.JJ.

 

This is to certify that the

thesis entitled

Influence of photoperiod and temperature on flowering of
Campanula carpatica 'Blue Clips', Coreopsis grandi-

flora 'Early Sunrise', Coreopsis verticillata 'Moonbeam',
Rudbeckia fulgida 'Goldsturm', and Lavandula angustifolia '

 

l
'Munstead' . presented by
Catherine Margaret Whitman
has been accepted towards fulﬁllment i
of the requirements for
Master ofme— degree in We

 

A‘ajor professor

Date ////(/ 7‘}/

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

 

L\".
CA‘
SK?

INFLUENCE OF PHOTOPERIOD AND TEMPERATURE ON FLOWERING OF
CAMPANULA CARPAH CA 'BLUE CLIPS' , COREOPSIS GRANDIFZORA 'EARLY
SUNRISE', COREOPSIS VERHCILLATA 'MOONBEAM', RUDBECKIA F ULGIDA

'GOLDSTURM' AND LAVANDULA ANGUSHFOLIA 'MUNSTEAD'

By

Catherine Margaret Whitman

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Horticulture

1995

 

BKF
CAA‘

Eﬁbt
sodu
‘Bhn
‘hky
ant)
The

ﬂou'

and.
Larg
incrc
Phom
Unde

0f pr

deVel

ABSTRACT
INFLUENCE OF PHOTOPERIOD AND TEMPERATURE ON FLOWERING OF
CAMPANULA CARPA TICA 'BLUE CLIPS', COREOPSIS GRANDIFLORA 'EARLY
SUNRISE', RUDBECKIA FULGIDA 'GOLDSTURM' AND LAVANDULA
ANGUSIYFOLIA 'MUNSTEAD'
By

Catherine Margaret Whitman

Effectiveness of supplementary light from cool white ﬂuorescent, high-pressure
sodium, incandescent, and metal halide lamps for ﬂowering in Campanula carpatica
‘Blue Clips’, Coreopsis grandiﬂora ‘Early Sunrise’, Coreopsis verticillata
‘Moonbeam’, and Rudbeckia fulgida ‘Goldsturm’ was compared. An irradiance of 1.0
umol m'ZS'l from any lamp tested was adequate for uniform ﬂowering in these species.
The inﬂuence of 5C treatments on ﬂowering in C. carpatica was determined. Time to
ﬂower was unaffected by cold treatments in 12- to 16-node plants, but decreased by
=10 days in 10— to 12—node plants cooled 14 weeks. The inﬂuence of 5C treatments
and photoperiod on ﬂowering in Lavandula angustifolia ‘Munstead’ was determined.
Larger initial plant size and increased duration of cold storage were associated with
increased percent ﬂowering and decreased time to ﬂower. Cold treatments reduced the
photoperiodic response in L. angustifolia. C. carpatica and L. angustifolia were forced
under 15, 18, 21, 24, or 27C, and the linear relationship between temperature and rate
of progress to ﬂowering was determined. Base temperatures and degree-days for

developmental stages were calculated and can be used to predict ﬂowering.

 

 

int
K0

Yu.

cou
Kat;

Olhe

ACKNOWLEDGMENTS

I would like to express my sincere thanks to Dr. Royal Heins for his generous
assistance, guidance, and support throughout my program.

The contributions of the other members of my guidance committee; Dr. Art
Cameron, Dr. Will Carlson, and Dr. Jan Zeevaart; are greatly appreciated. Their
valuable input and inspirations truly enriched this project.

I would also like to thank Henry Mast Greenhouses, and the other greenhouse
growers who supported this work, for their generosity which made this research
possible. Their contributions also included plant material, as well as ideas,
observations, and suggestions which beneﬁtted us all.

Many of my colleagues' assistance and support helped in both concrete and
intangible ways. A special thanks to Bill Argo, Beth Engle, Cheryl Hamaker, Paul
Koreman, Bin Liu, Erik Runkle, Tom Wallace, Cara Wallace, Mark Yelanich, and Mei
Yuan.

The numerous day-to-day tasks in the greenhouse that made this project possible
could not have been completed without Sarah Bommarito, Paul Booth, Carmen Diaz,
Katie Larkin, Nate McGinnis, Shawn Sevenski, Dan Tschirhart, Ryan Warner, and

others who kept the greenhouse running smoothly; and I thank you all.

iii

 

 

 

 

 

LE
115
SE(

SEC

SEC

SEC]

TABLE OF CONTENTS

LIST OF TABLES ........................................... v
LIST OF FIGURES ......................................... vii
SECTION 1, LITERATURE REVIEW .............................. 1
SECTION 11
Inﬂuence of long-day treatment light quality on ﬂowering of herbaceous
perennials ........................................... 27
Abstract ............................................ 29
Materials and Methods .................................. 33
Results ............................................. 37
Discussion .......................................... 40
Literature cited ....................................... 44
SECTION III
Inﬂuence of cold treatment and forcing temperature on ﬂowering of Campanula
carpatica 'Blue Clips' ................................... 62
Abstract ............................................ 64
Materials and Methods .................................. 68
Results ............................................. 72
Discussion .......................................... 73
Literature cited ....................................... 77
SECTION IV
Inﬂuence of cold treatment and forcing temperature on ﬂowering of Lavandula
angustifolia 'Munstead' .................................. 87
Abstract ............................................ 89
Materials and Methods .................................. 92
Results ............................................. 96
Discussion .......................................... 98
Literature cited ...................................... 103

iv

Tab]:

 

rd

 

I‘J

Table

LIST OF TABLES

Page
SECTION H

Characteristics of plant material. Experiments 1, II, and III started on 16 Sept.
1994, 15 Dec. 1994, and 15 March 1995, respectively. ............ 47

Parameters of nonlinear and linear regression analysis relating rate of progress
to ﬂowering to quantum ﬂux (400-700mn). b1 and b3 are slopes and b0 and b2
are intercepts. For nonlinear regression, b0, b1, b2. and b3 represent values for
Eqs. [2] and [3]. Iintersection is the irradiance at the intersection point of the two

lines in umol m“’-S'l and is calculated using Eq. [1]. ................ 48

Inﬂuence of light quality on mean total plant height at the time of ﬂowering.

Data includes the height of all plants that ﬂowered. ............... 49
SECTION III

Inﬂuence of 5C treatments on mean time to ﬂower, number of nodes formed
during forcing, number of ﬂower buds present at ﬂowering, and height at
ﬂowering of C. carpatica 'Blue Clips'. ....................... 79

Actual average daily air temperatures during the indicated developmental stages,
during forcing of C. carpatica 'Blue Clips'. .................... 80

Parameters of linear regression analysis relating forcing temperature to rate of
progress to visible bud and ﬂowering in C. carpatica ‘Blue Clips’. Intercept
and slope were used in Eqs. [2] and [3] to calculate base temperature (Tb) and
degree-days, or thermal time (°days). Visible bud and ﬂowering data for 128-
cell plants forced at setpoint 27C were not included in the calculations because
the plants were damaged by heat stress. ...................... 81

Inﬂuence of forcing temperature on time to ﬂower in C. carpatica ‘Blue Clips’.

Comparison of regression lines within each developmental stage, between the 3

plant sizes tested. .................................... 84
SECTION IV

Actual average daily air temperatures during the indicated developmental stage
of L. angustifolia ‘Munstead’. ........................... 105

Parameters of linear regression analysis relating forcing temperature to rate of
progress to visible bud and ﬂowering in L. angustifolia ‘Munstead’. Intercept

V

and slope were used in Eqs. [2] and [3] to calculate base temperature (Tb) and
degree—days, or thermal time (°days). Development was delayed in some plants
in the highest temperature treatments, so data for 128-cell plants forced at 23
and 26.2C, and for 50-cell and 5.5-cm plants forced at 26C was not included in
the calculations. ..................................... 106

Inﬂuence of forcing temperature on time to ﬂower in Lavandula angustifolia
‘Munstead’. Comparison of regression lines within each developmental stage,
between the 3 plant sizes tested, followed procedures described by Snedecor and
Cochran (1967). .................................... 107

vi

 

Figure

LIST OF FIGURES

Page
SECTION II

Spectral characteristics of light sources, average of 3 scans. Calculations of
RzFR (redzfar-red) ratio are based on the 1.0 umol m'ZS'l level (400-700 nm).
Band widths of 600-700 nm were used for red radiation, and 700-800 nm for
far-red radiation. (A) Cool white ﬂuorescent, Philips F48T12, 60 watts. Photon
ﬂux in 600-700 nm waveband = 0.25 pmol m‘zs"; RzFR = 8.8. (B) High
pressure sodium, General Electric Lucolox LU70/MED, 70 watts. Photon ﬂux
in 600-700 nm waveband = 0.40 umol m’zs"; RzFR = 5.9. (C) Incandescent,
Philips 100 watt bulb. Photon ﬂux in 600-700 nm waveband = 0.58 umol m'ZS'
1; RzFR =0.7. (D) Metal halide, VENTURE MH70W, 70 watts. Photon ﬂux
in 600-700 nm waveband = 0.27 umol m‘zs"; RzFR = 3.3. ........ 51

Effect of light source and intensity on rate of progress to ﬂowering in
Campanula carpatica 'Blue Clips'. Each symbol (0) represents one plant. Data
from the three replications were combined. The solid line (—) represents the
predicted rate of progress to ﬂowering based on Eqs. [2 and 3]. Plants which
did not ﬂower were not included in the analysis. The saturation illuminance (Isa)
is the quantum ﬂux at the point of intersection between the two lines. The rate
of progress to ﬂowering is constant at irradiances above 153,. None of the plants
under SD ﬂowered. Statistical analysis is presented in Table 2. ....... 53

Effect of light source on height at ﬂowering in (A) Campanula carpatica 'Blue
Clips' , (B) Coreopsis grandiﬂora 'Early Sunrise' , (C) Coreopsis verticillata
'Moonbeam' , and (D) Rudbeckia ﬁdgida 'Goldsturm'. Data includes the height
of all plants that ﬂowered. Letters denote signiﬁcant differences between
treatments (P = 0.05), using Duncan’s multiple range test. ......... 55

Effect of light source and intensity on rate of progress to ﬂowering in Coreopsis
grandiﬂora 'Early Sunrise'. Each symbol (0) represents one plant. Data from
the three replications were combined. The solid line (——) represents the
predicted rate of progress to ﬂowering based on Eqs. [2 and 3]. Plants which
did not ﬂower were not included in the analysis. The saturation illuminance (1,3,)
is the quantum ﬂux at the point of intersection between the two lines. The rate
of progress to ﬂowering is constant at irradiances above Isa. None of the plants
under SD ﬂowered. Statistical analysis is presented in Table 2. ....... 57

vii

Effect of light source and intensity on rate of progress to ﬂowering in Coreopsis
verticillata 'Moonbeam' , experiment 1. Each symbol (0) represents one plant.
For (A) and (B), the solid line (—) represents the predicted rate of progress to
ﬂowering based on linear regression. For (C) and (D), the solid line (—)
represents the predicted rate of progress to ﬂowering based on Eqs. [2 and 3].
Plants which did not ﬂower were not included in the analysis. The saturation
illuminance (15,“) is the quantum ﬂux at the point of intersection between the two
lines. The rate of progress to ﬂowering is constant at irradiances above 1”,.
None of the plants under SD ﬂowered. Statistical analysis is presented in
Table 2. ........................................... 59

Effect of light source and intensity on rate of progress to ﬂowering in Coreopsis
verticillata 'Moonbeam' , experiment 11. Each symbol (0) represents one

plant. The solid line ( ) represents the predicted rate of progress to ﬂowering
based on linear regression. Plants which did not ﬂower were not included in the
analysis. Statistical analysis is presented in Table 2. ................ 60

 

Effect of light source and intensity on rate of progress to ﬂowering in
Rudbeckia ﬁtlgida 'Goldsturm'. Each symbol (0) represents one plant. The
solid line (—) represents the predicted rate of progress to ﬂowering based on
linear regression. Plants which did not ﬂower were not included in the analysis.
Statistical analysis is presented in Table 2. ...................... 61

SECTION III

Inﬂuence of forcing temperature on time to ﬂower in C. carpatica ‘Blue Clips’.
Each symbol (0, I, orA) represents the mean of 10 plants. Open symbols (A)
represent data not included in the calculations because the plants were damaged
by heat stress. (A), (B), and (C) show days for the indicated developmental
stage. For (D), (E), (F): solid lines (——) represent predicted values for the rate
of progress to the indicated developmental stage, based on linear regression.
Statistical analysis and calculations are presented in Table 3. ......... 83

Inﬂuence of forcing temperature on (A) number of ﬂower buds, (B) average
ﬂower diameter, and (C) total plant height in C. carpatica ‘Blue Clips’. For (A)
and (C), each symbol (.,I, orA) represents the average of 10 plants. For
average ﬂower diameter (B), each symbol (0 or I) represents the average of 50
ﬂowers. Flowers on plants from 128-cell trays were not measured. Error bars
represent 95 % conﬁdence intervals. ......................... 90

viii

SECT

 

In)

 

U)

 

SECTION IV

1. Inﬂuence of cold treatments on ﬂowering of L. angustifolia ‘Munstead’. (A)
and (B) show ﬂowering percentage for 128-cell and 50—cell plants, respectively.
(C) and (D) show time to visible bud and ﬂower for 128-cell and 50-cell plants,
respectively. Error bars represent standard error of the means. ...... 109

2. Inﬂuence of cold treatments on ﬂowering of L. angustifolia ‘Munstead’. (A) and
(B) represent number of inﬂoresences visible per plant at the time of ﬁrst ﬂower
opening for 128-cell and 50-cell plants respectively. (C) and (D) show number
of nodes present on the main stem at ﬂowering for l28-cell and 50-cell plants
respectively. (B) and (F) are total plant height at ﬂowering for l28-cell and 50-
cell plants respectively. Error bars represent standard error of the means. 111

3. Inﬂuence of forcing temperature on ﬂowering of L. angustifolia ‘Munstead’.
Each symbol (A, O, or I ) represents the mean of 10 plants except for the 24-
setpoint and 27-setpoint treatments of 128-cell plugs. Seven and four plants
ﬂowered in those treatments, respectively. Open symbols represent data not
included in regression analysis. (A), (B), and (C) show days for the indicated
developmental stage. For (D), (E), (F), solid lines (—) represent predicted
values for the rate of progress to the indicated developmental stage, based on
linear regression. Statistical analysis and calculations are presented in
Table 2. ......................................... 118

4. Inﬂuence of forcing temperature on (A) number of ﬂower buds and (B) total
plant height in L. angustifolia ‘Munstead’. Each symbol (.,I, orA) represents
the mean of 10 plants, except in the 24-setpoint and 27-setpoint treatments for
128-cell plugs, where the symbols represent the mean of 7 and 4 plants,
respectively. Error bars represent 95 % conﬁdence intervals. ........ 120

ix

SECTION I

LITERATURE REVIEW

 

 

 

 

 

harod

excep
found
thecx
annu:
temp
ternp
desc:
mid
anif
quic
Elite
65p:
ofs
The
(Ch

Dir

ind:

LITERATURE REVIEW
Introduction and deﬁnitions.

Most biological processes proceed more quickly as temperatures rise. An
exception is the promotion of ﬂowering upon exposure to low temperatures which is
found in a wide range of plant species. As early as the 18008, research revealed that
the cold temperatures of winter were the factor that enabled biennials and winter
annuals, such as many grains, to ﬂower promptly after the return of warmer
temperatures (Chouard, 1960). Early studies emphasized the inﬂuence of low
temperatures on germinating seeds. A cool, moist pretreatment of grain seeds is
described in crop production reports by J .H. Klippart from Ohio State University in the
mid 18008 (summarized by Whyte, 1948). After irnbibition and exposure to an
artiﬁcial cold treatment, winter wheat seeds planted in the spring grew and ﬂowered
quickly like the spring wheat varieties (Chouard, 1960). This strategy was used
extensively in Russia during the 1930s to hasten the development of grain crops,
especially in regions with a short growing season. “Jar” is a Russian name for the god
of spring, and the phenomenon was termed “jarovization” by the Russian Lysenko .
The English translation is “vemalization” , “vernum” meaning “spring” in Latin
(Chouard, 1960).

Direct and indirect eﬁ‘ects of low temperatures.
Environmental factors can inﬂuence plant development either directly or

inductively. Direct effects evoke the plant response during exposure to the

3

environmental condition. Inductive effects are associated with plant responses that
occur some time after exposure to the environmental condition. Exposure to low
temperature can inﬂuence ﬂowering in several different ways.

Cold temperatures promote ﬂowering in many plants by breaking the dormancy
of existing buds, when reproductive structures are present prior to the onset of low
temperatures. Insufﬁcient duration of cold eliminates or reduces the ﬂowering
response. This response is found in many spring-ﬂowering woody perennials and
ﬂowering bulbs such as T ulipa and Narcissus (Chouard, 1960; Lede and
DeHertogh, 1993). Floral primordia of Paeonia are also initiated soon after anthesis
of the current year's ﬂowers and require a minimum of four weeks at 5.6C to break
dormancy (Byme and Halevy, 1986).

Flower induction, initiation, and development occur during exposure to low
temperatures in some species. Temperatures below 11C are required for ﬂower
initiation in Brassica oleracea gongylodes L. (kohlrabi), with the optimum being
between 2 and 8C (W iebe et al. , 1992). Flower initiation and development occurred
during 3C cold treatments in brussels sprouts (Stokes and Verkerk, 1951). Low
temperatures are also required during ﬂoral initiation in Allium cepa (onion), Iris cv.
Wedgewood, and Matthiola incana (stocks) (V ince-Prue, 1975). In M. incana, it was
found that low-temperature conditions must continue until the buds are visible in order
for their development to continue upon return to warmer temperatures (Kohl, 1957).
Floral bud initiation and development occur in Astilbe during storage at 2C

(Pemberton, 1992), and in Dicentra spectabilis during storage at 5 and 10C

(Hanchek, 1989).

In contrast to these direct effects, exposure of some species to low temperatures
induces the subsequent initiation and development of ﬂoral primordia. This inductive
inﬂuence of low temperatures (often -2 to SC) is the phenomenon known as
vernalization, generally deﬁned as a cold treatment that induces or hastens the capacity
for ﬂowering in a plant (Chouard, 1960; Vince-Prue, 1975; Zeevaart, 1963). Floral
primordia are not initiated until after the return of warmer temperatures. In many
species initiation will not occur until photoperiodic requirements are met as well.
Species exhibiting a vemalization response encompass a wide range of winter annuals,
biennials, and herbaceous perennials (V ince-Prue, 1975).

While promotion of ﬂowering by cold temperatures is readily observed, often it
is not possible to categorize the nature of this promotion in the absence of deﬁnitive
knowledge on the time of ﬂower initiation. In a diverse group of plants such as
herbaceous perennials, a spectrum of mechanisms may exist which do not ﬁt into the
categories described above. Studies of ﬂowering bulbs have established that the bulbs
or corms are never completely at rest, but are physiologically active even in the
absence of visible organogenesis. A similar situation may be present in other types of
plants (LeNard and DeHertogh, 1993).

In addition, it can be difﬁcult to determine if ﬂowering was truly hastened by a
cold treatment when forcing conditions for the plants are not constant through the
course of the experiment. Temperatures and irradiance in greenhouses change as the

seasons change, and this will inﬂuence the developmental rates of plants emerging from

5

cold treatments of different durations (Baskin and Baskin, 1989; Preston et al., 1983;
White et a1. , 1989). One way to document a hastening of ﬂowering is to count the
number of leaves formed before ﬂowering, but this is often not done.

Vemalization is deﬁned as the acquisition of the ability to ﬂower, or reduction
in time to ﬂower, and a range of obligate and facultative responses exists. No
ﬂowering will occur until the qualitative or obligate requirement for a cold treatment
has been fulﬁlled in some biennial and herbaceous perennial plants, including members
of the Umbelliferae family, Hyoscyamus, Beta spp. , Brassica spp. , Campanula spp. ,
Denothera spp. , Dianthus spp., and many others (Lang, 1965; Vince-Prue, 1975).

Facultative or quantitative responses are found in annuals and herbaceous
perennials, including many cereal grains, Brassica spp. , Pisum sativum, Oenothera
spp., Dianthus spp., and many others (Lang, 1965; Vince-Prue, 1975). Effectiveness
of vernalization in these plants is measured as a reduction of leaves formed before
ﬂowering, or a reduction in days to ﬂowering, which often becomes more pronounced
as the duration of low-temperature conditions increases. Unvernalized winter rye will
ﬂower after forming 22 leaves, but in vernalized plants, ﬂowering occurs after only six
to seven leaves are formed (Purvis, 1961). Vemalization does not inﬂuence subsequent
leaf production rate in Arabidopsis, and the cold treatment hastens ﬂowering by
reducing the number of leaves formed before anthesis (Bagnall, 1993).

