Quantitative contributions of blue light and PAR by Jeff Hunter

Journal of Experimental Botany, Vol. 57, No. 10, pp. 2379–2390, 2006 doi:10.1093/jxb/erj210 Advance Access publication 23 June, 2006

RESEARCH PAPER

Quantitative contributions of blue light and PAR to the photocontrol of plant morphogenesis in Trifolium repens (L.) Ange´lique Christophe*,†, Bruno Moulia‡ and Claude Varlet-Grancher Unite´ d’Ecophysiologie des Plantes Fourrage`res, INRA, Lusignan, F-86600, France Received 17 October 2005; Accepted 22 March 2006

Abstract Shade-avoidance is a major adaptive response of plants, and is usually considered to be controlled by phytochromes through the perception of changes in the red:far red light ratio. However, few studies on the effects of blue light (BL) and of light intensity [photosynthetically active radiation (PAR)] on light-grown plants have been conducted, especially concerning changes in PAR at constant BL. The objective here was to quantify the photocontrol of aerial morphogenesis by BL and PAR. Experiments were conducted varying BL and PAR independently, with three BL levels (4, 38, and 83 lmol m22 s21) at constant PAR (300 lmol m22 s21) and three PAR levels (338, 705, and 163 lmol m22 s21) at constant BL (36 lmol m22 s21). Effects on morphogenetic processes were analysed as quantitative modulations of ontogenic trends and response curves were produced. White clover (Trifolium repens L.) was used, as it is a typical shade-avoider displaying the whole syndrome of shade-avoidance in a purely vegetative stage. Morphological responses were strongly controlled by both BL and PAR changes, through antagonist effects on leaf appearance rate and additive effects on petiole elongation. All the other responses appeared to be the indirect consequences of changes in the leaf appearance rates. BL acted as a light signal for plant morphogenesis. However, the PAR control probably implicates two distinct mechanisms, such as a trophic effect and a signal. Both PAR and BL actions involved organ-speciﬁc differences, which are central in the control of the shadeavoidance responses.

Key words: Blue light, leaf appearance rate, PAR, petiole growth, photomorphogenesis, plasticity, shade-avoidance, Trifolium repens L.

Introduction Plants have evolved a high level of plasticity in response to light conditions. In many species, light-grown plants react to canopy shade by a suite of morphogenetic responses, involving (i) an enhancement of stem and petiole elongation and an associated reduction in lamina expansion, (ii) an increase in main axis development compared with that of existing branches, and (iii) a reduction in the number of branches (Smith and Whitelam, 1997). This suite of photomorphogenetic responses has been called the shadeavoidance syndrome (Grime, 1981) and is interpreted as an adaptive light-foraging behaviour (for reviews see de Kroon and Hutchings, 1995; Ballare´ et al., 1997). However, the way that these responses are controlled by shading is not yet fully understood. Many questions remain unanswered, such as, which spectral components of light are involved and to what extent? Generally, studies of the shade-avoidance syndrome are primarily concerned with the perception of the changes in the red far:red light ratio (R:FR) through phytochromes. This has now been well characterized (for reviews see Smith and Whitelam, 1997; Smith, 2000; Franklin and Whitelam, 2005). However, it is known that approximately half of the total shade-avoiding responses can also be observed under neutral shading at a constant R:FR ratio, thus involving other photocontrolling mechanisms ( Lo¨tscher and No¨sberger, 1997; Stuefer and Huber, 1998). More recently,

* To whom correspondence should be addressed. E-mail: christop@ensam.inra.fr y Present address: Laboratoire d’Ecophysiologie des Plantes sous Stress Environnementaux, UMR INRA – ENSAM, place Viala, F-34060 Montpellier cedex 1, France. z Present address: UMR PIAF, INRA, Site de Croue¨l, 234 avenue du Bre´zet, F-63039 Clermont-Ferrand cedex 02, France. Published by Oxford University Press [2006] on behalf of the Society for Experimental Biology.

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

2380 Christophe et al.

manipulation of light conditions during plant development and the use of various photoreceptor mutants have indeed revealed the specific role of the blue light (BL) fluence rate in the shade-avoidance syndrome, acting through cryptochromes (Ballare´ et al., 1991; Kozuka et al., 2005). However, despite significant advances in the understanding of the molecular mechanisms involved in BL perception (Lin and Shalitin, 2003) and in the BL control of seedling morphogenesis (Ahmad et al., 2002), there is still little information about the effects of BL on light-grown plants. And even fewer studies have been conducted on light-grown plants to separate BL fluence rate from changes in phytochrome photoequibrium and in photosynthetically active radiation (PAR) (Ballare´ et al., 1991; Gautier et al., 1997, 1998; Dougher and Bugbee, 2001a). Additionally, there is evidence indicating that PAR is also implicated in the control of plant morphogenesis. For example, light-grown plants displayed different responses in stem elongation or growth under two levels of PAR with no reduction in BL and in the R:FR ratio (Ballare´ et al., 1991; Dougher and Bugbee, 2001a). More recently, interactions between the BL responses and the sugarsensing pathway (which is likely to be PAR-dependent) were also observed by using a genetic approach on Arabidopsis using sugar-insensitive mutants (Kozuka et al., 2005). However, the effects of PAR are still rarely documented and no quantitative insights have been given on the relative importance of BL and PAR responses in a given amount of shading. Indeed, most of studies on the effects of PAR on plant morphogenesis are focused upon the so-called ‘neutral shading’, which reduces light homogeneously in the 400–700 nm waveband with a constant phytochrome photoequilibrium but inducing a proportional reduction of BL. Interpretation of the effects of PAR and BL on plant morphogenesis in these studies is complicated because the light sources used and the BL:PAR ratios vary considerably. Plants may respond simultaneously, but in different ways, to changes in BL and in PAR so these studies can only yield circumstantial conclusions about the photocontrol of shade-avoidance responses. Another central question concerns the photocontrol of the different, and possibly antagonistic responses, of the plant parts involved in the adaptive shade-avoidance syndrome, for example, stimulation of stem elongation and inhibition of lamina expansion (Kozuka et al., 2005). Several hypotheses can be proposed to explain such differences. The simplest one is based on the fact that plant growth during ontogenesis is generally allometric rather than isometric. So an effect of light on the rate of plant development would change the absolute growth rates of organs, whereas the allometric relationships would be kept constant (Wright and Mc Connaughay, 2002). This potential confusion between plasticity in growth rates and in allometries could be overcome by taking plant ontogeny into account (through an index of developmental stages)

