 9101 100 17.1 22.1 3/05/01 100 18.1 22.5 3127/01 100 18.3 23.1 L4 1129/01 100 16.4 22.9 211 9101 100 18.5 23.0 3105101 100 19.7 23.3 3127/01 100 20.3 23.7 Gaura L1 211 9101 100 5.6 22.1 3105101 100 5.6 21.8 ‘Siskiyou Pink’ 3120/01 100 6.5 23.1 L2 211 9101 100 9.7 21.4 3105101 100 10.1 21.8 3120/01 100 11.5 22.2 L3 211 9101 100 17.5 22.0 3105101 100 17.6 22.1 3120/01 100 17.5 22.4 L4 2119/01 100 18.4 23.0 3105101 100 19.1 23.2 3120/01 100 19.5 23.0 Lavandula L1 3/05/01 80 5.8 22.7 . , 4102/01 60 6.6 23.0 H'd°°‘° 3'“9' 5102/01 100 6.1 24.3 L2 3105/01 89 10.3 21.9 4102/01 75 12.7 23.0 5102/01 100 12.6 23.4 L3 3/05/01 100 17.7 22.3 4102/01 100 18 23.2 5102/01 100 15.8 22.3 L4 3/05/01 100 19.2 23.2 4102/01 78 20.2 23.8 _ 5/02/01 100 18.6 23.1 63 .moEoEEG .moEoEEGm .moEoEEGm ficEmEEGc .mOEcEEGm Econ. >96 >96 >9m c896 :35 5.6.... omm__6n_ .o:.m 986.1. 82896 829566 99066:. 3.695 5.695 9.555 .993 9.69.3626 Earn: 29.3.: EmE:-.Ecm £9.63-.an 99696 :66... 926:9.3 mun mmv ouv 9v o v $020893... .993 oocmom96=£ 06A 0? mun own omA Lea 996E 99669: 99.960... 5.695 952 952 6:26:96 .993 mA mA mA 9N N; h3E2: 89w 251 33635. EBB 29% 29% £95: 99.85 Be: .6Eoecs. 95.6 M 629566 99308 6.695 5.695 952 6:30:90. .993 So own :6 2A Eu 9v Eo o..v 982m .33 .6 290.... m m m; m-.. 932 985 .33 69 aco6 69 6806 69 6066 69 S 5.56 xca 6.8 53°F. ..o>_c> 6cm. £93: £95: Earn: Em..a:-.Ecm 9969a Em... E:...99...E mm...Eo< m a m u a 93.92 8.88 2.5. I u m 0:39. fictmEDZ 5.025930. 61.0.3 $5636 cm...nmmoahfimfimmmfiuewflwcma .oam 0606.1?9fiflcmsmmuycem uwccmm :. c9093“. :6 . mt... Em: >=mb . m 0:39. can more. . . 8.26.6. 2.9m . comma co. 28 . .o 9.85 Be 0 m . 3.22. 6 seem 826.9 9.8 ...e>_e> 8m. llrlllllll Ill I I I I l I l l I l 10 11 12 13 14 15 16 17 18 19 Week of Year Figure 1 Levels for each lighting treatment from 29 Jan to 4 June 2001. The light treatments were: 50% shading of ambient light plus PPF of 100 pmol m 2-s‘ (L;1) ambient light plus PPF of 20 umol m 2-’s (L2); ambient light plus PPF of limol-'m2 9 (L3); and ambient light plus PPF of 150 pmol m' 2-‘S 1(l..4) 65 Figure 2. Quality ratings for Achillea (A), Gaura (B) and Lavandula (C) as described in Table 2 after 8, 6, and 7 weeks of forcing, respectively. Numerical ratings were determined following the first forcing period for each species and were used to assess quality in subsequent forcing periods. 66 Inflorescence Number Quality Rating Dry Mass (9 plant‘) Stem Length (cm) #0! 0) °C)—INC» N O .3 O Achillea Gaura Lavandula i ' ‘y= 0.147x + 0.296 y= 0.18x +0431 i i , =o.90 p<0.0001 i 8:077 p<0.0002 I .L H y= 0.136x +1.33 r2=0.62 p<0.0003 A B c 1 y= 0.829X + 2.62 “0.1013 =0.64 p<0.0006 I i y=0.722x+9.386 E i E i} =0.67 p<0.0001 D‘ E F ‘ y= 0.958x + 2.034 ‘y= 0.148x + 0.200 y= 0.28): + 1.465 , 3:032 p<0.0001 i . =0.56 p<0.0044 r2=0.51 p<0.0092 M”. “Ff/Eh I y= —0.320x + 21.935 y= -0.16x + 10.50 y= -0.36x + 17.387 ITO-5° {2500? .r’=0.44 p<0.0175 r2=0.59 p<0.0036 r [it i’ 0 0 0‘3? . _ . u t:- t. a, a, , é - W 4 [S 3 0 £323] J M K ML 0 5 10 15 20 0 5 10 15 20 0 5 10 15 20 Average DLI (mol m’zd") Figure 3. Quality ratings, lateral inflorescence number at anthesis, dry mass accumulation and stem length measured at visible bud (shaded squares) and anthesis (open squares) for Achillea ‘Red Velvet‘, Gaura 'Siskiyou Pink‘ and Lavandula ‘Hidcote Blue‘ as a function of daily light integral. Regression lines are shown for significant correlations only. Error bars represent 95% confidence intervals. Quality ratings and dry maSS were determined for Achillea, Gaura and Lavandula after 8, 6, and 7 weeks of forcing, respectively. 67 60- 40* 20‘ Days from Forcing to \fisrble Bud Achillea 'Red Velvet' m, j y= -O.599x + 37.538 A 12:0.584 p<0.0006 Gaura 'Siskiyou Pink' M y= -0.292x + 21.944 =0.85 p<0.0001 Lavandula 'Hidcote Blue' +3.3 p<0. 3048 {his Flo—Q 04 g 601 m 20) .26 £3 40- £5 £2 a 204 >5 G o 0. m 601 .s g2 0) “-2 40« g: ..~;< '09. : 20< o 0 Figure 4. Time from forcing to visible bud, from forcing to anthesis, and from visible bud to anthesis for Achillea millefolium 'Red Velvet', Gaura Iindheimeri 'Siskiyou Pink' and Lavandula angustifolia 'Hidcote Blue' grown under varying Average DLl (mol m‘zd’1) _ 4,22“ + 20932 y= -0.102x + 27037 p<0 2002 020.74 “00004 11:0.37 p<0.0366 .o o ' I M g i . . 3 , y= -0 34x + 55 48 y= -0.517x + 42.92 ‘ r2=0 28 p<0 0357 . G r2=0.86 p<0.0001 “05707 I 0 5 10 15 20 0 5 10 15 20 0 5 10 15 20 average daily light integrals. Regression lines are shown for significant correlations only. Error bars represent 95% confidence intervals. 68 CHAPTER III: CHARACTERIZING THE VERNALIZATION RESPONSE OF ACHILLEA ‘MOONSHINE’ 69 Evidence of a facultative vernalization response in Achillea ‘Moonshine’ Beth A. Fausey and Arthur C. Cameron Department of Horticulture, Michigan State University, East Lansing, MI 48824- 1325 Abstract Growth and flowering responses of Achillea clypeolata Sm. x A. ‘Taygetea’ ‘Moonshine’ rooted cuttings were monitored in a 20°C greenhouse with a 16-h long day (LD) photoperiod following a broad range of temperature treatment (-2.5 to 20°C) for varying durations (0 to 8 weeks). All plants flowered, except two that died at -2.5°C, when held at -—2.5 to 20°C for 0, 2, 4, 6, or 8 weeks. Leaf unfolding rate increased linearly from 0 leaves per day at —2.5°C to 0.25 leaves per day at 20°C. Plants held at -2.5 to 10°C formed a similar total number of leaves beneath the inflorescence although time to visible bud was hastened by up to 15 d following 6 and 8 weeks at 5 to 125°C. Total leaf number was elevated at warmer temperatures and was associated with slight delay in time to visible bud. Plants generally formed 2 inflorescences, yet inflorescence number doubled following 6 or 8 weeks at 5 and 75°C. While ‘Moonshine’ cuttings collected from vegetative stock plants grown under extended short days (SD) do not have an obligate vernalization requirement for flowering, vernalization at 5 to 125°C hastened flowering and in some cases increased inflorescence number. 70 Plants can be held at a wide range of temperatures for up to 8 weeks before transfer to inductive LD; however, temperature and photoperiod influence developmental rate and subsequent time to flower. It is recommended that rooted cuttings be transferred from propagation and bulking to inductive LD for consistent, predictable and efficient production of Achillea ‘Moonshine’. Introduction Achillea ‘Moonshine’ is a popular garden perennial that is widely produced in North America. Achillea ‘Moonshine’ does not flower freely in the garden, and the first and only flush of flowers results from the terminal and then lateral inflorescences that develop in early summer. Once flowering has ended, plants remain vegetative until the following year even if spent inflorescences are removed. In contrast, other perennial yarrows such as ‘Anthea’, ‘Coronation Gold’ and some A. millefolium cultivars continue to flower throughout the summer provided spent inflorescences are removed (DiSabato-Aust, 1998). Nauseida et al. (2000) reported that ‘Moonshine’ is an obligate long day plant with a critical photoperiod of approximately 14 hours; plants remain largely vegetative under SD (S13 h) and flower best under LD (216 h). Further investigations by Enfield et al. (2002) suggested that newly rooted ‘Moonshine’ cuttings collected from stock plants under SD do not benefit from exposure to low temperatures at 5°C and can be programmed to reach visible bud in approximately 3 weeks and to flower in 6 to 7 weeks following transfer to LD at 20°C. Other studies with related A. millefolium cultivars also reported no cold 71 requirement for flowering and quantitative LD photoperiodic responses (Fausey et al., 2005; Zhang et al., 1996). Despite these reports, complex dormancy and flower induction processes can influence production of Achillea ‘Moonshine’. We have observed sporadic flowering of ‘Moonshine’ under SD and failure of plants to flower after transfer to inductive LD. Enfield (2002) observed that vegetative stock plants grown under a prolonged 9-h photoperiod for 15 weeks failed to flower within 9 weeks following transfer to fully inductive LD conditions and suggested that prolonged SD (215 weeks) induced an obligate cold requirement for flowering in these plants. However, we observed that these plants eventually flowered after 24 weeks under LD without a cold treatment. Although vernalization is not required by many perennial species for flowering, low temperatures often promote flowering characteristics such as reduced time to flower and enhanced flower number. For example, Agastache ‘Blue Fortune’ and Rudbeckia fulgida Ait. ‘Goldsturrn’ flower more quickly with fewer nodes following exposure to 5°C compare to noncooled plants (Fausey et al., 2002; Runkle et al., 1999). In this study, we investigated whether Achillea ‘Moonshine’ cuttings collected from vegetative stock plants grown under extended SD (227 weeks) would require vernalization. We also set out to determine which, if any, temperatures promote growth and flowering characteristics. Growth and developmental responses of cuttings were monitored and characterized in response to a broad range of temperature treatments (-2.5 to 20°C) for varying durations (0 to 8 weeks). 72 Materials and Methods Stock plant and cutting management. Stock plants were developed from plants first grown in a 20 :l: 2°C glass greenhouse under extended 9-h SD (227 weeks). Stock plants were transplanted into 7.6 L containers on 08 Nov. 2002, and vegetative cuttings with 4 to 5 nodes were collected on 20 Dec. 2002 and 22 Feb. 2003. Cuttings were dipped into 1500 ppm indole-3—butyric acid and 750 ppm naphthalene acetic acid rooting hormone (Dip ‘N Grow, Clackamas, Ore.) and placed in 72-cell plug trays (50-mL cell volume) filled with a commercial peat-perlite media (Sure-Mix, Michigan Grower Products, Galesburg, Mich.). Cuttings were rooted for 3 weeks under 9-h photoperiod in a glass propagation house with air and media temperatures of 23 and 26°C, respectively. Once rooted, plants were grown in the plug tray for an additional 3 weeks in a 20°C growth chamber with an 11-h photoperiod receiving a photosynthetic photon flux (PPF) of 150 pmol'm‘z's'1 (~6 mol'm'z'd") before transfer to temperature treatments. Treatments and plant culture. Plugs were held at temperatures from -2.5 to 20°C at 25°C increments for 0, 2, 4, 6, and 8 weeks. Plants were illuminated with a PPF of 25 (-2.5 to 25°C) or 100 umol'm'z's'1(5 to 20°C) provided by fluorescent and incandescent lamps for 11 h'd'1 providing ~1 and 4 mol'm'z'd“, respectively. All plants entered treatments on the same date and were removed after 2, 4, 6 or 8 weeks. 