A range of responses can be found within a species. Different cultivars of
winter wheat reveal obligate, facultative, and neutral responses to the cold treatment

(Gardner and Barnett, 1990). Some wheat cultivars classiﬁed as spring wheats show

6

slight but signiﬁcant vernalization responses (Jedel et al., 1986). Evaluation of 32
Arabidopsis ecotypes also revealed a variety of responses (Karlsson et al., 1993). In
Lilium longiﬂorum, a qualitative requirement for low temperatures must be fulﬁlled
before a subsequent quantitative response becomes apparent (Miller, 1993).

Seed vernalization.

Different species are capable of being vemalized in many different ways. The
seeds of many species exhibiting a facultative response, such as cereal crops, Brassica
spp. , Pisum, and others, can be vemalized after they have begun imbibition. A water
content of 40-50 percent of the seed dry weight is required in cereals, 70 percent in
radish, and 60-160 percent in Arabidopsis (Chouard, 1960; Purvis, 1961; Vince-Prue,
1975). The vemalized condition generally persists through drying of the seeds in many
species, including cereal grains (Lang, 1965; Purvis, 1961).

The seeds of some species can become vemalized during development on the
mother plant. Developing ears of cereal grains including rye can be vemalized at any
stage after fertilization has occurred (summarized by Purvis, 1961). Pisum plants
were successfully vemalized while developing in pods on the maternal plant (Reid,
1979). In studies with wheat, plants with developing heads were transferred to cold
temperatures at various intervals following anthesis. The seeds from these plants were
planted, and the ﬂowering response was observed. The developing seeds were most
responsive to the cold treatment approximately 10-12 days after fertilization, during the
early stages of embryogenesis, and became less sensitive to the low-temperature

treatment as they matured (Krekule, 1987).

7

In biennials, cold treatment of the seeds does not completely fulﬁll the
vernalization requirement. Seed vernalization of Lunaria and Daucus carota (carrot),
was associated with a higher percentage of ﬂowering and more rapid ﬂowering after
subsequent plant vernalization (W ellensiek, 1965). Effectiveness of seed vernalization
in genotypes of wheat varies, with some varieties capable of being completely
vemalized as an imbibed seed, and others requiring at least some exposure to low
temperatures as a growing plant (Salisbury et al., 1979).

Requirements of the vernalization process.

In all plant tissues that have been studied, oxygen, carbohydrates, and water
must be present for vernalization to occur (Chouard, 1960). Exclusion of oxygen by
submerging cereal grain tissues in water or exposing them to a nitrogen atmosphere
during vernalization treatments delayed ﬂowering. Excised embryos of winter cereals
or radishes and meristematic tissues of carrot or cabbage respond optimally to
vernalizing temperatures only if supplied with carbohydrates (Lang, 1965; Purvis,
1961). Some level of respiratory activity appears to be necessary for vernalization.
Plant age and vernalization.

Sensitivity to vernalization varies in plants as they develop. The duration of
vernalization required to saturate the response in wheat decreases as the plants
germinate and develop to the eight- leaf stage (Wang et al., 1995). Lilium longiﬂorum
bulbs that are harvested later in the season are more responsive to vernalization,
perhaps because they are more mature (Miller, 1993). Wild-type Arabidopsis plants

responded maximally to vernalization when treatments were given to imbibed seeds,

8

and minimally when treatments were given to plants 5 - 10 days after germination. In
fca-I mutants, the phase of maximal sensitivity lasted for 25 days after germination
(Chandler and Dean, 1994).

Many biennials and perennials must attain a certain size before they are capable
of any vernalization response. These plants must complete a “ juvenile” period before
they will respond to a cold treatment. The juvenile period of Hyoscyamus niger is
approximately 10 days (Sarkar, 1958); of Lunaria, eight weeks (Zeevaart, 1963). In
Centaurea dtﬁ‘usa, 50 percent of the plants responded to vernalization at 4C after 13
leaves had formed (Thompson and Stout, 1991). Juvenility in carrot cv. Chantenay Red
Cored ended after 8-12 leaves were initiated (Atherton et al. , 1990). Length of
juvenility in Aquilegia x hybrida varied among cultivars, as did requirements for cold
treatment. To obtain 100 percent ﬂowering in 'McKana's Giant' plants, at least 12
leaves and 10 weeks of storage at 4.5C were required. For 'Fairyland,‘ 15 leaves and
only four weeks of cold storage were required for 100 percent ﬂowering to occur.
'Crimson Star' reached 100 percent ﬂowering after at least 15 leaves were formed and
eight weeks of cold storage were completed (Shedron and Weiler, 1982).

Conditions that promote growth can shorten the juvenile period. High-intensity
supplemental lighting of Lunaria plants prior to vernalization increased subsequent
ﬂowering and reduced time to ﬂower after vernalization, responses attributed to
increased photosynthesis (W ellensiek, 1965; Zeevaart, 1963). Timing of ﬂowering in
some facultative biennials may be related to the accumulation of a critical amount of

food reserves. However, work with Oenothera erythrosepala showed that ﬂowering

9

would not occur until the diameter of the rosette exceeded 9 centimeters, regardless of
chronological age or dry weight of the taproot. In this species, a critical leaf area
seems to be necessary to perceive inductive photoperiods after the vernalization
requirement has been met (Kachi and Hirose, 1983).

Temperature.

Vemalization is a cumulative process, the effectiveness of which depends on the
temperature experienced and the duration of the vernalization period. As in many
biochemical reactions, an optimum temperature range exists (often -2 to SC), and
higher or lower temperatures are less effective. In Secale cereale (winter rye),
exposure to any temperatures between 1 and 7C will be equally effective (Purvis,
1961). Optimum temperatures in wheat differ among cultivars, ranging from 0 to 15C
(Berry et al. , 1986). In carrots, rates of bolting and ﬂowering increased linearly with
vernalization temperatures from —1 to 6.5C and decreased linearly as temperatures
increased from 7 to 16C (Atherton et al. , 1990). Base, optimum, and maximum
vernalization temperatures were determined to be -1, 6.5, and 16C, respectively
(Atherton et al. , 1990). Studies of Lilium longiﬂorum showed that temperatures below
21C can fulﬁll the vernalization requirement of the bulbs (Weiler and Langhans, 1968).
Exposure to temperatures of 2 to 10C accelerates ﬂowering and decreases the number
of leaves and ﬂowers formed (Miller, 1993). In Hyoscyamus niger, temperatures
between 0C and 17C are effective but the optimum temperature for vernalization varies
with the duration of treatment. The optimum temperature was 10C when treatments

lasted 7-15 days, and 3C to 6C for treatments of 42 days. (Lang, 1965). The

10

vernalization response becomes saturated after a particular duration of cold which
ranges from 10 to 100 days between species (summarized by Thomas and Vince-Prue,
1984).

Altemation of vernalization temperatures with moderately warmer temperatures
does not diminish the vernalization response already accumulated, and will enhance the
response in some species (Chouard, 1960; Wellensiek, 1964). Wheat plants can be
vemalized while growing under cool night temperatures with a 20C day temperature,
although the inﬂuence of low temperatures on overall development rate confounded the
response in some genotypes (Berry et al., 1986).

Devemalization.

High temperatures during or soon after vernalization can reduce or reverse the
inductive effect. This process is termed devemalization and, like vernalization, is a
cumulative process that is inﬂuenced by the temperature experienced and the duration
of the exposure. The ranges of inductive and devemalizing temperatures can be very
close together, as in some winter cereals in which only a narrow range of temperatures
(12 to 18C) has neither vernalizing nor devemalizing effects (Lang, 1965). Typically,
temperatures approaching 30C or higher result in devemalization (Thomas and Vince-
Prue, 1984). Flower stalks on Cheiranthus cheiri plants vemalized at 2 or 5 C
reverted to vegetative growth when plants were transferred to 22 C, and aborted when
transferred to 32 C (Diomaiuto-Bonnand, 1972). Devemalization by short-day
conditions has been described in Beta and Oenothera biennis (V ince-Prue, 1975).

Many plants become progressively less susceptible to devemalization as the

11

duration of vernalizing conditions increases. An interval of moderate temperatures
between the low inductive temperatures and the high devernalizing temperatures also
tends to stabilize the vemalized state (Lang, 1965). Interactions between light levels
and stability of vernalization have also been described. In Chrysanthemum, high-
temperature treatments did not devernalize basal shoots, but complete devemalization
was induced by 8 weeks exposure to low intensity illumination (25 foot candles, 5.0
umol m‘zs") (Schwabe, 1955). In Arabidopsis, exposure to light during either
vernalizing or devernalizing conditions reduces devemalization (N app-Zinn, 1984).
However, red or far-red irradiation during devemalization of winter rye did not
signiﬁcantly affect the amount of devemalization (Friend, 1965).

Site of vernalization.

Generally the apical meristem is considered the site of perception of vernalizing
temperatures. Experiments with celery, beet, and Chrysanthemum showed that
ﬂowering only occurred when the growing point was chilled (V ince-Prue, 1975).
Localized temperature treatments conﬁrmed that the shoot tip is the site of perception
in Ihlaspi arvense plants (Metzger, 1988). Warming the roots of Chinese cabbage
plants had no inﬂuence on the effectiveness of vernalization (Rietze and Wiebe, 1988).

Other experiments suggest that dividing cells anywhere in the plant can be
vemalized. Leaf and root cuttings of Lunaria biennis were given cold treatments, and
plants regenerated from these cuttings ﬂowered. When the basal 0.5 centimeter of the
petiole was removed from the leaf cuttings, ﬂowering did not occur. Cells in this

region were regenerating and apparently were the site of perception of the cold

 

 

 

ten

Clll

pre

pla

ma
the
grc
ﬂo
sht
(13}

for

C01:

[W

12

temperatures. When intact plants were vemalized, leaf cuttings from younger leaves
that were expanding during the cold treatment were vemalized much more effectively
than those from mature leaves. Vemalization was most effective when cells were
dividing during the cold treatment (W ellensiek, 1964).

The presence of actively dividing cells may not be necessary for vernalization in
Ihlaspi arvense. The shoot tip was the site of perception in this species; however, leaf
cuttings of mature leaves from vemalized plants developed ﬂowering shoots. The
presence of dividing cells during the cold treatment was minimized by treating intact
plants, using mature leaves, and removing the basal 1 to 2 centimeters of the petiole.
Many cell types may be capable of being vemalized in Ihlaspi arvense (Metzger,
1988).

In some species, the vemalized state can be very stable and perpetuated through
many mitotic divisions. Plants vemalized as embryos of peas, wheat, and rye maintain
their thermoinduced state through maturation of the seed, germination, and subsequent
growth of the plant ( Krekule, 1987; Reid, 1979). Vernalized Hyoscyamus plants
ﬂowered promptly when exposed to long days even after more than 190 days under
short-day conditions. No decrease in ﬂowering response occurred until more than 300
days after the end of the cold treatment. All leaves present after 190 days had been
formed after vernalization (Lang, 1986). Cells that are vemalized apparently transmit
this condition to cells that arise from them mitotically. Vemalization is generally not
considered to be maintained through a meiotic division, but experiments with peas

revealed a small but signiﬁcant effect on the progeny of vemalized parents (Reid,

 

 

 

 

 

P0

du
gn
u?
bu
for

pr:

V8

C0

Vei

Pm

dex

13

1979).
Polycarpic or perennial plants.

Annuals and biennials are monocarpic plants that ﬂower once and then die. In
contrast to the considerable persistence of the vemalized state in some species,
perennial or polycarpic plants must be revernalized each year if they are to bloom.
Three hypotheses that can explain this requirement have been proposed.

The ﬁrst hypothesis is based on the observation that some perennial plants grow
vegetatively in a rosette form, and ﬂowers are formed on long stems above the rosette.
Some of the buds on the crown of these plants apparently do not become induced
during natural vernalization. In Geum urbanum, the terminal bud produces vegetative
growth, while axillary buds form the inﬂorescence. Under natural conditions, only
axillary buds of a certain maturity are capable of being vemalized, while the terminal
bud never differentiates into reproductive growth. If these plants are artiﬁcially chilled
for 40-50 weeks, eventually all buds including the terminal will become vemalized and
produce ﬂowers, and the plant will die. In other perennials, the terminal bud becomes
vemalized but axillary buds do not (Chouard, 1960; Vince-Prue, 1975).

The second hypothesis suggests that warm temperatures or other environmental
conditions during the summer months may cause some buds to devernalize (Chouard,
1960). Buds located in the basal portion of Chrysanthemum plants were found to be
vemalized in the spring, but not vemalized at the end of the summer (Schwabe, 1955).
Prolonged exposure to low intensity illumination (25 footcandles, 5 .0 umol m‘zs")

devernalized the basal shoots in Chrysanthemum plants, so low light levels under the

l4

plant canopy may inﬂuence the process (Schwabe, 1955).

The third hypothesis suggests that the vemalized state is not always transmitted
indefinitely to all cells formed by a vemalized plant. While shoots arising from
vemalized buds also are vemalized, the condition is sometimes not transmitted to the
buds formed on those shoots (Chouard, 1960).

Interactions between vernalization and photoperiod.

Interactions between vernalization requirements and photoperiodic responses are
varied and complex. Some plants respond most strongly to vernalization if they
experience short days before the cold treatment, while others prefer long days or
continuous light (reviewed by Napp-Zinn, 1984).

Exposure to certain daylengths during vernalization can modify the
effectiveness of the cold treatment. Long days, short days plus night interruptions, or
continuous light during vernalization can increase the vernalization effect in some
species and decrease it in others (reviewed by Napp-Zinn, 1984). In seed-vemalized
Arabidopsis, vernalization was more effective in darkness than in either eight-hour or
24-hour days, provided with ﬂuorescent lamps (Chandler and Dean, 1994). In Poa
pratensis and celery, long—day conditions during vernalization inhibited induction,
although long days after vernalization promote ﬂowering (V ince-Prue, 1975; Thomas
and Vince-Prue, 1984). In winter wheat, some experiments have demonstrated that
vernalization was more effective under long days than short days when irradiance
levels were low (approximately 2000 lx). The inﬂuence of photoperiod was reduced

when light intensity was increased, suggesting that a certain amount of carbohydrates

15

must be available for vernalization to take place (Krekule, 1961; Krekule, 1987).
Increasing light intensity during vernalization increased the percentage of 10—week old
Cardamine pratensis plants ﬂowering, but had little inﬂuence on ﬂowering percentage
in 38-week old plants (Pierik, 1967).

Little information is available about the inﬂuence of light quality during
vernalization. In Cardamine pratensis, long days during the cold treatment promoted
ﬂowering slightly when provided by incandescent lights, but not when ﬂuorescent
lights were used (Pierik, 1967). Red radiation delivered during vernalization of winter
rye accelerated ﬂowering slightly in several experiments, and far-red accelerated
ﬂowering slightly after 1, 3, or 4 weeks of irradiation. Radiation during vernalization
did not inﬂuence ﬂowering when given during treatments that fully vemalized the grain
(Friend, 1965).

Many plants that respond to vernalization are long-day plants. Vemalization of
the seed for 8 weeks lowered the critical photoperiod in spinach (var. Nobel) from 14
to 8 hours (V litos and Meudt, 1955). A similar response has been found in a number
of herbaceous perennials, including Iris (Buxton and Mohr, 1969), Dicentra spectabilis
(Lopes and Weiler, 1977), Echinops 'Taplow Blue', Achillea millefolium 'Rosea', and
Physostegia virginiana 'Summer Snow' (Iversen, 1989), Chrysanthemum x superbum
Bergmans (Shedron, 1980), and Lavandula angustifolia and Lobelia speciosa (Engle,
unpublished data; Runkle, unpublished data). Lang (1965) has listed some
modiﬁcations of photoperiodic requirements in a number of species.

Unvernalized late-ﬂowering mutants of Arabidopsis showed a strong response

16

to light quality. Under a 16- hour photoperiod, time to ﬂowering was doubled when
ﬂuorescent lamps were used, compared to ﬂuorescent plus incandescent lamps. After
vernalization, this response to light quality was largely eliminated. Studies with other
ecotypes also showed that vernalization reduces the response to light quality and
photoperiod (Bagnall, 1993).

In some cold-requiring long-day plants such as Campanula medium, short-day
treatments can replace vernalization prior to the long-day conditions. The mechanisms
of short-day and vernalization induction appear to be different because the young plants
became sensitive to short days one month sooner than to vernalization. In addition,
fewer long days were required for ﬂoral initiation when following short days than when
following vernalization (Wellensiek, 1960). Coreopsis grandiﬂora Nutt. has also been
classiﬁed as a short-long-day plant, in which vernalization can replace the short-day
treatment. Seven weeks of vernalization was somewhat more effective than 7 weeks of
short days in terms of ﬂowering percentage (Ketellapper and Barbaro, 1966). Long
days can substitute for quantitative vernalization responses in Lilium longiﬂorum once
the initial qualitative requirement for two weeks of temperatures below 21C has been
satisﬁed (Miller, 1993).

Light quality during long days inﬂuences the ability of photoperiod to substitute
for vernalization in Arabidopsis. Time to ﬂowering and leaf number at ﬂowering
increase as the ratio of redzfar red light increases. Leaf production rate is not
inﬂuenced by light quality (Bagnall, 1993).

Mechanisms.

17

Vemalization is an unusual phenomenon because it occurs more rapidly as
temperatures decrease, unlike most other biological processes. Research has yielded
descriptive information about the response and suggested a variety of hypotheses, but
uncovered little information about the biochemical basis of vernalization. No
substances that form during vernalization and induce ﬂowering have yet been isolated.
Analysis of unvemalized, partially vemalized, and fully vemalized wheat seeds showed
that changes in rRNA synthesis occur during the cold treatment. An rRNA that is
synthesized at low temperatures but decomposes at 25C was detected (Paldi and Devay,
1983).

Genetics.

Genes that inﬂuence vernalization have been identiﬁed in several species. In
wheat, at least ﬁve loci may be involved, and there are also reports of extranuclear
inheritance. Genes inﬂuencing photoperiodic responses are entirely separate from the
vernalization genes. The situation is very complex since wheat is a hexaploid. In
Pisum, six genes are implicated in ﬂowering, and two of these are responsible for both
photoperiodic and vernalization requirements. In Arabidopsis, up to 24 genes may be
involved (Napp-Zinn, 1987). In Secale cereale, only one gene is responsible for the
vernalization requirement, which results in a clear distinction between spring and
winter rye (Purvis, 1939). One gene also determines the vernalization response in
Hyoscyamus niger (Lang, 1986).

Researchers have suggested that the interaction of competing processes may

control vernalization. If the optimum temperatures differ for the inhibitory and

18

promotive processes, there may be certain temperatures at which levels of promoters
are greater than those of inhibitors (Lang, 1965). More recent work with peas has
suggested that a low temperature inﬂuence on gene expression may play a part.
Exposure to low temperatures or continuous light appears to repress activity of a gene
(Sn) that may encode a ﬂowering inhibitor. Other genes encode a promoter, and
inﬂuence the ratio of inhibitor to promoter. Vemalization may also change the
sensitivity of the shoot apex to the inhibitor or promoter (Reid and Murfet, 1975).
Similar mechanisms have been suggested for Lathyrus odoratus. When one of two Dn
alleles are present, a substance that delays ﬂowering is produced. After vernalization,
ﬂowering is promoted. Vemalization also appears to alter the sensitivity of the shoot
apex to ﬂoral initiation. Plants with only dn alleles show some vernalization response,
suggesting that a ﬂoral promoter may be involved as well (Ross and Murfet, 1986).
The control of gene expression is not well understood, but patterns of DNA
methylation appear to play a role. Methylation of cytosines in the promoter region of a
gene may be associated with a lack of transcription of that gene. In Arabidopsis and
cell cultures of Nicotiana plumbaginifolia, cold treatments resulted in a substantially
reduced level of methylation. Treatment with the demethylating agent 5-azacytidine
hastened ﬂowering in Thlaspi arvense and late-ﬂowering ecotypes and mutants of
Arabidopsis, but had no effect on ﬂowering in mutants that do not respond to
vernalization (Burn et al. , 1993). Gamma rays are a presumed demethylating agent,
and in winter wheat, exposure to 5-azacytidine or to gamma rays partially substituted

for the cold treatment and signiﬁcantly hastened ﬂowering (Brock and Davidson,

19

1994).
Grafting experiments.

