THE UNIVERSITY OF WESTERN AUSTRALIA FACULTY OF NATURAL AND AGRICULTURAL SCIENCES SCHOOL OF PLANT BIOLOGY

Effects of sclerophylly on photosynthesis and gas diffusion

Foteini Hassiotou (B.Sc. Hons)

This thesis is presented for the degree of Doctor of Philosophy

SEPTEMBER 2009

ABSTRACT

Sclerophylly comprises a suite of structural traits that result in tough long-lived leaves, but which also have the potential to influence leaf photosynthetic performance. Sclerophyllous traits such as leaf dry mass per area (LMA), the abundance of sclerified tissues and cell wall thickness, have been shown to influence the conductance to CO2 diffusion in the mesophyll

(gm), and through it, the rates of CO2 assimilation per unit leaf area (Aarea). However, key aspects of the photosynthetic process at the high end of the LMA spectrum and the conditions in which photosynthesis takes place at the tissue and cellular level are not well understood. The present study focused on the impact of leaf structure on CO2 diffusion and photosynthesis in the genus Banksia, which displays a great diversity of leaf morphologies, with the aim to determine whether high-LMA leaves differ from lower-LMA leaves in the organisation of the mesophyll or if the mesophyll itself is also different in its physiology.

A prominent leaf feature of many Banksia species is the presence of epidermal invaginations called crypts on the abaxial surface, which host the stomata. Stomatal crypts have been assumed to have a transpiration-reducing function. However, the occurrence of species with crypts in both wet and arid environments suggests that the primary role of these structures may not be moderation of water loss. The diffusion resistance of stomatal crypts was estimated in ten Banksia species using simple equations formulated for perforated or porous layers, and was also modelled in detail using finite-element modelling. Crypts reduced leaf transpiration by less than 15% compared with non-encrypted, superficially positioned stomata. Moreover, the trichomes that are often present within the crypts, and have also been assumed to reduce transpiration, had virtually no influence on transpiration. An alternative hypothesis was formulated that crypts facilitate CO2 diffusion to adaxial palisade cells in thick leaves, which was supported by evidence showing that stomatal encryption becomes more pronounced as leaf thickness and other indicators of sclerophylly increase. Furthermore, the possibility that crypts increase photosynthetic water-use efficiency was examined using an electrical resistance analogue model. This showed that crypts improve water-use efficiency only when the diffusivities for water vapor and CO2 in the crypts differ from those at the stomatal level. It was also

[iii]

ABSTRACT

demonstrated that the greater the part of the resistance that is due to stomata and crypts relative to mesophyll, the greater the benefit for diffusion of CO2 relative to water vapour.

Interrelationships between leaf structural traits and photosynthetic characteristics were investigated in 49 Banksia species and subsets of this group, and the contributions of the two components of LMA, leaf thickness and density, to the variability in LMA observed were determined. Leaf thickness and density contributed similarly to variation in LMA, but to different extents in different species, indicating that there are various ways to be sclerophyllous in this genus. The increasing amount of leaf structural tissues with increasing LMA resulted in lower mass-based chlorophyll, nitrogen and thus, photosynthesis (Amass) at high LMA. In contrast, mesophyll volume fraction and Aarea were independent of LMA. The lack of a correlation between Aarea and LMA was probably due to the increase in mesophyll volume per unit leaf area and the concurrent decrease in mesophyll-based photosynthesis (Ames), with LMA.

gm was estimated in seven Banksia species of diverse leaf structure using gas exchange combined with chlorophyll fluorescence measurements. gm decreased with increasing LMA and mesophyll cell wall thickness. However, CO2 concentration at the sites of carboxylation (Cc) was remarkably stable across the wide range of LMA examined, thus low gm could not explain the lower Ames at high LMA. It is hypothesised that the latter may reflect lower Rubisco concentrations, lower Rubisco specific activity or lower Rubisco activation state at the high end of the LMA spectrum.

gm was shown to be influenced not only by leaf structure, but also by environmental factors, such as CO2 and irradiance. In accordance with previous studies on mesophytes, elevated CO2 concentrations and low irradiance reduced gm in the short term. This was found using both the fluorescence and the on-line carbon isotope discrimination methods for estimating gm, and may be associated with gm regulation by membrane channels permeable to CO2, called cooporins. Although both methods agreed in the response of gm to

CO2 and irradiance, the values of gm obtained with the isotopic method were consistently higher than those obtained with the fluorescence method. This discrepancy between the two

[iv]

ABSTRACT

methods is in agreement with recent studies that measured gm with both methods in mesophytes. Sensitivity analyses of the data obtained in Banksia showed that this inconsistency between the two methods may be caused by: (a) likely underestimation of gm due to overestimation of electron transport rate by the fluorescence method, and (b) uncertainty in the used value of the photo-compensation point in the absence of mitochondrial respiration and of the fractionation by Rubisco and PEPC. A likely overestimation of gm by the isotopic method due to potential effects of O2 concentration on gm, may also explain the discrepancy between the two methods.

This study has highlighted that the relationship between photosynthesis and leaf structural traits, such as LMA, is not simple. Greater amount of supporting tissues can explain the lower Amass, but detailed analyses of the proportions of the different tissues per unit leaf area and volume are important in tackling the effect of leaf structure on Aarea. Despite the physiological limitations caused by leaf structure in high-LMA leaves, adaptations that compensate for part of the effects of leaf anatomy on Aarea have been suggested. This new perspective of the effects of leaf structural traits on photosynthesis, transpiration and CO2 diffusion has significantly advanced our understanding of the physiology of high-LMA leaves and of leaf adaptations that can maximise the photosynthetic potential of leaves of a certain structure. It has also highlighted the complexity of diffusion through stomatal crypts, giving new insight into the functional significance of these leaf structures. Finally, it has extended our knowledge of the diffusive limitations in the mesophyll for species at the high end of the LMA spectrum. Future research must focus on the improvement of the available methodology to estimate these limitations and on their regulation at the cell level.

[v]

[vi]

ACKNOWLEDGEMENTS

I am very grateful to my three supervisors, Dr Erik Veneklaas, Dr John Evans and Dr Martha Ludwig, for their kindness, patience, continuous support and encouragement during my PhD, and for making this experience an enjoyable and fruitful one. I am very appreciative of the long hours they spent reviewing my chapters and discussing my experiments, approaches and questions.

I am very grateful to Martha, for she was my first contact in UWA for almost a year before I migrated to Perth, giving me support and encouragement even before the commencement of my PhD. Martha was the one who put me in contact with Erik as soon as I arrived in Perth.

Erik is a wonderful person and has been a great supervisor; his assistance in all aspects and stages of my PhD as well as during my teaching internship will never be forgotten. Erik has inspired me to be flexible, look into the detail of everything and think more collectively on the relationship between structure and function. He has always been by me in moments of pressure, has given me excellent advice on my experiments and funding applications, and has helped and encouraged me to develop new contacts that would enable me to further pursue my research goals. One of these contacts was my third supervisor, John.

John is a leading authority in the field of CO2 diffusion within leaves and he runs one of the best gas exchange laboratories in Australia. Erik had encouraged me to send my research proposal to John at the beginning of my PhD. When I was about half way in my degree, I asked John if I could visit his laboratory in Canberra and perform some measurements there using equipment that was not available in WA. John accepted me wholeheartedly and has since been a great supervisor for me. He has put up with my questions and supported me both during and after my two visits to his laboratory and the long-day measurements of mesophyll conductance. Despite being so far from Perth, I have always felt he was around to support me.

Thanks are also extended to all the agencies that have provided me with funding to accomplish my research aims. The Australian Government is acknowledged for an Australian Postgraduate Award. Travel grants that enabled me to visit and work in John‟s laboratory and participate in conferences were a Mary Janet Lindsay of Yanchep Memorial Fund, an ARC-NZ Research Network for Vegetation Function grant, a Grieve Memorial travel grant, a Robertson Fellowship, a UWA travel grant and a School of Plant Biology travel grant. Special thanks to Dr Jaume Flexas, who gave me the opportunity to participate in an international workshop on mesophyll conductance in Palma de Mallorca in 2008. I also owe many thanks for this to Professor Hans Lambers who put me in contact with Jaume, but also for his continuous support during my PhD. I am also very grateful to the UWA Teaching Internship committee for giving me the opportunity to do a Teaching Internship in 2008, which has added significantly to my professional development.

[vii]

ACKNOWLEDGEMENTS

I thank all my colleagues, especially those I shared a room with during the last years, for advice and their cheerfullness at hard times. I believe that the environment one works in plays a very important role in one‟s work success. I would also like to thank those who have contributed to my thesis either as reviewers of my chapters or providing me with help during measurements and equipment: Professor Hans Lambers, Professor Marilyn Ball, Dr Paul Kriedemann, Ms Stephanie McCaffery, Professor Susanne von Caemmerer, Dr Michael-Saam Renton and Dr Anita Roth-Nebelsick. I am very grateful to Anita, who has done the finite element modeling of stomatal crypts. Although we have never met or talked face-to-face, her kind and understanding manner in her emails, her invaluable help with the analysis of the crypt effect on transpiration and her prompt responses have been greatly appreciated.

Special thanks to Dr Peta Clode and Mr John Murphy of the Centre for Microscopy, Characterisation and Analysis (UWA) for their assistance with the microscopical techniques done at UWA, and to Dr Cheng Huang for his assistance with cryo-scanning electron microscopy at ANU. I would also like to express my gratitude to the Plant Biology office staff, especially Dr Renu Sharma, for their continuous support during this journey.

I cannot thank enough my husband, Demetrius, for always being there for me. He was the one who visited UWA first whilst I was still completing my undergraduate degree in Greece, and through him I got in contact with Hans, Martha and Erik. You have tolerated me during my long work hours, especially towards the end of this thesis, you have supported me in difficult hours, you have inspired me to not stop trying. Finally, and most importantly, I thank with all my heart my Father, who has been the reason, inspiration and motivation behind everything I have done. To Him and to Demetrius I dedicate this thesis.

[viii]

TABLE OF CONTENTS

COVER PAGE ...... I

THESIS ABSTRACT ...... III

ACKNOWLEDGEMENTS ...... VII

TABLE OF CONTENTS ...... IX

THESIS DECLARATION AND PUBLICATION LIST ...... XIII

ABBREVIATION LIST ...... XV

CHAPTER 1: GENERAL INTRODUCTION ...... 1

1.1 BACKGROUND ...... 2

1.2 SCLEROPHYLLY: CONTROVERSY OVER ITS ADAPTIVE VALUE ...... 2

1.3 LEAF ANATOMY AND GAS EXCHANGE ...... 6

1.4 THE MODEL GROUP ...... 9

1.5 RESEARCH OBJECTIVES ...... 10

CHAPTER 2: STOMATAL CRYPTS MAY FACILITATE DIFFUSION OF CO2 TO ADAXIAL MESOPHYLL CELLS IN THICK SCLEROPHYLLS ...... 13

2.1 ABSTRACT ...... 14

2.2 INTRODUCTION ...... 14

2.3 MATERIALS AND METHODS ...... 16 2.3.1 Plant species and growth conditions ...... 16 2.3.2 Leaf structural traits...... 17 2.3.3 Crypt and stomatal characteristics ...... 17 2.3.4 Crypt conductance ...... 19 2.3.5 Photosynthetic measurements ...... 22 2.3.6 Modeling the effect of crypts on WUE ...... 23 2.3.7 Statistical analyses ...... 24

2.4 RESULTS ...... 25 2.4.1 Effects of crypts on gas exchange ...... 25 2.4.2 Correlations between crypts, stomata and leaf structure ...... 28

[ix]

TABLE OF CONTENTS

2.5 DISCUSSION ...... 36 2.5.1 Stomata contribute significantly more to leaf resistance than crypts ...... 36 2.5.2 Do crypts primarily function to reduce transpiration? ...... 38

2.5.3 A role of crypts in facilitating diffusion of CO2 ...... 39

CHAPTER 3: STOMATAL CRYPTS HAVE SMALL EFFECTS ON TRANSPIRATION: A

NUMERICAL MODEL ANALYSIS ...... 43

3.1 ABSTRACT ...... 44

3.2 INTRODUCTION ...... 44

3.3 MATERIALS AND METHODS ...... 46 3.3.1 Study species ...... 46 3.3.2 Microscopy ...... 48 3.3.3 Simulation method ...... 48 3.3.4 Model setup ...... 49

3.4 RESULTS...... 55 3.4.1 Spatial pattern of the humidity gradient ...... 55 3.4.2 Stomatal and crypt conductance, and transpiration ...... 57

3.5 DISCUSSION ...... 59

3.6 CONCLUSIONS...... 65

CHAPTER 4: PHOTOSYNTHESIS AT AN EXTREME END OF THE LEAF TRAIT SPECTRUM:

HOW DOES IT RELATE TO HIGH LEAF DRY MASS PER AREA AND ASSOCIATED

STRUCTURAL PARAMETERS? ...... 67

4.1 ABSTRACT ...... 68

4.2 INTRODUCTION ...... 68

4.3 MATERIALS AND METHODS ...... 70 4.3.1 Plant material and growth conditions ...... 70 4.3.2 Leaf morphology and anatomy ...... 71 4.3.3 Chemical composition ...... 72 4.3.4 Microscopy ...... 73 4.3.5 Photosynthetic measurements ...... 74

[x]

TABLE OF CONTENTS

4.3.6 Statistical analyses ...... 75

4.4 RESULTS ...... 77 4.4.1 LMA and its anatomical correlates ...... 77 4.4.2 Leaf structure, photosynthesis and mesophyll conductance ...... 82

4.5 DISCUSSION ...... 85 4.5.1 LMA and its anatomical correlates ...... 85 4.5.2 Leaf structure, photosynthesis and mesophyll conductance ...... 87

4.6 CONCLUSIONS ...... 90

CHAPTER 5: INFLUENCE OF LEAF DRY MASS PER AREA, CO2, AND IRRADIANCE ON MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS ...... 93

5.1 ABSTRACT ...... 94

5.2 INTRODUCTION ...... 94

5.3 MATERIALS AND METHODS ...... 97 5.3.1 Plant material and growth conditions ...... 97 5.3.2 Gas exchange and chlorophyll a fluorescence ...... 98

5.3.3 Calibration of the relationship between Jf and J ...... 100 CO 2

5.3.4 Estimation of gm using the „variable J method‟ ...... 101 5.3.5 Statistical analyses ...... 101

5.4 RESULTS ...... 103

5.5 DISCUSSION ...... 109

5.5.1 Effects of LMA on gm ...... 109

5.5.2 Effects of CO2 concentration and irradiance on gm ...... 112 5.5.3 Methodological issues ...... 113

5.6 CONCLUDING REMARKS ...... 115

CHAPTER 6: RESPONSE OF MESOPHYLL CONDUCTANCE MEASURED BY ON-LINE

CARBON ISOTOPE DISCRIMINATION TO CO2 AND IRRADIANCE IN SCLEROPHYLLS .... 117

6.1 ABSTRACT ...... 118

6.2 INTRODUCTION ...... 118

6.3 MATERIALS AND METHODS ...... 121

[xi]

TABLE OF CONTENTS

6.3.1 Plant material and growth conditions ...... 121 6.3.2 Gas exchange and carbon isotope discrimination measurements...... 121

6.3.3 Estimation of gm by on-line carbon isotope discrimination ...... 123 6.3.4 Gas exchange and chlorophyll fluorescence measurements ...... 125

6.3.5 Estimation of gm using the Variable J method ...... 125

6.3.6 Modeling of Jf overestimation and sensitivity analyses ...... 126

6.4 RESULTS...... 126

6.4.1 gm measured by on-line carbon isotope discrimination ...... 126

6.4.2 Comparison of methodologies that estimate gm ...... 131

6.5 DISCUSSION ...... 137

6.5.1 Response of gm to CO2 and irradiance using on-line carbon isotope discrimination ..... 137

6.5.2 Comparison of methodologies that estimate gm ...... 138

6.6 CONCLUDING REMARKS ...... 143

CHAPTER 7: GENERAL DISCUSSION ...... 145

7.1 STOMATAL CRYPTS ...... 146

7.2 SCLEROPHYLLY AND GAS EXCHANGE ...... 150

7.3 MESOPHYLL CONDUCTANCE AT THE HIGH END OF THE LMA SPECTRUM ...... 153

7.4 CONCLUSIONS ...... 156

BIBLIOGRAPHY ...... 157

APPENDICES ...... 183

APPENDIX 1 ...... 184

APPENDIX 2 ...... 187

APPENDIX 3 ...... 189

[xii]

THESIS DECLARATION AND PUBLICATION LIST

This thesis includes three published papers and one submitted manuscript that have been co-authored:

Chapter 2: Hassiotou F, Evans JR, Ludwig M, Veneklaas EJ (2009). Stomatal crypts

may facilitate diffusion of CO2 to adaxial mesophyll cells in thick sclerophylls. Plant, Cell and Environment 32: 1596-1611.

Chapter 3: Roth-Nebelsick A, Hassiotou F, Veneklaas EJ (2009). Stomatal crypts have small effects on transpiration: a numerical model analysis. Plant Physiology 151: 2018-2027.

Chapter 4: Hassiotou F, Renton M, Ludwig M, Evans JR, Veneklaas EJ (2009). Photosynthesis at an extreme end of the leaf trait spectrum: how does it relate to high leaf dry mass per area and associated structural parameters? Journal of Experimental Botany (submitted).

Chapter 5: Hassiotou F, Ludwig M, Renton M, Veneklaas EJ, Evans JR (2009).

Influence of leaf dry mass per area, CO2, and irradiance on mesophyll conductance in sclerophylls. Journal of Experimental Botany 60: 2303-2314.

The majority of the work associated with the production of these papers as well as the rest of this thesis is my own, with one exception: the finite-element modeling of stomatal crypts that is included in Chapter 3 was carried out by Dr Anita Roth-Nebelsick (State Museum of Natural History, Stuttgart), who has given her written consent for inclusion of this paper in the thesis for purposes of completeness. This paper/chapter was instigated by me based on histological analyses of Banksia leaves that were done for the purposes of this thesis. My contribution to this paper/chapter constitutes of formulation of aims and objectives, research plan, collection and analysis of data on leaf structure, and editorial input in different draft forms of the paper/chapter. The contribution of the co-authors in the published or submitted papers/chapters was primarily associated with the initial research direction, advice on experiments when required, and editorial input in draft forms. [xiii]

[xiv]

ABBREVIATION LIST

a average aperture of a crypt

A or Aarea net CO2 assimilation rate per unit leaf area

AChl net CO2 assimilation rate per unit chlorophyll

Amass net CO2 assimilation rate per unit leaf mass

Ames net CO2 assimilation rate per unit mesophyll

AM mesh area α leaf absorptance β fraction of quanta absorbed by Photosystem II C concentration of diffusing substance

Ca ambient CO2 concentration

Cc chloroplastic CO2 concentration

Ci intercellular CO2 concentration

Chlmes chlorophyll concentration per mesophyll volume cp air specific heat capacity CWA total wall surface area per crypt Γ* photo-compensation point in the absence of mitochondrial respiration D diffusion coefficient of CO2 in air d crypt depth

DL or LD leaf density * DL leaf density corrected for porosity

δ diffusivity ratio of water vapour to CO2 ΔT temperature difference between the mesh and the chamber air E evaporation rate ε thermal emissivity F flow rate gbl boundary layer conductance to water vapour gc crypt conductance gias conductance to CO2 diffusion through the mesophyll intercellular air spaces

[xv]

ABBREVIATION LIST

conductance to CO diffusion through the liquid phase, including cell walls, plasma g 2 liq membrane, cytosol, chloroplast envelope and stroma gL leaf conductance to CO2 gm mesophyll conductance to CO2 gp,c calculated mesh pore conductance gp,m measured mesh pore conductance gs stomatal conductance gs,c calculated stomatal conductance gs+c total leaf conductance

Gsystem system conductance ISD inter-stomatal distance J rate of electron transport (or diffusional flux in Chapter 3) J rate of electron transport obtained from gas exchange CO 2

Jf rate of electron transport calculated from chlorophyll fluorescence LDMC leaf dry matter content LMA leaf dry mass per area

LP adaxial palisade cell length LVA leaf volume per unit leaf area MVA mesophyll volume per unit leaf area n crypt density

Narea nitrogen per unit leaf area

Nmass nitrogen per unit leaf mass Φ gas exchange-based quantum yield CO 2

ΦPSII photochemical efficiency of Photosystem II p porosity of layer P porosity

PL leaf porosity (%) PNUE photosynthetic nitrogen use efficiency PPFD photosynthetic photon flux density

Rd respiration in the light RH relative humidity rbl resistance of the boundary layer

[xvi]

ABBREVIATION LIST

rc crypt resistance rm mesophyll resistance rs stomatal resistance rs+c leaf resistance

Rw and Rc total resistance to water vapour and CO2, respectively

Sc chloroplast surface area exposed to the intercellular air spaces SCV Single crypt volume SDC stomatal density per crypt

SDCWAB stomatal density per crypt wall surface area at the flat crypt bottom

SDCWAT stomatal density per total wall surface area of a crypt ζ Stefan-Boltzmann constant t thickness of layer

Ta chamber air temperature

TAB thickness of abaxial epidermis and hypodermis

TAD thickness of adaxial epidermis and hypodermis

TL or LT leaf lamina thickness

TM mesophyll thickness

Tm mesh temperature

Tw mesophyll cell wall thickness tortuosity

Vc maximum rate of RuBP (ribulose-1,5-bisphosphate) carboxylation

VG leaf gas volume

VL leaf fresh volume

VM mesophyll volume per unit leaf volume (%) VPD leaf to air vapour pressure difference WUE water-use efficiency

[xvii]

[xviii]

CHAPTER 1

GENERAL INTRODUCTION

1.1 BACKGROUND ...... 2

1.2 SCLEROPHYLLY: CONTROVERSY OVER ITS ADAPTIVE VALUE ...... 2

1.3 LEAF ANATOMY AND GAS EXCHANGE ...... 6

1.4 THE MODEL GROUP ...... 9

1.5 RESEARCH OBJECTIVES ...... 10

[1]

CHAPTER 1: GENERAL INTRODUCTION

1.1 BACKGROUND

Sclerophyllous vegetation occurs both in arid and humid environments, but is particularly common in Mediterranean-type ecosystems. It is characterized by hard (= sclero) leaves (= phyllo) displaying different combinations of structural traits that give them rigidity, thus conferring long leaf lifespans. However, the widespread occurrence of sclerophyllous species has created some debate over the functional significance of these traits, which range from sclerified and thickened epidermes, vascular bundle extensions, and sclereids to mesophyll cells with thick cell walls, and stomatal crypts. The role of, and the extent to which each of these structural traits influences leaf physiology, in particular gas exchange and photosynthesis, are not well understood. To improve our understanding of the functional significance and physiological consequences of sclerophylly, quantitative relationships between sclerophyllous traits and gas exchange were examined.

1.2 SCLEROPHYLLY: CONTROVERSY OVER ITS ADAPTIVE VALUE

The term “sclerophylly” was coined by Schimper (1903) to distinguish leaves of xerophytes from those exhibiting succulence or leaflessness. Terms such as tough and stiff have also been used to describe such leaves. More recently, Cowling and Campell (1983) described sclerophylls as coriaceous (leathery) and hard leaves, breaking when folded. Previous reports have highlighted some debate over the definition of sclerophylly (Hill, 1998; Edwards et al., 2000) and its evolutionary advantage regarding the relative importance of water versus nutrients (Mast and Givnish, 2002). Loveless (1961, 1962) and Hill (1998) define sclerophylly as the morphological response of leaves to low nutrient levels, particularly low phosphorus. One very common response to very low phosphorus levels is a small, evergreen, perennial habit, with an extensive root system (i.e. large root:shoot ratio) and small, highly fibrous leaves (Chapin, 1980; Hill, 1998). Many of the typical responses to low phosphorus (e.g. small leaves with very thick cuticles), generally regarded as sclerophyllous, resemble adaptations to restrict water loss (defined as xeromorphic) and this has led to inconsistent literature, where the two issues of low soil phosphorus and low water availability are not clearly separated (Hill, 1998). Sclerophylly occurs both in dry and/or nutrient-poor environments, which share a common characteristic

[2]

CHAPTER 1: GENERAL INTRODUCTION

in that they have the potential to be stressful. Such environments do not support high growth rates and plants in these environments typically demonstrate long leaf life spans, a trait which is usually expressed in the form of tough durable leaves. In the present study, sclerophylly is considered to be the combination of high leaf dry mass per area (LMA) and high dry matter content, due to thick and/or dense leaves with high levels of lignification. This is a general definition, which does not distinguish between phosphorus and water effects. It must be noted that certain non-sclerophyllous succulent species also have high LMA due to high leaf thickness (Poorter et al., 2009). Such species are not the focus of the present thesis. Although found worldwide, sclerophylly is particularly common in Mediterranean dense shrublands dominated by woody evergreen species (Specht, 1969; Edwards et al., 2000). The community of sclerophyllous evergreen shrubs is known as "chaparral" in California, "fynbos" in the Cape Province of South Africa, "matorral" in Chile, “kwongan” or "mallee" in Australia and "macchia" or "maquis" in the European Mediterranean basin, with some additional names existing in each country of the European Mediterranean basin (Specht, 1969; Di Castri, 1981; Valiente-Bonuet et al., 1998; Rotondi et al., 2003). These Mediterranean-type plant communities are a recognised reservoir of biodiversity (Myers et al., 2000; Rotondi et al., 2003). Sclerophyllous vegetation is also common on low-nutrient and ultramafic soils (igneous, containing magnesium and iron and only a very small amount of silica) in other climate types, including regions of high rainfall from temperate to tropical latitudes (Turner, 1994; Edwards et al., 2000) and is not only limited to shrubs, with many forest and woodland trees having hard leaves (Edwards et al., 2000). Sclerophylls possess a characteristic anatomical organization: they commonly are thick, fibrous and hairy, at least on the lower surface, with thick cuticles, epidermal, hypodermal and mesophyll layers, thick cell walls, often both in epidermal and mesophyll cells, and have abundant sclerification in the vascular bundle sheaths and extensions (Edwards et al., 2000; Mast and Givnish, 2002; Turner, 1994; Karabourniotis, 1998). Furthermore, individual or groups of stomata can be encrypted in several ways (Jordan et al., 2008), in shallow pits, deep longitudinal grooves or deep crypts, such as in the genus Banksia (Proteaceae). It is noteworthy that some of these characters are not restricted to sclerophylls and not all hard-leaved species possess all of these characters (Read et al., 2000). For

[3]

CHAPTER 1: GENERAL INTRODUCTION

instance, sclerophylls can be thick or dense or both. This explains the great variation in LMA (the product of leaf thickness and leaf density) that is found among sclerophylls (Read et al., 2000). However, it is not well understood which environmental factors lead to increased leaf thickness and which to increased leaf density. Additionally, it is not clear how increases in each of these two traits result: through changes in the amounts/volumes of certain tissue types, or cell walls, cell contents, cell size or porosity? The occurrence of anatomical characters that are considered sclerophyllous, often in different combinations, in a range of environments has led to the development of six hypotheses for their potential functional significance: (i) Sclerophylly is advantageous in arid environments (e.g. Schimper, 1903; Oertli et al., 1990; Turner, 1994; Niinemets, 2001). A number of studies have demonstrated that high- LMA sclerophylls have lower tissue elasticity compared with lower-LMA species (Sobrado, 1986; Lo Gullo and Salleo, 1988; Dreyer et al., 1990; Zobel, 1996; Nardini et al., 1996; Beckett, 1997; Groom and Lamont, 1997; Salleo et al., 1997; Zhang et al., 1998; Niinemets, 2001; Corcuera et al., 2002). Lower tissue elasticity, which is due to the lower elasticity of thicker cell walls (Corcuera et al., 2002), causes leaf water potential to drop more quickly upon leaf water loss. This will increase the driving force for water uptake, which will enable less-elastic, high-LMA leaves to sustain water uptake from drier soils (Corcuera et al., 2002; Mitchell et al., 2008). The above studies are based on bulk modulus elasticity, which is a whole-leaf trait. It is not known if elasticity, and therefore water relations differ between contrasting cell types in high-LMA leaves, e.g. sclerified cells with thick cell walls versus mesophyll cells with thinner cell walls. (ii) Sclerophylly is advantageous in oligotrophic soils (Loveless, 1961, 1962; Beadle, 1966; Turner, 1994); the leaf hardness of the sclerophyllous slow-growing species ensures longer leaf lifespans that reduce the annual nutrient requirements by increasing nutrient residence times. (iii) Sclerophylly offers protection from excess solar radiation (Jordan et al., 2005); specific scleromorphic structures, such as very thick cuticles, are associated with the leaf surface exposed to direct sunlight and have been proposed to protect the mesophyll from photodamage and photoinhibition.

[4]

CHAPTER 1: GENERAL INTRODUCTION

(iv) On the other hand, specific scleromorphic structures have been suggested to act as optical fibres, increasing the uniformity of the illumination within leaves (reviewed in Niinemets and Sack, 2006). Sclerified lens-like epidermal cells may focus the light in the leaf interior (Poulson and Vogelmann, 1990; Smith et al., 1997), whilst sclerified bundle sheath extensions in heterobaric leaves as well as sclereids may guide light deep inside a thick sclerophyll (Nikolopoulos et al., 2002; Karabourniotis, 1998). This may be advantageous in thick leaves considering that, due to the high efficiency of chlorophyll as a light-absorbing molecule, most of the intercepted radiation is often absorbed near the adaxial mesophyll layers (Buckley and Farquhar, 2004). However, trade-offs with the rates of photosynthesis may occur (Niinemets and Sack, 2006) since bundle sheath extensions cover up to 50% of the leaf surface, resulting in reduced apparent rates of photosynthesis per unit leaf area. Moreover, photosynthesis may also be restricted in heterobaric leaves by lack of lateral diffusion when stomata are arranged in distinct patches or are not uniformly open (Niinemets and Sack, 2006). (v) Sclerophylly protects leaves from herbivory (Turner et al., 1993; Turner, 1994). (vi) Although all the above hypotheses consider sclerophylly as a response to a specific abiotic or biotic stress, a more general hypothesis has also been put forward that considers sclerophylly a non-specific evolutionary response to a range of environmental conditions (Salleo and Nardini, 2000). Sclerophyllous traits can result in enhancement of leaf longevity by leaf protection and physical support (Chabot and Hicks, 1982; Turner, 1994). It is crucial for sclerophyllous plants to have a metabolic strategy based on a longer leaf lifespan with a low energy investment per year (Merino et al., 1984; Salleo et al., 1997), thus slow growth. Plants adapted to various stressful conditions grow slowly and share many of the basic characteristics of plants adapted to infertile soils, reinforcing the concept of interdependent physiological characteristics constituting a stress-tolerant adaptive strategy (Chapin, 1980). All these hypotheses are supported by at least some empirical autoecological evidence in certain species. It is likely that a generalisation across species is not possible since similar structure may have evolved in response to different selective pressures. In this thesis, the primary interest is to investigate general relationships of leaf structural traits with

[5]

CHAPTER 1: GENERAL INTRODUCTION

physiological attributes, irrespective of the cause of sclerophyllous anatomy, within a single phylogenetic group.

1.3 LEAF ANATOMY AND GAS EXCHANGE

Photosynthesis requires diffusion of CO2 into the leaf. In C3 plants, the diffusion of CO2 from the atmosphere to the active site of ribulose 1,5-bisphosphate carboxylase /oxygenase (Rubisco) follows a complex pathway involving as many as eight discrete resistance (the reciprocal of conductance) components (Nobel, 1999). Most commonly, with respect to resistance, this pathway is simplified into three main components: boundary layer and stomatal and mesophyll resistances (Bernacchi et al., 2002).

Boundary layer resistance (rb and its reciprocal, gb) depends on several leaf physical and environmental properties, such as leaf size, surface structures (trichomes, wax layers, crystals) and air movement around the leaf. Stomata facilitate gas exchange between the atmosphere and the leaf interior. These flexible pores are typically located on the surface of the photosynthesising organ (leaf or stem), in distinct groups or non-clustered, comprising part of the epidermis. Stomatal resistance (rs and its reciprocal, gs) is primarily influenced by stomatal pore numbers, size and degree of opening. However, modifications of stomatal arrangement influencing leaf resistance have been observed in leaves (e.g. Mast and Givnish, 2002) and photosynthesising stems (Gibson, 1983), in which stomata are sunken beneath the surface. In these modifications, proliferation of protoderm as multiple epidermis or formation of thick primary and secondary cuticle (Gibson, 1996) results in depressions of the epidermis where stomata, either individual or grouped (in invaginations called „crypts‟), are slightly (5-10 μm; Gibson, 1983; Yiotis et al., 2006) to deeply (up to more than 400 μm; Mohammadian, 2005) sunken. Thus, when stomata are sunken in abaxial epidermal surface depressions an additional resistance component is added, that of the superstomatal chamber. In the present study, this component is termed crypt resistance

(rc and its reciprocal, gc). Finally, the resistance in the mesophyll (rm and its reciprocal, gm) is influenced by the structure of the mesophyll, its gaseous component in the intercellular air spaces (rias and its reciprocal, gias) and/or its liquid component from the cell walls to the chloroplasts (rliq and its reciprocal, gliq). rm is affected by the environment, which

[6]

CHAPTER 1: GENERAL INTRODUCTION

influences structural/physical/chemical leaf properties (Flexas et al., 2008). The effects of the boundary layer and stomatal resistance on gas exchange have been studied extensively under a range of environmental conditions (Farquhar and Sharkey, 1982; Zeiger et al., 1987; Schuepp, 1993; Meinzer et al., 1997; Jones, 1998; Gutschick, 1999; Long and Bernacchi, 2003). However, our understanding of mesophyll resistance, its impact on photosynthesis and how it is influenced by leaf structure and the environment, is still poor. Furthermore, studies on stomatal crypts, their function and effects on gas exchange are scarce. These properties are of particular relevance to the functioning of sclerophylls. Sunken stomata are found in a number of species occurring in a range of habitats, from arid, e.g. Pinus species (Napp-Zinn, 1966; Burrows and Bullock, 1999; Boddi et al., 2002), Banksia species (Jordan et al., 2008) and Nerium oleander (Metcalfe and Chalk, 1979; Lösch et al., 1982), to wet or occasionally flooded environments, e.g. Sequoia sempervirens (Burgess and Dawson, 2004) and Symmeria paniculata (Waldhoff et al., 2002). Stomatal recessions vary in architecture, depth and width (Jordan et al., 2005). Stomatal encryption has often been considered a xerophytic feature, a structural adaptation that reduces water loss through reduced leaf transpiration (Turner, 1994; Hill, 1998; Rotondi et al,. 2003; Haworth and McElwain, 2008; Jordan et al., 2008). Despite the wide acceptance of this idea, however, direct evidence of the impact of stomatal depressions on transpiration, water-use efficiency or other aspects of gas exchange and leaf physiology is scarce (Lösch et al., 1982; Roth-Nedelsick, 2007). Plants that inhabit arid environments have other ways to control transpiration, e.g. via tightly controlled stomata. Moreover, crypts are present in species from both arid and wet environments (Brodribb and Hill, 1997), and this challenges the view that the sole or primary role of crypts is transpiration reduction.

Prior to techniques being developed to measure mesophyll conductance (gm), gm had been assumed to be large enough to have a negligible impact on photosynthesis (Farquhar et al.,

1980). However, gm has now been shown to be sufficiently small to significantly decrease the concentration of CO2 at the site of carboxylation (Cc) relative to that in the intercellular air space (Ci), thereby limiting photosynthesis (von Caemmerer and Evans, 1991; Harley et al., 1992; Loreto et al., 1992; Evans et al., 1994; von Caemmerer et al., 1994; Bernacchi et al., 2002; Warren, 2006). As the methodology to measure gm becomes more widely [7]

CHAPTER 1: GENERAL INTRODUCTION

available, increasing attention is being paid to understanding leaf internal diffusion (Warren, 2007; Flexas et al., 2008). Although some studies using helox (air where nitrogen has been replaced by helium) have indicated that gias (gas phase conductance) can account for 10-60% of gm, being more important in hypostomatous leaves (Parkhurst and Mott,

1990), other studies have shown that gliq (liquid phase conductance) is the main determinant of gm (von Caemmerer and Evans, 1991; Genty et al., 1998; Farquhar et al., 2001; Aalto and Juurola, 2002; Sharkey et al., 2007; Warren, 2007).

The effects of leaf structure on gm have often been studied through its relationship with LMA, an indicator of sclerophylly (Gratani and Varone, 2006; Flexas et al., 2008). A recent review by Flexas et al. (2008) summarising data from 17 studies showed that while low-

LMA mesophytic species present a wide range of gm values, leaf structure appears to strongly limit gm in evergreen species with high LMA. Flexas et al. (2008) reported a negative relationship between gm and LMA, with an upper bound of gm that extrapolates to zero at a LMA of 240 g m-2. However, since LMA in sclerophyllous plants can be much higher than 240 g m-2, there is a need to extend the range of measurements. During the last decade, it has been shown that leaf structure is not the only determinant of gm, as the latter shows fast responses to environmental factors such as soil water availability, salinity and temperature (Warren, 2007; Flexas et al., 2008). Recently, Flexas et al. (2007a) showed that gm decreased with increasing CO2 concentration in seven species and with decreasing irradiance in tobacco, but the reasons why there is variation between species and the mechanisms behind these responses are not clear yet. Recent studies suggest that the changes in gm with the environment happen through changes in leaf biochemistry. Bernacchi et al. (2002) pointed out that the observed Q10 of approximately

2.2 in tobacco shows that gm does not conform to CO2 transfer dominated by simple diffusion, but suggests that an enzyme or other protein-facilitated process is involved. The involvement of a protein in gm regulation supports the possibility of tight co-regulation of photosynthesis and gm. Likely candidates are an enzyme, carbonic anhydrase (CA), and some membrane channels, aquaporins (or “cooporins”), since recent studies show correlations between the regulation of these proteins and CO2 uptake (Price et al., 1994; Williams et al., 1996; Bernacchi et al., 2002; Terashima and Ono, 2002; Üehlein et al.,

[8]

CHAPTER 1: GENERAL INTRODUCTION

2003; Flexas et al., 2004; Hanba et al., 2004; Flexas et al., 2006; Üehlein et al., 2008; Miyazawa et al., 2008).

The two main methodologies that have been used for gm estimation are gas exchange combined with chlorophyll fluorescence (“fluorescence method”) (Harley et al., 1992) or carbon isotope discrimination measurements (“isotopic method”) (Evans, 1986). These methods are independent of each other, but they share a number of assumptions (Warren,

2007; Pons et al., 2009), raising the possibility of potential errors in estimating gm (Pons et al., 2009). For example, an assumption of both methods is the uniformity of the Ci and Cc across the leaf, which does not always occur (Terashima et al., 1988). An important assumption of the fluorescence method is that the fluorescence signal emanating from the adaxial leaf surface is representative of the whole leaf depth; this may lead to gm underestimation, especially in thick leaves. Similarly, an assumption of the isotopic method which may influence gm estimation is that gm is independent of O2 concentration. Consequently, there is a need to obtain comparative results of the two independent methods in more species and optimise the current methodologies. High LMA (through leaf thickness and/or density) and low leaf porosity may constrain maximum rates of CO2 assimilation through a limiting gm (Loreto et al., 1992; Evans et al., 1994; Parkhurst, 1994; Evans and von Caemmerer, 1996; Evans and Loreto, 2000; Terashima et al., 2001; Terashima et al., 2006). Surface properties of sclerophylls (wax layers, shape of the epidermal cells, cuticular thickening, trichomes, stomatal crypts) as well as specific scleromorphic structures (e.g. sclereids) can alter leaf optical properties (Myers et al., 1994, Baldini et al., 1997) and thus influence gas exchange. Although research shows that sclerophylls with high LMA tend to have relatively low gm values (Flexas et al., 2008) and rates of photosynthesis per leaf area (Wright et al., 2004), the extent to which these physiological parameters are influenced by the different sclerophyllous traits is not known, especially at the high end of the LMA spectrum.

1.4 THE MODEL GROUP

To tackle the gaps in knowledge identified above, the genus Banksia (family Proteaceae) was chosen as the model plant group. Sclerophylly within the southern-hemisphere family

[9]

CHAPTER 1: GENERAL INTRODUCTION

Proteaceae is an ancient phenomenon (Hill, 1998). The oldest proteaceous leaves known from the fossil record were probably scleromorphic (Hill, 1998). The primary centre of Proteaceae diversity is Australia (Wrigley and Fagg, 1989; Jordan et al., 2005). Banksia species are native to this region, with the majority of the 80 species (Lamont and Connell, 1996) being endemics of Australia's Southwest Botanical Province, one of the world's great centres of floristic endemism (Mast and Givnish, 2002). Banksia is a great example of sclerophylly and generally occurs in phosphorus- and water- stressed environments. The diverse leaf morphology and anatomy found in Banksia make it an excellent model to study how gas exchange co-varies with sclerophyllous traits. The leaves are entire, serrated or lobed, and vary greatly in size, from large broad forms to small needle-like leaves. They exhibit a range of sclerophylly. Many leaves possess dense coverings of trichomes, mainly on the abaxial surface but also on the adaxial surface while developing. The majority of Banksia also possess encrypted stomata (Mast and Givnish, 2002) and display great diversity in crypt architectures.

1.5 RESEARCH OBJECTIVES

The general aim of this study was to improve our understanding of the functional significance of sclerophyllous traits by determining their impact on CO2 diffusion and photosynthesis. Key aspects of the photosynthetic process in high-LMA sclerophylls and the conditions in which photosynthesis takes place at the tissue and cellular level have not been extensively addressed. It is unclear whether the photosynthetically active mesophyll cells of high-LMA leaves are different from those in lower-LMA leaves, and if so, whether this is because of the conditions in which they operate (CO2, light) or because they themselves are structurally and physiologically different. The overall question to address was: “Do high-LMA leaves differ in the organisation of the mesophyll, or is the mesophyll itself also different in terms of its physiology?” The study involved comparative analysis of leaves with different degrees of sclerophylly, as indicated through LMA, using closely related Banksia species. The advantage of focusing on one species-rich genus was the minimisation of anatomical and physiological variation due to phylogenetic differences,

[10]

CHAPTER 1: GENERAL INTRODUCTION

whilst at the same time a wide range of sclerophylly could be examined. The specific aims of the study were: (i) To investigate the impact of stomatal encryption on gas exchange and obtain further insight into the functional significance of stomatal crypts. Crypt and stomatal characteristics were examined in a range of Banksia species and were related to other sclerophyllous traits, e.g. LMA, leaf thickness and leaf density. A model of diffusion through crypts based on a porous layer was validated, and used to partition resistance to diffusion between stomata and crypts. The modelled diffusive resistance was used to examine the likely effects of crypts on water-use efficiency (Chapter 2). It was

hypothesised that crypts function to facilitate CO2 diffusion from the abaxial surface to adaxial palisade cells in thick leaves. Detailed finite-element models of diffusion through crypts of differing architecture were constructed and their effects on leaf transpiration were examined (Chapter 3). (ii) To examine the effects of leaf structure on photosynthesis at the high end of the LMA spectrum, and improve our understanding of the physiological consequences of sclerophylly and its functional significance. Quantitative relationships between leaf structural and physiological properties of Banksia species that cover a wide range of LMA were investigated. The contribution of the anatomical constituents of leaf traits to the variation observed in these traits as well as their relationship with photosynthesis

and gm were examined (Chapter 4).

(iii) To determine the extent to which LMA, CO2 concentration and irradiance influence gm in sclerophylls. Given that LMA values of Banksia species are between 130 and 500 g -2 m , this genus provides an excellent model to examine the effect of LMA on gm as well

as extend the current knowledge on gm for species at the high-LMA end. In this

context, seven Banksia species covering a wide range of LMA were selected, and gm

was estimated from CO2 response curves at two irradiances using combined gas exchange and chlorophyll fluorescence measurements (Chapter 5).

(iv) To compare the two main current methodologies that estimate gm. High LMA possibly

creates optical issues leading to uncertainty in the estimates of gm obtained with the “fluorescence” method. Thus, the latter were compared with estimates obtained with the “isotopic” method for three Banksia species of different LMA, to improve our

[11]

CHAPTER 1: GENERAL INTRODUCTION

understanding of the limitations of each method and potential associations with leaf structure (Chapter 6). The thesis concludes with a General Discussion (Chapter 7).