73 Following treatments, plugs were transplanted to 13-cm (1 .1-L) square containers filled with the above media and grown in a 20: 2°C glass greenhouse with a 16-h LD provided by 400-watt high pressure sodium (HPS) lamps from 0600 to 0800 HR and 1700 to 2200 HR. Instantaneous light in each treatment was measured at plant height with Ll-COR line quantum sensors (Ll-COR, Lincoln, Nebr.) connected to a CR10 datalogger (Campbell Scientific, Logan, Utah). Greenhouse air temperatures were controlled by a climate-control computer (Priva, Model CD750, De Lier, The Netherlands) and were monitored on each bench with 36-gauge (0.127-mm diameter) type-E thermocouples connected to a CR10 datalogger. Temperature and light measurements were collected every 10 s, and the hourly average was recorded. The average daily temperature and light integral from forcing to visible bud was calculated for run 1 and 2 (Table 1). All plants were irrigated with well water (EC of 0.70 mS'cm'1 and 105 mg Ca, 35 mg Mg, and 85 mg S'L") acidified with H2804 to a titratabie alkalinity of 130 mg CaCOgL‘1 and water soluble fertilizer providing 125N—12P—125K—1 3Ca mg'L'1 (30% ammoniacal N) plus 1.0Fe—0.5Mn—0.52n—0.5Cu—0.1B—0.1Mo mg'L' ‘ (MSU Special; Greencare Fertilizers, Chicago, III). Data collection and analysis. The number of unfolded leaves was recorded prior to and following temperature treatments. A leaf was considered unfolded and scored when it was held at 45° angle from vertical. The number of leaves that expanded during temperature treatments was used to calculate rate of leaf 74 formation at all temperatures. Additional leaves formed in the greenhouse below the inflorescence, date of first visible flower bud and inflorescence number at first open flower were recorded for all plants. The reciprocal of time to visible bud was calculated and used to determine rate to visible bud. This experiment was a completely randomized design replicated in time with eight observational units in each treatment. Run 1 began on 24 Jan 03 and run 2 began on 29 Mar 03. Plant responses (leaf number, time to flower) did not vary significantly (p<0.05) between replications and were pooled for analysis. Data were analyzed by ANOVA using PROC GLM and linear regression analysis was performed on treatment means using PROC REG procedures in SAS version 8.0 (SAS Institute, Cary, N.C.). Error bars represent 95% confidence intervals. Results and Discussion All surviving plants flowered following treatment at -2.5 to 20°C for 2, 4, 6, or 8 weeks, and control plants not exposed to temperature treatments readily resumed flowering when grown under LD. The only exceptions were two plants that died at —2.5°C, one after 6 and one after 8 weeks. Rate to visible flower bud was slightly delayed following treatment at - 2.5°C (Fig. 1A, B). Rate to visible bud was generally similar to control plants at all temperatures following 2 weeks treatment, and increased following greater durations of treatment at 5 to 125°C. Flowering reached a minimum after 6 and 8 weeks at 7.5 to 12.5°C, and the variability associated with rate of progress to 75 visible bud increased following these treatments. Treatment at 0, 2.5, 15, 17.5, and 20°C had little effect on rate to visible bud compared to the control. Control plants grown under LD and plugs grown at 11-h photoperiods with relatively low light for up to 8 weeks formed visible flower buds in 25 to 32 d (Fig. 1A). In two comparable studies, ‘Moonshine’ rooted cuttings grown for up to 6 weeks under 9, 10, and 12-h photoperiods formed visible flower buds in 25 to 36 d with a similar number of leaves after transfer to LD (Enfield, 2002; unpublished data). Modest leaf development occurred when plugs were held at -2.5 or 0°C, and the number of leaves that unfolded increased from 0 to 20°C and with increasing length of temperature treatment (Fig. 2A). When expressed as a developmental rate, leaf unfolding increased linearly from 0 leaves per day at - 25°C to 0.25 leaves per day at 20°C (r2=0.90) (Fig. 3). Control plants formed ~11 new leaves after transfer to LD, while 8 to 12 leaves unfolded in the greenhouse below inflorescences following 2 and 4 weeks at all treatment temperatures (Fig. 28). The fewest leaves formed following 6 and 8 weeks at 10°C (Fig. 28). The number of leaves that unfolded after transplant positively correlated with time from transplant to visible flower bud (r2=0.64) (Fig. 4). Each additional leaf that formed and unfolded in the greenhouse below the visible bud took approximately 3 days to do so. Despite the apparent promotive effects of low temperature on rate of progress to visible bud and minimum leaf number unfolded after transplant, the total number of leaves formed below each inflorescence was similar following treatment at -2.5 to 10°C (Fig. 2C). Total leaf number increased when plants 76 were held at temperatures 212.5°C, providing evidence that flowering was ultimately delayed in response to these temperature treatments. Total leaf number increased from 12 to 15 leaves following 2 weeks to 18 to 22 leaves following 6 and 8 weeks at 17.5 and 20°C (Fig. 2C). Similar leaf numbers were reported by Enfield (2002) for plants bulked under 10, 12, and 13-h photoperiods for 2, 4, or 6 weeks. Increasing the duration of bulking under these photoperiods resulted in a similar increase in time to flower in both studies (Fig. 4A). The environmental conditions required to produce multiple flowering stems on Achillea have not been well-defined. Nausedia et al. (2000) reported that lateral branching of yarrow often improved following cold treatment perhaps due to increased light levels during forcing, as a direct effect of cold, or from a reduction in apical dominance. Enfield (2002) suggested that temperatures below 20°C during the bulking period prior to inductive LD conditions may increase shoot number without pinching. In this study, inflorescence number was variable, and plants generally averaged one to two inflorescences following temperature treatments (Fig. 1C). Inflorescence number doubled following treatment at 5 and 75°C for 6 and 8 weeks compared to lesser durations, although this cannot be clearly explained. Inflorescence number also increased following 8 weeks at warmer temperatures perhaps due to larger plant size following these treatments. The hastening effect of temperature treatments on flowering after transplant may have resulted from flower initiation and partial development during vernalization. In previous studies, Achillea ‘Moonshine’ cuttings were held 77 at 5°C with a 9-h photoperiod providing 10 u'mol'm‘z's'1 (~0.3 mol'm‘z'd'1) (Enfield, 2002; Nausieda, 2000) while in the present research, we utilized an 11- h photoperiod and elevated light levels (~4.9 mol°m'2'd'1), conditions that may have promoted flower induction and development. Microscopic flower initials were observed on plugs cooled at 75°C for 8 weeks in a separate study, although no other temperature treatments were evaluated (data not shown). While rate to visible bud failed to hasten and total leaf number increased following treatment at temperatures 215°C, the delay in development was greatest at 20°C. Time to visible bud was delayed by ~5 d for plants held at 20°C with an 11-h photoperiod and limited light compared to control plants under LD and elevated light levels (Fig. 1A). Leaf number formed after transplant was similar for these two treatments providing evidence that plants held at warmer temperatures failed to initiate flowers during the temperature treatments (Fig. 28). Plants held at warmer temperature treatments likely had greater carbohydrate demands than were met under the growth chamber conditions causing slight flowering delays when transferred to the greenhouse. Flowering of Achillea millefolium ‘Red Velvet’ was delayed by ~9 d at 22°C when grown with an average daily light integral of 5 compared to 10 mol-m‘Z-d“, similar light levels encountered in this study (Fausey et al., 2005). Rooted vegetative cuttings taken from SD stock plants readily flowered in this study while stock plants held under SD for 215 weeks remained completely vegetative for up to 24 weeks following transfer to inductive LD conditions (Fig. 5). Why do ‘Moonshine’ stock plants not flower following transfer to inductive 78 LD? Extended exposure to SD appears to render stock plants less receptive to inductive LD for flowering, while removal of vegetative cuttings from these plants enables growth and rapid flowering. Cutting removal and rooting appears to eliminate inhibitory factors present elsewhere in the stock plant without requiring exposure to low temperatures for flowering. Although the origination and mechanism of action for this flowering block remain unclear, all stock plants remained vegetative under SD throughout the course of the experiment. Thus, actively growing rooted cuttings of Achillea ‘Moonshine’ do not have an obligate vernalization requirement for flowering under LD and should be used in production. ‘Moonshine’ plugs can be held at a wide range of temperatures for up to 8 weeks and likely longer, before transplanting and forcing to flower. Plant survival and performance were excellent in this study even though plugs were not preconditioned to low temperatures before treatment. Treatment at -2.5°C resulted in more variable leaf development and delayed flowering compared to other treatments (Figs. 1A, 28, C). Engle (1994) observed considerable variability in plant survival and subsequent regrowth of A. ‘Cloth of Gold’ when 128-cell plugs were stored at -2.5 or 0°C without pretreatment at 0 or 5°C for several weeks. It is likely that ‘Moonshine’ would also benefit from hardening plugs at 0 or 5°C for several weeks before exposure to lower temperatures, particularly when using small starting material. Vernalization of ‘Moonshine’ plugs at 5 to 125°C for 4 to 8 weeks promotes flowering (hastened rate to visible bud, inflorescence number), albeit 79 slightly, compared to noncooled plants. Therefore it is important to note that pre- forcing treatment conditions (temperature, duration, and photoperiod) may influence developmental rate and subsequent time to flower. For example, growers who receive plugs held at 5 to 125°C with an 11-h photoperiod for greater than 4 weeks will likely experience accelerated flowering by up to 15 d, while flowering may be delayed by a similar amount if held at cooler temperatures. In this study, high temperatures (15-20°C) coupled with 11-h photoperiod and low light levels delayed flowering by up to 7 days, depending on duration. immediately transferring plugs from propagation and bulking to inductive LD for flowering is the recommended protocol for consistent, predictable and efficient production of Achillea ‘Moonshine’. 80 Literature Cited DiSabato-Aust, T. 1998. The Well-Tended Perennial Garden. Timber Press, Portland, OR. Enfield, A. 2002. Flower induction and cultural requirements for quick-cropping of the herbaceous perennials Veronica spicata, Phlox paniculata, Leucanthemum xsuperbum, Achillea, Gaura Iindheimeri, and Campanula. Master’s Thesis. Michigan State University, East Lansing, MI. Enfield, A., E. Runkle, R. Heins, A. Cameron, and W. Carlson. 2002. Herbaceous Perennials: Quick-cropping Part II. Greenhouse Grower: 20(3):40-48. Engle, BE. 1994. Use of light and temperature for hardening of herbaceous perennial plugs prior to storage at -2.5°C. Master’s Thesis. Michigan State University, East Lansing, MI. Fausey, 8., E. Runkle, R. Heins, A.Cameron, and W. Carlson. 