<%BANNER%>

Genetic Analysis of Stem Hydraulic Properties, Foliar Carbon Isotope Composition, and Growth in a Pseudo-Backcross Popul...

Permanent Link: http://ufdc.ufl.edu/UFE0021871/00001

Material Information

Title: Genetic Analysis of Stem Hydraulic Properties, Foliar Carbon Isotope Composition, and Growth in a Pseudo-Backcross Population of Populus
Physical Description: 1 online resource (89 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: carbon, conductivity, huber, hydraulic, isotope, populus, sapwood, vessel
Forest Resources and Conservation -- Dissertations, Academic -- UF
Genre: Forest Resources and Conservation thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: There are many genetic studies on wood properties, growth and disease in Populus, but few genetic investigations of hydraulic properties and water-use efficiency. Because of the interactions between xylem hydraulic conductivity and leaf gas exchange, high hydraulic conductivity is hypothesized to be a prerequisite for rapid growth in forest trees (Tyree 2003). Two studies were conducted to examine genetic and phenotypic variation and map quantitative trait loci (QTL) for stem hydraulic properties, growth, foliar stable carbon isotope composition (delta13C), and interactions among these traits for 100 clonally propagated progeny genotypes (and the parents) of a pseudo-backcross population of Populus deltoides and P. trichocarpa x P. deltoides. Mean hydraulic vessel diameter (Dh), vessels per sapwood area (VSA), and hydraulic conductivity were calculated from image analysis of stem cross-sections. Plant biomass, growth increment, foliar delta13C and nutrient content were also quantified. QTL for growth and physiological traits were identified using composite interval mapping performed on separate single-tree maps of the mother and father of the progeny population. Stem diameter increment was positively genetically and phenotypically correlated with leaf specific hydraulic conductivity. Growth traits, specific hydraulic conductivity and xylem anatomy traits were moderately heritable, while delta13C was not heritable. Strong genetic correlations between hydraulic and growth traits suggest they share genes or biological pathways. Significant QTL were identified for all traits examined except Huber value, delta13C, foliar percent carbon, and root to shoot ratio. Co-localizations of QTL for hydraulic traits and growth increment provide a causative linkage and further support for Tyree?'s (2003) hypothesis. Overall, the genetic analysis revealed QTL that co-localize for multiple traits, suggesting loci with pleiotropic effects. This research is the first to quantify genetic control and to map QTL for hydraulic conductivity in an angiosperm tree species.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Peter, Gary F.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021871:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021871/00001

Material Information

Title: Genetic Analysis of Stem Hydraulic Properties, Foliar Carbon Isotope Composition, and Growth in a Pseudo-Backcross Population of Populus
Physical Description: 1 online resource (89 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: carbon, conductivity, huber, hydraulic, isotope, populus, sapwood, vessel
Forest Resources and Conservation -- Dissertations, Academic -- UF
Genre: Forest Resources and Conservation thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: There are many genetic studies on wood properties, growth and disease in Populus, but few genetic investigations of hydraulic properties and water-use efficiency. Because of the interactions between xylem hydraulic conductivity and leaf gas exchange, high hydraulic conductivity is hypothesized to be a prerequisite for rapid growth in forest trees (Tyree 2003). Two studies were conducted to examine genetic and phenotypic variation and map quantitative trait loci (QTL) for stem hydraulic properties, growth, foliar stable carbon isotope composition (delta13C), and interactions among these traits for 100 clonally propagated progeny genotypes (and the parents) of a pseudo-backcross population of Populus deltoides and P. trichocarpa x P. deltoides. Mean hydraulic vessel diameter (Dh), vessels per sapwood area (VSA), and hydraulic conductivity were calculated from image analysis of stem cross-sections. Plant biomass, growth increment, foliar delta13C and nutrient content were also quantified. QTL for growth and physiological traits were identified using composite interval mapping performed on separate single-tree maps of the mother and father of the progeny population. Stem diameter increment was positively genetically and phenotypically correlated with leaf specific hydraulic conductivity. Growth traits, specific hydraulic conductivity and xylem anatomy traits were moderately heritable, while delta13C was not heritable. Strong genetic correlations between hydraulic and growth traits suggest they share genes or biological pathways. Significant QTL were identified for all traits examined except Huber value, delta13C, foliar percent carbon, and root to shoot ratio. Co-localizations of QTL for hydraulic traits and growth increment provide a causative linkage and further support for Tyree?'s (2003) hypothesis. Overall, the genetic analysis revealed QTL that co-localize for multiple traits, suggesting loci with pleiotropic effects. This research is the first to quantify genetic control and to map QTL for hydraulic conductivity in an angiosperm tree species.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (M.S.)--University of Florida, 2008.
Local: Adviser: Peter, Gary F.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021871:00001


This item has the following downloads:


Full Text

PAGE 1

1 GENETIC ANALYSIS OF STEM HYDRAULIC PROPERTIES, FOLIAR CARBON ISOTOPE COMPOSITION AND GROWTH IN A PSEUDO BACKCROSS POPULATION OF Populus By BRIANNA L. MILES A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FL ORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 200 8

PAGE 2

2 200 8 Brianna L. Miles

PAGE 3

3 T o my grandfather, Gordon Smith

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank Drs. Gary Peter, Tim Marti n, Matias Kirst and Jeannine Cavender Bares for serving on my advisory committee. In particular, I want to express my gratitude to Dr. Peter for the countless hours spent reading and revising the many versions of my thesis, as well as helpful discussions of my research and valuable advice. Dr. Martin I wish to thank for his candid and lucid discussions of physiological processes, help with interpreting results, and for his great sense of humor. Thanks also go to Dr. Kirst for his assistance with understa nding the genetic perspective of my research, and the invaluable day to day encouragement he provided. I am especially indebted to Dr. Cavender Bares for motivating me to continue my formal education and for being a wonderful academic mentor and friend. I owe many people in the Forest Genomics and Tree Ecophysiology Labs a great deal of thanks for providing help with collecting and processing the large amount of data demanded by my project. For this, I would like to particularly thank Chris Dervinis, Dere k Drost, Emerita Golden, Carlos Gonzales, Xiabo Li Evandro Novaes, Dave Nolletti, Luis Osorio, as well as two great undergraduate assistants, Daniel Lambert and Jessie Wilson. I also want to thank Dr. Dudley Huber for his time and patience with explainin g statistical concepts and help with the design of the genetic analysis. This research was made possible by the financial support provided by the USFS Agenda 2020 # SRS 03 CA 11330136 245 and DOE Genomic Mechanisms of Carbon Allocation and Partitioning in Poplar #ER64114 1026645 0011741 Special thanks are extended to the great friends I have been blessed with during my time at UF: Chris Dervinis, Derek Drost, Greg Gorman, Evandro Novaes, Tania Quesada, Greg Starr, Gustavo Ramirez, Veronica Tienza Sanchez and Kathy Smith. Sharing laughs and beers with them was one of the best parts of my time at UF I also must acknowledge and give heartfelt thanks to my family: my parents Kris and Gini Smith, my in laws, Warren and Paula Miles

PAGE 5

5 Derek Smith, Ryan Miles, Michelle Iturrate, my grandparents and my best friend Toni Ann Herwig, for their continuing support of my life adventures. Also, I want to thank my dog, Paws, who always knows when I need him. Last, but certainly not least, I thank my husband, David Mil es, for many years of love, friendship and loyalty. I owe my success to his generous support during my academic career and to his amazing job with harvesting and processing samples.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDG MENTS ................................ ................................ ................................ ........... 4 LIST OF TABLES ................................ ................................ ................................ ...................... 8 LIST OF FIGURES ................................ ................................ ................................ .................... 9 ABSTRACT ................................ ................................ ................................ ............................. 10 CHAPTER 1 REVIEW OF TREE PHYSIOLOGY AND GENETICS: STEM HYDRAULIC PROPERTIES, FOLIAR CARBON ISOTOPES AND GROWTH ................................ ..... 1 2 Introduction ................................ ................................ ................................ ........................ 12 Plant Stem Hydraulic Properties ................................ ................................ ......................... 13 Water use Efficiency ................................ ................................ ................................ .......... 15 Water use Ef ficiency and Carbon Isotopes ................................ ................................ ......... 16 Role of Stem Hydraulics and Water use Efficiency in Tree Growth ................................ ... 18 Populus Background ................................ ................................ ................................ .......... 21 Stem Hydraulic Properties and Water use Efficiency in Populus ................................ ........ 22 Tree Physiology and Growth in a Genetic Context ................................ ............................. 25 Conclusion ................................ ................................ ................................ ......................... 28 2 GENETIC CONTROL OF STEM HYDRAULIC PROPERTIES AND CORRELATIONS WITH GROWTH IN Populus ................................ .............................. 30 Introduction ................................ ................................ ................................ ........................ 30 Materials and Methods ................................ ................................ ................................ ....... 34 Plant Material, Propagation and Growth ................................ ................................ ...... 34 Seedling Harvest, Biomass and Growth Measurements ................................ ............... 35 Foliar Carbon Isotope Composition ................................ ................................ ............. 36 Hydraulic Conductivity and Xylem Vessel Measurements ................................ ........... 36 Statistical Analysis ................................ ................................ ................................ ...... 39 Results ................................ ................................ ................................ ............................... 40 Measured and Theoretical Conductivity ................................ ................................ ...... 40 Variation in Growth, Hydraulic, and Physiological Traits ................................ ............ 42 Genetic Control of Growth, Hydraulic, and Physiological Traits ................................ 43 Phenotypic and Genetic Correlations ................................ ................................ ........... 43 Discussion ................................ ................................ ................................ .......................... 44 Genetic Control of Hydraulic Conductivity ................................ ................................ 44 Genetic Correlations between Xylem Anatomy, Hydraulic Conductivity and Growth ................................ ................................ ................................ .................... 45 Carbon Isotope Composition and Leaf Nutrient Content ................................ .............. 47 Specific Leaf Area, Huber Value, Hydraulics and Growth ................................ ........... 47

PAGE 7

7 Conclusion ................................ ................................ ................................ ......................... 48 3 QUANTITATIVE TRAIT LOCI ANALYSIS OF STEM HYDRAULIC PROPERTIES, FOLIAR CARBON ISOTOPES AND GROWTH IN Populus ................................ ........... 54 Introduction ................................ ................................ ................................ ........................ 54 Materials and Methods ................................ ................................ ................................ ....... 56 Plant material ................................ ................................ ................................ .............. 56 Experimental Design ................................ ................................ ................................ ... 57 Plant Propagation and Growth ................................ ................................ ..................... 57 Pesticide Treatments ................................ ................................ ................................ ... 58 Genotype Selection ................................ ................................ ................................ ..... 58 Seedling Harvest, Biomass and Growth Measurements ................................ ............... 59 Fol iar Carbon Isotope Composition ................................ ................................ ............. 60 Hydraulic Conductivity and Xylem Vessel Measurements ................................ ........... 60 Statistical Analysis ................................ ................................ ................................ ...... 62 Heritability ................................ ................................ ................................ .................. 62 Quantitative Trait Analysis ................................ ................................ .......................... 63 Results ................................ ................................ ................................ ............................... 63 Variation in Growth, Hydraulic, and Physiological Traits ................................ ............ 63 Scaling Growth, Hydraulic, and Physiological Traits ................................ ................... 64 Heritability of Growth, Hydraulic, and Physiological Traits ................................ ........ 64 Quantitative Trait Loci Results ................................ ................................ .................... 65 Discussion ................................ ................................ ................................ .......................... 66 Plant Hydraulics and Growth ................................ ................................ ....................... 66 Carbon Isotope Composition and Leaf Nutrient Concentration ................................ .... 68 Specific Leaf Area, Huber Value, Hydraulics and Growth ................................ ........... 68 Conclusion ................................ ................................ ................................ ......................... 69 4 SUMMARY ................................ ................................ ................................ ....................... 77 LIST OF REFERENCES ................................ ................................ ................................ .......... 79 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ..... 89

PAGE 8

8 LIST OF TABLES Table page 2 1 Trait means, standard errors ( SE ) and significance levels for 22 progeny genotypes and parental genotypes, Populus trichocarpa x P. deltoides ( P. t x d ) and P. deltoides ......... 49 2 2 Trait within family broad sense heritabilities (diagonal, bold type), genetic correlations (above diagonal), and phenotypic correlations (below diagonal) for progeny genotypes. ................................ ................................ ................................ ..................... 50 3 1 Trait means, standard errors ( SE ), significance levels and ranges for 100 progeny genotypes and parental genotypes, Populus trichocarpa x P. deltoides ( P. t x d ) and P. deltoides ................................ ................................ ................................ .................... 70 3 2 Trait within family broad sense heritabilities ( ) and significant QTL detected in the mother, Populus trichocarpa x P. deltoides ( P. t x d ), and father, P. deltoides maps. ................................ ................................ ................................ ............................. 71

PAGE 9

9 LIST OF FIGURES Figure page 1 1 A c i cur ve ................................ ................................ ................................ ........................... 29 2 1 Leaf specific hydraulic conductivity ( K L ) scales positi vely with diameter increment ( D INC ).. ................................ ................................ ................................ .......................... 51 2 2 Relationships between hydraulic vessel diameter ( D h ), vessels per sapwood area ( VSA ) and vessel area per sapwood area ( VA:SA ) ................................ ................................ ..... 52 2 3 Hydraulic vessel diameter ( D h ), open circles, scales positively ( r 2 = 0.7196, P <0.0001) and vessels per sapwood area ( VSA ), filled circles, scales negatively ( r 2 = 0.6280, P <0.0001) with diameter increm ent ( D INC ) ................................ ................................ .... 53 2 4 Specific leaf area ( SLA ) scales negatively with total biomass ( BIO ) ................................ .... 53 3 1 Leaf specific hydraulic conductiv ity ( K L ) scales positively with diameter increment ( D INC ) ................................ ................................ ................................ ............................ 72 3 2 Hydraulic vessel diameter ( D h ), open circles, scales positively ( r 2 = 0.4172, P <0.0001) and vessels per sapwood area ( VSA ), f illed circles, scales negatively ( r 2 = 0.1510, P <0.0001) with diameter increment ( D INC ) ................................ ................................ .... 72 3 3 Specific leaf area ( SLA ) scales negatively with total biomass ( BIO ) ................................ .... 73 3 4 QTL located in the Father Map ................................ ................................ .......................... 74 3 5 QTL located in the Mother Map ................................ ................................ ......................... 75 3 6 Signif icant (P<0.05) QTL for hydraulic conductivity and growth increment traits on linkage group (LG) 1 in the Mother Map ................................ ................................ ....... 76

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in P artial Fulfillment of the Requirements for the Degree of Master of Science GENETIC ANALYSIS OF STEM HYDRAULIC PROPERTIES, FOLIAR CARBON ISOTOPE COMPOSITION AND GROWTH IN A PS EU DO BACKCROSS POPULATION OF Populus By Brianna L. Miles May 2008 Chair: G ary F. Peter Major: Forest Resources and conservation There are many genetic studies on wood properties, growth and disease in Populus but few genetic investigations of hydraulic properties and water use efficiency Because of the interactions between xylem hydraulic conductivity and leaf gas exchange, high hydraulic conductivity is hypothesized to be a pre requisite for rapid growth in forest trees (Tyree 2003). Two studies were conducted to examine g en etic and phenotypic variation and map quantitativ e trait loci (QTL) for stem hydraulic properties, growth, foliar stable carbon isotope discrimination ( 13 C) and interactions among these traits for 100 clonally propagated progeny genotypes (and the parents) of a pseudo backcross population of Populus de ltoides and P. trichocarpa x P. deltoides Mean hydraulic vessel diameter ( D h ), vessels per sapwood area ( VSA ) and hydraulic conductivity were calculated from image analysis of stem cross sections. Plant biomass, growth increment and foliar 13 C and n utrient content were also quantified. QTL for growth and physiological traits were identified using composite interval mapping performed on separate single tree maps of the mother and father of the progeny population.

PAGE 11

11 Stem diameter increment was positivel y genetically and phenotypically correlated with leaf specific hydraulic conductivity. Growth traits, specific hydraulic conductivity and xylem anatomy traits were moderately heritable, while 13 C was not heritable. Strong genetic correlations between hydraulic and growth traits suggest they share genes or biological pathways. Significant QTL were identified for all traits examined except Huber value 13 C foliar percent carbon, and root to s hoot ratio. Co localizations of QTL for hydraulic traits and Overall, the genetic analysis revealed QTL that co localize for multiple traits, suggesting loc i with pleiotropic effects. This research is the first to quantify genetic control and to map QTL for hydraulic conductivity in an angiosperm tree species.

PAGE 12

12 CHAPTER 1 REVIEW OF TREE PHYSI OLOGY AND GENETICS: STEM HYDRAULIC PROPE RTIES, FOLIAR CARBON ISOTOP ES AND GROWTH Introduction Early photosynthetic organisms had a very simple structure and lived in an aquatic environment. This aqueous medium provided all nutrients essential for growth, as well as dissolved carbon dioxide for photosynthesis. As plants evo lved, moving from the aquatic environment to the terrestrial environment necessitated the development of specialized tissues to avoid water stress. Plants developed roots and stems which transport water and nutrients from the soil to the leaves that were protected by a waxy cuticle and have specialized pores for gas exchange These specializ ations helped prevent excessive water loss during leaf gas exchange necessary for photosynthesis. Thus, the trade off between traits minimizing water loss and those m aximizing photosynthesis, is a key issue in plant evolution and agronomic plant breeding for water use efficiency as water resources become increasingly scarce. Research addressing the genetic mechanisms of plant physiological processes is necessary for ex panding our understanding of plant function (Ackerly & Monson, 2003) This research has wide ranging consequences for agriculture, timber production, forest and ecosystem management. Forest trees have an important role in addressing climate change by providing ecosystem services such as carbon sequestra tion, and in the development of biofuels. A better understanding of tree physiological processes in a genetic framework is necessary to fully realize the potential of trees as a tool for mitigating greenhouse gas emissions. T he following pages review the literature on (1) relationships among stem hydraulic properties, foliar carbon isotopes, and growth; and (2) genetic control of these traits in trees, with special emphasis on the genus Populus

PAGE 13

13 Plant Stem Hydraulic Properties The physiological mechanisms regulating stem hydraulic properties, i.e. the movement of water and nutrients from roots to leaves, in woody plants have been the focus of much research (Lambers et al. 1998; Tyre e & Zimmerman, 2002) The m echanisms include hydraulic conductivity of xylem, stomatal conductance and xylem an atomy. The Cohesion Tension theory explains the movement of sap through the xylem based on the cohesive properties of water and the tension cr eated in the xylem by evaporation at the stomata on leaf surfaces during transpiration (Tyree, 1997) The tension between water molecules propagates along the continuous water column from the l eaf to the soil, causing the sap to move through the xylem vessels. The hydrogen bonding between adjacent water molecules (cohesion) keep the molecules toge ther as a continuous chain. This cohesion tension theory of sap flow in trees was formulated in 1894 by H.H Dixon and J. Joly (Pickard, 1981) The cohesion tension theory has been criticized by some (Canny, 1995; Zimmermann et a l. 2004) but is generally accepted as the domi nant theory explaining sap ascent in plants (Angeles et al. 2004) Stem hydraulic conductivity ( K h m 4 Pa 1 s 1 ) can be described by the Hagen Poiseuille equation (Equation 1 1), where r is the radius of the condu it (m), and is the dynamic viscosity of liquid (Pa s) at a given temperature. (1 1 ) This equation assumes that xylem conduits act as ideal capillary tubes with laminar flow (Tyree & Zimmerman, 2002) In general, the flow rate is proportional to the fourth power of the conduit radius, meaning sma ll increases in vessel diameter lead to substantial increases in conductivity (Tyree & Zimmerman, 2002) However, bigger vessels contain more dissolved gases and greater vessel diameter increases the susceptibility of xylem conduits to occlusion by a gas bubble, i.e.