In an effort to determine if a graft-transmissible substance is produced during
vernalization, grafting experiments have been performed with a number of species
(Lang, 1965). The long—day, cold-requiring variety of henbane has been used
extensively. Nonvernalized henbane plants were induced to ﬂower under long-day
conditions when vemalized plants were grafted onto them, suggesting that a ﬂower-
promoting substance termed "vernalin" was produced during the cold treatment. The
vemalized plants were photoinduced and ﬂowering when the grafts were made. The
majority of earlier successful grafting experiments were performed under conditions in
which the donor tissue was already ﬂowering, so it is possible that other ﬂower-
inducing substances that were not associated with vernalization were present (Chouard,
1960; Lang, 1965). One set of experiments with the short-day tobacco 'Maryland
Mammoth' , which does not require a cold treatment, showed that even under
noninductive photoperiods, 'Maryland Mammoth' donors did overcome the cold
requirement in henbane (Lang, 1965). Experiments with Lathyrus odoratus grown
under noninductive photoperiods demonstrated that vernalization reduced the
production of a graft-transmissible substance that inhibits ﬂowering (Ross and Murfet,
1986). Similar results were obtained in Pisum (Reid and Murfet, 1975). Grafting
experiments have been unsuccessful in many cases, however, and vernalization seems
only to affect a small region of the apex in Geum and Chrysanthemum morifolium

(Thomas and Vince-Prue, 1984). Whether the vemalized state leads to the production

 

It

to

ch

Cl"

Ar

30

Ol l.

20

of a mobile substance remains unclear (Burn et al., 1993).
Gibberellins.

Gibberellins have been associated with vernalization, but the nature of the
relationship is not clear. Levels of gibberellins were found to be higher after
vernalization in Chrysanthemum, Althaea, Raphanus, and winter wheat (V ince-Prue,
1975). Analysis of Thlaspi arvense showed that the level of one gibberellin-like
substance was dramatically inﬂuenced by low temperatures, but could not be causally
related to thermoinduced stem elongation (Metzger, 1985). Subsequent studies have
found that vernalization induced the activity of an enzyme catalyzing the conversion of
the gibberellin precursor kaurenoic acid to 7a-OH kaurenoic acid. This thermoinduced
change in gibberellin metabolism occurred only in the shoot tip, which is the site of
perception of vernalization in this species (Hazebroek et al. , 1993). Preliminary
evidence indicates that a correlation between vernalization and gibberellins exists in
Arabidopsis (Burn et al. , 1993). In Lunaria, no accumulation of gibberellin-like
material during vernalization was found (Zeevaart, 1968).

Applications of exogenous gibberellins can induce ﬂowering in some cold-
requiring plants grown under long-day conditions such as Hyoscyamus niger (Sarkar,
1958). Applications of gibberellic acid could not substitute for vernalization in
Coreopsis grandiﬂora, but could replace the requirement for long days (Ketellapper
and Barbaro, 1966). In many cases, gibberellin treatments cause stem elongation, but
no ﬂowering. In some species, gibberellin treatments have no inﬂuence on ﬂowering

or inhibit the process. A table summarizing these responses has been published by

 

 

 

 

 

 

ﬁe

21

Lang (1965). In Thlaspi arvense, GA3 treatments completely substituted for
thermoinduced stem growth. In addition, application of a gibberellin biosynthesis
inhibitor completely inhibited the elongation response (Metzger, 1985). Gibberellins
may be associated more closely with elongation in response to cold rather than with
ﬂowering (Thomas and Vince-Prue, 1984).

Induction by application of gibberellins differs from environmental induction in
several ways. When temperature or photoperiod induces ﬂowering, shoot elongation
and ﬂoral initiation occur simultaneously; but when gibberellins are used to induce the
process, elongation occurs before initiation. In species that can be vemalized as
imbibed seeds, gibberellin treatment is not effective as a substitute for low temperatures
(Vince-Prue, 1975).

Clearly, ﬂowering mechanisms are very complex. The enormous variety of
reproductive strategies that plants have evolved reﬂects the wide range of environments
they inhabit. Types of vernalization response vary within a species and sometimes
correspond to patterns of climatic conditions. Varieties of wheat native to subtropical
areas show very different vernalization responses than those from northern areas
(Krekule, 1987). Often requirements for vernalization are coupled with a response to
long days. A requirement for vernalization before ﬂowering ensures that mature plants
will ﬂower at the time of year when conditions are most favorable for development and
maturation of the seeds. Without this mechanism, seedlings might begin reproductive
development during the same summer as their germination and ﬂowering might occur

in the fall when insufﬁcient time remains for the seeds to develop.

 

 

Ba

Br.

Bu

LITERATURE CITED

Atherton, J ., J. Craigon,and E.Basher. 1990. Flowering and bolting in carrot. I.
Juvenility, cardinal temperatures, and thermal times for vernalization. J. Hort.
Sci. 65:423-429.

Bagnall, D. 1993. Light quality and vernalization interact in controlling late ﬂowering
in Arabidopsis ecotypes and mutants. Ann. Bot. 71:75-83.

Baskin, J. and C. Baskin. 1989. Ec0physiology of seed germination and ﬂowering in
Liatris squarrosa. Bull.Torrey Bot. Club 116:45-51.

Berry, G., P. Salisbury, and G. Halloran. 1986. Wheat lines: effects of night
temperature. Ann. Bot. 58:523-529.

Brock, R. and J. Davidson. 1994. 5-Azacytidine and gamma rays partially substitute
for cold treatment in vernalizing winter wheat. Env. Exper. Bot. 34: 195-199.

Burn, J., D. Bagnall, J. Metzger, E. Dennis, and W. Peacock. 1993. DNA
methylation, vernalization, and the initiation of ﬂowering. Proc. Natl. Acad.
Sci. USA 90:287-291.

Buxton, J. and H. Mohr. 1969. The effect of vernalization and photoperiod on
ﬂowering of tall bearded irises. HortScience 4:53-55.

Byme, T. and H. Halevy. 1986. Forcing herbaceous peonies. J. Amer. Soc. Hort. Sci.
111 :379-383.

Chandler, J. and C. Dean. 1994. Factors inﬂuencing the vernalization response and
ﬂowering time of late ﬂowering mutants of Arabidopsis thaliana (L.) Heynh. J.
Exp. Bot. 45: 1279-1288.

Chouard, P. 1960. Vemalization and its relations to dormancy. Annu. Rev. Plant
Physiol. 11:191-238.

Diomaiuto-Bonnand, J. 1972. Réversions de l’état reproducteur a l’état végétatif, chez

la Giroﬂée (Cheiranthus cheiri L. , Cruciferes), sous l’action de températures
moyennes ou élevées. CR. Acad. Sci. Ser. D. Sci. Nat. (Paris) 274:372-375.

22

 

Ge

H2

It:

It:

Kc

Kt

Lil

23

Friend, D. 1965. Interaction of red and far-red radiations with the vernalization process
in winter rye. Can. J. Bot. 43:161-170.

Gardner, F. and R. Barnett. 1990. Vemalization of wheat cultivars and a triticale.
Crop Sci. 30:166-169.

Hanchek, A. 1989. Growth and development of Dicentra spectabilis in relation to
storage temperature and harvest date. PhD Diss. Michigan State Univ. , East
Lansing.

Hazebroek, J ., J. Metzger, and E. Mansager. 1993. Thermoinductive regulation of
gibberllin metabolism in Thlaspi arvense L. Plant Physiol. 102:547-552.

Iversen, R. 1989. Greenhouse forcing 'of herbaceous garden perennials. PhD Diss.
Cornell Univ. , Ithaca.

Jedel, P., L. Evans, and R. Scarth. 1986. Vemalization responses of a selected group
of spring wheat (T riticum aestivum L.) cultivars. Can. J. Plant Sci. 66:1-9.

Kachi, N. and T. Hirose. 1983. Bolting induction in Oenothera erythrosepala Borbas
in relation to rosette size, vernalization, and photoperiod. Oecologia 60:6-9.

Karlsson, B., G. Sills, and J. Nienhuis. 1993. Effects of photoperiod and vernalization
on the number of leaves at ﬂowering in 32 Arabidopsis thaliana (Brassicaceae)
ecotypes. Amer. J. Bot. 80:646-648.

Ketellapper, H. and A. Barbaro. 1966. The role of photoperiod, vernalization and
gibberellic acid in ﬂoral induction in Coreopsis grandiﬂora Nutt. Phyton 23:33-
41.

Kohl, H. 1957. Flower initiation of stocks grown with several temperature regimens.
Proc. Amer. Soc. Hort. Sci. 72:481-484.

Krekule, J. 1961. The effect of photoperiodic regime on vernalization of winter wheat.
Biol. Plantarum 3:180-191.

Krekule, J. 1987. Vemalization in wheat, p. 159-169. In: J .G. Atherton (ed).
Manipulation of ﬂowering. Butterworths, London.

Lang, A. 1965. Physiology of ﬂower initiation, p. 1380-1536. In: W. Ruhland (ed.).
Handbuch der Pﬂanzenphysiologie. Springer-Verlag, Berlin.

Lang, A. 1986. Hyoscyamus niger, p. 144-186. In: A. Halevy (ed.). CRC handbook
of ﬂowering. CRC Press, Boca Raton.

24

LeNard, M. and A. DeHertogh. 1993. Bulb growth and development and ﬂowering, p.
29-43. In: A. DeHertogh and M. Lede (eds.). The physiology of ﬂower
bulbs. Elsevier Science Publishers, Amsterdam.

Lopes, L. and T. Weiler. 1977. Light and temperature effects on the growth and
ﬂowering of Dicentra spectabilis (L.) Lem. J. Amer. Soc. Hort. Sci. 102:388-
390.

Metzger, J. 1985. Role of gibberellins in the environmental control of stem growth in
Thlaspi arvense L. Plant Physiol. 78:8-13.

Metzger, J. 1988. Localization of the site of perception of thermoinductive
temperatures in Thlaspi arvense L. Plant Physiol. 88:424-428.

Miller, W. 1993. Lilium longiﬂorum, p. 391-422. In: A. DeHertogh and M. Lede
(eds.). The physiology of ﬂower bulbs. Elsevier Science Publishers,
Amsterdam.

Napp-Zinn, K. 1984. Light and vernalization, p. 75-88. In: D. Vince-Prue, B.
Thomas, K. Cockshull (eds.). Light and the ﬂowering process. Academic
Press,London.

Napp—Zinn, K. 1987. Vemalization - environmental and genetic regulation, p. 123-
132. In: J.G. Atherton (ed.). Manipulation of ﬂowering. Butterworths,
London.

Purvis, O. 1939. Studies in vernalization of cereals. V. The inheritance of the spring
and winter habit in hybrids of Petkus rye. Ann. Bot., 3:719-729.

Purvis, O. 1961. The physiological analysis of vernalization, p. 76-122. In: W.
Ruhland (ed.). Encyclopedia of Plant Physiology volume 16. Springer-Verlag,
Berlin.

Paldi, E. and M. Devay. 1983. The changes in the nature of rRNA synthesis in the
plumula of seedlings during vernalisation. Plant Sci. Lett. 30:69-75.

Pemberton, G. 1992. Factors affecting the growth and development of forced Dutch-
grown Astilbe. MS Thesis, North Carolina State Univ., Raleigh.

Pierik, R. 1967. Effect of light and temperatue on ﬂowering in Cardamine pratensis L.
Z. Pﬂanzenphysiol. 56:141-152.

25

Preston, M., J. Buston, R. Anderson, and H. Mohr. 1983. Effect of vernalization
duration and storage method on forcing of tall bearded irises. HortScience
18:455-456.

Reid, J. 1979. Flowering in Pisum: effect of the parental environment. Ann. Bot.
44:461-467.

Reid, J. and I. Murfet. 1975. Flowering in Pisum: the sites and possible mechanisms
of the vernalization response. J. Expt. Bot. 26:860-867.

Rietze, E. and H. Wiebe. 1988. The inﬂuence of soil temperature on vernalization of
Chinese cabbage. J. Hort. Sci. 63:83-86.

Ross, J. and I. Murfet. 1986. The mechanism of action of vernalization in Lathyrus
odoratus L. Ann. Bot. 57:783-790.

Salisbury, P., G. Berry, and G. Halloran. 1979. Expression of vernalization genes in
near-isogenic wheat lines: methods of vernalization. Can. J. Genet. Cytol.
21:429-434.

Sarkar, S. 1958. Versuche zur Physiologic der Vernalisation. Biol. Zentralbl. 77:1-49.

Schwabe, W. 1955. Factors controlling ﬂowering in the chryanthemum. J. Exp.Bot.
6:435450.

Shedron, K. 1980. Regulation of growth and ﬂowering of four herbaceous perennials:
Aquilegia x hybrida Sims, Aurinia saxatilis (L.) Desv., Chrysanthemum x
superbum Bergmans, and Lupinus spp. ‘Russell.' MS Thesis, Purdue Univ.,
Lafayette.

Shedron, K. and T. Weiler. 1982. Regulation of growth and ﬂowering in Aquilegia x
hybrida Sims. J. Amer. Soc. Hort. Sci. 107:878-882.

Stokes, P. and K. Verkerk. 1951. Flower formation in brussels sprouts. Meded.
Landbouwhogeschool, Wageningen 50: 143-160.

Thomas, B. and D. Vince-Prue. 1984. Juvenility, photoperiodism and vernalization,
p. 408-439. In: M. Wilkins (ed.). Advanced plant physiology. Pitrnan, London.

Tompson, D. and D. Stout. 1991. Duration of the juvenile period in diffuse knapweed
(Centaurea diﬁ‘usa). Can. J. Bot. 69:368-371.

Vince-Prue, D. 1975. Photoperiodism in plants. McGraw-Hill, Maidenhead.

26

Vlitos, A. and W. Meudt. 1955. Interactions between vernalization and photoperiod in
spinach. Contr. Boyce Thompson Inst. 18:159-166.

Wang, S., R. Ward, J. Ritchie, R. Fischer, and U. Schulthess. 1995. Vemalization
in wheat. I. A model based on the interchangeability of plant age and
vernalization duration. Field Crops Res. 41 :91-100.

Weiler, T. and R. Langhans. 1968. Determination of vernalizing temperatures in the

vernalization requirement of Lilium longiﬂorum (Thunb.) cv ‘Ace.' Proc.
Amer. Soc. Hort. Sci. 93:623-629.

Wellensiek, S. 1960. Flower formation in Campanula medium. Meded.
Landbouwhogeschool, Wageningen 60: 1-18.

Wellensiek, S. 1964. Dividing cells as the prerequisite for vernalization. Plant Physiol.
39:832-835.

Wellensiek, S. 1965. Recent developments in vernalization. Acta Bot. Neerl. 14:308-
314.

White, J ., D. Beattie, and E. Holcomb. 1989. Flowering studies with Aquilegia
cultivars. Acta Hort. 252:219-226.

Whyte, R. 1948. History of research in vernalization, p. 1-38. In: A. Mumeek and R.
Whyte (eds.). Vemalization and photoperiodism. Chronica Botanica
Co.,Waltham.

Wiebe, H., R. Habegger, and H. Liebig. 1992. Quantiﬁcation of vernalization and
devemalization effects for kohlrabi (Brassica oleracea convar. acephala var.
gongylodes L.) Sci. Hort. 50:11-20.

Zeevaart, J. 1963. Climatic control of reproductive development, p. 289-310. In: L.
Evans (ed.). Environmental control of plant growth. Academic Press, New
York.

Zeevaart, J. 1968. Vemalization and gibberellins in Lunaria annua L., p. 1357-1370.
In: F. Wightrnan and G. Setterﬁeld (eds.). Biochemistry and physiology of
plant growth substances. The Runge Press, Ltd., Ottawa.

SECTION II

INFLUENCE OF THE SPECTRAL QUALITY OF DAYLENGTH EXTENSIONS
ON FLOWERING OF CAMPANULA CARPATICA 'BLUE CLIPS', COREOPSIS
GRANDIFLORA 'EARLY SUNRISE', COREOPSIS VERTYCIILATA 'MOONBEAM',

AND R UDBECKIA F ULGIDA 'GOLDSTURM'

 

 

 

 

 

 

C2

.4:

7‘?

U5

Inﬂuence of the spectral quality of daylength extensions on ﬂowering of
Campanula carpatica 'Blue Clips' , Coreopsis grandtﬂora 'Early Sunrise' , Coreopsis

verticillata 'Moonbeam' , and Rudbeckia fulgida 'Goldsturm' .

Catherine M. Whitman‘, Royal D. Heinsz, Arthur Cameronz, and William Carlson2

Department of Horticulture, Michigan State University, E. Lansing, MI 48824

Additional index words. long-day plants, herbaceous perennials, photoperiod, redzfar-

red ratio.

Received for publication . We gratefully acknowledge funding by
Michigan Agricultural Experiment Station, Henry Mast Greenhouses, and other
greenhouse growers supportive of Michigan State University ﬂoriculture research. The
use of trade names in this publication does not imply endorsement of the products
named or criticism of similar ones not mentioned.

1 Graduate student.

2 Professor of Horticulture.

28

29

Abstract. The effectiveness of cool white ﬂuorescent (CWF), high-pressure sodium
(HPS), incandescent (INC), and metal halide (MH) lamps in inducing ﬂowering in
Campanula carpatica Jacq. ‘Blue Clips’, Coreopsis grandiﬂora Hogg ex Sweet.‘Early
Sunrise’, Coreopsis verticillata L. ‘Moonbeam’, and Rudbeckia fulgida
Ait.‘Goldsturm’ was compared. For each light source, lighting was delivered as a 7-hr
day extension with PPF ranging from 0.05 to 2.0 umol m‘zs“. Threshold irradiance
values for ﬂowering ranged from<0.05 to 0.37 umol m'ZS'l , depending on species.
Saturation irradiance values for C. carpatica ‘Blue Clips’ and C. grandiﬂora ‘Early
Sunrise’ were between 0.22 and 0.68 umol m'zs" for all lamps tested. An irradiance of
1.0 pmol m‘zsl was adequate for ﬂowering in all species. Time to ﬂower at irradiances
above the saturation points did not differ signiﬁcantly between lamp types in all species
tested. C. carpatica ‘Blue Clips’ and C. grandiﬂora ‘Early Sunrise’ plants under INC
lamps were signiﬁcantly taller than those in any other treatment. C. verticillata
‘Moonbeam’ plants grown under CWF lamps were approximately 10% taller than those
grown under HPS and INC. The heights of R. fulgida ‘Goldsturm’ plants did not

differ signiﬁcantly between light treatments.

30

The popularity of herbaceous perennials has increased in recent years. Combined
wholesale, retail, and landscape sales in 1992 were estimated by the Perennial Plant
Association at $222 million, with 81% of respondents reporting increases over 1991
levels (Rhodus, 1993). Traditionally perennials have been produced outdoors, and
many are not in ﬂower in the spring when the majority of garden plants are sold
(Armitage, 1994a; Schwarze, 1993). A plant in bloom has much greater sales appeal
than a non-ﬂowering plant. Consequently, producers of herbaceous perennials are
interested in production methods which will allow them to produce ﬂowering plants at
any season of the year.

Several popular species of herbaceous perennials are long-day plants (LDP), so
photoperiods exceeding a certain minimum length either are required for ﬂowering or
hasten ﬂowering (Beattie and German, 1985; Iversen and Weiler, 1994). Long days
can be provided by extending the day length to 14 to 16 hours, or by providing a 4-
hour night break (Armitage, 1994b; Vince-Prue and Canham, 1983).

Traditional photoperiodic lighting involves the use of incandescent lamps (INC)
(Vince-Prue and Canham, 1983). Incandescent lights are effective, inexpensive, and
simple to install; however their output in the 400-700 nm waveband is only 57-69
milliwatts per watt (=6%) of input power compared to 204 mW, 261 mW, and 265
mW (z20,26, and 27%) for cool white ﬂuorescent (CWF), high pressure sodium

(HPS), and metal halide (MH) lamps respectively (Thimijan and Heins, 1983). Use of

31

these other light sources in place of INC can reduce photoperiodic lighting energy
costs.