2002. Herbaceous Perennials: Agastache. Greenhouse Grower 20(9):74-82. Fausey, 8.A, R.D. Heins, and AC. Cameron. 2005. Daily light integral affects flowering and quality of greenhouse-grown Achillea, Gaura and Lavandula. HortScience 40(1):114-118. Nauseida, E, L. Smith, T. Hayashi, 8. Fausey, A. Cameron, R. Heins and W. Carlson. 2000. Forcing perennials: Achillea. Greenhouse Grower 18(5):53-64. Runkle, E.S., R.D. Heins, A.C. Cameron, and W.H. Carlson. 1999. Photoperiod and cold treatment regulate flowering of Rudbeckia fulgida ‘Goldsturrn’. HortScience 34(1):55-58 Zhang, 0., AM. Arrnitage, J.M. Affolter, and MA. Dirr. 1996. Environmental control of flowering and growth of Achillea millefolium L. ‘Summer Pastels’. HortScience 31 (3):364-365. 81 Table 1. Average daily temperature and light integral values for 5-week forcing period in the greenhouse following Achillea ‘Moonshine’ temperature treatments for Run 1 and 2. Weeks Average Daily Average Daily of Transplant Date Temperature Light Integral Treatment (°C) (mol'm'z'd'1) Run 1 Run 2 Run 1 Run 2 Run 1 Run 2 0 01/24/03 03/29/03 20.3 21.5 8.0 10.9 2 02/07/03 04/14/03 20.4 22.0 10.0 10.2 4 02/20/03 04/28I03 20.7 21.7 10.7 9.8 6 03/06l03 05l12/03 21.1 21.9 11.3 10.7 8 03/24/03 05/23/03 21.4 22.5 11.1 1 1.5 82 50 401 30 .4. ,., ,, .. -.... .,, ...... . _ u - I 20- Days to visible bud 0.10 0.08 1 0.06 , 0.041.- Rate to visible bud (1Id) 0.00 Inflorescence number o 1 1 1 1 1 1 1 4 -2.5 0.0 2.5 5.0 7.5 10.012.515.017.5 20.0 Temperature (°C) Figure 1. Days to visible bud (A), rate of progress to visible bud (B), and inflorescence number (C) for Achillea ’Moonshine' plants following 2, 4, 6, and 8 weeks at -2.5 to 20°C. Error bars represent 95% confidence intervals. Control plants reached visible bud in 25.5 days with a rate to visible bud of 0.039 and formed an average of 1.8 inflorescences (solid lines in A, 8, and C, respectively). 83 Leaves during treatment Leaves after transplant Total leaves unfolded 0 i 1 l t 1r i + i i i -2.5 0.0 2.5 5.0 7.5 10.012.515.017.5 20.0 Temperature (°C) Figure 2. Number of leaves unfolded during (A) and after (8) treatment and their sum (C) for Achillea 'Moonshine' plants following 2, 4, 6, and 8 weeks of storage at -2.5 to 20°C. Error bars represent 95% confidence intervals. Control plants unfolded 10911.3 leaves after transfer to the greenhouse (solid line, B and C). 84 0.30 - 1 «q y=0.018695+ 0.01148x O r ’=0.90 0.25 '1 0.20 4 Number of leaves unfolded per day .0 a? I l I I I I I -2.5 0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 20.0 Temperature (°C) Figure 3. The rate of leaf unfolding for Achillea 'Moonshine' plants held at -2.5 to 20°C for 2, 4, 6, and 8 weeks. Data from two replications were pooled for analysis. 85 50 O 2weeks E] 4weeks 40 d A 6weeks <> 0 8weeks "O 3 .Q 9 30 1 :9 .l’ > .9. w 20 - >5 (0 O 10 -1 y=-6.2988+ 3.2281x r’=0.64 0 l I I T 6 8 10 12 Leaf number following temperature treatments Figure 4. Correlation between the number of leaves unfolded following 2, 4, 6, or 8. weeks treatment at -2.5 to 20°C and the number of days required to reach Vlsible flower bud. Greenhouse temperatures were 20:1:2°C. 86 : l 1 l 1 r r i 1 l 1 Stock . . . SD >15 weeks I ‘ Tra'nsfer to LD 2 I . I 1 ' \ ; .1 ; LD ~12 weeks —;) LD ~24 weeks (no flowering) j ‘ (sparse flowering) Take cuttings . . and propagate . Z . . , l . . 1 under SD , j , I T i I I l . I No cold Transfer to L0 : Rooted plug SD lln flower} : {Sweeksi §6weeksi ' 0 2 3 4 5 6 ”5101214161620 22 24 Time (weeks) Figure 5. The flowering physiology of Achillea ‘Moonshine’ stock plants and cuttings. Stock plants grown under a short day photoperiod for >15 weeks and transferred to an inductive long day photoperiod flower in approximately 24 weeks with few inflorescences per plant. Vegetative cuttings taken from these stock plants grown under short days, if rooted and grown under inductive long day photoperiods will flower in 6 weeks at 20°C. 87 CHAPTER IV: DIFFERING VERNALIZATION RESPONSES OF VERONICA SPICA TA AND ISOTOMA AXILLARIS 88 Differing vernalization responses of Veronica spicata and Isotoma axillaris Beth A. Fausey and Arthur C. Cameron Department of Horticulture, Michigan State University, East Lansing, MI 48824- 1325, phone: 517-355-5191 ext 1451 ; fax:517-355-0249 Abstract Many herbaceous perennials are known to have a cold-requirement for flowering, yet we know little about their specific temperature responses and flower induction mechanisms. To determine the range of vernalization temperatures effective for flower induction of Veronica spicata L. ‘Red Fox’ and Isotoma axillaris Lindl., clonally propagated plants were held at -2.5 to 20°C for 0, 2, 4, 6, or 8 weeks or 0, 2.5, 5, 7.5, 10, 12.5, and 15 weeks, respectively, and growth and flowering were monitored under common greenhouse conditions (constant 20°C with an average daily light integral of ~5 mol-m'z-d"). All Veronica plants survived the transfer from 20 to -2.5°C remarkably well while Isotoma plants failed to survive prolonged exposure to temperatures less than 2.5°C. Both Veronica and Isotoma exhibited obligate vernalization requirements for flowering, yet node number, time to flower, and flowering characteristics following vernalization treatments differed with species. A minimum of 4 weeks at -2.5 and 0°C, 6 weeks at 25°C, and 8 weeks at 5 and 75°C was required for 89 complete (100%) flowering of Veronica while a minimum of 5 weeks at 5 to 10°C, 7.5 weeks at 12.5°C, and 10 weeks at 25°C were required for complete flowering of Isotoma. Shorter durations at each temperature resulted in lower flowering percentages. Maximal temperatures for vernalization of Veronica and Isotoma were 10 and 12.5°C, respectively. The rate of node formation during vernalization exhibited a curvilinear response in Veronica and a linear response in Isotoma to increasing temperature from 0 to 20°C. In Veronica, node number and flower timing were relatively fixed following up to 8 weeks of cold at each temperature while node number and time to flower of Isotoma generally decreased at each temperature with extended exposure. The greatest flowering response of Veronica was observed following all vernalization durations at -2.5°C based on the percentage of lateral nodes flowering. Conversely, based on percent flowering, temperatures of 5 to 10°C were equally effective in vernalizing Isotoma plants following 5 or more weeks, and vernalization at 125°C was equally effective following 10 or more weeks. Flower initiation in Veronica occurred uniformly following all treatments with complete flowering in 8 weeks while flower initiation and development in Isotoma likely occurred during temperature treatments following a minimum of 5 weeks at 5 to 10°. Intrinsic differences in growth or physiology under the same low temperature conditions may contribute to enhanced vernalization in one species (Veronica) and limited survival in another (Isotoma). These results further demonstrate the complexity of growth strategies and flower induction mechanisms in diverse species. 90 Introduction Many herbaceous winter annual, biennial and perennial plants require vernalization, prolonged exposure to low temperatures, for flower induction (Chouard, 1960). Vernalization has traditionally referred to low temperature treatments that promote flower initiation and development after plants have returned to warmer inductive conditions (Chouard, 1960; Thomas and Vince- Prue, 1984), though low temperatures may also directly promote flowering in some species, and flower primordia may form and develop while plants are still exposed to cold (Lang, 1965). In some cases, the inductive and direct effects of low temperatures are difficult to differentiate particularly when plants are exposed to prolonged cold treatments (Bemier et al., 1981). However, when cold- requiring species fail to receive adequate vernalization (inappropriate temperature or duration of exposure), partial or a complete lack of flowering may result. While the vernalization requirements and responses of numerous plant species have been studied, the bulk of our knowledge comes from extensive research on a relatively few plant species. In particular, numerous studies have evaluated the vernalization responses of annual and biennial crops such as winter rye, carrot, celery, sugar beet, and onion, along with the winter annual Arabidopsis thaliana Stockholm, biennials such as Hyoscyamus niger and Lunaria biennis, and the perennial Chrysanthemum mon'folium. These have been thoroughly summarized in several reviews, most of which are decades old (Chouard, 1960; Lang, 1952, Thomas and Vince-Prue, 1984). 91 Although comprehensive studies investigating broad temperature ranges and durations for therrnoinduction are largely lacking, several common characteristics of therrnoinduction have been observed and identified in vemalization-requiring species. In general, vernalization occurs over a broad range of effective temperatures from -5 to 15°C, and most species have exhibited an optimum response at 1 to 5 or 6°C (Chouard, 1960; Lang, 1965). Classical studies with biennial strains of Hyoscyamus niger illustrated the quantitative nature of vernalization, a process whose intensity at effective temperatures gradually increases with the length of exposure until saturation occurs (Lang, 1965; Vince-Prue, 1975). The effectiveness of vernalization is highly dependent on temperature following short durations of cold, yet all effective temperatures eventually promote similar flowering responses when the duration is sufficiently long (Lang, 1952). Additionally, as vernalization increases, flowering is generally accelerated through a reduction in final leaf number below the inflorescence and/or a reduction in time to flower (Lang, 1965). Genetic and molecular studies with Arabidopsis thaliana, as well as winter wheat, have provided new insight into the molecular mechanism of vernalization and the signaling pathways involved (Henderson et al., 2003; Sung and Amasino, 2004). However, very few studies have furthered our understanding of vernalization in polycarpic species. Many herbaceous perennials are known to have a cold-requirement for flowering, yet we know little about their specific temperature responses and flower induction mechanisms. An understanding of the relative effectiveness of vernalization temperatures and their duration is of 92 interest to ecologists as well as commercial horticulturalists investigating flower induction under field and controlled environment conditions. To better understand the role of vernalization in flower induction of diverse herbaceous perennial species, we developed a protocol to systematically evaluate and Characterize therrnoinduction under controlled environmental conditions. Here we focus on two species with differing vernalization responses, Veronica spicata and Isotoma axillaris. Veronica spicata (Scrophulariaceae) is an extremely cold hardy ornamental species (to -40°C) native to Eastern Europe and Northern Asia (Armitage, 1989; Griffiths, 1994). The species has Characteristic blue flowers; and horticulturally, many clonal hybrids are available including the pink-flowered ‘Red Fox’. Veronica spicata typically flowers for several weeks in early summer then remains vegetative until the following year. Enfield (2002) determined that Veronica ‘Red Fox’ had an obligate vernalization requirement. Isotoma axillaris (syn. Laurentia axillaris, Campanulaceae) is native to Queensland, Victoria, and New South Wales in eastern Australia (Bentham and Mueller, 1844). Plants naturally flower in late winter and early spring for several months and sporadically thereafter until autumn (Blomberry, 1977; Johnston, personal communication). Preliminary trials indicated that Isotoma had a near- obligate vernalization and long day requirement for flowering (Fausey et al, 2003a). The objectives of this study were to quantify and compare flower induction of Veronica spicata and Isotoma axillaris as a function of vernalization temperature and duration. 