PAGE 14

14 the formation of embolism, resulting from the nucleation of a gas bubble or release of existing bubbles (Sperry et al. 1994; Holbrook & Zwieniecki, 2005) In angiosperms, maximum c onductivity depends on vessel number and xylem conduit diameter but e mbolism inter vessel pit membrane morphology and vulnerability to drought induced cavitation may si gnificantly affect actual stem hydraulic conductivity (Hacke & Sperry, 2001; Sperry, 2003) Hydraulic conductivity is most often reported on a leaf ( K L K h /leaf area) or sapwood specific ( K S K h /sapwood area) basis. Compared to gymnosperms, a ngiosperms generally have higher and wider ranges of values for K L and K S ranging from 10 4 to 10 2 kg s 1 m 1 MPa 1 and 1 to 100 kg s 1 m 1 MPa 1 respectively; gymnosperms have K L values from 8x10 5 t o 3x10 4 kg s 1 m 1 MPa 1 and K S from 1 2 kg s 1 m 1 MPa 1 (Sperry, 2003) K h as well as percent loss of hydraulic conductivity ( PLC ) can be estimated using a low pressure flow method developed by Sperry et al. (1988) Stem segmen ts are fitted in to a tubing apparatus containing the perfusion solution which flows through the stems resulting from a pressure gradient (3 10 kPa P ) generated by the height difference between a raised reservoir and a lower reservoir. The flow throu gh is collected in the lower reservoir, which located on a digital balance to measure flow rate, F (kg s 1 ). K h is estimated as F multiplied by the length of the stem, and divided P However, the low pressure flow method is time prohibitive for experime nts with large sample sizes, given that stems need to be processed the same day they are harvested. Theoretical estimation of K h by the Hagen Poiseille equation is an alternative method. Theoretical conductivity ( K t ), calculated as the sum of the fourth power of vessel diameters, is commonly seen in the literature as a substitute for direct measurement of K h (Lewis & Boose, 1995; Lo Gullo et al. 1995; Villar Salvador et al. 1997; Solla & Gil, 2002) Most studies that use the theoretical method do not provide a comparison of results with the low

PAGE 15

15 pressure flow method. Howe ver, strong linear corre lations between measured and theoretical estimates of K h based on vessel anatomy have been demonstrated with theoretical typically being higher than measured values (Lo Gullo, 1991; Martre et al. 2000; Christensen Dalsgaard et al. 2007) For example, i n Festuca arundinacea leaf xylem, K h and K t were linearly correlated ( r = 0.86, P <0.05), but K t was 36 4.5% greater than K h (Martre et al. 2000) The main limitations of the theoretical method are that only maximum K h can be estimated because the extent of embolism is not known; and K t does not reflect pit mem brane conductivity, therefore two plants with the same K t could have different pit conductivity or embolism resulting in different K h Water use E fficiency When stomata open for carbon dioxide uptake, water is lost due to the water potential gradient be tween the leaf and the air, simultaneously creating the tension necessary for water transport. Soil water availability, xylem hydraulic conductivity and evaporative demand of the ambient air influences leaf water status, and can impose a gas exchange limi tation through stomatal control. Stomatal control therefore has a strong influence on carbon gain and growth of the plant. Water use efficiency ( WUE ) is an important ecological and agronomic trait affecting plant water status and is defined as the ratio of photosynthesis, or carbon assimilation, ( A ) to transpiration ( E ) rate. WUE has important implications for plant performance, particularly in water limited environments, but also when stomatal conductance ( g s ) is physiologically limited by the costs of water uptake and transport (Sperry et al. 2002) At the scale of a photosynthesizing leaf, WUE can be represented by carbon and water vapor diffusion gradients through stomata between the atmosphere and leaf intercellular spaces

PAGE 16

16 (1 2 ) In E quation 1 2 g CO2 and g H2O are stomatal conductances to CO 2 and H 2 O, respectively, c and w refer to concentrations of CO 2 and H 2 O, respectively, subscripts a and i are atmospheric and intercellular pools, respectively, and 1.6 is the ratio of the molecular diffusivity of CO 2 and H 2 O in air (Comstock et al. 2005) Since c a w a and w i generally can not be control led by the plant, c i use efficiency, largely determine s WUE In turn, c i depends on the CO 2 diffusion gradient created by stomatal opening and the carboxylation capacity of the leaf. Carboxylation, the binding of CO 2 to a CO 2 acceptor molecule, is the first step in carbon fixation during photosynthesis, and is catalyzed by r ibulosebisphosphate carboxylase oxygenase (Rubisco) in C 3 plants (Lambers et al. 1998) When plants are under water stress, they will reduce g s in order to maximize WUE (Lambers et al. 1998) An increase in WUE demands a decrease in c i resulting in a potential r ibulosebisphosphate (RuBP) limitation of the carboxylation reaction, and a strong correlation between maximum g s and maximum photosynthetic rate (Wong et al. 1979) Figure 1 1 describes the supply and demand relationship between carbon assimilation ( A ), intercellular CO 2 concentra tion, c i and stomatal conductance ( g s ). While carboxylation is RuBP saturated, A increases linearly with increases in c i (demand function), but A quickly plateaus as carboxylation becomes RuBP limited. The slope of the supply function relating c i and c a determines g s W ater use Efficiency and Carbon Isotopes An isotope is an atom whose nuclei contain the same number of protons but a different number of neutrons. Isotopes may be stable or unstable (radioactive isotopes) and usually one isotope is predomi nantly abundant. For example, the average atmospheric abundance of 12 C is 98.89%, while the average atmospheric abundance for 13 C is 1.11% (Griffiths, 1991) The

PAGE 17

17 isotope composition of the atmosphere is different from that of plant biomass, which allows the utilization of isotope composition in plants to examine physiological proc esses. Stable carbon isotopes provide a time integrated measure of plant physiological processes and plant environment interaction (Farquhar et al. 1989) Fractionation of the two most abundant carbon isotopes, 13 C and 12 C, during photosynthesis create an isot opic ratio ( 13 C/ 12 C) of assimilated carbon in plant tissues Stable carbon isotope composition ( 13 C expressed per mil, Equation 1 3 ) is the ratio of 13 C and 12 C, in the plant sample ( R sample ) compared to the same ratio in the international standard ( R standard h Carolina (Farquhar et al. 1989) ( 1 3 ) Diffusional and enzymatic fractionations are the two main processes that determine composition. Lighter 12 C containing CO 2 molecules diffuse more readily into the leaf than 13 C containing CO 2 molecules, resulting in a small diffusional frac tionation of 4.4 (Comstock et al. 2005) The majority of the fractiona tion (generally 27 CO 2 molecules containing 12 C over 13 C during carboxylation, resulting in discrimination against the heavier carbon isotope (Lambers et al. 1998) This relationship is represented by Equation 1 4, where a and b are discrimination constants for the fractionat ion due to diffusion in air (4.4 ) and carboxylation by Rubisco (27 ), respectively (Farquhar et al. 1982) ( 1 4 ) Carbon isotope discrimination ( ) can then be calculated from 13 C values (Equation 1 5), where a is the composition of the atmosphere (assumed to be 8 p is the composition of the plant sample (Farquhar et al. 1989)

PAGE 18

18 ( 1 5 ) Less discrimination against 13 C corresponds to lower and less negative 13 C Fa r quhar and colleagues pioneered the idea that 13 C and in plant foliar tissue could be used as a measure of WUE integrated over the life of the leaf as is negatively correlated with WUE (Farquhar et al. 1982; Fa rquhar & Richards, 1984) Since then, several studies have shown that variation in closely follows variation in the ratio c i /c a (reviewed in Brugnoli & Farquhar, 2000) Overall, variations in c i account for most variation in 13 C and Moreover, variation in photosynthetic capacity or stomatal conductance may cause variation in leaf 13 C by changing c i (Condon et al. 2004) Abiotic environmental variables such as climate, elevation and salinity as well as pl ant species, origin and age can also affect 13 C (Fa rquhar et al. 1989) Also, 13 C will vary among plants with different photosynthetic pathways, i.e. C 3 plants, which have Rubisco as their photosynthetic enzyme, as opposed to C 4 and crassulacean acid metabolism (CAM) plants, which have PEP carboxylase. In general, 13 C values will be much lower (more negative) for C 3 plants than C 4 and CAM plants, which show relatively little discrimination against 13 C (O'leary, 1988; Ehleringer & Osmond, 1989) Generally, 13 C values range from 36 to 23 for C 3 plants, and 10 to 18 for C 4 / CAM plants (Griffiths, 1991) Accordingly, in C 3 plants will range from 15 to 28, and from 2 to 10 in C 4 / CAM Role of Stem Hydraulics and Water use Efficiency in Tree Growth Photosynthesis, carbon allocation, growth, transpiration and hydraulic conductivity are interconnected. The xylem hydrauli c pathway from the roots to the crown must be sufficient to meet the transpirational demand of the leaves. The p ipe model (Shinozaki et al. 1964) describes

PAGE 19

19 the tree structure as a bundle of pipes in parallel which provide a pathway from roots to leaves and a constant quantity of conducting pipes supplies a given leaf area with water. The model also predicts a linear relationship between sapwood area and leaf mass or area and assumes this relationship remains constant throughout plant development. The a llocation betw een total leaf area ( LA ) and tree basal area or sapwood area (SA) conceptualized within the pipe model, is described by the ratio LA:SA and called the Huber value for the original work by Huber (Huber, 1928; Waring et a l. 1982; Mcdowell et al. 2002) LA:SA has been found to depend on site water balance, such as evaporative demand and soil water availability. A simple hydraulic at under conditions of increased evaporative demand plants may compensate by reducing LA:SA to prevent extreme water potential gradients that may reduce canopy conductance (Whitehead & Jarvis, 1981) A lso, a s height increases p lants can avoid severe xylem tension by decreasing LA:SA (Mcdowell et al. 2002) thereby increas ing leaf specific hydraulic conductivit y and decreas ing the maximum water tensions in the canopy. Due to the relationships between hydraulic conductivity and stomatal conductance, and between photosynthesis rates and growth, foliar 13 C may be expected to vary with hydraulic conductivity and plant growth rate. WUE influences the balance between xylem safety and efficiency and therefore is also expected to affect xylem structure and function (Sperry & Hac ke, 2002) In fact, i t has been proposed that hydraulic properties of xylem c an constrain leaf gas exchange, resulting in a larger WUE for individuals with smaller vessel cross section al area (Ponton et al. 2001) However, observed r elationships between WUE and growth have varied. In pine, correlations between 13 C and growth have been shown to be negative, positive, or uncorrelated, depending on species and environmental conditions (Brend el et al. 2002; Prasolova et al. 2003) L ack of a relationship between WUE and growth indicate s that variation

PAGE 20

20 in WUE is due to differences in stomatal conductance rather than xylem hydraulic properties (Rae et al. 2004; Marron et al. 2005; Monclus et al. 2005) whereas a negative or positive relationship suggests photosynthetic capacity drives WUE (Farquhar et al. 1989; Osorio & Pereira, 1994) Work on Acer saccharum by Yang and Tyree (199 3) provided evidence that shoot hydraulic architecture affects stomatal regulation and consequently may limit gas exchange. They concluded that as trees grow larger, the change in leaf water potential should limit stomatal conductance. Because of the in teractions between xylem hydraulic conductivity and leaf gas exchange, Tyree (2003) hypothesized that high hydraulic conductivity is actually a pre requisite for rapid growth in forest trees. By this reasoning, trees with inherently low hydraulic conducti vity will tend to experience more extreme within tree water potential gradients and more frequent stomatal closure, leading to lowered photosynthetic gas exchange and, in turn, reduced carbon gain and growth rates. Previous studies showed stem hydraulic c onductivity and growth rates were correlated between tree s pecies but not within species (Tyree et al. 1991; Machado & Tyree, 1994) A strong linear relationship between stem hydraulic conductivity and leaf photosynthetic capacity has also been observed across species, suggesting stem hydraulic conductivity constrains l eaf maximum photosynthetic rate (Brodribb & Field, 2000; Brodribb et al. 2002; Campanello et al. 2008) One study of dwarf and tall tree forms of Rhizophora mangle showed high plasticity of conductivity, vessel anatomy and growt h in response to nutrient additions, and the authors suggest hydraulic properties are key controls of growth (Love lock et al. 2006) Stem and root hydraulic conductivity were also correlated with growth potential in grafted apple trees ( Malus pumila ) (Atkinson et al. 2003) Th ese correlations between growth, hydraulic conductivity and leaf photosynthetic capacity across species suggests

PAGE 21

21 that some level of genetic control. Howev er, few studies have examined within species or genetic variation in hydraulic conductivity relative to growth in forest trees using a genetic approach. Populus Background The genus Populus (Salicaceae) which includes poplars, aspens and cottonwoods, is r epresented by approximately 29 species having distributions ranging across much of the northern hemisphere (Stettler et al. 1996) Populus species are ecol ogically and commercially important. They provide habitat and re colonize disturbed sites in their native ecosystems (Stettler et al. 1996; Williams & Cooper, 2005) Populus trees are also valued by the forestry industry, because of their fast growth and light colored wood. Managed stands, including Populus are being rec ognized as a key component of the carbon balance equation due to its expected carbon sequestration and increased growth responses to a global rise in atmospheric CO 2 (Gielen & Ceulemans, 2001; Taylor, 2002) Hybrid Populus has been used successfully in the commercial market as veneer, low grade lumber, pallets, formwork, wood composite, and fuel. Populu s are fast growing and can be vegetatively propagated, traits highly desirable to the forestry industry. Unfortunately, high productivity comes at the cost of a large water demand (Tschaplinski et al. 1994; Harvey & Van Den Driessche, 1997; Rood et al. 2003) Hybrid Populus plantations that are well watered with drip irrigation are fast growing, despite seasonal water deficits. Howeve r, due to their high demand for water, Populus spp. are not widely used as a crop tree. Populus is considered a model tree for studying growth, fundamental physiology and genetics (Stettler et al. 1996; Taylor, 2002) Th ere are several Populus species that are indigenous to North America, including Populus deltoides and P. trichocarpa from eastern and western North America, respectively These species differ greatly in geographic distribution as

PAGE 22

22 well as leaf morphology (Burns & Honkala, 1990) The adaptive range of P. trichocarpa is generally wetter and also slightly cooler than P. deltoides with a little overlap at moderat e moisture and temperate (Rood et al. 2003) This study considers the pse udo backcross of P. deltoides with an F1 hybrid of P. trichocarpa x deltoides. Stem Hydraulic Properties and Water u se Efficiency in Populus Populus species are typically diffuse porous or semi r ing porous in the case of P deltoides Vessel length distribution tests conducted by Harvey and van den Driessche (1997) found 98.5% of vessels to be less than 7 cm in length on average for P. trichocarpa x. deltoides clones. The same study reported average vessel diameters ranging from 31 to 37 m (Harvey & Van Den Driessche, 1997) Maximum vessel lengths in four naturally occurring populations of P. trichocarpa grown from cuttings were determined to be approximately 4 cm (Sparks & Black, 1999) Vessels are composed of a series of smaller conduits called vessel elements, which in Populus species, are bordered by simple pe rforation plates. Vessel element length has been estimated to be 58 m in P. trichocarpa Torr. & A. Gray (Stettler et al. 1996) Due to the anatomical str ucture of xylem vessels in Populus species, hydraulic resistance to laminar (longitudinal) flow is less than hydraulic resistance to lateral flow between vessels through i nter vessel bordered pits, which have pores on the order of 0.1 m for P. trichocarpa x. deltoides clones (Harvey & Van Den Driessche, 1997) I n Populus loss in xylem function (cavitat ion) usually begins at relatively high ( less negative ) stem water potential (around 1 MPa) and may be associated with large vessel size and high drought sensitivit y (Harvey & Van Den Driessche, 1997; Hukin et al. 2005) Tyree et al. (1992) reported a sharp increase in PLC caused by embolism at water potentials below 1.0 MPa in P. deltoides and absolute loss of condu ctivity by 2 MPa. Stettler et al. (1996) reported

PAGE 23

23 minimum stem xylem pressure without inducing runaway cavitation ranges from 1.05 to 2.2 7 (MPa), and minimum leaf xylem pressure for well watered conditions are 1.45 and 1.40 (MPa) for P. deltoides and P. trichocarpa respectively. In addition, the stomata of P. deltoides close more rapidly in response to declining leaf water potential tha n those of P. trichocarpa whose stomata may remain open even at low leaf water potentials (Braatne et al. 1992) These observations may explain the restriction of Populus species to riparian habitats. Interspe cific hybrids have been shown to be better able to cope with limited soil water availability (Braatne et al. 1992) Variation in hydraulic conductivity and vessel size has been observed in drought hardy and dro ught sensitive hybrid clones of P. trichocarpa and P. deltoides (Harvey & Van Den Driessche, 1997) Drought resistant clones had wider and longer vessels and greater K S than drought sensitive clones ( Harvey & Van Den Driessche, 1997) K S values averaged 5.7 and 7.8 kg s 1 m 1 MPa 1 for sensitive and hardy clones, respectively, with hardy clones having higher mean vessel diameter (36.5 m) than sensitive clones (33.9 m) (Harvey & Van Den Driessche, 1997) A study by Sparks and Black (1999) provided evidence of inter population variation in hydraulic conductivity and resistance to drought induced cavitation among P. trichocarpa populations from habitats with differing humidity and temperature gradients. Differences in leaf spec ific hydraulic conductance was also observed in two P. deltoides genotypes, although, no differences for sapwood specific hydraulic conductance, biomass or productivity were found (Samuelson et al. 2007) An ear ly investigation of WUE in Populus found a high degree of variation among and within clones of the same species when measured under high water availability (Blake et al. 1984) WUE was also correlated with foliar adaptations to minimize water loss as higher leaf