when studying adaptive responses of plant morphology to shading (Huber and Stuefer, 1997). Alternatively, growth allometries could be directly controlled by light-inducing contrasted plastic responses of the organs (Wright and McConnaughay, 2002). In that case, the contrasted plastic responses of the different organs could be explained by two hypotheses: (i) the photoperception mechanisms are the same, but the selection of morphogenesis responses is organ specific; or (ii) the photoperception sensitivity is organ-dependent (the different organs being more or less sensitive to the different spectral components of light). An approach to test these hypotheses is to consider light fluence-rate response curves. Indeed, if the hypothesis of organ-specific selection holds true, then all the responses that are selected in all the organs should display fluence-rate response curves with similar shapes (as they depend on the same process of photoperception). By contrast, if the photoperception itself is organ-dependent, then different fluence-rate response curves should be produced. However, until now, few BL response curves have been characterized (Wheeler et al., 1991; Dougher and Bugbee, 2001a), and possible changes in developmental rates caused by shading were not considered. Moreover, no PAR response curves independent of BL have ever been produced, as far as is known. The aim of this study was to identify and quantify the relative contributions of BL and PAR in the photocontrol of aerial morphogenesis of a typical shade-avoiding species (Trifolium repens L.). This species presented the advantage that the analysis of its shade-avoidance syndrome is not complicated by flowering time. Although this work required controlled environmental conditions, light conditions were selected to mimic those found in natural shade, in the range in which PAR and BL responses are likely to become significant (Ballare´ et al., 1991). Developmental ontogenic trends were quantified to investigate how morphogenetic processes are plastically affected by shading, separating responses of the rate of ontogenic development from specific plastic responses of organ growth. Additional light conditions were also studied to produce PAR and BL fluence-rate response curves to determine how each spectral component of light controls the different morphogenetic responses for each organ.

Materials and methods Deﬁning ecologically relevant artiﬁcial light treatments Each artificial light treatment was controlled to simulate the PAR and BL irradiances experienced by a plant that is shaded by a white clover canopy of a given leaf area index (LAI) in natural conditions. The PAR incident in the growth chambers was fixed to produce a typical natural incident light. The levels of PAR and BL irradiances to be applied in the artificial treatments were determined using the relationship between the fraction of transmitted light in the visible

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

BL and PAR photocontrols of aerial plant morphogenesis range and the LAI for a natural white clover canopy [PARtransmitted/ PARincident=exp(–0.594LAI); data compiled from Sinoquet et al., 1990] in order to correspond to that of natural shading (see below for details on the shade conditions retained). The fraction of light transmitted in the BL waveband was assumed to be similar to that of incident light, as the BL:PAR ratio has been shown to be independent of the amount of shading (Holmes, 1981; Messier and Bellefleur, 1988). As the aim was to vary independently BL and PAR fluence rates, each light treatment was designated as BLx_PARy where x is the fraction of light transmitted in the BL range and y that for PAR. When x and y are equal, the treatment corresponds to a neutral shade treatment that could be achieved under a natural canopy. Experimental design and artiﬁcial light treatments In the growth chambers, the incident light was fixed to simulate a typical day at the beginning of the main growing season for clover (early spring in central France; Simon et al., 1989), i.e. a mean daily incident PAR of 25.4 mol mÿ2 dÿ1, a mean daily BL of 7.6 mol mÿ2 dÿ1, and a 10 h photoperiod (10 year means, INRA Lusignan, France, 0.078E, 46.38N, meteorological data). Two neutral shade treatments, BL45_PAR45 and BL20_PAR20, were applied to study the effects of neutral shading. These fractions of transmitted light in BL and PAR correspond to ranges of natural shading (LAIs from 1.4 to 2.8) in which (i) PAR and BL responses are likely to become significant (Ballare´ et al., 1991) and (ii) shadeavoidance responses have been observed in natural conditions (Simon et al., 1989). A third treatment, BL20_PAR45, was used to measure quantitatively the separated contributions of PAR and BL reductions in this neutral shading (Fig. 1). The contribution of BL reduction at a constant PAR was assessed by comparing BL45_ PAR45 and BL20_PAR45 treatments. That of PAR reduction at a constant BL was assessed by comparing BL20_PAR45 and BL20_PAR20. Two additional light treatments, BL20_PAR100 and BL2_PAR45 (Fig. 1), were defined to establish PAR and BL fluence-rate response curves. BL20_PAR100, with no PAR reduction (i.e. with the same PAR level as that of the incident light) was preferred to a deep PAR reduction, to prevent the plants from being under the point of compensation of photosynthesis. BL2_PAR45 treatment, corresponding to 4 lmol mÿ2 sÿ1 BL in the present conditions, was retained as the effects of BL were sometimes found maximum for photon fluxes lower than 40 lmol mÿ2 sÿ1 BL (Wheeler et al., 1991; Dougher and Bugbee, 2001a). Moreover it is similar to the BL (–) treatment used by Gautier et al. (1997) in white clover. Fluence-rate response curves to BL at constant PAR could then be established considering BL45_PAR45, BL20_PAR45, and BL2_PAR45, and that of PAR at constant BL considering BL20_PAR20, BL20_PAR45, and BL20_PAR100. Each light treatment was applied in a separate growth chamber (Phytotron, Froid et Mesure, Angers, France), using a ceiling of metallic halide lamps (HQI, 400 W; Osram, France). In the BL45_PAR45 treatment no filter was used. In the other light treatments, PAR and BL irradiances required were obtained from a similar ceiling but using one layer of a filter (Lee Filters, ATOHAAS France SA Argenteuil, France; transmission curves of the filters are shown in Fig. 2A) on top of the plants. BL2_PAR45 was obtained by adding a deep straw filter (Lee Filter HT015); BL20_PAR100 by adding an amber-yellow filter (Lee filter 102); BL20_PAR45 by adding a straw tint filter (Lee Filter HT013), and BL20_PAR20 by adding a perforated reflector (Lee Filter 270). The light spectrum at plant level (Fig. 2B) was measured for each treatment using an LI-1800 spectroradiometer (Li-Cor Inc., Lincoln, NE, USA) and relevant actinic light fluxes were then computed for all the treatments (Table 1). For PAR, the standard definition of

2381

transmitted light in the PAR (%) BL20_PAR100 100

BL2 PAR45

BL20_PAR45

BL45_PAR45

BL reduction 45

PAR reduction

Neutral shading

20 BL20_PAR20

0 2

100

transmitted light in the BL (%)

Fig. 1. Experimental design. Each treatment was designated by BLx_PARy where x is the percentage light transmission in the BL range and y that for PAR. When x and y are equal, the treatment was describe as ‘neutral’. The comparison between treatments with different x was described as ‘BL reduction’, whereas that between treatments with different y was described as ‘PAR reduction’.