93 Materials and Methods Experiment 1. Vernalization Temperature and Duration Plant material. Vegetative Veronica and Isotoma stock plants were maintained in a 22°C growth chamber with a 13-h photoperiod (150 plmol'm'z's'1 from fluorescent and incandescent lamps from 0600 to 1900 HR). Veronica cuttings with 3 to 4 nodes were taken at 2-week intervals for 10 weeks beginning on 22 May 2003 and 22 Aug 2003. Isotoma cuttings with 3 to 4 nodes were taken at 2.5-week intervals beginning on 13 May 2003 for 17 weeks. Cuttings were dipped into 1000 ppm indole-3-butyric acid and 500 ppm naphthalene acetic acid rooting hormone (Dip ‘N Grow, Clackamas, Ore.) and placed in 72-cell plug trays (50-mL cell volume) filled with a commercial peat-perlite media (Sure-Mix, Michigan Grower Products, Galesburg, Mich). Cuttings were rooted for 2 weeks under 9-h photoperiod in a glass propagation house with air and media temperatures of 23 and 26°C, respectively. Once rooted, plants were grown for an additional 3 weeks in a 20°C growth chamber with a 13-h photoperiod receiving 150 umol'm‘z's'1 before transfer to temperature treatments. Veronica and Isotoma plants averaged 5 to 6 and 12 to 14 nodes, respectively, at the start of temperature treatments. Temperature treatments. Plants were held in plug trays and exposed to temperatures from -2.5 to 20°C at 25°C increments for 0, 2, 4, 6, or 8 weeks (Veronica) or 0, 2.5, 5, 7.5, 10, 12.5, and 15 weeks (Isotoma). During treatment, 94 plants were illuminated with 20 (-2.5 to 25°C) or 100 (5 to 20°C) umol'm'z's'1 provided by fluorescent and incandescent lamps for 11 h'd". Veronica plants entered temperature treatments at 2-week intervals, were removed and transplanted to the greenhouse on 6 Aug and 25 Nov 2003. Isotoma plants entered temperature treatments at 2.5-week intervals, were removed and transplanted in the greenhouse on 9 Sept 2003. Plant culture. All plants were irrigated as needed with reverse osmosis water and a water soluble fertilizer providing 125N, 12P, 100K, 65Ca, 12Mg mg-L’1 (12.6% nitrate nitrogen, 2% ammoniacal N) plus 1.0Fe, 0.5Mn, 0.5 Zn, 1.0Cu, 0.38, 0.1Mo mg-L’1 (MSU RO Special; Greencare Fertilizers, Chicago, Ill.). Following vernalization treatment, plants were transplanted to 13-cm (1 .1-L) square containers filled with the above media. Environmental control. Following treatment, plants were grown in a 20:1:2°C glass greenhouse with a 16-h photoperiod provided by 400-watt high pressure sodium (HPS) lamps from 0600 to 0800 HR and 1700 to 2200 HR. Instantaneous light in each treatment was measured at plant height with Ll-COR line quantum sensors (Ll-COR, Lincoln, Nebr.) connected to a CR10 datalogger (Campbell Scientific, Logan, Utah). Greenhouse air temperatures were controlled by a climate-control computer (Priva, Model CD750, De Lier, The Netherlands) and were monitored on each bench with 36-gauge (0.127-mm diameter) type-E thermocouples connected to a CR10 datalogger. Temperature 95 and light measurements were collected every 10 s, and the hourly average was recorded. Average daily temperature and light integrals were calculated for each species from date of transplant to average date of first open flower and were 208°C and 4.5 mol-m’z’d‘1 for Veronica and 209°C with 5.7 mol-m'z'd" for Isotoma. Experiment 2. Veronica Short Cold To further determine the minimum vernalization requirement for flowering of Veronica, plants were propagated at 4-d intervals as described in Experiment 1 and held at -2.5, 0 or 25°C for 0, 12, 16, 20, 24, 28, or 32 days. Plants were then transplanted and grown in the greenhouse under similar light and temperature conditions as previously described. Data collection and analysis. Node number prior to treatment, at transplant, and following transplant below the first visible flower bud were collected for all plants. For Veronica, the date of first visible inflorescence, the date of first open flower, and the number of reproductive and vegetative lateral shoots at each node were collected for all flowering plants at first open flower. For Isotoma, the date of first visible flower bud in the leaf axil (buds 0.2 mm in length), macroscopic visible bud (1 mm in length) and first open flower were collected for all flowering plants. Flower bud number at first open flower was also recorded for Isotoma. 96 Each experiment was a completely randomized design and was repeated in two runs over time. For Experiment 1, results from the second run for each species will be used in analyses and comparisons as greenhouse temperature and light levels were similar following transplant. Data were analyzed using PROC GLM general linear model procedures, and linear regression analysis was performed on treatment means using PROC REG procedures in SAS version 8.0 (SAS Institute, Cary, NC). Linear regression lines are presented only when the correlation was statistically significant (p<0.05). Error bars represent 95% confidence intervals. Results Flowering responses. To determine the range of vernalization temperatures effective for flower induction of Veronica and Isotoma, clonally propagated plants were held at -2.5 to 20°C for 0, 2, 4, 6, or 8 weeks or 0, 2.5, 5, 7.5, 10, 12.5, and 15 weeks, respectively, and growth and flowering were monitored under common greenhouse conditions (constant 20°C with an average daily light integral of ~5 moI-m'z-d"). Cuttings of both species were propagated from stock plants, rooted, and grown under strictly controlled photoperiod and temperature conditions to eliminate confounding effects of pre-vernalization growth environment on subsequent responses. Veronica and Isotoma exhibited obligate vernalization requirements for flowering. A minimum of 4 weeks at -2.5 and 0°C, 6 weeks at 25°C, and 8 weeks at 5 and 75°C was required for complete (100%) flowering of Veronica 97 (Fig. 18, F). In a subsequent study, the vernalization requirement for 90 to 100% flowering was a minimum of 20 days at -2.5°C, 28 d at 0°C, and 32 d at 25°C (Fig. 2A). A minimum of 5 weeks at 5 to 10°C, 7.5 weeks at 12.5°C, and 10 weeks at 25°C were required for complete flowering of Isotoma (Fig. 1F). Shorter durations at each temperature resulted in lower flowering percentages. Marginal flowering (30-50%) of Veronica occurred after 8 weeks at 10°C, and no plants flowered when vernalized at or above 125°C (Fig. 18). Marginal flowering (50-60%) of Isotoma plants that survived treatment at 0 and 25°C was also observed, and flowering of these plants was generally delayed (Figs. 1E, F, 3). Isotoma did not flower when vernalized at or above 15°C. Survival and growth during vernalization. Newly rooted plants were actively growing when moved from 20°C post-rooting conditions to vernalization treatments, and resulting plant survival at the low temperatures differed between species (Fig. 1A, E). All Veronica plants survived treatment at -2.5 to 20°C for up to 8 weeks while Isotoma plants did not survive 2.5 weeks at -2.5°C and had variable survival at 0 and 25°C (Figure 1E). Treatments resulting in death from freezing injury were removed from further analysis. For both species, the rate of node formation during vernalization was largely constant. In Veronica, the rate of node formation exhibited a curvilinear response in relation to increasing temperature with essentially no growth at -2.5 and 0°C (0.01 nodes/ day) and a maximum of 0.14 nodes per day at 20°C (Fig. 3A, 8). In contrast, the rate of node formation in Isotoma increased linearly from 98 0 nodes per day at 0°C to 0.2 nodes per day at 20°C (Fig. BC, D). Veronica formed between 0 and 6 nodes following treatment at -2.5 to 20°C for up to 8 weeks compared to Isotoma which formed nearly twice as many nodes at 20°C in a similar time period (Fig. 3A, C). Vernalization effects on subsequent growth and flowering characteristics. In Veronica vernalization temperature but not duration of treatment affected the number of nodes formed after transfer to the greenhouse (p<0.0001) (Fig. 1C). Node number after transplant increased nonlinearly following vernalization treatment above -2.5°C (Fig. 16) but did not correlate with time to flower (data not shown). All treatments resulting in complete flowering of Veronica plants did so in 53 to 58 days (data not shown). In contrast, node formation after transplant in Isotoma generally decreased with increasing duration at each vernalization temperature (Fig. 1G). The fewest nodes were observed following 15 weeks at 7.5 and 10°C and were associated with the fewest days from transplant to first open flower. Time to flower in Isotoma was highly correlated with node formation following vernalization (r2=0.93) (Fig. 4). Flowering of several plants that survived 7.5 and 10 weeks vernalization at 0°C was significantly delayed although plants formed a similar number of nodes as other treatments (Fig. 4). All reproductive Veronica plants formed a single inflorescence spike with vegetative or floral buds forming in the leaf axil of each node below the spike. Flower induction of nodes below the spike occurred in a basipetal fashion, and 99 induced buds developed into secondary inflorescences. The percentage of nodes forming secondary inflorescences was effectively used to measure the intensity of the vernalization response (Fig. 1D). Sixty to 80% of nodes flowered following 8 weeks of vernalization at -2.5°C and decreased linearly with increasing temperature (Fig. 1D). For every 25°C increase in vernalization temperature above -2.5°C, the percentage of flowering nodes decreased by 7 to 10%. A similar trend was observed in a second study (Fig. 28). Isotoma has an indeterminate flowering habit, but in contrast to Veronica, when induced, Isotoma plants form solitary axillary flowers. Flower buds measuring 1 mm in length were counted when the first flower opened, and the average number of flowers per shoot was determined for each treatment (Fig. 1H). Flower number for Isotoma varied with treatment but generally increased with increasing temperature above 0°C for treatments with 100% flowering (Fig. 1H). Temperature treatments at 0 and 25°C