PAGE 24

24 diffusive res istance in water efficient clones, while morphology and physiology did not help explain variation in WUE (Blake et al. 1984) A 1989 survey of exotic and natural poplar forests in Italy for 13 C found significant variation among P. alba and P. nigra clones, despite differences in environment and sex (Stettler et al. 1996) Another study in wes tern Washington with P. trichocarpa and P. deltoides and their hybrids, grown in a common garden on wet and dry plots showed less discrimination on the dry plot due to improvement of WUE though no correlation was found with productivity (unpublished data from Hinkley reported in Stettler et al. 1996) In a greenhouse study comparing six hybrid popla r clones ( P. trichocarpa and hybrids with P. deltoides and P. euramericana ), grown under different nitrogen and potassium treatments, drought resistant clones did not exhibit osmotic adjustment and had similar WUE (instantaneous gas exchange measurement) to drought susceptible clones under drought conditions (Harvey & Driessche, 1999) Contrasting these results both Liu and Dickmann (1996) and Ridolfi and Dreyer (1997) found poplar clones that were productive and showed increased WUE in response to drought. In a well watered field trial looking at leaf traits for fast growth in an F 2 population derived from P. trichocarpa and P. deltoides significant clonal variation for 13 C was detected, and ranged from 30.97 to 19.99 (Rae et al. 2004) The trial also suggested that this pedigree would be useful to study further in terms of selecting trees for breeding programs in varying environment s due to the high segregation of the population for 13 C Marron et al. (2005) reported significant clonal differences i between extremes, for P. deltoides x P. nigra hybrid clones. They concluded that it is possible to select poplar clones which have high productivity and high WUE

PAGE 25

25 biomass producti on. In a very similar study, Monclus et al. (2005) found large clonal variability for foliar and productivity in P. deltoides x P. nigra clones. They also demonstrated that and productivity were not correlated among young rooted cutt ings and larger sizes and ages of P. deltoides x P. nigra clones, even under very different growth conditions. The authors again suggested there is potential for improvement of WUE in Populus without sacrificing productivity, and clones could be selected based on leaf traits. Tree Physiology and Growth in a G enetic C ontext Ackerly and Monson (2003) illustrate the need for the investigation of plant physiological traits in a genetic context. Quantitative genetics explains how the variability in continuous traits, or phenotypes, including growth, 13 C and WUE is influenced by heritability (genetic control) and examines the role and efficiency of selection in directing these traits (Stettler et al. 1 996) Generally, heritability is the extent to which genetic differences among individuals contribute to the observed differences in phenotype. In a pseudo backcross population we calculate b road sense heritability, the ratio of the total genetic varian ce to the total phenotypic variance which includes dominance and epistasis effects. Herita bility ranges from 0 to 1, with values closer to 1 corresponding to higher heritability. Q ua n titative trait loc i (QTL) a re genomic region s associate d with a quanti tative trait and may contain one to several genes controlling that trait QTL mapping has facilitated investigations of the genetic architecture of traits important to breeders, such as growth, disease and WUE A few to several QTL explain the phenotypi c variation of a trait, and may be used to identify candidate genes controlling a trait (Hartl & Clark, 1997) Populus is the model tree for genomics research due to several favorable attributes, including diploidy (simplification of segregation patterns compared to polyploid species ) and

PAGE 26

26 relatively small genome size (485 Mb) which eases genetic mapping (Tuskan et al. 2006) In addition to being easy to propagate and fast growing, Populus has a substantial amount of genetic variation (Taylor, 2002; Wullschleger et al. 2002; B runner et al. 2004) which may be as high as one polymorphi sm per 50 base pairs (Cronk, 2005) Average total nucleotide diversity for five genes in P. tremula has been observed to be 0.0111 (Ingvarsson, 2005) which is relatively high compared to crop species, as well as oak ( Quercus ) and pine ( Pinus ). Na tural plant populations as well as crop varieties have displayed a range of WUE that are genetically heritable (Geber & Dawson, 1997; Richards et al. 2002) G enetic c ontrol and mapped QTL for 13 C and growth have been identified in several tree species but relationships between these traits have been variable. In pine, correlations between 13 C and growth have been shown to be negative, positive, or uncorrelated, depending on species and envi ronmental conditions (Brendel et al. 2002; Prasolova et al. 2003) A study of maritime pine ( Pinus pinaster ) determined 13 C and ring width were heritable (0.17 and 0.19, respectively), but the positive phenotypic correlation between the traits was not determined by the genetic component, but rather the environment (Brendel et al. 2002) This study identified two QTL for ring width and four for 13 C and these QTL did not co localize. However, 13 C and relative growth rate in 92 full sib willow ( Salix ) hybrids, negatively correlated in drought and positively correlated in well watered conditions, and three QTL for these traits were found to co l ocalize (Weih et al. 2006) In addition, s ignificant variation in 13 C and negative correlations between 13 C and growth were observed in water stressed Eucalyptus genotypes (Osorio & Pereira, 1994) QTL for growth and crown architect ure traits have been previously identified in Populus spp (Wu et al. 1998; Wu, 1998; Wullschleger et al. 2005) These studies generally identified one to a few QTL, explaining less than a quarter of the phenotypic variation in growth tra its. Still, despite

PAGE 27

27 finding clonal variation for 13 C and heritability of 0.46 to 0.71, a few studies found no correlation between 13 C and growth in various Populus sp. genotypes (Rae et al. 2004; Marron et al. 2005; Monclus et al. 2005) Though QTL for WUE and 13 C have been found in several species, determining whether the QTL are associated with constitutive or inducible variation in 13 C is difficult due to the appli cation of soil water deficit treatments in some studies (Condon et al. 2004) The importance of wood for construction, pulp and paper has motivated extensive research in gymnosperm s and to a lesser exte nt in angiosperm species on the genetic control of wood and fiber properties. In general, stem wood and fiber properties are under stronger genetic control than growth or disease resistance, and sufficient genetic variation exists to breed and select for improved properties. The best studied property is wood density, which in gymnosperms is related to tracheid dimensions and secondary wall thickness and therefore the most relevant to hydraulic conductivity. In gymnosperms, the heritabilities of wood dens ity (0.3 0.8) have been reported. Genetic correlations between growth and wood density range from weak to strong and depend on the species, population and cambial age. Multiple QTL have been mapped for wood density in Pinus taeda (Sewell et al. 2000; Brown et al. 2003) Pinus radiata (Devey et al. 2004) Pinus pinaster (Pot et al. 2006) Pinus caribaea x Pinus elliottii (Shepherd et al. 2003) hybrid larch (Arcade et al. 2002) and Eucalyptus (Grattapaglia et al. 1996) QTL for wood chemical composition have been mapped in Pinus taeda (Sewell et al. 2002) Populus (Zhang et al. 2006) and Eucalyptus (Kirst et al. 2004) Genetic investigations of stem hydraulic properties in trees are currently lacking. Only two studies have examined the genetic control of hydraulic traits, suc h as conduit diameter or hydraulic conductivity and these are limited to gymnosperms (Anekonda et al. 2002; Rosner et

PAGE 28

28 al. 2007) Moderate heritabilities and positive phenotypic and genotypic correlations between maximum sapwood specific hydraulic conductivity and tree growth rate were found among eight Norway spruce ( Picea abies (L.) Karst.) genotypes (Rosner et al. 2007) and 39 full sib families of Douglas fir ( Pseudotsuga menziesii var. menziesii (Mirb.) Franco) (Anekonda et al. 2002) In angiosperms, reports are limited to investigations of clonal variation. For example, ex amination of four willow ( Salix sp.) clones, varying by geographic origin, revealed differences in leaf specific hydraulic conductance (Wikberg & Ogren, 2004) In Eucalyptus clonal variation has been investigated for vessel number and size (Leal et al. 2003) No QTL studies of hydraulic traits are known. Conclusion Genetic studies of wood properties and growth in tree specie s have been substantial, and have determined genetic control and mapped QTL for several traits. The genetic control of WUE in trees has also been adequately covered, although to a lesser extent. Genetic studies on stem hydraulic properties are currently inadequate. In order to more fully understand the relationships between hydraulic conductivity, WUE and growth in trees, as well as possible genetic control and QTL a more comprehensive examination of these traits within a genetic framework is necessary.

PAGE 29

29 Fig. 1 1 A c i curve. CO 2 assimilation rate ( A ), as a function of the intercellular CO 2 concentration, c i represented by solid line, is the demand function. The dashed line represents the supply function (CO 2 diffusion from the atmosphere to intercellular spaces) as a straight line with a negative slope, and the magnitude of the slope of this line is equal to the stomatal conductance. The concentration value on the demand function where A =0 is the CO 2 compensation point ( function with the demand function is the net CO 2 assimilation rate (A*) at concentrations c i* and c a* ( Adapted and reproduced from (Lambers et al. 1998) ) demand function CO 2 Assimilation rate ( A mol m 2 s 1 ) Intercellular CO 2 concentration, c i RuBP saturated Supply function (slope =stomatal conductance) RuBP limited c a c i* Carboxylation capacity A

PAGE 30

30 CHAPTER 2 G ENETIC CONTROL OF ST EM HYDRAULIC PROPERT IES AND CORRELATIONS WITH GROWTH IN POPULUS Introduction Secondary xylem is a characteristic tissue in trees. In addition to the mechanical support and transport functions provided by xylem, this tissue also serves as a whole plant regulator of leaf gas exchange (Sperry, 2000) Water potential gr adients along the root to leaf xylem continuum are initiated when stomata open, allowing the diffusion of water vapor from leaves to the air and creating tension as the water column is pulled upward against the forces of gravity and frictional resistance f rom the xylem pathway itself. Lower resistance or higher conductance (resistance and conductance are inversely related) of xylem to water transport allows more water flow to the canopy at a given water potential gradient. The hydraulic conductivity of xy lem is directly tied to the collective structure and size distribution of xylem conduits: tracheids in gymnosperms, and vessels and vessel elements in angiosperms. Hydraulic conductivity increases rapidly with greater xylem vessel diameter because flow is proportional to the fourth power of conduit diameter (Tyree & Zimmerman, 2 002) Although larger vessels can supply the leaves with water more efficiently than narrow vessels, greater vessel diameter increases the susceptibility of xylem conduits to embolism and cavitation which decrease conductivity (Tyree & Zimmerman, 2002) and can lead to branch dieback (Williams & Cooper, 2005) Still, the hydraulic pathway in xylem must be sufficient to meet the transpirational demand of the leaves, while approaching a balance between xylem safety and efficiency (Tyree & Sperry, 1989; Tyree, 2003; Hacke et al. 2006) This balance is shifted toward effic iency (wider vessels) in mesic environments where competition for light leads to high leaf area, and toward safety (narrower vessels) in arid environments where low soil water availability determines plant water status (Kocacinar & Sage, 2003)

PAGE 31

31 When water potential gradients in the hydraulic pathway exceed threshold levels, the conduit water column begins to cavitate (S perry et al. 1996) rendering portions of the hydraulic pathway inoperable, and causing reduced hydraulic conductivity. If transpiration continues unchecked, feedbacks between reduced hydraulic conductivity and water potential will lead to "runaway cavi tation" and catastrophic failure of the entire conducting system (Tyree & Sperry, 1988) As plants approach the water potential associated with runaway cavitation, stomata begin closing, transpiration rate is reduced, and water potential is moderated. In this way, the ability of the xyle m to deliver water to the crown, and the regulation of water loss from leaves is highly coordinated (Sperry, 2000; Holbrook et al. 2002) Stomatal control also has a strong influence on carbon gain and plant growth. Leaf level water use efficiency ( WUE ), often m easured as foliar carbon isotope composition ( 13 C ), varies mostly with the internal leaf CO 2 concentration, which may change with variation in photosynthetic capacity or stomatal conductance (Condon et al. 2004) WUE has important implications for plant performance when the rate of CO2 uptake is reduced by stomatal closure associated with water stress. Due to these relationships between hydraulic conductivity and stomatal conductance, and between photosynthesis rates and growth, foliar 13 C may be expected to vary with hydraulic conductivity and plant growth rate. In addition, WUE influences the balance between xylem safety and efficiency and therefore is also expected to affect xylem structure and function (Sperry & Hacke, 2002) Because of the interactions between xylem hydraulic conductivity and leaf gas exchange, Tyree (2003) hypothesized that high hydraulic conductivity is actually a pre req uisite for rapid growth in forest trees. By this reasoning, trees with inherently low hydraulic conductivity will tend to experience more extreme within tree water potential gradients and more frequent

PAGE 32

32 stomatal closure, leading to lowered photosynthetic g as exchange and, in turn, reduced carbon hypothesis, but few studies have examined within species or genetic variation in hydraulic conductivity relative to growth in for est trees. A study of desert and montane ecotypes of Ponderosa pine ( Pinus ponderosa ) populations, no significant ecotypic or population differences in specific hydraulic conductivity, growth or biomass allocation were found (Maherali et al. 2002) However, moderate heritabilities and positive phenotypic and genotypic correlations between maximum sapwood specific hydraulic conductivity and tree growth rate were found among eight Norway spru ce ( Picea abies (L.) Karst.) genotypes (Rosner et al. 2007) and 39 full sib families of Douglas fir ( Pseudotsuga menziesii var. menziesii (Mirb.) Franco) (Anekonda et al. 2002) Examination of four willow ( Salix sp.) genotypes, varying by geographic origin, revealed differ ences in leaf specific hydraulic conductance (Wikberg & Ogren, 20 04) This was also observed in two P. deltoides genotypes, although, no differences for sapwood specific hydraulic conductance, biomass or productivity were found (Samuelson et al. 2007) V ariation in sapwood s pecific hydraulic conductivity has been observed among P. trichocarpa populations from habitats of differing humidity and temperature gradients (Sparks & Black, 1999) Additionally, genetic correl ations between 13 C and growth have been variable (Condon et al. 2004) In pine, correlations between 13 C and growth have been shown to be negative, positive, or uncorrelated, depending on species and environmental co nditions (Brendel et al. 2002; Prasolova et al. 2003) In contrast, s ignificant variation in 13 C and negative correlations between 13 C and growth were observed in water stressed Eucalyptus genotypes (Osorio & Pereira, 1994) Finally, despite finding clonal v ariation for 13 C a few studies found no correlation between 13 C and growth in various Populus sp. genotypes (Rae et al. 2004; Marron

PAGE 33

33 et al. 2005; Monclus et al. 2005) Ov erall, the underlying genetic mechanism of hydraulic conductivity has not been sufficiently explored in angiosperm trees, and investigations of stem hydraulic properties in relation to WUE and growth are lacking. Cottonwoods ( Populus spp.) are considered a mong the fastest growing angiosperm trees in temperate regions and are ecologically and commercially important (Stettler et al. 1996; Williams & Cooper, 2005) Besides being easy to propagate and fast growing, Populus has a substantial amount of genetic variation (Taylor, 2002; Wullschleger et al. 2002; Brunner et al. 2004) This study uses progeny from a pseudo backcross of P. deltoides with an F1 hybrid of P. trichocarpa x P. deltoides ( P. t x d ) Eastern cottonwood ( P. d eltoides ) and black cottonwood ( P. trichocarpa ) are mesic species, dependent on shallow alluvial groundwater linked to water from streams, and have diffuse porous xylem structure (Rood et al. 2003) However, these species vary in geographical distribution. P. deltoides of the Aigeiros section of Populus is found in drier, warmer sites at lower elevations and latitudes in eastern and central North America from Minnesota to Texas. In contrast, P. trichocarpa of the Tacamahaca section, is typically found in wetter, coo ler sites at higher elevations and latitudes in western North America from Alaska to California (Stettler et al. 1996) The geographic variation in adaptiv e ranges for these species suggests genetic variation in the inherent physiology of these species (Sparks & Black, 1999; Samuelson et al. 2007) A greenhouse experiment was conducted to qua ntify variation in hydraulic conductivity, xylem anatomy, WUE as measured by 13 C and growth, and examine the interactions among these traits. Three hypotheses were tested: (1) hydraulic conductivity and productivity are heritable, have a positive phenotypic correlation, and are genetically correlated (Tyree, 2003) ; (2) 13 C is heritable and has a negative phenotypic and genetic correlation with growth; (3)

PAGE 34

34 hydraulic vessel diameter and vessels per sapwood area are heritable and phenotypically and genetically correlated with growth. Materials and Methods Plant M aterial, P ropagation and G rowth All plant material was obtained from a pedigree produced for a concurrent study by the Natural Resources Research Institute (University of Minnesota). The pedigree is a pseudo backcross of the hybrid female parent 52 225 ( Populus t richocarpa 93 968 x P. deltoides ILL 101, P. t x d ) and male parent D124 ( Populus deltoides ), hereafter referred to as Family 52 124. The P. deltoides parent material of the female hybrid was taken from Illinois, while the pure P. deltoides parent was fro m northern Minnesota. The P. trichocarpa parent of the hybrid came from western Washington. In May 2006, the parental and 22 genotypes from the segregating population were clonally propagated as rooted cuttings. Apical meristem cuttings (10 cm) were tr eated with Shultz Take Root rooting hormone (Indole 3 butyric acid 1%), planted in peat pellets, and placed on benches with intermittent mist (2min/12min from 0600 to 1800) in the greenhouse, under 30% shade cloth. After approximately two weeks, the mis ters were turned off and the rooted cuttings were hardened off for three days. On June 16, 2006, cuttings (3 5 reps per genotype, for a total of 82 plants) were transplanted to 41 cm deep pots (TPOT2, Stuewe & Sons, Inc.,Corvallis, Oregon, USA) with Fafar d 4MIX soil (Canadian Sphagnum Peat 40%, processed pine bark and vermiculite) to allow for unrestricted root growth. A black mark was made with a felt tip pen on the stem, five centimeters from the base of the terminal bud. This served as a reference poi nt for measurement of diameter and height growth increments. The pots were initially watered from the top until saturated, given an initial dose of (Hocking, 1971) with 5 mM ammonium nitrate and S.T.E.M.