[12]

CHAPTER 2

STOMATAL CRYPTS MAY FACILITATE DIFFUSION OF CO2 TO ADAXIAL MESOPHYLL CELLS IN THICK SCLEROPHYLLS

2.1 ABSTRACT ...... 14

2.2 INTRODUCTION ...... 14

2.3 MATERIALS AND METHODS ...... 16 2.3.1 Plant species and growth conditions ...... 16 2.3.2 Leaf structural traits...... 17 2.3.3 Crypt and stomatal characteristics ...... 17 2.3.4 Crypt conductance ...... 19 2.3.5 Photosynthetic measurements ...... 22 2.3.6 Modeling the effect of crypts on WUE ...... 23 2.3.7 Statistical analyses ...... 24

2.4 RESULTS ...... 25 2.4.1 Effects of crypts on gas exchange ...... 25 2.4.2 Correlations between crypts, stomata and leaf structure ...... 28

2.5 DISCUSSION ...... 36 2.5.1 Stomata contribute significantly more to leaf resistance than crypts ...... 36 2.5.2 Do crypts primarily function to reduce transpiration? ...... 38

2.5.3 A role of crypts in facilitating diffusion of CO2 ...... 39

[Hassiotou F, Evans JR, Ludwig M, Veneklaas EJ (2009). Stomatal crypts may facilitate diffusion of CO2 to adaxial mesophyll cells in thick sclerophylls. Plant, Cell and Environment 32: 1596-1611]

[13]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

2.1 ABSTRACT

In some plants, stomata are exclusively located in epidermal depressions called crypts. It has been argued that crypts function to reduce transpiration; however, the occurrence of crypts in species from both arid and wet environments suggests that crypts may play another role. The genus Banksia was chosen to examine quantitative relationships between crypt morphology and leaf structural and physiological traits, to gain insight into the functional significance of crypts. Crypt resistance to water vapour and CO2 diffusion was calculated by treating crypts as an additional boundary layer partially covering one leaf surface. Gas exchange measurements of polypropylene meshes confirmed the validity of this approach. Stomatal resistance was calculated as leaf resistance minus calculated crypt resistance. Stomata contributed significantly more than crypts to leaf resistance. Crypt depth increased and accounted for an increasing proportion of leaf resistance in species with greater leaf thickness and leaf dry mass per area. All Banksia species examined with leaves thicker than 0.6 mm had their stomata in deep crypts. It is proposed that crypts function to facilitate CO2 diffusion from the abaxial surface to adaxial palisade cells in thick leaves. This and other possible functions of stomatal crypts, including a role in water use, are discussed.

2.2 INTRODUCTION

Sunken stomata, either singly or in groups, are located in depressions of the leaf surface that form shallow pits, deep longitudinal grooves or deep, narrow-mouthed crypts. Plant species that display sunken stomata occur in a range of habitats, from arid (e.g. Pinus species, Cistus species, Banksia species, Hakea species, Persoonia species, Grevillea striata, Franklandia fucifolia, Nerium oleander) to wet or occasionally flooded environments (e.g. Sequoia sempervirens, Symmeria paniculata) (Napp-Zinn, 1966; Metcalfe and Chalk, 1979; Lösch et al., 1982; Brodribb and Hill, 1997; Burrows and Bullock, 1999; Boddi et al., 2002; Waldhoff et al., 2002; Rotondi et al., 2003; Burgess and Dawson, 2004; Roth-Nebelsick, 2007; Jordan et al., 2008). The adaptive significance of stomatal encryption is still under debate, but crypts have often been considered a xerophytic feature, a structural adaptation that reduces water loss

[14]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

through reduced leaf transpiration (Turner, 1994; Hill, 1998; Rotondi et al., 2003; Haworth and McElwain, 2008; Jordan et al., 2008). This argument is supported by the fact that trichomes, which are also considered to reduce water loss via enhanced boundary layer resistance (Baldini et al., 1997; Ripley et al., 1999), are often present within the crypts in modest or great abundance (Metcalfe and Chalk, 1979; Lösch et al., 1982; Roth-Nebelsick, 2007; Hassiotou, unpublished data). Jordan et al. (2008) found that the evolution of very deep and narrow-mouthed stomatal crypts in Proteaceae was strongly linked to arid climatic conditions, whereas less pronounced encryption did not show any relation to climate. Despite the wide acceptance of the transpiration-reducing role of stomatal crypts (Turner, 1994; Hill, 1998; Jordan et al., 2008; Haworth and McElwain, 2008), direct evidence of their impact on transpiration (compared with that of stomata), water-use efficiency or other aspects of gas exchange and leaf physiology is scarce (Lösch et al., 1982; Roth-Nedelsick, 2007;). Based on Fick‟s First Law of diffusion, stomatal crypts are bound to impose some resistance to gas diffusion; however, this has only been modelled and not experimentally tested (Matthews, 2003; Chapter 3). Moreover, species with crypts do not necessarily have lower leaf conductance and transpiration, depending on their stomatal conductance (Lösch et al., 1982). Plants that naturally occur in arid environments have other ways to control transpiration, e.g. via tightly controlled stomata. Moreover, crypts are present in species from both arid and wet environments (Brodribb and Hill, 1997), and this challenges the view that the sole or primary role of crypts is transpiration reduction. The occurrence of crypts in many sclerophyllous taxa from wet environments suggests that stomatal encryption may be linked to sclerophylly rather than being a xerophytic character. Sclerophylly relates to a suite of traits that increase leaf toughness (Read et al., 2000) and is distinguished from xeromorphy (Turner, 1994; Hill, 1998; Rotondi et al., 2003; Raven et al., 2005; Haworth and McElwain, 2008). Many sclerophyllous traits are also xerophytic; however, sclerophylly is a worldwide phenomenon that is found both in arid and humid environments (Turner, 1994; Salleo et al., 1997). Hill and Merrifield (1993) and Hill (1994, 1998), based on fossil records, reported that in Banksia and Dryandra, sclerophylly evolved in the Late Palaeocene and preceded xeromorphy, which did not evolve until the Late Eocene. They suggested that sclerophylly evolved in the early Tertiary as a response

[15]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

to the nutrient-impoverished soils of the rainforests where these ancestral plant taxa occurred, and this pre-adapted these species to the arid conditions that appeared in Australia in the mid-Tertiary. Consequently, stomatal encryption may have evolved in response to factors other than plant water status, although it is inevitable that it also affects gas exchange. Given this ambiguity over the role(s) of stomatal crypts, the aim of the present study was to provide further insight into their functional significance by assessing quantitative relationships between stomatal crypt traits and leaf gas exchange characteristics. The genus Banksia was chosen for a case study. This genus belongs to the Proteaceae, where stomatal crypts appeared with the clade Cryptostomata that includes sclerophyllous species with long-lived leaves, common in south-western Australia (Mast and Givnish, 2002). In this region, the majority of 60 species of Banksia (Lamont and Connell, 1996) possess encrypted stomata (Mast and Givnish, 2002) and display great diversity of crypt architecture and leaf anatomy. Crypt architecture and density and the positioning of stomata were examined. These traits were compared with other sclerophyllous traits, e.g. leaf dry mass per area (LMA), leaf thickness and leaf mass density. A model of diffusion through crypts was validated using polypropylene meshes that resembled abaxial leaf surfaces with crypts, and was used to partition resistance to diffusion between stomata and crypts. The modelled diffusive resistance was used to examine the likely effects of crypts on water-use efficiency (WUE).

2.3 MATERIALS AND METHODS

2.3.1. Plant species and growth conditions Three- to five-year old plants of ten Banksia species were used (B. attenuata R.Br., B. candolleana Meisn., B. elderiana F.Muell. and Tate, B. hookeriana Meisn., B. ilicifolia R.Br., B. prionotes Lindl., B. quercifolia R.Br., B. repens Labill., B. serrata L.f. and B. victoriae Meisn.). The plants, except for B. serrata, were grown from seed under ambient conditions outdoors in 10-L pots containing a mixture of river sand and potting mix following their establishment during 2002 in Perth, Western Australia. About three weeks before measurements were taken, the plants were transferred to a controlled-temperature

[16]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

greenhouse (23°C day/18°C night). Mature plants of B. serrata were purchased from a local nursery in Canberra, Australian Capital Territory. Upon purchase, the plants were re- potted into 10-L pots containing a mixture of grey sand and potting mix and grown for two months prior to measurement in a greenhouse (25°C day/20°C night). For all measurements described below, youngest fully expanded leaves were used.

2.3.2 Leaf structural traits Three leaves per species, each leaf originating from a different plant, were sampled early in the morning, and leaf traits associated with sclerophylly were examined. Leaf lamina thickness (LT) was determined by light microscopy from transverse hand-sections that were also used for the measurement of crypt depth as described below. Leaf area (using a leaf area meter, LI-300A, Li-Cor, Lincoln, NE, USA) and leaf dry mass (after drying at 70 oC for three days) were measured, and leaf dry matter content, leaf dry mass per area (LMA) and leaf density (leaf density=LMA/LT) were calculated. The data set of Mast and Givnish (2002) was used to identify the stomatal position in 46 broad-leaved Banksia species in which leaf lamina thickness was subsequently measured. Species were categorised as those that had superficial stomata, shallow pits or deep crypts.

2.3.3 Crypt and stomatal characteristics Two leaves per species, each leaf originating from a different plant, were collected early in the morning, and immediately analysed with variable pressure scanning electron microscopy (ESEM) and/or cryo-scanning electron microscopy (CSEM). Leaves of B. serrata that were used for ESEM were initially frozen in liquid nitrogen and then transferred on dry ice from Canberra to Perth (Australia), where they were stored at -80 oC prior to ESEM. For CSEM, the procedure described by McCully et al. (2004) was followed. Segments of the leaf lamina from the middle part of each leaf were removed under liquid nitrogen, mounted on stubs with low-temperature Tissue-Tek (O.C.T. Compound cryostat specimen matrix, ProSciTech) and planed flat in the paradermal and transverse direction using a diamond knife in a cryo-microtome (Cryo-system Oxford CT1500, Oxford Instruments Ltd, Old Station Way, Eynsham, Oxford OX8 1TL, UK) at - 100 oC. Samples were then etched in the column of the CSEM (Cambridge S360,

[17]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

Cambridge Instruments Ltd, Viking Way, Bar Hill, Cambridge, CB3 8EL, UK) for 1-2 min at -90 oC to reveal cell outlines, sputter-coated with gold, and examined at 15 kV. Images of top and side views of crypts were captured using Microsoft Photodraw. For ESEM, several leaf discs (maximum diameter of 1 cm) were cut from both sides of the midrib, sampling both close to and away from the midrib and near the base and tip of the leaf. Scanning electron micrographs of the adaxial and abaxial surfaces of each leaf disc were obtained using a Phillips XL30 ESEM at 15 kV. Where abundant pubescence prevented clear identification of crypt boundaries, wood glue was applied in small areas of the abaxial leaf surface, from which leaf discs would be taken, prior to detachment of the leaf. After 8 hours the leaf was detached, the hardened wood glue was peeled off and leaf discs were prepared for ESEM. At least 33 scanning electron micrographs of different regions of each leaf were obtained and used for the quantification of: (i) the average number of crypts per unit leaf area (crypt density); (ii) the number of stomata per crypt, and per unit crypt bottom wall surface area; the latter was measured at the bottom of the crypts in a portion of the wall that was flat; (iii) the average aperture area of a crypt, which was obtained by measuring the aperture area of 100 crypts per species; (iv) stomatal length and inter-stomatal distance, measured at the bottom of the crypt and expressed in stomatal lengths. Stomatal number per unit leaf area was estimated from the average stomatal number per crypt and the average crypt density. Stomatal characteristics were examined in six species (B. attenuata, B. elderiana, B. ilicifolia, B. prionotes, B. quercifolia and B. serrata). These analyses were carried out using Image J (Abramoff et al., 2004). After the preparation of the leaf discs for ESEM, the rest of each leaf was stored at 4 oC for a maximum of two hours prior to sectioning for the determination of crypt depth and leaf lamina thickness (mm) at crypt sites using fluorescence microscopy. Transverse hand- sections from each leaf were prepared using carbon steel razorblades and examined using an optical fluorescence microscope (Zeiss Axioskop2, Zeiss, Oberkocken, Germany). Crypt depth was defined as the distance from the base of a crypt to the centre of the crypt aperture as viewed in transverse section, and was determined from median sections through a crypt. Images of 20-50 crypts per leaf were captured using a digital camera (Zeiss Axiocam with

[18]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

AxioVision software, Zeiss, Oberkocken, Germany) and analysed in Image J (Abramoff et al., 2004). The volume of a single crypt (SCV) and the total wall surface area per crypt (CWA) were calculated assuming a half-ellipsoid shape for each crypt (Hilbert and Cohn-Vossen, 1999): 4 (2r)d SCV 3 (1) 2 and

2 rd 2 d 2 r 2 2 r 2 sin 1 d 2 r 2 d CWA (2) 2 where r is crypt aperture radius and d is crypt depth. r was determined from crypt aperture area assuming a circular shape (area=πr2). In B. quercifolia, r was measured in fluorescence micrographs of transverse leaf views, as the widest part of a crypt, due to the irregular morphology of the crypt apertures. CWA and the number of stomata per crypt (measured in six species) were used to infer the number of stomata per wall surface area of a crypt.

2.3.4 Crypt conductance Current gas exchange methods do not allow separate measurement of stomatal and crypt -1 conductance. Therefore crypt conductance (gc, m s ) was calculated by treating the crypt as an additional boundary layer, using the general equation that predicts conductance through the latter (Nobel, 1999): Dp Dna g (3) c t d -5 2 -1 where D is the diffusion coefficient of CO2 in air (D=1.51 × 10 m s ), p is the porosity of the surface, which equals n×a, with n being the crypt density (number of crypts per unit leaf area, m-2) and a the average aperture area of a crypt (m2); t is the thickness of the layer, -1 which in this case equals d, the crypt depth (m). gc was then converted from m s to mmol -2 -1 -2 -1 -1 3 m s based on the relationship gc(mmol m s ) = gc(m s )×(P/RT)×10 (Nobel 1999), where P is the ambient air pressure (1 atm), R is the gas constant (8.205×10-5 m3 atm mol-1 K-1) and T is the ambient air temperature (21 oC=294.15 K). Sensitivity analyses addressing [19]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

Table 1. Morphological characteristics of the polypropylene meshes (Sefar AG, Thal, Switzerland) used to represent leaf surfaces with stomatal crypts for the estimation of crypt-like pore conductance. Mean values ± standard errors are given.

Mesh type Pore density Pore area Pore depth (mm-2) (mm2) (μm)

PP140-105 20.27 ± 0.12 0.011 ± 0.000 101 ± 2 PP120-125 19.11 ± 0.13 0.016 ± 0.001 106 ± 2 PP100-149 16.32 ± 0.07 0.022 ± 0.001 104 ± 1 PP80-177 9.47 ± 0.16 0.026 ± 0.001 165 ± 3 PP70-210 7.99 ± 0.02 0.038 ± 0.002 153 ± 3 PP60-250 5.00 ± 0.00 0.055 ± 0.001 227 ± 4 PP280 4.77 ± 0.03 0.088 ± 0.002 163 ± 3 PP405 2.16 ± 0.05 0.202 ± 0.003 221 ± 3

approximations relating to the crypt shape assumed by Eqn 3 have also been performed (see last section of Materials and Methods). Since Eqn 3 has not been previously used for the estimation of conductance through pores such as crypts, it was tested on polypropylene meshes. Eight types of polypropylene mesh were examined with pores of varying dimensions that resembled those of the crypts (Table 1, Suppl. Fig. S1 in Appendix 1). The conductance to water vapour through the pores of each mesh was both calculated (gp,c) using Eqn 3 and measured (gp,m). For the calculation of gp,c, scanning electron micrographs of each mesh were obtained using a Phillips XL30 ESEM, and pore density (number of pores per mesh area), the pore area and the strand diameter (representing the pore depth, d) were measured using Image J (Abramoff et al.,

2004; Table 1). gp,m was measured with a gas exchange system. The mesh (2×3 cm) was fitted on filter paper (2×3 cm) wetted with 150 μL of distilled water and backed with plastic tape. Immediately after wetting, the mesh with the filter paper was placed in the 6 cm2-chamber of a LI-6400 open gas exchange system (LI-6400-40, Li-Cor, Lincoln, NE, USA), with the mesh facing downwards, reflecting the location of the crypts on the leaf

[20]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

surface. Evaporation was recorded in the dark until the filter paper was completely dry. Evaporation rate at steady state (E, mmol m-2 s-1) was calculated as (LI-6400, 1998-2004): F(H O H O ) E 2 S 2 R (4) AM (1000 H 2OS ) -1 where F is the air flow rate (mmol s ), H2OS and H2OR are the sample (chamber) and -1 reference water vapour mole fractions (mmol H2O (mol air) ), respectively, and AM is the -4 2 -2 -1 mesh area (6 10 m ). The system conductance (Gsystem, mol m s ), which includes the conductance of the fixed boundary layer used (1.4 mol m-2 s-1, for the bottom surface only) and the pore conductance in series, was calculated at steady state as (LI-6400, 1998-2004): E Gsystem (5) H 2OL H 2OS -2 -1 where E is the evaporation rate (mmol m s ), H2OL is the mole fraction of water vapour at -1 saturation (mmol H2O (mol air) ) and H2OS is the mole fraction of water vapour in the -1 chamber (mmol H2O (mol air) ). The chamber thermocouple was withdrawn from the evaporating surface during the measurements to ensure that it would not get wet, and the mesh temperature was computed using an energy balance (LI-6400, 1998-2004). In short, the difference, ΔT, between the mesh temperature (Tm) and the chamber air temperature

(Ta) was computed as:

4 4 2 [(Tw 273) (Ta 273) ] 44.1E T Tm Ta 3 (6) 2cpGsystem 8 (Ta 273) where ε is the thermal emissivity (0.95), ζ is the Stefan-Boltzmann constant (5.67 10-8 W -2 -1 o o m K ), Tw is the chamber wall temperature ( C), Ta is the chamber air temperature ( C), E -2 -1 -1 - is the evaporation rate (mol m s ), cp is the specific heat capacity of the air (28 J mol K 1 -2 -1 ) and Gsystem is the system conductance (mol m s ). Since ΔT is dependent on Gsystem, o which is dependent on the mesh temperature (Tm), Tm was initially set at 20 C (Tm,initial) and the solver function of Microsoft Excel® (Microsoft Corporation 2007) was used to find 2 the calculated mesh temperature (Tm,final) for which (Tm,final - Tm,initial) = 0 by varying

Tm,initial. The mesh pore conductance (gp,m) was then calculated as: 1 gp,m (7) 1 1

Gsystem gbl

[21]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

where gbl is the boundary layer conductance to water vapour from one side of the mesh. gbl had a value of 1.4 mol m-2 s-1, which was determined by measuring evaporation from the surface of a filter paper backed with tape and facing downwards, without a mesh attached to it.

The sensitivity of gp,c to pore depth was calculated for five of the meshes that represented the range measured (Suppl. Fig. S2 in Appendix 1). The close agreement between gp,m and gp,c when pore depth was assumed to equal the strand diameter meant that it was the appropriate measure of pore depth for the mesh.

2.3.5 Photosynthetic measurements Gas exchange measurements were carried out on three leaves per species at a photosynthetic photon flux density (PPFD) of 1500 μmol quanta m-2 s-1 and at ambient -1 CO2 and O2 concentrations (380 μmol CO2 mol air and 21% O2), using a LI-6400 open gas exchange system (LI-6400-40, Li-Cor, Lincoln, NE, USA). Leaves were kept in the gas -2 -1 exchange chamber at high irradiance (1500 μmol quanta m s ) and low CO2 -1 concentration (100 μmol CO2 mol air) for at least 10 min before the commencement of the measurements, ensuring stomata were fully open and steady state was reached. At ambient CO2 concentration, 4-10 measurements of gas exchange, at least 7 sec apart, were recorded for each leaf, and the mean value of total leaf conductance (gs+c), representing the conductance to CO2 diffusion from the leaf surface to the substomatal cavity, was calculated. Since the stomata of the leaves are located in crypts, gs+c includes the conductance to CO2 diffusion of the crypts (gc) and the stomata (gs,m), as well as that of any surface trichomes. Stomatal conductance was also calculated for the six species in which stomatal characteristics were determined (see Crypt and stomatal characteristics above) using an equation that predicts stomatal conductance (in this case termed gs,c to differentiate from gs,m) based on stomatal morphology and densities on a leaf area basis (Nobel, 1999): Dna g (8) s,c d r where n is the number of stomata per unit leaf area, a the stomatal pore area, d the stomatal pore depth and r the stomatal pore radius. Eqn 8 is similar to Eqn 3, with an additional

[22]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

term, r, the “end correction” (Nobel, 1999) that attempts to include the bulging shape of the diffusion front, which adds a certain component to the total resistance of the pores. The number of stomata per unit leaf area was calculated from the number of stomata per crypt multiplied by the number of crypts per unit leaf area. The number of stomata per crypt was measured in high-magnification scanning and cryo-scanning electron micrographs obtained from leaves where the crypt trichomes were removed using the wood-glue peeling method. The stomatal pore area and radius were not measured in all species due to difficulties in obtaining high-magnification images of open stomata. Thus, these parameters were measured in stomata that were imaged when open and an average value (a=2.36 10-5 mm- 2 of leaf area and r=2 μm) was used for the estimation of gs,c in all the six species. In previous studies, stomatal pore depth has been regarded as equal to guard cell width (Eensalu et al., 2008); thus, guard-cell width was measured in high-magnification scanning electron micrographs and an average value for all six species (d=6.2 μm) was used. To estimate the relative importance of gc and gs (either gs,m or gs,c), reciprocals of the conductance values were taken, where rc is the crypt resistance, rs (either rs,m or rs,c) is the stomatal resistance and rs+c is the total leaf resistance. The relative magnitude of rc and rs was then determined.

2.3.6 Modelling the effect of crypts on WUE To examine the possible impact of stomatal crypts on water-use efficiency (WUE), a model was constructed, taking into account the partitioning of resistance between crypts and stomata obtained as described above. WUE (mmol mol-1) was computed as: A (C C ) WUE a i (9) E (VPD) where A is the CO2 assimilation rate, E is leaf transpiration rate, Ca is the atmospheric CO2 concentration, Ci is the CO2 concentration in the intercellular airspaces, VPD is the leaf to air vapour pressure difference, and δ is the diffusivity ratio of water vapour to CO2 in the stomatal pore and in the crypts, taking into account the different contributions of stomata and crypts to total leaf resistance [δ = (1.55 contribution of stomata)+(1.37 contribution -1 of crypts)]. Values of 380 and 250 μmol mol were assumed for Ca and Ci, respectively, -1 and 1.7 mmol mol for VPD. The diffusivity ratio of water vapour to CO2 in stomata is

[23]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

1.55 (25 oC) (Lide, 1990), and a ratio of 1.37 was assumed for crypts, which is the value used for the leaf boundary layer (Kays, 1966).

A second model considering theoretical values for the boundary layer conductance (gbl), stomatal conductance (gs), crypt conductance (gc) and mesophyll conductance (gm), was constructed to examine how crypts influence the overall resistance to CO2 relative to water vapour. The reciprocal of leaf conductance, leaf resistance (rs+c), was treated as the sum of crypt and stomatal resistance, rc + rs. -1 2 The resistance to water vapour of the boundary layer (rbl,w) was set at 0.5 mol m s -2 -1 (gbl,w=2 mol m s ), and that to CO2 (rbl,c) was thus 0.685 (rbl,c=rbl,w 1.37). The resistance -1 2 -2 -1 to CO2 in the mesophyll (rm) was set at 5 mol m s (gm=0.2 mol m s ). Leaf resistance to -1 2 water vapour (rs+c,w) was set at 5, 10, 20 and 40 mol m s. For each leaf resistance, two sets of values were used for the stomatal and the crypt resistance to water vapour (rs,w and rc,w, respectively), varying the partitioning between stomata and crypts. Resistances to water vapour and CO2 were computed (denoted by “c” as the second term of the subscript), assuming a diffusivity ratio of water vapour to CO2 of 1.37 and 1.55 for crypts and stomata, respectively. The total resistance to water vapour (Rw) was calculated as:

Rw rbl,w rs,w rc,w (10) and the total resistance to CO2 (Rc) as:

Rc rbl,c rs,c rc,c rm (11)

The ratio of the total CO2 resistance to the total water vapour resistance, Rc/Rw, was computed for each set of resistances.

2.3.7 Sensitivity and statistical analyses Eqn 3 assumes a cylindrical shape for the crypt. In reality, however, the crypt bottom may be slightly hemispheric (see Fig. 4). Comparing a flat-bottomed cylinder to one with a hemispheric end would suggest that this discrepancy would lead to an overestimate of crypt depth by at most 10%. To test for this, the estimate of stomatal resistance for each species was re-calculated when crypt resistance was varied by ± 20%. To test whether leaf lamina thickness was significantly different among species with crypts, shallow pits or superficial stomata, a one-way ANOVA was carried out followed by

[24]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

a TukeyHSD (Tukey Honest Significant Differences) test, using the R programming language (R Development Core Team, 2007). All the regression and sensitivity analyses were done in Microsoft Excel® (Microsoft Corporation 2007).

2.4 RESULTS

2.4.1 Effects of crypts on gas exchange

Measured leaf conductance (gs+c) varied 2-fold among species (ranging from 74 to 164 -2 -1 mmol CO2 m s ) and did not correlate with crypt traits. There was good quantitative agreement between calculated (gp,c) and measured (gp,m) pore conductances for polypropylene meshes (P<0.0001) (Fig. 1). This supported the use of Eqn 3 for calculating crypt conductance (gc), which varied 7-fold in the examined species (ranging from 203 to -2 -1 1513 mmol CO2 m s ).

The reciprocal of gc, rc, varied among species, but did not correlate with stomatal resistance, rs (Fig. 2A) or any leaf or crypt structural trait (Table 2). Interestingly, gc significantly increased with increasing inter-stomatal distance and decreased with increasing number of stomata per wall surface area of a crypt (Table 2).

4.0 3.5 Fig. 1. The results of testing Eqn 3 (see

)

-1

s 3.0 Materials and Methods) for the

-2 2.5 estimation of conductance of pores with

O m

2 2.0 dimensions similar to those of stomatal 1.5 crypts. The figure shows the relationship (mol H 1.0 p,m between measured (g ) and calculated g p,m 0.5 r2 = 0.94 P<0.0001 (gp,c) polypropylene mesh pore 0.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 conductance to water vapour. -2 -1 gp,c (mol H2O m s )

[25]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

As LMA increased, crypts accounted for a greater proportion of leaf resistance (Fig. 2). The average contributions of rc and rs,m to rs+c in the examined species was 23±3% and 77±3%, respectively, and their contributions to rs+c ranged from 10-38% and 62-90%, respectively. To assess the impact that the shape of the bottom of the crypt has on the partitioning of leaf resistance between stomata and crypts, the sensitivity to varying rc by ± 20% was calculated for each species using Eqn 3 (Suppl. Fig. S3 in Appendix 1). The majority of leaf resistance was always due to rs regardless of whether there was a 20% over- or under- estimation of rc by Eqn 3. The direct estimate of stomatal conductance (gs,c) calculated from stomatal traits for six species (Eqn 8) confirmed that stomata constitute 60-90% of leaf resistance (Fig. 2B, open symbols).

49.9 WUE 51.6 14 100

A rc 12 c B

r rs 80

s)

or

2

10 s

r

m 2 8 60

CO

due to -1 6 40

s+c 4 r

(mol r 20 2 % of

0 0 0 100 200 300 400 500 600 LMA (g m-2)

B. repens

B. serrata

B. ilicifolia

B. victoriae

B. prionotes B. attenuata

B. elderiana

B. quercifolia

B. hookeriana B. candolleana

Fig. 2. (A) Partitioning between crypt (rc, light grey) and stomatal (rs, dark grey) resistance in ten Banksia species. rs was estimated from total leaf resistance (rs+c) measured with a gas exchange system and rc. (B) Percentage contribution of stomata (circles) and crypts (squares) to total leaf resistance as a function of

LMA. Contribution is computed either using rs+c measured with a gas exchange system and rc (black symbols,

N=10) or rs calculated from the equation of Nobel (1999) based on stomatal morphology and densities per unit leaf area (white and grey symbols, N=6). The contribution of crypts and stomata tended to increase and decrease, respectively, as LMA increased (P=0.07). B. quercifolia (grey symbols) deviates from the rest of the species due to its different crypt aperture morphology and smaller area. WUE values show the consequence of increasing the contribution of crypt resistance from 20 to 30% of a given rs+c on modelled water use efficiency (mmol mol-1, see Eqn 9).

[26]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

Table 2. Correlation matrix of crypt, stomatal and leaf morphological and anatomical traits and photosynthetic parameters in Banksia species. The values are the correlation coefficients (r). ns: not significant, ***P≤0.001, **0.001

n a d SCV SDC SDCWAB SDCWAT ISD LMA LT LD LDMC a 0.921*** (-) d 0.819** (-) 0.781** (+)

SCV 0.948*** (-) 0.922*** (+) 0.941*** (+)

SDC 0.983*** (-) 0.902* (+) 0.919** (+) 0.993*** (+)

*** * * SDCWAB 0.757 ns 0.602 ns 0.978 (+) 0.879 (+) 0.840 (+)

SDCWAT 0.107 ns 0.467 ns 0.282 ns 0.021 ns 0.065 ns 0.358 ns

ISD 0.634 ns 0.279 ns 0.808* (-) 0.698 ns 0.645 ns 0.763 ns 0.678 ns

LMA 0.850** (-) 0.738* (+) 0.891*** (+) 0.912*** (+) 0.888* (+) 0.852* (+) 0.310 ns 0.873* (-)

LT 0.641* (-) 0.570 ns 0.874*** (+) 0.780** (+) 0.613 ns 0.896* (+) 0.404 ns 0.610 ns 0.779** (+)

LD 0.768** (-) 0.662* (+) 0.680* (+) 0.771** (+) 0.764 ns 0.522 ns 0.171 ns 0.784 ns 0.899*** (+) 0.430 ns

LDMC 0.810** (-) 0.740* (+) 0.801** (+) 0.841** (+) 0.836* (+) 0.545 ns 0.003 ns 0.710 ns 0.711* (+) 0.540 ns 0.653* (+)

** * gc 0.132 ns 0.111 ns 0.487 ns 0.274 ns 0.190 ns 0.565 ns 0.935 (-) 0.807 (+) 0.516 ns 0.604 ns 0.339 ns 0.254 ns

n: crypt density, a: crypt aperture area, d: crypt depth, SCV: single crypt volume, SDC: stomatal density per crypt, SDCWAB: stomatal density per crypt wall surface

area at the flat crypt bottom, SDCWAT: stomatal density per total wall surface area of a crypt, ISD: inter-stomatal distance, LMA: leaf dry mass per area, LT: leaf

lamina thickness, LD: leaf density, LDMC: leaf dry matter content, gc: crypt conductance.

[27]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

Table 3. Effect of crypts on the total diffusion pathway of water vapour and CO2 through a model leaf. rs,w and rc,w are the stomatal and crypt resistances to water vapour, respectively. rs,c and rc,c are the stomatal and crypt resistances to CO2, respectively. Rw is the total resistance to water vapour and Rc is the total resistance to CO2. Imp is the percentage improvement in the diffusion of

CO2 relative to water vapour in the leaf with crypts compared to the leaf with superficial stomata.

rc,w rs,w rc,c rs,c Rw Rc Rc/Rw Imp 0 5 0.0 7.8 5.5 13.4 2.44 5 0 6.9 0.0 5.5 12.5 2.28 7% 0 10 0.0 15.5 10.5 21.2 2.02 10 0 13.7 0.0 10.5 19.4 1.85 8% 0 20 0.0 31.0 20.5 36.7 1.79 20 0 27.4 0.0 20.5 33.1 1.61 10% 0 40 0.0 62.0 40.5 67.7 1.67 40 0 54.8 0.0 40.5 60.5 1.49 11%

The calculated effect of crypts on WUE showed that for species with the highest proportion of the total resistance associated with crypts (38%) WUE was 3.3% higher compared with the species with the lowest contribution of crypts (10%) (Fig. 2B). The greater the total resistance to CO2, the greater the potential improvement in WUE associated with crypts (Table 3).

2.4.2 Correlations between crypts, stomata and leaf structure While in some species (B. attenuata, B. candolleana, B. prionotes) the pattern of venation affected or determined the crypt distribution (Figs 3A, 3B, 3F), in most species, crypts were randomly distributed on the abaxial leaf surface (Figs 3C-3E, 3G-3J). A diagrammatic representation of a B. hookeriana transverse leaf section shows the crypt and stomatal arrangement (Fig. 4A). In the majority of species, trichome bases were observed on the abaxial leaf surface, indicating that the trichomes dropped off as leaves matured. In B. attenuata, intact abaxial surface trichomes were still present on the mature leaves (data not shown), creating an additional boundary layer, whereas in B. quercifolia the absence of both trichomes and trichome bases on the abaxial leaf surface was characteristic of the

[28]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

species (Fig. 3G). Crypt trichomes were observed in all species, typically located close to the crypt aperture (Figs 4C, 4E, 4H), but sometimes deeper within a crypt (Figs 4B, 4D, 4F, 4G). In some cases, trichomes were removed from the crypts using the wood glue peeling method, revealing the stomata (Figs 3A, 3I, 5A, 5B). This method also allowed the identification of a wax layer on the crypt walls of B. quercifolia (data not shown). Stomata were found only within the crypts, usually distributed across most of the crypt wall surface adjacent to mesophyll tissue, and displaying a cuticle cover that extended above the stomatal pore (Fig. 5). Stomata in crypts were invariably close to each other. The distance between stomata ranged from 0.3 stomatal lengths (B. elderiana, Fig. 5A) to 1.2 stomatal lengths (B. serrata, Fig. 5B) (Table 4). The number of stomata per unit crypt wall surface area was high and varied almost 2-fold (500-907 mm-2 crypt wall surface area) among the examined species. The length of stomata was quite conserved, ranging from 23.9±0.5 to 25.3±0.6 μm (Table 4). The number of stomata per crypt varied 4-fold, while the number of stomata expressed per unit leaf area varied 2-fold among the examined species (Table 4). The ten Banksia species that were chosen for examining relationships between crypt structure, stomatal traits and sclerophylly showed a 4-fold variation in LMA (134-507 g m- 2) and a 2-fold range in leaf lamina thickness (0.27-0.64 mm), leaf density (0.4-1 mg mm-3) and leaf dry matter content (34-72%).

Fig. 3. Scanning electron micrographs of abaxial leaf surfaces showing the stomatal crypt architecture and characteristics in the Banksia species. In most images, surface trichomes that remain on mature leaves (most are lost during development) have been removed with wood glue (see Materials and Methods), revealing the crypts among a plethora of trichome bases. However, crypt-filling trichomes have not been removed, unless stated otherwise, and consequently the stomata are less visible. (A) B. attenuata; many crypt-filling trichomes have been removed, allowing visibility of stomata (scale bar: 340 μm); (B) B. candolleana (scale bar: 420 μm); (C) B. elderiana (scale bar: 415 μm); (D) B. hookeriana (scale bar: 250 μm); (E) B. ilicifolia (scale bar: 500 μm); (F) B. prionotes (scale bar: 500 μm); (G) B. quercifolia; note the peculiar morphology and narrowness of the crypt aperture as well as the absence of surface trichome bases (scale bar: 200 μm); (H) B. repens (scale bar: 380 μm); (I) B. serrata; many crypt-filling trichomes have been removed allowing visibility of stomata (scale bar: 400 μm); (J) B. victoriae; note some crypt-filling trichomes extending well out of the crypt, onto the leaf surface (scale bar: 240 μm).

[29]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

A B

C D

E F

G H

I J

Fig. 3

[30]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

A B

C D

E F

G H

Fig. 4

[31]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

Fig. 4. Micrographs showing transverse sections through the leaf of selected Banksia species. (A) Diagrammatic representation of crypt and stomatal arrangement in a B. hookeriana leaf, based on the fluorescence micrograph of the transverse section shown in (B). Stomata are represented as white circles within the crypts; grey indicates epidermal/hypodermal tissue, vascular tissue and/or the sclerenchymatic vascular bundle extensions; the mesophyll is coloured beige (scale bars: 100 μm). This general structure was found in all the examined species, with the actual crypt morphology and stomatal characteristics and the amount of sclerenchymatic tissue relative to mesophyll tissue varying among the species. (C) B. attenuata (scale bar: 100 μm); (D) B. elderiana (scale bar: 120 μm); (E) B. prionotes; in the left areole two adjacent crypts have united creating a heart-shaped structure (scale bar: 100 μm); (F) B. quercifolia (scale bar: 60 μm); (G) B. serrata (scale bar: 60 μm); (H) B. victoriae (scale bar: 120 μm). (B-E and G and H, fluorescence micrographs; F, cryo-scanning electron micrograph). (Online version: blue fluorescence denotes sclerenchymatic – lignified – tissue; red fluorescence denotes photosynthesising cells with chlorophyll.)

A B

Fig. 5. Stomatal views within a crypt. (A) Cryo-scanning electron micrograph showing a paradermal section of a B. elderiana crypt, the examined species with the lowest inter-stomatal distance. Although most crypt trichomes had been removed with the wood glue method, a few remained (scale bar: 100 μm). (B) Scanning electron micrograph showing a crypt of B. serrata, the examined species with the highest inter-stomatal distance; note the trichome bases that have remained after removing crypt trichomes with the wood glue method (scale bar: 55 μm).

[32]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

Table 4. Stomatal traits of six Banksia species differing in leaf dry mass per area (LMA). Mean values ± standard errors are given. (* Crypt wall area refers to total crypt wall area per crypt. ** Crypt wall area refers to crypt wall area at the flat crypt bottom.)

Stomatal Inter-Stomatal Inter-Stomatal Stomatal Stomatal Stomatal Stomatal Species LMA length distance distance number number number number (mm-2 leaf (mm-2 crypt (mm-2 crypt (g m-2) (μm) (μm) (stomatal lengths) (per crypt) area) wall area*) bottom area**) B. attenuata 231 ± 11 24.7 ± 0.3 15.9 ± 1.9 0.6 18 ± 1 216 228 679 ± 28 B. elderiana 507 ± 45 23.9 ± 0.5 7.9 ± 0.9 0.3 35 ± 3 184 230 907 ± 109 B. ilicifolia 159 ± 15 23.9 ± 0.3 25.1 ± 2.4 1.0 14 ± 1 190 218 708 ± 31 B. prionotes 182 ± 15 24.5 ± 1.1 19.5 ± 2.9 0.8 16 ± 2 208 234 675 ± 27 B. quercifolia 215 ± 8 24.9 ± 0.0 16.6 ± 0.0 0.7 8 ± 2 116 267 576 ± 49 B. serrata 134 ± 13 25.3 ± 0.6 29.5 ± 2.5 1.2 15 ± 1 179 155 500 ± 15

Table 5. Stomatal crypt traits and leaf dry mass per area (LMA, an indicator of sclerophylly) of ten Banksia species. Mean values ± standard errors are given.

Species LMA Crypt density Single crypt area Crypt depth Crypt conductance -2 -2 2 -2 -1 (g m ) (mm ) (mm ) (μm) (mmol CO2 m s ) B. attenuata 231 ± 11 11.71 ± 0.15 0.0186 ± 0.0006 192 ± 2 709 ± 45 B. candolleana 354 ± 6 9.53 ± 0.14 0.0293 ± 0.0011 241 ± 3 724 ± 32 B. elderiana 507 ± 45 5.20 ± 0.07 0.0310 ± 0.0012 310 ± 3 325 ± 3 B. hookeriana 218 ± 13 16.72 ± 0.18 0.0084 ± 0.0003 177 ± 2 494 ± 25 B. ilicifolia 159 ± 15 13.80 ± 0.42 0.0145 ± 0.0004 173 ± 2 720 ± 6 B. prionotes 182 ± 15 12.68 ± 0.20 0.0166 ± 0.0005 169 ± 2 777 ± 68 B. quercifolia 215 ± 8 14.53 ± 0.05 0.0028 ± 0.0000 127 ± 3 203 ± 3 B. repens 471 ± 22 6.45 ± 0.07 0.0339 ± 0.0010 289 ± 4 474 ± 4 B. serrata 134 ± 13 11.83 ± 0.12 0.0213 ± 0.0007 105 ± 7 1513 ± 120 B. victoriae 291 ± 14 9.03 ± 0.18 0.0265 ± 0.0008 283 ± 3 529 ± 13

[33]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

20 A r2 = 0.722 Fig. 6. Relationships between leaf dry mass per ) 16 P<0.01 -2 area (LMA, an indicator of leaf structure) and 12 crypt morphological characteristics. LMA is 8 inversely correlated with crypt density (A), and 4

Crypt density (mm positively correlated with single crypt area (B), 0 crypt depth (C) and single crypt volume (D).

) -2 2 0.04 B LMA (g m )

0.03

0.02

0.01 r2 = 0.544 Crypt aperture area (mm 0.00 P<0.05

0.4 C -2 LMA (g m ) 0.3

0.2

0.1

Crypt depth (mm) r2 = 0.794

0.0 P<0.001

) 3 0.008 D LMA (g m-2)

0.006

0.004

0.002 r2 = 0.833

Single crypt volume (mm 0.000 P<0.001 0 100 200 300 400 500 600 LMA (g m-2)

Low crypt density correlated with large crypt aperture area and deep crypts (Tables 2, 5). There was a 3-fold variation in crypt density (ranging from 5 to 17 crypts mm-2 leaf area) and crypt depth (ranging from 0.1 to 0.3 mm) and a 12-fold variation in crypt aperture area (ranging from 0.003 to 0.034 mm2) among the examined species. Species with crypts of greater aperture area, depth and volume had stomata that were closer to each other (Table 2).

[34]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

Species with high LMA had both thicker and denser leaves with high dry matter content, and had fewer crypts per unit leaf area, which had a greater aperture area and extended deeper within the mesophyll (Figs 6, 7, Table 2). LMA and % leaf dry matter were consistently the best correlates of the above crypt traits (Table 2). Single crypt volume ranged from 0.0005 to 0.0064 mm3, increasing with increasing LMA, leaf lamina thickness, leaf density and % leaf dry matter (Fig. 6, Table 2).

0.4 Fig. 7. Crypt depth significantly

0.3 increased with increasing leaf lamina thickness in 10 Banksia species 0.2 (P<0.001).

0.1 Crypt depth (mm) r2 = 0.764 P<0.001 0.0 0.0 0.2 0.4 0.6 0.8 Leaf thickness (mm)

0.8 Fig. 8. Leaf lamina thickness of 46 Banksia species classified according to Mast and 0.6 Givnish (2002) based on stomatal position: species with superficial stomata (N=6), with 0.4 stomata located in shallow pits (N=6) and with stomata in deep crypts (N=34). 0.2 Thickness was measured with digital callipers. Differences in leaf thickness

Leaf lamina thicknessLeaf lamina (mm) 0.0 between the group with superficial stomata Superficially In shallow pits In deep crypts and that with deep crypts were statistically Stomatal position significant (P<0.05).

[35]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

High-LMA leaves had more stomata per crypt due to larger crypt size, as well as higher stomatal density per crypt bottom wall surface area (Tables 2, 4). The latter was on average, 3-fold higher than the stomatal density per wall surface area of a crypt (Table 4), which indicates that stomata were more concentrated at the bottom of the crypt than on the crypt walls. The study of 46 broad-leaved Banksia species revealed a trend for greater encryption with increasing leaf lamina thickness. Average leaf lamina thickness for species with superficial stomata, shallow pits and deep crypts was 0.32 ± 0.04, 0.42 ± 0.05 and 0.45 ± 0.02 mm, respectively. The difference in leaf lamina thickness between species with superficial stomata and species with crypts was significant (P<0.05) (Fig. 8). All species with leaf lamina thickness greater than 0.46 mm had either shallow pits or deep crypts and all species with leaf lamina thickness greater than 0.63 mm had deep crypts.

2.5 DISCUSSION

Crypts add to the overall gas diffusive resistance, but their contribution is significantly less than that of stomata. This suggests that the primary role of crypts is not reduction of transpiration. Instead, relationships between crypt morphology and leaf structure were revealed that give novel insight into the functional significance of stomatal encryption. Thicker leaves have fewer, but larger and deeper stomatal crypts. Stomata in these species form clusters where they are invariably close to each other. These findings suggest that crypts shorten the average path length for, and thus facilitate diffusion of, CO2 to the upper- surface palisade cells in thick leaves.