and at 125°C following 5 and 7.5 weeks resulted in marginal vernalization, and plants formed significantly fewer flowers. Discussion Optimum temperatures for vernalization of a species are relative and depend on the duration of therrnoinduction treatments as well as the method of determination (Lang, 1965). The greatest flowering response of Veronica was observed following all vernalization durations at -2.5°C based on the percentage of lateral nodes flowering. To our knowledge, no other species examined to date 100 has exhibited an optimum vernalization temperature below 2°C (Chouard, 1960; Lang, 1965; Atherton et al., 1990; Yeh et al., 1997). Conversely, temperatures of 5 to 10°C were equally effective in vernalizing Isotoma plants following 5 or more weeks of cold based on percent flowering, and vernalization at 125°C was equally effective following 10 or more weeks. Optimum vernalization temperatures of 3 to 7°C have been described for carrot, cineraria, beet, Petkus winter rye, and Arabidopsis thaliana Stockholm while slightly higher optimum temperatures of 5 to 10°C have been reported for Hyoscyamus niger and onion (Atherton et al., 1990, Bemier et al., 1981; Brewster, 1987; Chouard, 1960; Lang, 1965; Yeh et al., 1997). Lang (1965) cited occurrences of vernalization in winter rye, Hyoscyamus, onion and Rhipsalidopsis at temperatures of 15 to 17°C, and Lange (1992) found that Lilium Iongiflorum could be vernalized at or below 15°C. Maximal temperatures for vernalization of Veronica and Isotoma were 10 and 12.5°C, respectively. Although complete flowering of Veronica did not occur at 10°C with up to 8 weeks vernalization, we expect the response to become saturated with increased durations of cold. Isotoma exhibited an obligate vernalization requirement in this study, but there is evidence that Isotoma can flower via a vemalization-independent, autonomous pathway. We observed flowering of nonvernalized plants when grown for extended durations (2150 days) under higher greenhouse and light conditions than encountered in this study (Fausey and Cameron, unpublished data). Anecdotal reports from gardeners indicate that Isotoma will flower in late 101 summer if started from seed in early spring, although the environmental history of these plants is unclear. Attempts to flower inadequately vernalized Isotoma plants (0 and 2.5 weeks at 7.5°C) in controlled environment chambers with high light quantity (9 to 26 mol-m’z-d'" resulted in minimal flowering (0 to 15%) after 12 weeks of growth at 20°C. While we do not fully understand the conditions that allow for flowering of nonvernalized plants, Isotoma plants likely proceed towards flowering at a faster rate when grown under high light and temperature conditions. Although growth at any temperature would presumably result in flowering when given enough time, vernalization at 5 to 10°C clearly promoted the most rapid, uniform flowering of Isotoma in these studies. The effectiveness of vernalization treatments has been determined empirically by several methods. The proportion of plants flowering in a given population following vernalization treatment readily represents the quantitative nature of therrnoinduction (Thomas and Vince-Prue, 1984; Lange, 1992). Other methods to evaluate induction include evaluation of floral morphology, node and flower development, and the time required to reach a particular stage of development. For example, Lang (1965) generalized that for most species, the number of nodes formed below the inflorescence after transfer to warmer inductive conditions decreased with increasing effectiveness of a vernalization treatment until an optimal level was reached. This optimal level was described by Brooking (1996) in wheat as the point of saturation where leaf number failed to further decrease with increasing durations of cold. Additionally, saturation of the vernalization response has also been determined when additional exposure 102 to cold failed to impact minimum time to flower and/or maximum flower (Brooking, 1996). These methods have been adapted and combined to create flowering indexes that determine the effectiveness of vernalization treatments in species such as Lilium and Hatiora (syn. Rhipsalidopsis)(Lange, 1992; Rohwer, 2002) Node number, time to flower, and flowering Characteristics of Veronica and Isotoma responded differently to vernalization treatments, and the treatment effectiveness was determined by differing criterion. Vernalization responses approached but did not show conclusive evidence of reaching saturation in either species. In Veronica, node number and flower timing were relatively fixed following up to 8 weeks of cold at each temperature. This suggests that flower initiation in Veronica occurred uniformly following all treatments with complete flowering in ~ 8 weeks. Because node number and flower timing were generally similar following all durations of treatment at -2.5 to 5°C, the proportion of nodes below the inflorescence spike that were induced to flower better represented treatment effectiveness with prolonged vernalization. We did not achieve 100% flowering of all nodes in this study but this would likely occur outdoors following prolonged winters and natural saturation of the cold requirement. We also expect flowering of all nodes under controlled conditions provided the vernalization duration extended beyond 8 weeks of cold, although the exact cold requirement for saturation was not determined here. In contrast to Veronica, the effectiveness of vernalization treatments in Isotoma were evaluated using commonly employed methods. Node number and 103 time to flower of Isotoma generally decreased at each temperature with extended exposure (Fig. 4). Flower initiation first occurred following a minimum of 5 weeks vernalization at 5 to 10°C. Plants held for this minimum duration unfolded ~ 9 nodes and formed visible flower buds ~ 1mm in length 22 d after transplant (Fig. 16, data not presented). In contrast, plants vernalized for an additional 10 weeks formed 4 to 6 nodes, and flower buds were visible 8 days after transplant (Fig. 1G, data not presented). Isotoma will flower under both long and short day photoperiods once the cold requirement for flowering is satisfied. Although plants were not dissected to determine whether flower primordia were present following vernalization treatments, initiation and development likely occurred following a minimum of 5 weeks of vernalization at 5 to 10°C with an 11-h photoperiod. In fact plants cooled for 15 weeks at 10°C formed a similar total number of nodes below the first flower as those cooled for 5 weeks (Figs. 1G, SC). In addition to Isotoma, vernalization promoted earlier flowering in Achillea ‘Moonshine’ through reductions in node number below the inflorescence and time to flower (Fausey and Cameron, 2005). Actively growing rooted cuttings of Achillea ‘Moonshine’ do not have an obligate vernalization requirement for flowering yet time to visible bud reached a minimum after 6 and 8 weeks at 7.5 to 125°C. Like Isotoma, Achillea also exhibited a linear rate of node development from 0 nodes per day at -2.5°C to 0.25 nodes per day at 20°C (r2=0.90), and the hastening effect of vernalization on flowering likely resulted from flower initiation and partial development during vernalization treatments. 104 We do not know whether lateral buds in Veronica need to be of a certain age or developmental state before they can be readily vernalized or how a greater proportion of lateral buds become vernalized and capable of flowering at lower temperatures. Enfield (2002) found that plants required 3 weeks of growth following removal of the apex for vegetative shoots to perceive vernalization. Veronica and Isotoma plants were actively growing and not acclimated to low temperatures prior to vernalization treatments. Veronica tolerated the transfer from 20 to —2.5°C remarkably well while Isotoma plants failed to survive prolonged exposure to temperatures less than 2.5°C. Veronica is very cold hardy and can survive outdoors to -40°C while hardiness of Isotoma has not been fully determined but is suggested to approach -10°C (Griffiths, 1994). Engel (1994) found that 20 of 24 herbaceous perennial species benefited from exposure to 0 or 5°C for several weeks prior to low temperature storage at -2.5°C, and in arabidopsis it is well known that specific genes induce cold tolerance. Ecologically, a vernalization requirement could determine a species native range. In North America, both Veronica spicata and Achillea ‘Moonshine’ routinely survive and flower following winters in hardiness zones 3 to 5 where minimum temperatures may reach —29 to -40°C (Griffiths, 1994). The growth range of both species extends southward where minimum annual temperatures on average reach -12°C. Isotoma is innately less cold hardy than Veronica and Achillea, which limits its northern range. Yet it appears that Isotoma plants perceive cool to moderate temperatures for vernalization while maintaining the ability to flower if therrnoinduction does not occur. This ability to flower without 105 vernalization is presumably advantageous for species such as Isotoma growing in regions where cold exposure may be unpredictable and limited during winter months, such as its native range in Australia. While we largely do not know how plants perceive and integrate vernalizing temperatures to signal flower induction after transfer to warmer conditions, plants may fine-tune their vernalization requirements to adapt to regional environments. Intrinsic differences in growth or physiology under the same low temperature conditions may contribute to enhanced vernalization in one species (Veronica) and limited survival in another (Isotoma). Veronica plants did not form additional nodes when vernalized at less than 5°C while growth of Isotoma readily occurred at or above 0°C. The results presented here further demonstrate the complexity of growth strategies and flower induction mechanisms in diverse species. Flower induction is more tightly controlled in Veronica than Isotoma (and Achillea). While Achillea and Isotoma respond to vernalization, they are capable of flowering without a cold treatment and display more latitude in environmental conditions that promote flowering (F ausey and Cameron, 2005). Once vernalization is complete, all three species may flower under short or long day photoperiods. Whether required or not, vernalization promoted the most rapid, uniform flowering of each species reported here. 106 Literature Cited Armitage, A. 1989. Herbaceous Perennial Plants: A Treatise on their Identification, Culture, and Garden Attributes. Varsity Press, Inc. Athens, Georgia. Bentham, G. and F. Mueller. 1844. Flora Australiensis: A Description of the Plants of the Australian Territory. 4:121-137. Reeve, London. Bemier, G., J. Kinet, and RM. Sachs. 