PAGE 35

35 micronutrients (Soluable Trace Element Mix, The Scotts Company, USA) and placed randomly in a checkerboard arrangement on a flood bench (8 ft x 12 ft). The ebb n flow benches were f supplemented with 5 mM ammonium nitrate. Temperatures in the fan and pad cooled greenhouse ranged between 22C and 38C, and interior photosynthetically active radi ation ranged up to 1200 mol s 1 m 2 (over the waveband 400 700 nm) during the daily 14 hours of natural irradiance. Seedling Harvest, Biomass and Growth Measurements Seedling initial diameters and heights were recorded approximately 30 days after up pott ing. Sixty days after up potting, the seedlings were harvested between 0900 and 1700. On the day of harvest, at about 0900, one leaf in the middle of the live crown, was harvested from each seedling and immediately flash frozen in liquid nitrogen for 13 C %C %N analysis. Time of harvest, and seedling final diameters and heights were recorded. Leaves harvested from the main stem were stored in plastic zipper bags at 4C until leaf area was measured a few days after harvest with a leaf area meter (LI 1 000; Licor Inc., Lincoln, NB, USA). Shoots were cut at the root collar and immediately re cut with sharp anvil pruning shears under water, removing at least 5cm from the cut end. Stem segments at least 10cm long were stored in 15mL conical tubes with dei onized, distilled water in a cooler with ice for up to 5 hours until hydraulic conductivity could be measured. All leaves, sylleptic (lateral) branches, stems and woody roots were dried in paper envelopes at 65C, and weighed for calculation of total plan t biomass ( BIO g) and root to shoot ratio ( Root:Shoot ). Height ( H INC ) and diameter ( D INC ) growth increments, were calculated as final (60 days growth) minus initial measurements (30 days growth). Specific leaf area ( SLA, m 2 g 1 ) was

PAGE 36

36 calculated as main s tem leaf area per dry weight. A previous study of Populus grown under similar conditions showed sylleptic leaf weight was on average 65% of total sylleptic weight (Cooke et al. 2005) Sylleptic leaf area was estimated by calculating sylleptic leaf weight from total sylleptic weight (leaves and stems) an d multiplying by SLA Total leaf area ( LA ) was calculated as the main stem leaf area plus the sylleptic leaf area. Foliar Carbon Isotope Composition After drying at 65C for at least two days, the leaf samples were ground to a homogeneous consistency with a coffee grinder. Subsamples of approximately 0.01g were placed in glass scintillation vials and re dried at 65C. Prepared leaf samples were sent to the Cornell University Stable Isotope Laboratory (Ithaca, NY) to determine 13 C %C %N and C to N ratio ( C:N ) quantified with a continuous flow isotope ratio mass spectrometer (Finnigan MAT Delta Plus; Finnigan MAT, Bremen, Germany). 13 C was calculated (Equation 2 1) relative to the international standard Pee Dee Belemnite lim estone (Craig, 1954) where R sample and R standard are the 13 C/ 12 C ratios of the sam ple and standard, respectively ( 2 1) Hydraulic Conductivity and Xylem Vessel Measurements Hydraulic conductivity was determined two ways: first by the low pressure flow method (Sperry et al. 1988) on a subset of individuals, and secondly by the theoretical calcu lation of hydraulic conductivity (detailed later) on all individuals. Prior to the low pressure flow measurements, the stems were allowed to equilibrate to room temperature (25C) in the lab, and re cut under water with fresh razor blades. All stems were cut to at least 7 cm in length (ranging from 7.8 to 10 cm) based on previous vessel length distribution tests for P. t richocarpa x P. deltoides genotypes that showed 98.5% of vessels were less than 7cm long (Harvey & Van Den

PAGE 37

37 Driessche, 1997) Stem segments contained multiple nodes, which were wrapped with parafilm to prevent leaks during measurement. Average stem segment diameter ranged from 2 to 7 mm, and the segment was located within 15 cm above the root collar. To control for ion effects on pit membrane conductivity, the perfusion solution contained 20 mM KCl, in distilled, deionized water, deaera ted by sparging with helium (to reduce embolism formation), filtered to 0.2 adjusted to pH 2 with HCl (to control microbial growth) (Wheeler et al. 2005) Stems were connected under water to a hydraulic apparatus containing the perfusion solution, and measurements were not taken unti l the zero pressure (background) flow was zero. Flow of the solution through the stems at low pressure (4 kPa), to prevent flushing native embolism, was recorded on a balance (Fisher Scientific XA Analytical Balance; Fisher Scientific, Hampton, NH, USA) c onnected to a computer and converted to initial conductivity ( K native kg s 1 m MPa 1 ). Embolisms were then flushed with a higher pressure (>100kPa) produced by a syringe mounted in a caulk gun, and the flow recorded again and converted to maximum conduct ivity ( K max kg m MPa 1 s 1 ). Percent loss of conductivity ( PLC ) was calculated (Equation 2 2). PLC = 100*( K max K native )/ K max ( 2 2) After conductivity measurements were completed, all stems were placed in 15 mL conical tubes with 50% ethanol in deio nized, distilled water, and stored at 4C until cross sections were made. Cross cm from the upstream end of the stem with a vibratome (Leica VT1000 S; Leica Microsystems, Wetzlar, Germany) and mounted in deionized distilled water. Images of the xylem were captured with QCapture Suite V2.60 (QIMAGING, Surrey, BC, Canada) by a digital camera (Retiga 1300; QIMAGING, Surrey, BC ) attached to a light microscope (Olympus Ix70; Olympus, Tokyo, Japan) at 3 magnification. Each image was imported into Image Pro Plus 4.0

PAGE 38

38 (Media Cybernetics, Bethesda, MD, USA) for analysis. Sapwood area ( SA mm 2 ) was measured manually using the drawing tool. Huber value ( LA:SA m 2 cm 2 ) was calculated as LA divided by S A In each cross section, vessel area was measured by automated tracing and, when needed, manual drawing of the inner perimeter of the vessel lumen. Vessels with less than 0.000314 mm 2 lumen area, corresponding to a circle of equivalent area with a diamete r ( d therefore have little effect on total conducting capacity (Solla & Gil, 2002) Dividing total vessel area by sapwood area gave the ratio VA:SA (mm 2 mm 2 ). The individual vessel areas were converted to diameters ( d ) and counted ( n ), and vessels per sapwood area ( VSA count per mm 2 ) and mean hydraulic diameter ( D h d 4 ) n 1 ) 1/4 ) were calculated (Tyree & Zimmerman, 2002) To determ ine theoretical conductivity, d was used to calculate lumen resistivity (Sperry et al. 2005) for each vessel as in Equation 2 3, with low pressure flow meter measurements (8.9x10 10 MPa s). ( 2 3) Lumen conductivity for each vessel was calculated as the inverse of R L summed (conductances in parallel are additive) to determine theoretical conductivity, K t in m 4 MPa 1 s 1 (Tyree & Zimmerman, 2002) K t values were converted to the same units as K max (kg m MPa 1 s 1 ) by multiplying by 1000 kg m 3 H 2 O. Sapwood specific conductivity ( K S = K t /SA kg m 1 MPa 1 s 1 ) and leaf specific conductivity ( K L = K t /LA kg m 1 MPa 1 s 1 where LA is the total leaf area distal to the stem segment) were also calculated.

PAGE 39

39 Statistical Analysis Regression analysis was used to relate K t to K max and also to relate hydraulic tra its to growth and carbon isotope composition. Means of all ramets per genotype were used in the plots and regression analyses, which were performed with SigmaPlot version 10.0 (Systat Sofware, Inc., San Jose, CA, USA). Analysis of variance (ANOVA) was pe rformed using JMP version 6.0.3 (SAS Institute Inc, Cary, NC, USA) to assess significant ( P <0.05) genotypic differences between the parents and among the progeny for physiological and growth traits. One outlier per 74 measurements for five traits, and fou r outliers for Root:Shoot were identified by ASReml (see description below) and removed prior to performing regression analyses and ANOVA. Within family broad sense heritability and phenotypic and genetic correlations between physiological traits among th e progeny were calculated using the software package ASReml (ASReml 2.0; VSN International, Hemel Hempstead, UK) (Gilmour et al. 2007) This software fits linear mixed models using Residual Maximum Likelihood (REML). Variance components were estimated using a linear mixed mode l (Equation 2 4), where y is the predicted trait (random response variable), is the vector of fixed effects with incidence matrix X u is the vector of random effects with incidence matrix Z and e is the vector of residual errors, where the residual err or is unique to the observation (no residual covariance). ( 2 4) For analyses of single traits, the trait mean was the only fixed effect, and row (R) and column (C) (position on the greenhouse bench) and genotype (G) were rando m effects. A bivariate model ( Equation 2 4) was used to generate genetic correlations between traits. In the bivariate model, y is a vector containing observations for two traits (T1 and T2), contains the mean for each trait and the i th random variable has the following distribution: ~ MVN( 0 G i I i ), where G i : is defined

PAGE 40

40 by Equation 2 5, and I i has the dimension of the number of levels for the random variable, is the variance for T1, and is the covariance for T1 and T2, for i = G, R, C, and e G i = ( 2 5) Outliers and non significant random effects identified in the output from the linear mixed model were removed and the analysis rerun prior to calculation of wi thin family broad sense heritability and correlations. Within family broad sense heritability ( containing additive and non additive genetic variation Equation 2 6 ), was calculated by dividing the genetic variance ( ) by the total phenotypic variance (genetic variance, and residual variance, ) for each trait. ( 2 6) We also generated ph enotypic ( ) and genetic ( ) correlations (Equation 2 7, Equation 2 8). ( 2 7 ) ( 2 8) Standard errors ( SE ) for and correlations were s imultaneously generated by ASReml through Taylor series approximation for the variance of a ratio. Results Measured and Theoretical Conductivity In plant stems, low pressure flow is the standard method used to measure hydraulic conductivity (Sperry et al. 1988) However, the low pressure flow method is time prohibitive

PAGE 41

41 for experiments with large sample sizes, given that stems need to be processed the same day they are harvested. An alternative method that can be used with large numbers of samples is the theoretical estimation of K h from x ylem morphology measurements. Previous studies have shown strong linear correlations between measured and theoretical estimates of K h based on vessel anatomy (Lo Gullo, 1991; Martre et al. 2000) Theoreti cal methods (calculated as the sum of the fourth power of vessel diameters) are commonly seen in the literature as a substitute for direct measurement of K h (Lewis & Boose, 1995; Villar Salvador et al. 1997; Solla & Gil, 2002) The main limitations of the theoretical method are that only maximum K h can be estimated because the extent of embolism is not known; and K t does not reflect pit membrane conductivity, therefore two plants with the same K t could have different pit conductivity, resulting in different K h Using 29 young poplar stem s, we established K t as a good predictor of K max ( K max = 1.46* K t non significant intercept, r 2 =0.9266, P <0.0001), although K t consistently under estimated K max by about 46%. Plotting the residual against the predicted values revealed no discernable trend (data not shown). K max could have been over estimated if the vessel length assumption was incorrect, suggesting a decrease in the number of intact vessels and end plates, and effectively reducing resistance to flow (Sperry in press) However, all stems were longer than the maximum vessel length estimated previously for P. trichocarpa x deltoides (Harve y & Van Den Driessche, 1997) with stem length averaging 8.5 cm, so most vessels are expected to have remained intact. Significant positive affects of the concentration of KCl in the perfusion solution on pit membrane conductivity have been previously no ted (V an Ieperen et al. 2000; Zwieniecki et al. 2001; Domec et al. 2007) However, our tests showed no change in K max with or without KCl addition. The imaging of stem cross sections for K t may be the main contributor

PAGE 42

42 to the observed difference between K ma x and K t When the objective was changed from 3 x to 10 x a 13% increase in mean hydraulic vessel diameter on average was found. This small increase could be a significant source of error because conductivity is proportional to d 4 and a 10% increase in d iameter results in an approximately 46% increase in conductivity (Tyree & Z immerman, 2002) Overall, the strength of the relationship between K max and K t and the fact that the overestimate was consistent across the range of conductivity, support our conclusion that K t is an excellent predictor of the more difficult to measure K max In addition, with the low pressure flow method, PLC averaged <3%, and the maximum observed PLC was about 9% suggesting that watering twice a day was sufficient to minimize embolism and that K t would be similar to K h in our experiment. Thus, all hyd raulic conductivity and specific conductivity results presented are based on K t measurements. Variation in Growth, Hydraulic, and Physiological Traits The parents showed no significant differences in mean growth ( D INC or H INC Table 2 1). However, the P. t x d parent had significantly larger mean D h and lower mean VSA than the P. deltoides parent. Hydraulic conductivity traits were not significantly different, though the P. t x d hybrid parent had slightly higher K S Also, foliar C:N was significantly hi gher in the hybrid parent compared to the P. deltoides parent, while %N %C and 13 C were not significantly different. The progeny, which often exceeded the phenotypic values of the parents, differed significantly for most phenotypes (Table 2 1). Signific ant genotypic variation in growth and allocation parameters were detected, including D INC H INC Root:Shoot and BIO as well as SLA and LA:SA Significant variations in hydraulic traits were also found for D h VSA K S and K L Like the parents, foliar %C and %N were not significantly different among the progeny, while differences in C:N were significant. Variation in 13 C was due to only one genotype that was

PAGE 43

43 significantly different ( P = 0.0006) from the remaining progeny, which did not vary significantly for 13 C ( P = 0.5266). Therefore 13 C was not considered in the genetic analysis. Overall, significant phenotypic variation for growth, xylem anatomy and hydraulic conductivity was observed in the parents and progeny from Family 52 124. Genetic Control of Growth, Hydraulic, and Physiological Traits W ithin family broad sense heritabilit ies ( ) w ere calculated to estimate the proportion of the phenotypic variance due to genetic differences (Table 2). D INC H INC BIO and LA:SA showed moderate heritability (Table 2 2 ), while %C was not heritable ( was zero), and %N and 13 C showed marginal heritabilities of 0.140.14 and 0.150.12, respectively (not shown in Table 2 2). VSA had the highest of any trait and was double the heritability of D h K S showed similar moderate heritability compared to K L SLA showed moderate as well (0.320.15, not shown in Table 2 2). Phenotypic and Genetic Correlations Genetic and phenotypic correlations were determined among heritable growth, allocation and hydraulic traits (Table 2 2). Pheno typically, hydraulic conductivity was positively correlated with diameter and height increment. The strongest phenotypic correlation occurred between D INC and K L (Table 2 2, Fig. 2 1). Diameter increment had a much weaker genetic correlation with K S and height increment was not genetically correlated with either K S or K L Biomass showed a weak positive phenotypic correlation with the hydraulic conductivity traits, but essentially no genetic correlation was found between these traits. The amount of vess el lumen area per sapwood area can be influenced by the size of the vessels, the number of vessels per unit sapwood area or a combination of both. In this study, vessel area per sapwood area ( VA:SA ), was independent of vessel size ( D h ), while positively

PAGE 44

44 c orrelated with vessel number per sapwood area ( VSA ) (Fig. 2 2a). D h and VSA were strongly negatively correlated both phenotypically and genetically (Table 2 2, Fig. 2 2b), and the two highest values of VSA were found in the slowest growing progeny genotyp es. Both D h and VSA also had strong phenotypic and genetic correlations with D INC with fewer large vessels correlated with greater diameter increment (Table 2 2, Fig. 2 3). K L also had a very strong positive genetic correlation with D h (0.820.11), and much weaker negative correlation with VSA However, since K L is calculated from vessel diameter, these traits are not completely independent of each other. Vessel anatomy traits were also genetically correlated with LA : SA with the strongest being betwee n VSA and LA : SA (0.690.16). Specific leaf area ( SLA ) is sometimes correlated productivity in Populus species (Marron et al. 2005; Monclus et al. 2005) In our study, total biomass was negatively correlated with SLA phenotypically ( 0.710.07) and genetically ( 0.850.14) (not shown in Table 2 2), indicating thicker leaves are associated with greater total biomass production (Fig. 2 4). Discussion Genetic Control of Hydraulic Conductivity In the study population hydraulic conductivity was under moderate genetic control, with for specific hydraulic conductivity and productivity ranging from 0.29 to 0.45 (Table 2 2). Few other studies have found genetic differences in hydraulic conductance in angiosperms grown under optimal conditions. Willow ( Salix sp.) genotypes fro m northern and southern regions in Europe (Sweden and Greece), showed no significant differences between genotypes in whole plant leaf specific hydraulic conductance measured prior to drought (Wikberg & Ogren, 2004) However, if shoots were considered separately from roots, the northern genotype had higher shoot condu ctance and the southern genotype had higher root conductance (Wik berg &

PAGE 45

45 Ogren, 2004) In a study of two P. deltoides genotypes, significant ( P <0.05) clonal differences in leaf specific hydraulic conductance and foliar N were reported, but not for sapwood specific hydraulic conductance, LA:SA foliar 13 C biomass and productivity, although several traits differed significantly with irrigation and fertilizer treatment (Samuelson et al. 2007) However, our results in Populus are supported by previous work in gymnosperms that showed genetic variation in xylem hydraulic conduct ivity (Anekonda et al. 2002; Rosner et al. 2007) Douglas fir seedlings showed individual and family heritabilities of 0.25 and 0.56, respectively, for stem cross sectional area specific hydraulic conductivity (Anekonda et al. 2002) The same study found a significant estima ted family mean correlation of 0.35 between hydraulic conductivity and stem diameter (Anekonda et al. 2002) Similarly, broad sense heritability was 0.310.16 for sapwood cross sectional area specific hydraulic conductivity in Norway spruce, and this trait also had significant phenotypic (0.48) and genetic (0.99) correlations with stem diameter (Rosner et al. 2007) Vulnerability of xylem to drought induced cavitation would be an important component to add to this study but this challenging because of the large sample sizes necessary for genetic analysis. Investigation of vulnerability is limited with respect to genetic variation and control. Examination of four populations of Douglas fir on an environmental gradient found differences in vulnerability to drought induced ca vitation, with highest vulnerability occurring in the most mesic population (Kavanagh et al. 1999) Vulnerability may be expected to be high in Populus species, due to low drought tolerance and phreatophytic habit (Hukin et al. 2005) Genetic Correlations between Xylem Anatomy, Hydraulic Conductivity and Growth A strong genetic correlation between hydraulic conductivity and productivity, particularly K L and diameter increment, was observed for young Populus trees grown in optimal conditions (Table 2 2, Fig. 2 1). This genetic correlation suggests that genes that regulate hydraulic