2.5.1 Stomata contribute significantly more to leaf resistance than crypts Crypt conductance has been considered in theoretical models (Matthews, 2003; Chapter 3), but there is no published experimental methodology to estimate crypt conductance. To add to our current understanding of the implications of crypts for gas diffusion and water-use efficiency (WUE), the assumption that crypts can be treated as an additional boundary layer was tested by comparing the conductance of meshes calculated with Eqn 3 with measured mesh conductance. The good agreement in the values from these two methods (Fig. 1) suggests that the use of Eqn 3 is appropriate for estimating the crypt conductance of leaves.

[36]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

Gibson (1983) used an adaptation of the equation for calculating stomatal conductance (Nobel, 1974) to approximate the conductance of superstomatal chambers, shallow or deeper, where stomata are located in the photosynthesising stems of some desert plants. He -2 -1 found a range of 180 to 2000 mmol CO2 m s for deep and narrow-mouthed to shallow superstomatal chambers, which is in agreement with the values of crypt conductance of leaves obtained in the present study (Table 5).

The contribution of stomatal resistance (rs,m) to measured leaf resistance (rs+c) was found to be significantly higher (on average, 77%) than that of the calculated crypt resistance (rc) (on average, 23%) in the ten species examined. Eqn 3 is likely to provide an upper bound for the estimation of rc because a hemispherical end to the crypt and the location of stomata over much of the crypt surface (Fig. 4A) would reduce the effective crypt depth and the resistance for water vapour diffusion. The effect of varying crypt resistance by ± 20% is shown in Suppl. Fig. S3 in Appendix 1. This is in very good agreement with the conclusions of studies that modelled gas diffusion through crypts (Matthews, 2003). In these previous studies, all partitioning analyses were done for fully open stomata, i.e. the role of stomata was minimised and that of the crypts was maximised. It must be noted that the thickness of the boundary layer will influence the relative importance of stomata in controlling transpiration (Matthews, 2003) which in turn will influence the relative importance of the crypts (Matthews, 2003). Future studies are focusing on the effect of crypt architecture on gas diffusion using detailed 3-dimensional finite element modelling (Roth-Nebelsick et al., in preparation). Crypts account for an increasing proportion of leaf resistance as sclerophylly, indicated through LMA, increases. This suggests that crypts may have evolved to improve water-use efficiency (WUE). In this study, one outlier species, B. quercifolia, had a very narrow crypt aperture (0.003 mm2) compared with those of the other species (0.008-0.034 mm2; Table 3) -2 -1 -2 and the lowest gc (203 mmol m s ) despite its relatively low LMA (215 g m ). This species also had unusual crypt aperture morphology compared with the other Banksia species examined (Fig. 3), but similar to that of B. oreophila (Hassiotou, unpublished data), to which it is phylogenetically closely related (Mast and Givnish, 2002). The difference in WUE between species exhibiting different ratios of stomatal to crypt resistance is dependent on the assumed diffusivity ratio of water vapour to CO2 in the [37]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

crypts compared with that in the stomatal pores. If there is no difference in the diffusivity ratio, then crypts have no effect on WUE. However, when a diffusivity ratio of 1.37 was assumed for the crypts, the same as in the boundary layer (Kays, 1966), WUE improved by

3.3% in the species with the highest contribution of crypts to rs+c compared with the species with the highest contribution of stomata to rs+c (Fig. 2B). In reality, the diffusivity ratio in the crypts could be anywhere between 1.37 and 1.55 at 25 oC, with the latter being the ratio in the stomata (Lide, 1990). When leaf resistance (rs+c) increased relative to mesophyll and -1 2 boundary layer resistance, from 5 to 40 mol H2O m s, the ratio of total resistance to CO2 to total resistance to water vapour decreased by 34% (Table 3). That is, the greater the part of the resistance that is due to stomata and crypts relative to the mesophyll, the greater the benefit for the diffusion of CO2 relative to water vapour. Again, this is due to the different diffusivity ratio of CO2 to water vapour in the crypts relative to stomata, and also to the additional resistance to CO2 relative to water vapour in the mesophyll. Thus, while mesophyll resistance reduces photosynthesis, and therefore WUE, the addition of a significant resistance that applies to both CO2 and water (rc) will reduce the impact of mesophyll resistance on WUE. Thus, having the additional resistance at the level of crypts rather than at the stomatal level results in a small positive effect on WUE.

2.5.2 Do crypts primarily function to reduce transpiration? Crypts create a diffusive resistance in addition to stomata; however, leaves have other equally effective ways to reduce transpiration (e.g. fewer and smaller stomata) that are also more flexible (because stomata are variable resistances). Species with crypts do not necessarily have lower transpiration rates, suggesting that they are not adapted to have constitutively low transpiration. Moreover, the significantly higher contribution of stomata than that of crypts to leaf resistance suggests that the primary role of crypts is not reduction of transpiration. This raises the question, what is the functional advantage of stomatal crypts? There may not be a single purpose, as a given leaf structural trait may be multi- functional, serving different purposes in different environments (Press, 1999). Also, while crypts may serve a specific primary function, they have many secondary consequences for a leaf‟s physiology.

[38]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

Stomata positioned in crypts may be more protected from mechanical damage caused by abiotic or biotic stresses than stomata located at the leaf surface (Haworth and McElwain, 2008). Another role of crypts may be a reduced exposure of stomata to harsh environmental conditions, and thus moderation of the responsiveness and effectiveness of stomata by creating a microclimate and “buffering” the conditions between the atmosphere and the stomata. This agrees with the modelling studies of Roth-Nebelsick (2007), which showed that in a sunken stoma, the relative humidity above the stomatal pore was about 75% compared with 63% for a non-sunken stoma, when the ambient relative humidity was 50%. There may also be hydraulic consequences of crypts, as Brodribb et al. (2007) demonstrated that a reduced distance between veins and the evaporating surfaces improves leaf hydraulic conductance. This reduces water potential gradients at a given transpiration rate, possibly allowing higher stomatal conductance, which would be a favourable trait for plants with reliable access to water. The critical difference between leaves with and without crypts is not the magnitude of the resistance to gas exchange, but the difference in diffusivities for water vapour and CO2 in crypts versus stomata, as discussed above, and the fact that in a leaf with crypts, part of the resistance becomes “fixed”. What would be the advantage of a fixed resistance? It is speculated that crypts enhance safety, in the sense that in a species with crypts, even when stomata are wide open, there is still some resistance to diffusion imposed by the crypts compared with a species with superficial stomata. Therefore, if 10% of stomata failed to close, the increase in leaf conductance would only be 6-9% in the Banksia species examined, since the contribution of stomata to total leaf resistance was 62-90%. Similarly, if 10% of stomata were blocked, the decrease in leaf conductance would be less than 10%. Hence, crypts buffer the leaf during fine-scale regulation of leaf conductance: the conditions at the stomatal level are less extreme and the consequences of opening or closing the stomata are moderated.

2.5.3 A role of crypts in facilitating diffusion of CO2 Crypt morphology and stomatal traits vary in parallel with sclerophyllous traits. Leaf dry mass per area (LMA) is a leaf morphological trait that has been used frequently as an indicator of sclerophylly (Sobrado and Medina, 1980; Witkowski and Lamont, 1991;

[39]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

Gratani and Varone, 2006) and is shown to be closely correlated with other leaf structural traits, such as leaf lamina thickness, leaf density and percentage leaf dry matter in Banksia species (Table 2). This study has shown that high-LMA, thicker and denser leaves had fewer crypts per unit leaf area, but with greater volume, aperture area and depth into the mesophyll. This was in accordance with the enhanced encryption with increasing leaf lamina thickness in 46 Banksia species (Fig. 8).

Since mesophyll resistance to CO2 diffusion increases with increasing LMA (Hassiotou et al., 2009a or Chapter 5) and leaf thickness (Hassiotou, unpublished data) in these sclerophyllous Banksia species, deeper crypts may have evolved as a response to reduce the resistance by shortening the average path length for CO2 diffusion, thus facilitating CO2 diffusion to well-lit actively photosynthesising palisade cells near the adaxial surface. Supportive of this argument was the increased stomatal density on the bottom crypt walls, which are closer to palisade mesophyll, compared with the side walls. Stomata on the crypt side walls would be more effective for facilitating CO2 diffusion locally to abaxial mesophyll cells. This proposed role for crypts in facilitating diffusion assumes that diffusion through mesophyll is slower than through hairy crypts because of the tortuous path and narrow intercellular channels of the former. It is thought that early in its evolution, sclerophylly was related to slow growth, whereby the toughness of leaves ensured longer leaf life spans that reduced annual nutrient losses (Chapin, 1980). This leaf anatomy was associated with lower rates of photosynthesis (Flexas et al., 2008). It is hypothesised that the evolution of stomatal crypts facilitated the diffusion of CO2 to adaxial mesophyll layers in thick leaves. The enhanced encryption with increasing leaf thickness in combination with the presence of stomata both on the side walls and the bottom wall of a crypt in all the examined species further suggest that stomatal encryption allows for multiple pathways to the sites of photosynthesis. Highly sclerophyllous species (with high LMA, leaf lamina thickness and dry matter content) had more stomata per crypt. This was due to both the greater crypt wall surface area and the lower inter-stomatal distance of high-LMA leaves. This signifies that high- LMA species with larger crypts have more stomata per crypt not only because a larger crypt can accommodate more stomata, but also because stomata are more densely packed within the larger crypt. Decreasing inter-stomatal distance may cause the vapour cups above the

[40]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

stomata to overlap. The number of stomata per crypt might be excessive, particularly in high-LMA leaves, and perhaps not all of them are active at the same time. Given the long life span of these leaves, high stomatal number per crypt may act as a “backup” mechanism in case some stomata become inactive (e.g. dust-blocked) to ensure that the gas exchange ability of the leaf is not compromised in the longer term. Alternatively, thick leaves may have more stomata per crypt to compensate for the resistance imposed by it. Interestingly, Banksia species with higher numbers of stomata per crypt have larger mesophyll volume per areole (Hassiotou unpublished data). This observation indicates that stomatal density matches photosynthetic capacity in leaves with stomatal crypts, which is similar to leaves with superficial stomata (Flexas et al., 2008). The number of stomata per mm2 leaf area (116-216) was within the range previously reported for species with crypts (144-388 mm-2; Mohammadian, 2005), but was relatively low compared with stomatal densities of hypostomatous tree species with superficial stomata (ranging from 216-713 stomata mm-2; Gindel, 1969; Ridge et al., 1984; Beerling and Chaloner, 1993; Kelly and Beerling, 1995; Mohammadian, 2005). Higher stomatal densities have been associated with smaller stomata (Gratani and Varone, 2004), but despite higher stomatal densities of the species reported by Gindel (1969) and Ridge et al. (1984) compared with Banksia, they either have similar or larger stomatal pore lengths. Leaf trichomes were observed both on the abaxial leaf surface and inside the crypts of Banksia species. Crypt trichomes are often different in density and type compared with leaf surface trichomes (e.g. in B. attenuata); therefore, a distinct function of trichomes in crypts is likely. Trichomes that cover the crypt aperture may prevent dust from blocking the crypt and/or stomatal pores, as has been shown for stomatal pores in mangroves (Paling et al., 2001), and in Dryandra praemorsa, B. baxteri, B. caleyi, B. media and B. menziesii (Matthews 2003). Since leaves of B. species live for up to 13 years (Witkowski et al., 1992), the prevention of dust entering the crypts is particularly important. Trichomes have often been considered a xerophytic trait (Baldini et al., 1997; Ripley et al., 1999) and thus may add to gas diffusive resistance. However, Roth-Nebelsick et al. (Chapter 3) generated a detailed physical model of vapour diffusion in B. ilicifolia and found a reduction of less than 10% in crypt conductance due to trichomes.

[41]

CHAPTER 2: STOMATAL CRYPTS AND CO2 DIFFUSION

The present study considered the impact of stomatal crypts on gas diffusion as well as the relation of crypt architecture to leaf structure, and it presented evidence suggesting a function of crypts in the facilitation of CO2 diffusion to the upper-surface palisade mesophyll cells in thick leaves. The development of methodologies to measure crypt conductance would add greatly to our current understanding of the effect of crypts on gas diffusion. Lacking this, modelling of crypts of differing architecture and stomatal and trichome characteristics will give insights into the function of these structures and their effects on leaf physiology.

[42]

CHAPTER 3

STOMATAL CRYPTS HAVE SMALL EFFECTS ON TRANSPIRATION: A NUMERICAL MODEL ANALYSIS

3.1 ABSTRACT ...... 44

3.2 INTRODUCTION ...... 44

3.3 MATERIALS AND METHODS ...... 46 3.3.1 Study species ...... 46 3.3.2 Microscopy ...... 48 3.3.3 Simulation method ...... 48 3.3.4 Model setup...... 49

3.4 RESULTS ...... 55 3.4.1 Spatial pattern of the humidity gradient ...... 55 3.4.2 Stomatal and crypt conductance, and transpiration ...... 57

3.5 DISCUSSION ...... 59

3.6 CONCLUSIONS ...... 65

[Roth-Nebelsick A, Hassiotou F, Veneklaas EJ (2009). Evaluation of the influence of stomatal crypts and crypt trichomes on transpiration: A numerical model analysis. Plant Physiology 151: 2018-2027]

[43]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

3.1 ABSTRACT

Stomata arranged in crypts with trichomes are commonly considered to be adaptations to aridity due to the additional diffusion resistance associated with this arrangement; however, information on the effect of crypts on gas exchange, relative to stomata, is sparse. In this study, three-dimensional Finite-Element models of encrypted stomata were generated using commercial Computational Fluid Dynamics software. The models were based on crypt and stomatal architectural characteristics of the species Banksia ilicifolia, examined microscopically, and variations thereof. In leaves with open or partially closed stomata, crypts reduced transpiration by less than 15% compared to non-encrypted, superficially positioned stomata. A larger effect of crypts was found only in models with unrealistically high stomatal conductances. Trichomes inside the crypt had virtually no influence on transpiration. Crypt conductance varied with stomatal conductance, boundary layer conductance and ambient relative humidity, as these factors modified the three-dimensional diffusion patterns inside crypts. It was concluded that it is unlikely that the primary function of crypts and crypt trichomes is to reduce transpiration.

3.2 INTRODUCTION

Stomatal structure varies widely among plant taxa (Meidner and Mansfield, 1968). Since stomata represent the main interface for gas exchange between the leaf interior and the atmosphere, it is generally believed that their morphological and architectural features represent adaptations to environmental factors that affect transpiration and photosynthesis. In many xeromorphic or sclerophyllous species, stomata are clustered in depressed epidermal areas called “stomatal crypts” (Napp-Zinn, 1973; Metcalfe and Chalk, 1979). These invaginations of the leaf epidermis represent a prominent structural feature that is usually considered to be an adaptation to drought (Lösch et al., 1982; Larcher, 2003), under the assumption that the total leaf resistance to diffusion is increased by adding a crypt component to the stomatal component, referred to in this paper as crypt resistance or crypt conductance (resistance = 1/conductance). Moreover, trichomes that are often present inside the crypts may further increase crypt resistance.

[44]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

Studies concerning the significance of the impact of the recession of stomata on leaf transpiration are sparse considering the general acceptance of this idea. Lösch et al. (1982) studied transpiration and stomatal resistance of various Mediterranean perennials, including Nerium oleander, representing a classic example of stomatal crypts filled with trichomes. The transpiration rate of N. oleander was among the highest values in the examined species, which suggests that stomatal crypts are not necessarily associated with high leaf resistances. Recent evidence indicates that the evolution and presence of sunken and encrypted stomata is not restricted to dry habitats. Conifers with sunken stomata (which often have stomatal plugs), for example, occupy a range of habitats (Brodribb and Hill, 1997). Moreover, in a study of the gas exchange of the cloud forest species Drimys winteri, there was no evidence for a significant increase in resistance caused by sunken stomata (Feild et al., 1998). In a recent study, the association of sunken stomata with drought was investigated by analysing the evolutionary history of Proteaceae (Jordan et al., 2008). No straightforward coupling of stomatal architecture to climate was found, but a more complex form-function relationship. Only very pronounced encryption was associated with drought, whereas many other sunken or encrypted stomatal types did not appear to be drought-related (Jordan et al., 2008). Early models of leaf diffusive resistance demonstrated the importance of stomatal dimensions and architectural variations of the leaf epidermis. A well-established mathematical model for the inter-relationship between stomatal structure and resistance by Parlange and Waggoner (1970) calculates the resistance based on the number of stomata per unit leaf area and the stomatal pore area, length and depth, and takes into account that the gradient of humidity (or CO2) “bulges out” from the stomatal pore. This effect results in a virtual prolongation of the pore channel. In a single sunken stoma, the resistances of the stoma and the stomatal antechamber are additive (Lee and Gates, 1964). If the stoma is deeply sunken and the cross-sectional area of the antechamber is similar to the area of the open pore, then the additional resistance of the antechamber will be significant. Treating a catena of resistances as additive is, however, only appropriate if 1) the architecture can be considered as being composed of connected ducts of simple architecture (e.g. cylinders), and 2) the single channel components are “tight”, i.e. there is no gas exchange across the channel walls. Gas-exchanging pores are distributed all over the walls of crypts; thus

[45]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

proper assessment of diffusion resistance in such spaces requires a three-dimensional (3D) modelling approach. Further complexity of gas diffusion through crypts arises from the presence of trichomes within these structures. Ignoring the trichomes may lead to serious errors with respect to resistance since their presence turns the space from a simple cavity into a fibrous porous medium. Extensive research in certain technical disciplines, such as filter or textile technology, has documented the importance of fibre density, diameter and arrangement for the permeability of fibrous materials (Tomadakis and Sotirchos, 1993; Lawrence and Liu, 2006; Shen and Chen, 2007). In this study, transpiration through stomatal crypts was examined using a computer simulation approach. Banksia ilicifolia R.Br. (Proteaceae), representing a typical example of stomatal encryption, was chosen as a case study. Microscopic analyses of the crypt architecture as well as the stomatal and crypt/trichome characteristics led to the generation of a 3D Finite-Element model of a B. ilicifolia crypt. In this model, the effect of the crypt and crypt trichomes on transpiration was examined. Subsequently, model variations that considered a range of crypt architectures and stomatal characteristics were generated, with the aim of investigating the partitioning between the stomatal and the crypt resistance in situations of differing crypt and stomatal arrangement, similar to those found in nature.

3.3 MATERIALS AND METHODS

3.3.1 Study species Initial model parameters were based on a species of the genus Banksia (Proteaceae), which is an excellent example of stomatal encryption, with the majority of its species possessing crypts (Mast and Givnish, 2002). The species Banksia ilicifolia R.Br. has a crypt architecture that is representative for the genus. Diffusion was first modelled for a typical crypt, after which morphological variations both at the crypt and stomatal levels were introduced, and their effects on transpiration were examined. B. ilicifolia is a shrub or tree (up to 10 m high) that is relatively common in the south-west of Australia and occurs within 70 km of the coast in open woodlands on sandy soils and low-lying flats (George, 1996). Its leaves have serrated edges and non-recurved margins.

[46]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

Crypt and stomatal traits were measured on young fully mature leaves of four-year old plants. The plants were grown from seed in 10-L pots containing a mixture of river sand and potting mix, in Perth (Australia) and were maintained outdoors. Microscopic examination confirmed that stomata were exclusively situated in crypts on the abaxial leaf surface, as was previously reported by Mast and Givnish (2002). In mature leaves, trichomes were almost absent from the adaxial surface, where trichome bases indicated that trichomes fell off during leaf development; however, the abaxial leaf surface was pubescent, with trichomes mainly situated inside the crypts and often extending outside the crypts covering part of the surface (Fig. 1).

A B

C D

Fig. 1. Leaf and stomatal crypt anatomy of Banksia ilicifolia. (A, B) Scanning electron micrographs of the abaxial leaf surface before (A) and after (B) partial removal of trichomes (scale bars: 250 μm). (C) Scanning electron micrograph showing a single crypt after removal of trichomes; trichome bases and stomata are clearly visible (scale bar: 35 μm). (D) Cryo-scanning electron micrograph of a leaf transverse section; a crypt is highlighted (scale bar: 92 μm).

[47]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

3.3.2 Microscopy Optical, scanning and cryo-scanning electron micrographs were used to analyze 100 crypts of young, fully mature leaves of B. ilicifolia, using Image J software (Abramoff et al., 2004). The following parameters were determined: crypt diameter, crypt depth, number of stomata, trichome diameter, trichome number and stomatal dimensions. Crypt diameter was calculated from the measured crypt area assuming a circular shape for all crypts. Crypt area was measured in variable-pressure scanning electron micrographs (ESEM, Philips XL30) of the abaxial leaf surface after partial removal of surface trichomes by applying wood glue that was peeled off once hardened. Crypt depth was measured in optical transverse sections obtained using fluorescence microscopy (Zeiss Axioplan 2). The number of stomata per crypt, trichome diameter, trichome number and stomatal dimensions were measured using both scanning (ESEM Philips XL30) and cryo-scanning (CSEM Cambridge S360, Cryo- system Oxford CT1500) electron microscopy, capturing top and side views of crypts. The leaf and crypt anatomy of B. ilicifolia is shown in Fig. 1.

3.3.3 Simulation method For diffusion in a complex structure comprising sources and/or sinks, Fick´s law in its general form must be used: J D gradC (1) with J = diffusional flux, D = diffusion coefficient, C = concentration of diffusing substance and grad = the differential operator (∂/∂x, ∂/∂y, ∂/∂z), with grad C representing the concentration gradient in all three spatial directions. The mathematical description of the diffusion process is complete if the principle of mass conservation is included. Then the diffusion equation reads as (t = time): c div(D gradC) (2) t It is usually impossible to solve Eqn 2 for complex structures containing sources or sinks. This problem can be assessed by numerical methods, such as the Finite Element (FE) Method. This mathematical scheme converts the partial differential equation into a set of algebraic equations (Zienkiewicz and Taylor, 1989). Application of this method requires the subdivision of the structure considered into a mesh consisting of elements with a simple

[48]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

geometry, such as triangles for two-dimensional problems, or tetrahedrons for 3D problems. The calculation is then performed for each element. Since the elements making up the considered structure are connected at their nodes, an approximate solution is achieved for the entire structure. The degree of exactness increases with the number of elements and converges to the exact value. In practice, the number of elements is increased until the solution does not change significantly. 3D FE models with a high spatial resolution can now be run on common PC´s. In this study, the commercial FE (Finite Element)-based programme FIDAP (Version 8.7, ANSYS, USA) was applied. FIDAP is a CFD (Computational Fluid Dynamics) programme that includes a diffusion module. The meshes were generated by using the mesh generator GAMBIT (2.1, ANSYS, USA). The results were calculated for the steady state.

3.3.4 Model setup

For the first model (CS50A) representing the typical morphology of a B. ilicifolia crypt, the parameters are shown in Table 2. The crypt was modeled as a cubic shape with slightly curved walls and bottom (Fig. 2A). Crypt depth was defined as the maximum extension from the crypt aperture to the crypt bottom and crypt diameter as the maximum distance between opposite side walls. Generation of an FE-mesh required a slight simplification of the trichomes that were modeled as growing out of the crypt walls and turning towards the crypt aperture (Fig. 2B). In accordance with the SEM studies (Fig. 1), 14 stomata were positioned inside the crypt with two stomata being placed at each side wall and 6 stomata located at the bottom of the cavity (Fig. 2B). Temperature for all simulations was set at 20°C, and the model was devised to be isothermal. External concentration of water vapour was set at a relative humidity (RH) of 50% and that at the internal stomatal pore was set at 99%. This corresponds to a vapour pressure deficit of 1.17 and 0.02, respectively. The internal pore is defined here as the plane where the stoma leads into the leaf interior. The plane where the stoma leads into the crypt or the external environment (in leaves without crypts) is termed external stomatal pore. The stomata of these basic models are modelled as rectangular pores with a length of 9 µm, width of 4 µm and depth of 15 µm.

[49]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

Table 2. Crypt and stomatal parameters of the FE model variations. For explanations of the model names see Table 1.

Crypt parameters Stomatal parameters

Diameter Depth Number of Trichome Trichome Number of Pore width Pore length Pore depth Models Porosity (%) (μm) (μm) trichomes diameter (μm) orientation stomata (μm) (μm) (μm)

CS50A 140 166 36 9 ascending 94 14 4 (standard) 9 15

CS50H 140 166 37 6 and 8 horizontal 94 14 4 (standard) 9 15 CS , CSB , CSB , 50 50 65 140 166 No trichomes - - 100 14 4 (standard) 9 15 CSB80

RSB50 140 138.5 No trichomes - - 100 14 4 (standard) 9 15

CNB50, CNB65, CNB80 140 166 No trichomes - - 100 14 1.8 (narrow) 9 15

CLB50, CLB65, CLB80 140 166 No trichomes - - 100 14 10 (large) 20 10

SSB50, SSB65, SSB80 na na na na na na na 4 (standard) 9 15

SLB50, SLB65, SLB80 na na na na na na na 10 (large) 20 10

[50]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

A B

C D

Fig. 2. Graphical representations of some FE models. (A) 3D representation of the FE mesh of the

Banksia ilicifolia model CS50A. The holes in the outer walls show the position of trichomes. The “boxes” on the crypt wall are the stomata. (B) Transparent representation of the model showing the trichome architecture of model CS50A. (C) Transparent representation of the model showing the trichome architecture of model CS50H. Here, the trichomes are arranged parallel to the stomatal aperture (stomata not shown). (D) Model CSB50 as an example of a model with a boundary layer. The boundary layer is represented by an air space above the leaf surface (green).

Variations of the basic model included models without trichomes, with trichomes in different orientations (Fig. 2C), with narrower and larger stomata, and with reduced crypt depth. Moreover, the effects of different ambient RH and the addition of a leaf boundary layer (Fig. 2D) were examined for crypts with different stomatal sizes. For a description of the parameters of the different models, see Table 2.

[51]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

In all models with trichomes, porosity was kept constant despite the difference in trichome arrangement. Porosity (P) is defined as the fractional value of the air with respect to the entire crypt volume: V P p (3) V with Vp = volume occupied by air and V = entire crypt volume. A local boundary layer was constructed assuming a leaf of 2 cm width and a distance from the leading edge of 1 cm, at a moderate wind velocity of 0.32 m s-1. For these conditions, the local wind velocity close to the leaf surface can be calculated according to the following equation (Vogel, 1996):

U U 0.32yU (4) y x with Uy = wind velocity at a perpendicular distance y from the surface, U = ambient velocity, x = distance from the leading edge in the wind direction, ρ = air density, μ = viscosity of air. The oncoming wind carried the ambient humidity of 50%. In the models to which no boundary layer was added, the crypt aperture was directly exposed to ambient humidity. This corresponds to conditions of high wind velocity and turbulence intensity which prevent the development of a significant boundary layer. To evaluate the impact of the crypts on transpiration, models were devised in which stomata were not encrypted but located on the leaf surface. Stomatal density per unit leaf area corresponded to the mean stomatal density per unit leaf area of B. ilicifolia leaves (190 mm-2). In these models, stomata were of standard or large size (dimensions listed in Table 2), a boundary layer was present and ambient RH was 50, 65 or 80%, as in the corresponding variants of the crypt model. The program calculates RH at each mesh element for the steady state situation. From this humidity matrix, the diffusion through a single stoma, and therefore transpiration rate for each stoma of the crypt, can be calculated using Fick‟s law (Eqn 1). Transpiration rate per crypt was computed by summing the transpiration rates of all individual stomata.

Transpiration per leaf area (Jleaf) was obtained by multiplication of the transpiration rate per crypt with the crypt density (ncrypt = number of crypts per unit leaf area):

[52]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

J leaf J crypt ncrypt (5) Stomatal and crypt conductances were derived by dividing the transpiration rate by the corresponding humidity gradients. For the stomatal conductance (gs) of the superficial stomata, the humidity gradient included the superstomatal air layer but not the boundary layer (Parlange and Waggoner 1980; Vesala, 1998):

J leaf gs (6) (cinternal pore cair layer )

The height of the air layer (Hair layer) (was calculated according to Vesala (1998) (with nstomata = stomatal density and r = radius of stomatal pore): 1 H air layer (7) 4nstomatar The conductance of the stomata situated in the crypt was calculated based on the humidity gradient between internal and external pore. A superstomatal air layer was not considered, because the crypt stomata were – unlike those of the superficial stomata – part of a composite duct (stoma-crypt) and therefore the gradient outside a stoma represented the gradient inside the subsequent duct section “crypt”.

Crypt conductance (gc) was calculated by dividing the transpiration rate per leaf area by the average difference between water vapour concentration at the external stomatal pores and the ambient water vapour concentration, that is, the humidity in the outer atmosphere:

J leaf gc (8) (cexternal pore cambient air ) Using ambient air is appropriate for a single pore (stoma or crypt) since the superstomatal air layer is undefined in this case (Parlange and Waggoner 1970; Vesala, 1998, see also Eqn 7). To check the model, the conductances of the superficial stomata models were compared with estimates obtained with the equation of Parlange and Waggoner (1970). To this end, the width of the stomatal rectangle was multiplied by 4/ to correct for the area difference between a rectangular and an elliptical pore (Parlange and Waggoner, 1970).

[53]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

The quantitative influence of the crypts on transpiration rate, termed “crypt effect”, was calculated by relating the transpiration rate of the crypt models to the transpiration rates of the models with superficial stomata at the same ambient RH and stomatal size:

J leaf Crypt effect [%] 1 superficial 100 (9) J leaf

A B

C D

Fig. 3. Humidity patterns inside four crypt models. (A) CS50A: standard stomata, ascending trichomes; (B) CS50: standard stomata, no trichomes; (C) CSB50: standard stomata, no trichomes, boundary layer present; (D) CNB50: narrow stomata, no trichomes, boundary layer present. The legends provide the colour code for RH (%).

[54]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

3.4 RESULTS

3.4.1 Spatial pattern of the humidity gradient The 3D spatial humidity distribution in steady state conditions is presented in Fig. 3 for four models, namely with/without trichomes, with/without boundary layer and with standard and narrow stomata. For all models, the distinct “vapor concentration cups” around the stomata are visible. This is true for the stomata located at the crypt bottom as well as for the stomata seated at the crypt side walls. Further away from the stomata, the humidity contours merge into a contour shell. This results in vapour shells that are bulged towards the crypt bottom with a shape similar to a “hanging cloth”. For the models including a boundary layer (Figs 3C and 3D), the development of a vapour cup around the crypt aperture is clearly visible. The similarity of the humidity patterns in models that differ only with respect to absence/presence of trichomes (Fig. 3A versus 3B) indicates that trichomes do not significantly influence transpiration. The differences between models shown in Figs 3C and 1D with respect to the humidity gradient are due to their different stomatal sizes.

100

A CS50A B CSB50 CS CS 90 50 50 CS50H CNB50 80

%

RH 70

60

50 15 15 181 0 30 60 90 120 150 180 0 50 100 150 200 250 300 350 400 Diffusion path ( m) Diffusion path ( m)

Fig. 4. Gradient of relative humidity (RH) along a line transect from (A) an internal stomatal pore (at 0 µm) to the crypt opening for CS50A (standard stomata, ascending trichomes), CS50 (standard stomata, no trichomes) and CS50H (standard stomata, horizontal trichomes); and (B) from an internal stomatal pore (at 0 µm) to the top of the boundary layer for

CSB50 (standard stomata, no trichomes, boundary layer present) and CNB50 (narrow stomata, no trichomes, boundary layer present). The transition from the external stomatal pore into the crypt is indicated by a dotted line at 15 μm and the transition from the crypt to the boundary layer is indicated by a dotted line at 181 μm. [55]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

Table 1. Summary of all FE crypt models listing the parameters that were varied and the results. Models with crypts carry code C, while models with stomata on the leaf surface have code S. Model 1 is the basic model and best representation of Banksia ilicifolia crypts with fully open stomata. Models 1-3 explore the effect of crypt trichomes, where code A represents trichomes that are ascending towards the crypt aperture, and code H represents trichomes that are horizontally oriented. Models 4-12 explore the effect of stomatal conductance at ambient relative humidities of 50, 65 and 80% (indicated as subscripts in the model code). Stomatal size codes are N (narrow), S (standard) and L (large). Stomatal dimensions are given in Table 2. Model 13 explores the effect of reduced crypt depth, indicated by the code R replacing the C.

CRYPT MODELS 1 2 3 4 5 6 7 8 9 10 11 12 13

CS50A CS50H CS50 CNB50 CNB65 CNB80 CSB50 CSB65 CSB80 CLB50 CLB65 CLB80 RSB50 Stomatal size S S S N N N S S S L L L S Trichomes A H ------Boundary layer No No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Relative humidity (%) 50 50 50 50 65 80 50 65 80 50 65 80 50 Crypt depth 166 166 166 166 166 166 166 166 166 166 166 166 138.5 Transpiration rate 3.91 3.89 3.97 2.05 1.35 0.67 3.58 2.31 1.15 11.7 7.47 3.74 3.74 (mmol m-2 s-1) Stomatal conductance 0.37 0.35 0.37 0.17 0.17 0.17 0.37 0.37 0.37 3.30 3.30 3.30 0.37 (mol m-2 s-1) Crypt conductance 1.66 1.62 1.78 1.52 1.25 1.07 1.17 1.02 0.93 1.51 1.41 1.36 1.37 (mol m-2 s-1) Crypt effect (%) 15.1 14.8 14.9 48.4 48.5 48.5 11.3 ANALOGUES WITH STOMATA ON SURFACE

SSB50 SSB65 SSB80 SLB50 SLB65 SNB80 Transpiration rate 4.22 2.71 1.35 22.6 14.5 7.26 (mmol m-2 s-1) Stomatal conductance 0.39 0.39 0.39 2.26 2.26 2.26 (mol m-2 s-1)

[56]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

Gradients of RH within stomatal pores, crypts and the leaf boundary layer varied considerably among models. The steepest gradient was always located along the stomatal channel. Directly above the stomatal pores, the gradients became much flatter, and even flatter in the boundary layer (Fig. 4; for model parameters refer to Table 1). The steepness of the stomatal gradient increased with decreasing stomatal pore size, as expected. For the models with standard and narrow stomata, the stomatal gradients (reflecting the stomatal conductances) led to low average crypt humidity, particularly in the absence of a boundary layer (Fig. 4A). The presence of trichomes in the crypts (model CS50A and CS50H versus model CS50) modified the humidity gradient only slightly (Fig. 4A). The addition of a boundary layer did not have a large effect on total conductance and transpiration: transpiration dropped by 10% in CSB50 compared with CS50 (Table 1). The presence of a boundary layer also reduced the crypt gradient. Crypt conductance was 34% lower in

CSB50 compared with CS50 (Table 1).

3.4.2 Stomatal and crypt conductance and transpiration To check the model approach, the conductances of the models with superficial stomata were compared with estimates obtained by applying the equation of Parlange and Waggoner (1970). For example, the model with superficial stomata of standard size had a stomatal conductance of 0.39 mol m-2 s-1, which compared favorably with a stomatal conductance of 0.35 mol m-2 s-1 resulting from applying the equation of Parlange and Waggoner (1970). The stomatal conductance values were different for the models with different stomatal sizes, but were hardly affected by crypt or boundary layer properties, as expected (Table 1). The dominant role of the stomata is evident in Fig. 5, showing a strong relationship between stomatal conductance and transpiration rate, with minor effects of trichomes and boundary layers. In models with standard stomata the presence of a boundary layer decreased transpiration by 10% and the presence of trichomes decreased transpiration by 2% (Table 1). Only in models with "large" stomata which had unrealistically high stomatal conductances did crypts have substantial effects on transpiration. Changes in ambient RH affected transpiration rate, but not the magnitude of the effect that crypts have on transpiration rate (“crypt effect” in Table 1; compare models CSB50, CSB65 and CSB80). [57]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

CS A CS50A

) CS50A 25 CS50 -1 50 CS s CS50H CS50H

-2 CSBCS50H 20 50 CSBCNB50 CNB50 CNB50 CLB50 15 CS SSB5050

CS50H SLB5050 10 CSBCS 50H CS50 CNBCSB50 CS5050H CNBCSB 50 5 50 CNB50

Transpiration rate (mmol m (mmol rate Transpiration 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 -2 -1 Stomatal conductance (mol m s )

Fig. 5. Transpiration rate against stomatal conductance of crypt models CS50A (standard stomata, vertical trichomes), CS50 (standard stomata, no trichomes), CS50H (standard stomata, horizontal trichomes), CSB50 (standard stomata, no trichomes, presence of boundary layer), CNB50 (narrow stomata, no trichomes, presence of boundary layer), CLB50 (large stomata, no trichomes, presence of boundary layer), SSB50 (superficial standard stomata) and SLB50 (superficial large stomata). Open symbols represent the models without a boundary layer, whereas filled symbols the models including a boundary layer (black: crypt stomata, grey: superficial stomata).

Crypts with the same dimensions showed, contrary to stomatal conductances, significant variation in conductance (Table 1). The presence of trichomes decreased crypt conductance by only 5-9% (compare models CS50A, CS50H and CS50; Table 1). Decreasing crypt depth by 16% led to an approximately proportional increase of crypt conductance of 15% (model

RSB50), as expected. Interestingly, factors which also showed a large influence on crypt conductance were stomatal conductance, boundary layer conductance and RH. At all RHs, crypt conductances were lowest in models with intermediate stomatal conductances. Crypt conductances were 13-23% higher for models with narrow compared with standard stomata, corresponding to a 50% decrease in stomatal conductance (Table 1). When RH was raised from 50% to 80%, crypt conductances decreased by 30%, 21% and 10% for narrow, standard and large stomatal models, respectively (Table 1). The presence of a

[58]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

boundary layer decreased crypt conductance by 34% (models CS50 and CSB50), although it is worth noting that transpiration rate and the RH gradient in the crypt decreased by only 10%. Crypt conductance thus depended not only on the structure of the crypt, but also on the water vapour sources (stomata) and on the environmental conditions.

3.5 DISCUSSION

The results of this numerical study provide detailed accounts of the effect of stomatal encryption and crypt trichomes on the humidity profile, total leaf conductance and transpiration. Of course crypts and trichomes will also affect overall conductances to CO2, but the analysis of these effects involves further complexities (see below). The models demonstrate that, at realistic stomatal conductances, stomatal encryption causes only minor reductions in leaf transpiration. This small effect occurs despite a near-doubling of the epidermal surface area and very strong clustering of stomata in the crypts. Internal crypt surface area in B. ilicifolia is about five times larger than the crypt pore area, and crypts occupy about a fifth of the abaxial side of the leaf. The very high density of stomata at the bottom of the crypts represents a very moderate stomatal density when expressed per leaf surface area (190 mm-2). In fact, species in the Banksia genus that have stomata on the surface tend to have more stomata per unit leaf surface area than species with crypts (Hassiotou et al., 2009b or Chapter 2). Conductances estimated by the model are in reasonable agreement with empirical values for B. ilicifolia. Maximum measured stomatal conductance of well-watered plants of the species is 0.36 mol m-2 s-1 (Hassiotou, unpublished data), which is very close to the value -2 -1 of 0.37 mol m s in the models with open stomata of standard size (e.g. CS50A). Crypt conductances cannot be measured in leaves, but the values obtained here are consistent with estimations using validated simpler equations (discussed below). In the models with standard stomata, which represent B. ilicifolia leaves with open stomata

(models CSB50-CSB80), the crypt effect was only 15%. As stomata close, the stomatal contribution to total resistance increases and crypt effects become even smaller. Jordan et al. (2008) found that in the Proteaceae, deep stomatal crypts evolved preferably in dry climates. Those authors defined deep crypts as having a depth that is at least twice the

[59]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

diameter of the crypt opening. According to this criterion, the B. ilicifolia crypts do not fall in the deep crypt category, and may thus not have evolved as an adaptation to dry environments. If increased crypt depth results in a roughly proportional decrease of crypt conductance, as suggested by model RSB50, increasing the depth of the basic model CS50A to a value equaling twice the diameter would yield a crypt conductance of approximately 1.15 mol m-2 s-1, which is still almost three times as high as the maximum stomatal conductance measured for this species. While the contribution of crypts to total leaf conductance is generally small, their influence is maximized at high stomatal conductance, i.e. when stomata are large and numerous within crypts. Interestingly, stomatal densities at the crypt bottom, which are up to 900 mm- 2 in Banksia species, are positively correlated with crypt depth (Hassiotou et al., 2009b or Chapter 2). Therefore, crypt effects are largest in the species with deep crypts that are known to have evolved in dry environments (Jordan et al., 2008), supporting the idea that there may be a water-saving advantage in these more extreme forms of crypts, perhaps when such species respond to temporarily enhanced water availability by fully opening their stomata. It is to be expected that the spatial distribution of stomata inside a crypt has some influence on the effect of crypts on transpiration. For example, if stomata are located more closely to the crypt aperture, transpiration is somewhat higher (with all other parameters constant). An important and novel outcome of the models is that crypt conductance is quite dependent on stomatal conductance and the external environment, i.e. on the factors that define the rates at which water vapour can enter and leave the crypt. Changes in crypt conductance without changes in crypt structure are caused by complex changes in humidity gradients at different transpiration rates. Crypt conductance decreased when transpiration rates decreased due to a higher atmospheric humidity or due to the presence of a boundary layer, whereas the change in crypt conductance with increasing stomatal conductance was more complex. It should be reiterated at this point that the composite stomata-crypt diffusion pathway is not a simple resistance-in-series duct. Changing the stomatal pore area, with all other parameters remaining constant, will not only change stomatal conductance but also the spatial humidity distribution inside the crypt, affecting crypt conductance. An impact of boundary layer thickness on crypt conductance is to be expected since the development of a

[60]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

boundary layer above the crypt allows for the development of water vapour cups above the crypt apertures. The presence of a boundary layer decreases the transpiration rate and increases the influence of the crypt on diffusion. With a thick boundary layer, a crypt will therefore impede transpiration more strongly than under conditions that favor a thin boundary layer (for example, with high wind velocities). It must be noted, however, that better quantitative assessment of the influence of boundary layers on crypt conductance will require the modeling of populations of crypts on leaf surfaces, in order to properly describe the RH surfaces. Such a model would also allow the definition of vapor cups around crypts and a clear delimitation between crypt resistance and boundary layer resistance, which was not possible in the single-crypt model. A decrease in stomatal conductance, shown in models with narrow stomata, caused an increase in crypt conductance, whereas a decrease in boundary layer conductance (model

CSB50 compared with CS50) caused a decrease in crypt conductance. This means that under conditions which are usually associated with aridity, partial stomatal closure and high boundary layer conductance, crypts appear to offer no great advantage with respect to water conservation. The limited impact of crypts on total leaf resistance is consistent with the observation that species with crypts can actually have high rates of transpiration (Lösch et al., 1982; Mohammadian, 2005), and that therefore crypts are not adaptations for inherently low rates of transpiration. In environments where inherently low rates of transpiration have adaptive value, plants can achieve that with low stomatal density and smaller stomata, or by stomatal regulation that is sensitive to environmental and plant signals, which is a more flexible way to control transpiration than through a fixed resistance component. Ignoring the presence of crypts in measurements of gas exchange will overestimate the stomatal component of leaf resistance and underestimate the magnitude of stomatal responses to environmental conditions or physiological status. However, given the relatively minor influence of crypts on gas diffusion, it could be argued that for some applications an approximate value of crypt resistance is sufficient, rather than an elaborate model-derived value. One possible approach is to use the Parlange and Waggoner (1970) equation that is commonly used for stomata. In the case of a crypt with the standard dimensions (Table 2), this equation yields a crypt conductance of 0.9 mol m-2 s-1, which is

[61]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

somewhat lower than most models in Table 1. An even simpler approach that considers diffusion through an impermeable layer that is perforated with pores having the dimensions and density of the crypts in the leaves (Hassiotou et al., 2009b or Chapter 2), yields a crypt conductance of 1.5 mol m-2 s-1. Such deviations from the values obtained by detailed 3D modeling are probably similar to the inaccuracy associated with boundary layer conductance estimates. When greater accuracy is required, 3D modeling is recommended. A clearly counterintuitive result of this study was the small influence of crypt trichomes on transpiration. Neither their presence nor their spatial arrangement had a substantial influence on crypt resistance. A crypt with trichomes represents a fibrous porous material. Of fundamental importance for the diffusion through a porous material is its porosity. Obviously, diffusion of a gas can only occur in the void space of a porous material, and the restriction of the gas movement increases with decreasing void space. Furthermore, the presence of solid material leads to a deviation of the diffusion paths from straight lines, or in other words the molecules are forced to travel around the solid elements. This average increase of the diffusion path length is termed “tortuosity” ( ) (Shen and Chen, 2007): l e (10) l with le = path length in the porous material, l = shortest path length. The usual practice of considering the impeded diffusion in porous materials is to scale the diffusion coefficient of a gas with . The actual value of depends on porosity (P), which is defined as the fractional value of the air with respect to the entire crypt volume, and on the structure of the porous material. A collection of different approaches to = f(P) can be found in Shen and Chen (2007). One suitable approach for fibrous materials is provided by the numerical treatment of Tomadakis and Sotirchos (1993), which gives the tortuosity for random arrays of freely overlapping cylinders: 1 ln(P) (11) The scaling of the diffusion coefficient reads as: DP Deff 2 (12) with Deff = the effective diffusivity.