1981. The Physiology of Flowering. Volume I: The Initiation of Flowers. CRC Press, Boca Raton, FL. Blomberry,A.M. 1977. Australian Native Plants. Pg. 275. Angus and Robertson, London. Brewster, J.L. 1987. Vernalization in the onion: a quantitative approach. In: Manipulation of Flowering, J.G. Atherton, ed. Butterworths, London. Brooking, I. R. 1996. Temperature response of vernalization in wheat: A developmental analysis. Annals of Botany 78:507-512. Enfield, A. 2002. Flower induction and cultural requirements for quick-cropping of the herbaceous perennials Veronica spicata, Phlox paniculata, Leucanthemum xsuperbum, Achillea, Gaura Iindheimeri, and Campanula. Masters Thesis. Michigan State University, East Lansing, MI. Fausey, B.A., A.C. Cameron, and RD. Heins. 2003. Temperature, photoperiod and light quantity impact flowering of Laurent/a axillaris, a new crop for Michigan growers. HortScience 38(5):721. Fausey, BA. and AC. Cameron. 2005. Vernalization is not required for but promotes flowering of Achillea ‘Moonshine’. Manuscript in preparation. Griffiths, M. 1994. Index of Garden Plants. Timber Press, Portland, OR. Henderson, I.R., C. Shindo, and C. Dean. 2003. The need for winter in the switch to flowering. Annual Rev. Genet. 37:371-392. Lang, A. 1952. Physiology of Flowering. Ann. Review of Plant Physiology 32265-306. Lang, A. 1965. Physiology of flower initiation. In: Encyclopedia of Plant Physiology (ed. W. Ruhland). Pp. 1371-1576. Springer-Verlag, Berlin. Lange, N. 1992. Modeling flower induction in Lilium/ongiflorum. Masters Thesis, Michigan State University, East Lansing, MI. 107 Napp-Zinn, K. 1969. Arabidopsis thaliana (L.) Heynh. In: The Induction of Flowering. Ed. L. T. Evans, Cornell University Press, Ithaca, NY. Rohwer, C. 2002. Flowering Physiology of Hatiora. MS. Thesis, Michigan State University, East Iansing, Ml. Sung, S. and RM. Amasino. 2004. Vernalization in Arabidopsis thaliana is mediated by the PHD finger protein VIN3. Nature 427:159-164. Thomas, B. and D. Vince-Prue. 1984. Juvenility, photoperiodism, and vernalization. In: Advanced Plant Physiology, M. B. Wilkins (ed.). Pp. 408- 439. Pitman Publishing, London. Vince-Prue, D. 1975. Vernalization. In: Photoperiodism in Plants. MGraw—Hill, Berkshire, UK. 108 Figure 1. Survival percentage (A,E), flowering percentage (B,F), nodes formed after transplant below the inflorescence (C, G), and percent of flowering lateral shoots (D) or flower number per shoot (H) for Veronica spicata 'Red Fox' (AD) and lsotoma axillaris (E-H) held at -2.5 to 20°C for 2 t0 8 or 2.5 to 15 weeks, respectively. Node and flower number for lsotoma are reported for treatments with 100% survival. No control plants flowered. Error bars represent 95% confidence intervals. 109 «Box. 322:» 8:88.85 9: 26.8 308.2828 _m>_>5w Co 88:82.3 aerosol Emamcg Bum BEE moooz “oozm .oq 286: :30 0 0 0 5 0 5 0 0 0 0 0 0 0 0 w 0 0 2 0 . . . . . 4| 8 6 4 2 O 1 8 4 2 0 1 1 8 6 4 2 0 2 4| 1 0 0 a m . s 1. a. m Meeeemee m Mm saw” 0 m w w 5 w B 5. . 5. 0 2. 5 . . , r. + E .m . seem as .Mmmmm e n a m mm 3% 4 .040 ., m A w w 0M ___ .. __ .1 . 2468 ... 4 lie/will r W . mo . n n . . W mxxx.W 0. n +2.: a a. H . 0? H332 4 H 0,. H877.. . .. L +.. L.“ . .. .1 a. w A 1 _W M . A . . .c -.0 s\ p mm.memomwmmmoemec4zommmwmo 6828.3 .6225 mom-.2828 9.826.“. 8:88.85 9: 26.8 289$ 9.826: E880 Emamcg Eta 3:28 832 -2.5 0.0 2.5 5.0 7.5 10.012.515.017.520.0 Temperature (°C) -2.5 0.0 2.5 5.0 7.5 10.012.515.017.520.0 110 100 a) 80" U) B c 8 60» (D Q U) .g 40. Q) 3 2 “- 20- 0. Percentage of Laterals Flowering Days of treatment Figure 2. Flowering percentage (A) and percent of lateral shoots that flowered (B) for Veronica spicata 'Red Fox’ held at -2.5, 0, or 25°C for 12 to 32 days. Each symbol represents the mean of 16 plants. 111 0.0202. 00:00::8 $8 E00050. 0.00 .o..m. 20280000. 8.00.5 we 2 0N. .o m 9 N .2 Doom 2 0N- .0 0.0: 5 .0. 0.20.0.0 02.800. 0:0 Am :4. .xon. 00m. 0.00.30 0o.Eo.0> 8.. 5 :8 20.500... @550 :20th 000: .6 0.0. 05 0:0 8 .$ E02000: 0.20.0088 0:..30 00.2.8 008: .6 .0952 .m 0.29“. 8... 0.30.0QE0... odm mm: 0.9. m.N_. 0.2. m.\. 9.0 md od mN- odN mut- o.m_. 95. 0.2. 95 o.m 9N 0.0 9N. - b l-l p b d d 1 u d 4 u 1 d 4 q d q q a q q u d cod . . ood mod . 8.0 N 0 a; ‘ .0; 3 cm... . . em... 0 m. 0.00 w , Edda. . 0880 x052. «wane... . mm... 0 x . . -u . + . + . -u m o . .. v . . . . . t . m x0039. |¢| m. N m 00.5 05. [4| w. m. 9.00.59. Iol 9.00.50 |0| 6 a 9.0030 lol 9.002; lml m m 9.00; 0d Iol 8.0025 lol w .w: m—s ._ ..-11-lire-11......-.-..-31+ - . . : . r m—s nUU p 05360. 0 00..=oso> < 112 100 g 60 -- .,, . . .. m o z .9 g 0 2.5 weeks 8 40 ‘~-- ~ "0"”5'weeks I 7.5 weeks Cl 10 weeks A 12.5 weeks y=16.7625 + 5.1705x r2=0.9266 O I l I I 2 4 6 8 10 12 Nodes formed after transplant Figure 4. Correlation between node number after transplant and days to flower for Isotoma axillan's plants stored at -2.5 to 20°C for 2.5 to 15 weeks. Treatments with less that 100% survival are presented with (+) inside symbol and were not included in regression analysis. 113 CHAPTER V: CHARCTERIZING THE VERNALIZATION RESPONSES OF THREE ARABIDOPSIS THALIANA ACCESSIONS 114 Characterizing the Vernalization Responses of Three Arabidopsis thaliana Accessions Beth Anne Fausey‘, Steven van Nocker‘, Douglas Schemskez, Jon Agren3, and Arthur Cameron1 1Department of Horticulture, Michigan State University, East Lansing, MI 2Department of Plant Biology, Michigan State University, East Lansing, MI 3Evolutionary Biology Centre, Uppsala University, Uppsala, Sweden Introduction Natural variation in photoperiod and vernalization responses influence flowering time of geographically diverse Arabidopsis thaliana (L.) Heynh. accessions (Johansen et al, 2000; Napp-Zinn, 1985; Nordborg and Bergelson, 1999). While it has been documented that northern populations of species such as Beta vulgaris spp. maritime have greater vernalization requirements than southern populations that experience warmer, shorter winter seasons (Boudry et al., 2002), the correlation between latitude, vernalization requirement, and flowering time has been largely unsupported in arabidopsis (Napp-Zinn, 1985). A comparison of 40 arabidopsis accessions showed a north-south distribution in vernalization response with the majority from northern latitudes having a late- flowering phenotype (vernalization-responsive) while those from central and eastern Europe had early-flowering phenotypes; however, flower timing of these 115 locations (Johansen et al., 2000). In two similar studies, flowering of accessions from Asia, Europe, and North America varied widely in response to day length and vernalization and exhibited little geographical correlation (Karlsson et al., 1993; Nordberg and Bergelson, 1999). For example, all Swedish and Finnish accessions responded strongly to either seed or rosette cold treatment at 4°C, but so did accessions from Germany, the Midwestern United States and England while a NonNegian accession exhibited little response (Nordberg and Bergelson, 1999). Arabidopsis is a common weed often not at equilibrium with the local environment and, in addition to genetic factors, local climate and environmental conditions play key roles in flowering responses (Schemske, personal communication). The effectiveness of arabidopsis vernalization varies with temperature and duration of exposure. Napp-Zinn (1969) reported that vernalization of imbibed seed of the winter annual strain Stockholm occurred at —3.5 to 4°C with treatment effectiveness varying with temperature and duration of exposure; 4°C was identified as the most effective temperature for cold treatments of 14 to 38 days. As a result, the majority of studies investigating ecotypic differences in vernalization responses and requirements have evaluated cold treatment at 4 or 5°C (Karlsson et al., 1993; Lee and Amasino, 1995; Johansen et al., 2000). The effectiveness of vernalization has also been reported to vary with plant age; the greatest effect occurs with imbibed seed and plants between 45 and 90 days old (Napp-Zinn, 1969). However, seed and rosette cold treatments were not always equally effective. A short cold treatment of 3 d at 4°C generally 116 had a minor effect in decreasing time to flowering of imbibed seed compared with a longer 27 d treatment (Nordberg and Bergelson, 1999). The 27—d cold treatment was generally more effective at the seed stage than a 38—d cold treatment for rosettes, although the magnitude of difference in timing varied with accession (Nordberg and Bergelson, 1999). In general, accessions that responded to cold treatment as rosettes also responded to seed cold treatment. In light of this data, and because this better represents winter annual accessions overwintering as rosette plants, we vernalized 6-week-old (42 day) plants in this study. The objective of this research was to compare growth and vernalization responses of three accessions, the common laboratory strain Columbia, as well as two naturally occurring populations, Rodasen from Sweden and Bolsena from Italy under controlled environmental conditions. We quantified growth and development characteristics during and following vernalization at a broad range of temperatures (-2.5 to 15°C), including leaf formation, flowering time, and inflorescence development. A second objective was to compare flowering responses of the Swedish and Italian populations when grown in experimental gardens located within the climatic region of each locality. The Swedish and Italian populations represent, respectively, the northern and southern limits of the natural geographic range of arabidopsis. Given the striking climatic differences between these regions, we expect differential adaptation in flowering and vernalization requirements. This experiment provides a direct test of the relative role of genetics and 117 environmental factors on flowering. Furthermore, we compare our results from controlled environment studies with those obtained from common garden studies. Materials and Methods Plant Materiel. Seeds were collected in June 2002 from north-central Sweden, at Redasen (N 62°48’ E 18°12’), located 60 km northeast of Sundvall, and in April 2003 from central Italy, near Bolsena (N 42°39’ E 12°00’), located 100 km NW of Rome. These sites are 2, 440 km apart, and were chosen to represent the northern and southern limits of the natural geogeaphic range of Arabidopsis thaliana. The seeds used for the greenhouse and field experiments were the first- generation, selfed progeny of plants grown in the MSU growth chambers in the summer of 2003. For ROdésen, all seed was obtained from a single parent, while for Bolsena, seed was pooled from two full sibs derived from a single maternal parent. All seed was stored at room temperature in the dark. Controlled environment studies. Growth Conditions. Seeds of all three accessions were imbibed in distilled water overnight at 4 °C and planted in 72-cell plug trays (50-mL cell volume) filled with a commercial peat-perlite media (Sure-Mix, Michigan Grower Products, Galesburg, Mich.) (on 2l13/O4, 3/05/04, 3/26/04, and 4/16/04). Seedlings were started at 3-week intervals and grown for six weeks under short days in a controlled environment chamber at 20 ’C with 150 umol-m'z-s‘1 PAR provided by cool white fluorescent and incandescent lamps for 11 