The output of incandescent lamps is rich in far-red light, (700-800 nm), the
ratio of red (600-700 nm):far-red being less than one (Lane et al. , 1965). Far-red
light has several horticulturally undesirable effects on plant morphology including the
promotion of stem elongation, and suppression of lateral branching (Downs et al. ,
1958; Moe and Heins, 1990; Vince-Prue and Canham, 1983). Several other lamp
types, such as CWF, HPS, and MH, deliver proportionately less far-red than red light
compared to incandescent lamps (deGraaf-van der Zande and Blacquiére, 1992;
Thirnijan and Heins, 1983). The use of photosynthetic lighting with lamps other than
INC offers the potential for ﬂower induction with less stem elongation.

Lighting used to extend the photoperiod must be of adequate intensity
(Summerﬁeld and Roberts, 1987; Vince-Prue, 1975). Each LDP has a threshold
irradiance below which ﬂowering is not hastened or induced by the light treatment, and
a saturation irradiance above which the time to ﬂower becomes constant (Summerﬁeld
and Roberts, 1987). Between these two light levels, the rate of progress to ﬂowering
(inverse of time to ﬂower) for a LDP is often linearly related to the irradiance
(Summerﬁeld and Roberts, 1987). For incandescent lighting, the threshold irradiance
for ﬂoral induction in many species of LDP lies between 1 lux (0.1 footcandle or 0.02
umol m'zs“) (400-700nm) and 50 lux (5 footcandles or 1 um'zs") (Vince-Prue, 1975).
For example, the threshold irradiance of incandescent lighting for ﬂoral induction in

Callistephus chinensis Cass. was reported as 0.1 footcandle (0.02 umol m'zs")

32

(W ithrow and Benedict, 1936). In Cicer arietinum L., the threshold irradiance and
saturation irradiance values under incandescent lamps are 1 to 2 lux and 3 to 8 lux
(0.02 to 0.04 umol m’zs‘l and 0.06 to 0.16 umol mas"), respectively (Summerﬁeld and
Roberts, 1987) while in Lens culinaris L. the values are 60 to 70 lux and 720 to 1150
lux (1.2 to 1.4 umol m‘zs‘1 and 14.4 to 23 umol m‘zs") (Summerﬁeld et al., 1984).
Flowering response, determined by ﬂower number, was saturated in Gypsophila
paniculata L. at 60 to 100 lux (1.2 to 2.0 pmol m‘zs“) under incandescent lamps (Shillo
and Halevy, 1982). Horticulturally, determination of threshold and saturation
irradiances is important to avoid inadvertant induction in nearby greenhouse sections,
and so that uniform ﬂowering will be induced without providing surplus light.

The spectral quality of photoperiodic lighting is also important. In LDP,
research has shown that light in the far-red portion of the spectrum is important for
ﬂower induction (reviewed by Thomas, 1993). Mixtures of red (600-700nm) and far-
red (700-800) light, when delivered as day extensions following eight hours of natural
daylight, were more effective in hastening the ﬂowering of Dianthus caryophyllus L.
‘Puritan’ and Lactuca sativa L.‘Cheshunt Vb’ than either light source alone (Vince et
al. , 1964). After eight hours of natural daylight, eight-hour day extensions with far-red
did not induce ﬂowering in Lolium temulentum L., and the response to red-light day
extensions was limited, but mixtures of red and far-red light resulted in rapid ﬂowering
(Vince, 1965). The action spectrum for ﬂoral development, quantiﬁed as apex
elongation, in light-grown T riticum aestivum L. under 16-hour day—extensions showed

two maxirna, one at 660 nm and one at 716 nm (Carr-Smith et al. , 1989).

33

Artiﬁcial light sources vary in the ratio of redzfar-red light they deliver, and in
their effectiveness for inducing ﬂowering of LDP when used as day extensions
following natural daylight. Flowering of Anethum graveolens L., T riticum compactum
Host var. Little Club, and Hordeum vulgare L. Var. Colsess I were compared under
eight-hour day extensions with incandescent and cool white ﬂuorescent lamps at 6.0
pmol m‘zs", and ﬂowering was delayed under ﬂuorescent lamps by 28, 20, and 44 days
respectively (Downs et al. , 1958). Experiments with Hyoscyamus niger L. , Beta
vulgaris L. , Anethum graveolens 1., and Petunia hybrida Vilm. found that cool white
ﬂuorescent lamps delayed ﬂowering by 10 to 20 days compared to three other light
sources with lower ratios of redzfar-red light, when delivering 8-hour day extensions
(Lane et al., 1965). At the 0.2 umol m’zs'1 level, eight-hour day extensions with
incandescent lamps hastened ﬂowering by 20 to 40 days in Petunia hybrida Vilm.
‘Pink Cascade’, when compared to high pressure sodium, cool white ﬂuorescent, metal
halide, and mercury lamps (Cathey and Campbell, 1975) while cool-white ﬂuorescent
lamps did not induce ﬂowering in Gypsophila paniculata L, but were effective when
combined with incandescent lamps (Shillo and Halevy, 1982). However, Callistephus
chinensis ‘Kometa pink’ displayed similar ﬂowering responses under incandescent, cool
white ﬂuorescent, low pressure sodium, or compact discharge lamps (deGraaf-van der
Zande and Blacquiere, 1992). The ﬂowering response in Campanula isophylla Moretti
was similar under either incandescent lamps or ﬂuorescent lamps (Moe and Heins,

1990).

34

In this study, experiments were conducted to compare the effectiveness of four
lamp types on ﬂower induction in LDP by establishing threshold and saturation light
intensities, and to determine the effect of lamp type on plant height.

Materials and Methods

Plant Culture. Seedlings of C. carpatica Jacq. ‘Blue Clips’ and C. grandiﬂora
Hogg ex Sweet. ‘Early Sunrise’ were acquired from a commercial plug producer
approximately four weeks after sowing. Seedlings of R. ﬁtlgida Ait. ‘Goldsturm’
growing in ﬁfty-cell trays (85 ml cell volume) were acquired when approximately 20
weeks old. Cuttings of C. verticillata L. ‘Moonbeam’ were propagated on 28 June
1994 for use in experiment I, and plants in 72—cell trays (50 ml cell volume) were
acquired from a commercial producer for use in experiment 11. Characteristics of the
plant material are presented in Table 1. Seedlings were subsequently transplanted into
10-cm pots (470 ml), or 13 cm (1.1 liter) for R. fulgida ‘Goldsturm’, using a
commercial soilless media containing composted pine bark, horticultural vermiculite,
Canadian Sphagnum peat moss, processed bark ash, and washed sand (MetroMix 510,
Scotts-Sierra Horticultural Products Company, Marysville, Ohio). Plants were top-
watered as necessary with 7 mM N from a 20 N-4.4 P-16.6 K all—purpose water
soluble fertilizer (20-10-20), Peter’s professional Peat-lite special (Grace-Sierra
Horticultural Products Company, Milpitas, CA). Prior to the start of long-day
treatments, plants were maintained under a 9-hr photoperiod (SD) which was provided
by pulling blackcloth from 1700 to 0800 daily. When ambient greenhouse light levels

dropped below 400 umol m'ZS'l during the 0800 to 1700 time period, HPS lights were

35

turned on automatically by an environmental control computer, providing z 50 pmol m'
23" of supplemental light at plant level.

Light Sources & Treatments. Lights were installed at one end of four benches,
75-90 cm above the bench, so that an irradiance gradient from 0.05 to 2.0 umol m'ZS'l
(400-700 nm) was established as measured by a LI-Cor quantum sensor model LI—189
(Li-COR Inc., Lincoln, Nebraska), oriented horizontally at plant height. Irradiance
was measured at the beginning of each of the experiments, and checked midway
through the ﬁrst experiment.

The spectrum of each light source was determined at the 2 umol m'ZS’l bench
position using a Li-Cor LI-1800 portable spectroradiometer (Li-COR Inc., Lincoln,
Nebraska) with the remote sensor pointed directly toward the lamp, and oriented 45 °
from the horizontal position. Calculations of RzFR ratio used band widths of 600-700
nm for red radiation, and 700-800 nm for far-red radiation. The data in Figure 1
represent the average of 3 scans.

Fifteen plants of each species were arranged along the length of each bench,
with 15 plants also remaining in SD. Black cloth was pulled at 1700 and opened at
0800 on all benches. Lights were turned on at 1700 and remained on until 2400,
providing a 7-hr day extension.

The experiment was replicated in time with experiments I, II, and III starting on
16 Sept. 1994, 15 Dec.1994, and 15 March 1995, respectively. C. carpatica ‘Blue
Clips’ and C. grandiﬂora ‘Early Sunrise’ were included in all three experiments, C.

verticillata ‘Moonbeam’ was included in experiments I and II, and R. ﬁtlgida

36

‘Goldsturm’ was included in experiment 11. Date of the ﬁrst visible bud and date of
opening of the ﬁrst ﬂower were recorded for each plant, and days to visible bud and
ﬂowering were calculated. At the time of ﬂowering, total plant height, number of
visible ﬂower buds, and number of leaves on the main stem were determined.

Temperature control. Air temperatures on each bench were monitored with two
36—gauge thermocouples connected to a CR10 datalogger (Campbell Scientiﬁc, Logan,
Utah). The datalogger collected temperature data every 10 seconds and recorded the
hourly average. Nighttime air temperatures under blackcloth on solid aluminum
benches can be 3.5C cooler than greenhouse air temperatures due to radiant heat loss to
the greenhouse glazing material (Heins and Faust, 1994). To provide uniform
temperature conditions, the datalogger controlled a 1500W electric heater under each
bench which provided supplemental heat as needed throughout the night. Actual
average daily air temperature was 20.6C for all three experiments. The maximum
difference in average daily air temperature between any two benches within each
experiment was 0.8C, 0.6C, and 0.2C for Experiments 1, II, and 111, respectively.
Data Analysis

Days to ﬂower was converted to the rate of progress to ﬂowering by taking the
reciprocal. To determine the saturation irradiance, the relationship between irradiance
and rate of progress to ﬂowering was developed using a multiphase function with two
linear components as proposed by Argo and Biernbaum (1995) . The functions used

were

37

 

b - b
Iintersection : b2 _ b0 [1]

l 3
if i I < Iintersection then R] : b0+ b11 [2]
if : I > Iintersection then R2 : b2 + b31 [3]

where the I value is log irradiance and R is the rate of progress to ﬂowering. Iintersection
is the intersection point of the two lines where the R values are equal and is calculated
using Eq. [1]; b1 and b3 are the slopes, and b0 and b2 are the y-intercepts of Eqs. [2]
and [3] respectively. Initial estimates for the parameters were obtained from a graph of
the observed data. Estimates for b0, b1, b2, and b3, based on irradiance, were obtained
using the SAS nonlinear regression procedure NLIN (SAS Institute, 1989). In cases
where b3 was <0, data were reanalyzed with b3 set at 0.

In cases where two functions could not be ﬁt to the data, linear regression was

used to develop the relationship between irradiance and rate of progress to ﬂowering.

Results

Campanula carpatica ‘Blue Clips’. None of the plants under SD ﬂowered.
Some individuals ﬂowered at the 0.05 to 0.1 umol m'zsl levels under all light sources,
indicating that the threshold irradiance for ﬂowering in this species is below 0.05 umol
m'ZS'l (Fig. 2). As irradiances increased, ﬂowering was hastened, with the response
becoming saturated at irradiances of 0.44, 0.37, 0.25, and 0.68 umol m'ZS‘1 under

CWF, HPS, INC, and MH lamps respectively (Table 2). Time to ﬂower at irradiances

38

above the saturation point averaged 51 days for all treatments, and did not differ
signiﬁcantly between lamp types. Plants in INC light treatments were signiﬁcantly
taller than those under CWF, HPS, or MH lamps (Table 3).

Coreopsis grandiﬂora ‘Early Sunrise’. None of the plants under SD ﬂowered.
Some individuals ﬂowered at the 0.05 to 0.1 ,umol m'zs1 levels under all light sources,
indicating that the threshold irradiance for ﬂowering in this species is below 0.05 pmol
m’zs'l (Fig.3). At higher light intensities, ﬂowering was hastened, with the response
becoming saturated at irradiances of 0.22, 0.25, 0.43, and 0.50 umol m‘zs‘l under
CWF, HPS, INC, and MH lamps respectively (Table 2). Time to ﬂower at irradiances
above the saturation point averaged 60 days, and did not differ signiﬁcantly between
lamp types. Some plants at irradiances above Isat did not ﬂower within the time frame
of these experiments. Plants treated with INC light were signiﬁcantly taller than those
under CWF, HPS, and MH lamps (Table 3).

C. verticillata ‘Moonbeam’. None of the plants under SD ﬂowered. Flowering
response was somewhat dissimilar between the two replications, so the data were
analyzed separately. In Experiment I (Fig. 4), all plants under CWF and HPS
treatments ﬂowered at approximately the same time, with a saturation irradiance
apparently s 0.05 umol m‘zs“. Under INC lamps, some individuals ﬂowered at the
0.05 to 0.1 umol m‘ZS'1 level, and as irradiance increased, ﬂowering was hastened until
the response became saturated at the 0.27 umol m'zsl level (Table 2). All individuals
under MH lamps ﬂowered, but ﬂowering was delayed at irradiances below 0.19 pmol

m‘zs". Time to ﬂower at irradiances above the saturation point averaged 57 days and

39

did not differ signiﬁcantly between lamp types. In Experiment 11 (Fig. 5), ﬂowering
was not induced in individuals exposed to light levels less than 0.17, 0.22, 0.37, and
0.17 umol m'ZS‘l under CWF, HPS, INC, and MH lamps, respectively. Under CWF
and HPS lamps, rate of progress to ﬂowering increased linearly as light levels
increased, so saturation irradiance apparently was not reached. Time to ﬂowering for
plants in the CWF treatment decreased from approximately 70 days to 60 days as
irradiance increased from 0.17 to 2.0 umol m'zs", and averaged 65 days. In the HPS
treatment, time to ﬂower decreased from 66 to 58 days as irradiance increased from
0 .22 to 2.0 umol m'zs“, and averaged 63 days. Rate of progress to ﬂower under INC
and MH lamps did not increase signiﬁcantly at irradiances above 0.37 and 0.17 umol
m‘zs", respectively. At irradiances above the saturation points, time to ﬂower averaged
66 days under INC and MH lamps, and did not differ signiﬁcantly between the
treatments.

Under the conditions of these experiments, the threshold irradiance in C.
verticillata ‘Moonbeam’ was below 0.37 umol m'zs". The saturation irradiance values
were dissimilar between the two experiments, apparently below 0.05 umol m‘zs1 in
some cases and above 2.0 umol m‘zsl in others. Plants treated with CWF lamps were
approximately 3.6 cm taller than those grown under HPS and INC. The average height
Of plants treated with MH lamps was 2 cm greater than those of the HPS and INC
treatments (Table 3).

R. fulgida ‘Goldsturm’. None of the plants under SD ﬂowered. Plants exposed

to light levels less than 0.37, 0.37, 0.28, and 0.28 umol m'zs", under CWF, HPS,

40

INC, and MH lamps respectively, did not ﬂower(Fig. 6). These results indicate that the
threshold irradiance for this species is approximately 0.28 to 0.37 umol m‘zs". Rate of
progress to ﬂowering increased linearly under HPS and MH lamps as light intensity
increased throughout the range tested (Table 2). Under HPS lamps, time to ﬂower
decreased from approximately 142 to 97 days as irradiance increased from 0.37 to 2.0
umol m‘zs". Time to ﬂower for plants in the MH treatments decreased from 138 to 97
days as irradiance increased from 0.28 to 2.0 pmol m'zs". Time to ﬂower under CWF
and INC lamps did not change signiﬁcantly as irradiance increased above 0.37 and
0.28 umol m'zs", respectively. Time to ﬂower under CWF and INC lamps averaged
118 days, and was not signiﬁcantly different between the two lamps. The saturation
irradiance for HPS and MH lamps was not reached under the conditions of this
experiment. The saturation irradiance for CWF and INC lamps could not be
distinguished from the threshold irradiance in those treatments. The average heights of
plants in all light treatments were not signiﬁcantly different (Table 3).
Discussion

Under the conditions of this experiment, all tested lamps were effective for
ﬂower induction in these species of LDP. When saturation irradiances could be
determined for a species, they differed by more than 100% between lamps. However,
days to ﬂower at irradiances above the saturation points did not differ signiﬁcantly
between lamps in any species tested. Horticulturally, all the lamps tested are adequate

for ﬂoral induction in these crops.

41

Lowest saturation irradiance values were obtained under INC lamps for C.
carpatica ‘Blue Clips’ and under CWF for C. grandiflora ‘Early Sunrise’. Previous
research with other species has often found that lamps with a high redzfar-red ratio,
such as CWF, were less effective than lamps with a low redzfar-red ratio, such as INC
(Cathey and Campbell, 1975; Downs et a1, 1958; Lane et al., 1965; Shillo and Halevy,
1982). Mixtures of red and far-red light are highly effective (Vince et al., 1964;
Vince, 1965). All lamps included in this work emit some radiation in both the red and
far-red wavebands, which apparently was sufﬁcient for ﬂoral induction in these
species. These ﬁndings are similar to results reported by deGraaf-van der Zande and
Blacquiere, (1992) and Moe and Heins (1990).

Threshold irradiance values in this experiment (< 0.05 to 0.37 umol m'ZS'l) are
similar to values published for other LDP. The sensitivity of these species to low light
levels indicates that growers of herbaceous perennials should consider the potential for
light pollution, which may promote ﬂowering at times when vegetative growth is
desired. In The Netherlands, light levels up to 0.10 umol m'ZS'l were measured in
greenhouses adjacent to those using supplemental lighting during the night (Bakker and
Blacquiere, 1992).

Saturation irradiance values in this experiment were between 0.22 to 0.68 umol
m'ZS‘l for all lamps for C. carpatica ‘Blue Clips’ and C. grandiﬂora ‘Early Sunrise’.
These values are within the range of 0.02 to 1.0 umol m'ZS'1 described by Vince-Prue
(1975) for many species of LDP, and similar to those reported for C. arietinum

(Summerﬁeld and Roberts, 1987) and G. paniculata (Shillo and Halevy, 1982).

42

An irradiance of 1.0 umol m'ZS'l is more than adequate for ﬂoral induction in
C. carpatica ‘Blue Clips’, C. grandiﬂora ‘Early Sunrise’, C. verticillata ‘Moonbeam’,
and R. fulgida ‘Goldsturm’. In R. fulgida, ﬂowering was hastened by up to 30 days
by increasing the irradiance from 1.0 to 2.0 pmol m‘zs‘1 , so producers of this crop
could beneﬁt by providing at least 2.0 pmol m'zs". Although not tested here, further
increases in irradiance may reduce the time to ﬂower for R. ﬁtlgida since saturation
irradiance values of 14 to 23 umol m'zs‘l have been reported for other LDP, such as
Lens culinaris (Summerﬁeld et al., 1984).

A light level of 10 footcandles, or 2.0 pmol m’zs", is commonly used for
photoperiodic control of crops such as poinsettias and chrysanthemurns (Ecke and
Hartley, 1991; Ball, 1991). Producers who currently utilize photoperiodic lighting for
poinsettias and Chrysanthemums can also use their equipment for these species of
herbaceous perennials.

Far-red light strongly enhances stem elongation and suppresses lateral branching
(Moe and Heins, 1990). INC lamps have the lowest red:far-red ratio of the lamps
tested in these experiments, and both C. carpatica ‘Blue Clips’ and C. grandiﬂora
‘Early Sunrise’ were signiﬁcantly taller when treated with INC lamps compared to all
other lamps. Stem elongation in R. fulgida ‘Goldsturm’ was apparently unaffected by
the red:far-red ratio. In C. verticillata ‘Moonbeam’, plants grown under HPS and
INC were shorter than those treated with CWF and MH, a pattern of response that did

not correspond with the red:far-red ratios of the lamps. For at least some crops, use of

43

lamps with a high red:far-red ratio can reduce ﬁnal plant height compared to lamps
with a low red:far-red ratio.

Use of CWF, HPS, or MH lamps in place of INC lighting offers beneﬁts in
energy savings as well. For example, using values presented by Campbell and Cathey
(1980); to provide 1 pmol m‘zs‘1 (400-700 nm) the electrical input that must be
provided per square meter of growing space are 2.4, 1.5, and 2.4 W for CWF, HPS,

and MH lamps, respectively, compared to 7.0 W for INC.

LITERATURE CITED

Argo, W. and J. Biernbaum. 1995. The effect of lime, irrigation-water source, and
water-soluble fertilizer on root-zone pH, EC, and macronutrient management of
container root media with irnpatiens. J. Amer. Soc. Hort. Sci. (In review).