[62]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

If the approach of Tomadakis and Sotirchos (1993) for freely overlapping cylinders is used -5 for models CS50A and CS50H, then Deff = D × 0.94 = 2.28×10 . If this result is compared with the simulated effects of the trichomes on crypt conductance, then the values of 0.93 and 0.91 are obtained for the ratios of crypt conductances between CS50A and CS50, and

CS50H and CS50, respectively. Thus, the simulation results agree well with the technical literature on gas diffusion through fibrous materials. This topic will be pursued further with respect to the importance of the detailed fiber arrangement, since in CS50A trichomes were arranged differently to CS50H. A comprehensive collection of equations for Deff for different fiber arrangements is provided by Tomadakis and Sotirchos (1993). These data were obtained by various authors in mathematical-analytical, numerical and experimental studies. Random arrangements of cylinders and cylinders situated perpendicularly or parallel to the concentration gradient were considered. In accordance with Tomadakis and Sotirchos (1993), at porosities larger than 0.8, the detailed spatial arrangement of trichomes has almost no influence on conductance. Tomadakis and Sotirchos (1993) reported that densely packed fibrous fillings can substantially decrease the effective diffusivity, as this decreases in a non-linear way with decreasing porosity. Whether very dense crypt trichome fillings are common in nature is not known. Microscopic observations indicate that the trichome density of B. ilicifolia is probably representative of that of most Banksia species (Hassiotou, unpublished data). In the Proteaceae, Jordan et al. (2008) found no correlation between the presence of trichomes and aridity, suggesting that crypt trichomes are not primarily an adaptation related to plant water use. If dense trichome fillings occur in crypts, only a detailed study of the crypt porosity will provide a realistic estimation of the effective diffusivity of these structures. Any increase of the diffusion resistance for water vapour out of the leaf will also hinder

CO2 diffusion into the leaf. If structures like crypts have evolved as adaptations to optimize plant water use, it must be through a greater effect on water loss than on CO2 uptake. For simple structures like stomata, the relationship between resistances for water vapour and

CO2 is given by the relative values of their diffusivities. For a complex structure like a stomatal crypt, it may not be appropriate to directly calculate resistance to the inward flux for CO2 from the resistance to the outward flux of H2O, because the 3D diffusion space of

[63]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

the crypt may cause larger differences in the pathways of water vapour and CO2 compared with superficial stomata.

For example, for the model CS50, RH at the external stomatal pores varied between 56.9 and 61.1%, despite the fact that all internal pores had identical RH (internal pore = stomatal pore leading into the leaf interior). In this context, it is important to consider that CO2, compared with water vapour, experiences additional resistances in the mesophyll before reaching the chloroplasts. It is therefore to be expected that these additional resistances and the local concentration gradients inside the crypt caused by the 3D diffusion space would lead to substantial differences at the internal stomatal pores with respect to CO2 concentration. It is thus not possible at the moment to estimate the effects of crypts on CO2 diffusion. Mesophyll resistance is increasingly being recognized as a significant factor in photosynthesis (Flexas et al., 2008) and it has been shown to be relatively high in thick

Banksia leaves (Hassiotou et al., 2009a or Chapter 5). The diffusion of CO2 through the tortuous mesophyll air spaces is a component of mesophyll resistance. In leaves with crypts, this pathway is considerably shortened because stomata in crypt walls are much closer to the palisade mesophyll than stomata that would be positioned on the abaxial leaf surface. In this context, deep crypts with high porosity may actually facilitate diffusion of

CO2 to palisade cells. Evidence supportive of this was presented by Hassiotou et al. (2009b) (Chapter 2), who showed that in the genus Banksia encryption increases with increasing leaf lamina thickness. Whether or not the diffusion resistance within the intercellular airspaces affects photosynthesis is dependent on the leaf anatomy and porosity, the cell shape and packing and the pattern of stomatal openings (Morison et al., 2005). Facilitated diffusion by crypts may be particularly beneficial for thick leaves with densely packed mesophyll cells. Expansion of the crypt models to include substomatal cavities and mesophyll air spaces will give new insight into the potential benefits of these anatomical traits for the diffusion of CO2 and water vapour and their impact on photosynthesis and water-use efficiency. Stomatal crypts and crypt trichomes may have adaptive value unrelated to their effects on gas diffusion. Alternative roles of stomatal encryption are discussed in Hassiotou et al. (2009b) (Chapter 2), and include stomatal protection from mechanical damage (Haworth

[64]

CHAPTER 3: MODELLING OF GAS DIFFUSION THROUGH CRYPTS

and McElwain, 2008) and from harsh climatic conditions. Trichomes may provide a barrier to invasion by pathogens and blockage by dust or liquid water (Levin, 1973; Brewer et al., 1991; Paling et al., 2001). In future studies, it is worthwhile to examine the possible consequences of temperature gradients across the transpiration path in crypts. Our models assumed isothermal conditions. Banksia species are sclerophyllous and frequently exposed to intense insolation. Such conditions may cause temperature gradients in these thick and dense leaves, because heat conductivity of leaves is small (Vogel, 2009). Stomata placed at the bottom of crypts may be less exposed to the extreme temperatures of the leaf surface, particularly in dynamic conditions such as fluctuating insolation. Leaf to air temperature gradients strongly affect evaporation and are themselves affected by evaporative cooling. This feedback behavior may be influenced by crypts, which cause spatial separation of the surfaces that exchange heat and water vapour.

3.6 CONCLUSIONS

The present study shows that under conditions of limited soil moisture, when stomata will tend to close, crypts do not actually contribute much to water conservation. In contrast, when plants have higher stomatal conductances, the influence of crypts on water loss might be more significant. Such situations would be expected in arid-zone plants that respond to rainfall pulses, or perhaps during periods of high early-morning stomatal conductances after overnight rehydration. Trichomes had only minor effects on diffusion in the modeled crypts and thus do not appear to have a water-saving function. Future studies must focus on the effects of crypts with different characteristics (architecture, stomatal density and distribution, as well as trichome density) on diffusion of water vapour and CO2, and explicitly include mesophyll diffusion of CO2. Deep invaginations into the mesophyll may facilitate diffusion of CO2 to the sites of assimilation, especially in thick leaves with densely packed mesophyll cells. The influence of crypts on total leaf resistance depends on the detailed anatomy and on stomatal and boundary layer resistances. The question of the actual benefits of encrypted stomata is not fully resolved. Model simulations are useful for exploring diffusion processes within complex 3D leaf structures where approximations based on electrical resistivity networks are potentially flawed.

[65]

[66]

CHAPTER 4

PHOTOSYNTHESIS AT AN EXTREME END OF THE LEAF TRAIT SPECTRUM: HOW DOES IT RELATE TO HIGH LEAF DRY MASS PER AREA AND ASSOCIATED STRUCTURAL PARAMETERS?

4.1 ABSTRACT ...... 68

4.2 INTRODUCTION ...... 68

4.3 MATERIALS AND METHODS ...... 70 4.3.1 Plant material and growth conditions ...... 70 4.3.2 Leaf morphology and anatomy ...... 71 4.3.3 Chemical composition ...... 72 4.3.4 Microscopy ...... 73 4.3.5 Photosynthetic measurements ...... 74 4.3.6 Statistical analyses ...... 75

4.4 RESULTS ...... 77 4.4.1 LMA and its anatomical correlates ...... 77 4.4.2 Leaf structure, photosynthesis and mesophyll conductance ...... 82

4.5 DISCUSSION ...... 85 4.5.1 LMA and its anatomical correlates ...... 85 4.5.2 Leaf structure, photosynthesis and mesophyll conductance ...... 87

4.6 CONCLUSIONS ...... 90

[Hassiotou F, Renton M, Ludwig M, Evans JR, Veneklaas EJ (2009). Photosynthesis at an extreme end of the leaf trait spectrum: how does it relate to high leaf dry mass per area and associated structural parameters? Journal of Experimental Botany (submitted)]

[67]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

4.1 ABSTRACT

Leaf dry mass per area (LMA) is a composite parameter relating to a suite of structural traits that have the potential to influence photosynthesis. However, the extent to which each of these traits contributes to variation in LMA and photosynthetic rates is not well understood, especially at the high-LMA end of the spectrum. To improve our understanding of the physiological consequences of leaf structure at this end of the spectrum, interrelationships between leaf structural and photosynthetic characteristics were investigated in 49 Banksia species (Proteaceae) and subsets of this large group, displaying a wide range in LMA of 134-507 g m-2. Leaf thickness and density contributed similarly to variation in LMA. High LMA was associated with more structural tissue, lower mass-based chlorophyll and nitrogen concentrations, and therefore lower mass-based photosynthesis. In contrast, there was no association between LMA and area-based photosynthesis, despite an increase in mesophyll volume per area with increasing LMA, due to a reduction in photosynthetic rate per unit mesophyll volume at high LMA. Mesophyll conductance decreased with increasing LMA as mesophyll cell wall thickness increased. Adaptations such as stomatal crypts and bundle sheath extensions may partly counteract the effect that increasing LMA has on the distribution of CO2 and light across these thick leaves.

4.2 INTRODUCTION

In multi-species analyses, area-based photosynthetic rate correlates poorly with dry mass per unit leaf area (LMA), whereas mass-based photosynthesis shows a clear decline with increasing LMA (Reich et al., 1997; Wright et al., 2004). While the first observation may be explained by the greater proportion of structural (non-photosynthetically active) tissue per unit leaf dry mass, which is also expressed as lower mass-based nutrient concentrations

(Chapin, 1980), it is less clear how high-LMA leaves are able to fix CO2 at rates that are similar to those of low-LMA leaves that are usually found on fast-growing plants. LMA is a key structural trait that measures the investment of dry mass per unit of light- intercepting leaf area and is widely used as an indicator of plant ecological strategies (Westoby et al., 2002; Wright et al., 2004). High LMA can be due to a thick leaf or high leaf density, or both (Witkowski and Lamont, 1991). High-LMA leaves are often hard, and

[68]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

referred to as sclerophylls (Turner, 1994), although succulent species can also display high LMA values due to high leaf thickness (Poorter et al., 2009). In the present study, LMA and its relationship with photosynthesis is discussed in the context of hard, thick and dense leaves of a wide range of LMA, with robust construction, which confers long lifespans. Despite the general anatomical organisation of high-LMA leaves, which are thick and/or dense, fibrous and often hairy, at least on the abaxial surface (Turner, 1994; Read et al., 2000; Mast and Givnish, 2002), the structural traits at the tissue and cell level that contribute to high LMA are particularly diverse and include bundle fibre caps, lignified bundle sheaths, vascular bundle extensions, lignified leaf margins, very thick cuticles, lignified hypodermal structures associated with the adaxial and/or abaxial surfaces, sclereids within the mesophyll, sclereids associated with vein endings and thick cell walls (Dillon, 2002; Terashima et al., 2006). It must be noted that some of these characters are not restricted to high-LMA leaves, and not all high-LMA species possess all of these characters (Read et al., 2000). In other words, different combinations of the above leaf traits can result in high LMA (Read et al., 2000; Read and Sanson, 2003), and this explains the great variation in this trait that is usually found among hard leaves (Read et al., 2000), even within the same genus (Hassiotou et al., 2009a,b or Chapter 5 and 2 respectively). While it is clear that variation in leaf thickness and density is due to the number of cell layers (photosynthetic or not) and the relative amount of cell types, respectively, the relative importance of these structural traits in determining thickness, density and LMA is not well understood.

High LMA has been associated with low conductance to CO2 diffusion from the substomatal cavity to the chloroplasts (mesophyll conductance, gm), which can restrict the rate of CO2 assimilation (Loreto et al., 1992; Evans et al., 1994; Parkhurst, 1994; Evans and von Caemmerer, 1996; Evans and Loreto, 2000; Terashima et al., 2006; Hassiotou et al., 2009a or Chapter 5). Moreover, surface properties of high-LMA leaves, including wax layers, epidermal cell shape, cuticular thickening, trichomes and stomatal crypts, as well as specific scleromorphic structures, such as sclereids, can alter leaf optical properties (Myers et al., 1994, Baldini et al., 1997) and thus influence gas exchange. High-LMA leaves have low concentrations of key nutrients such as nitrogen, but whether this is simply due to “dilution” by the presence of more structural tissue, or also applies to the

[69]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

photosynthetically active mesophyll, is not known. In fact, it is unclear whether the photosynthetically active mesophyll cells of high-LMA leaves differ from those in lower- LMA leaves, and if so, whether this is because of the conditions in which they operate

(CO2, light) or because they are structurally and/or physiologically different. To advance our understanding of the physiological consequences of leaf structure expressed through LMA, leaf structural properties and gas exchange were investigated in 49 Banksia species (Proteaceae) and subsets of this large group. This genus, being predominantly endemic to Australia, was chosen on the basis of the great leaf structural diversity that it displays (LMA=134-507 g m-2, Hassiotou et al., 2009a or Chapter 5). The following questions were asked: 1. How much of the variation in LMA is explained by leaf thickness and density, and which are the most important anatomical correlates of these two traits? 2. Is photosynthesis related to leaf structural traits and how do these relate to leaf chlorophyll and nitrogen concentration? 3. What are the morphological/anatomical correlates of mesophyll conductance?

4.3 MATERIALS AND METHODS

4.3.1 Plant material and growth conditions Three- to five-year old plants of 49 broad-leaved (as opposed to needle-leaved) Banksia species were used (Appendix 2). The plants, except for B. integrifolia L.f., B. paludosa R.Br. and B. serrata L.f., were grown from seed in 10-L pots containing a mixture of river sand and potting mix in Perth (Australia), outdoors (with an average annual temperature and average daily solar exposure of 19°C and 20 MJ m-2, respectively; Australian Government, Bureau of Meteorology) until about three weeks before the measurements, when they were transferred to a controlled-temperature greenhouse (23°C day/18°C night). Mature plants of B. integrifolia, B. paludosa and B. serrata were purchased from a nursery in Canberra (Australia). Upon purchase, the plants were re-potted into 10-L pots containing a mixture of grey sand and potting mix and grown for two months prior to measurements in a greenhouse in Canberra (25°C day/20°C night). Measurements were done on the 49

[70]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

species, unless otherwise stated (Appendix 2), using in all cases the youngest fully expanded leaves.

4.3.2 Leaf morphology and anatomy Three leaves per species, from different plants, were sampled early in the morning. Projected area (using a leaf area meter, LI-300A, Li-Cor, Lincoln, NE, USA) and dry mass of the leaf lamina were measured after removing the midrib and petiole where present. Leaf dry mass per area (LMA) was calculated and leaf density (DL) was computed from:

LMA TL DL (1) where TL is leaf lamina thickness. The latter was measured with digital callipers at 5-10 different positions on each leaf. Based on the relationship between LMA and TL and DL, subsets of species that covered the range of LMA of the 49 species were chosen for further analyses (Appendix 2).

In three leaves per species, for 14 species (Appendix 2), leaf porosity (PL) was measured by determining leaf buoyancy before and after vacuum infiltration of the leaf air spaces with water, using the method of Raskin (1983) and the equations modified by Thomson et al. (1990). In short, fresh leaf volume (VL), leaf gas volume (VG) and PL were estimated as:

WL, air WL, water VL (2)

WL,after WL, before VG (3)

VG 100 PL (4) VL where WL,air and WL,water are the weights of the leaf in air and water, respectively; WL,after and WL,before are the weights of the submerged leaf holder with the leaf after and before vacuum infiltration, respectively; and ρ is the density of water (1 mg mm-3 at 25°C). The measurement of leaf volume allowed the computation of a second estimate of leaf density which is independent of LMA (DL,V), and which was used to calculate the density of * the leaf tissues excluding the gas volumes (leaf density corrected for porosity, DL ):

WL DL, V (5a) VL

[71]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

* WL DL (5b) (VL VG ) where WL is leaf dry weight.

4.3.3 Chemical composition

Nitrogen concentration (Nmass) was measured in the leaf blade (excluding the midrib) in 17 species (Appendix 2) using gas chromatography. Analyses of 14 species were done at the Western Australian Biogeochemistry Centre (University of Western Australia, Perth). Samples from the other three species (B. integrifolia, B. paludosa and B. serrata) were analysed at the Research School of Biological Sciences (Australian National University, Canberra). Finely ground leaf dry matter was used from three leaves per species from three different plants, except for B. attenuata and B. ilicifolia where one leaf was analysed. Narea was subsequently calculated (Narea=Nmass LMA).

The fraction of nitrogen allocated to Rubisco (RN/N) was estimated (Appendix 2) as: M N V R R R c k n N cat R (6) N N area where Vc is the rate of carboxylation (see next sections) MR is the molecular mass of -1 Rubisco (0.55 g Rubisco (μmol Rubisco) ), kcat is the catalytic turnover number at 25°C -1 -1 (3.5 mol CO2 (mol Rubisco sites) s ; von Caemmerer et al., 1994), nR is the number of -1 catalytic sites per mole of Rubisco (8 mol Rubisco sites (mol Rubisco) ), NR is the nitrogen concentration of Rubisco (11.4 mmol N (g Rubisco)-1) and N is the nitrogen content per unit leaf area (mmol N m-2). This calculation provides a minimum estimate as it assumes full Rubisco activation (Harrison et al., 2009).

Total chlorophyll content (Chlarea) was determined in 12 species (Appendix 2) using three leaves per species from three different plants, sampled early in the morning and analysed immediately. Leaf segments were excised and their areas were measured with a leaf area meter (LI-300A, Li-Cor, Lincoln, NE, USA). Within 5 min of sampling, the leaf segments were finely ground with liquid nitrogen using a cold mortar and pestle and were subsequently extracted with 100% cold methanol. The extract was clarified by centrifugation at 1,600 g (Beckman, AvantiTM J-25 Centrifuge, USA) for 20 min at 4°C. To

[72]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

avoid condensation on the cuvette while taking measurements, the samples were stored in the dark at room temperature for 5 min. Absorbance was measured with a spectrophotometer (Graphicord UV-240, Shimadzu, Kyoto, Japan) at three wavelengths (710, 665.2 and 652.4 nm) and the equations of Wellburn (1994) were used for the estimation of Chla and Chlb, and from these of total chlorophyll concentration per unit leaf area (Chlarea) and dry mass (Chlmass) were calculated. The fraction of nitrogen allocated to thylakoids (TN/N), including pigment-protein complexes, the components of electron transport and ATPase, was estimated based on Chlarea and Narea, assuming 50 mol of nitrogen per mol of chlorophyll (Evans, 1989).

4.3.4 Microscopy Cryo-scanning electron microscopy (CSEM) and fluorescence microscopy (Zeiss Axioskop2, Zeiss Axiocam with AxioVision software, Zeiss Oberkocken, Germany) were used to obtain transverse views of leaf laminas originating halfway from the leaf tip in samples from two leaves per species, from different plants. Analyses were done in Image J

(Abramoff et al., 2004). Leaf thickness (TL), mesophyll thickness (TM) and the thickness of the abaxial epidermis plus hypodermis (TAB) were measured from fluorescence micrographs taken at the same magnification in a subset of 10 species (Appendix 2), and the mean of at least six measurements was used. These measurements were confirmed with CSEM. The combined thickness of the adaxial epidermis and hypodermis (and pseudohypodermis where present) (TAD) was estimated by subtracting the average TM and

TAB from TL. Leaf lamina thickness and mesophyll thickness do not take into account the presence of stomatal crypts. Thus, micrographs of transverse leaf views obtained with fluorescence microscopy at the same magnification, were used to calculate leaf volume per area (LVA) and mesophyll volume per area (MVA) which exclude the volumes taken by crypt voids.

The width of an areole (Wa) and the cross-sectional area of non-photosynthetic tissue per areole (A1) (including the adaxial and abaxial epidermal and hypodermal tissues as well as the vascular bundles and their sclerified extensions) and of mesophyll tissue per areole (A2) (including photosynthetic cells and intercellular airspaces) were measured. A mean of at

[73]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

least four measurements for each of the above parameters was obtained. LVA and MVA were calculated as: A A LVA 1 2 (7a) Wa A MVA 2 (7b) Wa

Mesophyll volume per unit leaf volume (VM, %) was calculated as:

A2 MVA VM 100 100 (8) A1 A2 LVA Usually one layer, but sometimes locally two layers, of adaxial palisade mesophyll is present in Banksia leaves. The length of adaxial palisade cells (LP) was measured as the mean of at least seven measurements in transverse views of five species (Appendix 2) obtained with CSEM at the same magnification. Wall thickness of palisade and spongy mesophyll cells was measured in six species

(Appendix 2) and mean mesophyll cell-wall thickness was calculated (Tw). Leaves of these species were frozen in liquid nitrogen and high-magnification images of the cell walls were obtained with CSEM following McCully et al. (2004). Segments of the leaf lamina from the middle part of each leaf were excised under liquid nitrogen, mounted on stubs with low-temperature Tissue-Tek (O.C.T. Compound cryostat specimen matrix, ProSciTech), planed flat in the paradermal and transverse direction using a diamond knife in a cryo- microtome (Cryo-system Oxford CT1500, Oxford Instruments Ltd, Old Station Way, Eynsham, Oxford OX8 1TL, UK) at -100°C, etched in the column of the CSEM (Cambridge S360, Cambridge Instruments Ltd, Viking Way, Bar Hill, Cambridge, CB3 8EL, UK) for 1-2 min at -90°C to reveal cell outlines, sputter-coated with gold, and examined at 15 kV. Images were captured using Microsoft Photodraw and analysed in Image J (Abramoff et al. 2004).

4.3.5 Photosynthetic measurements Gas exchange measurements were carried out for 18 species (Appendix 2) using three leaves per species from different plants, at a photosynthetic photon flux density (PPFD) of -2 -1 -1 1500 μmol quanta m s and at 380 μmol CO2 mol air, with a LI-6400 open gas exchange

[74]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

system (LI-6400-40, Li-Cor, Lincoln, NE, USA). Leaves were kept in the gas exchange -2 -1 chamber at high irradiance (1500 μmol quanta m s ) and low CO2 concentration (100 -1 μmol CO2 mol air) for at least 10 min before the commencement of the measurements, ensuring stomata were fully open and steady state was reached. At ambient CO2 concentration, 4-10 measurements of gas exchange, at least seven seconds apart, were recorded for each leaf and the mean value of net CO2-assimilation rate was calculated and -2 -1 expressed on a leaf area basis (Aarea, μmol m s ), on a leaf mass basis (Amass=Aarea/LMA, -1 -1 -1 -1 nmol g s ), per unit Chl (AChl=Aarea/Chl, μmol g s ) per unit mesophyll volume -3 -1 (Ames=Aarea/MVA, μmol m s ) and per unit nitrogen (photosynthetic nitrogen-use -1 -1 efficiency, PNUE=Amass/Nmass, nmol g s ). Combined gas exchange and chlorophyll fluorescence measurements (Harley et al., 1992) were conducted and mesophyll conductance (gm) was calculated in seven species (Appendix 2) as described in Hassiotou et al. (2009a or Chapter 5).

4.3.6 Statistical analyses Three complementary approaches were used to identify the extent to which the two determinants of the key structural trait LMA (TL and DL) contribute to its variation across 49 species (Appendix 2). The first approach was the log-log scaling slope analysis that has been used previously (e.g. Poorter and van der Werf, 1998; Poorter et al., 2009) to estimate the relative contribution of possible explanatory variables (such as TL and DL) to variation in a particular variable of interest (such as LMA). The basis of this method is the relationship LMA=TL×DL and thus log(LMA)=log(TL)+log(DL). If the log of an explanatory variable (in this case either TL or DL) is regressed linearly against the log of the variable of interest (in this case LMA), then a slope coefficient value of close to 1 indicates that the particular explanatory variable used is largely responsible for variation in the variable of interest, whereas a value close to 0 indicates that the particular explanatory variable used is not responsible for much of the observed variation in the variable of interest (Poorter and van der Werf, 1998; Poorter et al., 2009). However, this method has a potential limitation: while a large contribution from an explanatory variable always results in a slope close to 1, cases are possible in which a variable with a slope close to 1 does not actually contribute very much to the variability in the variable of interest. This method was applied in our

[75]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

study to enable comparison with previous data, but due to its potential limitation, the contribution of TL and DL to variation in LMA was also evaluated using an alternative method that entailed calculating the correlation coefficient (r) for the linear regression described above (log(TL) or log(DL) against log(LMA)). In this method, a resulting r value close to 1 also generally indicates a high contribution and a value close to 0 a small contribution. However, there are also cases for this method where a r value close to 1 does not indicate a high contribution, namely cases when two explanatory variables were highly correlated. To identify situations where the two explanatory variables are highly correlated with each other, but one is more variable than the other, and thus accounts for more variability, a third approach was used whereby the variances of the log-transformed variables were computed and their percentage contribution to the sum of their variances was calculated. For example, the percentage variance contribution for DL was obtained from: Variance(log D ) L 100 (9) Variance(log DL ) Variance(logTL ) Using these three complementary methods enabled us to obtain a more complete picture of the relative contributions of each of TL and DL to LMA. In most cases, the results of the three methods would be expected to support each other (slope close to 1, r close to 1 and percentage variance contribution close to 100%), but in particular cases discrepancies between these methods would highlight issues that needed further investigation (such as high correlation between explanatory variables). These three approaches were also used to examine the main determinants of the variation in * DL (DL and PL) in 14 species (Appendix 2), by modelling DL as:

P D 1 L D * (10) L 100 L and in TL (TM, TAB and TAD) in 10 species (Appendix 2). Since the relationship between TL and its components is not multiplicative (TL = TM + TAB +TAD), the above methods were applied to the original values of the different thicknesses, without log-transformation.

To understand the reasons behind the variability in Aarea, two models were considered. The first model aimed at assessing if variation in Aarea was due mostly to differences in the

[76]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

-2 - amount of mesophyll tissue or in the mesophyll‟s photosynthetic activity. Aarea (μmol m s 1) was thus modelled as:

Aarea VM Ames LVA (11) 3 -3 where VM is the mesophyll volume faction (m m ), Ames is CO2-assimilation rate per mesophyll (μmol m-3 s-1) and LVA is leaf volume per area (m3 m-2). The second model for

Aarea aimed at assessing if variation in Aarea was related to differences in the amount of -2 -1 chlorophyll or in the photosynthetic rate per unit chlorophyll. Aarea (μmol m s ) was thus modelled as:

Aarea Chlmes AChl MVA (12) -3 where Chlmes is the chlorophyll concentration per mesophyll volume (g m ), AChl is CO2- assimilation rate per chlorophyll (μmol g-1 s-1) and MVA is the mesophyll volume per unit leaf area (m3 m-2). These two models were converted from multiplicative to additive linear models by taking the log of the various variables.

To examine whether TAD was significantly different to TAB, a paired T-test was carried out (Microsoft Excel® 2007, Microsoft Corporation).

4.4 RESULTS

4.4.1 LMA and its anatomical correlates Among the 49 broad-leaved Banksia species examined, leaf dry mass per area (LMA) varied 4-fold (134-507 g m-2), which was associated with a 4-fold variation in leaf lamina -3 thickness (TL; 193-700 μm) and a 3-fold variation in leaf density (DL; 0.41-1.17 mg mm ).

The variance contribution method showed that TL explained 54% and DL 46% of the variability in LMA (P<0.001) which was consistent with the log-log scaling slope and the correlation coefficient (r) analyses (Table 1). Some species had high LMA due to their high

DL and others due to their high TL, whilst in some high LMA was due to both (Fig. 1). For -2 example, both B. coccinea and B. quercifolia had a LMA of 215 g m , but a TL of 0.50 and -3 0.38 mm, and a DL of 0.4 and 0.6 mg mm , respectively.

Thicker leaves, with high volume per area (LVA), had thicker mesophyll (TM), adaxial

(TAD) and abaxial (TAB) epidermis and hypodermis, greater mesophyll volume per area

(MVA) and longer adaxial palisade cells (LP) (Fig. 2, Table 1). All the three statistical

[77]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

Table 1. Analyses of the relative contribution of explanatory variables to measured structural and physiological variables. Log-log scaling slope analysis (Slope), correlation coefficient analysis (r), and analysis of the percentage contribution of the variance of each explanatory variable to the total summed variance of the explanatory variables (Contribution to Variance Sum, %). N: species number. ***: P<0.001, **: P<0.01, *: P<0.05, ns: not significant; stars indicate the significance of the slope parameter i.e. whether the explanatory variable contributes significantly to the response variable; the indication of significance for the contribution to variance sum (%) is the result of a two sample F-Test for equality of variance and thus indicates the significance of the difference in variance between the explanatory variables (where there are three explanatory variables, the order of significance indicates the following pairs: first with second variable, first with third variable, second with third variable).

Contribution to N Slope r Variance Sum (%) *** DL to LMA 49 0.43 46 0.54 *** ns TL to LMA 49 0.57 54 0.66

* *** DL to DL 14 0.76 87 ** 0.97 ** (1-PL) to DL 14 0.24 13 0.79

*** *** TM to TL 10 0.74 86 0.96 * ** TAB to TL 10 0.04 0.5 0.67 * *** TAD to TL 10 0.22 13 0.71

Ames to Aarea 10 1.27 ns 53 ns 0.61 * LVA to Aarea 10 -0.38 ns 34 -0.23

VM to Aarea 10 0.11 ns 13 ns 0.11

AChl to Aarea 9 1.00 ns 48 ns 0.42 Chlmes to Aarea 9 0.40 ns 34 ns 0.20

MVA to Aarea 9 -0.19 ns 17 ns -0.13

* DL: leaf density; LMA: leaf dry mass per area; TL: leaf lamina thickness; DL : leaf density corrected for porosity; PL: percentage leaf porosity; TM: mesophyll thickness; TAB: thickness of abaxial epidermis and hypodermis combined; TAD: thickness of adaxial epidermis and hypodermis combined; Ames: net CO2 assimilation rate per unit mesophyll; Aarea: net

CO2 assimilation rate per unit leaf area; VM: mesophyll volume per unit leaf volume (%); AChl: net CO2 assimilation rate per chlorophyll; Chlmes: chlorophyll concentration per mesophyll volume.

analyses used to examine the contributions of the thickness of the different leaf layers to TL concurred in that TM explained most of the variability in TL compared with TAD and TAB

(Table 1). [78]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

2.9 0.2

2.8 0.1

)

2.7 0.0 -3

m) 2.6 -0.1

(

L

T 2.5 -0.2 mm (mg

L

D

log 2.4 -0.3

log 2.3 -0.4

2.2 -0.5 2.0 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 -2 logLMA (g m )

Fig. 1. Relationship between log10-transformed leaf lamina thickness (TL, circles) or leaf density

(DL, squares) and leaf dry mass per area (LMA) in 49 Banksia species. Grey symbols show the 7 species used for the measurement of mesophyll conductance, while crossed symbols represent the 10 species examined by microscopy (see Appendix 2 for species names). (For the relationship 2 between TL and LMA, slope is 0.57 and r is 0.43 (P<0.001); for the relationship between DL and LMA, slope is 0.43 and r2 is 0.30 (P<0.001).)

Mesophyll tissue represented on average 74% of leaf lamina thickness (based on TM/TL) and 58% of leaf volume (based on MVA/LVA). High TL, LVA, TM and MVA were associated with high LMA (P≤0.01) (data not shown). TAD varied 3-fold among the examined species and was significantly higher than TAB (P<0.001), which varied 2-fold. Both TAD and TAB increased with increasing LMA, although this was significant (P=0.008) only for TAB (data not shown). * Leaf density corrected for porosity (DL ) tended to increase with increasing thickness of the different leaf layers, but none of these relationships was significant. Leaf porosity (PL) tended to decrease with increasing thickness of the different leaf layers, although only its relationship with TL was significant (P=0.016). PL varied 3.5-fold among the species, ranging from 6% (in B. elderiana; TL= 0.63 μm) to 22% (in B. littoralis; TL= 0.22 μm)

[79]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

350 Fig. 2. (A) Mesophyll volume per MVA= 0.52LVA+22 300 A r2 = 0.65, P = 0.005 leaf area (MVA) and thickness of

m) 250

( adaxial and abaxial epidermis and

200 hypodermis (TAD+AB) against leaf

AD+AB T 150 volume per leaf area (LVA) in 10

or 100 Banksia species. (B) Palisade cell

MVA 50 TAD+AB = 0.34LVA-5.6 length (LP) against LVA in five B. r2 = 0.47, P = 0.030 0 species (see Appendix 2 for species 0 200 250 300 350 400 450 500 550 names). LVA ( m) 160 140 B

120

100

m) 80

(

P

L 60

40

20 LP = 0.64LVA-191 r2 = 0.93, P = 0.009 0 0 300 350 400 450 500 550 LVA ( m)

2 (Fig. 3A). DL was positively correlated with leaf dry matter content (r =0.28, P=0.023), * * which was similar but not quite significant for DL . DL and PL were positively (P<0.001) and negatively (P=0.0016) correlated with DL, and contributed 87% (P<0.001) and 13%

(P<0.001), respectively, to the variability in DL as indicated through the variance contribution method (Table 1). This was consistent with the log-log scaling slope (Table 1). * High-LMA leaves were less porous (Fig. 3A) and had higher DL (data not shown).

Mesophyll volume was approximately 60% of leaf volume (VM), irrespective of LMA (Fig.

3B). High-LMA, denser leaves had thicker mesophyll cell walls (Tw) (Fig. 4A).

[80]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

25 A r2 = 0.30 Fig. 3. (A) Relationship between leaf porosity P = 0.043 20 (PL) and leaf dry mass per area (LMA) in 14 Banksia species. Open circles show the 10 15

(%)

L species examined by microscopy. (B) P 10 Mesophyll volume per unit leaf volume (VM) 5 against LMA in 10 species (see Appendix 2 for 0 species names). r2 = 0.00 100 B LMA (g m-2) P = 0.990 80

60 (%)

M

V 40 20 0 0 100 200 300 400 500 600 LMA (g m-2)

600 A 500 Fig. 4. Relationship of leaf dry mass per

) area (LMA) (A) and mesophyll conductance

-2 400

(gm) (B) with mesophyll cell wall thickness

(g m 300 (Tw) in 6 Banksia species (see Appendix 2 for

LMA 200 LMA = 1379Tw - 161 species names). 100 r2 = 0.84 P = 0.01 0 200 B

)

-1

s 150

-2

100

(mmol m (mmol

m g 50 r2 = 0.96 P < 0.001 0 0.00 0.20 0.25 0.30 0.35 0.40 0.45 0.50 Tw ( m)

[81]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

4.4.2 Leaf structure, photosynthesis and mesophyll conductance

CO2-assimilation rate per unit leaf area (Aarea) and leaf conductance (gL, which in the case of species with crypts comprises stomatal and crypt conductance) correlated poorly with LMA (Fig. 5A, 5B) in 18 species. In the subset of seven species in which mesophyll conductance (gm) was measured, Aarea and gL tended to decrease with increasing LMA, while gm strongly decreased with LMA (Fig. 5C). CO2-assimilation rate per unit leaf mass

(Amass) showed a strong negative correlation with LMA (Fig. 6A), while CO2-assimilation

35 A ) 30 Fig. 5. Relationships of net CO2

-1

s

-2 25 assimilation rate per unit leaf area (Aarea)

m

2 20 (A), leaf conductance (gL: comprising 15 stomatal conductance and crypt

mol CO

( 10 conductance in species with stomatal

area 2 A 5 r = 0.07 P = 0.28 crypts) (B), and mesophyll conductance 0 (gm) (C) with leaf dry mass per area (LMA) B LMA (g m-2)

)

-1 in 18 Banksia species. Open circles: 7

s 300

-2

m species in which gm was measured; filled

2 200 circles: all other species (see Appendix 2 for species names). (C) is redrawn with

(mmol CO 100

L

g r2 = 0.06 permission from Hassiotou et al. (Influence P = 0.34 0 of leaf dry mass per area, CO2 and C LMA (g m-2) irradiance on mesophyll conductance in 200

) -1 sclerophylls, Journal of Experimental

s -2 150 Botany, 2009, volume 60, pages 2303- 100 2314, by permission of Oxford University

(mmol m

m g 50 Press). r2 = 0.7 P = 0.02 0 0 100 200 300 400 500 600 -2 LMA (g m )

[82]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

rate per unit nitrogen (photosynthetic nitrogen-use efficiency, PNUE) tended to decrease with increasing LMA, but this was not significant (Fig. 6C). No correlation was found between CO2-assimilation rate per unit chlorophyll (AChl) and LMA (data not shown), while

CO2-assimilation rate per unit mesophyll (Ames) decreased with LMA (Fig. 7). The decrease in gm with increasing LMA that was observed in seven Banksia species was better 2 2 correlated with DL (r =0.76, P=0.01) than with TL (r = 0.34, P=0.17). Lower gm was strongly correlated with thicker mesophyll cell walls (Fig. 4B).

200 A

) Fig. 6. Relationships of net CO2

-1

s 150 -1 assimilation rate per unit leaf mass (Amass)

g

2 (A), nitrogen content (Nmass) (B), and 100 photosynthetic nitrogen use efficiency

(nmol CO 50 (PNUE) (C) with leaf dry mass per area

mass 2 A r = 0.69 (LMA) in 18 Banksia species. Open P < 0.0001 0 circles: 7 species in which gm was B LMA (g m-2) 1.25 measured; filled circles: all other species

1.00 (see Appendix 2 for species names).

(%) 0.75

mass N 0.50

0.25 r2 = 0.47 P = 0.002 0.00

C LMA (g m-2)

)

-1 300

s -1 200

(nmol g

100 PNUE r2 = 0.06 P = 0.33 0 0 100 200 300 400 500 600 LMA (g m-2)

[83]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

140 Fig. 7. Net CO2 assimilation rate 120

) per unit mesophyll volume (Ames) -1 100

s

-3 against leaf dry mass per area 80 (LMA) in 10 Banksia species (see 60 Appendix 2 for species names).

(nmol m

mes 40

A 20 r2 = 0.42 P = 0.043 0 0 100 200 300 400 500 600

LMA (g m-2)

0.3 A Fig. 8. Fraction of nitrogen

allocated to Rubisco (NR/N) (A) and

0.2 to thylakoids (Nthylakoid/N) (B) against

N

/ leaf dry mass per area (LMA) in 6

R N and 12 Banksia species, respectively 0.1 (see Appendix 2 for species names). r2 = 0.59 P = 0.075 0.0 B LMA (g m-2) 0.5 0.4

N

/ 0.3

thylakoid

N 0.2

0.1 r2 = 0.16 P = 0.2 0.0 0 100 200 300 400 500 600 LMA (g m-2)

[84]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

Nitrogen concentration (Nmass) varied 4-fold and decreased with increasing LMA (Fig. 6B).

Nmass was rather low in all species (0.27-1.1%). As expected, Amass was positively 2 associated with Nmass (r =0.49, P=0.0018). The fractions of nitrogen allocated to Rubisco and thylakoids were independent of LMA (Fig. 8). Chlorophyll concentration per leaf area tended to be higher in high-LMA species with thicker mesophyll (r2=0.22, P=0.08).

The two models of Aarea as a function of CO2-assimilation rate per unit mesophyll (Ames),

LVA and VM or as a function of CO2-assimilation rate per chlorophyll (AChl), chlorophyll concentration per mesophyll volume (Chlmes) and mesophyll volume per unit leaf area

(MVA), showed that Ames and AChl were the most important explanatory variables in these two models, respectively (variance contribution 53% and 48%, r=0.61 and 0.42, slope=1.27 and 1, respectively). Nevertheless, the explanatory value of Ames was almost significant (P=0.07) and that of AChl was not significant (P=0.26) (Table 1).

4.5 DISCUSSION

Many comparative studies examining the variability in leaf structure and its effect on leaf physiology consider diverse species from different genera differing in LMA (Poorter and Evans, 1998; Wright et al., 2004; Flexas et al., 2008; Harrison et al., 2009; Poorter et al., 2009). In the present study, phylogenetic variation was minimised by focusing on one genus (Banksia) with a great leaf structural diversity that allowed establishing quantitative relationships between LMA of a wide range and photosynthetic characteristics at the high end of the LMA spectrum.

4.5.1 LMA and its anatomical correlates -2 Among the 49 Banksia species examined, LMA (134-507 g m ), TL (193-700 μm) and DL (0.41-1.17 mg mm-3) varied 4-, 4- and 3-fold, respectively, which is indicative of the broad range of leaf structure that is represented in this genus. Niinemets et al. (2009) found a 4.7- -2 fold variation in LMA (66-313 g m ) and a 2.5-fold variation in TL (274-594 μm) and DL (0.29-0.56 mg mm-3) across 35 Australian sclerophyllous species from 20 genera. Poorter et al. (2009) reported a 4-fold variation in leaf volume per area (equivalent to TL) (100-700 -3 μm) and a 7-fold variation in DL (0.1-0.6 mg mm ) in a dataset containing woody and herbaceous species from three functional groups. In their dataset, most of the variation in

[85]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

LMA within functional groups is attributed to DL, while differences in LMA between sclerophylls and mesophytes are usually due to variation in leaf thickness (Poorter et al.,

2009). Log-log scaling-slope analysis of the contribution of DL and TL to LMA in species from three functional groups showed that 80% and 20% of the differences in LMA were explained by these two variables, respectively (Poorter et al., 2009). The larger role of DL in the dataset of Poorter et al. (2009) is likely to be due to the fact that the range in DL was much greater in their dataset than that in our 49 Banksia species (7-fold and 3-fold respectively), whereas the ranges in TL were very similar (approximately 4-fold in both datasets). Moreover, the relationship between DL and LMA is fairly similar for different functional groups, whereas the relationship between TL and LMA differs between functional groups, such that TL becomes a poorer predictor of LMA in the combined dataset. It is also noteworthy that the ranges of DL, TL and LMA of the Banksia species all extend to values far beyond the dataset of Poorter et al. (2009).

The considerable variability in DL and TL indicates that even within the same genus there are various ways of achieving high LMA, with potential ecological significance. Niinemets et al. (2009) found that density tended to increase with decreasing water availability, and thickness increased with decreasing soil fertility in a comparison of Australian species from sites that differed in water and nutrient availability. A number of previous studies have also reported increases in leaf thickness with decreasing soil fertility as well as with other factors, such as decreasing rainfall and humidity and increasing irradiance (Beadle, 1966; Nobel et al., 1975; Chabot and Chabot, 1977; Givnish, 1978; Sobrado and Medina, 1980).

High irradiance can result in increased TL through the development of thicker epidermal tissues that confer photoprotection (Witkowski and Lamont ,1991; Jordan et al., 2005).

High irradiance can also lead to high DL (Chabot and Chabot, 1977) through addition of dense, sclerified tissues that increase the uniformity of illumination within thick leaves (Poulson and Vogelmann, 1990; Karabourniotis, 1998), although these tissues may also play other roles, such as providing support and enhancing the rigidity of long-lived high- LMA leaves. * DL was an important determinant of LMA in Banksia leaves through its effect on DL. * Increases in DL can result from increases in the proportion of non-photosynthetic supporting tissue, especially sclerified cells, and/or a general tendency for cells to have

[86]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

more structural mass. The latter can be due to thicker cell walls, but also to larger surface to volume ratios of smaller cells. In Banksia, even mesophyll cells of high-LMA species had thicker cell walls compared with low-LMA species: a 4-fold range in LMA was accompanied by a 2-fold range in Tw. Interestingly, mesophyll volume per unit leaf volume

(VM) did not correlate with LMA or its anatomical correlates, indicating that high-LMA leaves are not photosynthetically „disadvantaged‟ by a lower VM. Leaf volume per area was lower than leaf thickness. This is a reflection of the presence of stomatal crypts, which may facilitate diffusion by locally reducing the distance between the abaxial leaf surface and the adaxial palisade cells (Hassiotou et al., 2009b or Chapter 2). High-LMA leaves, with thicker mesophyll and more elongated adaxial palisade cells were less porous. Leaf porosity was negatively correlated with crypt volume and depth (Hassiotou, unpublished data), which suggests that increased leaf porosity and crypt size may have a similar function in facilitating CO2 diffusion in thick leaves. Leaf thickening can occur through (i) addition of mesophyll cell layers, (ii) elongation of mesophyll cells and/or (iii) addition of non-photosynthetic supporting tissue in the epidermal and hypodermal layers. In Banksia, the most important contributor to TL, and through it to LMA, is TM. Microscopic observations suggest that elongation of adaxial palisade cells was a major contributor to mesophyll thickening (Fig. 2B). It was also observed that abaxial palisade-like cells are present in Banksia leaves alongside the crypts, a pattern that appears to be more common in high-LMA leaves, but more research is needed to elucidate their contribution to leaf thickening.