h-d". After 6 weeks, plants 118 were randomly transferred to vernalization chambers set to -2.5, O, 2.5, 5, 7.5, 10, 12.5 and 15 11°C. Replicate numbers for each accession ranged from 5 to 16, depending upon treatment. Plants vernalized at —2.5 to 2.5 °C and those at 5 to 15 °C were grown with 25 and 100 pmol-m'z-s'1 PAR, respectively, provided by cool white fluorescent and incandescent lamps for 11 h-d". Plants were vernalized for O, 3, 6, or 9 weeks and were irrigated when needed with nutrient solution providing 125N-12P—1 OOK—65Ca-12Mg mg-L‘1 (12.6% nitrate nitrogen, 2% ammoniacal N) plus 1.0Fe—O.5Mn—0.5Zn—1 .OCu-O.BB-0.1Mo mg-L'1 (MSU Special; Greencare Fertilizers, Chicago, Ill.) Air temperature in each chamber was monitored with 36-gauge (0.127-mm diameter) type-E thermocouples connected to a CR10 datalogger (Campbell Scientific, Logan, Utah). Measurements were collected every 10 s and hourly averages recorded. All plants were removed from the chambers, transplanted into 5.5-inch (1.1 L) containers with the same media described above, and randomly placed on greenhouse benches on 28 May 2004. Plants were grown at 20:: 4 °C with a 16- h photoperiod provided by day extension lighting from 400-watt high pressure sodium lamps from 0600 to 0800 HR and 1700 to 2200 HR. Additional supplemental lighting provided 100 pmol-m'Z-s'1 only when ambient greenhouse light levels dropped below 290 pmol-m'Z-s'1 from 0800 to 1700 HR. Instantaneous light in each treatment was measured at plant height with a line quantum sensor (Ll-COR, Lincoln, Nebr.) connected to a CR10 datalogger. Greenhouse air temperatures were controlled by a climate-control computer (Priva, Model CD750, De Lier, The Netherlands) and were monitored on each 119 bench with 36-gauge (0.127-mm diameter) type-E thermocouples connected to a CR10 datalogger. Temperature and light measurements were collected every 10 s, and hourly averages recorded. Plants were irrigated as needed with the above nutrient solution. During the course of the experiment, the 125°C chamber malfunctioned and all data were removed from analysis. Rodésen plants vernalized at 15°C for 3 weeks died in the greenhouse due to drought stress. Measurements. The number of primary rosette leaves greater than 3 mm in length were recorded before transfer to vernalization chambers and upon placement in the greenhouse following vernalization. Additional leaves formed in the greenhouse below the inflorescence, the date of inflorescence appearance in the rosette, and date of first open flower were recorded for all reproductive plants. The total number of inflorescences arising from the basal rosette and the number of flowers along the primary inflorescence were counted 28 d after the first plant of each accession and vernalization duration flowered in the greenhouse. Time to visible inflorescence, time to first open flower, and time from visible inflorescence to first open flower were calculated and used in analyses. Several Columbia and Bolsena plants formed visible inflorescenoes during 9 weeks vernalization at 10 and 15°C, and these plants were removed from analysis. 120 Statistical Analysis. Data were analyzed using general linear model procedures (PROC GLM) in SAS version 8.0 (SAS Institute, Cary, NC). Error bars represent 95% confidence intervals. Common garden studies. To examine the relative importance of genetic and environmental factors on flowering time in these populations, we conducted reciprocal transplant experiments in gardens located within the climatic regions of the field localities. The field experiment in Sweden was initiated on September 24, 2003 at the botanical garden of Uppsala University, located approximately 330 Km south of Rodésen. Seeds were planted in 5 x 5cm pots (10 seeds/pot) containing a sand and soil mixture, with 50 pots for each accession and 20 pots as controls. The pots were randomized and placed in a cold frame at 2 cm spacing between pots. The field experiment in Italy was begun on October 29, 2003 near the town of Castelnuovo di Porto, located approximately 90 km SW of Bolsena, Italy and 30 km north of Rome. A 2.5 x 6.5m plot was established using soil collected from a nearby road cut. Round plastic rings, 10 cm in diameter were distributed in four rows at approximately 15 cm spacing, with 30 rings for each accession and 20 rings as controls. Each of the experimental rings received 100 seeds. Data loggers shielded from direct sunlight with reflective covers (Hobo Pro Temp/External Temp, Onset Computer Corp., Boume, MA, USA) recorded hourly air and soil temperatures at both sites throughout the growing season. One logger was used in Sweden and two in Italy. At both sites, the number of plants 121 initiating their first infloresenoe and the number of plants in flower were determined at approximately 14-d intervals following the first sign of infloresence development. Results Growth and development during vernalization. Plants grew continuously during the 9-week vernalization treatment and formed 1 to 100 new leaves, depending on temperature and accession, before transfer to long days (Fig.1). Rate of leaf formation of each accession did not differ with temperature at 3, 6 and 9 weeks vernalization which infers constant leaf formation over a 9-week period at each temperature, and data were pooled for analysis (Fig. 1). Plants grew, albeit slowly, even at -2.5 °C, and leaf formation rates increased from about 0.05 to over 1 leaf per day while cooled at -2.5 and 15 °C (Fig. 1). At -2.5 to 5°C, accession differences were not apparent, but at warmer temperatures, leaf formation rates were consistently greater for Columbia followed by Bolsena and Rodésen (Fig. 1). Growth and flowering after vernalization treatments. Flowering differences between accessions were observed. Columbia had no vernalization requirement, Bolsena had a facultative to nearly obligate, and Rodésen had an obligate vernalization requirement for flowering. Only 20% of Bolsena plants and no Rodésen plants flowered without exposure to a minimum of 3 weeks at -2.5 to 15 °C (Table 1). Following all vernalization treatments, Rodésen consistently 122 flowered later than Columbia and Bolsena and produced fewer inflorescences and flowers per plant (Figs. 2, 3). For all accessions, the number of leaves formed after vernalization decreased to a minimum following 9 weeks vernalization (Fig. 2). After 9 weeks vernalization, plants formed 15 to 30 new leaves below the inflorescence (Fig. 2). Due to significant interactions between accession, temperature and vernalization duration, (P<0.001) no clear relationship between leaf number and these variables could be determined. Reproductive timing of all accessions appeared to be influenced more by duration of vernalization than by vernalization temperature. Time from transplant to visible inflorescence decreased with increasing length of vernalization, and the magnitude of response varied with accession and temperature (Fig. 2). The greatest differences in time to visible inflorescence between individual accessions were observed following 3 weeks vernalization (Fig. 2). Rodésen achieved the greatest reduction in time to visible inflorescence with increasing vernalization followed by Bolsena and Columbia. Differences in timing following 3 and 9-week vernalization treatments were minimum at -2.5°C and maximum at 10 and 15°C. In all, Columbia, Bolsena, and Rodésen plants vernalized for 9 weeks formed visible inflorescences up to 12, 18, and 25 days earlier, respectively, than those vernalized for only 3 weeks. Despite large differences in plant size and total leaf number at flower, reproductive timing of accessions was remarkably similar across temperature after vernalization (Fig. 2). For example, ROdésen plants cooled for 9 weeks at - 123 2.5 and 15°C formed one and 71 leaves during vernalization and 29 and 31 leaves, respectively, after vernalization. Visible inflorescences were recorded in 19-21 days after placement in the greenhouse, though at flower, plants vernalized at -2.5 were nearly one-third the size of those vernalized at 15°C. Slight differences in flower timing (4-5 days) were observed across the range of temperatures investigated for all accessions following 3 weeks vernalization and for Columbia and Bolsena following 6 weeks vernalization (Fig. 2). ROdasen exhibited the greatest temperature response following 6 and 9 weeks vernalization with maximum differences in time to visible inflorescence of 10 and 14 days, respectively. Time to visible inflorescence was generally minimum at 5 to 10°C and maximum at -2.5 and 15°C (Fig. 2) following 6 weeks (Rodésen only) and 9 weeks (all accessions) vernalization, although differences were not always statistically significant. Time to visible inflorescence was most variable for Rodasen and to a lesser degree Bolsena, while Columbia plants flowered uniformly regardless of vernalization temperature or length of cold treatment (Fig. 2). Neither Rodasen nor Bolsena bloomed completely (0 to 20%) without vernalization (Table 1). Increasing vernalization from 3 to 9 weeks greatly improved uniformity in subsequent time to visible inflorescence of both accessions (Fig. 2). The flowering response of each accession was quantified by measuring the number of inflorescences originating from the basal rosette and the number of flowers that formed along the primary inflorescence (Fig. 3). Inflorescence and flower number trends were similar for Columbia and Bolsena following 6 and 9 124 weeks vernalization. At all temperatures evaluated, Rodasen formed the same or fewer inflorescences and flowers than Columbia or Bolsena following vernalization. In general, inflorescence and flower number of Bolsena and Columbia increased following vernalization at -2.5 and 0°C while Redasen exhibited the greatest flowering response at these temperatures. Flower and inflorescence number were most variable following vernalization at 15°C. Garden trials. Bolsena and Rddésen plants were grown in common garden sites in Uppsala, Sweden and Belmonte, Italy. Plants in Sweden reached peak flowering in early June, while flowering occurred in late April in Italy (Table 2, personal observation). Both accessions had nearly completed flowering in Italy when bolting began in the Swedish plots (Table 1). At both locations. Bolsena bolted and flowered slightly faster than Rodasen, but differences were not significant at the second observation. Discussion Vernalization responsiveness. While Bolsena had a nearly obligate cold requirement for flowering and Columbia flowered completely without cold, Rodasen absolutely required cold for flowering. Interestingly, all accessions responded to the broad range of temperatures investigated, and as few as 3 weeks at —2.5 to 15°C were effective for vernalization. All accessions survived 9 weeks at —2.5°C, though it is noteworthy that none were acclimated to low temperatures before treatement. In general, the greatest variability in responses was achieved following vernalization at -2.5 and 15°C. The upper end of 125 effective temperatures