Armitage, A. 1994a. Spring into forcing perennials. Greenhouse Grower 12: 77-78.
Armitage, A. 1994b. Spring into perennials. Greenhouse Grower 13:92-94.

Bakker, J. and T. Blacquiere. 1992. Reﬂected supplementary lighting can affect the
crops of neighbours. Acta Hort. 327:95.

Ball, V. 1991. Chrysanthemums, p. 435-467. In: V. Ball (ed.). Ball Red Book. Ball
Publishing, West Chicago, IL.

Beatty, D. and R. German, Sr. 1985. Perennials, p. 510-525. In: J. Mastalerz and E.
Holcomb (eds.). Bedding Plants 111. Pennsylvania Flower Growers.

Carr-Smith, H., Johnson, C., and B. Thomas. 1989. Action spectrum for the effect of
day-extensions on ﬂowering and apex elongation in green, light-grown wheat
(T riticum aestivum L.). Planta 179: 428-432

Cathey, H. and L. Campbell. 1975. Effectiveness of ﬁve vision-lighting sources on
photo-regulation of 22 species of ornamental plants. J. Amer. Soc. Hort. Sci.
100265-71.

Cathey, H. and L. Campbell. 1980. Light and lighting systems for horticultural plants,
p. 491-537. In: J. Janick (ed.). Horticultural reviews. AVI publishing,
Westport, CN.

De Graaf-van der Zande,M. and T. Blacquiere. 1992. Light quality during longday
treatment for poinsettia and china aster. Acta Hort. 327:87-93.

Downs, R., Borthwick, H., and A. Piringer. 1958. Comparison of incandescent and
ﬂuorescent lamps for lengthening photoperiods. Proc. Amer. Soc. Hort.Sci.
71:568-578.

Ecke, P. and D. Hartley. 1991. Poinsettias, p. 695-734. In: V. Ball (ed.). Ball Red
Book. Ball Publishing, West Chicago, IL.

44

45

Heins, R. and J. Faust. 1994. Radiant cooling leads to cooler temperatures under black
cloth. HortScience 29:503. (Abstr.)

Iversen, R. and T. Weiler. 1994. Strategies to force ﬂowering of six herbaceous garden
perennials. HortTechnology 4:61-65.

Lane, H., Cathey, H., and L. Evans. 1965. The dependence of ﬂowering in several
long-day plants on the spectral composition of light extending the photoperiod.
Amer. J. Bot. 52:1006-1014.

Moe, R. and R. Heins. 1990. Control of plant morphogenesis and ﬂowering by light
quality and temperature. Acta Hort. 272281-89.

Rhodus, T. 1993. Views on management. Perennial Plants; The quarterly newsletter of
the Perennial Plant Association, Summer 1993 26-34.

SAS Institute. 1989. SAS/STAT user’s guide, version 5, vol. 2, SAS Institute, Inc.,
Cary, NC.

Schwarze, D. 1993. Perennials. Minn. Flower Growers Bull. 42:30-39.

Shillo, R. and A. Halevy. 1982. Interaction of photoperiod and temperature in
ﬂowering-control of Gypsophila paniculata L. Scientia Hort. 16:385-393.

Summerﬁeld, R., Muehlbauer,F., and E. Roberts. 1984. Controlled environments as
an adjunct to ﬁeld research on lentils (Lens culinaris). III. Photoperiodic
lighting and consequences for ﬂowering. Expl. Agric. 20: 1-18.

Surnrnerﬁeld, R. and E. Roberts. 1987. Effects of illuminance on ﬂowering in long-
and short-day grain legumes: a reappraisal and unifying model, p. 203-223. In:
J .G. Atherton (ed.). Manipulation of ﬂowering. Butterworth’s, London.

Thimijan, R. and R. Heins. 1983. Photometric, Radiometric, and Quantum light units
of measure: a review of procedures for interconversion. HortScience 18:818-
822.

Thomas, B. 1993. The role of phytochrome and other photoreceptors in the control of
ﬂowering in long-day plants. Flowering Newsletter 16:6-10.

Vince, D., Blake, J ., and R. Spencer. 1964. Some effects of wave-length of the
supplementary light on the photoperiodic behavior of the long-day plants,
carnation and lettuce. Physiol. Plant. 17:119-125.

46

Vince, D. 1965. The promoting effect of far-red light on ﬂowering in the long day
plant Lolium temulentum. Physiol. Plant 18: 474-482

Vince-Prue, D. 1975. Photoperiodism in plants. McGraw-Hill, London.

Vince-Prue, D. and A. Canham. 1983. Horticultural signiﬁcance of
photomorphogenesis, p. 518-544. In: W. Shropshire, Jr. and H. Mohr (eds.).
Photomorphogenesis. Springer-Verlag, Heidelberg.

Withrow, R. and H. Benedict. 1936. Photoperiodic responses of certain greenhouse
annuals as inﬂuenced by intensity and wavelength of artiﬁcial light used to
lengthen the daylight period. Plant Physiol. 11:225-249.

47

Table 1. Characteristics of plant material. Experiments 1, II, and III started on 16
Sept. 1994, 15 Dec. 1994, and 15 March 1995, respectively.

 

Species

 

Expt. I
Campanula carpatica ‘Blue Clips’
Coreopsis grandiﬂora ‘Early Sunrise’

Coreopsis verticillata ‘Moonbeam’

 

Expt. II
Campanula carpatica ‘Blue Clips’

Coreopsis grandiﬂora ‘Early Sunrise’
Coreopsis verticillata ‘Moonbeam’

Rudbeckia ﬁdgida ‘Goldsturm’ ’

 

Expt. Ill
Campanula carpatica ‘Blue Clips’

9

Coreopsis grandiﬂora ‘Early Sunrise

Sow date Age at forcing Average leaf #
(wks) at start of
treatment
7June 1994 14 16.3
7 June 1994 14 24.5
Cuttings 11 z
propagated 28
June 1994.
8 Sept. 1994 14 11.4
8 Sept. 1994 14 16.7
June 1994 =28 11.1
7 Dec. 1994 14 8.6
7 Dec. 1994 14 13.6

 

‘ Not recorded.
3' LD started 3 Feb. 1995.

48

8.. >

duo—m gunmen 8 Eo>8 8 Son 88 Beach 33 3 a
note cum—8:88. BBEEAE x

as gammxéimm; m8 noun—:28... 83 2:? NM 2:. a
83. 88—8 85 5 20282030 mo 53:52 N

 

 

 

 

 

 

t... .888 8 - #88“ 888 888“ 888 - - 82
a .88 8 - 88H .88 88H 88 - - U7:
4... .888 8 - ~88“n 88 8.8“ 888 - - mm:
a .88 8 - 88.3U 888 88“ 888 - - "Eu .Eaﬁoo. 8383 .x
a .88 8 - 88.8H 888.9 88“ 288 - - m:
8.28 8 - 888“ 888 88a :88 - - 07:
.. .38 A - 88“.H 888 888“ 288 - - mam
: .88 8 - M881.r 888 8.3n :88 - - ”Eu 59588882. 3838:: .9
88 2 28 .8 888“ £88 828% g8 “88“.” 888 :2
~88 : R8 ,8 N88“ 288 :88“ ~38 v88“ :88 07:
8.88 E - 888aH 88 88H :88 - - man
2.28 2 - ~88“ M8? 88“ :88 - - "Eu 3m .5888? 3888.52 u
88 8 88 .8 N88“ :88 888“ 88 ~88“H 288 :2
88 X 9.8 .8 8.3H 288 888“ 888 88“ $88 oz“
88 mm 28 38 88H :88 888“ :88 88H «88 mm:
88 2 88.8 88“ 88 ~88“ :88 888“ 888 M88“ 288 "$6 .858 ram. 8888888 .o
5.8 mm 88 .8 88a 888 N88“ 888 ~88ur :88 :2
88 mm 28 8888H 88 88a 88 $88..n 888 888“ 288 07:
88 mm R8 .8 8888 888 :88“ :88 88% 288 mm:
88 8 3.8 28.8H 88 888“ 888 88.8H 88 .88.3288 ":5 .35 Ba. 8838 u
_. was .83 p.845 85 : Eb: it: as: who:
. 53...... + 88258.8: + 38.5.55 83 “2888
. Nu . oz . _ a 8 .8 on

 

 

.E .5 m8»: coda—3:3 mm 28 7848 881 E 8:: 92 85 he

“Eon E58835 85 88 3588.5 85 ﬂ 5853 .5 88 E .85 “8 825, 8:88qu 6 was .3 .5 an £2883.— BoEan Sm 8.3835 28 No new on
can mono—8 8.8 mp 88 3 A5833 56 88896 8 35:6: 8 8888a Lo 28 wean—2 £8588 =28»:on 88:: 28 385—8: do ﬁauaﬁmm :8. 838A.

49

Table 3. Inﬂuence of light quality on mean total plant height at the time of ﬂowering.
Data includes the height of all plants that ﬂowered.

 

 

Lamp C. carpatica C. grandiﬂora C. verticillata R. fulgida
'Blue Clips' 'Early Sunrise' 'Moonbeam' 'Goldsturm'
(cm) (cm) (cm) (cm)
CWF 13.6bz 26.0b 41.1a 33.0a
HPS 13.3b 25.1b 37.7b 29.6a
INC 16.2a 31.9a 37.5b 36.7a
MH 12.5b 23.7b 39.9ab 31.4a

 

’ mean separation in columns by Duncan's multiple range test, P s 0.05.

50

Figure 1. Spectral characteristics of light sources, average of 3 scans. Calculations of
R:FR (red:far-red) ratio are based on the 1.0 umol m'ZS'l level (400-700 nm). Band widths
of 600-700 nm were used for red radiation, and 700-800 nm for far-red radiation. (A)
Cool white ﬂuorescent, Philips F48T12, 60 watts. Photon ﬂux in 600-700 nm waveband
= 0.25 umol m'zs"; R:FR = 8.8. (B) High pressure sodium, General Electric Lucolox
LU70/MED, 70 watts. Photon ﬂux in 600-700 nm waveband = 0.40 umol mas"; R:FR
= 5.9. (C) Incandescent, Philips 100 watt bulb. Photon ﬂux in 600-700 nm waveband =
0.58 umol m‘zs"; R:FR =0.7. (D) Metal halide, VENTURE MH70W, 70 watts. Photon

ﬂux in 600-700 nm waveband = 0.27 pmol m‘zs"; R:FR = 3.3.

51

 

 

 

 

 

 

 

(A) Cool white ﬂuorescent (B) High pressure sodium
*7 0.15 -—
w
‘2‘
E
'7 0.10 —
E I
c
g 0.05 .—
1 M
0.00 - _-- #J L

 

 

 

 

 

 

 

| l l I l
400 500 600 700 800 400 500 600 700 800

nm nm

(C) Incandescent (D) Metal halide
‘7 0.15 —
we:
E
'E 0.10 —
c
B
E 0.05 —- n
0-00 “ ‘ I I I I I

52

Figure 2. Effect of light source and intensity on rate of progress to ﬂowering in
Campanula carpatica 'Blue Clips'. Each symbol (0) represents one plant. Data from the
three replications were combined. The solid line (——) represents the predicted rate of
progress to ﬂowering based on Eqs. [2 and 3]. Plants which did not ﬂower were not
included in the analysis. The saturation illuminance (1,8,) is the quantum ﬂux at the point
of intersection between the two lines. The rate of progress to ﬂowering is constant at
irradiances above 183,. None of the plants under SD ﬂowered. Statistical analysis is

presented in Table 2.

53

com
00F
hm
om
ov

BuueMou oi sAea

com
00—.
no
om
ow

.053 m. E._oE1_>>o_on¢+ F.m. ~-E._oE3 90:
x8: E3530

F. N-

md vd md Nd

 

_ _ _ _._1

_ _ A _
ON v9 dé md mod

_.d dd md Vd md Nd

 

_ _ _ _
_ _ _ _
oN v.8 o8 md

.93

mod

rd dd

II

 

F.m.~.E._oE1 mod n a...._

 

F.m.~..E._oE_.. mmd u a».

 

 

 

 

 

 

 

 

 

 

8:2 .922 EV

 

 

 

F...w.~.E._oE: nmd u an.

EoommucmoE 5V

 

 

a.

 

 

F.m.~.E._oE1 Sud u an.

 

 

 

 

 

 

 

 

 

 

 

Esﬁom 93on :9... Amv

 

 

 

E8383: 222, 600 2d

 

 

dddd
mood
dwdd
05d
dmdd
mwod

dddd
mood
owod
mrod
omod

mmod

(g 1.0/L) BugieMou o; sserﬁord to area

54

Figure 3. Effect of light source and intensity on rate of progress to ﬂowering in Coreopsis
grandiflora 'Early Sunrise'. Each symbol (0) represents one plant. Data from the three
replications were combined. The solid line (—) represents the predicted rate of progress
to ﬂowering based on Eqs. [2 and 3]. Plants which did not ﬂower were not included in the
analysis. The saturation illuminance (153,) is the quantum ﬂux at the point of intersection
between the two lines. The rate of progress to ﬂowering is constant at irradiances above
I None of the plants under SD ﬂowered. Statistical analysis is presented in Table 2.

sat '

55

ddN
dd F

CON
V1.00

Buglemog oi sAea

ddN
dd F
Nd
dm
dv

.0698 F.m.~.E._oE: .323 3+ F.m. ~-E._oE.._V mo:
x3: E35880

 

 

 

 

 

 

 

 

 

 

 

 

dd vd dd Nd _..d d.d md vd dd Nd —..d dd
.. _. V M} .. :4 40th
ON v._. d.—. dd. d.N v... d._. dd.
-m.~-E._oE: omd u .8. -m.~-E._oE: mvd n .3.

a..-

o
a a I
”II. o noon
0 o
o
.l.

 

82.9. .38.... .9

 

 

Eoomoncmos 6V

 

 

O O!

 

 

O O Ti

 

-m.~-E._oE: mNd u .8.

 

 

 

-m.~-E._oE: NNd u .a_

 

 

 

in

 

 

 

 

 

E288 Bummea :91 Amv

 

 

E8353: 32.5 600 day

 

 

 

dddd
mddd
d_.dd
m_.dd
ded
mNdd

dddd
mddd
d..d.d
m_.dd
ded
mNdd

(:l 1.0/L) 5U113M0u or sserﬁord to area

56

Figure 4. Effect of light source and intensity on rate of progress to ﬂowering in Coreopsis
verticillata 'Moonbeam', experiment I. Each symbol (0) represents one plant. For (A)
and (B), the solid line ( ) represents the predicted rate of progress to ﬂowering based on
linear regression. For (C) and (D), the solid line (—) represents the predicted rate of
progress to ﬂowering based on Eqs. [2 and 3]. Plants which did not ﬂower were not
included in the analysis. The saturation illuminance (1,3,) is the quantum ﬂux at the point
of intersection between the two lines. The rate of progress to ﬂowering is constant at
irradiances above 153,. None of the plants under SD ﬂowered. Statistical analysis is
presented in Table 2.

 

 

57

ddN
dd _.
no
on

dv

Buriamou or s/iec]

ddN
dd 3
Nd
dm

dv

_o>onm F.m.~.E._oEq_ .323 3+ F.w. N.E._oE3 do:
x3: E25830

md vd md Nd
_ . . . a

rd dd md vd md Nd

 

 

. _ . _ _
oN v... o3 md mod

 

_ _ _ __._1L

_ _ _
d.N v... d._. md mdd

3d dd

I dddd

 

 

F..m.~.E._oE1 Ed .I. a... on

wood

 

 

. “mm
F.m.~.E._oE1 N o n _ o

 

rel-L

 

IIPI-islbnx.»

.

odd
I mwdd

 

 

ded

 

 

mgmc Ham—2 ADV

 

 

 

EoommucmoE ADV

mNdd

 

 

 

 

 

 

 

dddd

 

mddd

 

 

dFdd

 

 

till-.14...»

 

0

«Juli

mvdd

 

 

ded

 

 

Eauom 9.385 :91 Amv

 

 

 

mNdd

 

E08903: 923 600 3d

 

( :1 ion) Suriamou or ssarﬁord 40 area

58

Figure 5. Effect of light source and intensity on rate of progress to ﬂowering in Coreopsis
verticillata'Moonbeam' , experiment 11. Each symbol (0) represents one plant. The solid
line (—) represents the predicted rate of progress to ﬂowering based on linear regression.
Plants which did not ﬂower were not included in the analysis. Statistical analysis is
presented in Table 2.

59

ddN
dd _.

no

0
l0

0
‘-

ddN

Buriamorr or slieq

dd _.
mm
on

dv

.988 F.......~.E..oE.. .323 3+ F.8. ~-E..oE_.V do:
x8: E3550

md vd md Nd

__ __ __ J _ 1..
d.N v... d... md mdd

Pd dd md vd md Nd

 

_ _ _ _

_ _ _
d.N V... d... md mdd

...d dd

_ i- 888

 

 

mddd

 

 

iIoILo 0+

 

 

d..d.d

 

e on

mwdd

 

 

ded

 

28sz .905. And

 

 

 

Emomoncmos .od

 

mNdd

 

 

 

J

 

 

 

 

 

I dddd

 

mddd

 

 

 

 

stills...

 

drdd

. o
.\¢\.J..4f- 288

 

 

ded

 

Eauom 9339... no... Add

 

 

 

283.53: 3.52. .80 2V

 

mNdd

 

 

( _.] 1.0/L) 5U!J€MO|} or ssarﬁord ro area

60

Figure 6. Effect of light source and intensity on rate of progress to ﬂowering in
Rudbeckia ﬁtlgida 'Goldsturm'. Each symbol (0) represents one plant. The solid line
(—) represents the predicted rate of progress to ﬂowering based on linear regression.
Plants which did not ﬂower were not included in the analysis. Statistical analysis is
presented in Table 2.

61

ddN

ddw

[\
(D

Buriamou or s/ieg

ddN

dd_.

Nd

Hm>onm F.m.~.E..oE.. .322. 3+ F.w. N.E..oE...o o2.
xoo E3320

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mo to no No ...o o.o no to no No to o.o
_. .. _. .. .1 _. .. __ .. II ooo.o
oN we o3 mo moo o.N v.3 o3 mo moo
mooo
i
owoo
mroo
82.2.22). .9 28828:. .9
i] 0'1 dddd
mooo
\ \‘O\I|D\O
00 .
Vol. 888
mroo

 

 

 

Eauom oSwmoa no.1 Add

 

 

 

E8853: 9...; .80 2V

 

 

(:LLGII.) ﬁugiamou or ssarﬁord ro area

SECTION HI

INFLUENCE OF COLD TREATMENT AND FORCING TEMPERATURE ON

FLOWERING OF CAMPANULA CARPATICA 'BLUE CLIPS'

Influence of cold treatment and forcing temperature on ﬂowering of Campanula
carpatica ‘Blue Clips’.

Catherine M. Whitman‘, Royal D. Heins“, Arthur Cameronz, and William Carlson2

Department of Horticulture, Michigan State University, E. Lansing MI 48824

Additional index words: vernalization, herbaceous perennial, long-day plant

Received for publication . We gratefully acknowledge funding by
Michigan Agricultural Experiment Station, Henry Mast Greenhouses, and other
greenhouse growers supportive of Michigan State University ﬂoriculture research. The
use of trade names in this publication does not imply endorsement of the products
named or criticism of similar ones not mentioned.

1 Graduate student.

2 Professor of Horticulture

63

64

Abstract. The inﬂuence of cold treatments on ﬂowering in Campanula carpatica Jacq.
‘Blue Clips’ was determined. Plants with 10 to 12 nodes and 12 to 16 nodes, in 128-
cell (10 ml cell volume) and 50-cell (85 ml cell volume) trays, respectively, were
stored at 5C for 0, 2, 4, 6, 8, 10, 12, or 14 weeks. They were then transplanted and
forced in a 20C greenhouse under a 4-hr night interruption (2200-0200) long day. Time
to visible bud and time to ﬂowering in 128—cell plants decreased approximately 10 days
after 14 weeks at 5C, but the trends were not signiﬁcant in 50-cell plants. The number
of ﬂower buds on 128-cell plants showed no signiﬁcant response to cold treatments,
while ﬂower bud number increased by approximately 60% in 50—cell plants as duration
of cold treatments increased. Final plant height at ﬂowering of 128-cell plants showed
no signiﬁcant trends, but height of 50—cell plants decreased 20% with increasing cold.
To determine the relationship between forcing temperature and time to ﬂower in
C. carpatica ‘Blue Clips’, 3 plant sizes were forced under a 4-hr night interruption
(2200-0200) at setpoints of 15C, 18C, 21C, 24C, and 27C. Plants ﬂowered more
quickly at higher temperatures, but number of ﬂowers and diameter of ﬂowers were
reduced. Days to visible bud and ﬂowering were converted to rates, and base
temperature (Tb) and thermal time to ﬂowering (°days), were calculated. Average Tb
for forcing to visible bud stage was 2.1C and for forcing to ﬂower, 0.0C. Calculated
°days to visible bud were 455, and°days to ﬂower, 909. These values can be used to

predict time to ﬂower at different forcing temperatures.