Given that Aarea and VM (%) did not correlate with LMA, does this indicate that the increase in mesophyll volume per area with increasing LMA is associated with a roughly proportional decrease in photosynthesis per unit mesophyll? Are there limits to how much photosynthetic tissue per area a leaf can have before it becomes inefficient in some way?

4.5.2 Leaf structure, photosynthesis and mesophyll conductance LMA has often been the trait of interest when looking at relationships between leaf structure and photosynthesis, but since it is a product of two anatomical traits that often vary independently (TL and DL) and that may influence photosynthesis differently, a great variability is found in the relationship between LMA and photosynthesis (Niinemets and

[87]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

Sack, 2006). In the present study, Amass showed a strong negative correlation with LMA in 18 Banksia species, which can be at least partly attributed to the fact that high-LMA species have more structural material per unit dry weight, as indicated through their higher dry matter content. The lower Amass in combination with the lower Nmass at high LMA can explain the weak relationship obtained between photosynthetic nitrogen-use efficiency

(PNUE=Amass/Nmass) and LMA. This is in contrast with previous studies, which reported a strong decrease of PNUE with increasing LMA (Poorter and Evans, 1998; Hikosaka, 2004), attributing this relationship to the lower Narea and higher Rubisco specific activity of low- LMA leaves, and the increased allocation of nitrogen to non-photosynthetic (structural) relative to photosynthetic (e.g. Rubisco) components in high-LMA leaves. However, Harrison et al. (2009), in a comprehensive study of 25 species covering a 10-fold range in LMA, showed that the fraction of nitrogen allocated to cell walls is independent of LMA. Moreover, they found that the relationship between the fraction of nitrogen allocated to Rubisco and LMA is curvilinear: at low LMA, the fraction of nitrogen associated with Rubisco decreases with LMA – explaining the negative correlation between PNUE and LMA that has been found in previous studies – down to a stable level above a LMA of 130 g m-2. All the Banksia species examined in the present study had LMA higher than 134 g m-2; thus, the absence of a strong correlation between PNUE and LMA in these species is consistent with the findings of Harrison et al. (2009). Interestingly, PNUE was high compared with previous studies on other species (Reich et al., 1991; Sobrado, 2009). This is similar to the finding of a high photosynthetic phosphorus use efficiency in Banksia species (Denton et al., 2007). The physiological basis for these high nutrient-use efficiencies is unresolved, but the adaptation is presumably vital for these species, which occur on some of the most nutrient-impoverished soils in the world (Richardson et al., 2004). Interestingly, neither the fraction of nitrogen allocated to Rubisco nor that allocated to thylakoids correlated with LMA (Fig. 8B). This suggests that, although high-LMA leaves do have lower leaf Nmass, they are not „disadvantaged‟ by reduced allocation of nitrogen to photosynthetic machinery as a result of an increased proportion of leaf nitrogen in supporting tissue. The observed low Nmass is consistent with the species‟ high LMA and their oligotrophic habitats (Reich et al., 1991; Wright et al., 2001; Niinemets et al., 2009).

[88]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

High-LMA species had higher Narea and lower Nmass, with the latter being more strongly correlated with LMA. This is in agreement with the global trends (Wright et al., 2004). Since a large percentage of leaf N is directly associated with the photosynthetic machinery

(Wright et al., 2004), the negative relationship between Nmass and LMA is probably due to the more numerous thick-walled cells and sclerified tissues in high-LMA leaves, and is consistent with the general trend of species at the high-LMA end of the spectrum (Wright et al., 2004). It would be worthwhile to measure N concentrations of mesophyll tissue across the range of LMA. Our data do not enable us to estimate mesophyll N concentrations; however, Chl per mesophyll volume was estimated based on the assumption that all leaf chlorophyll is located in the mesophyll. Chlmes did not decrease with LMA. A similar pattern is expected for N per mesophyll.

In contrast with Amass and consistent with global trends (Wright et al., 2004; Poorter et al.,

2009), Aarea was independent of LMA, whereas Ames decreased since MVA increased with

LMA. The chloroplastic CO2 concentration (Cc) was remarkably stable across the LMA range examined (Hassiotou et al., 2009a or Chapter 5), so this does not explain a lower

Ames. Evans et al. (2009) reported a positive relationship between mesophyll resistance per unit of exposed chloroplast surface area and mesophyll cell-wall thickness (Tw) and a negative relationship between the rates of photosynthesis per unit of exposed chloroplast surface area, Ac, and Tw. To the extent that Ames reflects Ac, the data for Banksia confirm this trend. Lower Ames may offset the impact of the increase in Tw in high-LMA leaves to moderate the CO2 drawdown from the substomatal cavity to the sites of carboxylation. A similar relationship was found by Terashima et al. (2006). A lower Ames of high-LMA leaves may reflect lower Rubisco concentrations, lower Rubisco specific activity or lower Rubisco activation state. Greater investment in photosynthetic machinery may not be advantageous in the extremely nutrient-impoverished and seasonally dry habitats of Banksia species where economic use of nutrients is vital and partial stomatal closure is common in the dry season (Veneklaas and Poot, 2003).

An increasing body of evidence shows that gm is an important factor limiting photosynthesis in C3 plants (Flexas et al., 2008; Evans et al., 2009). In seven Banksia species, gm was indeed significantly lower in high-LMA compared with low-LMA species

(Fig. 5C and Hassiotou et al., 2009a or Chapter 5). The negative relationship between gm [89]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

and LMA was mainly associated with DL and not with TL, since the latter correlated poorly with gm. This finding is consistent with data in Niinemets et al. (2009) for a set of 35 Australian species from 20 genera.

Variability in Aarea was explained by Ames and AChl to a greater extent than by VM, MVA and

Chlmes, and although none of the correlations with Aarea were significant at P=0.05, these trends indicate that the photosynthetic capacity of the tissue is more responsible for the variation in Aarea than the amounts of photosynthetically active tissue. Interestingly, Aarea in the examined species reached values that were comparable with many mesophytic species of lower LMA (Flexas et al., 2008). Denton et al. (2007) found similar photosynthetic rates in field-grown Banksia plants.

In addition to lower investment in photosynthetic machinery, lower Ames in high-LMA leaves could be due to irregular distribution of CO2 and light across these leaves. Interestingly, there are some indications of anatomical and physiological mechanisms that may reduce the negative effects of the structure of thick and dense leaves on CO2 diffusion and light transmission. Increased presence of bundle sheath extensions and other sclerenchymatous tissues in high-LMA leaves facilitates light transmission to deeper leaf layers (Poulson and Vogelmann, 1990; Smith et al., 1997; Nikolopoulos et al., 2002; Karabourniotis, 1998), improving the uniformity of illumination across thick leaves. Stomatal crypts, present in most Banksia species, in which crypt depth increases with LMA and leaf thickness (Hassiotou et al., 2009b or Chapter 2), facilitate CO2 diffusion to adaxial palisade cells.

4.6 CONCLUSIONS

The detailed analyses of the specific leaf structural and physiological traits contributing to variation in Aarea in Banksia leaves have provided new insights into the relationship between Aarea and LMA at the high-LMA end of the spectrum. These leaves have large amounts of dense tissues that are not photosynthetically active, and therefore it is not a surprise that they have lower Amass. Our analysis of the factors that contribute to variation in

Aarea, however, shows that high-LMA leaves actually have more mesophyll per unit leaf area, but that the photosynthetic capacity of this tissue is lower. The net result is that area-

[90]

CHAPTER 4: SCLEROPHYLLY AND PHOTOSYNTHESIS

based photosynthetic rates do not change with LMA. There is evidence that the lower photosynthetic capacity of the mesophyll of high-LMA leaves is not due to lower nitrogen or chlorophyll concentrations. Future research must focus on how the investment in photosynthetic machinery (e.g. Rubisco) changes with LMA, as a potential explanation of the lower Ames of high-LMA leaves. Moreover, the distribution of CO2, light and Rubisco across these thick leaves may not be uniform, thus influencing the overall pattern of photosynthetic capacity. Structural adaptations such as bundle sheath extensions that enhance light transmission (Nikolopoulos et al., 2002) and stomatal crypts that enhance

CO2 diffusion into the leaves (Hassiotou et al., 2009b or Chapter 2) may be important in minimising the negative effects of leaf structure on area-based photosynthesis.

[91]

[92]

CHAPTER 5

INFLUENCE OF LEAF DRY MASS PER AREA, CO2, AND IRRADIANCE ON MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

5.1 ABSTRACT ...... 94

5.2 INTRODUCTION ...... 94

5.3 MATERIALS AND METHODS ...... 97 5.3.1 Plant material and growth conditions ...... 97 5.3.2 Gas exchange and chlorophyll a fluorescence ...... 98

5.3.3 Calibration of the relationship between Jf and J ...... 100 CO 2

5.3.4 Estimation of gm using the „variable J method‟ ...... 101 5.3.5 Statistical analyses ...... 101

5.4 RESULTS ...... 103

5.5 DISCUSSION ...... 109

5.5.1 Effects of LMA on gm ...... 109

5.5.2 Effects of CO2 concentration and irradiance on gm ...... 112 5.5.3 Methodological issues ...... 113

5.6 CONCLUDING REMARKS ...... 115

[Hassiotou F, Ludwig M, Renton M, Veneklaas EJ, Evans JR (2009). Influence of leaf dry mass per area, CO2, and irradiance on mesophyll conductance in sclerophylls. Journal of Experimental Botany 60: 2303- 2314]

[93]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

5.1 ABSTRACT

Leaf photosynthesis (A) is limited by mesophyll conductance (gm), which is influenced by both leaf structure and the environment. Previous studies have indicated that the upper bound for gm declines as leaf dry mass per area (LMA, an indicator of leaf structure) -2 increases, extrapolating to zero at a LMA of about 240 g m . No data exist on gm and its response to the environment for species with LMA values higher than 220 g m-2. In this study, laboratory measurements of leaf gas exchange and in vivo chlorophyll a fluorescence were used concurrently to derive estimates of gm in seven species of the Australian sclerophyllous genus Banksia covering a wide range of LMA (130-480 g m-2).

Irradiance and CO2 were varied during those measurements to gauge the extent of environmental effects on gm. A significant decrease of gm with increasing LMA was found. gm declined by 35-60% in response to increasing atmospheric CO2 concentrations at high irradiance, with a more variable response (0-60%) observed at low irradiance, where gm was on average 22% lower than at high irradiance at ambient CO2 concentrations. Despite considerable variation in A and LMA between the Banksia species, the CO2 concentrations -1 in the intercellular air spaces (Ci, 262±5 μmol mol ) and in the chloroplasts (Cc, 127±4 μmol mol-1) were remarkably stable.

5.2 INTRODUCTION

Photosynthesis requires diffusion of CO2 into the leaf. The diffusion pathway has several components. Turbulent diffusion outside the leaf carries CO2 towards the leaf surface where it passes through a laminar boundary layer. Entry into the leaf is restricted by stomatal pores, which are generally situated on the leaf surface, but are sometimes below the leaf surface in crypts. From the substomatal cavity, CO2 then diffuses through the intercellular air spaces to mesophyll cell walls where it dissolves into the liquid phase. CO2 diffuses through the cell walls and cytosol and into the chloroplast where it combines with RuBP (ribulose-1,5- bisphosphate) and enters the photosynthetic carbon reduction cycle. Until recently, research focused on the resistance caused by stomata (Parkhurst, 1994; Evans and von Caemmerer, 1996; Morison and Lawson, 2007; Warren, 2007). For convenience, conductance through the mesophyll (gm) had been assumed to be infinite

[94]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

prior to techniques being developed to measure gm. However, now that methodology to measure gm is more widely available, increasing attention is being paid to understanding leaf internal diffusion, as it actually accounts for more than 40% of the decrease in CO2 concentration between the atmosphere and the sites of carboxylation (Warren, 2007; Flexas et al., 2008).

The structure of the mesophyll can be a major determinant of gm through either its gaseous component in the intercellular air spaces (gias) or its liquid component from the cell walls to the chloroplasts (gliq). Mesophyll porosity and thickness can affect gias (Parkhurst, 1994; Evans and von Caemmerer, 1996), which is dependent on the length, diameter and tortuosity of the diffusion pathway, i.e. it is determined by the depth of the mesophyll layer, the position of stomata relative to mesophyll cells, and the size, shape and packing of mesophyll cells. Morison et al. (2005) computed the vertical gias of 56 species surveyed by

Slaton and Smith (2002), using the diffusion coefficient of CO2 in still air, mesophyll -2 -1 porosity and leaf thickness, and found a range of gias values of 0.1-1.9 mol m s . Parkhurst and Mott (1990) used helox (air where nitrogen has been replaced by helium) to measure the resistance of the gaseous path to diffusion, and found that gias accounted for

10-60% of gm, being more important in hypostomatous leaves. On the other hand, Genty et al. (1998) have shown that gias contributes only a small proportion to the total drawdown in the concentration of CO2 in the mesophyll, and that gliq is the main determinant of gm, which has also been supported by other investigators (von Caemmerer and Evans, 1991; Aalto and Juurola, 2002; Sharkey et al., 2007; Warren, 2007). Leaf anatomical traits such as mesophyll cell surface area and chloroplast surface area exposed to intercellular air spaces (Evans et al., 1994; Evans and Loreto, 2000), chloroplast rearrangements (Sharkey et al., 1991; Tholen et al., 2007) and cell wall thickness (Terashima et al., 2006) can all impact on gliq. Nobel et al. (1975) argued that cell walls impose a major resistance to photosynthesis, which suggests that sclerophylls (hard leaves) with thick cell walls may have lower gm. Interestingly, in the early stages of gm research, von Caemmerer and Evans (1991) and Lloyd et al. (1992) found that sclerophylls, such as Eucalyptus blakelyi,

Macadamia integrifolia, Citrus limon and Citrus paradisi, had gm values between 0.1-0.25 mol m-2 s-1, which overlapped with the range observed for thin and soft leaves of tobacco and bean with similar rates of photosynthesis. Loreto et al. (1992) found that seven

[95]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

sclerophylls had low values of gm and low rates of photosynthesis, but the relationship between photosynthesis and gm did not vary between sclerophyllous plants and mesophytes. Leaf dry mass per area (LMA) is a leaf morphological trait that has often been used as an indicator of sclerophylly (Gratani and Varone, 2006) and a recent review by

Flexas et al. (2008) examined the relationship between LMA and gm, summarising data from 17 studies. While low-LMA mesophytic species present a wide range of gm values, leaf structure appears to strongly limit gm in evergreen species with high LMA (Flexas et al., 2008). It is evident that gm decreases with increasing LMA, extrapolating to zero at a LMA of 240 g m-2. However, since LMA in sclerophyllous plants can be much higher than 240 g m-2, there is a need to extend the range of measurements. During the last decade, it has been shown that leaf structure is not the only determinant of gm. Interestingly, gm seems to respond to environmental factors such as soil water availability, salinity and temperature (Warren, 2007; Flexas et al., 2008). Bernacchi et al.

(2002) argued that the temperature response coefficient (Q10) of approximately 2.2 for gm that they found for tobacco leaves suggests a role for proteins in the path of CO2 diffusion. Current research is focusing on two proteins, carbonic anhydrase (CA) and cooporins. CA has been shown to have a modest effect on gm (Price et al., 1994; Williams et al., 1996).

However, Gillon and Yakir, (2000) suggested that the role of CA in gm regulation may be more important when gm is low, such as in sclerophyllous species. Moreover, research on cooporins, initially using inhibitors, then antisense and over-expression genotypes, suggests that certain cooporins located in the plasma and chloroplast membranes are permeable to CO2 and are involved in the regulation of gm (Terashima and Ono, 2002; Üehlein et al., 2003; Hanba et al., 2004; Flexas et al., 2006; Üehlein et al., 2008;

Miyazawa et al., 2008). The findings of Flexas et al. (2007a) that gm responded rapidly to changes in atmospheric CO2 concentration in six species and that gm was reduced in tobacco when measured under low irradiance show that part of the resistance pathway can be reversibly altered within minutes, which further supports an active regulation of gm. Given that LMA values of Banksia species are between 130 and 500 g m-2 (Hassiotou, unpublished data), this genus provides an excellent model to examine the effect of leaf structure as indicated through LMA on gm as well as extend the current knowledge on gm for species at the high-LMA end. In this context, seven Banksia species covering a wide

[96]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

range of LMA were selected and mesophyll conductance was estimated from A-Ci curves at two irradiances using combined gas exchange and chlorophyll fluorescence measurements.

5.3 MATERIALS AND METHODS

5.3.1 Plant material and growth conditions Three- to five-year old plants of seven Banksia species that covered a wide range of LMA were used (Table 1). The experiment was done in two phases: (a) in September-October 2007 in Canberra, eastern Australia (Southern Hemisphere spring), measurements were carried out on three species (Banksia integrifolia L.f., Banksia serrata L.f. and Banksia paludosa R.Br.), using plants purchased from a local nursery. Upon purchase, the plants were re-potted into 10-L pots containing a mixture of grey sand and potting mix and grown for two months prior to the measurements in a greenhouse (25oC day/20oC night); and (b) in November-December 2007 in Perth, Western Australia (Southern Hemisphere spring to summer), measurements were done on four species (Banksia solandri R.Br., Banksia elderiana F.Muell. and Tate, Banksia attenuata R.Br., and Banksia repens Labill.), using plants in 10-L pots containing river sand and potting mix. In 2002, seeds of the above four Banksia species were germinated and plants were grown outdoors until about three weeks before the measurements, when they were transferred to a controlled-temperature greenhouse (23oC day/18oC night).

Table 1. Leaf dry mass per area (LMA) of the Banksia species examined. Values are averages ± standard errors of three replicates per species.

Species LMA (g m-2) B. serrata 134 ± 12.5 B. integrifolia 171 ± 3.6 B. attenuata 217 ± 1.4 B. solandri 238 ± 3.6 B. paludosa 262 ± 2.6 B. repens 269 ± 8.2 B. elderiana 478 ± 8.2

[97]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

5.3.2 Gas exchange and chlorophyll a fluorescence Combined gas exchange and chlorophyll a fluorescence measurements (the so called “fluorescence method”, Evans and von Caemmerer, 1996) were carried out using the youngest fully expanded leaves. Three leaves were measured from each species, each leaf originating from a different plant, with the exception of B. integrifolia and B. paludosa, where three leaves of the same plant were used. For each leaf, two sets of measurements were taken: (a) the CO2 response of gas exchange and chlorophyll fluorescence measured simultaneously, at ambient O2 concentration and at two irradiances (500 and 1500 μmol -2 -1 quanta m s ); and (b) the CO2 response of gas exchange and chlorophyll fluorescence -2 measured simultaneously, at 2% O2 and at two irradiances (500 and 1500 μmol quanta m s-1). At the end of the measurements, all leaves were collected, leaf area (using a leaf area meter, LI-300A, Li-Cor, Lincoln, NE, USA) and leaf dry mass (after drying at 70oC for three days) were measured, and leaf dry mass per area (LMA) was calculated. For (a) and (b), a LI-6400 open gas exchange system was used with an integrated 2 cm2 chamber fluorometer (LI-6400-40, Li-Cor, Lincoln, NE, USA). The photochemical efficiency of

Photosystem II (ΦPSII) was determined by measuring steady-state fluorescence, Fs, -2 -1 followed by a multiple saturating pulse of 7000 μmol quanta m s (Fm') according to Genty et al. (1989):

Fm ' Fs ΦPSII (1) Fm '

Before the measurements were taken, CO2 and H2O values were zeroed and fresh drierite (and soda lime if necessary) was used. Leaf temperature was between 21-23oC, except for B. integrifolia, where the leaf temperature was 18.7-19.4oC. Leaf to air vapour pressure difference (VPD) was kept between 1.2-1.4 kPa; this range had no effect on stomatal conductance (gs). Leaves were kept in the gas exchange chamber at high irradiance (1500 -2 -1 -1 μmol quanta m s ) and low ambient CO2 concentration (Ca) (100 μmol mol ) for at least 30 min before the commencement of the simultaneous measurement of chlorophyll fluorescence and the response of net leaf photosynthesis (A) to the intercellular CO2 concentration (Ci) (A-Ci response), ensuring stomata were fully open and steady state was reached. The CO2 response of photosynthesis and chlorophyll fluorescence measurements

[98]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

-1 -1 were started at a Ca of 380 μmol mol , then the Ca was decreased to 50 or 100 μmol mol -1 -1 and subsequently increased at 50 μmol mol intervals to 800 μmol mol . At each Ca, 2-4 measurements were recorded, at least 1 min apart. This was sufficient time for all leaf parameters to stabilise after the pulse of CO2. Between different Ca values, gas exchange and chlorophyll fluorescence were allowed to stabilise before the next pulse was given. For -2 a particular leaf, the CO2 response was recorded at high irradiance (1500 μmol quanta m s-1) and then at low irradiance (500 μmol quanta m-2 s-1) without removing the leaf from the gas exchange chamber and after allowing the stomata to re-open. Tests showed that the order of the high and low irradiances had no effect on the CO2 response. Possible leakage into and out of the cuvette was checked and the correction approach suggested by Flexas et al. (2007b) was deemed unnecessary. To improve the gasket seal around the midrib for some leaves, an additional flexible modelling compound (Terastat) was used. Lateral diffusion through the leaf is unlikely to have happened due to the heterobaric nature of these leaves.

The C3 photosynthesis model (Farquhar et al., 1980) was fitted to the A-Ci measurements, whereby A is the minimum of the RuBP-saturated rate of photosynthesis (Ac) and the

RuBP-limited rate (Aj):

A min( Ac , Aj ) Rd (2) where Ac and Aj are computed as:

* Vc (Ci Γ ) Ac Rd (3) Ci Kc (1 O / Ko )

* J (Ci Γ ) Aj * Rd (4) 4Ci 8Γ * Vc is the maximum rate of RuBP carboxylation, Γ is the photo-compensation point in the absence of mitochondrial respiration, Rd is respiration in the light, Kc and Ko are the

Michaelis-Menten constants for CO2 and O2, respectively, O is the O2 concentration and J is the rate of electron transport.

[99]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

5.3.3 Calibration of the relationship between Jf and J CO 2 The rate of electron transport estimated from chlorophyll fluorescence prior to any calibration (Jf, uncal) is given by the equation:

J f,uncal ΦPSII PPFD (5) where PPFD is the photosynthetic photon flux density, α is leaf absorptance and β represents the proportion of quanta absorbed by Photosystem II (this has been assumed to be 0.5 for C3 plants; Ögren and Evans, 1993). A Taylor integrating sphere (Taylor, 1935) and a spectroradiometer (LI1800, Li-Cor, Lincoln, NE, USA) were used to obtain estimates of α for the LED light source. Two leaves per species (originating from two different plants, with the exception of B. paludosa and B. integrifolia, where two leaves of the same plant were used) were measured and their average absorptance was used as the α value of that species.

The relationship between Jf, uncal and the rate of electron transport obtained from gas exchange ( J ) bears some fundamental uncertainties. Part of these uncertainties can be CO 2 eliminated by measuring α and β (although the latter is rarely measured; Warren, 2006). However, even when these measurements are made, there is still much uncertainty in the relationship between Jf and which may originate from alternative electron sinks and/or the fact that the fluorescence signal primarily emanates from the upper mesophyll layers and thus, most probably, is not representative of the whole leaf, in contrast with the gas exchange signal (Warren, 2006). To partially deal with the first uncertainty, fluorescence was calibrated by relating ΦPSII to A measured under non-photorespiratory conditions. Φ was calculated as = 4*(A+Rd)/PPFD. This approach assumes that CO 2 under low O2 conditions, photorespiration is suppressed; hence the true relationship between Jf and can be established (Epron et al., 1995). Thus, A-Ci curves (for Ca of

100, 200, 400, 800 and for some species 1000 μmol mol-1), combined with chlorophyll -2 -1 fluorescence measurements, were generated at 2% O2 at both low (500 μmol quanta m s ) -2 -1 and high (1500 μmol quanta m s ) irradiances. The CO2 responses under non- photorespiratory conditions were collected after the CO2 responses under photorespiratory

[100]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

conditions for a particular leaf, without removing it from the gas exchange chamber. Initial tests showed that this order of data collection did not influence the results.

The relationship between Φ and ΦPSII at 2% O2 at each irradiance was used to calibrate CO 2

Jf, which was calculated as:

J f [(ΦPSII s) c] PPFD (6) where s is the slope and c is the intercept of the relationship between and ΦPSII at 2%

O2.

5.3.4 Estimation of gm using the Variable J method The “variable J method” (Harley et al., 1992), based on simultaneous measurements of gas exchange and chlorophyll fluorescence, was employed to obtain estimates of gm for each -2 -1 -2 - point of the A-Ci curves at low (500 μmol quanta m s ) and high (1500 μmol quanta m s 1 ) irradiances. Initially, the A-Ci curves were converted to A-Cc curves using the equation:

Γ [Jf 8(A Rd )] Cc (7) Jf 4(A Rd ) gm was then calculated as: A gm (8) Ci Cc -2 -1 * -1 A Rd of 0.8 μmol m s and a Γ of 36.9 μmol mol (Brooks and Farquhar, 1985), adjusted using the temperature dependence stated by the authors, were used. For the temperature corrections, the average temperature of the leaf during the construction of the A-Ci curve * was used. A sensitivity analysis was also done to examine the effects of different Rd and Γ on gm.

5.3.5 Statistical analyses -2 -1 To test whether gm at 500 μmol quanta m s (gm500) was significantly lower than gm at -2 -1 1500 μmol quanta m s (gm1500), a paired T-test was carried out comparing gm for the two irradiances at ambient CO2 concentration. This test was done separately for each species and by considering all the replicates of all species. A regression analysis, based on the

mCi+k exponential model gm = e , was done to determine whether gm decreased significantly

[101]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

with increasing Ci. The data from each leaf were fitted to this model to examine whether the slope parameter m was significantly different to zero in each case. Moreover, the relationship between LMA and the rate at which gm decreased with increasing Ci was examined using a mixed-effects model to predict log(gm), with Ci and LMA as the explanatory variables and individual leaves as random effects, and then considering the significance of the Ci-LMA interaction. The above regression analyses were carried out using the R programming language (R Development Core Team, 2008), the first using the generalised linear modelling function with Gaussian errors and an exponential link function, and the second using the linear mixed-effects modelling function (Pinheiro et al., 2007).

50 A B

)

-1 40

s

-2

)

-1

s

m

2 30

-2

20 mol m

A (

mol CO mol 10

(

A 3000 C D 250

)

-1

s 200

-2 150

mol m mol 100

(

f J 50

0 0 100 200 300 400 500 600 70000 100 200 300 400 500 600 700 C ( mol mol-1) C ( mol mol-1) i i

Fig. 1. Response of net photosynthesis (A) and electron transport rate estimated from fluorescence

(Jf) to intercellular CO2 concentration (Ci) in Banksia solandri (A, C) and B. integrifolia (B, D), at a PPFD of 1500 μmol quanta m-2 s-1 (open symbols) and 500 μmol quanta m-2 s-1 (filled symbols). Measurements from three replicate leaves (circles, squares, triangles) of each species at each PPDF are shown.

[102]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

5.4 RESULTS gm was estimated in seven Banksia species using gas exchange and chlorophyll fluorescence at different CO2 concentrations and irradiances. Results from two of the species studied, B. solandri and B. integrifolia, are presented in greater detail. The relationships between CO2 assimilation rate per unit leaf area (A) or electron transport rate

(Jf) and intercellular CO2 concentration (Ci) are shown for three replicate leaves (Fig. 1).

Greater Ci values were reached in B. integrifolia compared with B. solandri because stomatal conductance declined less at high ambient CO2 concentration (Ca). For both species, CO2 assimilation rates were reduced across the range of Ci values used at a PPFD of 500 compared with 1500 μmol quanta m-2 s-1. The transition from Rubisco to RuBP -1 regeneration limitation of A at a Ci of about 300 μmol mol is apparent from the calculated rate of electron transport (Fig. 1B, D).

50 A B

)

-1 40

s

-2

m

2 30

20 T = 22.8 mol CO leaf Tleaf = 18.7 Vc = 98, 73 Vc = 110, 76 ( 10 A J = 174, 124 J = 207, 143

Rd = 3.9, 2.7 Rd = 6.8, 3.2 0 0 100 200 300 400 500 600 70000 100 200 300 400 500 600 700 C ( mol mol-1) C ( mol mol-1) i i

Fig. 2. Responses of net photosynthesis (A) to leaf intercellular CO2 concentration (Ci) fitted to the

C3 model of photosynthesis (Farquhar et al., 1980) in a leaf of Banksia solandri (A) and a leaf of B. integrifolia (B) at a PPFD of 1500 μmol quanta m-2 s-1 (open circles) and 500 μmol quanta m-2 s-1 o -2 -1 (filled circles). Tleaf ( C) is the leaf temperature, Vc (μmol m s ) is the RuBP-carboxylation rate, J -2 -1 -2 -1 (μmol m s ) is the electron transport rate and Rd (μmol m s ) is the respiration rate in the light. For each of these parameters, the first number corresponds to the value of the parameter at a PPFD of 1500 μmol quanta m-2 s-1, whereas the second number is the value at a PPFD of 500 μmol quanta m-2 s-1.

[103]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

0.30 A B 0.25

0.20

2

CO 0.15

0.10

0.05

0.00 0.00 0.10 0.20 0.30 0.40 0.50 0.000.600.00 0.10 0.20 0.30 0.40 0.50 0.60

PSII PSII

Fig. 3. Calibration curves for Banksia solandri (A) and B. integrifolia (B) showing the relationship between electron transport rate estimated from gas exchange and electron transport rate estimated from fluorescence under non-photorespiratory conditions (2% O2). Data are expressed as gas exchange-based quantum yield (Φ = 4(A + Rd)/PPFD) versus photochemical efficiency of CO 2

Photosystem II (ΦPSII). Each calibration was done at four Ca values (100, 200, 400 and 800 μmol mol-1) and at PPFDs of 1500 μmol quanta m-2 s-1 (open symbols) and 500 μmol quanta m-2 s-1 (filled symbols). Measurements from three replicate leaves (circles, squares, triangles) of each species at each PPFD are shown.

In the RuBP regeneration-limited region, Jf was essentially constant for B. integrifolia, but gradually increased for B. solandri. In some species (e.g. B. repens) the RuBP -1 regeneration-limited region was not reached at a Ca of 800 μmol mol . A did not decrease at high CO2 concentrations in any of the leaves measured. Rubisco-limited and RuBP regeneration-limited curves were fitted to A-Ci curves for one replicate leaf of each of the two species (Fig. 2). The RuBP regeneration-limited region was probably not reached by -1 the B. solandri leaf at a Ca of 800 μmol mol , but a curve was fitted through the two highest Ci points by way of illustration. The apparent transition between the two limitations -1 occurred at a Ci of 350 to 400 μmol mol , which was higher than that indicated from Jf. While the Rubisco-limited region is well described by the model, the model curves fit poorly around the transition region. Comparison of the relationship between the gas exchange-based quantum yield ( ) and the photochemical efficiency of Photosystem II

[104]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

(ΦPSII) at 2% O2 and different irradiances (Fig. 3) revealed that separate calibrations were required for each leaf at each irradiance to accurately estimate Jf, and thus gm (Table S1 in Appendix 3).

A two-fold range in CO2 assimilation rate per unit leaf area was found between the Banksia species examined, with a trend to lower rates as LMA increased (Fig. 4A). Mesophyll conductance declined from 0.17 mol m-2 s-1 in B. attenuata to 0.09 mol m-2 s-1 in B. elderiana as LMA increased fourfold (Fig. 4B). Increasing LMA did not alter the draw- down in the CO2 concentration from the atmosphere to the sites of carboxylation (Ca-Cc), and no correlation was found between LMA and the draw-down in the CO2 concentration from the atmosphere to the substomatal cavities (Ca-Ci) or from the substomatal cavities to the sites of carboxylation (Ci-Cc) (Fig. 4C, D, E). Strong correlations were found between

A and both gs and gm (Fig. 5). Scatter in these relationships was also evident, as both B. integrifolia and B. attenuata had low gm relative to A. -2 -1 At 1500 μmol quanta m s , a significant decrease in gm with increasing Ci was found in all the replicates of the seven species examined (Table S2 in Appendix 3). gm declined by -1 35-60% in response to increasing CO2 concentration (Ca values of 50 to 800 μmol mol ) at 1500 μmol quanta m-2 s-1, with a more variable response (0-60%) at 500 μmol quanta m-2 s- 1 . Despite the variability of gm amongst some replicates at very low and/or very high CO2 concentrations, a decline in gm with increasing CO2 concentration was evident at both irradiances in B. integrifolia and at high irradiance in B. solandri (Fig. 6). At 500 μmol -2 -1 quanta m s , a significant decrease in gm with increasing ambient CO2 concentration was seen in all the species examined, with the exception of B. solandri (all replicates) (Fig. 6C) and two out of three replicates of B. paludosa, for which trends were absent (Table S2 in

Appendix 3). Mixed-effects modelling showed that the interaction between LMA and Ci was not significant (P=0.07) at high irradiance (1500 μmol quanta m-2 s-1) but highly significant (P<0.0001) at low irradiance (500 μmol quanta m-2 s-1), indicating that the rate at which gm decreased with increasing Ci was more pronounced in species with high LMA.

To illustrate the sensitivity of the gm calculation to the fluorescence calibration, two calculations are shown for a B. solandri leaf at two irradiances (Fig. 7): 1. fluorescence was calibrated using the slope and intercept of the Φ - ΦPSII relationship under non- CO 2

[105]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

35 Fig. 4. Relationship between (A) rate of A r²=0.44 ) 30 -1 net photosynthesis (A), (B) mesophyll

s

-2 25 conductance (gm), (C) the draw-down in

m 2 20 the CO2 concentration from the 15 atmosphere to the substomatal cavities mol CO 10

( (Ca-Ci), (D) the draw-down in the CO2 A 5 0.250 concentration from the substomatal B r²=0.69 LMA (g m-2) cavities to the sites of carboxylation 0.20

) -1 (Ci-Cc), and (E) the draw-down in the

s

-2 0.15 CO2 concentration from the atmosphere 0.10 to the sites of carboxylation (Ca-Cc),

(mol m

m g 0.05 and leaf dry mass per area (LMA) at a PPFD of 1500 μmol quanta m-2 s-1 and 0.00300

-2 C LMA (g m ) ambient CO2 concentration in the seven

) 250 -1 Banksia species examined (B. serrata: 200 squares, B. integrifolia: crosses, B. 150

mol mol attenuata: circles, B. solandri: triangles

( i 100

C

- a up, B. paludosa: stars, B. repens: C 50 triangles down, B. elderiana: 3000

D LMA (g m-2) diamonds).

) 250 -1 200

150

mol mol

( c 100

C

-

i

C 50

3500 E LMA (g m-2)

)

-1 300

mol mol mol 250

(

c

C

- a 200 C 150 0 100 200 300 400 500 600 LMA (g m-2)

[106]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

0.4

) 0.3

-1

s

-2

m

2 0.2

(mol CO g 0.1

0.0 0 5 10 15 20 25 30 35 -2 -1 A ( mol CO m s ) 2

Fig. 5. Relationship of mesophyll conductance (gm) (filled symbols) and stomatal conductance (gs) (open symbols) with rates of net photosynthesis (A) at a PPFD of 1500 μmol quanta m-2 s-1. A was -1 normalised to a common intercellular CO2 concentration (Ci of 262 μmol mol ), which was the average Ci of the seven Banksia species examined at ambient CO2 concentration (Ca), to take into account any variation in gs. (B. serrata: squares, B. integrifolia: crosses, B. attenuata: circle, B. solandri: triangles up, B. paludosa: stars, B. repens: triangles down, B. elderiana: diamonds). A significant positive correlation was found between gs-A (P<0.01) and gm-A (P<0.05).

photorespiratory conditions; 2. the rate of electron transport was calculated from Eqn 5 analytically. Surprisingly, under high irradiance, gm values calculated from Eqn 5 were more stable over the Ci values tested than when the calibrated fluorescence was used. The -1 latter resulted in a high gm value at a Ci of around 150 μmol mol . Under low irradiance, -1 gm values calculated from Eqn 5 increased dramatically once Ci exceeded 500 μmol mol , whereas gm values calculated from calibrated fluorescence were consistent over the Ci range tested. The uncertainty of gm estimates increases at low and high CO2 concentrations. * A sensitivity analysis of the effects of Rd and Γ on the estimation of gm showed that the lower the Ci the greater the impact of these parameters on gm (Table S3 in Appendix 3). * Although the degree to which Rd and Γ influenced the estimation of gm changed with

[107]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

0.6 A B 0.5

)

-1

s 0.4

-2 0.3

(mol m 0.2

m

g 0.1

0.00.50.0 C D 0.4

)

-1

)

s

-1 -2 0.3 s

-2

0.2 (mol m

m

g

(mol m

m g 0.1

0.0 0 100 200 300 400 500 600 70000 100 200 300 400 500 600 700 C ( mol mol-1) C ( mol mol-1) i i

Fig. 6. Response of mesophyll conductance (gm) to intercellular CO2 concentration (Ci) at a PPFD of 1500 μmol quanta m-2 s-1 (A, B) and 500 μmol quanta m-2 s-1 (C, D) in Banksia solandri (A, C) and B. integrifolia (B, D). Measurements from three replicate leaves (circles, squares, triangles) of each species at each PPFD are shown.

irradiance, this was of minor importance (Table S3 in Appendix 3). Regardless of Ci, gm was always more sensitive to changes in Rd at lower than at higher irradiance. The effect of * Γ on gm, however, was somewhat greater at higher than at lower irradiance at Ci values -1 lower than 200 μmol mol , but the opposite was found at Ci values higher than 200 μmol -1 * -1 mol . Reduction of Γ from 36.9 to 28 μmol mol reduced gm by 22.5% and 25.2% on average at 1500 and 500 μmol quanta m-2 s-1, respectively, whereas increasing Γ* from 36.9 -1 to 40 μmol mol increased gm by 11.7% and 14.3% on average at 1500 and 500 μmol quanta m-2 s-1, respectively. -2 For each leaf, A-Ci curves were measured at PPFD levels of 1500 and 500 μmol quanta m -1 s which allowed us to examine whether gm varied with irradiance (Fig. 8). On average, gm

[108]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

0.6 Fig. 7. Effect of chlorophyll fluorescence A 0.5 calibration on mesophyll conductance (gm) in a

) -1 0.4 s leaf of Banksia solandri. The response of gm to

-2 0.3 leaf intercellular CO2 concentration (Ci) is

(mol m -2 -1

m 0.2

g presented at a PPFD of 1500 μmol quanta m s 0.1 (A) and 500 μmol quanta m-2 s-1 (B). Open 0.00.7 symbols show gm calculated from uncalibrated 0.6 B fluorescence, whereas filled symbols show gm ) 0.5

-1

s

-2 calculated from calibrated fluorescence. 0.4

0.3

(mol m

m

g 0.2 0.1 0.0 0 100 200 300 400 500 600 C ( mol mol-1) i

-2 -1 - at a PPFD of 500 μmol quanta m s (gm500) was 22% less than that at 1500 μmol quanta m 2 -1 s (gm1500). This compares to a decrease in A of 27%. The decrease in gm with decreased irradiance was significant (P<0.01 or P<0.05) for all the species studied, with the exception of B. solandri, in which one out of the three leaves measured had higher gm500 than gm1500.

5.5 DISCUSSION

The combined effects of leaf structure (LMA), ambient CO2 concentration and irradiance on gm were investigated in sclerophylls of the genus Banksia with a wide range of LMA.

Whilst gm decreased with increasing LMA, it scaled with A such that gm imposed a similar limitation on A in all species. After discussing the species differences in leaf attributes, I then deal with dynamic changes of gm in response to CO2 and irradiance and uncertainties associated with the method.

5.5.1 Effects of LMA on gm

In a recent review, Flexas et al. (2008) drew an upper bound for gm that decreased linearly -2 with increasing LMA, reaching zero at a LMA value of about 240 g m . Our gm values for

[109]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

0.25

) 0.20

-1

s -2 0.15

(mol m 0.10

500

m g 0.05

0.00 0.00 0.05 0.10 0.15 0.20 0.25

-2 -1 gm1500 (mol m s )

Fig. 8. Effect of irradiance on mesophyll conductance (gm). Mesophyll conductance at ambient CO2 -2 -1 -2 -1 concentration and a PPFD of 500 μmol quanta m s (gm500) and 1500 μmol quanta m s (gm1500) for all replicates of six Banksia species is shown (B. serrata: squares, B. integrifolia: crosses, B. attenuata: circles, B. solandri: triangles up, B. paludosa: stars, B. repens: triangles down). The dotted line is the 1:1 line.

Banksia species with LMA values ranging from 130-480 g m-2 lie well above this upper bound. Therefore, the notional upper bound is redrawn as a concave curve with an unknown asymptotic value at LMA greater than 500 g m-2 (Fig. 9).

Banksia species with high LMA values showed significantly (P<0.05) lower gm than low- -2 -1 LMA species at 1500 μmol quanta m s . Leaf structure may constrain gm (Fig. 4) either through gias (conductance to diffusion in the mesophyll intercellular air spaces), gliq (conductance to diffusion through the cell walls, the cytosol and chloroplast envelope) or both. Lower mesophyll porosity and greater leaf thickness of the species with higher LMA may reduce gias. Chloroplast surface area exposed to the intercellular air spaces (Sc) has been shown to positively correlate with gm (Evans et al., 1994; Terashima et al., 2006) and one would expect Sc to increase with increasing LMA. However, high-LMA species may have photosynthetic cells with thicker cell walls which would reduce gliq (Kogami et al., 2001; Terashima et al., 2006, Evans et al., in preparation). The highest-LMA species for which gm has been examined, B. elderiana, has a mesophyll cell wall thickness of about

[110]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

0.5

0.4

)

-1

s

-2 0.3

0.2

(mol m

m

g 0.1

0.0 0 100 200 300 400 500 -2 LMA (g m )

Fig. 9. Relationship between mesophyll conductance (gm) and leaf dry mass per area (LMA) in the absence of stress in different species. Open symbols represent leaves of high-altitude plants, old leaves, shade leaves and other leaves, data that have been compiled from 17 different studies (Flexas et al., 2008). The Banksia species examined in this study are presented with filled symbols. Taking into account only the data presented in open symbols, Flexas et al. (2008) drew an upper bound for gm that decreased linearly with increasing LMA, extrapolating to zero at a LMA value of -2 about 240 g m . The current study has extended the LMA range where gm has been examined to a LMA value of 480 g m-2. Thus, this notional upper bound is redrawn as a concave curve with an unknown asymptotic value at LMA greater than 500 g m-2.

0.35 μm (Hassiotou, unpublished data), which is towards the upper end of the 0.15-0.4 μm range that has been previously reported (Hanba et al., 1999, 2001, 2002). However, it is very unlikely that mesophyll cell wall thickness is proportional with LMA, as the LMA of B. elderiana is eight times greater than the tree leaf LMA values from the above studies for which cell wall thickness was measured.

The range of gm values recorded in this study at ambient CO2 concentration and high irradiance (0.084-0.169 mol m-2 s-1) for Banksia species falls within the range reported previously for woody evergreen species (Flexas et al., 2008). Interestingly, Ci-Cc did not differ noticeably among the species despite a four-fold range in LMA and a two-fold range

[111]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

in A. It is evident from anatomical observations (data not shown) that variation in LMA is due to changes in leaf thickness, or structural sclerified tissue relative to mesophyll tissue and/or mesophyll cell wall thickening. Increased investment in structural tissue will have no consequence on a mesophyll trait like gm. However, if higher LMA is a result of mesophyll cell wall thickening, this would reduce gm per unit of exposed mesophyll surface. This reduction could be offset by increases in exposed mesophyll surface associated with thicker leaves, which could maintain a constant rate of photosynthesis per unit leaf area. To maintain the constant draw-down in CO2 concentration between the intercellular airspaces and the chloroplasts (Ci-Cc) that was observed, it is concluded that leaves with thicker mesophyll cell walls and greater mesophyll cell surface area must have a lower amount of Rubisco per unit of chloroplast surface area exposed to intercellular airspace.