PAGE 46

46 conductivity may also regulate growth in Populus The positive direction of this correlation also hat high hydraulic conductivity is a prerequisite for fast growth in angiosperm trees. The difference in xylem anatomical traits between the parents was consistent with their geographical background. The P. trichocarpa material originated from a wetter, cooler climate, and had greater D h and lower VSA than the P. deltoides (Table 2 1). However, the progeny mean for D h and VSA was closer to that of the pure P. deltoides parent. This is expected because differences in alleles betwe e n the parents of the h ybrid parent are bigger than the differences in alleles between the two P. deltoides species of the pseudo backcross These differences in the parents and the strong negative relationship among the progeny between VSA and D h (Fig. 2 2) suggest a possible trade off in hydraulic architecture. There must be a tradeoff between vessel size and the proportion of other non conducting cell types, such that when vessel size per sapwood area increases fiber and ray cell number and/or area decreases. Previous inves tigations have not seen such differences in vessel anatomy. No significant differences in vessels per sapwood area and vessel diameter were observed between four geographically diverse populations of European beech ( Fagus sylvatica ) (Borghetti et al. 1993) Also, in a study of two varieties of Metrosideros polymorpha (Myrtaceae) from different geographic ranges in the Hawaiian islands, no significant intraspecific variation in vessel diameter was found (Hoof et al. 2007) Another study of 17 clones from mass selected Eucalyptus globulus t rials, found clonal variation accounted for 30% of the total variation of vessel proportion (Leal et al. 2003) Strong correlations between xylem anatomy and productivity traits, such as VSA and D h with D INC (Fig. 2 3), suggest vascular anatomy drives productivity when water is not limited. Generally, investment in fewer, larger vessels results in higher plant growth and higher hydraulic

PAGE 47

47 conductivity. The strong genetic correlations between leaf specific hydraulic conductance and mean hydraulic vessel diameter supports the idea that xylem structure in mesic speci es, such as poplars, favors efficiency (wider vessels) rather than safety (Kocacinar & Sage, 2003) Carbon Isotope Composition and Leaf Nutrient Conten t Our results show a lack of genetic variation for 13 C in the progeny population with of 0.15 0.12 This result is contrary to another well watered study with 29 young P. deltoides x P. nigra genotypes, where broad sense heritabil ity for foliar carbon isotope (Monclus et al. 2005) Expected genotypic variation in 13 C may be due to differences in stomatal conductance or photosynthetic capacity or both (Farquhar et al. 1989) Also, vapor pressure deficit (VPD) drives stomatal conductance and circulation fans in the greenhouse created an environment where VPD was not expe cted to vary. Assuming this was the case, then stomatal conductance and therefore transpiration, a component of WUE did not vary among the progeny While we did not quantify these traits directly, the lack of significant variation for foliar %N among th e progeny or parents suggests photosynthetic capacity is not varying in this experiment. Because foliar %N and %C did not vary significantly in the study population, the importance of the significant variation among the progeny and parents for foliar C:N may be limited. Increasing the number of genotypes in the study population may result in observation of greater genetic variation, in 13 C for example. S pecific Leaf Area, Huber Value, Hydraulics and Growth Significant genetic variation for SLA and LA:S A in the progeny, as well as moderate to strong implies these traits are also under genetic control. A previous study of two varieties of Metrosideros polymorpha (Myrtaceae), from different geographic ranges in the Hawaiian islands found significant intraspecific variation in SLA but not for LA:SA (Hoof et al.

PAGE 48

48 2007) Producing more wood per unit leaf area (lower LA:SA ) has been consi dered to improve plant nutrient and water storage capacity (Callaway et al. 1994) This is supported by the negative phenotypic and genetic correlations observed between LA:SA and D INC and BIO Our results showing biomass and SLA are negatively correlated (Fig. 4) support a previous study of cottonwood (Marron et al. 2005) Thus, low SLA characterized highly productive genotypes in our study. This relationship is attributed to increased density or size of mesophyll cells in low SLA leaves resulting in high CO 2 a ssimilation and thus high productivity (Marron et al. 2005) However, other studies in cottonwoods showed no correlation between the SLA and biomass (Rae et al. 2004; Monclus et al. 2005) Conclusion In conclusion, theoretical hydraulic conductivity by vessel area measurement provided a conservative and consistent estimate of measured hydraulic conductivity. The significant genetic va riation found for vessels per sapwood area, hydraulic vessel diameter, and specific hydraulic conductivity suggests these physiological traits are under genetic control in Populus and our results are the first to show genetic control of hydraulic conducti vity in angiosperm trees. Genetic examination of the vulnerability of xylem to drought induced cavitation could provide further insight to the genetic control of physiological traits. A QTL study would be a logical next step. Co localization between K h and growth would provide a causative linkage between

PAGE 49

49 Table 2 1 Trait means, standard errors ( SE ) and significance levels for 22 progeny genotypes and parental genotypes, Populus trich ocarpa x P. deltoides ( P. t x d ) and P. deltoide s Progeny P. t x d P. deltoides Abbreviation Trait (units) Mean (SE) P Range Mean (SE) P Mean (SE) D INC Diameter increment growth (mm) 4.00 (0.280) *** 1.61 5.24 5.61 (0.350) ns 4.66 (0.560) H INC Height increment growth (cm) 58.24 (2.999) ** 26.50 79.28 71.33 (10.328) ns 80.78 (6.505) SLA Specific leaf area (m 2 g 1 ) 0.0 292 (0. 0 00 7) ** 0.0 2 28 0.0 359 2 75 (0. 11) ns 2 94 (0. 06) LA Total leaf area (m 2 ) 0.207 (0.023) *** 0.045 0.492 LA:SA Huber value (m 2 cm 2 ) 1.13 (0.038) *** 0.88 1.53 SA Sapwood area (mm 2 ) 19.52 (2.408) *** 3.87 53.44 33.62 (6.501) ns 23.69 (3.771) D h Mean hydraulic vessel diameter ( m) 38.51 (0.552) 32.15 42.35 44.33 (0.891) 40.43 (0.630) VS A Vessels per sapwood area (count mm 2 ) 116.68 (3.887) *** 90.21 165.62 90.33 (3.136) ** 112.53 (2.030) K t x 10 4 Theoretical HC (kg m Mpa 1 s 1 ) 1.54 (0.215) *** 0.20 4.49 3.25 (0.662) ns 2.02 (0.425) K S Sapwood specific HC (kg m 1 Mpa 1 s 1 ) 7.17 (0.310) ** 4.63 9.95 9.63 (0.615) ns 8.33 (0.518) K L x 10 4 Leaf specific HC (kg m 1 Mpa 1 s 1 ) 6.61 (0.392) ** 3.90 10.41 13 C Foliar carbon isotope composition ( ) 31.51 (0.083) ns 32.59 30.97 31.28 (0.266) ns 31.94 (0.161) %N Fol iar percent nitrogen (%) 4.73 (0.061) ns 4.28 5.27 4.65 (0.201) ns 5.11 (0.098) %C Foliar percent carbon (%) 45.86 (0.154) ns 44.44 47.41 48.72 (0.336) ns 47.41 (0.423) C:N Foliar carbon to nitrogen ratio 9.77 (0.123) 8.57 10.79 10.54 (0.435) 9.28 (0.117) BIO Total plant biomass (g) 13.98 (1.983) *** 2.44 41.51 ROOT:SHOOT Root to shoot ratio (g g 1 ) 0.0 580 (0.00 23) ** 0.0 3 90 0.0 783 P values from ANOVA results for significant differences in trait means among the progen y or between the parents are indicated by: ns, non significant; P P P eaf area and biomass data was not measured in the parents ( ).

PAGE 50

50 Table 2 2 Trait within family broad sense h eritabilit ies (diagonal, bold type), genetic correlations (above diagonal), and phenotypic correlations (below diagonal) for progeny genotypes. Trait D INC H INC LA:SA D h VSA K S K L BIO D INC 0.45 (0.13) 0.88 (0.08) 0.59 (0.21) 0.89 (0.09) 0.77 (0.12) 0.4 5 (0.28) 0.71 (0.18) 0.92 (0.06) H INC 0.89 (0.03) 0.31 (0.13) 0.37 (0.29) 0.68 (0.19) 0.83 (0.11) 0.24 (0.35) 0.27 (0.33) 0.79 (0.14) LA:SA 0.59 (0.10) 0.46 (0.12) 0.50 (0.13) 0.50 (0.27) 0.69 (0.16) 0.03 (0.34) 0.66 (0.19) 0.47 (0.26) D h 0.81 ( 0.05) 0.77 (0.06) 0.48 (0.11) 0.29 (0.13) 0.68 (0.18) 0.73 (0.17) 0.82 (0.11) 0.64 (0.22) VSA 0.82 (0.05) 0.78 (0.06) 0.65 (0.09) 0.68 (0.08) 0.60 (0.12) 0.01 (0.32) 0.38 (0.26) 0.71 (0.16) K S 0.51 (0.10) 0.48 (0.10) 0.18 (0.14) 0.87 (0.04) 0.2 6 (0.13) 0.29 (0.13) 0.19 (0.34) K L 0.66 (0.08) 0.56 (0.10) 0.62 (0.09) 0.88 (0.03) 0.48 (0.12) 0.34 (0.13) 0.34 (0.30) BIO 0.86 (0.03) 0.79 (0.05) 0.38 (0.13) 0.66 (0.08) 0.67 (0.08) 0.40 (0.12) 0.48 (0.11) 0.41 (0.14) Standard errors are in pa rentheses next to each heritability or correlation. Correlations that are not biologically applicable ( ). See Table 1 for trait abbreviations.

PAGE 51

51 Fig. 2 1 Leaf specific hydraulic conductivity ( K L ) scales positively with diameter increment ( D INC ). Points are progeny genotype means. Line is a linear regression with r 2 = 0.4164, P =0.0012.

PAGE 52

52 Fig. 2 2 Relationships between hydraulic vessel diameter ( D h ), vessels per sapwood area ( VS A ) and vessel area per sapwood area ( VA:SA ). Points in all plots are progeny genotype means and error bars represent one standard error. (a) VSA scales positively with VA:SA ; line is a linear regression with r 2 = 0.2373, P =0.0215 (b) D h scales negatively with VSA ; line is a linear regression of with r 2 = 0.4250, P =0.0010.

PAGE 53

53 Fig. 2 3 Hydraulic vessel diameter ( D h ), open circles, scales positively ( r 2 = 0.7196, P <0.0001) and vessels per sapwood area ( VSA ), filled ci rcles, scales negatively ( r 2 = 0.6280, P <0.0001) with diameter increment ( D INC ). Points are progeny genotype means. Fig. 2 4 Specific leaf area ( SLA ) scales negatively with total biomass ( BIO ). Points are progeny g enotype means. Line is a linear regression with r 2 = 0.5488, P <0.0001.

PAGE 54

54 CHAPTER 3 Q UANTITATIVE TRAIT LO CI ANALYSIS O F STEM HYDRAULIC PRO PERTIES, FOLIAR CARBON ISOTOP ES AND GROWTH IN POPULUS Introduction The importance of trees for carbon sequestration and biofuels as a means to address climate change is prominent and growing (Millard et al. 2 007; Rae et al. 2008) For trees to play a more prominent role in mitigating greenhouse gas emissions, more information is needed in general about plant physiologica l traits, and in particular their genetic control (Ackerly & Monson, 2003) G enetic analysis is an important approach with which to dissec t fundamental questions in plant physiology For example, Tyree (2003) hypothesized that high hydraulic conductivity is required for fast growth in forest trees. Another question is the extent to which WUE is under genetic control or is an inducible trai t, triggered by the environment (Condon et al. 2004) Yet another debate involves the causes behind tradeoff s in v ascular hyd raulic efficiency and safety of xylem anatomy. Quantitative genetics explains how the variability in plant phenotypes is influenced by heritability, and examines the role and efficiency of selection for improving these traits (Stettler et al. 1996) Quantitative traits, such as growth, are very important to breeders and are essential to plant genetic improvement programs. C onventional quan titative trait loci (QTL) mapping has facilitated investigations of the genetic architecture of growth for several tree species (Grattapaglia et al. 1996; Wu et al. 1998; Brendel et al. 2002; Wullschleger et al. 2005; Weih et al. 2006) These studies generally identified one to a few QTL explaining less than a third of the phenotypic variation in growth traits of pine, poplar and eucalyptus. Heritability and QTL for wood property traits have also been identified, and most research has focused on wood density (Grattapaglia et al. 1996; Sewell et al. 2000; Arcade et al. 2002; Brown et al. 2003;

PAGE 55

55 Shepherd et al. 2003; Pot et al. 2006) but has also included wood cell ulose and lignin content (Sewell et al. 2002; Zhang et al. 2006) However, for physiological traits in trees g enetic control and mapped QTL are limited to carbon isotope composition ( 13 C ), an intrinsic measure of water use efficiency ( WUE ) (Brendel et al. 2002; Ponton et al. 2002; Prasolova et al. 2005; Weih et al. 2006) and to leaf chlorophyll content (Parelle et al. 2007) Only two studies have examined the genetic control of hydraulic traits, such as conduit diameter or hydraulic conductivity and these are limited to gymnosperms (Anekonda et al. 2002; Rosner et al. 2007) No QTL studies of hydraulic traits have been reported. Over recent decades, the genus Populus ( Salicaceae ) has been considered a model tree for studying growth, fundamental physiology and genetics (Stettler et al. 1996; Taylor, 2002) Populus trees are also valued by the forestry industry, becaus e of their fast growth and light colored wood. Managed stands, including Populus are being recognized as a key component of the carbon balance equation due to its expected carbon sequestration and increased growth responses to a global rise in atmospheri c CO 2 (Gielen & Ceulemans, 2001; Taylor, 2002) In addition to being easy to propagate and fast grow ing, Populus has a substantial amount of naturally occurring genetic variation (Taylor, 2002; Wullschleger et al. 2002; Brunner et al. 2004) G enetic vari ation within and among Populus species exists (Wu et al. 1997) and may be as high as one polymorphi sm per 50 base pairs (Cronk, 2005) Average total nucleotide diversity for five genes in P. tremula has been observed to be 0.0111 (Ingvarsson, 2005) which is relatively high compared to crop species, as well as with oak ( Quercus ) and pine ( Pinus ). Populus is also the model tree for genomics research due to several favorable attributes, including diploidy (simplif ication of segregation patterns compared to polyploid species) and relatively small genome size (485 Mb) which facilitates genetic mapping (Tuskan et al. 2006)

PAGE 56

56 The availability of a draft sequence of the Populus trichocarpa genome (Tuskan et al. 2006) enabl es the identification of candidate genes located within QTL intervals for most traits. While genetic studies on wood properties, growth and disease in Populus have been substantial (Wu et al. 1997; Frewen et al. 2000; Netzer et al. 2002; Tagu et al. 2005) investigations of the genetic mechanisms of stem hydraulic properties a nd WUE of Populus are currently lacking. Previous evidence with a small number of progeny (22) from this pseudo backcross (Chapter 2) showed genetic control of stem hydraulic and growth traits, and found strong genetic correlations between productivity a nd hydraulic conductivity suggesting that some of the same genes regulate both traits. If some of the genes controlling stem hydraulics and growth co (Tyree, 2003) A QTL analysis was conducted to investigate the genetic mechanisms controlling stem hydraulic traits, foliar c arbo n isotope composition, and growth. In our study population from the pse udo backcross of P. deltoides with an F1 hybrid of P. trichocarpa x deltoides we hypothesize that: 1. Significant variation in stem and leaf water relations properties exist. 2. QT 13 C. 3. Some QTL controlling hydraulic conductivity and increment growth co localize. 13 C co localize. 5. Some QTL contro 13 C co localize. M aterials and Methods Plant material All plant material was obtained from a pedigree produced for a concurrent study by the Natural Resources Research Institute (University of Minnesota). The pedigree i s a pseudo backcross of the hybrid female parent 52 225 ( Populus trichocarpa 93 968 x P. deltoides ILL 101, P. t x d ) and male parent D124 ( Populus deltoides ), hereafter referred to as family 52 124.

PAGE 57

57 The P. deltoides parent material of the female hybrid w as taken from Illinois, while the pure P. deltoides parent was from northern Minnesota. The P. trichocarpa parent of the hybrid came from western Washington. Experimental Design The seedlings used for this study were grown as part of a larger greenhouse experiment which included two nitrogen treatments, 0 mM and 25 mM ammonium nitrate, and 397 genotypes. However, our study was limited to the higher nitrogen treatment, and a subset of 100 genotypes (selection description in a later section). The full exp eriment had partially balanced incomplete design, with 3 replications and two treatments of 397 genotypes (on 18 ebb n flow benches). Among the 397 genotypes, 367 genotypes had three ramets, and 30 genotypes had two ramets. H11 11, a well characterized P. trichocarpa x P. deltoides hybrid, was used to fill gaps in the bench and as a plant growth indictor. H11 11 was not used for the analysis in the experiment. On each bench, 130 genotypes were arranged in a checker board pattern in a single plant plot des ign. The size of each bench was approximately 1.8 m x 2.5 m, with10 rows and 13 columns, having 10 plants per column and 13 plants per row. Temperatures in the fan and pad cooled greenhouse ranged between 22C and 38C, and interior photosynthetically act ive radiation ranged up to 1200 mol s 1 m 2 (over the waveband 400 700 nm) during hours of natural irradiance, and was supplemented by fluorescent lighting for a total of 16 hours of daylight. Plant P ropagation and G rowth In early August 2006, the parenta l and 397 progeny genotypes from the segregating population were clonally propagated as rooted cuttings. Apical meristem cuttings (10 cm) were treated with Shultz Take Root rooting hormone (Indole 3 butyric acid 1%), planted in peat pellets. At this ti me a black mark was made with a felt tip pen on the stem, 2.5 centimeters from

PAGE 58

58 the base of the terminal bud. This served as a reference point for measurement of diameter and height growth increments. The planted cuttings were placed on benches with inter mittent mist (2min/12min from 0600 to 1800) in the greenhouse, under 30% shade cloth. After approximately two weeks, the misters were turned off and the rooted cuttings were hardened off for three days. Cuttings (3 ramets per genotype) were transplanted over 3 days to 41 cm deep pots (TPOT2, Stuewe & Sons, Inc.,Corvallis, Oregon, USA) with Fafard 4MIX soil (Canadian Sphagnum Peat 40%, processed pine bark and vermiculite) to allow for unrestricted root growth. Upon transplanting, pots were initially water ed from the top until saturated, given 500mL (Hocking, 1971) with 5 mM ammonium nitrate and S.T.E.M. micronutrients (Soluable Trace Eleme nt Mix, The Scotts Company, USA) and placed on ebb n flow benches. The benches were flooded twice daily for approximately 30 minutes at 0800 and weeks. After five week s, benches were flooded once daily for approximately 30 minutes at 1500 supplemented with 25 mM ammonium nitrate. The treatment was applied over six days to the three reps. Pesticide T reatments Right after transplanting the pla nts, a systemic pesticide Marathon (Olympic Horticulture Products) was applied in granular form to the soil surface of each pot. Additional p esticides applied weekly include M pede (Dow Agroscience) and Tame Orthene (Whitmire Micro Gen). Pesticide applicat ions were essential to maintain healthy plants given the high density of plants in the greenhouse. Genotype Selection T o facilitate timely completion of our physiological phenotyping, a subset of 100 genotypes representing the 100 most informative recombin ations were selected with the MapPop

PAGE 59

59 software package (Vision et al. 2000; Xu et al. 2005) available online ( http://www.bio.unc.edu/faculty/vision/lab/mappop/ ) The 100 genotype subsample was selected from all 394 genotypes in the experiment, excluding genotypes with ramet s that died during the experiment genotypes displaying growt h beyond 2.5 standard deviations relative to the regression line predicted value and genotypes for which aneuploidy was apparent in scoring of the gels. Selections were based on SSBL function run for 300 seconds and a randomization factor of 5. Previous experiments with the software indicate d that adding time or additional randomization ha d no effect on the output. Seedling Harvest, Biomass and Growth Measurements Seedling initial diameters and heights were recorded approximately five weeks after up pott ing over 6 days, corresponding to the order of treatment application. 10 weeks after up potting, the seedlings were harvested between 0800 and 1700 over 6 days. Time and day of harvest, and seedling final diameters and heights were recorded. Height ( H I NC ) and diameter ( D INC ) growth increments were calculated as final minus initial measurements Three leaves in the middle of the live crown were harvested from each seedling and immediately flash frozen in liquid nitrogen for 13 C %C %N analysis. One additional leaf, in the middle of the live crown, was harvested, placed in a zipper plastic bag and stored at 4C until the leaf could be scanned (HP Scanjet 5400C HP, Palo Alto, CA ) for calculation of single leaf area. The ma in shoot was cut at the black m ark and a stem segment approximately 5 cm long was placed in a 15 mL conical tube with 50% ethanol in deionized, distilled water, and stored at 4C until cross sections were made. All leaves, sylleptic (lateral) branches, st ems and woody roots were dried in paper envelopes in a 65C drying room or a freeze drier (Freezezone 18L Bulk Tray Dryer, Labconco, Kansas City, Missouri), and we ighed for calculation of total plant biomass ( BIO g) and root to shoot ratio ( Root:Shoot ).