65

The Campanulaceae family includes over 700 species of annual, biennial, and
perennial plants. There are over 300 species within the Campanula genus, many of
which are popular garden plants (Bailey et al. , 1976). Campanula is one of the 20 top
selling genera of herbaceous perennials grown in the United States (Rhodus, 1993).

Campanula carpatica Jacq. is a popular herbaceous perennial which grows as a
compact mound, with numerous blue or white bell-shaped ﬂowers. The species is
native to the Carpathian Mountains of Eastern Europe and is hardy from USDA zones
3 to 8 (Nau, 1993). In North America, C. carpatica has traditionally been produced
outdoors and is not in bloom in the spring when the majority of garden plants are sold.
Consumers prefer to purchase plants in ﬂower, consequently producers are interested
in understanding the ﬂowering requirements of this species and developing production
schedules.

Exposure to a period of low temperatures is one of the environmental cues that
regulates development and ﬂowering in many plant species native to temperate latitudes
(Roberts and Summerﬁeld, 1987). Some species of herbaceous perennials require a
cold treatment for ﬂowering, while in others, low temperatures hasten or improve
uniformity of ﬂowering (Iversen, 1994). In some cases, cold temperatures promote
ﬂowering by breaking the dormancy of existing buds. Floral primordia of Paeonia are
initiated soon after anthesis of the current year's ﬂowers and require a minimum of
four weeks at 5.6C to break dormancy (Byme and Halevy, 1986). Flower induction,
initiation, and development occur during exposure to low temperatures in some species.

Floral bud initiation and development occurred in Astilbe during storage at 2C

66

(Pemberton, 1992) and in Dicentra spectabilis during storage at 5 and 10C
(Hanchek, 1989).

In some species, low temperatures induce the subsequent initiation and
development of ﬂoral primordia. This response is termed vernalization, generally
deﬁned as a cold treatment that induces or hastens the capacity for ﬂowering in a plant
(Chouard, 1960). Hastening of ﬂowering can be measured directly by calculating time
to ﬂower, and has also been correlated with a reduction of the number of leaves formed
before ﬂowering in grains (Chouard, 1960) and Arabidopsis thaliana L. (Bagnall,
1993). A cold treatment is not required for ﬂowering in C. carpatica, but Armitage
(1995) reported that cooling the cultivar ‘White Clips’ at 2C for 12 weeks accelerated
subsequent ﬂowering by == 50 days compared to uncooled plants, under a 16-hr
photoperiod provided with day-extension lighting. In preliminary experiments, we have
not observed this response to cold treatments in C. carpatica ‘Blue Clips’ when grown
under a 4-hr night interruption, but observed a hastening of ﬂowering of approximately
10 days when plants were forced under day—extension lighting (unpublished).

Many species native to temperate latitudes are long-day plants (LDP), so
photoperiods exceeding a certain minimum length either are required for ﬂowering or
hasten ﬂowering (Roberts and Summerﬁeld, 1987). C. carpatica is an obligate LDP
(Drushal, 1991). Recommended forcing conditions include extending the daylength to
a minimum of 16 hours, or providing a 4-hr night-break with incandescent lamps, at

10-20 footcandles (2 to 4 pmol m‘zs‘l) (Drushal, 1991).

67

Production time for any crop is related to temperatures supplied during forcing.
At temperatures below a species-speciﬁc minimum, Tb, the time to ﬂower (1‘) is
inﬁnity. Temperatures above a ceiling value, Toe, are detrimental to development and
will delay ﬂowering. In the range of temperatures between Tb and Tc, , the relationship
between mean temperature and rate of development is often linear (Roberts and
Surnmerﬁeld, 1987). Base and optimum temperatures vary within species and between
species, and are related to climatic origin (Roberts and Summerﬁeld, 1987). The Tb
calculated for leaf unfolding in Hibiscus rosa-sinensis was 9.8C (Karlsson et al. , 1991);
minimum temperature for leaf growth of Saintpaulia ionantha was 8C (Faust and
Heins, 1993); and Tb was z8.5C in three cultivars of Phaseolus vulgaris (Roberts and
Summerﬁeld, 1987). H. rosa-sinensis, S. ionantha, and P. vulgaris are all native to
tropical or sub-tropical climates. For Lilium longiﬂorum, native to temperate climates,
the base temperature was estimated to be approximately 1.1C (Karlsson et al., 1988).

Temperatures which result in most rapid ﬂowering may not optimize plant
appearance. Recommended forcing temperatures for C. carpatica are approximately
13 to 17C (Armitage, 1994; Drushal, 1991; Madsen and Madsen, 1986; Nau, 1993);
however, the actual time to ﬂower at different forcing temperatures has not been
quantiﬁed.

Information on the time required for ﬂowering is critical for development of
effective production schedules for any crop. In order to predict time to ﬂower, it is
important to determine if C. carpatica ‘Blue Clips’ exhibits a vernalization response,

and how forcing temperature inﬂuences development. The objectives of these

68

experiments were to clarify the inﬂuence of cold treatments on subsequent ﬂowering,
and to understand the relationship between temperature and time to ﬂower in C.
carpatica ‘Blue Clips’.

Materials and Methods

General. During forcing, plants were grown in a commercial soilless media
containing composted pine bark, horticultural vermiculite, Canadian Sphagnum peat
moss, processed bark ash, and washed sand (MetroMix 510, Scotts-Sierra Horticultural
Products Company, Marysville, Ohio). Plants were top-watered as necessary with 7
mM N from a 20N-4.4P-16.6K all—purpose water soluble fertilizer (20-10-20), Peter’s
professional Peat-lite special (Grace-Sierra Horticultural Products Company, Milpitas,
CA). When the ambient greenhouse photosynthetic photon ﬂux (PPF) dropped below
400 pmol m'zs", high pressure sodium lights were turned on automatically by an
environmental control computer, providing z 50 umol m‘zs‘1 PPF at plant level.

All cold treatments were delivered in a 5C cooler, lighted for 9 hours per day
with cool white ﬂuorescent lamps (VHOF96T12; Philips, Bloomﬁeld, N .J .) at
approximately 10 umol m'zs". While in the cooler, plugs were watered with well water
(340 mg calcium bicarbonate per liter) acidiﬁed (93% HZSO4) to a titratable alkalinity
of 100 mg calcium bicarbonate per liter.

Experiment 1 - cold treatments. Two plug sizes were tested. Seedlings grown
in 50-cell trays (85 ml cell volume) were received from a commercial producer on 27
October 1994 when they were approximately 20 weeks old, and had 12 to 16 nodes.

Seedlings grown in 128—cell trays (10 ml cell volume) were received 21 December

69

1994, and had 10 to 12 nodes. Upon arrival, ten plants of each size were removed
from the plug tray, thinned to a single plant per cell (singulated), and transplanted. The
128-cell seedlings were transplanted into 10-cm containers (470 ml), and the 50—cell
plants into 2.2-liter containers. Plants were placed in the greenhouse at 20 C, and
grown under a 4-hr night interruption (NI) from 2200 to 0200. The NI was supplied
with 60 W incandescent lamps at a PPF of 3 to 5 umol m'ZS'1 as measured by a LI-Cor
quantum sensor model LI-189 (Li-COR Inc., Lincoln, Nebraska). Black cloth was
pulled at 1700 and opened at 0800 on all benches. The remaining plants in the plug
trays were placed in a 5C cooler. Ten plugs of each size were removed from the cooler
at two-week intervals, singulated, transplanted, and placed under N1 in the greenhouse.
Date of the ﬁrst visible bud (when ﬂower bud was approximately 2 mm long) and date
of opening of the ﬁrst ﬂower were recorded for each plant, and days to visible bud and
ﬂowering were calculated. At the time of ﬂowering, total plant height, number of
visible ﬂower buds, and number of nodes under the terminal ﬂower on the main stem
were determined. The number of nodes formed on the main stem during forcing was
calculated.

Temperature control. Air temperatures on each bench were monitored with two 36-
gauge thermocouples connected to a CR10 datalogger (Campbell Scientiﬁc, Logan,
Utah). The datalogger collected temperature data every 10 seconds and recorded the
hourly average. Night time air temperatures under blackcloth on solid aluminum
benches can be 3.5C cooler than greenhouse air temperatures due to radiant heat loss to

the greenhouse glazing material (Heins and Faust, 1994). To provide uniform

 

7O

temperature conditions, the datalogger controlled a 1500W electric heater under each
bench which provided supplemental heat as needed throughout the night. Actual
average daily air temperature was 20.4 C. The maximum difference in actual average
daily air temperature between any two treatments throughout the experiment was 0.6C.
Data analysis. Analysis of variance was used to determine the relationship between
duration of cold treatment and time to visible bud, time to ﬂowering, number of buds
and nodes at ﬂowering, nodes formed during forcing, and total plant height.
Experiment 2 - Forcing temperature. Three plant sizes were tested. The 50-cell
seedlings were received from a commercial producer on 27 October 1994 when they
were approximately 20 weeks old, and had 12 to 16 nodes. They were placed into 3
5C cooler for 13 weeks until 23 January 1995, then singulated, transplanted into 13-cm
square containers (1.1 liter), and moved into the different treatments. The 128—cell
seedlings were received 21 December 1994, and had 10 to 12 nodes. They were
placed in a 5C cooler for 15 weeks until 7 April 1995, when they were singulated,
transplanted into 10-cm containers, and placed into the different treatments. The 5.5
cm plants (190 ml cell volume) were received on 17 January 1995, transplanted into
2.2-liter containers, and placed into the different greenhouses on 4 February 1995.
Ten plants of each size were placed into each of ﬁve greenhouses set to 15C, 18C,
21C, 24C, and 27C. Plants received natural daylengths with a 4-hr NI, 2200 to 0200,
provided by HPS lamps which delivered approximately 90 umol m'zs".

Temperature control. Temperatures in each greenhouse were controlled with a Priva

environmental computer. Actual air temperatures were recorded every ﬁfteen minutes

71

by a CR-lO datalogger. Actual average daily air temperatures were determined, and
used in all calculations.

Data analysis. Date of the ﬁrst visible ﬂower bud (when bud was approximately 2 mm
long) and date of opening of the ﬁrst ﬂower were recorded for each plant, and days to
visible bud and ﬂowering were calculated. At the time of ﬂowering, total plant height
and number of visible ﬂower buds were determined. Base temperature (Tb) and
thermal time, or degree-days, were calculated using equations presented by Roberts and
Summerﬁeld (1987). The mean time to ﬂower in each treatment was converted to a rate
by taking the reciprocal, and the relationship between the rate of progress to ﬂowering

(1/f) and the mean temperature T in °C was determined:

—: bo+b T [1]

where b0 and b1 are constants. Once b0 and b1 had been calculated, the base

temperature, Tb, was calculated as:

Tb : __ [2]
and thermal time, or degree—days was determined by:

o 1
Cd : _
b1 [3]

72

Within each developmental stage, slopes and intercepts of regression equations for the
three plant sizes were compared using the procedure by Snedecor and Cochran (1967).
Results

Experiment 1 - Screen. The number of new nodes formed on the main stem
during forcing did not change signiﬁcantly after cold treatments in plants from either
128-cell or 50-cell trays (Table 1). No other trends were common to both plant sizes.
Time to visible bud and time to ﬂowering decreased signiﬁcantly in 128-cell plants
after cold treatments, whereas the trends were not signiﬁcant in 50—cell plants. The
number of ﬂower buds on 128-cell plants showed no signiﬁcant response to cold
treatments, while ﬂower bud number increased signiﬁcantly in 50-cell plants as
duration of cold treatments increased. Final plant height at ﬂowering of 128-cell plants
showed no signiﬁcant trends, but height of 50-cell plants decreased signiﬁcantly with
increasing cold.

Experiment 2 - Forcing Temperatures. Actual average daily air temperatures
during forcing are presented in Table 2. Plants ﬂowered more quickly at higher
temperatures (Figure 1). Time to ﬂower at z 16C averaged approximately 57 days, and
at z26C, 36 days.

Rates of progress to visible bud and ﬂowering were linear within the range of
temperatures tested (Figure 1). There were small but signiﬁcant differences in
estimated Tb values for the different plant sizes (Table 3). Since the regression
parameters showed signiﬁcant differences between the plant sizes in only 5 out of 18

comparisons (Table 4), the data for the 3 plant sizes were pooled and the average actual

73

daily temperatures used in the calculations. Average base temperatures for all sizes
and developmental stages were between -1.0 and 2.0C (Table 3). Within each
developmental stage, the slopes of regression lines for the three plant sizes were not
signiﬁcantly different, indicating that the °days required for each stage were the same
for all plant sizes tested.

The average number of ﬂower buds per plant decreased as forcing temperatures
increased (Figure 3). Flower diameter on both 50-cell and 5.5-cm plants decreased as
forcing temperatures increased from z 16C to :26C. Average plant height in 50-cell
plants and 5.5-cm plants did not vary with temperature. The 128-cell plants forced at
= 16 and 19C were approximately 2 to 4 cm taller than those in all other treatments.
Discussion

Time to visible bud and time to ﬂower in the 128-cell plants decreased after 14
weeks cooling at 5C, but time to visible bud and time to ﬂower in the 50—cell plants did
not change. These ﬁndings contrast with those of Armitage (1995), who reported a
reduction in time to ﬂower from = 150 days for uncooled plants to = 100 days after 12
weeks at 2C for the cultivar ‘White Clips’. According to temperature data provided by
Armitage, forcing temperatures in the greenhouse were not consistent throughout the
course of his experiment. Plants that were cooled experienced warmer forcing
temperatures than uncooled plants, which would certainly hasten ﬂowering in the
cooled plants. A forcing schedule for C. carpatica 'Karl Foerster' suggested by
Madsen and Madsen (1986) states that plants will be ready for sale 8 to 10 weeks after

the start of forcing under long days at 15C, similar to the time to ﬂower we observed.

74

A hastening of ﬂowering in C. carpatica after cold treatments may be more
evident when plants are subsequently forced under day-extension treatments (DE)
rather than NI treatments. Preliminary experiments have revealed a reduction of time to
ﬂower after 15 weeks at 5C of approximately 10 days when 128-cell plants were forced
under DE, and approximately 4 days when plants were forced under N1 (E. Runkle,
personal communication).

Flower bud number on the 50-cell plants increased with cold treatment but
since time to ﬂowering was not reduced, it is difﬁcult to conclude that this is a
vernalization response. An increase in ﬂower number after vernalization was reported
in some species of Dianthus, but was accompanied by a hastening of ﬂowering
(Chouard, 1960). The increase in ﬂower bud number reported here may reﬂect the
higher natural light levels experienced by plants forced in the spring, although 128-cell
plants did not respond similarly.

The two plant sizes tested displayed conﬂicting responses to the cold treatments.
However, in general, holding plugs of C. carpatica 'Blue Clips' at 5C for up to 14
weeks inﬂuenced subsequent time to ﬂower by 10 days or less. Exposure to cold
temperatures had no detrimental effects on any of the characteristics evaluated in this
work.

To complete a developmental process in a plant, the plant must experience a
speciﬁc number of units of thermal time (°days) above the base temperature
characteristic of that process (Roberts and Summerﬁeld, 1987). Once the base

temperature and the amount of thermal time required for a developmental stage are

75

known, predicted time for the developmental stage can be calculated. Thermal time is
commonly used to schedule planting and harvest of fruit and vegetable crops (Roberts
and Summerﬁeld, 1987). For C. carpatica ‘Blue Clips’, calculated °days to visible
bud were 455 and°days to ﬂower, 909. To calculate time to visible bud or ﬂower at
any temperature between Tb and Tee, the degree-days required for that developmental
stage are divided by the degrees provided above the Tb . For example, time to ﬂower
for plants forced at 18C (Tb = 0.0C) is estimated to be 909°days/18.0° C = 50 days.

C. carpatica ‘Blue Clips’ is native to temperate alpine regions, and the plants in
this experiment were most attractive, in terms of the number and size of ﬂowers, when
forced at temperatures at or below 21C. For growers, plant appearance must be
balanced against the longer production time required with cooler forcing temperatures.
Forcing at 20C will hasten ﬂowering by approximately 5 days compared to forcing at
1 8C.

Plants from the 50—cell trays bloomed approximately ﬁve days earlier than
either the larger or smaller seedlings. The 5.5-cm plants had been pruned back before
Shipment, which apparently delayed their development. We expect that cutting back
plants of any size would delay their ﬂowering comparably. Slopes of the regression
lines for each stage of development did not differ signiﬁcantly between plug sizes,
indicating that all plant sizes responded similarly to the forcing temperatures.

Larger plugs generally resulted in taller plants with more ﬂowers. Seedlings
fI‘om 128-cell trays developed into well-proportioned ﬂowering potted plants in the 10-

cm containers, and larger plugs were appropriate for the 2.2-liter containers. C.

76

carpatica is grown extensively as a ﬂowering pot plant in Northern Europe (Madsen
and Madsen, 1986), and could certainly occupy a similar niche in the marketplace in

North America.

LITERATURE CITED

Armitage, A. 1994. Spring into perennials. Greenhouse Grower 13:92-94

Armitage, A. 1995. Forcing perennials for early spring sales. Bedding Plants
Foundation Inc. Res. Rpt. F-9501.

Bagnall, DJ. 1993. Light quality and vernalization interact in controlling late
ﬂowering in Arabidopsis ecotypes and mutants. Ann. Bot. 71: 75-83.

Liberty Hyde Bailey Hortorium, 1976. Hortus Third. Macmillan, New York.

Byme, T. and H. Halevy. 1986. Forcing herbaceous peonies. J. Amer. Soc. Hort. Sci.
111:379—383.

Chouard, P. 1960. Vemalization and its relations to dormancy. Annu. Rev. Plant
Physiol. 11: 191-238.

Drushal, N. 1991. Campanulas. p. 422-423 in Ball Red Book, V. Ball, Ed., Ball
Publishing, Chicago.

Faust, J. and R. Heins. 1993. Modeling leaf development of the African violet
(Saintpaulia ionantha Wendl.) J. Amer. Soc. Hort. Sci. 118:747-751.

Hanchek, A. 1989. Growth and development of Dicentra spectabilis in relation to
storage temperature and harvest date. PhD Diss. Michigan State Univ., East

Lansing.

Heins, R. and J. Faust. 1994. Radiant cooling leads to cooler temperatures under black ,
cloth. HortScience 29:503. (Abstr.)

Iversen, R. and T. Weiler. 1994. Strategies to force ﬂowering of six herbaceous garden
perennials. HortTechnology 4:61-65.

Karlsson, M., Heins, R., and J. Erwin. 1988. Quantifying temperature-controlled leaf
unfolding rates in 'Nellie White’ Easter lily. J. Amer. Soc. Hort. Sci. 113:70-

74.

Karlsson, M., Heins, R., Gerberick, J ., and M. Hackmann. 1991. Temperature-driven
leaf unfolding rate in Hibiscus rosa-sinensis. Scientia Hort. 45:323-331.

77

78

Madsen, P. and K. Madsen. 1986. Campanula- a bright star in the future. BPI News.
May, p. 7.

Nau, J. 1993. Ball Culture Guide. Ball Publishing, Batavia, Illinois.

Pemberton, G. 1992. Factors affecting the growth and development of forced Dutch-
grown Astilbe. MS Thesis, North Carolina State Univ., Raleigh.

Rhodus, T. 1993. Views on management. Perennial Plants. Summer 1993 p.26-34.

Roberts, E. and R. Summerﬁeld. 1987. Measurement and prediction of ﬂowering in
annual crops. p. 17-51. In: J.G. Atherton (ed.). Manipulation of ﬂowering.
Butterworth’s, London.

Snedecor, G. and W. Cochran. 1967. Statistical methods. 6th edition. The Iowa State
Univ. Press, Ames.

79

Table 1. Inﬂuence of 5C treatments on mean time to ﬂower, number of nodes formed
during forcing, number of ﬂower buds present at ﬂowering, and height at ﬂowering of
C. carpatica 'Blue Clips'.