The stability in Cc found across a considerable range of LMA is consistent with co-variation of gm with photosynthetic capacity. The relationship between gm and A was linear, which is consistent with what has often been reported in the literature for both mesophytic and sclerophyllous species (von Caemmerer and Evans, 1991; Loreto et al., 1992; Evans and von Caemmerer, 1996).

5.5.2 Effects of CO2 concentration and irradiance on gm

The variation in gm with CO2 that was observed in seven Banksia species was consistent with responses reported by Flexas et al. (2007a) for six species (Olea europaea, a hybrid of Vitis berlandieri × Vitis rupestris, Cucumis sativus, Arabidopsis thaliana, Nicotiana tabacum, Limonium gibertii). The relative response of gm to CO2 (as a percentage of maximum values) in the above species that was presented by Flexas et al. (2008) was very similar to that of the Banksia species studied here. The physiological explanation for the reduced gm at high CO2 concentration is not known. Consistent with optimisation of resource use, it is to be expected that gs would increase at low CO2 concentrations, to allow for more efficient photosynthesis, and would decrease at CO2 concentrations that are not limiting photosynthesis in order to reduce water loss from the leaf through transpiration; however, it is not clear why gm would decrease at high CO2 concentrations. Is there a disadvantage to high CO2 diffusion or a cost related to maintenance of high gm? This is still

[112]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

unknown. Recent evidence involving cooporins in gm regulation was given by Miyazawa et al. (2008) who showed that the decrease in gm in tobacco leaves acclimated to long-term water stress was not associated with a decrease in cooporin content. Rather, there was a loss of mercury sensitivity which was suggested to reflect deactivated cooporins. The cost- related hypothesis behind the decrease of gm at high CO2 concentrations remains to be elucidated.

Irradiance was found to also impact on gm. At ambient CO2 concentration, gm at low irradiance (500 μmol quanta m-2 s-1) was on average, 22% lower than that at 1500 μmol -2 -1 quanta m s . The effect of irradiance on gm in our high-LMA species was much less than that reported for Nicotiana tabacum (Flexas et al., 2007a). The speed and reversibility of the changes in gm suggest a change in membrane permeability. Cooporins may provide a mechanism for this through rapid changes in their gating, insertion or removal from membranes (Katsuhara et al., 2008). However, it is also possible that the irradiance dependence of gm is a consequence of the experimental technique used, since fluorescence samples the leaf differently to gas exchange. This is discussed in the next section.

5.5.3 Methodological issues

The “fluorescence method” that is often used to derive gm bears some assumptions that may explain the dependence of gm on irradiance observed in this study. This analysis treats the leaf as a “big chloroplast” which assumes that the ratio of photosynthetic capacity to light absorbed is the same for each chloroplast. In the few cases where profiles of photosynthetic properties have been measured with respect to depth from the leaf surface, the ratio of photosynthetic capacity to light absorbed is only constant with depth through a leaf in green light (Evans and Vogelmann, 2003, 2006). Since chlorophyll fluorescence and photosynthesis were measured under red light, CO2 fixation occurs predominantly near the adaxial surface under low irradiance, with fixation occurring deeper within the mesophyll under high irradiance (Evans and Vogelmann, 2003, 2006). Assuming that gias >> gliq and that the profile of gliq through the leaf is proportional to the profile of Rubisco, under dim red light there is a large draw-down (Ci-Cc) near the adaxial surface decreasing towards the abaxial surface. As irradiance increases, additional CO2 fixation occurs deeper within the mesophyll, which increases the draw-down deep within the mesophyll. However, the

[113]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

profile of chloroplasts sampled by chlorophyll fluorescence stays the same. Consequently, - while CO2 assimilation rate increases by 33% under 1500 compared to 500 μmol quanta m 2 s-1, the apparent draw-down increases by only 15% (Evans, unpublished data). This translates into a gm500 value that is 15% less than the gm1500 value, consistent with the trend shown in Fig. 8. Inevitably, there are many assumptions in this modelling approach in the absence of detailed anatomical information and measurements of gm at such a fine scale. A more comprehensive modelling effort will be presented elsewhere as it is beyond the scope of this paper.

Another consequence of the „big chloroplast‟ model is that the estimation of gm involves the application of Fick‟s Law to an average point in the mesophyll. In reality, there are distributed sinks, and a network of resistances in series and parallel would be a better analogy. While this would confer more complex behaviour, it is also less tractable given the limited spatially resolved information available. In contrast to our finding that gm estimated using the fluorescence method varies with irradiance, gm estimated by carbon isotope discrimination was found to be independent of irradiance in wheat leaves (Tazoe et al., 2009). Given that both the fluorescence and isotope methods for calculating gm require a number of assumptions, that the short-term response of gm to irradiance has only been observed for tobacco (Flexas et al., 2007a) and Banksia (Fig. 8), but not wheat (Tazoe et al., 2009) and that both methods have not been applied to the same leaves, more experimentation is needed to elucidate the response of gm to irradiance. The fluorescence signal was calibrated for each leaf to avoid the assumptions needed when calculating Jf. The sensitivity of gm to this calibration is illustrated in Fig 7. The non-zero intercept of the relationship between Φ and ΦPSII at a low O2 concentration may be CO 2 indicative of alternative electron sinks. These could include pathways in the chloroplasts that use reduced ferredoxin or NADPH (Genty and Harbinson, 1996), the oxaloacetate- malate shuttle (Scheibe, 1987), nitrate assimilation and biosynthetic activities, or cyclic electron flow around Photosystem I (Heber et al., 1978; Furbank and Horton, 1987). * A sensitivity analysis of the effects of Rd and Γ on the estimation of gm was presented * (Table S3 in Appendix 3). Although the absolute values of gm are affected by Rd and Γ , -1 these effects are significantly (P<0.01) lower at Ci values between 200 and 400 μmol mol .

[114]

CHAPTER 5: MESOPHYLL CONDUCTANCE IN SCLEROPHYLLS

* Most importantly, this exercise showed that error in neither Rd nor Γ altered the dependence of gm on CO2, irradiance or LMA.

5.6 CONCLUDING REMARKS

This study extended the range of LMA values where gm had been measured from 220 to 480 g m-2, presenting the effects of both leaf structure (through LMA) and key environmental factors (CO2 and irradiance) on gm in sclerophyllous Banksia species. Increased CO2 concentration and decreased irradiance caused gm to significantly decline. gm showed a strong correlation with LMA, with high-LMA species having significantly lower gm than low-LMA species. Interestingly, across the LMA range that was examined and across a wide variation in photosynthetic capacity, the leaves had a similar draw-down in the CO2 concentration from the atmosphere to the chloroplasts, i.e. they had similar Ci and Cc values. This supports the notion that variation in gs and gm across different species and functional groups results in a highly conserved Cc. Future studies must focus on how leaf structural parameters such as cell wall thickness and exposed chloroplast surface area influence gm.

[115]

[116]

CHAPTER 6

RESPONSE OF MESOPHYLL CONDUCTANCE MEASURED BY ON-LINE

CARBON ISOTOPE DISCRIMINATION TO CO2 AND IRRADIANCE IN SCLEROPHYLLS

6.1 ABSTRACT ...... 118

6.2 INTRODUCTION ...... 118

6.3 MATERIALS AND METHODS ...... 121 6.3.1 Plant material and growth conditions ...... 121 6.3.2 Gas exchange and carbon isotope discrimination measurements ...... 121

6.3.3 Estimation of gm by on-line carbon isotope discrimination ...... 123 6.3.4 Gas exchange and chlorophyll fluorescence measurements ...... 125

6.3.5 Estimation of gm using the Variable J method ...... 125

6.3.6 Modeling of Jf overestimation and sensitivity analyses ...... 126

6.4 RESULTS ...... 126

6.4.1 gm measured by on-line carbon isotope discrimination ...... 126

6.4.2 Comparison of methodologies that estimate gm ...... 131

6.5 DISCUSSION ...... 137

6.5.1 Response of gm to CO2 and irradiance using on-line carbon isotope discrimination...... 137

6.5.2 Comparison of methodologies that estimate gm ...... 138

6.6 CONCLUDING REMARKS ...... 143

[117]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

6.1 ABSTRACT

Mesophyll conductance (gm) is an important factor that limits photosynthesis in C3 plants.

Recently, controversial results on the short-term responses of gm to CO2 and irradiance have been reported, which may be related to the methodology used. Previously, the response of gm to CO2 and irradiance was examined in sclerophyllous Banksia species using combined gas exchange and chlorophyll fluorescence measurements (Chapter 5). The aim of the present study was to re-examine this response using a second independent method, namely on-line carbon isotope discrimination. On average, gm measured with the isotopic method decreased by 32% and 43% at 1500 and 200 μmol quanta m-2 s-1 -1 respectively, when CO2 increased from 400 to 1000 μmol mol air. The decline in gm -2 -1 measured with the fluorescence method with increasing CO2 at 1500 μmol quanta m s was 61%. The response of the isotopically measured gm to irradiance was curvilinear in most leaves, showing a 25% decline from 1500 to 200 μmol quanta m-2 s-1. Although the response of gm to CO2 and irradiance was similar with the two methods used, gm was always higher when estimated with the isotopic method. Analyses of the sensitivity of gm estimates to assumed parameters indicated that while methodological limitations generally do not influence the shape of the gm response to CO2 and irradiance, the absolute values of gm are sensitive to certain assumptions in both methods.

6.2 INTRODUCTION

Mesophyll conductance (gm) restricts C3 photosynthesis by limiting CO2 diffusion from the substomatal cavity to the sites of carboxylation, resulting in a Cc (chloroplastic CO2 concentration) that is lower than Ci (CO2 concentration in the intercellular air spaces).

Initially, it had been assumed that the drawdown Ci-Cc was negligible and could be ignored

(Farquhar et al., 1980). However, with the development of methods to measure gm, it has been revealed that the mesophyll accounts for more than 40% of the CO2 drawdown between the atmosphere and the sites of carboxylation (Warren, 2007; Flexas et al., 2008). gm is arguably the most complex component of total leaf conductance from the atmosphere to the chloroplasts. Mesophyll resistance, rm (the reciprocal of gm), includes a series of diffusion barriers from gaseous to liquid phase: the intercellular air spaces, the cell wall,

[118]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

the plasmalemma, the cytosol, the chloroplast envelope and chloroplast stroma (Pons et al.,

2009). Very few studies have experimentally separated the gaseous (gias) and liquid (gliq) phases of gm (Parkhurst and Mott, 1990; Genty et al., 1998). Parkhurst and Mott (1990) found an increase in CO2 assimilation rate at a given Ci in helox (air where nitrogen has been replaced by helium) as compared with normal air, concluding that gias is not infinite and may significantly influence gm, being more important in hypostomatous leaves. Genty et al. (1998) showed that helox did not alter CO2 assimilation rate at a given Cc, arguing that gliq is the main determinant of gm, which has also been supported by other investigators (von Caemmerer and Evans, 1991; Aalto and Juurola, 2002; Sharkey et al., 2007; Warren,

2007). Both gias and gliq are determined by the structure of the mesophyll. The length, diameter and tortuosity of the diffusion pathway can affect gias (Parkhurst, 1994; Evans and von Caemmerer, 1996), whereas mesophyll traits such as mesophyll cell surface area and chloroplast surface area exposed to intercellular air spaces (Evans et al., 1994; Evans and Loreto, 2000), chloroplast rearrangements (Sharkey et al., 1991; Tholen et al., 2007) and cell wall thickness (Terashima et al., 2006) can impact on gliq. With evidence accumulating that gm responds to environmental factors, the importance of gliq for gm is highlighted. gm responds in the short-term to soil water availability, salinity, temperature (Warren, 2007;

Flexas et al., 2008) and more recently, to CO2 concentration (Flexas et al., 2007; Hassiotou et al., 2009a/Chapter 5; Vrábl et al., 2009; Yin et al., 2009) and irradiance (Flexas et al., 2007; Hassiotou et al., 2009a/Chapter 5).

Although the mechanisms behind the regulation of gm by the environment are still unclear, attention has focussed on membrane channels that facilitate water movement, the so called aquaporins (Maurel, 1997). Certain aquaporins located in the plasma and chloroplast envelope membranes have been shown to be permeable to CO2 (Uehlein et al., 2003, 2008); hence the term cooporins has been coined (Terashima et al., 2006). Research on cooporins using transgenic plants indicates that they can influence gm and photosynthesis (Terashima and Ono, 2002; Üehlein et al., 2003; Hanba et al., 2004; Flexas et al., 2006; Üehlein et al., 2008; Miyazawa et al., 2008). Altered activity of cooporins in response to environmental changes may provide an explanation for the rapid short-term responses of gm.

[119]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

Nevertheless, fundamental questions relating to the accuracy and implications of the methodologies used to infer gm are still unanswered, introducing uncertainty about the observed values and responses of gm to the environment. Three principal techniques have been employed to estimate gm, all using gas exchange measurements, alone (“curvature method”; Ethier and Livingston, 2004), or in combination with chlorophyll fluorescence (“fluorescence method”; Harley et al., 1992) or carbon isotope discrimination measurements (“isotopic method”; Evans et al., 1986). The curvature method is usually preferred only when no other alternative is available (Pons et al., 2009), thus the fluorescence and isotopic methods have been most commonly used, especially the former, for which the necessary equipment is more widely available (Evans, 2009). Due to the need to obtain accurate estimates of gm and since the isotopic method is considered to be less sensitive to errors and more reliable (Pons et al., 2009), more studies using the isotopic method are now emerging. Both the isotopic and the fluorescence methods rely on a number of assumptions, some shared between them, and are subject to sources of error, to which gm is shown to be sensitive (Pons et al., 2009). These assumptions and/or errors may lead to apparent responses of gm to environmental changes (Evans, 2009; Pons et al., 2009).

Although gm estimates from the two methods sometimes agree (Pons et al., 2009), this is not always the case, especially at low CO2 concentrations, where the limitations of the techniques are greater (Vrábl et al., 2009). Flexas et al. (2007), Hassiotou et al.

(2009a)/Chapter 5, Vrábl et al. (2009) and Yin et al. (2009) observed a decrease in gm with increasing CO2 concentration in a number of species. These studies used the fluorescence method, except for Flexas et al. (2007), who used the isotopic as well as the fluorescence method in tobacco, and Vrábl et al. (2009), who used both methods in Helianthus annuus.

In contrast, Tazoe et al. (2009) did not observe any response of gm to CO2 concentration in wheat using the isotopic method. Similarly, the observed response of gm to irradiance has been variable. Flexas et al. (2007), Hassiotou et al. (2009a) (Chapter 5) and Yin et al.

(2009) found a decrease in gm with decreasing irradiance in tobacco, in seven Banksia species and in wheat, respectively, using the fluorescence method. In contrast, Tazoe et al.

(2009), using the isotopic method, did not find any response of gm to irradiance in wheat.

Although it is possible that the response of gm to CO2 and/or irradiance may be species- [120]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

dependent, this cannot explain the discrepancy observed in wheat between the studies of Tazoe et al. (2009), who used the isotopic method, and Yin et al. (2009), who used the fluorescence method. This discrepancy may be due to the different methodology, cultivar or experimental protocol.

More experimentation is needed to resolve the controversy in the estimates of gm obtained with the two methods and in the response of gm to CO2 and irradiance. The different responses of different species may be indicative of genetic variability, which should be examined more rigorously (Flexas et al., 2009). The simultaneous use of two independent methods can increase confidence in the estimated gm (Pons et al., 2009) and its response to

CO2 and irradiance, and give further insight into the limitations associated with each method. In Chapter 5, the response of gm to CO2 and irradiance was investigated in sclerophyllous Banksia species using the fluorescence method (Hassiotou et al., 2009a/Chapter 5). This study focussed on the isotopic method in three Banksia species and compared the results with those obtained using the fluorescence method.

6.3 MATERIALS AND METHODS

6.3.1 Plant material and growth conditions Three- to four-year old plants of three Banksia species (B. integrifolia, B. serrata and B. solandri) were used. These species were chosen on the basis that they display diverse leaf structure, which is ideal in the investigation of methodological limitations in estimating gm and their relation to leaf structure. The plants, growing outdoors in Canberra (Australia) in 10-L pots containing a mixture of grey sand and potting mix, were transferred a week prior to measurements (March-April 2009) to a controlled-temperature greenhouse (25 oC day/20 oC night).

6.3.2 Gas exchange and carbon isotope discrimination measurements Concurrent gas exchange and carbon isotope discrimination measurements were carried out in young fully expanded leaves placed in the 6 cm2 chamber of a LI-6400 open gas exchange system and illuminated by a red-blue LED light source (Li-Cor, Lincoln, NE, USA). For on-line carbon isotope discrimination measurements, gas exchange was coupled to a tunable diode laser (TDL) absorption spectrometer (TGA100A, Campbell Scientific,

[121]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

13 12 Inc., USA), which measures absolute concentrations of CO2 and CO2 in air samples

(Bowling et al., 2003). Input gases (N2 and O2) were mixed using mass flow controllers. The gas stream was divided with part of the gas being used for calibration of each measurement cycle. Another part of the gas mixture was fed to the inlets of two LI-6400 consoles which added pure CO2. The airstreams entering and leaving the leaf chamber were subsampled by the TDL to measure the isotopic composition (Barbour et al., 2007). The flow rate through the leaf chamber was maintained at 200 μmol s-1. Tests showed that increased flow rate of 400 μmol s-1 did not affect the estimation of mesophyll conductance. o Leaf temperature was 25-27 C and the O2 concentration of the air was 2% throughout the measurements. Each 4-min measurement cycle measured 12 gases in a sequence consisting of a zero gas, six CO2-concentration calibration steps, a reference gas (from a compressed air cylinder), then reference and sample gases from the two LI-6400s. Each gas was sampled for 20 s with the last 10 s being averaged for the calculations. The gas reached 99.95% of its true value at this time, with the flow rate of 150 ml min-1 and 20 mbar pressure in the TDL.

CO2 and light response curves were generated from 4-11 leaves per species, with at least 7 measurements taken at each condition. Each leaf was initially left in the chamber to stabilise and fully open its stomata at ambient CO2 and O2 concentration and at the photosynthetic photon flux density (PPFD) of 1500 μmol quanta m-2 s-1. Subsequently, the -1 chamber was fed with 2% O2 and measurement started either at low (200 μmol mol ) or -1 high (1000 μmol mol ) CO2, with stepwise increase or decrease. Tests showed that the direction of CO2 response did not affect the measurements. PPFD was then changed to 200 -2 -1 μmol quanta m s and a CO2 response was generated again. For some leaves, full light response curves were generated, with stepwise measurements between 1500 and 200 μmol quanta m-2 s-1. To minimise the probability of leakage, the leaf chamber was sealed with flexible modelling compound (Terastat). No leaks were found using the air-blowing method (see manual “Using the LI-6400”, LI-COR Biosciences of LI-6400-40, Li-Cor, Lincoln, NE, USA). Leaks through the gas exchange gasket were also tested by comparing dark -1 -1 respiration rates at low (200 μmol mol ) and high (1000 μmol mol ) CO2 concentration at the end of the light and CO2 responses (Fig. 1). The small differential between sample and [122]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

12 reference CO2 was insignificant compared with that present in the light during isotopic measurements.

0.0 -0.5 -1.0 -1.5 -2.0 -2.5 -3.0 -3.5

C] Reference - Sample (ppm) -4.0

12 [ 0 200 400 600 800 1000 1200 12 [ C] Sample (ppm)

Fig. 1. Two examples of testing for cuvette leakage at two extreme CO2 concentrations (200 and -1 12 1000 μmol mol ). Relationship between reference minus sample concentration of CO2 and sample 12 concentration of CO2 in the dark, in two leaves of Banksia integrifolia (filled and open circles).

Measurements were conducted at 2% O2 in the LI-6400 chamber using a TDL absorption spectrometer.

6.3.3 Estimation of gm by on-line carbon isotope discrimination Carbon isotope discrimination (Δ) during gas exchange was calculated according to Evans et al. (1986) as:

13 13 ( Co Ce ) Δ 13 13 13 (1) 1000 Co ( Co Ce ) where δ denotes carbon isotope composition with respect to the standard Pee Dee

Belemnite, ξ = Ce/(Ce – Co), and Ce and Co are the CO2 concentrations of dry air entering and leaving the leaf chamber, respectively. The boundary layer conductance in the chamber was high, thus it was ignored so that Δ also equals:

[123]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

* Ca Ci Ci Cc Cc eRd k fΓ Δ a ai b (2) Ca Ca Ca Ca where Ca, Ci and Cc are the CO2 concentrations of the atmosphere, the intercellular * airspaces and the sites of carboxylation, respectively. Rd is respiration in the light and Γ is the photo-compensation point in the absence of mitochondrial respiration. a is discrimination due to diffusion through stomata (4.4‰), ai is discrimination during hydration and diffusion through water (1.8‰), b is discrimination in the carboxylation reaction by Rubisco and PEPC, and the value of 30‰ was used (Evans et al., 1994). e describes the discrimination during day respiration and when the isotopic signature of the source differs from that during current photosynthesis, e should be replaced by e′:

13 13 e e Ctank Catmosphere (3)

13 13 δ Catmosphere was -8‰ and δ Ctank was either approximately -24‰ or -5‰. f in Eqn 2 is the discrimination during photorespiration (11.6‰; Lanigan et al., 2008), and k is the carboxylation efficiency. k = Vc/Cc, with Vc being the rate of RuBP carboxylation: Vc = (A * + Rd)/(1 – Γ /Cc) (von Caemmerer and Farquhar, 1981). Moreover, Ci – Cc = A/gm, with A being the CO2 assimilation rate per unit leaf area. Thus, Eqn 2 can be written as:

* * Ci A e Rd (Ci A gm Γ ) fΓ Δ a (b a) (b ai ) (4) Ca gmCa (A Rd )Ca Ca

For each measurement, Eqn 4 and Δ calculated from Eqn 1 were used to infer gm as: A e R (A C ) (b a ) d a i C A R g a d (5) m C e R (C Γ * ) fΓ * Δ a (b a) i d i Ca (A Rd )Ca Ca * If it is assumed that Ci=Cc and the terms involving Rd and Γ are ignored, Eqn 2 becomes:

(b a)Ci Δi a (6) Ca where Δi is the expected discrimination assuming gm is infinite. From Eqn 4 and 6, the deviation in observed discrimination (Δ) from expected discrimination under the assumption that mesophyll conductance is infinite (Δi) can be calculated: fΓ * A Δi Δ (b ai ) (7) Ca gmCa [124]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

where ε is the respiratory term: e R (C A g Γ * ) d i m (8) (A Rd )Ca

6.3.4 Gas exchange and chlorophyll fluorescence measurements Concurrent gas exchange and chlorophyll fluorescence measurements were carried out for some of the same B. integrifolia and B. solandri leaves used in the on-line carbon isotope discrimination measurements. For each leaf, two sets of measurements were taken using a LI-6400 open gas exchange system with an integrated 2 cm2 chamber fluorometer (LI-

6400-40, Li-Cor, Lincoln, NE, USA): (a) the CO2 response of gas exchange and chlorophyll fluorescence measured simultaneously, at ambient O2 concentration and at -2 -1 1500 μmol quanta m s ; and (b) the CO2 response of gas exchange and chlorophyll -2 -1 fluorescence measured simultaneously at 2% O2 and at 1500 μmol quanta m s . The latter was used to calibrate the relationship between the rate of electron transport estimated from chlorophyll fluorescence and that obtained from gas exchange measurements, as described previously (Hassiotou et al., 2009a/Chapter 5).

Measurements were done as described in Hassiotou et al. (2009a) (Chapter 5). The CO2 response of photosynthesis and chlorophyll fluorescence measurements were started at a Ca -1 -1 of 380 μmol mol , then Ca was decreased to 100 μmol mol and subsequently stepwise increased to 1000 μmol mol-1. Leaf temperature was held between 23-24oC. At the end of the measurements, all leaves were collected, leaf area (leaf area meter LI-300A, Li-Cor, Lincoln, NE, USA) and leaf dry mass (drying was done at 70oC for three days) were measured, and leaf dry mass per area (LMA) was calculated.

6.3.5 Estimation of gm using the Variable J method The “variable J method” (Harley et al., 1992), based on simultaneous measurements of gas exchange and chlorophyll fluorescence, was employed to obtain estimates of gm for each -2 -1 point of the A-Ci curves at 1500 μmol quanta m s . Initially, the A-Ci curves were converted to A-Cc curves using the equation:

Γ [Jf 8(A Rd )] Cc (9) Jf 4(A Rd )

[125]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

where Jf is electron transport rate obtained from chlorophyll fluorescence. gm was then calculated as: A gm (10) Ci Cc

For the calculation of gm using either the isotopic (gm,i, Eqn 5) or the fluorescence method

(gm,f, Eqn 8), Rd was approximated by measuring respiration in the dark for each leaf. For * -1 gm,f, a Γ of 36.9 μmol mol (Brooks and Farquhar, 1985), adjusted using the temperature * dependence at 21% O2 stated by the above authors, was used. For gm,i, Γ was initially calculated assuming 21% O2, and was then converted to the respective value at 2% O2, * * using the relationship: Γ 2% = Γ 21% × (2/21). For the temperature corrections, the average temperature of the leaf during the construction of the A-Ci curve was used.

6.3.6 Modelling of Jf overestimation and sensitivity analyses The likely effects of an overestimated electron transport rate obtained with the fluorescence method (Jf) on both the absolute values of gm and its responses to CO2 and irradiance were investigated in a total of 25 leaves of seven Banksia species examined here as well as in

Hassiotou et al. (2009a) (Chapter 5). It was assumed that the low-O2-calibrated Jf calculated with the fluorescence method was overestimated by 10%, thus it was compared with a Jf 0.9 times this value. Based on this new Jf, Cc and gm were re-calculated using Eqn

7 and 8, respectively, and the new gm estimates and their response to CO2 and irradiance were compared with the original estimates. To test the sensitivity of the isotopic method to the assumed parameters for fractionation associated with Rubisco and PEPC (b), day respiration (e) and photorespiration (f) as well as Γ*, b was varied between 27-33‰ and e, f and Γ* were varied between half and double the values used in the original estimation of gm (see above).

6.4 RESULTS

6.4.1 gm measured by on-line carbon isotope discrimination -2 -1 At 2% O2, CO2 assimilation rate (A) was lower at 200 μmol quanta m s compared with 1500 μmol quanta m-2 s-1 (Fig. 2). At both PPFDs, A showed the typical hyperbolic relationship with CO2 concentration. At both PPFDs and in all leaves examined, the RuBP [126]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

30 A Fig. 2. Response of net CO2 assimilation rate (A)

) 25

-1

s to intercellular CO2 concentration (Ci) -2 20

m 2 determined with the on-line carbon isotope 15 discrimination method (Evans et al., 1986).

mol CO 10

( Measurements were taken at 2% O2 and at 1500 A 5 μmol quanta m-2 s-1 (filled circles) and 200 μmol 0 -2 -1 B -1 quanta m s (open circles), in a leaf of Banksia 30 Ci ( mol mol )

)

-1

s 25 integrifolia (A), B. serrata (B) and B. solandri

-2

m 2 20 (C). 15 mol CO 10

(

A 5

0 C -1 30 Ci ( mol mol ) )

-1

s 25 -2

m 2 20

15

mol CO 10

(

A 5 0 0 200 400 600 800 1000 C ( mol mol-1) i

- regeneration-limited region was reached within the highest Ca considered (1000 μmol mol 1 ). A did not decrease at high CO2 concentration in any of the leaves measured. The rates of -2 -1 CO2 assimilation were high (12-27 and 7-17 μmol CO2 m s at high and low irradiance, respectively, at ambient CO2 concentration, with an average rate of 20±0.8 and 10±0.5 -2 -1 μmol CO2 m s , at each irradiance respectively), but they were measured under non- photorespiratory conditions. As photosynthetic rates increased with increasing irradiance, the deviation between observed and predicted discrimination (Δi – Δ) became greater, consistent with the expectation that the drawdown Ci-Cc increased. The deviation also increased with decreasing Ca. Generally, the variability observed in Δi – Δ was greater at

[127]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

-1 high CO2 concentrations (1000 μmol mol ) and low photosynthetic photon flux density (PPFD) (Fig. 3). A increased with increasing PPFD reaching saturation at 1500 μmol -2 -1 -1 -1 quanta m s at 400 μmol CO2 mol air, but not at 1000 μmol CO2 mol air (Figs 4A and 4B).

12 A Fig. 3. Relationship between the 10

)

o deviation in observed discrimination (Δ)

% 8 from expected discrimination under the 6 assumption that mesophyll conductance

epsilon ( epsilon C 200 mol mol-1 4 a is infinite (Δi), minus the term associated -1 i Ca 400 mol mol -1 with respiration in Equation 4 2 Ca 750 mol mol C 1000 mol mol-1 * a e R (C A g Γ ) 0 ( d i m ), and the 8 B (A Rd )Ca

) o product of A/Ca (A: net CO2 assimilation % 6 rate; Ca: ambient CO2 concentration).

4 A, B and C show an example leaf for

epsilon ( epsilon Banksia integrifolia, B. serrata and B.

i 2 solandri, respectively. Measurements were conducted at 2% O , at four 0 2 8 C photosynthetic photon flux densities

)

o (200, 500, 1000 and 1500 μmol quanta

% 6 -2 -1 -1 m s ) and four Ca: 200 μmol mol (filled circles), 400 μmol mol-1 (open 4

epsilon ( epsilon circles), 750 μmol mol-1 (open squares)

i 2 and 1000 μmol mol-1 (filled squares).

0 0.00 0.02 0.04 0.06 0.08 0.10 -2 -1 A/Ca (mol m s )

[128]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

40 35 A B

)

-1

s 30

-2

m 25

2 20 15

mol CO

( 10

A 5 0 C PPFD ( mol quanta m-2 s-1) D PPFD ( mol quanta m-2 s-1)

) 0.4

-1

s

-2

m 0.3

2

0.2

(mol CO

L

g 0.1

0.0 0.30 E PPFD ( mol quanta m-2 s-1) F PPFD ( mol quanta m-2 s-1)

) 0.25

-1

s

-2 0.20

0.15

(mol m

m 0.10

g

0.05

0.00 0 500 1000 1500 0 500 1000 1500 2000 -2 -1 -2 -1 PPFD ( mol quanta m s ) PPFD ( mol quanta m s )

Fig. 4. Response of net CO2 assimilation rate (A), mesophyll conductance (gm) and leaf conductance (gL) to photosynthetic photon flux density (PPFD) determined with the on-line carbon isotope discrimination method (Evans et al., 1986). Measurements were conducted at 2% O2 and -1 -1 CO2 concentrations of 400 μmol mol (open circles) and 1000 μmol mol (filled circles), in a leaf of Banksia integrifolia (A, C, E) and B. serrata (B, D, F).

[129]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

0.5 0.4 A B

0.4 )

-1

) 0.3

s

-1

-2

s

-2 0.3 m

2 0.2 0.2

(mol m

m

g

(mol CO

0.1 L

0.1 g

0.0 0.0 C -1 D -1 Ci ( mol mol ) Ci ( mol mol )

)

0.3 0.15 -1

)

s

-1

-2

s

-2

m

2 0.2 0.10

(mol m

m

(mol CO g 0.05 0.1 L

g

0.0 0.00 E -1 F C ( mol mol-1) 0.5 Ci ( mol mol ) i 0.3

)

-1

)

s

-1 0.4 -2

s

m

-2

0.2 2 0.3

(mol m 0.2

m

0.1 (mol CO

g

L 0.1 g

0.0 0.0 0 200 400 600 800 1000 0 200 400 600 800 1000 C ( mol mol-1) C ( mol mol-1) i i

Fig. 5. Response of mesophyll conductance (gm) and leaf conductance (gL) to intercellular CO2 concentration (Ci) determined with the on-line carbon isotope discrimination method (Evans et al., 1986). Measurements were taken at 1500 μmol quanta m-2 s-1 (filled circles) and 200 μmol quanta m-2 s-1 (open circles), in a leaf of Banksia integrifolia (A, B), B. serrata (C, D) and B. solandri (E, F). Since in these species stomata are positioned in depressions of the leaf epidermis called stomatal crypts, gL shown includes the conductance through both the stomata and the stomatal crypts. The response of gL to Ci reflects the response of stomatal conductance to Ci.

Leaf conductance (gL) tended to decline with increasing CO2 in some leaves, but in others it remained relatively stable across the CO2 range examined (Fig. 5). gL increased with increasing PPFD at both CO2 concentrations examined (Figs 4C and 4D). Mesophyll

[130]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

conductance (gm) declined with increasing CO2 concentration (Fig. 5). On average, this -1 decline was 32% and 43% from 400 to 1000 μmol CO2 mol air at 1500 and 200 μmol -2 -1 quanta m s , respectively. The response of gm to irradiance was variable even within a species (or a single plant): in most leaves, an initial increase in gm with PPFD was followed by saturation or small decrease in gm at high PPFD (e.g. Fig. 4E), whilst in a few - leaves gm continually increased with increasing PPFD (e.g. Fig. 4F at Ca of 400 μmol mol 1 -2 -1 ). At ambient Ca, gm was higher at 1500 than at 200 μmol quanta m s in all but two of the 24 leaves measured (Fig. 6). On average, gm declined by 25% and 37% from 1500 to -2 -1 -1 200 μmol quanta m s at 400 and 1000 μmol CO2 mol air, respectively.

6.4.2 Comparison of methodologies that estimate gm

The responses of CO2 assimilation rate to CO2 and irradiance were typical hyperbolic.

Higher rates of CO2 assimilation were measured at 2% O2 than at 21% O2, due to suppression of photorespiration, which distinguishes the A-Ci curves of the isotopic method from those of the fluorescence method in Figs 7A and 7C. gm decreased with increasing

CO2 concentration with both the isotopic and the fluorescence method used to estimate gm

(Figs 5 and 7, Table 1). The percentage decline at high CO2 (from 400 to 1000 μmol CO2 -1 mol ) of gm estimated with the fluorescence method was almost double that obtained with the isotopic method.

0.6 Fig. 6. Relationship between gm -2 -1 0.5 measured at 1500 μmol quanta m s

) -1 (g ) and that measured at 200 μmol s m1500

-2 0.4 -2 -1 quanta m s (gm200), at ambient CO2 0.3 concentration, using the on-line carbon

(mol m (mol 0.2 isotope discrimination method (Evans

m1500 g 0.1 et al., 1986). All individual leaves

0.0 measured (24) in three species 0.0 0.1 0.2 0.3 0.4 0.5 0.6 (Banksia integrifolia, B. serrata and B. g (mol m-2 s-1) m200 solandri) are included.

[131]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

30 0.5 A B

) 25 0.4

-1

)

s

-1

-2

20 s

m 0.3 -2

2 15 0.2

(mol m

mol CO

10 m

g

(

A 5 0.1

0 0.0 C -1 D -1 Ci ( mol mol ) Ci ( mol mol ) ) 25 0.4

-1

)

s

-1 -2 20

s

m 0.3 -2

2 15 0.2

(mol m

mol CO 10

m

g

(

A 5 0.1

0 0.0 0 200 400 600 800 0 200 400 600 800 -1 C ( mol mol ) C ( mol mol-1) i i

Fig. 7. Comparison of two independent methods that estimate mesophyll conductance (gm).

Response of net CO2 assimilation rate (A) and gm to intercellular CO2 concentration (Ci), measured either using the on-line carbon isotope discrimination method (Evans et al., 1986) at 2% O2 (filled circles) or the fluorescence method (Harley et al., 1992) at 21% O2 (open circles) at 1500 μmol quanta m-2 s-1. (A, B) Banksia integrifolia, (C, D) B. solandri.

Although the direction of the response of gm to CO2 was similar with the two methods used to derive gm, the absolute values of gm differed between the two methods (Figs 7B and 7D).

In particular, gm was always higher when estimated with the isotopic method. This is also -2 -1 shown in Fig. 8, where gm at ambient Ca and 1500 μmol quanta m s is compared between the two methods in three Banksia species, including the estimates of gm for these species from Hassiotou et al. (2009a) (Chapter 5) using the fluorescence method. The estimates of gm from the two independent fluorescence measurements were in good agreement, but were half that estimated using the isotopic method.

[132]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

Table 1. Mesophyll conductance (gm) estimated with the on-line carbon isotope discrimination method (Evans et al., 1986) at two photosynthetic photon flux densities (PPFD: 200 and 1500 μmol -2 -1 -1 quanta m s ) and CO2 concentrations (Ca: 400 and 1000 μmol mol ), and with concurrent gas exchange and chlorophyll fluorescence measurements (Harley et al., 1992) at 1500 μmol quanta m- 2 -1 s , in Banksia integrifolia, B. serrata and B. solandri. The last row shows the average gm for these three B. species with each method and at each CO2 and irradiance, for a total of 24 leaves for the -2 -1 -1 isotopic method (or 20 leaves at 200 μmol quanta m s and Ca of 1000 μmol mol ) and 3-4 leaves for the fluorescence method.

Method Isotopic Isotopic Fluorescence

PPFD (μmol 200 1500 1500 quanta m-2 s-1)

-1 Ca (μmol mol ) 400 1000 400 1000 400 1000 B. integrifolia 0.26 ± 0.01 0.14 ± 0.02 0.32 ± 0.01 0.23 ± 0.02 0.13 ± 0.00 0.06 ± 0.01

B. serrata 0.16 ± 0.02 0.12 ± 0.02 0.25 ± 0.03 0.18 ± 0.02 - -

B. solandri 0.29 ± 0.02 0.16 ± 0.02 0.39 ± 0.02 0.24 ± 0.02 0.15 ± 0.06 0.05

Average 0.24 ± 0.01 0.14 ± 0.01 0.33 ± 0.01 0.22 ± 0.01 0.14 ± 0.02 0.05 ± 0.01

Sensitivity analysis of the data obtained with the fluorescence method showed that a 10% -2 - overestimation of Jf would cause 10-60% underestimation of gm at 1500 μmol quanta m s 1 -2 -1 and 5-40% at 500 μmol quanta m s across the CO2 range examined (Figs 9A and 9B) in Banksia species. For B. solandri, gm would be underestimated by on average, 36% at 1500 μmol quanta m-2 s-1 and 12% at 200 μmol quanta m-2 s-1. The shape of the response of gm to CO2 did not change much at low irradiance, but a dramatic change was observed at high irradiance where gm initially increased with increasing CO2 and then stabilised, instead of decreasing with CO2. In the B. integrifolia leaf, gm would be underestimated by on average, 14% at 1500 μmol quanta m-2 s-1 and 39% at 500 μmol quanta m-2 s-1. In this case, the dramatic change in the response of gm to CO2 was observed at low irradiance where gm initially decreased and then increased with CO2.

[133]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

0.5 Isotopic method Fluorescence method Fluorescence method 0.4 Hassiotou et al. (2009a)

)

-1

s

-2 0.3

0.2

(mol m

m

g 0.1

0.0 B. integrifolia B. solandri B. serrata

Fig. 8. Comparison of two independent methods that estimate mesophyll conductance (gm) at -2 -1 ambient CO2 concentration and 1500 μmol quanta m s . Black bars show gm estimated using the on-line carbon isotope discrimination method (Evans et al., 1986) (average of 13 leaves in Banksia integrifolia, six leaves in B. solandri and five leaves in B. serrata). Light grey bars show gm estimated using the fluorescence method (Harley et al., 1992) (average of two leaves per species).

Dark grey bars show gm estimated previously in these species by Hassiotou et al. (2009a) (Chapter 5) using the fluorescence method.

In a total of 20 leaves of seven B. species examined here and in Hassiotou et al. (2009a)

(Chapter 5), a 10% overestimation in the calibrated Jf resulted in, on average, 18±2% -2 -1 -2 -1 underestimation of gm at 1500 μmol quanta m s and 26±3% at 500 μmol quanta m s -1 -1 over a range of Ca from 200 to 800 μmol mol (Ci 150-450 μmol mol ) (Fig. 9C). At a Ca -1 -1 of 200 μmol mol (Ci=102-172 μmol mol ), a gm underestimation of, on average, 21±3% and 18±2% was found at high and low irradiance, respectively. At a Ca of 380-400 μmol -1 -1 mol (Ci=187-350 μmol mol ), the respective underestimation was 16±3% and 24±3% and -1 -1 at a Ca of 680-700 μmol mol (Ci=300-577 μmol mol ), 18±2% and 26±7%. At low irradiance, the underestimation of gm tended to increase with Ca, whereas at high irradiance no general trend was observed.

[134]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

0.20 A Fig. 9. Effect of an assumed 10% overestimation of electron transport

) 0.15

-1 rate on mesophyll conductance (gm)

s -2 as determined by the fluorescence 0.10 method (Harley et al., 1992). (A, B)

(mol m

m

g 0.05 Response of gm to intercellular CO2

concentration (Ci) in a leaf of (A) 0.00 Banksia solandri and (B) B. 0 200 400 600 800 1000 integrifolia. Circles show gm -1 Ci ( mol mol ) estimated with measured electron

0.5 B transport rate and triangles gm estimated with „true‟ electron

) 0.4

-1

s transport rate if we assume that the

-2 0.3 measured rate is overestimated by 10%. Filled symbols show the

(mol m 0.2

m -2 - g response at 1500 μmol quanta m s 0.1 1, whereas open symbols in (B) show 0.0 the response at 500 μmol quanta m-2 0 100 200 300 400 500 600 700 s-1. (C) Average underestimation of C ( mol mol-1) i g (%) across a C range of 150-450 50 m i PPFD 1500 -1 -1 C μmol mol (Ca 200-800 μmol mol ) m PPFD 500

g 40 at 1500 μmol quanta m-2 s-1 (black -2 -1 30 bars) and 500 μmol quanta m s (grey bars) in seven Banksia species 20 examined previously by Hassiotou

10 et al. (2009a) (Chapter 5). Species % underestimation of % underestimation are arranged in order of increasing 0 leaf dry mass per area (LMA 134- 269 g m-2).

B. repens B. serrata

B. solandri

B. paludosa

B. attenuata

B. elderiana B. integrifolia

[135]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

13 The sensitivity analysis of the isotopic estimates of gm to the assumed parameters of C fractionation by Rubisco and PEPC (b), during day respiration (e) and photorespiration (f) as well as of Γ* is shown in Fig. 10. Halving of e and halving or doubling of f and Γ* did not cause an average change in gm greater than 3% at high irradiance and 6% at low irradiance. Doubling of e caused an average decrease in gm of 8% at high irradiance and

19% at low irradiance. The greatest change occurred at very high or very low Ca.

Decreasing b had the greatest effect on gm, which increased by on average, 84% and 100% at high and low irradiance, respectively, from a b=30‰ to a b=27‰. Increasing b from 30 to 33‰ caused on average, a 30% decrease in gm. The variations of the above parameters did not change the shape of the gm response to CO2 or irradiance (Fig. 10).

0.8 0.8 A b = 27 B b = 27 0.7 b = 28 0.7 b = 28 b = 29 b = 29 b = 31 b = 31 ) 0.6 0.6 b = 33 -1 b = 33

s gm original g original 0.5 m -2 0.5 (b = 30) (b = 30) 0.4 0.4 0.3

(mol m (mol 0.3

m g 0.2 0.2 0.1 0.1 0.0 0.0 e = -8 e = -8 C -1 D -1 e = -32 0.30 e = -32 0.5 Ci ( mol mol ) Ci ( mol mol ) f = 5.8 f = 5.8 f = 23.2 f = 23.2 ) 0.25 * * -1 G = 1.79 G = 1.79 0.4 * s G = 7.16 G* = 7.16 -2 0.20 gm original gm original 0.3 0.15

(mol m (mol 0.2 m 0.10

g 0.1 0.05

0.0 0.00 0 200 400 600 800 0 200 400 600 800 1000 -1 C ( mol mol-1) C ( mol mol ) i i

Fig. 10. Sensitivity analysis of the parameters assumed in the estimation of mesophyll conductance

(gm) with the on-line carbon isotope discrimination method (Evans et al., 1986) in a leaf of Banksia integrifolia. The response of gm to intercellular CO2 concentration (Ci) is shown at 1500 μmol

[136]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

-2 -1 -2 -1 quanta m s (A, C) and at 200 μmol quanta m s (B, D) for gm estimated using the parameters stated in the Materials and Methods (“gm original” – filled circles and thick line) or using variations of b (fractionation by Rubisco and PEPC) (A, B), e (fractionation during respiration in the light), f (fractionation during photorespiration) and Γ* (photo-compensation point in the absence of mitochondrial respiration) (C, D). b was varied in increments between 27‰ and 33‰, with a value of 30‰ used in the original gm; e was calculated according to Wingate et al. (2007) and was varied between half and double that value; f was assumed to be 11.6‰ (Lanigan et al., 2008) and was varied between half and double that value; Γ* was assumed to be 3.51 μmol mol-1 at o 2% O2 at 25 C (Brooks and Farquhar, 1985) and was varied between half and double that value.