the PPFD (photosynthetic photon flux density at 400–700 nm) was used. For BL, fluence rate was measured as the photon fluence rate in the 350–500 nm waveband (Ahmad et al., 2002; Lin and Shalitin, 2003). In order to eliminate the confusion between BL and photosynthetic effects, the photosynthetic efficiency (light spectrum times the action spectrum of the leaf photosynthetic yield; Sager et al., 1988) was computed for all the light treatments and controlled to be decorrelated with BL (Table 1). In all treatments, the phytochrome photoequilibrium was maintained at its maximal ‘unshaded’ values (i.e. slightly higher than that of incident natural solar radiation), the UV fluence rate was negligible (Fig. 2) and the photoperiod was 10 h. Note that, from the commercially available filters, it was not possible to produce the levels of BL and PAR fluence rates matching exactly the targeted percentage of light transmitted. The greatest deviation from the values was 6% (Table 1). Plant material and growth conditions Cuttings of a single clone of white clover (Trifolium repens L. cv. Huia) were grown in individual pots (2031238 cm) filled with sterilized sand (Gautier et al., 1997). During the pre-experiment period, all the cuttings were grown together and subjected to the BL45_PAR45 treatment (Table 1). After 2 weeks, plants were selected for uniformity of (i) internode and petiole lengths and (ii) unfolding foliar stage using the decimal scale proposed for white clover by Carlson (1966) (10 stages from 0.1 when the leaf first becomes visible to 1 when the leaf is fully unfolded). Eight plants were then randomly assigned to each light treatment. Pots were watered liberally by excess solution eight times a day, using a complete nutrient solution containing nitrogen (Gastal and Saugier, 1986) to avoid any water or nutrient limitation. Relative air humidity

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

2382 Christophe et al.

Lee Filter 102 80

Light transmission (%)

data were used to calculate the usual indices of development representing the number of leaves of a given axis (cumulated Carlson stage; Gautier et al., 1997) and of the whole plant (plant cumulated Carlson stage; Christophe et al., 2003), in decimal units. Note that the development of the whole plant depends on both (i) the rate of leaf appearance on the axes and (ii) the rate of branch production.

100

Lee Filter HT013

40 Lee Filter 270 20 Lee Filter HT015 0 300

400

500

600

700

800

Wavelength (nm)

Photon flux density (µmol m-2 nm-1 s-1)

Rate of branch production per bud: The rate at which newly formed axillary buds were produced on new lateral branches was estimated by the slope of the relationship between the number of branches and the number of leaves of the whole plant produced over the same time interval (in Carlson decimal units), as proposed by Davies (1974) (and also called the site filling).

BL PAR

6 BL20 PAR100

BL2_PAR45 2 BL45_PAR45 BL20_PAR45 0 300

400

BL20_PAR20 500

600

Rate of leaf appearance: In each treatment, the rates of leaf appearance (RLA; leaf 8Cÿ1 dÿ1) on the whole plant and on each axis were estimated by fitting the number of leaves (in Carlson decimal units) that were visible on the whole plant and on each axis against thermal time. Thermal time was calculated as the cumulative degree-days from emergence, assuming a base temperature of 0 8C (Simon et al., 1989; Gautier et al., 1997). The balance between the rate of leaf appearance on the lateral branches and that on the main axis was estimated by the slope of the relationship between the number of leaves of each branch (in Carlson decimal units) and the number of leaves of the main axis (in Carlson decimal units) produced at the same time.

700

800

wavelength (nm)

Fig. 2. (A) Light transmission curves of the filters used in the experimental design. Continuous line, Lee Filter 270; dashes, Lee Filter HT015; dots, Lee Filter 102; dash-dot-dots, Lee Filter HT013. (B) Spectral photon distribution of incident radiation in the different treatments. Continuous line, BL20_PAR45; long dashes, BL20_PAR100; dots, BL20_PAR20; dash-dots, BL45_PAR45; dashdot-dots, BL2_PAR45.

was maintained constant at 80%. Air temperature was measured at plant height using six thermocouples per growth chamber and maintained at 23 8C day/20 8C night. A more complete description of the growth conditions is presented in Table 1. Note that with a 10 h photoperiod all the plants in all the treatments remained vegetative. Plant measurements Plant morphology: White clover (Trifolium repens L.) is a clonal stoloniferous plant with plagiotropic axes (main axis and lateral branches) and orthotropic petioles bearing leaves (a leaf consists of three leaflets). An axis can be viewed as a succession of phytomers comprising a node, an internode, an axillary bud, a leaf, and two nodal root primordia (Bell, 1984). Number of leaves: The number of leaves and their unfolding stage (using the Carlson decimal scale; Carlson, 1966) were recorded every 3 d, on each axis of the cutting, for eight plants per treatment. These

Plant growth: For each plant, the lengths of (i) all the internodes and (ii) the midribs and the petioles of all the leaves were measured every 3 d using callipers. For each treatment, the individual leaf area was estimated from an allometric relationship between the leaflet area and the midrib length determined from a sample of leaves using an Li-Cor planimeter (Li-3100, Li-Cor Inc.). In order to study plastic growth responses, the plant axis length and the plant leaf area were compared between treatments at the same plant developmental stage (Christophe et al., 2003). Thus the rates of the plant axis elongation and of the plant leaf expansion were estimated by fitting the plant axis length and the plant leaf area, respectively, against the number of leaves of the whole plant (in Carlson decimal units). Lengths of petioles were compared between treatments at their final stage of leaf development (when maximal length is achieved). Only the petiole lengths of leaves 3 and 4 of the main axis are given, as they were the only petioles that completed the entire visible growth during the light treatments. Statistical analysis Final plant measurements were compared using a one-way analysis of variance (general linear model with treatment as a fixed factor; procedure GLM in SAS). The regression and analysis of variance of the various ontogenicdependent variables were determined using a mixed general linear model, with the mixed procedure (Proc Mixed) in SAS version 6.12 (Littell et al., 1996). The effects of individual plants (nested within the treatment) were added as random effects in the error of the model, using the repeated option in Proc Mixed. The variance–covariance matrix of the error was specified by a ‘spatial’ power structure [SP (Pow)]. SP (Pow) takes into account uneven intervals in the data sets. For the present data sets, it displayed the highest Akaike’s information criterion (AIC) and Schwarz’ Bayesian criterion (SBC) (as defined in SAS version 6.12; Littell et al., 1996). For the study of developmental kinetics, kinetics were fitted using a linear or polynomial regression model. The light treatments were first tested as fixed effects on the heterogeneity of the parameters of this model (slopes, higher order coefficients). If this test was significant, the parameters were compared using the estimate statement; otherwise a covariance analysis was performed. When the common slope was not different from zero, the model was reduced to a (mixed) analysis of variance, with the light treatment effects as fixed effects.