 

 

 

 

 

 

 

 

 

Weeks Plug Days Nodes FLW Height
of 5C size VB FLW VB to FL Planting Final New Count (cm)
0 128 31 48 17 11 25 14 28 16
2 34 56 21 12 28 16 17 16
4 34 54 20 ll 27 16 32 18
6 25 51 26 10 25 15 22 18
8 32 53 21 12 30 18 21 16
10 24 42 18 10 25 14 23 14
12 25 44 19 13 29 16 26 15
14 25 42 17 12 27 15 25 16
Average 29 49 20 l 1 27 16 24 16
Signiﬁcance
Weeks 5C *** *** *** NS NS NS
Weeks Plug Days Nodes FLW Height
of 5C size VB FLW VB to FL Planting Final New Count (cm)
50 26 48 21 15 34 20 33 22
2 27 48 21 17 34 18 31 22
4 27 49 22 15 30 15 33 19
6 29 50 21 12 30 18 26 20
8 26 47 20 14 31 17 32 18
10 30 50 20 12 28 16 35 18
12 28 47 19 12 30 18 45 20
14 25 47 20 13 30 17 52 18
Average 27 48 20 13 31 18 36 19
Signiﬁcance
Weeks 5C NS NS NS NS *** ***

 

TVS, *, **, *** not signiﬁcant, signiﬁcant at P<0.05, 0.01, and 0.001, respectively.

80

Table 2. Actual average daily air temperatures during the indicated developmental stages, during forcing
of C. carpatica 'Blue Clips'.

 

 

Plant size Force date Set point Forcing to Visible bud Forcing to
temperature visible bud to ﬂower ﬂower
C C C C
128-cell 7 April 1995 15.0 16.3 17.2 16.7
18.0 19.4 19.2 19.4
21.0 21.1 21.2 21.1
24.0 23.6 23.7 23.6
27.0 27.7 26.9 27.4
50-cell 23 Jan 1995 15.0 15.5 16.0 15.6
18.0 18.4 19.4 18.7
21.0 20.7 21.3 20.9
24.0 22.8 22.9 22.9
27.0 25.9 26.1 25.9
5.5-cm 4 Feb 1995 15.0 15.4 15.9 15.6
18.0 18.5 19.1 18.8
21.0 20.6 21.2 20.8
24.0 22.8 23.1 22.9

27.0 25.8 26.6 26.1

81

Gone UHNOQSW ..

526888. ..88 ea .88 8.on a 888.88.... 888.8: .8 ......... .t .._. .mz .

 

 

.....:..oo.o ooo o .oooduﬂ :ood ﬂood n.n Noood 830: 8 waged
Loo o9. no. omoodH mNood moood H Dood 830d 8 can 83.83
13d mm». ﬁn Nooodu.H NNood ovood H mmood- 25 83.8..» 8 3.88 8.09.
iﬁod ooo a..- .ooodH :ood NNoodu.H ﬂood 83o... o. 3.88
.Iowd mmv o..- oooodhH mmood odeuH omood .838... 8 25 83.83
*......oo.o Ev oN .oood H vmood Rood”.H good. 95 8.2.83 9 wage Eu Wm
2:..ood 9% EN ooood“.h ﬂood hoooduH Nmood- 830: o. wEBE
...oo.o mum do. oooodH .mood odeH N. .od. 838: 8 was 83.83 A88: 2N:
imod oom W. mooodu.H oNood vooodH oNood- was 83.83 8 waged =8-om
+wod mmw wd .ooodhH ﬂood wNoodH ooood- 830d 8 mascot.
mzood 28 ed. oooodu.H vmood m:o.ou.H Noood .830... 8 25 83.8..» 38.88: 2d:
mzowd mmv 8.. mooodH NNood 38.9.48 omood. .82. 83.8? 8 waged =3.wN_
a m 88?. Co a... 0.. 28838.53: 885.83: 8mm; 3:25.233— onv. ESE

.8. 2.8.8

o5 38285

.8858 28.. .3 command 283 8:83 85 8:88p 82.8.3.8 85 5 8.8285 8: 825 Own .5098 on .888 388:. .693. So and wﬁbaod EB
25 83.85 .Amxmooo one .8885 .5 88.8.8883 can 2.5 28.32.88 83 88.8.8 8 mm. 98 E 8.5 E .88: 283 2.0.8 28 2.888.: $9.0
82m. SEES .0 E 3.83% 98 .83 83.83 8 8.8on do .38 9 888388 3.88 3.3.2 8.8385. 5.888%: good: do Masada; .m oEﬁ

82

Table 4. Inﬂuence of forcing temperature on time to ﬂower in C. carpatica ‘Blue
Clips’. Comparison of regression lines within each developmental stage, between the 3
plant sizes tested.

 

Rate to visible bud 128 vs 50-cell slopes F = 0.1649 (df = 1,5) NS
intercepts F = 8.4410 (df = 1,6) *

128 vs 5.5-cm slopes F = 0.2302 (df = 1,5) NS
intercepts F = 6.7647 (df = 1,6) *

50-cell vs 5.5-cm slopes F = 2.0000 (df = 1,6) NS

intercepts F = 46.141 (df = 1,7) **

Rate visible bud to flower 128 vs 50-cell slopes F = 0.6347 (df = 1,5) NS

intercepts F = 0.5245 (df = 1,6) NS

128 vs 5.5-cm slopes F = 0.0033 (df = 1,5) NS

intercepts F = 1.7824 (df = 1,6) NS

50-cell vs 5.5-cm slopes F = 0.8763 (df = 1,6) NS

intercepts F = 0.5597 (df = 1,7) NS

Rate to ﬂower 128 vs 50-cell slopes F = 0.0173 (df = 1,5) NS
intercepts F = 50.828 (df = 1,6) **

128 vs 5.5-cm slopes F = 0.4493 (df = 1,5) NS

intercepts F = 0.0758 (df = 1,6) NS

50-cell vs 5.5-cm slopes F = 1.5398 (df = 1,6) NS

intercepts F = 20.815 (df = 1,7) **

 

NS, *, **, *** not signiﬁcant, signiﬁcant at P<0.05, 0.01, and 0.001, respectively.

83

Figure 1. Inﬂuence of forcing temperature on time to ﬂower in C. carpatica ‘Blue
Clips’. Each symbol (0, I, orA) represents the mean of 10 plants. Open symbols (A)
represent data not included in the calculations because the plants were damaged by heat
stress. (A), (B), and (C) show days for the indicated developmental stage. For (D),
(E), and (F): solid lines (—) represent predicted values for the rate of progress to the
indicated developmental stage, based on linear regression. Statistical analysis and
calculations are presented in Table 3.

84

.0. 93.2380...
.1. om mm or v

_. ow mm or 3

 

 

 

 

9.526.“. 2 9:20". .9

9.532“. 2 8m 285 .m:

 

 

 

 

=8-va ldl ........... .1
__oo-om lol
Eo-m.m ............

 

 

 

 

gm o_n_w_> 0. @520“. 2.

 

ood

Nod

wod

mod

on
oo
o...

REP/L

$1190

85

Figure 2. Inﬂuence of forcing temperature on (A) number of ﬂower buds, (B) average
ﬂower diameter, and (C) total plant height in C. carpatica ‘Blue Clips’. For (A) and
(C), each symbol (.,I, orA) represents the average of 10 plants. For average ﬂower
diameter (B), each symbol (0 or I) represents the average of 50 ﬂowers. Flowers on
plants from 128-cell trays were not measured. Error bars represent 95% conﬁdence
intervals.

86

 

 

 

 

 

   

 

 

 

 

 

 

 

 

 

 

) H )W 1 \IL
A m n m B m C. T
5. mu 2 . m 6
H Hyl T 2
. I 2
m . 2
m m I 8
m m 1
H u H H m n u n u H u H u 4
. . . . _ . . . 4 . _ . . . . . . . . . 1
O O O O 0 0 0 5 4 3 2 1 4 2 O 8 6 4 2 O
4 2 0 8 6 4 2 2 2 2 1 1 1 1 1
4| 1 1
.mnEzc dam .Eov .mﬁEmi 530.“. ES £99. .55

Temperature (C)

SECTION IV

INFLUENCE OF COLD TREATMENTS, PHOTOPERIOD, AND FORCING
TEMPERATURE ON FLOWERING OF LAVANDULA ANGUSHFOLIA

'MUNSTEAD'

Inﬂuence of cold treatments, photoperiod, and forcing temperature on ﬂowering of
Lavandula angusafolia ‘Munstead’.

Catherine M. Whitman‘, Royal D. Heinsz, Arthur Cameronz, and William Carlson2

Department of Horticulture, Michigan State University, E. Lansing MI 48824

Additional index words: vernalization, herbaceous perennial, long-day plant

Received for publication . We gratefully acknowledge funding by
Michigan Agricultural Experiment Station, Henry Mast Greenhouses, and other
greenhouse growers supportive of Michigan State University ﬂoriculture research.
The use of trade names in this publication does not imply endorsement of the products
named or criticism of similar ones not mentioned.

1 Graduate student.

2 Professor of Horticulture

88

89

Abstract. The inﬂuence of cold treatments and photoperiod on ﬂowering of 8-to 11-
node and 18-to 23-node Lavandula angustifolia Mill. ‘Munstead’ plants from 128-cell
(10 ml cell volume) and 50-cell trays (85 ml cell volume), respectively, was
determined. Plants were stored at 5C for up to 15 weeks, then forced under a 9-hr
photoperiod (SD), or under a 4-hr night interruption (NI) (2200-0200) photoperiod at
20C. Increasing durations of cold treatment were associated with an increase in
ﬂowering percentage in plants from both 128-cell and 50-ce11 trays. Flowering
percentage was greater under N1 than SD. Time to visible bud (VB) and to ﬂowering
(FLW) generally decreased with increasing durations of cold treatment. Plants under
N1 reached FLW 3 to 7 days before those under SD. The 128-cell plants that ﬂowered
produced 1 to 2 inﬂoresences in all treatments. Inﬂoresence number in the 50-cell
plants increased with increasing durations of cold treatment, and in those cooled for 15
weeks, was increased under N1 compared to those under SD.

To determine the relationship between forcing temperature and time to ﬂower in
Lavandula angustifolia ‘Munstead’, plants growing in 3 sizes of plugs were forced
under a 4—hr NI (2200-0200) at setpoints 15C, 18C, 21C, 24C, and 27C. Plants
generally ﬂowered more quickly at higher temperatures, time to FLW decreasing from
77, 71, and 60 days at the lowest temperature (=15.6C) to 46, 40, and 36 days at the
highest temperature (z 26C) for 128-cell, 50-cell, and 5.5-cm (190 ml pot volume)
plants respectively. The 128-cell plants ﬂowered 5 to 10 days later than 50—cell plants,
and plants from 50-cell trays ﬂowered 5 to 10 days later than plants from 5.5 cm pots,

under all forcing temperatures. The Tb for forcing to VB stage was 7.1C, 10.7C, and

90

8.7C for 128-cell, 50-cell, and 5.5-cm plants, respectively, and for forcing to FLW, T,
was 4.7C, 5.4C, and 5.6C respectively. Calculated thermal time, or °days required for
forcing to VB were 357 days, 178 days, and 178 days for 128-cell, 50-cell, and 5.5-
cm plants, respectively; and for forcing to FLW, 833 days, 769 days, and 625 days,

respectively.

91

Lavandula angustifolia Mill. is native to the Mediterranean region, and is cultivated
both for its aromatic oil which is used in perfumery, and as an ornamental for its
purple ﬂowers and attractive grey-green foliage (Bailey, 1976). It is hardy in USDA
zones 5 to 9 (Ball, 1991). Botanically, L. angustifolia is considered a shrub (Bailey,
1976), but is commonly produced and marketed with herbaceous perennials (Ball,
1991). Like most herbaceous perennials in North America, L. angustifolia has
traditionally been produced outdoors and is not in bloom in the spring when the
majority of garden plants are sold. Since current consumers prefer to purchase plants
in ﬂower, producers are interested in understanding the ﬂowering requirements of this
species.

Exposure to a period of low temperatures is one of the enviromnental cues that
regulates development and ﬂowering in many plant species (Roberts and Summerﬁeld,
1987). Some species of herbaceous perennials require a cold treatment for ﬂowering,
while in others, low temperatures hasten or improve uniformity of ﬂowering (Iversen,
1994). We have been unable to ﬁnd information in the literature on the response of L.
angustifolia to cold treatments. In preliminary experiments, we found that L.
angustifolia required exposure to low temperatures for subsequent ﬂowering in the
greenhouse, and that larger seedlings are more sensitive to the inductive inﬂuence of
cold treatments (unpublished data).

A requirement for a period of low temperatures is frequently coupled with a
photoperiodic requirement, generally for long-days, so photoperiods exceeding a

certain minimum length either are required for ﬂowering or hasten ﬂowering (Roberts

92

and Summerﬁeld, 1987). In preliminary experiments, we found that L. angustifolia
demonstrated a facultative response to night-interruption (NI) lighting (unpublished
data).

Production time for any crop is related to temperatures supplied during forcing.
At temperatures below a species—speciﬁc minimum, Tb, the time to ﬂower (7') is
inﬁnity. Temperatures above a ceiling value, Tee, are detrimental to development and
delay ﬂowering. In the range of temperatures between Tb and Tc, , the relationship
between mean temperature and rate of development is often linear (Roberts and
Summerﬁeld, 1987). However, temperatures that result in most rapid ﬂowering may
not optimize plant appearance. We were unable to ﬁnd any recommended temperatures
for production of L. angustifolia as a ﬂowering plant, but for greenhouse production
as an herb, temperatures of 15 to 18C are suggested (Laskey, 1991).

We are unaware of any information in the literature on the ﬂowering
requirements of L. angustifolia, or on time to ﬂower. The objectives of these
experiments were to determine the inﬂuence of plant size, cold temperatures, and
photoperiod on subsequent ﬂowering; and to quantify the effect of forcing temperature
on time to ﬂower.

Materials and Methods

General. Plants were grown in a commercial soilless media containing
composted pine bark, horticultural vermiculite, Canadian Sphagnum peat moss,
processed bark ash, and washed sand (MetroMix 510, Scotts-Sierra Horticultural

Products Company, Marysville, Ohio). Plants were top-watered as necessary with 7

93

mM N from a 20N-4.4P-l6.6K all-purpose water soluble fertilizer (20-10—20), Peter’s
professional Peat-lite special (Grace-Sierra Horticultural Products Company, Milpitas,
CA). When the ambient greenhouse photosynthetic photon ﬂux (PPF) dropped below
400 pmol m‘zs", high pressure sodium lights were turned on automatically by an
environmental control computer, providing z 50 umol m'zs‘1 PPF at plant level.

All cold treatments were delivered in a 5C cooler, lighted from 0800 to 1700
with cool white ﬂuorescent lamps (VHOF96T12; Philips, Bloomﬁeld, NJ.) at
approximately 10 pmol m‘zs". While in the cooler, plugs were watered with well water
(340 mg calcium bicarbonate per liter) acidiﬁed (93% H2804) to a titratable alkalinity
of 100 mg calcium bicarbonate per liter.

Experiment 1 - cold treatments. Two plug sizes were tested. Seedlings growing in 50-
cell trays (85 ml cell volume) with 18 to 23 nodes were received from a commercial
producer on 27 October 1994, approximately 20 weeks after sowing. Seedlings in 128-
cell trays (10 ml cell volume) with 8 to 11 nodes were received 27 October 1994.
Upon arrival, twenty plants of each size were removed from the plug tray, thinned to a
single plant per cell (singulated), and transplanted. The 128-cell seedlings were
transplanted into 10-cm containers (470 ml), and the 50-cell plants into 2.2-liter
containers. Plants were placed in the greenhouse at 20 C. Ten plants of each size
were placed under a 9-hr photoperiod (SD). The remaining ten were placed under a 9-
hr photoperiod with a 4-hr night interruption (NI) from 2200 to 0200. The NI was
supplied with incandescent lamps at a PPF of 3 to 5 pmol m’ZS'l as measured by a LI-

Cor quantum sensor model LI-189 (Li-COR Inc., Lincoln, Nebraska). Black cloth was

94

pulled at 1700 and opened at 0800 on all benches. Twenty plugs of each size were
removed from the cooler at ﬁve-week intervals, singulated, transplanted, and placed in
the greenhouse. Half were placed under SD and half under N1. Date of the ﬁrst
visible bud (VB) (when inﬂoresence was approximately 2 mm long), and date of
opening of the ﬁrst ﬂower (FLW) were recorded for each plant, and days to visible bud
and ﬂowering were calculated. At the time of opening of the ﬁrst ﬂower, total plant
height, number of visible inﬂoresences, and number of nodes on the main stem were
determined.

Temperature control. Air temperatures on each bench were monitored with two 36-
gauge thermocouples connected to a CR10 datalogger (Campbell Scientiﬁc, Logan,
Utah). The datalogger collected temperature data every 10 seconds and recorded the
hourly average. Nighttime air temperatures under blackcloth on solid aluminum
benches can be 3.5C cooler than greenhouse air temperatures due to radiant heat loss to
the greenhouse glazing material (Heins and Faust, 1994). To provide uniform
temperature conditions, the datalogger controlled a 1500W electric heater under each
bench which provided supplemental heat as needed throughout the night. Actual
average daily air temperature throughout the course of the experiment was 20.4 C.
The maximum difference in actual average daily air temperature between any two
treatments throughout the experiment was 0.7 C.

Data analysis. Analysis of variance was used to relate weeks of cold treatment and
photoperiod to time to ﬂower, number of inﬂoresences and nodes present at ﬂowering,

and ﬁnal plant height.

95

Experiment 2 - Forcing temperature. Three plant sizes were tested. Seedlings in 50-
cell trays with 19 to 23 nodes were received from a commercial producer on 27
October 1994, approximately 20 week after sowing. They were placed into a 5C cooler
for 13 weeks until 22 January 1995, then transplanted into 13-cm square containers
(1.1 liter), and moved into the different treatments. Seedlings in 128—cell trays were
received 27 October 1994, with 7 to 9 nodes. They were placed in 3 5C cooler for 13
weeks until 21 January 1995, when they were transplanted into lO-cm containers and
placed into the different treatments. Plants growing in 5.5 cm pots (190 ml pot
volume) were received on 17 January 1995, transplanted into 2.2-liter containers, and
placed into the different greenhouses on 4 February 1995. Ten plants of each size were
placed into each of ﬁve greenhouses set to 15C, 18C, 21C, 24C, and 27C. Plants
received natural daylengths with a 4-hr NI, 2200 to 0200, provided by HPS lamps
which delivered approximately 90 umol mas".

Temperature control. Temperatures in each greenhouse were controlled with a Priva
environmental computer. Actual air temperatures were recorded every ﬁfteen minutes
by a CR-10 datalogger. Actual average daily air temperatures were determined, and
used in all calculations.

Data analysis. Date of the ﬁrst visible bud (when inﬂoresence was approximately 2
mm long) and date of opening of the ﬁrst ﬂower were recorded for each plant, and
days to visible bud and ﬂowering were calculated. At the time of opening of the ﬁrst
ﬂower, total plant height, number of additional visible inﬂoresences, and number of

nodes on the main stem were determined. Base temperature (Tb) and thermal time, or

96

degree-days, were calculated using equations presented by Roberts and Summerﬁeld
(1987). The average time to visible bud or ﬂower within each treatment was converted
to a rate by taking the reciprocal, and the relationship between the rate of progress to

visible bud or ﬂowering (10‘) and the mean temperature T in °C was determined:

—= b +bT [1]

where b0 and b1 are constants. Once b0 and b1 had been calculated, the base

temperature, Tb, was calculated as:

Tb : -_ [2]

and thermal time, or degree-days was determined by:

Td=ﬁ [a
Results

Experiment 1 - Cold treatments and photoperiod. Flowering percentage in the
8-to ll-node plants from 128-cell trays did not reach 100 in any treatment (Figure 1).
Increasing durations of cold treatment were associated with an increase in ﬂowering
percentage in plants from both 128-cell and 50-cell trays. Flowering percentage for

both plant sizes tested was greater under N1 than SD.

97

Time to VB and to FLW generally decreased with increasing durations of cold
treatment (Figure 1). Plants under NI generally reached FLW 3 to 7 days before those
under SD.

The 128-cell plants that ﬂowered produced 1 to 2 inﬂoresences in all treatments
(Figure 2). Inﬂoresence number in the 50-cell plants cooled for 15 weeks was
increased under N I compared to those plants forced under SD.