6.5 DISCUSSION

Long-term (days to weeks) responses of gm to environmental variables, such as water stress and temperature, have been well established (Warren, 2007). Recently, gm was shown to rapidly respond to changes in CO2 concentration and irradiance (Flexas et al., 2007). Since the study of Flexas et al. (2007), the response of gm to these factors was examined in more species using either one or both of the two independent methods known to date that provide an estimate of gm. Interestingly, controversial results have been reported. In Chapter 5, the response of gm to CO2 and irradiance was examined in Banksia species of varying leaf structure using the fluorescence method (Hassiotou et al., 2009a/Chapter 5). However, given the variation in responses in the literature, the measurements were repeated estimating gm by on-line carbon isotope discrimination.

6.5.1 Response of gm to CO2 and irradiance using on-line carbon isotope discrimination -1 gm declined in response to increasing CO2 from 400 to 1000 μmol mol by on average, 32% at high irradiance (1500 μmol quanta m-2 s-1) and 43% at low irradiance (200 μmol quanta m-2 s-1) in three Banksia species. Two previous studies have examined this response using the isotopic method for estimating gm (Tazoe et al., 2009; Vrábl et al., 2009). Our results agree with those of Vrábl et al. (2009), who observed a decreasing gm with increasing CO2 in sunflower, but not with those of Tazoe et al. (2009), who found that gm was independent of CO2 in wheat.

[137]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

In 21 out of 24 leaves gm varied with irradiance. In particular, gm increased with increasing irradiance in a curvilinear way in most leaves, saturating above 1000 μmol quanta m-2 s-1, both at high and low CO2 concentrations. There was little variation in Ci with changing irradiance. The response of gm to irradiance measured isotopically was previously examined only in wheat by Tazoe et al. (2009), who reported a gm independent of irradiance, contrary to our findings in Banksia species. This may reflect phylogenetic differences in the response of gm to irradiance and CO2 between different species or methodological limitations. The latter will be discussed in the next section.

The physiological explanation of the observed response of gm to CO2 and irradiance is not clear; however, hypotheses discussed before (Flexas et al., 2007, 2008) suggest changes in the activity of aquaporins (or cooporins) induced by the CO2 concentration or the light regime. Another explanation may relate to chloroplast movements in response to irradiance. It is well established that low irradiance causes chloroplasts to move to positions where light absorption is increased (Brugnoli and Björkman, 1992; Del Hierro et al., 2000). It is possible that these movements alter the surface area of chloroplasts exposed to intercellular air spaces, and through it the liquid component of gm (gliq), causing or contributing to the decline in gm at low irradiance (Tholen et al., 2008). Although cooporins and/or chloroplast movements may explain the physiological basis of the short-term response of gm to CO2 and irradiance, it is not known if such a response has adaptive value or is an inevitable consequence of leaf functioning. It can be hypothesized that if there is a cost related to maintenance of high gm, there may be no justification for that expense when CO2 is abundant or when there is not enough light to fix it. Further studies are needed to give insight into the ecological significance and explanation of the gm response to CO2 and irradiance.

6.5.2 Comparison of methodologies that estimate gm

The two independent methods used to estimate gm (isotopic and fluorescence) showed similar responses of gm to CO2 in Banksia species. The magnitude of these responses, however, differed: a greater decline was observed at high irradiance when the fluorescence method was used (61% compared with 32% with the isotopic method). Vrábl et al. (2009) also reported a gm dependent on CO2 using both methods. This has also been observed in

[138]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

six other species (Flexas et al., 2007) and in wheat (Yin et al., 2009) using the fluorescence method. In the study of Tazoe et al. (2009), however, where the isotopic method was used in wheat, gm did not show any response to CO2.

There are fewer reports of the response of gm to irradiance. Flexas et al. (2007) and Yin et al. (2009) reported a decrease in gm with decreasing irradiance in tobacco and wheat, respectively, using the fluorescence method. In contrast, Tazoe et al. (2009) found no response in wheat using the isotopic method. For Banksia species, gm varied with irradiance when measured by either method (see also Hassiotou et al., 2009a/Chapter 5). The explanation for different responses is unknown and more comparative results in a range of species from different functional groups are needed.

Both the isotopic and the fluorescence method for estimating gm rely on a number of assumptions and possible measurement errors to which gm is sensitive (Pons et al., 2009). It is possible that these assumptions/errors are partially responsible for the apparent responses of gm to CO2 and irradiance. Estimation of gm is most sensitive at the lower (<400) and very high (>1000) Ca end. Even if we do not consider these Ca regions, the agreement of the two independent methods in the present study and in the study of Vrábl et al. (2009) gives confidence to a decreasing gm with increasing CO2 and possibly with decreasing irradiance. Nevertheless, some uncertainty is still associated with the magnitude of this response and the absolute values of gm.

Some previous studies that estimated gm using both methods in the same species have found good agreement between these estimates (Warren et al., 2004; Flexas et al., 2006, 2007). However, that has not always been the case; for example, the recent studies by Yin et al. (2009), who used the fluorescence method in wheat, and by Tazoe et al. (2009), who - used the isotopic method in the same species, estimated a gm of about 0.17 and 0.45 mol m 2 -1 -2 -1 -1 s , respectively, at 1500 μmol quanta m s and a Ca of 350 μmol mol . In this comparison, however, the discrepancy between the two studies may be due to reasons other than methodological differences, e.g. the different wheat cultivars used by each of these two studies and/or differences in the way the material was handled and measurements were -2 -1 carried out. Vrábl et al. (2009) estimated a gm of 0.38 mol m s with the fluorescence -2 -1 -2 -1 method and 0.51 mol m s with the isotopic method at 800 μmol quanta m s and a Ca -1 of 400 μmol mol in sunflower. The above correspond to a 62% and 25% decrease in gm in [139]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

wheat and sunflower, respectively, with the fluorescence method compared with the isotopic method at the same conditions. This agrees with the deviation in gm associated with the methodology used that was found in the present study. gm was always lower when estimated with the fluorescence method; on average, it was 53% lower at ambient CO2 and high irradiance, with this deviation increasing at high CO2 concentration. A 53% difference in gm between the two methods would result in 29% difference in Cc, at a Ci of 250 μmol -1 -2 -1 -2 -1 mol , A of 18 μmol m s and gm, isotopic of 0.35 mol m s . This discrepancy between the methods may be associated with false assumptions of each method, which may differently influence the absolute values of gm and/or its response to CO2 and irradiance in leaves of differing structure. Due to the assumption of the fluorescence method that the adaxial chloroplasts are representative of the whole leaf (Warren, 2007), Jf may be overestimated, especially if there are errors associated with the Jf calibration at low O2: e.g. if photorespiration has not been fully suppressed. The sensitivity analysis in the present study showed that an overestimation of Jf by 10% caused an underestimation of gm by on average, 10-60% at high irradiance and 5-40% at low irradiance, in Banksia species. It was also shown that overestimation of Jf can alter the response of gm to CO2 and irradiance. Although this was generally species- or leaf-dependent, at high irradiance the average underestimation of gm was somewhat greater than at low irradiance, where this underestimation also tended to increase with Ca. This model demonstrated how sensitive gm is to a small overestimation of

Jf and highlighted that limitations of the fluorescence method for estimating gm are likely to be associated with the lower gm values obtained herein and in previous studies with this method compared with the on-line carbon isotope discrimination method. It was hypothesised that the limitations of the fluorescence method discussed above may change with leaf structure, given that the adaxial chloroplasts may be less representative of the whole leaf depth in thicker leaves. This hypothesis was tested using the fluorescence data of Hassiotou et al. (2009a) (Chapter 5) for Banksia species of LMAs between 134-478 g m-2 and leaf thicknesses between 317-670 μm. Although it was expected that the proportional difference between uncalibrated Jf and gross photosynthesis [4×(A+Rd)] at 2%

O2 would be greater in thicker leaves, no correlation was found between this difference and leaf thickness, LMA or photosynthetic capacity in these species (data not shown). This

[140]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

suggests that there is no systematic error associated with the fluorescence method in relation to leaf thickness in Banksia species. Non-uniform Ci and Cc across the leaf (Terashima et al., 1988), however, may cause differential responses of thin and thick leaves to errors of the fluorescence as well as the isotopic method, since both methods share this assumption. This merits further attention, although at present we can only address this through modelling involving a number of assumptions.

If errors associated with the methodology used to estimate gm are of different magnitude in thicker leaves, the relationship between gm and LMA may be different than that previously presented by Flexas et al. (2008), and subsequently revised by Hassiotou et al. (2009a)

(Chapter 5). In particular, gm may not decrease with LMA to the extent previously thought. Fig. 11 shows a revised version of this relationship – including data of the present study and more data that have recently become available – where gm measured with the fluorescence method is distinguished by that measured with the isotopic method with different symbols. It is noteworthy that at the low-LMA end gm values of the two methods overlap, whereas at the high-LMA end the fluorescence estimates of gm are clearly lower than the isotopic, which is particularly evident for the Banksia species. To complicate matters, although the isotopic method is generally considered to be more robust, it is not certain that the isotopic estimates of gm are entirely accurate. This method shares with the fluorescence method the assumption of Ci and Cc uniformity across the leaf

(Pons et al., 2009), and that gm is independent of O2. This assumption is necessary in order to conduct measurements at low O2 concentration where the carbon isotope signal is maximised, for higher precision (if a mass spectrometer membrane inlet system is used) or where the contribution of photorespiration is minimised (if a TDL system is used). If, however, gm responds to O2, and in particular, if it increases with decreasing O2, as it does with CO2, then it is likely that the gm estimates obtained with this method are somewhat overestimated. Since the assumption of a gm independent of O2 is also used in the calibration of the fluorescence method, this is a potential error of the fluorescence as well as the isotopic method. A significant flow of electrons to O2 or a metabolic pathway that cannot be seen under low O2 during the fluorescence calibration routine could influence the gm estimation. The above emphasise the need to examine if and how gm responds to O2 concentration.

[141]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

0.6

0.5

)

-1 0.4

s

-2 0.3

(mol m 0.2

m

g 0.1

0.0 0 100 200 300 400 500 600 -2 LMA (g m )

Fig. 11. Relationship between mesophyll conductance (gm) and leaf dry mass per area (LMA) in the absence of stress in different species. Circles include leaves of high-altitude plants, old leaves, shade leaves and other leaves, data that have been presented in Flexas et al. (2008), Niinemets et al. (2009), Yin et al. (2009) and Vrábl et al. (2009). Squares include Banksia species examined in this study and by Hassiotou et al. (2009a) (Chapter 5). Filled symbols show gm calculated from the on-line carbon isotope discrimination method (Evans et al., 1986), while all other symbols show gm calculated using the fluorescence method (Harley et al., 1992). Crossed squares show gm obtained with the fluorescence method by Hassiotou et al. (2009a) (Chapter 5) in B. integrifolia, B. serrata and B. solandri, the three B. species where both methods have been used.

The deviation between gm estimated from the fluorescence and isotopic methods may be associated with assuming incorrect values for parameters. Small differences in Γ*, e.g. of -1 the order of 28-40 μmol mol at 21% O2, do not significantly influence gm at Ci values between 200 and 400 μmol mol-1, as was shown by Hassiotou et al. (2009a) (Chapter 5). It * -1 is necessary to increase Γ to 50 μmol mol to obtain gm values close to those obtained with the isotopic method, but the response of gm to CO2 and irradiance was not influenced by the value assumed for Γ* (Fig. 10C). Regarding the isotopic method, a sensitivity analysis was conducted for a leaf of B. integrifolia, which showed that although the absolute values of gm may be somewhat influenced by the parameters used, especially at

[142]

CHAPTER 6: MESOPHYLL CONDUCTANCE, CO2 AND IRRADIANCE

very low and high Ca, the response of gm to CO2 and irradiance is not affected (Fig. 10). It was reasonable to conclude that the values assumed for e, f and Γ* do not significantly -1 influence the estimation of gm, at least between a Ca of 300 and 700 μmol mol . Reducing b had the greatest effect on the absolute values of gm. There are few in vivo estimates of b so it is unknown how much variation could be expected between different species.

6.6 CONCLUDING REMARKS

All studies to-date using the fluorescence method to estimate gm have found that gm decreases with increasing CO2 and decreasing irradiance. Two previous studies (Flexas et al., 2007; Vrábl et al., 2009) and the present study that used the isotopic method were consistent with the above. By contrast, Tazoe et al. (2009) did not observe any dependence of gm on CO2 or irradiance in wheat. More experimentation is needed to elucidate this discrepancy, which may be due to a number of reasons, e.g. phylogenetic and/or methodological differences. It was also shown that in the Banksia species examined gm generally shows a saturating response to irradiance. In this and previous studies comparing both methods in the same species (Flexas et al., 2007; Vrábl et al., 2009), the isotopic method yielded higher gm estimates than the fluorescence method. The fluorescence method may overestimate Jf since it samples mainly chloroplasts of the adaxial leaf side, which is poorly representative of the whole leaf depth. It was shown that a 10% overestimation of calibrated Jf can result in up to 60% underestimation of gm in Banksia leaves. Moreover, this underestimation changed with CO2 and irradiance, being more pronounced at high irradiance and high CO2. The isotopic method also has its own limitations. Although most of the assumed parameters in this method do not significantly influence the absolute values of gm, it relies on the value assumed for the fractionation by

Rubisco and PEPC. Moreover, the assumption that gm is independent of O2 may be related to the higher gm obtained with this method. Since this assumption may also influence the low-O2 calibration in the fluorescence method, studies on the response of gm to O2 concentration deserve further attention.

[143]

[144]

CHAPTER 7

GENERAL DISCUSSION

7.1 STOMATAL CRYPTS ...... 146

7.2 SCLEROPHYLLY AND GAS EXCHANGE ...... 150

7.3 MESOPHYLL CONDUCTANCE AT THE HIGH END OF THE LMA SPECTRUM ...... 153

7.4 CONCLUSIONS ...... 156

[145]

CHAPTER 7: GENERAL DISCUSSION

Leaf structure is a major determinant of CO2 diffusive conductance, and through it, of photosynthesis; however, our understanding of where the main limitations to photosynthesis reside is still poor. Important insights into the effects of leaf structure on gas exchange and photosynthesis have been gained by the study of sclerophylls (Smith et al., 1997; Evans and Vogelmann, 2006; Niinemets et al., 2009), which include extreme examples of structural properties, such as high leaf densities and thicknesses and extensive sclerification. Species that display a great diversity of leaf morphologies and anatomies, with a wide range of sclerophylly, often indicated through leaf dry mass per area (LMA), are excellent candidates for the study of the relationship between leaf structure and function. To this end, the genus Banksia (Proteaceae) was chosen and the impact of the diverse sclerophyllous characters found in this genus on CO2 diffusion and photosynthesis was examined through direct experimentation and modelling. Limitations to diffusion from the atmosphere to the chloroplast level were considered. Many Banksia species have a structural feature that represents a diffusion resistance that is not normally present: the abaxial leaf epidermis forms invaginations called “stomatal crypts” and stomata are exclusively located there. The effect of crypts on gas exchange was evaluated using two approaches, and the relationships of these structures with other sclerophyllous traits were examined, generating a new hypothesis for the role of stomatal encryption. Limitations to photosynthesis due to leaf chemistry and whole leaf anatomy were investigated. The effects of leaf anatomy on mesophyll conductance were further addressed in species covering a wide range of LMA, in response to elevated CO2 concentration and high irradiance, using two independent methods.

7.1 STOMATAL CRYPTS

Almost all the reports referring to sunken stomata, either grouped in crypts or not, consider these stomatal arrangements to be adaptations that reduce water loss through reduced leaf transpiration (Sobrado and Medina, 1980; Turner, 1994; Groom et al., 1994, 1997; Hill, 1998; Rotondi et al., 2003; Yiotis et al., 2006, 2008; Haworth and McElwain, 2008; Rossini Oliva et al., 2009). This assumption probably originates from the xeric habitat of many of the species possessing sunken stomata or stomatal crypts. However, these

[146]

CHAPTER 7: GENERAL DISCUSSION

structures also occur in species from mesic environments and rainforests (Brodribb and Hill, 1997; Waldhoff et al., 2002; Burgess and Dawson, 2004; Jordan et al., 2008). Under flooded conditions, crypts may facilitate plant survival through the retention of air bubbles around the leaf, thus allowing gas exchange between the water and the trapped air during submergence (Waldhoff et al., 2002). Stomatal recession in non-xerophytes challenges the assumed role of transpiration reduction. Previous research has shown that species with sunken stomata do not necessarily have lower leaf conductance (Lösch et al., 1982). Furthermore, a recent investigation about the interrelationship between species of the genus Banksia and their habitat conditions revealed no straightforward correlation between the presence of crypts and water availability (Jordan et al., 2008). It is therefore necessary to perform comparative analyses of the impacts of superstomatal chambers and the stomata on gas exchange to draw firm conclusions on the possible role(s) of these structures. The present study, using two novel techniques to examine gas diffusion through crypts, found no evidence linking stomatal encryption to water economy. Firstly, a simplified, one- dimensional approach was used, whereby crypt resistance was approximated by analyses analogous to electrical resistivity networks. Secondly, the three-dimensional arrangement of stomata located in crypts was considered in a more sophisticated finite-element modelling approach. Both approaches agreed in that stomata contributed significantly more than crypts to leaf resistance. At situations characteristic of water stress (low stomatal conductance and high boundary layer conductance) the effect of crypts on transpiration was negligible. Furthermore, the modelling approach used here allowed quantification of the effect of crypt trichomes, which are often considered a xerophytic feature, on gas exchange. Trichomes were found to have virtually no influence on transpiration. Thus, the few studies that have so far evaluated the effect of superstomatal chambers, compared with stomata, on gas exchange in photosynthesising stems (Gibson, 1983) or leaves (Matthews, 2003; Chapters 2 and 3), do not support a function for these structures in moderating water loss. A novel outcome of this study was the observation that crypt conductance depends not only on the detailed structural arrangement, but also on stomatal and boundary layer conductances, i.e. on the rates at which water vapour enters and leaves the crypt. It was thus concluded that estimating crypt conductance through resistance-in-series approaches is

[147]

CHAPTER 7: GENERAL DISCUSSION

a simplification, which may somewhat deviate from the actual crypt conductance, depending on the given leaf boundary layer and the dynamic stomatal conductance. Given, however, the relatively minor influence of crypts on gas diffusion, it could be argued that for some applications an approximate value of crypt resistance is sufficient, rather than an elaborate model-derived value. Since stomatal resistance represents the crucial and a more flexible limitation on transpiration, crypts may perform other roles. The observation that crypts do not represent the large diffusion resistance that was assumed to reduce transpiration, and the implication that therefore CO2 also diffuses easily through crypts, led to the idea that crypts may actually facilitate CO2 diffusion to photosynthetic cells. Thicker leaves were found to have deeper crypts, which gave rise to the hypothesis that stomatal crypts assist CO2 diffusion to well-lit actively photosynthesising palisade cells near the adaxial surface. The proposed role for crypts in facilitating diffusion assumes that diffusion is restricted by the tortuous path and narrow intercellular channels through the mesophyll in comparison with diffusion through pubescent crypts, given that diffusivity decreases in a non-linear way with decreasing porosity (Tomadakis and Sotirchos, 1993). However, this crypt function may not apply to slightly sunken individual stomata. In the latter case, protection of stomata from harsh conditions (abiotic and biotic stress), which may also be the case in more deeply sunken stomata, and/or facilitation of gas exchange in submerged leaves (Waldhoff et al., 2002) appear to be more appealing hypotheses explaining their functional significance. In fact, crypts may be multifunctional. The alternative possibility that crypts improve WUE was examined through one- dimensional modelling exercises. Improved WUE can be achieved if diffusion of water is more affected by crypts than diffusion of CO2 due to different diffusivities of these gases in the crypts compared with stomata. If this diffusivity difference exists, it will also result in a greater benefit for the diffusion of CO2 relative to water vapour when the part of the resistance that is due to stomata and crypts relative to mesophyll is greater. Thus, crypts may partially compensate for the high mesophyll resistance of high-LMA leaves. As shown in Chapter 3, however, due to the complexity of the diffusion pathway in crypts, it is not an easy task to elucidate differences in the concentration gradients of water vapour and CO2 within these structures. For simple structures like stomata, the relationship between

[148]

CHAPTER 7: GENERAL DISCUSSION

resistances for water vapour and CO2 is given by the relative values of their diffusivities. For a complex structure like a stomatal crypt, it may not be appropriate to directly calculate resistance to the inward flux for CO2 from the resistance to the outward flux of water vapour, because the three-dimensional diffusion space of the crypt may cause larger differences in the pathways of water vapour and CO2 compared with superficial stomata.

Moreover, additional resistances apply only to CO2 in the mesophyll. Thus, further studies must focus on examining the diffusion of water vapour and CO2 through crypts, including mesophyll diffusion of CO2, and re-investigating the effect of crypts on WUE by three- dimensional modelling approaches. Alternative or additional functions of sunken stomata and/or stomatal crypts may relate to stomatal microclimates created by superstomatal chambers, which “buffer” the conditions between the atmosphere and the stomata. Findings in Chapter 3 agree with the modelling studies of Roth-Nebelsick (2007) that showed that in a sunken stoma the relative humidity above the stomatal pore was about 75% compared with 63% for a non-sunken stoma, when the ambient relative humidity was 50%. Elevated and more “buffered” relative humidities inside the crypts may confer advantages related to better stomatal regulation in dry climates: stomata in crypts would be exposed to relative humidities less extreme, and perhaps closer to optimal, than those on the leaf surface. Furthermore, crypts may influence leaf hydraulic conductance, as Brodribb et al. (2007) recently showed that a reduced distance between veins and the evaporating surfaces improves leaf hydraulic conductance. In conclusion, this thesis has shown that the previously assumed transpiration-reducing role of crypts is conjectural and unlikely to be their primary function. Although based on leaf trait interrelationships crypts appear to benefit CO2 diffusion, the physiological significance of crypts is not yet fully resolved. Multi-functionality of these structures cannot be excluded. Three-dimensional modelling highlighted the complexity of diffusion through these structures and the need of further experimentation to unravel their effects on WUE. Finite element modelling involving the spatial array boundary layer-crypt-stomata- mesophyll can give new insights into the physiological consequences of stomatal crypts.

[149]

CHAPTER 7: GENERAL DISCUSSION

7.2 SCLEROPHYLLY AND GAS EXCHANGE

In the present study, sclerophylly was represented by LMA, a pivotal leaf trait (Cunningham et al., 1999) that measures the investment of dry mass per unit of light- intercepting leaf area (Wright et al., 2004) and has been widely used as an indicator of sclerophylly (Sobrado and Medina, 1980; Witkowski and Lamont, 1991; Groom and Lamont, 1999; Westoby et al., 2002; Gratani and Varone, 2006). LMA varied 4-fold across the 49 Banksia species examined. Interestingly, the different contributions to LMA of leaf thickness and density in different species showed that even within the same genus there are various ways to be sclerophyllous, highlighting that different combinations of leaf traits can result to sclerophylly. In contrast to a recent multi-genus review which found that leaf density was the major factor associated with variation in LMA (Poorter et al., 2009), in the genus Banksia, leaf thickness and density contributed similarly to variation in LMA. This may be explained by the greater range in leaf density and much wider array of life forms and an associated wide range of leaf lifespans considered by Poorter et al. (2009). Contrary to the expectation, the leaf fraction that was mesophyll did not correlate with LMA or its anatomical components, indicating that high-LMA leaves are not „disadvantaged‟ by less mesophyll volume per leaf volume. Evidently, while structural and supporting tissue increase in high-LMA leaves, as has also been found in other studies (Castro-Díez et al., 2000), so too does mesophyll thickness. Consequently, there were no correlations between LMA and mesophyll volume fraction, or the fraction of nitrogen allocated to Rubisco and thylakoids.

In accordance with previous studies (Reich et al., 1997; Wright et al., 2004), Amass showed a strong negative correlation with LMA, highlighting the increased presence of structural tissue in high-LMA leaves. In addition, consistent with global trends (Wright et al., 2004;

Poorter et al., 2009), Aarea was independent of LMA, despite the fact that mesophyll volume per area increased with LMA, which is in accordance with previous studies on evergreen species (Castro-Díez et al., 2000). This was due to the concurrent decrease in the photosynthetic capacity of the mesophyll with increasing LMA. It was hypothesised that the lower photosynthetic rates per mesophyll in thicker, denser, high-LMA leaves are probably due to lower Rubisco concentrations, lower Rubisco specific activity or lower Rubisco

[150]

CHAPTER 7: GENERAL DISCUSSION

activation state, and this is an area where future research can focus. Furthermore, photosynthetic capacity can be influenced by non-uniform distribution of CO2 and light in these tissues. Patchiness of photosynthetic activity and of factors that determine it is likely to happen in Banksia leaves, in which stomatal crypts may create local gas microenvironments, whereas the great leaf thickness may be responsible for irregular distribution of absorbed irradiance. Interestingly, compensatory mechanisms of these effects are present in these leaves. Increased presence of bundle sheath extensions and other sclerenchymatic tissues in high-LMA leaves can facilitate light transmission to deeper leaf layers (Poulson and Vogelmann, 1990; Smith et al., 1997; Nikolopoulos et al., 2002; Karabourniotis, 1998), improving the uniformity of illumination across these thick leaves. Chlorophyll per unit mesophyll tended to increase with LMA, indicating potentially more efficient light capture. Stomatal crypts facilitate CO2 diffusion to adaxial palisade cells.

These mechanisms may help in achieving the relatively high Aarea of these species compared with other sclerophyllous species. As explained in the Introduction of this thesis, a range of LMA can be considered to reflect different ways of packaging photosynthetically active cells. As LMA increases, photosynthetic capacity spreads across a larger leaf area, where mesophyll cells are surrounded by more structural tissue. What is the reason behind the different packaging of high-LMA leaves? A number of hypotheses about the functional significance of high-LMA sclerophylls have been put forward (see General Introduction). Banksia species have evolved in a continent that underwent radical changes, from wet rainforests to drylands in mid Tertiary (Hill, 1998). Their fossil record shows that sclerophylly was present in rainforests in the early Tertiary, and Hill and Merrifield (1993) and Hill (1994, 1998) based on earlier literature (Loveless, 1962) suggested that sclerophylly was initially a response to the nutrient impoverished soils of these forests. Thus, poor nutrition-induced sclerophylly “pre-adapted” these ancestral taxa to the arid conditions that appeared in mid Tertiary. Today, these species inhabit a range of Australian environments, wet or dry, nutrient- poorer or nutrient-richer habitats, although generally Australian soils are considered to be nutrient-poor (Cunningham et al., 1999). It is possible that some sclerophyllous characters are closely associated with soil infertility, whereas others with other kinds of stresses, such as water shortage. At the same time, some sclerophyllous characters may have developed

[151]

CHAPTER 7: GENERAL DISCUSSION

in response to more than one environmental driver, leading to the observation of the same trait in different environments. Cunningham et al. (1999), for example, using phylogenetically independent contrasts in a dataset of perennial evergreen flora of New South Wales (Australia), found that leaf thickness and density increased similarly as drought and soil fertility decreased. However, vascular sclerification was more consistently associated with drought than with soil fertility, whereas epidermal thickening evolved repeatedly in response to nutrient-impoverished soils. Close associations between other leaf structural traits and specific environmental conditions may also exist, and this deserves further study. Thicker cell walls, for example, may be more closely related to seasonal drought than soil fertility, given the suggested increase in the driving force for water uptake due to less-elastic, more rigid cell walls, after a decline in relative water content (Corcuera et al., 2002; Mitchell et al., 2008). A factor that must also be taken into account in these interpretations is the light environment of these habitats. Banksia species are generally exposed to high light intensities throughout most of the year. Jordan et al. (2005) suggested that certain scleromorphic features may protect the leaves from excess irradiance. In the present study, the combined thickness of the adaxial epidermis and hypodermis (and pseudohypodermis, where present) was compared with that of the abaxial side and was found to be significantly higher (P<0.001). Considering that the orientation of these leaves is between horizontal and inclined but not fully vertical, more light may reach the adaxial than the abaxial leaf surface. The significantly thicker epidermis and hypodermis of the adaxial surface may have a photoprotective function (Jordan et al., 2005). Further research can explore relationships between the light environment and leaf orientation, the thickness of the abaxial epidermis and/or hypodermis, as well as the presence/abundance of elongated palisade-like mesophyll cells on the abaxial surface near the crypts, which was observed in the present study. It becomes clear therefore, that different sclerophyllous characters may be more or less associated with particular environmental drivers, such as soil fertility, water availability, light intensity, etc., but these differences converge in favouring high-LMA hard leaves (Cunningham et al., 1999). Banksia species inhabit a range of stressful environments, thus sclerophylly would be expected to be an adaptation that contributes to the fitness of these

[152]

CHAPTER 7: GENERAL DISCUSSION

species under harsh conditions, ultimately favouring long leaf life spans. Indeed, Banksia leaves have been reported to live up to 13 years (Witkowski et al., 1992). However, sclerophyllous anatomy also appears to compromise photosynthesis, as indicated through a reduced rate of CO2 assimilation per unit mesophyll, potentially due to lower investment in Rubisco. Greater investment in photosynthetic machinery may not be advantageous in the extremely nutrient-impoverished and seasonally dry habitats of Banksia species where economic use of nutrients is vital and partial stomatal closure is common in the dry season (Veneklaas and Poot, 2003).

7.3 MESOPHYLL CONDUCTANCE AT THE HIGH END OF THE LMA SPECTRUM

Stomatal conductance was independent of LMA (Fig. 5b in Chapter 4). Since the limitations to photosynthesis in the mesophyll have been shown to be more important in high-LMA than in low-LMA leaves (Flexas et al., 2008), attention was paid to mesophyll conductance (gm). Remarkably, across the wide LMA range that was examined, the leaves had a similar draw-down in the CO2 concentration from the atmosphere to the chloroplasts. This led to the conclusion that variation in stomatal and mesophyll conductance across different species and functional groups results in a highly conserved Cc. gm estimated using the “fluorescence method” (Harley et al., 1992) was significantly lower in high-LMA species, and was accompanied by a trend of decreasing Aarea with increasing LMA in this subset of species. These results agree with previous studies (Flexas et al., 2008), confirming that gm is limited by leaf structure at the high end of the LMA spectrum, thereby limiting photosynthesis. The findings of this thesis have improved the previously published relationship between gm and LMA (Flexas et al., 2008), by replacing the linear decrease of the gm notional upper bound by a concave curve with an unknown asymptotic value at LMA -2 above 500 g m . The low gm at high LMA was found to be associated with thicker mesophyll cell walls. Since, however, the chloroplast surface area exposed to intercellular air spaces as well as gias (the air phase component of gm) can also influence gm, measuring these traits in high-LMA leaves merits attention.

[153]

CHAPTER 7: GENERAL DISCUSSION

Currently, there are two independent methods to estimate gm, which sometimes yield different values of gm. For this reason, gm was also estimated with the second method, which is based on carbon isotope discrimination and gas exchange measurements (Evans et al., 1986), and the results were compared with those obtained with the fluorescence method. Interestingly, the two methods yielded different values of gm, namely the gm estimates obtained with the isotopic method were up to 74% higher than those obtained with the fluorescence method. This inconsistency was clear in some previous studies (Tazoe et al., 2009; Vrábl et al., 2009; Yin et al., 2009), but it was not observed in others (Warren et al., 2004; Flexas et al., 2006, 2007a). It was hypothesised that this discrepancy between the two methods was due to assumptions of both or either methods and sensitivity analyses were done together with theoretical modelling to examine this. In the isotopic method, although most assumed parameters do not significantly influence the absolute values of gm, the assumption of the isotope fractionation by Rubisco and PEPC (termed b) can significantly affect gm. Usually, the value of 30‰ is used for b based on values measured for tobacco of (29.5‰; Evans et al., 1994) and wheat (32‰; von Caemmerer and Evans, 1991). Hence, future studies must focus on the quantification of the isotope fractionation by Rubisco and PEPC in more species. In the fluorescence method, small variations of the photo-compensation point in the absence of mitochondrial respiration, Γ* -1 (of the order of 28-40 μmol mol at 21% O2), do not influence the gm estimation; however, * -1 greater deviations (e.g. an increase in Γ to 50 μmol mol ) can increase gm to values closer to those obtained with the isotopic method. This can potentially explain the discrepancy between the two methods, signifying that it is vital that Γ* is properly estimated. The assumption of the fluorescence method that the fluorescence signal emanating from the adaxial surface is representative of the whole leaf depth can overestimate electron transport rate estimated from fluorescence (Jf). Modelling showed that a 10% overestimation of Jf results in up to 60% underestimation of gm in Banksia leaves. This is even more likely to happen in thicker leaves, in which the chloroplasts of the adaxial leaf side sampled by the fluorescence are poorly representative of the whole leaf depth. If that is the case, the apparent relationship of decreasing gm with LMA that has been shown previously (Flexas et al., 2008; Chapter 5) will change: gm will not be as low as previously thought and may not decrease as fast with LMA (Fig. 11 in Chapter 6). In addition, the isotopic method has its

[154]

CHAPTER 7: GENERAL DISCUSSION

own limitations, which may also be partially responsible for the higher gm estimates obtained with this compared with the fluorescence method. In particular, the assumption that gm is independent of O2 may influence the gm estimation, since the isotopic measurements are done at low O2 concentration. Thus, future research must focus on whether and how gm responds to varying O2 levels. Finally, stomatal patchiness, if any, may introduce some error in the calculation of gm with either method, through errors in the estimation of intercellular CO2 concentration (West et al., 2005; Pons et al., 2009). Further experimentation is needed to investigate the possibility and effects of stomatal patchiness on the calculation of gm. This can be assessed by thermal imaging at low O2, and is likely for Banksia leaves in which stomata are arranged in distinct patches (crypts).

Flexas et al. (2007a) were the first to report a gm responding to short term changes in CO2 and irradiance. Since then, a number of studies have reported contradictory results (Tazoe et al., 2009; Vrábl et al., 2009; Yin et al., 2009), which may be related to the methodology used to estimate gm in each study, since, as was also shown above, the two independent methods currently available often yield different values of gm. Moreover, the response of gm to CO2 and irradiance had not been studied in sclerophyllous high-LMA species. To advance our understanding of the likely effects of leaf structure and of limitations of each method on the responses of gm to CO2 and irradiance, both methods mentioned above were used to derive estimates of gm in three diverse Banksia species. gm declined with increasing

CO2 and decreasing irradiance with both methods used. This was consistent with all previous studies, except for Tazoe et al. (2009), who did not find any response of gm to either CO2 or irradiance using the isotopic method in wheat. Although the absolute values of gm changed between the two methods, its response to CO2 and irradiance was similar with both methods. This adds more confidence on the observation that gm responds to CO2 and irradiance in the short term. The physiological advantage, if any, and mechanism behind the response of gm to CO2 and irradiance are still unknown. Recent research is paying attention to cooporins, membrane channels, which may be involved in the regulation of gm in response to environmental drivers (Terashima and Ono, 2002; Üehlein et al., 2003; Hanba et al., 2004; Flexas et al., 2006; Üehlein et al., 2008; Miyazawa et al.,

2008). Further experimentation is needed to elucidate the regulation of gm at the cell and organelle level, and give insight into its adaptive value.

[155]

CHAPTER 7: GENERAL DISCUSSION

7.4 CONCLUSIONS

This study has made a step forward regarding our knowledge on the causes of variation in leaf structure and its consequences for gas exchange at an extreme end of the LMA spectrum. Moreover, it has raised new questions on the physiological benefits and effects of leaf structures, the causes of functional trait interrelationships and their association with photosynthesis, and finally, the accuracy of methodologies widely used to estimate essential parameters in photosynthetic modelling, such as mesophyll conductance. Put into a wider ecophysiological perspective, the findings of the present thesis complement our knowledge of the physiology and water relations of sclerophyllous plants and of adaptations that enhance plant performance under resource-poor conditions.

[156]

BIBLIOGRAPHY

[157]

BIBLIOGRAPHY

Aalto T, Juurola E. 2002. A three-dimensional model of CO2 transport in airspaces and mesophyll cells of a silver birch leaf. Plant, Cell and Environment 25: 1399-1409.

Abramoff MD, Magelhaes PJ, Ram SJ. 2004. Image Processing with ImageJ. Biophotonics International 11: 36-42.

Baldini E, Facini O, Nerozzi F, Rossi F, Rotondi A. 1997. Leaf characteristics and optical properties of different woody species. Trees 12: 73-81.

Barbour MM, Farquhar GD, Hanson DT, Bickford CP, Powers H, McDowell NG. 2007. A new measurement technique reveals temporal variation in delta O18 of leaf- respired CO2. Plant Cell and Environment 30: 456-468.

Beadle NCW. 1966. Soil phosphate and its role in molding segments of the Australian flora and vegetation, with special reference to xeromorphy and sclerophylly. Ecology 47: 992-1007.

Beckett RP. 1997. Pressure-volume analysis of a range of poikilohydric plants implies the existence of negative turgor in vegetative cells. Annals of Botany 79: 145-152.

Beerling DJ, Chaloner WG. 1993. Stomatal density responses of Egyptian Olea europaea

L. leaves to CO2 change since 1327 BC. Annals of Botany 71: 431-435.

Bernacchi CJ, Portis AR, Nakano H, von Caemmerer S, Long SP. 2002. Temperature response of mesophyll conductance. Implications for the determination of Rubisco enzyme kinetics and for limitations to photosynthesis in vivo. Plant Physiology 130: 1992-1998.

Beyer WH. 1987. CRC Standard Mathematical Tables, 28th ed. Boca Raton, FL: CRC Press.

[158]

BIBLIOGRAPHY

Boddi S, Morassi Bonzi L, Calamassi R. 2002. Structure and ultrastructure of Pinus halepensis primary needles. Flora 197: 10-23.

Bowling DR, Sargent SD, Tanner BD, Ehleringer JR. 2003. Tunable diode laser absorption spectroscopy for stable isotope studies of ecosystem-atmosphere CO2 exchange. Agricultural and Forest Meteorology 118: 1-19.

Brodribb T, Hill RS. 1997. Imbricacy and stomatal wax plugs reduce maximum leaf conductance in Southern Hemisphere conifers. Australian Journal of Botany 45: 657-668.

Brodribb TJ, Field TS, Jordan GJ. 2007. Leaf maximum photosynthetic rate and venation are linked by hydraulics. Plant Physiology 144: 1890-1898.

Brooks A, Farquhar GD. 1985. Effect of temperature on the CO2/O2 specificity of ribulose-1,5-bisphosphate carboxylase/oxygenase and the rate of respiration in the light. Planta 165: 397-406.

Brugnoli E, Björkman O. 1992. Chloroplast movements in leaves: Influence on chlorophyll fluorescence and measurements of light-induced absorbance changes related to ΔpH and zeaxanthin formation. Photosynthesis Research 32: 23-35.

Buckley TN, Farquhar GD. 2004. A new analytical model for whole-leaf potential electron transport rate. Plant, Cell and Environment 27: 1487-1502.

Burgess SSO, Dawson TE. 2004. The contribution of fog to the water relations of Sequoia sempervirens (D.Don): foliar uptake and prevention of dehydration. Plant, Cell and Environment 27: 1023-1034.

Burrows GE, Bullock S. 1999. Leaf anatomy of wollemi pine (Wollemia nobilis, Araucariaceae). Australian Journal of Botany 47: 795-806.

[159]

BIBLIOGRAPHY

Castro-Díez P, Puyravaud JP, Cornelissen JHC. 2000. Leaf structure and anatomy as related to leaf mass per area variation in seedlings of a wide range of woody plant species and types. Oecologia 124: 476-486.

Chabot BF, Chabot JF. 1977. Effects of light and temperature on leaf anatomy and photosynthesis in Fragaria vesca. Oecologia 26: 363-377.

Chabot BF, Hicks DJ. 1982. The ecology of leaf lifespans. Annual Review of Ecology and Systematics 13: 229-259.

Chapin FS. 1980. The mineral nutrition of wild plants. Annual Review of Ecology and Systematics 11: 233-260.

Corcuera L, Camarero JJ, Gil-Pelegrín E. 2002. Functional groups in Quercus species derived from the analysis of pressure-volume curves. Trees 16: 465-472.

Cowling RM, Campbell BM. 1983. The definition of leaf consistence categories in the fynbos biome and their distribution along an altitudinal gradient in the south eastern Cape. Journal of South African Botany 49: 87-101.

Cunningham SA, Summerhayes B, Westoby M. 1999. Evolutionary divergences in leaf structure and chemistry: comparing rainfall and soil nutrient gradients. Ecological Monographs 69: 569-588.

Del Hierro AM, Kronberger W, Richter H. 2000. Chloroplast movements in leaves of two evergreen sclerophyllous plants (Ilex aquifolium and Prunus laurocerasus). Plant Biosystems 3: 297-304.

Denton MD, Veneklaas EJ, Freimoser FM, Lambers H. 2007. Banksia species (Proteaceae) from severely phosphorus-impoverished soils exhibit extreme efficiency in the use and re-mobilisation of phosphorus. Plant, Cell and Environment 30: 1557-1565.

[160]

BIBLIOGRAPHY

Di Castri F. 1981. Mediterranean-type shrublands of the world. In: Di Castri F, Goodall DW, Specht RL, eds. Ecosystems of the world. Mediterranean-type shrublands. Elsevier Scientific Publishing Company.

Dillon RJ. 2002. The diversity of scleromorphic structures in the leaves of Proteaceae. Honours thesis, University of Tasmania, Hobart, Australia.

Doley D. 2004. Evolution of plant water relations research in Australia. Functional Plant Biology 31: 405-414.

Dreyer E, Bousquet F, Ducrey M. 1990. Use of pressure volume curves in water relation analysis on woody shoots: influence of rehydration and comparison of four European oak species. Annals of Forest Science 47: 285-297.

Edwards C, Read J, Sanson G. 2000. Characterising sclerophylly: some mechanical properties of leaves from heath and forest. Oecologia 123: 158-167.

Eensalu E, Kupper P, Sellin A, Rahi M, Sõber A, Kull O. 2008. Do stomata operate at the same relative opening range along a canopy profile of Betula pendula? Functional Plant Biology 35: 103-110.

Ehleringer J. 1980. Leaf morphology and reflectance in relation to water and temperature stress. In: Turner NC, Kramer PJ, eds. Adaptation of plants to water and high temperature stress. John Wiley and Sons, Inc., 295-308.

Ehleringer JR, Mooney HA. 1978. Leaf hairs: Effects on physiological activity and adaptive value to a desert shrub. Oecologia 37: 183-200.

Epron D, Godard D, Cornic G, Genty B. 1995. Limitation of net CO2 assimilation rate by interna1 resistances to CO2 transfer in the leaves of two tree species (Fagus sylvatica L. and Castanea sativa Mill.). Plant, Cell and Environment 18: 43-51.

[161]

BIBLIOGRAPHY

Ethier GJ, Livingston NJ. 2004. On the need to incorporate sensitivity to CO2 transfer conductance into the Farquhar-von Caemmerer-Berry leaf photosynthesis model. Plant, Cell and Environment 27: 137-153.

Evans JR. 1989. Photosynthesis and nitrogen relationships in leaves of C3 plants. Oecologia 78: 9-19.

Evans JR, Kaldenhoff R, Genty B, Terashima I. 2009. Resistances along the CO2 diffusion pathway inside leaves. Journal of Experimental Botany 60: 2235-2248.

Evans JR, Loreto F. 2000. Acquisition and diffusion of CO2 in higher plant leaves. In: Leegood RC, Sharkey TD, von Caemmerer S, eds. Photosynthesis: Physiology and Metabolism. Kluwer Academic Publishers, Dordrecht, The Netherlands, 321-351.

Evans JR, Sharkey TD, Berry JA, Farquhar GD. 1986. Carbon isotope discrimination measured concurrently with gas exchange to investigate CO2 diffusion in leaves of higher plants. Australian Journal of Plant Physiology 13: 281-292.

Evans JR, Vogelmann TC. 2003. Profiles of 14C fixation through spinach leaves in relation to light absorption and photosynthetic capacity. Plant, Cell and Environment 26: 547-560.