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

BL and PAR photocontrols of aerial plant morphogenesis

2383

Table 1. Environmental conditions Characteristics of light treatments were determined at plant height, using a spectroradiometer (Li-1800, Li-Cor, Lincoln, NE, USA). The phytochrome equilibrium state (/c) and the photosynthetic efficiency (light spectrum multiplied by photosynthetic yield) were calculated as described by Sager et al. (1988). Mean air temperature was monitored at plant height. Each treatment was designated as BLx_PARy where x is the percentage light transmission in the BL range and y that for PAR. The exact fractions of transmitted light were calculated afterwards from the exact percentage transmission through filtering. Treatments Mean air temperature day/night (and standard deviation) Relative humidity Light source Photoperiod Photosynthetic efficiency (350–750 nm) Photosynthetic photon flux density (400–700 nm) Mean daily PAR Blue light fluence rate (350–500 nm) Mean daily BL Phytochrome photoequilibrium Transmitted light in the PARa Transmitted light in the BLa

BL2_PAR45 Tair (8C)

22.860.5/20.160.3

80 HQI+a Lee filter HT015 10 317

ÿ1

hd y lmol mÿ2 sÿ1 PPFD lmol mÿ2 sÿ1

338

mol mÿ2 dÿ1 BL lmol mÿ2 sÿ1 mol mÿ2 dÿ1 /c

12.2 4

(% of the incident light) (% of the incident light)

0.14 0.78

1.8

BL20_PAR100 23.060.2/20.060.2

BL20_PAR45 23.160.9/20.160.3

BL45_PAR45 22.760.6/2060.1

BL20_PAR20 23.060.4/20.060.1

HQI+a Lee filter 102 HQI+a Lee filter HT013

HQI

HQI+a Lee filter 270

646

334

287

141

705

301

333

163

25.4 36 1.3 0.78 100 17

10.8 38 1.37 0.78

12 83 3 0.77

5.9 36 1.3 0.77

42.5

The ‘incident light’ reference corresponded to the natural incident light during a typical day in early spring, with a mean daily PAR=25.4 mol mÿ2 d , a mean daily BL=7.6 mol mÿ2 dÿ1. a

ÿ1

Whenever a statistically significant effect was found for at least one light treatment, fluence-rate response curves were constructed using means values (and standard error).

process, the effects of the neutral shading and the contribution of the effects of PAR and BL shading were analysed by quantifying modulations of these ontogenic trends.

Results

Ontogenic development trends In all the treatments, the plants had a similar initial developmental stage and their number of leaves increased exponentially during development. As a consequence, the rate of the ontogenic development can be easily quantified by estimating the slope of the changes in the number of leaves on a natural log scale against thermal time (Fig. 3A). By contrast, the rate of leaf appearance on the main axis (RLA, slope of the curve in Fig. 3B) was constant over time for the period considered (23 d). The balance between the rate of leaf appearance on the primary branches and that on the main axis (i.e. the slope of the curve in Fig. 3C) was also constant over the time. Additionally, as there was no significant difference in this ratio between the first four branches, the data were pooled. In all the treatments, this ratio was below 1, indicating that the production of leaves of the main axis was always superior to that of the primary branches. The relationship between the total number of leaves and the total number of branches was complex and exhibited a cubic pattern when the plant had <10 branches (Fig. 3D). This suggests that the rate of branch production

Morphology of shaded plants

At the end of experiment, the effects of neutral shading on plant morphology were tested by comparing plants grown under the two neutral shade treatments (Table 2). Shaded plants had produced about half the number of leaves and branches than plants grown under high-light conditions. Shaded plants also presented a reduction in the number of leaves on the branches and in the total leaf area. By contrast, the shaded plants produced petioles much longer than plants under high-light conditions. The number of leaves on the main axes and the total stem length were not affected by shading (Table 2). Neutral shading induced changes in both plant development and growth, leading to important differences in plant morphology. Morphogenetic processes were thus studied during ontogeny in order to separate responses of the rate of the ontogenic development and their possible indirect consequences on organ growth from direct plastic responses of organs. Then, for each morphogenetic

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

2384 Christophe et al. Table 2. Comparison of the morphology of plants grown under two neutral shade treatments BL45_PAR45 and BL20_PAR20, after 23 d (i.e. 786 degree-days) Values are means of eight plants (6 standard deviation). Data were analysed by a one-way ANOVA and the asterisks indicate significant differences between treatments (P <0.05). BL45_PAR45

BL20_PAR20

Probability value P

Total no. of leaves (decimal units) Total no. of branches No. of leaves in the main axis (decimal units) No. of leaves in the first four primary branches (decimal units) Total leaf area (mm2) Total stem length (mm) Length of petiole 4 (mm)

29.768.2 7.863.2 8.261.3 3.760.8 11714.861310.9 198.86106.2 80.265.9

16.7*62.6 4.8*60.8 8.160.8 2.1*60.9 5612.3*61327.9 99.3638.4 113.9*613.7

0.0001 0.0045 0.8335 0.0001 0.0027 0.065 0.0004

Number of leaves of the main axis (in decimal units)

Log (total number of leaves, in decimal units)

Light treatment

300

400

500

600

700

14 12 10 8 6 4 2 0

800

300

400

Time (degree days)

600

700

800

14 12

Total number of leaves (in decimal units)

Number of leaves of the fourth primary branches (in decimal units)

500

Time (degree days)

10 8 6 4

5 2 0

Number of leaves of the main axis (in decimal units)

Total number of branches

Fig. 3. Characteristics of leaf production and branching. Each point represents an individual plant. Open squares, BL45_PAR45; closed circles, BL20_PAR45; closed triangles,, BL20_PAR20. The lines are the mixed linear regression. (A) Changes in the total number of leaves (in Carlson decimal units) with thermal time. Dashed line: BL20_PAR20, log(y)= –0.516(60.148)+0.0041(60.0002)x; continuous line: BL45_PAR45, BL20_PAR45, log(y)= –0.872(60.091)+0.0054(60.0001)x. (B) Changes in the number of leaves on the main axis (in Carlson decimal units) with thermal time. Continuous line: BL20_PAR45, y= –2.058(60.201)+0.0145(60.0003)x; dashed line: BL45_PAR45, BL20_PAR20, y= ÿ2.058(60.201)+0.0122(60.0004)x. (C) Changes in the number of the first four primary branches plotted as a function of the number of leaves on the main axis (in Carlson decimal units). Continuous line: BL20_PAR45, y= –3.341(60.121)+0.768(60.021)x; dashed-and-dotted line: BL45_PAR45, y= –3.341(60.121)+0.880(60.05)x; dashed line: BL20_PAR20, y= –3.341(60.121)+0.695(60.05)x. (D) Changes in the total number of leaves (in Carlson decimal units) with the total number of branches. Continuous line: y=3.168(6 0.186)+2.627(60.415)x–0.430(60.197)x2+0.08(60.02)x3.