PAGE 60

60 T he scanned leaf images were imported into Image Pro Plus 4.0 (Media Cybernetics, Bethesda, MD, USA) to determine leaf area of a single leaf. Specific leaf area ( SLA, m 2 g 1 ) was calculated as single leaf area per unit dry weight. A subsequent study of a subset of genotypes from the same progeny, grown under the same conditions showed sylleptic leaf weight was on average 71 .4% of total sylleptic weight. Sylleptic leaf area was estimated by calculating sylleptic leaf weight from total sylleptic weight (lea ves and stems) and multiplying by SLA Main stem leaf area was determined by multiplying main stem leaf weight by SLA Total leaf area ( LA ) was calculated as the main stem leaf area plus the sylleptic leaf area. Foliar Carbon Isotope Composition The dried leaf samples were ground to a homogeneous consistency with a coffee grinder. Sub samples of approximately 0.01g were placed in glass scintillation vials and re dried at 65C. Prepared leaf samples were sent to the Cornell University Stable Isotope Laboratory (Ithaca, NY) to determine 13 C %C %N and C to N ratio ( C:N ) quantified with a continuous flow isotope ratio mass spectrometer (Finnigan MAT Delta Plus; Finnigan MAT, Bremen, Germany). 13 C was calculated (Equation 3 1) relative to the intern ational standard Pee Dee Belemnite limestone (Craig, 1954) where R sample and R stan dard are the 13 C/ 12 C ratios of the sample and standard, respectively ( 3 1) Hydraulic Conductivity and Xylem Vessel Measurements Cross from the black m ark end of the stem segment with a vibratome (Leica VT1000 S; Leica Microsystems, Wetzlar, Germany) and mounted in deionized distilled water. Images of the xylem were captured with Stereo Investigator software by a digital camera (CX9000, MBF

PAGE 61

61 Bioscience, Williston, VT ) attached to a digital microscope (Leica DM 4000B, Leica Microsystems, Wetzlar, Germany ) at 5 magnification. Each tiled image was imported into Image Pro Plus 4.0 (Media Cybernetics, Bethesda, MD, USA) for analysis. Sapwood area ( SA mm 2 ) was measured manually using the drawing tool. Huber value ( LA:SA m 2 cm 2 ) was calculated as LA divided by SA Theoretical hydraulic conductivity was calculated from xylem vessel measurements. In previous tests of 29 young poplar stems, we established K t as a good predictor of maximum conductivity ( K max K max = 1.46* K t non significant intercept, r 2 =0.9266, P <0.0001), although K t consistently under estimated K max by about 46%. In each cross section, vessel area was measured by automated tracing and, wh en needed, manual drawing of the inner perimeter of the vessel lumen. Vessels with less than 0.000314 mm 2 lumen area, corresponding to a circle of equivalent area with a diameter ( d diameter contribute l ess than 0.01% of conductivity and therefore have little effect on total conducting capacity (Solla & Gil, 2002) Dividing total vessel area by sapwood area gave the ratio VA:SA (mm 2 mm 2 ). The individual vessel areas were converted to diameters ( d ) and counted ( n ), and vessels per sapwood area ( VSA count per mm 2 ) and mean hydraulic diameter ( D h d 4 ) n 1 ) 1/4 ) were calculated (Tyree & Z immerman, 2002) To determine theoretical conductivity, d was used to calculate lumen resistivity (Sperry et al. 2005) for each vessel as in Equation 3 2, measurements (8.9x10 10 MPa s). ( 3 2) Lumen conductivity for each vessel was calculated as the inverse of R L summed (conductances in parallel are additive) to determine theoretical conductivity, K t in m 4

PAGE 62

62 MPa 1 s 1 (Tyree & Zimmerman, 2002) K t values were multiplied by 1000 kg m 3 H 2 O to attain the common units of kg m MPa 1 s 1 Sapwood specific conductivity ( K S = K t /SA kg m 1 MPa 1 s 1 ) and leaf specific conductivity ( K L = K t /LA kg m 1 MPa 1 s 1 ) were also calculated. Statistical Analysis Regression analysis was used to relate hydraulic traits to growth and carbon isotope composition. Means of all reps per genot ype were used in the plots and r egression analyses which were performed with SigmaPlot version 10.0 (Systat Sofware, Inc., San Jose, CA, USA). Analysis of variance (ANOVA) was performed using JMP version 6.0.3 (SAS Institute Inc, Cary, NC, USA) to assess significant ( P <0. 10 ) genotypic differences between the parents and among the progeny genotype means for physiological and growth traits. Heritability Within family broad sense heritability for several growth and physiological traits were calculated usin g the software package ASReml (ASReml 2.0; VSN International, Hemel Hempstead, UK) (Gilmour et al. 2007) This software fits linear mixed models using Residual Maximum Likelihood (REML). Variance components were estimated using a linear mixed model (Equation 3 3), where y is t he predicted trait (random response variable), is the vector of fixed effects with incidence matrix X u is the vector of random effects with incidence matrix Z and e is the vector of residual errors, where the residual error is unique to the observatio n (no residual covariance). ( 3 3) The trait mean was the only fixed effect, and rep, bench, and row and column (position on the greenhouse bench) and genotype were random effects. Within family broad sense heritability ( containing additive and non additive genetic variation Equation 3 4 ), was calculated by

PAGE 63

63 dividing the genetic variance ( ) by the total phenotypic variance (genetic variance, and residual var iance, ) for each trait ( 3 4 ) Standard errors ( SE ) for were simultaneously generated by ASReml through Taylor series approximation for the variance of a ratio. Quantitative Trait Analysis QTL for growth and physiological traits were identified using composite interval mapping performed with QTL Cartographer V2.5 (Wang et al. 2007) on separ ate single tree maps of the mother and father of family 52 124 ( Wu et al. submitted ). The resolution of the mother map is approximately 15 cM. Phenotypic data were converted to least square means prior to running the analysis (Frewen et al. 2000) The standard model (model 6) was used, with a walk speed of 2 cM, and significance level of P <0.05 determined by performing 1000 permutation tests. The magnitude of the QTL effect was calculated as the percentage variance explained (PVE). The likeliho od ratio (LR) was converted to an equivalent log of odds (LOD) score by multiplying LR by 0.2171 (Frewen et al. 2000) Results Variation in Growth, Hydraulic, and Physiological Traits The parents of family 52 124 showed few significant differences in the traits examined (Table 3 1). The P. t x d hybrid parent had larger D h ( P <0.10) than the P. deltoides parent, while VSA did not differ. Leaf specific and stem specific hydraulic conductivity traits were higher ( P <0.10, and P <0.05, respectively) in the P. t x d hybrid parent. Also, foliar C:N was higher in the hybrid parent compared to the P. deltoides parent, mostly because %N was lower, and %C was higher in the hybrid parent. Among the progeny genotypes, there were differences ( P <0.05)

PAGE 64

64 in all tra its examined except %C (Table 3 1). The range of progeny genotype mean values for most traits exceeded the range of parental genotype means. There was significant genotypic variation in growth and allocation parameters, including D INC H INC Root:Shoot, and BIO as well as SLA and LA:SA The hydraulic traits D h VSA K S and K L also varied significantly among the progeny. Unlike the parents, foliar %C was not significantly different among the progeny. Differences in C:N were significant, and most likely due to significant differences in %N Variation in 13 C was significantly different ( P = 0.05) in the progeny. Overall, significant phenotypic variation for growth, xylem anatomy and hydraulic conductivity was observed in the parents and progeny from Family 52 124. Scaling Growth, Hydraulic, and Physiolo gical Traits Diameter growth increment scaled positively with leaf specific hydraulic conductivity (Fig 3 1; r 2 = 0.4839, P <0.0001) and with mean hydraulic vessel diameter (Fig 3 2; r 2 = 0.4172, P <0.0001 ). Diameter growth increment also showed a weak but significant negative relationship with vessels per sapwood area (Fig. 3 2; r 2 = 0.1510, P <0.0001). However, D h and VSA were very weakly if at all correlated ( r 2 = 0.0448, P =0.0346). Both D h and VSA scaled positively with VA : SA ( r 2 = 0.5103, P <0.0001, and r 2 = 0.2588, P <0.0001, respectively). In addition, specific leaf area scaled positively with total biomass (Fig. 3, r 2 = 0.4682, P <0.0001), and LA:SA scaled negatively with D INC ( r 2 = 0.4664, P <0.0001). Heritability of Growth, Hydraulic, and Physio logical Traits W ithin family broad sense heritabilit ies ( ) were calculated to estimate the proportion of the phenotypic variance due to genetic differences (Table 3 2). The growth and allocation traits, including D INC H INC BIO Ro ot:Shoot and LA:SA showed moderate heritability, ranging from 0.29 to 0.43 (Table 3 2). Leaf traits %N and C:N were also moderately

PAGE 65

65 heritable, while 13 C, %C, and SLA showed the lowest heritabilities, ranging from 0.07 to 0.15, respectively. D h had the highest of any trait, at 0.53. VSA and K S had the next highest heritability (0.46 and 0.47, respectively), while heritabilities for K L and K t were lower (0.40 and 0.30). Q uantitative Trait Loci Results Significant ( P <0.05) QTL were identified for all traits examined except LA:SA 13 C %C and Root:Shoot (Table 3 2) Many more significant QTL were identified in the map of the mother (21 QTL on 10 linkage groups) than that of the father (7 QTL on 6 linkages groups) (Figs. 3 4 3 5 and 3 6 ). The reduced num b er of QTL identified in the paternal map was expected due to the greater differences in segregating alleles for the parents of the hybrid mother than the differences in the two P. deltoides genotypes contributing the population. %N had the most QTL with four QTL on four separat e linkage groups explaining a total of approximately 46 percent of the phenotypic variance. Although the QTL for %N were the most numerous for a one trait, they were not individually a s strong as QTL for growth and hydraulic traits. Two QTL were identifi ed in the father map and one in the mother map for both stem diameter and height growth increments, explaining up to 16.3 and 17.5 percent of the phenotypic variance, respectively, per QTL. One QTL for total biomass, explaining 10.56% of the phenotypic va riance, was found on linkage group seven in the map of the father, but no QTL for this trait was identified in the map of the mother. One QTL for SLA was identified in each map but on different linkage groups. The QTL for SLA explained 23 percent of the phenotypic variance when mapped in the maternal parent (this was the strongest QTL mapped), but only 10 percent in the paternal map.

PAGE 66

66 Several QTL for the hydraulic traits were mapped. A significant QTL in linkage group three of the father map was identif ied for K t Interestingly this QTL may be co localized with D h and diameter growth increment, although this is not definitive due to the location of the QTL at the end of the linkage group (Fig. 3 4). In the mother map, multiple QTL in different linkage groups were identified for several hydraulic traits (Table 3 2, Fig 3 5 ). Most notably, in linkage group one, QTL for D INC H INC D h K t K S and K L were identified, and QTL for D INC H INC and D h co localized (Table 3 2, Fig 3 6 ). On linkage group six, VSA and KS are co localized. Also, on linkage group 8 in the mother map, QTL for D h and K S co localized (Table 3 2, Fig 3 5 ). This QTL for D h explained 19.24 percent of the phenotypic variance and was the strongest QTL found in the mother map. The comb ined total percent of phenotypic variance explained by the two QTL for D h was 38.11. QTL for VSA K t K S and K L explained 33.70, 23.88, 34.11 and 25.58 percent of the phenotypic variance, respectively. D iscussion Plant Hydraulics and Growth In the 100 pro geny of the study population, hydraulic conductivity was under moderate genetic control, with for specific hydraulic conductivity ranging from 0.40 to 0.47 (Table 3 2). Other studies have found clonal differences in hydraulic condu ctivity in angiosperm trees (Wikberg & Ogren, 2004; Samuelso n et al. 2007) but no reports of heritability or QTL analysis have yet been reported. Although only two QTL were found for D h the percent of the phenotypic variance explained was the highest per QTL than any other trait, except SLA (Table 3 2). The s trong phenotypic correlation between K L and D INC (Fig. 3 1), as well as multiple QTL for specific hydraulic conductivity traits and co localization between D h and D INC and K t and D INC (Table 3 2, Figs. 3 4, 3 5, and 3 6 ) show that stem growth and hydrauli c traits are under

PAGE 67

67 common genetic control, which strongly supports (Tyree, 2003) that high hydraulic conductivity is a prerequisite for fast growth. Our results in Populus are supported by previous work in gymnosperms that showed genetic variation (heritability) in xylem hydraulic conduct ivity, as well as phenotypic and genetic corr elations with stem diameter (Anekonda et al. 2002; Rosner et al. 2007) The large variation in D h and VSA among the progeny was not surprising due to differences in the environmental distribution of the two species in the parents. The P. trichocarpa material originated from a wetter, milder climate, than th e P. deltoides However, between the parents, D h was only marginally different while VSA did not differ. For D h, the mean of the progeny was between the parental means, and the progeny mean for VSA was higher, although perhaps not significantly, than tha t of the parents. The multiple QTL for D h and VSA explained approximately two fifths and a third, respectively, of the phenotypic variance. This finding is similar to another study of 17 clones from mass selected Eucalyptus globulus trials, found clonal variation accounted for 30% of the total variation of vessel proportion (Leal et al. 2003) Strong correlations between xylem anatomy and productivity traits, such as D h with D INC (Fig. 3 2), suggest that when water is not limited vascular anatomy drives stem growth Generally, investment in larger diameter vessels results in higher stem growth and higher hydraulic conductivity. The co localization of QTL for sapwood specific hydraulic conduct ivity and mean hydraulic vessel diameter (Table 3 2, Fig. 3 5 ) supports the idea that xylem structure in mesic species, such as poplars, favors efficiency (wider vessels) rather than safety (Kocacinar & Sage, 2003) Other wood traits, such as wood specific gravity have show n strong genetic

PAGE 68

68 control and overlapping QTL with growth in Eucalyptus (Grattapaglia et al. 1996) but this was not the case in hybrid pine (Shepherd et al. 2003) Carbon Isotope Composition and Leaf Nutrient Con centration in 13 C was found, it was not heritable in the progeny population, with of 0.07 0. 06 and no significant QTL were identified for this trait. This result is contrary to another well watered study with 29 young P. deltoides x P. nigra genotypes, where broad (Monclus et al. 2005) Expected genotypic variation in 13 C may be due to differences in stomatal conductance or photosynthetic capacity or both (Farquhar et al. 1989) While we did not quantify these traits directly, the significant variation for foliar %N among the progeny suggests photosynthetic capacity is varyi ng in this experiment. Foliar nitrogen content is strongly correlated with photosynthetic capacity as a large portion is incorporated into Rubisco, and photosynthetic capacity varies strongly with foliar %N in Populus (Lambers et al. 1998; Cooke et al. 2005) However, this did not translate to genetic control of 13 C in our population grown with ample nitrogen. It is therefore likely that stomatal conductance is co varying proportionally with photosynthetic capacity in the progeny, producing low genetic variation in 13 C. This suggests that WUE in Populus may be an inducible trait that responds to environmental cues, rather than being under strong genetic control. Also, because foliar %N but not %C varied significantly in the study population, the significant variation among the progeny and parents for foliar C:N is due to variation in %N S pecific Leaf Area, Huber Value Hydraulics and Growth Significant genetic variation for SLA and LA:SA in the progeny, as well as moderate to strong implies these traits are also under genetic contro l, but QTL were identified for SLA

PAGE 69

69 (Table 3 2), and not for LA:SA A previous study of two varieties of Metrosideros polymorpha (Myrtaceae), from different geographic ranges in the Hawaiian islands, found significant intraspecific variation in SLA but not for LA:SA (Hoof et al. 2007) Producing more wood per unit leaf area (lower LA:SA ) has been considered to improve plant nutrient and water storage capacity (Ca llaway et al. 1994) This is supported by the significant negative phenotypic correlation ( r 2 = 0.4664, P <0.0001) between LA:SA and D INC Our results showing biomass and SLA are negatively correlated (Fig. 3 3) support a previous study in Populus deltoi des x Populus nigra clones (Marron et al. 2005) Thus, low SLA characterized highly productive genotypes in our study. This relationship was attributed to increased density or size of mesophyll cells in low SLA leaves resulting in high CO 2 a ssimilation and th us high productivity (Marron et al. 2005) However, other studies in cottonwoods showed no correlation between the SLA and biomass (Rae et al. 2004; Monclus et al. 2005) Additionally, QTL found for SLA and total biomass did not co localize. C onclusion In conclusion, the moderate heritability and significant QTL for vessels per sapwood area, hydraulic vessel diameter, and specific hydrau lic conductivity suggests these physiological traits are under genetic control in Populus Our results are the first to map QTL for hydraulic conductivity in an angiosperm tree species. Co localizations between hydraulic traits and growth increment provid Overall, the genetic analysis revealed QTL that co localize for multiple traits, suggesting a locus with pleiotropic effects. At the same time, sin ge trait specific QTL occur acr oss linkage groups. Future research should include a plan to identify candidate genes within identified QTL intervals for conductivity and vessel anatomy traits. Increasing the number of genotypes in the study population may result in observation of greater genetic variation, in 13 C for example.