Evans JR, Vogelmann TC. 2006. Photosynthesis within isobilateral Eucalyptus pauciflora leaves. New Phytologist 171: 771-782.

Evans JR, von Caemmerer S. 1996. Carbon dioxide diffusion inside leaves. Plant Physiology 110: 339-346.

Evans JR, von Caemmerer S, Setchell BA, Hudson GS. 1994. The relationship between

CO2 transfer conductance and leaf anatomy in transgenic tobacco with reduced content of Rubisco. Australian Journal of Plant Physiology 21: 475-495.

[162]

BIBLIOGRAPHY

Farquhar GD, Sharkey TD. 1982. Stomatal conductance and photosynthesis. Annual Review of Plant Physiology 33: 317-345.

Farquhar GD, von Caemmerer S, Berry JA. 1980. A biochemical model of photosynthetic CO2 assimilation in leaves of C3 species. Planta 149: 78-90.

Feild TS, Zwieniecki MA, Donoghue MJ, Holbrook NM. 1998. Stomatal plugs of Drimys winteri (Winteraceae) protect leaves from mist but not drought. Proceedings of the National Academy of Sciences USA 95: 14256-14259.

Flexas J, Diaz-Espejo A, Berry JA, Cifre J, Galmés J, Kaldenhoff R, Medrano H, Ribas-Carbo M. 2007b. Analysis of leakage in IRGA‟s leaf chambers of open gas exchange systems: quantification and its effects in photosynthesis parameterization. Journal of Experimental Botany 58: 1533-1543.

Flexas J, Diaz-Espejo A, Galmés J, Kaldenhoff R, Medrano H, Ribas-Carbo M.

2007a. Rapid variations of mesophyll conductance in response to changes in CO2 concentration around leaves. Plant, Cell and Environment 30: 1284-1298.

Flexas J, Ribas-Carbo M, Diaz-Espejo A, Galmés J, Medrano H. 2008. Mesophyll conductance to CO2: current knowledge and future prospects. Plant, Cell and Environment 31: 601-621.

Flexas J, Ribas-Carbó M, Hanson DT, Bota J, Otto B, Cifre J, McDowell N, Medrano H, Kaldenhoff R. 2006. Tobacco aquaporin NtAQP1 is involved in mesophyll conductance to CO2 in vivo. The Plant Journal 48: 427-439.

Furbank RT, Horton P. 1987. Regulation of photosynthesis in isolated barely protoplasts: the contribution of cyclic photophosphorylation. Biochimica et Biophysica Acta 894: 332- 338.

[163]

BIBLIOGRAPHY

Genty B, Briantais JM, Baker N. 1989. The relationship between the quantum yield of photosynthetic electron transport and quenching of chlorophyll fluorescence. Biochimica et Biophysica Acta 990: 87-92.

Genty B, Harbinson J. 1996. Regulation of photosynthetic light utilisation. In: Baker NR, ed. Photosynthesis and the Environment, Kluwer Academic Publishers, Dordrecht, 67-99.

Genty B, Meyer S, Piel C, Badeck F, Liozon R. 1998. CO2 diffusion inside the leaf mesophyll of ligneous plants. In: Garab G, ed. Photosynthesis: Mechanisms and Effects, Kluwer Academic Publishers, 3961-3966.

George AS. 1996. The Banksia Book. Kangaroo Press, Australia.

Gibson AC. 1983. Anatomy of photosynthetic old stems of nonsucculent dicotyledons from North American deserts. Botanical Gazette 144: 347-362.

18 Gillon JS, Yakir D. 2000. Internal conductance to CO2 diffusion and C OO discrimination in C3 leaves. Plant Physiology 123: 201-213.

Gindel I. 1969. Stomatal number and size as related to soil moisture in tree xerophytes in Israel. Ecology 50: 263-267.

Givnish TJ. 1978. Ecological aspects of plant morphology: leaf form in relation to environment. In: Satteer R, ed. Theoretical plant morphology (Acta Biotheoretica, Vol. 27). Linden University Press, The Hague, 83-142.

Gratani L, Varone L. 2004. Adaptive photosynthetic strategies of the Mediterranean maquis species according to their origin. Photosynthetica 42: 551-558.

Gratani L, Varone L. 2006. Long-time variations in leaf mass and area of Mediterranean evergreen broad-leaf and narrow-leaf maquis species. Photosynthetica 44: 161-168.

[164]

BIBLIOGRAPHY

Groom PK, Lamont BB. 1997. Xerophytic implications of increased sclerophylly: interactions with water and light in Hakea psilorrhyncha seedlings. New Phytologist 136: 231-237.

Groom PK, Lamont BB. 1999. Which common indices of sclerophylly best reflect differences in leaf structure? Ecoscience 6: 471-474.

Groom PK, Lamont BB, Kupsky L. 1994. Contrasting morphology and ecophysiology of co-occurring broad and terete leaves in Hakea trifurcata (Proteaceae). Australian Journal of Botany 42: 307-320.

Gutschick VP. 1999. Biotic and abiotic consequences of differences in leaf structure. New Phytologist 143: 3-18.

Hanba YT, Kogami H, Terashima I. 2002. The effect of growth irradiance on leaf anatomy and photosynthesis in Acer species differing in light demand. Plant, Cell and Environment 25: 1021-1030.

Hanba YT, Miyazawa S-I, Kogami H, Terashima I. 2001. Effects of leaf age on internal

CO2 transfer conductance and photosynthesis in tree species having different types of shoot phenology. Australian Journal of Plant Physiology 28: 1075-1084.

Hanba YT, Miyazawa S-I, Terashima I. 1999. The influence of leaf thickness on the

CO2 transfer conductance and leaf stable carbon isotope ratio for some evergreen tree species in Japanese warm-temperate forests. Functional Ecology 13: 632-639.

Hanba YT, Shibasaka M, Hayashi Y, Hayakawa T, Kasamo K, Terashima I, Katsuhara M. 2004. Overexpression of the barley aquaporin HvPIP2;1 increases internal

CO2 conductance and CO2 assimilation in the leaves of transgenic rice plants. Plant and Cell Physiology 45: 521-529.

[165]

BIBLIOGRAPHY

Harley PC, Loreto F, Di Marco G, Sharkey TD. 1992. Theoretical considerations when estimating the mesophyll conductance to CO2 flux by the analysis of the response of photosynthesis to CO2. Plant Physiology 98: 1429-1436.

Harrison MT, Edwards EJ, Farquhar GD, Nicotra AB, Evans JR. 2009. Nitrogen in cell walls of sclerophyllous leaves accounts for little of the variation in photosynthetic nitrogen-use efficiency. Plant, Cell and Environment 32: 259-270.

Hassiotou F, Evans JR, Martha L, Veneklaas EJ. 2009b. Stomatal crypts may facilitate diffusion of CO2 to adaxial mesophyll cells in thick sclerophylls. Plant, Cell and Environment (DOI: 10.1111/j.1365-3040.2009.02024.x).

Hassiotou F, Ludwig M, Renton M, Veneklaas EJ, Evans JR. 2009a. Influence of leaf dry mass per area, CO2 and irradiance on mesophyll conductance in sclerophylls. Journal of Experimental Botany 60: 2303-2314.

Haworth M, McElwain J. 2008. Hot, dry, wet, cold or toxic? Revisiting the ecological significance of leaf and cuticular micromorphology. Palaeogeography, Palaeoclimatology, Palaeoecology 262: 79-90.

Heber U, Egneus H, Hanck U, Jensen M, Koster S. 1978. Regulation of photosynthetic electron transport and photo-phosphorylation in intact chloroplasts and leaves of Spinacia oleracea L. Planta 143: 41-49.

Hilbert D, Cohn-Vossen S. 1999. Geometry and the Imagination. New York: Chelsea.

Hill RS. 1994. History of the Australian vegetation. Cambridge University Press, Cambridge, UK.

Hill RS. 1998. Fossil evidence for the onset of xeromorphy and scleromorphy in Australian Proteaceae. Australian Systematic Botany 11: 391-400.

[166]

BIBLIOGRAPHY

Hill RS, Merrifield HE. 1993. An early tertiary macroflora from West Dale, southwestern Australia. Alcheringa 17: 285-326.

Jones HG. 1998. Stomatal control of photosynthesis and transpiration. Journal of Experimental Botany 49: 387-398.

Jordan GJ, Dillon RA, Weston PH. 2005. Solar radiation as a factor in the evolution of scleromorphic leaf anatomy in Proteaceae. American Journal of Botany 92: 789-796.

Jordan GJ, Weston PH, Carpenter RJ, Dillon RA, Brodribb TJ. 2008. The evolutionary relations of sunken, covered, and encrypted stomata to dry habitats in Proteaceae. American Journal of Botany 95: 521-530.

Karabourniotis G. 1998. Light-guiding function of foliar sclereids in the evergreen sclerophyll Phillyrea latifolia: a quantitative approach. Journal of Experimental Botany 49: 739-746.

Katsuhara M, Hanba YT, Shiratake K, Maeshima M. 2008. Expanding roles of plant aquaporins in plasma membranes and cell organelles. Functional Plant Biology 35: 1-14.

Kays WM. 1966. Convective Heat and Mass Transfer. McGraw-Hill, New York.

Kelly CK, Beerling DJ. 1995. Plant life form, stomatal density and taxonomic relatedness: a reanalysis of Salisbury (1927). Functional Ecology 9: 422-431.

Kogami H, Hanba YT, Kibe T, Terashima I, Masuzawa T. 2001. CO2 transfer conductance, leaf structure and carbon isotope composition of Polygonum cuspidatum leaves from low and high altitudes. Plant, Cell and Environment 24: 529-538.

Lamont BB, Connell SW. 1996. Biogeography of Banksia in southwestern Australia. Journal of Biogeography 23: 295-309.

[167]

BIBLIOGRAPHY

Lanigan GJ, Betson N, Griffiths H, Seibt U. 2008. Carbon isotope discrimination during photorespiration and carboxylation in Senecio. Plant Physiology 148: 2013-2020.

Larcher W. 2003. Physiological plant ecology, 4th edn. Cambridge University Press, Cambridge.

Lawrence CA, Liu P. 2006. Relation of structure, properties and performance of fibrous media for gas filtration. Chemical Engineering Technology 29: 957-967.

Lee R, Gates DM. 1964. Diffusion resistance in leaves as related to their stomatal anatomy and micro-structure. American Journal of Botany 51: 963-975.

Lide DR. 1990. CRC Handbook of chemistry and physics. CRC Press, Inc.

Lloyd J, Syvertsen JP, Kriedemann PE, Farquhar GD. 1992. Low conductances for

CO2 diffusion from stomata to the sites of carboxylation in leaves of woody species. Plant, Cell and Environment 15: 873-899.

Lo Gullo MA, Salleo S. 1988. Different strategies of drought resistance in three Mediterranean sclerophyllous trees growing in the same environmental conditions. New Phytologist 108: 267-276.

Long SP, Bernacchi CJ. 2003. Gas exchange measurements, what can they tell us about the underlying limitations to photosynthesis? Procedures and sources of error. Journal of Experimental Botany 54, 2393-2401.

Loreto F, Harley PC, Di Marco G, Sharkey TD. 1992. Estimation of mesophyll conductance to CO2 flux by three different methods. Plant Physiology 98: 1437-1443.

[168]

BIBLIOGRAPHY

Lösch R, Tenhunen JD, Pereira JS, Lange OL. 1982. Diurnal courses of stomatal resistance and transpiration of wild and cultivated Mediterranean perennials at the end of the summer dry season in Portugal. Flora 172: 138-160.

Loveless AR. 1961. A nutritional interpretation of sclerophylly based on differences in the chemical composition of sclerophyllous and mesophytic leaves. Annals of Botany 25: 168- 184.

Loveless AR. 1962. Further evidence to support a nutritional interpretation of sclerophylly. Annals of Botany 26: 551-561.

Mast AR, Givnish TJ. 2002. Historical biogeography and the origin of stomatal distributions in Banksia and Dryandra (Proteaceae) based on their cpDNA phylogeny. American Journal of Botany 89: 1311-1323.

Matthews P. 2003. The function of stomatal crypts in relation to gas exchange and airborne dust. Honours thesis, School of Earth and Environmental Sciences, Discipline of Environmental Biology, University of Adelaide, Australia.

Maurel C. 1997. Aquaporins and water permeability of plant membranes. Annual Reviews of Plant Physiology and Plant Molecular Biology 48: 399-429.

McCully ME, Canny MJ, Huang CX. 2004. The management of extracellular ice by petioles of frost-resistant herbaceous plants. Annals of Botany 94: 665-674.

Meidner, H, Mansfield TA. 1968. Physiology of Stomata. McGraw-Hill: London.

Meinzer FC, Andrade JL, Goldstein G, Holbrook NM, Cavelier J, Jackson P. 1997. Control of transpiration from the upper canopy of a tropical forest: the role of stomatal, boundary-layer and hydraulic architecture components. Plant, Cell and Environment 20: 1242-1252.

[169]

BIBLIOGRAPHY

Merino J, Field C, Mooney HA. 1984. Construction and maintenance costs of mediterranean-climate evergreen and deciduous leaves. II. Biochemical pathway analysis. Acta Oecologica 5: 211-229.

Metcalfe CR, Chalk L. 1979. Anatomy of the dicotyledons. Vol. I. Systematic anatomy of the leaf and stem, 2nd edition. Oxford, Clarendon Press.

Mitchell PJ, Veneklaas EJ, Lambers H, Burgess SSO. 2008. Leaf water relations during summer water deficit: differential responses in turgor maintenance and variation in leaf structure among different plant communities in south-western Australia. Plant, Cell and Environment 31: 1791-1802.

Miyazawa S-I, Yoshimura S, Shinzaki Y, Maeshima M, Miyake C. 2008. Deactivation of aquaporins decreases internal conductance to CO2 diffusion in tobacco leaves grown under long-term drought. Functional Plant Biology 35: 553-564.

Mohammadian M. 2005. An investigation of the functions of leaf surface modifications in the Proteaceae and Araucariaceae. PhD thesis, School of Earth and Environmental Sciences, Discipline of Environmental Biology, University of Adelaide, Australia.

Myers N, Mittermeier RA, Mittermeier CG, Dafonseca GAB, Kent J. 2000. Biodiversity hotspots for conservation priorities. Nature 403: 853-858.

Myers DA, Vogelmann TC, Bornman JF. 1994. Epidermal focusing and effects on light utilization in Oxalis acetosella. Physiologia Plantarum 91: 651-656.

Morison JIL, Gallouët E, Lawson T, Cornic G, Herbin R, Baker NR. 2005. Lateral diffusion of CO2 in leaves is not sufficient to support photosynthesis. Plant Physiology 139: 254-266.

[170]

BIBLIOGRAPHY

Morison JIL, Lawson T. 2007. Does lateral gas diffusion in leaves matter? Plant, Cell and Environment 30: 1072-1085.

Napp-Zinn K. 1966. Anatomie des Blattes. II. Blattanatomie der Gymnospermen. Berlin, Gebrüder Bornträger.

Nardini A, Lo Gullo MA, Tracanelli S. 1996. Water relations of six sclerophylls growing near Trieste (Northeastern Italy): has sclerophylly a unequivocal functional significance? Giornale Botanico Italiano 130: 811-828.

Niinemets U. 2001. Global scale climatic controls of leaf dry mass per area, density and thickness in trees and shrubs. Ecology 82: 453-469.

Niinemets Ü, Sack L. 2006. Structural determinants of leaf light-harvesting capacity and photosynthetic potentials. Progress in Botany 67: 385-419.

Niinemets Ü, Wright IJ, Evans JR. 2009. Leaf mesophyll diffusion conductance in 35 Australian sclerophylls covering a broad range of foliage structural and physiological variation. Journal of Experimental Botany 60: 2433-2449.

Nikolopoulos D, Liakopoulos G, Drossopoulos I, Karabourniotis G. 2002. The relationship between anatomy and photosynthetic performance of heterobaric leaves. Plant Physiology 129: 235-243.

Nobel PS. 1974. Introduction to biophysical plant physiology. Freeman W.H., San Francisco.

Nobel PS. 1999. Physiological and environmental plant physiology. San Diego, CA: Academic Press.

[171]

BIBLIOGRAPHY

Nobel PS, Zaragoza LJ, Smith WK. 1975. Relation between mesophyll surface area, photosynthetic rate and illumination level during development for leaves of Plectranthus parviflorus Henkel. Plant Physiology 55: 1067-1070.

Oertli JJ, Lips SH, Agami M. 1990. The strength of sclerophyllous cells to resist collapse due to negative turgor pressure. Acta Oecologica 11: 281-289.

Ögren E, Evans JR. 1993. Photosynthetic light response curves. I. Influence of CO2 partial pressure and leaf inversion. Planta 189: 182-190.

Paling EI, Humphries G, McCardle I, Thomson G. 2001. The effects of iron ore dust on mangroves in Western Australia: Lack of evidence for stomatal damage. Wetlands Ecology and Management 9: 363-370.

Parkhurst DF. 1994. Diffusion of CO2 and other gases inside leaves. New Phytologist 126: 449-479.

Parkhurst DF, Mott KA. 1990. Intercellular diffusion limits to CO2 uptake in leaves. Studies in air and helox. Plant Physiology 94: 1024-1032.

Parlange J-Y, Waggoner PE. 1970. Stomatal dimensions and resistance to diffusion. Plant Physiol 46: 337-342.

Pinheiro J, Bates D, Debroy S, Sarkar D, The R Core team. 2007. nlme: Linear and Nonlinear Mixed Effects Models. R package version 3: 1-85.

Pons TL, Flexas J, von Caemmerer S, Evans JR, Genty B, Ribas-Carbo M, Brugnoli

E. 2009. Estimating mesophyll conductance to CO2: methodology, potential errors, and recommendations. Journal of Experimental Botany 60: 2217-2234.

[172]

BIBLIOGRAPHY

Poorter H, Evans JR. 1998. Photosynthetic nitrogen-use efficiency of species that differ inherently in specific leaf area. Oecologia 116: 26-37.

Poorter H, Niinemets Ü, Poorter L, Wright IJ, Villar R. 2009. Causes and consequences of variation in leaf mass per area (LMA): a meta-analysis. New Phytologist 182: 565-588.

Poorter H, van der Werf A. 1998. Is inherent variation in RGR determined by LAR at low irradiance and by NAR at high irradiance? A review of herbaceous species. In: Lambers H, Poorter H, van Vuuren MMI, eds. Inherent Variation in Plant Growth. Backhuys Publishers, Leiden, The Netherlands, 309-336.

Poulson ME, Vogelmann TC. 1990. Epidermal focusing and effects upon photosynthetic light-harvesting in leaves of Oxalis. Plant, Cell and Environment 13: 803-811.

Press MC. 1999. The functional significance of leaf structure: a search for generalizations. New Phytologist 143: 213-219.

Price D, von Caemmerer S, Evans JR, Yu JW, Lloyd J, Oja V, Kell P, Harrison K, Gallagher A, Badger M. 1994. Specific reduction of chloroplast carbonic anhydrase activity by antisense RNA in transgenic tobacco plants has a minor effect on photosynthetic CO2 assimilation. Planta 193: 331-340.

R Development Core Team. 2007. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07- 0, URL http://www.R-project.org.

Raskin I. 1983. A method for measuring leaf volume, density, thickness and internal gas volume. Hortscience 18: 698-699.

Raven PH, Evert RF, Eichorn SE. 2005. Biology of plants. WH Freeman, New York, USA.

[173]

BIBLIOGRAPHY

Read J, Edwards C, Sanson GD, Aranwela N. 2000. Relationships between sclerophylly, leaf biomechanical properties and leaf anatomy in some Australian heath and forest species. Plant Biosystems 134: 261-277.

Read J, Sanson GD. 2003. Characterizing sclerophylly: the mechanical properties of a diverse range of leaf types. New Phytologist 160: 81-99.

Reich P, Uhl C, Walters MB, Ellsworth DS. 1991. Leaf lifespan as a determinant of leaf structure and function among 23 amazonian tree species. Oecologia 86: 16-24.

Reich PB, Walters MB, Ellsworth DS. 1997. From tropics to tundra: global convergence in plant functioning. Proceedings of the National Academy of Sciences USA 94: 13730- 13734.

Richardson SJ, Peltzer DA, Allen RB, McGlone MS, Parfitt RL. 2004. Rapid development of phosphorus limitation in temperate rainforest along the Franz Josef soil chronosequence. Oecologia 139: 267-276.

Ridge RW, Loneragan WA, Bell DT, Colquhoun IJ, Kuo J. 1984. Comparative studies in selected species of Eucalyptus used in rehabilitation of the Northern jarrah forest, Western Australia. II. Wood and Leaf Anatomy. Australian Journal of Botany 32: 375-386.

Ripley BS, Pammenter NW, Smith VR. 1999. Function of leaf hairs revisited: The hair layer on leaves of Arctotheca populifolia reduces photoinhibition, but leads to higher leaf temperatures caused by lower transpiration rates. Journal of Plant Physiology 155: 78-85.

Rossini-Oliva S, Valdés B, Leidi EO. 2009. Accumulation and in vivo tissue distribution of pollutant elements in Erica andevalensis. Science of the Total Environment 407: 1929- 1936.

[174]

BIBLIOGRAPHY

Roth-Nebelsick A. 2007. Computer-based studies of diffusion through stomata of different architecture. Annals of Botany 100: 23-32.

Rotondi A, Rossi F, Asunis C, Cesaraccio C. 2003. Leaf xeromorphic adaptations of some plants of a coastal Mediterranean macchia ecosystem. Journal of Mediterranean Ecology 4: 25-35.

Salleo S, Nardini A. 2000. Sclerophylly: evolutionary advantage or mere epiphenomenon? Plant Biosystems 134: 247-259.

Salleo S, Nardini A, Lo Gullo MA. 1997. Is sclerophylly of Mediterranean evergreens an adaptation to drought? New Phytologist 135: 603-612.

+ Scheibe R. 1987. NADP malate dehydrogenase in C3 plants: regulation and role of a light- activated enzyme. Physiologia Plantarum 71: 393-400.

Schimper AFW. 1903. Plant geography upon a physiological basis. Transl. Fisher WR, Clarendon Press, Oxford.

Schuepp PH. 1993. Leaf boundary layers. New Phytologist 125: 477-507.

Sharkey TD, Bernacchi CJ, Farquhar GD, Singsaas EL. 2007. In Practice: Fitting photosynthetic carbon dioxide response curves for C3 leaves. Plant, Cell and Environment 30: 1035-1040.

Sharkey TD, Vassey TL, Vanderveer PJ, Vierstra RD. 1991. Carbon metabolism enzymes and photosynthesis in transgenic tobacco (Nicotiana tabaccum L.) having excess phytochrome. Planta 185: 287-296.

Shen L, Chen Z. 2007. Critical review of the impact of tortuosity on diffusion. Chemical Engineering Science 62: 3748-3755.

[175]

BIBLIOGRAPHY

Slaton MR, Smith WK. 2002. Mesophyll architecture and cell exposure to intercellular air space in alpine, desert, and forest species. International Journal of Plant Sciences 163: 937-948.

Smith WK, Vogelmann TC, DeLucia EH, Bell DT, Shepherd KA. 1997. Leaf form and photosynthesis. Do leaf structure and orientation interact to regulate internal light and carbon dioxide? Bioscience 47: 785-793.

Sobrado MA. 1986. Aspects of tissue water relations and seasonal changes of leaf water potential components of evergreen and deciduous species coexisting in tropical dry forest. Oecologia 68: 413-416.

Sobrado MA. 2009. Cost-benefit relationships in sclerophyllous leaves of the „Bana‟ vegetation in the Amazon region. Trees 23: 429-437.

Sobrado MA, Medina E. 1980. General morphology, anatomical structure, and nutrient content of sclerophyllous leaves of the „Bana‟ vegetation of Amazonas. Oecologia 45: 341- 345.

Specht RL. 1969. A comparison of the sclerophyllous vegetation characteristic of the Mediterranean type climates in France, California, and Southern Australia. Australian Journal of Botany 17: 293-308.

Taylor AH. 1935. Errors in reflectometry. Journal of the Optical Society of America 25: 51-56.

Tazoe Y, von Caemmerer S, Badger MR, Evans JR. 2009. Light and CO2 do not affect the mesophyll conductance to CO2 diffusion in wheat leaves. Journal of Experimental Botany 60: 2291-2301.

[176]

BIBLIOGRAPHY

Terashima I, Hanba YT, Tazoe Y, Vyas P, Yano S. 2006. Irradiance and phenotype: comparative eco-development of sun and shade leaves in relation to photosynthetic CO2 diffusion. Journal of Experimental Botany 57: 343-354.

Terashima I, Ono K. 2002. Effects of HgCl2 on CO2 dependence of leaf photosynthesis: evidence indicating involvement of aquaporins in CO2 diffusion across the plasma membrane. Plant and Cell Physiology 43: 70-78.

Terashima I, Wong SC, Osmond CB, Farquhar GD. 1988. Characterization of non- uniform photosynthesis induced by abscisic acid in leaves having different mesophyll anatomies. Plant and Cell Physiology 29: 385-394.

Tholen D, Boom C, Noguchi K, Terashima I. 2007. The effects of chloroplast movement on CO2 transfer conductance in Arabidopsis thaliana. Plant and Cell Physiology 43: S95- S95.

Tholen D, Boom C, Noguchi K, Ueda S, Katase T, Terashima I. 2008. The chloroplast avoidance response decreases internal conductance to CO2 diffusion in Arabidopsis thaliana leaves. Plant, Cell & Environment 31: 1688-1700.

Thomson CJ, Armstrong W, Waters I, Greenway H. 1990. Aerenchyma formation and associated oxygen movement in seminal and nodal roots of wheat. Plant, Cell and Environment 13: 395-403.

Tomadakis M, Sotirchos SV. 1993. Transport properties of random arrays of freely overlapping cylinders with various orientation distributions. Journal of Chemical Physiology 98: 616-626.

Turner IM. 1994. Sclerophylly: primarily protective? Functional Ecology 8: 669-675.

[177]

BIBLIOGRAPHY

Turner IM, Choong MF, Tan HTW, Lucas PW. 1993. How tough are sclerophylls? Annals of Botany 71: 343-345.

Üehlein N, Lovisolo C, Siefritz F, Kaldenhoff R. 2003. The tobacco aquaporin NtAQP1 is a membrane CO2 pore with physiological functions. Nature 425: 734-737.

Üehlein N, Otto B, Hanson DT, Fischer M, McDowell N, Kaldenhoff R. 2008. Function of Nicotiana tabacum aquaporins as chloroplast gas pores challenges the concept of membrane CO2 permeability. Plant Cell 20: 648-657.

Valiente-Banuet A, Flores-Hernàndez N, Verdù M, Davila P. 1998. The chaparral vegetation in Mexico under nonmediterranean climate: the convergence and Madrean Tethyan Hypotheses reconsidered. American Journal of Botany 85: 1398-1408.

Veneklaas EJ, Poot P. 2003. Seasonal patterns in water use and leaf turnover of different plant functional types in a species-rich woodland, south-western Australia. Plant and Soil 257: 295-304.

Vogel S. 1996. Life in Moving Fluids, Ed 2. Princeton University Press, Princeton.

von Caemmerer S, Evans JR. 1991. Determination of the average partial pressure of CO2 in chloroplasts from leaves of several C3 plants. Australian Journal of Plant Physiology 18: 287-305. von Caemmerer S, Evans JR, Hudson GS, Andrews TJ. 1994. The kinetics of ribulose 1,5-bisphosphate carboxylase/oxygenase in vivo inferred from measurements of photosynthesis in leaves of transgenic tobacco. Planta 195: 88-97. von Caemmerer S, Farquhar GD. 1981. Some relationships between the biochemistry of photosynthesis and the gas exchange of leaves. Planta 153: 376-387.

[178]

BIBLIOGRAPHY

Vrábl D, Vašková M, Hronková M, Flexas J, Šantrůček J. 2009. Mesophyll conductance to CO2 transport estimated by two independent methods: effect of variable

CO2 concentration and abscisic acid. Journal of Experimental Botany 60: 2315-2323.

Waldhoff D, Furch B, Junk WJ. 2002. Fluorescence parameters, chlorophyll concentration, and anatomical features as indicators for flood adaptation of an abundant tree species in Central Amazonia: Symmeria paniculata. Environmental and Experimental Botany 48: 225-235.

Warren CR. 2006. Estimating the internal conductance to CO2 movement. Functional Plant Biology 33: 431-442.

Warren CR. 2007. Stand aside stomata, another actor deserves centre stage: the forgotten role of the internal conductance to CO2 transfer. Journal of Experimental Botany 59: 1475- 1487.

Warren CR, Livingston NJ, Turpin DH. 2004. Water stress decreases the transfer conductance of Douglas-fir (Pseudotsuga menziesii) seedlings. Tree Physiology 24: 971- 979.

Wellburn AR. 1994. The spectral determination of chlorophylls a and b, as well as total carotenoids, using various solvents with spectrophotometers of different resolution. Journal of Plant Physiology 144: 307-313.

West JD, Peak D, Peterson JQ, Mott KA. 2005. Dynamics of stomatal patches for a single surface of Xanthium strumarium L. leaves observed with fluorescence and thermal images. Plant, Cell and Environment 28: 633-641.

Western Australian Herbarium 1998–. FloraBase – The Western Australian Flora. Department of Environment and Conservation. http://florabase.dec.wa.gov.au/.

[179]

BIBLIOGRAPHY

Westoby M, Falster DS, Moles AT, Vesk PA, Wright IJ. 2002. Plant ecological strategies: some leading dimensions of variation between species. Annual Review of Ecology and Systematics 33: 125-159.

Williams TG, Flanagan LB, Coleman JR. 1996. Photosynthetic gas exchange and 13 18 16 discrimination against CO2, and C O O in tobacco plants modified by an antisense construct to have low chloroplastic carbonic anhydrase. Plant Physiology 112: 319-326.

Wingate L, Seibt U, Moncrieff JB, Jarvis PG, Lloyd J. 2007. Variations in 13C discrimination during CO2 exchange by Picea sitchensis branches in the field. Plant, Cell and Environment 30: 600-616.

Witkowski ETF, Lamont BB. 1991. Leaf specific mass confounds leaf density and thickness. Oecologia 88: 486-493.

Witkowski ETF, Lamont BB, Walton CS, Radford S. 1992. Leaf demography, sclerophylly, and ecophysiology of two Banksias with contrasting leaf life spans. Australian Journal of Botany 40: 849-862.

Wright IJ, Cannon K. 2001. Relationships between leaf lifespan and structural defences in a low-nutrient, sclerophyll flora. Functional Ecology 15: 351-359.

Wright IJ, Reich BP, Westoby M, Ackerly DD, Baruch Z, Bongers F, Cavender-Bares J, Chapin T, Cornelissen JHC, Diemer M et al. 2004. The worldwide leaf economics spectrum. Nature 428: 821-827.

Wrigley JW, Fagg M. 1989. Banksias, Waratahs, Grevilleas and other plants in the Australian Proteaceae family. Collins Publishers Australia, 80-121.

[180]

BIBLIOGRAPHY

Yin X, Struik PC, Romero P, Harbinson J, Evers JB, van der Putten PEL, Vos J. 2009. Using combined measurements of gas exchange and chlorophyll fluorescence to estimate parameters of a biochemical C3 photosynthesis model: a critical appraisal and a new integrated approach applied to leaves in a wheat (Triticum aestivum) canopy. Plant, Cell and Environment 32: 448-464.

Yiotis C, Manetas Y, Psaras GK. 2006. Leaf and green stem anatomy of the drought deciduous Mediterannean shrub Calicotome villosa (Poiret) Link. (Leguminosae). Flora 201: 102-107.

Yiotis C, Psaras GK, Manetas Y. 2008. Seasonal photosynthetic changes in the green- stemmed Mediterranean shrub Calicotome villosa: a comparison with leaves. Photosynthetica 46: 262-267.

Zeiger E, Farquhar GD, Cowan IR. 1987. Stomatal function. Stanford, USA: Stanford University Press.

Zhang WH, Prado C, Zhang WH. 1998. Water relations balance parameters of 30 woody species from Cerrado vegetation. Journal of Forest Research 9: 233-239.

Zienkiewicz OL, Taylor RL. 1989. The Finite Element Method, 4th edn. London, McGraw-Hill.

Zobel DB. 1996. Variation of water relations parameters with extended rehydration time, leaf form, season, and proportion of leaf. Canadian Journal of Forest Research 26: 175- 185.

[181]

[182]

APPENDICES

[183]

APPENDIX 1

APPENDIX 1

A B

C D

E F

G H

Fig. S1. Polypropylene mesh types used for testing the Nobel (1999) equation of stomatal pore conductance for pores as large as those of stomatal crypts. (A) PP140-105; (B) PP120-125; (C) PP100-149; (D) PP80-177; (E) PP70-210; (F) PP60-250; (G) PP280; and (H) PP405. Table 1 shows the morphological characteristics of each mesh type. Scale bars: 1 mm.

[184]

APPENDIX 1

) 7.00

1

-

s 2

- 6.00 O m O 2 5.00

4.00

3.00

2.00

1.00

Pore conductance (mol H (mol conductancePore 0.00 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Strand thickness (μm)

Fig. S2. Sensitivity analysis of the effect of pore depth as approximated by strand thickness on mesh pore conductance. The symbols indicate the measured pore conductance at the strand thickness used originally in the calculations – see Table 1 (green square: PP100-149; purple diamond: PP60-250; blue triangle: PP70-210; red circle: PP405; black cross: PP140-105). The lines indicate the effect of varying strand thickness by ± 50% on the calculation of pore conductance by Eqn 3, and their colours match the colours of the symbols (green line: PP100-149; purple line: PP60-250; blue line: PP70-210; red line: PP405; black line: PP140-105). Measured conductance closely matches the calculated conductance assuming a pore depth approximated by strand thickness. Overall, measured conductance exceeded the prediction slightly, which suggests that the mesh pore depth appears to be slightly lower that the one used. A crypt is an elliptical prism, with a rounded end, whereas the mesh flares out at both the top and the bottom. Thus, for real crypts, since the flaring at the entrance is not observed, the slight overestimation of pore depth would not occur.

[185]

APPENDIX 1

1

0.8

0.6

L

r

/

s r 0.4

0.2

0 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3

-1 2 Apparent rc (mol m s)

Fig. S3. Sensitivity analysis of the effect of over- or under-estimating crypt resistance (rc) by 20% on the resistance partitioning. The figure shows how stomatal resistance (rs), expressed as a fraction of total leaf resistance (rL), varies as rc varies in 10 Banksia species (filled circle: B. serrata; open circle: B. victoriae; filled square: B. attenuata; open square: B. ilicifolia; filled triangle: B. candolleana; open triangle: B. prionotes; filled diamond: B. repens; open diamond: B. elderiana; cross: B. quercifolia; x: B. hookeriana). The majority of leaf resistance is always due to stomatal resistance.

[186]

APPENDIX 2

APPENDIX 2. List of the 49 Banksia species examined (for nomenclature see Western Australian Herbarium, 1998).

Analysis 1 2 3 4 5 6 7 8 9 10 Species B. aculeata + B. aemula + B. ashbyi + B. attenuata + + + + + + + + + B. baueri + B. baxteri + B. benthamiana + B. brownii + B. burdettii + B. caleyi + B. candolleana + + + + + + B. chamaephyton + B. coccinea + + + + + B. dentata + B. dryandroides + + B. elderiana + + + + + + + + + B. elegans + B. epica + B. gardneri + B. goodii + B. grandis + B. hookeriana + + + + + + + B. ilicifolia + + + + + + + B. integrifolia + + + + + + B. laevigata + B. lemanniana + B. lindleyana + B. littoralis + + + + + B. media + B. menziesii + B. oblongifolia + B. oligantha + B. oreophila + + + + + B. paludosa + + + + + B. petiolaris + B. pilostylis + B. praemorsa + B. prionotes + + + + + + B. quercifolia + + + + B. repens + + + + + + + + + B. rosserae + B. sceptrum + B. seminuda + B. serrata + + + + + + + B. solandri + + + + + + + + + + B. speciosa + B. spinulosa + + + + + B. verticillata + B. victoriae + + + + + +

[187]

APPENDIX 2

1: 49 species (Fig. 1): Leaf dry mass per area and its relationship with leaf density and thickness; 2: 18 species (Fig. 5a,b, 6a): gas exchange measurements (CO2 assimilation rate and leaf conductance); 3: 17 species (Fig. 6b,c): nitrogen content and photosynthetic nitrogen use efficiency; 4: 14 species (Fig. 3a): leaf volume and porosity; 5: 12 species (Fig. 8b): nitrogen allocated to thylakoids; 6: 10 species (Fig. 2a, 3b, 7): thickness of the different leaf layers, mesophyll volume per unit leaf volume and net CO2 assimilation rate per unit mesophyll; 7: 7 species (Fig. 5c): mesophyll conductance; 8: 6 species (Fig. 4): mesophyll cell wall thickness; 9: 5 species (Fig. 2b): palisade cell length; 10: 6 species (Fig. 8a): nitrogen allocated to Rubisco.

[188]

APPENDIX 3

APPENDIX 3

Table S1. Slopes and intercepts of the calibration curves (gas exchange-based quantum yield,Φ , versus photochemical efficiency of Photosystem CO 2 -2 -1 -2 -1 II, ΦPSII) at low (500 μmol quanta m s ) and high (1500 μmol quanta m s ) PPFD of replicate leaves of the Banksia species examined. Species are shown in order of increasing leaf dry mass per area (LMA).

Species Slopes Intercepts Replicate 1 2 3 1 2 3 PPFD 500 μmol quanta m-2 s-1 B. serrata 0.562 0.815 0.782 -0.069 -0.119 -0.140 B. integrifolia 0.743 0.924 0.876 -0.134 -0.193 -0.190 B. attenuata 0.911 1.163 1.243 -0.124 -0.176 -0.189 B. solandri 0.622 0.673 0.694 -0.085 -0.102 -0.105 B. paludosa 0.534 0.504 0.579 -0.069 -0.060 -0.074 B. repens 0.709 0.700 0.714 -0.077 -0.095 -0.092 B. elderiana 0.564 0.527 0.562 -0.038 -0.043 -0.056 PPFD 1500 μmol quanta m-2 s-1 B. serrata 0.593 0.519 0.564 -0.025 -0.019 -0.025 B. integrifolia 0.583 0.744 0.676 -0.035 -0.062 -0.050 B. attenuata 0.847 0.811 0.868 -0.042 -0.043 -0.046 B. solandri 0.542 0.626 0.569 -0.023 -0.033 -0.028 B. paludosa 0.423 0.505 0.575 -0.009 -0.023 -0.027 B. repens 0.669 0.512 0.613 -0.022 -0.018 -0.030 B. elderiana 0.763 0.916 0.624 -0.024 -0.032 -0.023

[189]

APPENDIX 3

Table S2. Exponential regression analysis of the response of mesophyll conductance (gm) to leaf intercellular CO2 concentration (Ci) showing the significance (P-value) of the slope, the slope (m) and the intercept (k) of the gm versus Ci curves, as well as the % difference (% D) in gm at Ci of 200 -1 mCi+k and 300 μmol mol , in the replicate leaves of the seven Banksia species examined at two PPFDs. The regression equation is gm = e . A subset of -1 Ci values between 180 and 450 μmol mol was considered from each curve. Species are shown in order of increasing leaf dry mass per area (LMA).

PPFD (μmol quanta m-2 s-1) 500 1500 Species P-value m k % D P-value m k % D B. serrata 1.64E-06*** -1.93E-03 -2.038 -17.6 8.73E-05*** -1.72E-03 -1.593 -15.8 1.76E-05*** -1.62E-03 -1.751 -15.0 3.35E-05*** -1.32E-03 -1.551 -12.4 na na na na 1.09E-08*** -4.24E-03 -0.588 -34.6 B. integrifolia 7.57E-03** -9.35E-04 -1.699 -8.9 2.90E-03** -1.24E-03 -1.489 -11.7 3.37E-05*** -2.18E-03 -1.359 -19.6 1.07E-05*** -2.35E-03 -1.166 -20.9 2.33E-04*** -1.78E-03 -1.293 -16.3 1.92E-05*** -2.48E-03 -1.039 -22.0 B. attenuata 5.09E-05*** -2.05E-03 -1.594 -18.5 8.78E-04*** -1.46E-03 -1.357 -13.6 2.59E-03** -2.00E-03 -1.195 -18.1 3.19E-05*** -1.82E-03 -1.222 -16.6 1.33E-05*** -2.09E-03 -1.690 -18.9 3.60E-05*** -1.93E-03 -1.348 -17.6 B. solandri 0.356 ns -2.02E-04 -1.832 -2.0 5.50E-04*** -3.08E-03 -0.772 -26.5 0.297 ns 1.64E-04 -1.819 1.7 4.71E-05*** -2.20E-03 -1.312 -19.7 0.799 ns -9.54E-05 -1.712 -0.9 6.49E-06*** -2.02E-03 -1.163 -18.3 B. paludosa 0.640 ns -2.54E-04 -1.630 -2.5 2.24E-07*** -2.95E-03 -1.037 -25.5 0.300 ns 3.88E-04 -2.167 4.0 2.41E-05*** -2.58E-03 -1.256 -22.7 3.18E-05*** -1.10E-03 -2.045 -10.4 1.69E-04*** -2.20E-03 -1.453 -19.7 B. repens 5.87E-05*** -1.25E-03 -2.586 -11.8 3.95E-03** -1.18E-03 -2.084 -11.1 2.26E-03** -1.08E-03 -2.261 -10.2 4.07E-02* -8.21E-04 -1.997 -7.9 3.69E-02* -9.56E-04 -2.348 -9.1 3.76E-02* -6.43E-04 -2.010 -6.2 B. elderiana na na na na 1.40E-05*** -2.28E-03 -1.868 -20.4 na na na na 1.09E-02* -1.56E-03 -1.925 -14.4 na na na na 1.98E-03** -2.38E-03 -1.729 -21.2

***: P<0.001, **: P<0.01, *: P<0.05; ns: not significant slope; na: not analysed, due to some negative gm values being calculated, which prevented this data set from being considered. [190]

APPENDIX 3

Table S3. Sensitivity analysis of the effect of respiration in the light (Rd) and the photo- compensation point in the absence of mitochondrial respiration (Γ*) on the estimation of mesophyll conductance (gm) in the Banksia species examined. The % difference ± standard error between gm -2 -1 -2 -1 calculated assuming a Rd of 0.8 μmol m s and a Rd of 1.6 or 2.4 μmol m s is presented in columns “Doubling Rd” or “Tripling Rd”, respectively. Similarly, the % difference ± standard * -1 error between gm calculated assuming a Γ , before the temperature correction, of 36.9 μmol mol and a Γ* of 28 or 40 μmol mol-1 is presented in columns “Γ* 28 – 36.9” or “Γ* 36.9 – 40”, respectively. A negative percentage indicates reduction of gm. The ecologically meaningful range of intercellular CO2 concentration (Ci) is highlighted in bold.

-1 * * Ci (μmol mol ) Doubling Rd Tripling Rd Γ 28 – 36.9 Γ 36.9 – 40 PPFD 500 μmol quanta m-2 s-1

Ci < 100 41.1 ± 15.3 104.7 ± 33.8 -38.2 ± 3.6 39.5 ± 11.2

100 ≤ Ci < 200 13.7 ± 2.5 40.0 ± 10.5 -31.0 ± 2.0 21.3 ± 3.0

200 ≤ Ci < 300 5.9 ± 0.8 12.9 ± 1.8 -25.2 ± 1.8 14.3 ± 1.5

300 ≤ Ci < 400 5.2 ± 0.8 11.2 ± 1.8 -25.5 ± 2.4 15.3 ± 2.2

Ci ≥ 400 6.2 ± 1.4 14.1 ± 3.4 -28.3 ± 3.6 22.4 ± 5.4 PPFD 1500 μmol quanta m-2 s-1

Ci < 100 29.6 ± 7.2 81.8 ± 22.0 -38.9 ± 2.9 47.9 ± 19.6

100 ≤ Ci < 200 13.2 ± 2.9 30.4 ± 6.0 -33.0 ± 2.0 27.1 ± 6.3

200 ≤ Ci < 300 3.7 ± 0.6 7.9 ± 1.2 -22.5 ± 1.1 11.7 ± 1.0

300 ≤ Ci < 400 2.1 ± 0.2 4.3 ± 0.5 -18.0 ± 0.8 8.4 ± 0.5

Ci ≥ 400 1.6 ± 0.2 3.2 ± 0.4 -15.6 ± 0.8 7.0 ± 0.5

[191]