(i.e. the inverse of the slope of this curve) was not constant over the developmental range investigated (up to six branches). This rate increased rapidly during the early

stage of plant development, reaching a peak when the plant had two branches, and then declined continuously (data not shown). Considering growth of the whole plant, the total

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

BL and PAR photocontrols of aerial plant morphogenesis

leaf area increased proportionally with the number of leaves of the whole plant (Fig. 4A), whereas the plant axis length increased as a three-quarters power law function of the total number of leaves (Fig. 4B). Thus a three-quarter power allometry between total internode length and total leaf area per plant was maintained during development. Despite significant differences in the final petiole length of leaves 3 and 4 (Fig. 4C) there was no significant interaction between shade treatments and leaf rank (P=0.3944) and all the petioles on all the branches were similarly affected, even when correcting for total leaf number (data not shown). Effects of neutral shading and PAR and BL reductions on development of the whole plant

The neutral shading significantly slowed down the relative rate of the ontogenic development of the plant (by –0.0013, –24%, Fig. 3A; Table 3), and this was shown to be caused entirely by the reduction in PAR (Table 3). In the present conditions, the rate of branch production was not signifi-

cantly affected by either neutral shading or by BL and PAR reductions (Fig. 3D). Only the rate of leaf appearance (RLA) on the different axes was differentially affected by the treatments (Fig. 3B, C; Table 3). Under the neutral shading, the RLA on the main axis was not significantly modified (Table 3) and its mean value was 0.0122 leaf 8Cÿ1 dÿ1. However, individually, both BL and PAR reductions affected the RLA on the main axis (Table 3). BL reduction significantly increased RLA by +0.002 leaf 8Cÿ1 dÿ1 whereas, on the contrary, PAR reduction decreased RLA by –0.002 leaf 8Cÿ1 dÿ1. This explains the absence of neutral shading effects. The neutral shading significantly reduced the ratio of the rates of leaf appearance on the main axis and primary branches by 21% (Table 3). This was the consequence of a reduction of the RLA on the primary branches (–0.002 leaf 8Cÿ1 dÿ1, ÿ20%). BL reduction decreased this ratio by 13% (Table 3), which was due to a greater increase of the RLA on the main axis (three times more) than that on the branches (+0.0004 leaf 8Cÿ1 dÿ1, +4%). PAR reduction B

10000

Total internode length (mm)

Total leaf area (mm2)

8000

6000

4000

2000

Total number of leaves (in decimal units)

250

200

150

100

Total number of leaves (in decimal units)

140 120

Final petiole length (mm)

2385

phytomer 3 phytomer 4

100 80 60 40 20 0

BL45_PAR45 BL20_PAR45 BL20_PAR20

Fig. 4. Characteristics of plant growth. Each point represents an individual plant. Open squares, BL45_PAR45; closed circles, BL20_PAR45; closed triangles, BL20_PAR20. The lines are the mixed linear regression. (A) Changes in the total leaf area per plant as a function of the total number of leaves (in Carlson decimal units). Continuous line: y=351.93(610.70)x. (B) Changes in the total internode length per plant as a function of the total number of leaves (in Carlson decimal units). Continuous line: loge(y)=1.326(60.03)loge(x)+0.867(6 0.07). (C) Final length of petioles 3 and 4 of the main axis.

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

2386 Christophe et al. Table 3. Parameters of the plant ontogenic processes under the three light treatments BL45_PAR45, BL20_PAR45, and BL20_PAR20 For each ontogenic parameter, the effects of neutral shading were assessed by comparing BL45_PAR45 and BL20_PAR20, the effects of corresponding BL reduction were assessed by comparing BL45_PAR45 and BL20_PAR45 and those of corresponding PAR reduction by comparing BL20_PAR45 and BL20_PAR20. The values correspond to the estimated parameters (mean 6standard error). The letters indicate significant differences between treatments (P <0.05). Parameters of plant ontogenic development

Relative rate of the plant ontogenic development (leaf leafÿ1 8Cÿ1dÿ1) Rate of leaf appearance on the main axis (leaf 8Cÿ1 dÿ1) Ratio of the rates of leaf appearance on the main axis and on the first four primary branches Rate of the total leaf area expansion (mm2 leafÿ1) Mean petiole length of the phytomers 3 and 4 on the main axis (mm)

also decreased this ratio by 10.5% (Table 3) but, in this case, it was due to a greater reduction of the RLA on the branches (–0.003 leaf 8Cÿ1 dÿ1, –23%) than that on the main axis. Effects of neutral shading and PAR and BL reductions on plant growth No significant modification in the rate of the plant leaf expansion and of the plant axis elongation was observed in response to neutral shading, BL and PAR reductions (Fig. 4; Table 3). By contrast, the final petiole length of the mature petioles was significantly increased by the neutral shading (Fig. 4C; Table 3). Neutral shading induced a significant increase of 28 mm of the petiole length of leaves 3 and 4 of the main axis (corresponding to an increment of 35%). BL reduction increased the petiole length of leaves 3 and 4 by 12.5 mm (corresponding to an increment of 16%). PAR reduction increased the petiole length of leaves 3 and 4 by 15.5 mm (corresponding to an increase of 17%).

Light treatments BL45_PAR45

BL20_PAR45

BL20_PAR20

0.005560.0002 a 0.012160.0006 a 0.8960.05 a

0.006060.0002 a 0.014560.0003 b 0.77760.02 b

0.004260.0002 b 0.012360.0005 a 0.69560.05 c

356.64641.6 a 78.662.0 a

347.85622.30 a 91.062.0 b

376.14617.8 a 106.562.8 c

the rates of leaf appearance on the main axis and on the branches did not show such a saturating pattern, although it seemed to display damped increments between 300 and 750 lmol mÿ2 sÿ1. Lateral axes probably displayed a different range and shape of PPFD sensitivity than the main axis. Also surprising, the rate of leaf expansion was never significantly affected by the levels of PAR in the conditions (P=0.2395), contrary to the petiole. The rate of leaf appearance on the main axis increased for moderate BL reductions (Fig. 5B) and then levelled off with BL fluence rates lower than 38 lmol mÿ2 sÿ1, or at least reach some maximum for a deep BL reduction. The response of the ratio between the rates of leaf appearance on the main axis and on the branches was distinct from that of the rate of leaf appearance on the main axis, as noted for PAR response, but to a higher extent. Indeed a deep BL reduction decreased very significantly the ratio on the leaf appearance rates. Contrary to the PAR responses, BL reduction increased both the rate of leaf expansion and the petiole final length to similar extents of, respectively, +35% and +40% (between extreme treatments).