for vernalization appears to be 15°C as Bolsena and Redasen control plants grown at 20°C were not vernalized. However, control plants were not grown for an additional 3, 6, or 9 weeks at 20°C as other vernalization treatments. Vernalization of winter rye, henbane, onion, easter lily and Rhipsalidopsis also occurs at temperatures of 15 to 17°C (Lang, 1952, Lange,1992) Vernalization is a quantitative process and earlier flowering is promoted with increasing lengths of vernalization treatment until the response is saturated. Although complete saturation of the vernalization response was not determined in this study, several Bolsena and Columbia plants initiated flowers during 9 weeks vernalization at 10 and 15°C, indicating that the vernalization and photoperiod requirements for flowering were fulfilled. Lee and Amasino (1995) found no differences in leaf number below the inflorescence in late and early flowering Columbia lines grown under short days after 30 to 40 days of seed vernalization, although complete saturation of the response required 80 days of seed cold treatment at 4°C. In our study, plants were vernalized for up to 63 days as rosettes with 10 to 20 leaves, depending on accession, and further cold may be needed for complete saturation of the response. Photoperiod and flowering. Arabidopsis is a quantitative long day plant that flowers most rapidly under long day photoperiods. Accessions have been categorized as summer annual (early flowering) or winter annual (late flowering) due to their flowering response under long days (Napp-Zinn, 1969). Winter 126 annual accessions typically require or are responsive to a vernalizing cold treatment that reduces flower timing under long days compared to noncooled plants. Why did Bolsena and Columbia plants flower during vernalization treatments and not Rodésen? Differences in photoperiodic responsiveness of each accession following vernalization may explain differences in flower initiation and flower timing. The photoperiod required for most rapid flower initiation of the accessions Bolsena and Rodésen is unknown. The photoperiod required for most rapid flowering of Bolsena is likely to be less than 16 hours because plants are exposed to ~12 hour photoperiods when flowering naturally occurs in Italy (early March). In fact, some Bolsena and Columbia plants initiated flowers when vernalized at 15°C for nine weeks and most likely initiated flowers at 5, 7.5 and 10 °C as time from transplant to first visible inflorescence for these treatments was less than 5 days. For Rodasen, photoperiods during likely times of flower induction are ~ 14 to 15 hours from April 1 onwards in Sweden, and it is possible that plants do not flower under shorter photoperiods. No Rodésen plants initiated flowers during the 11-h vernalization treatment, even after 9 weeks vernalization. In the garden trial, the photoperiod in Italy was approximately 12 h at the time of bolting, and Rodésen plants began to flower (Table 1). If 12 h is below the photoperiod for rapid flowering of ROdasen, adequate vernalization may have either reduced the accession’s sensitivity to photoperiod or high irradiance in the spring modified the photoperiodic requirement. Both cases have been documented in winter-annual strains of arabidopsis (cited by Napp-Zinn, 1985). 127 Reproductive output. Inflorescence and flower number for the winter annual accession Rodésen lagged behind Columbia and Bolsena after 49 d in the greenhouse. Pigliucci and Marlow (2001) found the reproductive output of a winter annual accession to vary with early flowering accessions over different season lengths. Reproductive output of the winter annual accession lagged behind the spring ephemeral accessions after 30 and 44 d seasons, caught up to them by 58 d and outranked them after 86 d season. Had we allowed a longer growing duration, it is possible that Rddésen could have produced as many inflorescences as Columbia and Bolsena. 128 Literature Cited Boudry, P., H. McCombie, and H.Van Dijk. 2002. Vernalization requirement of wild beet Beta vulgaris ssp. maritime: among population variation and its adaptive significance. Journal of Ecology 90:693-703. Johanson, U., J. West, C. Lister, S. Michaels, R. Amasino, and C. Dean. 2000. Moloecular analysis of FRIGIDA, a major determinant of natural variation in Arabidopsis flowering time. Science 290:344-347. Karlsson, B. H., G.R. Sills, J. Nienhuis. 1993. Effects of photoperiod and vernalization on the number of leaves at flowering in 32 Arabidopsis thaliana (Brassicaceae) accessions. American Journal of Botany 80(6):646-648. Lang, A. 1965. Physiology of flower initiation. In: Encyclopedia of Plant Physiology (ed. W. Ruhland). Pp. 1371-1576. Springer-Verlag, Berlin. Lange, N. 1992. Modeling flower induction in Lilium/ongiflorum. MS. Thesis, Michigan State University, East Lansing, MI. Lee, I. And R.M. Amasino. 1995. Effect of vernalization, photoperiod, and light quality on the flowering phenotype of Arabidopsis plants containing the F RIGIDA gene. Plant Physiol. 108:157-162. Napp-Zinn, Klaus. 1969. Arabidopsis thaliana (L.) Heynh. In: The Induction of Flowering: Some case histories. L.ZT.Evans, editor. Ithaca, NY. Napp-Zinn, Klaus. 1985. Arabidopsis thaliana. In H.A. Halevy, editor. Handbook of Flowering Voume 1, 492-503. CRC Press, Boca Raton, FL. Nordberg, M. and J. Bergelson. 1999. The effect of seed and rosette cold treatment on germination and flowering time in some Arabidopsis thaliana (Brassicaceae) accessions. American Journal of Botany 86(4):470-475. Sheldon, C.C., D.T. Rouse, E. J. Finnegan, W.J. Peacock, and ES. Dennis. 2000. The molecular basis of vernalization: the central role of FLOWERING LOCUS C (FLC). PNAS 97(6):3753-3758. Stenoien, H.K., C. B. Fenster, H. Kuittinen, and O. Savolainen. 2002. Quantifying latitudinal clines to light responses in natural populations of Arabidopsis thaliana (Brassicaceae). American Journal of Botany 8(10):1604-1608. 129 Stinchcombe, J.R., C. Weining, M. Ungerer, K.M. Olsen, C. Mays, S.S. Halldorsdottir, M.D. Purugganan, and J. Schmitt. 2004. A latitudinal cline in flowering time in Arabidopsis thaliana modulated by flowering time gene FRIGIDA. PNAS 101(13): 4712-4717. 130 .0_0>_0:0 So... 00580. 0.02. 0.00 0:0 00:28:20... 09:05 N 8.8. - 8.8. 8.8. 8.8. 8.8. 8.8. 8.8. 8 8:8. - 8:8. 88. 88. 8:8. 8:8. 88. 0 8.8 - 8.8. 8.8. 88. 8.8. 8.8. 8.8. m 8008: E8 8 8:8. - 8:8. 8:8. 8:8. 8:8. 8:8. 8:8. 8 8:8. - 8:8. 8:8. 8:8. 8:8. 8:8. 8:8. 0 E8. - E8. E8. E8. E8. E8. 88. m 0828 8.8 8 8:8. - 8:8. 8:8. 8:8. 8:8. 8:8. 8:8. 8 8:8. - 8:8. 8:8. 8:8. 8:8. 8:8. 8:8. 0 8:8. 8:8. - 8:8. 8:8. 8:8. 8:8. 8:8. 8:8. 0 05528 8:8. 8 8.8 8.0. New. 8.8. 0.. 8.0 0.... 8.8 0.0- cwwflwmocnw 8.0089... 8.: 0.30.0:E0h ..:0.:.00.. :000 .0. 00:5: 0.02.80. .3 0026.6. 0080.50.08 85.032“. .008 .0 0.0.. 0.0; 0.88 .988 .000. o. 0.0 - .0 :o_.0~__0:.0> 0x002. m .o .o .m .o 0. 0008.0 0:200080 0.030302. 00.... .o. 0080.50.08 85.032“. .. 030.. 131 .88.: .o .88 .89: .0 200.....90 .0 .000580821 E- .2. wz mz t... 0300585 . o H 8.0 3.0 H .06 0 :00003. . o H 8.0 mod H Ed 0 050.8 05.0.52“. mz m2 wz 3... 00:85:85 . . 3.0 H mm... mod H 00.0 :00002... . o H mod 0 H 8.: mod H 5.0 050.3 09.8 500.5% 0.00%.... 02 z... 8:08.80 «o... H 2.: mod H mud c :00003. no... H 0...: 3.0 H 3.0 0 050.8 0:00.52“. 02 02 2... .8080290 mod H No mod H and o H 3.0 :00003. mod H on... mod H 8.0 mod H owd 050.8 09.8 50.. .0.=o§0m 08.. 0 .0... 00 .0... o. .8... 8 __.:< 0 5.0.2 00 5.0.2 o. 03.0.m 0.00 3.00000... 05.0.52“. .0:o...o:o.: 00::0.0:0.. .oo.-0.0_..00-0:_00.0 :0 00.03080 0.0.5 0.00.- .- .>_0>_.00:00. 0.00::3 0:0 0.:o:._0m .0 500.30: .0: 0.0.: on 0:0 cm H z :02: 0.0 5.0 0.00:0.0 H 0:00.). .808 :. 00.:0.:. woow :_ 500.5 0.08:: 0:0 5.0.. .0.:oE_0m .0 .:0:._.0:x0 :00.0m 0... :_ .050: 0. 0000.0 0:0 00:00 .0... 0.:0.: .o :o...o:o.: 0:... .N 0.8.0... 132 133 .oodv .oodv ~86 .oodv .oodv .oodv ._..>..< .oodv .oodv .oodv .oodv .oodv .oodv .r.> .oodv .oodv .oodv ..oo.ov .oodv .oodv ....< .oodv .oodv .oodv .oodv .oodv .oodv 0.3.0.0:E0... .oo.ov .oodv .oodv .0 .od .oodv .oodv >.< .oodv .oodv .oodv .oodv .oodv .oodv :o_.0~_.0:.0> .oodv .oodv .oodv mmmod .oodv .oodv 5.000004. 00:0000.o=:_ .0080: 020000.22. .0082: 3005.059 8:05.05? 5.06:: 00:0000.o=:_ 0.0.0.5 .00. .0..0 3:30 0058 :0 0.0.50. “. .05... 0. 0E. .- .0.o... 00500.. 00>00.. 05.000000 __0 .0 0.05.0:0 <>Oz< .0006 So... 00:.0> “.00.. .. 0.00... 5.0.:0E0.:::w 134 m _.00.0 t 00.0 0000.0 .00.0v 00 ..0.0 _.00.0v ..x> _.00.0v .00.0v _.00.0v .00.0v _.00.0v .00.0v 0.:.0.0:E0.. 00 _.0.0 .00.0v .00.0v 0000.0 .00.0v _.00.0v :o_.0N._0E0> cmmauom .00.0v .00.0v N000.0 .00.0v .00.0v F00.0v ..x> .00.0v 0000.0 _.00.0v .00.0v 00.0 F .0 _.00.0v 0.30.0:E0... hmwod _.00.0v _.00.0v N000.0 _.00.0v _.00.0v :o..0~..0:.0> 0:00.00 _.00.0v _.00.0v _.00.0v _.00.0v 000.0v _.00.0v ._.x> 000.0v _.00.0v 000.0v _.00.0v 000.0v _.00.0v 0.3.0.0:E0k 000.0v .00.0v _.00.0v 0000.0 _.00.0v _.00.0v :o_.0N__0E0> 0.0.0200 00:0000.o=:_ .0080: 00:0000.o..:_ .0005: 5000.089 5000.059 5.08:: 00:0000.o..:_ 0.0.03 .00. 0.0 0:030 00.300 5.000004. :0 0.052“. .08... o. 0...... .05.. 0000.. 0000.. 08.000000 .0:0.>_0:. .0 0.05.0:0 <>Oz< 50.. 00:.0> .QQ .N 0.00... 5.0.:0E0_:::w 120 + Columbia 100 - —I.—. Bolsena . . . .. —A— Rodasen Leaf number during vernalization Leaf formation rate (leaves/day) 0.0 I I I I I I 1 I -2.5 0.0 2.5 5.0 7.5 10.0 12.5 15.0 Temperature C Figure 1. The number of leaves > 3mm formed during the 9-week vernalization treatment and the rate of leaf formation for Columbia, Bolsena, and Rodésen accessions grown at -2.5 to 15°C for 3, 6 and 9 weeks. Leaf formation rates were nonsignificant for duration of cooling and were pooled for each accession at each temperature. Error bars represent 95% confidence intervals. 135 80 + Comma _ .40 + Bolsena 60 1 30 40 20 .5 o 1 0 0 ° .____ 6weeks _40 8 0 0 «E» 601 9 o i 330 GE 8 '6 m 40 ' . 20 3 2 E 20‘ ~10 .9 3 e 0.. (6 g o . - . . . . . . . . . . . . . . 0 D 9weeks 4o 60 - » 30 40 3 20 20 4 -10 0 0 -2.5 0.0 2.5 5.0 7.5 1o.o12.515.o -2.5 0.0 2.5 5.0 7.5 10.012.515.o Temperature (°C) Figure 2. Number of leaves formed in the greenhouse below the inflorescence after vernalization and the number of days from placement in the greenhouse to visible inflorescence for all accessions following 3, 6, and 9 weeks vernalization. Error bars represent 95% confidence intervals. Visible inflorescences were recorded after 25 d for nonvernalized Columbia plants, and one nonvernalized Bolsena plant reached visible bud in 32 d. 136 .m_m>.2:_ 853:8 $3 9865 m5: .o..m_ .3526: Ema _o..:oo mcomBm m:o gm :8me 02 2950: N._, H md ucm m8:8mo.o::_ v.~ H ad .0 889m :m 8:..8 Doom um 2m: mEma _o.Eoo £98200 .:o=mN__m:.m> .o 383 m .wtm 5688... :08 .8 Ev 8:88.85 2.25.: 05 :o 9250: B .852: 05 .8: 2: .883: 8:8mo.o=:_ :30... .m 93$ 0 0.29an8. 0.9. mNF 0.0? ms o.m mN o.o md. 9m? 99 0.9 ms 0.0 md o.o mN- o . _ . . . . t . . . . . . . _ 0 ¢ 4 4. m I 4 1 t. W H m .. w 2 m o a . a S .M, \ a m m 8 m w \I. I .l \ U m. mm H J 9 : ; w 1:3a8:,.1§.|;§i; . . on m. 238+ J ow m ,, $328,194, , _ . 5 . ; <5 8 137 RRRRRRRRR lllllilllilllillllllllllllll 12