PAGE 70

70 Table 3 1 Trait means, standard errors ( SE ), significance levels and ranges for 100 progeny genotypes and parental genotypes, Populus trichocarpa x P. deltoides ( P. t x d ) and P. deltoide s Progeny P. t x d P. deltoides Abbreviation Trait (units) Mean (SE) P Range Mean (SE) P Mean (SE) D INC Diameter increment growth (mm) 1.78 ( 0.07 ) *** 0.32 3.52 1.88 ( 1.01 ) 1.81 ( 0.62 ) H INC Height increment growth (cm) 48.08 ( 1.12 ) *** 14.90 75.50 41.23 ( 10.24 ) 57.77 ( 13.38 ) SLA Specific leaf area (m 2 g 1 ) 0.0252 ( 0.0005 ) ** 0.0165 0.0400 0.0234 ( 0.0036 ) 0.0298 ( 0.0108 ) LA Total leaf area (m 2 ) 0.164 ( 0.005 ) ** 0.067 0.290 0.168 ( 0.051 ) 0.115 ( 0.038 ) LA:SA Huber value (m 2 cm 2 ) 1.47 ( 0.06 ) *** 0.64 3.68 1.15 ( 0.13 ) 0.96 ( 0.07 ) SA Sapwood area (mm 2 ) 12.95 ( 0.56 ) *** 4.18 33.66 14.44 ( 4.47 ) 12.64 ( 4.59 ) D h Mean hydraulic vessel diameter ( m) 34.19 ( 0.26 ) *** 26.47 39.56 36.65 ( 1.56 ) + 31.41 ( 1.18 ) VSA Vessels per sapwood area (count mm 2 ) 1 59.50 ( 1.95 ) *** 100.52 214.49 148.05 ( 12.50 ) 146.94 ( 14.92 ) K t x 10 4 Theoretical HC (kg m Mpa 1 s 1 ) 0.89 ( 0.06 ) ** 0.12 3.23 1.11 ( 0.43 ) 0.51 ( 0.19 ) K S Sapwood specific HC (kg m 1 Mpa 1 s 1 ) 6.25 ( 0.17 ) *** 1.86 11.19 7.32 ( 0.65 ) ** 3.90 ( 0. 32 ) K L x 10 4 Leaf specific HC (kg m 1 Mpa 1 s 1 ) 5.34 ( 0.27 ) *** 1.23 13.91 6.50 ( 0.85 ) + 4.14 ( 0.51 ) 13 C Foliar carbon isotope composition ( ) 31.27 ( 0.10 ) 33.81 27.85 31.01 ( 0.52 ) 31.93 ( 0.60 ) %N Foliar percent nitrogen (%) 4.29 ( 0.04 ) ** 3.14 5.25 4.06 ( 0.28 ) + 4.93 ( 0.17 ) %C Foliar percent carbon (%) 47.54 ( 0.11 ) 44.10 50.67 48.48 ( 0.51 ) 46.09 ( 0.56 ) C:N Foliar carbon to nitrogen ratio 11.32 ( 0.12 ) ** 8.99 15.61 12.07 ( 0.94 ) + 9.38 ( 0.30 ) BIO Total plant biomass (g) 14 .06 ( 0.62 ) *** 5.72 34.23 14.16 ( 5.45 ) 11.01 ( 4.79 ) ROOT:SHOOT Root to shoot ratio (g g 1 ) 0.0827 ( 0.0027 ) *** 0.0408 0.1764 0.1003 ( 0.0254 ) 0.0699 ( 0.0171 ) P values from ANOVA results for significant differences in trait means between the pare nts and among the progeny are indicated by: + P P P P

PAGE 71

71 Table 3 2 Trait within family broad sense heritabilit ies ( ) and significant QTL detected in the mo ther, Populus trichocarpa x P. deltoides ( P. t x d ), and father, P. deltoide s maps. Progeny P. t x d map P. deltoides map Trait (SE) LG Loc LOD PVE LG Loc LOD PVE D INC 0.3916 ( 0.0771 ) 1 4.12 3.05 16.28 3 4 2.09* 0.60 2.76 3.1 4 9.71 15.64 H INC 0.3117 ( 0.0707 ) 1 4.22 3.14 17.46 4 16 0.46 0.11 2.65 4.24 13.70 15.03 SLA 0.1476 ( 0.0735 ) 13 1.06 4.89 23.29 1b 0.06* 2.73 10.02 LA:SA 0.4304 ( 0.0635 ) D h 0.5347 ( 0.060 8 ) 1 8 4.08 0.22 4.93 5.98 19.24 18.87 3 2.09* 2.52 9.32 VSA 0.4573 ( 0.0729 ) 3 6a 9 0.00* 0.22 1.23* 3.05 3.02 3.97 9.87 10.28 13.54 K t 0.2958 ( 0.0548 ) 1 4 1.45 1.83 3.26 3.17 11.28 12.60 3 2.09* 2.59 9.94 K S 0.4656 ( 0.0616 ) 1 6a 8 1.05 0.04 0.22 3.41 3.32 3.40 12.29 11.64 10.18 K L 0.4047 ( 0.0618 ) 1 16 1.44! 0.50 3.63 2.95 12.62 12.96 13 C 0.0727 ( 0.0593 ) %N 0.2522 ( 0.0640 ) 2 8 9 17 1.85 2.01! 1.23* 0.14 3.31 3.47 3.90 3.65 11.31 1 2.15 11.93 10.85 %C 0.1127 ( 0.0595 ) C:N 0.2417 ( 0.0685 ) 2 17 1.83 0.14 3.65 4.08 13.29 12.53 BIO 0.2945 ( 0.0656 ) 7 0.68 2.64 10.56 ROOT:SHOOT 0.3904 ( 0.0736 ) Standard errors are italicized in parentheses next to each heritability. LG, linkage group (chromosome); Loc, location of QTL in M organs; LOD, log of odds score; PVE, percent phenotypic variance explained by the QTL. See Table 1 for trait definitions. QTL is located at one end of the chromosome QTL had a double peak above the threshold line; location, LOD and PVE are the values f or the peak with the highest LOD. No significant QTL detected.

PAGE 72

72 Fig. 3 1 Leaf specific hydraulic conductivity ( K L ) scales positively with diameter increment ( D INC ). Points are progeny genotype means. Line is a line ar regression with r 2 = 0.4839, P <0.0001. Fig. 3 2 Hydraulic vessel diameter ( D h ), open circles, scales positively ( r 2 = 0.4172, P <0.0001) and vessels per sapwood area ( VSA ), filled circles, scales negatively ( r 2 = 0.1510, P <0.0001) with diameter increment ( D INC ). Points are progeny genotype means.

PAGE 73

73 Fig. 3 3 Specific leaf area ( SLA ) scales negatively with total biomass ( BIO ). Points are progeny genotype means. Line is a line ar regression with r 2 = 0.4682, P <0.0001.

PAGE 74

74 LG 1b LG 3 LG 4 LG 7 LG 16 Fig. 3 4 QTL located in the Father Map All traits with significant ( P <0.05) QTL LG, linkage group. LOD, log of odds score.

PAGE 75

75 LG 1 LG 2 LG3 LG4 LG 6 LG8 LG9 LG13 LG16 LG17 Fig. 3 5 QTL located in the Mother Map. All traits with significant ( P <0.05) QTLs. LG, linkage group. LOD, log of odds score.

PAGE 76

76 LG 1 Fig. 3 6 Significant (P<0.05) QTL for hydraulic conductivity and growth increment traits on linkage group (LG) 1 in the Mother Map LG, linkage group. LOD, log of odds score.

PAGE 77

77 CHAPTER 4 SUMMARY Genetic studies of wood properties and growth in tree species have been substantial, and have determined genetic contro l and mapped QTL for several traits. The genetic control of WUE in trees has been less adequately covered. Genetic studies on stem hydraulic properties are currently inadequate. In order to more fully understand the relationships between hydraulic condu ctivity, WUE and growth in trees, as well as possible genetic control and QTL two studies were conducted in order to give a more comprehensive examination of these traits within a genetic framework The first study examined stem hydraulic traits, foliar carbon isotope composition and growth in 22 gentoypes of a pseudo backcross population of P. deltoides with an F1 hybrid of P. trichocarpa x deltoides T heoretical hydraulic conductivity by vessel area measurement provided a conservative and consistent e stimate of measured hydraulic conductivity. The significant genetic variation found for vessels per sapwood area, hydraulic vessel diameter, and specific hydraulic conductivity suggests these physiological traits are under genetic control in Populus and our results are the first to show genetic control of hydraulic con ductivity in angiosperm trees. The second study examined stem hydraulic traits, foliar carbon isotope composition and growth in 100 gentoypes of the same pseudo backcross population of P. de ltoides with an F1 hybrid of P. trichocarpa x deltoides T he moderate heritability and significant QTL for vessels per sapwood area, hydraulic vessel diameter, and specific hydraulic conductivity suggests these physiological traits are under genetic contr ol in Populus Our results are the first to map QTL for hydraulic conductivity in an angiosperm tree species. Co localizations between hydraulic traits

PAGE 78

78 hypothesis. Ov erall, the genetic analysis revealed QTL that co localize for multiple traits, suggesting a locus with pleiotropic effects. At the same time, singe trait specific QTL occur accoss linkage groups. Future research should include a plan to identify candidat e genes within identified QTL intervals for conductivity and vessel anatomy traits. Increasing the number of genotypes in the study population may result in observation of greater genetic variation, in 13 C for example.

PAGE 79

79 LIST OF REFERENCES Ackerly DD, Monson RK 2003 Waking the sleeping giant: the evolutionary foundations of plant function. Int. J. Plant Sci. 164(3 Suppl.) :S1 S6. Anekonda TS, Lomas MC, Adams WT, Kavanagh KL, Aitken SN 2002 Genetic variation in drought hardiness o f coastal Douglas fir seedlings from British Columbia. Canadian Journal of Forest Research Revue Canadienne De Recherche Forestiere 32 :1701 1716. Angeles G, Bond B, al. e 2004 The Cohesion Tension Theory. New Phytologist 163 :451 452. Arcade A, Faivre R ampant P, Paques LE, Prat D 2002 Localisation of genomic regions controlling microdensitometric parameters of wood characteristics in hybrid larches. Annals of Forest Science 59 :607 615. Atkinson CJ, Else MA, Taylor L, Dover CJ 2003 Root and stem hydr aulic conductivity as determinants of growth potential in grafted trees of apple ( Malus pumila Mill.). Journal of Experimental Botany 54 :1221 1229. Blake TJ, Tchaplinski TJ, Eastham A 1984 Stomatal control of water use efficiency in poplar clones and hy brids. Canadian Journal of Botany 62 :1344 1351. Borghetti M, Leonardi S, Raschi A, Snyderman D, Tognetti R 1993 Ecotypic variation of xylem embolism, phenological traits, growth parameters and allozyme characteristics in Fagus sylvatica Functional Ecol ogy 7 :713 720. Braatne JH, Hinckley TM, Stettler RF 1992 Influence of soil water on the physiological and morphological components of plant water balance in Populus trichocarpa Populus deltoides and their F1 hybrids. Tree Physiology 11 :325 339. Brende l O, Pot D, Plomion C, Rozenberg P, Guehl JM 2002 Genetic parameters and QTL analysis of 13 C and ring width in maritime pine. Plant Cell and Environment 25 :945 953. Brodribb TJ, Field TS 2000 Stem hydraulic supply is linked to leaf photosynthetic cap acity: evidence from New Caledonian and Tasmanian rainforests. Plant, Cell & Environment 23 :1381 1388. Brodribb TJ, Holbrook NM, Gutierrez MV 2002 Hydraulic and photosynthetic co ordination in seasonally dry tropical forest trees. Plant, Cell & Environm ent 25 :1435 1444. Brown GR, Bassoni DL, Gill GP, Fontana JR, Wheeler NC, Megraw RA, Davis MF, Sewell MM, Tuskan GA, Neale DB 2003 Identification of quantitative trait loci influencing wood property traits in loblolly pine ( Pinus taeda L.). III. QTL veri fication and candidate gene mapping. Genetics 164 :1537 1546.

PAGE 80

80 Brugnoli E, Farquhar GD 2000 Photosynthetic fractionation of carbon isotopes. In: Photosynthesis: physiology and metabolism. Advances in Photosynthesis. Eds. Leegood RC, Sharkey TDCaemmerer Sv Kluwer Academic Publishers, Dordrecht. Brunner AM, Busov VB, Strauss SH 2004 Poplar genome sequence: functional genomics in an ecologically dominant plant species. Trends in Plant Science 9 :49 56. Burns RM, Honkala BH 1990 Silvics of North America: 1. Conifers; 2. Hardwoods. Agriculture Handbook 654. [Online book]. Available from: www.na.fs.fed.us/spfo/pubs/silvics_manual/table_of_contents.htm U.S. Department of Agriculture Forest Service, Washington, DC. 877 p. Callaway RM, Delucia EH, Schlesinger WH 1994 Biomass allocation of montane and desert Ponderosa pine an analog for response to climate change. Ecology 75 :1474 1481. Campanello PI, Gatti MG, Goldstein G 2008 Coordination between water transport efficiency and photosynthetic capacity in canopy tree species at different growth irradiances. Tree Physiology 28 :85 94. Canny M 1995 A new theory for the ascent of sap: cohesion supported by tissue pressur e. Annals of Botany 75 :343 357. Christensen Dalsgaard KK, Fournier M, Ennos AR, Barfod AS 2007 Changes in vessel anatomy in response to mechanical loading in six species of tropical trees. New Phytologist 176 :610 622. Comstock JP, McCouch SR, Martin BC Tauer CG, Vision TJ, Xu Y, Pausch RC 2005 The effects of resource availability and environmental conditions on genetic rankings for carbon isotope discrimination during growth in tomato and rice. Functional Plant Biology 32 :1089 1105. Condon AG, Richa rds RA, Rebetzke GJ, Farquhar GD 2004 Breeding for high water use efficiency. Journal of Experimental Botany 55 :2447 2460. Cooke JEK, Martin TA, Davis JM 2005 Short term physiological and developmental responses to nitrogen availability in hybrid popl ar. New Phytologist 167 :41 52. Craig H 1954 Carbon 13 in plants and the relationship between carbon 13 and carbon 14 variations in nature. J. Geol. 62 :115 149. Cronk QCB 2005 Plant eco devo: the potential of poplar as a model organism. New Phytologist 166 :39 48. Devey ME, Carson SD, Nolan MF, Matheson AC, Riini CT, Hohepa J 2004 QTL associations for density and diameter in Pinus radiata and the potential for marker aided selection. Theoretical and Applied Genetics 108 :516 524.

PAGE 81

81 Domec J C, Meinzer FC Lachenbruch B, Housset J 2007 Dynamic variation in sapwood specific conductivity in six woody species. Tree Physiology 27 :1389 1400. Ehleringer JR, Osmond CB 1989 Stable isotopes. In: Plant physiological ecology: field methods and instrumentation. C hapman & Hall, London. Farquhar GD, Ehleringer JR, Hubick KT 1989 Carbon isotope discrimination and photosynthesis. Annual Review of Plant Physiology and Plant Molecular Biology 40 :503 537. Farquhar GD, O'Leary HH, Berry JA 1982 On the relationships b etween carbon isotope discrimination and intercellular carbon dioxide concentrations in leaves. Australian Journal of Plant Physiology 9 :121 137. Farquhar GD, Richards RA 1984 Isotopic composition of plant carbon correlates with water use efficiency of wheat genotypes. Australian Journal of Plant Physiology 11 :539 552. Frewen BE, Chen THH, Howe GT, Davis J, Rohde A, Boerjan W, Bradshaw HD 2000 Quantitative trait loci and candidate gene mapping of bud set and bud flush in Populus Genetics 154 :837 845 Geber MA, Dawson TE 1997 Genetic variation in stomatal and biochemical limitations to photosynthesis in the annual plant, Polygonum arenastrum Oecologia 109 :535 546. Gielen B, Ceulemans R 2001 The likely impact of rising atmospheric CO 2 on natural and managed Populus : a literature review. Environmental Pollution 115 :335 358. Gilmour AR, Gogel BJ, Cullis BR, Welham SJ, Thompson R 2007 ASReml User Guide Release 2.0. VSN International Ltd, Hemel Hempstead,UK. Grattapaglia D, Bertolucci FLG, Penchel R, Sederoff RR 1996 Genetic mapping of quantitative trait loci controlling growth and wood quality traits in Eucalyptus grandis using a maternal half sib family and RAPD markers. Genetics 144 :1205 1214. Griffiths H 1991 Applications of stable isotope technology in physiological ecology. Functional Ecology 5 :254 269. Hacke UG, Sperry JS 2001 Functional and ecological xylem anatomy. Perspectives in Plant Ecology Evolution and Systematics 4 :97 115. Hacke UG, Sperry JS, Wheeler JK, Castro L 2006 Scal ing of angiosperm xylem structure with safety and efficiency. Tree Physiology 26 :689 701. Hartl DL, Clark AG 1997 Principles of population genetics. Sinauer Associates, Inc., Canada. 542 p. Harvey HP, Driessche Rvd 1999 Nitrogen and potassium effects on xylem cavitation and water use efficiency in poplars. Tree Physiology 19 :943 950.