PAR and BL ﬂuence-rate response curves

The morphogenetic parameters shown to be significantly affected by at least one light treatment were studied further by the analysis of the fluence-rate response curves, including the two additional light treatments BL20_PAR100 and BL2_PAR45. Plant axis length and branching production were not affected under the present treatments (P=0.2691 and P=0.074, respectively). The shape of the fluence-rate response curves varied between responses (Fig. 5A, B). Two response curves to PAR (the rate of leaf appearance on the main axis and the petiole length; Fig. 5A) displayed no significant changes between 300 and 750 lmol mÿ2 sÿ1 despite a 2.5-fold difference in irradiance, whereas they displayed clear and opposite responses under 300 lmol mÿ2 sÿ1. These two responses thus probably tend to saturate for irradiance over 300 lmol mÿ2 sÿ1 or, at least, reach a maximum between 300 and 750 lmol mÿ2 sÿ1. Surprisingly, the ratio between

Discussion Varying BL and PAR independently within ecologically relevant ranges The action spectra of phytochromes, cryptochromes, and photosynthetic efficiency largely overlap (Scha¨fer et al., 1983). Therefore, uncoupling experimentally blue light (BL) and photosynthetically active radiation (PAR) within a realistic broad-waveband light without affecting phytochrome photoequilibirum (/c) is not simple. The experimental design reported here is the first as far as is known that was constructed (i) to quantify the relative contributions of BL and PAR in the photocontrol of plant morphogenesis, and (ii) to control the level of artificial light irradiance in BL and PAR ranges to correspond to that of natural shading (as defined with a relationship between the fraction of transmitted light and the leaf area index

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

BL and PAR photocontrols of aerial plant morphogenesis

2387

0.020 0.018 0.016

0.014 a

0.012 0.010

200

400

PAR (µmol

600

m-2

800

Ratio of the leaf appearance rates

Leaf appearance rate on the main axis (leaves degree-days-1)

0.8 0.7

0.6 0.5 0.4

200

400

600 -2

800

-1

PAR (µmol m s ) 120

Final length of petiole 3 (mm)

Rate of leaf expansion (mm2 leaves-1)

0.9

s-1)

550 500 450 a

400

350 300

1.0

200

400

PAR (µmol

600

m-2

110 100 90 80

70 60

800

200

400

600

800

PAR (µmol m-2 s-1)

s-1)

0.020 0.018 a

0.016 a 0.014

b 0.012 0.010

BL (µmol

m-2

100

Ratio of the leaf appearance rates

Leaf appearance rate on the main axis (leaves degree-days-1)

0.8 0.7 0.6 a 0.5 0.4

100

BL (µmol m-2 s-1) 120

500

Final length of petiole 3 (mm)

Rate of leaf expansion (mm2 leaves-1)

0.9

s-1)

550

450 b

400 350 300

1.0

BL (µmol

m-2

s-1)

100

110 100 90

70 60

BL (µmol

m-2

100

s-1)

Fig. 5. Fluence-rate response curves of morphogenetic processes: the rate of leaf appearance on the main axis, the ratio of the rates of leaf appearance on the branches and on the main axis, the rate of the leaf expansion and the final length of petiole 3 of the main axis. Each point is the mean of eight plants and the vertical bars represent the standard error of the mean. The different letters indicate significant differences between treatments (P <0.05). (A) Effects of three PAR levels: 163 lmol mÿ2 sÿ1 (BL20_PAR20); 301 lmol mÿ2 sÿ1 (BL20_PAR45); 705 lmol mÿ2 sÿ1 (BL20_PAR100). (B) Effects of three BL levels: 4 lmol mÿ2 sÿ1 (BL2_PAR45); 38 lmol mÿ2 sÿ1 (BL20_PAR45); 83 lmol mÿ2 sÿ1 (BL45_PAR45).

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

2388 Christophe et al.

(LAI; see Materials and methods). It is a generalization of the method of ‘light equivalence’ designed to analyse the BL effects at constant /c using two monochromatic lights (Scha¨fer et al., 1983). Dougher and Bugbee (2001a) produced BL variations for two levels of PAR by combining different light sources and filters in a way similar to the one used in the present study. But, in their experimental design, the relative contribution of the effects of BL and of PAR in shading could not be fully analysed. Note also that the present approach can be conducted over the whole range of plant development (the light treatments can be applied at different times during plant development) and the whole range of possible shade. Indeed, it would be interesting to extend the experimental design (with additional light treatments) to study the effects of /c, and the possible interactions between the different spectral components in the control of shade-avoidance in natural ‘green shading’. Moreover, such experiments could be repeated using other species where photoreceptor mutants are available (for example, Arabidopsis thaliana; Kozuka et al., 2005). Such a combination with genetic analysis could be very useful to analyse the mechanisms of photocontrol as the two approaches are complementary. Note, however, that the use of the LAI reference for defining the range of natural shading might not be ecologically meaningful for a ruderal plant such as Arabidopsis thaliana that may never live in canopies. Neutral shading produced typical shade-avoidance responses

Neutral shading primarily slowed down ontogenic plant development. At the end of the experiment (i.e. at the same chronological age), shaded plants were smaller than plants under high-light conditions with a total number of leaves reduced by half. Additionally, shaded plants showed a reduction in the final number of branches, a reduction in the development of existing branches than that of the main axis, and an enhancement of the petiole elongation associated with a reduction in total leaf area per plant. All these responses are consistent with previous work on the effects of neutral shading in a range of species (Lo¨tscher and No¨sberger, 1997; Stuefer and Huber, 1998; Tsukaya et al., 2002) and are typical of what is termed the shadeavoidance syndrome (Smith and Whitelam, 1997). BL and PAR control the shade-avoidance syndrome through antagonist effects on leaf production and additive effects on petiole elongation PAR and BL reductions were differently involved in producing the shade-avoidance responses of the plants grown under neutral shading. PAR reduction induced a decrease in the leaf production on the main axis, probably through the reduction of photosynthesis. This inhibitory