The ﬁnal node number and height in 128-cell plants under N1 were similar
whether plants were cooled for 10 or 15 weeks (Figure 2). Final node number and
height in 50-cell plants generally decreased as duration of cold treatment increased.

Experiment 2 - Forcing temperatures. Actual average daily air temperatures
during forcing are presented in Table 1. Plants generally ﬂowered more quickly at
higher temperatures (Figure 3). Time to ﬂower at z 16C averaged approximately 69
days, and at z26C, 41 days.

The 128-cell plants ﬂowered 5 to 10 days later than 50-cell plants, and plants
from 50-cell trays ﬂowered 5 to 10 days later than plants from 5.5-cm pots, under all
forcing temperatures.

All plants ﬂowered in all treatments except for plants at z26C, where only
40% of plants ﬂowered and 128-cell plants at 23C, where 70% ﬂowered. Since
ﬂowering was adversely affected in those treatments, and Eqs. [2] and [3] are only
valid for temperatures below the optimum or ceiling temperature, visible bud and
ﬂowering data for those plants were not included in the calculations. Rates of progress

to visible bud and ﬂowering were linear between 16C and 21C for 128—cell plants, and

98

between 16C and 23C for 50-cell and 5.5-cm plants (Figure 3). The Tb for forcing to
visible bud stage were between 7.1C and 10.7C, and for forcing to ﬂower, between 4.7
and 5.6C (Table 2). Slopes of regression lines for the 3 plant sizes were not
signiﬁcantly different from each other within each developmental stage (Table 3),
indicating that the °days required for each stage were the same for all plant sizes
tested.

The average number of inﬂoresences per plant increased as initial plant size
increased and forcing temperatures decreased (Figure 4). Final plant height in 128-
cell, 50-cell, and 5.5-cm plants was 5, 6, and 12 cm greater respectively when forced
at =15C than when forced at z26C. The 5.5-cm plants had been cutback before
shipping, resulting in a ﬁnal plant height below that of the 50—cell plants.

Discussion. Preliminary experiments had suggested that plant size inﬂuences the
ﬂowering response of L. angustifolia ‘Munstead’. The results of these experiments
also indicate that 128-cell plants with 8 to 11 nodes required a longer cold treatment to
induce ﬂowering than 50—cell plants with 18 to 23 nodes. Flowering percentage did not
reach 100% for the 128-cell plants in any treatment.

Many biennials and perennials must attain a certain size before they are capable
of ﬂoral induction. Juvenility is deﬁned as an early developmental phase during which
a plant is insensitive to conditions that later promote ﬂower initiation (Bernier et al.,
1981). A juvenile phase has been described in several herbaceous perennials including
some cultivars of Aquilegia x hybrida. To obtain 100 percent ﬂowering in 'McKana's

Giant' plants, at least 12 leaves and 10 weeks of storage at 4.5C were required while

99

for 'Fairyland,’ 15 leaves and only four weeks of cold storage were required. (Shedron
and Weiler, 1982). For maximum ﬂowering in Heuchera sanguinea, plants should
have approximately 19 nodes before cold storage (Yuan,1995). A juvenile phase
appears to exist in L. angustifolia ‘Munstead’, but its duration can not be determined
from these experiments.

Exposure to a period of low temperatures is a primary factor inﬂuencing
ﬂowering in L.angustifolia 'Munstead'. Flowering percentage, time to ﬂower, number
of inﬂoresences, number of nodes present at ﬂowering, and total plant height all were
inﬂuenced by cold treatments.

Exposure to low temperature can inﬂuence ﬂowering of herbaceous perennials
in several different ways. Cold temperatures promote ﬂowering in many plants by
breaking the dormancy of existing buds. Floral primordia of Paeonia are initiated after
anthesis of the current year's ﬂowers and require a minimum of four weeks at 5.6C to
break dormancy (Byrne and Halevy, 1986). Flower induction, initiation, and
development occur during exposure to low temperatures in some species. Floral bud
initiation and development occurred in Astilbe during storage at 2C (Pemberton, 1992)
and in Dicentra spectabilis during storage at 5 and 10C (Hanchek,1989).

In many species, exposure to low temperatures induces the subsequent initiation
and development of ﬂoral primordia. Vemalization is generally deﬁned as a cold
treatment that induces or hastens the capacity for ﬂowering in a plant (Chouard, 1960).
Floral primordia are not initiated until after the return of warmer temperatures.

We observed vegetative growth from the apical meristem on uncooled

100

Langustifolia ‘Munstead’ plants that did not ﬂower. Since the inﬂoresence of this plant
is a terminal inﬂoresence, it is unlikely that ﬂower buds were present prior to cold
treatments. While low temperatures clearly inﬂuence ﬂowering of L. angustifolia
‘Munstead’, we did not determine in these experiments whether initiation occurs during
or after cold treatments.

When 50—cell plants were cooled for 10 or 15 weeks, ﬂowering percentages in
plants forced under SD were 60% and 100% , respectively. Extended exposure to SC
apparently eliminated any photoperiodic requirement for ﬂowering of L.angustifolia
‘Munstead’. A similar response has been found in a number of herbaceous perennials,
including Iris (Buxton and Mohr, 1969), Dicentra spectabilis (Lopes and Weiler,
1977), Echinops ‘Taplow Blue,’ Achillea millefolium ‘Rosea,’ and Physostegia
virginiana ‘Summer Snow’ (Iversen, 1989), and Chrysanthemum x superbum Bergmans
(Shedron, 1980).

Forcing temperatures have a signiﬁcant effect on rates of development in all
plants. At temperatures below a certain species-speciﬁc value, no growth will occur,
and at temperatures above optimum, development is delayed or ceases. Optimum
temperature ranges vary within species and between species, and are related to climatic
origin (Roberts and Summerﬁeld, 1987). The Tb calculated for Hibiscus rosa-sinensis
was 9.8C (Karlsson et al. , 1991), minimum temperature for leaf growth of Saintpaulia
ionantha was 8C (Faust and Heins, 1993), and T, was =8.5C in three cultivars of
Phaseolus vulgaris (Roberts and Summerﬁeld, 1987). H. rosa-sinensis, S. ionantha,

and P. vulgaris are all native to tropical or sub-tropical climates. The base temperature

 

101

was estimated to be approximately 1.12C in Lilium longiﬂorum (Karlsson et al., 1988),
a species native to temperate climates. Calculated base temperatures for L. angustifolia
‘Munstead’ for forcing to the visible bud stage were 7.1 to 10.7C, and for forcing to
ﬂower, 4.7 to 5.6C. These values reﬂect the moderate Mediterranean climate in the
regions where L. angustifolia is native. Slopes of regression lines for the 3 plant sizes
were not signiﬁcantly different from each other within each developmental stage,
indicating that different plant sizes responded similarly to forcing temperatures (Table
3). Intercepts of regression lines were signiﬁcantly different in some comparisons,
reﬂecting the fact that the plants were at different physiological stages at the start of
forcing.

To complete a developmental process in a plant, the plant must experience a
speciﬁc number of units of thermal time (°days) above the base temperature
characteristic of that process (Roberts and Summerﬁeld, 1987). Once the base
temperature, and the amount of thermal time required for a developmental stage are
known, predicted time for the developmental stage can be calculated. Thermal time is
commonly used to schedule planting and harvest of fruit and vegetable crops (Roberts
and Summerﬁeld, 1987) and ﬁeld crops such as maize (Tollenaar et al., 1979).

In L. angustifolia, degree days required for reaching the visible bud stage
varied between 178 and 357, and for ﬂowering, 625 to 833 between 15C and 22C. The
128-cell plants required more thermal time than the 50—cell or 5.5-cm plants, because
they were at a less advanced physiological starting point when forcing began. In order

to calculate time to visible bud or ﬂower at any temperature between Tb and Tee, the

102

degree-days required for that developmental stage are divided by the degrees provided
above the Tb. For example, time to ﬂower in 50—cell plants forced at 18C (Tb = 5.4C)
is 769°days/12.6° C z 60 days.

In conclusion, recommendations for ﬂowering L. angustifolia 'Munstead' in
approximately 8 weeks include using initial plant material with more than 11 nodes.
Plants with 18 to 23 nodes are adequate. For rapid uniform ﬂowering, provide a
minimum of 10 weeks of 5C. Flower number will be increased if the plants are
subsequently forced under a 4-hr NI. To reach ﬂowering in 8 weeks, forcing
temperatures should be approximately 14 degrees above the Tb of 5C, or 19C
(769°days/ 56 days = 14) Inﬂoresence number decreased with higher forcing
temperatures.

The 5.5-cm plants used in this experiment had been cut back before shipping,
and we observed that their regrowth was compact, uniform, and attractive. Although
not speciﬁcally tested in this work, observations suggest that pruning or pinching may
be beneﬁcial when forcing other sizes of L. angustifolia ‘Munstead' as ﬂowering

plants.

LITERATURE CITED
Armitage, A. 1994. Spring into perennials. Greenhouse Grower 13:92-94.
Liberty Hyde Bailey Hortorium, 1976. Hortus Third. Macmillan, New York.

Ball, V. 1991. Perennials. p. 677-689 in Ball Red Book 15th edition, V. Ball, Ed.,
Ball Publishing, Chicago.

Bernier, G., Kinet, J ., and R. Sachs. 1981. Age and ﬂower initiation. p. 106-115 in
The physiology of ﬂowering, Vol.1. CRC Press, Boca Raton, FL.

Buxton, J. and H. Mohr. 1969. The effect of vernalization and photoperiod on
ﬂowering of tall bearded irises. HortScience 4:53-55.

Byme, T. and H. Halevy. 1986. Forcing herbaceous peonies. J. Amer. Soc. Hort. Sci.
111:379-383.

Chouard, P. 1960. Vemalization and its relations to dormancy. Annu. Rev. Plant
Physiol. 11: 191-238.

Faust, J. and R. Heins. 1993. Modeling leaf development of the African violet
(Saintpaulia ionantha Wendl.) J. Amer. Soc. Hort. Sci. 118:747-751.

Hanchek, A. 1989. Growth and development of Dicentra spectabilis in relation to
storage temperature and harvest date. PhD Diss. Michigan State Univ., East
Lansing.

Heins, R. and J. Faust. 1994. Radiant cooling leads to cooler temperatures under black
cloth. HortScience 29:503. (Abstr.)

Iversen, R. 1989. Greenhouse forcing of herbaceous garden perennials. PhD Diss.
Cornell Univ., Ithaca.

Karlsson, M., Heins, R., and J. Erwin. 1988. Quantifying temperature-controlled leaf
unfolding rates in 'Nellie White’ Easter lily. J. Amer. Soc. Hort. Sci. 113:70-
74.

Karlsson, M., Heins, R., Gerberick, J ., and M. Hackmann. 1991. Temperature-driven
leaf unfolding rate in Hibiscus rosa-sinensis. Scientia Hort. 45:323-331.

Laskey, M. 1991. Herbs. p. 577-581 in Ball Red Book 15th edition, V. Ball, Ed., Ball

103

104

Publishing, Chicago.

Lopes, L. and T. Weiler. 1977. Light and temperature effects on the growth and
ﬂowering of Dicentra spectabilis (L.) Lem. J. Amer. Soc. Hort. Sci. 102:388-
390.

Pemberton, G. 1992. Factors affecting the growth and development of forced Dutch-
grown Astilbe. MS Thesis, North Carolina State Univ., Raleigh.

Roberts, E. and R. Summerﬁeld. 1987. Measurement and prediction of ﬂowering in
annual crops. p. 17-51. In: J .G. Atherton (ed.). Manipulation of ﬂowering.
Butterworth’s, London.

Shedron, K. 1980. Regulation of growth and ﬂowering of four herbaceous perennials:
Aquilegia x hybrida Sims, Aurinia saxatilis (L.) Desv., Chrysanthemum x
superbum Bergmans, and Lupinus spp. ‘Russell.’ MS Thesis, Purdue Univ.,
Lafayette.

Shedron, K. and T. Weiler. 1982. Regulation of growth and ﬂowering in Aquilegia x
hybrida Sims. J. Amer. Soc. Hort. Sci. 107:878-882.

Snedecor, G. and W. Cochran. 1967. Statistical methods. 6th edition. The Iowa State
Univ. Press, Ames.

Tollenaar, M., Daynard, T. and R. Hunter. 1979. Effect of temperature on rate of leaf
appearance and ﬂowering date in maize. Crop Sci., 19:363-366.

Yuan, M. 1995. Effect of juvenility, temperature, and cultural practices on ﬂowering
of ﬁve herbaceous perennials. MS Thesis, Michigan State Univ., E. Lansing.

105

Table 1. Actual average daily air temperatures during the indicated developmental
stage of L. angustifolia ‘Munstead’.

 

 

Set point Forcing to Visible bud Forcing to

Plant size Force date temperature visible bud to ﬂower ﬂower
(C) (C) (C) (C)

128-cell 21 Jan 1995 15.0 15.4 15.8 15.6
18.0 18.5 19.1 18.7

21.0 20.7 21.1 20.8

24.0 22.9 23.1 23.0

27.0 26.0 26.5 26.2

50-cell 22 Jan 1995 15.0 15.4 15.7 15.6
18.0 18.2 19.0 18.7

21.0 20.5 20.8 20.8

24.0 22.6 23.0 22.8

27.0 26.2 25.9 26.0

5.5 cm 4 Feb 1995 15.0 15.4 15.8 15.6
18.0 18.4 19.1 18.8

21.0 20.3 21.0 20.9

24.0 22.7 23.1 22.9

27.0 25.5 26.5 26.1

 

 

106

8382.8. .8... .5 «adv. a .888 .888... .8 ....... .. .mz .

 

 

.880 ognﬁm N
*mod mNo 8m mooodu.” o.oo.o .ooodu.H owood. 830.. o. 8.88
mzowd oo... o... ..ooodH mNood o.m.odn.H .N.o.o- 88.6.. o. .85 8......>
mzood or. ...w ..oo.o”.H omood ooNodH mwvod- .85 e36.> o. 8.88 :8 Wm
**wod ooh 8m .ooodn.H m.oo.o ..Nood“.H ohood- .250... o. 8.88
food ooo do- .ooodH ..oo.o w.oo.ou.H oo.od .36... o. .85 03...... Amoco: mN-o.V
*wod w... hd. ooood”.H omood M...o.ou.H oomod- 85 037.3 o. 8.88 ..oo-om
mzvod mm... b... mooodH N.ood mmood”.H omood- .026... 8 8.88
good o2. Nd- Nooodn.H m.ood mmood“.H oo.o.o 830.. 9 95 oBE> 30.8: o....
.8..de nmm .6 N.oo.oH wNood oNNodNH ooNod- 85 28m.> 8 8.88 ..oodN.
0.. 28338:. 8858: 9.8.
q. 93?. .0. a... on. 8.2m .3. 2.0888. owﬁm 8588.33. 3% 88.x.

 

.80....828 2.. 8 3.5.08 .0: 83 UN 8 .888 $8.... 8?m.m .8. ..oo.om .8 .8. .UNdN .8. N 8 .888 $5...

..oo.wN. .8 Bad 8 858.88 8882.88 82%... o... 8 38.... 688 8 8.8.0.. 83 8688.025 .658... 68.. .888... .o
6.98.08on .8. 3.... 8882.88 83 «8.3.8 o. .m. .8. .N. .mom 8 .88 88> 8.05 .88 .3868. ..38822. 838.8%...» ..N
8 8.830... .8. .85 03...? o. 8088.. D8 88 8 8888.83 8.88 w8§8 £338. 80.8088 .38. .0 88088.... .N 0.8....

107

Table 3. Inﬂuence of forcing temperature on time to ﬂower in Lavandula angustzfolia
‘Munstead’. Comparison of regression lines within each developmental stage, between
the 3 plant sizes tested, followed procedures described by Snedecor and Cochran

(1967).

 

Rate to visible bud 128 vs 50-cell slopes F = 5.5276 (df = 1,3) NS
intercepts F = 9.4234 (df = 1,4) *
128 vs 5.5-cm slopes F = 1.8564 (df = 1,3) NS
intercepts F = 19.649 (df = 1,4) *
50-cell vs 5.5-cm slopes F = 0.0000 (df = 1,4) NS
intercepts F = 10.4373 (df = 1,5) *
Rate visible bud to ﬂower 128 vs 50—cell slopes F = 0.8220 (df = 1,3) NS
intercepts F = 52.838 (df = 1,4) **
128 vs 5.5-cm slopes F = 1.5209 (df = 1,3) NS
intercepts F = 0.0537 (df = 1.4) NS
50-cell vs 5.5-cm slopes F = 3.8447 (df = 1,4) NS
intercepts F = 3.7091 (df = 1,5) NS
Rate to ﬂower 128 vs 50-cell slopes F = 0.4327 (df = 1,3) NS
intercepts F = 10.445 (df = 1,4) *
128 vs 5.5-cm slopes F = 0.8185 (df = 1,3) NS
intercepts F = 19.167 (df = 1,4) *
50-cell vs 5.5-cm slopes F = 0.6437 (df = 1,4) NS
intercepts F = 14.688 (df = 1,5) *

 

z N S, *, not signiﬁcant, signiﬁcant at P <0.005, respectively.

108

Figure 1. Inﬂuence of cold treatments on ﬂowering of L. angustifolia ‘Munstead’. (A)
and (B) show ﬂowering percentage for 128—cell and SO-cell plants, respectively. (C)
and (D) show time to visible bud and ﬂower for 128-cell and SO-cell plants,
respectively. Error bars represent standard error of the means.

 

 

109

o m a 3mm;

 

 

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

 

am .2332“. .6. .

_z 9:35.“. I?
am .25 gems :01 .
_z .25 297.5 .0.

 

 

 

 

 

 

 

 

 

nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn

 

 

 

 

 

__mo-om __mo..mN_‘

OOOO
CDCOV'N

on:

sileg

ﬁuuamou wowed

110

Figure 2. Inﬂuence of cold treatments on ﬂowering of L. angustifolia ‘Munstead’. (A)
and (B) represent number of inﬂoresences visible per plant at the time of ﬁrst ﬂower
opening for 128-cell and SO-cell plants respectively. (C) and (D) show number of nodes
present on the main stem at ﬂowering for 128-cell and SO-cell plants respectively. (B)
and (F) are total plant height at ﬂowering for 128-cell and SO—cell plants respectively.
Error bars represent standard error of the means.

111

50-cell

128-cell

 

 

 

 

 

 

 

(A)
+so

 

 

(D)

 

 

 

1
(C)

 

wm_

t

m 4.

 

-0— so

 

 

 

 

 

2086420
11

mmocmmmzocE .6 59:32

w s w
0 0 O
5 4 3

mono: ho 59:52

 

O
2

 

 

(F)

 

 

 

 

 

(E)

 

 

 

_ I
O O 0 0
5 4 3 2
$9 2%:

H: -

15

1O

1 5
Weeks at 5 C

10

112

Figure 3. Inﬂuence of forcing temperature on ﬂowering of L. angustifolia
‘Munstead’. Each symbol (A, O, or I ) represents the mean of 10 plants except for
the 24-setpoint and 27-setpoint treatments of 128-cell plugs. Seven and four plants
ﬂowered in those treatments, respectively. Open symbols represent data not included
in regression analysis. (A), (B), and (C) show days for the indicated developmental
stage. For (D), (E), (F), solid lines (—) represent predicted values for the rate of
progress to the indicated developmental stage, based on linear regression. Statistical
analysis and calculations are presented in Table 2.

113

Q 239388.

mm mm m? 3. mm mm w_. 3. mm mm 9. S.
_ r _ _ _ _ _ _ _ 00.0

 

No.0

wed

Rep/L

mod

mod

 

 

 

 

sAeq

__mo-cm Iol D ........................ 1 cm
__mo-mNF 141

 

 

 

 

 

.. @5532“. 2 Ego“. Q 3526.“. 9 3m gems av..mamm_m_._w_..>ms.mm_gmm_...Ei om

 

 

 

114

Figure 4. Inﬂuence of forcing temperature on (A) number of ﬂower buds and (B) total
plant height in L. angustifolia ‘Munstead’. Each symbol (.,I, orA) represents the
mean of 10 plants, except in the 24-setpoint and 27-setpoint treatments for 128-cell
plugs, where the symbols represent the mean of 7 and 4 plants, respectively. Error
bars represent 95% conﬁdence intervals.

 

Inﬂoresence number

Plant height (cm)

115

 

 

oooooooooooooooooooooooooooooooooooooooooooooooo

 

 

 

+ 5.5-cm
—l— 50-cell

+ 128-cell

 

 

 

 

 

 

Temperature (C)