PAGE 82

82 Harvey HP, van den Driessche R 1997 Nutrition, xylem cavitation and drought resistance in hybrid poplar. Tree Physiology 17 :647 654. Hocking D 1971 Preparation and u se of a nutrient solution for culturing seedlings of lodgepole pine and white spruce with selected bibliography. In: Northern Forestry Research Centre Information Report Nor X L. Canadian Forest Service, Department of the Environment, Edmonton, Canada. Hol brook NM, Zwieniecki MA 2005 Vascular transport in plants. In Physiological Ecology Series Ed. Mooney HA. Elsevier Academic Press, Stanford, California. 564 p. Holbrook NM, Zwieniecki MA, Melcher PJ 2002 The dynamics of "dead wood": maintenance of wate r transport through plant stems. Integr. Comp. Biol. 42 :492 496. Hoof J, Sack L, Webb DT, Nilsen ET 2007 Contrasting structure and function of pubescent and glabrous varieties of Hawaiian Metrosideros polymorpha (Myrtaceae) at high elevation. Biotropica doi:10.1111/j.1744 7429.2007.00325.x Huber B 1928 Weitere quantitative Untersuchungen ber das Wasserleitungssytem der Pflanzen. Jahrb Wiss Bot 67 :877 959. Hukin D, H. Cochard, E. Dreyer, D. Le Thiec, Bogeat Triboulot MB 2005 Cavitation vulnerability in roots and shoots: does Populus euphratica Oliv., a poplar from arid areas of Central Asia, differ from other poplar species? Journal of Experimental Botany 10 :1093 2000. Ingvarsson PK 2005 Nucleotide polymorphism and linkage disequilibrium within and among natural populations of European Aspen ( Populus tremula L., Salicaceae). Genetics 169 :945 953. Kavanagh KL, Bond BJ, Aitken SN, Gartner BL, Knowe S 1999 Shoot and root vulnerability to xylem cavitation in four populations of Douglas fir seedlings. Tree Physiology 19 :31 37. Kirst M, Myburg AA, De Leon JPG, Kirst ME, Scott J, Sederoff R 2004 Coordinated genetic regulation of growth and lignin revealed by quantitative trait locus analysis of cDNA microarray data in an interspecific backcross of euc alyptus. Plant Physiology 135 :2368 2378. Kocacinar F, Sage RF 2003 Photosynthetic pathway alters xylem structure and hydraulic function in herbaceous plants. Plant Cell and Environment 26 :2015 2026. Lambers H, III FSC, Pons TL 1998 Plant Physiologica l Ecology. Springer Verlag Inc., New York. 540 p. Leal S, Pereira H, Grabner M, Wimmer R 2003 Clonal and site variation of vessels in 7 year old Eucalyptus globulus IAWA Journal 24 :185 195.

PAGE 83

83 Lewis AM, Boose ER 1995 Estimating volume flow rates through xylem conduits. American Journal of Botany 82 :1112 1116. Liu Z, Dickmann DI 1996 Effects of water and nitrogen interaction on net photosynthesis, stomatal conductance, and water use efficiency in two hybrid poplar clones. Physiologia Plantarum 97 :507 512. Lo Gullo MA 1991 Three different methods for measuring xylem cavitation and embolism: a comparison. Ann Bot 67 :417 424. Lo Gullo MA, Salleo S, Piaceri EC, Rosso R 1995 Relations between vulnerability to xylem embolism and xylem conduit dimensions in young trees of Quercus cerris Plant Cell and Environment 18 :661 669. Lovelock CE, Ball MC, Choat B, Engelbrecht BMJ, Holbrook NM, Feller IC 2006 Linking physiological processes with mangrove forest structure: phosphorus deficiency limits canopy dev elopment, hydraulic conductivity and photosynthetic carbon gain in dwarf Rhizophora mangle Plant Cell and Environment 29 :793 802. Machado JL, Tyree MT 1994 Patterns of hydraulic architecture and water relations of two tropical canopy trees with contras ting leaf phenologies Ochroma pyramidale and Pseudobombax septenatum Tree Physiology 14 :219 240. Maherali H, Williams BL, Paige KN, Delucia EH 2002 Hydraulic differentiation of Ponderosa pine populations along a climate gradient is not associated wit h ecotypic divergence. Functional Ecology 16 :510 521. Marron N, Villar M, Dreyer E, Delay D, Boudouresque E, Petit J M, Delmotte FM, Guehl J M, Brignolas F 2005 Diversity of leaf traits related to productivity in 31 Populus deltoides Populus nigra clo nes. Tree Physiology 25 :425 435. Martre P, Durand J L, Cochard H 2000 Changes in axial hydraulic conductivity along elongating leaf blades in relation to xylem maturation in tall fescue. New Phytologist 146 :235 247. McDowell N, Barnard H, Bond BJ, Hinc kley T, Hubbard RM, Ishii H, Kostner B, Magnani F, Marshall JD, Meinzer FC, Phillips N, Ryan MG, Whitehead D 2002 The relationship between tree height and leaf area:sapwood area ratio. Oecologia 132 :12 20. Millard P, Sommerkorn M, Grelet G A 2007 Envi ronmental change and carbon limitation in trees: a biochemical, ecophysiological and ecosystem appraisal. New Phytologist 175 :11 28. Monclus R, Dreyer E, Delmotte FM, Villar M, Delay D, Boudouresque E, Petit J M, Marron N, Brechet C, Brignolas F 2005 Pr oductivity, leaf traits and carbon isotope discrimination in 29 Populus deltoides x P. nigra clones. New Phytologist 167 :53 62.

PAGE 84

84 Netzer DA, Tolsted DN, Ostry ME, Isebrands JG, Riemenschneider DE, Ward KT 2002 Growth, yield, and disease resistance of 7 t o 12 year old poplar clones in the north central United States. U.S. Dept. of Agriculture, Forest Service, North Central Research Station, St. Paul, Minnesota, p. 33. O'Leary MH 1988 Carbon isotopes in photosynthesis. BioScience 38 :325 336. Osorio J, Pe reira JS 1994 Genotypic differences in water use efficiency and 13 C discrimination in E. globulus Tree Physiology 14 :871 882. Parelle J, Zapater M, Scotti Saintagne C, Kremer A, Jolivet Y, Dreyer E, Brendel O 2007 Quantitative trait loci of tolerance to waterlogging in a European oak ( Quercus robur L.): physiological relevance and temporal effect patterns. Plant Cell and Environment 30 :422 434. Pickard WF 1981 The ascent of sap in plants. Progress in biophysics and molecular biology 37 :181 229. Po nton S, Dupouey J L, Breda N, Dreyer E 2002 Comparison of water use efficiency of seedlings from two sympatric oak species: genotype environment interactions. Tree Physiology 22 :413 422. Ponton S, Dupouey J L, Breda N, Feuillat F, Bodenes C, Dreyer E 2001 Carbon isotope discrimination and wood anatomy variations in mixed stands of Quercus robur and Quercus petraea Plant, Cell & Environment 24 :861 868. Pot D, Rodrigues JC, Rozenberg P, Chantre G, Tibbits J, Cahalan C, Pichavant F, Plomion C 2006 Q TLs and candidate genes for wood properties in maritime pine ( Pinus pinaster Ait.). Tree Genetics & Genomes 2 :10 24. Prasolova NV, Xu ZH, Lundkvist K 2005 Genetic variation in foliar nutrient concentration in relation to foliar carbon isotope compositio n and tree growth with clones of the F1 hybrid between Slash pine and Caribbean pine. Forest Ecology and Management 210 :173 191. Prasolova NV, Xu ZH, Lundkvist K, Farquhar GD, Dieters MJ, Walker S, Saffigna PG 2003 Genetic variation in foliar carbon iso tope composition in relation to tree growth and foliar nitrogen concentration in clones of the F1 hybrid between Slash pine and Caribbean pine. Forest Ecology and Management 172 :145 160. Rae AM, Pinel MPC, Bastien C, Sabatti M, Street NR, Tucker J, Dixon C, Marron N, Dillen SY, Taylor G 2008 QTL for yield in bioenergy Populus : identifying GE interactions from growth at three contrasting sites. Tree Genetics & Genomes 4 :97 112. Rae AM, Robinson KM, Street NR, Taylor G 2004 Morphological and physiologi cal traits influencing biomass productivity in short rotation coppice poplar. Canadian Journal of Forest Research 34 :1488 1498.

PAGE 85

85 Richards RA, Rebetzke GJ, Condon AG, van Herwaarden AF 2002 Breeding opportunities for increasing the efficiency of water use and crop yield in temperate cereals. Crop Science 42 :111 121. Ridolfi M, Dreyer E 1997 Responses to water stress in an ABA unresponsive hybrid poplar ( Populus koreana x trichocarpa cv. Peace). III. Consequences for photosynthetic carbon assimilation. N ew Phytologist 135 :31 40. Rood SB, Braatne JH, Hughes FMR 2003 Ecophysiology of riparian cottonwoods: stream flow dependency, water relations and restoration. Tree Physiology 23 :1113 1124. Rosner S, Klein A, Muller U, Karlsson B 2007 Hydraulic and me chanical properties of young Norway spruce clones related to growth and wood structure. Tree Physiology 27 :1165 1178. Samuelson LJ, Stokes TA, Coleman MD 2007 Influence of irrigation and fertilization on transpiration and hydraulic properties of Populus deltoides Tree Physiology 27 :765 774. Sewell MM, Bassoni DL, Megraw RA, Wheeler NC, Neale DB 2000 Identification of QTLs influencing wood property traits in loblolly pine ( Pinus taeda L.). I. Physical wood properties. Theoretical and Applied Genetics 101 :1273 1281. Sewell MM, Davis MF, Tuskan GA, Wheeler NC, Elam CC, Bassoni DL, Neale DB 2002 Identification of QTLs influencing wood property traits in loblolly pine ( Pinus taeda L.). II. Chemical wood properties. Theoretical and Applied Genetics 104 : 214 222. Shepherd M, Cross M, Dieters MJ, Harding K, Kain D, Henry R 2003 Genetics of physical wood properties and early growth in a tropical pine hybrid. Canadian Journal of Forest Research Revue Canadienne De Recherche Forestiere 33 :1923 1932. Shinoza ki K, Yoda K, Hozumi K, Kira T 1964 A quantitative analysis of plant form: the pipe model theory. I. Basic analysis. Jpn J Ecol 14 :97 105. Solla A, Gil L 2002 Xylem vessel diameter as a factor in resistance of Ulmus minor to Ophiostoma novo ulmi Fore st Pathology 32 :123 134. Sparks JP, Black RA 1999 Regulation of water loss in populations of Populus trichocarpa : the role of stomatal control in preventing xylem cavitation. Tree Physiology 19 :453 459. Sperry JS 2000 Hydraulic constraints on plant g as exchange. Agricultural and Forest Meteorology 104 :13 23. Sperry JS 2003 Evolution of water transport and xylem structure. International Journal of Plant Sciences 164 :S115 S127. Sperry JS, Donnelly JR, Tyree MT 1988 A method for measuring hydraulic conductivity and embolism in xylem. Plant Cell and Environment 11 :35 40.

PAGE 86

86 Sperry JS, Hacke UG 2002 Desert shrub water relations with respect to soil characteristics and plant functional type. Functional Ecology 16 :367 378. Sperry JS, Hacke UG, Oren R, Comstock JP 2002 Water deficits and hydraulic limits to leaf water supply. Plant Cell and Environment 25 :251 263. Sperry JS, Hacke UG, Wheeler JK 2005 Comparative analysis of end wall resistivity in xylem conduits. Plant Cell and Environment 28 :456 4 65. Sperry JS, Nichols KL, Sullivan JEM, Eastlack SE 1994 Xylem embolism in ring porous, diffuse porous, and coniferous trees of northern Utah and interior Alaska. Ecology 75 :1736 1752. Sperry JS, Saliendra NZ, Pockman WT, Cochard H, Cruiziat P, Davis S D, Ewers FW, Tyree MT 1996 New evidence for large negative xylem pressures and their measurement by the pressure chamber method. Plant Cell and Environment 19 :427 436. Stettler RF, Bradshaw HD, Heilman PE, Hinckley TM 1996 Biology of Populus and its i mplications for management and conservation. Ottawa, Ontario, Canada: NRC Research Press. Tagu D, Bastien C, Faivre Rampant P, Garbaye J, P.Vion, Villar M, Martin F 2005 Genetic analysis of phenotypic variation for ectomycorrhiza formation in an interspe cific F1 poplar full sib family. Mycorrhiza 15 :87 91. Taylor G 2002 Populus : Arabidopsis for forestry. Do we need a model tree? Annals of Botany 90 :681 689. Tschaplinski TJ, Tuskan GA, Gunderson CA 1994 Water stress tolerance of black and eastern cot tonwood clones and four hybrid progeny. 1. Growth, water relations, and gas exchange. Canadian Journal of Forest Research 24 :364 371. Tuskan GA, DiFazio S, Jansson S, Bohlmann J, Grigoriev I, Hellsten U, Putnam N, Ralph S, Rombauts S, Salamov A, Schein J, Sterck L, Aerts A, al. e 2006 The Genome of Black Cottonwood, Populus trichocarpa (Torr. & Gray). Science 313 :1596 1604. Tyree MT 1997 Cohesion Tension theory of sap ascent: current controversies. Journal of Experimental Botany 10 :1753 1765. Tyree M T 2003 Hydraulic limits on tree performance: transpiration, carbon gain and growth of trees. Trees 17 :95 100. Tyree MT, Alexander J, Machado J L 1992 Loss of hydraulic conductivity due to water stress in intact juveniles of Quercus rubra and Populus d eltoides Tree Physiology 10 :411 415. Tyree MT, Snyderman DA, Wilmot TR, Machado JL 1991 Water Relations and Hydraulic Architecture of a Tropical Tree ( Schefflera morototoni ) Data, Models, and a

PAGE 87

87 Comparison with 2 Temperate Species ( Acer saccharum and Thuja occidentalis ). Plant Physiology 96 :1105 1113. Tyree MT, Sperry JS 1988 Do woody plants operate near the point of catastrophic xylem dysfunction caused by dynamic water stress? Plant Physiology 88 :574 580. Tyree MT, Sperry JS 1989 Vulnerability of xylem to cavitation and embolism. Annual Review of Plant Physiology and Plant Molecular Biology 40 :19 36. Tyree MT, Zimmerman MH 2002 Xylem Stucture and the Ascent of Sap. New York: Springer Verlag. van Ieperen W, van Meeteren U, van Gelder H 2000 Fluid ionic composition influences hydraulic conductance of xylem conduits. Journal of Experimental Botany 51 :769 776. Villar Salvador P, Castro Diez P, Perez Rontome C, Montserrat Marti G 1997 Stem xylem features in three Quercus (Fagaceae) species alo ng a climatic gradient in NE Spain. Trees 12 :90 96. Vision TJ, Brown DG, Shmoys DB, Durrett RT, Tanksley SD 2000 Selective Mapping: A Strategy for Optimizing the Construction of High Density Linkage Maps. Genetics 155 :407 420. Wang S, Basten CJ, Zeng Z B 2007 Windows QTL Cartographer 2.5. Department of Statistics, North Carolina State University. Raleigh, NC. http://statgen.ncsu.edu/qtlcart/WQTLCart.htm Waring R, Schroeder P, Oren R 1982 Application of the pipe model theory to predict canopy leaf area. Canadian Journal of Forest Research 12 :556 560. Weih M, Ronnberg Wastljung A C, Glynn C 2006 Genetic basis of phenotypic correlations among growth traits in hybrid willow ( Salix dasycla dos x S. viminalis ) grown under two water regimes. New Phytologist 170 :467 477. Wheeler JK, Sperry JS, Hacke UG, Hoang N 2005 Inter vessel pitting and cavitation in woody Rosaceae and other vesselled plants: a basis for a safety versus efficiency trade off in xylem transport. Plant Cell and Environment 28 :800 812. Whitehead D, Jarvis PG 1981 Coniferous forests and plantations. In: Water deficits and plant growth. Ed. Kozlowski TT. Academic Press, New York, p. 49 152. Wikberg J, Ogren E 2004 Interrel ationships between water use and growth traits in biomass producing willows. Trees 18 :70 76. Williams CA, Cooper DJ 2005 Mechanisms of riparian cottonwood decline along regulated rivers. Ecosystems 8 :382 395.

PAGE 88

88 Wong SC, Cowan IR, Farquhar GD 1979 Stoma tal conductance correlates with photosynthetic capacity. 282 :424 426. Wu R, Bradshaw HD, Stettler RF 1997 Molecular genetics of growth and development in Populus (Salicaceae). 5. Mapping quantitative trait loci affecting leaf variation. American Journal of Botany 84 :143 153. Wu R, Bradshaw HD, Stettler RF 1998 Developmental quantitative genetics of growth in Populus Theoretical and Applied Genetics 97 :1110 1119. Wu RL 1998 Genetic mapping of QTLs affecting tree growth and architecture in Populus : i mplication for ideotype breeding. Theoretical and Applied Genetics 96 :447 457. Wullschleger S, Yin TM, DiFazio SP, Tschaplinski TJ, Gunter LE, Davis MF, Tuskan GA 2005 Phenotypic variation in growth and biomass distribution for two advanced generation p edigrees of hybrid poplar. Canadian Journal of Forest Research Revue Canadienne De Recherche Forestiere 35 :1779 1789. Wullschleger SD, Jansson S, Taylor G 2002 Genomics and forest biology: Populus emerges as the perennial favorite. Plant Cell 14 :2651 2 655. Xu Z, Zou F, Vision TJ 2005 Improving quantitative trait loci mapping resolution in experimental crosses by the use of genotypically selected samples. Genetics 170 :401 408. Yang SD, Tyree MT 1993 Hydraulic resistance in Acer saccharum shoots and its influence on leaf water potential and transpiration. Tree Physiology 12 :231 242. Zhang D, Zhang Z, Yang K 2006 QTL analysis of growth and wood chemical content traits in an interspecific backcross family of white poplar ( Populus tomentosa x P bollea na ) x P tomentosa Canadian Journal of Forest Research Revue Canadienne De Recherche Forestiere 36 :2015 2023. Zimmermann U, Schneider H, Wegner LH, Haase A 2004 Water ascent in tall trees: does evolution of land plants rely on highly metastable state? N ew Phytologist 162 :575 615. Zwieniecki MA, Melcher PJ, Michele Holbrook N 2001 Hydrogel control of xylem hydraulic resistance in plants. Science 291 :1059 1062.

PAGE 89

89 BIOGRAPHICAL SKETCH Brianna L. Miles was born in 1978 in Takoma Park, Maryland, in the U nited States of America. She spent her early years in Adelphi, MD and then moved to Odenton, MD where she lived until graduating from Arundel High School in 1997. Brianna attended the University of Maryland at College Park, and graduated in 2001 from the College of Agriculture with a Bachelor of Science in environmental science and policy. After graduation, she worked as an intern and later technician at the Smithsonian Environmental Research Center, in Edgewater, MD, in the GIS, Forest Ecology and Ecoph ysiology Labs. In 2003, Brianna moved to Minneapolis, Minnesota, and worked at the University of Minnesota, St. Paul, as a Junior Scientist, under the supervision of Dr. Jeannine Cavender degree at the University of Flori da in 2005. After graduation, Brianna plans to move to Colorado with her husband David and dog Paws, where she will work as a research scientist and continue to explore and expand her understanding of tree structure and function.