effect of PAR was more pronounced on the branches. By contrast, a BL reduction protected the leaf production on the main axis, but its promoting effects on the rate of leaf appearance declined steeply on the branches. These responses of the rate of leaf appearance to BL, in an axisdependent way is in accordance with previous reports in white clover (Gautier et al., 1998). As a result, in neutral shading, the combined contributions of the antagonistic PAR and BL effects maintained the leaf production on the main axis, and decreased it in the branches, as described in the shade-avoidance syndrome. Increasing shade might not change this trend as BL stimulation of the leaf production on the main axis levelled off in deeper BL reduction. Thus, differential PAR and BL controls of leaf production appear to be important in the establishment of the linear growth pattern involved in light-foraging behaviour. By contrast to the antagonistic effects of PAR and BL reductions on leaf production, BL and PAR reductions displayed additive promoting effects on petiole elongation. Both contributed to the increase in petiole length by the neutral shading studied, with an almost equal share. BL reduction accounted for 45% and PAR reduction for 55% of the neutral shading effects. The effect of BL reduction on petiole elongation confirms previous reports (Gautier et al., 1997; Kozuka et al., 2005). But, the quantification of a PAR control in this shade-avoidance response is a novel and major finding. Increased shade might not change the additive behaviour of the BL and PAR effects as the response curves for petiole length tended to be parallel with increasing shade (Fig. 5). Effects of neutral shading on branching and on leaf area versus stem growth were only the consequence of the effects of PAR on leaf appearance The strong decrease of the leaf appearance rates due to PAR reduction, despite its mitigation by BL effects, clearly produced indirect effects on branching. No direct effect of PAR and BL reductions was found on the relative rate of bud outgrowth. These results indicated that the effects of shade on branching usually reported in the literature for dicots (for example, for white clover; Lo¨tscher and No¨sberger, 1997), are probably an indirect consequence of the effects on the leaf appearance rate reducing the amount of axillary buds. By the same token, a clear allometry between total stem length and leaf area per plant was found during development (Fig. 4). Except in the deep BL reduction, no direct effect of BL or PAR was found on total stem length or leaf area developmental trend (Fig. 4). The effects of neutral shading on leaf area versus stem growth were only an allometric consequence of the reduction of the leaf appearance rate by decreased PAR. In the literature, there is controversy concerning the effects of neutral shading on internode elongation (for a review see de Kroon and

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

BL and PAR photocontrols of aerial plant morphogenesis

Hutchings, 1995) and on lamina expansion (Lo¨tscher and No¨sberger 1997; Tardieu et al., 1999). This could be due to the fact that the ranges of neutral shading were different and that, generally, possible changes in the rates of leaf appearance were not monitored. The present results demonstrated that an analysis of the growth responses during ontogenesis is crucial to identify plastic responses directly affected by shading (see also Huber and Stuefer, 1997; Wright and McConnaughay, 2002) and to get response curves informative about the underlying mechanisms of perception. PAR and BL actions also involved organ-speciﬁc differences in sensitivity

The responses of five types of organs were studied: main apices, lateral apices, internodes, petioles, and laminas. Fluence-rate response curves differed both qualitatively and quantitatively between organs and between processes. For example, the response of lateral apices differed from that of primary apices for both PAR and BL. This difference in the shape of the response curves might track possible mechanistic differences. A diversity of shapes was also described for other responses to BL reduction by Dougher and Bugbee (2001a). Thus, in light-grown plants, the control of the shade-avoidance responses might not be explained by similar mechanisms of photoperception and signalling pathways which are just on or off depending on the organ. In both BL and PAR responses, the control involves organand process-specific quantitative differences in sensitivity. Two distinct mechanisms for light in the PAR waveband? BL reduction had a global enhancing effect on the production of leaves and on petiole growth. By contrast, PAR reduction had an opposite effect on the production of leaves, whereas it also enhanced petiole growth (even when corrected for changes in developmental rates). This suggests two distinct mechanisms for the action of PAR. The negative effect on the rate of leaf appearance is likely to correspond to the expected reduction of growth and development due to decreased photosynthesis, i.e. a trophic control (Ryle et al., 1992). Indeed the response curve for the rate of leaf appearance to PAR displays a downward concavity and an upper asymptote, like the response curve of the net CO2 assimilation. From the literature, there are two putative candidates for PAR control of petiole elongation. The first one is a photon-counting effect mediated by phytochromes (i.e. the high irradiance responses mode of phytochrome action; Casal et al., 1998; Nagy and Scha¨fer, 2002). Indeed, in the present conditions, both R and FR fluence rates are highly correlated with PAR (although the R:FR ratio and /c were constant). This would be consistent with previous inferences suggesting that spectral

2389

wavebands distinct from BL (350–500 nm) may be directly implicated in the photomorphogenetical control of growth (Ballare´ et al., 1991; Dougher and Bugbee, 2001b). The alternative candidate for PAR control is sugar signalling. Indeed, Kozuka et al. (2005), using ABA and ethylene pathway mutants in Arabidopsis, argued that sugar signalling (which is likely to be PAR-dependent) could be involved in petiole and leaf blade responses and interacted with BL perception in an organ-specific manner. Concerning the effects of BL reduction, they are unlikely to be neither trophic-mediated responses, nor phytochromemediated high irradiance responses, as no correlation exists in this work between fluence rates in BL and in any other spectral waveband. This is in accordance with the already well-substantiated evidence that BL can trigger light signalling pathways for plant morphogenetic processes acting through specific photoreceptors like cryptochromes or phototropins. Moreover, although interactions between cryptochromes and the low fluence-rate mechanism of phytochrome perception have been documented genetically and molecularly (Nagy and Scha¨fer, 2002; Lin and Shalitin, 2003), they are probably not involved here as the phytochrome photoequilibrium (/c) was kept constant in this experiment. Control by BL was thus likely to involve the BL signalling pathway on its own.

Conclusion The experimental design presented here is suited to quantify the relative contribution of each spectral component of natural shading on plant morphogenesis, as illustrated here for BL and PAR. Separating and quantifying the morphogenetic effects of BL and PAR reductions revealed that both are involved in the control of the contrasted responses that produce the shade-avoidance syndrome. However, only two responses were directly controlled by both BL and PAR reductions: leaf appearance and petiole extension. All the other typical responses to shading were indirect consequences of changes in the leaf appearance rates. BL reduction was promotive on both responses, whereas PAR reduction was inhibitory for leaf appearance and promotive on petiole expansion. However, the sensitivity to BL or PAR varied between the different apices. These differences are very important in producing the shade-avoidance syndrome. This demonstrates that the quantitative aspects of BL and PAR sensitivity have to be considered to understand the control of the shade-avoidance responses, in addition to phytochrome-mediated responses to the R:FR ratio.

Acknowledgements We thank Dr F Tardieu, Dr S Cookson, the two anonymous referees for fruitful comments on the manuscript, and S Cookson for revising the English text.

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009

2390 Christophe et al.

References Ahmad M, Grancher N, Heil M, Black RC, Baldissera G, Galland P, Lerdemer D. 2002. Action spectrum for cryptochrome-dependent hypocotyl growth inhibition in Arabidopsis. Plant Physiology 129, 774–785. Ballare´ CL, Scopel AL, Sańchez RA. 1991. Photocontrol of stem elongation in plant neighbourhoods: effects of photon fluence rate under natural conditions of radiation. Plant, Cell and Environment 14, 57–65. Ballare´ CL, Scopel AL, Sańchez RA. 1997. Foraging for light: photosensory ecology and agricultural implications. Plant, Cell and Environment 20, 820–825. Bell AD. 1984. Dynamic morphology: a contribution to plant population ecology. In: Dirzo R, Sarukhan J, eds. Perspectives on plant population ecology. Sunderland, UK, Sinauer Associates, 48–65. Carlson GE. 1966. Growth of clover leaves: developmental morphology and parameters at ten stages. Crop Science 6, 293–294. Casal JJ, Sańchez RA, Botto JF. 1998. Modes of action of phytochromes. Journal of Experimental Botany 49, 127–138. Christophe A, Moulia B, Varlet-Grancher C. 2003. A quantitative analysis of the three dimensional spatial colonization by a plant as illustrated by white clover (Trifolium repens L.). International Journal of Plant Science 164, 359–370. Davies A. 1974. Leaf tissue remaining after cutting and regrowth in perennial ryegrass. Journal of Agricultural Science 82, 165–172. Dougher TAO, Bugbee B. 2001a. Differences in the response of wheat, soybean and lettuce to reduced blue light radiation. Photochemistry and Photobiology 73, 199–207. Dougher TAO, Bugbee B. 2001b. Evidence for yellow light suppression of lettuce growth. Photochemistry and Photobiology 73, 208–212. de Kroon H, Hutchings MJ. 1995. Morphological plasticity in clonal plants: the foraging concept reconsidered. Journal of Ecology 83, 143–152. Franklin KA, Whitelam GC. 2005. Phytochromes and shadeavoidance responses in plants. Annals of Botany 96, 169–175. Gastal F, Saugier B. 1986. Alimentation azoteé et croissance de la fe´tuque e´leveé. I. Assimilation du carbone et re´partition entre organes. Agronomie 6, 157–166. Gautier H, Varlet-Grancher C, Baudry N. 1997. Effects of blue light on the vertical colonization of space by white clover and their consequences for dry matter distribution. Annals of Botany 80, 665–671. Gautier H, Varlet-Grancher C, Baudry N. 1998. Comparison of horizontal spread of white clover (Trifolium repens L.) grown under two artificial light sources differing in their content of blue light. Annals of Botany 82, 41–48. Grime J. 1981. Plant strategies in shade. In: Smith H, ed. Plant strategies in shade. London: Academic Press, 159–186. Holmes MG. 1981. Spectral distribution of radiation within plant canopies. In: Smith H, ed. Plant and the daylight spectrum. Proceedings of the first international symposium of the British Photobiology Society. London: Academic Press, 147–157. Huber H, Stuefer JF. 1997. Shade-induced changes in the branching pattern of a stoloniferous herb: functional response or allometric effect? Oecologia 110, 478–486.

Kozuka T, Horiguchi G, Kim GT, Ohgishi M, Sakai T, Tsukaya H. 2005. The different growth responses of the Arabidopsis thaliana leaf blade and the petiole during shade avoidance are regulated by photoreceptors and sugar. Plant Cell Physiology 46, 213–223. Lin CT, Shalitin D. 2003. Cryptochrome structure and signal transduction. Annual Review of Plant Biology 54, 469–496. Littell RC, Milliken GA, Stroup WW, Wolfinger RD. 1996. SAS system for mixed models. In: SAS system for mixed models. Cary, NC: SAS Institute Inc. Lo¨tscher M, No¨sberger J. 1997. Branch and root formation in Trifolium repens is influenced by the light environment of unfolded leaves. Oecologia 111, 499–504. Messier C, Bellefleur P. 1988. Light quantity and quality on the forest floor of pioneer and climax stages in a birch-beechsugar maple stand. Canadian Journal of Forestry Research 18, 615–621. Nagy F, Scha¨fer E. 2002. Phytochromes control photomorphogenesis by differentially regulated, interacting signaling pathways in higher plants. Annual Review of Plant Biology 53, 329–355. Ryle GJA, Powell CE, Davidson IA. 1992. Growth of white clover, dependent on N fixation, in elevated CO2 and temperature. Annals of Botany 70, 221–228. Sager JC, Smith WO, Edwards JL, Cyr KL. 1988. Photosynthetic efficiency and phytochrome photoequilibria determination using spectral data. Transactions of the ASAE 31, 1882–1889. Scha¨fer E, Fukshansky L, Shropshire WJ. 1983. Action spectroscopy of photoreversible pigment systems. In: Shropshire Jr W, Mohr H, eds. Photomorphogenesis. Berlin: Springer-Verlag, 39–68. Simon JC, Gastal F, Lemaire G. 1989. Compe´tition pour la lumie`re et morphologie du tre`fle blanc (Trifolium repens L.): e´mission des feuilles et des ramifications. Agronomie 9, 383–389. Sinoquet H, Moulia B, Gastal F, Bonhomme R, Varlet-Grancher C. 1990. Modeling the radiative balance of the components of a well-mixed canopy: application to a white clover–tall fescue mixture. Acta Oecologica 11, 469–486. Smith H. 2000. Phytochromes and light signal perception by plants: an emerging synthesis. Nature 407, 585–591. Smith H, Whitelam GC. 1997. The shade avoidance syndrome: multiple responses mediated by multiple phytochromes. Plant, Cell and Environment 20, 840–844. Stuefer JF, Huber H. 1998. Differential effects of light quantity and spectral light quality on growth, morphology and development of two stoniferous Potentilla species. Oecologia 117, 1–8. Tardieu F, Granier C, Muller B. 1999. Modelling leaf expansion in a fluctuating environment: are changes in specific leaf area a consequence of changes in expansion rate. New Phytologist 143, 33–43. Tsukaya H, Kozuka T, GyungTae K. 2002. Genetic control of petiole length in Arabidopsis thaliana. Plant and Cell Physiology 43, 1221–1228. Wheeler RM, Mackowiak CL, Sager JC. 1991. Soybean stem growth under high-pressure sodium with supplemental blue lighting. Agronomy Journal 83, 903–906. Wright D, McConnaughay KDM. 2002. Interpreting phenotypic plasticity: the importance of ontogeny. Plant Species Biology 17, 119–131.

Downloaded from http://jxb.oxfordjournals.org by on 18 October 2009